pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
34927022 | PMC8649731 | pmc | 7,170 | {
"abstract": "Summary Wearable thermoelectrochemical cells have attracted increasing interest due to their ability to turn human body heat into electricity. Here, we have fabricated a flexible, cost-effective, and 3D porous all-polymer electrode on an electrical conductive polymer substrate via a simple 3D printing method. Owing to the high degree of electrolyte penetration into the 3D porous electrode materials for redox reactions, the all-polymer based porous 3D electrodes deliver an increased power output of more than twice that of the film electrodes under the same mass loading using either n-type or p-type gel electrolytes. To realize the practical application of our thermocell, we fabricated 18 pairs of n-p devices through a series connection of single devices. The strap shaped thermocell arrangement was able to charge up a commercial supercapacitor to 0.27 V using the body heat of the person upon which it was being worn and in turn power a typical commercial lab timer.",
"conclusion": "Conclusions In conclusion, we have fabricated novel wearable multi-unit thermocells assembly through 3D-printing all polymer-based flexible electrodes. We also matched serial thermocells and brought about an open voltage of 0.27 V to charge up a supercapacitor and store the harvested energy when operating with a ΔT = 10°C. Moreover, the demonstration of a wearable thermocell, which can harvest body heat, charge commercial supercapacitors, and even power a lab timer, shows the great potential of our wearable device in practical applications. This work provides a platform for the future development of 3D-printable integrated wearable device systems.",
"introduction": "Introduction With the rapid development of next-generation wearable electronics, such as portable devices and soft electric devices, there is strong demand for lightweight, wearable, and environmentally friendly energy devices, such as piezoelectric nanogenerators, thermoelectric generators and thermoelectrochemical cells ( Dargusch et al., 2020 ; Shi et al., 2020 ; Khan et al., 2021 ; Jia et al., 2021 ; Tian et al., 2019 ; Zhang et al., 2021 ). Human body heat is an accessible, relatively consistent, and environmentally friendly power source with a temperature difference (ΔT) between human skin and ambient environment ( Oh et al., 2016 ; Zhong et al., 2014 ). The most convenient strategy to use this low-grade thermal energy is to convert thermal to electricity. Conventional thermoelectric generators (TEGs) utilizing the Seebeck effect are mostly dependent on thermoelectric (TE) materials such as semiconductors or electrical conducting polymers. These TE materials are either expensive or exhibit low Seebeck Coefficient (S e , S e = ΔV/ΔT, the open voltage is ΔV, and the temperature difference is ΔT) in the range of several hundreds of μVK −1 ( Khan et al., 2016 ; Bux et al., 2010 ), limiting their application in wearable electronics for harvesting of low-grade body heat ( Khan et al., 2016 ). Alternatively, thermoelectrochemical cells (also called thermogalvanic cells or thermocells) can generate a larger thermal voltage, which is caused by a temperature-dependent entropy change during the electron transfer process between the redox couples and the electrode ( Liu et al., 2021 ; Romano et al., 2012 ). The simple device structure, the low-cost of electrode and electrolyte materials, and the relatively large S e has made thermoelectrochemical cells a promising candidate to efficiently harvest low-grade body heat. The performance of thermoelectrochemical cells (TEC) is greatly dependent on the electrode materials employed ( Dupont et al., 2017 ; Hu et al., 2010 ; Im et al., 2014 ). Electrodes in wearable thermoelectrochemical cells should meet a number of requirements including low cost, large surface area, high electrical conductivity, extraordinary flexibility, porous architecture, and high thermal conductivity ( Dupont et al., 2017 ; Hu et al., 2010 ; Im et al., 2016 ). In previous studies, platinum (Pt) electrodes have been conceived as an ideal electrode material for wearable thermocells because of the high electrical conductivity and high catalytic activity of the material ( Im et al., 2016 ; Zhou et al., 2021 ). However, the high-cost of platinum greatly limits its use as a commercially available electrode material. Carbon based materials, including carbon cloth, carbon nanotubes (CNTs) and graphene ( Romano et al., 2013 ), could provide large active surface areas and high electrical conductivities, which lead to large numbers of reaction sites and fast electron transfer kinetics for the redox couples. These advantages could increase the obtainable current density of thermocell devices ( Kang et al., 2012 ). However, there are limitations in the processing of these carbon materials into a homogeneous dispersion and subsequebtly in making electrodes from such dispersions ( Romano et al., 2013 ; Liu et al., 2020 ). Conducting polymers (CPs) are promising materials for wearable electronics because of their high electrical conductivity, intrinsic flexibility, relative low-cost, light weight, and ease of preparation ( Liu et al., 2020 ). The suitability of CPs as electrode materials for thermocell devices has been investigated in recent years. For example, poly (3,4-ethylenedioxythio-phene): polystyrenesulfonate (PEDOT:PSS), one of the most popular CPs, has been reported to provide attractive alternative to platinum electrodes ( Yuk et al., 2020 ). Owing to a low charge transfer resistance, films of PEDOT:PSS showed performance that was comparable to carbon-based materials for a thermocell application ( Wang et al., 2020 ). However, current PEDOT:PSS electrodes are generally in a thin film form prepared via techniques including drop-casting ( Wang et al., 2020 ), ink-jet printing ( Perinka et al., 2013 ), and screen printing ( Sinha et al., 2017 ), in which the penetration of electrolyte, ion transfer rate and ion accessible surface area are greatly limited. As a result, it is hard to further increase the performance (especially the current & power output) of the thermocell using these electrode configurations. To address this issue, a PEDOT:PSS electrode with a 3D porous structure is required. Herein, we developed a PEDOT:PSS-based ink with rheological properties (in terms of viscosity, shear thinning, and shear yielding) desirable for 3D printing, which allows the direct printing of PEDOT:PSS with controlled line spacing and interlayer spacing porosity to produce well defined structures. The printed porous PEDOT:PSS electrode with an interaxial design (the printed direction rotated through 45° for each successive layer) could provide surface and cross-sectional area, which enables a high degree of electrolyte penetration into the electrode for redox reactions to occur and give rise to increased current, open voltage, and power density. Meanwhile, we also find that a thin PEDOT:PSS film prepared via drop-casting technique, can be integrated with the printed structure to act as the current collector. This configuration exhibits superior performance in comparison to conventionally sputter coated platinum films in terms of endurance over repeated bending cycles and thermoelectrochemical performance (i.e., current & power output). Hence, an all polymer-based electrode with a 3D printed porous PEDOT:PSS structure integrated with a thin PEDOT:PSS film, for thermocell applications has been successfully fabricated. In addition, the prepared all polymer electrodes display excellent and comparable performance in both n-type (PVA-FeCl 2/3 ) and p-type (CMC-K 3/4 Fe (CN) 6 ) gel electrolytes, so a matched pair of n-p cell was efficiently achieved. To demonstrate the practical application of our thermocell, we also fabricated 18 pairs of n-p devices connected in series arranged in a strap shaped thermocell assembly. The assembled device was able to charge up a commercial supercapacitor to 0.27 V using the body heat of the person upon which it was being worn and in turn power a typical commercial lab timer.",
"discussion": "Results and discussion Selection criteria of electrical conductive substrates for 3D printing of electrode materials It is widely known that electrode materials in energy devices are usually coated on a conductive thin film ( e.g. platinum, gold, stainless steel, etc), which act as the current collector to collect electrical current generated at the electrodes ( Antiohos et al., 2011 ). In most thermocell studies, sputter coated platinum electrodes have been used due to the high electrical conductivity and high catalytic activity behaviour that platinum provides ( Dupont et al., 2017 ). However, in wearable applications, devices are usually bent over large angles (close to 180°) or even twisted as they conform to the wearer during movement ( Im et al., 2014 ; Liu et al., 2020 ; Yang et al., 2016 ). Here we found sputter coated platinum (Pt) is not stable under these conditions and would peel off from the substrate after 10 bending cycles (see Figure 1 A). As a result, the thermoelectrochemical performance ( i.e., Seebeck coefficient, current density, and power density for short term & long term) in both n-type and p-type electrolyte is greatly affected as shown in Figures 1 B, 1C, and S1 . Hence, sputter coated Pt is not a stable conductive substrate upon which to integrate 3D printed active electrode materials for wearable thermocell devices. To address this problem, we utilized a thin PEDOT:PSS film (2 mg cm −2 ) as an alternative cast conductive substrate for 3D printing. This thin PEDOT:PSS film is reported as a flexible and electrically conductive electrode material for thermocell, which can be easily fabricated via a simple drop-coating technique (by unifying the concentration and then controlling the volume of the drop per unit area)( Liu et al., 2020 ; Wijeratne et al., 2017 ). After an optimization study of PEDOT:PSS films with mass loadings ranging of from 0.5 mg cm −2 to 4 mg cm −2 (see Figures S2 A and S2B and thickness in Table S1 ), the thermocell device made from 2 mg cm −2 PEDOT:PSS film electrode delivered the highest thermoelectrochemical performance in both n and p type electrolytes. The recorded performance was compatible with that observed from a device made from 100 nm Pt electrodes ( Figures 1 B and 1C). Meanwhile, the performance of thermocells made with PEDOT:PSS films were almost the same after multiple bending cycles. Therefore, the PEDOT:PSS film with a mass loading of 2 mg cm −2 was selected for the cast conductive substrate. Figure 1 Selection criteria of electrical conductive substrates for 3D printing of electrode materials (A) Microscope images of 100 nm Pt and 2 mg cm −2 PEDOT: PSS film before and after bended (scale bar size is 100μm). (B) TEC performance: current & power versus voltage of the devices made from 100 nm Pt, bended 100 nm Pt, 2 mg cm −2 PEDOT: PSS film and 2 mg cm −2 bended PEDOT: PSS film in n-type electrolyte (ΔT = 10°C). (C) TEC performance: current and power versus voltage of the devices made from 100 nm Pt, bended 100 nm Pt, 2 mg cm −2 PEDOT: PSS film and 2 mg cm −2 bended PEDOT: PSS film in p-type electrolyte (ΔT = 10°C). Preparation of printable PEDOT: PSS ink To prepare homogenous printable ink dispersions, a commercial PEDOT:PSS pellets was used in this study. PEDOT:PSS pellets and diethylene glycol (DEG) were dispersed in deionized (DI) water with a weight ratio of (PEDOT:PSS):DEG = 1:1.86 and homogenized by a Thinky Mixer (see experimental details and Figure S3 ). Here, DEG is used as a secondary dopant to enhance the electrical conductivity (increasing from 3.2 S cm −1 to 230 S cm −1 ), as well as to induce accessible microscale pores within resultant PEDOT:PSS films ( Liu et al., 2018 ). As the concentration of PEDOT:PSS increased, the prepared suspension transformed from liquid state to a gelled and printable state ( Figure S4 A), a result of the entanglement of PEDOT:PSS fibrils at high concentrations ( Yuk et al., 2020 ). Rheological measurements of the PEDOT:PSS inks demonstrate the transition from low concentration PEDOT:PSS (10 mg mL −1 to 50 mg mL −1 ) with low viscosity (1.75 Pa . s to 2297 Pa . s at 0.01 s −1 shear rate) to high concentration PEDOT:PSS (50 mg mL −1 to 120 mg mL −1 ) with high viscosity (2297 Pa . s to 25,148 Pa . s at 0.01 s −1 shear rate ( Figure S4 B). In addition, all prepared dispersions exhibited a shear-thinning behaviour as well as an increase in shear-yielding stress from 0.90 Pa to 794.5 Pa with increasing PEDOT:PSS concentration from 10 mg mL −1 to 50 mg mL −1 ( Figures S4 and S5 ). For dispersions with low concentration (10 to 50 mg mL −1 ), the 3D printed ink would spread laterally on the substrate upon printing because of the low viscosity and yield stress. However, the use of concentration of PEDOT:PSS above 100 mg mL −1 would not continuously flow through an extrusion nozzle resulting in to clogging a printing nozzle (see Figure S5 F) due to aggregated PEDOT:PSS ( Yuk et al., 2020 ). The electrical conductivity of the thermocell electrodes is one of the key elements affecting performance ( Dupont et al., 2017 ) It is a clear advantage to use as high a concentration of PEDOT:PSS as possible ( Kim et al., 2002 ; Tian et al., 2017 ; Yuk et al., 2020 ). However, the objective of high electrical conductivity has to be a compromise with processability of the prepared suspensions. Hence, we found that PEDOT:PSS ink with a concentration of 100 mg mL −1 which exhibited favourable rheological properties with a viscosity of 15,485 Pa . s at a shear rate of 0.01S −1 and shear yield stress of 572 Pa, can be considered as an optimized ink to be processed through extrusion based 3D printing. 3D printed PEDOT:PSS ink on electrical conductive substrates The prepared ink could be printed via our 3DREDI being manufactured by TRICEP at University of Wollongong (see Figure 2 A). As 3D printed patterns can affect the electrical and mechanical properties of final energy devices performance ( Li et al., 2020 ; Wang et al., 2018 ), a newly interaxial pattern (in this design, the printed direction rotated through 45 degrees for each successive layer) was printed with speed of 200 mm min −1 and nozzle size of 0.15 mm were utilized. The designed layer-by-layer printing patterns are shown in Figure S6 A with a line spacing of 0.2 mm. The rotating 45° interaxial pattern is designed to create high porosity through both the surface and cross-sections of the multi-layer PEDOT:PSS structure. This is because of the lower electrical resistance resulted from more electrical channels and the robust structure with higher modulus in 3D printed structure with an interaxial angle of 45°( Huang et al., 2018 ). Figure 2 B shows SEM images of 3D printed porous PEDOT:PSS electrode (surface and cross-section). This 3D porous patterned architecture is expected to increase the ionically accessible surface area of the electrode and facilitate the penetration of electrolyte into the active electrode materials. We further compared the TEC performance of PEDOT:PSS film and the 3D interaxial porous PEDOT:PSS electrode in both n and p type electrolyte systems, Figures 2 C and 2D. The PEDOT:PSS film mass loading was 12 mg cm −2 . The mass loading of the 3D printed interaxial porous PEDOT: PSS was 10 mg cm −2 which was printed onto a 2mg cm −2 conductive substrate, so the total 3D interaxial porous PEDOT: PSS electrode is 12 mg cm −2 (the mass loading of 1 layer 3D-printed PEDOT: PSS electrode is 1 mg cm −2 ). Electrolyte systems were p-type: CMC-K 3/4 FeCN 6 , and n-type: PVA-FeCl 2/3 , respectively, with a system temperature difference (ΔT) of 10°C (T H = 35°C & T C = 25°C). Owing to the open and porous structure ( Figure 2 B) as well as the good wettability of PEDOT:PSS ( Wang et al., 2020 ), the thermocell made from the 3D printed PEDOT:PSS electrode exhibited an enhanced thermoelectrochemical performance compared to PEDOT:PSS film with increasing current from 8.2 A m −2 to 13.0 A m −2 and power density from 12.2 mW m −2 to 25 mW m −2 for n type and 15.2 A m −2 to 27.5 A m −2 and power density from 30.2 mW m −2 to 70 mW m −2 for p type ( Figures 2 C and 2D). In addition, the use of 3D electrodes also increases the open voltage compared to the 2D film electrode from 5.86 mV to 6.90 mV in n type device and from 7.40 mV to 9.21 mV in p type device. This result is because of the 3D porous structure enabling greater heat transfer efficiency from electrode surface to the electrode/electrolyte interface ( Liu et al., 2020 ), and thus the real temperature difference between the cold side and hot side of 3D printed electrode/electrolyte interface is increased. Figure 2 Preparation and characterization of electrodes for p-type and n-type cells (A) The process of 3D printing flexible interaxial PEDOT: PSS electrode. (B) Surface and cross-sectional SEM images of printed PEDOT: PSS (scale bar size is 100μm for surface and cross-section, scale bar size is 10μm for 3D printed fiber). (C) TEC performance: current & power versus voltage of the devices made from 3D porous structured PEDOT: PSS and PEDOT: PSS film in n type system. (D) TEC performance: current & power versus voltage of the devices made from 3D porous structure PEDOT: PSS and PEDOT: PSS film in p type system. For long-term current output of both PEDOT:PSS film and 3D porous structure PEDOT:PSS, the short circuit current (I soc ) over time was measured and evaluated ( Figures S6 B and S6C). It was observed that the current output dropped dramatically and stabilized at a value of approximately 1 A m −2 after 30 min. This was due to the redox couples’ low diffusion rate in gel electrolyte systems ( Wu et al., 2017 ). Owing to the higher driven potential for ion diffusion, the long-term current density of the 3D printed porous PEDOT:PSS electrode was still higher than that of the PEDOT:PSS film electrode (1.25 A m −2 versus 1.05 A m −2 in n type) and (1.15 A m −2 versus 0.94 A m −2 in p type). In addition, cyclic voltammograms (CV) with scan rate of 10 mV s −1 were performed, where the faradic peak current density provides insight into the electroactive surface area (ESA) of the electrode and could greatly affect the thermocell devices performance ( Romano et al., 2013 ). It is clearly seen that the 3D printed porous structure PEDOT:PSS exhibited higher peak current density than the 2D film under the same mass loading (10 mg cm −2 ) in both types ( Figure S7 A for n type and Figure S7 C for p type). Table S2 indicates that the 3D porous PEDOT:PSS electrodes could provide larger electroactive surface area than the 2D film electrode so as to enhance the current density. Electrochemical impedance spectroscopy (EIS) was also performed. The equivalent series resistance (ESR, the intercept of the curve with the x-axis of the Nyquist plot) of 3D porous structure PEDOT:PSS was slightly reduced compared with 2D PEDOT:PSS film in both types ( Figures S7 B and S7D). These measurements indicate that the 3D porous structure PEDOT:PSS electrode gave the benefit of lowering the activation barrier in thermocell reactions. The improved reaction characteristics also led to the ESR being significantly reduced from 5.4 to 5.2 Ω in n type and from 11.1 to 9.6 Ω in p type as shown in Figures S7 B and S7D insert images. In addition, the charge transfer resistance (R ct , the diameter of the semicircle) can clearly be seen for the 3D porous PEDOT:PSS (3.7 Ω) and PEDOT: PSS film (4.32 Ω) in n type; 3D porous PEDOT: PSS (8.1Ω) and PEDOT: PSS film (8.4Ω) in p type, respectively. These observations from EIS agree with the CV results. Optimizing electrode for both n and p type cells and pairing the optimized n–p cells Then, we printed various layers of 3D porous PEDOT: PSS on PEDOT: PSS thin film substrate (2 mg cm −2 ) as electrode materials for both n and p types. In order to improve thermocell performance, we further increased the printed layer of PEDOT:PSS from 10 layers to 30 layers (equals to 10 mg cm −2 to 30 mg cm −2 of mass loading, electrode thickness in Table S1 ) to facilitate an increase in electroactive sites ( Figure S8 A). The thermoelectrochemical performance in n type electrolyte was improved with increasing layers of 3D printed PEDOT:PSS, while it reached a plateau when the number of layers went beyond 20. This was ascribed to the restricted ion penetration into the outer part of electrode materials which was not directly exposed to the electrolyte. In addition, higher layers of electrode also inhibited the effective heat transfer, as indicated by the decreased open circuit voltage. For the long-term performance of the n-type optimized thermocell device (see Figure S8 B), the high current output decreased initially, which is a disadvantage of all gel electrolyte based thermocells ( Liu et al., 2020 ). However, it is noted that the 3D printed electrodes optimized for n type thermocells can exhibit a higher long-term J SC (1.65 vs 1.50 A m −2 ). This represents a 10% improvement upon our previous reported work using laser-etched PEDOT:PSS film as electrode ( Liu et al., 2020 ). Meanwhile, CV curves (scan rate of 10 mv s −1 ) in Figure S8 C also clearly indicate the best electroactive properties of the 20-layer PEDOT:PSS electrode. Electrochemical impedance spectroscopy (EIS) was also carried out in the n-type electrolyte, where it was clearly found that the ESR was significantly reduced from 6.37 Ω (2 mg cm −2 PEDOT:PSS film) to 5.13 Ω (20 layers of 3D printed PEDOT:PSS on top of PEDOT:PSS thin film), denoted as 20-layer PEDOT:PSS) ( Figure S8 D inset). The Nyquist plot also shows that the charge transfer resistance (the diameter of the semicircle) of 4.16 Ω and 5.3 Ω for 20-layer PEDOT:PSS and 2 mg cm −2 PEDOT:PSS film. These observations from EIS and CV are consistent with the results of the thermoelectrochemical performance. Therefore, 20-layer PEDOT:PSS is the optimized electrode for n-type thermocells. After the optimization for n-type thermocells, thermoelectrochemical performance of 3D printed electrodes in p-type gel electrolytes were also investigated. It was found that 10 layers of PEDOT:PSS on 2 mg cm −2 PEDOT:PSS film exhibits the best performance ( Figure S9 A). The current density was also improved by 11% (1.26 vs 1.13 A m −2 in Figure S9 B) when compared to our previously reported PEDOT:PSS-edge functionalized graphene/carbon nanotube electrode (PEDOT: PSS/EFG/CNT)( Liu et al., 2020 ). CV and EIS results also confirmed 10-layer PEDOT: PSS as the optimized electrodes with highest electroactive behaviour and lowest resistances ( i.e., ESR & charge transfer resistance), as shown in Figures S9 C and S9D. The thermoelectrochemical performance of thermocell devices is greatly dependent on the temperature difference between the two electrodes, so the devices were tested under various ΔT ( Dupont et al., 2017 ). Here, the cold side or the warm side was kept constant as 10°C or 30°C respectively, and then the other side increased from 10°C to 30°C or decreased from 30°C to 10°C to create controlled ΔT's in the range of 0°C to 20°C. The open voltage of the optimized n and p type cells increased or decreased almost linearly with ΔT ( Figures 3 A and 3C). Thus, the effective S e of n and p type cells was calculated to be 0.65 mV K −1 and 9 mV K −1 , respectively. Current density also increased linearly with ΔT, whereas power density has a linear relationship with (ΔT) 2 as shown in Figures 3 A, 3C, and 3D, which indicates that the individual optimized n and p cells were stable at various temperature gradients. Figure 3 TEC performance: current & power versus voltage of optimized p–n cell at different ΔT (A) and (B) optimized n type TEC performance at different ΔT. (C) and (D) optimized p type TEC performance at different ΔT. (E) and (F) optimized n and p types connected in series TEC performance at different ΔT. When multiple cells are assembled into a series circuit, the current output is rate limited by the individual cell with the least performance and the open voltage is the sum of all the individual cells ( Yang et al., 2016 ). To effectively connect the n type and p type thermocells in series, single devices with similar current output should be matched for one pair of n-p thermocells. We found that, our optimized n type and p type cells just in line with this requirement with comparable current densities of 26.1 A m −2 and 27.5 A m −2 at ΔT = 10°C respectively, and therefore were selected for a matched n-p thermocells ( Figures 3 B and 3D). Thus, a pair of n–p cells connected in series was fabricated ( Figure 3 E insert) and the thermoelectrochemical performance characterized ( Figures 3 E and 3F). The n-p cell performance between ΔT's of 5–20°C were examined, yielding linear V versus I and parabolic P versus V relationships at all temperature differences. The n-p cell exhibited current density of 26.0 A m −2 and open circuit voltage of 15.5 mV at ΔT = 10°C, where the current density is same with n and p type single devices and the voltage value is the sum of n and p type single cells ( Figure S10 A). In addition, long-term output current of the n-p cell was consistent with that of the individual optimized n type and p type. All the results demonstrate the efficient matching of the optimized n and p type devices for n-p cells ( Figure S10 B). Prototyping multiple n–p cells To further promote application, flexible multi- n-p cell assemblies (up to 18 pairs) were fabricated as shown in Figures 4 A and 4B. To facilitate the heat transfer between the assembled device and the environment, polyimide tape (PI), the support substrate used for PEDOT:PSS electrode, was adhered to aluminium foil. Sputter coated Pt was used as the interconnection between single devices and copper tape was used to connect multiple n-p cells to external electrical connections. Wearable polydimethylsiloxane (PDMS, SE 1700 clear base and catalyst) was 3D-printed as the spacer to encapsulate the gel electrolyte. The power density of the multi-cell arrangement was measured and found to increase from 10.5 μW (1 pair) to 45 μW (18 pairs) at ΔT = 10°C. The open voltage also increased from 15.5 mV (1 pair) to 270 mV (18 pairs) ( Figures 4 C and S10 C). However, because of the increased resistance between fabricated thermocell connections, the current output inevitably decreased when the number of n-p pairs increased (see Figure S10 C). These n-p cell arrays could charge up various commercial electrochemical supercapacitors (C = 1, 4.7, 22, 47, and 100 mF) to more than 200 mV (see Figure 4 D). It was also observed that due to the relatively decreased long-term performance of our thermocells employing gel electrolyte, the charging rate of supercapacitors (C = 100 mF) decreased along with time. Notably, this is the first time that a flexible thermocell device has been fabricated using the 3D printing of all polymer electrodes. Figure 4 Device evaluation of p-n cells connected in series (A) Photo demo of thermocell arrays with serial 6 pairs n-p cells. (B) Prototyping thermo-electrochemical device of 18 pairs n-p cells. (C) When ΔT = 10°C, 1 to 18 serial pairs of n-p cells TEC performance and (D) 18 serial pairs of n-p cells charging up commercial super-capacitors with different capacitance. Wearable n-p cells To harvest body heat, a wearable thermocell that can conform to a curved body surface is preferred. To achieve this aim, we designed a strap shaped wearable thermocell as shown in Figure 5 A. The fabricated device is shown in Figure 5 B. The benefit of this design is that the top electrodes of the wearable strap shaped thermocell could move freely so that the strap arrangement could bend around the contours of the wearer. This ensured that the bottom electrodes of the device could conform closely to the skin of the wearer ( Figure 5 C). As a demonstration ( Figure 5 D), we have attached the strap shaped thermocell array to a human body. The wearable thermocell could charge up a 100 mF supercapacitor, and then power a lab timer when coupled with a voltage booster. The results indicated the practical application of our wearable thermocells in powering wearable electronics utilizing low grade human body heat. Figure 5 Wearable n-p cells (A) Design of wearable thermocell. (B) Fabrication of wearable thermocell. (C) Flexible strap shaped thermocell. (D) The demonstration of a strap shaped thermocell charge supercapacitor and light a lab timer screen utilizing body heat. Conclusions In conclusion, we have fabricated novel wearable multi-unit thermocells assembly through 3D-printing all polymer-based flexible electrodes. We also matched serial thermocells and brought about an open voltage of 0.27 V to charge up a supercapacitor and store the harvested energy when operating with a ΔT = 10°C. Moreover, the demonstration of a wearable thermocell, which can harvest body heat, charge commercial supercapacitors, and even power a lab timer, shows the great potential of our wearable device in practical applications. This work provides a platform for the future development of 3D-printable integrated wearable device systems. Limitations of the study Compared with previous thermocells, the polymer wearable thermocells developed in this work have the following advantages. First, a compatible high electrical conductivity polymer film can work as an underlying conductive substrate, to replace traditional and expensive platinum electrodes. Second, a 3D printable polymer ink with suitable rheological properties was developed for high-performance electrodes in both n and p type thermocell devices, which enables facile and cost-effective fabrication of thermocells. Thirdly, a serial arrangement of 18 pairs of n-p devices enabled charging of commercial supercapacitors up to 0.27 V sufficient to power a commercial lab timer. Last but not least, we also claim that this is the first time that a 3D-printed all-polymer electrode thermocell device was utilized for harvesting body heat. Despite this progress, we note the need for further optimizations of our printing system. It is necessary to develop printing techniques with a high resolution of minimum feature size down to 10 μm in the future, which can further increase the porosity and surface area of electrode materials. Furthermore at present, only inks of electrode materials are developed in this work. In the near future, it is urgent to develop inks consisting of other components (e.g., electrolyte & encapsulation materials). Fully integrated manufacturing requires multi-materials dispersion approaches."
} | 7,691 |
39204837 | PMC11359260 | pmc | 7,172 | {
"abstract": "Textile-based thermoelectric (TE) devices are being investigated to power smart textiles autonomously. While previous research has focused on a solid system where the required junctions are fabricated into the device, there has been limited attention given to replacing these TE systems reliably. This work looks at a newer approach to the construction and demonstration of a wearable thermoelectric structure that employs three-dimensional knitted spacers to increase the temperature difference where the TE junctions are detachable and disposable. This system features positive and negative junctions which can be removed while maintaining its excellent voltage generation in low ΔT and good Seebeck coefficients. A mathematical model simulates the potential energy outputs and maximum power points generated, which can be used to increase the device’s performance for future wearable sensing applications.",
"conclusion": "5. Conclusions A flexible, knitted 3D-spacer fabric was successfully designed and demonstrated, with detachable and disposable parts that can be seamlessly knitted as part of a wearable sensing device. Thermal imaging confirmed that some heat is transferred through the knitted spacer structure, which needs to be controlled. However, it maintains the temperature difference across the TE configuration and results in sufficient output voltage. This demonstrates that knitted 3D-spacer structures are ideal for TE wearable devices, as they offer higher ΔT than single-layer structures. While this device features detachable junctions, the contact resistance between the snaps that hold the junctions onto the knitted device is low enough that there is no change in the resulting voltage or Seebeck coefficients generated by low ΔT. This 11-junction system produces a 1.52 mV open-circuit voltage of electrical energy, a current of 9.05 µA, a maximum power point of 70 µW, and a Seebeck coefficient of 62.79 µV·K−1 at 312 K. The device offers facile fabrication methods, and its disposable parts do not detract from its performance, making it a viable alternative to the current systems that require the entire system to be replaced if something breaks down.",
"introduction": "1. Introduction Electronic textiles (e-textiles) are increasingly popular, and there is a drive to find ways of powering these devices autonomously. One way this can be accomplished is through thermoelectric (TE) systems, which can convert a temperature difference to electrical energy via the Seebeck effect [ 1 ]. The Seebeck effect is when positive and negative materials are connected at a junction, the p-n junction, are subjected to a temperature difference across the boundary of the device and create electrical energy through this difference [ 2 ]. To use wearable TE systems in medical, sports, and protective applications, the systems need to be flexible, lightweight, have the ability to work in low temperatures, and, if worn by humans as a garment, comfortable [ 3 , 4 , 5 , 6 ]. Wearable TE systems require different characteristics than those used in other applications, such as hybrid vehicles, as the temperature difference (ΔT) is much smaller [ 7 , 8 ]. The structure of the wearable device influences how effective the system can be, as it can give a natural temperature difference based on its thickness and makeup. A knitted three-dimensional (3D) spacer fabric is ideal because the conductive TE system can be seamlessly integrated into a structure that already has a designated thickness and density. However, while the research into advanced, material-based knitted TE systems is ongoing, researchers have not been able to account for the high electrical resistance in these systems, so the researchers demonstrate the TE device performing at higher ΔTs than what can be used for wearable technology [ 9 , 10 , 11 , 12 ]. 3D-spacer knitted structures are being investigated for TE application because their structures have natural gaps between the boundary yarns, which allows for an added temperature difference between the skin and the atmosphere. However, there is very little research into a 3D-spacer knit structure that features knitted-in p-n wire conductive pathways. Other researchers have created the 3D structure and then embroidered, sewn, or deposited the conductive material post-fabrication [ 13 , 14 , 15 ]. Li et al. created a 3D spacer knit for TE application using PEDOT:PSS and constantan where the device was glued to the spacer-knit structure [ 16 ]. With 100 units connecting the two sides of the spacer fabric, the results showed an output voltage of around 0.5 mV at a 20 K temperature difference. Another group doped 3D spacer fabrics post-fabrication in aluminum-doped zinc oxide and silver, then threaded a conductive wire through the knit to connect the two sides, creating a device that generated a power of 0.2 µW [ 15 ]. This device did investigate a low ΔT of 10 K, with the results showing around 0.5 mV and 26 µA for voltage and current, respectively. Only one research group, Dallmann et al., used a single-fabrication method for their 3D spacer fabric, resulting in a power output of 1.78 µW [ 17 ]. However, the results for this spacer knit are captured when the ΔT is from 32 to 65 K, which is too high of a temperature difference to be worn on humans. Yang et al. show a knitted TE device with high stretchability made from silver and silver selenide which was then sewn onto a knit structure [ 18 ]. While this device featured excellent bending capabilities, this process is not a single-knit process; it involves creating strips of TE material and manually inserting them into the knitted fabric. This research does feature a good voltage output of four mV at a ΔT of 15 K. Kim et al. also embedded conductive yarn into a knit structure [ 19 ]. They used wet-spun graphene to create the p-n junctions, with a low voltage output of 0.2 mV at a 10 °C temperature difference and no reported power output. Also, while a knitted system is seamless, there is a problem when the junctions wear down or disintegrate. If this happens, the fabric will need to be disposed of and a new one made, creating waste products. Some researchers have encapsulated their advanced materials to protect them from wearing down [ 20 ]. Dong et al., for example, created their TE device and then encapsulated it with a silica gel, polydimethylsiloxane, and polymethylmethacrylate [ 21 ]. This kept the device from being damaged during washing and maintained its high flexibility. However, this device was fully encapsulated, requiring copper connections added post-fabrication and creating areas where the device could be damaged. Therefore, creating a practical, knitted, wearable device featuring detachable and disposable parts is important to the feasibility of wearable TE systems. In this work, we present a wearable 3D-spacer garment and TE system featuring detachable and disposable p-n junctions. This 11-junction device offers good output voltage, current, and Seebeck coefficients at low-temperature applications. This device’s facile manufacturing process and its detachable parts increase its reliability as a practical system for long-term wearable applications.",
"discussion": "4. Results and Discussion 4.1. Device Structure The knitted spacer structure resists heat transfer and can maintain a higher ΔT than a fabric that is only one layer thick. Conductive k-type thermocouple wires, made from Alumel and Chromel, were knitted through the structure in a single course, with exterior points found on both sides of the fabric [ 23 ]. These conductive wires were chosen for their knittability and the strength of the welded p-n junction ( Figure S1 ). The thermocouple legs knitted into the device were 20 mm in length and were cut at 10 mm increments to be soldered to the snap closures, which are common sewing notions. Nickel-plated snap closures are electroconductive and ideal separators if the TE device needs to be detached or replaced, and they maintain low electrical resistance ( Figure 4 a). By interrupting the electrical circuit of the TE device with the snap, the structure can maintain its electrical resistance while allowing the p-n junctions to be detached and replaced when necessary, as shown in schematic Figure 4 b. The final TE device used in this research had 11 p-n junctions evenly spaced along both sides of the conductive wire’s knitted course, with six pairs of junctions reaching from the cold to the hot side of the device, Figure 4 c. The initial test of the device on a heated plate is shown in Figure 4 d. 4.2. Thermoelectric Capabilities The electrical resistance of the TE device is critical to the results, as it impacts the voltage output, the Seebeck coefficient, and the generated current. The overall resistance of the device measured 30.87 ohms (Ω) before washing and 33 Ω after 10 wash cycles. Due to the device’s detachable junctions, the overall resistance cannot vary when the junctions are replaced to maintain the device’s TE efficiency. Therefore, the junctions were replaced 100 times, and the resistance was taken each time ( Figure S2, Table S2 ). These results showed that the resistance remained under 80 Ω but did vary due to altered surface contact between the different snaps. However, the snap notions did not damage the nickel coating, indicating long-life application. The device’s resistance was also captured during a bending test to understand how deformation would affect the resistance of the device. It showed that the device’s resistance reduced during bending and then increased when the device was unbent ( Figure S3 ). This is because of the contact resistance increase during the bend as the garment contracted around the snaps. When a TE junction is exposed to a temperature difference, there is a thermal diffusion of electrons from the hotter side to the cooler side because the increased entropy in the heated electrons transports them to the colder side [ 6 ]. The electrons also move from the n-type material, which has a surplus of electrons, to the p-type material, which is deficient in electrons, creating a flow of energy. However, if the heat transfer occurs through the device, it cannot maintain the temperature difference and gradually decreases electron transport. Thus, direct thermal diffusion of the device can severely limit the device’s TE capabilities. A knitted 3D-spacer structure can stop this effect by providing thermal insulation. Figure 5 shows the thermal imaging of the device’s surface when heated to a steady-state temperature and the overall heat transfer from the hot side to the cooler side. Area 1 is the knit fabric and Area 2 is the replaceable p-n junctions. During the heating period, when the temperatures were at a steady state, the results show that the spacer fabric managed to maintain a temperature difference of 15.5 °C between its hot side and cooler side ( Supplementary Video S1 ). These results show the spacer knit’s excellent thermal insulation, allowing its TE performance to remain constant through different human skin temperatures and activities. The temperature difference in the TE system creates an output open-circuit voltage that can be analyzed. To test the capabilities of this device, it was connected to a multimeter and heated to temperatures from 36 °C to 40 °C, resulting in ΔTs of 13 to 17 degrees, respectively. The device was heated from room temperature, 23 °C, to different temperatures and held until a steady output voltage was reached, as seen in Figure 6 a. The resulting voltages ranged from 0.7 to 1.52 mV for ΔT 13 to 17 °C, respectively. The test shows that the highest open-circuit output voltage occurs with the highest ΔT, proving that the thermoelectric capabilities are linked to the temperature difference. Subsequently, the device was tested to analyze the device’s heating and cooling properties. The device was heated from room temperature to just below 40 °C with a ΔT from 4 to 17 degrees, as shown in Figure 6 b. The overall output open-circuit voltage increased with the ΔT but reached its highest voltage, 1.56 mV, at a ΔT of 14 after the heating was turned off. This shows a slight delay between heating the device and the improved voltage from the system. When the device cooled, it maintained a higher voltage than when it was heated, most likely because of thermal inertia in the device. The Seebeck coefficient (α) captures the efficiency of the TE materials and system, determined from the relationship, α = ∆ V / ∆ T [ 2 ]. The p-type material has a Seebeck coefficient of 11.54 µV·K−1 and the n-type material, −6.82 µV·K−1 at 295.15 K. The device’s Seebeck coefficient was determined for temperatures 303.15 to 313.15 K in single-degree increments, with results measured across a range varying from 24.7 to 67.7 µV·K−1, respectively, as shown in Figure 7 a. This proves that the Seebeck coefficient is temperature-dependent and increases with the overall ΔT applied to the device. For wearable applications, temperatures up until 312 K are suggested, giving a Seebeck coefficient of 62.79 µV·K−1, higher than other examples of knitted 3D-spacer TE devices and at lower temperatures [ 15 , 16 , 24 ]. To analyze the efficiency of the TE system with varying numbers of paired junctions, the device was tested at different temperature differences with one to six paired junctions being heated. This resulted in different output voltages and different Seebeck coefficients. The device was again heated to 36, 37, 38, 39, and 40 °C, resulting in a ΔT from 13 to 17 degrees, respectively, as seen in Figure 7 b. These results show the higher number of paired junctions created a higher output voltage and Seebeck coefficient. For one pair of junctions, the resulting output voltage and Seebeck coefficient ranged from 0.08 mV and 5.13 µV·K−1 at ΔT 13 as compared to 0.16 mV at 8.3 µV·K−1 at ΔT 17. When all junctions were used, the output voltage increased from 0.71 mV to 1.52 mV at ΔT 13 and 17, respectively. However, the Seebeck coefficient decreased from 62.7 µV·K−1 to 55.7 µV·K−1 at ΔT 13 and 17, respectively. This is due to the ΔT value increasing faster than the ΔV value and having a higher impact on the resulting coefficient. This confirms that a lower ΔT allows the device to perform more efficiently than a higher ΔT. Similar tests were performed to analyze the performance of the device, with different numbers of paired junctions. Figure 7 c shows that the current and voltage increased with the number of junctions and the ΔT. The higher the number of paired junctions and ΔT, the higher the current, with the highest current being 9.04 µA for all junctions maintained at 17 °C temperature difference. This generated current is higher than other reported wearable TE systems that use a similar number of junctions, as seen in Table 1 , due to the metal in the conductive paths and the tight connection between the snaps in the detachable system. After confirming the capabilities of the device on a heater, the device was tested on the human body. The three areas tested, thigh, upper arm, and torso, did not reach the anticipated 36–40 °C described in the other literature [ 5 , 27 ]. The skin’s temperature reached 31.1 °C, 32.5 °C, and 32.8 °C for each respective area. This may be due to the locations chosen to test the device and because the radiated body temperature of a human is different from the core temperature. This is because the skin’s epidermis acts as a barrier, reducing the skin’s temperature more than the expected radiated temperature. Therefore, the overall performance of the device was lower than expected because of the lower ΔT. The results were recorded from the initial placement of the device until a steady-state output voltage was achieved, as seen in Figure 7 d. The thigh demonstrated a ΔT of 6.2 °C and resulted in a voltage output of 0.29 mV, the upper arm had a ΔT of 7.6 degrees and an output voltage of 0.32 mV, and the torso had a ΔT of 7.9 degrees and an output voltage of 0.33 mV. The torso offered the greatest output voltage and should be considered a prime location for TE wearable devices. For each area, the Seebeck coefficient was calculated at 48.89 µV·K−1, 42.43 µV·K−1, and 41.88 µV·K−1 for the thigh, upper arm, and torso, respectively. 4.3. Analysis with Simulations Using Equations (6) and (7) from the modeling section, a linear relationship between the simulated current and the voltage can be graphed, as shown in Figure 8 a. From this, a power curve can be determined, and a maximum power point (P max ) can be found for the single paired junction at different temperature differences. With an increase in ΔT, the current and voltage increase, raising the P max . The calculated P max for ΔTs from 13 to 17 vary from 12.08 to 53.96 µW, respectively; see Figure 8 a. To validate the mathematical model, the device’s current and voltage were empirically measured at varying ΔT with different R L . Figure 8 b–f compares the empirical data from a single paired junction heated to different ΔTs with the mathematical model’s generated information. The empirical data is semi-linear, not linear like the mathematical model, due to heat trapped in the device, heat leakage, humidity, and other environmental factors. At each ΔT, the empirical data performed better than the model. With ΔT of 13 degrees, Figure 8 b, the device had an empirical p max 37 µW higher than predicted. Similarly, ΔT 14, 15, 16, and 17 empirically had higher p max results of 24.5, 22, 21, and 18 µW, respectively. As the ΔT increased, the percentage error between the empirical and mathematical data decreased, and the two voltages drew together. This shows that the mathematical model became more accurate at higher temperatures. This may be because the detachable contact points in the system swell when heated, reducing the contact resistance in the device. The relationship, while different, between the two datasets shows that the mathematical model does represent a version of the device’s performance and can be used to calculate potential voltage, current, and power generation. This comparison demonstrates that the device performs better than mathematically predicted but is relatively accurate at ΔT 17."
} | 4,564 |
25927577 | PMC4463603 | pmc | 7,173 | {
"abstract": "Harnessing hydrogen competently through wastewater treatment using a particular class of biocatalyst is indeed a challenging issue. Therefore, biohydrogen potential of real-field wastewater was evaluated by hybrid fermentative process in a single-stage process. The cumulative hydrogen production (CHP) was observed to be higher with distillery wastewater (271 mL) than with dairy wastewater (248 mL). Besides H 2 production, the hybrid process was found to be effective in wastewater treatment. The chemical oxygen demand (COD) removal efficiency was found higher in distillery wastewater (56%) than in dairy wastewater (45%). Co-culturing photo-bacterial flora assisted in removal of volatile fatty acids (VFA) wherein 63% in distillery wastewater and 68% in case of dairy wastewater. Voltammograms illustrated dominant reduction current and low cathodic Tafel slopes supported H 2 production. Overall, the augmented dark-photo fermentation system (ADPFS) showed better performance than the control dark fermentation system (DFS). This kind of holistic approach is explicitly viable for practical scale-up operation.",
"conclusion": "4. Conclusions This study demonstrated the feasibility of biological H 2 generation from distillery and dairy wastewater treatment in a single stage hybrid system. The hybrid system comprising dark and photo-fermentation facilitated higher H 2 production with distillery wastewater over dairy waste because of its high organic fractions compared with dairy waste, which was protein-rich. An increment of about 40% was noticed in H 2 production in photo-augmented system with simultaneous reduction in accumulated VFA. This study interprets the performance of hybrid process in a single system which is likely to provide the crucial information for the development of a full scale integrated photo-fermentation system.",
"introduction": "1. Introduction Recently, a great deal of attention has been paid to the biological production of hydrogen (H 2 ) as alternative and eco-friendly fuel throughout the world [ 1 , 2 ]. Microbial conversion of substrate to H 2 and volatile fatty acid (VFA) by anaerobic fermentation is a complex series of biochemical reactions manifested by diverse group of selective bacteria [ 3 , 4 ]. Accumulation of organic acid metabolites inhibits the H 2 production process and makes the H 2 production process unfavorable by limiting the substrate degradation. Further utilization of the organic acids towards H 2 production is thermodynamically feasible only if there is an additional energy input [ 5 ]. This energy input can be in the form of electricity in microbial electrolysis cell (MEC) [ 6 , 7 , 8 ] or in the form of light in two-stage photofermentation [ 9 , 10 ] or augmentation of photosynthetic bacteria with dark fermentative culture in a single-stage hybrid system [ 11 ]. Photofermentataion can be carried out with a wide variety of organic substrates such as carbohydrates; lactate, malate, benzoate and sucrose which are utilized by different species of phototrophic bacteria as electron donors for H 2 production [ 12 , 13 , 14 , 15 ]. Exploitation of wastewater as substrate for H 2 production with concurrent wastewater treatment is an attractive and effective way of tapping clean energy from renewable resources in a sustainable approach. This provides dual environmental benefits in the direction of wastewater treatment along with sustainable bioenergy (H 2 ) generation [ 16 ]. Molasses-based distilleries generate 8–15 L of wastewater having high chemical oxygen demand (COD) (100–126 g/L) for every litre of the alcohol produced [ 17 ]. Distillery wastewater generated in the form of spent wash or spillage is one of the most complex and strongest industrial organic effluents. It possesses high concentration of biodegradable organic material, such as sugars, lignin, hemicelluloses, dextrin, resins and organic acids [ 18 ]. Dairy wastewater contains complex organics, such as polysaccharides, proteins and lipids, which on hydrolysis form sugars, amino acids, and fatty acids [ 19 ]. In subsequent acidogenic reaction, these intermediate products are converted to volatile fatty acids (VFA), which are further degraded by acetogens, forming VFA, CO 2 , and H 2 . High organic load and persistent color associated with the distillery and wastewater pose a serious problem to the environment and treatment of such kind of wastewater is challenging [ 20 ]. The technologies currently used by distilleries and dairies for treatments of wastewater are biomethanation followed by two-stage biological treatment, concentration and incineration [ 18 ]. High organic load, absence of toxic chemicals and availability of large quantities of wastewater may be considered as potential sources for biohydrogen production by integration [ 21 , 22 ]. In this context, a hybrid strategy comprising dark-photofermentation was investigated using designed synthetic wastewater (DSW) as previously reported [ 11 ]. In this study, an attempt was made to harvest biohydrogen using real field wastewater (distillery and dairy wastewater obtained from brewery and milk processing industries, respectively) by combining dark and photo-fermentation in a single stage hybrid system using mixed anaerobic bacteria and photosynthetic bacteria.",
"discussion": "2. Result and Discussion 2.1. Bio-Hydrogenesis 2.1.1. Dark-Fermentation (DFS) The experimental data depicted feasibility of H 2 production by utilizing distillery and dairy wastewater as substrate ( Figure 1 ). Acidogenesis of distillery wastewater in control (CDi) resulted in H 2 production with an initial value of 84 mL during the start-up phase and thereafter gradually increased with cycle operation and reached a maximum consistent value of 133 mL. A similar trend of acidogenesis was observed when the control (CDa) was operated with dairy wastewater. During start-up, the CDa reactor produced 7 mL, and thereafter gradually increased to a maximum of 144 mL. Steady increments in the H 2 evolution is attributed to the acclimatization and enrichment of H 2 producing acidogenic bacteria. These experimental results evidenced relatively higher H 2 production with distillery wastewater than dairy wastewater because of high carbohydrate content in distillery compared with high protein and fat content in dairy waste. Dark-fermentation process involves VFA production as co-metabolites during conversion of organic substrates to H 2 (Equations (1)–(3)) [ 3 ]. The production of VFA affects the buffering capacity that can inhibit the functioning of acidogenic bacteria; perhaps the decline in HPR was consequently noticed.\n C 6 H 12 O 6 + 2H 2 O → 2CH 3 COOH + 2CO 2 + 4H 2 (1) \n C 6 H 12 O 6 + 2H 2 O → CH 3 CH 2 CH 2 COOH + 2CO 2 + 2H 2 (2) \n 3C 6 H 12 O 6 → 4CH 3 CH 2 COOH + 2CH 3 COOH + 2CO 2 + 2H 2 O (3) Figure 1 Temporal profile of Cumulative Hydrogen production (CHP) in dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa) as function of time with dairy and distillery wastewater. 2.1.2. Hybrid Dark-Photo Fermentation Another set of experiments (EDi and EDa) were carried out where dark fermentation was augmented with photosynthetic bacteria (PSB) ( Figure 1 ). This is designed as hybrid dark-photo fermentation system (ADPFS) operated in a single stage with both the wastewaters, separately. A remarkable performance was noticed with ADPFS relative to controls. Comparing both augmented systems, cumulative H 2 was noted as higher in EDi (271 mL) than in EDa (248 mL), which is almost double than the controls (CDi, 133 mL and CDa, 114 mL). In the ADPFS, an interesting observation was noticed i.e. , the H 2 production was maximum at the 48th h unlike in the control at 24th h. Higher H 2 production with ADPFS can be attributed to the co-existence of photosynthetic bacteria with dark-fermentative microflora, which facilitates dark fermentation as well as photofermentataion. The increased duration of H 2 production is attributed to the role of PSB which are competent in VFA consumption thereby minimizing the acidic stress. The short-chain organic acids (VFA) released during dark-fermentation gets further metabolized to H 2 by photosynthetic bacteria thereby resulting in higher H 2 production. Theoretically one mole of acetate can produce four moles of H 2 while one mole of butyrate can produces 10 moles of H 2 by photofermentation (Equations (4)–(6)) [ 3 , 11 ]. However, the H 2 production efficiency and yield was found to depend on the type of wastewater and amount of organic content which is biodegradable.\n 2CH 3 COOH + Light→ 2CO 2 + 4H 2 (4) \n CH 3 CH 2 CH 2 COOH + 3H 2 O + Light → 4CO 2 + 10H 2 (5) \n HCOOH + Light → CO 2 + H 2 (6) 2.2. Pigments and Biomass Biomass concentration of photosynthetic bacteria was calculated indirectly via bacteriochlorophyll estimation. Bacteriochlorophyll ( BChl ) is a pyrrole derivative specific to photosynthetic bacteria and plays a major role in anoxygenic photosynthesis [ 23 ]. To confirm the growth of photosynthetic bacteria during ADPFS operation of all experimental variants, BChl was analyzed ( Figure 2 ). Control systems (CDi and CDa) did not contain PSB throughout the study. On the other hand, augmented systems (EDi and EDa) showed growth of PSB. Among the two experimental runs, EDa showed significant increments in photosynthetic BChl over a cyclic operation and a maximum of 84 µg/mg at 72 h was noticed. Dairy wastewater is rich in protein and is a good source of nitrogen for the PSB biomass growth ( Figure 3 ). However, in the case of the EDi system, a decline in PSB growth was noticed at the end of each cycle. Therefore, prior to the start of each batch, a fixed inoculum volume of 10 mL was added to the ADPFS and at the end of the batch 26 µg/mg (72 h) was noted. The deficiency in PSB growth is possibly attributed to the absence of protein in distillery wastewater. Besides, the decrease in PSB biomass is attributed to the acidic shock caused by VFA present in the system. BChl is pH sensitive and at acidic pH structural and functional aspects of BChl to PhBChl was previously reported [ 11 ]. In the present experiment, the pH drop and VFA present in the system was sufficient to trigger pheophytinization of the Bchl which automatically hinders bacterial photosynthetic activity. Pheophytinization is a bio-physio-chemical process where at low pH and high proton (H + ) concentration, the central metal ion (Mg 2+ ) of BChl gets bleached out and is replaced by H + ion [ 11 , 24 , 25 , 26 ]. Decrease in pH was noticed in DFS (CDi and CDa) due to anaerobic fermentation and release of VFA. But, in case of hybrid ADPFS (EDi and EDa) the pH increase was noticed due to the consumption of VFA ( Figure 4 ). In this regard, to revive the batch, an aliquot of PSB inoculum was added to the system. Figure 2 Pigment analysis (bacteriochlorophyll) of dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa) as function of time with distillery and dairy wastewater. Figure 3 Variation in protein content in dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa) as a function of time with distillery and dairy wastewater. Figure 4 Changes in pH profile as function of time with dairy and distillery wastewater in dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa). 2.3. Total Volatile Fatty Acids and Composition Both the wastewaters contained a certain amount of VFA prior to the start-up of experiments; distillery wastewater (2500 ± 200 mg/L) and dairy wastewater (2744 ± 200 mg/L). Thereafter, when fed to control systems (CDi and CDi) and ADPFS (EDi and EDa) changes in total VFA concentration and its composition were observed. In CDi operation, VFA concentration increased to a maximum value of 2519 ± 200 mg/L at 12 h and thereafter decreased to 1690 mg/L at the end of the batch. Similarly, in CDa, VFA increased to a maximum value of 2962 mg/L at 12 h and thereafter declined to 1898 mg/L at the end of the batch. These variations were much seen in case of EDi and EDa where the initial/final total VFA concentrations were 2356/880 and 2744/872 mg/L, respectively ( Figure 5 ). VFA production is associated with conversion of organic fraction to acid intermediates in the anaerobic microenvironment. Change in the concentration of acid metabolites can affect the system buffering capacity and higher acid concentrations will inhibit the function of acidogenic bacteria specifically, H 2 production. Therefore, individual composition was also analyzed using HPLC which assists in tracking the biochemical pathway of the biocatalyst during process operation. Acetate or butyrate pathway favors H 2 production, while propionic acid is not as favorable for H 2 production [ 27 ]. In the present study, a higher proportion of acetic acid along with butyric acid and propionic acid was observed in the controls (CDi and CDa). Initial concentrations of observed VFA (CDi-acetate, 1457 mg/L; propionate, 469 mg/L; butyrate, 463 mg/L; CDa-acetate, 1459 mg/L; propionate, 472 mg/L; butyrate, 466 mg/L) were changed by the end of the batch (CDi-acetate, 1225 mg/L; propionate, 327 mg/L; butyrate, 412 mg/L; CDa-acetate, 1228 mg/L; propionate, 330 mg/L; butyrate, 415 mg/L) ( Figure 6 ). The temporal profile of these acids did not show much variation indicating the inefficiency of the dark-fermentative consortia. However, a remarkable change in the acid composition profile was noticed in the same wastewaters when fed to the experimental set-up inoculated with PSB The initial concentrations of acetic, butyric and propionic were (EDi-acetate, 1357 mg/L; propionate, 469 mg/L; butyrate, 563 mg/L; EDa-acetate, 1427 mg/L; propionate, 449 mg/L; butyrate, 663 mg/L) decreased by the end of the batch (EDi-acetate, 445 mg/L; propionate, 227 mg/L; butyrate, 272 mg/L; CDa-acetate, 545 mg/L; propionate, 227 mg/L; butyrate, 272 mg/L). VFA removal was about 62% in the EDi system and 68% in the EDa system. Remarkably, total VFA in ADPFS removed about 50% less than control DFS. The above experiment (ADPFS) documented the functional role of PSB in utilizing the VFA for its growth and maintenance. Interestingly, dairy wastewater showed higher removal of VFA in spite of less H 2 production which was due to the utilization of these VFAs towards the biomass production. Besides, dairy wastewater contains high protein content which is a good nitrogen source for PSB growth. Although VFA removal was observed in distillery wastewater, PSB growth declined with batch time and also the protein content was minimal, which supports maintenance but not growth [ 3 , 11 ]. Figure 5 Total volatile fatty acids (VFA) profile in dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa) as a function of time with dairy and distillery wastewater. Figure 6 Compositional analysis of volatile fatty acids composition as a function of time with distillery and dairy wastewater in dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa). 2.4. Substrate Degradation Substrate degradation efficiencies (based on COD) showed variation as a function of biocatalyst and complexity of wastewater. EDi operation showed COD removal efficiency of 18% at the 12th h which improved with time (24 h-24.32%, 36 h-43.52%; 48 h-50.80%; 60 h-56.80%) and approached a maximum value of 56.8% at the 72nd h. EDa operation showed COD removal efficiency of 14% at the 12th h which improved with time (24 h-21.72%, 36 h-29.81%; 48 h-37.69%; 60 h-41.31%) and approached a maximum value of 45.42% at 72nd h. CDi operation showed COD removal efficiency of 5.36% at the 12th h which improved with time (24 h-17.31%, 36 h-31.20%; 48 h-34.69%; 60 h-39.63%) and approached a maximum value of 39.63% at 72nd h. CDa operation showed COD removal efficiency of 6.89% at the 12th h which improved with time (24 h-10.85%, 36 h-16.84%; 48 h-20.84%; 60 h-28.84%) and approached a maximum value of 35.42% at the 72nd h ( Figure 7 ). Almost 20% increment in substrate degradation efficiency was observed with augmented operation which also demonstrated the effective functioning of photosynthetic consortia in treating wastewater. CDi and CDa operation showed comparatively lower substrate degradation. Microbial fermentation generates energy-rich reducing power (NADH, etc. ), which subsequently gets re-oxidized during respiration with simultaneous generation of biological energy molecules (ATP) in the presence of a terminal electron acceptor (TEA). Anaerobic respiration has the ability to utilize a wide range of organic compounds by the acidogenic pathway and generates VFA in association with H 2 . Hydrogenase plays an important role for the generation of H 2 . Under anaerobic conditions, photosynthetic bacteria use sunlight as a source of energy and produce H 2 and CO 2 by degrading organic molecules [ 1 ]. The observed higher H 2 production in hybrid system might be attributed to consumption of VFA by photosynthetic consortia towards additional H 2 . Light absorption by bacteriochlorophyll (BChl) molecules initiates e − transfer from a reaction center to quinine pool (QA) and then to the cytochrome subunit for generating a H + gradient, which finally gets reduced to H 2 . The ability of PSB to trap energy over a wide range of the light spectrum without producing oxygen and its versatility in utilizing various substrates like acetate, butyrate and propionate makes the hybridization of photo fermentation with dark fermentation a feasible and viable approach [ 2 , 11 , 28 , 29 ]. Figure 7 Substrate utilization in terms of chemical oxygen demand (COD) removal as function of time with distillery and dairy wastewater in dark-fermentative system (DFS-CDi, CDa) and augmented dark-photo fermentative system (ADPFS-EDi, EDa). 2.5. Bio-Electrocatalytic Analysis Electron-transfer reactions are integral components in every act of microbial metabolism. These electron discharges are studied using bio-electrochemical techniques viz., cyclic voltammetry (CV). This is a highly versatile technique which allows understanding of the electron transfer from redox mediator species of the biocatalyst into solution using a three electrode set-up. Understanding the behavior of biocatalyst and its nature by scanning a definite potential window during H 2 production is a significant aspect of this study. Voltammogram profiles ( vs. Ag/AgCl) showed significant variation in both control (DFS) and experimental (ADPFS) set-up with both the wastewaters. Reduction current (RC) was relatively higher compared to the oxidation current (OC) in ADPFS operation, indicating a feasible anaerobic microenvironment for the evolution of hydrogen ( Table 1 ). Analogous observations relating CHP was noticed in ADPFS (EDi and EDa) which was previously discussed. In the case of control DFS (CDi and CDa) as well the RC was higher than OC, and similarly, H 2 production was relatively higher in distillery wastewater than dairy wastewater. The catalytic redox currents varied with batch time and the peak values were corroborating RC with maximum CHP and/or OC with substrate degradation. In CDi, the RC gradually increased from the 0th h (0.07 µA) and increased to a peak value at 24th h (0.10 µA) and thereafter decreased at 72nd (0.06 µA). In CDa, RC decreased along the batch process (0.081 µA at 0th and 0.064 µA at 72nd h) which is not favorable for H 2 production. Likewise in case of EDi, the RC gradually increased from the 0th h (0.23 µA) to a peak value at the 48th h (0.40 µA) and thereafter declined at the 72nd h (0.37 µA). The disparity between CDi and EDi is the presence of PSB that involves acid consumption and consequently results in higher H 2 than DFS. While, EDa responded in the same manner as in case of CDa, i.e. , the RC values decreased from the batch start-up (0th h, 0.09 µA) till the end of the batch (72nd h, 0.053 µA) ( Figure 8 ). ijms-16-09540-t001_Table 1 Table 1 Comprehensive results of bioelectrochemical analysis carried out for both the control (DFS) and experimental systems (ADPFS). \n Time (h) OC (µA) RC (µA) β c (V/dec) β a (V/dec) R p (kΩ) EDi 0 0.25 0.23 0.219 0.408 15,216 24 0.24 0.29 0.263 0.625 11,473 48 0.27 0.40 0.157 0.352 14,734 72 0.24 0.37 0.165 0.377 15,173 CDi 0 0.12 0.07 0.257 0.544 15,038 24 0.10 0.10 0.121 0.557 16,325 48 0.08 0.08 0.059 0.667 15,440 72 0.06 0.06 0.070 0.632 15,290 EDa 0 0.07 0.09 0.683 0.158 16,430 24 0.06 0.06 0.205 0.622 17,580 48 0.041 0.055 0.165 0.607 12,300 72 0.042 0.053 0.155 0.726 16,480 CDa 0 0.067 0.081 1.770 0.576 13,330 24 0.060 0.064 0.212 0.584 18,890 48 0.054 0.060 0.192 0.608 18,136 72 0.050 0.064 0.184 0.681 17,020 DFS, dark fermentation system; ADPFS, augmented dark photo fermentation system; OC, oxidation current; RC, reduction current; β c , reduction slope; β a , oxidation slope; R p , polarization resistance. Further, these CV were analyzed for Tafel slopes which reveal the bioelectro-kinetic behavior of the biocatalyst in terms of exchange current density and electron transfer coefficients (oxidative β a , reductive β c ). The electron transfer during redox reactions between the biocatalyst and solid electrode need to overcome different barriers referred to as overpotentials. Higher oxidation slope suggests the requirement of higher activation energy that makes oxidation less favorable and vice versa. The same relationship applies to the reduction slope. Remarkable variation was seen in these slope values both with batch time and operating process condition (specifically nature of biocatalyst and wastewater used). However, overall the β c values were lower than the β a values which supports for a reductive microenvironment which is congenial for H 2 production. Rate of change in β c indicated the bioelectro-kinetics of biocatalyst used and variable profiles were noted with control and experimental systems. Interestingly, this change corroborated the RC of CV and CHP at that particular batch time. Besides, β c values were lower in control than in ADPFS, which indicates that dark-fermentative consortia have more tenacity for H 2 production. But, this was not observed so in this study because of the VFA accumulation, which might have disturbed the biocatalyst buffer capacity leading to a drop in pH. On the other hand, in ADPFS the VFA are utilized in the process of H 2 production. Figure 8 Cyclic voltammograms of dark-fermentative system (DFS- CDi, CDa) and augmented dark-photo fermentative system (ADPFS- EDi, EDa) operated with distillery and dairy wastewater. Besides, polarization resistance ( R p ) refers to the electron transfer from the biocatalyst at the solution electrode interface which is derived from Tafel analysis. The higher the R p , the lower will be the electron transfer making the process less favorable. In this study, the R p values are comparatively on a higher side because the electrode assembly is not continuously put in the system which assists in biofilm formation and for better electron conduction. The redox catalytic currents measured are due to suspended biocatalyst in the electrolyte, and in this manner heavy electron losses and resistances (referred to overpotentials) are noticed. The Tafel analysis therefore reports high activation loss (referred to as R p ). However, the solution conductivity is a criterion which showed considerable variation with the nature of wastewater used and the biocatalyst composition. In this study, R p values were relatively lower with distillery wastewater than dairy wastewater and besides, these values were higher in control DFS than experimental ADPFS ( Table 1 ). Notably the polarization resistance affects the Hydrogen Evolution Reaction (HER) kinetics which is consecutively influenced by the microbial catalytic capability within the tested potential region [ 30 ]. Perhaps, this could be a possible reason for high RC response and correspondingly higher H 2 production."
} | 6,046 |
39946539 | PMC11848305 | pmc | 7,174 | {
"abstract": "Significance For photoautotrophs, light is both an essential driver of photosynthesis and a potential threat when in excess. Our research unveils the photoreceptor-induced Light Harvesting complex-Like 4 (LHL4) as a protagonist in the photoprotective response against harmful effects of high light (HL) in Chlamydomonas reinhardtii . LHL4 is a critical component of the core of the photosynthetic machinery where it limits the production of damaging molecules, ensuring cell survival under HL. Thus, we provide insights into the molecular and mechanistic strategies employed by photosynthetic microalgae confronted with light stress, improving our understanding of their acclimation responses.",
"conclusion": "Conclusion Our study reveals a key role for the LHL4 protein in photoprotection. LHL4 protects PSII monomers by associating with the core complex via two antenna proteins, CP43 and CP47. This protection takes place during the early stages of photodamage upon HL exposure, but also under prolonged stress conditions when light intensity is too high. Protection conferred by LHL4 occurs following cell exposure to UV-B and HL, which concomitantly activate the UVR8 and PHOT photoreceptor signaling pathways that converge likely at the COP1 level in Chlamydomonas. Our phylogenetic analysis indicates that LHL4 is exclusively present in green microalgae. However, its crucial function in photoprotection of these organisms suggests the existence of similar actors in other phototrophs. Consistently, specific High-Light-Inducible Proteins (HLIPs) have been identified in cyanobacteria, where they associate with CP47 and intermediates of PSII assembly modules ( 56 ). Remarkably, cyanobacteria lacking HLIPs are viable but experience extreme light-induced stress levels, suggesting the presence of protective mechanism in prokaryotic photosynthetic organisms similar to the one conferred by Chlamydomonas LHL4. Investigating potential candidates in other microalgae and flowering plants may unlock a deeper understanding of photoprotection mechanisms across diverse photosynthetic life forms.",
"discussion": "Results and Discussion LHL4 Is Highly Induced upon UV-B. We analyzed the effect of UV-B on the membrane-enriched proteome of Chlamydomonas and identified several proteins potentially involved in photoprotection among those accumulating under 16 h UV-B ( Fig. 1 A ). After UV-B exposure, 46 proteins showed a significant increase in abundance, and nine proteins showed a significant decrease in abundance. Several proteins that showed increased accumulation in response to UV-B are involved in PSII biogenesis, stability, or repair [Filamentous Temperature-Sensitive-like H (FTSH-like) and Degradation of periplasmic proteases ( 40 ), and members of the High Chlorophyll Fluorescence (HCF) family ( 41 )], as well as in NPQ (the two LHC-like proteins LHCSR1 and LHCSR3) ( SI Appendix , Table S1 ). Interestingly, we identified a third LHC-like protein, namely LHL4, enriched more than seven times under UV-B compared to control ( Fig. 1 A and SI Appendix , Table S1 ). In a parallel RNA-Seq analysis, LHL4 was also identified as highly induced at the transcript level in response to 1 h exposure to UV-B, in a UVR8-dependent manner, along with 831 genes ( Fig. 1 B and SI Appendix , Fig. S1 and Table S2 ). While LHCSRs were detected in both experiments, PSBS was identified among the top-induced genes in our transcriptome dataset ( SI Appendix , Table S2 ), as expected ( 12 , 25 ), but showed only limited accumulation at the protein level in the UV-B-treated samples ( SI Appendix , Table S1 ), consistent with its reported limited stability over time ( 11 , 14 ). In total, 44 out of the 46 proteins accumulating in response to UV-B were also found to be transcriptionally induced ( Fig. 1 C and SI Appendix , Table S3 ). LHL4 is a 285-amino acid protein with three predicted transmembrane domains that differs from the other LHC-Like proteins (including the ELIP family) because it contains an exceptionally long predicted loop between the second and third transmembrane domains ( Fig. 1 D ) ( 42 , 43 ). LHL4 homologs are uniquely found in green microalgae ( 44 , 45 ), where they are closely related to PSBS, but belong to a different clade ( Fig. 1 E ). Fig. 1. LHL4 is induced by UV-B in Chlamydomonas. ( A ) Mass spectrometry (MS)-based quantitative comparison of membrane-enriched proteomes from Chlamydomonas exposed to 16 h of supplemental UV-B [Low Light (LL, 20 µmol photons m −2 s −1 ) + UV-B (0.2 mW cm − 2 )] compared to untreated control (LL). Volcano plot displaying the differential abundance of proteins in the membrane proteomes analyzed by MS-based label-free quantitative proteomics. The volcano plot represents the −log 10 ( P -value), (limma P -value, y axis) plotted against the log 2 FC (LL+UV-B vs. LL, x axis) for each quantified protein. Green and red dots represent proteins significantly enriched in LL+UV-B and in LL samples, respectively (log 2 FC ≥ 1 and −log 10 ( P -value) ≥ 2.11, corresponding to a Benjamini–Hochberg FDR < 1%). Dots representing LHL4, LHCSR1, and LHCSR3 proteins are indicated. ( B ) RNA-Seq analysis of Chlamydomonas exposed to 1 h LL+UV-B compared to LL. Volcano plot displaying the differential abundance of transcripts by representing the −log 10 (FDR), ( y axis) plotted against the log 2 FC (LL+UV-B vs. LL, x axis). Dots representing LHL4 , PSBS1 / PSBS2 (encoding PSBS), LHCSR1 , and LHCSR3.1 / LHCSR3.2 (encoding LHCSR3) transcripts are indicated. ( C ) Venn diagram showing the overlap of proteins (blue) and transcripts (orange) significantly enriched in LL+UV-B ( SI Appendix , Tables S1 and S2 ). ( D ) Scaled schematic representation of LHC-Like protein sizes and the locations of predicted chlorophyll a/b binding domains (green, predicted using the Superfamily database) and transmembrane domains (gray motifs) in LHL4 and in consensus Chlamydomonas ELIP, LHCSR, and PSBS like protein sequences. The red arrow indicates the presence of an exceptionally long loop in LHL4 compared to the other proteins. aa, amino acids. ( E ) Phylogenetic tree generated from aligned protein sequences of LHC-Like LHCSRs, LHL4s, PSBSs, and ELIPs homologs. Proteins from Chlamydomonas and Arabidopsis are approximatively emphasized by green and gray dots, respectively. Due to their proximity, the two ELIP dots for Arabidopsis overlap. LHL4 Expression Is Controlled by UVR8 and PHOT Photoreceptors and Depends on CrCOP1 and CrCO. LHL4 gene expression was found to be rapidly and transiently induced in response to both UV-B and HL ( Fig. 2 A ) ( 25 , 42 , 43 ). LHL4 showed a higher induction by UV-B than HL at both transcript and protein levels. The LHL4 protein level rapidly increased up to 4 h and remained stable for 8 h upon both light treatments ( Fig. 2 B ). However, the LHL4 abundance rapidly decreased when cells were returned to low light (LL), unlike LHCSR1 and LHCSR3, which level remained stable for at least 8 h posttreatment ( Fig. 2 C ) ( 46 ). Fig. 2. LHL4 expression is controlled by UVR8 and PHOT photoreceptor signaling pathways. ( A ) RT-qPCR analysis of LHL4 expression in WT 137C cells grown under LL (20 µmol photons m −2 s −1 ) and then transferred to LL supplemented with UV-B (0.2 mW cm − 2 , LL+UV-B), or HL (300 µmol photons m −2 s −1 ) for the indicated times. The data were normalized to the LHL4 levels at time 0 for each condition. Individual data points of biological replicates and means ± SD are shown ( n = 4). ( B ) Immunodetection of LHL4 in cells exposed for up to 8 h under LL, LL+UV-B, or HL. ATP synthase beta subunit (ATPB) levels were used as loading control. ( C ) Comparative stability of LHL4 and LHCSRs proteins. Immunodetection of LHL4, LHCSR1, LHSCR3, and ATPB in cells sequentially exposed to LL (20 μmol photons m −2 s −1 ) supplemented with UV-B (0.2 mW cm −2 ) for 16 h and then HL (300 μmol photons m −2 s −1 ) for 4 h. Cells were finally placed under LL (20 μmol photons m −2 s −1 ) for the indicated times. ATPB was used as loading control. ( D – F ) RT-qPCR analysis of LHL4 expression in ( D ) uvr8 and phot , ( E ) cop1 hit1 , and ( F ) crco and crblz3 cells grown under LL (0), or exposed for 1 h to LL+UV-B, or HL. Data are normalized to levels in respective WTs (strains 137C, CC124, CC5325) under LL (“0”). Values of independent measurements and means ± SD are shown ( n = 3). ( G – I ) Immunodetection of LHL4 in uvr8 and phot , ( H ) cop1 hit1 , ( I ) crco and crblz3 , as well as their respective WT cells grown under LL (0), or 6 h LL+UV-B, or HL. ATPB was used as loading control. As LHL4 expression is induced in response to UV-B and HL, we examined the accumulation of LHL4 transcripts and LHL4 proteins in uvr8 and phot mutants. UV-B-dependent induction of LHL4 expression was abolished in the uvr8 mutant, but not affected in a phot mutant ( Fig. 2 D ). Consistently, UV-B-induced LHL4 protein accumulation was absent in uvr8 but comparable to wild type (WT) in phot cells ( Fig. 2 G ). Under HL, a comparably weak LHL4 induction was observed in both photoreceptor mutants and WT cells ( Fig. 2 D ). However, accumulation of LHL4 protein was reduced in response to HL in the phot mutant compared to WT and uvr8 ( Fig. 2 G ), similar to LHCSR3 ( SI Appendix , Fig. S2 A ) ( 17 ). Altogether, our data demonstrate that the induction of LHL4 accumulation by UV-B and HL is mediated by UVR8 and at least partially PHOT, respectively. We thus examined the involvement of downstream signaling components on LHL4 induction and LHL4 accumulation. We first used the cop1 hit1 loss-of-function mutant strain that contains an Arg-1256-to-Pro mutation (CrCOP1 R1256P ) in the C-terminal WD40 domain ( 12 , 33 ). Under both UV-B and HL, the cop1 hit1 mutant exhibited a much weaker induction of LHL4 transcripts and reduced LHL4 protein accumulation in comparison to WT ( Fig. 2 E and H ). This result is similar to the reduced accumulation of LHCSR1 and LHCSR3 ( SI Appendix , Fig. S2 B ), as well as PSBS ( 12 ), confirming that the UV-B and HL signaling pathways converge at the level of COP1. It is also of note that cop1 hit1 is more strongly affected in HL-induced LHCSR3 and LHCSR1 accumulation than phot ( SI Appendix , Fig. S2 A and B ), suggesting that additional HL-induced signaling pathway(s) converge at COP1. We next investigated transcription factors potentially involved in LHL4 regulation upon UV-B and HL by comparing LHL4 levels in crco ( 28 , 32 ) and crblz3 [CrBLZ3 is a putative AtHY5 ortholog, ( 39 )] mutants with WT. At the transcriptome-wide level, we found that both CrCO and CrBLZ3 are required for UV-B-regulated gene expression with CrCO playing a major role ( SI Appendix , Fig. S3 A and B and Table S2 ). In agreement, accumulation of LHL4 mRNA and LHL4 protein in response to both UV-B and HL was severely impaired in crco but was similar between crblz3 and WT ( Fig. 2 F and I and SI Appendix , Fig. S2 C and Table S2 ). We conclude that UV-B- and HL-dependent accumulation of LHL4 depends on UVR8 and partially on PHOT, respectively, and in both cases involves CrCO-dependent transcriptional activation of LHL4 expression. LHL4 Binds the PSII Monomer upon UV-B Exposure. To elucidate the integration of LHL4 within the Chlamydomonas photosynthetic apparatus, we performed biochemical analysis under nondenaturing conditions by Blue native polyacrylamide gel electrophoresis (BN-PAGE) of solubilized thylakoid complexes isolated from WT cells exposed to UV-B. We found that LHL4 was only detected upon UV-B exposure and migrated in the LHCII trimer region, as well as in two different spots in proximity to the PSII monomer/cytochrome b 6 f bands ( Fig. 3 A ). We then used complexome profiling ( 47 ) to analyze the protein composition of these two spots by cutting this specific region of the gel into six slices and examining their protein content using mass spectrometry (MS) ( Fig. 3 B ). LHL4 was found to be localized in two different organizations of the PSII monomer. The lower band (band 5) corresponded to a PSII monomer containing only one of the two core antennae, specifically CP47, known as the RC47 intermediate complex during PSII assembly ( 48 ). The upper band (band 1) consisted of the full PSII monomer with both core antennae, CP43 and CP47. LHL4 was also detected in LL-acclimated samples with the same accumulation pattern ( Fig. 3 B ), albeit at a much lower level compared to UV-B-acclimated samples. This suggests that a basal level of the protein is also constitutively present within the PSII monomer in nonstressed samples. To identify potential interactions of LHL4 within the PSII monomer, we then used a combination of BN-PAGE and cross-linking MS (XL-MS) analysis. This approach allows capturing protein–protein interaction by creating covalent bonds between amino acids in close proximity and can be directly applied to complexes previously separated via BN-PAGE ( 49 ). Importantly, the chemical cross-linking reaction did not produce any noticeable difference on the presence of major photosynthetic complexes ( SI Appendix , Fig. S4 A ). XL-MS data acquired on these specific gel bands were first validated by mapping the cross-links on known PSII and cytochrome b 6 f structures ( SI Appendix , Fig. S4 B – D ), where most fall within an acceptable 20 Å distance cut-off ( SI Appendix , Fig. S4 C ) ( 50 ). We detected two cross-links of LHL4 with both CP43 and CP47, these are compatible with two putative positions predicted by using AlphaFold2 multimer ( 49 ) ( SI Appendix , Fig. S5 ). The distances of these two reproducible cross-links with CP43 and CP47 are of ~12 and ~20 Å, respectively ( SI Appendix , Fig. S5 ), thus validating both structural models. Taken together, these data suggest that LHL4 binds to a fully assembled monomer on both inner antennae ( Fig. 3 C ), producing a complex at higher molecular weight detected by immunodetection ( Fig. 3 A ) and complexome profiling (band 1, Fig. 3 B ). Alternatively, LHL4 may bind to a partially assembled RC47 on CP47 only, resulting in a smaller complex ( Fig. 3 A and band 5 in Fig. 3 B ). In both cases, LHL4 binds to the site where the second PSII monomer attaches, suggesting that the protein may need to be removed prior to dimer assembly. We propose that LHL4 acts as or is a part of a specific antenna for the PSII monomer, capable of binding to the complex early in its assembly process by a direct connection to the two core antenna proteins CP43 and CP47. Fig. 3. LHL4 localizes with PSII monomers at the dimerization interface interacting with the inner PSII antenna proteins CP43 and CP47. ( A ) BN-PAGE of thylakoids extracted from cells exposed to LL (20 µmol photons m −2 s −1 ) or LL supplemented with UV-B (LL +UV-B; 0.2 mW cm −2 ). Second dimension blots indicate LHL4 (red circles), and its absence in LL cells. A blue circle indicates a nonspecific band detected by the LHL4 antibody in both WT and lhl4 . ( B ) The middle section of the gel was divided into six parts, as shown in the Left image. The relative abundance of LHL4 ( Left graph) and the main components of the PSII monomer ( Right graph; UV-B acclimated cells) were assessed by MS. n = 4 ( C ) LHL4 interactors CP43 and CP47 uncovered by cross-linking and in-gel digestion of the BN-PAGE band. The two AlphaFold2 pairwise predictions of LHL4-CP43/47 are shown overlaid on the PSII monomer structure (PDB 6KAD, light blue) and both are localized at the dimerization interface. LHL4 Prevents PSII Photoinhibition upon Exposure to HL. The sequence homology of LHL4 with LHCSR and PSBS, its accumulation in response to both UV-B and HL, as well as its location within the PSII monomer may suggest a role for LHL4 in photoprotection. To investigate this possibility, we generated lhl4 knock-out mutants using CRISPR-Cas9 and evaluated their tolerance to HL compared to WT. LL-acclimated cells were exposed to HL (900 µmol photons m −2 s −1 ) for 1 h, followed by 1 h of recovery under LL. Their photosynthetic capacity was monitored as the maximum quantum yield of PSII (F v /F m ). Lower F v /F m values indicate, but are not limited to, PSII damage due to photoinhibition ( 51 ). lhl4 mutants showed no difference in PSII antenna size compared to WT ( SI Appendix , Fig. S6 ) but exhibited impaired recovery from light stress ( Fig. 4 A and B ). To further characterize the role of LHL4, we conducted parallel experiments using lincomycin to inhibit the synthesis of chloroplast-encoded proteins of PSII, effectively blocking its repair system. Upon lincomycin treatment, we assume that recovery is linked to reversibly photodamaged PSII, as suggested before ( 52 ). On the contrary, the additional recovery observed without the inhibitor likely reflects irreversible photodamaged PSII, which can only be repaired via assembly of de novo synthesized PSII proteins ( 52 ). As expected, we observed a significant reduction in the recovery of both strains compared to untreated samples upon addition of lincomycin. While the inhibition rate was the same, a large difference in the recovery between WT and lhl4 became evident ( Fig. 4 A and B ). This finding rules out a direct role for LHL4 in the repair process, in agreement with a parallel report ( 53 ). Instead, it implies a role of LHL4 in PSII recovery from light-mediated degradation, which would stem from increased pool of reversibly photodamaged PSII. By binding to the PSII monomer at the core antenna subunits CP43 and CP47, we propose that LHL4 might protect the PSII monomer pool from degradation during HL. This could help preserve its structural integrity and facilitate the assembly of functional PSII dimers, leading to a more efficient recovery. Fig. 4. LHL4 protects PSII and facilitates cell survival under HL. WT and lhl4 cells were grown under LL (20 µmol photons m −2 s −1 ). ( A ) Cells were exposed to HL (900 µmol photons m −2 s −1 ) for 1 h and then placed under LL for 1 h and their maximum PSII quantum yield (Fv/Fm) was monitored after 1 min of dark relaxation. The data presented are representative of four biological replicates. Errors bars represent the SE of the means of three technical replicates. ( B ) The rate of recovery was calculated using the data from panel A by determining the slope of the first 30 min of recovery under LL and 1 min of dark relaxation. n = 4. The asterisk indicates statistical difference (*** P < 0.001, **** P < 0.0001). ( C ) WT and lhl4 cells were exposed to HL (600 µmol photons m −2 s −1 , red light) and 1 O 2 accumulation was monitored over 80 min of exposure. n = 3. Asterisks indicate statistical differences (* P < 0.05). ( D ) Photographs of WT and lhl4 cultures were taken after LL acclimation (0) and then after 1, 1.5, and 2 h under 2,000 µmol photons m −2 s −1 . A defect in photoprotection generally leads to ROS production during photosynthesis, mainly singlet oxygen ( 1 O 2 ) at the PSII level ( 8 ). We measured 1 O 2 levels in WT and lhl4 during HL exposure ( Fig. 4 C ). During early HL exposure (20 min), i.e., when LHL4 expression is still very low ( Fig. 3 B ), 1 O 2 levels are comparable in both strains. However, we observed a significant overaccumulation of 1 O 2 in the lhl4 strain compared to the WT under prolonged stress, while LHL4 accumulated at a higher level in the WT ( Fig. 4 C ). Our data suggest that the protection of PSII monomers by LHL4 facilitates PSII renewal and helps limiting photoinhibition and 1 O 2 production. Accordingly, in case of higher (2,000 µmol photons m −2 s −1 ) and prolonged light stress exposure, in contrast to WT, lhl4 cultures were unable to survive, resulting in complete bleaching within 2 h of exposure ( Fig. 4 D ). To cope with the harmful effects of light stress, photosynthetic organisms can activate NPQ. We investigated the interplay between the NPQ-driven and LHL4-mediated protection. First, we monitored the phenotype of LHL4 during the first 3 h of HL exposure. During this phase, cells begin to accumulate the LHCSR3 protein, which is responsible for NPQ activity ( 54 , 55 ). We observed a stable difference in the recovery between WT and lhl4 cells during the first 2 h of HL ( SI Appendix , Fig. S7 A ). At this point, NPQ became activated to a similar extent in both strains ( SI Appendix , Fig. S7 B ). This observation suggests that LHL4 does not directly contribute to this process in Chlamydomonas. Alternatively, NPQ capacity could compensate, at least partially, the absence of protection conferred by LHL4 in the mutant. After 3 h of exposure, NPQ was highly activated, leading to a higher Fv/Fm and similar fluorescence recovery between the two strains. Overall, these results suggest that LHL4 primarily catalyzes a recovery from light stress when cells have not yet developed their NPQ capacity. We further dissected the impact of NPQ on LHL4-mediated protection, focusing on UV-B acclimated cells that had fully developed their NPQ capacity ( SI Appendix , Fig. S7 C ). We observed that LHL4 accumulation in UV-B-acclimated cells was significantly lower after 1 h of HL exposure compared to LL-acclimated cells ( SI Appendix , Fig. S7 D ). Additionally, the lhl4 strain showed no significant difference in Fv/Fm compared to the WT during the first hour of light stress and subsequent recovery, which was significantly higher than in nonacclimated cells ( SI Appendix , Fig. S7 E and F ). These results confirm that when cells have acquired NPQ capacity, energy dissipation via NPQ prevents photoinhibition, reducing PSII damage and turnover, and limits LHL4 induction, which is not essential under these conditions. We further evaluated the impact of LHL4 under long-term exposure and stronger HL intensity (2,000 µmol photons m −2 s −1 ) in UV-B acclimated samples. Under these conditions, LHL4 levels are maintained at a very high level ( SI Appendix , Fig. S7 G ). Both WT and lhl4 cells retained their green pigmentation longer than LL-acclimated cells ( SI Appendix , Fig. S7 H vs. Fig. 4 D ). However, after 8 h of prolonged exposure, lhl4 failed to endure the stress and bleached, whereas the WT cells remained green ( SI Appendix , Fig. S7 H ). These results suggest that under prolonged and extremely high irradiance, NPQ alone is no longer sufficient to prevent photodamage. Therefore, the cumulative effect over time of the absence of PSII monomers-driven LHL4 protection affects cell survival. This complementary role of NPQ and LHL4 in these extreme conditions underscores the importance of LHL4 in priming UV-B-induced photoprotection in Chlamydomonas , alongside LHCSRs ( 12 )."
} | 5,708 |
37276409 | PMC10268325 | pmc | 7,175 | {
"abstract": "Significance Coral reefs are biodiversity hotspots that are in decline due to stressors associated with climate change. Thus, a critical research goal is to achieve a deeper understanding of basic coral biology to inform conservation efforts. This goal has been difficult to achieve due to the lack of genetic tools for corals. Here, we used CRISPR/Cas9 mutagenesis to show that a predicted bicarbonate transporter that apparently evolved specifically in the common ancestor of the stony corals is indeed required for formation of the calcium-carbonate skeleton in young coral colonies. The ability to do such genetic analysis in adult corals should now allow critical tests of hypotheses about gene function and the generation of stable, genetically modified lines for research and conservation.",
"discussion": "Discussion Although it has been hypothesized that passive diffusion of CO 2 to the site of calcification and subsequent conversion to bicarbonate may occur ( 32 ), the requirement for SLC4γ for septum formation suggests strongly that bicarbonate transport is needed to reach the concentrations needed for calcium-carbonate deposition. Moreover, the strong phenotypes of SLC4γ -deficient animals also demonstrate that SLC4γ function cannot be supplied by the other A. millepora SLC4 family members, indicating that SLC4γ has been specialized during evolution to play its distinctive role. This specialization presumably occurred after the coral-specific gene duplication of SLC4β, as indicated by the differences in SLC4γ and SLC4β expression during the onset of biomineralization in young polyps. SLC4β and SLC4γ proteins are also differentially expressed in adult S. pistillata, where SLC4γ is highly expressed in calicoblastic cells and SLC4β appears to be expressed ubiquitously, suggesting that their specialization of function may persist during biomineralization in adult corals ( 27 , 32 ). Although our analysis shows that the function of SLC4γ is required for calcification, the details of the bicarbonate transport remain unclear. SLC4γ is found at both the apical and basal surfaces of calicoblasts ( 27 , 32 ) and may be involved in transporting bicarbonate into the calicoblasts, out of the calicoblasts to the site of calcification, or both. A goal of future research will be to answer such questions ( 9 ). For example, in future experiments, the cellular nature of the bicarbonate transport disrupted in the SLC4γ mutants might be determined by measuring the transport of isotope-labeled bicarbonate in wild-type and mutant polyps. Coral species differ dramatically in skeletal morphology ( 37 ), but the molecular underpinnings of these differences are unknown. Our analyses indicate that SLC4γ is present in all stony corals and is required for skeleton formation in at least one complex coral, A. millepora . The most parsimonious explanation of these results is that SLC4γ was neofunctionalized in the last common ancestor of the stony corals to play a role in biomineralization that has since been conserved. Consistent with this interpretation, treatment with a broad-acting chemical inhibitor of bicarbonate-anion transporters reduces calcification rates in the robust coral S. pistillata ( 15 , 29 ). However, there may be differences in the specific functions of SLC4γ across coral species. Indeed, other transporters associated with skeleton formation have different localization patterns in complex and robust corals ( 31 ). The striking differences between robust and complex corals in the genomic arrangement of SLC4γ and SLC4β ( Fig. 1 B ) might affect the positions of these genes relative to transcriptional enhancers, topologically associated domains, and other regulatory elements, and thus affect their expression and effects on morphology ( 38 – 40 ). It may be possible to explore this hypothesis and whether SLC4γ function is conserved by comparing robust and complex coral species for SLC4γ expression at a single-cell level and the phenotypes associated with SLC4γ mutations. The stereotypical pattern of early-forming septa should allow the detection of subtle defects produced by such mutations and thus make young coral polyps a sensitive model for study of the genetic basis of biomineralization. By coupling CRISPR/Cas9 genome editing with amplicon sequencing of individuals, we were able to directly link genotype to phenotype in individual coral animals. The extent of the correlation was remarkable, given that, in principle, an individual could have a low overall frequency of mutations but a strong phenotype, or the reverse, depending on whether the relevant clones of cells (in this case, presumably calicoblasts) carried the mutations in the mosaic animals. This study also extends the range of CRISPR/Cas9 methods for corals by demonstrating that it is possible to generate and analyze the phenotypes of mutant juveniles (and, by extension, adults). These advances should enable direct and rigorous genetic analyses of various ecologically important traits such as the determinants of algal-symbiont specificity and heat tolerance. As corals continue to decline from stressors associated with climate change, it will be particularly important to analyze the molecular mechanisms of coral stress-response pathways, including those involved in response to reactive-oxygen species ( 41 , 42 ), immune stress ( 43 , 44 ), and unfolded-protein stress ( 45 , 46 ), all of which have been associated with the breakdown of the symbiosis during heat-induced bleaching. A more detailed understanding of the genetic basis of ecologically important traits will be valuable in developing and evaluating novel reef management and restoration strategies ( 47 – 50 ). Furthermore, by combining CRISPR/Cas9-based genetic manipulations with the ability to maintain corals long-term and achieve spawning throughout the year in the laboratory ( 51 , 52 ), it should be possible to initiate genetic studies throughout the year and to generate mutant coral lines, which in turn should allow at least the following: i) The long-term investigation of the cellular and developmental functions of target genes without relying on seasonal coral-spawning events. ii) The sharing of mutant lines among research groups. iii) The generation of double mutants and reporter lines through genetic crosses. These technical advances should open the way to a more rigorous understanding of coral molecular and cellular biology, which should be a vital resource in the battle to save these ecologically critical organisms from the ravages of climate change ( 50 )."
} | 1,653 |
37655360 | PMC10466179 | pmc | 7,176 | {
"abstract": "Superhydrophobic porous materials exhibit remarkable stability and exceptional efficacy in combating marine oil spills and containing oily water discharges. This work employed the multi-template high internal phase emulsion method to fabricate a multi-template porous superhydrophobic foam (MTPSF). The materials were characterized through SEM, IR spectroscopy, contact angle measurement, and an electronic universal testing machine. Moreover, the materials' oil–water separation capability, reusability, and compressibility were thoroughly evaluated. The obtained results demonstrate that the material displays a water contact angle of 143° and an oil contact angle of approximately 0°, thus exhibiting superhydrophobic and superoleophilic properties. Consequently, it effectively facilitates the separation of oil slicks and heavy oil underwater. Furthermore, the MTPSF conforms to the second kinetic and Webber–Morris models concerning the oil absorption process. MTPSF exhibits an outstanding oil absorption capacity, ranging from 39.40 to 102.32 g g −1 , while showcasing reliable reusability, high recovery efficiency, and excellent compressibility of up to 55%. The above exceptional attributes render the MTPSF highly suitable for oil–water separation applications.",
"conclusion": "4. Conclusions In this study, porous superhydrophobic foam (MTPSF) was fabricated by applying the multi-template high-internal-phase emulsion method. The material with a high water contact angle of 143°, a water sliding angle of 6°, an oil contact angle of about 0°, and a porosity of 97.39% is a superhydrophobic–superoleophilic three-dimensional porous material. The OSPs and EAC added to the emulsion system could construct smaller-diameter pores, and the former also stabilised the emulsion precursors. MPTF-30 demonstrated excellent oil absorption capacity, with a saturation absorption multiplicity ranging from 39.40 g g −1 to 102.32 g g −1 for the tested oils. The sorption process of material aligned well with the pseudo-second kinetics model and the Webber–Morris model. Furthermore, material exhibited remarkable reusability and toughness. It could undergo 10 absorption–desorption cycles without significant damage and endure up to 55% ultimate compression. In conclusion, MTPSF is an ideal material for future oil–water separation applications due to its outstanding performance.",
"introduction": "1. Introduction With the rapid development of human civilization and the economy, the demand for petroleum resources has been increasing. The ocean, which covers approximately 71% of the global area, contains abundant oil and gas resources and has gradually become a vital source of petroleum reserves. However, oil spills often occur during the exploration, extraction, transportation, processing, and storage of oil. Two oil spills 1 occurred on June 4th and 17th, 2011, at the B and C platforms of the Penglai 19-3 oil field, resulting in a total discharge of 3300 barrels of crude oil into Bohai Bay. The oil spills not only have fatal impacts on the plankton, 2–5 benthic invertebrates, 6 and fish 7,8 in the ocean, but they also pose a significant risk to nearby birds, 9 mammals, 10 and reptiles. 11 The adverse effects of oil spills on marine ecosystems are severe and long-lasting. Plankton, which forms the base of the marine food chain, is especially vulnerable to oil spills. Oil spills can cause the death of planktonic organisms, which, in turn, affects the entire food chain, leading to the death of larger fish and marine mammals. Benthic invertebrates, which live on or under the ocean floor, are also highly susceptible to the toxic effects of oil spills. These organisms play a crucial role in the marine ecosystem, as they help recycle nutrients and maintain the marine environment's balance. The death of benthic invertebrates and plankton may lead to the collapse of the entire ecosystem. Oil spills also have a significant impact on the health and survival of birds, mammals, and reptiles that inhabit the coastlines and nearby islands. These animals are exposed to the toxic effects of oil spills when they come into contact with contaminated water or when they ingest contaminated prey. The ingestion of oil can cause damage to the internal organs and can lead to death. Several commonly 12 used methods for removing oil spills include physical absorption, chemical methods, thermal combustion, and biodegradation. Physical absorption 13 involves using oil-absorbing materials to absorb, recover, and reuse the spilled oil. This method effectively removes oily wastewater from the marine environment, eliminating secondary pollution risks during subsequent treatment. 14 Environmental experts consider physical absorption a highly efficient and feasible method for treating marine oil spills, owing to the high absorption efficiency and ease of oil recovery of absorbent materials. Hethnawi 15 et al. have proposed a novel approach for synthesizing titanomagnetite (NTM) nanoparticles through a green method that involves varying the initial concentration and temperature of iron precursors. The resulting NTM nanoparticles have shown promising properties for oil spill remediation. The water contact angle of the nanoparticles was observed to be 141.86°, indicating their excellent hydrophobicity, while the maximum adsorption capacity for crude oil was determined to be 38 ± 2 g g −1 . Guo 16 et al. developed a novel method to synthesize graphene/hydroxypropyl methylcellulose composite aerogels (RGA/HPMCs) using a two-step hydrothermal reduction technique combined with ice templates. The resulting RGA/HPMCs possess high photothermal conversion efficiency, allowing them to rapidly increase their surface temperature by utilizing solar energy and significantly reducing the viscosity of crude oil for efficient adsorption. The saturation absorption capacity of RGA/HPMCs for crude oil was as high as 124.43 g g −1 and could be achieved within 22 minutes. The mechanical properties of the RGA/HPMCs were evaluated through 300 cycles of compression experiments under 70% strain, and no significant damage was observed. Moreover, the aerogels exhibited good reusability, with more than 90% of the initial adsorption capacity maintained after 5 cycles of absorption–extrusion desorption. The results suggest that RGA/HPMCs have great potential as a highly efficient and reusable adsorbent for the remediation of crude oil spills. Ouyang 17 et al. have reported on the successful modification of sponge surfaces using poly(cyclotriphosphazene nitrile- co -bisphenol AF) (PZAF), resulting in a modified sponge (Sponge@PZAF) with a high water contact angle (WCA) of 153° and a high absorption capacity of 71.04 to 135.19 g g −1 for various greases and organic solvents within 30 s. The fine three-dimensional mesh structure of the sponge aids in the efficient and rapid adsorption of oil and grease, facilitating full saturation of the chamber. After 10 cycles of oil absorption–desorption recovery, Sponge@PZAF exhibited an absorption capacity as high as 69.76–132.44 g g −1 , indicating excellent recoverability. In previous work, 18 our research group successfully developed superhydrophobic foam materials (OSPs@Foam) with a porous structure based on a high internal phase emulsion of Span 80 and oyster shell powders (OSPs) synergistic stabilizers. This compound showed excellent lipophilic hydrophobicity and absorption–desorption, enabling selective absorption and oil recovery from oil-containing water for effective oil–water separation. However, the material's mechanical properties were suboptimal, with a tendency towards rupture. In order to address this limitation, compressible superhydrophobic foams will be constructed using a multi-template high-internal-phase emulsion method in this study. Unlike the conventional single-template high-internal-phase emulsion method, which applies only a single substance as the pore-making template (usually pure water 19,20 or organic solvents 21 ), this study utilizes pure water, ethyl acetate (EAC), and OSPs as multiple templates. During the material fabrication, the water and EAC will be removed from the emulsion system through the drying procedure, and the OSPs will be eliminated through a reaction with hydrochloric acid. The application of multi-template high-internal phase emulsion method could construct more pores within the material and enhance its compressibility. The resulting material will be characterized for density, porosity, microstructure, chemical composition, wettability, and mechanical properties and tested for oil–water separation ability and reusability. This study aims to improve the existing foam material and enable its use in more demanding applications.",
"discussion": "3. Results and discussion 3.1 Microstructure and pore size distribution The microstructures of MTPSF-10, MTPSF-20, MTPSF-30, and OSPs@Foam are depicted in Fig. 3 . All materials possess three-dimensional porous structures. The pore size statistics for MTPSF-10, MTPSF-20, MTPSF-30, and OSPs@Foam are presented in Table 2 . As the amount of internal phase water increases, MTPSF's median and average values of pore diameters decrease, along with an increase in the cumulative frequency of pores with a diameter less than 3 μm. Compared to OSPs@Foam, MTPSF-30 exhibits even smaller median and average values, as well as larger cumulative frequency values for pores smaller than 3 μm. That can be attributed to the utilization of the multi-template high internal phase method in the production of MTPSF-30. Fig. 3 (a 1 –d 1 ) SEM images of MTPSF-10, MTPSF-20, OSPs@Foam and MTPSF-30; (a 2 –d 2 ) expanded SEM images of MTPSF-10, MTPSF-20, OSPs@Foam and MTPSF-30; (a 3 –d 3 ) pore size distribution and cumulative frequency of MTPSF-10, MTPSF-20, OSPs@Foam and MTPSF-30. The pore size distribution for samples Samples The diameter of pores Mean (CI 95% a )/μm Median/μm <3 μm/% MTPSF-10 3.29(0.228) 2.77 54.3 MTPSF-20 2.72(0.177) 2.23 67.3 OSPs@Foam 2.53(0.176) 2.06 74.0 MTPSF-30 2.46(0.236) 1.95 75.3 a Confidence interval (CI 95%) was calculated based on multiple experiments using the same sample aliquot. The pores in MTPSF originate from various mechanisms. Firstly, at high temperatures, the water inside the material can break through the material's skeleton, resulting in pore formation. 23 The increased proportion of water in the internal phase leads to thinning the emulsion's organic phase film. This thinning makes the skeleton formed by the material thinner and more susceptible to breakage by the internal water. Additionally, a higher proportion of water means increased impact energy, favouring pore formation. Secondly, during the polymerization reaction of DVB and St, the organic phase undergoes a volume shrinkage from liquid to solid, resulting in the formation of numerous pores. 24,25 As the proportion of water in the internal phase increases, the spherical shape of the aqueous droplets in the emulsion transitions into various polyhedral forms, and even rhombic dodecahedra, 26 and the area of each facet of the droplet will be reduced accordingly. During monomer polymerization, each facet of the polyhedron has an equal probability of forming pores. Therefore, droplets with more facets generate more pores and smaller pore diameters. Thirdly, EAC in the organic phase can break through the material's skeleton at high temperatures, creating pores. Fourthly, OSPs embedded within the material's skeleton react with hydrochloric acid and migrate out of the material during subsequent acidification treatment. This migration leaves behind pores in the previously occupied OSP locations. In summary, the multi-template high internal phase method employed in MTPSF facilitates the construction of a larger number of pores with smaller diameters. 3.2 Chemical characteristics \n Fig. 4 displays the infrared spectra of MTPSF-30, DVB and St. In the IR spectra of MTPSF-30, several notable peaks can be observed. The peaks at 3063 cm −1 and 3025 cm −1 correspond to the C–H stretching vibrations on the benzene ring. Peaks at 1605 cm −1 , 1495 cm −1 and 1450 cm −1 are attributed to the backbone vibrations on the benzene ring. Furthermore, peaks at 880 cm −1 , 755 cm −1 and 703 cm −1 indicate C–H out-of-plane bending vibrations. The appearance of the 880 cm −1 and 755 cm −1 peaks suggests that interposition substitution predominates in the benzene ring substitution of MTPSF-30. Additionally, DVB exhibits an 845 cm −1 peak, indicating the presence of para -substitution on DVB. Fig. 4 FTIR spectra of MTPSF-30, DVB and St. MTPSF-30 possesses two characteristic peaks, namely, 2923 cm −1 and 2850 cm −1 , which arise from the stretching vibrations of hydrocarbon-saturated bonds. These peaks are absent in DVB and St. On the other hand, DVB and St exhibit distinctive peaks at 1634 cm −1 and 994 cm −1 , which are not present in MTPSF-30. The peak at 1634 cm −1 corresponds to the stretching vibration of the carbon–carbon double bond, while the peak at 994 cm −1 is attributed to the C–H out-of-plane bending vibration. The presence of the 994 cm −1 peak suggests that DVB and St are monosubstituted. Based on the above information, it can be inferred that DVB and St produce MTPSF-30 via a polymerization reaction. 3.3 Physical characteristics The densities, porosities, and internal phase proportions of MTPSF-10, MTPSF-20, MTPSF-30 and OSPs@Foam are listed in Table 3 . As the internal phase proportion increases, the density of the material drops while the porosity goes up. Notably, MTPSF-30, produced through multi-template high internal phase method, exhibits a lower density and higher porosity compared with OSPs@Foam. An ideal superhydrophobic porous material should possess lower density and higher porosity in the field of oil–water separation. These characteristics facilitate the material's transport and enhance the oily wastewater's adsorption. The physical characteristics of materials Samples Density (CI 95% a )/g cm −3 Porosity (CI 95% a )/% Internal phase proportion/% MTPSF-10 0.0406(3.17 × 10 −4 ) 93.4(0.431) 94.8 MTPSF-20 0.0210(4.55 × 10 −4 ) 95.4(0.679) 97.3 MTPSF-30 0.0144(4.91 × 10 −4 ) 97.4(0.652) 98.2 OSPs@Foam 0.0159(5.56 × 10 −4 ) 96.3(0.445) 98.2 a Confidence interval (CI 95%) was calculated based on multiple experiments on different samples. 3.4 Wettability \n Fig. 5 presents the water contact angles of MTPSF-30, MTPSF-20, MTPSF-10 and OSPs@Foam as 143°, 142°, 138° and 151°, respectively, showcasing their superhydrophobic–superoleophilic properties with water sliding angle of 6° and an oil contact angle of approximately 0°. Notably, MTPSF derived from improvements of OSPs@Foam exhibit a minor decline in water contact angle, likely due to the incorporation of OSPs, which produce the rough surface structure of foams. Subsequently, the surface roughness is compromised by acidizing, which reduces the water contact angle. Fig. 5 (a 1 –d 1 ) Water contact angle of MTPSF-10, MTPSF-20, MTPSF-30 and OSPs@Foam; (a 2 –c 2 ) ethanol contact angle of MTPSF-30; (a 3 –c 3 ) water sliding angle of MTPSF-30. 3.5 Oil–water separation In Fig. 6(a) and (b) , we conducted a test wherein ethanol treated with methyl yellow was dropped onto the surface of the material, resulting in the rapid orange-yellow dyeing of the surface, indicating the material's lipophilic properties. Conversely, when distilled water was dropped onto the material's surface, the formation of water droplets was observed. Only slight tilting of the material led to the rapid rolling off of the water droplets, demonstrating the material's hydrophobic characteristics. Fig. 6 (a and b) Schematic diagram of water and ethanol drops on the surface of MTPSF-30; (c–e) MTPSF-30 absorbed toluene on the water surface; (f–h) MTPSF-30 absorbed CTC under the water. The oil–water separation capability of MTPSF was assessed using two distinct models: the floating light oil model and the submerged heavy oil model. As demonstrated in Fig. 6(c)–(e) , MTPSF absorbs toluene stained with Sudan II on the water surface, clarifying the oily water body within two minutes. Similarly, Fig. 6(f)–(h) depicts the material's buoyancy overcoming by tweezers to make contact with the CTC stained with Sudan II under the water surface for adsorption. Upon contact with CTC, the absorbed CTC squeezes out the air, forming bubbles in the water body and completely removing the CTC in under a minute, leaving behind clear water. These oil–water separation experiments confirm that MTPSF possesses exceptional oil–water separation abilities, highlighting their potential for widespread applications. 3.6 Oil absorption capacity The oil absorption capacity (OAC) is a crucial characteristic to assess the ability of materials to absorb oil, and the OAC of different samples (MTPSF-10, MTPSF-20, MTPSF-30 and OSPs@Foam) for various oils is shown in Fig. 7 . The OAC of the samples ranged from 39.40 to 102.32 g g −1 , with MTPSF-30 exhibiting excellent oil absorption ability. MTPSF-30's OAC for CTC reached an impressive 102.32 g g −1 , which was on account of the large density of CTC. A comparison of MTPSF-10, MTPSF-20 and MTPSF-30 showed that the OAC of the material rose with higher water content in the internal phase, which provides more space for the oil to be absorbed. 27 On the other hand, a comparison of MTPSF-30 and OSPs@Foam revealed that constructing more pores led to a more powerful oil-absorbing capacity. Table 4 compares the OAC of MTPSF with other materials, and it is evident that MTPSF has superior oil-absorbing capability. Fig. 7 The oil absorption capacity of samples for various oils. Comparison among oil absorption capacities of other materials Reference Materials Oils OAC (g g −1 ) Zhu 28 Modified wood aerogel DCM 25.1 Satria 29 PFU-Fe-SA DCM 38.0 Bi 30 Spongy graphene DCM 86.0 Yang 31 PU sponge Petroleum ether 25.0 Wang 32 PDMS-SiO 2 @PU-Econea sponge Ethanol 42.0 Gharehasanloo 33 CS@PLA foam Trichloromethane 48.0 This work MTPSF DCM 87.6 Petroleum ether 39.4 Ethanol 46.2 Trichloromethane 96.7 3.7 Recovery efficiency The recovery efficiency of MTPSF was evaluated for different types and viscosities of oils, which is a crucial parameter for measuring the material's ability to recover oil. The results are illustrated in Fig. 8 , and the recovery efficiency ranges from 85.67% to 99.12%. Notably, the recovery efficiency of MTPSF for petroleum ether was an outstanding 99.12%, whereas for lubricating oil and corn oil, the values were only 85.67% and 86.76%, respectively. This discrepancy can be attributed to the oil's viscosity, as oils with higher viscosity are more challenging to desorb through centrifugation, leading to higher recovery costs. High recovery efficiency is an important factor in reducing recovery costs and increasing the practical application value of the material. Fig. 8 Recovery efficiency of MTPSF-30 for different viscosity oils. 3.8 Reusability The absorbed oil inside the MTPSF can be recovered by simple centrifugation and squeezing. The results of repeated adsorption–desorption cycles of MTPSF for ethanol and diesel by centrifugation are presented in Fig. 9(e) and (f) . In Fig. 9(e) , the initial oil absorption capacity of MTPSF-30 for ethanol is measured at 36.1 g g −1 . This value significantly differs from the OAC shown in Fig. 7 (46.2 g g −1 ). Such discrepancies can be attributed to two key factors. One is that the materials used in the two experiments were cut differently. The material used in Fig. 7 was cut into cylinders, while the material used in Fig. 9(e) was cut into rectangles. The second is that even though these materials were prepared by the same recipe and preparation method, there were still large fluctuations in the multiplicity of their absorption of ethanol. The material underwent ten cycles of the adsorption–desorption process, exhibiting its durability and resistance to structural damage. After ten cycles, the OAC of the material dropped by 13.0% and 13.6% for ethanol and diesel, respectively, resulting in residual rates of 2.36 g g −1 and 2.71 g g −1 for ethanol and diesel inside the material. The decrease in OAC can be attributed to the shrinkage of pores inside the material during centrifugal desorption. Additionally, diesel has a larger viscosity than ethanol, allowing its retention inside the material during centrifugal desorption, which accumulates gradually during cycling. However, the MTPSF-30's oil absorption capacity does not decrease dramatically and can desorb most oil absorbed inside the material. Compared to other studies, 34,35 MTPSF demonstrates excellent reusability and practicality in the field of oil–water separation, mainly due to their high number of cycles. Fig. 9 (a–d) Absorption–desorption cycles of MTPSF-30 by centrifugation; (e and f) reusability of MTPSF-30 for the absorption of ethanol and diesel within 10 cycles by centrifugation; (g–i) absorption–desorption of MTPSF-30 by squeezing. 3.9 Adsorption kinetics To eliminate the influence of oil density on the adsorption of MTPSF-30, we introduce the parameter K , which represents the progress of adsorption. Its expression is shown in eqn (6) , 6 where, Q e represents the saturation absorption capacity of MTPSF-30, and Q t represents the absorption capacity of MTPSF-30 at time t . We conducted adsorption experiments using the intermittent adsorption method and measured the adsorption kinetic curves of MTPSF-30 for various oils, as depicted in Fig. 10 . Over time, the adsorption curve of MTPSF-30 became flat, and the adsorption rate decreased sharply. The saturation adsorption time of MTPSF-30 for the oil under test ranged between 195 and 7775 seconds. This saturation adsorption time was observed to increase with the oil's viscosity. A comparison of the adsorption kinetic curves for MTPSF-30 and OSPs@Foam showed that MTPSF-30's shorter saturation time for ethanol adsorption could be attributed to its higher pore quantity, which would offer a stronger capillary force for oil adsorption. Fig. 10 (a and b) Adsorption kinetic curves of MTPSF-30 for various oils; (c) ethanol adsorption kinetic curve of MTPSF-30 and OSPs@Foam. When the material comes into contact with the oil, the initial oil molecules in contact with the material are adsorbed first. These molecules enter the interior of the material through capillary forces. However, oil molecules with higher viscosity, experiencing stronger intermolecular forces, cannot quickly penetrate the material's interior and tend to remain on the surface, reducing the effective contact area and significantly decreasing the adsorption rate. Additionally, as the oil molecules on the material's surface are adsorbed, molecules from other areas migrate to the surface, enabling subsequent adsorption. Higher oil viscosity leads to stronger intermolecular forces between the oil molecules, resulting in a slower migration rate and lower adsorption rate of the material for the oil. Moreover, oil with higher viscosity that enters the material demonstrates a relatively slower diffusion rate within the pores, which consequently results in a lower adsorption rate by the material for oils of higher viscosity. To simulate the adsorption process of MTPSF-30, we employed the pseudo-first-order kinetics, pseudo-second-order kinetics, Elovich, and Webber–Morris models. The simulation results, presented in Fig. 11 and ESI, † indicated that the adsorption process of MTPSF-30 aligns more closely with the second-order sorption kinetics and Webber–Morris models, with R 2 values exceeding 0.9477 for all oils. Webber–Morris model 36 presented in this study describes a process in which the adsorption rate is influenced by intraparticle or film diffusion. The model equation is as follows: 7 Q t = k i t 1/2 + C where, Q t represents the absorption capacity of the material at time t , k i is the intraparticle diffusion rate constant, and C is the intercept on the vertical axis. Fig. 11 (a 1 and a 2 ) The pseudo-first-order kinetics model of oil sorption on MTPSF-30; (b 1 and b 2 ) the pseudo-second-order kinetics model of oil sorption on MTPSF-30; (c 1 and c 2 ) the Elovich model of oil sorption on MTPSF-30; (d 1 and d 2 ) the Webber–Morris model of oil sorption on MTPSF-30. The fitted data demonstrated that the plotted straight line for the MTPSF does not pass through the origin, indicating that C is not 0. This observation implies that the absorption of MPTSF for oil is restricted by both the diffusion rate through the film and the internal diffusion rate. Since MTPSF has an impressive porosity and can be considered as an internally empty material. Oil molecules that penetrate the interior of the MPTSF-30 undergo lateral or longitudinal diffusion driven by capillary forces and intermolecular forces, eventually reaching the centre and top of the material. Furthermore, as MTPSF absorbs more viscous oil, the k i value decreases, thereby reducing the absorption rate of MPTSF. 3.10 Compressibility The compressibility of different samples (MTPSF-10, MTPSF-20, MTPSF-30 and OSPs@Foam) was evaluated through stress–strain curves, as shown in Fig. 12 . MTPSF-30 showed the best compressibility among all samples, with an ultimate strain of 55% of initial height. The ultimate strain of MTPSF-10, MTPSF-20, and OSPs@Foam was below that of MTPSF-30, at 7%, 15%, and 41%, respectively. In Fig. 13 , compression experiments on OSPs@Foam and MTPSF-30 were conducted using a weight, and the results revealed that OSPs@Foam were completely crushed. At the same time, MTPSF-30 not only did not get destroyed but also achieved a slight rebound after withstanding compression. The increase in the proportion of internal phase water and the application of the multiple-perforated method can improve the compressibility of the material. Increasing the proportion of internal phase water creates more spaces inside the material that can be compressed under external stress, delaying crushing. On the other hand, MTPSF-30 employs the multi-template high internal phase method to construct an extensive number of pores within the material. This feature enables it to absorb a significant amount of external stress, thereby effectively improving the material's compressibility. Fig. 12 The stress–strain curves of all samples. Fig. 13 The compression experiment of OSPs@Foam (a) and MTPSF-30 (b)."
} | 6,625 |
29030583 | PMC5640656 | pmc | 7,178 | {
"abstract": "Nacre, a composite made from biogenic aragonite and proteins, exhibits excellent strength and toughness. Here, we show that nacreous sections can exhibit complete brittle fracture along the tablet interfaces at the proportional limit under pure shear stresses of torsion. We quantitatively separate the initial tablet sliding primarily resisted by nanoscale aragonite pillars from the following sliding resisted by various microscale toughening mechanisms. We postulate that the ductility of nacre can be limited by eliminating tablet interactions during crack propagations. Our findings should help pursuing further insights of layered materials by using torsion.",
"introduction": "Introduction Nacre is a hierarchical material with extraordinary strength and toughness 1 – 12 . Composed of ~95% aragonite and ~5% protein, nacreous structures exhibit orders of magnitude higher toughness than their brittle aragonite constituents primarily due to nano- and microscale brick-and-mortar structures 1 , 4 , 13 – 17 . Different toughening mechanisms have been proposed to explain the mechanical behavior of nacre. At the nanoscale, the sliding of nacreous tables are resisted by nanopillars (i.e., mineral bridges) 18 – 21 , contacts of nanoasperities 20 , 22 – 24 and the unfolding of protein chains 25 . At the microscale, interactions of wavy tablets 26 , 27 , crack deflections and pulling out of tablets 28 provide essential resistance to the nacreous deformation. Despite prior studies, it is still difficult to quantify how various nanoscale toughening mechanisms contribute to the shear resistance in the initial sliding of nacreous tablets. It is also unclear how nacre exhibits brittle behavior under different stress conditions. In this work, we employ pure shear stresses of torsion to investigate the mechanical properties of nacre in red abalone, and to further elucidate the evolution of toughening mechanisms during the deformation of nacre. We focus on separating adjacent nacreous tablets using pure shear stresses of torsion 29 , 30 . We created composite dog-bone shaped specimens using the pure nacreous sections without any growth layers 21 , 31 – 33 from a red abalone shell. The hexagonal surfaces of aragonite tablets are perpendicular to the cylindrical axis (Fig. 1b ). For a shaft under torsion, pure shear stresses exist in every two-dimensional cross-sectional planes over the entire gauge section. This feature enables the nanoscale interfaces between nacreous tablets to be tested successfully. The shear stress and strain in torsion are described by 1 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$\\tau =\\frac{T\\cdot \\rho }{J}$$\\end{document} τ = T ⋅ ρ J \n 2 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$${s}_{max}={\\rm{\\Delta }}\\theta \\cdot r={\\gamma }_{max}\\cdot l$$\\end{document} s m a x = Δ θ ⋅ r = γ m a x ⋅ l Where τ is the shear stress on the horizontal plane of the gauge section, T is the torque, r is the gauge section radius, ρ is the distance between an arbitrary point on the cross section to the center varying from 0 to r , J = 0. 5πr \n 4 is the polar moment of inertia, γ \n max is the shear strain at the external edge, s \n max , Δ θ , and l are the sliding distance at the external edge, twisting angle, and height of two vertically adjacent tablets, respectively. Since nacre specimens are transversely isotropic, shear stresses increase linearly from zero in the center to the maximum in the external edge before the specimen fails. τ \n b represents the shear stress distribution before the specimen failure. Once the external aragonite tablets fail, large gradients of shear stresses occur radially due to the rapid decrease in the polar moment of inertia, which is a fourth-order function of the cylindrical radius. These gradients promote fast crack propagations radially until the specimen fails completely. τ \n a represents the shear stress distribution after the specimen failure. In this study, the failure of specimens is achieved when the interfacial shear strength is reached. The maximum tablet sliding distance upon failure is measured as the maximum rotational distance between two vertically adjacent brick layers at the external edge. The shear strain is the ratio between the sliding distance and the height of adjacent tablets (Fig. 1c ). In the experiments (Fig. 1a ), the axial load was precisely controlled (~0.22 N), and the catastrophic failure after peak load was detailed using data collected at a high sampling rate (200,000 samples per second). Figure 1 Schematics of the experimental system. ( a ) The dual system that includes an axial-torsional load cell (4.4 kN and 56.4 Nm) at the sampling rate of 50 samples per second and a torsional load cell (2.8 Nm) at the sampling rate of 200,000 samples per second. ( b ) Distributions of pure shear stresses of torsion before ( τ \n b ) and after ( τ \n a ) the failure of nacre specimens. Arrows 1 and 2 denote the sliding and crack directions, respectively. ( c ) A close view of two tablets at the external edge of the gauge section subjected to the maximum shear stress ( τ \n m ) before failure. This geometrical model was used in the finite element analysis.",
"discussion": "Results and Discussion Pure shear stresses generate complete brittle fracture of nacre A total of five successful composite dog-bone shaped nacre specimens were created due to the scarce of pure nacre sections in red abalone shells. For each specimen, the torque-rotation curve collected during quasi-static monotonic torsional tests exhibits an increase and a sharp drop (Fig. 2a ). The increase segments are almost linear since R 2 values of the linear fitting are above 0.99 for all specimens (Fig. 2c ). The shear strength is 41.5 ± 14.7 MPa at 2.0 ± 0.8% strain. The variation is due to the specimen locations in the red abalone shell. The post-peak curves are nearly vertical, showing that catastrophic failures happen in a short time (Fig. 2a and b ). In comparison, single-crystal aragonite minerals were shaped to the same size and orientation of nacreous specimens. The shear strength of aragonite is 14.5 ± 2.2 MPa at 1.4 ± 0.2% strain. Although the R 2 values of aragonite segments are close to unity, the relatively flat drop (Fig. 2b ) indicates that crack propagations are affected by the spiral fractural surfaces (Fig. 3d ). For the same geometry, cracks travel faster between the nacreous tablets than in aragonite minerals since the dropping periods of nacre specimens last shorter (5.0 ± 2.0 milliseconds) than those of aragonite specimens (20.0 ± 0.6 milliseconds). By discretizing the twisting angle of the gauge section to the rotational angle of two adjacent tablets, the measured sliding distance at the external edge in the initial stage is ~20 nm (20.0 ± 8.4 nm) that agrees with prior estimations 22 . Figure 2 Monotonic torsional tests for nacre and aragonite specimens. ( a ) Shear stress-stress curves. ( b ) Post-peak drop periods with respect to time. ( c ) R 2 values of the linear-fitting for the increase segments. ( d ) Sliding distances of adjacent nacreous tablets. \n Figure 3 Fractographic characterization. ( a ) A representative dog-bone shaped nacre specimen. ( b ) and ( c ) Optical and SEM images of the fractured nacre half, respectively. ( d ) A microCT image of a fractured aragonite half under torsion, showing a 45° helical fracture. ( e ) A closer view of broken edges and interlayer spiral transitions of nacreous specimens. ( f ) A detail look at inter- and transtablet breakage of nacreous tablets under pure shear stresses of torsion. The scale bars are 1 mm in ( a ), ( b ), ( c ), ( d ), 50 µm in ( e ), and 5 µm in ( f ). \n Fractographic characterization proves interfacial fracture in nacre We observed two flat surfaces over the entire 3 mm-diameter circular cross sections of nacreous specimens after failure (Fig. 3a,c ). Under the white light, the flat surface exhibits a combination of green (λ≈510 nm) and yellow (λ≈570 nm) colors, which is different from the iridescence of the inner surface of red abalone shells. The reason is that uniform horizontal tablet (~500-nm thick) layers underneath reinforce lights with specific wavelengths over the entire fracture surface. Elevated tablet sections appear on the left (Fig. 3b and c ). This is because pure shear stress condition is no longer maintained once cracks start to propagate after the failure of external tablets. As a result, cracks kink from one interface to another that eventually separate the specimen. Detail SEM image (Fig. 3e ) shows that large areas of tablets are exposed and are interlaced with broken edges that transit spirally from one layer to the next 20 . The broken edges (Fig. 3f ) include both the smooth intertablet delamination and brutal transtablet breakage. However, these sharp edges are different from the blunted edges of polished nacre specimens 34 , demonstrating that tablets and spiral connections are quickly removed during crack propagations. The microCT image (Fig. 3d ) clearly shows the different brittle fracture of aragonite specimens. Aragonite specimens exhibit a classical 45-degree helical fracture perpendicular to the principal tensile stress, while nacreous specimens fracture sharply through the interfaces between aragonite tablets. Mathematical modeling quantifies nanoscale toughening mechanisms To detail the nanoscale toughening mechanisms in the initial sliding stage, we created isotropic linear elastic finite element models to quantify the contribution of nanopillars, nanoasperities and protein chains to the shear resistance. A contour of an exposed tablet (Fig. 4a ) shows that stresses in nanopillars and nanoasperities (E = 100 GPa) are substantially higher than those of protein chains (E = 20 MPa) due to the significantly different elastic moduli. The lower and upper bounds of shear stress curves (Fig. 4b ) correspond to nanopillar densities of 1.4 and 5.6/µm 2 , respectively. The mean shear strength (~41.5 MPa) curve corresponds to a density of ~2.2/µm 2 . By decoupling the contribution of each mechanism with respect to various tablet moduli (E = 80 to100 GPa) and pressures (p = −35 to 35 kPa), Fig. 4c shows that nanopillars contribute to more than 95% of the shear resistance, while nanoasperities and protein chains contribute limitedly in the initial sliding stage. High shear stresses in the middle section of nanopillars (~1.0 GPa) enable their breakage at the end of the initial sliding stage. Compared to the shear strength of aragonite mineral specimens (~14.5 MPa), nacreous nanopillars exhibit much higher shear resistance, demonstrating the ‘smaller-is-stronger’ size effect down to the nanoscale. Figure 4 Mathematical modeling. ( a ) Mises stress contours of nanopillars, nanoasperities and organic matrices at the end of the initial sliding. ( b ) Predictions of the upper, lower and average shear stress lines in the torsional tests. ( c ) Contributions of nanopillars, nanoasperities and organic matrices to the shear resistance of nacre in the initial sliding stage. E1, E2 and E3 correspond to the elastic moduli of 100, 90 and 80 GPa, respectively. N, C and T denote zero, compressive (−35 kPa) and tensile (35 kPa) stresses on tablets, respectively. The scale bar is 500 nm in ( a ). \n The Achilles heel of nacre If we define the intact nanopillars, nanoasperities and protein chains before sliding as ‘mortar’, the ductile or brittle behavior of nacre is highly dependent on how bricks (tablets) perform after mortar sections fail. The mineral bridges, nanoasperities and protein chains are activated to resist sliding in the beginning. When mineral bridges break after tablets slide about twenty nanometers, vertical distances between tablets decrease (Fig. 5a and b ). In most stress conditions, contact areas between tablets increase gradually and substantially as cracks propagate completely or partially parallel to the sliding direction (Fig. 5c ). The toughening mechanisms then change to the nanoasperity contact and protein deformation in nanoscale, and to interactions of wavy tablets, tablet pulling-out and crack deflections in microscale 1 , 27 , 28 (Fig. 5d ). For example, when nacre is in tension along the tablets 18 , 24 , 35 , mortar sections fail first (the linear increment), and some tablets start to touch each other (the nonlinear increment). Then, microscale toughening mechanisms are triggered at various locations continuously (the extended stress plateau). Similar behavior exists in nacre under compression 36 , 37 , bending 24 , 45-degree shear 24 , 26 or direct shear 26 , 38 . However, when nacre is under torsion normal to aragonite layers, the sliding of tablets (tangential) is perpendicular to the crack propagation (radial), and the large stress gradients enable the specimen to fail quickly (Fig. 1b ). Thus, nacre exhibits completely interfacial brittle fracture since there is little chance for brick interactions after the mortar sections fail (Fig. 5e ). The Achilles heel of nacre is to avoid triggering microscale toughening mechanisms that can induce the ductile behavior. The discretization nature of torsional loads enable us to study nano- and microscale material behavior using relatively large specimens. Torsion can be particularly advantageous to study interlayer/interfacial mechanical behavior of layered materials. Figure 5 Schematics of the evolution of toughening mechanisms in the nacre deformation. ( a ) The beginning of tablet sliding. ( b ) Initial sliding up to the breakage of nanopillars. ( c ) Decrease in spacing induces more tablet contact. ( d ) Tablet geometries (waviness, pulling out, etc.) contribute to the shear resistance in the microscale toughening stage. Diagram ( e ) shows the direction of the crack propagation is perpendicular to tablet sliding under pure shear stresses of torsion, in which ( c ) and ( d ) are not fully triggered. \n In summary, by applying pure shear stresses of torsion, we exhibit the linear response in the initial sliding stage of nacreous tablets. Mathematical modeling shows that nanopillars contribute dominantly to the shear resistance, while nanoasperity contact and protein chains contribute limitedly in this stage. Complete brittle fracture observed between tablet interfaces in torsion is convincing proof that microscale toughening mechanisms are not triggered to promote ductile behavior of nacreous structures. These findings open an exciting perspective into studying mechanical properties of natural and artificial layered materials using pure shear stresses of torsion. Future effort could also be extended to study the interactions between nanoparticles that build the nacreous tablets, such as rotations, orders, protein strengthening, and deformation twinning."
} | 3,815 |
35157069 | PMC8902694 | pmc | 7,179 | {
"abstract": "Abstract Nutrient pollution is linked to coral disease susceptibility and severity, but the mechanism behind this effect remains underexplored. A recently identified bacterial species, ‘ Ca . Aquarickettsia rohweri,’ is hypothesized to parasitize the Caribbean staghorn coral, Acropora cervicornis , leading to reduced coral growth and increased disease susceptibility. Aquarickettsia rohweri is hypothesized to assimilate host metabolites and ATP and was previously demonstrated to be highly nutrient-responsive. As nutrient enrichment is a pervasive issue in the Caribbean, this study examined the effects of common nutrient pollutants (nitrate, ammonium, and phosphate) on a disease-susceptible genotype of A. cervicornis . Microbial diversity was found to decline over the course of the experiment in phosphate-, nitrate-, and combined-treated samples, and quantitative PCR indicated that Aquarickettsia abundance increased significantly across all treatments. Only treatments amended with phosphate, however, exhibited a significant shift in Aquarickettsia abundance relative to other taxa. Furthermore, corals exposed to phosphate had significantly lower linear extension than untreated or nitrate-treated corals after 3 weeks of nutrient exposure. Together these data suggest that while experimental tank conditions, with an elevated nutrient regime associated with coastal waters, increased total bacterial abundance, only the addition of phosphate significantly altered the ratios of Aquarickettsia compared to other members of the microbiome.",
"conclusion": "Conclusions Nutrient enrichment significantly effects coral microbiome dysbiosis and increases disease susceptibility, though the effects of individual nutrient constituents and shifts in N:P ratios remain under-characterized. Previous work has demonstrated a positive response of the parasitic genus Aquarickettsia to a combined nutrient treatment with corresponding negative effects on coral growth rates (Shaver et al. 2017 ). Results of the present study, however, suggest that Aquarickettsia responds most significantly to phosphate enrichment. Microbial diversity decreased over the course of nutrient enrichment in phosphate- and combined-treated samples, reflective of increased dominance of Aquarickettsia , and differential abundance of Aquarickettsia was only significant in these treatments. Fragment linear extension after 6 weeks of exposure to these treatments was also significantly lower than in other treatments. Alpha diversity decreased in nitrate-treated samples as a result of a significant decline in minor taxa. While results from quantitative PCR showed an increase in Aquarickettsia across all samples throughout the course of experimentation, significant shifts in individual taxa were not observed in ammonium-treated and untreated samples. This suggests that an increase in total bacterial abundance occurred in all samples, but only amendment with phosphate was sufficient to alter dominance of the predicted parasite Aquarickettsia over other bacterial species.",
"introduction": "Introduction The scleractinian coral holobiont, composed of diverse bacterial, archaeal, viral, and eukaryotic microorganisms, directly and indirectly influences the health, nutrition, and development of the coral organism (Rosenberg et al. 2007 , Krediet et al. 2013 , Bourne et al. 2016 , Webster and Reusch 2017 ). These taxa perform numerous services for their host including nitrogen fixation, sulfur cycling, and protection against pathogens (Bourne et al. 2016 , Glasl et al. 2016 ). The coral holobiont plays an essential role in coral responses to changing environments and likely mediates host resilience to stress, as environmentally induced changes in microbiome structure are hypothesized to facilitate host adaptation (Bourne et al. 2016 , Webster and Reusch 2017 ). While microbiome responses to short-term stressors (such as thermal stress) may be reversed with the removal of the stressor (Bourne et al. 2008 , Ziegler et al. 2019 , Maher et al. 2020 ), cumulative and long-term stressors (such as nutrient enrichment) can shift the coral microbiome from mutualistic to pathogenic (Bourne et al. 2016 ; McDevitt-Irwin et al. 2017 ). Different coral species may have different strategies to prevent this shift: while the high microbial diversity harbored by some corals may buffer some negative effects of environmental stress by providing functional redundancy and the potential for microbiome restructuring, other corals maintain a core microbial repertoire of beneficial species and may possess higher capacity for resisting pathogens (Ziegler et al. 2019 ). While the well-documented decline in coral diversity and coverage worldwide (Huang and Roy 2015 , Hughes et al. 2017 ) can primarily be attributed to warming ocean temperatures and coral disease, these stressors are exacerbated by localized and manageable stressors including nutrient pollution (Donovan et al. 2021 ). These stressors interact to cause changes in coral physiology and microbiome composition that contribute to low stress resilience and high mortality (Zaneveld et al. 2016 ; Vega Thurber et al. 2012 , 2014 , Pawlik et al. 2016 , Shaver et al. 2017 ). Corals thrive in oligotrophic tropical water due to their closely managed exchange of nutrients with the endosymbiotic dinoflagellate Symbiodiniaceae, a relationship that depends on low available nitrogen to stimulate highly efficient phosphorus cycling (Wooldridge 2010 , Shantz and Burkepile 2014 ). This symbiosis is disrupted by external N and P inputs from human activities, which are so significant that anthropogenic nutrient input has surpassed natural nutrient sources over the past century (Bennett et al. 2001 , Shantz et al. 2016 ). Field surveys and manipulative experiments alike have suggested that the severity, rapidity, and frequency of coral diseases is also related to local nutrient concentrations (Bruno et al. 2003 ; Voss and Richardson 2006 , Vega Thurber et al. 2014 ). Further, Vega Thurber et al. ( 2014 ) demonstrated that nutrient enrichment increased both the prevalence and severity of disease in enriched corals as well as the prevalence of bleaching, and that after cessation of nutrient exposure, bleaching and disease prevalence returned to baseline levels. Nutrient enrichment induces the overgrowth of macroalgae and turf algae (De'ath and Fabricius 2010 , Lapointe and Bedford 2011 ) with demonstrated negative effects on coral health (Pratte et al. 2018 ; Vega Thurber et al. 2012 ). We previously showed that a common bacterial symbiont of Acropora cervicornis , ‘Ca. Aquarickettsia rohweri ’, proliferates in response to nutrient-enriched conditions and was strongly associated with reduced coral growth rates (Shaver et al. 2017 ), and that increased abundances of the order Rickettsiales were associated with increased disease prevalence and tissue loss in other coral species (Table S1 (Supporting Information) of Zaneveld et al. 2016 ). These putative parasites are hypothesized to influence Acroporid disease susceptibility through the overconsumption of host and symbiont nutritional and energy resources (Klinges et al. 2019 ). The relationship between Aquarickettsia and Caribbean acroporids appears to be long-established, as Rickettsiales-like organisms were found in all histological samples of these coral species since 1975 (Peters et al. 1983 , Miller et al. 2014 ). Despite this, Aquarickettsia does not appear to coevolve with its coral host, but rather varies phylogenetically by geographic region, suggesting that this species is highly environmentally responsive (Baker et al. 2021 ). The primary nutrient inputs to oligotrophic reef systems, nitrogen (as nitrate and ammonium) and phosphorus, each have distinct effects on coral physiology. Importantly, anthropogenic nutrification of these systems often leads to not only an increase in these constituents but also an alteration of the ratio between their concentrations (Brodie et al. 2011 , Wiedenmann et al. 2013 ). As nutrient enrichment has become more pervasive in reef environments and often co-occurs with other stressors, disentangling the effects of individual stressors in situ is complex. We conducted a manipulative 6-week nutrient enrichment experiment in aquaria, allowing us to examine impacts of individual chemical constituents of nutrient pollution, specifically ammonium, phosphate, and nitrate in isolation and in concert, on the coral microbiome and coral health. To measure changes in key members of the A. cervicornis microbiome we used four different abundance metrics: relative abundance (% of total microbial community) from 16S amplicon data, quantitative PCR of a gene specific to the dominant taxon Aquarickettsia , analyses of differential abundance (shifts in abundance of individual taxa compared to other taxa) using ANCOM-II, and beta-binomial regressions using the R package corncob to model differential abundance of individual ASVs. We found that while phosphate enrichment stimulated an increase in the coral-associated parasite Aquarickettsia relative to other taxa, all forms of nutrient enrichment led to an increase in total bacterial abundance such that measured absolute abundance of Aquarickettsia increased significantly but these changes were found to be insignificant when examined using differential abundance metrics.",
"discussion": "Discussion \n Aquarickettsia \n rohweri dominates the microbiome of disease-susceptible A. cervicornis genotypes Numerous studies have demonstrated the surprising dominance of the parasitic genus Aquarickettsia in microbiomes of A. cervicornis , and in particular, of disease-susceptible genets of this species (Shaver et al. 2017 , Rosales et al. 2019 , Klinges et al. 2020 , Baker et al. 2021 , Aguirre et al. 2022 , Miller et al. 2020 , Casas et al. 2004 , Gignoux-Wolfsohn et al. 2020 ). Results from this study were consistent with previous work, as populations of Aquarickettsia were found to comprise 74.98% of the microbiome in samples of A. cervicornis genotype 50, a disease-susceptible genotype (Muller et al. 2018 ). We previously showed that disease-resistant genotypes of A. cervicornis , in contrast to disease-susceptible genotypes, are characterized by a diverse, even microbiome with no single member exceeding 11% relative abundance (Klinges et al. 2020 ). Several studies suggest that high microbial diversity provides the host with greater defense resources to combat microbial invaders (Bourne et al. 2016 , West et al. 2019 ) and that exposure to many different bacterial taxa increases host plasticity and ability to respond to changing environmental conditions (West et al. 2019 ; Zilber-Rosenberg and Rosenberg 2008 ). Voolstra and Ziegler ( 2020 ) argue that the microbiome flexibility and diversity in many Acropora species contributes to their ability to acclimate and adapt to environmental stressors, while leaving them more vulnerable to disease. They theorized that peak ecological capacity could be reached if genera with high physiological plasticity reach higher microbiome flexibility, by means of probiotics or genetic engineering. In contrast, if a coral known to lack high physiological plasticity were to lose microbiome flexibility, adaptability to stressors would be reduced even further. The consistent dominance of Aquarickettsia in all recently sequenced samples of A. cervicornis is concerning, as some studies have shown that the overabundance of a single taxon may result in the destabilization of the remaining bacterial community through effects on host immunity (reviewed in Rooks and Garrett 2016 , and Hooper et al. 2012 ), further reducing the potential for adaptability in these corals. Impacts of nutrient enrichment on coral physiology and microbiome structure Nutrient enrichment alters the coral microbiome primarily through increases in opportunistic species that may contribute to disease (Thompson et al. 2015 , Zaneveld et al. 2016 , Shaver et al. 2017 , Wang et al. 2018 ). While nutrient enrichment alone does not often induce mortality, the interaction of enrichment with other stressors such as thermal stress and loss of herbivorous fish species has been demonstrated to prolong bleaching and increase coral mortality (Zaneveld et al. 2016 , Wang et al. 2018 ). Anthropogenic nutrification of oligotrophic reefs often leads to not only an increase in the primary constituents of terrestrial runoff (ammonium, nitrate, and phosphate) but also an alteration of the ratio between their concentrations (Brodie et al. 2011 , Wiedenmann et al. 2013 ). While anthropogenic pollution tends to deliver more nitrate, ammonium is primarily derived from fish excretion (Allgeier et al. 2017 ). As such, we investigated the impacts of increased levels of each constituent on microbial community composition and coral health. While nitrogen enrichment leads to the proliferation of the coral algal symbiont Symbiodiniaceae (Muscatine et al. 1989 , Cunning and Baker 2013 ) phosphate alone does not affect symbiont density. We nonetheless found that symbiont density (as measured visually by changes in coral color) increased across all treatments in this experiment. Nitrate enrichment has been hypothesized to affect the translocation of nutrients between Symbiodiniaceae and the coral host (Shantz and Burkepile 2014 , Shantz et al. 2016 ), and it has been suggested that the host controls symbiont density through nitrogen and phosphorus limitation to maximize carbon return from this symbiosis (Wooldridge 2010 ). Further, overabundance of dissolved inorganic nitrogen leads to phosphate limitation in corals as a result of algal proliferation, causing increased susceptibility to light- and temperature-induced bleaching (Wiedenmann et al. 2013 , Rosset et al. 2017 ). Nitrate has been demonstrated to decrease coral growth rates, while ammonium did not (Shantz and Burkepile 2014 ). Excess phosphate, in contrast to nitrate, has minimal impact on coral physiology and may increase stress tolerance (Wiedenmann et al. 2013 , Shantz and Burkepile 2014 ). In fact, during thermal stress, phosphate uptake is increased in order to maintain symbiont density and carbon translocation (Ezzat et al. 2016 ). Ammonium, which is often naturally derived in reef systems from fish excretion, may enhance coral growth (reviewed in Shantz and Burkepile 2014 ). In light of these data, we hypothesized that nitrate enrichment would lead to the lowest coral growth rates over the course of enrichment and would lead to the greatest impact on microbiome structure. While nitrate treatment led to a trend of decreased coral growth rates, only exposure to the combination treatment (including nitrate, phosphate, and ammonium) significantly reduced coral growth rates compared to untreated samples. The addition of nitrate led to the significant decrease in minor taxa, but not to significant changes in Aquarickettsia . Phosphate treatment was the driving nutrient constituent in shifts of Aquarickettsia , with corals exposed to a combined treatment containing phosphate also exhibiting shifts in this taxon (Fig. 4C and D ). Treatment with ammonium did not significantly impact any individual microbial taxon (Fig. 4A ), nor did it significantly impact growth rates, consistent with the minimal effects observed from frequent, low-level introduction of ammonium to reef environments in the form of fish excretion (Shantz and Burkepile 2014 , Rice et al. 2019 ). Differences in alpha and beta diversity and in relative and absolute abundance of Aquarickettsia between 3 and 6 weeks of nutrient exposure were largely insignificant across all conditions. Importantly, this suggests that 3 weeks of exposure to tank conditions or nutrient treatment was sufficient to shift microbiomes to a new stable state. Phosphate enrichment shifted microbial populations further towards dysbiosis Phosphate and combined nutrient treatments led to a decrease in community dispersion by only 3 weeks of exposure (Figure S6, Supporting Information), and removal of Aquarickettsia from the dataset erased this effect (Figure S7, Supporting Information), suggesting that significant increases in Aquarickettsia were the driver of these shifts in community composition. While the addition of environmental stressors has been found to increase community dispersion in other studies (Zaneveld et al. 2017 ; Maher et al. 2019 ; Klinges et al. 2020 ), we found that dispersion generally decreased over the course of nutrient exposure. Zaneveld et al. ( 2017 ) posited that increases in dispersion resulting from stress are suggestive of microbiome instability resulting from stochastic shifts differing by individual. We argue that the trends observed in this study represent a takeover by an opportunistic species that excludes beneficial members from the microbiome, resulting in a different form of microbiome instability. Though dispersion decreased overall with enrichment, we found the magnitude of observed increase in the dominant constituent of A. cervicornis microbiomes differed by nutrient constituent. Differential abundance analyses revealed that abundance of Aquarickettsia was significantly altered in phosphate and combined treatments. Corals exposed to combined treatment demonstrated decreased growth rates over the course of the experiment, consistent with the hypothesized parasitism of essential host metabolites by Aquarickettsia . As decreases in growth rate as well as microbial community shifts were largely insignificant for ammonium and nitrate treatments, the majority of the response to combined nutrient treatment can be attributed to the phosphate in this treatment. This microbial response to phosphate and combination nutrient enrichment is consistent with our previous predictions of Aquarickettsia life strategy, namely, that Aquarickettsia is a parasite, dependent on nutritional supplementation from the algal symbiont (Klinges et al. 2019 ), and responds positively to nutrient enrichment (Shaver et al. 2017 ). As algal symbiont density was visually observed to increase over the course of experimentation, it is likely that Aquarickettsia utilized surplus amino acids and sugars produced by elevated symbiont populations. We previously identified annotations for a complete PhoR–PhoB two component system in the genome of A. rohweri (Klinges et al. 2019 ). This system, notably absent in the genomes of closely related species in the order Rickettsiales, allows for sensing of extracellular inorganic phosphate concentrations. Paired with evidence that phosphate stimulates abundance of Aquarickettsia , it is likely that the PhoR–PhoB system plays a role in cellular proliferation or upregulation of virulence factors in this species. The Pho regulon has been found to modulate the expression of genes involved in survival response ( via production of poly P, ppGpp, and RpoS), virulence, motility, and biofilm formation through quorum sensing in various strains of pathogenic bacteria (Lamarche et al. 2008 ). The transcriptional-response regulator PhoB translates signals of phosphate depletion or enrichment into gene activation or repression. Recently, the histidine kinase PhoR was previously found to be involved in the upregulation of swarming, flagellar motility, and T3SS expression in the pathogen Vibrio parahaemolyticus (Zhang et al. 2020 ). Control of gene activity by this regulon has been suggested to play a role in life strategy in bacteria experiencing large shifts in phosphate levels, leading to a switch between a motile swimming phase and a biofilm-forming or intracellular phase. Aquarickettsia encodes a nearly complete flagellar assembly and no evidence has been found of vertical transmission of this species, suggesting it may have a brief free-living life stage (Baker et al. 2021 ). It is, therefore, possible that high phosphate levels act as a cue for upregulation of virulence and host cell infiltration, as well as increased cell replication due to an abundance of phosphate for the production of nucleic acids. Multiple measures of Aquarickettsia abundance elucidate its dynamics in response to nutrient enrichment Regardless of nutrient constituent or concentration, the absolute abundance of the putative bacterial parasite Aquarickettsia increased with time. This trend also included untreated samples, confounding an association of Aquarickettsia populations with nutrient type. Although examination of Aquarickettsia relative abundance over the course of experimentation indicated that phosphate treatment induced the greatest shift in Aquarickettsia abundance, analyses of differential abundance of this taxon were also performed due to known issues with relative abundance data. Relative abundance metrics are easily confounded by shifts in one taxonomic group—even if the absolute abundance of most taxa remains unchanged, increases in one taxon will deflate the relative abundance of other taxa (Mandal et al. 2015 , Gloor et al. 2017 ). Differential abundance analysis confirmed that shifts in Aquarickettsia relative to other taxa were not significant in any nutrient treatment besides phosphate or combined treatments. Differential abundance tools such as ANCOM-II and corncob may reflect more accurately shifts in individual taxa, as they account for the inherent issues with compositional data, including excess zeroes, though each of these tools utilizes a slightly different method to identify differentially abundant taxa. While total bacterial abundance was not quantified in this study, the inconsistencies between differential abundance data and absolute abundance data suggest that an increase in total bacterial abundance occurred throughout the course of the experiment, such that absolute abundance of Aquarickettsia increased in untreated, nitrate, and ammonium-treated samples, but proportions of this taxon compared to other species did not change significantly. This increase in total bacterial abundance may be attributed to the elevated nutrient levels across all treatments, as the nearshore water used in this study was found to have higher levels of nitrate than Mote's nursery near Looe Key Reef (Lapointe et al. 2019 ) from which coral fragments were sourced (Figure S1, Supporting Information), though phosphate and ammonium levels were comparable. Thus, while nitrate levels were elevated ∼4-fold significantly compared to Looe Key reef conditions in all samples, phosphate levels were only enriched compared to reef conditions in phosphate-treated or combination-treated samples. It is therefore possible that we did not capture the true response of these corals to nitrate enrichment, as they had been allowed to acclimate to high-nutrient conditions prior to experimental manipulation. While Looe Key Reef is not truly oligotrophic (Lapointe et al. 2019 ), it is possible that the transplant of these coral individuals from a nutrient-limited system to a nutrient-rich system stimulated a bacterial bloom in all samples. Nutrient enrichment has been demonstrated to lead to increased algal growth, leading to the increased production of algal exudates. These exudates are high in dissolved sugars, which rapidly stimulate bacterial growth (Nelson et al. 2013 , Bourne et al. 2016 ). The observed bacterial bloom appeared to affect all bacterial taxa indiscriminately, as differential abundance analysis found no taxa were significantly enriched or depleted in untreated samples after 6 weeks of exposure to tank conditions in comparison to T0. While previous work in Mediterranean coastal oligotrophic waters found that bacterial biomass responded most dramatically to increases in total nutrient concentrations (Sipura et al. 2005 , Rahav et al. 2018 ), studies performed in the Florida Bay indicated that this system was highly phosphate-limited due to excess nitrogen and carbon inputs (Fourqurean et al. 1993 , Cotner et al. 2000 ). It is, thus surprising that overall bacterial abundance appeared to increase in response to the nitrate-enriched, but not phosphate-enriched, conditions in unamended aquarium water, though no single taxon shifted significantly in respect to other taxa. Conclusions Nutrient enrichment significantly effects coral microbiome dysbiosis and increases disease susceptibility, though the effects of individual nutrient constituents and shifts in N:P ratios remain under-characterized. Previous work has demonstrated a positive response of the parasitic genus Aquarickettsia to a combined nutrient treatment with corresponding negative effects on coral growth rates (Shaver et al. 2017 ). Results of the present study, however, suggest that Aquarickettsia responds most significantly to phosphate enrichment. Microbial diversity decreased over the course of nutrient enrichment in phosphate- and combined-treated samples, reflective of increased dominance of Aquarickettsia , and differential abundance of Aquarickettsia was only significant in these treatments. Fragment linear extension after 6 weeks of exposure to these treatments was also significantly lower than in other treatments. Alpha diversity decreased in nitrate-treated samples as a result of a significant decline in minor taxa. While results from quantitative PCR showed an increase in Aquarickettsia across all samples throughout the course of experimentation, significant shifts in individual taxa were not observed in ammonium-treated and untreated samples. This suggests that an increase in total bacterial abundance occurred in all samples, but only amendment with phosphate was sufficient to alter dominance of the predicted parasite Aquarickettsia over other bacterial species."
} | 6,472 |
38626194 | PMC11020763 | pmc | 7,181 | {
"abstract": "The move from a free-living environment to a long-term residence inside a host eukaryotic cell has profound effects on bacterial function. While endosymbioses are found in many eukaryotes, from protists to plants to animals, the bacteria that form these host-beneficial relationships are even more diverse. Endosymbiont genomes can become radically smaller than their free-living relatives, and their few remaining genes show extreme compositional biases. The details of how these reduced and divergent gene sets work, and how they interact with their host cell, remain mysterious. This Unsolved Mystery reviews how genome reduction alters endosymbiont biology and highlights a “tipping point” where the loss of the ability to build a cell envelope coincides with a marked erosion of translation-related genes.",
"conclusion": "Conclusions Here, we have highlighted several unsolved mysteries related to how endosymbiont-host systems function. These are difficult problems to solve. It is one thing to purify a protein complex from Escherichia coli , or an organelle from Saccharomyces cerevisiae , it is quite another to do these same experiments in a tiny insect, without genetic control, where the endosymbiont of interest exists only in a small, specialized tissue. The issue, of course, is that it was only by the study of strange insects and difficult-to-isolate protists in the first place that allowed these organelle-adjacent endosymbionts to be discovered. So we have no choice, and it is not all bad news. If genomics was the gateway into tipping-point endosymbiont biology, advances in chemical biology (such as click-chemistry), structural biology (such as cryo-EM), and genetics (such as CRISPR and RNAi) are the tools that will make the cell biological and biochemical study of nonmodel endosymbiont-host systems accessible, if not easy. Creative uses of these and other technologies over the next several years will allow at least partial answers into how endosymbionts work with so few genes.",
"introduction": "Introduction Host-beneficial endosymbiosis, where an organism stably maintains an unrelated organism inside some or all of its cells, is a complex process. The host must calm its immune system sufficiently to not kill its endosymbiont or overreact to the sustained presence of a foreign cell. The host must extract what it needs—often nutrition, energy, or protective molecules—from the endosymbiont without taking so much that it imperils the health of its smaller partner. How the endosymbiont experiences host cell restriction is difficult to assess, but genome reduction, gene loss, and rapid rates of sequence evolution are near-universal outcomes of the process. The endosymbionts with the smallest genomes have lost so many genes that they lack the ability to do much at all on their own and start to resemble mitochondria and chloroplasts more than typical bacteria. How these hosts and endosymbionts integrate their biochemical and cell biological processes is largely unknown. In this Unsolved Mystery, we briefly outline which bacteria become endosymbionts, which genes are retained in endosymbionts, and what happens to the genome and protein composition of endosymbionts. We focus on endosymbionts with genomes less than 200 kilobases (kb) in length because these organisms seem to have crossed a tipping point where gene loss has been so extensive that it is unclear how fundamental bacterial processes are carried out. We then highlight 4 unsolved mysteries related to the function of endosymbionts with highly reduced genomes: how to build a cell boundary with no genes to make it; how to transport molecules across the cell envelope with no genes to make transporters; how to make proteins when missing key translation-related genes; and how proteins function with extreme amino acid compositional biases."
} | 952 |
25897488 | PMC4410645 | pmc | 7,182 | {
"abstract": "Horizontal gene transfer (HGT) plays a key role in bacterial evolution, especially with respect to antibiotic resistance. Fitness costs associated with mobile genetic elements (MGEs) are thought to constrain HGT, but our understanding of these costs remains fragmentary, making it difficult to predict the success of HGT events. Here we use the interaction between P. aeruginosa and a costly plasmid (pNUK73) to investigate the molecular basis of the cost of HGT. Using RNA-Seq, we show that the acquisition of pNUK73 results in a profound alteration of the transcriptional profile of chromosomal genes. Mutations that inactivate two genes encoded on chromosomally integrated MGEs recover these fitness costs and transcriptional changes by decreasing the expression of the pNUK73 replication gene. Our study demonstrates that interactions between MGEs can compromise bacterial fitness via altered gene expression, and we argue that conflicts between mobile elements impose a general constraint on evolution by HGT.",
"discussion": "Discussion In this study, we used P. aeruginosa PAO1 and the small plasmid pNUK73 as a model system to investigate the origin of the costs of HGT and the nature of its subsequent compensatory adaptation. We found that the cost of pNUK73 was generated by the expression of the plasmid replication protein gene rep , which in turn produced massive changes in the expression of the PAO1 genome (749 genes), including the activation of the SOS response and the expression of genes that are associated with stalled chromosomal replication. Remarkably, the alteration produced by pNUK73 in the expression of PAO1 genome is larger than the effect of general transcriptional regulators such as LasR or AmpR 33 34 . The high-level expression of rep depends on the presence of a putative helicase (PA1372) and two putative protein kinases (PA4673.15-16), which showed strong signatures of HGT and no apparent biological role in PAO1. The inactivation of the helicase or one of the kinases genes completely restored the changes in gene expression and fitness associated with pNUK73 acquisition. To the best of our knowledge, this is one of the few studies dissecting the mechanistic basis of cost and adaptation in a bacterium/mobile element interaction at a genetic and transcriptomic level 35 . One of the most important limitations of this study is that although we clearly showed the link between the mutations in the helicase and kinase and the reduction of rep expression, we could not elucidate the specific interactions among these genes driving the high level expression of rep . Our results suggest that the helicase and kinase interact with each other, since mutations in both helicase and kinase lead to the underexpression of PA3194 and PA3887 genes. In addition, helicases are known to interact with plasmid replication proteins 36 37 . Taken together, these lines of evidence suggest that the kinases may be responsible for the activation of the helicase through phosphorylation, which could in turn interact with pNUK73 replication protein, leading to the derepression or activation of rep expression. Further experimental work will be necessary to elucidate the nature of these interactions. Our results clearly show that the interactions between recently acquired genes are responsible for the cost of HGT in this model. Interactions between MGEs have previously been shown to influence bacterial fitness; for example, epistatic interactions between co-occurring plasmids in the same bacterial strain have been shown to both buffer and aggravate the fitness cost of plasmid carriage 38 39 . Unfortunately, the mechanistic basis of epistasis has not been elucidated in these studies. Interactions between MGEs that influence fitness should play a key role in the persistence of MGEs in bacterial populations; positive interactions that ameliorate the cost of MGE carriage should increase the stability of MGEs in bacterial populations, while negative interactions that exacerbate the costs of MGE carriage should drive the loss of mobile elements. Why do mobile elements interact with each other? One possibility is that MGEs have specifically evolved systems to cooperate or compete with other MGEs (for examples, see refs 40 , 41 ), and the interactions between them could translate into a fitness alteration for the host. This explanation requires an evolutionary history between the MGEs. Alternatively, it is possible that interactions between MGEs arise as a spurious accident. By definition, HGT brings together genes that have different evolutionary histories, and there is no a priori reason to expect that these genes should interact with each other in a mutually beneficial way. Even if the different acquired genes have evolved together and could interact with each other, it is likely that these interactions would produce different results in a new species compared to in the original host due to the different genetic circuitry. Therefore, under the accident hypothesis the recently acquired genes are likely to produce negative effects on the host. We argue that our model system provides an example for the accident hypothesis. We found that although the helicase and kinases seemed to interact with each other, they produced a negligible impact on the expression profile and fitness of PAO1, essentially acting as pure genetic parasites. In the presence of pNUK73, however, these two genes interact to induce high-level expression of the pNUK73 Rep protein, triggering cytotoxic effects that lead to tragic consequences for all the parties involved in this interaction. In fact, it is only in combination that the three acquired genes produced a big cost in PAO1, while any pairwise combinations produced no extra cost 20 . We speculate that these types of costly accidental interactions between MGE are probably frequent in bacteria and may restrict evolution via HGT. The precise molecular bases of these interactions are difficult to predict due to the lack of experimental evidence in current literature. However, it is likely that the fitness cost derived from these accidental interactions is generally based on the destabilization of fine-tuned host cellular networks such as DNA replication, as is the case in our model system. In this work, we have elucidated the mechanisms implicated in the cost of the small plasmid pNUK73 in P. aeruginosa PAO1, as well as the mechanistic basis for the alleviation of this cost via compensatory evolution. We found that the detrimental effects produced by this plasmid arose from interactions between recently acquired genetic elements in PAO1. Even though the results presented here are only applicable to this model system, several lines of evidence support the idea that the interactions between recently acquired genetic elements may play a central role in the cost of HGT. Future experimental and bioinformatic work will be necessary to provide further support to this hypothesis."
} | 1,746 |
24903884 | null | s2 | 7,183 | {
"abstract": "DNA assembler is an efficient synthetic biology method for constructing and manipulating biochemical pathways. The rapidly increasing number of sequenced genomes provides a rich source for discovery of gene clusters involved in synthesizing new natural products. However, both discovery and economical production are hampered by our limited knowledge in manipulating most organisms and the corresponding pathways. By taking advantage of yeast in vivo homologous recombination, DNA assembler synthesizes an entire expression vector containing the target biosynthetic pathway and the genetic elements needed for DNA maintenance and replication. Here we use the spectinabilin clusters originated from two hosts as examples to illustrate the guidelines of using DNA assembler for cluster characterization and silent cluster activation. Such strategies offer unprecedented versatility in cluster manipulation, bypass the traditional laborious strategies to elicit pathway expression, and provide a new platform for de novo cluster assembly and genome mining for discovering new natural products."
} | 272 |
34081101 | PMC8358223 | pmc | 7,185 | {
"abstract": "Abstract Annotated genome sequences provide valuable insight into the functional capabilities of members of microbial communities. Nevertheless, most studies on the microbiome in animal guts use metagenomic data, hampering the assignment of genes to specific microbial taxa. Here, we make use of the readily culturable bacterial communities in the gut of the fruit fly Drosophila melanogaster to obtain draft genome sequences for 96 isolates from wild flies. These include 81 new de novo assembled genomes, assigned to three orders (Enterobacterales, Lactobacillales, and Rhodospirillales) with 80% of strains identified to species level using average nucleotide identity and phylogenomic reconstruction. Based on annotations by the RAST pipeline, among-isolate variation in metabolic function partitioned strongly by bacterial order, particularly by amino acid metabolism (Rhodospirillales), fermentation, and nucleotide metabolism (Lactobacillales) and arginine, urea, and polyamine metabolism (Enterobacterales). Seven bacterial species, comprising 2–3 species in each order, were well-represented among the isolates and included ≥5 strains, permitting analysis of metabolic functions in the accessory genome (i.e., genes not present in every strain). Overall, the metabolic function in the accessory genome partitioned by bacterial order. Two species, Gluconobacter cerinus (Rhodospirillales) and Lactiplantibacillus plantarum (Lactobacillales) had large accessory genomes, and metabolic functions were dominated by amino acid metabolism ( G. cerinus ) and carbohydrate metabolism ( La. plantarum ). The patterns of variation in metabolic capabilities at multiple phylogenetic scales provide the basis for future studies of the ecological and evolutionary processes shaping the diversity of microorganisms associated with natural populations of Drosophila .",
"introduction": "Introduction Significance The metabolic capability of microorganisms can be inferred from genome sequence data but metagenomics and related -omics methods widely used to study complex microbial communities, including microbiomes in animal guts, cannot assign specific metabolic functions to specific taxa with certainty. Our analysis of the genome sequence of 96 bacterial isolates from the gut microbiome of Drosophila fruit flies identified considerable metabolic variation at multiple taxonomic levels, ranging from substantial among-order differences to strain-level variation for several species. The assignment, in this study, of function to taxon for members of a complex gut microbiome provides the basis for future studies on ecology and evolution of bacterial metabolism in gut microbiomes. Animal gut microbiomes are complex assemblages of microorganisms which mediate diverse functions that impact host physiology, behavior, and fitness ( Nicholson et al. 2012 ; Sommer and Bäckhed 2013 ; Huang et al. 2015 ; Rolhion and Chassaing 2016 ; Thaiss et al. 2016 ; Read and Holmes 2017 ; Qiao et al. 2019 ; Turkiewicz et al. 2019 ). Most interactions between the microbiome and the animal host are based on the metabolic capabilities of microbiome members, with traits ranging from degradation and fermentation of host-inaccessible substrates to synthesis of key nutrients for the host, detoxification of harmful dietary constituents and recycling of metabolic waste products, and effects on host signaling pathways ( Hooper et al. 2002 ; Engel and Moran 2013 ; Ankrah and Douglas 2018 ). Investigation of the relationship between traits and taxonomic identity of gut microorganisms has shown that many metabolic traits are functionally redundant and can be shared by closely and distantly related microbiome members ( Heintz-Buschart and Wilmes 2018 ; Louca et al. 2018 ). This finding is largely based on metagenomic studies, where the taxonomic composition of the microbiome is uncontrolled and variable ( Huttenhower et al. 2012 ; Lozupone et al. 2012 ). Functional redundancy can ensure sustained function (also known as ecosystem resilience) of the gut microbiome during perturbations that reduce the abundance or function of specific taxa and alter the overall microbiome composition ( Allison and Martiny 2008 ; Heintz-Buschart and Wilmes 2018 ). Evolutionary changes, which can occur within ecological timeframes, can also affect the relationship between taxonomy and function. In particular, phylogenetically divergent taxa may share a metabolic trait by gain of function through horizontal gene transfer (HGT), and closely-related taxa may differ in functional traits by differential gene deletions and by functional divergence of a recently duplicated gene ( Louca et al. 2018 ). Two examples illustrate these processes. The first is the bile salt hydrolase gene, which is involved in lipid homeostasis and antimicrobial effects. This gene is widespread across bacterial taxa in the human microbiome (most prevalent among the Firmicutes) with evidence of HGT events among different lactobacilli and Listeria monocytogenes ( Jones et al. 2008 ; Kumar et al. 2012 ; Chand et al. 2017 ). Secondly, in the honey bee gut microbiome, the distribution of a glucoside hydrolase gene family (genes involved in degradation of hemicellulose in pollen) in Bifidobacterium spp. is the result of gene duplication and deletion events ( Zheng et al. 2019 ). The apparent ubiquity of functional redundancy, however, is open to question. Functional composition analyses often rely on broad metabolic annotations that can encompass multiple pathways ( Langille 2018 ). These methods can fail to detect biologically important differences in metabolic function of gene families, as demonstrated, for example, in Proteobacteria of the human gut microbiome ( Bradley and Pollard 2017 ). Compounding these problems, within-species variation in metabolic function can be widespread, such that metabolic traits important to the host are displayed by only a subset of strains or are mediated by pathways distributed across two or more different strains ( Douglas 2020 ). For example, Bifidobacterium longum , a member of the microbiome of the human infant, has a large accessory genome with variable incidence of genes involved in transport and degradation of human milk oligosaccharides, implicating some, but not all, strains of this species as important to human milk metabolism ( Vatanen et al. 2019 ). Intraspecific variation requires identification of not only the pangenome (i.e., total genetic capabilities) of a species, but also how the functional traits are distributed across different strains ( Tettelin et al. 2005 ; Brockhurst et al. 2019 ; Van Rossum et al. 2020 ). The goal of this study was to investigate how primary metabolism functions of a gut microbiome map onto bacterial phylogeny. We used the gut microbiome of Drosophila melanogaster for this analysis because, unlike the microbiome of many animals, most of the Drosophila- associated bacteria are readily culturable ( Douglas 2019 ). Relative to metagenome-assembled genomes, genome sequences of the individual bacterial isolates enable higher quality assembly and increased resolution of phylogenomic patterns ( Van Rossum et al. 2020 ). More generally, Drosophila is a fast-emerging system to investigate ecological and evolutionary questions regarding animal-associated microbiomes ( Broderick and Lemaitre 2012 ; Erkosar et al. 2013 ; Wong et al. 2016 ; Douglas 2019 ) and there are indications that, as for the mammalian gut microbiome, the Drosophila metagenome displays incongruence between functional traits and taxonomic composition ( Newell et al. 2014 ; Petkau et al. 2016 ; Adair et al. 2018 ; Consuegra et al. 2020 ; Kang and Douglas 2020 ). However, the relationship between taxonomy and distribution of traits has not been robustly tested. For our analysis, we focused on bacterial taxa isolated from natural populations of Drosophila , which are associated with rotting fruits ( Markow 2015 ). The gut microbiome of wild Drosophila is dominated by members of the bacterial orders Enterobacterales, Lactobacillales, and Rhodospirillales, although the relative abundance of the different taxa varies among individuals and collections ( Chandler et al. 2011 ; Adair et al. 2018 ; Walters et al. 2020 ; Wang et al. 2020 ). Long-term laboratory cultures of Drosophila were not used because their gut microbiome is of low diversity ( Cox and Gilmore 2007 ; Staubach et al. 2013 ; Wong et al. 2013 ; Obadia et al. 2018 ) and can be functionally different from wild populations ( Winans et al. 2017 ; Bost et al. 2018 ). The great majority of published studies on the genome sequences of Drosophila gut microorganisms have concerned bacterial taxa derived from laboratory lines ( Broderick and Lemaitre 2012 ; Matos and Leulier 2014 ) with few sequences available from field isolates ( table 1 ). Therefore, this study was initiated by the isolation of bacteria from field-collected Drosophila . In total, we isolated and sequenced the genomes of 81 bacterial strains associated with wild Drosophila . We performed comparisons of metabolic traits among all field-isolated strains, and then examined the metabolic pangenomes of prevalent species to assess the scale of within-species variation. Within this panel of bacteria, the three bacterial orders were strongly differentiated by primary metabolic functions, and a subset of species also displayed strain-level variation in metabolism-related genes. The taxonomically variable traits include functions likely to be adaptive for utilization of the sugar-rich rotting fruit environment and are predicted to influence Drosophila physiology and performance. Table 1 Bacterial Strains Used in Comparative Genomics Analyses Order Family Genus Species (Strain ID) No. Strains Sequenced (no. flies) Publicly Available Strains Enterobacterales Enterobacteriaceae \n Citrobacter \n sp. (C) 1 (1) \n Enterobacter \n asburiae (Ea) 1 (1) ludwigii (El) 1 (1) mori (Em) 1 (1) sp. (E) 1 (1) \n Klebsiella \n michiganensis (Km) 1 (1) variicola (Kv) 1 (1) Erwiniaceae \n Pantoea \n dispersa (PAd) 2 (1) sp. (PA) 1 (1) \n \n Tatumella \n \n \n sp. #1 (T) \n \n 6 (6) \n sp. #2 (T) 1 (1) Morganellaceae \n \n Providencia \n \n alcalifaciens (PRa) 1 b burhodogranariea (PRb) 1 b \n rettgeri (PRr) \n \n 4 (4) \n \n 1b \n sneebia (PRs) 1 b sp. (PR) 3 (3) Yersiniaceae \n Nissabacter \n archeti (Na) 1 (1) \n Serratia \n rubidaea (Sr) 1 (1) Lactobacillales Lactobacillaceae \n Lacticaseibacillusa \n paracasei (LApa) 1 (1) 1 c \n \n Lactiplantibacillus a \n \n \n plantarum (LApl) \n \n 5 (5) \n 1 d \n Leuconostoc \n citreum (LEc) 1 e mesenteroides (LEm) 1 (1) pseudomesenteroides (LEp) 1 (1) suionicum (LEs) 1 (1) \n \n Levilactobacillus a \n \n \n brevis (LAb) \n \n 5 (5) \n \n Weissella \n cibaria (Wc) 1 f minor (Wm) 1 (1) Streptococcaceae \n Lactococcus \n lactis (Ll) 1 g Rhodospirillales Acetobacteraceae \n \n Acetobacter \n \n cibinongensis (Ac) 1 h indonesiensis (Ai) 1 h okinawensis (Aok) 2 (1) orientalis (Aor) 2 h persici (Ap) 3 (2) \n thailandicus (Ath) \n \n 4 (4) \n \n 1h \n tropicalis (Atr) 1 h \n \n Gluconobacter \n \n albidus (Ga) 1 (1) \n cerinus (G8c) \n \n 13 (5) \n japonicus (Gj) 1 (1) \n kondonii (Gk) \n \n 6 (5) \n sp. #1 (G) 3 (2) sp. #2 (G) 1 (1) sphaericus (Gs) 3 (2) wancherniae (Gw) 3 (1) Note .—Prevalent species (detected in four or more flies and represented by >4 strains in our data set) used for pangenome analyses are in bold. a Genus formally known as Lactobacillus. b Galac and Lazzaro (2012) . c Hammer et al. (2017) . d Petkau et al. (2016) . e Wright et al. (2017) . f Ricks et al. (2017) . g Chaston et al. (2014) . h Winans et al. (2017).",
"discussion": "Discussion A robust understanding of the relationship between the taxonomic identity and functional traits of microorganisms is essential for detailed analyses of the ecological and evolutionary processes that shape microbial communities. This relationship is particularly important for the microbial communities in animal guts because microbial function can influence many host traits, but the pattern and scale of the effect of variation in taxonomic composition on microbial function are poorly understood. This study on the comparative genomics of bacteria isolated from the guts of wild Drosophila focused on bacterial metabolic traits, which have been implicated in the metabolic health and fitness of animal hosts ( McFall-Ngai et al. 2013 ; Visconti et al. 2019 ), including Drosophila ( Chaston et al. 2014 ; Newell et al. 2014 ; Bost et al. 2018 ; Consuegra et al. 2020 ). Two key results were obtained. First, representatives of the three dominant bacterial orders (Enterobacterales, Lactobacillales, and Rhodospirillales) can be differentiated by key metabolic traits, based on annotations and homology of metabolism-related genes. Second, evidence for within-species variation in metabolic functions was obtained, including functions relevant to utilization of the sugar-rich habitats and interactions with the Drosophila host. Here, we consider these two issues in turn. Our finding that the variation in metabolic function partitions by the three bacterial orders of gut bacteria ( fig. 3 ) reflects the differences in lifestyles of the bacteria. Important for interpretation of these results, these differences relate exclusively to the panel of genomes isolated from Drosophila guts, comprising members of just one, two, and four families for Rhodospirillales (five families in total on NCBI), Lactobacillales (five families in total on NCBI), and Enterobacterales (nine families in total on NCBI), respectively ( table 1 ). The diversity of taxa studied are functionally restricted by the conditions in the Drosophila gut, including physical instability, hypoxia (but not anoxia), low pH, and immunological defenses ( Lemaitre and Miguel-Aliaga 2013 ; Douglas 2018 ). A further potential issue is that some taxa in the Drosophila gut microbiome may be intractable to cultivation but the magnitude of this difficulty is likely low because the taxa in the genome panel ( table 1 ) match well to the results from cultivation-independent studies on Drosophila collected from the same habitats in New York State ( Adair et al. 2018 ; Bost et al. 2018 ; Kang and Douglas 2020 ). Further studies are required to assess whether these conclusions apply to flies in other locations. The key lifestyle features of Acetobacteraceae (Rhodospirillales) relate to their adaptation to high sugar habitats, such as the rotting fruits utilized by Drosophila ( Lievens et al. 2015 ). The distinctive metabolic features identified in this study ( fig. 3 B ) relate to aerobic fermentation of exogenous sugars via processes dependent on the tetrapyrrole derivative pyrroloquinoline quinone ( Matsutani and Yakushi 2018 ) and the capacity to utilize simple inorganic and organic nitrogenous substrates for the synthesis of amino acids required for protein synthesis and proliferation ( Sainz et al. 2017 ). Similarly, all but one of the Lactobacillales in this study comprised members of the family Lactobacillaceae and have the functional traits of fermentative metabolism, especially of sugars and other organic compounds, including terpenes and nucleotides ( Duar et al. 2017 ). Many of the products from these metabolic pathways are likely to be important for Drosophila growth and physiology, as illustrated by the evidence that the amino acids produced by Acetobacter may promote Drosophila larval development ( Consuegra et al. 2020 ). On the contrary, the Enterobacterales associated with Drosophila are taxonomically and functionally more diverse ( fig. 1 D and E ). The lifestyles represented by the Enterobacterales in our panel likely include both free-living bacteria associated with the food ingested by the flies and taxa that may be pathogenic to Drosophila , for example, some strains of P. rettgeri ( Galac and Lazzaro 2011 ; Adair et al. 2018 ). This metabolic diversity probably accounts for the single metabolic trait (i.e., arginine, urea cycle, and polyamine metabolism pathway) that partitions with the Enterobacterales ( fig. 3 B ). Unlike the Acetobacteraceae and Lactobacillaceae, relatively little is known about the dynamics of Enterobacterales and other γ-Proteobacteria in the Drosophila gut, beyond the observations that γ-Proteobacteria are generally not detectably beneficial, or can be detrimental, to Drosophila (e.g., Galac and Lazzaro 2011 ; Chaston et al. 2014 ), and that host filtering processes may limit their abundance in the gut ( Wang et al. 2020 ). The association of Enterobacterales with the urea cycle and polyamine synthesis raises the possibility that the association of these bacteria with Drosophila may be facilitated by their capacity to utilize Drosophila waste urea as a nitrogen source and to tolerate hostile conditions in the gut via polyamine-mediated stabilization of the genome and membranes. Microbiome-mediated polyamine production has also been implicated in microbiome effects on human health ( Tofalo et al. 2019 ), but the role of this class of metabolites in Drosophila -microbe interactions has not been investigated. The parallel analysis of within-species variation, conducted on seven species with at least five sequenced genomes, provided the opportunity to assess the scale of among-strain genetic and functional variation in metabolism, including metabolic traits with known effects on Drosophila nutritional physiology and performance (e.g., Shin et al. 2011 ; Chaston et al. 2014 ; Winans et al. 2017 ; Judd et al. 2018 ; Kang and Douglas 2020 ). For this analysis, we used two approaches. First, we compared between-species genetic variation ( fig. 4 D ), which was congruent with annotation-based analysis in figure 3 B . Of the top gene functions found to vary by species, only a few are known to be relevant determinants of Drosophila physiology and some were functionally redundant across disparate taxa ( supplementary table S4 , Supplementary Material online). Several genes involved utilization of Drosophila nitrogenous waste products were identified, primarily among Gluconobacter spp., and these capabilities may allow the taxa to use host nitrogenous waste for their own growth. The second analysis focused on identifying functions enriched in the accessory genome of each species. Interestingly, the majority of the genes that differed within species related to carbohydrate digestion and fermentation as well as carboxylic acid and short chain fatty acid metabolism. The enrichment of carbohydrate metabolism genes is also supported by published pangenome analyses of La. plantarum and P. rettgeri ( Galac and Lazzaro 2012 ; Martino et al. 2016 ). Taken together, the identified gene functions are suggestive of survival in sugar-rich rotting fruit environment that is enriched by the waste products of Drosophila larvae and possibly adults ( Lievens et al. 2015 ; Winans et al. 2017 ; Storelli et al. 2018 ). Rotting fruit provide an energy-rich but ephemeral resource colonized by numerous microorganisms. In this environment, there is strong selective pressure to utilize carbon sources due to exploitative competition and the release of toxic metabolic by-products by co-occurring microbes (e.g., citrate lyase gene functions can be involved in acid stress in lactobacilli; Martin et al. 2005 ). Although we did not sample strains from rotting fruits, various studies indicate that there is frequent cycling of microbes between wild Drosophila gut and the external environment ( Blum et al. 2013 ; Inamine et al. 2018 ; Pais et al. 2018 ), and that this likely limits taxonomic and functional differentiation between strains in Drosophila and the external environment ( Winans et al. 2017 ; Bueno et al. 2019 ; Wang et al. 2020). A related issue is the taxonomic and functional differences between bacteria in the natural environment and associated with laboratory cultures of Drosophila. The limited data available have not identified fixed differences between laboratory-derived bacteria and field isolates, although a higher incidence of genes coding uric acid degradation in laboratory isolates, and of motility genes in wild isolates has been reported in one study of Acetobacteraceae ( Winans et al. 2017 ). Much of the knowledge of microbiome effects on Drosophila metabolism has focused on bacteria isolated from laboratory flies (e.g., Shin et al. 2011 ; Chaston et al. 2014 ; Newell and Douglas 2014 ; Consuegra et al. 2020 ), and future work would benefit from the inclusion of wild-derived bacterial strains. This study also raises methodological issues. One issue relates to the utility of 16S rRNA gene sequence data for taxonomic identification and inference of functional traits. Our analysis reinforces the conclusion of many previous studies, including research on microbiomes, that 16S data can be insufficiently precise to discriminate functionally different microorganisms because functionally important sequences are gained, lost or modified by mutation more rapidly than 16S sequence change ( Koeppel and Wu 2013 ; Ellegaard and Engel 2016 ; Lladó Fernández et al. 2019 ). 16S rRNA gene sequence evolution can also yield phylogenetic patterns that are incongruent with patterns from phylogenomic data, as illustrated for several taxa in figure 2 as well as other bacterial orders ( Maayer et al. 2019 ). Although not explored in this study, other housekeeping genes, for example, gyrB, rpoB , have been suggested as alternatives to 16S rRNA gene for amplicon-based microbiome studies ( Moeller et al. 2016 ; Ogier et al. 2019 ). For these reasons, inferring function from 16S gene surveys ( Langille et al. 2013 ) is less satisfactory than genomic and metagenomic data. A second issue relates to the key limitation of genomic data, that these data provide the genetic capacity for function, and the realized capacity is dictated by gene expression, enzyme activity, and pattern of flux through the metabolic network of individual microbial cells and the microbial community ( Heintz-Buschart and Wilmes 2018 ). In microbiomes, as in other complex microbial communities, the metabolic traits of individual bacterial taxa can be strongly dependent on the identity and metabolic activity of other co-occurring microorganisms, such that the metabolic function of any taxon can be resolved most effectively by a community approach (e.g., Fischer et al. 2017 ; Douglas 2020 ; Henriques et al. 2020 ; McMullen et al. 2020 ). A final issue relates to the use of draft genomes in comparative genomics analyses. Draft genomes include poorly sequenced regions of the genome (e.g., due to repetitive regions and mobile genetic elements) and can have genes split across contigs ( Ricker et al. 2012 ). For pangenome analyses, these limitations can lead to genes being miscategorized as accessory (i.e., not present in all strains). In the present analysis, many of the genomes were of draft status, and therefore some designations of accessory functions may be inaccurate. Nevertheless, the highly significant positive correlation between strain diversity and the accessory genome size (supplementary fig. S6 B , Supplementary Material online) indicates strongly that the observed variation in pangenome size has a biological basis. We conclude by considering how this study informs our understanding of metabolic trait distribution among members of animal gut microbiomes. The taxonomic and functional composition of animal gut microbiomes are influenced by diet, host, and co-occurring microorganisms. By identifying the microorganisms that mediate different functions and their evolutionary relationships, this study provides a basis to understand and predict microbiome functions, which is the foundation for rationally designed routes to manipulate microbiomes for treatment of metabolic disease and application of probiotics ( Bauer et al. 2015 ). The identification of variation in metabolic functions at different phylogenetic scales in this study provides the basis for future studies to determine the ecology and evolution of microbiome functions of Drosophila in natural settings."
} | 6,069 |
26296163 | null | s2 | 7,186 | {
"abstract": "The underlying physical properties of microfluidic tools have led to new biological insights through the development of microsystems that can manipulate, mimic and measure biology at a resolution that has not been possible with macroscale tools. Microsystems readily handle sub-microlitre volumes, precisely route predictable laminar fluid flows and match both perturbations and measurements to the length scales and timescales of biological systems. The advent of fabrication techniques that do not require highly specialized engineering facilities is fuelling the broad dissemination of microfluidic systems and their adaptation to specific biological questions. We describe how our understanding of molecular and cell biology is being and will continue to be advanced by precision microfluidic approaches and posit that microfluidic tools - in conjunction with advanced imaging, bioinformatics and molecular biology approaches - will transform biology into a precision science."
} | 245 |
34938574 | PMC8671620 | pmc | 7,187 | {
"abstract": "Abstract The leakage of industrial oil and organic wastewater discharge has caused serious damage to the natural environment and ecology. Therefore, implementation of a low‐cost and high‐performance adsorbent material is of great significant. This work reports the preparation of superhydrophobic rock wool (RW) for efficient clean‐up of oil and organic solvents. The modified RW is prepared by coating a commercial RW with reduced graphene oxide (RGO) nanosheets under hydrothermal treatment. To improve the adhesion between the RGO nanosheets and RW, a film of chloroperene rubber is deposited on the RW surface followed by modification with RGO. The modified RW possesses superhydrophobicity and superoleophilicity with a water contact angle of 152.4°, and it is used for separation of oil–water mixture. The modified RW exhibits excellent mechanical elasticity and durability when compared with commercial one, and the adsorbed oils are recycled by simple squeezing. Its oil adsorption capacities are maintained above 95%, after several compression cycles. Importantly, the modified RW exhibits excellent photothermal properties which are beneficial for the separation of high‐viscosity oils. Owing to low costs, versatility, and scalability in production, the modified RW can be regarded as a suitable choice for large‐scale oil/water separation.",
"conclusion": "3 Conclusion We have reported preparation of modified RW using RGO with excellent superhydrophobic property. Addition of CR into the porous structure of RW further enhanced the surface roughness and hydrophobicity. The RW without CR coating demonstrated severe mechanical damage, while the RW with CR coating exhibited excellent mechanical elasticity and durability. The modified RW exhibited the adsorption capacity from 520 to 1030 wt%, depending on the density and viscosity of the oils and organic solvents. Due to the higher density of modified RW, the adsorption capacity of modified RW is lower than that of unmodified one. We also showed that the adsorption rate with the high viscosity engine oil is lower than that of low viscosity oil. The modified RW efficiently separated oil/water mixtures with separation efficiency higher than 95%. Owning to easy modification procedure, low cost, high separation efficiency, excellent durability, and recyclability, the modified RW may be a promising candidate for removing oil spills on a large scale.",
"introduction": "1 Introduction Oil‐spill accidents and discharge of industrial wastewater have caused serious environmental issues. The spilled oil and contaminated water can be transported by ocean currents to large distances, creating permanent effects on the marine ecosystem. In the Sanchi oil spill accident in 2018, one of the most polluting oil spill accident into the 21st century, ≈136 000 tons of oil spilled into the East China. [ \n \n 1 \n \n ] This accident imposed devastating impacts on the environment. In this regard, the cleanup strategies of oil spill should be implemented to significantly reduce the environmental damages. Conventional strategies including skimming, [ \n \n 2 \n \n ] bioremediation, [ \n \n 3 \n \n ] in situ burning, [ \n \n 4 \n \n ] and adsorption [ \n \n 5 \n \n ] have been widely used for oil–water separation. Among these approaches, adsorption is considered the most effective method with simplicity and low cost for oil pollutions removal. [ \n \n 6 \n \n ] Several adsorbent materials such as carbon aerogels, [ \n \n 7 \n \n ] graphene sponges, [ \n \n 8 \n \n ] and magnetic foams [ \n \n 9 \n \n ] have been used for this purpose due to their large pore volume and strong adsorption capacity. However, complicated preparation processes, high costs, and scalability challenges restrict the practical applications of these adsorbent materials. In this regard, the use of 3D porous materials such as wools and sponges is more effective. [ \n \n 10 \n , \n 11 \n \n ] \n Rock wool (RW) is an inorganic fiber material with the characteristics of high tensile strength, high chemical resistance, and excellent incombustibility. [ \n \n 12 \n \n ] It has been widely used as thermal insulator in buildings and industrial equipment. RW can be a good candidate for oil/water separation owning to its porous structure, low density, and low cost. However, due to the mixed hydrophilic/oleophilic properties, the original RW adsorbs both water and oil simultaneously. Thus, it is not suitable for selective adsorption of oil. The modification of original RW could improve its hydrophobicity. Hao et al. [ \n \n 13 \n \n ] recently used polydimethylsiloxane (PDMS)/silica nanoparticle as modifying agent for preparation of superhydrophobic RW. However, PDMS/silica nanoparticle used as modifier is quite expensive which limits its large‐scale applications. Graphene materials including graphene and reduced graphene oxide (RGO) are promising candidates as modifier considering their hydrophobicity. [ \n \n 14 \n , \n 15 \n \n ] Carbon lattice of graphene and its derivatives efficiently repels polar molecules and enhances affinity towards oil and organic solvents. [ \n \n 16 \n , \n 17 \n \n ] Graphene materials can increase the roughness of a substrate when used as coating. Presence of small amounts of oxygen‐containing functional groups on RGO increases affinity toward polar solvents. [ \n \n 16 \n \n ] Jamsaz and Goharshadi [ \n \n 18 \n \n ] used polyurethane sponge coated with RGO and orthoaminophenol for oil–water separation. The prepared sponge exhibited the adsorption capacities of 80 times of its own weight for several oil and organic solvents. RGO can be produced in large scale and low price. The π‐electron delocalization on RGO leads to the broadband absorption of the solar spectrum. [ \n \n 19 \n \n ] Thus, it can be heated up under natural sunlight to decrease oil viscosity and collect high‐viscosity spilled oil to facilitate the separation process. Wang et al. [ \n \n 20 \n \n ] reported RGO nanosheets coated melamine sponge with excellent photothermal property. Upon light irradiation, the surface temperature of RGO‐melamine sponge increased rapidly and lowered oil viscosity. In this work, for the first time, we used RGO‐coated RW for selective adsorption of oil from oil–water mixtures. However, modified RW with RGO can be degraded in high‐viscosity oils because of weak adhesion between RGO and RW surface and the poor mechanical stability of the rough surface structure. To further enhance hydrophobicity and recyclability, chloroprene rubber (CR) was deposited on the RW surface followed by the second modification with RGO. The elasticity of CR is crucial for preparation of recyclable modified RW. CR shows better mechanical properties than other resins and adhesive materials and allows fast oil recovery by simple squeezing method. Zulfiqar et al. [ \n \n 21 \n \n ] created a robust superhydrophobic surface using CR as an adhesive to bind sawdust particles on glass slide and used it for oil/water separation. The introduction of CR layer to RW structure further enhanced the adhesion of RGO into the RW surface and increased the roughness, flexibility, and durability. The resulted RGO/CR‐RW composite possesses desirable hydrophobicity and oleophilicity.",
"discussion": "2 Results and Discussion 2.1 Preparation The schematic diagrams of preparation superhydrophobic RW and the possible interactions are shown in Scheme \n 1 \n . To improve the mechanical properties and hydrophobicity, CR adhesive layer was deposited on the surface of the RW by a simple dip‐coating method. CR generates a layer of coating on the RW surface. Then, a secondary layer of RGO nanosheets was covered on the RW framework. The CR can bond RGO nanosheets onto the RW surface using H‐bonds via the interaction between the Cl atom and O of functional groups such as COOH and OH and enhances the bonding energy between RGO and RW. [ \n \n 22 \n \n ] The hydrophobic RGO/CR composite created the hierarchical rough structure with low‐surface‐energy property on the RW surface. The hydrophobic carbon lattice of RGO could interact with a wide range of organic solvents, hydrocarbons and oils through π–π conjugation, while it efficiently repels polar molecules such as water. Also, it can increase the roughness of a surface when used as a coating and further enhance the surface hydrophobicity. The small amounts of oxygen containing functional groups present on RGO enhance the affinity toward polar solvents. Scheme 1 Schematic diagrams of preparation superhydrophobic RW. 2.1.1 Factors Affecting the Preparation of RGO/CR‐RW To obtain an optimal preparation, we studied the effect of hydrothermal reduction time, hydrothermal reduction temperature, graphene oxide (GO) concentration, and CR concentration. The samples were prepared under different hydrothermal temperature (100, 200, and 400 °C) and different hydrothermal time (1, 2, and 4 h). The hydrophobicity of the modified RW is mainly related to the reduction degree of GO. In 100 °C, the modified RW had somewhat hydrophilic properties. In the low temperature, the oxygen‐containing functional groups on the GO surface were less removed. In addition, the high hydrothermal temperature (400 °C) affects the physical state of RW, thereby affecting the surface energy of RW. In addition, GO was not completely reduced in a short time. With increasing hydrothermal time, the degree of reduction of GO gradually increased and the prepared RW obtained proper hydrophobicity. Thus, appropriate hydrothermal condition for preparation of modified RW was 200 °C for 2 h. Additionally, high concentration of GO on the RW structure might block the pores, leading to reduction of the adsorption capacity. The superhydrophobicity RW increased in presence CR due to the reduced surface energy, while excessive CR concentration did not contribute to the superhydrophobicity of RW. High concentration of CR led to a collapse of overall macrostructure of RW. 2.2 Characterization The morphology of the pristine and the modified RW is presented in Figure \n 1 \n . The scanning electron microscopy (SEM) images of the pristine RW show a smooth structure (Figure 1a 1 –a 3 \n ). Figure 1b 1 –b 3 \n shows the modified RW with RGO nanosheets without CR layer, in which RGO nanosheets are heterogeneously adhered to the RW surface resulting in a relatively rough surface. High‐magnification SEM image (Figure 1b 3 \n ) displays a large number of RGO sheets on the framework of the RW. However, there is not strong binding force between RGO nanosheets and RW surface, resulting in poor stability. Figure 1c 1 –c 3 \n shows a thin layer of CR covered on the RW surface with strong adhesion the RGO nanosheets onto the RW surface. RGO nanosheets are embedded in CR layer provides hierarchical rough structure. The introduction of CR onto the RW structure further enhanced the roughness. This rough structure is due to the agglomeration of RGO nanosheets onto the RW surface under the adhesion of CR. Figure 1 SEM images of a 1 – a 3 ) RW; b 1 –b 3 ) RGO‐RW; c 1 –c 3 ) RGO/CR‐RW. Right and left columns show 20, 10, and 2 µm magnification, respectively. The Fourier transform infrared (FTIR) spectra of the modified and unmodified RW are depicted in Figure \n 2 a . Different steps of RW modifications are shown in this figure. The strong absorption band around 990 cm −1 is attributed to the existence of oxide compounds of Al, Mg, Ca, and Si on the surface of the RW. [ \n \n 12 \n \n ] There is a weak band at 3460 cm −1 attributed to —OH groups on RW structure. These functional groups are both involved in adsorption of oil and water. After modification with RGO, the intensity of band corresponding to —OH decreased. The absorption bands at 1615 and 1704 cm −1 are assigned to the presence of stretching and bending vibrations of —C=C and —C=O groups, respectively. In the case of RGO/CR‐RW, the intense bands at 2900 and 1450 cm −1 are attributed to the stretching and bending vibrations of C—H groups in CR. [ \n \n 23 \n \n ] \n Figure 2 a) FTIR spectra of RW, RGO‐RW, and RGO/CR‐RW; b) XRD patterns of RW, RGO, and RGO/CR‐RW; c) TGA curves of RW and RGO/CR‐RW. To further confirm the successful modification of RW, the structural properties of both pristine and modified RW were also determined using X‐ray diffraction (XRD) analysis. As shown in Figure 2b , a characteristic peak (2θ = 30.5°) is found in the XRD pattern of RW, suggesting the existence of CaO, MgO, SiO 2 , and Al 2 O 3 compounds. [ \n \n 12 \n \n ] The XRD pattern of RGO shows a strong and sharp diffraction peak at 2θ = 22.6° and a relatively weak diffraction peak at 2θ = 43.8°. [ \n \n 24 \n \n ] The XRD pattern of RGO/CR‐RW shows an intense characteristic peak at 2θ = 30.3°. It can be observed that the RGO/CR‐RW have diffraction pattern characteristics similar to RW, suggesting that modification have few effects on the crystalline structures of RW. The thermal stability of the RW and RGO/CR‐RW were determined by TGA and is shown in Figure 2c . Obviously, the pristine RW had no clear weight loss in the temperature range of 25–800 °C indicating high thermal stability of the unmodified RW. The RGO/CR‐RW display only a weight loss of ≈3% at 300–500 °C, which is probably due to the decomposition of O‐containing functional groups in RGO and degradation of CR. [ \n \n 25 \n , \n 26 \n \n ] It can be concluded that presence of RGO/CR as modifier on the RW surface does not make a significant difference in thermal stability of RW. 2.3 Superhydrophobicity The surface hydrophobic properties were investigated by measuring the WCA of the pristine and modified RW. The pristine RW showed superhydrophilicity with a WCA of 0°. The results also verified that the contact angle for directly grafting RGO onto the RW surface was only 138.5 ± 3.2°, which did not have superhydrophobic property. When the CR was coated on RW surface, the WCA was increased to 152.4 ± 0.2°, changing the property from hydrophobic to superhydrophobic surface ( Figure \n 3 a ). This can be attributed to the intensification of roughness obtained from RGO nanosheets when coupled with low surface energy of adhesive materials (Figure 1 ). This result confirms that the CR adhesive layer plays an important role in the preparation of superhydrophobic RW. As shown in Figure 3b , unlike the RGO‐RW, the RGO/CR‐RW composite turned black because of RGO nanosheets being completely coated on the RW surface. It should be noted that the water droplet could maintain spherical shape on the surface of modified RW for more than 600 s, exhibiting the high hydrophobicity of RGO/CR‐RW composite (Figure 3c ). Figure 3 a) WCA of RGO‐RW and RGO/CR‐RW; b) images of RGO‐RW and RGO/CR‐RW; c) photographs of a water droplet on RGO/CR‐RW during 600 s exposure period. \n Figure \n 4 a shows that water and oil droplets are both immediately adsorbed onto the pristine RW (water dyed with methylene blue). The water droplets are spherical on the RGO/CR‐RW and easily roll off the surfaces, indicating the superhydrophobic nature of the modified RW (Figure 4b ). Additionally, the modified RW exhibited not only superhydrophobicity but also superoleophilicity (Figure 4c ). Figure 4d shows the water droplets readily departed from the RGO/CR‐RW surface immersed in n ‐hexane (Video S1 , Supporting Information). This indicated that the superhydrophobicity of RGO/CR‐RW composite was retained even in the presence of an organic solvent. Figure 4 a) Hydrophilicity and oleophilicity of pristine RW; b) superhydrophobicity of RGO/CR‐RW; c) oleophilicity of RGO/CR‐RW; d) superhydrophobicity of RGO/CR‐RW in n ‐hexane; e) mirror reflection of RGO/CR‐RW under water. The superhydrophobic nature of modified RW was further verified by the formation of a silver mirror‐like surface when RGO/CR‐RW immersed in water using an external force (Figure 4e ). This appearance is attributed to entrapping air between the water and surfaces of the adsorbent which prevents water from entering the pores. [ \n \n 27 \n \n ] These results showed that the RGO/CR‐RW is not only desirable oil‐adsorbent but also possess oil–water separation ability. 2.4 Mechanical Property To examine the stability of modified RW under multiple immersions and mechanical squeezing cycles, the mechanical properties of the modified RW were determined. Figure \n 5 a shows that the modified RW is compressed with 200 g weight (repeated 30 times). After removing the weight, the original shape of the RW was restored. Compressive strain efficiency (η′, %) is calculated by the following equation [ \n \n 28 \n \n ] \n \n (1) \n η ′ = H 0 − H C H 0 \n where H \n 0 represents the height of the original RW; H \n C is the height of the compressed RW. The modified RW exhibited the compressive strain efficiency of 50% confirming that the modified RW is remarkably flexible. This is mainly due to the existence of CR adhesive materials on RW structure. Also, the composition of CR with RGO could further improve the mechanical properties of RW. Figure 5 a) Mechanical strength test images of RGO/CR‐RW before and after loading a counterweight; b,c) appearance changes of modified and pristine RW after five squeezing cycles in engine oil. After several cycles of adsorption–desorption, the modified RW showed good recycling performance. Figure 5b,c shows that the modified RW has maintained its shape in high viscosity oil after five squeezing cycles, whereas the pristine RW shows severe mechanical damage by a simple squeezing due to its brittle nature. Ultimately, the shape of the modified RW returned to its normal size after the drying process. Therefore, RGO/CR‐RW is a suitable choice for multiple oil adsorption–squeezing cycles. To evaluate the compressive durability and cyclic properties of the RW and RGO/CR‐RW, cyclic compression testing at 50% strain was conducted ( Figure \n 6 \n ). The modified RW completely restored to its original state after the external force was removed, indicating its great compressive durability and cyclic properties. However, pristine RW cannot be completely returned to its original height (Figure 6a ). It shows about 70% recyclability. Figure 6b shows that the maximum compressive stress decreased from 21.4 to 17.3 kPa after ten cycles, indicating excellent mechanical robustness of the modified RW. Also, the unloading curve could return to the initial point after ten cycles, which suggested that the modified RW possessed outstanding elasticity. This performance makes the modified RW as a recyclable adsorbent for oil and organic solvents by manual squeezing. Figure 6 a) Cyclic compression curve of RW and RGO/CR‐RW at 50% strain. b) Cyclic compression curve of RGO/CR‐RW. 2.5 Oil Adsorption and Oil–Water Separation The high porosity of RW makes it a promising adsorbent for a wide range of oils and organic solvents. Figure \n 7 a shows the adsorption capacity of the pristine and modified RW for various types of oils and organic solvents, including chloroform, acetone, hexane, petrol, kerosene, and engine oil. The adsorption capacity of modified RW was lower than that of pristine RW because the density of modified RW is much larger than the density of the pristine RW. According to Table \n 1 \n , the density and porosity of modified RW are 266.0 kg m −3 and 89.3%, respectively [calculated using Equation ( 3 )]. Modification of RW significantly increased its weight. The higher density of the modified RW reduced its porosity. Chen et al. [ \n \n 29 \n \n ] investigated the effect of melanin sponge density on its adsorption capacities. They also concluded that the adsorption capacity of modified melanin sponge was lower than that of the unmodified sponges due to higher density of modifier. Figure 7 a) Adsorption capacity of RW and RGO/CR‐RW in several types of oils and organic solvents; b) separation efficiency of RW and RGO/CR‐RW in several types of oils and organic solvents. Table 1 Density and porosity of the RW and RGO/CR‐RW Samples ρ [kg m −3 ] Φ [vol%] RW 83.3 96.7 RGO/CR‐RW 266.0 89.3 John Wiley & Sons, Ltd. Figure 7a also shows that the adsorption capacity of modified RW for organic solvents is higher than those of viscous oils such as engine oil. The modified RW exhibited the adsorption capacity ranges from 520 to 1030 wt%, depending on the density and viscosity of the oils and organic solvents. The adsorption capacities for liquids with lower density and higher viscosity were decreased. Low density liquids decrease the weight gain of the RW. Liquids with higher viscosity and surface tension resist their diffusion into porous sponges. These results are in agreement with previous reports. [ \n \n 11 \n , \n 30 \n \n ] It should be noted that the adsorption capacity of modified RW in the present work was much lower than those of several graphene‐based adsorbent materials such as graphene aerogels [ \n \n 31 \n , \n 32 \n \n ] and graphene/polyurethane sponges. [ \n \n 14 \n , \n 33 \n \n ] However, most of these adsorbent materials such as graphene aerogels for oil adsorption have high production cost and complex preparation processes. Low cost of the raw materials and easy fabrication of the modified RW will make it cost effective for scale‐up purposes. On the other hand, the modified RW used in this study exhibited greater performance when compared with some adsorbent materials reported in the literature such as PDMS sponges. [ \n \n 30 \n , \n 34 \n \n ] \n Table \n 2 \n lists the performance of the reported adsorbents compared to the modified RW in this work. Table 2 Comparison of RGO/CR‐RW performance with reported adsorbent materials Samples Adsorption capacity [g g −1 ] CA [°] Cost Ref. PDMS‐graphene sponge 2.2–8.0 126.5 \n 500 mg graphene: 100 USD \n 1 g PDMS: 23.7 USD \n 100 mL DMF: 20.5 USD \n \n https://www.sigmaaldrich.com \n \n Consumption rate: 50 mg graphene, 20 mL DMF, 40 g PDMS \n (specific surface area: 31 m 2 g −1 ) \n Total: 962 USD ≈ 31.0 USD m −2 g −1 \n \n \n [ \n \n 30 \n \n ] \n Graphene‐PU 29.0–33.0 152.0 \n 500 mg graphene: 100 USD \n 250 g cellulose nanowhiskers: 37.6 USD \n 1 m 3 : 100 USD \n 1 EA PU (150 × 150 × 10 mm 3 ): 84.7 USD \n \n https://www.sigmaaldrich.com \n \n Consumption rate: PU (40 × 40 × 20 mm 3 ), 1 g graphene, 5 g cellulose nanowhiskers \n (surface area: 6.4 cm 3 ) \n Total: 211.3 USD ≈ 33.0 USD cm −3 \n \n \n [ \n \n 14 \n \n ] \n Graphene/MWCNT‐PDA aerogel 125.0–533.0 – \n 25 g graphite powder: 11.5 USD \n 1 L H 2 SO 4 : 73.7 USD \n 1 L H 2 O 2 : 6.4 USD \n 500 g NaNO 3 : 43.3 USD \n 25 g KMnO 4 : 18.10 USD \n 1 g MWCNT‐COOH: 127 USD \n 100 g PDA: 268 USD \n \n https://www.sigmaaldrich.com \n \n Consumption rate: 100 mg MWCNT‐COOH, 200 mg PDA, 50 mg GO \n (Surface area: 140 m 2 g −1 ) \n Total: 13.23 USD ≈ 0.1 USD m −2 g −1 \n \n \n [ \n \n 31 \n \n ] \n PDMS sponge 2.5–13.0 122.6 ± 1.7 \n 25 g graphite powder: 11.5 USD \n 1 L H 2 SO 4 : 73.7 USD \n 1 L H 2 O 2 : 6.4 USD \n 500 g NaNO 3 : 43.3 USD \n 25 g KMnO 4 : 18.10 USD \n 500 g APTS: 126 USD \n 100 g DCC: 53.1 USD \n 100 g DMAP: 110 USD \n 1 g PDMS: 23.7 USD \n \n https://www.sigmaaldrich.com \n \n Consumption rate: 5 g PDMS, 0.2 g GO, 2.6 mL APTS, 4 g DCC, 0.3 g DMAP (surface area: 8 cm 3 ) \n Total: 123 USD ≈ 15.4 USD cm −3 \n \n \n [ \n \n 34 \n \n ] \n SiO 2 /PDMS‐RW 6.0–13.0 152.9 \n 50 g SiO 2 nanoparticles: 83.5 USD \n 100 mL PDMS‐OH: 68.6 USD \n 25 mL Tetraethoxysilane (TEOS): 28 USD \n 5 g Dibutyltin dilaurate (DBTDL): 15.2 USD \n \n https://www.sigmaaldrich.com \n \n 1 m 2 RW: 1.8 USD \n \n https://www.alibaba.com \n \n Consumption rate: RW (4 × 4 × 2 cm 3 ), 2 g PDMS‐OH, 0.4 g TEOS, 0.04 g DBTDL, 0.8 g SiO 2 \n \n (Surface area: 64 cm 3 ) \n Total: 50.7 USD ≈ 0.8 USD cm −3 \n \n \n [ \n \n 13 \n \n ] \n RGO/CR‐RW 5.2–10.3 152.4 ± 0.2 \n 25 g graphite powder: 11.5 USD \n 1 L H 2 SO 4 : 73.7 USD \n 1 L H 2 O 2 : 6.4 USD \n 500 g NaNO 3 : 43.3 USD \n 25 g KMnO 4 : 18.10 USD \n 250 g CR: 127 USD \n 250 mL toluene: 20.56 USD \n \n https://www.sigmaaldrich.com \n \n 1 m 2 RW: 1.8 USD \n \n https://www.alibaba.com \n \n Consumption rate: RW (3 × 2 × 2 cm 3 ), 2 mg GO, 300 µL CR, 40 mL toluene \n (Surface area: 42 cm 3 ) \n Total: 3.2 USD ≈ 0.08 USD cm −3 \n \n Present work John Wiley & Sons, Ltd. The superhydrophobic and superoleophilic features of the modified RW are essential for selective oil adsorption in oil–water mixture. The oil–water separation efficiency (η, wt%) is calculated by the following equation [ \n \n 28 \n \n ] \n \n (2) \n η = m 1 − m 2 m 3 \n where m \n 1 is the total weight of oil and water before separation; m \n 2 is the weight of water after separation; and m \n 3 is the weight of oil before separation. The oil–water separation efficiency was calculated to be above 95% for all oils and organic solvents with the ratio of oil to water 1:5 (Figure 7b ). Separation efficiency for pristine RW was much lower than that of modified RW (an average 65% for oils and organic solvents). This indicates that the modified RW can effectively separate oil from water. The adsorption rate of the RGO/CR‐RW composite is related to the solvents’ viscosity. It was really fast for low‐viscous oil and organic solvents. As shown in Figure \n 8 a 1 \n – a 3 \n , the petrol was immediately removed from the water surface by RGO/CR‐RW composite. The adsorption rate of RGO/CR‐RW composite in the same volume of high viscous engine oil was much lower and some excess oil remained on the water surface (Figure 8b 1 \n – b 3 \n ). The RGO/CR‐RW composite adsorbed large volume of viscous engine oil floated on the water surface within 60 s. Figure 8 Adsorption process of a 1 –a 3 ) petrol and b 1 –b 3 ) engine oil floated on water with RGO/CR‐RW. Video S2 (Supporting Information) demonstrates the separation of chloroform–water mixture by gravity. The chloroform was completely collected in an Erlenmeyer, while the water (dyed with methylene blue) was blocked in the balloon. No water could be seen in the Erlenmeyer. This indicates that the modified RW can separate the mixture of oil and water continuously. 2.6 Photothermal Properties Photothermal properties of RW and RGO/CR‐RW composite were compared to confirm the effectiveness of the modified RW in adsorption of highly viscous oil under natural sunlight irradiation. As the high‐viscosity oils are difficult to be adsorbed by the porous materials, increasing the temperature of the high‐viscosity oils can reduce their viscosity effectively. [ \n \n 27 \n , \n 35 \n \n ] With increasing the temperature, the viscosity of oil decreased and the oil adsorption rate increased. [ \n \n 36 \n , \n 37 \n \n ] Due to the outstanding thermal conductivity and effective photothermal properties corresponding to black RGO nanosheets, [ \n \n 38 \n \n ] the modified RW can be heated up under illumination. The temperature changes of the pristine and modified RW were evaluated using a thermal infrared imager camera under illumination on (1.0 kW m −2 ) and off conditions ( Figure \n 9 a ). The surface temperature of modified RW increased rapidly under light illumination from 25 to 76 °C within 100 s. However, the temperature rise of pristine RW was lower than that of the modified RW. When light was off, the surface temperature of modified RW dropped faster than pristine ones which can be attributed to efficient heat dissipation (Figure 9b ). The corresponding infrared thermal images of the pristine and modified RW under illumination showed that the modified RW is more effective in adsorption of highly viscous oil under natural sunlight. Figure 9c shows a highly viscous crude oil droplet on the surface of the modified RW. Under light illumination, the crude oil viscosity decreased and the modified RW completely adsorbed the crude oil after 10 min. The crude oil droplet remains almost unchanged without illumination for more than 60 min. Figure 9 a) Thermal images of RW and RGO/CR‐RW under light on and off conditions; b) surface temperature changes of RW and RGO/CR‐RW with time under light on and off conditions; c) adsorption of a crude oil droplet on the surface of the modified RW under light on and off conditions (power density: 1.0 kW m −2 illumination). 2.7 Recyclability The modified RW have a flexible structure, and thus squeezing is appropriate for its recycling. To investigate the recyclability of the RGO/CR‐RW composite, it was immersed in engine oil and chloroform and then squeezed to remove the adsorbed oil. This process was repeated several times. In order to remove residual oil, the RGO/CR‐RW composite was heated to decrease oil viscosity after squeezing. The chloroform was completely removed in each cycle. As Figure \n 10 a suggests, the adsorption capacity for chloroform did not change for five cycles and retained above 95% of the initial adsorption capacity. In the contrary, the adsorption capacity was declined in viscous engine oil after the first cycle because of the diffusion of oil within the RW pores (Figure 10b ). The adsorption capacity of the RGO/CR‐RW composite in viscous engine oil up to the fifth cycle remained about 88% of the initial adsorption capacity. After five cycles of recycling, the adsorption capacity decreased. This may be attributed to the removal of some of RGO from the RW surface and slight damage of the surface structure in the repeated squeezing and drying process. Figure 10 Recyclability of RGO/CR‐RW in a) chloroform and b) engine oil in five cycles."
} | 7,314 |
25617645 | PMC4305289 | pmc | 7,190 | {
"abstract": "Previous explanations of computations performed by recurrent networks have focused on symmetrically connected saturating neurons and their convergence toward attractors. Here we analyze the behavior of asymmetrical connected networks of linear threshold neurons, whose positive response is unbounded. We show that, for a wide range of parameters, this asymmetry brings interesting and computationally useful dynamical properties. When driven by input, the network explores potential solutions through highly unstable ‘expansion’ dynamics. This expansion is steered and constrained by negative divergence of the dynamics, which ensures that the dimensionality of the solution space continues to reduce until an acceptable solution manifold is reached. Then the system contracts stably on this manifold towards its final solution trajectory. The unstable positive feedback and cross inhibition that underlie expansion and divergence are common motifs in molecular and neuronal networks. Therefore we propose that very simple organizational constraints that combine these motifs can lead to spontaneous computation and so to the spontaneous modification of entropy that is characteristic of living systems.",
"introduction": "Introduction The principles of biological computation are not well understood. Although the Turing Machine and related concepts [ 1 – 3 ] have provided powerful models for understanding and developing technological computing, they have provided less insight for biological computation because they generally assume that the machines themselves, as well as their initial program and data are granted as input. In contrast, the organization of states and transitions of the biological process arise out of phylogenetic and ontogenetic configuration processes and execute autonomously without the intervention of an intelligent external programmer and controller being necessary to supply already encoded organizationally relevant information. Our goal here is to make steps towards understanding biological computation [ 4 – 6 ], by considering the behavior of a simple non-linear dynamical system composed of asymmetrically inter-connected linear-threshold neurons. We suppose that such computations entail a mapping from some input towards a limited (low entropy) region of phase space, which is the solution [ 7 ]. We do not suppose that the computational goal is known—only that computation must conform to this basic entropy reducing process. Here we describe the organizational constraints that make such spontaneous computation possible. Previous authors have explained neural network computation in terms of the convergence of special dynamical systems, and emphasized the attractors to which they converge [ 8 – 13 ]. For example, Hopfield [ 9 , 10 ] has shown how and why the dynamics of symmetrically connected neurons with saturating outputs converge to attractor states; and others have offered similar insights for symmetrically connected linear threshold neurons [ 14 – 16 ]. However, interactions between inhibitory and excitatory neurons are clearly asymmetric, making these studies ill suited to study biological computation. To the extent that asymmetrical networks have been considered at all, this has been through assumptions that reduce asymmetrical networks to approximate symmetry. By contrast, we consider here the dynamics of fully asymmetrical networks, and discover that asymmetry contributes strongly to computational behavior. The scope of our work is restricted to recurrent neural networks with asymmetric coupling that express an important and ubiquitous behavior: soft winner-take-all (sWTA) dynamics. We present a formal account of the response of these networks to exogenous perturbations and use a form of non-linear stability analysis (contraction analysis [ 17 ]) to characterize the itinerant transients than ensue, and which have useful interpretations in terms of neuronal computation and information theory. Contraction Theory offers a more flexible framework than the conventional Lyapunov approach to non-linear stability (See Methods for details). This is particularly the case for non-autonomous systems such as our network, in which external inputs can vary with time. We explore particularly the behavior of network computation during the non-equilibrium phase, when the network is traversing its state-space seeking for a solution. We show that the ability of the network to explore potential solutions depends on highly unstable ’expansion’ dynamics driven by recurrent excitation. This expansion is steered and constrained by negative divergence of the dynamics, which ensures that the dimensionality of the solution space continues to reduce until an acceptable solution manifold is reached. The system then ’contracts’ stably on this manifold [ 17 ] towards its final solution trajectory, which is not necessarily converging to a fixed point. We argue that the simple principle of unstable expansion constrained by negative divergence provides the central organizing drive for more general autonomous biological systems from molecular networks, through neurons, to society. Consider a simple network of non-linear neuron-like elements whose task it is to compute the solution to some problem. The states of the computation are encoded in the activations (firing rates) of the neurons, and the computational transitions between these states arise out of their synaptic interactions. The overall trajectory resulting from the successive transitions through its state space express its computation [ 18 – 20 ]. In current technological systems the hardware states of the system are encoded on binary nodes whose discrete states are imposed by signal restoring [ 21 ] circuitry [ 19 ]. This signal restoration is achieved by extremely high gain, so that a small input bias will drive the node into saturation at one of its two voltage limits. Biology rarely commands such sharply demarcated states and transitions. Instead, molecular and electrophysiological activation functions are often approximately sigmoidal (eg Hill functions, voltage dependent conductance, neuronal current-discharge curves, etc). However, neuronal systems do not typically run in saturation. The typical activation of a neuron is thresholded below, and above this threshold it makes use of only the lower part of its dynamic range. It very rarely enters saturation at the upper end of its activation range. Therefore, a suitable model for neuronal activation is a thresholded linear one. That is, their activity is bounded from below, but their positive activity is essentially unbounded (over any practical range of discharge). This is a very well studied model [ 14 – 16 , 22 ]. As in our previous work, the neuronal network model is composed of thresholded linear neuron-like units coupled through positive (excitatory) and negative (inhibitory) connections (see Fig. 1a ). The unbounded positive range of neuron activation implies that the global stability of networks of these neurons must arise out of their collective interactions rather than from saturation of their individual activation functions as assumed by for example [ 9 , 10 ]. The key interaction here is the inhibitory feedback, which must at least ensure that not all neurons can simultaneously increase their activation [ 16 ]. Previous studies of such models have focused on the mathematically more tractable case in which the connections between neurons are symmetrical, and have no transmission delays. Our networks, by contrast, need not be symmetrical and may have transmission delays. Indeed, the asymmetry of connections will be used to computational advantage, not offered by symmetrical networks. 10.1371/journal.pcbi.1004039.g001 Figure 1 Circuit motifs and simple circuit composed of motifs. (A) Circuit motifs are excitatory self-recurrence (top) and shared inhibition (bottom). I i denotes an external input. (B) Connectivity of a simple WTA circuit, consisting of two excitatory units that compete through shared inhibition. (C) More compact notation to denote the circuit shown in (B). Each excitatory element receives an external input.",
"discussion": "Discussion The contribution of this paper has been to explore the fundamental role of instability in driving computation in networks of linear threshold units. Previous studies of computation in neural networks have focused on networks of sigmoidal units with symmetrical connectivity. Our networks of asymetrically connected LTNs draw attention to important features of computation that were not apparent in these previous models. The conditional selective behavior crucial for computation depends on the threshold nonlinearity of the LTN. However, in order to make use of these non-linearities the network must express substantial gain. Because the activation of LTNs is unbounded for positive inputs, the network can in principle produce very high activations through unstably high gain. In these networks, computation is expressed as passage through a sequence of unstable states. It is this dynamical trajectory by which the network computes [ 1 , 2 , 31 ]. Despite this essential instability, the system does not escape, but remains bounded in its behavior. In this paper we have analyzed why this is so. We find that the instabilities are self limiting, and that the overall process of computation is systematically quenched by Gaussian divergence. Contraction analysis provides explicit tools to quantify both instantaneous rates of exponential convergence to limiting states or trajectories, and divergence rates from specific subspaces. Here, we use these tools to analyze the unstable phase of the dynamics. This phase is crucial, because computation is inseparable from instability. Here we have made steps towards characterizing and explaining these phenomena. The type of dynamical system we consider can implement soft-WTA type behavior, amongst others. This makes our framework applicable to the extensive body of literature on this type of network [ 32 – 38 ]. While simple, the soft-WTA is a powerful computational primitive that offers the same computational power than a multi-layer perceptron [ 32 ]. Key aspects of what makes WTA-networks powerful are high network gain, which allows computations that require sparsification, and also provides stability. While the computational power of WTAs has long been recognized and exploited to implement simple cognitive behaviors [ 23 , 29 , 39 ], it has remained unclear what it means to compute in such networks. Here, we provide such understanding in terms of a dynamical system. This system is physically realizable by realistic neurons and their connections. Other work in this direction has focused on abstract mathematic models [ 36 , 40 – 43 ], and less on physically realizable dynamical computation. More recently, others [ 44 – 46 ] have offered useful models for understanding the principles whereby the brain may attain cognition, but these approaches do not offer methods for implementing such algorithms as physical computation in neuronal circuits. The advances of this paper can be seen in contrast with classical assumptions concerning the form of activation functions, continuous sensitivity to input, and symmetry of connections. For example, the behavior of our LTN networks can be contrasted with networks of the kind originally proposed by Hopfield [ 20 ] that allow no self-connections ( w \n ii = 0, ∀ i ), have symmetric connectivity ( w \n ij = w \n ji ), and their activation function is bounded on both sides. This guarantees bounded dynamics by construction, allowing such networks to express high gain by virtue of a steep activation function rather than through connections of the network. However, a consequence of this is that when it operates with high gain the network operates in saturation and thus becomes insensitive to input apart from initial conditions. Such networks have neither negative divergence nor rotational dynamics, which together with insensitivity to external input severely restricts their computational abilities as well as systematic design. Importantly, our networks are continuously sensitive to their inputs. These external inputs are a combination of signal and noise and can transfer the network from one subspace to an other at any point of time and this transfer can be against the gradient imposed by negative divergence. Non-autonomous systems continuously interact with their environment, for which continuous sensitivity to input is crucial. Systems of asymmetrically interacting linear threshold units are well suited for this situation. This is because their non-saturating units make the system adaptive to the input amplitudes and sensitivity to inputs is conditional on the current state, i.e. only the inputs contributing to the dynamics of the currently active state influence the dynamics. Although there has been a considerable amount of work on symmetric networks, biological neuronal networks are always asymmetric because of inhibitory neurons. Also, the inhibitory inputs to an excitatory neuron can be substantially stronger than the excitatory inputs. This results in particularly strong asymmetry, a property with many implications for computation in such networks [ 47 ]. The theoretical study of networks with defined cell types (excitatory or inhibitory) thus requires asymmetric connectivity. Previous studies have used infinitely fast all-to-all inhibition to circumvent this problem, which results in symmetric connectivity but lacks defined cell types. Such networks allow dynamically bounded activity for linear threshold units [ 16 , 48 ]. Apart from being biologically unrealistic, such networks can only express limited gain and are thus computationally very limited [ 47 ]. By contrast, our networks express high gain and dynamic instabilities during the exploration phase. Their asymmetric connections provide the rotational dynamics that keep their activity bounded despite this high gain. It is worth noting that many powerful algorithms, such as e.g. the Kalman filter [ 49 , 50 ] also rely on negative feedback and strongly asymmetric connectivity. The dynamics of the exploration phase are highly structured because the different forbidden subspaces are systematically related to one another. Indeed, the subspaces are ordered in a hierarchy through which the dynamics proceed. At any point in this hierarchy only a limited and known set of subspaces can be entered next (unless the external input changes). The systematic understanding of the unstable dynamics driving exploration can be used to steer and modify the computational trajectory while it is in process, rather than only when a solution has been found. The network can influence its environment continuously as a function of the forbidden subspaces it traverses, for example by executing a specific action whenever a particular subspace is entered. This feature can be used to make the computations of several networks dependent on each other. For example. to enforce dependencies between several ongoing computations such as, “all solutions must be different”. The connections of the network are the constraints imposed on the computation. The more connections per neuron, the fewer possible solutions exist and the harder (slower) the computation is. From this point of view, the networks we describe perform constraint satisfaction, which is a hard computational problem and which has been proposed as an abstract model of computation [ 51 , 52 ]. Connections can be inserted systematically to forward program specific algorithms and behaviors [ 23 , 24 , 29 ], randomly or a combination thereof. Either way, the system will compute [ 53 ], but in the former case will execute specific algorithms while in the later the algorithm is unknown. The constraints active at any point of time depend on the state of the network as expressed by the effective connectivity of the network expressed by the switching matrix. Every time the network changes state, the switching matrix changes. Dynamically, the same concept can be applied: the effective Jacobian jointly expresses all the currently activity constraints for a given state. Only if the possible state(s) of a network are known is it possible to determine the effective Jacobian. An important implication is that to understand the underlying algorithm that drives the computation performed by a group of neurons knowledge of the structural connectivity is not sufficient [ 54 – 56 ]. This is because connectivity alone does not determine the possible states of the network. The circuit motifs and parameter bounds we describe guarantee that collections of these circuits will also possess forbidden and permitted subspaces and thus are computing. By collections we mean multiple copies of the same motifs that are in addition connected to each other, as for example in the random network discussed in the results. This is important because collections of these motifs will compute automatically, a property we refer to as collective computation. This makes it practical to design large-scale computing systems without having to perform global analysis to guarantee both the type of instability required for computation as well as stability of the solutions. It is important to note that one need not commit to a particular circuit motif beyond guaranteeing that both forbidden and permitted subspaces exist in the way we define them. While a network composed of such motifs but otherwise connected randomly will always compute, the individual states do not have meaning nor is the algorithm that the network computes known. However, the states that the network proceeds through while it computes are systematically related to each other. Consequently, assigning a meaningful interpretation to a few key states will make all states meaningful. A similar approach is used in reservoir computing, where states are first created and only later assigned with meaning by learning mechanisms [ 57 ]. A key next step will be to discover how linking specific forbidden subspaces with motor actions that in turn change the input to the system allow a computation to remain continuously sensitive to the environment while it proceeds. An other next step is to discover how several interacting systems can bias each others computations systematically to reach a solution that is agreeable to all while satisfying the local constraints of each computation. The unstable positive feedback and cross inhibition that underly expansion and divergence are common motifs found in many molecular, cellular and neuronal networks [ 58 ]. Therefore all such systems follow the very simple organizational constraints that combine these motifs. This will lead such circuits to compute spontaneously and thereby to reduce their state entropy as is characteristic of living systems."
} | 4,726 |
27019743 | null | s2 | 7,191 | {
"abstract": "Host-associated microbiomes influence host health. However, it is unclear whether genotypic variations in host organisms influence the microbiome in ways that have adaptive consequences for the host. Here, we show that wild accessions of "
} | 59 |
33293404 | PMC7743110 | pmc | 7,192 | {
"abstract": "In this work, an original methodology was developed that quantifies bioenergetically and physiologically feasible net ATP yields for large numbers of microbial metabolic pathways and their variants under different conditions. All variants are evaluated, which ensures global optimality in finding the pathway variant(s) leading to the highest ATP yield.",
"conclusion": "Conclusions. The automated pathway analysis method developed in this work provides an unprecedented capability to evaluate large numbers of pathway configurations. This allows for the evaluation of any known, and even postulated, biochemistry to theoretically determine the feasible pathways (physiologically and thermodynamically) with the highest ATP yield. The method, entirely mechanistic and largely founded on first principles, brings insights for the study of energy-limited microbial metabolisms. Propionate oxidation was evaluated in the entire domain of possible pathway variations within the known biochemistry and the thermodynamic and physiological feasibility, applied to all reaction steps and all metabolite concentrations. Under a scenario of optimum environmental conditions, the oxidation of propionate via the Smithella pathway yields the most ATP, and the methylmalonyl-CoA pathways can generate sufficient ATP for growth only under a cyclical pathway configuration with pyruvate (P 2 ). Under a scenario of typical methanogenic conditions, the oxidation of propionate via the lactate and via the hydroxypropionyl-CoA pathways appears to yield the most ATP. Extremely low P H2 values (below the minimum reasonable physiological limits) appear to be required to sustain syntrophic growth coexistence with methanogens if propionate is oxidized to acetate and three hydrogens (P 1–6 ), while this is not observed for the Smithella stoichiometry and pathways (P 7 ). This implies that IET via dissolved hydrogen is not feasible under pathways P 1–6 and must occur via alternative mechanisms that could include formate or direct electron transfer (e.g., via conductive pili). Conversely, dissolved hydrogen appears a feasible IET if the propionate oxidation goes via the Smithella pathway (P 7 ). The very different results predicted under most favorable or methanogenic typical conditions suggest that local concentrations or spatial variability via microbial aggregates must be occurring to explain the literature observations for syntrophic propionate oxidation.",
"introduction": "INTRODUCTION Propionate oxidation to acetate and hydrogen is a highly endergonic reaction under standard conditions (Δ G 01 = +76.1 kJ/mol propionate [ 1 ]). The reaction, however, can become exergonic and yield sufficient energy for net ATP production only at very low hydrogen partial pressures (P H2 ) (1 to 10 Pa) typically found in methanogenic environments ( 2 – 4 ). Although the reduction reaction of CO 2 to methane via hydrogen is highly exergonic under standard conditions (Δ G 01 = −131 kJ/mol [ 1 ]), under typical methanogenic conditions (very low P H2 ) the reaction falls much closer to equilibrium, with actual quantities of energy available between −15 and −40 kJ/mol ( 3 ). Volatile fatty acid (VFA) oxidizers and methanogens are both known to grow very close to thermodynamic equilibrium ( 5 ). Due to these bioenergetic limitations, propionate oxidation is believed to occur primarily under syntrophic association with hydrogen-scavenging microorganisms ( 6 ). The specialized nature of methanogenic archaea, such as hydrogenotrophic methanogens, which are able to grow only on a very few substrates (hydrogen and/or formate) ( 4 , 7 ), makes them dependent on other microorganisms for their supply of substrate. Both syntrophic reactions can proceed simultaneously within a narrow range of concentrations if dissolved hydrogen is the interspecies electron transfer (IET) mechanism. This range is known as the methanogenic niche. The fact that, under methanogenic conditions, the amount of energy available from either of the two syntrophic reactions is smaller than the minimum needed for one ATP unit of synthesis via substrate-level phosphorylation (SLP) implies that metabolic energy conservation must be driven by chemiosmotic transmembrane proton (or its equivalent, such as sodium or potassium) translocations ( 1 ). Numerous studies have focused on elucidating the catabolic pathways of propionate oxidation to acetate, for which numerous different possible pathways have been described, including (i) propionate oxidation via the methylmalonyl-coenzyme A (CoA) pathway, which has been extensively studied ( 4 , 6 , 8 – 16 ), (ii) propionate oxidation via lactate ( 12 , 17 , 18 ), or (iii) propionate oxidation via hydroxypropionyl-CoA ( 12 , 17 ). Propionate oxidizers that use the methylmalonyl-CoA pathway are, however, the only ones that have been isolated ( 6 ). Propionate oxidation via an alternative butyrate- and acetate-yielding pathway has also been reported ( 19 , 20 ). In addition, significant work has been done on the elucidation of electron transfer mechanisms between syntrophic partners, propionate (or butyrate) oxidizers with methanogens. Different mechanisms for IET have been proposed to occur via hydrogen and/or formate. Although IET via hydrogen has been identified as more suitable than formate due to its higher diffusivity ( 21 ), IET via formate has also been proposed when microorganisms do not grow in aggregates, given its much higher solubility ( 6 , 8 , 9 , 22 , 23 ). Formate and hydrogen production in the same microorganism have also been proposed to take place at different reaction sites, with (i) formate produced at the reoxidation step of menaquinone from the oxidation of succinate to fumarate and (ii) hydrogen produced in the reoxidation of the NADH from the malate oxidation to oxaloacetate and the ferredoxin reduction of pyruvate to acetyl-CoA reactions, respectively ( 4 , 10 ). The hypothesis of both IET-capable species being simultaneously produced is supported by faster observed growth in the presence of syntrophic methanogens that metabolize both hydrogen and formate ( 24 , 25 ). Formate has been suggested to serve in those cases as a temporary electron sink ( 11 ). Although detailed thermodynamic studies have been conducted on individual reactions present in related microbial catabolic pathways ( 26 – 31 ), the complete understanding of many microbial conversions remains unachieved. This is largely due to the lack of clarity on the different possible pathway variants and/or mechanisms that drive endergonic reactions. Pathway variants in this work consist of any possible configurations compatible with known biochemistry and physiology and are defined in terms of which intermediate metabolites (including which specific electron carriers) are involved and in terms of the mechanisms and locations in which energy conservation by proton translocations or SLP take place within the pathway. A comprehensive bioenergetic evaluation of a very large set of pathway variants is presented in this work for propionate oxidation as well as for its syntrophic counterpart, hydrogenotrophic methanogenesis. The impact that intermediate metabolite concentrations have on the bioenergetics of the reaction step is central to determine the feasibility of each pathway variant and the quantification of its net ATP yield. The syntrophic pathway evaluation for an ample range of hydrogen partial pressures is specifically targeted to understand the limits of the IET mechanism and of the methanogenic niche within which syntrophic propionate oxidizers and methanogens both can simultaneously sustain growth.",
"discussion": "RESULTS AND DISCUSSION The pathway variants with the highest ATP yield, along with their corresponding metabolite concentration profiles and proton translocation configurations, are presented and discussed. Propionate oxidation. Figures 1 and 2 present (for scenarios Opt and Met , respectively [see Materials and Methods]) the results of the positive net ATP yield of the feasible propionate oxidation pathways for different values of the physiological parameters and environmental conditions around the default values (see Table 3 ). The net ATP yields presented are the maxima found for each pathway considering all possible combinations of electron carriers and all possible configurations for proton translocations ( Tables 1 and 2 ). TABLE 1 The complete set of pathway reactions considered for propionate oxidation to acetate per equations 1 and 2 a a The reactions considered for electron carrier regeneration are also included (eC Reg ). The numbers under the pathways (P n ) indicate the order in which reactions occur in each pathway. Lower (LL) and upper (UL) limits for the number of proton translocations in a specific reaction step and ATP consumption/production as the substrate-level phosphorylation (SLP) for each reaction are indicated. TABLE 2 The set of reactions considered for CO 2 reduction with H 2 to methane a a The reactions considered for electron carrier regeneration are also included (eC Reg ). The numbers under the pathways (P n ) indicate the order in which reactions occur in each pathway. Lower (LL) and upper (UL) limits for the number of proton translocations in a specific step are indicated. No ATP consumption/production via substrate-level phosphorylation (SLP) was reported for this reaction. Results under the scenario Opt ( Fig. 1a ) show that, at the default value of −50 kJ/mol of Δ G ATP , the pathway of Smithella propionica (P 7 ) is the one yielding the most net ATP. It appears that a higher or lower Δ G ATP value would lead to lower efficiency of the P 7 pathway. The lactate and hydroxypropionyl-CoA pathways appear to be those capable of producing the most ATP for the oxidation of propionate under the stoichiometry of equation 1 . The methylmalonyl-CoA pathway (arguably the most frequently reported for propionate oxidation) appears to be feasible in most cases only via the cyclical configuration with pyruvate (R 5 in Table 1 ). This is in agreement with several literature observations ( 9 , 11 , 16 ). Alternative configurations for the methylmalonyl-CoA pathway (P 1 and P 3 ) appear feasible only for Δ G ATP values of −55 kJ/mol.The value of the H + /ATP ratio ( Fig. 1b ) shows that small energy quanta (up to an optimum H + /ATP ratio of 13/3) could favor the efficiency of the propionate oxidation pathways ( equation 1 ) (P 1–6 ) with highly efficient energy conservation from the total available in the lactate pathway (see Fig. S5 in the supplemental material).Conversely, the Smithella pathway (P 7 ) appears to yield diminishing net ATP production when the quantum of energy becomes smaller. The concentration values of free CoA ( Fig. 1c ) do not appear to impact the efficiency of the lactate pathway in the range of values evaluated, but they show an optimum range (1 to 10 mM) for the methylmalonyl-CoA (P 2 ) and the Smithella (P 7 ) pathways. Values for intracellular pH ( Fig. 1d ) appear to show an optimum for the methylmalonyl-CoA (P 2 ) and the Smithella (P 7 ) pathways at a neutral pH, while the rest of the configurations (P 4–6 ) appear to be unaffected by pH. The detailed effect of the intracellular pH on the lactate pathway can be seen in Fig. S6 . FIG 1 Net ATP produced for each pathway under different physiological (a to d) and environmental (e to h) parameter values is shown for the optimal concentrations for propionate degraders (scenario Opt ). Only the parameter indicated below each graph is modified with respect to the reference conditions ( Table 3 ). The horizontal dark lines showing the total catabolic energy available in each pathway (−Δ G cat ) allow for pathway efficiency visualization. Regarding the differences in the environmental conditions considered ( Table 3 ), it is worth noting that these imply changes in the overall catabolic energy available. Lower values of P H2 ( Fig. 1e ) make the overall reaction more exergonic, and potentially more net ATP can be produced. Due to the different amounts of hydrogen produced in the two propionate oxidation stoichiometries ( equations 1 and 2 ), their total catabolic energies available are different and are impacted differently by P H2 , even intersecting at some values. At a P H2 higher than 30 Pa, the Smithella pathway (P 7 ) remains the most exergonic with respect to the other pathways (P 1–6 ) and is favored for higher net ATP yield production ( Fig. 1e ). TABLE 3 Set of physiological parameters and environmental conditions evaluated for propionate oxidizers and hydrogenotrophic methanogens a Parameter Unit Value Physiological ΔG ATP kJ/mol −40 −45 −50 (a)(b) −55 −60 −65 H + /ATP 9/3 (b) 10/3 (a) 11/3 12/3 13/3 14/3 15/3 CoA-SH mM 10 −3 10 −2 10 −1 1 (a) 10 CoM-SH mM 10 −3 10 −2 10 −1 1 (b) 10 H 4 MPT mM 10 −3 10 −2 10 −1 1 (b) 10 pH in 6 6.5 7 (a)(b) 7.5 8 Environmental conditions H 2 Pa 0.01 0.13 (a) 1.62 31.63 (b) 316.3 CO 2 mM 10 −3 10 −2 10 −1 1 10 (a)(b) T b o C 25 (b) 35 (a) 45 55 pH out 6 7 (a)(b) 8 a The values indicated were evaluated for each parameter individually, leaving all other parameters at the default reference value (in boldface). Label (a) refers to default values for propionate oxidizers, and label (b) refers to default values for hydrogenotrophic methanogens. CoA-SH and CoM-SH refer to the concentrations of their free forms. b For hydrogenotrophic methanogenesis, a sensitivity analysis was not performed for temperature. All ATP yields are computed with enthalpies at 25°C. Analogously, for the dissolved external CO 2 concentration (also a product of the overall reaction), the lower its concentration, the more catabolic energy is available for pathways P 1–6 . However, under the environmental conditions of scenario Opt , the net ATP yield remains unaffected for all pathways. It appears that even at high CO 2 concentrations the carboxylation step (R 4 ) does not proceed; therefore, methylmalonyl-CoA pathways (P 1 and P 3 ) do not yield any net ATP. For propionate oxidation via methylmalonyl-CoA to be feasible, the configuration needs to involve no carboxylations, as is the case in the cyclical configuration (P 2 and R 5 ). The impact of temperature ( Fig. 1g ) can be observed in terms of higher temperature leading to more exergonic overall reactions (due to increases in entropy under these stoichiometries), allowing for more pathway variants to obtain positive and higher net ATP yields. The extracellular pH ( Fig. 1h ) shows almost no impact due to the very similar acidity (pK a values) for propionate and acetate (substrate and product of the overall reaction) ( equation 1 ), while it seems to affect the Smithella pathway (P 7 ) due to the impact on the bioenergetics of the reaction of the protons (which is one of the products of equation 2 ). Results under the scenario Met ( Fig. 2 ) correspond to typical methanogenic conditions under which much lower energy is available for microbial growth than under the previous optimum conditions of the scenario Opt . Therefore, lower net ATP yields are observed for every pathway under scenario Met . Under these conditions, the Smithella pathway reaction ( equation 2 ) does not have sufficient available energy to translocate one proton and yields no net ATP. FIG 2 Net ATP produced for each pathway under different physiological (a to d) and environmental (e to h) parameter values is shown for a typical methanogenic environment (scenario Met ). Only the parameter indicated below each graph is modified with respect to the reference conditions ( Table 3 ). The horizontal dark lines showing the total catabolic energy available in each pathway (−Δ G cat ) allow for pathway efficiency visualization. The results consistently present the lactate and the hydroxypropionyl-CoA pathways (P 4 and P 6 ) as the ones biochemically and thermodynamically capable of yielding the most ATP from propionate oxidation under the methanogenic conditions. Only for Δ G ATP values of −65 kJ/mol ( Fig. 2a ) does it appear that the oxidation via methylmalonyl-CoA (P 2 ) could yield a similar net ATP. The H + /ATP ratio ( Fig. 2b ) shows that a small energy quantum (up to an optimum H + /ATP ratio of 14/3) could favor the efficiency of the lactate pathway. The concentration of free CoA ( Fig. 2c ) does not appear to impact the efficiency of the lactate pathway in the range of values evaluated. The net ATP yields by the lactate and the hydroxypropionyl-CoA pathway (P 4–6 ) appear to be unaffected by intracellular pH within the range covered. The different environmental conditions considered ( Table 3 ) show tendencies similar to those in the scenario Opt . Lower values for P H2 ( Fig. 2e ) make the overall reaction more exergonic, and potentially more net ATP can be produced. However, even at very low P H2 values (1.62 Pa), such as those found in methanogenic environments, the energy available is below the energy threshold for one net proton translocation. As in scenario Opt , CO 2 concentration does not impact the net ATP yield for the pathways considered ( Fig. 2f ). Analogously to scenario Opt , the impact of temperature in scenario Met ( Fig. 2g ) can be observed in terms of higher temperature leading to a more exergonic overall reaction (due to increases in entropy under these stoichiometries), allowing for a higher net ATP yield for the lactate and hydroxypropionyl-CoA pathways (P 4,6 ). The extracellular pH ( Fig. 2h ) shows almost no impact due to the very similar acidity (pK a values) for propionate and acetate (substrate and product of the overall reaction). To enable detailed pathway and bottleneck analyses, the intermediate metabolite concentration profiles of all feasible reactions are provided by the automated method developed for the pathway evaluation. In Fig. 3 , the profile is shown for the pathway variants that appeared to yield the most net ATP, namely, propionate oxidation via lactate (P 4 ), at three different partial pressures of hydrogen (P H2 ). FIG 3 Pathway metabolite concentrations in the propionate oxidation pathway via lactate (P 4 ) at different hydrogen partial pressures (P H2 ). Symbols in gray (top) indicate the logarithmic concentration of each metabolite as labeled in the upper axis. Concentrations outside the physiological limits fall in the shaded red area. Green and red bars (middle) indicate energy conservation reactions in which either energy is recovered or consumed to fuel a reaction via proton translocations. Darker green bars indicate ATP production via substrate-level phosphorylation. Yellow bars (bottom) indicate Gibbs free energy dissipations (losses) at that step in the pathway. The default physiological parameters and environmental conditions from Table 3 were used (other than that for P H2 ). Figure 3 shows how all metabolites remain within physiological limits for all the P H2 values evaluated. As the catabolic energy decreases with increasing product concentration (P H2 ), less net energy in the form of translocated protons can be recovered by the cell, particularly in the reoxidation of FADH 2 . For those reactions with products potentially exceeding the maximum physiological concentrations thermodynamically (e.g., pyruvate to acetyl-CoA), energy is dissipated (as described in Materials and Methods). Figure 3 also clearly illustrates the energetic bottlenecks of the lactate pathway (steps leading to very low product concentrations), namely, (i) the oxidation of propionyl-CoA to acryloyl-CoA, for which the influx of two protons is needed, and (ii) the conversion of lactate to pyruvate, highly sensitive to the P H2 value. Hydrogenotrophic methanogenesis. In Fig. 4 , the results obtained from the evaluation of the hydrogenotrophic methanogenesis pathway are presented in terms of net ATP yield as a function of different physiological parameters and environmental conditions around the default values in Table 3 . FIG 4 Net ATP equivalents produced in the hydrogenotrophic methanogenesis pathway for different physiological parameters (a to e) and environmental conditions (f to i). In each plot only one parameter, as indicated, is modified with respect to the default conditions from Table 3 . Temperature (h) could only be evaluated at 25°C due to unavailable enthalpies of formation data for several key components present in the pathway. During hydrogenotrophic methanogenesis, no ATP is produced by substrate-level phosphorylation, and the value of Δ G ATP only impacts the size of the energy quantum in reactions with proton translocation. As shown in Fig. 4 , the reaction has no sensitivity to Δ G ATP in the range from −45 to −60 kJ/mol, while an energy quantum of 40 kJ/mol or smaller appears to allow for one additional net proton translocation. At the default reference Δ G ATP , the optimum H + /ATP ratio appears to be 11/3. Intracellular pH values lower than 7 appear to decrease the net ATP, while no effect is shown from the concentrations of CoM or H 4 MPT within the evaluated ranges. As for the case of propionate oxidation, different environmental conditions imply differences in the overall catabolic energy available (with the exception of extracellular pH, since no net acidity was produced or consumed). Since hydrogen and CO 2 ( Fig. 4f and g ) are substrates of the hydrogenotrophic methanogenesis reaction, the higher their concentration the higher the catabolic energy available and, potentially, the higher the net ATP recovered. Syntrophic propionate oxidation and methanogenesis: methanogenic niche. The pathway evaluation method developed was also applied to gain insight into the syntrophic growth of propionate oxidizers and hydrogenotrophic methanogens. The maximum net ATP yield achievable by each of the two microbial groups was evaluated as a function of the dissolved hydrogen concentration (or its corresponding partial pressure), widely accepted as the syntrophic link for interspecies electron transfer (IET). The energetically equivalent values for alternative possible IET mechanisms are also shown, namely, the concentration of formate and the electron potential for any possible direct IET. The values shown for both are in thermodynamic equilibrium with the corresponding P H2 . All other default parameters as per Table 3 were used for both microbial groups. The methanogenic niche was evaluated under the two environmental conditions as previously defined, namely, scenario Opt ( Fig. 5 ) and scenario Met ( Fig. 6 ). FIG 5 ATP yields under optimal conditions for propionate degraders (scenario Opt ) for propionate oxidizers (top) and hydrogenotrophic methanogens (at 25°C; middle) as functions of P H2 are shown in bars corresponding to the feasible reactions with positive net ATP yields. Horizontal lines indicate the available catabolic energy, in purple for propionate oxidation to acetate (P 1–6 ) and in black for the Smithella pathway (P 7 ). A range of hydrogen partial pressures is shown where both microbial functional groups (a propionate degrader and a hydrogenotrophic methanogen) could sustain growth and coexist (shaded green area). The bottom plot shows the total energy dissipated (lost) in each complete syntrophic reaction. Additional axis for alternative IET via formate and direct electron transfer shows their values of concentration and voltage equivalent (in equilibrium) with the hydrogen concentrations and pressures shown. FIG 6 ATP yields under methanogenic conditions (scenario Met ) for propionate oxidizers (top) and hydrogenotrophic methanogens (at 25°C; middle) as functions of P H2 are shown in bars corresponding to the feasible reactions with positive net ATP yields. Horizontal lines indicate the available catabolic energy, in purple for propionate oxidation to acetate (R P1–6 ) and in black for the Smithella pathway (R P7 ). A range of hydrogen partial pressures is shown where both microbial functional groups (a propionate degrader and a hydrogenotrophic methanogen) could sustain growth and coexist (shaded green area). The bottom plot shows the total energy dissipated (lost) in each complete syntrophic reaction. Additional axis for alternative IET via formate and direct electron transfer shows their values of concentration and voltage equivalent (in equilibrium) with the hydrogen concentrations and pressures shown. Although in scenario Opt the conditions for propionate oxidation are optimal, propionate oxidizers following the pathways P 1–6 (stoichiometry as per equation 1 ) show a limited syntrophic coexistence range of P H2 (or equivalent alternative IET) between 1.2 and 100 Pa. A complete evaluation of all possible pathway variant feasibilities shows that, within this syntrophic P H2 range, propionate oxidation can generate net ATP only via the lactate or the hydroxypropionyl-CoA pathway (P 4–6 ). The methylmalonyl-CoA pathway (P 2 ) was shown to be able to generate only net ATP at P H2 of 3.6 Pa or below. The very low values of P H2 for syntrophic growth coexistence correspond to dissolved hydrogen concentrations between 10 −8 and 10 −6.1 M, below the defined minimum physiological limit of 10 −6 M. This poses a fundamental problem if we consider that, for a bacterial cell volume of circa 1 μm 3 , the number of hydrogen molecules present inside a cell within this concentration range would be as few as 6 to 480. Such small numbers imply a kinetic impossibility for methanogenesis to actually occur. This supports the idea that IET between syntrophic partners should occur through alternative or additional mechanisms other that via dissolved hydrogen. Sustained growth for the methanogenic syntrophic partner, if based solely on dissolved hydrogen as the electron donor, appears theoretically impossible according to this analysis. The equivalent concentrations of formate (taken as thermodynamic equilibrium with hydrogen) are shown in Fig. 5 . An alternative formate IET mechanism appears feasible with concentrations above the defined lower physiological limit (1 μM). Observations of highly expressed enzymes for the reoxidation of quinone or ferredoxin that produce formate in propionate oxidizers such as Pelotomaculum ( 16 ) are in support of formate as an IET mechanism. Direct electron transfer via conductive materials at potentials between −270 and −325 mV also appears feasible ( Fig. 5 ). The syntrophic growth coexistence P H2 range if the propionate oxidation takes place via the Smithella pathway (P 7 ) is, however, much wider, reaching feasible P H2 values of up to 11,000 Pa. This corresponds to dissolved hydrogen concentrations up to 10 −4 M, well within physiological limits. These results indicate that the IET mechanism for syntrophic propionate oxidation under the stoichiometry from equation 1 is infeasible via dissolved hydrogen and feasible via formate or a direct electron transfer alternative. At the same time, the results indicate that dissolved hydrogen is a feasible IET mechanism if the propionate oxidation takes place via the Smithella stoichiometry and pathway (P 7 ). The syntrophic coexistence niche was also evaluated under the typically observed conditions in methanogenic environments of scenario Met (less favorable for propionate oxidation). The results under these conditions are shown in Fig. 6 . Interestingly, under scenario Met , the conditions are so restricted energetically that the net ATP from all pathways is substantially lower. As opposed to the scenario Opt , shown in Fig. 5 , the syntrophic coexistence of the P H2 range becomes even narrower, and the Smithella pathway (P 7 ) does not even seem to be possible. Propionate oxidation seems possible only via hydroxypropionyl-CoA and syntropy via IET, unlike dissolved hydrogen. The Smithella pathway (P 7 ) never makes any net ATP feasible due to the catabolic energy available being lower than the minimum required for single-proton translocation. These dramatically different results between the conditions of the two scenarios suggest that the actual concentrations encountered under local conditions (e.g., by microorganisms growing within aggregates, such as a granules) must fall between both or have a spatial variability and differ substantially from those measured in the bulk liquid of anaerobic digestion reactors. Conclusions. The automated pathway analysis method developed in this work provides an unprecedented capability to evaluate large numbers of pathway configurations. This allows for the evaluation of any known, and even postulated, biochemistry to theoretically determine the feasible pathways (physiologically and thermodynamically) with the highest ATP yield. The method, entirely mechanistic and largely founded on first principles, brings insights for the study of energy-limited microbial metabolisms. Propionate oxidation was evaluated in the entire domain of possible pathway variations within the known biochemistry and the thermodynamic and physiological feasibility, applied to all reaction steps and all metabolite concentrations. Under a scenario of optimum environmental conditions, the oxidation of propionate via the Smithella pathway yields the most ATP, and the methylmalonyl-CoA pathways can generate sufficient ATP for growth only under a cyclical pathway configuration with pyruvate (P 2 ). Under a scenario of typical methanogenic conditions, the oxidation of propionate via the lactate and via the hydroxypropionyl-CoA pathways appears to yield the most ATP. Extremely low P H2 values (below the minimum reasonable physiological limits) appear to be required to sustain syntrophic growth coexistence with methanogens if propionate is oxidized to acetate and three hydrogens (P 1–6 ), while this is not observed for the Smithella stoichiometry and pathways (P 7 ). This implies that IET via dissolved hydrogen is not feasible under pathways P 1–6 and must occur via alternative mechanisms that could include formate or direct electron transfer (e.g., via conductive pili). Conversely, dissolved hydrogen appears a feasible IET if the propionate oxidation goes via the Smithella pathway (P 7 ). The very different results predicted under most favorable or methanogenic typical conditions suggest that local concentrations or spatial variability via microbial aggregates must be occurring to explain the literature observations for syntrophic propionate oxidation."
} | 7,641 |
36793505 | PMC9912012 | pmc | 7,193 | {
"abstract": "Deterministic and stochastic processes are believed to play a combined role in microbial community assembly, though little is known about the factors determining their relative importance. We investigated the effect of biofilm thickness on community assembly in nitrifying moving bed biofilm reactors using biofilm carriers where maximum biofilm thickness is controlled. We examined the contribution of stochastic and deterministic processes to biofilm assembly in a steady state system using neutral community modelling and community diversity analysis with a null-modelling approach. Our results indicate that the formation of biofilms results in habitat filtration, causing selection for phylogenetically closely related community members, resulting in a substantial enrichment of Nitrospira spp. in the biofilm communities. Stochastic assembly processes were more prevalent in biofilms of 200 µm and thicker, while stronger selection in thinner (50 µm) biofilms could be driven by hydrodynamic and shear forces at the biofilm surface. Thicker biofilms exhibited greater phylogenetic beta-diversity, which may be driven by a variable selection regime caused by variation in environmental conditions between replicate carrier communities, or by drift combined with low migration rates resulting in stochastic historical contingency during community establishment. Our results indicate that assembly processes vary with biofilm thickness, contributing to our understanding of biofilm ecology and potentially paving the way towards strategies for microbial community management in biofilm systems.",
"conclusion": "5 . Conclusion The assembly of MBBR biofilms from planktonic communities involves a combination of stochastic and deterministic processes that vary in their relative importance with thickness. The role of dispersal in community assembly of mature biofilms appears to be small. The migration rate derived from the neutral model suggests that dispersal played a minor role in assembly, and that it did not vary between biofilm thicknesses. We suggest that in established biofilms subjected to shear forces, surface area is likely a more relevant factor than thickness in determining migration rates. Thus, drift is the main stochastic process driving the assembly of biofilms, and it can explain a large proportion of the community composition, particularly for biofilms of 200 µm and thicker. In addition to drift, selection also played a role in biofilm assembly. All biofilms exhibited habitat filtering compared to the suspended source community, with greater prevalence in thinner biofilms. Phylogenetic analysis points towards different selective pressures acting on biofilms of different thicknesses. In thin biofilms, homogeneous conditions throughout the biofilms and between replicates resulted in uniform communities in which closely related community members were selected for. For thicker biofilms, dissimilar replicate communities in which diverse members were present may have been due to variable selection or drift. Functionally, this resulted in thin biofilms with extremely efficient nitrification and thicker biofilms with less efficient nitrification but improved micropollutant degradation [ 27 ]. Better efforts to test neutral community models using well-defined source communities will reveal the conditions under which neutral processes play an important role [ 29 , 54 ], and similarly, using defined source communities in null modelling can provide insights into the strength and type of selective processes acting on communities. Few studies to date have used a defined source community to test the neutral community model, and even fewer have estimated migration rates. The quantification of migration rates is further complicated by the challenge of defining local community sizes and/or effective community sizes. A deeper consideration of what constitutes a microbial community and measurement of migration rates in different types of environments is needed. Further work on biofilm community assembly should aim to describe community assembly processes over time to unify neutral and niche-based models of biofilm succession [ 53 , 54 ] and to determine if steady state biofilm communities exist, or if they are in a constant state of flux.",
"introduction": "1 . Introduction The assembly of microbial communities is driven by a combination of niche-based processes, in which environmental conditions and microbial interactions impose selection, and stochastic processes, including random birth and death events (i.e. drift), dispersal and diversification [ 1 , 2 ]. Both deterministic and stochastic processes can play important roles in community assembly, with their relative importance depending on the strength of selection and the rate of stochastic dispersal. A number of recent studies on engineered microbial environments have described a combined role for stochastic and deterministic processes in community assembly [ 3 – 9 ]. Developing an understanding of the relative importance and contributions of these processes to community assembly under various conditions or designs could inform strategies for engineering and managing microbial communities and to predict patterns of resistance and resilience in the face of disturbance. We posit that deterministically assembled communities will have very different responses to disturbance compared to stochastically assembled communities, so an understanding of the controls on community assembly can improve our ability to make predictions and manage communities. The biofilm lifestyle is believed to be the most prevalent form of microbial life [ 10 ]. Biofilms have important ramifications in a range of fields including medical microbiology, industrial microbiology and environmental engineering. Despite the importance of the biofilm lifestyle, the ecological processes involved in the assembly of biofilms from planktonic communities remain unclear. Engineered systems provide ideal model systems for studying the assembly of complex microbial communities due to the ability to closely control environmental variables and monitor dispersal. Biofilm-based microbial biotechnologies for drinking water and wastewater treatment often achieve similar contaminant removal efficiencies compared to suspended systems, with a lower footprint and less sludge production [ 11 ]. The enhanced microbial retention that occurs in biofilms allows for the growth of slow-growing organisms, which may result in improved functionality [ 12 ]. In wastewater treatment systems, biofilm communities tend to be more biodiverse than suspended ones, which can result in increased functionality and stability [ 13 – 15 ]. Different studies examining community assembly processes in biofilms draw diverging conclusions. For example, the assembly of mature stream biofilms appears to be mainly deterministic [ 16 , 17 ], while the formation of biofilms on microbial electrolysis cells (MECs) in wastewater is dominated by stochastic processes [ 5 , 8 , 18 ]. The factors driving these diverging observations are not well understood, and many studies include additional selective pressures in addition to biofilm formation that can confound the identification of assembly processes involved solely in the process of biofilm development [ 3 , 6 ]. Biofilm thickness varies substantially in natural and engineered systems and is generally thought of as an emergent property, which is determined by hydrodynamic forces such as shear, substrate loading rates and the types of organisms present [ 19 ]. Although the role of biofilm thickness in biological treatment processes has been widely discussed and modelled [ 20 – 22 ], few studies have considered biofilm thickness as a parameter that can be controlled [ 23 – 27 ]. Controlling biofilm thickness could prove useful for managing microbial communities in engineered environments. We therefore sought to examine the relative importance of deterministic and stochastic processes in the assembly of biofilm communities, and to investigate if controlling biofilm thickness influences community assembly processes. Biofilm formation can induce competition for space, which is likely strongest in the thinnest biofilms. With increasing biofilm thickness, nutrient gradients become larger, creating contrasting microenvironments which can support a greater diversity of metabolic lifestyles, but may result in additional selective pressures on the organisms present. A previous study in which biofilms of two thicknesses were compared emphasized the importance of deterministic processes in driving the differences in biofilms of different thicknesses; however, stochastic processes were not explicitly considered [ 26 ]. We hypothesize that there is a substantial role for stochastic processes in biofilm assembly, and that while selection likely occurs in all biofilms, the type of selection may vary. We expect that resource and diffusion limitation will generate greater competition in biofilms than in planktonic communities. Intense competition for space and a homogeneous environment in thin biofilms could result in competitive exclusion, while increased variable selection may be observed in thicker biofilms due to environmental heterogeneity [ 28 ]. We examined established biofilms of five thicknesses ranging from 50 to 500 µm to assess differences in assembly processes between biofilms across a range of diffusion limitation, resource gradients and spatial constraints. We used Sloan's neutral community model to assess the extent of stochastic processes in biofilm assembly from the suspended metacommunity [ 29 ]. We then assessed the extent and type of deterministic processes acting on the communities using a null-modelling approach and identified the organisms driving these effects. Lastly, we considered the calculation of migration rates using the neutral model and considered the implications of migration rates on the importance of dispersal in microbial community assembly.",
"discussion": "4 . Discussion Our results point towards a combined role of stochastic and deterministic processes in the assembly of MBBR biofilm communities, with biofilm thickness driving their relative importance. Importantly, the assembly of a large proportion (approximately half) of the members of biofilms 200 µm and thicker can be explained by stochastic processes, accounting for up to 75% of biofilm relative abundance ( table 1 ), indicating that stochastic processes play a major role in biofilm assembly. In addition, habitat filtering, resulting in, in particular, the enrichment of nitrite-oxidizing Nitrospira spp. relative to the influent was observed in all biofilms ( figure 4 ). Environmental differences between maximum biofilm thicknesses appeared to impact the importance of and the type of deterministic processes involved in community assembly. Thin biofilms were more similar to each other than expected by chance suggesting a highly uniform environment in which phylogenetically similar organisms are selected for, reflecting a homogeneous selection regime [ 28 ]. Conversely, thick biofilms tended towards phylogenetic overdispersion which could be explained by either deterministic or stochastic processes. A variable selection regime due to variations in substrate gradients within and between replicate communities could explain this observation [ 28 ]. On the other hand, given the low migration rates and greater role of stochastic processes in thicker biofilms, stochastic attachment and low migration rates followed by drift could create historical contingency effects resulting in differences between replicate community compositions, as has been suggested for the developing rumen microbiome [ 46 ] and MECs [ 8 ]. Distinguishing between these two scenarios driven by either deterministic or stochastic processes is not possible, as both could cause the observed patterns. The differences in community assembly processes in biofilms of different thicknesses also resulted in functional differences where thin biofilms had high nitrification rates and thicker biofilms were less efficient at nitrification, but showed improved transformation of diverse micropollutants [ 27 ]. Another study comparing both thin (50 µm) and thick (400 µm) biofilm communities in an MBBR for nitrogen removal found that the communities of different thicknesses were distinct and attributed this to deterministic processes [ 26 ]. While our results also show distinct biofilm communities with thickness, our results give a more nuanced view of assembly processes by considering the source community composition, explicitly considering neutral processes and examining biofilm thickness in greater resolution. Both studies agree that biofilm thickness can influence both community composition and function, and we suggest that this is related to different community assembly processes between thin and thick biofilms. The reactors in this study were inoculated and operated with effluent wastewater. The addition of ammonia to the effluent generated a selective pressure for nitrifying bacteria and against heterotrophs, and this manifested in the enrichment of Nitrospira spp., and the maintenance of high numbers of Nitrosomonas spp. in all biofilms. However, other selection pressures were minimized. The wastewater community used was well adapted to the environment provided. The good agreement with the neutral community model suggests that many microbes are well adapted to the biofilm lifestyle, and that biofilm formation in itself is a relatively stochastic process. This is not particularly surprising if we consider that biofilms are the dominant lifestyle of microbes in nature [ 10 ]. Our observations are also in agreement with a previous study in which high levels of stochasticity were observed in the assembly of syntrophic fatty acid degrading biofilms when other selection pressures were weak [ 7 ]. Compared to studies that used the neutral community model to study assembly of biofilm communities, we find a better fit to the model than drinking water biofilter communities on granular activated carbon and sand [ 3 ] and glass beads [ 47 ] but poorer fits than MECs in wastewater [ 5 ]. The studies on drinking water, however, included additional selective regimes that likely contributed to increased importance of deterministic processes in community assembly. In particular, the low- and high-nitrite feeding regimes in [ 47 ] resulted in only slightly poorer fits with the neutral model (Spearman ρ , low nitrite = 0.39, high nitrite = 0.46) compared to the biofilms described here ( ρ = 0.495). Given that different selective regimes were deliberately imposed by varying substrate concentration in that previous work, it is surprising that the thin biofilms described in our study exhibited a much poorer fit with the neutral model ( ρ = 0.295). Thus, limiting biofilm thickness to this extent may impose considerable selective pressure. We propose that hydrodynamic forces and physical shearing at the biofilm surface, combined with competition for space, may provide an explanation for the stronger selection observed in thin biofilms. The surface area to volume ratio of the thin biofilm was between 4- and 10-fold higher than that of the thicker biofilms. Previous work has shown the importance of hydrodynamic forces as a selective parameter in microbial communities [ 48 ]. Based on previous modelling efforts, it is believed that 50 µm and thinner biofilms would not have substantial substrate gradients [ 26 ], and thus the environment would be homogeneous with depth. If hydrodynamic forces were strong at the surface of the biofilms, it is likely that only the fastest growing taxa that were capable of persisting despite hydrodynamic shear and abrasion, or alternatively were able to penetrate deeper biofilm layers, would persist in the thinnest biofilms. 4.1 . Migration to the biofilms is low The rate of migration of members from the source community to the biofilm can be determined using the neutral model if the size of the target community ( N T ) is known. We considered each carrier as a single microbial community and quantified community size using 16S rRNA gene-based qPCR. We used the method described by Sloan et al . [ 44 ] to calculate the true migration rate of a community when a subsample of the community (of size N s ) is used to calibrate the model. As determined by this method, m , the probability that an empty space in the community will be filled by an organism from the source community, was low for all biofilms, ranging from 7.38 × 10 −8 to 1.17 × 10 −7 ( figure 1 f ). Extremely low m values have also been noted by other studies that used realistic values of N T [ 9 ]. However, few studies have calculated migration rates predicted by the neutral model, and those that have often ignore sampling effects, and use the collected sample size as N T , rather than the true size of the target community. When using the sample size (37 378 sequences after rarefaction) in order to compare to other studies, the migration rates in this study varied from 0.0033 to 0.0067. By comparison, migration rates estimated for other environments tend to be much higher, including the Humber estuary (0.7 for both AOB and denitrifiers), the respiratory tract (0.2) and a sewage treatment plant (0.1) [ 29 ]. Migration rates in the human gut were somewhat lower, but still an order of magnitude greater than in this study (0.032–0.063) [ 49 ]. However, these differences in reported migration rates appear to be inversely related to the sample size with samples as small as 13 clones [ 29 ]. Indeed, in another study of neutral assembly of microbial communities in tree holes, N T was estimated based on tree hole volume and microbial density, and the migration rate was several orders of magnitude lower, estimated at 1 × 10 −6 [ 50 ]. Sloan et al . [ 44 ] recalculated m for several environments taking into account the sampling effect. Corrected m values for the Humber estuary and sewage samples are 7.0 × 10 −8 and 1.55 × 10 −9 , respectively, given an N T of 10 9 cells, even lower than for the biofilm communities in the present study. The migration rates in all of these studies seem low intuitively, but since migration rates in microbial communities have not been measured experimentally, it is not possible to comment on whether this is an accurate estimate. Accurate estimations of N T and m could enhance our ability to understand community assembly processes leading to predictive microbial community management. The effective community size ( N ec ) is a relatively new concept that could be used to reflect ecologically relevant microbial community sizes to use as N T [ 51 ]. This method may also provide more realistic estimates of m going forward. Mature biofilms on carriers with a maximum biofilm thickness would be difficult to colonize once established, as migrating cells would quickly be sloughed off due to hydrodynamic and physical forces acting at the biofilm surface. Had we sampled at earlier timepoints during biofilm establishment, before steady state, we may have observed higher migration rates when these forces were weaker. Supporting this idea, a study examining biofilm succession from initial establishment found that bacterial community assembly was driven primarily by stochastic processes, but that a high level of replacement occurred as biofilms developed [ 52 ]. In the mature biofilms that we sampled, we never observed a pattern between maximum thickness and migration rate. This indicates that biofilm thickness did not influence the migration rate, and we suggest that surface area is likely to be a more important control on migration in biofilms experiencing high shear forces. 4.2 . Steady state communities were not observed The high variability in community structure between biofilms of the same thickness reveals that substantial changes in community structure occurred over time ( figure 3 ; electronic supplementary material, figure S2) even though biofilms were only sampled once nitrogen removal had reached steady state [ 27 ]. Since the community had reached a functional steady state, we expected that the community composition would be similarly consistent; however, this was not the case. Replicate communities were always similar to each other, but composition varied over time. This is not believed to be due to variation in the influent community (wastewater treatment plant effluent) as the influent microbial community composition was observed to be highly consistent even between collected batches (electronic supplementary material, figure S2). Given the low migration rates, it is not expected that the influent composition would have had a major impact on biofilm composition once they were established. Thus, the observed variations in biofilm composition are likely due to drift or selection. Variation of biofilm communities over time is predicted by models of biofilm succession based on both neutral and niche perspectives [ 53 , 54 ]. However, these models focus mainly on the initial establishment of biofilms and not on changes occurring after the initial colonization and development stages. One study that examined the establishment of MBRs over 30 days also observed temporal variation in communities and in assembly processes; however, given the short timeframe of that study, early biofilm succession may still have been occurring in the communities [ 6 ]. Further work examining long-term biofilm succession is needed to define the relative importance of drift, migration and various deterministic processes over time. A drawback of the experimental design of this study is that the thinnest biofilms (50 µm) were in a different reactor than the thicker biofilms, started 45 days later. It is unlikely that the observed differences can be explained entirely by the existence of two reactors as general trends in reactor performance showed consistently increasing nitrification rate and decreasing micropollutant removal with thickness. The feed and thus metacommunity entering the reactors were identical (and were stable over time), as were the hydraulic-loading characteristics. The experimental design of the reactors was such that the HRT was short (2 h), thereby minimizing homogenizing dispersal between carriers. Additional insights could be gained by repeating this experiment with all carrier thicknesses in a single reactor, and all carrier thicknesses in individual reactors."
} | 5,640 |
36879809 | PMC9985049 | pmc | 7,194 | {
"abstract": "Summary Epoxy resin is widely used in various fields of the national economy due to its excellent chemical and mechanical properties. Lignin is mainly derived from lignocelluloses as one of the most abundant renewable bioresources. Due to the diversity of lignin sources and the complexity as well as heterogeneity of its structure, the value of lignin has not been fully realized. Herein, we report the utilization of industrial alkali lignin for the preparation of low-carbon and environmentally friendly bio-based epoxy thermosetting materials. Specifically, epoxidized lignin with substituted petroleum-based chemical bisphenol A diglycidyl ether (BADGE) in various proportions was cross-linked to fabricate thermosetting epoxies. The cured thermosetting resin revealed enhanced tensile strength (4.6 MPa) and elongation (315.5%) in comparison with the common BADGE polymers. Overall, this work provides a practicable approach for lignin valorization toward tailored sustainable bioplastics in the context of a circular bioeconomy.",
"introduction": "Introduction Nowadays, plastic provides great convenience and fast services for daily life while consuming large amounts of petrochemical resources. 1 , 2 , 3 To ensure the sustainable utilization of the resources, the development of renewable materials has attracted increasing interest. Lignin, the most abundant resource with renewable aromatic structures, is a typical heterogeneous biopolymer with a complex chemical structure and broad molecular weight distribution. 4 , 5 , 6 The random polymerization of three monomers during biosynthesis is the main reason for the heterogeneity of lignin. 7 , 8 Plants on Earth synthesize 150 billion tons of lignin each year. As a result, lignin is deservedly the most abundant renewable bioresources. Among them, the pulp and paper industry produced about 150–180 million tons of industrial lignin annually, but only 2% are commercially used, 9 , 10 such as energy, 11 chemicals, 12 , 13 , 14 polymers, 15 and carbon materials. 16 There are mainly two ways for lignin valorization. One strategy was to deconstruct lignin into phenolic compounds with low molecular weights by the strategies of catalytic and pyrolytic approaches. 17 , 18 , 19 , 20 , 21 Another alternative route is the crosslinking or blending of lignin with other monomers or polymers to produce thermoset resins, 22 , 23 adhesives, 24 , 25 foams, 26 and thermoplastics. 27 , 28 Currently, lignin has been developed to participate in the manufacture of various materials, such as composites, 29 , 30 hydrogels, 31 , 32 thermosets, and thermoplastic material. 33 Among them, the preparation of thermoset materials is one of the most promising uses. Epoxy resins are used in various industries with excellent chemical and mechanical properties. 34 , 35 , 36 The significance of exploring bio-based polymers is to replace existing petroleum-based polymers, more importantly, to provide significant performance advantages over existing products. According to previous reports, lignin-based thermosetting materials include phenol/formaldehyde resins, 37 , 38 polyurethanes, 39 epoxy, 22 , 40 and thiol-vinyl resins. 41 Epoxy resins are primarily made from petrochemicals and used in adhesives, coatings, and composite materials due to their versatile features. 42 Diglycidyl ether bisphenol A (DGEBA) is a commonly used epoxy resin, which is cured under different conditions by adding hardeners to form epoxy resins. 43 Bisphenol A, the main raw material of DGEBA epoxy resin, 44 has been reported to cause endocrine disorders that threaten the health of fetuses and children. Obesity due to metabolic disorders and cancer are also thought to be linked. 45 Therefore, it is urgent to use renewable aromatic sources to replace BADGE in the preparation of epoxy resins. Industrial lignin, with high aromatic content and low molecular weight dispersion index, is suitable for replacing BADGE in epoxy resin. There are three means to mix lignin with epoxy resins: 1) mixing with petroleum-based epoxy resin, 46 , 47 2) epoxidation of lignin, 48 , 49 and 3) epoxidation of lignin after modification. 22 , 48 , 50 , 51 , 52 , 53 In order to prevent bisphenol A from affecting humans and the environment, it is essential not only to ban its use in materials, such as food packaging, 43 , 54 but also to find alternative, renewable, and sustainable raw materials to replace BADGE in epoxy formulations. Replacing BADGE with lignin still remains challenges due to its high molecular weight, different types of hydroxyl groups, and low solubility in organic solvents and water. 22 Herein, we report a route to fabricate lignin-based thermosetting epoxy resins with excellent mechanical strength by solvent fractionation and lignin modification. Firstly, industrial lignin was purified by acetone separation, and then ethylene oxide structure was introduced into the lignin framework to improve the reactive sites of lignin. Subsequently, the obtained lignin fraction was applied as a substitute for petroleum-based chemical BADGE to fabricate thermosetting resin material. Notably, the acetone-soluble lignin sample revealed better processability as compared to the pristine lignin due to its excellent homogeneity and low molecular weight. Accordingly, the objective of this work was to address the petroleum consumption and environmental issues by replacing the petroleum-based component of epoxy resins with lignin.",
"discussion": "Discussion In conclusion, this work prepared a lignin-based thermosetting epoxy resin with better mechanical strength than commercial BADGE. First, the molecular weight, polydispersity, and structural heterogeneity of lignin were significantly reduced by acetone fractionation. Subsequently, a large amount of propylene oxide structure was introduced into the lignin framework to improve the reactivity. Finally, lignin was used as a substitute for petroleum-based chemical BADGE to prepare epoxy thermosetting resin materials. This polymer material was fabricated by mixing lignin and BADGE in different proportions and crosslinking as well as curing with a flexible polyether diamine (M w 400 g/mol). It was found that the elongation at break and tensile strength of lignin-based epoxy resin were improved by 95.8% and 50.6% as compared with commercial bisphenol A with the addition of 5% lignin content. This work bridges lignin and traditional thermosetting resins, affording a sustainable strategy that prepares excellent epoxy resins benefit from the advantages of the addition of lignin, and will vastly broaden the application of lignin-based epoxy resin. Limitations of the study The main content of our work is the fabrication of thermosetting epoxy resin from industrial lignin. However, the amount of lignin substitute for bisphenol A diglycidyl ether is limited. We should further improve the interface compatibility of lignin and increase the amount of lignin to reduce the cost. Such tasks are currently ongoing in our laboratory."
} | 1,754 |
21556065 | PMC3130554 | pmc | 7,197 | {
"abstract": "Many compounds being considered as candidates for advanced biofuels are toxic to microorganisms. This introduces an undesirable trade-off when engineering metabolic pathways for biofuel production because the engineered microbes must balance production against survival. Cellular export systems, such as efflux pumps, provide a direct mechanism for reducing biofuel toxicity. To identify novel biofuel pumps, we used bioinformatics to generate a list of all efflux pumps from sequenced bacterial genomes and prioritized a subset of targets for cloning. The resulting library of 43 pumps was heterologously expressed in Escherichia coli , where we tested it against seven representative biofuels. By using a competitive growth assay, we efficiently distinguished pumps that improved survival. For two of the fuels ( n -butanol and isopentanol), none of the pumps improved tolerance. For all other fuels, we identified pumps that restored growth in the presence of biofuel. We then tested a beneficial pump directly in a production strain and demonstrated that it improved biofuel yields. Our findings introduce new tools for engineering production strains and utilize the increasingly large database of sequenced genomes.",
"introduction": "Introduction Existing biofuels, such as ethanol and esters of linear fatty acids, can be used in only limited quantities in current gasoline, diesel and jet engines. Recently, there have been several reports of efforts to engineer microorganisms to produce advanced biofuels that can be used in larger quantities in our existing transportation infrastructure. Short to medium chain (C 4 –C 12 ) alcohols such as butanol, isopentanol and geraniol are superior to ethanol as gasoline replacements ( Peralta-Yahya and Keasling, 2010 ). Longer, branched-chain compounds (C 9 –C 23 ) such as geranyl acetate and farnesyl hexanoate are better biodiesel alternatives than linear esters because branching reduces their freezing point. In addition, cyclic alkenes such as limonene and pinene serve as precursors to jet fuel ( Harvey et al, 2010 ; Peralta-Yahya and Keasling, 2010 ). While microbial biosynthetic routes to most of these compounds exist, the biofuels have known antimicrobial activity ( Fischer et al, 2008 ). For the production to be cost effective, yields must exceed native tolerance levels, necessitating the development of stress-tolerant strains ( Supplementary Text ). Product toxicity is a common problem in strain engineering for biotechnology applications. Work on ethanol production has shown that alleviating toxicity is necessary for maintaining and maximizing yield ( Alper et al, 2006 ; Jarboe et al, 2007 ). Thus, it is crucial that we improve tolerance in parallel with the development of metabolic pathways for the production of next-generation biofuels. Microbes have several strategies for addressing biofuel toxicity ( Isken and de Bont, 1998 ; Ramos et al, 2002 ). Here we focus on efflux pumps, a class of membrane transporters that export toxins from the cell using the proton motive force ( Putman et al, 2000 ; Nikaido and Takatsuka, 2009 ). Efflux pumps in Gram-negative bacteria are composed of three proteins: an inner membrane protein responsible for substrate recognition and proton exchange, a periplasmic linker and an outer membrane channel. All subunits are essential for function and the corresponding genes are commonly arranged together in an operon. Relatively few solvent-resistant efflux pumps have been previously characterized. Examples include ttgABC, ttgDEF, ttgGHI ( Rojas et al, 2001 ) and srpABC ( Kieboom et al, 1998a ) from Pseudomonas putida . These pumps appear to be specific to solvents ( Kieboom et al, 1998b ; Isken and de Bont, 2000 ), while more general multidrug efflux pumps such as acrAB-tolC from E. coli ( Nikaido and Takatsuka, 2009 ) export a broad range of substrates, including solvents. All known solvent-resistant efflux pumps in Gram-negative bacteria fall into the hydrophobe/amphiphile efflux (HAE1) family of resistance-nodulation-division pumps ( Tseng et al, 1999 ). Sequenced bacterial genomes include many efflux pumps and present a largely unexplored resource for discovering novel pumps with potential for use in engineering fuel tolerance. Here we take a systematic approach to screen a library of primarily uncharacterized heterologous pumps for engineering biofuel-tolerant host strains. We then demonstrate that expression of a heterologous pump can increase the yield of a biofuel production strain.",
"discussion": "Results and discussion Using E. coli as our engineering host, we asked whether heterologously expressed efflux pumps could reduce toxicity by exporting biofuel from the cell. We constructed a database of all HAE1 pumps from sequenced bacterial genomes (Materials and methods). Using this set, we performed a bioinformatics screen to compare regions that are predicted to be responsible for substrate specificity to those of TtgB, a well-characterized solvent-resistant pump. This metric allowed us to rank the complete set of pumps and select a subset that represented a uniform distribution of candidates ( Supplementary Figure S1 , Supplementary Methods ). To construct the library, efflux pump operons were amplified from the genomic DNA of the selected bacteria, cloned into a vector, and transformed into an E. coli host strain (Materials and methods). In total, our library contains 43 efflux pumps, most of which have not been previously characterized for biofuel or solvent tolerance. Although tolerance and export of intracellularly produced biofuel is the ultimate goal, we hypothesized that testing for tolerance to exogenous biofuels would identify pumps with the potential to export biofuel from the cell. Similar strategies have been used previously to improve production. For example, mutations in Saccharomyces cerevisiae that improved ethanol tolerance led to an increase in production ( Alper et al, 2006 ). In addition, an evolved isobutanol-tolerant strain of E. coli improved growth and production when grown under isobutanol stress ( Atsumi et al, 2010 ). It should be noted that yields from production strains can exceed the inhibitory concentrations of exogenous biofuels. For example, isobutanol inhibits E. coli growth at 8 g/l, but strains continue to produce up to 20 g/l in stationary phase after growth stops ( Atsumi et al, 2008 ). In order to efficiently screen the efflux pumps against biofuel candidates, we devised a competition-based strategy to select for pumps that improved biofuel tolerance ( Figure 1A ). When a survival or fitness phenotype can be used, competitive growth experiments provide an effective selection strategy ( Lynch et al, 2007 ; Ho et al, 2009 ). Efflux pump expression strains were grown individually and then pooled, so that all strains were represented in equal proportion. This pooled culture was then grown both with and without biofuel and maintained through serial dilutions every 10–14 h. At each dilution time point, plasmids from the culture were isolated and a custom microarray was used to quantify the amount of each efflux pump plasmid remaining in the culture (Materials and methods). A mathematical model of competitive growth was used to guide experimental design ( Figure 1B ). Because strains expressing pumps that help to mitigate biofuel toxicity will have a growth advantage, these strains will dominate the co-cultures after only a small number of dilution cycles ( Supplementary Figure S2 ). In order to experimentally validate our predictions, we first asked if the composition of the competing cultures changed over time. When the pooled culture was grown without any biofuel, all pumps were represented equally, indicating that no strain had a particular advantage ( Figure 2A ). This remained true over the course of the 96-h experiment, showing that under these induction conditions, any burden of pump expression was roughly equivalent for all strains. In contrast, when the pooled culture was grown in the presence of an inhibitory biofuel such as geranyl acetate, some efflux pumps conferred a distinct advantage ( Figure 2B ). Although all strains started out with equal representation, after 38 h the population composition changed, with cells containing the advantageous pumps becoming an increasingly large proportion of the population. The efflux pumps that enhanced tolerance to geranyl acetate originated from a variety of hosts and include both known and previously uncharacterized pumps. Next, we asked whether competing the cultures in the presence of different biofuels would identify unique sets of resistant efflux pumps for each biofuel. With respect to the compounds tested, our results fall broadly into two classes: (i) biofuels that are toxic, but where pumps do not reduce toxicity and (ii) biofuels where the pumps do reduce toxicity. Both n -butanol and isopentanol fall into the first class of fuels ( Figure 3A and B ). This could be because none of the pumps in the library export these compounds or, alternatively, the rate of export may not be sufficient to counteract intracellular accumulation ( Lim and Nikaido, 2010 ). Pumps improved tolerance for the second class of fuels. We saw a distinct separation between the competition winners and those pumps that were not beneficial ( Figure 3C ). The competition survivors and their relative abundances in the culture were biofuel-specific. To control for the possibility that pumps were mutated over the course of the competition assay and to further characterize their individual performance, we retransformed sequenced pump plasmids into the E. coli host strain and retested for tolerance improvements after 14 h of growth. For each biofuel, we tested strains expressing the top-performing pumps individually and compared their survival relative to a pump-free control. All competition winners outperformed the control strain ( Figure 3D ). Furthermore, comprehensive tests with one fuel verified that all competition survivors had increased tolerance relative to the control ( Supplementary Figure S3 ) and outperformed non-winning strains ( Supplementary Figure S4 ). The pumps that survived the competition are ideal candidates for future biofuel engineering efforts. Next, we asked whether pumps that improved tolerance also enhanced biofuel production. Two pumps consistently survived the limonene competition: the native E. coli pump AcrAB and a previously uncharacterized pump from Alcanivorax borkumensis . We focused on the A. borkumensis pump and tested it in a limonene production strain. Notably, strains expressing the pump produced significantly more limonene than those with no pump ( Figure 3E ). These results provide an important proof-of-principle demonstration that efflux pumps that increase tolerance to exogenous biofuel can improve the yield of a production host. It should be noted that current limonene production levels are not yet toxic to cells. Irrespective of the toxicity, an effective export pump may serve to relieve end product inhibition of metabolic pathway enzymes and result in an improvement in production. As yields improve, export pumps may play an increasingly important role. Beneficial efflux pumps can come from a variety of unrelated bacteria. Our results identified marine microbes such as A. borkumensis , Marinobacter aqueolei and Pseudoalteromonas haloplanktis to be valuable sources of biofuel-tolerant efflux pumps. Pumps from these organisms are similar (40–69% sequence homology) to several known solvent-tolerant pumps, but it is not clear whether their substrate specificity is narrow (similar to TtgB) or broad (similar to AcrB) ( Supplementary Table S2 ). Interestingly, several of the biofuel-tolerant pumps have only the inner membrane and periplasmic proteins present in their operons (Materials and methods), meaning that they must successfully recruit a native E. coli outer membrane protein for export. As pumps can work in concert to export substrates ( Lee et al, 2000 ; Segura et al, 2003 ), multi-pump constructs may produce further increases in tolerance. Additional methods for improvement, such as codon optimization or directed evolution, can be used to fine-tune these pumps for a specific target. This opportunity for optimization is underscored by the fact that the native E. coli pump AcrB appeared as a competition winner for many of the biofuels assayed. It is possible that increases in expression by sequence optimization may improve non-native pumps. Furthermore, for biofuel production from renewable materials, toxic by-products of lignocellulosic biomass pretreatment comprise another factor that affects biofuel yields ( Jarboe et al, 2007 ; Pienkos and Zhang, 2009 ). Pumps may also provide a potential mechanism for excluding these inhibitors. As metabolic engineering efforts continue to increase biofuel production titers, it will be crucial to develop strategies for increasing tolerance. This is especially important for production of bulk commodities such as biofuels, where relatively high titers of an inhibitory compound are required, as was found to be the case in industrial production of 1,4-butanediol and 1,3-propanediol in E. coli ( Burk, 2010 ; Zeng and Biebl, 2010 ). Efflux pumps show great promise as biofuel transporters and are a valuable tool for engineering production strains. Our strategy of bioprospecting for heterologous tolerance mechanisms also provides a widely applicable method for the rapidly advancing field of biofuel research."
} | 3,393 |
27774985 | PMC5079060 | pmc | 7,198 | {
"abstract": "The subterranean world hosts up to one-fifth of all biomass, including microbial communities that drive transformations central to Earth's biogeochemical cycles. However, little is known about how complex microbial communities in such environments are structured, and how inter-organism interactions shape ecosystem function. Here we apply terabase-scale cultivation-independent metagenomics to aquifer sediments and groundwater, and reconstruct 2,540 draft-quality, near-complete and complete strain-resolved genomes that represent the majority of known bacterial phyla as well as 47 newly discovered phylum-level lineages. Metabolic analyses spanning this vast phylogenetic diversity and representing up to 36% of organisms detected in the system are used to document the distribution of pathways in coexisting organisms. Consistent with prior findings indicating metabolic handoffs in simple consortia, we find that few organisms within the community can conduct multiple sequential redox transformations. As environmental conditions change, different assemblages of organisms are selected for, altering linkages among the major biogeochemical cycles.",
"discussion": "Discussion Microbial communities across various environments have been documented to contain thousands of different species, most of which occur at low abundance, and thus are members of the ‘rare biosphere' 20 . Because rare organisms are difficult to characterize genomically, the overall functioning of microbial communities has remained largely unknown. In this study, we demonstrate the ability to genomically describe thousands of microorganisms from a single ecosystem and bring to light aspects of the microbial community metabolic network. In addition, we defined the metabolic capacities of 1,297 organisms represented by 2,540 genomes. We show that metabolic plasticity involving the use of multiple electron donors and acceptors appears to be extremely common in microorganisms in the studied terrestrial subsurface system. A wide metabolic repertoire is likely to be important in the face of the natural environmental perturbations that occur at this site, such as seasonal snowmelt-induced fluctuations in the water table that move the oxic/anoxic interface. In spite of redox metabolic plasticity, we found that the majority of organisms probably lack the ability to perform multiple sequential redox transformations within a pathway. This result expands on prior research that has described syntrophic interactions 4 40 41 . Thus, it appears that organisms often work in cohorts to turn biogeochemical cycles. Further, the organisms that mediate individual reaction steps display a multitude of combinations of metabolic traits, and different organisms proliferate as conditions change ( Fig. 3 ; Supplementary Data 1 and 9 ). Thus, selection for different organisms to carry out specific steps in redox pathways has the potential to change the ways in which biogeochemical cycles are cross-linked. Metabolic handoffs to a wide variety of potential recipients, in combination with the potential for cycles within cycles, provide very high levels of complexity and flexibility. This modular ‘plug and play' strategy enables an enormous variety of system configurations and likely confers ecosystem resilience in the face of perturbation. Recognition of the importance of metabolic handoffs motivates new thinking about how biogeochemical processes should be modelled. Specifically, based on genomic information, individual reaction steps should be explicitly assigned to different organisms. Although this will increase model complexity and require detailed consideration of fluxes, such modifications will be essential to capture effects that can arise from metabolic handoffs, such as ‘leakage' of reaction intermediates following perturbations ( Fig. 6 ). Leakage is likely when ecosystem discordance arises from lags in activation of microbial community members responsible for sequential steps in a biogeochemical cycle. This is analogous to the uncoupling that occurs when climate warming causes early flowering that is out of sync with insect hatching, leading to pollination failures 42 . Such phenomena are little known in microbial ecosystems, but could give rise to large fluxes of climate-relevant intermediate compounds. Examples include pulses of N 2 O following influx of ammonium-rich water 43 or decrease in oxygen availability 44 . Another important finding, from the perspective of development of both conceptual and quantitative models of biogeochemical processes, is the possibility of ‘cycles within cycles'. These could short-circuit the elemental cycles as they are traditionally conceived 45 (for example, where the most reduced form, for example, S 2− , N 3− , is presumed to be converted to the most oxidized form, S 6+ , N 5+ and vice versa). For example, we conclude that the inter-conversion of elemental sulfur and sulfide may be a prominent cycle within the larger sulfur cycle in this system. A similar phenomenon could also occur in the nitrogen cycle, when nitrate is reduced to nitrite by bacteria that have no further capacity for denitrification 46 , resulting in a substrate that could be oxidized back to nitrate by nitrite oxidizers. We observed no correlation between the number or relative abundance of organisms mediating a particular step of a pathway and the total energy yields associated with that step ( Supplementary Data 11 ). This would suggest that thermodynamic considerations alone do not control selection for the set of pathway steps that occur in organisms. The trait distribution data ( Fig. 4 ) highlight an example of where a cycle occurs within a larger cycle: the oxidation of sulfide to elemental sulfur, which can be converted back to sulfide rather than oxidized to sulfite and sulfate. The direct oxidation of sulfide (S 2− ) to elemental sulfur (S 0 ) is mediated by two different enzymes, sulfide:quinone oxidoreductase ( sqr ) 47 and flavocytochrome c sulfide dehydrogenase ( fcc ) 48 , which were present in 11% (groundwater) and 27% (sediment) of the recovered genomes. Elemental sulfur may also be produced as a byproduct of thiosulfate disproportionation by the sox enzyme system if soxCD are lacking 49 . Significantly, genes for elemental sulfur reduction were present in 17% (groundwater) and 22% (sediment) of the genomes, whereas the capacity for elemental sulfur oxidation was present in only 4% (groundwater) and 13% (sediment) of the genomes. The tremendous novelty of microorganisms observed in the aquifer ecosystem highlights the potential for biological discovery in the terrestrial subsurface. Given the novel phylogenetic diversity of the studied organisms, the genomes reported here represent a vast treasure-trove that could be mined for biotechnological applications and for potential strategies for genome-enabled cultivation of novel organisms. The findings relating to metabolic network topology will guide future in silico studies of inter-organism metabolic networks 50 , and may have application in trait-based ecosystem models that are needed to predict the impacts of changing environmental conditions on biogeochemical cycles 51 ."
} | 1,802 |
35559674 | PMC9106289 | pmc | 7,203 | {
"abstract": "Synthetic composite materials constructed by hybridizing multiple components are typically unsustainable due to inadequate recyclability and incomplete degradation. In contrast, biological materials like silk and bamboo assemble pure polymeric components into sophisticated multiscale architectures, achieving both excellent performance and full degradability. Learning from these natural examples of bio-based “single-component” composites will stimulate the development of sustainable materials. Here, we report a single-component “Silk nacre,” where nacre’s typical “brick-and-mortar” structure has been replicated with silk fibroin only and by a facile procedure combining bidirectional freezing, water vapor annealing, and densification. The biomimetic design endows the Silk nacre with mechanical properties superior to those of homogeneous silk material, as well as to many frequently used polymers. In addition, the Silk nacre shows controllable plasticity and complete biodegradability, representing an alternative substitute to conventional composite materials.",
"introduction": "INTRODUCTION High-performance composite materials are in great demand in many fields, such as building construction, automobile manufacturing, aircraft technology, and biomedical engineering ( 1 – 5 ). To this end, the strategy of hybridizing multiple components, including metals, ceramics, and polymers, is generally used. Despite the progress achieved in improved performance for such composite materials, concerns over their sustainability increase rapidly ( 6 ), due to the difficulty of removing individual components for recycling and incomplete degradation. In particular, environmental issues associated with the pollution of waste plastics and synthetic composites have aroused increasing attention worldwide ( 7 – 9 ). Therefore, constructing more high-performance and sustainable composites from bio-based polymers becomes increasingly vital, where many natural polysaccharides (cellulose, chitin, and starch) or proteins (silk, collagen, and keratin) can serve as promising building blocks ( 1 , 6 ). Biological materials usually achieve excellent performance by building sophisticated multiscale architectures, despite their limited selection over components ( 10 – 16 ). This strategy is particularly evident in pure polymeric materials like silk and bamboo, which are also categorized as “single-component” composite in the sense that they have composite architectures. For example, with pure biopolymer, silk achieves both high strength and toughness by building nanosized crystalline phase embedded in an amorphous matrix ( 17 ). The same strategy applies to bamboo, where biopolymer matrix is reinforced by hierarchically aligned cellulose fibrils ( 11 ). Learning from these natural examples will stimulate the development of high-performance and sustainable composite materials. This has become increasingly important as typical synthetic composites are difficult to recycle such as glass fiber–reinforced composites used in wind power blades and tires reinforced with functional nanofillers. Hence, the natural strategy of building high-performance materials by efficient integration of single components into composite architectures represents a more sustainable solution than the typical strategy of hybridizing multiple components. However, the achievement of using this bioinspired strategy is still far from satisfaction in terms of material performances, mainly due to our limited capability of mimicking the sophisticated architectures of biological materials. Here, as a proof of concept, a silk-based nacre-like composite material, denoted as “Silk nacre,” is constructed by using one single component (silk fibroin) as the building block while mimicking natural nacre’s composite structure. Generally speaking, nacre is formed by living animals (mostly mollusks) in the natural environment and presents as a composite material, which is reported to consist of about 95 weight % (wt %) aragonite (CaCO 3 ) and ~5 wt % biopolymers ( 2 , 15 ). The layered aragonite platelets are bonded by a thin layer of biopolymer, forming a brick-and-mortar–like architecture. Nacre’s hierarchical structures and the strong interfacial interactions between the inorganic “bricks” and the organic “mortar” endow it with remarkable mechanical properties, combining both high strength and toughness. In contrast to common organic-inorganic nacre-like composites, the bricks and the mortar in the Silk nacre derive from the same component, which are compactly integrated without any other glue. The composite structures of the Silk nacre endow it with mechanical properties superior to those of a homogeneous silk material, with its bending strength, modulus, and strain significantly improved by 67, 37, and 19%, respectively. Moreover, the Silk nacre shows controllable plasticity, which is usually difficult to achieve regarding strong and bulk nacre-like composites. In addition, the Silk nacre is completely biodegradable with the enzyme pronase E at 37°C, indicating its favorable environmental friendliness. We believe that this work paves an effective way to develop high-performance and sustainable materials with a simple component and a facile manufacturing procedure.",
"discussion": "DISCUSSION In summary, we propose and successfully demonstrate a bioinspired strategy to construct high-performance and sustainable composite materials with single polymeric component. Despite its simple composition (pure silk fibroin), our Silk nacre achieves excellent mechanical properties by mimicking nacre’s brick-and-mortar structure. Together with its controllable plasticity and complete biodegradability, our Silk nacre suggests a great opportunity to partly substitute conventional composite materials that are increasingly suffering from difficult recycling and incomplete degradation. This work will not only provide an insightful perspective for better understanding the sophisticated multiscale architectures of biological materials but also paves an effective way for the development of high-performance and sustainable composites with a simple component and a facile manufacturing procedure."
} | 1,542 |
28592993 | PMC5460468 | pmc | 7,204 | {
"abstract": "Background Feedstock cost is a substantial barrier to the commercialization of lignocellulosic biorefineries. Poplar grown using a short rotation coppice (SRC) system has the potential to provide a low-cost feedstock and economically viable sugar yields for fuels and chemicals production. In the coppice management regime, poplars are harvested after 2 years’ growth to develop the root system and establish the trees. The biomass from these 2-year-old trees is very heterogeneous, and includes components of leaf, bark, branch, and wood chip. This material is quite different than the samples that have been used in most poplar bioconversion research, which come from mature trees of short rotation forestry (SRF) plantations. If the coppice management regime is to be used, it is important that feedstock growers maximize their revenue from this initial harvest, but the heterogeneous nature of the biomass may be challenging for bioconversion. This work evaluates bioconversion of 2-year-old poplar coppice and compares its performance to whitewood chips from 12-year-old poplar. Results The 2-year-old whole tree coppice (WTC) is comprised of 37% leaf, 9% bark, 12% branch, and 42% wood chip. As expected, the chemical compositions of each component were markedly different. The leaf has a low sugar content but is high in phenolics, ash, and extractives. By removing the leaves, the sugar content of the biomass increased significantly, while the phenolic, ash, and extractives contents decreased. Leaf removal improved monomeric sugar yield by 147 kg/tonne of biomass following steam pretreatment and enzymatic hydrolysis. Bioconversion of the no-leaf coppice (NLC) achieved a 67% overall sugar recovery, showing no significant difference to mature whitewood from forestry plantation (WWF, 71%). The overall sugar yield of NLC was 135 kg/tonne less than that of WWF, due to the low inherent sugar content in original biomass. An economic analysis shows the minimum ethanol selling price required to cover the operating cost of NLC bioconversion was $1.69/gallon. Conclusions Leaf removal resulted in significant improvement in overall monomeric sugar production from SRC biomass. Leaf removal is essential to achieve good yields in bioconversion of poplar. Economic analysis suggests the NLC could be a reasonable feedstock provided it can be obtained at a discounted price. Electronic supplementary material The online version of this article (doi:10.1186/s13068-017-0829-6) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions Growing poplar using a SRC system requires the trees to be harvested after 2-year growth such that the root system is well established. Bioconversion of WTC composed of (whitewood) chip, bark, branch, and leaf from this initial harvest were evaluated and compared to bioconversion of WWF. It was found that leafy material makes up more than a thrid of WTC but has a low sugar content, and high phenolic, ash, and extractives contents. Leaf removal significantly changed the chemical composition of coppice sample, and improved monomeric sugar yield by 147 kg/tonne after steam pretreatment and enzymatic hydrolysis. With the presence of bark and branch, the NLC achieved 67% monomeric sugar recovery, showing no significant difference compared to that of WWF (71%). The overall sugar yield of NLC was 135 kg/tonne less than that of WWF, however, due to the low inherent sugar content of the NLC. These findings demonstrate that it is essential to remove the leaves prior to pretreatment to ensure a better overall sugar yield. A minimum ethanol selling price to cover operating expenses of $1.69/gallon was established from the economic analysis, assuming the NLC feedstock is available for its fuel value. The growth and maturity of the wood are limited in the first cycle. We would anticipate that the following rotations will have more wood chips, less bark/branch (the second cycle is showing 60% whitewood content in recent harvest), and therefore a higher overall sugar content. In addition, several practices could potentially enhance the compositional characteristics of short rotation poplar and fortify the sugar content, such as choosing higher sugar composition hybrid/clone, modifying the management strategy, and/or mixing with other sugar-rich feedstocks.",
"discussion": "Results and discussion Physical and chemical compositions of SRC biomass As Table 2 shows, WTC consisted of 37% leaf, 9% bark, 12% branch, and 42% chip. Chips comprised less than half of the dry WTC, while leaves comprised over one-third of the total dry biomass. The relatively small chip content was mainly due to the tree age and management practice [ 36 ]. In a conventional poplar plantation, leaf, bark, and branch components are considered low value and may be discarded after harvest. Table 2 Physical and chemical compositions of four components in SRC poplar and SRF poplar Physical composition (%) Chemical composition (%) Arabinan Galactan Glucan Xylan Mannan Total sugar Total phenolics Acetic acid Ash Extractives WTC Leaf 37 2.9 ± 0.1 2.3 ± 0.1 13.1 ± 0.6 3.9 ± 0.2 0.6 ± 0.1 22.7 ± 0.1 39.8 ± 0.1 2.3 ± 0.3 10.5 ± 0.1 27.6 ± 1.3 Bark 9 4.0 ± 0.1 2.1 ± 0.1 24.4 ± 0.1 4.1 ± 0.1 0.6 ± 0.1 35.2 ± 0.1 32.9 ± 0.3 2.6 ± 0.2 6.9 ± 0.2 27.7 ± 0.5 Branch 12 2.9 ± 0.1 1.8 ± 0.1 23.9 ± 0.1 7.2 ± 0.4 0.9 ± 0.1 36.8 ± 0.1 29.2 ± 0.4 3.4 ± 0.2 5.7 ± 0.1 21.9 ± 0.9 Chip 42 0.5 ± 0.1 0.7 ± 0.1 38.5 ± 0.3 14.3 ± 0.1 1.8 ± 0.2 55.8 ± 0.2 24.2 ± 0.2 4.4 ± 0.3 1.3 ± 0.1 12.3 ± 0.4 WWF N/A 0.4 ± 0.0 0.4 ± 0.1 46.5 ± 0.9 13.1 ± 0.2 2.6 ± 0.1 63.0 ± 0.4 25.3 ± 0.2 1.2 ± 0.1 0.7 ± 0.1 5.2 ± 0.4 ± All data are represented as the mean of triplicates with standard deviation \n The chemical compositions of the four main components are listed in Table 2 . It was found that the chemical compositions across the four main components were significantly different. Among all components, chip represented the best general chemical composition for biofuel production, with the highest sugar content (55.8%), the lowest total phenolic content (24.2%), and just 1.3% ash. In contrast, leaf had the lowest total sugar (22.7%) but the highest total phenolic and ash contents (39.8% and 10.5%, respectively). In bark and branch, the total sugar, total phenolic, and ash contents were in between those of chip and leaf. As Table 2 reveals, the WWF exhibited significantly higher total sugar content, and lower acetic acid, ash, and extractives contents than all the components from WTC. Although the chip was most suitable for bioconversion among four WTC components, it was still less desirable than WWF. In particular, WWF had a 7.2% higher total sugar content, 0.7% lower ash content, and 7.1% less extractives compared to the chip from WTC. Previous studies demonstrated that the radius, length, and cell wall thickness of fibers increase as tree ages, and as a result the chemical composition differs between juvenile and mature wood [ 37 ]. Our observations confirmed the fact that juvenile wood is lower in cellulose, but higher in lignin, ash, and extractives than mature wood [ 24 , 38 ]. The low sugar content leaf consisted of 37% of the total dry WTC biomass (Table 2 ). Consequently, the presence of leaf increased the non-carbohydrate constituent and reduced the proportion of convertible sugar in WTC. Additionally, high phenolic content, ash, and extractives in the leaves will have negative impacts on the bioconversion process [ 39 – 42 ]. The leaves appeared to be problematic. So we conducted a leaf separation to change the raw biomass chemical composition and examine the impacts of leaf removal in bioconversion. Table 3 presents the compositional analysis of NLC, WTC, LC, and WWF. Leaf removal significantly changed the physical composition of SRC biomass, increasing chip composition to 67%. The original biomass, WTC, was composed of 41.3% total sugar, 32.1% total phenolics, 3.7% acetic acid, 5.5% ash, and 21.0% extractives (Table 3 ). Besides glucan and xylan, the WTC consisted of 5.0% of minor sugars, which were mainly attributed to the presence of bark, branch, and leaf, as these sugars are rarely observed in mature poplar wood [ 37 ]. Following leaf removal, the sugar content increased by 8.2%, while total phenolic, ash, and extractives contents decreased by 5.3, 2.1, and 4.3%, respectively. Leaf removal generated a feedstock (NLC) with a 49.5% total sugar content. The total sugar content of the NLC was still 13.6% lower, and the ash and extractives contents were 4 and 2 times higher, respectively, compared to WWF. Table 3 Chemical and physical compositions of NLC, WTC, LC, and WWF Physical composition (%) Chemical composition (%) Leaf Bark Branch Chip Arabinan Galactan Glucan Xylan Mannan Total sugar Total phenolics Acetic acid Ash Extractives NLC 0 15 18 67 1.5 ± 0.1 1.3 ± 0.1 33.2 ± 0.5 11.9 ± 0.3 1.6 ± 0.1 49.5 ± 0.2 26.8 ± 1.0 4.3 ± 0.3 3.4 ± 0.0 16.7 ± 0.0 WTC 37 9 12 42 2.1 ± 0.1 1.7 ± 0.1 26.9 ± 0.5 9.4 ± 0.3 1.2 ± 0.1 41.3 ± 0.2 32.1 ± 1.0 3.7 ± 0.1 5.5 ± 0.1 21.0 ± 0.6 LC 100 0 0 0 2.9 ± 0.1 2.3 ± 0.1 13.1 ± 0.6 3.9 ± 0.2 0.6 ± 0.0 22.7 ± 0.1 39.8 ± 0.1 2.6 ± 0.5 10.5 ± 0.1 27.6 ± 1.3 WWF 0 0 0 100 0.4 ± 0.0 0.4 ± 0.1 46.5 ± 0.9 13.1 ± 0.2 2.6 ± 0.1 63.1 ± 0.4 25.3 ± 0.2 1.2 ± 0.1 0.7 ± 0.1 5.2 ± 0.4 ± All data are represented as the mean of triplicates with standard deviation \n Chemical composition of water-insoluble fraction (WIF) after pretreatment Following pretreatment and liquid–solid separation of all samples, the compositions of the solid, water-insoluble fraction (WIF), and the liquid, water-soluble fraction (WSF), were analyzed. Expressed as percent of dry matter, Table 4 shows that the trends of the WIF chemical constituents were generally consistent with the compositional data in the raw biomass. The sugar and phenolic contents of three coppice samples ranged from 17.9% to 55.2% and from 37.4% to 50.5%, respectively. Comparing WTC and NLC shows that leaf removal increased the WIF sugar content by 13.8% and lowered the phenolic and ash contents by 6.2% and 3.8%, respectively. Typically, WIF of LC contained the lowest sugar content, but the highest phenolic and ash contents. Although leaf removal significantly enhanced the sugar content and reduced the phenolic and ash contents of WIF, the NLC sugar content was still 8.7% lower, and phenolics and ash were 6.2% and 1.9%, respectively, higher than WWF. Table 4 Chemical composition of WIF after steam pretreatment of poplar samples (as percentages of the solid weight) Chemical composition (%) Arabinan Galactan Glucan Xylan Mannan Total sugar Total phenolics Acetic acid Ash NLC 0.1 ± 0.1 0.1 ± 0.1 52.2 ± 0.3 2.4 ± 0.3 0.5 ± 0.1 55.2 ± 0.1 37.4 ± 0.1 1.6 ± 0.2 2.0 ± 0.1 WTC 0.1 ± 0.1 0.4 ± 0.1 36.6 ± 0.1 3.6 ± 0.3 0.7 ± 0.1 41.4 ± 0.1 43.6 ± 0.1 1.9 ± 0.1 5.8 ± 0.3 LC 0.4 ± 0.1 0.9 ± 0.2 13.1 ± 2.9 3.2 ± 0.6 0.4 ± 0.1 17.9 ± 0.6 50.5 ± 0.2 2.5 ± 0.9 9.8 ± 0.4 WWF 0.0 ± 0.0 0.0 ± 0.0 62.4 ± 2.4 1.1 ± 0.1 0.4 ± 0.1 63.9 ± 0.3 31.2 ± 0.1 0.0 ± 0.0 0.1 ± 0.1 ± All data are represented as the mean of triplicates with standard deviation \n Chemical composition of water-soluble fraction (WSF) after pretreatment The amount of sugars, degradation products, and the pH in the WSF after pretreatment were measured. Table 5 shows that sugar yields in the WSF varied across different poplar samples. Consistent with the results in Table 4 , the majority of minor sugars resided in the WSF, as hemicellulose was mostly dissolved during pretreatment for all poplar samples. Sugar yields are presented in the unit of kg/tonne, implying the total soluble sugars released during pretreatment. For the coppice samples, the glucose and xylose yields ranged between 15.8 to 63.5 and 6.9 to 96.0 kg/tonne, respectively. Minor sugars, including arabinose, galactose, and mannose, represented a non-negligible composition in the WSF, and their yields ranged between 7.6–9.7, 5.6–8.5, and 1.1–14.3 kg/tonne, respectively. The trends of total sugars in WSF matched the sugar content in original biomass. As the amount of the individual sugars decreased from NLC to LC (Table 5 ), the total sugars decreased from 192.1 to 37.0 kg/tonne in the WSF. The monomeric sugar percentage in the WSF is critical because it indicates the amount of direct fermentable sugars. Most sugars in the WSF were recovered in monomeric form except for the LC biomass. As shown in Table 5 , leaf removal significantly changed the total amount of dissolved sugar in WSF—from 108 kg/tonne of WTC to 192 kg/tonne of NLC—an increase of 83.8 kg/tonne. Meanwhile, leaf removal increased the proportion of monomeric sugars by 12%. The fermentable sugar yield in WSF was markedly increased by removing leaves. The sugar yield in WSF of WWF, however, was 18.2 kg/tonne higher than NLC. Interestingly, the monomeric sugar contents of WWF and NLC were about the same at 82% and 83%, respectively. Table 5 Sugar yield in WSF after steam explosion (expressed as kg/tonne raw biomass) Arabinose Galactose Glucose Xylose Mannose Total sugars Total % mon a \n Total % mon Total % mon Total % mon Total % mon Total % mon NLC 9.7 ± 0.3 100 8.5 ± 0.2 90 63.5 ± 2.5 78 96.0 ± 1.2 81 14.3 ± 1.1 79 192.1 ± 0.1 82 WTC 9.1 ± 1.2 100 7.5 ± 0.7 71 39.2 ± 4.1 64 45.5 ± 2.4 71 7.0 ± 0.3 65 108.3 ± 0.3 70 LC 7.6 ± 0.2 60 5.6 ± 0.8 23 15.8 ± 2.1 41 6.9 ± 1.2 10 1.1 ± 0.1 33 37.0 ± 0.2 36 WWF 2.9 ± 0.2 93 5.4 ± 0.4 79 62.9 ± 0.7 76 122.2 ± 8.9 87 16.8 ± 1.0 85 210.3 ± 0.5 83 ± All data are represented as the mean of triplicate measurement \n a “% mono” describes the percentage of soluble sugar presented in monomeric form \n It is shown in Table 6 that the degradation products, including acetic acid, furfural, 5-hydroxymethyl furfural (HMF), and phenolics, were found at different amounts in the WSFs. Leaf removal increased the amount of acetic acid, furfural, and HMF in the WSF by 10.0, 8.3, and 0.4 kg/tonne, respectively (Table 6 ). It is known that with an increase of pretreatment severity, more sugars will be solubilized as monosaccharides and potentially more will be degraded into furans [ 29 ]. All samples were steam-pretreated at the same reaction temperature and residence time with the same SO 2 loading, however, the pH of WSF decreased from 3.1 to 1.9 as a result of leaf removal (Table 6 ). The lower pH in the WSF from the NLC is partially a result of the higher acetic acid concentration compared to that from WTC. In addition, the ash in the leaf appears to provide some buffering capacity to the WSF [ 43 ]. Figure 3 shows the titration curve of three coppice samples in comparison with the blank (deionized water). It appears that for any level of acid addition, the pH of extractant from the leafy material is higher than the blank or any of the other biomass type, demonstrating the larger buffering capacity of the leaves. Table 6 Acetic acid, furans, phenolics yields (expressed as kg/tonne raw biomass), and pH of WSF after steam explosion pH Acetic acid Furfural HMF a \n Phenolics NLC 1.9 48.8 ± 1.5 14.8 ± 1.5 2.2 ± 0.1 23.2 ± 2.0 WTC 3.1 38.8 ± 1.3 6.5 ± 0.9 1.8 ± 0.2 21.9 ± 1.8 LC 3.7 25.2 ± 0.9 0.5 ± 0.1 0.6 ± 0.1 28.1 ± 0.2 WWF 1.6 33.5 ± 0.6 5.2 ± 0.4 1.6 ± 0.1 8.0 ± 0.4 ± All data are represented as the mean of triplicate measurement \n a 5-hydroxymethyl furfural \n Fig. 3 Titration curves with 0.1 M H 2 SO 4 for water extracts of NLC, WTC, LC, and the deionized water (blank) \n Sugar yield and recovery after steam pretreatment Following steam pretreatment, total sugar yield was calculated for each of the poplar samples by combining the sugars in WIF (Table 4 ) and WSF (Table 5 ). Figure 4 presents the total sugar yield for each poplar sample. The yields ranged from 200 to 656 kg/tonne. Corresponding to the compositional characteristics of raw biomass (Table 3 ), NLC achieved the highest sugar yield of 483 kg/tonne among the three coppice samples, WTC had an intermediate sugar recovery of 390 kg/tonne, whereas LC had the lowest sugar recovery of 200 kg/tonne. WWF recovered 656 kg/tonne sugar in total because of its high sugar content in the original biomass. Leaf removal improved the total sugar yield by 93 kg/tonne after steam pretreatment. Interestingly, the sugar recovery was similar across all poplar samples, ranging from 85 to 92%. Although the total sugar yield was much lower in NLC, its sugar recovery showed no significant difference from WWF (Fig. 4 ), indicating that the large difference in sugar recoveries was mainly due to the different original sugar contents (Table 3 ). Fig. 4 Sugar yield (kg/tonne), expressed as total mass of sugar per unit raw biomass, and sugar recovery (%), expressed as total mass of sugar per unit original sugar after pretreatment, of NLC, WTC, LC, and WWF. Error bars indicate standard deviation from triplicate measurements \n Enzymatic hydrolysis of water-insoluble fraction (WIF) Following steam pretreatment, the enzymatic digestibility of WIF samples was evaluated at 5% (w/v) consistency with 5 FPU/g cellulose enzyme loading. Figure 5 highlights the differences in digestibility between poplar samples after 72 h of saccharification. Results in Fig. 5 reveal that NLC had the highest overall sugar conversion, a 72.7% cellulose to glucose conversion, and a 54.7% xylan to xylose conversion. The WTC had lower hydrolysis conversions of 52.6% glucose conversion and 36.3% xylose conversion. Notably, leaf removal improved the enzymatic hydrolysis, resulting in 20% higher glucose conversion and 18% higher xylose conversion. LC had the lowest glucose conversion (23.5%) and xylose conversion (19.5%). By comparison, the glucose conversion of WWF (73.5%) was similar to NLC, while the xylose conversion was lower (34.1%). Fig. 5 72 h cellulose to glucose and xylan to xylose conversion of steam-pretreated NLC, WTC, LC, and WWF at 5% (w/v) solids consistency with 5 FPU/g cellulose and 10 CBU/g cellulose enzyme loading. Error bars indicate standard deviation from triplicate measurements \n The digestibility of pretreated coppice samples demonstrated an improvement in sugar conversion with leaf removal and underscored the low digestibility of steam-pretreated LC. The low yield and digestibility of the LC are attributed to the enzyme inhibition of phenolic compounds and ash. Phenolic compounds have been reported to inhibit and/or deactivate cellulase and β -glucosidases [ 41 , 42 ]. The steam-pretreated coppice samples, especially WTC and LC, had relatively high phenolic contents of 43.6 and 50.5%, respectively (Table 4 ). Contrary to whitewood chips where more phenolic compounds are lignified, the foliage phenolic compounds are present in the form of low molecular weight polyphenols [ 33 ] such as tannin and some flavonoids. These low molecule polyphenols can be easily broken into monomeric phenolics during pretreatment, resulting in stronger enzyme inhibition [ 42 ]. It has also been suggested that ash in steam-pretreated biomass, especially the metal ions, hindered the action of cellulase and β-glucosidases once it exceeds certain content thresholds [ 40 ]. Compared to NLC, the two-fold greater WIF ash content of WTC could partially explain the 20% lower cellulose conversion and 18% lower xylan conversion. Leafy material with its high phenolic and ash content strongly inhibits enzymatic hydrolysis. Prior leaf removal will be necessary to achieve high saccharification yields at modest enzyme loadings. Overall sugar yield and recovery after steam pretreatment and enzymatic hydrolysis The overall monomeric sugar available after pretreatment and enzymatic hydrolysis determines the total amount of fermentable sugars from bioconversion, and is calculated by adding monomeric sugars in WSF and hydrolyzed WIF (Fig. 6 ). For each biomass, bioconversion efficiency is also expressed in two formats—the overall monomeric sugar yield at kg/tonne demonstrates the monomeric sugar from per unit biomass, and overall monomeric sugar recovery (in percentage) represents the monomeric sugar from original sugar in raw biomass. Within all three coppice samples, NLC had the highest overall sugar yield of 363 kg/tonne and LC had the lowest of 66 kg/tonne. The highest overall sugar yield of NLC is a direct result of the high sugar content in original biomass (Table 3 ), high sugar recovery in WSF and WIF, and the most efficient WIF digestibility. Leaf removal increased the overall sugar yield by 147 kg/tonne, which accounts for 40% relative increase from the WTC (Fig. 6 ). Not surprisingly, only 27% original sugars were recovered from LC. NLC recovered 67% sugar after pretreatment and enzymatic hydrolysis, showing no significant difference with WWF (71%). It appears that the branch and bark in the NCL does not impair the bioconversion sugar recovery efficiency. The inherent sugar content was lower in original NLC biomass, however, resulting in 136 kg/tonne less total sugar yield of NLC compared to WWF (Fig. 6 ). Fig. 6 Overall monomeric sugar yield (kg/tonne), expressed as monomeric sugar per unit raw biomass, and overall monomeric sugar recovery (%), expressed as monomeric sugar per unit original sugar after pretreatment and enzymatic hydrolysis, of NLC, WTC, LC, and WWF. Error bars indicate standard deviation from triplicate measurements \n The differences in sugar yields for NLC and WTC show that leafy material hinders the bioconversion sugar recovery. Moreover, since leaf had an extremely high moisture content (around 70%, data not shown), the mass proportion of leaf will be even higher in fresh biomass. From a logistic point of view, the high moisture content will accelerate microbial-driven deterioration of the biomass during transportation and storage [ 6 ]. Considering the difficulty in removing leaf from the harvested poplar mixture (as we experienced), we suggest a leaf separation operation during harvest. Thompson et al. [ 44 ] have developed an air classification method to separate the heterogeneous biomass mixture into anatomical or visually unique fractions. Ideally, a similar air separation mechanism can be integrated into the coppice harvester to efficiently classify and collect both leaf and no-leaf components during harvest. Economic assessment With the assumptions that NLC poplar could be purchased for its heating value—$53/tonne—we calculated the ethanol selling price required to cover the operating cost using the yield data in the current study. The simulated biorefinery used 700,000 tonne of feedstock per year and produced 144 million l/year (38 million gallon/year) at an ethanol conversion of 206 l/dry tonne feedstock (49 gallon/dry US ton feedstock). The minimum ethanol selling price required to cover the operating cost was found to be $1.69/gallon, which is approximately the same as the current ethanol selling price ($1.65/gallon, 2017). Despite the low bioconversion yield, it appears that the NLC biomass could be a reasonable feedstock provided it can be obtained at a discounted price reflecting its lower quality. The economics of using short rotation coppice (SRC) poplars may be improved if higher value materials can be produced from leaves separated before bioconversion. In a recent review study, Devappa et al. [ 45 ] summarized a number of value-added natural chemicals which can be extracted and purified from tree residues including leaf. Tree foliage is also recognized as a high-quality supplement for animal feed [ 46 ]. Traditionally, poplar leaf has been used as a crude protein fodder resource for livestock, particularly for ruminant animals [ 13 , 47 ]. In this current study, we were able to obtain about 12% of crude protein from the LC following a sonication protein extraction method (data not shown). Returning leafy material to the tree farm could also have environmental benefits. Ecologically, foliage serves several important functions in plantation soil [ 48 ]. It improves soil structure and increases solid porosity for better aeration and moisture holding capacity. In addition, upon decomposition, foliage litter returns the minerals and organics to the site, and becomes a source of plant nutrients [ 48 ]. Complete leaf removal during whole tree harvest might increase the potential of soil erosion, site degradation, and accelerate nutrient withdrawals [ 49 ]. Harvesting trees during the dormant season or using a defoliant would ensure the tree farm realizes the environmental benefits of the leafy material."
} | 6,126 |
35560875 | PMC9089752 | pmc | 7,205 | {
"abstract": "Correction for ‘Preparation of superhydrophobic surfaces with micro/nano alumina molds’ by Takashi Yanagishita et al. , RSC Adv. , 2018, 8 , 36697–36704."
} | 39 |
23922937 | PMC3724788 | pmc | 7,208 | {
"abstract": "Mutualisms are common in nature, though these symbioses can be quite permeable to cheaters in situations where one individual parasitizes the other by discontinuing cooperation yet still exploits the benefits of the partnership. In the Rhizobium -legume system, there are two separate contexts, namely nodulation and nitrogen fixation processes, by which resident Rhizobium individuals can benefit by cheating. Here, we constructed reversible and irreversible mutations in key nodulation and nitrogen-fixation pathways of Rhizobium etli and compared their interaction with plant hosts Phaseolus vulgaris to that of wild type. We show that R. etli reversible mutants deficient in nodulation factor production are capable of intra- specific cheating, wherein mutants exploit other Rhizobium individuals capable of producing these factors. Similarly, we show that R. etli mutants are also capable of cheating inter- specifically, colonizing the host legume yet contributing nothing to the partnership in terms of nitrogen fixation. Our findings indicate that cheating is possible in both of these frameworks, seemingly without damaging the stability of the mutualism itself. These results may potentially help explain observations suggesting that legume plants are commonly infected by multiple bacterial lineages during the nodulation process.",
"introduction": "Introduction Rhizobia can form symbiosis nodules with legumes in which the bacteria fix atmospheric nitrogen into ammonia that can be utilized by the host plant. Two-way chemical signaling is required for this process. For example, Rhizobium etli is able to identify flavonoid substances released by its host legume Phaseolus vulgaris signaling it to produce nodulation factors (Nod), which in turn induce cortical cell division and form symbiosis nodules, where nitrogen is fixed [1] . Nodulation and nitrogen fixation is achieved by nod and nif / fix gene clusters, which play pivotal regulatory roles in symbiosis [2] . All Rhizobia strains contain the common nodulation genes nodDABC , which encode a transcription factor, as well as enzymes that are required to produce Nod factors [3] . In bacteroids that rhizobia differentiate in the infection thread, nif and fix genes are expressed through the FixL/FixJ two-component regulatory system and the transcriptional activator NifA [4] which leads to the production of nitrogenase and ultimately nitrogen fixation. Due both to the ability of Rhizobia to form symbiotic nodules on host legumes and their ability to fix inert atmospheric nitrogen (N 2 ) to plant-usable ammonia, mutualistic interactions are commonly studied within the Rhizobia-host system [5] . In this context, Rhizobium individuals cooperate with the plant host by fixing nitrogen into usable ammonia while benefiting from receiving host-derived resources, such as carbon [5] . Nitrogen-fixation, however, is an energetically costly process, and can be predicted to benefit Rhizobia only if the plant supplies efficient resources to outweigh the cost of ammonia production on the bacteria [6] . Given this cost, opportunity exists for non-cooperating partners to parasitize these social dynamics by reaping benefits (carbon or other resources) while failing to supply nitrogen-fixation in return [7] . It has been demonstrated that one way of mitigating this potential burden is through the differential allocation of host resources in attempts to punish these non-fixing “cheaters” within a community, though the mechanism of these sanctions is unknown [8] . Though these studies show that plants can reduce rhizobial nodule occupancy in scenarios where whole nodule cheating is forced, they do not address how plants respond to increased parasitism in the event of nodules containing a mixed population of both cooperating and non-cooperating strains, potential in nature given estimates of 12% to 32% of field soybean nodules containing double occupancy [9] . Similar dynamics can be studied within the colonization process itself. For effective colonization, Rhizobia secrete nodulation factors (Nod factors) which induce cortical cell division in their host legume [1] . For instance, previous studies have shown that it is possible for mutants deficient in nodulation to form mixed nodules with Nod factor-producing strains [10] , suggesting that Rhizobia can hitch-hike within their own species during the nodule-formation process. This system, in turn, can be permeable to cheaters (nodulation deficient mutants) that benefit by colonizing with other individuals, allowing for host colonization and resource sequestration without the burden of producing Nod factors themselves. If a strong metabolic cost to produce these Nod factors exists, mutants deficient in nodulation can be expected to arise at high frequencies, as what typically happens with any common good [11] . As with nitrogen-fixation, the selective advantage gained by mutants deficient in nodulation factor production would depend upon the metabolic cost of production. In this study, we constructed reversible and irreversible mutations in the key nod and nif genes of R. etli and investigated their nodulation and N 2 -fixation behaviors when they were inoculated individually or co-inoculated with wild type strains. We show that R. etli is capable of both intra- specific cheating through the mutation of a key gene in the nodule-formation pathways as well as inter- specifically cheating their host through the mutation of a gene essential to nitrogen fixation, suggesting a mechanism for the existence of multiple bacterial strain populations in nature as well as a reason for their persistence.",
"discussion": "Discussion Social cheating can be common in microbial populations [25] – [28] . Cheating individuals, acting selfishly, can persist within a population despite being detrimental to the long-term survival of that population [29] . Similarly, “cheaters” can sometimes evade host pressure to cooperate in mutualistic relationships, particularly in rhizobium-legume interactions where mutualism efficiency, at least as defined in terms of nodule productivity, remains intact [5] . Here, we study induced cheating in the plant symbiont Rhizobium etli through two important processes thought necessary for success of the symbiosis: nodule formation and nitrogen fixation. Our results indicate that both intra- and inter- specific cheating is facilitated by cooperating strains in this system. \n Intra- specific cheating, occurring when individuals within a population seemingly parasitize other individuals of the same species, has been shown to occur in a variety of natural systems [25] – [28] . In Rhizobium -host interactions, this cheating is predicted in situations such as nitrogen fixation to be thwarted by differential host sanctions imparted on non-cooperating nodules, likely due to allocating fewer resources to nodules supplying the host plant with little or no nitrogen [8] ; however, few studies have examined the role of intra- specific cheating during the colonization process, a context potentially preceding the appropriation of resources and thus having little influence from the host. Here, we show that it is possible for mutants deficient in nodulation factor (Nod factors) production to colonize host nodules only if co-colonizing with Nod factor producing individuals, effectively “hitch-hiking” during the colonization process. We tested this using nodB mutants created by two different mechanisms ( Fig. 1A ) and theorized that those created by double-crossover deletions should not have the potential to revert to wild type during colonization, whereas those mutants created by single-crossover events could revert. In this context, deletion strains ( ΔnodB ) could not colonize host legumes by themselves, though mutants could co-exist in roughly 30% of nodules when co-inoculated with wild type strains ( Fig. 3A ), suggesting that host legumes can be primed for nodule production by strains producing Nod factors. The ability of nod deletion mutants to colonize only in the presence of wild type strains can imply somewhat of a redundancy to nodulation factor production in environments where wild type individuals are common. Though deletion strains could not colonize by themselves, we investigated the colonization of nodB mutants created by single-crossover mutations ( ::nodB ) because of the potential for these strains to revert to wild type through subsequent homologous recombination events. Nodule occupancy levels for experiments in which ::nodB was co-inoculated with wild type strains were similar as those seen in the ΔnodB -wild type co-inoculations ( Fig. 3A ). Though nodule formation was possible when legumes were inoculated with only these mutants ( ::nodB ) ( Fig. 2A ), 97.2% of ::nodB mutants found in these nodules had reverted to wild type, which is predicted by the fact that nodulation factor production is necessary for the establishment of nodules on host legumes [3] . Taken together with the necessity for wild type co-colonization in ΔnodB mutants, it is likely that these reversion events in ::nodB mutants are again simply necessary to establish nodule formation. As a whole, these data suggest that nodB mutants can potentially forego producing nodulation factors and instead rely on the production from neighboring individuals. Depending on the cost of nodulation factor production to wild type strains, these findings could have implications for a tragedy of the commons in which no individuals are able to colonize hosts due to a lack of nodulation factor production, though we have not surveyed isolates of R. etli to investigate whether nod mutants are common in natural systems. Within the host nodule, it has been predicted that non-cooperating Rhizobium can be readily controlled, likely due to differential host sanctions meant to punish nodules deficient in nitrogen fixation [5] . Studies have shown that both Rhizobia occupancy per nodule and Rhizobia occupancy per nodule mass have decreased when forcing cheating by substituting an atmospheric N 2 :O 2 mixture with Ar:O 2 \n [8] , though these experiments effectively force the entire rhizobia nodule population to cheat and may not represent host sanctions in the event of a mutant lineage deficient in nitrogen-fixation arising within an otherwise cooperating population. To test the effects of this, we created nifA mutants, again by two different mechanisms, hypothesizing that those mutants created by double-crossover deletions ( ΔnifA ) should not have the potential for reversion, while those created by single-crossover events ( ::nifA ) could possibly revert to wild type. Surprisingly, single inoculation experiments with either ::nifA or ΔnifA mutants produced nodule numbers and sizes similar to those seen in wild type controls, contrary to what is predicted by host sanctions ( Fig. 5A ). Similarly, single inoculation experiments showed that neither ::nifA nor ΔnifA single inoculation experiments resulted in significantly different nodule weights compared to wild type controls ( Fig. 5B ). Although it is unclear why the host does not select against non-cooperating nodules, it is possible that host-derived sanctions could be detected after a longer growth period. Though host sanctions are typically studied in contexts where whole nodules are forced to cheat, legume plants are typically infected with several bacterial lineages [5] . Within individual nodules, it could be expected that mutant bacterial lineages deficient in nitrogen-fixation could endure a selective advantage due to diminished metabolic burden, sweeping to high frequencies or fixation within the population. To study this, legumes were co-inoculated with either wild type strains and their ::nifA reversible mutant derivatives or wild type and their ΔnifA deletion derivatives. Surprisingly, in both cases, co-inoculations formed more nodules per plant ( Fig. 5A ) compared to wild type and single-inoculations of each of the mutants, though the average weight of each nodule remained similar ( Fig. 5B ). Within the nodules, roughly 50% contained only the mutant strain and approximately 40% contained a mixed population of wild type and mutants (for both ::nifA and ΔnifA ). From these mixed populations, competitive index was measured and both mutants were found to have selective advantages relative to wild type strains, as expected by the decreased metabolic costs endured by lacking nitrogenase activity. Similarly, in ::nifA reversible mutants, reversion rate was low both in single inoculations as well as in mixed inoculations, likely due to the costs associated with reverting. Interestingly, nitrogenase activity in co-inoculation experiments showed similar levels to those seen in wild type. While nodules inoculated with mutant strains by themselves could not reduce acetylene, the high levels of nitrogenase activity seen in mixed inoculations could suggest that wild type strains can increase fixation of atmospheric nitrogen, though we have not investigated expression levels of nifA in these instances. As host sanctions would likely be subjected to the nodule as a whole, this finding could imply that, in a mixed population of cooperating and non-cooperating strains, cooperating individuals could increase nitrogenase activity to account for the deficit due to cheaters, likely in attempts to avoid sanctions imparted on the nodule as a whole, thus punishing both cooperating and non-cooperating strains. In this study, we show both that Rhizobium mutants can cheat during the nodule colonization process as well as suggest mechanisms for the stable persistence of these cheaters within the population. As it is likely that natural Rhizobium -plant mutualisms contain many distinct bacterial lineages, this study implies a potentially relevant manner in which to study this social parasitism in the natural context, as well as a reason for the persistence of many lineages within a population."
} | 3,522 |
18760359 | null | s2 | 7,212 | {
"abstract": "The extremely thermoacidophilic archaea are a particularly intriguing group of microorganisms that must simultaneously cope with biologically extreme pHs (< or = 4) and temperatures (Topt > or = 60 degrees C) in their natural environments. Their expanding biotechnological significance relates to their role in biomining of base and precious metals and their unique mechanisms of survival in hot acid, at both the cellular and biomolecular levels. Recent developments, such as advances in understanding of heavy metal tolerance mechanisms, implementation of a genetic system, and discovery of a new carbon fixation pathway, have been facilitated by the availability of genome sequence data and molecular genetic systems. As a result, new insights into the metabolic pathways and physiological features that define extreme thermoacidophily have been obtained, in some cases suggesting prospects for biotechnological opportunities."
} | 232 |
36297830 | PMC9611852 | pmc | 7,213 | {
"abstract": "Polymer film coating with a highly hydrophobic surface property is a practical approach to prevent fouling of any structures in the marine environment without affecting marine microorganisms. The preparation of a polymer coating, from a simple and easy method of solution blending of hydrophobic polydimethylsiloxane elastomer and hydrophilic polyurethane with SiO 2 , was carried out in this study, with the aim of improving characteristics, and the coating demonstrated economic feasibility for antifouling application. Incorporation of SiO 2 particles into PDMS and PDMS/PU polymer film improved mechanical properties of the film and the support fabrication of micropatterns by means of a soft lithography process. Observations from field emission scanning electron microscope (FESEM) of the PDMS/SiO 2 composite film revealed a homogeneous morphology and even dispersion of the SiO 2 disperse phase between 1–5 wt.%. Moreover, the PDMS film with 3 wt.% loading of SiO 2 considerably increased WCA to 115.7° ± 2.5° and improved mechanical properties by increasing Young’s modulus by 128%, compared with neat PDMS film. Additionally, bonding strength between barnacles and the PDMS film with 3 wt.% of SiO 2 loading was 0.16 MPa, which was much lower than the bonding strength between barnacles and the reference carbon steel of 1.16 MPa. When compared to the previous study using PDMS/PU blend (95:5), the count of barnacles of PDMS with 3 wt.% SiO 2 loading was lower by 77% in the two-week field tests and up to 97% in the eight-week field tests. Subsequently, when PDMS with 3 wt.% SiO 2 was further blended with PU, and the surface modified by the soft lithography process, it was found that PDMS/PU (95:5) with 3 wt.% SiO 2 composite film with micropatterns increased WCA to 122.1° ± 2.9° and OCA 90.8 ± 3.6°, suggesting that the PDMS/PU (95:5) with 3 wt.% SiO 2 composite film with surface modified by the soft lithography process could be employed for antifouling application.",
"conclusion": "4. Conclusions In this work, nontoxic polymer film coatings made from PDMS with SiO 2 and PDMS/PU with SiO 2 composite film were successfully prepared to provide antifouling performance through a simple technique in which SiO 2 was initially added to the PDMS matrix phase to enhance mechanical property of the film in order to support the fabrication of micro-/nano-structures by the soft lithography process. When SiO 2 was introduced to PDMS, the phase morphology of the polymer composite clearly illustrated good distribution from 1 to 5 wt.% of SiO 2 addition. Furthermore, at 3–5 wt.% of SiO 2 addition, hydrophobicity and mechanical properties of the material were enhanced. More importantly, the PDMS with SiO 2 composite film prevented fouling attachment and lower barnacle adhesion strength. The PDMS with 3 wt.% SiO 2 composite film showed decreased adhesion strength of fouling on surface, with 85–90% lower adhesion strength than the reference carbon steel. Subsequently, SiO 2 was introduced to the PDMS/PU (95:5) blend to produce PDMS/PU (95:5) with 3 wt.% of SiO 2 composite film. Based on the best composite film behavior result, 3 wt.% of SiO 2 was applied to the PDMS/PU blend (95:5) film. PDMS/PU (95:5) with 3 wt.% of SiO 2 composite film with ridge pillar micropatterns increased water contact angle to 122.1° ± 2.9° and oil contact angle to 90.8° ± 3.6°. Moreover, the addition of 3 wt.% SiO 2 enhanced the Young’s modulus to twice that of the neat PDMS. This study suggests a simple and easy method for the development of an environmentally safe and effective antifouling film that would be useful for the marine industry.",
"introduction": "1. Introduction Generally, biofouling is formed due to the attachment of microorganisms (bacteria, algae, and fungi) and macroorganisms (barnacles, sponges, seaweed, etc.) to the surface that is covered with water [ 1 ], which is one of the major causes of physical damage to surfaces of marine structures and equipment. Biofouling has been, and continues to be, a global issue owing to its substantial environmental and economic effects. For example, the major issues caused by marine fouling in terms of economic effects are the increase in fuel consumption, of up to 40%, and overall expenses, of up to 77%, from the increase in hydrodynamic drag. These increased costs arise from increasing weight, and reducing ship speed, due to the attached organisms. Costs also arise out of corrosion. Furthermore, there are the environmental effects, such as the introduction of non-native species into local ecosystems [ 2 ]. The estimated cost of transportation delays, hull repair, cleaning and general maintenance are estimated to be about $150 billion per year [ 3 ]. As a result, industries are paying very close attention to the finding of solutions for this issue. The fouling process typically involves the formation of biofouling on the surface of materials, which is caused by the adsorption of organic and inorganic macromolecules immediately after immersion, resulting in the formation of a conditioning film. Then, microbial cells are rapidly transported to other surfaces and form a microbial film on the surface. After that, the development of a more complex community of multicellular species means the biofilm develops with a chemical oxygen gradient and corrosion processes occur on the surface. Antifouling (AF) coatings have grown in popularity as appropriate substitutes for the standard metals with decreased corrosion effects. The antifouling mechanism adopts several methods to reduce the amount of fouling on the surfaces, such as physical and chemical methods. The chemical method focuses on biocide release and the physical method focuses on surface modification to reduce the attachment of foulants to the surface. Paints containing dispersed biocides (e.g., arsenic and mercury oxide) were the major focus for marine AF applications in the mid-nineteenth century [ 1 ]. Later, the prohibition of organo-mercury and organo-arsenic chemicals, motivated by mounting safety and health concerns, meant a decline in the use of these paints. Consequently, tributyltin (TBT) was introduced as an antifouling agent. Tributyltin was one of the most widely used chemicals for antifouling paints, based on leaching biocide. Nevertheless, environmental concerns about TBT were raised in the late 1980s, when oysters showed considerable shell thickening, and some marine creatures became locally extinct (e.g., Nucella), owing to their inability to reproduce as a result of the coating. Tin bioaccumulation was discovered in fish, seals and even ducks [ 4 ]. In 2003, the international maritime organization (IMO) banned tributyltin from antifouling application [ 5 ]. Since then, antifouling paints have been developed using copper oxide pigment, thiocyanate, cuprous bromide and so on. Copper oxide was chosen over other pigments because it was less costly and easily dispersed. However, it has been proven to have bioaccumulation and biomagnification effects on blue mussels [ 6 ]. Non-toxic, non-biocide-release based approaches for antifouling coatings aims to modify hydrophobic surfaces with nano and micro patterns to create rough surfaces. Surface roughness plays an important role in hydrodynamic performance of antifouling (AF) coatings and influences the settlement behavior of fouling, because the surface architecture of rough surfaces reduces available contact area and the ability for water droplets to adhere to the surface, due to the stable air cushion entrapped at liquid/solid interfaces, leads to higher hydrophobicity, or lower surface energy, which significantly minimizes biofouling [ 7 ]. So, the attached and adhering foulants are more easily removed from the surfaces at low hydrodynamic shears, such as the external force on ship movements, currents and waves. On the other hand, barnacles on the surface of untreated metal form biofilm and rapidly grow on the metal surface. Finally, biofilm develops with a chemical oxygen gradient, with the innermost layer in contact with the metal exhibiting an anoxic environment and the most superficial layer retaining large concentrations of oxygen. At this stage, anaerobic microorganisms, associated with metal corrosion processes, are dominant [ 8 ]. Regarding hydrophobic polymers, polydimethylsiloxane (PDMS) is one of the non-toxic and inexpensive polymers widely studied for antifouling coating [ 9 , 10 ]. It is preferred because the main chains of PDMS consist of siloxane bonds and side chains of methyl groups. The high bond energy and bond angle of the siloxane bond structures offer superior heat stability and flexibility for PDMS. The side chains of the methyl groups are hydrophobic (water contact angle (WCA) 107°–110°) and non-polar, leading to low surface energy and excellent fouling release [ 11 ]. Furthermore, because of its low shrinkage rates and ease of penetration into micropatterns, PDMS is suited for the soft lithography process [ 12 ]. The soft lithography process is a simple, cost-effective and scalable method for modifying PDMS surfaces into highly hydrophobic surfaces or superhydrophobic surfaces (WCA > 150°) with micro- and nano-patterns, or nanostructures, such as the surface of a lotus leaf, for antifouling application and corrosion resistance on metal surfaces [ 13 , 14 , 15 ]. However, PDMS does not have high enough mechanical properties, and when the PDMS surface was modified with micropatterns using the soft lithography process, often the micropatterns would collapse [ 9 ]. Inorganic materials, such as alumina (Al 2 O 3 ), titanium dioxide (TiO 2 ), zirconium (ZrO 2 ), silver (Ag) and silica (SiO 2 ) could be used to improve the mechanical properties, roughness and antifouling of the polymer surface [ 16 ]. Although each nanoparticle is highly efficient, pricing limits are a challenge in the industry [ 17 ]. Silica (SiO 2 ) is one of the most convenient and widely utilized materials, due to its being inexpensive, having inert reactivity, well-known chemical characteristics, good mechanical properties and a mature preparation technique. Moreover, the modulus values of the PDMS elastomers are enhanced as an increasing amount of silica is introduced [ 18 ]. Although hydrophobic surfaces, such as the PDMS coating, have shown high water repellence and potential against many marine organisms, oils found in marine environments attach to the surface of the coating, thereby reducing efficiency and performance of the polymer coating, and allowing slime composed of diatoms and bacterial attachments, which limits the usage of polymer coating in practical applications. Consequently, although the hydrophobic characteristic is the most important factor for polymer coating for antifouling applications, amphiphobic surface polymers, which have both hydrophilic and hydrophobic properties, have piqued interest due to their unique wetting behavior and potential applications [ 19 ]. Thus, to improve the amphiphobic surface properties of the coating, various techniques have been employed, such as inducing a hierarchical surface, and grafting or blending with a hydrophilic polymer. In 2018, Bhalani and co-workers prepared poly(vinyl pyrrolidone) and induced hierarchical surface morphology on a poly(vinylidene fluoride) membrane with the prepared poly(vinyl pyrrolidone) to obtain a material with superhydrophilic and underwater oleophobic characteristics [ 20 ]. Bhalani et al. reported the preparation of low toxic polymers for antifouling using a blended membrane of PVDF-blend from poly(vinylidene fluoride) and poly(methyl methacrylate)-co-poly(chloromethyl styrene). The modified membranes, obtained by covalent functionalization of PVDF, that were prepared had increased hydrophilicity and decreased protein fouling properties [ 21 ]. However, grafting and hierarchical surface modification required a variety of chemicals and it was a very complicated method. Polymer blending is one of the best solutions for producing antifouling materials, since it is a less expensive and quicker means to induce surface enrichment with unique properties [ 22 ]. It involves the combination of two or more polymers to give applications that benefit from the excellent properties of each polymer. A blend of different quality polymers has made it more cost-effective to create novel materials with the desired qualities, mechanical properties, abrasive resistance, etc. [ 23 ]. Polyurethane has a wide range of applications, such as its use in biomedical applications, the automotive industry, building, textiles and others, due to its non-toxic, non-flammable, environment-friendly characteristics, and its good mechanical properties, economical manufacturing process and wide range of other properties [ 24 , 25 ]. In addition, many articles have reported on polyurethane modification for decreased protein adsorption and cell adhesion [ 26 ]. In our previous study, a highly hydrophobic surface polymer coating for antifouling application made from PDMS/PU blends was prepared and characterized. In this study, polymer coatings for antifouling application from PDMS with different amounts of SiO 2 were prepared and studied. Mechanical properties and hydrophobicity of PDMS composite films were improved with the addition of different amounts of SiO 2 . This was aimed at presenting a simple procedure for preparing nontoxic material and an easily processable polymer coating for antifouling. Subsequently, PDMS with 3 wt.% SiO 2 was further blended with PU. The performances of the PDMS/SiO 2 and PDMS/PU (95:5) with SiO 2 composite and PDMS/PU (95:5) with SiO 2 composite modified surface with soft lithography process were studied in terms of increased mechanical and surface properties.",
"discussion": "3. Results and Discussion 3.1. PDMS with SiO 2 Composite 3.1.1. Morphology of the PDMS/SiO 2 Composite The morphologies of neat polydimethylsiloxane (PDMS), and polydimethylsiloxane (PDMS) composite, with PDMS as the matrix phase and SiO 2 as the disperse phase, were observed using field-emission scanning electron microscopy (FE-SEM), and are shown in Figure 1 . The rough surface of the cross-section, as shown in neat PDMS, indicated the ductile characteristic of the material [ 31 ]. Compared with the neat PDMS surface, the PDMS composite surface, with SiO 2 particles from 1 to 10 wt.%, showed microroughness, which correlated positively with filler content [ 32 ]. The dispersion of particles inside the matrix was more uniform and homogeneous from 1 to 5 wt.% SiO 2 . The average diameter of dispersed SiO 2 particles was about 0.2–0.3 µm. The quantity of SiO 2 particle agglomerations and their size increased as the silica loading increased. Particle agglomeration occurred when the amount of SiO 2 increased to 7 wt.% and 10 wt.% by weight, as shown in the SEM image (see Figure 1 ). It was also demonstrated that with increased wt.% of SiO 2 particle loadings, the probability of bulk formation increased, resulting in a loss of material properties owing to the heterogeneous dispersion of silica particles in the matrix [ 33 , 34 ]. 3.1.2. Water Contact Angle (WCA) of the PDMS/SiO 2 Composite Water contact angle (WCA) was used to evaluate the hydrophobic surface characteristic, based on surface wettability of the PDMS and PDMS/SiO 2 composites. To achieve antifouling properties, the surface wettability of the material, determined by water contact angle, should be greater than 90°, which is the characteristic of a hydrophobic surface. The contact angle of water droplets on neat polydimethylsiloxane surface had a value of 107.6° ± 1.7°, as shown in Figure 2 . The addition of 1 wt.% of SiO 2 gave a WCA value of the PDMS/SiO 2 composite of 108.3° ± 2.5°, which was relatively similar to that of the neat PDMS. The addition of 1 wt.% SiO 2 was found to have little effect on the WCA of the composite. This might have been due to the small amount of SiO 2 added which was insufficient to reach the dispersion limit to produce a surface architecture of hierarchically rough surfaces. Similar behavior was also observed by Tapasa, K. and coworker [ 35 ]. However, WCA increased to 115.7° ± 2.5° when the ratio of SiO 2 was 3 wt.% and slightly decreased to 113.9° ± 1.8° when the ratio of SiO 2 was 5 wt.%. This might have been because of the synergistic impact of the surface roughness and the hydrophobic matrix [ 32 ]. The water contact angle value of the composites further decreased when wt.% SiO 2 content increased to 7 wt.% at 107.2° ± 1.3° and 10 wt.% at 105.3° ± 2.1°, which was probably due to the agglomeration of the SiO 2 particles. Consequently, WCA of the PDMS/SiO 2 composites was highest when the wt.% of SiO 2 was at 3 wt.%. From the result, adding SiO 2 was one way to increase the hydrophobicity of the polymer surface. It was also possible that addition of SiO 2 to the PDMS/PU blended (95:5) films would also further enhance the hydrophobic characteristics of polymer blend films, since the contact angle of water droplets on PDMS/PU blend (95:5) films in the earlier studies showed a range of WCA between 103.4° ± 3.8° to 91.4° ± 0.8° [ 12 ]. Furthermore, it is widely acknowledged that converting the hydrophobic surface of polymer film to a highly hydrophobic surface is a complex manufacturing procedure, which may involve complicated processing techniques, such as the use of plasma, etching, electrochemical reactions, decomposition etc. [ 36 , 37 ]. In this work, the soft lithography process was subsequently employed to further improve the hydrophobic characteristic of the materials to obtain highly hydrophobic surfaces for antifouling applications by means of a simple and cost-effective approach. 3.1.3. Barnacle Measurements Neat PDMS and PDMS/SiO 2 composite films were further examined in a short-term marine field test at Koh Sichang Marine Science Research Centre of Chulalongkorn University (13°09′10.6″ N 100°49′02.6″ E) in Thailand from April to June to study antifouling performance of the composite films by determining the number of barnacles and barnacle adhesion strength on the surfaces. The number of barnacles of each sample was compared with carbon steel (JIS SS400) as a reference material, using a digital microscope ( Figure 3 ). After 2 weeks, the blank control carbon steel was fully covered by fouling organisms, including barnacles. on the surface ( Figure 4 ). Moreover, corrosion and rusting were also observed on the surface of the carbon steel. Barnacles on the surface of carbon steel presented at approximately 3.1 ± 0.6 marine barnacles/cm 2 . After 4 and 8 weeks, this decreased from 2.5 ± 0.2 to 0.9 ± 0.4 marine barnacles/cm 2 , which was due to the increase in the size of barnacles. However, the number of barnacles of the neat PDMS in sea water after 2 weeks was approximately 0.9 ± 0.1 marine barnacles/cm 2 for the samples facing away from the shore ( Figure 5 ). When immersed longer, some biofoulings disappeared from the surface of the neat polydimethylsiloxane, since fouling was more easily released from the surface of the neat PDMS samples due to weak adhesion between the surfaces [ 38 ]. Furthermore, PDMS with 3 wt.% SiO 2 exhibited slightly less barnacles than the blank control carbon steel and the neat PDMS samples. The marine organism counts decreased by 77% at 2 weeks and 97% at 8 weeks, when compared to the reference carbon steel. Apart from the number of barnacles on the surface, barnacle adhesion strength, which relates to the barnacle release property, was also another significant factor linked to antifouling application that was investigated. Barnacle adhesion strength was calculated by dividing barnacle adhesion force with the barnacle base area. The result showed that the PDMS with 3 wt.% SiO 2 composite exhibited lower barnacle adhesion strength, which was 85–90% less than barnacle adhesion strength on carbon steel ( Figure 6 ). Furthermore, when compared to previous studies, the barnacle counts of 3 wt.% SiO 2 and 5 wt.% SiO 2 were lower than the barnacle counts on PDMS/PU blend (95:5) films, although barnacle adhesion strength remained rather similar [ 39 ]. From the results, surface roughness played an important role in hydrodynamic performance of the antifouling (AF) coating and influenced the settlement behavior of fouling. The surface architecture of the rough surfaces reduced available contact area and the ability for water droplets to adhere to the surface, due to the stable air cushion entrapped at liquid/solid interfaces, leading to higher hydrophobicity or lower surface energy, which significantly minimized the biofouling. So, the attached and adhering foulants were more easily removed from the surfaces at low hydrodynamic shears, such as the external forces due to the ship movement, currents and waves. It was indicated that all PDMS/SiO 2 composite surfaces had a significantly lower number of barnacles and lower barnacle adhesion strength than carbon steel. So, it was clearly illustrated that the PDMS/SiO 2 composite film surface improved both antifouling attachment and fouling release performance. 3.1.4. Mechanical Properties Studies of PDMS/SiO 2 Composite Another important property for the application of an antifouling coating film are mechanical properties, such as tensile strength. The strength of the neat PDMS and PDMS/SiO 2 composite films, and mechanical properties of both neat PDMS and PDMS/SiO 2 composites were studied. The results obtained from the tensile tests for the neat PDMS and PDMS/SiO 2 composites with various weight percentages of SiO 2 are presented in Figure 7 , which is a plot between stress and strain of the different composite films. The neat PDMS film showed an elastic property with a Young’s modulus of 1.50 ± 0.18 MPa. Young’s modulus of the composite with the addition wt.% of SiO 2 in the PDMS matrix phase at 3, 5, 7 and 10 wt.% was 2.92 ± 0.96 MPa, 3.07 ± 0.59 MPa, 2.68 ± 1.91 MPa and 2.30 ± 0.39 MPa, respectively. It is interesting to note that further increasing of the SiO 2 content from 7 wt.% caused SiO 2 particles to agglomerate (as observed from SEM results), and, thus, the optimum wt.% of SiO 2 for improving mechanical characteristics of the composites was between 3–5 wt.% [ 39 ]. These results showed that polydimethylsiloxane elastomer filled with SiO 2 , as a dispersal phase, improved the mechanical properties of the polymer film and still retained good properties of PDMS, in terms of resistance to deformation, as well as the opportunity to build microstructures using the soft lithography process [ 40 ]. 3.1.5. Atomic Force Microscopy (AFM) of PDMS/SiO 2 Composite To support WCA results, surface roughness of neat PDMS, PDMS with 3 wt.% SiO 2 and PDMS with 5 wt.% SiO 2 composite films were studied by means of an atomic force microscope (AFM). Surface roughness is a key factor to increase water contact angle of a surface because the surface architecture of hierarchically rough surfaces reduces available contact area and the ability for water droplets to adhere to the surface [ 41 ]. The surface roughness of the PDMS film could be further improved by introducing SiO 2 particles and the material could become more hydrophobic after the addition of SiO 2 particles in the PDMS matrix. AFM was employed to collect topography images and measure roughness of the scanned samples by showing a brighter phase and a darker phase, with the brighter phase showing a higher modulus than the darker phase [ 42 ]. The AFM topography images of the PDMS with SiO 2 composite film presents the dark areas as PDMS and the bright areas as SiO 2 ( Figure 8 ). The neat PDMS, PDMS/SiO 2 composite with 3 and 5 wt.% SiO 2 exhibited surface roughness in increasing order at nanoscale polymeric structures, according to the AFM results. Figure 8 illustrates the AFM result, which indicates a roughness value of less than 10 nm for the neat PDMS and the surface roughness of the film increased by adding SiO 2 particles. Moreover, the relative modulus of the polymeric film was also improved when 3 wt.% SiO 2 , and 5 wt.% of SiO 2 was added to the PDMS matrix. 3.2. PDMS/PU with SiO 2 Composite 3.2.1. Morphologies and Compositions of the PDMS/PU with SiO 2 Composite From the results of PDMS with SiO 2 composites in the previous section, the optimum wt.% of SiO 2 for improving the mechanical properties of PDMS was 3 wt.% of SiO 2 , so this was employed to further improve the PDMS/PU (95:5) blend film. It gave the lowest number of barnacle attachments, low bonding strength of barnacles and high contact angle of the film after fabricating with micropatterning, using the soft lithography process from previous work [ 10 ]. The main objective of using SiO 2 particles in a low viscous liquid polymer blend was ease of the process by improving surface wettability and surface strength. The surface morphology of the PDMS/PU (95:5) with SiO 2 composite was observed with field-emission scanning electron microscopy (FESEM). The PDMS/PU (95:5) film without SiO 2 filler clearly illustrated homogenous distribution of PU in PDMS phase, in which the average diameter of dispersed polyurethane particles was about 8.3 ± 5.6 µm. Nonetheless, PDMS/PU (95:5) with SiO 2 composite showed that silica particles caused decrease in size of PU droplet diameter and resulted in a finer dispersion of PU in the PDMS matrix to around 2.7 ± 1.7 µm (see Figure 9 ). It was confirmed that the addition of SiO 2 filler modified and improved PU compatibility with the hydrophobic PDMS matrix [ 32 ]. Energy-dispersive X-ray spectroscopy (EDX) mapping spectrum of PDMS/PU (95:5) with SiO 2 composite indicated the characteristic elemental distribution (C, O and Si) (see Figure 10 ). It was observed that the oxygen (O) weight percentage in the PDMS/PU with SiO 2 composite increased from 14.24 wt.% to 31.68 wt.% and weight percentage of carbon atoms decreased from 53.47 wt.% to 32.27 wt.% with the addition of SiO 2 into the PDMS/PU (95:5) blend. It was also illustrated that when SiO 2 was added to the PDMS/PU blend, the silicon (Si) weight percentage of the surface increased from 32.29 wt.% to 36.05 wt.%, as shown in Table 2 . 3.2.2. Water Contact Angle (WCA) of the PDMS/PU (95:5) with SiO 2 Composite The effect of the addition of SiO 2 on the surface wettability of PDMS/PU (95:5) with SiO 2 composite is shown in Figure 11 . The surface characteristic of the neat PDMS film was hydrophobic, with a WCA value of about 107.6° ± 1.7°, and the surface of neat PU was hydrophilic, with a WCA value of about 86.9° ± 3.2°. When PDMS was blended with PU, it was discovered that the PDMS/PU blended (95:5) film showed higher WCA than the neat polyurethane film at 103.4° ± 3.8° [ 39 ]. Moreover, with the addition of the SiO 2 in PDMS/PU (95:5) with 3 wt.% SiO 2 composites, the water contact angle (WCA) value of the composite surface increased to 112.62° ± 4.9°. This increased WCA might have been caused by the addition of SiO 2 particles to the polymer composite film which enhanced surface roughness of the composite, resulting in increased WCA. 3.2.3. Mechanical Properties Study of the PDMS/PU (95:5) with SiO 2 Composite The neat polydimethylsiloxane film demonstrated a Young’s modulus of 1.50 ± 0.18 MPa. After blending with polyurethane, mechanical properties of the PDMS/PU (95:5) polymer blend films were improved, due to the introduction of rigid polyurethane. From the graphs, it can be observed that the Young’s modulus of the PDMS/PU blend (95:5) film increased to 1.75 ± 0.35 MPa, which was superior to the neat PDMS. Subsequently, addition of 3 wt.% SiO 2 to the PDMS/PU with SiO 2 composite increased the Young’s modulus to twice that of the neat PDMS (see Figure 12 ). These results showed that PDMS/PU (95:5) with 3 wt.% SiO 2 improved the mechanical properties of the PDMS/PU composite film making it more suitable for subsequent fabrication with the soft lithography process to impart micropatterns on the surface. 3.3. Characteristics of PDMS/PU (95:5) with 3 wt.% SiO 2 Composite Film with Micropatterns Fabricated by Soft Lithography Process 3.3.1. Morphology of PDMS/PU (95:5) with 3 wt.% SiO 2 Composite Film with Micropatterns Fabricated by the Soft Lithography Process The soft lithography process was used to impart microstructures or micro patterns on the surface of PDMS/PU (95:5) with 3 wt.% SiO 2 composite film to further improve the hydrophobic characteristics of the polymer composite film. SEM was used to examine the microstructures or micropatterns on the PDMS/PU (95:5) with 3 wt.% SiO 2 composite surface (see Figure 13 ). The top view of the composite film revealed consistent pillar structures on the surface, indicating that the soft lithography process was able to further modify the PDMS/PU (95:5) with 3 wt.% SiO 2 composite film. 3.3.2. Water Contact Angle (WCA) and Oil Contact Angle (OCA) of PDMS/PU (95:5) with 3 wt.% SiO 2 Composite Film with Micropatterns Fabricated by Soft Lithography Process Surface wettability of PDMS, PDMS/PU (95:5) with 3 wt.% SiO 2 composite film and PDMS/PU (95:5) with 3 wt.% SiO 2 composite film with micropatterns fabricated by soft lithography process are shown in Figure 14 . The surface of the neat PDMS film surface gave WCA of about 107.6° ± 1.7° and OCA of about 57.2° ± 5.4°. However, PDMS/PU (95:5) with 3 wt.% SiO 2 composite film showed slightly higher WCA and OCA than the neat PDMS film at about 112.6° ± 4.9° and 71.7° ± 5.4°. PDMS/PU (95:5) with 3 wt.% SiO 2 composite film with micropatterns fabricated by soft lithography process gave WCA and OCA of 122.1° ± 2.9° and 90.8° ± 3.6°. The results showed that the soft lithography process could further improve surface properties of the polymer composite film, making it more suitable for antifouling application."
} | 7,485 |
28301612 | null | s2 | 7,216 | {
"abstract": "The Gold King Mine spill in August 2015 released 11 million liters of metal-rich mine waste to the Animas River watershed, an area that has been previously exposed to historical mining activity spanning more than a century. Although adsorption onto fluvial sediments was responsible for rapid immobilization of a significant fraction of the spill-associated metals, patterns of longer-term mobility are poorly constrained. Metals associated with river sediments collected downstream of the Gold King Mine in August 2015 exhibited distinct presence and abundance patterns linked to location and mineralogy. Simulating riverbed burial and development of anoxic conditions, sediment microcosm experiments amended with Animas River dissolved organic carbon revealed the release of specific metal pools coupled to microbial Fe- and SO"
} | 207 |
36134938 | PMC9496521 | pmc | 7,218 | {
"abstract": "Controlled, reversible attachment is widely spread throughout the animal kingdom: from ticks to tree frogs, whose weights span from 2 mg to 200 g, and from geckos to mosquitoes, who stick under vastly different situations, such as quickly climbing trees and stealthily landing on human hosts. A fascinating and complex interplay of adhesive and frictional forces forms the foundation of attachment of these highly diverse systems to various substrates. In this review, we present an overview of the techniques used to quantify the adhesion and friction of terrestrial animals, with the aim of informing future studies on the fundamentals of bioadhesion, and motivating the development and adoption of new or alternative measurement techniques. We classify existing methods with respect to the forces they measure, including magnitude and source, i.e., generated by the whole body, single limbs, or by sub-structures. Additionally, we compare their versatility, specifically what parameters can be measured, controlled, and varied. This approach reveals critical trade-offs of bioadhesion measurement techniques. Beyond stimulating future studies on evolutionary and physicochemical aspects of bioadhesion, understanding the fundamentals of biological attachment is key to the development of biomimetic technologies, from soft robotic grippers to gentle surgical tools.",
"introduction": "1. Introduction Controlled reversible attachment is a key adaptation across diverse terrestrial animal groups that exhibit various locomotory modes and encounter complex three-dimensional environments. Sticking to vertical or overhanging substrates requires a combination of strong adhesion (i.e., attachment force perpendicular to a substrate) and strong friction (i.e., attachment force parallel to a substrate) [ 1 ]. Among spiders, insects, tree frogs, and geckos, various versatile attachment strategies have evolved. The adhesive pads on the limbs of geckos and spiders rely on what is commonly referred to as ‘dry’ adhesion, thought to be dominated by weak intermolecular forces [ 2 , 3 , 4 ], while those of insects and tree frogs are believed to rely also on what is referred to as ‘wet’ adhesion—liquid-mediated interactions, such as capillary and viscous forces [ 5 , 6 , 7 ]. In addition to the adhesive pads on their limbs, animals may utilize other body parts to control or aid their attachment, such as generating friction through other tarsal segments in insects [ 8 ] or through the belly in tree frogs [ 9 , 10 ], or using claws to mechanically interlock with asperities on substrates [ 11 , 12 ]. These mechanisms have been studied in animals that vary in size across several orders of magnitude—from insects and spiders of a couple of milligrams to geckos and tree frogs of several hundreds of grams in mass [ 13 ]. Some animals can rapidly establish and reverse attachment, with stride frequencies of up to 10 steps per second for geckos or even 100 steps per second for mites [ 8 ]. To achieve such rapid reversibility, animals presumably control the strength of their attachment via shear-sensitive adhesive pads and control peeling by varying the angle between their limb and the substrate [ 8 , 14 , 15 ]. Furthermore, there is increasing evidence that shearing and peeling also contribute to self-cleaning during locomotion [ 16 , 17 , 18 , 19 ]. The fundamental understanding of rapid and reversible attachment of biological systems can inform many biomimetic applications that benefit humans in daily life. Reversible adhesion finds applications in sticky tapes, robotic grippers [ 20 , 21 , 22 ], and climbing robots [ 23 , 24 , 25 ]. The development of surgical tools may be inspired by the strategies and mechanisms used by animals, specifically for the manipulation of delicate and slippery tissues inside the human body [ 26 , 27 , 28 ]. Other applications can be found in agriculture and architecture, such as the development of grippers for autonomous harvesting robots [ 29 ], protecting crops from animal pests [ 30 , 31 ], improving pollination of flowers [ 32 , 33 ], protecting buildings from termites [ 34 ], or safeguarding people from disease vectors such as mosquitoes and ticks [ 35 , 36 , 37 ]. Accurate measurements of adhesion and friction forces are crucial for unravelling the fundamental mechanisms of biological adhesion, or bioadhesion. In order to understand and transfer the underlying principles of bioadhesion into biomimetic applications, physicochemical models of attachment need to be developed and validated against experimentally measured attachment forces, or derived parameters such as normal or shear stresses. As adhesive forces correlate strongly with contact area [ 38 ], normalizing adhesion forces to average adhesive stresses using contact areas provides a scale-independent representation of adhesive capacity [ 3 ]. However, measuring these parameters accurately poses a number of challenges. To measure maximum adhesion and friction performance, one needs to detach the animal from a substrate through external forcing. These external forces can be applied globally, as a field, like gravitational or centrifugal forces, or locally by pulling on parts of the animal, for example through a tether. Such forces can be applied to the entire animal, one of its organs, or its sub-structures. In force measurements of live animals, behavior needs to be considered. When an animal moves freely it might employ behavioral strategies that are different than when it is perturbed, constrained, or sedated. Isolating individual limbs (or sub-structures) can help to control for animal behavior; however, extrapolating measurements on a single limb to the whole animal may lead to errors due to assumptions and oversimplifications. For example, in some animals, it has been found on the limb-level that larger adhesive pads generate stronger adhesion per unit area [ 13 , 39 , 40 , 41 ], which, however, may be explained through behavioral adaptations on the whole-organism-level (i.e., active shearing of the pad for adhesion control; [ 41 ]). Given the many parameters that can influence adhesion and friction, such as temperature, humidity, and substrate properties, as well as the hierarchy of biological attachment devices ( Figure 1 ), many factors need to be considered in the design of a bioadhesion study. In this review, we give an overview of the methods used for measuring contact forces in animal attachment studies, and discuss their trade-offs. This review limits itself to methods used in studies on terrestrial animals because they have direct implications for applications that humans encounter in their daily, (mostly) terrestrial life. However, many of the methods presented here are also used in studies on aquatic bioadhesive systems. We conclude this review with a novel perspective on force measurement methods focusing on force magnitudes and how they are generated by and/or applied to the animal. To this extent, we will review the most-used force measurement methods considering whole animals ( Figure 1 A), isolated limbs ( Figure 1 B), and their sub-structures ( Figure 1 C), and whether the animals experience global or local forcing. Additionally, we address relevant parameters that can be measured, controlled, and varied in the different methods. This overview provides guidance for scientists that are new to the field of bioadhesion, and presents key challenges in measurement methodology that need to be overcome to advance the field. To assist those new to the field, we also provide a glossary of technical terms at the end.",
"discussion": "3. Discussion In the previous section, we presented a broad overview of existing methods to study the attachment of terrestrial animals (see Table 1 for a summary). When deciding on a method for a new study, one should consider a few questions. What parameters need to be measured (e.g., force, contact area, stress)? What are the magnitudes of the parameters to be measured? Is the method suitable for the animal of interest? Does the method provide the freedom to choose and/or vary experimental conditions (e.g., substrate characteristics)? Does the method limit the behavior of the animal? Are there alternative methods available for the study? In this section we present relevant considerations when selecting a method. First, we consider some of the limitations of the most prevalent methods with respect to scale and subject, e.g., species and body part. Then, the trade-offs in selecting a method for a study are discussed. Lastly, we present outlooks for future development and the general implications of animal adhesion studies in science and society. 3.1. Limitations When deciding on a method, it is critical to consider the size of the animal and magnitudes of the attachment forces it can generate. Figure 4 shows a regime map of the most common adhesion and friction force measurement methods. Only AFM, 2D (biaxial) FTs, tethers, rotation platforms, and force centrifuges are included. Force platform studies are excluded because they include both whole animal and limb measurements, as well as 1D, 2D, and 3D force measurements, and so are difficult to compare. To our knowledge, there are insufficient previous studies ( n < 3 ) available in the literature to make meaningful estimates of the regimes of photo-elastic gelatin and optic tactile sensors. However, their limitations were discussed in the previous section. In Figure 4 , measured force is plotted against subject mass as reported in the reviewed studies. The data shows two distinct trends: (1) whole animal studies follow constant safety factor (SF) lines, and (2) body part measurements are limited by sensor precision. The measurable ranges are also outlined in Table 1 and in more detail in the foregoing section. As noted before, rotation platform limits are explained well by the animal’s SF, which should be considered during the design of an experiment. For the other whole animal force measurement methods, tethers and force centrifuge measurements, SF bounds are suggested as well by the reviewed studies. Tethered studies are not effective when SF < 1 , since animals that can not sustain their own weight through friction will likely start slipping when pulling their own body weight. There is a considerable overlap between tethers and force centrifuge studies, suggesting both are capable of studying the same animal species, and expected SF or animal weight does not need to be considered when choosing between the two. However, Federle et al. [ 57 ] report higher adhesion forces for ants when measured using a centrifuge compared to a tether. They speculate that tethers (i.e., local forcing) may affect an animal’s posture and natural response more than a centrifuge (i.e., global forcing). Considering body part measurements, there is a clear gap between AFM and FTs. The bounds of these methods are set by sensor limitations. Reviewed studies and sensor limitations suggest a gap in the {1 µN, 10 µN} range, above the maximum flexible probe range of AFM and below the sensor precision of FTs. When forces in this range are expected, extra consideration should be taken in designing the measurement setup. Notably, both methods are suitable for any type of animal and are not limited by animal weight because these methods are used to investigate limbs or their sub-structures. 3.2. Trade-Offs in Study Design In addition to animal size and expected force magnitudes, there are other factors to consider when deciding on a method to measure bioadhesion. First, one needs to determine if measurements should be carried out on whole animals or their limbs or sub-structures. Measurements with whole animals are influenced by behavior (e.g., motivation) and body kinematics (e.g., posture). However, investigating behavior may shed light on the postures and kinematics that animals use to promote attachment. For example, observations on tree frogs found that when attaching to overhanging substrates they spread their limbs away from their body to presumably minimize the angle between their limbs and substrate to prevent peeling [ 15 ]. While some behaviors promote attachment, there are others that may hinder it. Bioadhesion measurements only work when animals attach to substrates and do not jump or fly away. Insects capable of flight may need to be incapacitated by gluing or trimming their wings to prevent escape. In their study with moths, Al Bitar et al. [ 59 ] had to cut the insects’ wings to prevent them from fleeing during measurements using force centrifuges. Such modifications allow measuring of attachment forces, but may affect the animal’s behavior and response to external stimuli. For fundamental studies into the physicochemical basis of attachment, bioadhesion measurements are best carried out with individual limbs or their sub-structures, where animal behavior can be controlled for. These measurements enable control over kinematics and mechanics, and thus may provide a deeper insight into the mechanisms underlying the generation of adhesion and friction. For example, previous work using individual limbs has found that the adhesive pads of geckos, tree frogs, and insects are shear-sensitive and generate increased adhesion under enhanced shear loading [ 8 ]. The linear relationship between shear force and adhesive force would be impossible to observe with whole animals. By working with individual limbs and biaxial FTs, the shear forces were controlled while adhesive forces measured. In another example, the adhesive forces generated by a single gecko seta were carefully measured using AFM [ 2 ]. Then, the measured forces were compared with predictions from an analytical model of van der Waals forces (i.e., the interaction forces between the molecules on the seta and the substrate) to test if such intermolecular forces underpin gecko adhesion [ 3 ]. This finding motivated the development of gecko-inspired, micro-structured adhesives that stick without glue by also exploiting van der Waals forces [ 89 ]. Therefore, bioadhesion studies using limbs or their sub-structures have the potential to generate fundamental knowledge of great importance for the design of biomimetic adhesives. As stated in the introduction, in order to measure attachment performance, an animal needs to experience an external force that works against the adhesion and friction it can generate. This external force can be applied globally, as a field, or locally, and the way it is applied can significantly influence the study. Global forcing is typically done using gravitational or centrifugal forces. These force fields act on the whole animal uniformly and simulate the forcing that an animal may experience when attaching to vertical or overhanging substrates. Local forcing acts on individual body parts. While such forcing is not typically experienced by animals in day-to-day life, it enables the isolation of individual limbs (and their sub-structures) and provides minimalist ways to measure maximum attachment performance, e.g., tethered studies require only a thin wire and force sensor. Finally, the parameters that need to be measured and controlled, i.e., the dependent and independent variables, respectively, should be identified. Table 1 outlines the dependent and independent variables that were measured and controlled in previous studies. Based on this, tethers, force transducers, and AFM are the most versatile methods. They enable variation and control of independent variables, especially substrate properties and interaction kinematics as well as mechanics. Force platforms and optical methods are the most limited with respect to independent variables. This is primarily because the substrates cannot be controlled or varied due to requirements dictated by the methods, e.g., force platforms have sensors embedded and optical methods require substrate transparency. 3.3. Beyond Adhesion and Friction Measurements While this review focuses primarily on techniques used for measuring forces, there are other parameters that need to be measured to fully grasp the attachment of a given animal. Theoretical models of contact mechanics and attachment can help identify underlying physicochemical mechanisms, but require validation through comparisons with experimental observations. Typically, the models predict adhesion and friction forces that can be compared to measured values; however, the models also depend on additional parameters as inputs. One particularly important parameter needed in theoretical models of contact mechanics is the distance between the adhesive pad (or its sub-structures) and substrate. The magnitude of this distance could help determine which types of interactions are dominant or negligible. For example, for 10- μ m spherical particles under dry conditions, electrostatic forces from the net charge on the particles dominate for distances greater than 100 nm, electrostatic forces from local charge patches dominate for distances between 10 and 100 nm, and van der Waals forces dominate for distances less than 10 nm [ 90 ]. Furthermore, if there is fluid present, measuring fluid film thickness can help determine if the fluid acts like a lubricant or enhances friction. These distances can be measured through interference reflection microscopy (IRM). This technique was first developed to measure how close cells are to substrates [ 91 ], but was later used with tree frog toe pads [ 76 , 92 ]. In tree frogs, it was found that while mucus is present on the toe pads, parts of the surface features on frog toes are in quasi-direct contact with the substrate, with separation distances between 0 and 35 nm [ 76 ], indicating a potential contribution of van der Waals forces or other ‘dry’ interactions in tree frog attachment. Additionally, it was found that there is an intermediate fluid film thickness (∼200 nm) that enhances friction compared to a fully wet (lubricating) or fully dry state [ 92 ]. Fluids covering the contact surface are an inherent part of many bioadhesive systems. For example, tree frog toes—as the whole amphibian body—are covered with a watery mucus [ 93 ], and insects secrete a viscous emulsion onto their adhesive pads. While these fluids help to prevent skin and cuticle from drying out and may have anti-bacterial and anti-fungal properties [ 94 , 95 ], their implications in bioadhesion are still being investigated. The physical and chemical properties of these fluids have been measured using various techniques. To measure the fluid’s viscosity, methods were adopted from the field of rheology. For tree frog mucus, laser optical tweezers were used to measure the viscous force exerted on a trapped particle by the mucus [ 76 ]. The viscosity of insect pad fluid was measured by placing small tracer particles in a drop of the fluid and recording the dampening of the particle’s Brownian motion (or thermal fluctuations) through the fluid’s viscosity [ 96 ]. For chemical characterisation of the fluid, several techniques have been used. In tree frogs, cryo-histochemistry, attenuated total reflectance-infrared spectroscopy, and sum frequency generation spectroscopy have been used. From the measurements, it was found that the mucus on the toe pads is chemically similar to the mucus secreted by other body parts, including the belly [ 93 ]. In insects, gas chromatography and mass spectrometry have been used to characterize the chemical composition of their secreted fluids [ 97 ]. From this characterization, it was found that, like in tree frogs, the fluid secretions on the adhesive pads are chemically similar to those secreted throughout the rest of the body [ 98 ]. Surface tension is another important physical property of a bioadhesive fluid, as the capillary forces associated with it can be dominant at small spatial scales. However, to our knowledge, this property so far has been measured only indirectly through contact angle measurements [ 58 , 99 , 100 ]. Contact angle, or the angle between the substrate and fluid meniscus, quantifies the ‘wettability’ of a fluid on a substrate. For insects and tree frogs, this contact angle has been found to be quite small (~10°) on a wide variety of substrates, so the adhesive fluid appears to be highly wetting regardless of substrate chemistry [ 58 , 99 , 100 ]. Recent studies of insects have made assumptions of the surface tension of the fluid given that it is comprised of hydrocarbons [ 101 , 102 ]. This assumed, approximate value sufficed for these studies since the models provided leading order analyses of the capillary interactions. For more detailed and accurate models, direct characterization will be required. The material properties of the pad tissues, setae, or spatulae are also important for understanding bioadhesion. Animals stick to a wide variety of substrates, including smooth and rough ones. For rough substrates, the adhesive pads should conform to asperities in order to form a large area of close contact. A pad’s ability to conform to rough substrates is dictated by its physical properties, especially its stiffness or Young’s modulus. This property can be measured using micro- or nano-indentation, where the adhesive pad is compressed by a small probe and its stress response is measured, or using optical techniques, like confocal laser scanning microscopy [ 103 ]. Using such techniques, it has been found that setae on the adhesive pads of beetles are stiffer at the base and softer at the tip [ 103 ]. Similarly, the smooth adhesive pads of insects exhibit softer tissues in the outer layers and stiffer tissues underneath [ 104 ]. On the other hand, for tree frogs, it was found that the outer layers of the toe pads are stiffer than internal tissues [ 105 , 106 ]. Pad stiffness not only influences conformability, but may also affect the strength of adhesion. Classical experiments measuring the adhesion between a spherical indenter and flat substrate found that adhesion increases with material stiffness [ 107 ]. Similarly, the attachment force of fiber-reinforced adhesives such as gecko toes is proportional to the tensile stiffness of the fiber-reinforcement [ 108 ]. Therefore, there seems to be a trade-off between having soft pad tissues to conform to rough substrates and having stiff tissues to generate strong adhesion. In geckos and tree frogs, blood sacks have been observed immediately underneath the adhesive skin surface. Blood pressure may be controlled in these sacks to help tune pad stiffness [ 106 , 109 ]. Having such control could enable geckos and tree frogs to easily conform to rough or non-flat substrates using soft tissues and then stiffen the tissues to promote strong adhesion. A similar mechanism has been exploited by synthetic adhesives that use phase changing liquid metals [ 110 ]. AFM is a very versatile method that allows more than just contact force measurements. Many studies that investigate the effects of substrate properties on attachment use AFM to measure roughness, or to image surface sub-structures. Alternatives for measuring surface roughness of biological samples, such as scanning electron microscopy (SEM), are prone to artefacts from the preparation steps, such as shrinkage or drying, and are not suitable for living animals [ 105 ]. AFM can also be used for indentation experiments. Micro-indentation using FTs with a motorized stage is sufficient for larger structures, such as whole tree frog toes [ 106 ]. However, for smaller structures, AFM is required, for example to measure the stiffness of epithelial cells and local friction profiles over single pillars on tree frog toes [ 88 ], or the stiffness of the adhesive tarsal setae of ladybird beetles [ 103 ]. While the physical and chemical properties of adhesive pads and their fluid secretions are important for developing physicochemical models of adhesion and friction, the ways in which contact is established and released, i.e., pad and limb kinematics, can significantly influence attachment and detachment. Previous work has found that animals may be able to control adhesion by varying shear forces [ 8 , 65 ]. In addition to controlling adhesion via shear, tree frogs have been observed to spread out their limbs away from their body in response to increased loads [ 15 ]. By spreading their limbs, they not only promote shearing but also decrease the angle between their limbs and substrate. Just like in sticky tapes, minimizing this angle may prevent peeling. For insects, it has been found that attachment and detachment occur at different time scales [ 111 , 112 ]. Specifically, adhesive pads move quicker during detachment, which is believed to help conserve the secreted fluid. A faster separation velocity ensures that less fluid is deposited on the substrate. Additionally, a slower approach during attachment may help generate intimate contact and reduce the gap between pad and substrate to increase adhesion and friction forces [ 112 ]. 3.4. Perspectives Based on the reviewed data, we could map established force measurement methods to show their effectiveness and limitations, as summarized in Table 1 and Figure 4 . From this analysis, we find that studying attachment for the large and slow no longer poses a problem. The frontier lies at the small and fast. Measuring small and fast processes still poses a considerable challenge given the trade-offs in spatial and temporal resolutions for cameras and sensors. There is renewed interest in optical methods during the past decade [ 9 , 10 , 47 , 51 ]. With visual data processing technologies, data storage and transfer capacities, and optic systems ever improving, optics-based methods seem promising, like the optic-tactile sensor developed by Eason et al. [ 52 ] to directly measure adhesive stress. Quantifying adhesive and frictional stresses can help reveal the true performance of biological adhesives, since it provides a scale-independent measure of adhesion and friction and captures the exact contact stress distribution. Typically, adhesive pads are asymmetric and limbs are rarely oriented completely parallel or perpendicular to a substrate; therefore, forces are applied with offsets that induce moments and cause imbalances in contact stress distribution. Direct measurements of contact stress distribution can pinpoint where stress concentrations occur to reveal how the adhesive may fail and how limb kinematics influence adhesion and friction. However, to our knowledge, optic-tactile sensors are the only ones capable of contact stress measurements at the moment. Measuring adhesive and frictional stresses across various animals could contribute significantly to our understanding of the scaling of adhesive performance in biological systems [ 13 ]. In this review, we have largely skipped over micro-electromechanical sensors (MEMS). Interest in MEMS for measuring attachment seemingly faded in the past decade, but MEMS might be key in exploring the realm of fast and small. A MEMS force plate for studying insect locomotion developed by Bartsch et al. [ 113 , 114 ] has barely been cited in actual animal studies. The same holds for a biaxial MEMS cantilever design by Lin & Tramer [ 115 ]. This raises the question: is MEMS irrelevant to bioadhesion research, or have developments in MEMS design gone unnoticed in bioadhesion research? Bioadhesion has always been a fascinating subject to study for biologists and engineers alike. Their work over the last decades resulted in various insights into these remarkable mechanics, attracting an ever-increasing interest from various other disciplines. Electrical engineers, (soft) roboticists, medical engineers, material scientists, and ecologists all benefit from discoveries in bioadhesion and work to tackle multidisciplinary problems, such as protecting honey bees, preventing animal pests, or developing new soft grippers for various applications."
} | 6,985 |
33927732 | PMC8076870 | pmc | 7,220 | {
"abstract": "Microalgae are used as a source of lipids for the production of biofuels. Most algae produce neutral lipids under stress conditions. Here, lipid accumulation by the unicellular alga Chlamydomonas reinhardtii was examined during cultivation under iron-limiting conditions. Severe iron stress caused the cells to accumulate a significant amount of lipid, specifically triacylglycerols (TAGs), by compromising the growth. Semi-quantitative measurements by Fourier transform infrared (FTIR) spectroscopy showed an increase in both carbohydrate and lipid content in iron-stressed C. reinhardtii cells compared to control. Analysis by flow cytometry and thin layer chromatography confirmed that severe iron deficiency-induced TAG accumulation to fourfold higher than in iron-replete control cells. This accumulation of TAGs was mostly degraded from chloroplast lipids accompanied by overexpression of diacylglycerol acyltransferase (DGAT2A) protein. Furthermore, liquid chromatography-mass spectrometry (LC-MS) analysis demonstrated significantly enhanced levels of C16:0, C18:2, and C18:3 fatty acids (FAs). These results indicate that iron stress triggers the rapid accumulation of TAGs in C. reinhardtii cells. The enhanced production of these lipids caused by the iron deficiency may contribute to the efficient production of algal biofuels if we escalate to the photobioreactor’s growth conditions.",
"conclusion": "Conclusion Iron (Fe 2+ ) is an essential cofactor for photosynthesis since most photosynthetic complexes contain iron. If algal cells are propagated in a medium depleted of iron, photosynthetic efficiency is reduced, but cells remain viable after 72 h in culture. In response to iron deficiency, C. reinhardtii cells accumulate significant amounts of lipids. Here, we observed that cells were grown in low-iron conditions accumulated lipids to levels up to fourfold higher than iron-replete controls. Transmission electron micrographs showed large LDs formed in cells subjected to the most severe iron starvation. Furthermore, we detected a significantly increased accumulation of TAGs in severe iron deficiency about sixfold higher than the control. We also analyzed cellular FA content, which demonstrated that iron starved cells accumulated more saturated than unsaturated FAs. Notably, saturated FAs are the preferred substrate for biodiesel production. In summary, it indicates that severe iron deficiency hampers growth and triggers the glycolipid breakdown and high TAG accumulation levels. FAs accumulated in TAG are likely derived from the various cellular metabolic networks under iron deficiency. The accumulation of TAG is high in severe iron deficiency conditions which is very important for the feedstock and biodiesel. Hence, this study could be very important for the society to improve the fishery and poultry industries using algae as a feedstock. Therefore, the photobioreactors can enhance the growth and lipid accumulation on an industrial scale that could be important in producing biofuels or aquafeed.",
"introduction": "Introduction Photosynthetic algae capture solar energy and store it as chemical energy in the form of starch and lipids, particularly triacylglycerols (TAGs) ( Merchant et al., 2007 ). The transesterification of TAGs achieves biodiesel production with methanol in the presence of a suitable catalyst (e.g., H 2 SO 4 ). Most microalgae accumulate increased levels of TAGs under nutrient deficiency, such as nitrogen starvation. However, stress conditions like this are also accompanied by decreased growth rates. For the economical production of biodiesel from algae, increased lipid levels must be combined with a high percentage of algal growth. Various stress conditions like nitrogen starvation ( Rios et al., 2015 ; Lin et al., 2018 ), iron starvation ( Gorain et al., 2013 ), copper stress ( Hamed et al., 2017 ), heat stress ( Ördög et al., 2016 ), pH ( Abinandan et al., 2019 ), and high-light stress ( Fan and Zheng, 2017 ) have been reported to induce lipid accumulation in microalgae effectively. The nutritional requirements for the accumulation of TAGs by Dunaliella tertiolecta were investigated ( Chen et al., 2011 ). The soluble form of iron (Fe 2+ ) is very limited in soil. Its low bioavailability is a significant problem for photosynthetic organisms ( Glaesener et al., 2013 ) because iron is essential in the biochemical pathways of plants and microalgae. In oxygenic photosynthesis, iron acts as a cofactor in all-electron transfer reactions and enhances the active photosynthetic reaction centers in algae. A lack of iron decreases photosystem (PS)II function, inter-photosystem electron transport, carbon fixation rates and ultimately affects the growth of cyanobacteria and algae ( Singh et al., 2005 ; Devadasu et al., 2016 ). Limitation of micronutrients (e.g., zinc and iron) results in the conversion of membrane lipids into individual fatty acids (FAs), which leads to the formation of lipid droplets (LDs) in D. tertiolecta ( Chen et al., 2011 ). The impact of N, S, P, and Mg deficiency on microalgal metabolism has been studied in Chlamydomonas reinhardtii ( Çakmak et al., 2014 ). Nitrogen starvation severely inhibits algal growth and induces TAG accumulation in C. reinhardtii cells ( James et al., 2011 ). Iron stress has a wide range of effects on the endoplasmic reticulum (ER) and its membrane structures in this organism, which indicates its potential influence on TAG accumulation. Therefore, we hypothesized that two-stage iron starvation, i.e., cells initially subjected to iron deficiency are further cultured in an iron-depleted medium to cause severe iron starvation, might effectively induce lipid accumulation in C. reinhardtii . This so-far untested treatment regime could offer a simple way of increasing lipid content in microalgae. The eukaryotic microalga, C. reinhardtii , is a model organism used extensively to study photosynthesis and primary metabolism ( Merchant et al., 2007 ). Its genome has been sequenced; it is easy to grow in culture and has a well-characterized response to adverse environmental conditions. Therefore, this organism was selected to investigate the influence of two-stage iron starvation on TAG accumulation. We recently reported the effect of iron stress conditions on cellular biomass and the proteins involved in lipid biosynthesis in C. reinhardtii ( Devadasu et al., 2019 ). In the present study, we examined the effect of two-stage iron starvation on growth rate and lipid production in this species. Cells cultivated in iron-deficient conditions were also examined for carbon partitioning into lipid and carbohydrate. Severe iron starvation conditions significantly increased the production of TAGs, which will help to make biofuel production for more efficient.",
"discussion": "Results and Discussion Influence of Iron Deficiency on the Growth of Chlamydomonas reinhardtii Tris-acetate-phosphate (TAP) medium, the standard growth medium for the culture of C. reinhardtii , contains 18–20 μM iron ( Kropat et al., 2011 ). To examine the effect of iron-deficiency on growth, cells propagated in TAP were washed twice with TAP−Fe (iron-free medium) and then transferred to the same medium or TAP+Fe (iron-replete medium) as a control ( Figure 1A ) to give an optical density (OD 750 ) of approximately 0.2. The growth of these cultures was then monitored over 4 days. In iron-sufficient conditions, the OD was doubled in 2 days and reached a maximum value (OD 750 = 1.01 ± 0.18) after 72 h. In the inoculum, traces of iron were utilized by the cells for initial multiplication in the iron-deficient medium, so genuine iron stress only occurred after the first cell division. The OD of the iron-deprived culture decreased (OD 750 = 0.23 ± 0.02) over the early 2 days, but it then increased over the following 4 days to a maximum value of less than half iron-replete culture (OD 750 = 0.42 ± 0.04). Moreover, since iron is a cofactor in several enzymes such as hydrogen dehydrogenase, nitrite reductase and superoxide dismutase, normal cellular metabolism required to maintain ATP production is depressed in cells lacking iron. FIGURE 1 Growth and biomass measurements of Chlamydomonas reinhardtii cells cultured under control and iron-deficient conditions. (A) The optical density (OD) of control culture [20 μM iron (Fe 2+ ) concentration], compared with those grown under iron deficiency and severe iron deficiency conditions. The statistical analysis was carried out from day 2 to day 6 of control and iron deficiency conditions. (B) Total biomass measured in the same cultures harvested after 4 days of growth. Values are mean ± SD. Statistical analysis was performed by Student t -test (non-paired) and P values were represented as ** P < 0.01; *** P < 0.001. Error bar indicates the standard deviation of three biological replicates. Further, the iron-limited culture (1st generation) was then used to inoculate into the same iron-free medium to see further consequences (severe iron starvation) (2nd generation). In severe iron-deficient culture, the cells were wholly starved due to lack of iron in the medium. During the severe iron starvation, the cells were in the dividing stage, and the photosynthesis process was reduced significantly as they do not have sufficient iron content in the cell ( Devadasu et al., 2021 ). The cell growth was reduced after 4 days (OD = 0.313 ± 0.03) in severe iron deficiency. Total biomass in dw was up to 2.1 g/L in cells grown in iron-replete medium ( Figure 1B ). In comparison, cell biomass in the iron-limited culture was 1.0 g/L, whereas the cultures with severe iron-deficiency were further reduced to 0.45 g/L ( Figure 1B ). Under optimal conditions, microalgae can grow and produce higher biomass with low lipid content. In contrast, nutrient limitation can accumulate high levels of lipid content but with low biomass ( Tan and Lee, 2016 ). A previous study on nitrogen starvation in C. reinhardtii showed that cell growth was severely reduced, causing a decrease in total biomass, whereas increased lipid content ( Park et al., 2015 ). Two major obstacles to the production of biodiesel from microalgae is (1) poor growth in culture leading to low cell density with low cellular lipid content and (2) stress conditions promotes neutral lipid content while decreasing growth ( James et al., 2011 ). Previous reports show that total lipid levels in C. reinhardtii cells increased dramatically by iron starvation ( Urzica et al., 2013 ; Devadasu et al., 2019 ). Since continuous exposure of cells to iron limitation causes plastid and ER stress and is highly likely to result in the degradation of cells (apoptosis). It is tempting to speculate that apoptosis may be involved in this cellular degradation. This process has recently been demonstrated in Chlamydomonas, and severe ER stress leads to apoptosis in many organisms ( Howell, 2013 ). Other reports from nitrogen and phosphate starvation lead to LDs and TAG through apoptotic mechanism ( Couso et al., 2018 ; Masclaux-Daubresse et al., 2020 ) indeed, in our case, also the apoptotic mechanism would have happened. Therefore LDs and TAG could have also occurred through the apoptotic mechanism under severe iron deficiency, and the results can be seen below. Visualization of Intracellular LDs by Confocal and Transmission Electron Microscopy The accumulation of LDs in C. reinhardtii cells was examined using confocal and transmission electron microscopy (TEM) ( Figure 2 ). In iron deficiency, the size of the cells was decreased, and they displayed abnormal morphology. Cells were grown in an iron-replete (control) medium did not show any visible accumulation of lipid bodies ( Figure 2A ). In contrast, lipid production was induced in an iron-deficient medium, and LDs were present in cells in both iron deficiency and severe iron-limited cultures. Our study supports the earlier reports that the accumulation of LDs was seen in iron deficiency conditions ( Urzica et al., 2013 ; Devadasu et al., 2019 ). Interestingly, severe iron deficiency (almost no iron in the cells or medium) caused the appearance of more abundant lipids and larger lipid bodies ( Figure 2A ), which has important implications for industrial applications. Previously, lipid body accumulation was observed in Chlamydomonas cells grown under nitrogen starvation conditions ( James et al., 2011 ). C. reinhardtii cells accumulated lipids up to 45–50% of their total NR fluorescence ( Moellering and Benning, 2010 ; James et al., 2011 ). In the present study, we observed lipid accumulation of 65% after 72 h of severe iron starvation compared with 45% in cells from the iron-limited 1st generation culture. This high lipid content in cells from severely iron starved cultures may make them a valuable starting material for biofuel production. TEM showed decreased unstacked thylakoid membranes in these cells. Also, the LDs observed in cells subjected to severe iron starvation (marked with “L”) were more extensive than those grown in the iron deficiency conditions ( Figure 2B ). FIGURE 2 Microscopic examination is of Chlamydomonas reinhardtii cells cultured under iron-replete and iron-deficient conditions. (A) Confocal fluorescence images of cells stained with Nile Red (NR) show the location and size of lipid droplets (scale bar = 10 μm). (B) Transmission electron microscopy (TEM) images showed lipids accumulated within cells under iron deficiency and severe iron deficiency conditions (scale bar = 1 μm). Symbols represent, Th, thylakoids; L, lipid droplet; N, nucleus; S, starch. Neutral Lipid Analysis by NR Staining It has been shown that nutrient limitation causes decreased cell division in microalgae, and most studied species divert FAs into TAG accumulation ( Sharma et al., 2012 ). To examine the effect of iron deficiency on neutral lipid accumulation at the population level, 10,000 C. reinhardtii cells from each group (i.e., control and iron-deficient) were stained with NR after 72 h in culture, and fluorescence intensity was analyzed using flow cytometry. Many more cells with raised fluorescence levels (threefold) were present in the iron-deficient conditions. The severely iron-deprived culture showed up to sixfold fluorescence values than the control culture ( Figure 3A ). These fluorescence data indicate that cells subjected to severe iron deficiency are more stressed than control cells, which induces significant lipid accumulation ( Figure 3A ). Therefore, it is probable that the iron deficiency-induced lipid accumulation and the nitrogen starvation-induced pathway of LD formation are similar and involve the degradation and remodeling of chloroplast membrane lipids to form TAGs ( Siaut et al., 2011 ). Notably, biomass was decreased while NR fluorescence increased when cells were grown in severe iron-depleted conditions for up to 72 h ( Figure 1B ). Decreased biomass and increased lipid content were observed in both iron deficiency stages, but higher levels of lipid were accumulated under severe iron starvation. The effect of iron limitation on lipid accumulation was observed from cell growth, but lipids’ accumulation was highest at fourth-day cultures ( Figure 3B ). These results indicate that the metabolism of iron and carbon are interlinked in C. reinhardtii . Neutral lipid accumulation in iron-deficient conditions occurs primarily through photosynthetic carbon fixation via the Calvin-Benson cycle ( Velmurugan et al., 2014 ; Devadasu et al., 2019 ). FIGURE 3 Flow cytometry (FACS) and total lipid content analysis of Chlamydomonas reinhardtii cells cultured under iron-replete and iron-deficient conditions. (A) The mean fluorescence intensity of neutral lipids stained with NR was measured by FACS. (B) Total lipids recovered by organic extraction were quantified in cells over 4 days of growth. Values are mean ± SD. Statistical analysis was performed by Student t -test (non-paired) and P values were represented as ** P < 0.01; *** P < 0.001. Error bar indicates the standard deviation of three biological replicates. Neutral lipid content of cells during growth under iron-replete and -deficient conditions was monitored by staining with NR. Samples of culture were taken throughout each experiment (12–72 h), and cell densities were normalized so that an equal number of cells (3 × 10 6 ) were stained ( Supplementary Figure 1 ). As in previous studies, NR staining was employed to evaluate the neutral lipid content of C. reinhardtii cells and characterize the intracellular lipid bodies ( James et al., 2011 ). Based on plate reader assay after 72 h of cell growth, the 1st and severe iron-deficient cells exhibited twofold and threefold increases in neutral lipid content, respectively, compared with the iron sufficient control cells. A previous study on C. reinhardtii cells cultured under N, S, and P limitation conditions also found an increased neutral lipid content ( Çakmak et al., 2014 ). In our earlier investigation of iron limitation in C. reinhardtii (traces of iron still present), cells’ neutral lipid content was increased ( Devadasu et al., 2019 ). Similarly, TAG accumulation was observed when C. reinhardtii cells grown in iron deficiency ( Urzica et al., 2013 ). The total lipid content of cells was also determined following organic extraction and the results compared with NR staining data ( Figure 3B ). The level of lipid accumulation caused by severe iron deficiency was higher than that resulting from the first generation of iron deficiency ( Urzica et al., 2013 ; Devadasu et al., 2019 ). Therefore, severe iron starvation conditions could be a valuable means of increasing the yield of neutral lipids in microalgae, which would benefit biodiesel production. Lipid and Carbohydrate Measurements by FTIR To corroborate the NR fluorescence and total lipid data, we used FTIR to monitor carbon partitioning between lipid and carbohydrate in C. reinhardtii cells cultured under iron-deficient conditions ( Figure 4 ). Based on the vibrational stretches due to peptide bonds, carbohydrates and lipid molecules, we plotted lipid ratios: amide I and carbohydrate: amide I, according to previous reports by Dean et al. (2010) . Previous reports have demonstrated that microalgae are grown under N, S, and P deprivation display decreased protein content ( Kilham et al., 1997 ; Cakmak et al., 2012 ). Our results also showed a drastic decline in C. reinhardtii cells’ protein content under 2nd stage severe iron starvation compared with 1st stage iron limitation ( Figure 4 ). The FTIR spectra ( Figure 4A ) showed a strong absorption peak for the carbohydrate region (C–O–C) compared with weaker absorption for the protein amide I and amide II bands (C=O and N–H, respectively), and also for the lipid band (CO). The decrease in protein content produced by other elemental deprivations suggests that photosynthetic energy is used to synthesize more lipid and carbohydrate for storage, which correlates with our data from neutral lipid and carbohydrate measurements ( Figures 3 , 4 ). Carbon storage as lipid and carbohydrate was measured in cells cultured under all conditions after 72 h. In the 1st stage of iron-deficiency, cells showed excess storage of carbon as starch due to increased operation of the Calvin-Benson cycle of carbon fixation as previously reported ( Devadasu et al., 2019 ). This resulted in a substantial increase in the carbohydrate: amide I ratio (fourfold) ( Figure 4B ). The decrease in carbohydrate content in the severe iron deficiency may be due to carbon fixation toward lipid synthesis for energy storage. Lipid content was increased in both 1st and severe iron deficiency cells after 3 days of growth: the lipid: amide I ratio increased twofold in the former while a comparatively moderate increase of threefold occurred in the latter ( Figure 4C ). FIGURE 4 Analysis of carbon storage in Chlamydomonas reinhardtii cells cultured under control and iron-deficient conditions by Fourier transform infrared (FTIR). (A) Representative FTIR spectra of cells. (B) Lipid: amide I ratios over the course of 4 days in culture. (C) Carbohydrate: amide I ratio for the iron-deprived culture. Three independent cultures propagated in each medium were examined, and Error bars represened the means ± SD ( n = 3). The significance values were compared with the control and they represented as *** P < 0.001 for both iron deficiency conditions in panel (B) ; *** P < 0.001 for iron deficiency condition; ** P < 0.01 value for severe iron deficiency condition in panel (C) . In contrast, cells from the severe iron deficiency culture with 1st stage of iron deficiency showed a decrease (twofold) in the carbohydrate: amide ratio after 3 days of growth compare to iron sufficient condition (control) ( Figure 4C ). The results of a previous study suggest that the disassociation of glycolipids [monogalactosyldiacylglycerol (MGDG) and digalactosyldiacylglycerol (DGDG)], into their individual FAs and these FAs recycling to contributes to TAG accumulation in C. reinhardtii ( Urzica et al., 2013 ). Similary, in nitrogen starvation the chloroplast membrane lipids have converted to TAG accumulation in C. reinhardtii ( Yang et al., 2018 ). We assume that FAs dissociated from glycolipids enter the Kennedy pathway for the synthesis of TAGs. An analysis of lipids extracted from cells from iron-deficient cultures indicated that polar lipids, especially MGDG and DGDG, are degraded more rapidly in these conditions ( Supplementary Figure 2 ). Therefore, it is likely that these two major class of lipids of FAs are diverted to participate in the formation of TAGs. Interestingly, the other polar lipids, DGTS and SQDG, are also decreased in severe iron deficiency, and possibly these galactolipids could have dissociated into individual FAs. A previous study on the effects of N supplementation showed growth, and it produced enhanced cell growth in microalgae and decreased intracellular lipid content ( Huang et al., 2013 ). Iron-induced TAG accumulation was observed following the alteration of chlorophyll content, indicating that iron deficiency affects chloroplast membranes without influencing lipid accumulation. These results are corroborated by TEM images of cells showing disturbance of thylakoid stacks ( Figure 2B ), which indicates that membrane lipids were converted to neutral lipids. It may be concluded that FTIR is a rapid and accurate means of confirming NR staining results when determining the quality of microalgae as a biodiesel feedstock. Thin-Layer Chromatography Analysis of Accumulated TAGs Previous studies showed that C. reinhardtii gets TAGs when grown under micronutrient limitation ( Deng et al., 2011 ; Kropat et al., 2011 ). Another report showed epecifically, the TAG accumulations were observed in iron deficiency ( Urzica et al., 2013 ). However, the accumulation of TAGs in severe iron deficiency is not known so far. Following in severe iron deficiency conditions caused enhanced lipid production by C. reinhardtii cells under photoheterotrophic conditions, we next semi-quantified the accumulated TAGs using TLC. Total lipids were extracted from cells following growth under control, iron deficiency, and severe iron deficiency for 72 h ( Figure 5 ). TLC analysis of these lipid extracts confirmed increased TAG accumulation during iron starvation ( Figure 5A ). The TAG content (semi quanitiave) of control C. reinhardtii cells contained 0.036 mg/g dw of lipid. In comparison, cells cultured under the 1st stage of iron deficiency showed up to lipid content (threefold) (fold increased compare to control), and 2nd stage iron deficiency had lipid contents of fourfold. These results confirmed that two-stage iron stress is a promising method for inducing lipids accumulation, specifically TAGs, in microalgae ( Figure 5B ). It will be necessary to determine whether this strategy can be employed in high throughput photobioreactors to generate TAGs on an industrial scale. FIGURE 5 Thin-layer chromatography (TLC) analysis of total triacylglycerol (TAG) content and semi-quantitative measurement of fatty acids in Chlamydomonas reinhardtii cells cultured under control and iron-deficient conditions. (A) Lipids were extracted from equal dry cell weights (dws), and similar concentrations were loaded in each lane. (B) Semi-quantitative measurements of TAGs bands from the cells under control and iron-deficient conditions. Data are the mean values from three replicates ( n = 3) expressed as a percentage of dw. The data values were done with Student’s t -test. Error bars represent the means ± SD ( n = 3). *** P < 0.001. Characterization of Lipids by LC-MS Analysis Our recent report showed that C. reinhardtii cells subjected to 1st generation iron deficiency contain increased saturated FA levels ( Devadasu et al., 2019 ). To determine the nature of lipids accumulation under iron deficiency conditions, we used LC-MS to characterize the FA composition. Extracted total FAs were analyzed by derivatization (transesterification), and the resultant fatty acid methyl esters (FAMEs) were analyzed by LC-MS. C. reinhardtii cultured under the two iron deficiency stages contained increased amounts of monounsaturated and saturated FAs compared with cells grown in iron-replete conditions ( Supplementary Table 1 ). Levels of both FA types were higher in cells experiencing the most severe iron starvation. TAGs are rich in saturated FAs (16:0, 18:0), so their accumulation might be due to an increase in this class of FA as stated earlier ( James et al., 2011 ; Urzica et al., 2013 ). A previous study demonstrated that saturated FAs form up to 20% of dw in C. reinhardtii cells grown under nitrogen-deficient conditions ( James et al., 2011 ). Thus, the content of saturated FAs, which are useful for biofuels’ production, appears to be increased more by iron deficiency ( Supplementary Table 1 ). Our analysis demonstrated that the levels of all primary FAs found in C. reinhardtii cells (C16:0, C16:3, C18:0, C18:3, and C18:4) are significantly increased by iron starvation. Studies on iron starvation in C. reinhardtii leads to the most dramatic impact on MGDG levels that decreased very rapidly due to activation of MGDG-specific lipase, which predominantly acts on newly synthesized MGDG ( Li et al., 2012 ), which suggests that it might be involved in re-shuffling of saturated FA from MGDG to TAG accumulation. Other reports also stated that the FA desaturation mechanism is played by diiron enzymes known as FA desaturases. These FA desaturases require electrons from NADPH and reduced ferredoxin ( Wada et al., 1993 ; Urzica et al., 2013 ). Thus, due to lack of iron, FA desaturases are the most sensitive targets and inhibit FA’s desaturation in C. reinhardtii ; hence more saturated FA content was observed ( Supplementary Table 1 ). Moreover, other essential FAs, such as omega-3, α-linolenic acid (18:3), linoleic acid (18:2), and the omega-6 FA, oleic acid (18:1), were detected in cells subjected to iron deficiency. Alterations in FA composition have been reported previously in response to environmental changes, including temperature, pH and nitrogen deficiency in Chlamydomonas sp. ( Poerschmann et al., 2004 ; James et al., 2011 ; Çakmak et al., 2014 ). The abundance of TAGs in severe iron deficiency could be an excellent strategy to boost biofuel production from green algae. Iron Starvation Induces DGAT2A Expression Triacylglycerols can also serve as storage lipids in both plants and algae. The final step in TAG biosynthesis, the incorporation of hydroxylated FAs, is catalyzed by diacylglycerol acyltransferase (DGAT) ( Zhang et al., 2009 ). Six genes encoding this enzyme are present in Chlamydomonas, and increased expression of DGAT2A has been reported in nitrogen-starved cells ( Wase et al., 2015 ). Previously, we observed increased DGAT2A protein level expression in C. reinhardtii cells during the 1st stage of iron deficiency ( Devadasu et al., 2019 ). We also observed increased expression of DGAT2A protein level during severe iron starvation (0–72 h) ( Figure 6 ). It is known that DGAT2A catalyzes the incorporation of diacylglycerol with FA into TAG ( Zhang et al., 2009 ). FIGURE 6 Expression analysis of diacylglycerol acyltransferase (DGAT2A) levels in Chlamydomonas reinhardtii cells grown in iron-deprived as well as severe iron-deficient cultures grown from 0, 24, 48, and 72 h by immunoblotting. Total proteins were extracted, and 5 μg of protein were loaded per lane after quantification. Immunostaining of Histone (H) 3 was used as a loading control (C). DGAT2A bands were quantified with Image J software. Experiments were done with three biological replicates ( n = 3)."
} | 7,221 |
35211146 | PMC8860905 | pmc | 7,221 | {
"abstract": "Soft rot Pectobacteriaceae (SRP), typical of Pectobacterium and Dickeya , are a class of Gram-negative bacterial pathogens that cause devastating diseases on a wide range of crops and ornamental plants worldwide. Quorum sensing (QS) is a cell-cell communication mechanism regulating the expression of specific genes by releasing QS signal molecules associated with cell density, in most cases, involving in the vital process of virulence and infection. In recent years, several types of QS systems have been uncovered in Dickeya pathogens to control diverse biological behaviors, especially bacterial pathogenicity and transkingdom interactions. This review depicts an integral QS regulation network of Dickeya , elaborates in detail the regulation of specific QS system on different biological functions of the pathogens and hosts, aiming at providing a systematic overview of Dickeya pathogenicity and interactions with hosts, and, finally, expects the future prospective of effectively controlling the bacterial soft rot disease caused by Dickeya by quenching the key QS signal.",
"conclusion": "Conclusion Quorum sensing is a “language” for microbial communication to regulate the group behavior of microbes. The vital role of QS in bacterial virulence has attracted considerable interest from researchers, making it a promising novel target for the prevention and control of QS-mediated bacterial infections ( Joshi et al., 2016 ). Such novel disease control strategy, called quorum quenching (QQ), is distinguished from other disease biocontrol measures in that QQ disrupts signal-mediated QS by inactivating QS signal or interfering with signal production or perception, not acting on the main growth factors of the pathogens; thus, it would not cause selective pressure on the survival of pathogens. In this study, QS systems that regulate the pathogenesis of Dickeya are systematically revealed which not only modulate the virulence of Dickeya but also affect the drug resistance and adaptability to the environment. They are differential in chemical structures, biosynthesis pathways, signal transduction pathways, and regulation mechanisms. For the widely conserved classical AHL-QS system, it mainly affects bacterial motility and biofilm formation and regulates the adaptability of Dickeya spp. to the surrounding environment. Except in D. solani strains that cause blackleg of potato disease in Western Europe, the AHL-QS system regulates the virulence of the pathogens ( Potrykus et al., 2018 ). Given that VFM and putrescine are ubiquitous in Dickeya spp. and they function as major regulatory systems modulating virulence of Dickeya spp., we suggest focusing on quenching these two systems for prevention and control of bacterial soft rot on crops caused by Dickeya. From current research results, no obvious evidence reveals that there is a relationship between AHL and VFM systems. However, since the VFM signal actions at a relatively low cell density (OD600 below 1.2), while the AHL signal functions when the cell density is high, we think that some gene(s) may be responsible for the switching between these two QS systems. For the putrescine signal, some of the regulon overlays with those regulated by the AHL signal, but no current evidence indicates any interplay between them. The interactions between different QS systems in Dickeya need more in-depth investigations to draw a more accurate and clear conclusion.",
"introduction": "Introduction Soft rot Pectobacteriaceae (SRP) belonging to the genera Pectobacterium and Dickeya ( Charkowski et al., 2012 ) are emerging parasitic pathogens listed in the top ten important bacterial phytopathogens in the world ( Mansfield et al., 2012 ). In addition to causing bacterial soft rot, these pathogens also cause blackleg of potato, stalk rot of maize, and foot rot of rice, resulting in considerable economic damage to vegetable and ornamental plant production worldwide. Previously, Dickeya was grouped into the genus Erwinia containing all plant-pathogenic Enterobacteriaceae, but in 2005, it was reclassified as a new genus Dickeya ( Samson et al., 2005 ). Currently, twelve species are included in the Dickeya genus, including Dickeya dianthicola , Dickeya dadantii , Dickeya zeae , Dickeya chrysanthemi , Dickeya paradisiaca , Dickeya solani , Dickeya aquatica , Dickeya fangzhongdai , Dickeya poaceaephila , Dickeya lacustris , Dickeya undicola , and Dickeya oryzae ( Samson et al., 2005 ; van der Merwe et al., 2010 ; Brady et al., 2012 ; Parkinson et al., 2014 ; van der Wolf et al., 2014 ). Diverse phenotypic differentiation and complicated pathogenic mechanisms have been revealed in different D. zeae and D. solani strains. Recent studies compared the characteristics of D. solani strains isolated from countries with different climate conditions and found higher activities of cell wall degrading enzymes (CWDEs) and virulence in Polish strains than in Finland and Israel strains ( Golanowska et al., 2017 ). D. oryzae and D. zeae strains isolated from rice, banana, and ornamental clivia in China showed different types of phytotoxins produced, including the zeamine I and zeamine II specific in some of the D. oryzae strains, as well as in D. solani strains ( Zhou et al., 2011 , 2015 ; Cheng et al., 2013 ; Hellberg et al., 2015 ), and another toxin produced by D. zeae banana strain MS2 but not the MS3 strain ( Hu et al., 2018 ). Some D. zeae strains from different sources infect plant hosts in different ranges ( Hu et al., 2018 , 2021 ). Moreover, different structures of virulence factor modulating (VFM)-quorum sensing (QS) signals were implicated between D. dadantii 3937 and D. oryzae EC1 and D. zeae banana strains ( Lv et al., 2019 ). Also, the functions of acyl-homoserine lactone (AHL) signals on the virulence of hosts are different in D. oryzae EC1 and D. zeae MS2 ( Feng et al., 2019 ). The diversity of strains and the complexity of pathogenic mechanisms increase the difficulty of disease prevention and control in fields. The achievement of the successful infection of Dickeya on plants depends on a complex range of virulence factors, including plant cell wall degrading enzymes (PCWDEs) ( Hugouvieux-Cotte-Pattat et al., 2014 ), lipopolysaccharides (LPS), extracellular polysaccharide (EPS) ( Condemine et al., 1992 ), iron carriers ( Expert, 1999 ), pigment indigoidine, type III secretion system (T3SS) ( Yang et al., 2002 , 2004 ; Yap et al., 2005 ), and cell motility and adhesion associated with plants ( Hussain et al., 2008 ; Chen et al., 2016 , 2020 ). Motility, which is regulated by the AHL-QS signal, is a secondary virulence determinant of Dickeya , whereas pectinases in PCWDEs, which are regulated by the VFM-QS signal, are the primary virulence determinants of Dickeya , participating and dominating the macerating soft rot process of the pathogen in plant tissues. The capacity of synthesizing and secreting pectinases is modulated by complex and interconnected circuits involving multiple regulatory pathways. A recent study in our laboratory indicated that putrescine is a transkingdom communication signal modulating cell motility, biofilm formation, and virulence of D. oryzae EC1 ( Shi et al., 2019 ). At the moment of current research, QS signals including the AHL signal, the VFM signal, and the crucial signal putrescine have been shown to participate in the intraspecific and transkingdom cell-cell communication, regulating the infection and the colonization of Dickeya toward host plants."
} | 1,895 |
36075917 | PMC9458730 | pmc | 7,222 | {
"abstract": "Operation of temperature sensors over extended temperature ranges, and particularly in extreme conditions, poses challenges with both the mechanical integrity of the sensing material and the operational range of the sensor. With an emissive bendable organic crystalline material, here we propose that organic crystals can be used as mechanically robust and compliant fluorescence-based thermal sensors with wide range of temperature coverage and complete retention of mechanical elasticity. The exemplary material described remains elastically bendable and shows highly linear correlation with the emission wavelength and intensity between 77 K to 277 K, while it also transduces its own fluorescence in active waveguiding mode. This universal new approach expands the materials available for optical thermal sensing to a vast number of organic crystals as a new class of engineering materials and opens opportunities for the design of lightweight, organic fluorescence-based thermal sensors that can operate under extreme temperature conditions such as are the ones that will be encountered in future space exploration missions.",
"introduction": "Introduction Exploration of extreme environments, such as for example the possibility to sustain life at very low temperatures on Earth or in outer space requires robust cryogenic engineering materials that will be able to respond to a number of requirements 1 – 3 . Cooling has an inevitable, and sometimes drastic effect on the properties of both biogenic and artificial materials, as is well known with liquefaction or solidification of natural gas 4 , deceleration or cessation of cell metabolism 5 , 6 , and low-temperature superconductivity 7 , 8 . The oil and gas, machining, chemical industry, food, biomedicine, and aerospace sectors are in increasing demand for cryogenically stable and robust temperature sensing materials 9 – 11 . Low-temperature environments also imply the need for accurate measurement of low temperatures, which currently relies on either thermally controlled resistance 12 or integrated circuit (IC) temperature sensing 13 . However, devices based on resistive measurement may fail due to the opening of the electrical circuit by mechanical separation of the components or as a result of materials’ aging. While the IC sensors are comparatively more robust, they typically cover only a narrow temperature range for measurement. While being effective for specific applications, a common drawback of both of these sensing technologies is that their physical components are made of stiff and brittle materials, and are therefore prone to mechanical damage in devices that are exposed to vibrations or shock over prolonged periods of time. A long sought-after alternative to these materials are fundamentally new soft, light, mechanically compliant, and robust temperature-sensing materials that can be used for accurate and reliable thermal measurement. Organic fluorescent crystals based on π-conjugated small organic molecules with photoluminescent properties have been reported 14 . In a crystalline condensed state, both the intermolecular interactions and the molecular arrangement are known to affect the luminescence 15 – 17 . Within that context, molecular organic crystals from π-conjugated molecules have been studied extensively as promising emissive materials due to the effects of molecular aggregation on the emission 18 . However, organic crystals have not been traditionally considered the materials of choice for temperature measurement. The recent realization of their chemical versatility, anisotropy in structure and properties, and long-range structural order has led to increased recognition of these compounds as a new materials class for organic optoelectronic components, such as resonators 19 , 20 , circuits 21 – 24 , and lasers 25 , 26 . A particularly important aspect of their property profile that has been highlighted only recently is their pronounced elasticity 27 , 28 . This newly realized facet is central to their mechanical compliance and has been recently explored for electronics applications that require lightweight materials, such as optical waveguides for passive transduction of information in both the visible and near-infrared spectral regions 29 – 33 . Opportunities for application of the fluorescence of some flexible organic crystals have also been implicated 34 , 35 for active signal transmission 36 – 39 . In line with the increasing demand for resilient cryogenic materials, we sought to develop lightweight, flexible materials that can be used for sensing or optical information transduction at low temperatures. The material that we report here combines all of these assets. Specifically, crystals of the organic crystalline material that we describe display reversible, reproducible, and strong temperature-induced shifts of their fluorescence, which translates into an opportunity for reproducible optical temperature measurement based on fluorescence. Within the broader set of available luminescence thermometry techniques 40 , this work brings a new class of materials that could be considered as an alternative to the currently used fluorescent thermometers.",
"discussion": "Results and discussion Preparation and characterization The material described here, ( E )−1-(3,5-bis(trifluoromethyl)phenyl)−3-(1-ethyl-1 H -indol-3-yl)prop-2-en-1-one (compound 1; Fig. 1a ), was discovered during our broader screening of materials in an effort to identify organic crystals that are both elastic and emissive. It was synthesized in two steps with a 72% yield (Supplementary Figs. 1 and 2 ). Long, yellow needle-like crystals of about 1.5 cm in length were readily prepared by liquid-phase diffusion (Fig. 1b ). The material has a good thermal stability and does not undergo a phase transition up to its melting point at 467 K (Supplementary Fig. 3 ). The crystals of compound 1 are reversibly elastic; they can be bent without any visible damage (Supplementary Movie 1 ), as it was confirmed by the scanning electron microscopic images of bent specimens that were fixed in their bent state for inspection (Fig. 1c ). No obvious defects could be observed on the crystal surface even after bending the crystals up to 5000 times (Supplementary Fig. 4 ). The crystals absorb light with a maximum at 415 nm, and emit bright green fluorescence with a maximum emission at about 540 nm and a quantum yield of 0.16 at 298 K (Fig. 1d ). The fluorescence quantum yield is 0.30 at 77 K, which is higher compared to room temperature due to the decreased non-radiative rate at low temperature (Supplementary Table 1 ). Their bulk elastic modulus at 298 K was found to be 1.6 GPa by tensile measurements (Fig. 1e ) 41 . The elasticity of the crystals is preserved even at low temperatures, and they can be bent repeatedly even when they are immersed in liquid nitrogen (Fig. 1f–i ). Fig. 1 Structure, elasticity, and fluorescence of compound 1. a Chemical structure of compound 1. b Photograph of crystals of compound 1 under daylight. c SEM images of a bent crystal. The region in the boxed area is shown as a zoomed-in image in the inset. d Absorption (broken line) and emission (solid line) spectra of crystals of compound 1. e Stress-strain curve of the crystal. f , g Photographs showing crystal bending in air. h , i Photographs showing crystal bending in liquid nitrogen. j – m Change in emission color of the crystals from 298 K to 77 K (the images shown in panels f ‒ m were recorded under UV light for better contrast). Temperature-induced change of fluorescence During the mechanical characterization at low temperature, we noticed that the color of emission of crystals of compound 1 changes visibly from green to orange when they are transferred from room temperature to liquid nitrogen (Fig. 1j–o ; Supplementary Movie 2 ). The temperature-dependent fluorescence spectra show that the color change is due to a red-shift of the emission maximum upon cooling, whereupon the emission intensity gradually increases (Fig. 2a ; Supplementary Fig. 5 ). The maximum emission wavelength changes from 540 to 580 nm upon cooling from 277 to 77 K, and the dependence is linear in the temperature range from 77 to 277 K ( λ max /nm = –0.21 T /K + 596.91, R 2 = 0.9948; Fig. 2b ). More quantitatively, when plotted on a CIE 1931 color space diagram 42 this change in color of the emitted light is reflected in the change of the CIE coordinates from (0.38, 0.59) to (0.51, 0.48) (Fig. 2c ). At 77 K, the emission intensity increases to 1.79 times compared to that at 277 K. The change of emission intensity with temperature is also linear within a certain temperature range ( I / I max = –0.0022 T /K + 1.14, R 2 = 0.9711; Fig. 2d ). This linear and strong response of the fluorescence of compound 1 with temperature favors this material as a temperature-sensing medium. The lowest detectable temperature of compound 1 is lower than that of some metal sensors 43 , 44 , however, the sensitivity (maximum emission wavelength vs. temperature) is higher than other organic materials with temperature-dependent fluorescence 45 , 46 . In order to verify whether the red-shift of compound 1 is caused by triplet luminescence, the lifetimes at 298 and 77 K were determined (Supplementary Fig. 6 ). The lifetime of the emission of compound 1 at 77 K is 17.99 ns, which is longer than that at 298 K (6.13 ns), confirming that the emission does not originate from a triplet state. The lifetime and quantum yield of bent crystals were also recorded and indicate that the bending does not have any significant effect on non-radiative pathways (Supplementary Fig. 6 , Supplementary Table 2 ). Fig. 2 Temperature dependence of the emission of compound 1. a Variable-temperature emission spectra of a crystal of compound 1. b Correlation between crystal emission wavelength and temperature in the temperature range 77–277 K. The straight broken line shows a linear fit to the data. c The CIE coordinates of crystal emission bands at different temperatures. The region in the boxed area is shown as a zoomed-in image in the inset. d Correlation between the crystal fluorescence intensity ratio and temperature in the temperature range 77–277 K. The straight broken line shows a linear fit to the data. e , f Emission spectra were measured at one end of the crystal at 298 K ( e ) and 77 K ( f ). The values 0‒5 mm, corresponding to curves with different symbols, represent the distance from the excitation site to the point of measurement. The excitation was performed at 355 nm. Optical-waveguiding capability The remarkable flexibility of the compound 1 crystals at both ambient and low temperatures appears as a prospective platform for their application in thermally resistant (opto)electronics. Similar to other flexible crystals 47 , 48 , single crystals were tested as active optical waveguides (Fig. 2e ). The optical loss coefficient of a straight crystal at room temperature was found to be 0.16 dB mm ‒1 (Supplementary Fig. 7 ). A bent crystal had a nearly identical optical loss, 0.17 dB mm ‒1 (Supplementary Fig. 8 ). This result indicates that the crystal bending does not have a significant effect on the optical waveguiding performance, as is expected from the retention of the crystal integrity upon bending, described above. The cooling did not affect the waveguiding ability of the crystal (Fig. 2f ); at 77 K, the optical loss factors were found to be 0.17 dB mm ‒1 for a straight crystal and 0.20 dB mm ‒1 for a bent crystal (Supplementary Figs. 9 and 10 , Supplementary Movie 3 ). Temperature sensing using fluorescence We hypothesized that the combination of linear dependence of the emission wavelength on the temperature and the favorable mechanical flexibility of compound 1 at low temperature could carry some potential for the development of flexible temperature sensors. Along this line of thought, a crystal of about 1.0 cm in length was selected, and liquid nitrogen was dropped continuously on a small area at one end (Fig. 3a ). The cold spot was excited with a laser, and the optical output was collected and analyzed at both ends of the crystal (Supplementary Fig. 11 ). Consistent with the temperature difference, the cold end of the crystal emitted orange light, while the opposite end that was at higher temperature emitted green light. The temperature of the warm end was determined to be about 230 K (Supplementary Figs. 12 and 13 ). However, the emission wavelength maximum of the spectra recorded at both ends was 580 nm, showing that only the orange light was transduced (Fig. 3b ). We note that the optical waveguide is in active mode here. The photoluminescence (PL) spectra of crystals of millimeter size at 77 and 298 K were also recorded. Comparing the PL spectrum at 77 K with the optical waveguide signal at 77 K, we conclude that the peak shapes are almost identical (Supplementary Fig. 14d ). This result indicates that the optical signal propagating in the waveguide is not affected during its propagation. Other two crystals that have been reported earlier (compounds 2 and 3) were also investigated to confirm this property of organic crystals (Supplementary Fig. 15 ). The position of laser excitation was then changed, and the crystal was excited at the higher temperature end, while it was being cooled at the opposite end (Fig. 3c ). Contrary to the previous case, the warm end emitted green light and the cold end emitted orange light. The spectra of the fluorescence remained identical at both ends, with a maximum emission at 550 nm (Fig. 3d ). These results demonstrate that the crystal can be indeed cooled locally, and it does not equilibrate thermally during the experiment. More importantly, when the crystal is used as a medium for optical transduction, the output signal depends only on the temperature at the point of excitation; the output is not affected even when the intermediate section of the crystal is at lower temperature. Fig. 3 Temperature-dependent optical waveguiding. a Schematic diagram and real picture of a crystal waveguide at low temperature. The gray dashed box denotes the area where the liquid nitrogen was dropped (77 K). b Emission spectra recorded at the excitation end (77 K) and at the output end of the crystal (230 K). c Schematic diagram and real picture of a crystal waveguide at room temperature. d Emission spectra recorded at the excitation end (230 K) and at the output end of the crystal (77 K). e Emission spectra of the crystal optical waveguide at different temperatures. f CIE 1931 color space coordinates of crystal optical waveguide signals at different temperatures. The region in the boxed area is shown as zoomed-in image in the inset. g Correlation between the emission intensity ratio and temperature in the temperature range 77–277 K. The straight broken line shows a linear fit to the data. h Correlation between the crystal waveguide emission wavelength and temperature in the temperature range 77–277 K. The straight broken line shows a linear fit to the data. In another experiment, a 1.5 cm-long crystal was selected, and one of its ends was brought into contact with evaporating liquid nitrogen. This cool end was excited with a laser at 355 nm, and the spectrum of the output light was recorded at the other end (Supplementary Fig. 16 ). As the liquid nitrogen evaporated, the excited end of the crystal gradually warmed up, and the output spectrum changed (Fig. 3e ). As expected from the other experiments above, upon warming the position of the maximum emission was blue-shifted while its intensity decreased, as it is illustrated by changes in the respective CIE coordinates (Fig. 3f ). This is in accordance with the results obtained by cooling, as shown in Fig. 3 g and h . The changes in maximum emission wavelength and intensity with temperature on warming are linear ( λ max /nm = –0.19 T /K + 595.00, R 2 = 0.9921; I / I max = –0.0015 T /K + 1.12, R 2 = 0.9964). These experiments demonstrate that the crystal does not only respond to temperature changes by emitting light, but it also transduces and outputs the temperature information by acting as an optical waveguide. The measurement was also performed after the crystal was repeatedly bent, and the results show that the sensitivity was practically unaffected even after the crystal was bent 1000 times (Supplementary Fig. 17 , Supplementary Table 3 ). The signal from the crystal is not affected by the temperature of the section through which the optical signal is transmitted. The crystal’s output reflects only the temperature of the point of excitation, and this property favors this material, and in principle also other emissive elastic organic crystals, as an active sensing medium for luminescence thermometry. Computational modeling of the fluorescence To investigate the root cause of the temperature dependence of fluorescence, density functional theory (DFT) calculations were performed on a molecule of compound 1. Supplementary Fig. 18 shows the frontier molecular orbitals based on the ground-state (S 0 ) geometry of compound 1. Obviously, the LUMO, HOMO, and HOMO-1 are mostly π-orbitals, while the HOMO-2 is an n-orbital and distributed over the carbonyl group and its surroundings. The HOMO-1 is localized on the indole ring and the HOMO spreads over the indole ring and the adjacent vinyl group. In contrast, the LUMO is delocalized over the entire molecular backbone. Vertical excitations from S 0 to S 1 , S 2 , and S 3 are dominated by the transitions from the HOMO, HOMO-2, and HOMO-1 to LUMO, respectively. The S 2 and S 3 excitation energies are close to the S 1 excitation energy (Supplementary Table 4 ). Deeper insight into the fluorescence properties necessitated closer inspection of the electronic structures in the excited-state optimized geometries. The nature of the frontier orbitals in the excited state is hardly altered; moreover, the S 1 and S 2 excitations still correspond to the HOMO and HOMO-2 to LUMO transitions, respectively. Interestingly, in the S 3 geometry, besides the HOMO-1 → LUMO transition, the HOMO → LUMO transition has a significant contribution (14.8%) to the S 3 excitation, with significant oscillator strength ( f ≈ 0.30) (Fig. 4a ). Because of the relatively close energies between the S 1 and S 3 states (Supplementary Tables 4 and 5 ), the S 3 population will be increased when the temperature is elevated, which is beneficial to enhance the luminescence from S 3 and results in blue-shift in the emission (Fig. 4b ). This result supports the experimental observations. Fig. 4 Molecular orbitals related to the temperature-dependent emission. a Frontier orbitals and transition contributions for S 1 , S 2 and S 3 states in the respective optimized geometries (energy unit: eV). b Energy level diagram showing the changes in the emission energy and the accompanying blueshift of the emission with increasing temperature caused by the thermal population of the higher-energy S 3 state. Relation to the crystal structure The change in emissive properties of any crystal is also inevitably related to its crystal structure, and therefore the structure was determined at 298 and 100 K (Supplementary Fig. 19 , Supplementary Table 6 , Supplementary Data 1 and 2 ). At 298 K, the crystal is in the monoclinic system (space group P 2 1 / n ) with Z = 4 and cell parameters a = 12.1007(9) Å, b = 4.8876(4) Å, c = 31.714(2) Å, and β = 91.548(3)°, and unit cell volume V = 1875.0(2) Å 3 . The torsion angles around the two single chemical bonds between the benzene and the indole rings are 12.2° and 1.3° (Supplementary Fig. 19 ). The molecules are stacked in a parallel fashion and form columnar structures by π···π interactions with an intermolecular distance of 3.758 Å along the crystallographic [010] direction (Fig. 5a ). Upon cooling to 100 K, the crystal symmetry is retained, but the cell parameters change to a = 12.0224(7) Å, b = 4.8149(3) Å, c = 31.4650(18) Å and β = 92.271(2)°, and the unit volume is V = 1819.97(19) Å 3 . The unit cell shrinks by 0.7% along the a axis, 1.5% along the b axis, and 0.8% along the c axis. The torsion angles between the benzene ring and the indole ring remain very close, 11.0° and 1.4° respectively, and thus the high degree of planarity of the molecule is retained at low temperature (Supplementary Fig. 19 ). The stacking distance of the molecules is slightly reduced, to 3.579 Å (Fig. 5b ). Since the molecular geometry is practically conserved, similar to other cases 49 , 50 , the change in emission can be attributed to the decreased π···π distance, which is related to the significant shrinking of the lattice along the b axis. Furthermore, the enhanced intensity of emission at low temperatures is related to the restriction of non-radiative transition pathways and enhancement of the competing radiative processes 51 – 53 . In addition, the interaction energy among the molecules at 298 and 100 K were both calculated. The energy frameworks were constructed by using Crystal Explorer 54 and the B3LYP hybrid functional with the 6–31G(d,p) basis set, where semiempirical dispersion was included by using D2 version of Grimme’s dispersion, and they are visualized in Fig. 5c . Based on the preceding discussion, the structural changes observed upon cooling indicate that the unit cell contraction along the b axis is the most likely contributor to the appealing change in emission. The calculated interaction energy between molecule 1 and molecule 2 of –66.4 kJ mol ‒1 at room temperature was found to decrease to –67.5 kJ mol ‒1 at 100 K (Table 1 ). This decrease in energy is in accord with the strengthening of the π···π interaction and is also consistent with the observations from the crystal structures. Fig. 5 Effect of temperature on the structure and energy framework analysis of compound 1. a , b Molecular geometry and packing at 298 K ( a ) and 100 K ( b ). The arrows indicate the torsional angles discussed in the text. c Energy bars showing the interaction energies. The blue bars represent the interactions between molecules, and their thickness indicates the value of interaction energy. d Energy frameworks viewed along the b axis are consistent with the qualitative requirements for bending. e Parallel packing structure, and the purported expansion and contraction of the outer and inner arcs of the crystal that are suggested to occur during the bending of the crystal. Table 1 Intermolecular interaction energy of compound 1 Temperature (K) Method E total (1‒2) (kJ mol ‒1 ) E total (1‒3) (kJ mol ‒1 ) 100 B3LYP/ 6-31 G(d,p) ‒67.5 ‒52.4 298 B3LYP/ 6-31 G(d,p) ‒66.4 ‒47.4 The intermolecular interaction energy between different molecules (molecules 1–3, as shown in Fig. 5c ). The crystal structure is also directly related to the elastic properties of compound 1. The thicker blue bars in Fig. 5c correspond to the π···π interactions along the [010] direction. The C–H···O interactions ( d = 2.600 Å, θ = 148.8° and d = 2.418 Å, θ = 135.6°) and F···F interactions ( d = 2.870 Å) among the molecular layers in the c direction can also be identified in the energy frameworks, while the thinner blue bars among the layers correspond to the F···F interactions (Fig. 5d ). As explained above, the interactions along the c axis are likely responsible for the extraordinary flexibility of the crystal. The intermolecular interactions along the a axis participate in the ‘bendable’ (100) plane (Supplementary Fig. 20 ). These supramolecular interactions could effectively absorb and dissipate the applied stresses, thereby rendering the crystal more susceptible to stress. In elastically bendable crystals, the stretching of the outer arc of the crystal and separation of the respective molecules, and the concomitant contraction of the inner layer and decrease of the distance between the molecules are thought to ensure reversibility of structural changes upon bending, that is, the elasticity of the crystal (Fig. 5e ) 55 – 66 . Although the thermal changes in the structure cannot be taken as an indication of the structural changes that actually occur during bending 67 , upon cooling these interactions in compound 1 are maintained with only a slight decrease in the respective distances (F···F, d = 2.861 Å; C–H···O, d = 2.532 Å, θ = 131.8° and d = 2.335 Å, θ = 140.0°) due to the nearly isotropic contraction of the crystal. The result is also consistent with the slight decrease of intermolecular interaction energy at 100 K compared with 298 K (Table 1 ). This retention of overall interactions might be the reason for the retention of the compliant mechanical properties at low temperature. In summary, a flexible organic crystalline material is reported here that was found to exhibit temperature-dependent spectral changes which could be used for measurement of temperature by optical means in a wide temperature range, particularly at low temperatures. The fluorescence of the material displays linearity in response with respect to both the maximum and intensity of the emission. Variable-temperature experiments on the optical waveguides properties further demonstrated that these organic crystals can convert the temperature at the excited position into a more stable optical signal output. The extraordinary elasticity of the crystal is a desirable property, as it is expected to translate into better durability and resistance to mechanical damage when it is being used as an optical waveguide. An attempt was made to rationalize the temperature dependence of the emission and the elastic properties of the crystal by theoretical calculations and analysis of its crystal structure. The thermal population of a higher-lying S 3 state was identified as the most likely factor behind the change in emission energy. This work expands the scope of application of elastic organic crystals as optical waveguides at low temperatures and provides an alternative route to the development of fluorescence thermometric devices by using soft and light organic crystalline materials as active sensing medium."
} | 6,581 |
38161372 | PMC10755848 | pmc | 7,223 | {
"abstract": "Selective lignin\ndepolymerization is a key step in lignin valorization\nto value-added products, and there are multiple catalytic methods\nto cleave labile aryl–ether bonds in lignin. However, the overall\naromatic monomer yield is inherently limited by refractory carbon–carbon\nlinkages, which are abundant in lignin and remain intact during most\nselective lignin deconstruction processes. In this work, we demonstrate\nthat a Co/Mn/Br-based catalytic autoxidation method promotes carbon–carbon\nbond cleavage in acetylated lignin oligomers produced from reductive\ncatalytic fractionation. The oxidation products include acetyl vanillic\nacid and acetyl vanillin, which are ideal substrates for bioconversion.\nUsing an engineered strain of Pseudomonas putida ,\nwe demonstrate the conversion of these aromatic monomers to cis , cis -muconic acid. Overall, this study\ndemonstrates that autoxidation enables higher yields of bioavailable\naromatic monomers, exceeding the limits set by ether-bond cleavage\nalone.",
"conclusion": "Conclusions This work demonstrates that catalytic autoxidation\nmay be used\nto generate aromatic monomers from C–C linked dimers and oligomers\nderived from lignin. This concept is demonstrated here by acetylating\nphenol-rich RCF oil and then conducting aerobic oxidation of the oligomeric\nfraction of poplar RCF oil with a Co/Mn/Br catalyst mixture in acetic\nacid. This reaction yields a collection of oxygenated aromatic monomers\nthat represent a 17% increase in monomer yield compared to the RCF\nprocess alone. Additionally, P. putida strains can\nutilize these oxidation mixtures to generate muconic acid in quantitative\nyield, which is a bioprivileged molecule that can be converted into\nperformance-advantaged biopolymers and direct replacement biobased\nchemicals, such as adipic acid and terephthalic acid. 32 , 35 While these data demonstrate the viability of catalytic autoxidation\nto increase monomer yields from lignin, a limitation in our current\nexperimental setup is manifested in the product degradation chemistry\nthat is likely occurring in parallel to the productive C–C\nbond cleavage. One possible pathway for arene degradation may proceed\nvia a phenolic intermediate, as related ring-opened products have\nbeen reported in various oxidative degradation reactions of phenols, 51 − 55 and the detected dimethoxybenzoquinone and methoxymaleic acid products\nmay result from such a reaction. To overcome this limitation, flow\nchemistry could allow for improved control over the reaction conditions\nand residence time, thereby enabling improved selectivities and yields\nto aromatic products.",
"introduction": "Introduction Lignin is one of Earth’s most abundant\nnatural polymers,\nand it is synthesized via oxidative radical coupling reactions of\nmonolignols that give rise to a polymer with aryl–ether bonds\nand several types of carbon–carbon (C–C) linkages. 1 Lignin depolymerization to aromatic monomers\nis one of the most sought-after contemporary approaches to derive\nvalue from lignin. Many effective methods have been developed to this\nend, 2 − 8 and near-theoretical monomer yields are now accessible, based on\ncleavage of β-O-4 aryl-ether linkages. 9 − 15 The inability of these methods to cleave refractory C–C bonds\nin lignin, such as those present in the 5–5, β–1,\nβ–β, and β–5 linkages, severely limits\nthe aromatic monomer yield accessible from lignin ( Figure 1 ). Hardwood lignin often exhibits\nhigh β-O-4 content and yields of aromatic monomers between 30\nand 40 wt % are often attainable. Considerably lower yields are obtained\nfrom lignins with lower aryl–ether bond content, including\nthose from softwoods, grasses, and extracted lignins from biorefinery\nand pulping processes. 4 , 16 , 17 Figure 1 Overall\nlignin conversion approach presented in this work, featuring\nan oxidative C–C bond cleavage process to generate oxidized\nmonomers that are suitable for biological funneling to a single product. Reductive catalytic fractionation (RCF) of lignin\nis among the\nmost effective methods available for conversion of lignin into aromatic\nmonomers. 18 However, like most lignin depolymerization\nmethods, RCF is largely limited to aryl–ether bond cleavage\nand generates an oligomeric byproduct that is rich in C–C linkages.\nAccess to higher yields of aromatic monomers from lignin necessitates\nmethods for cleavage of the C–C bonds, such as those in β–1,\nβ–5, β–β, and 5–5 linkages. 1 , 2 Recent efforts have explored homogeneous thermal catalysis, 19 photocatalysis, 20 and catalytic cracking. 21 An important\nadvance was reported recently by Samec et al., who demonstrated a\ntandem RCF/oxidation sequence to increase monomer yields. Specifically,\nlignin oligomers obtained from RCF treatment of a birch feedstock\nwere treated with a superstoichiometric oxoammonium reagent. This\nreagent, which operates via a hydride transfer mechanism, led to cleavage\nof linkages containing C–C bonds and selectively generated\n2,6-dimethoxybenzoquinone as a product in 18 wt % yield with respect\nto the oligomeric feedstock. 18 Here,\nwe report a complementary strategy to achieve C–C\ncleavage in lignin that leverages catalytic autoxidation and radical\nreaction pathways. A cobalt/manganese/bromide cocatalyst system provides\nthe basis for the industrial autoxidation of p -xylene\nto terephthalic acid, 22 and analogous conditions\nhave been used to support C–C cleavage in both simple hydrocarbons\nand synthetic plastics. 23 − 25 Key mechanistic steps in these\nreactions include hydrogen atom transfer, radical trapping by O 2 , and β-scission of intermediate alkoxyl radicals. Br\nand oxygen-centered radicals contribute to hydrogen-atom transfer\nfrom the organic substrate in the catalytic oxidation process. The\nCo and Mn catalysts support the autoxidation through activation of\nO 2 , oxidation of HBr, and selective conversion of reaction\nintermediates, such as organic hydroperoxides, into the oxidized products. 26 , 27 Lignin has been subjected to related conditions, but monomer yields\ndid not exceed the monomer content of the lignin substrate. 28 , 29 We postulated, however, that lignin oligomers derived from RCF treatment\nof biomass would be more amenable to such treatment and undergo successful\nconversion into aromatic monomers. The results outlined herein validate\nthis hypothesis, showing that a Co/Mn/Br-based catalyst system converts\nRCF-derived oligomers from poplar into aromatic monomers, which are\nthen used as substrates for bioconversion to cis , cis -muconate, a precursor to bioderived polymers ( Figure 1 ). 30 − 35 By controlling the catalytic conditions, the yield of aromatic products\ncan be maximized while limiting overoxidation to quinone-based products\nthat are not amenable to biological funneling to cis , cis -muconate. This pairing of catalytic aerobic\noxidation that supports C–C bond cleavage and biological funneling\noffers a strategy to obtain higher yields of single products from\nlignin.",
"discussion": "Results and Discussion Phenol Acetylation Is Key to Enabling C–C\nBond Cleavage\nvia Autoxidation As an initial test of the targeted C–C\nbond cleavage chemistry, we explored the oxidation of model aromatic\nsubstrates ( Scheme 1 , Tables S17 and S18 in the Supporting Information , SI ). Upon subjecting 4-propylguaiacol ( 1 ) to catalytic\nCo/Mn/Br salt mixtures (3 mol % Co(OAc) 2 ·4H 2 O, 3 mol % Mn(OAc) 2 ·4H 2 O, 3 mol % NaBr)\nwith heating at 120 °C for 2 h under 6 bar O 2 , we\nobserved no C–C bond cleavage and instead recovered 1 in 87(12)% yield. This result is consistent with previous work describing\nthe antioxidant properties of phenols 36 and highlights the importance of phenol protection in autoxidation.\nIn contrast, acetyl 4-propylguaiacol ( 2 ) can be converted\nto acetyl vanillic acid and acetyl vanillin in 47(8)% yield under\nthe same oxidation conditions. Similarly, the oxidation of acetyl\npropylsyringol ( 3 ) gives a total yield of acetyl syringic\nacid and acetyl syringaldehyde at 42(12)%. These results on 2 and 3 demonstrate the viability of autoxidation\nconditions for simplified G- and S-type lignin models. Scheme 1 Oxidation\nof Model Substrates: 4-Propylguaiacol ( 1 , top ), Acetyl 4-Propylguaiacol ( 2 , Middle ), and Acetyl 4-Propylsyringol ( 3 , Bottom ) Mol % yields shown\nas value(standard\ndeviation) and determined by LC-MS or UHPLC quantification. Application of Autoxidation on Model Lignin\nDimers We subsequently subjected various representative dimer\nmodels to\nthe same autoxidation conditions and were able to detect monomeric\nproducts in most cases ( Scheme 2 , Tables S19–S21 ). As a\nrepresentative β-1 dimer, diacetyl bivanillyl ( 4 ) was subjected to the same reaction conditions (3 mol % Co(OAc) 2 ·4H 2 O, 3 mol % Mn(OAc) 2 ·4H 2 O, 3 mol % NaBr, 120 °C for 2 h under 6 bar O 2 ), which afforded C–C bond cleavage products in in 64(2)%\nyield. The acetyl-protected β-5 model ( 5 ) was oxidized\nto yield acetyl vanillic acid and the corresponding aldehyde in 8(3)%\nyield. Overall, the oxidations of compounds 4 and 5 demonstrate that C–C bond cleavage can be accessed\non dimer models featuring common C–C linkages found in RCF\noil. Specifically, cleavage of the C benzylic –C bonds\nis observed to form the corresponding benzoic acid monomer. For the\n5–5 dimer model ( 6 ), full consumption of 6 was observed, but only trace products were detected. Scheme 2 Oxidation of Model Dimer Substrates of β–1 ( 4 ), β–5 ( 5 ), and 5–5 ( 6 ) Linkages Mol % yields shown\nas value(standard\ndeviation) and determined by LC-MS or UHPLC quantification. Preparation of a Lignin Dimer and Oligomer-Rich\nStream for Autoxidation On the basis of these promising model\ncompound results, we sought\nto apply this autoxidation approach to a realistic lignin stream.\nRCF of biomass affords an “RCF oil” that contains monomers,\nbut is also rich in dimers and oligomers that exhibit 5–5,\nβ–1, β–5, and β–β linkages. 37 − 39 This RCF oil provides an ideal substrate to investigate the utility\nof Co/Mn/Br-catalytic autoxidation to supplement the yield of aromatic\nmonomers accessible from lignin. RCF oil was prepared by subjecting\nextractives-free poplar biomass to 5 wt % Ru/C under 30 bar H 2 in methanol for 6 h at 225 °C. RCF monomers were quantified\nby GC-FID to determine a total monomer content of 1.8 mol monomer/g\nRCF oil, Tables S1–S3 . The aromatic\nmonomer yield on a total lignin oil basis from this experiment was\n34 wt %, similar to previous work. 40 Using\nthis oil, we functionalized free OH groups as phosphites 41 and conducted quantitative analysis by 31 P NMR spectroscopy, which yielded a phenolic content of 4.2\nmmol/g RCF oil and an aliphatic OH content of 2.4 mmol/g RCF oil, Figure S1 . With the goal of protecting the phenolic\nfunctionalities, the RCF oil was subsequently derivatized via treatment\nwith excess acetic anhydride and pyridine at 40 °C to yield acetyl-protected\nRCF oil. GC-FID quantification of acetyl monomers in the acetyl RCF\noil demonstrate that 78% of RCF monomers are recovered upon acetylation, Tables S4–S5 . GPC traces of the RCF oil\nand acetyl-derivatized material exhibit very similar profiles ( Figures S2–S5 ), suggesting that acetylation\ndoes not alter the distribution of monomer, dimer, and oligomer fractions\nin the oil. Treatment of the acetylated RCF oil under the phosphite\nOH functionalization conditions and 31 P NMR analysis 41 confirmed the absence of phenolic and aliphatic\nOH groups, Figure S1 . Subjecting\nacetyl poplar RCF oil, containing the full distribution\nof monomers, dimers and higher molecular-weight components, to 4 wt\n% Co(OAc) 2 ·4H 2 O, 4 wt % Mn(OAc) 2 ·4H 2 O, and 2 wt % NaBr at 6 bar O 2 in\nacetic acid yielded 21(4) wt% of monomers after heating at 120 °C\nfor 2 h. Acetyl vanillic acid and acetyl syringic acid are the major\nproducts formed, but the quantity of oxidation monomers following\noxidation (0.13 mmol/g) is lower than the initial monomer content\nof the starting material (1.4 mmol/g), which may be attributable to\nproduct degradation during oxidation. To gauge the stability\nof the oxidation products in our autoxidation\nconditions, we subjected acetyl vanillin, acetyl vanillic acid, acetyl\nsyringaldehyde and acetyl syringic acid to similar conditions with\n3 mol % Co(OAc) 2 ·4H 2 O, 3 mol % Mn(OAc) 2 ·4H 2 O, and 3 mol % NaBr. Quantification of\nthe acid and aldehyde products by UHPLC determined 75(2)% and 79(5)%\nof acetyl vanillin and acetyl vanillic acid, respectively, were recovered\nas acetyl vanillic acid. Similarly, acetyl syringaldehyde and acetyl\nsyringic acid were recovered in 63(5)% and 92(6)% as acetyl syringic\nacid ( Scheme 3 ; Table S6A ). These data are consistent with competing\ndegradation of aromatic species during catalytic conditions. In the\noxidation mixtures of each of these four compounds, the corresponding\nphenolic carboxylic acid (i.e., vanillic acid or syringic acid) is\ndetected as a minor product (<∼20%, Table S6B ). Furthermore, trace quantities of methoxymaleic\nacid (<∼2%) are detected in the oxidation of acetyl vanillin,\nacetyl syringaldehyde, and acetyl syringic acid, and 2,6-dimethoxybenzoquinone\nis detected in the oxidation of acetyl syringaldehyde (9%) and acetyl\nsyringic acid (3%). These data are consistent with the degradation\nof the aromatic rings. One possible mechanism for degradation could\ninvolve the initial deprotection of the acetyl-protected substrate\nto generate the phenolic congener, which can be further oxidized to\nform the corresponding benzoquinone derivative. Deprotection of the\nacetyl groups likely occurs through hydrolysis, as water is a byproduct\nunder Mid-Century (MC) oxidation conditions. 22 Benzoquinone species are known to undergo oxidative ring-opening\nto form maleic acid derivatives, 42 which\nmay potentially undergo further oxidation under the studied conditions. Scheme 3 Product Stability Reactions under Autoxidation Conditions with Mol\n% Yields Shown As Value (Standard Deviation) Yield determined\nby LC-MS\nquantification. To circumvent the problem\nof monomer degradation, we separated\nthe acetyl RCF oils monomers by vacuum distillation, as previously\ndone by Samec et al., 18 and subsequently\nexplored the oxidation of the acetyl-protected oligomers. Distillation\nof the acetyl RCF oil (ca. 50 mbar, 250 °C) afforded the acetyl\nmonomer distillate as a pale-yellow oil (56 wt %), and the acetyl\noligomers (43 wt %) remained in the boiling flask as a brown solid.\nGC-FID analysis revealed that 95% of acetyl RCF monomers were recovered\nin the distillate, and only trace amounts of diacetyl 4-propanolguaiacol\nand diacetyl 4-propanolsyringol remained the acetyl oligomer fraction\n( Table S5 and Figure S6 ). GPC analysis\nof the distillate (acetyl monomer fraction) confirmed the presence\nof only lower molecular weight species, while analysis of the acetyl\noligomer fraction contains higher molecular weight compounds ( Figure 2 A). Furthermore,\nthe GPC data of the acetyl monomer fraction exhibits intensity at\na very similar MW range as that of authentic standards of the main\nmonomeric components in acetyl RCF oil, consistent with the distillate\nbeing primarily composed of acetyl RCF monomers ( Figure 2 B). Phosphite functionalization\nand 31 P NMR analysis of the acetyl monomer and acetyl dimer\nfractions revealed the absence of free phenolic or aliphatic OH groups,\nsuggesting that the acetyl groups remained intact in both fractions\nduring distillation, Figure S1 . Previous\nstudies have revealed the identity of dimers and oligomers present\nin the RCF-derived substrate, which mostly comprise compounds that\nexhibit 5–5, β–1, β–β, and\nβ–5 linkages. 38 , 39 Figure 2 (A) GPC traces of acetyl\nRCF oil, acetyl monomer fraction, and\nacetyl oligomer fraction. Wt % values expressed as the weight of acetyl\nmonomer or oligomer fraction/weight of acetyl RCF oil. (B) GPC trace\nof acetyl monomer fraction ( top ) and acetyl RCF monomer\nstandards ( bottom ). Autoxidation of Dimers and Oligomers in RCF Lignin Oil We\nnext sought to study the effects of different reaction parameters\non the oxidation of the acetyl oligomer substrate ( Figure 3 , Tables S7–S10 ). Reactions were conducted by treating the acetyl\noligomer substrate with mixtures of Co(OAc) 2 ·4H 2 O, Mn(OAc) 2 ·4H 2 O, and NaBr, and\nheating the acetic acid solutions under 6 bar O 2 gave mixtures\nof monomers comprising aromatic and non-aromatic structures as shown\nin Figure 3 . The quantification\ndata of the individual molecules are presented in the SI . Figure 3 Reaction optimizations showing total yields for the autoxidation\nof the acetyl oligomer substrate at variable (A) Co and Mn loadings\n([Co]=[Mn]), (B) NaBr loadings, (C) temperatures, and (D) reaction\ntimes. Standard conditions for optimization utilize x = 0.003 mmol, y = 0.003 mmol, z = 2 h, T = 120 °C for 20 mg of acetyl oligomer\nsubstrate. For the standard conditions, oxidation yields are an average\nof four runs. Values from all other conditions are from single runs.\nNumerical data for this figure and individual compound yields for\neach bar are provided in Tables S7–S10 (products quantified by GC-FID and LC-MS). A study of Co and Mn catalyst loadings showed that\nincreasing their\nloadings in a 1:1 molar ratio from 0.001 mmol to 0.003 increases the\ntotal yield of oxidation products from 5 wt % to 13 wt %. Between\n0.003 and 0.009 mmol loadings, the yields range from 12 to 19 wt %\n( Figure 3 A). An increase\nin monomer yield was observed upon increasing NaBr loadings from 0.001\nmmol (9 wt % products) to 0.003 mmol (13 wt % products), above which\nthe monomer yields modestly varied between 13 and 15 wt % ( Figure 3 B). Regarding the\neffect of reaction temperature, a large increase in yield was observed\nwhen the reaction was run at 120 °C (13 wt % products) compared\nto 100 °C (2 wt % products, Figure 3 C). More carboxylic acid products (acetyl\nsyringic acid and acetyl vanillic acid) are formed compared to the\ncorresponding aldehyde products (acetyl syringaldehyde and acetyl\nvanillin) when the temperature is increased from 120 to 140 °C,\nbut the overall yields are comparable (13 wt % products). Further\nincrease in temperature up to 180 °C resulted in comparable overall\nyield of aromatic products but increased nonaromatic products, such\nas 2,6-dimethoxybenzoquinone. Thus, while higher temperatures enable\na moderate increase in overall monomer yield, the quantity of aromatic\naldehyde and carboxylic acid products compounds suitable for downstream\nbiological funneling does not significantly increase. This may be\ndue, at least in part, to more favorable ring-opening oxidation pathways\nat higher temperatures. Regarding the effect of reaction time, while\nproduct yields increase from 0–2 h reaction time, yields decrease\nat longer times ( Figure 3 D), likely also due to oxidative product degradation. On the\nbasis of these studies, we used 4 wt % Co(OAc) 2 ·4H 2 O, 4 wt % Mn(OAc) 2 ·4H 2 O, 2 wt %\nNaBr in acetic acid, at 120 °C for 2 h for subsequent\nreactions. These conditions yielded a total of 0.59(7) mmol/g of acetylated\noligomers (13 wt %) of monomeric products with acetyl vanillic acid\nand acetyl syringic acid being the main components identified by UHPLC\nand LC-MS ( Figure 4 , Table S11 ). As only very small quantities\nof RCF monomers are present in the starting acetyl oligomer materials\n(0.4 wt %), the monomers generated by oxidation demonstrates the ability\nof these conditions to achieve C–C bond cleavage of dimers\nand higher molecular weight species. Overall, the average yield of\nadditional oxidation products produced through oxidative C–C\nbond cleavage is 0.24 mmol/g of acetylated RCF oil. This quantity\nreflects a 17% increase in the acetylated monomer yield, relative\nto the 1.4 mmol monomers/g from the original acetylated RCF oil ( eqs 1 and 2 ). 1 2 Figure 4 Monomer content in the\nstarting acetyl oligomer material and resulting\noxidation mixture (products quantified by UHPLC and LC-MS). Oxidation\nyields are an average of four runs. Wt % values expressed as the weight\nof total oxidation products/weight of acetyl oligomer substrate. Numerical\ndata for this figure and the quantities of the additional monomeric\nproducts in the gray bar are provided in Table S11 . It is also important to note that\nthe catalytic conditions used\nhere with acetic acid, 2 h residence time, and 120 °C with Co/Mn/Br\nare directly inspired by those of the MC process to manufacture terephthalic\nacid from p -xylene at ∼80 MMT per year scale.\nPrevious research on MC oxidation conditions indicate that acetic\nacid can undergo degradation to CO or CO 2 , but the losses\nare minimal even when run up to 175–225 °C, which, as\nwe show in our work, is higher than the temperature needed for C–C\nbond cleavage in lignin. 43 It is also known\nthat the Co/Mn/Br oxidation catalyst can be reused for many years\nwith little loss in activity. 22 We are\noptimistic that the same beneficial features will be true for the\npresent system. Biological Funneling of Oxidation Products\nfrom Acetyl RCF Oligomers The monomers produced from the\noxidation of poplar RCF oligomers\nresemble ideal substrates for biological funneling, wherein heterogeneous\nmixtures of aromatic monomers are catabolized to a single product. 30 , 31 , 33 , 44 − 46 Previous metabolic engineering of the aromatic catabolic\nbacterium Pseudomonas putida KT2440 has demonstrated\nthe utilization of S-type monomers (syringate and syringaldehyde)\nas a source of carbon and energy 47 and\nthe conversion of G-type monomers (vanillate and vanillin) to the\nvalue-added chemical cis , cis -muconic\nacid. 48 To leverage this catabolic capability,\noxidation products from poplar acetyl RCF oligomers ( Figure S7; Table S12 ) were treated with aqueous base (NaOH)\nto precipitate the metal catalysts and hydrolyze acetyl groups to\npromote bioavailability. The resulting mixture consisted of phenolic\nmonomers and acetate ( Table S13 ), both\nof which can be catabolized by P. putida . 49 , 50 It is important to note that P. putida can also\ngrow with acetylated aromatic compounds (acetyl benzoate and acetyl\nvanillate) as the sole source of carbon and energy ( Figure S8 ) and convert them to biomass or products ( Figure S9 ), but these were not present in the\nbase-treated substrate. Engineered strains of P. putida ( Table S14 ) were used to consume and convert\nthe four major oxidation products in base-treated oxidation mixtures,\nnamely syringate, syringaldehyde, vanillate, and vanillin. First,\nshake flask experiments with P. putida strain CJ486 47 demonstrated the ability of the strain to rapidly\nconsume all four of these as model compounds ( Figure S10; Figure S11A ). The same strain was then cultivated\nin minimal medium with 10% v/v base-treated, acetyl RCF oxidation\nmixtures, and all aromatic monomers were once again consumed within\n24 h ( Figures S11B–D ). Next, P. putida strain CJ781 48 was\nemployed to convert these mixtures to muconate ( Figure 5 A). As expected,\nthe model compounds syringate and syringaldehyde were consumed as\nsources of carbon and energy via 3- O -methylgallate\n(3MGA), while vanillate and vanillin were converted to muconate at\n100% molar yield ( Figure 5 B, Figure S12 ), demonstrating viability\nof the engineered pathway. Similar outcomes were observed when strain\nCJ781 was cultivated in minimal medium with 10% v/v base-treated,\noxidized acetyl RCF oil. Syringate and syringaldehyde were consumed\nwith little to no accumulation of 3MGA, and vanillate and vanillin\nwere converted to muconate at 100% molar yield for all replicates\n( Figure 5 C; Figure S12 ). Furthermore, evaporation and abiotic\nconversion of aromatic compounds in the RCF-derived substrates was\nnegligible, as evidenced by a lack of compositional changes in cell-free\nmedia incubated under the same conditions ( Figure S13 ). Figure 5 (A) Metabolic pathway for oxidation products from acetyl\nRCF oil\noligomers in P. putida strain CJ781. Base treatment\nof the oxidation products hydrolyzed acetyl groups from aromatic monomers,\nyielding a mixture of syringaldehyde, syringate, vanillin, and vanillate.\nSyringaldehyde and syringate are converted to biomass (growth) via\n3 -O -methylgallate, gallate, and 4-oxalomesaconate.\nVanillin and vanillate are converted to muconate via protocatechuate\nand catechol. (B) Strain CJ781 cultivated in M9 minimal medium with\n1 mM/each of the model compounds syringaldehyde, syringate, vanillin,\nvanillate. (C) Strain CJ781 cultivated in M9 minimal medium with 10%\nv/v oxidation products from acetyl RCF oil. Gallate and catechol were\nnot detected at significant concentrations in any of the experiments,\nand 4-oxalomesaconate was not measured. All cultures contained 5 mM\nglucose at time zero, and glucose was fed to a concentration of 5\nmM every 24 h to support growth. Error bars represent the standard\ndeviation from the mean of three biological replicates. Numerical\ndata are provided in Table S15 , and additional\nreaction replicates for the experiment in (C) are shown in Figure S12 ."
} | 6,276 |
36824217 | PMC9941318 | pmc | 7,224 | {
"abstract": "Intelligent sensor systems are essential for building modern Internet of Things applications. Embedding intelligence within or near sensors provides a strong case for analog neural computing. However, rapid prototyping of analog or mixed signal spiking neural computing is a non-trivial and time-consuming task. We introduce mixed-mode neural computing arrays for near-sensor-intelligent computing implemented with Field-Programmable Analog Arrays (FPAA) and Field-Programmable Gate Arrays (FPGA). The combinations of FPAA and FPGA pipelines ensure rapid prototyping and design optimization before finalizing the on-chip implementations. The proposed approach architecture ensures a scalable neural network testing framework along with sensor integration. The experimental set up of the proposed tactile sensing system in demonstrated. The initial simulations are carried out in SPICE, and the real-time implementation is validated on FPAA and FPGA hardware.",
"conclusion": "5. Conclusion The mixed signal hardware for a neural network based on sensor-neuron crossbars using an FPAA and FPGA cluster is the focus of the study in the article. The sensor-neuron crossbar neural network shows the analog domain computation for the input layer and the digital domain computation for dense layers. An equivalent LIF circuit is designed using CAM and is implemented on Anadigm AN231E04 ICs. The proposed sensing module is then used to implement a tactile sensing application for a Braille and Morse character identification system. The simulation results show that the proposed model is accurate and power-efficient in the temporal domain. The FPAA platform enables complex circuit design much more easily using configurable analog modules. The proposed prototyping approach helps to optimize the mixed-signal sensor-neural network designs before being deployed for on-chip implementations.",
"introduction": "1. Introduction Field-Programmable Analog Arrays (FPAA) are the analog counterparts to the more popular Field-programmable gate arrays (FPGA) (Farsa et al., 2019 ; Azghadi et al., 2020 ; Yu et al., 2020 ). The ability to program and configure FPGAs has resulted in numerous applications being developed in a short period. In this modern era, reconfigurable computing largely considers only the digital VLSI implementations and the fact is that mostly people turn a blind eye toward the possibilities with analog computing (Azghadi et al., 2020 ; Yu et al., 2020 ; García Moreno et al., 2021 ). In contrast, most sensors detect signals in the analog domain and require analog interface circuits for further processing. Furthermore, the progress in edge artificial intelligent computing has forced the inclusion of more computing modules next to sensors for efficient data processing. This makes a strong case for considering analog computing as a natural approach to be used next to sensors. The FPAA processors consist of a set of reconfigurable analog circuit blocks (Sekerli and Butera, 2004 ). These blocks consist of switched capacitor logic that can be programmed to realize various analog computing operations. Such a system can easily build multipliers, adders, and integral and differential operations. FPAA applications that involve signal processing or data converters can find immediate applications to be used in conjunction with sensors. Another possibility is to implement intelligent data processing using analog neural networks. Various neural networks and neuron models are implemented with FPAA (Rocke et al., 2005 ; Maher et al., 2006 ; Schlottmann and Hasler, 2011 ). Commercially available FPAA AN221E04 was used to build a 2-input, 1-output, 5-intermediate neuron model in Rocke et al. ( 2005 ) and Maher et al. ( 2006 ). A feed-forward neural network trained with the MNIST dataset is implemented using the AN231E04 FPAA Anadigm in García Moreno et al. ( 2021 ). The implementation of FPAA of neuron models such as Hodgkin Huxley and FitzHugh-Nagumo neurons was successfully tested in the past (Zhao and Kim, 2007 ; Joubert et al., 2012 ; Khanday et al., 2019 ; Natarajan and Hasler, 2019 ). These success stories indicate the wide possibilities with FPAA-based computing. This paper explores the combined use of FPGA and FPAA arrays as a prototyping tool to test an integrated solution for real-time tactile sensing, recognition, and classification. This uses the popular neuron model, “Leaky Integrate and Fire,” for the first neural network layer implemented on FPAAs. The remaining neural network layers are implemented in the digital domain using FPGAs. This takes the best of both worlds, where the sensing layer is analog while the remaining layers responsible for classification are implemented in FPGAs. The major contributions of the work are (1) to convey the practical demonstration of the use of tactile sensing with FPAAs, (2) to show a unique scalable array architecture built with FPAAs for near-sensor computing, and (3) to exhibit the possibilities of mixed-signal pipelines sequentially built on FPAA and FPGA to create large-scale neural networks next to sensors.",
"discussion": "3. Results and discussions The proposed mixed-mode neural computing is experimentally demonstrated on a system for identifying braille and morse code symbols ( Figure 5 ). The sensor-neuron crossbar array's input layer acts as a tactile patch for blind users to press the braille and the morse code characters. In the Braille system, each character is represented by 6 points (D1, D2, D3, D4, D5, D6) (Chithra et al., 2022 ). Some of the braille symbols have the same representations. For example, the character A and the number 1 have the same representations. Hence, we use two additional dots (D7, D8) to differentiate them. Thus, we represent 125 braille characters. In the case of Morse code, the repetitive combination of dots and dashes forms alphabets and numbers. Here, 10 dots are used for representing morse code each column representing either dots or dashes. The selection dots (D11, D12) represent braille and morse code selection. The designed system implemented 62 characters (capital letters, small letters, and numbers) for morse code. Hence the data set consists of 40 instances of 187 different symbols of braille and morse code. Hence, the tactile sensing system is implemented using a sensor-neuron crossbar size of 6 × 2. Each dot represents one cell in the sensor-neuron crossbar array. Table 2 shows the characters implemented in the proposed tactile sensing system. The difference in character implementations of braille and morse code are presented in Table 3 . Figure 5 (A) FPAA equivalent circuit model of fully integrated memristive model (Jin and Cui, 2019 ) and (B) The pinched hysteresis characteristics of the memristive model. Table 2 Braille and morse code character implementation on LIF-neuron crossbar array, 1/0 denotes touch sensor is pressed/not pressed. \n { D 11 , D 12 } \n \n { D 7 , D 8 } \n \n Characters \n 0, 0 0, 0 Braille alphabet capital (27 symbols) 0, 0 0, 1 Braille small (26 symbols) 0, 0 1, 0 Braille words (46 words) 0, 0 1, 1 Braille numbers, punctuation & symbols (26 symbols) 1, 1 Morse alphabet capital (26 symbols) 0, 1 Morse Small (26 symbols) 1, 1 Morse code numbers (10 symbols) Table 3 Braille and morse character implementation using the sensor array patch, 1/0 denotes touch sensor is pressed/not pressed. \n Character \n \n { D 1 , D 2 , D 3 , D 4 , D 5 , D 6 , D 7 , D 8 , D 9 , D 10 , D 11 , D 12 } \n \n Braille characters \n A {1, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0} a {1, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0} 1 {1, 0, 0, 0, 0, 0, 1, 1, 0, 0, 0, 0} \n Morse code \n A {1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 1} a {1, 0, 0, 1, 0, 0, 0, 0, 0, 0, 1, 0} 1 {1, 0, 0, 1, 0, 1, 0, 1, 0, 1, 1, 1} 3.1. Training—Simulations The preliminary neuron circuit is simulated using SPICE tools and the equivalent circuit of the same is implemented in FPAA using different CAMs. The real-time implementation of neuron sensor crossbar arrays for the Braille character recognition system uses a cluster of 12 FPAA chips. The weighted summation of the current from each cell is taken through the HL and VL lines. The input layer consists of 6 preneurons and 8 postneurons. The programming of weights requires training the neurons taking into account hardware variability. Simulations are carried out using equivalent models to estimate the weight values. This experiment uses a capacitive touch sensor module (TTP223 touch-sensing IC) and fully integrated memristive model. The generated dataset is used to train the subsequent dense layers using Python program. The measured values of FPAA crossbar output are prone to different types of noise. The noise in the measured data affects the boundary points of 187 class of symbols of braille and morse code. Any small deviations will lead to change in the whole feature set combination. Hence a min-max scalar based preprocessing technique is adopted to remove the noise factors that might be affected at the boundary points. In min-max normalization, the noisy data is scaled up/down using a range based on averaging. Figure 6 shows the first column current values for characters A and B, respectively. For each character, we take 40 samples. The measured data is normalized into a symmetric range after preprocessing. For example, the data range of the measured value of character A is between 2.5 and 3.1, as shown in Figure 6 . With min-max preprocessing, the data range for character A is limited between 2.7 and 2.9. This helps to remove the noise factors affecting the boundary points. Figure 6 Min-max normalized data for i 1 readout of LIF-TMS crossbar array for (A) Character A and (B) Character B. The training for the FPAA crossbar data is done with and without preprocessing technique. The dataset consists of 7480 samples of braille and morse code, i.e., 40 samples of 187 symbols. The generated data set is used to train the subsequent dense layers using the Python program. The trained model consists of 2 hidden layers and an output layer. The first layer has 8 × 14 nodes with relu activation function. The output layer uses a softmax activation function of 187 node size. The trained parameters of the dense layer are then implemented on myRIO FPGA for the real-time implementation. 3.2. Inference—Hardware implementation The proposed design is implemented on the commercially available Anadigm AN231E04 IC. The parasitic capacitance of the hardware components introduces noise in the output of each of the CAM modules, as shown in Figure 7 . Hence, the output spikes have noise content in them as shown in Figure 7D . The chip clock frequency is 4 MHz, and each CAM clock frequency is 62.5 KHz. The spikes occur only when someone touches the touchpad of the tactile/touch sensor. This reference potential acts as a trigger for the sensor neuron model to generate a spike. Figure 7 Hardware implementation output of LIF FPAA implementation with an offset of 1 V: (A) Sum/Diff CAM output, (B) integrator CAM output, (C) the Comparator CAM output with a reference voltage of 0.3 V, and (D) the final output of the sensor-LIF module. The layer1 output is read through AIO pins on the myRIO. The output of the sensor-neuron crossbar array is measured and used to train the dense layers to perform the classification. The noise in the measured data from the sensor-neuron crossbar array affects the boundary points of 187 classes of symbols of braille and morse code. Any small deviations will lead to a change in the whole feature set combination. Hence a min-max scalar-based preprocessing technique is adopted to normalize the measured values (more details in the Supplementary material ). Training of the sensor data is done with and without noise removal. The trained ANN model contains five dense layers with input, hidden, and output layers. The relu activation function is used for all layers, and the output layer uses the softmax activation function (Krestinskaya et al., 2020 ; Newns et al., 2020 ). The trained model is then implemented on myRIO FPGA for real-time implementation. As discussed in Table 1 , the implementation consists of real-time target VI and FPGA target VI. The LABVIEW application of real-time target VI and FPGA target VI is shown in Figure 8 . The open FPGA VI reference will cause FPGA VI to start running ( Table 1 ). The while loop makes the system run continuously for real-time applications. The single-cycle timed loop structures are always used in an FPGA VI, which will execute all functions within one tick of the clock, here we use a 40 MHz global clock. The linear algebra matrix multiply function block and high-throughput add function block work only inside the single-cycle timed loop. For dense layer1, M 1 = 1, N 1 = 8 and P = 14. Correspondingly, the FIFO depths are 8, 112 and 14, respectively, for X , W , and B . The output of the linear algebra matrix multiply function block is a column vector. Hence, the bias values read from the FIFO are converted to arrays of size M 1 × P , here 1 × 8. The bias addition is done using the high-throughput add function block. The output of the present layer forms the input to the subsequent layers. The graphical programming of real-time target VI and FPGA VI is presented in Figure 8 . Figure 8 Single dense layer implementation on LABVIEW application. (A) Real-Time target VI and (B) FPGA target VI. Table 4 shows the average relative error, RE avg of hardware dense layer output in comparison with software results. RE avg can be calculated as 1 P 1 ∑ l = 1 P 1 | Y s l - Y h l | Y s l , where Ys l and Yh l denote the software and hardware results of the output layer. The table shows the relative output error for varying FIFO integer length on myRIO. Each symbol in the braille alphabet has different combinations of dots, i.e, for braille character “A” all the FPAA array output is zero except Vo 1 and Vo 3 . This creates sparsity in the subsequent dense layers and output layer. Whereas for braille character “Y,” only Vo 7 and Vo 8 are zeros (according to Table 2 ). Hence there are variations in RE avg for different symbols as shown in Table 4 . The results show that 16 bits representation shows comparable performance with software results. Hence the FIFOs for FPGA implementation are defined for a word length of 16 bits. The system testing accuracy is 65% for braille characters and 75% for morse characters with hardware noise ( Table 5 ). With the preprocessed input data using Min-Max normalization, the performance accuracy is improved to 96 and 98%, respectively, for braille and morse code character detection. Table 5 shows the values of precision (P), recall (R), and F1 score (F1) for 5 characters. Table 4 Average relative error, RE avg , of hardware dense layer output in comparison with software results with varying FIFO integer length. \n Integer word length \n \n Symbol \n \n 2 bits \n \n 4 bits \n \n 8 bits \n \n 12 bits \n \n 16 bits \n A 0.13 0.03 0.01 0 0 D 0.46 0.34 0.23 0.13 0.03 I 0.43 0.22 0.11 0.06 0.01 T 0.89 0.56 0.34 0.19 0.06 Y 1.05 0.72 0.46 0.26 0.09 Table 5 Testing accuracy of Braille and morse recognition system with sensor-neuron crossbar array. \n Braille characters \n \n Morse code \n With noise \n Acc = 65% With filter \n Acc = 96% With noise \n Acc = 75% With filter \n Acc = 98% \n P \n \n R \n \n F1 \n \n P \n \n R \n \n F1 \n \n P \n \n R \n \n F1 \n \n P \n \n R \n \n F1 \n A 1.0 1.0 1.0 1.0 1.0 1.0 0.12 1.00 0.22 0.60 1.00 0.57 D 0.33 1.0 0.50 0.5 1.0 0.67 0.38 1.00 0.55 1.00 1.00 1.00 J 0.0 0.0 0.0 0.22 1.00 0.36 0.40 1.00 0.50 0.8 1.00 0.89 T 0.56 1.0 0.71 1.0 1.0 1.0 0.6 0.58 0.88 1.00 1.00 1.00 Y 1.0 1.0 1.0 1.0 1.0 1.0 1.00 1.00 1.00 1.00 1.00 1.00 Avg 0.58 0.69 0.60 0.90 0.922 0.88 0.71 0.72 0.72 0.92 0.96 0.95 P, R, F1, and Acc denote precision, recall, F1 score, and accuracy, respectively.\n\n4. Discussions 4.1. Scalability in rapid prototyping The proposed mixed-signal development kit is a rapid prototyping solution for different neural computing applications. The paper presents the implementation of braille character and moorse code recognizing system using the same prototype designed. The system can be easily scaled up by adding FPAA arrays and reprogramming the FPGA. This makes the system more flexible in implementing other neural tactile sensing applications. 4.2. Neural architecture optimization The proposed mixed signal processing helps to reduce the required power consumption by optimizing the neural architecture. In the conventional method, the sensed data from each sensors are transmitted for neural computing. i.e., for an M rows and N columns sensor array, the conventional method takes M × N inputs for neural processing. Whereas, our proposed method shown in Figure 1 only needs M + N inputs for neural processing. The reduction in number of inputs directly reduces the size of neural network architecture. Table 6 demonstrates the neural architectural comparison of existing mixed signal prototyping with the conventional implementation techniques using either FPAA or FPGA. The results in the table show there is approximately 30% reduction in the neural architecture size for the mixed signal implementation to achieve the same accuracy of 96% for the braille and 98% for the morse code character recognition system. Table 6 Mixed signal development kit for tactile application: a comparison with the existing implementation methods. \n Number of neurons \n \n Method \n Hidden layer1 , N 1 Hidden layer2 , P Output layer , P 1 FPAA (García Moreno et al., 2021 ) 12 19 187 FPGA (Azghadi et al., 2020 ) 12 19 187 Proposed (FPAA+FPGA) 8 14 187"
} | 4,421 |
30410980 | PMC6202650 | pmc | 7,227 | {
"abstract": "Marine\norganisms such as mussels have mastered the challenges in\nunderwater adhesion by incorporating post-translationally modified\namino acids like l -3,4-dihydroxyphenylalanine (DOPA) in adhesive\nproteins. Here we designed a catechol containing elastomer adhesive\nto identify the role of catechol in interfacial adhesion in both dry\nand wet conditions. To decouple the adhesive contribution of catechol\nto the overall adhesion, the elastomer was designed to be cross-linked\nthrough [2 + 2] photo-cycloaddition of coumarin. The elastomer with\ncatechol moieties displayed a higher adhesion strength than the catechol-protected\nelastomer. The contact interface was probed using interface-sensitive\nsum frequency generation spectroscopy to explore the question of whether\ncatechol can displace water and bond with hydrophilic surfaces. The\nspectroscopy measurements reveal that the maximum binding energy of\nthe catechol and protected-catechol elastomers to sapphire substrate\nis 7.0 ± 0.1 kJ/(mole of surface O–H), which is equivalent\nto 0.10 J/m 2 . The higher dry and wet adhesion observed\nin the macroscopic adhesion measurements for the catechol containing\nelastomer originates from multiple hydrogen bonds of the catechol\ndihydroxy groups to the surface. In addition, our results show that\ncatechol by itself does not remove the confined interstitial water.\nIn these elastomers, it is the hydrophobic groups that help in partially\nremoving interstitial water. The observation of the synergy between\ncatechol binding and hydrophobicity in enabling the mussel-inspired\nsoft adhesive elastomer to stick underwater provides a framework for\ndesigning materials for applications in tissue adhesion and moist-skin\nwearable electronics.",
"conclusion": "Conclusion We have used polymers\nwith well-defined chemistry, JKR geometry\nfor adhesion measurements, and interface-sensitive SFG spectroscopy\nto understand the role of catechol in underwater adhesion of mussel-inspired\npolymers. Having both the elastomers of similar bulk mechanical properties,\nbut a difference in only the surface-active catechol groups being\nprotected and deprotected, helped to identify the importance of the\ndihydroxyl groups and the hydrophobic groups in adhesion. The dry\nadhesion of deprotected elastomer is higher than the protected elastomer.\nDirect probing of the contact interface using SFG spectroscopy reveals\nthat the strength of the acid–base interactions of polar groups\nwith surface O–H groups on the sapphire substrate is very similar\nfor both protected and deprotected elastomers. The stronger adhesion\nstrength obtained from macroscopic measurements can be explained by\nthe dihydroxy chemistry of catechol despite similar interaction strength\nas observed in SFG spectra. The energy required to break multiple\nbonds concurrently is higher than the sequential breaking of monodentate\nbonds of similar strength, thus increasing the adhesion strength of\ndeprotected elastomer compared to protected elastomer. In wet\nenvironment, the deprotected elastomer has a higher adhesion\ncompared to protected elastomer. The SFG experiments reveal that the\npresence of “liquid-like” water in the contact interface\nfor both the polymers. The SFG and adhesion data combined show the\ninterplay of hydrophobicity and catechol binding to enhance adhesion.\nBoth elastomers have a patchy contact underwater. The protected and\ndeprotected elastomers have a similar water contact angle (similar\nhydrophobicity) and similar decrease in wet adhesion compared to dry\nadhesion indicating the assistance of hydrophobicity in achieving\nsimilar patchy contact underwater. The formation of patchy contact\nbetween a hydrophobic and hydrophilic interface underwater is consistent\nwith the previous studies. 66 , 78 The higher adhesion\nstrength of the deprotected elastomer is due to multimodal bonding\nof catechol with the hydroxylated surface. Hence, the interplay of\nhydrophobicity to remove water from the interface and the role of\ncatechol in forming multiple hydrogen bonds are proposed from the\ncombination of SFG spectroscopy and adhesion measurements in this\nstudy. Although the synergistic effect of lysine to displace cations\nand water from the surface for enhanced interfacial binding of catechol\nis also essential to understand the adhesion of Mfp, our current focus\nwas to study the importance of hydrophobicity in improving underwater\nadhesion of mussel-inspired polymers. 79 , 80 In the future,\nthe critical understanding of the interplay of hydrophobicity and\nmussel-inspired dihydroxy chemistry in synthetic adhesives can be\nput to use in designing adhesives for applications in tissue adhesion\nwhere the presence of moisture or physiological fluids cannot be avoided.",
"introduction": "Introduction Mussels,\ncaddisflies, sandcastle worms, barnacles, and many other\nspecies use biological adhesives to form interfacial bonds to various\nsubstrates in the presence of water, 1 − 4 and among these examples, the mussel holdfast\nis one of the most well-studied biological adhesive system. 5 − 7 Mussels secrete a series of adhesive proteins which solidify to\nform the byssus thread—the holdfast of mussels. 8 The byssus thread comprises of more than 15 different mussel\nfoot proteins (Mfp), 9 − 12 and among them, a set of low molar mass (<11 kDa) foot proteins\n(Mfp -3F, -3S, and -5) are found at the adhesive interface. 4 , 13 These low molecular weight proteins are rich in l -3,4-dihydroxyphenylalanine\n(DOPA, more than 20 mol %). 7 , 13 − 16 The DOPA-rich proteins adhere strongly to various surfaces (adhesion\nenergy ≈ 3–15 mJ/m 2 ), and it has been suggested\nthat the strong adhesion of DOPA is due to its Janus-like nature. 17 − 22 Depending on the surface, the catechol binding group of DOPA is\nproposed to either interact through hydrogen bonding (to polar and\nmetal oxide surfaces), coordination bonds (to metal oxides and metallic\nsurfaces), or hydrophobic interactions (to nonpolar surfaces). 17 , 19 , 20 , 23 − 28 Single molecule atomic force microscopy (AFM) experiments 23 , 29 − 32 have shown that catechol interacts with TiO 2 surfaces\nwith a binding energy of 22 kcal/mol, which is similar to the bond\nenergy of a covalent single bond. 23 When\nthe surfaces of main group metal oxides such as SiO 2 and\nAl 2 O 3 were studied, lower adhesion strengths\nwere found, and it has been suggested that they interact exclusively\nthrough hydrogen bonding in ambient conditions. 30 Although AFM studies provide detailed information on binding\nenergies, there are unresolved inconsistencies in the values of DOPA-TiO 2 pull-off forces in the literature. 23 , 29 , 32 Inspired by the strong interactions\nof catechol to various surfaces,\na substantial amount of adhesive polymers containing catechol have\nbeen synthesized and examined for their ability to improve adhesion\nto surfaces. 33 These mussel-inspired adhesives\nhave been shown to adhere to various surfaces such as metal oxides\nand biological tissues. 34 − 39 But only in a few of these studies have adhesive joints been constructed\nin the presence of water. 34 , 40 − 42 Moreover, most of these studies in the DOPA-polymer literature utilize\nlapshear strength measurements to demonstrate the importance of catechol\nin increasing adhesion strength. 41 , 43 , 44 The lapshear test cannot accurately distinguish the\ncontribution of interfacial interactions and cohesive strength to\nadhesion. Catechol can also play an important role in increasing the\ncohesive strength of mussel-inspired polymers. 45 For example, catechol can be oxidized and cross-linked\nto increase the cohesive strength of the material, thereby making\nit difficult to distinguish the role of catechol in interfacial adhesion. 39 , 46 , 47 Also, these adhesion studies\nlack control samples of similar bulk properties 48 and without catechol groups, which is necessary to understand\nthe role of catechol in increasing underwater interfacial adhesion.\nHerein, taking these factors into consideration and inspired by our\nrecent design of a photocurable, mussel-inspired adhesive with remarkable\nunderwater lapshear adhesion strength (0.65 ± 0.09 MPa), 49 we have designed an experimental approach to\ntest the role of catechol in underwater adhesion. To conclusively\nidentify the role of catechol in dry and wet interfacial\nadhesion, we have designed two polymers containing either catechol\n(deprotected) or protected-catechol (protected) pendant groups. To\ndecouple the interfacial and cohesive contributions to the overall\nadhesion, the polymers were designed with pendant coumarin units,\nwhich undergo a [2 + 2] photocycloaddition to provide catechol independent\ncross-linking. 49 − 52 Both these polymers have similar glass transition temperatures and\nviscoelastic properties, thereby minimizing the influence of bulk\nphysical property variations on adhesion measurements. We have chosen\nprotected elastomer as our control instead of varying the molar ratio\nof the catechol groups because the later would vary the bulk physical\nproperties and make the interpretation of adhesion results difficult. Johnson Kendall Roberts (JKR) adhesion geometry 53 is used to measure force (pull-off force) and work required\nto separate two substrates by bringing a glass lens in contact with\nthe elastomeric adhesives in dry and wet conditions. The JKR adhesion\ntest in conjunction with low pull-off velocities helps in reducing\nthe contribution from the energy spent in deforming the bulk material\nfrom the total work done in separating the two substrates. 54 Because the adhesive is cross-linked and soft\n( G ′ ≈ 7 kPa), we obtain good molecular\ncontact between the adhesive and the glass probe, thus reducing the\neffect of roughness. In addition to adhesion strength measurements,\nwe have used interface-sensitive sum frequency generation (SFG) spectroscopy\nto directly probe the contact interface in dry and wet conditions.\nSFG is a second-order nonlinear optical spectroscopic technique that\nprovides direct information on the concentration and orientation of\nthe interfacial molecules. 55 , 56 SFG probes a depth\nof a few nanometers compared to hundreds of nanometers probed by the\ninfrared or Raman spectroscopic technique. We have combined the interface-sensitivity\nof SFG with an experimental design of protected or deprotected elastomers\nplaced in contact with a sapphire prism in either dry conditions or\nin the presence of water. By observing the shift of the sapphire O–H\npeak after bringing in contact with the protected or deprotected elastomers,\nwe directly measured the interactions of catechol with the surface\nO–H groups and the role it plays in hydrogen bonding with a\nhydrophilic surface. 57 , 58 The current work provides the\nframework for the design of effective underwater adhesive elastomers\nbased on the information provided from SFG and adhesion measurements.\nAlso, this study gives critical insight into the interplay of catechol\nin hydrogen bonding and polymer hydrophobic groups in removing interstitial\nwater as a means of providing effective underwater adhesion.",
"discussion": "Results\nand Discussion Elastomer Design To create an elastomer,\na statistical\ncopolyester was synthesized using a N , N ′-diisopropylcarbodiimide (DIC) assisted polyesterification\nreaction of three N -functionalized diols and sebacic\nacid ( Scheme S1 and Figure 1 ). 59 The first\ndiol with pendant aliphatic hydrocarbons ( S ) provides\nhydrophobicity and lowers the glass transition temperature ( T g ≈ – 45 °C), which makes\nit easier to spread at room temperature. 52 The second coumarin diol ( C ) undergoes [2 + 2] cycloaddition\nwhen exposed to UV light of wavelength (λ) ∼350 nm, which\nconverts the viscous polymer to an elastomer. 50 , 51 The third component is an acetonide protected-catechol diol ( D pr ), which is deprotected under acidic conditions\nto provide catechol and is expected to increase underwater adhesion\nupon deprotection. 60 The feed ratio of\nmonomers to form the polymer was chosen to be S : C : D pr 65:5:30 (mol %), and from 1 H NMR ( Figure S1A ) the actual composition\nwas calculated to be 63:5:32 and the molar mass as detected from GPC\n( M n,GPC ) of the protected polymer was\n11.3 kDa (dispersity = 1.6). The polymer with catechol groups (deprotected)\nwas obtained by the reaction of protected polymer (500 mg) with trifluoroacetic\nacid (5.0 mL) in methylene chloride (10 mL) for 2 h at room temperature\nunder N 2 (details provided in the Supporting Information ). The disappearance of protons from the 1,2-acetonide\ngroup (δ = 1.63 ppm, -C(C H 3 ) 2 ) in the 1 H NMR spectrum ( Figure S1B ) and the disappearance of acetonide C–H bend signal at 1498\ncm –1 along with the appearance of a broad absorption\nband at 3100–3650 cm –1 assigned to O–H\ngroups of catechol in the FT-IR spectrum ( Figure S2 ) indicate the successful deprotection of acetonide groups.\nChoosing the protected elastomer as the control avoids the problem\nof comparing polymers with a different molar mass and/or T g ( Table S1 ). From the rheological\nmeasurements of cross-linked protected and deprotected polymers, the\nfrequency responses of storage ( G ′ ) and loss ( G ′′ )\nmoduli were quantified ( Figure S3 ). The\nrheological response of both the elastomers after cross-linking were\nsimilar, indicating the appreciable similarity of bulk behavior of\nthe protected and deprotected elastomers and minimal interaction between\ncatechol units themselves. The usage of coumarin for cross-linking\nalso gives similar modulus for the elastomers (discussed in detail\nlater), which is a prerequisite for comparing the adhesive properties\nsince the adhesion values could be influenced by both interfacial\nand bulk mechanical properties. Figure 1 Chemical structures of the polymers designed\nfor the adhesion and\nspectroscopic measurements. Adhesion Measurement The adhesion of a hemispherical\nglass lens to protected and deprotected elastomers coated on oxidized\npolydimethylsiloxane (PDMS) elastomer was tested (PDMS provides an\nelastic backing; see Supporting Information for details). The polymers coated over PDMS elastomers were then\ncross-linked using UV-A irradiation (λ = 350–420 nm,\nintensity on the substrate = 50 mW/cm 2 ) for 10 min to form\nan elastomer. To ensure a uniform coating of elastomer on PDMS sheets,\nthe elastomer-coated PDMS sheets were analyzed by fluorescence microscopy.\nThe elastomer-coated films fluoresced under a DAPI filter in contrast\nto bare PDMS, confirming the uniform coverage of the elastomer on\nthe PDMS sheets ( Figure 2 A). Figure 2 Characterization of the substrates used for adhesion force measurements.\n(A) (L–R) Fluorescence microscopy images of PDMS sheet, protected\nand deprotected elastomers coated PDMS sheets before and after periodate\ntreatment. The images shown are under 5× magnification and the\nscale bars (white box at bottom right) in the images correspond to\n400 μm. (B) The UV–vis absorption spectra of the polymer\nfilms on quartz substrates before and after cross-linking. The presence\nof UV absorption peak λ max ≈ 280 nm indicates\nthe presence of unoxidized catechol after cross-linking. (C) Single\nbounce ATR/FT-IR spectra of PDMS sheet and both elastomers coated\non PDMS sheets recorded by exposing the coated side to the IR beam.\nThe spectra of the elastomer-coated sheets match those of the respective\npolymers and are different from the FT-IR spectra of the PDMS sheet. Since catechol is prone to oxidation\nreactions, it is important\nto show that the catechol moieties are intact after exposure to the\ncross-linking conditions. 61 The chemical\nstabilities of the protected and deprotected elastomers on UV-A exposure\nwere analyzed by UV–vis and ATR/FT-IR absorption spectroscopies\nafter the polymer films were exposed to UV-A irradiation. The UV–vis\nspectra of elastomers show that the catechol absorption peak (π\n→ π*, λ max = 280 nm) remains intact\nafter cross-linking ( Figure 2 B), indicating the stability of catechol upon UV-A exposure\n(upon oxidation, red shift is expected). 62 The absorption bands (λ max = 310, 320, and 340\nnm) corresponding to coumarinyl groups disappear with UV-A exposure\nconfirming the completion of cross-linking reactions. 51 Single bounce ATR/FT-IR absorption spectra of elastomers\nwere also collected by placing the elastomer-coated side of the substrates\ntoward the IR beam. The characteristic absorption peaks in Figure 2 C match the IR absorption\nsignature of corresponding polymers ( Figure S2 ). In the protected elastomer spectrum, it is seen that the acetonide\nprotection group (C–H bend, 1498 cm –1 ) remains\nintact after cross-linking reactions. The results from characterization\nof the elastomers provided confidence for the following experiments\nwhich examine the role of catechol in dry and wet adhesion. Figure 3 A shows\nthe schematic diagram of the in-house experimental set up used for\nthe adhesion force measurements using a JKR geometry. During the adhesion\nstrength measurements, a hemispherical glass lens is brought in contact\nwith the elastomers in the absence (dry) and presence (wet) of water\nto a preload of −1 mN. The lens is then retracted back after\nletting it equilibrate to measure the maximum force for separating\nthe contact, which is recorded as the pull-off force ( Figure 3 B). 53 Figure 3 Adhesion\nstrength measurements. (A) Schematic diagram of the home-built\nset up showing the JKR geometry used for adhesion force measurements.\n(B) Representative force runs of the protected and deprotected elastomers\nin dry and wet environments showing the force and work done to separate\nthe two surfaces. (C) Pull-off forces (left axis), work of adhesion\n(right axis), and (D) work done to separate the protected and deprotected\nelastomers in dry and wet environments from the glass surface when\nloaded to −1 mN force followed by steady hold for 3 min and\nunloaded at a rate of 0.4 μm/s. The data represented here are\npresented as mean ± standard deviation (SD), and “ * ” represents the statistical significance among\nthe samples using a Tukey mean comparison test ( p < 0.05). Error bars (SD) are evaluated using at least three measurements\nfor each condition. Higher forces are required to separate the contact\nof the glass lens with the deprotected elastomer than the contact\nof the protected elastomer in dry and wet environments. There is a\nreduction in work of adhesion for both the elastomers in the presence\nof water. Figure 3 C shows\nthe pull-off force (left axis) and work of adhesion (right axis) calculated\nusing the JKR model for the protected and deprotected elastomers in\ndry and wet conditions. Under dry conditions, the work of adhesion\nof the deprotected elastomer (1.80 ± 0.21 N/m) is significantly\nhigher than the protected elastomer (0.51 ± 0.01 N/m). During\nthe experiments, the loading and unloading cycles were monitored using\nan optical microscope from which it was observed that the mode of\nfailure for protected and deprotected elastomers was adhesive and\ncohesive for the dry measurements, respectively ( Figure S5 ). Hence it is possible that the interfacial adhesion\nstrength of the deprotected elastomer could be even higher than what\nis described here. Since the protected and deprotected elastomers\nhave a similar mechanical response, it is safe to conclude that the\ndifferences are due to interfacial interaction of the elastomers. In wet conditions, both protected (0.15 ± 0.03 N/m) and deprotected\n(0.38 ± 0.05 N/m) elastomers showed appreciable adhesion in contrast\nto the PDMS sheet (∼0 N/m) ( Figure S4C ). This indicates that in these experiments PDMS is not contacting\nthe glass substrate. However, in wet conditions both the elastomers\nhave a significant decrease in adhesion strength as compared to their\ncorresponding dry values. Catechol containing polymers have been shown\nto have enhanced adhesion to mica surfaces under acidic conditions\nsince they are known to oxidize readily at higher pH. 19 To investigate the pH dependence of adhesion, we measured\nthe adhesion strength of our elastomers at pH 3, 6.5, and 9 ( Figure S4A ). For deprotected elastomer, the adhesion\nstrength is insensitive to pH changes (based on statistical comparison),\nwhereas the protected elastomer showed a reduction in adhesion (0.06\n± 0.04 N/m) at pH 9. At pH 9, the glass surface becomes more\nnegative and can lead to electrostatic repulsion between hydrophobic\npolymers and the glass surface. This repulsion of the similar charge\ndensities might be the reason for reduced underwater adhesion of protected\nelastomer at higher pH. 63 However, we did\nnot observe any statistical difference in the pull-off forces for\nthe deprotected elastomer with the increase in pH. We speculate that\nat a higher pH, the hydroxyl group of catechol groups deprotonates\nto form either coordination bonds with silicon or quinone, which then\nact as an efficient hydrogen bond acceptor to the surface hydroxyl\ngroups and maintain the adhesion strength. 20 , 64 After wet measurements at pH 9 were performed, both the elastomers\nwere submerged in an aqueous solution of 10 mM sodium periodate (NaIO 4 ) for 2 h. Periodate (IO 4 – ) treatment\ncan cause oxidation of catechol to form quinone and its tautomer and\ndecrease adhesion of the deprotected elastomer. 19 , 43 We observed that the deprotected elastomer does not stick after\nperiodate treatment, but the protected elastomer still retains its\nadhesion compared to pH 9 ( Figure S4A ).\nIt is possible that the deprotected elastomer undergoes extensive\noxidation during periodate treatment, resulting in the loss of adhesion,\nwhile the protection prevents this oxidation reaction. But, it was\nintriguing that upon oxidation, the deprotected elastomer did not\nretain adhesion comparable to the protected elastomer. We found that\nthe periodate treatment not only caused the catechol oxidization reaction\nbut also ruptured the elastomer film and exposed the oxidized PDMS\nsheet. This introduces roughness which was evident in fluorescence\nmicroscopy images ( Figure 2 A). Both roughness and exposure of PDMS lead to loss of underwater\nadhesion. A similar trend for the work of adhesion was observed\nwhen the\nwork done was calculated by integrating the area under the unloading\ncurve of force as a function of displacement ( Figure 3 D). The time axis of the graph in Figure 3 B can be converted\ninto displacement by multiplying time with the unloading rate of 0.4\nμm/s. This work comprises the energy required to break the interfacial\nbonds and also the elastic/viscous work done in stretching the elastomer. 54 From Figure 3 D, it can be observed that more work is required to\nseparate the contact for the deprotected elastomer than the protected\nelastomer. To further investigate the detailed adhesion mechanisms\nof these elastomers, we performed SFG spectroscopy experiments of\nthe contact interface, which are described in the following section. SFG Spectra of the Elastomer-Substrate Contact Interface For SFG experiments, we have used total internal reflection geometry\nto probe the contact interface in both dry and wet conditions. The\nexperimental details are provided in the methods section. The generation\nof SFG signals requires a breakdown in the symmetry of dipole orientation,\nand this happens only for the interfacial molecules at the contact\ninterface. This interface selectivity allows us to interpret the presence\nor absence of molecules at the contact interface. For example, if\nthe contact is dry, we should not observe water bands between the\n3100–3600 cm –1 (or 2300–2700 cm –1 for D 2 O) region. If water is present at\nthe contact interface, then the location of the water peak indicates\nthe nature of hydrogen bonding of the confined water. Since the bulk\nis centrosymmetric, the SFG signals are interface-specific and are\nnot swamped by the signals from the bulk elastomers. A sapphire\n(Al 2 O 3 ) prism-like glass ( Figure S6 ) with surface hydroxyl functional groups (O–H)\nwas brought in contact with the protected and deprotected elastomer-coated\nPDMS lenses in dry and wet environments ( Figure 4 A). We expect SFG signals to be generated\nfrom only a few nanometer-thick interfacial layer between the elastomer\nand the sapphire substrate. In addition, the shift in the surface\nsapphire O–H peak can be used to calculate the strength of\nthe acid–base interactions (hydrogen bonding is a subset of\nacid–base interactions). 57 For example,\nin the previously reported study, the ester groups in poly(methyl\nmethacrylate) showed stronger interaction (O–H peak is shifted\nto 3580 cm –1 compared to free O–H peak at\n3720 cm –1 ) than functional groups in polystyrene\n(3645 cm –1 ) when in contact with a sapphire substrate. 58 Similarly, since the interfacial adhesion of\nthe deprotected elastomer is higher, we speculated that there might\nbe a larger shift for the deprotected elastomer as compared to protected\nelastomer, which would indicate a stronger bond between the hydroxyl\ngroups of catechol and the sapphire. Figure 4 SFG spectra of elastomer–substrate\ncontact interface. (A)\nSchematic of total internal reflection geometry used to probe the\nmechanical contact of elastomer-coated PDMS lens with a sapphire prism\nin dry and wet (D 2 O) environments. (B) Top and bottom panels\nshow SFG spectra (SSP polarization) of protected and deprotected elastomers\nin contact with a sapphire substrate in dry and wet (D 2 O) conditions, respectively. Spectra were collected in two regions\n2700–3200 cm –1 and 3100–3800 cm –1 separately and plotted together to show the differences.\nThese spectra show the hydrocarbon signature from the elastomer and\nsapphire hydroxyl groups. (C) SFG spectra (SSP polarization) in the\nO–D stretching region of the protected and deprotected elastomers\nin contact with a sapphire substrate under wet (D 2 O) conditions.\nFeatures in the region from 2200–2800 cm –1 indicate the presence of D 2 O in the contact region. Figures 4 B, C, and S7 show\nthe SFG spectra in both SSP ( Figure 4 B and C) and PPP\n( Figure S7 ) polarizations (the polarization\ncombination is for three beams: SFG, visible, and IR, respectively,\nwhere S and P are components of electric field perpendicular and parallel\nto the plane of incidence). Depending upon the polarization combination,\ninformation about the molecular group orientation can be inferred. 65 Figure 4 B shows the SSP spectra comparison of both elastomers in dry\n(top panel) and wet (bottom panel) environments in the hydrocarbon\nregion (C–H stretch, 2800–3100 cm –1 ) and the sapphire region (O–H stretch, 3200–3800 cm –1 ). The wet measurements were done in D 2 O to avoid the overlap of signal from O–H stretches of water\n(H 2 O) and sapphire. In both elastomers, the peak at ∼\n2960 cm –1 represents the vibration of the aliphatic\nside chain and polymer backbone C–H groups. The contact of\nPDMS with sapphire substrates results in a very different SFG spectrum,\nand this again confirms that the protected or deprotected elastomer\nlayers are intact upon contact with the sapphire substrate. 66 , 67 The sapphire O–H peak positions for both dry contacts of\nthe protected and deprotected elastomers are very similar. On the\nbasis of three independent measurements, using three different lenses\ncoated with elastomers from two different molecular polymers, the\naveraged peak position of sapphire O–H region was 3552 ±\n6 cm –1 for protected elastomer and 3557 ± 22\ncm –1 for the deprotected elastomer. These peak positions\nwere obtained by fitting the data using a Lorentzian equation (details\nin Supporting Information ). Surprisingly,\nthe similar shifts in sapphire O–H peak indicate a similar\nstrength of acid–base interactions for both elastomers. The shift in sapphire O–H was further scrutinized with a\nfirst moment analysis of the peak distribution (details in Supporting Information ). The average sapphire\nO–H shift was found to be 3557 ± 10 cm –1 and 3557 ± 12 cm –1 for protected and deprotected\nelastomers in dry contact, respectively using the first moment analysis.\nThis confirms the identical acid–base interaction strength\nof both protected and deprotected elastomers. On the basis of this\nshift of sapphire free O–H (3707 cm –1 , obtained\nexperimentally) and using the Badger–Bauer equation (energy\nof interaction, Δ H = m ×\nΔν + C , where m = 1.09\n× 10 –2 kcal/mol cm, C = 0.03\n± 0.01 kcal/mol for sapphire, and Δν is the shift\nof the O–H peak), we estimate that the interaction corresponds\nto an adhesion energy of 7.0 ± 0.1 kJ/(mole of O–H groups). 57 A sapphire surface typically has around nine\nO–H groups per nm 2 , and hence macroscopically, this\ninteraction can contribute to a maximum interfacial threshold energy\n( G 0 ) of 0.10 J/m 2 for both\nelastomers assuming all the surface O–H groups are participating\nequally in this interaction. 54 , 58 The similarity\nin the shift of the O–H peak for protected\nand deprotected elastomers was unexpected. We were anticipating higher G 0 for the deprotected elastomer than the protected\nelastomer from the Badger–Bauer equation-based calculations\nin accordance with the observations from the adhesion measurements\n( Figure 3 C). To investigate\nmore, we measured the interaction of catechol- d 2 in CHCl 3 - d (0.07 M) adsorbed\non sapphire substrate using SFG. Interestingly, the peak of sapphire\nO–H interacting with catechol was observed at ∼3593\n± 2 cm –1 ( Figure S8 ). The bimodal peak observed in Figure S8 is for catechol- d 2 and CHCl 3 - d interacting with the sapphire O–H groups.\nThe position of CHCl 3 - d peak is at the\nsimilar location with or without adding catechol- d 2 . This indicates that the presence of catechol- d 2 does not alter the interaction strength of\nCHCl 3 - d groups. The shift in the position\nof the O–H peak as result of the interaction of catechol to\nsapphire O–H was intermediate to acetone (∼3610 cm –1 ) and pyridine (∼3575 cm –1 ) interactions. 57 Also, the lower O–H\nshift in the case of catechol than the elastomers suggest that the\nother polar groups present in the elastomer may interact more strongly\nthan the catechol. For the wet contact ( Figure 4 B, bottom panel), there is a notable decrease\nin the intensity\nof sapphire O–H region as compared to hydrocarbon signature.\nTo understand this change in intensity, we need to also compare the\nchanges in the peak in the O–D stretching region ( Figure 4 C). Both protected\nand deprotected elastomers showed two distributions of peaks: one\ncorresponding to O–D stretching in “liquid-like”\nwater at ∼2500 cm –1 and another corresponding\nto O–D stretching next to hydrophobic interface at ∼2650\ncm –1 . 63 The higher the\nnumber of hydrogen bonds per water molecule, the O–D or the\nO–H stretch mode moves to lower wavenumbers. So, the “liquid-like”\nwater peak is ∼2500 cm –1 (∼3400 cm –1 for H 2 O) compared to more strongly hydrogen\nbonded peak for ice, which is ∼2400 cm –1 (∼3200\ncm –1 for H 2 O). If there are fewer numbers\nof hydrogen bonds compared to liquid-water, the peaks are shifted\nto ∼2600 cm –1 (∼3500 cm –1 for H 2 O), whereas the non-hydrogen bonded O–D\npeak appears at ∼2700 cm –1 (∼3700\ncm –1 for H 2 O). The decrease in the intensity\nof sapphire O–H region (∼ 3550 cm –1 ) as compared to hydrocarbon signature along with the distribution\nof O–D stretch next to the hydrophobic interface (∼2650\ncm –1 ) could be due to exchange of the sapphire proton\nto deuterium (O–H → O–D) after exposing the sapphire\nto deuterated water. The exchange can cause decrease in sapphire O–H\nintensity and is expected to show peaks in the 2600–2750 cm –1 region. The second reason could be the presence of\nweakly hydrogen bonded water between the elastomer and sapphire interface. 68 This weakly hydrogen bonded water layer may\nnot decrease adhesion. Zhou et al. also observed similar water structure\n(∼2630–2700 cm –1 ) at the polyurethane-sapphire\ninterface after exposing the sample to low humidity and the presence\nof this water layer did not completely disrupt the polyurethane-sapphire\ninteractions. 69 In both these scenarios,\nthe exchange of surface O–H to O–D or presence of weakly\nhydrogen bonded water (∼2650 cm –1 ) may not\ndisrupt adhesion of the protected or deprotected elastomers when contacted\nwith the sapphire substrate in the presence of water. The important\nobservation here is the presence of the O–D peak near 2500\ncm –1 corresponding to “liquid-like”\nconfined water, which can disrupt underwater adhesion. Interestingly,\nin the hydrocarbon region ( Figure 4 B, bottom panel) an aromatic =C–H\nsignature at ∼3030 cm –1 was observed only\nfor the deprotected elastomer in wet contact. This signature observed\nin the SFG spectra of the deprotected elastomer exclusively in wet\nconditions indicates the presence of catechol at the underwater contact\ninterface. 70 , 71 This peak also overlaps with\nthe features from other unsaturated hydrocarbons in the polymer side\nchains. The lack of such a signature for the dry and underwater contact\nof protected elastomer and dry contact of deprotected elastomer confirms\nthe presence of catechol groups at the underwater contact interface. Now relating these spectroscopic observations to adhesion measurements,\nwe can understand the precise role of catechol in increasing adhesion.\nThe similar acid–base interactions for both elastomers negate\nthe possibility of hydroxyl groups of catechol forming stronger acid–base\nbonds with hydrophilic substrates compared to other polar groups present\nin the polymer. This leaves the following three possible explanations\nfor the increased adhesion of the deprotected elastomer. First, the\ndifferences observed in adhesion strength measurements could originate\nfrom the differences in the deformation during loading. To scrutinize\nsuch possibility, we measured the effective modulus of the contacting\nsurfaces. The load dependent contact deformation of the elastomer-lens\nduring the approach and retraction was captured by observing the changes\nin contact area as a function of force using an optical microscope\nover the set up shown in Figure 3 A. This data along with a JKR model was used to measure\nthe effective modulus ( K ) of contacting surfaces\nin dry condition, 53 which were calculated\nto be ∼1.9 and ∼1.5 MPa for the protected and deprotected\nelastomers coated PDMS sheets, respectively ( Figure S9 ). The similar values of K ( Figure S9 ) and rheological properties ( Figure S3 ) imply that the differences in adhesion\nmeasurements arise from the interfacial adhesion and not because of\ndifferences in the bulk properties. Since the elastomers display adhesion\nhysteresis ( Figure S9 ), the adhesion strength\nreported here may also contain the energy spent in stretching interfacial\nchains and potentially a contribution from the energy dissipated in\nstretching the bulk polymeric chains during the unloading cycle. Therefore,\nthe work of adhesion measured using the JKR geometry during pull-off\nis not equal to but is proportional to the interfacial threshold strength\n( G 0 ) of the protected or deprotected elastomers.\nSince the protected and deprotected elastomers have similar moduli\nand bulk properties, we expect that the ratio of G 0 for these two elastomers would be similar to the corresponding\nratio of their work of adhesion. 54 The second possibility is that the catechol moieties can form multiple\nhydrogen bonds (multimodal acid–base interactions), 72 and breaking multiple bonds simultaneously would\nrequire higher energy as proposed by the single molecule AFM and surface\nforce apparatus measurements of catechol containing molecules. 20 , 31 Our experimental evidence substantiates this model. By comparing\nthe sapphire O–H peak shifts in the adsorption of catechol\n( Figure S8 ) and the elastomer-sapphire\ncontacts ( Figure 4 B),\nthe interaction of catechol hydroxyl groups (3593 ± 2 cm –1 ) was observed to be weaker than the interaction of\nboth protected (3552 ± 6 cm –1 ) and deprotected\n(3557 ± 12 cm –1 ) elastomers. In the SFG measurements,\nthe maximum sapphire O–H peak shift corresponds to the strongest\nmonomodal interaction of functional groups with sapphire. Therefore,\nwe can conclude that the interaction strength of individual hydroxyl\ngroups of catechol is lower than some of the functional groups in\nthe elastomers. However, with two adjacent hydroxyl groups, catechol\ncan form multiple weaker monomodal interactions in a localized area\nto constitute multimodal interactions, and as a result, the elastomer\nwith catechol groups show a higher adhesion strength than the protected\nelastomer. The third possibility is that the presence of catechol\nin the deprotected\nelastomer increases the number of polar groups (∼60 mol % increase).\nThis increases the number of potential hydrogen bonds that can be\nformed by the deprotected elastomer compared to the protected elastomer.\nSince the shift in the sapphire O–H peak of both the elastomers\nis very similar, the level of polar interactions with the substrate\nis identical. Hence, we conclude that the localized multiple hydrogen\nbonding by dihydroxy groups is primarily responsible for the increase\nin the interfacial adhesion in dry conditions. The probability of\nother interactions such as metal-coordination is unlikely, as it has\nbeen shown that catechol interacts exclusively through hydrogen bonding\nwith SiO 2 and Al 2 O 3 surfaces. 31 In the case of wet contact of both the\nelastomers, “liquid-like”\nwater is present ( Figure 4 C, bottom panel, peak ∼2500 cm –1 ).\nSubtle changes in the water structure have been shown to influence\nthe interfacial phenomena such as adhesion and friction. 73 The statistically similar O−D spectral\nsignatures in the wet contact of both the elastomers eliminate the\npossibility of water structure causing the differences in adhesion\nstrength. The presence of “liquid-like” water and the\nsignificant underwater adhesion indicate that the contact is patchy.\nThere are certain regions where the elastomers are in direct contact\nwith the sapphire substrate and other regions where there is “liquid-like”\nwater trapped between the elastomers and the sapphire substrate. 57 , 74 This patchy contact interface explains the drop in underwater adhesion\ncompared to the dry adhesion. Besides, catechol itself does not play\nan important role in removing interfacial water next to hydrophilic\ninterface, which is consistent with the observations by Kirpat et\nal. 75 The observed catechol signature in\nunderwater contact of the deprotected elastomer ( Figure 4 B, bottom panel, peak ∼3030\ncm –1 ) is from the population of catechol that is\ninteracting with water. The nature of the underwater contact\ndepends on the surface energy\nof the materials. 74 For example, two hydrophobic\nsurfaces make true molecular contact after removing interfacial water,\nand the contact between two hydrophilic surfaces retains a thin film\nof water at the interface, 73 , 76 whereas a hydrophobic\nmaterial makes patchy contact with hydrophilic surfaces. 66 In our case, both elastomers are hydrophobic\n(polymer water contact angles >95°, Table S1 ) and are in contact with a hydrophilic surface, and hence\nthe contact area is expected to be patchy. Perhaps a patchy contact\ncould also form as the draining of interstitial water requires more\ntime. If this was the case, increasing the contact time should increase\nthe dry molecular contact and thus lead to higher adhesion. 66 Figure S4B shows\nthe work of adhesion at contact equilibration times of 0.5, 3, and\n15 min. The underwater adhesion values remain lower than those measured\nin dry contact, indicating that the kinetics of drainage is not playing\nan important role in underwater adhesion. The deprotected elastomer\nshowed higher underwater adhesion as compared to the protected elastomer.\nSince the percentage reduction in the adhesion between dry and wet\ncontacts for both protected (∼72%) and deprotected (∼79%)\nelastomers are similar, we infer that the fraction area of patchy\nwater contact is similar for both elastomers. From the analysis\nof adhesion measurements and SFG spectra, we\npropose an overall model for the underwater adhesion for these elastomers\n( Figure 5 ). Regardless\nof the presence of catechol, both elastomers form patchy contact with\nhydrophilic substrates underwater. The wet patches which contain water\n(blue) between the polymer and substrate constrain the adhesion. The\ndry patches (yellow) are true contacts between the polymer and substrate\nwhich provide underwater adhesion. 66 , 74 We anticipate\nthat the patches are smaller than the resolution of the optical probes\n(less than microns). Both the protected and deprotected elastomers\nsucceed in contacting the substrate underwater. However, the interfacial\nstrength of the dry patch of deprotected elastomer is higher due to\nthe localized multimodal interactions of O–H groups in catechol\nwith the sapphire substrate, resulting in higher underwater interfacial\nadhesion than the protected elastomer. We have reported recently that\na hydrophilic adhesive with catechol does not adhere underwater compared\nto a relatively hydrophobic adhesive without catechol groups. 49 Therefore, for effective underwater adhesion,\ninitially the polymer should make interfacial contact with the surface\nand remove bound water, which in this case is achieved by a conformable\nhydrophobic adhesive. Also, polar functional groups with strong interfacial\ninteractions are essential to display significant adhesion underwater.\nIf we use hydrophobic PDMS in contact with hydrophilic substrates,\nwe observe patchy contact. However, the wet adhesion is very weak\nfor PDMS ( Figure S4C ) due to the absence\nof strong interfacial polar interactions present in mussel-inspired\ncatechol polymers. Additionally, the hydrophobic functional groups\nin Mfp 19 and mussel-inspired polymers 77 have also been shown to improve adhesion by\nshielding catechol groups from oxidation reactions. Figure 5 Proposed model for the\nadhesion mechanism in wet condition (pH\n= 6.5) for the deprotected elastomer. The diagram on the left side\nis a representative underwater contact of the deprotected elastomer\nwith hydrophilic substrate. In the dry patch (top right), the elastomer\nis in contact with sapphire and can interact with substrate through\nacid–base interactions. Catechol makes multimodal hydrogen\nbonds with the substrate. In the wet patch (bottom right), water interferes\nthe interaction of the elastomer with the substrate."
} | 10,639 |
34845026 | PMC8670470 | pmc | 7,228 | {
"abstract": "Significance Almost all organisms replicate by growing and then shedding offspring. Some molecules also replicate, but by moving rather than growing: They find and combine building blocks into self-copies. Here we show that clusters of cells, if freed from a developing organism, can similarly find and combine loose cells into clusters that look and move like they do, and that this ability does not have to be specifically evolved or introduced by genetic manipulation. Finally, we show that artificial intelligence can design clusters that replicate better, and perform useful work as they do so. This suggests that future technologies may, with little outside guidance, become more useful as they spread, and that life harbors surprising behaviors just below the surface, waiting to be uncovered.",
"discussion": "Discussion The ability of genetically unmodified cells to be reconfigured into kinematic self-replicators, a behavior previously unobserved in plants or animals, and the fact that this unique replicative strategy arises spontaneously rather than evolving by specific selection, further exemplifies the developmental plasticity available in biological design ( 1 – 8 ). Although kinematic self-replication has not been observed in extant cellular life forms, it may have been essential in the origin of life. The amyloid world hypothesis ( 31 ), for instance, posits that self-assembling peptides were the first molecular entity capable of self-replication, and would thus represent the earliest stage in the evolution of life, predating even the RNA world. Unlike self-replicating RNAs which template themselves during replicative events, amyloid monomers can form seeds which produce a variety of amyloid polymorphs, yielding either larger or smaller “offspring” depending on peptide availability, kinematics, and thermodynamic conditions. This variation is similar to modern-day prions, where self-propagating misfolded proteins are capable of forming aggregates of multiple sizes and polymorphisms ( 32 ). Although reconfigurable organisms are not a model for origin of life research, which strives to describe the first information unit capable of self-replication, they may shed light on its necessary and sufficient initial conditions. Traditional machine self-replication is assumed to require a constructor, a copier, a controller, and a blueprint to describe all three ( 15 ). However, there are no clear morphological or genetic components in the organisms described here that map onto these distinct structures. The concept of control in reconfigurable organisms is further muddied by their lack of nervous systems and genetically modified behavior. This suggests that reconfigurable organisms may in future contribute to understanding how self-amplifying processes can emerge spontaneously, in new ways and in new forms, in abiotic, cellular, or biohybrid machines, and how macroevolution may proceed if based on kinematic rather than growth-based replication. Today, several global challenges are increasing superlinearly in spatial extent ( 33 ), intensity ( 34 ), and frequency ( 35 ), demanding technological solutions with corresponding rates of spread, adaptability, and efficacy. Kinematic self-replication may provide a means to deploy a small amount of biotechnology that rapidly grows in utility, but which is designed to be maximally controllable ( 36 ) via AI-designed replicators. Even if the behaviors exhibited by reconfigurable organisms are currently rudimentary, such as those shown in past ( 10 ) and this current work, AI design methods have been shown to be capable of exploiting this flexibility to exaggerate these behaviors and, in future, possibly guide them toward more useful forms."
} | 930 |
33042172 | PMC7527422 | pmc | 7,229 | {
"abstract": "Soil microorganisms influence biotic and abiotic stress tolerance of crops. Most interactions between plant symbiotic and non-symbiotic soil microorganisms and plants occur in the rhizosphere and are sustained through plant exudation/rhizodeposition. Bioaugmentation, i.e. , the introduction or amplification of certain plant beneficial microbes ( e.g. , entomopathogenic fungi) into the rhizosphere, could contribute to controlling insect crop pests and replacing chemical, environmentally unfriendly insecticides. Wireworms, the soil-burrowing larval stages of click beetles (Coleoptera: Elateridae), are major pests of crops including maize, wheat and potatoes, worldwide. Alternative strategies for controlling wireworms are needed because several chemical pesticides used successfully in the past are being phased out because of their ecotoxicity. Therefore, virulence to Agriotes \n lineatus L. wireworms and plant beneficial traits of entomopathogenic fungi were investigated in a series of laboratory experiments. Tested taxa included environmentally retrieved Metarhizium brunneum Petch. (two strains), M. robertsii Bisch., Rehner & Humber (Hypocreales: Clavicipitaceae), and Beauveria brongniartii (Sacc.) Petch. and commercially formulated B. bassiana (Bals.-Criv.) Vuill. (Cordycipitaceae) and Bacillus \n thuringiensis Berliner 1915 var. kurstaki . In-house reared larvae were dipped in conidial suspension, and maize and wheat seeds were coated with fungal conidia. Metarhizium brunneum strains 1154 and 1868 significantly increased wireworm mortality. Fungi were significantly more often re-isolated from maize than wheat rhizoplanes in laboratory assays. The strains tested were rarely isolated as endophytes. Metarhizium brunneum strain 1154 stimulated wheat growth, while M. robertsii 1880 stimulated maize growth, whereas M. \n brunneum 1868 and others did not affect root or shoot length or plant biomass significantly in laboratory settings. Metarhizium \n brunneum strain 1868, re-isolated most often from maize rhizoplane, caused the highest wireworm mortality. It was further evaluated whether M. \n brunneum 1868 can protect maize varieties FeroXXY, LG 34.90 and Chapalu from wireworm damage and promote plant growth at field conditions. Plants of all three varieties stemming from seeds treated with conidia of M. brunneum 1868 showed significantly less wireworm damage 3 to 4 weeks after sowing (5- to 6-leaf stage) resulting in a significantly higher initial maize stand. However, only in the variety LG 34.90 a significant increase of the maize stand was observed at harvest time.",
"conclusion": "Conclusions The investigated entomopathogenic fungi exhibited multifaceted functions ( Vega et al., 2009 ), i.e. pathogenicity to wireworms, rhizosphere competence and some growth promoting effects of maize and wheat plants in lab settings. However, observed effects were either depending on the EPF strain, plant species or variety. Metarhizium brunneum strain 1868 reduced Agriotes herbivory and increased initial plant stands of all three maize varieties tested in field settings. Interestingly, pre-harvest maize stand density and yield were increased only in one of the three varieties. Similar differential interactions between EPF and different plant varieties were reported by Canassa et al. (2020) for two root-inoculated strawberry varieties. This highlights the problems of generalizations and warrants further studies on the mechanisms of plant × fungus interactions. Insect repellence ( Meyling and Pell, 2006 ; Kabaluk and Ericsson, 2007a ) mediated through fungal rhizosphere competence may be the underlying mechanism for the measured increase of plant biomass. The ability of tested fungi to improve stand and robustness of young maize plants could contribute to wireworm stress resilience as this pest limits maize growth especially after crop emergence and less towards harvest time ( Taupin, 2007 ).",
"introduction": "Introduction Wireworms (Coleoptera: Elateridae) damage potato and other crops including wheat and maize. They start feeding on seed potatoes shortly after planting without causing plant losses initially. However they reduce the market quality of the produce as they penetrate into near harvest potatoes ( Benjamin et al., 2018 ). Once potatoes are damaged, also secondary microbial infections occur and the yield of entire potato crops can become unmarketable in high pest pressure areas or organic production settings ( Brandl et al., 2017 ). Due to their hidden life cycle belowground, wireworms can hardly be controlled, especially in organic farming, where persistent, non-specific soil insecticides cannot be used ( Schepl and Paffrath, 2007 ; Brandl et al., 2017 ; Benjamin et al., 2018 ). In maize and wheat, wireworms target germinating seeds and young sprouts, what results in typical herbivory symptoms such as leaf drilling holes and dead central leaves. However, in case of severe infestations, plant stand and yield can be significantly decreased ( Reddy et al., 2014 ; Furlan et al., 2017 ). It has been emphasized that the abandonment of ecotoxicologically problematic soil insecticides may increase wireworm-related problems ( Parker and Howard, 2001 ; van Herk and Vernon, 2013 ). Organophosphates, organochlorines, and carbamates effectively controlled wireworms in the second half of the 20 th century ( Merrill, 1952 ). However, due to their ecotoxicity ( Costa, 2015 ), biomagnification in non-target organisms ( Mitra et al., 2011 ), and yearlong availability in soils ( Wilkinson et al., 1976 ), these pesticides are no-longer used in agriculture. Newer types of chemical insecticides used in the past two decades included pyrethroids, phenyl pyrazoles, and neonicotinoids ( Jeschke et al., 2011 ), which in some cases function via pest repellency or morbidity, rather than mortality ( van Herk et al., 2008 ; Vernon et al., 2009 ). However some of the neonicotinoids and phenyl pyrazoles are already prohibited due to negative effect on bees and other pollinators ( Zhang and Nieh, 2015 ), aquatic invertebrates and fish ( Werner and Moran, 2008 ) or beneficial spiders and mites ( Douglas and Tooker, 2016 ). Accordingly, several nonchemical methods for wireworm control were proposed including crop rotation ( Willis et al., 2011 ), crop residue removal, biofumigation ( Furlan et al., 2010 ), weeding ( Schepl and Paffrath, 2007 ), trap and cover crop use ( Rogge et al., 2017 ), mechanical soil disturbance and biological control ( Reddy et al., 2014 ; la Forgia and Verheggen, 2019 ). Also entomopathogenic fungi (EPF) can significantly reduce insect pest pressures. Typically, they penetrate the insect cuticle, paralyse and destructively colonize insect bodies ( Yousef et al., 2018 ). Several commonly occurring species of Metarhizium and Beauveria are known to be effective against wireworms ( Kabaluk and Ericsson, 2007a ; Kabaluk and Ericsson, 2007b ; Kabaluk et al., 2013 ; Razinger et al., 2013 ; Reddy et al., 2014 ; Brandl et al., 2017 ; Rogge et al., 2017 ; Benjamin et al., 2018 ; Razinger et al., 2018b ). They also infest wireworms as naturally occurring soil fungi ( Kabaluk et al., 2005 ). In addition to causing pathogenicity in pest insects, EPF also have other beneficial functions as they enhance plant growth and mineral nutrition and exclude phytopathogens form rhizosphere niches ( Herbst et al., 2017 ; Rivas-Franco et al., 2019 ; Ahmad et al., 2020 ). Various studies have stressed that plants may actively shape their root microbial communities through rhizosphere depositions ( Dennis et al., 2010 ; Canarini et al., 2019 ) and some have speculated that rhizosphere colonizing entomopathogenic fungi could protect plants from plant insect pests in tritrophic interactions ( Vega et al., 2009 ; Bruck, 2010 ; Steinwender et al., 2015 ). Therefore, we hypothesized that utilizing rhizosphere competent EPF could contribute to potentially long-lasting pest management solutions, and decrease the amount of required biopesticide product. Accordingly, we are trying to identify microbial agents with plant beneficial metabolic or ecological traits that can be bioaugmented in the rhizosphere( Compant et al., 2010 ). In previous studies we screened the virulence of several EPF species ( Razinger et al., 2013 ; Razinger et al., 2018b ). The aim of this study was to apply EPF onto maize kernels as a one-step prophylactic strategy to protect maize plantlets during germination and sprouting against wireworm herbivory. In addition, the plant × microbe interactions were investigated in a series of laboratory experiments.",
"discussion": "Discussion Of the several EPF strains tested in the laboratory, M. brunneum 1868 showed the highest virulence to Agriotes lineatus wireworms and was most often re-isolated from washed pieces of maize roots that emerged from seeds coated with conidia of that strain. Frequency of retrieved re-isolation events suggests that M. brunneum 1868 is rhizosphere competent. Based on these findings we evaluated its potential to reduce damages of wireworms in field settings. Maize plants stemming from kernels inoculated with M. brunneum 1868 conidia showed significantly less wireworm damage at emergence in the field experiment resulting in significantly higher undamaged maize stand in springtime. While the early plant belowground herbivory avoiding effect could be measured in all three varieties, increased fresh ear yield and fresh aboveground biomass was observed only in variety LG 34.90 at harvest time. While chemical insecticides have an immediate effect on wireworm fitness, no such immediate effect can be expected through inoculating seeds or roots with entomopathogenic fungi ( Razinger et al., 2013 ; Razinger et al., 2018b ). Even after inoculating insect larvae directly with dense conidial suspensions, long incubation is required for observing an effect on insect larvae. In the laboratory experiments, wireworm mortality reached less than 60% after 8 weeks ( \n Figure 1 \n ), whereas wireworms damage corn seedlings already within three to four weeks after sowing. It is therefore possible that the reduction of Agriotes herbivory by M. brunneum is due to larval repellence or other mechanisms. Also Kabaluk and Ericsson (2007a) reported that wireworms were not killed but repelled by Metarhizium anisopliae (Metchnikoff) Sorokin contaminated soil and that repellence increased with the conidial concentration in soil in laboratory experiments. Based on these results they postulated that the plant stand density in field settings increased possibly due to larval repulsion ( Kabaluk and Ericsson, 2007b ). Similar maize stand density increase was also observed when an encapsulated M. brunneum formulation, registered for wireworm management in potatoes ( Brandl et al., 2017 ), was tested against wireworms in maize (prof. S. Vidal, personal communication). Metabolite production by endophytes of other species of the Clavicipitaceae or endophyte mediated production of volatile organic compounds has frequently been discussed as a mechanism resulting in herbivory repellence (reviewed in Johnson et al., 2016 ). Similarly, predatory bugs such as Anthocoris nemorum L. sense the presence of Beauveria bassiana after leaf inoculation with conidia of that species ( Meyling and Pell, 2006 ). Meyling and Pell (2006) also observed that A. nemorum avoided contact with the thus inoculated leaves. It is thus possible that the high conidial inoculum mediated presence of M. brunneum can protect crop plants from wireworms non-parasitically. \n Metarhizium brunneum did not stimulate early maize growth in the here described laboratory settings, which was also reported by Kabaluk and Ericsson (2007b) for M. \n anisopliae . Specifically, no effect of fungal treatment was observed on the speed of seed germination in maize and wheat. The only significant effect on the speed of seed germination was attributed to CMC-treatment used in the seed coating procedure, revealing a need for additional experiments and especially improved formulation for testing possible effects of Metarhizium on plant growth. By using a simple inoculation technique of immersing seeds in conidial suspension for 2 h, Ahmad et al. (2020) reported a M. robertsii mediated increase of maize height and aboveground biomass compared to control plants in laboratory experiments. Ahmad et al. (2020) also observed a much higher proportion of endophytically colonized maize leaves as we did, and they calculated a positive correlation between plant height and aboveground biomass and the proportion of endophytic root and leaf colonization by M. robertsii . Metarhizium brunneum strain 1868 well developed on maize root surfaces ( \n Figure 4 \n ) and clearly better than on cauliflower ( Razinger et al., 2014 ) or broccoli ( Herbst et al., 2017 ) roots. This tight association with maize roots is in sync with the report by Hu and Bidochka (2020) , who suggested that Metarhizium has a preference for monocots such as barley and corn. As a rhizosphere colonizer, Metarhizium spp. might also protect corn from other detrimental factors like soil pathogens ( Kabaluk and Ericsson, 2007b ; Vega et al., 2009 ). Furthermore, a higher frequency of endophytic colonization was detected in maize roots (18.8%) compared to leaves (8.3%). This could be the result of fungal preference towards different tissues within the plant host, i.e. plant root preference by Metarhizium species as postulated by Behie et al. (2015) . Laboratory results are often inconsistent with field trials ( Kabaluk et al., 2005 ; Kölliker et al., 2011 ; Sufyan et al., 2017 ). To achieve the highest EPF mediated control effectiveness, many factors must be considered, such as landscape properties, soil characteristics, crop type, etc. In addition, different Agriotes species can be differently susceptible to a certain entomopathogenic taxon or individual strain or formulation product ( Kölliker et al., 2011 ). The challenge is thus to find a strain or a mixture of different strains that performs equally well in different environmental conditions and against different pests. One might consider using multiple EPF strains of the same species to cover a wider range of ecological conditions, but such a strategy would be very difficult to put into practice due to registration constraints ( Gadhave et al., 2016 ; Humber, 2016 )."
} | 3,626 |
30949148 | PMC6437101 | pmc | 7,233 | {
"abstract": "Cyanobacteria are supposed to be promising photosynthetic microbial platforms that recycle carbon dioxide driven into biomass and bioproducts by solar energy. Glycogen synthesis serves as an essential natural carbon sink mechanism, storing a large portion of energy and organic carbon source of photosynthesis. Engineering glycogen metabolism to harness and rewire carbon flow is an important strategy to optimize efficacy of cyanobacteria platforms. ADP-glucose pyrophosphorylase (GlgC) catalyzes the rate-limiting step for glycogen synthesis. However, knockout of glgC fails to promote cell growth or photosynthetic production in cyanobacteria, on the contrary, glgC deficiency impairs cellular fitness and robustness. In this work, we adopted a theophylline-responsive riboswitch to engineer and control glgC expression in Synechococcus elongatus PCC7942 and achieved flexible regulation of intracellular GlgC abundance and glycogen storage. With this approach, glycogen synthesis and glycogen contents in PCC7942 cells could be regulated in a range from about 40 to 300% of wild type levels. In addition, the results supported a positive role of glycogen metabolism in cyanobacteria cellular robustness. When glycogen storage was reduced, cellular physiology and growth under standard conditions was not impaired, while cellular tolerance toward environmental stresses was weakened. While when glycogen synthesis was enhanced, cells of PCC7942 displayed optimized cellular robustness. Our findings emphasize the significance of glycogen metabolism for cyanobacterial physiology and the importance of flexible approaches for engineering and understanding cellular physiology and metabolism.",
"conclusion": "Conclusion Glycogen synthesis is an essential natural carbon sink mechanism found in cyanobacteria, that stores carbon and energy derived from photosynthesis. Engineering the glycogen metabolism of cyanobacteria is required for rewiring the intracellular carbon flow and to optimize the efficacy of the photosynthetic platform. It has previously been reported that the glgC -knock-out strategy impaired cellular fitness, therefore, in this work we adopted a theophylline-responsive riboswitch system to control glgC expression in PCC7942. Through the regulation of theophylline concentrations, GlgC abundance and glycogen storage could be reversibly regulated in a range from 40 to 300% of the wild type level. With this bidirectional regulation approach, a more flexible understanding of the glycogen metabolism effects on cellular fitness and robustness in cyanobacteria could be obtained. When glycogen storage was reduced, the growth of PCC7942 under normal conditions was maintained, while tolerance to NaCl and high pH stress was inhibited. When glycogen storage was elevated, improved stress-tolerances were endowed in the engineered strain. Our results further highlight the significance of flexible and dynamic regulation tools for engineering and understanding cellular physiological and metabolic activities.",
"introduction": "Introduction Cyanobacteria are photoautotrophic prokaryotes, that are widespread in diverse ecosystems, including the ocean, fresh water, and terrestrial environments ( Waterbury et al., 1979 ). They evolved oxygenic photosynthesis, an efficient system converting solar energy and CO 2 into organic compounds ( Hohmann-Marriott and Blankenship, 2011 ). Currently, cyanobacteria produces 10–20% of the organic carbon on Earth, playing an essential role in global carbon and nitrogen cycles ( Flombaum et al., 2013 ; Rousseaux and Gregg, 2014 ). A large portion of the photosynthetically assimilated carbon source, which excesses the normal requirements of cell growth and metabolism in cyanobacteria, is stored in the form of glycogen, a high molecular branched α-polyglucan ( Ball and Morell, 2003 ; Nakamura et al., 2005 ). Its pool can account for up to 50% of the total cellular biomass under specific environmental conditions ( Aikawa et al., 2014 ; Song et al., 2016 ). During the night or in the absence of a carbon source, carbon and energy stored in the glycogen is mobilized and used for the central metabolism ( Stal and Moezelaar, 1997 ; Guerra et al., 2013 ). In addition, there are hints that glycogen metabolism protects cyanobacteria against unfavorable environmental conditions ( Suzuki et al., 2010 ; Grundel et al., 2012 ; Hickman et al., 2013 ). During the last decade, cyanobacteria were increasingly recognized as efficient photosynthetic platforms that can used to recycle CO 2 into biomass and bioproducts, using solar energy. The use of cyanobacteria for biotechnological application is also supported by their simple structure, rapid growth, and existing tools for genetic manipulation ( Angermayr et al., 2009 ; Lu, 2010 ). To improve the efficacy of photosynthetic platforms, a promising strategy would be engineering glycogen metabolism in cyanobacteria for better rewiring and to harness the organic carbon output from the Calven-Benson-Bassham cycle ( Carrieri et al., 2012 ). As shown in Figure 1A , cyanobacterial glycogen synthesis starts from the precursor glucose 1-phosphate (G-1-P), which is stepwisely converted into glycogen by the action of ADP-glucose pyrophosphorylase (GlgC, also termed glucose-1-phosphate adenylyltransferase) catalyzing ADP-glucose (ADP-G) formation, glycogen synthase (GlgA) incorporating glucose monomers into the growing 1-4 α-linked glucose polymer, and branching enzyme introducing 1-6 branches connecting the linear polyglucose chains. The activity of GlgC represents the rate-limiting step and controls the glycogen accumulation process ( Ball and Morell, 2003 ; Grundel et al., 2012 ). Thus, many attempts have been published to knock-out the glgC gene in cyanobacteria, to eliminate glycogen accumulation in diverse cyanobacterial strains. However, impaired glycogen accumulation severely affected cell physiology, including reduced photosynthesis, growth, respiration, and cellular robustness in facing environmental stresses ( Miao et al., 2003 ; Suzuki et al., 2010 ; Carrieri et al., 2012 ; Grundel et al., 2012 ; Guerra et al., 2013 ; Hickman et al., 2013 ). In many cases, deficient glycogen synthesis decreased rather than increased the productivity of heterologous pathways in engineered cyanobacterial strains ( Davies et al., 2014 ; Jacobsen and Frigaard, 2014 ; Li et al., 2014 ; van der Woude et al., 2014 ; Work et al., 2015 ). In summary, when the glycogen synthesis pathway was rigidly and completely blocked (through deletion of glgC ), carbon flow can neither be effectively rewired toward cell growth nor to desired non-natural metabolites, indicating that more efforts and approaches are still required to understand and harness cyanobacterial glycogen metabolism. Figure 1 Development of a theophylline-responsive glgC -expression system in Synechococcus elongatus PCC7942. (A) Schematic representation of glycogen metabolism in cyanobacteria. GlgP, glycogen phosphorylase; GlgC, glucose-1-phosphate adenylyltransferase or ADP-glucose pyrophosphorylase; GlgA, glycogen synthase; G-1-P, glucose-1-phosphate; G-6-P, glucose-6-phosphate; ADP-G, ADP-glucose; PPP, pentose phosphate pathway; EMP, Embden-Meyerhof-Parnas pathway, glycolytic pathway; CBB cycle, Calvin-Benson-Bassham cycle; Pi, inorganic phosphate; PiPi, pyrophosphate. (B) Construction strategy of a theophylline-responsive riboswitch control system on glgC expression in PCC7942. KmR , kanamycin resistance gene; PglgC , native promoter sequence of glgC. (C) Genotype identification of the PCC7942-XC1 and PCC7942-WT by PCR. In this work, we aimed to engineer and decipher the physiological function of cyanobacteria glycogen metabolism in a flexible and regulatable mode. To this end, a theophylline-responsive riboswitch was adopted to control the expression of glgC in Synechococcus elongatus PCC7942 (hereafter referred to as PCC7942), permitting controllable down- and up-regulated glycogen synthesis and storage in the same system. Based on this flexible approach, the influence of glycogen metabolism on cyanobacteria cellular physiology was explored. The data we obtained in this work supported the positive role of glycogen synthesis and contents on cyanobacteria cellular fitness and robustness toward environmental stresses.",
"discussion": "Results and Discussion Construction of a Theophylline-Responsive glgC Expression System in PCC7942 GlgC catalyzes the rate-limiting step of glycogen synthesis ( Figure 1A ) and is generally supposed to maintain control over all glycogen metabolism activities in cyanobacteria ( Ball and Morell, 2003 ; Grundel et al., 2012 ). Thus, we selected glgC ( Synpcc7942_0603 ) in PCC7942 as the target to engineer and regulate. To develop artificial control over the glgC expression, we adopted a synthetic theophylline-responsive riboswitch system ( ENYC4 ) ( Nakahira et al., 2013 ) to achieve strict regulation of intracellular protein abundance. As shown in Figure 1B , a cassette containing kanamycin-resistance gene ( KmR ), a Ptrc promoter and a theophylline-responsive riboswitch region ( Ptrc-ENYC4 ) was inserted into the chromosome of PCC7942 between the glgC CDS region and the native promoter sequence. The integration was confirmed by PCR ( Figure 1C ) and DNA sequencing. The PCC7942 mutant carrying the KmR - Ptrc-ENYC4-glgC cassette on the chromosome was termed as PCC7942-XC1 (XC1), while the wild type strain of PCC7942 was termed as PCC7942-WT (WT) as a control. Theophylline-Dose Regulated glgC Expression and Glycogen Storage in PCC7942-XC1 To evaluate the effects of theophylline-responsive riboswitch on controlling glgC expression, three different concentrations (0, 110, 1100 μM) of theophylline were supplemented into BG11 culture medium of PCC7942-WT and PCC7942-XC1, and the GlgC abundances in the two strains were determined by western-blot. As shown in Figure 2A (XC1 region), GlgC concentrations in PCC7942-XC1 were correlated with the theophylline concentrations. GlgC abundance in XC1 strain cells without theophylline addition (XC1-T0) was significantly reduced, compared to that of the PCC7942-WT (WT-T0). When theophylline concentrations were supplemented at 1100 μM, the GlgC concentrations (XC1-T1100) were much higher than that of the control (WT-T0, T110, T1100), indicating that expression of glgC was successfully controlled by the regulation of theophylline-dose. Figure 2 Theophylline-dose regulated GlgC abundances and glycogen storage in PCC7942-XC1. (A) Western blot assay for GlgC concentrations in WT strain and XC1 strain under serial theophylline concentrations (T0, no theophylline addition; T110, 110 μM theophylline; T1100, 1100 μM theophylline). In PCC7942, glgC gene encodes a protein with MW of about 48 kDa, consistent with the band locations marked in the PVDF membrane. (B) Glycogen contents in WT strain and XC1 strain under serial theophylline concentrations. The data shown are averages of at least three independent biological repeats, and standard deviation bars are also shown. Glycogen storage in XC1 strain could also be successfully regulated by the theophylline-dose. As shown in Figure 2B and Supplementary Figure S1 , when no theophylline was supplemented, glycogen contents in the cellular biomass of XC1 was approximately 40% of that in the wild type control (6.4 mg/L/OD730 for XC1-T0 and 15.6 mg/L/OD730 for WT-T0), which is consistent with the lowered GlgC abundances in XC1-T0. When theophylline concentrations were increased to 110 μM, the glycogen contents in XC1 (19.8 mg/L/OD730) was about 25% higher than that of the wild type control (15.8 mg/L/OD730 for WT-T110). When the theophylline concentrations were further increased to 1100 μM, glycogen contents in XC1 (39.4 mg/L/OD730 for XC1-T1100) were increased threefold higher than the wild type control in the same conditions (13.9 mg/L/OD730 for WT-T1100). In summary, glycogen synthesis and storage in PCC7942-WT cells were not significantly altered in response to theophylline addition, while the glycogen contents of XC1 cells could be regulated in a range from about 40 to 300% of the wild type level by theophylline-dose regulation. It has previously been reported that glycogen storage in cyanobacteria could be down-regulated by some recently existing interfering approaches. In Synechocystis sp. PCC6803, Yao et al. (2016) adopted a CRISPR interference (CRISPRi) tool to knock down the expression of glgC . Through an inducible expression of the dCAS9 element, the transcription of glgC was decreased by 90%, while the glycogen contents were reduced by 70% compared to the same in the wild type control under nitrogen deprivation conditions. Sun and colleagues developed small RNA regulatory tools based on paired termini RNAs and an exogenous MicC scaffold combined with the Hfq chaperone (Hfq-MicC). The sRNA regulatory system repressed expression of glgC by over 90%, and resulted in a 75% decreased glycogen content ( Sun et al., 2018 ). In PCC7942, Huang and colleagues also adopted the CRISPRi approach for down-regulated expression of glgC , and successfully reduced glycogen contents by 75%. In general, intracellular glycogen contents in cyanobacteria could be decreased by 70 to 90% with the CRISPRi strategy and a small RNA regulatory tool, permitting controllable repression of glgC and reduction of glycogen contents. In comparison, the riboswitch approach adopted in this work enabled a bidirectional regulation of glycogen synthesis and storage, meaning both up-regulation and down-regulation could be achieved in a single system. Through dose-regulation of the signal molecules, physiological characteristics of the same cyanobacteria strain containing reduced or over-accumulated glycogen could be directly evaluated and compared, which would benefit the comprehensive understanding of glycogen metabolism effects on cyanobacteria cellular fitness and robustness. Reduced Glycogen Synthesis Did Not Impair Cell Growth of PCC7942 Previously, it has been generally reported that abolished glycogen synthesis and storage would lead to significant retarded cell growths in diverse cyanobacteria strains, including Synechocystis sp. PCC6803 ( Miao et al., 2003 ), Synechococcus elongatus PCC7942 ( Suzuki et al., 2010 ), and Synechococcus sp. PCC7002 ( Guerra et al., 2013 ). Thus, cell growths of PCC7942-XC1 were first characterized under gradually increased theophylline concentrations. As shown in Figure 3A , when no theophylline was added to induce the synthesis of glycogen, cell growths of PCC7942-XC1 were indistinguishable from that of PCC7942-WT. After 10 days of cultivation, no significant difference was observed on cell densities of the two strains with different glycogen contents ( Figure 2B ). When the theophylline concentration was increased to 110 μM, cell growths of the two strains were still maintained on a similar level ( Figure 3B ). While, as shown in Figure 3C , when the theophylline significantly positively-relates to 1100 μM, slight growth advantages of XC1 cell growth were observed over that of PCC7942-WT, indicating that increased glycogen contents might compensate the inhibitory effects on PCC7942 growth from high concentrations of theophylline. Figure 3 Growth assays of PCC7942-WT and PCC7942-XC1 under induction of gradually increasing theophylline concentrations. (A) No theophylline was added; (B) 110 μM theophylline was added; (C) 1100 μM theophylline was added. Optical densities under 730 nm (OD730) were used to calculate cell concentrations. The data utilized are averages of measurements from at least three independent biological repeats, and the standard deviations bars are shown. To further explore the influence of regulated glycogen synthesis and storage on cyanobacterial cellular physiology, we assayed oxygen evolution rates in PCC7942-XC1 and the wild type control. As shown in Figure 4A , when light intensities were set at a low level (46 μmol photons/m 2 /s), which was utilized in cultivation, the oxygen evolution rate in XC1-T0 was similar to that of the wild type control (WT-T0). And that phenomenon was consistent with similar growth profiles between the XC1 strain and the wild type control when no theophylline was supplemented. Figure 4 Oxygen evolution of PCC7942-WT and PCC7942-XC1 cells under different illumination. L-46 represented light intensities of 46 μmol/m 2 /s; L-2878 represented light intensities of 2878 μmol/m 2 /s. The data shown are averages of at least three independent measurements, and standard deviation bars are also shown. (A) PCC7942-WT and PCC7942-XC1 under 0 μM theophylline; (B) PCC7942-XC1 under induction of 0 μM, 110 μM, 1100 μM theophylline, respectively. Additionally, when high light intensities (2878 μmol photons/m 2 /s) were provided, oxygen evolution activities in the wild type strain of PCC7942 were significantly increased, while the increase in XC1 was much more limited, indicating that the capacities to tolerate and utilize strong illuminations of PCC7942 cells were impaired by the reduced glycogen synthesis and storage. The results were consistent with a previous observation in a glgC -null mutant of PCC7942 ( Suzuki et al., 2010 ). A hypothesis was proposed that organic carbon output rates exceeded the consuming rates of cellular growth in cyanobacteria, while excessed carbon and energy would prevent higher carbon fixation rates in photosynthesis ( Oliver and Atsumi, 2015 ). Glycogen storage serves as a natural carbon sink mechanism and can remove restrictions on photosynthesis. In addition, strong illuminations would also cause physiological impairments on cyanobacteria. Glycogen metabolism also plays an important role in maintaining cellular robustness when facing high light stress followed by oxygenic stress ( Suzuki et al., 2010 ; Grundel et al., 2012 ). Thus, the weakened cellular robustness caused by reduced glycogen metabolism in XC1 (T0) might also contribute to the inhibited photosynthesis activities compared to that of the PCC7942-WT cells. The inhibited oxygen evolution activities under high light intensities could be relieved by the addition of theophylline and increased glycogen synthesis in XC1 ( Figure 4B ), confirming an important role of glycogen synthesis and storage in tolerating strong illumination stress and buffering carbon and energy excess from photosynthesis. As has been generally discovered and accepted in previous research, completely abolishing glycogen synthesis and storage, would inhibit cyanobacterial cellular growths and photosynthesis activities, even when cultivated under normal conditions with continuous illuminations ( Miao et al., 2003 ; Suzuki et al., 2010 ; Grundel et al., 2012 ). However, our results provides evidence showing that when glycogen synthesis and storage is reduced rather than completely removed, the defects on cell growth and cellular physiology under normal conditions can be avoided. In addition, the important role of glycogen synthesis and storage for buffering the excess of light energy input was also confirmed in this work. Glycogen Synthesis Promoted Cellular Tolerance to Environmental Stresses Glycogen metabolism has been proposed to play an important role for cyanobacteria, to resist multiple environmental stresses ( Preiss, 1984 ; Suzuki et al., 2010 ; Grundel et al., 2012 ; Hickman et al., 2013 ). Thus, we also evaluated the stress tolerances dynamics of the XC1 strains with theophylline-concentration regulated glycogen synthesis and storage. Previously it has been reported that knockout of the glgC gene in PCC6803 ( Miao et al., 2003 ), PCC7942 ( Suzuki et al., 2010 ), and PCC7002 ( Guerra et al., 2013 ) resulted in decreased cellular tolerance to NaCl stress. In this work, we observed similar results. As shown in Figure 5A , when 0.3 M NaCl was added to stressed PCC7942 cells, the XC1 strain without theophylline induction (XC1-T0) showed significant reduced cell growth compared with the WT-T0 control. The cell growth of XC1-T0 under 0.3 M NaCl stress continued for less than 2 days and entered into a stationary phase (OD730 ∼ 0.6), while that of WT-T0 continued for over 5 days, OD730 reaching up to 0.9. When theophylline was supplemented to induce glycogen synthesis and accumulations, growth capacities of the XC1 strain facing NaCl stress were gradually recovered and enhanced to levels even surpassing those of the WT controls ( Figure 5B,C ). While as for PCC7942-WT, cell growth was slightly retarded by increasing concentrations and toxicities of theophylline ( Figure 5B,C ). In our previous research, we reported that, as the main compatible solute of PCC7942 to resist osmotic stress, synthesis of sucrose is closely linked to glycogen accumulations ( Qiao et al., 2018 ), which might be a reason for the advantages of the XC1 strain with increased glycogen synthesis. Figure 5 Growths of PCC7942-WT and PCC7942-XC1 facing NaCl stress under theophylline concentrations of (A) 0 μM, (B) 110 μM, and (C) 1100 μM. 0.3 M NaCl was supplemented into the culture medium in Day 2 as the arrow pointed. The OD730 data utilized are averages of measurements from at least three independent biological repeats, and standard deviation bars are also shown. Asterisksg49 indicate significant differences between the data of WT and XC1 (Student’s t -test, P < 0.05). We also explored the effects of glycogen synthesis and contents on cellular tolerance to high pH stress. Similar to the growth pattern in high salt conditions, when the pH value of the culture medium was increased to 11, growth of XC1 (without theophylline induction, XC1-T0) was significantly decreased ( Figure 6A ). Addition of 110 μM theophylline, under which the glycogen content of XC1 cells was comparable with that of the wild type control, relieved the growth defects in XC1 ( Figure 6B ), indicating that glycogen synthesis and storage is an important approach of cyanobacteria to resist alkaline stress. Figure 6 Growth of PCC7942-WT and PCC7942-XC1 facing high pH stress under theophylline concentrations of (A) 0 μM, (B) 110 μM. To adjust pH values of the culture medium to 11.0, NaOH would be added in day 2 as the arrow pointed. The OD730 data utilized are averages of measurements from at least three independent biological repeats, and standard deviation bars are also shown. Asterisks’ indicate significant differences between the data of WT and XC1 (Student’s t -test, P < 0.05). Stress-responsive cellular activities required additional supply of energy and materials to maintain homeostasis. Glycogen storage served as an essential substrate for respiration, thus fulfilling emergent requirements ( Guerra et al., 2013 ). It has been generally accepted that deficiency of glycogen synthesis would result in impaired tolerance to abiotic stresses ( Preiss, 1984 ; Suzuki et al., 2010 ; Grundel et al., 2012 ; Hickman et al., 2013 ). In this work, our results further provided solid data supporting the positive role of glycogen metabolism in cellular robustness."
} | 5,794 |
28389638 | PMC5429687 | pmc | 7,234 | {
"abstract": "Genome-scale metabolic models (GSMMs) constitute a platform that combines genome sequences and detailed biochemical information to quantify microbial physiology at the system level. To improve the unity, integrity, correctness, and format of data in published GSMMs, a consensus IMGMD database was built in the LAMP (Linux + Apache + MySQL + PHP) system by integrating and standardizing 328 GSMMs constructed for 139 microorganisms. The IMGMD database can help microbial researchers download manually curated GSMMs, rapidly reconstruct standard GSMMs, design pathways, and identify metabolic targets for strategies on strain improvement. Moreover, the IMGMD database facilitates the integration of wet-lab and in silico data to gain an additional insight into microbial physiology. The IMGMD database is freely available, without any registration requirements, at http://imgmd.jiangnan.edu.cn/database.",
"conclusion": "Conclusion The IMGMD database (http://imgmd.jiangnan.edu.cn/database) provides a platform that integrates the names of metabolites and metabolic reactions from common biochemical databases and existing model repositories. This database includes 328 models for 139 microorganisms and provides 265 standardised models for downloading. Based on a homologous sequence alignment method, models can be reconstructed automatically in the IMGMD database, which can accelerate the process of model construction. Furthermore, IMGMD provides a pathway mining tool for pathway design and a mutation library for strain optimisation. Compared with other GSMM databases, the IMGMD database is specific for microorganisms. It is user-friendly and feature-rich; accordingly, the scientific community can easily use and extend the knowledge base. Thus, IMGMD will be a useful database for the design phase of systems metabolic engineering. Future developments include integration of the COBRA Toolbox, which will allow users to directly simulate gene deletion or over-expression, on the IMGMD platform. Besides, the IMGMD database is maintained by our lab and will be updated annually, to keep pace with the advances of GSMMs.",
"introduction": "Introduction Genome-scale metabolic models (GSMMs) are a kind of a mathematical model that integrates multiple types of omics data, such as genomics, transcriptomics, proteomics, and metabolomics. GSMMs can clarify the relations among genes, proteins, and reactions. Models can be used to describe all biochemical reactions, metabolites, and genes involved in the metabolism of a specific organism. They have been used to decipher metabolic, regulatory, and signaling networks at the whole-organism level 1 – 3 . Since the first GSMM of Haemophilus influenzae Rd was constructed in 1999 4 , more than 300 GSMMs for over 100 organisms have been built 5 , 6 . GSMMs have been developed for over 15 years, and four steps are involved in their construction: creation of a draft model, manual refinement, conversion to a mathematical format, and network evaluation 6 – 9 . Nonetheless, owing to differences in nomenclature 10 , integrity, and correctness 11 as well as the format 12 of published GSMMs, these models cannot be directly applied by other researchers. To reduce the manual labour needed for model construction and to increase the quality of GSMMs, some databases and tools, such as BIGG Models 11 and MetaNetX 13 , have been developed. Nonetheless, they are limited by the quality and quantity of models. For example, BIGG Models includes only 80 high-quality GSMMs (as of 30 November 2016), which is far from the number of published models. In MetaNetX, only 24 (14.7%) models have been published, which were validated by experimental results ( http://www.metanetx.org/cgi-bin/mnxweb/repository ). In this study, we built a database named In silico Microbial Genome-scale Metabolic Models (IMGMD) in the LAMP (Linux + Apache + MySQL + PHP) system. It provides a platform for integration and standardisation of all published microbial GSMMs. In IMGMD, users are not only able to browse and download standardised GSMMs but can also reconstruct GSMMs automatically. In addition to pathway mining and mutation library functions, users can access information that can guide pathway design and metabolic target identification.",
"discussion": "Results and Discussion Database content and web interface IMGMD (http://imgmd.jiangnan.edu.cn/database/) has a user-friendly website for the following applications: (1) It can be used to download standardised GSMMs; this module integrates model-related information, such as gene–protein–reaction relations, genome information, and references (Fig. 1A ). (2) It enables auto-reconstruction of GSMMs; this tool is based on homology alignments, and only sequences that meet a threshold are used for model construction. Additionally, transport proteins and sub-cellular location are identified for further model refinement (Fig. 1B ). (3) It can be applied to explore potential pathways; using this function, users can explore the potential pathways from one metabolite to another in a certain GSMM (Fig. 1C ). (4) It guides metabolic engineering; the mutation library includes in silico and in vivo metabolic engineering results, and accordingly, it provides guidance for target searches (Fig. 1D ). Figure 1 Summary of four functional modules in IMGMD. ( A ) The ‘model browser’ function; ( B ) the process of model auto-reconstruction; ( C ) pathway mining in a certain model organism; ( D ) gathering experimental and simulated results to identify metabolic engineering targets. \n All web interfaces of the IMGMD database were tested in various browsers, such as Google Chrome, Mozilla Firefox, Internet Explorer, Opera, and Safari on Windows or Linux platforms. Despite minor differences in appearance, all tools functioned normally in all the tested browsers and on all platforms. Among the browsers tested, Google Chrome and Mozilla Firefox provided the best user experience. Hence, we recommend that users access the database using one of these two browsers. The ‘model browse’ function in IMGMD Using model browse, users can browse, search, and download almost all published microorganism models. From the main page of model browse, basic model information, such as the number of genes, reactions, and metabolites can be accessed. Using the search bar, models can be queried by organism name, model name, kingdom, or year of publication. We chose Saccharomyces cerevisiae as an example to demonstrate the use of ‘Search for organism’. All 8 S . cerevisiae models are returned. Then, by clicking on ‘ Saccharomyces cerevisiae S288c’ for model i ND750, a user can find detailed information about the organism (e.g., strain, genome information, and ORFs), model (e.g., model name, cell compartments, model download, and in silico media for simulation), and reference (e.g. reference name, journal name, and publication date; Fig. 2 ). The genome information is linked to the NCBI database 14 , which contains the genome assembly and annotation report for a microorganism. ORFs are linked to the protein sequence downloaded from the UniProt database 15 . The in silico media are linked to the MediaDB database 16 , a database of microbial growth conditions in defined media, which can be applied as the constraint condition for metabolic model growth. These standardised GSMMs in IMGMD can be further applied to many analyses using the COBRA Toolbox 17 – 20 . The ‘model browse’ module attempts to integrate scattered data on organisms, models, and literature, and promotes the establishment of GSMM standardisation. Figure 2 The ‘models browse’ module in IMGMD. Search results for Saccharomyces cerevisiae in model browse, and the detailed information on model i ND750. \n The ‘model auto-construction’ function in IMGMD Five steps are needed to construct a model in IMGMD: (1) choosing three models for reference; (2) uploading the genome sequence; (3) choosing a threshold (eukaryotic: identity ≥40%, identity ≤10E-30; prokaryotic: identity ≥30%, identity ≤10E-6); (4) entering an e-mail address to receive results (optional); (5) submitting the job to the IMGMD database. Once the job is complete, the results contain three parts, including the model, transport proteins, and prediction of protein subcellular localisation (Fig. 3 ). Model construction is automatically implemented on the basis of the sequence alignment results. After protein sequences are submitted, the local BLASTP program will calculate the sequence similarity. Sequences that meet the established threshold are automatically screened using a Python script written in our lab. Based on the local Blast results, genes with high similarity are replaced in the reference models. Additionally, transport proteins are identified according to the alignment results, using the TCDB database 21 . For eukaryotic organisms, WoLF PSORT 22 was chosen, whereas for prokaryotic organisms (gram-positive, gram-negative, or Archaea), PSORTb 23 was employed to predict protein subcellular localisation. Figure 3 Flow chart for model construction in the IMGMD database. \n Although some software or platforms for model auto-reconstruction have been developed, including ModelSEED 24 , RAVEN 25 , COBRA Toolbox 26 , SuBliMinal 27 , these tools have their advantages and disadvantages 12 . For instance, ModelSEED ( http://modelseed.org/ ) is a Web service that includes the RAST genome annotation tool. Based on the annotation results, a model for a specific organism can be reconstructed automatically. Given that the RAST service ( http://rast.nmpdr.org/rast.cgi ) can annotate only prokaryotes, ModelSEED has limited applicability to eukaryotes. Besides, model construction by ModelSEED will take a long time, according to the job numbers. IMGMD is also a web platform that serves for model construction. It is based on the results of genome homologous alignment. Users can upload a target organism’s genome sequence and choose relevant parameters. After submission of the job to IMGMD, results will be returned within 1 day. Nonetheless, a model constructed by IMGMD is a draft model. It still needs to be further processed to obtain a GSMM. The COBRA Toolbox is based on the Matlab platform, which is commonly used for model construction. The COBRA Toolbox requires users to have basic Matlab knowledge and an advanced computer configuration for model analysis (Table 1 ). Table 1 Selected characteristics of software platforms for reconstruction and simulation of metabolic networks. Model SEED RAVEN COBRA Toolbox SuBliMinal IMGMD Input Genome annotated in RAST Annotated genome sequence GSMM Species name Species genome sequence Reference Database SEED KEGG N/A KEGG, MetaCyc IMGMD Interface Web Matlab Matlab Command Line Web License Free Free (requires a Matlab license) Free (requires a Matlab license) Free Free Output SBML, Excel SBML, Excel SBML, Excel SBML Excel Supports Simulations Yes Yes Yes No No \n Pathway mining function in IMGMD In this module, users can explore metabolic pathways at three levels. (1) According to the input metabolites as substrates and products, total pathways from a substrate to product in a GSMM can be output. For example, in the Mortierella alpina model i CY1106 28 , 21 pathways exist from glucose to pyruvate, indicating that in addition to the basic glycolysis pathway in M . alpina (according to the KEGG pathway 29 ), other pathways also could generate pyruvate. On the web page of pathway-mining results, information about the substrate and production can be linked to some metabolic databases, like KEGG 29 , ModelSEED 24 , ChEBI 30 , and PubChem 31 . Besides, on the page of detailed pathway information, reactions participating in a pathway are shown, including Reaction ID, Formula, Genes, Subsystem, and EC numbers (Fig. 4 ). (2) Comparisons between two or more GSMMs help to understand phenotypic characteristics based on metabolic pathway differences. When comparing the pathway differences between two Archaea, Methanococcus maripaludis ( i MM518) 32 and Methanosarcina barkeri ( i MG746) 33 , there were 8 and 12 pathways from glucose to pyruvate, respectively. (3) Pathways that generate highly valuable products may exist in typical organisms. To mine these potential pathways, users can choose all collected models for the search, and then choose reactions in which species and corresponding genes can serve as references for a target strain to guide strain design. Considering these three levels, the function of pathway mining may be useful in synthetic biology and systems metabolic engineering. Figure 4 Pathway mining results for the i CY1106 model from glucose to pyruvate. A summary of all pathways found in the i CY1106 model from glucose to pyruvate, and detailed reaction information for a pathway in model i CY1106. \n Mutation library function in IMGMD The pathway prediction tool enables new pathway design for metabolic engineering; additionally, the mutation library function can be used for optimisation of the host strain. It can help to identify targets that couple cell growth with product formation, e.g., targets for gene upregulation, downregulation, and gene deletion 34 , 35 . In IMGMD, a library that combines in vivo and in silico results to guide metabolic engineering was created. Organisms, models, and genes can be used as keywords to search for mutation information. For example, in a search for mutation information with model i AF1260, 217 results can be found. The effect of a knockout of b4025 , which encodes glucosephosphate isomerase (pgi, EC: 5.3.1.9) in E . coli , the growth rate and production rate can be viewed on another webpage (Fig. 5 ). According to the information on this new page, when galactose serves as a carbon source, the in silico growth decreases by 36.1%, while the in vivo growth rate increases by 12.0% 36 (Table 2 ). Furthermore, the amino acid sequence and nucleic acid sequence of gene b4025 were also included (Fig. 5 ). The EC number of 5.3.1.9 is linked to BRENDA database for more detailed information. Figure 5 \n In silico and in vivo results on the b4025 deletion in the i AF1260 model, and detailed information on b4025 in E . coli . \n Table 2 Results of the b4025 deletion in E . coli collected by IMGMD 36 . strain glucose as carbon source galactose as carbon source growth rate H 2 production (mol/mol) growth rate H 2 production (mol/mol) Scaled OD 600 \n Scaled OD 600 \n \n in silico \n \n in vivo \n \n in silico \n \n in vivo \n \n in silico \n \n in vivo \n \n in silico \n \n in vivo \n wide type 1.31 1.31 1.713 1.58 0.83 0.83 1.74 1.48 \n Δ \n pgi \n 1.19 0.02 1.740 0 0.53 0.93 1.835 1.16 \n pgi (b4025): encoding glucose-6-phosphate isomerase (EC: 5.3.1.9), which can catalyse d -glucose 6-phosphate into d -fructose 6-phosphate in glycolysis pathway. \n In this module, 950 total mutation results were collected by literature mining. Additionally, 885 results (93.2%) were related to various knockout strategies, involving different algorithms, such as OptKnock 19 , GDLS 37 , ReacKnock 38 , DBFBA 39 , BAFBA 40 , and RobustKnock 41 . The remaining results are related to gene upregulation or downregulation 42 . Combined with the pathway mining and mutation library modules, IMGMD can be used to guide systems metabolic engineering, for both pathway screening and for target identification."
} | 3,851 |
18794915 | null | s2 | 7,235 | {
"abstract": "In this Analysis we use published 16S ribosomal RNA gene sequences to compare the bacterial assemblages that are associated with humans and other mammals, metazoa and free-living microbial communities that span a range of environments. The composition of the vertebrate gut microbiota is influenced by diet, host morphology and phylogeny, and in this respect the human gut bacterial community is typical of an omnivorous primate. However, the vertebrate gut microbiota is different from free-living communities that are not associated with animal body habitats. We propose that the recently initiated international Human Microbiome Project should strive to include a broad representation of humans, as well as other mammalian and environmental samples, as comparative analyses of microbiotas and their microbiomes are a powerful way to explore the evolutionary history of the biosphere."
} | 221 |
35495540 | PMC9042051 | pmc | 7,238 | {
"abstract": "With the background of contemporary art, using comprehensive materials to create artworks is becoming more and more common. The new era of digital image-based copperplate artworks, using photosensitive lithography, has given traditional art forms new life and greater popularity in the digital age. However, the patterns and textures of the works created by the new techniques are generally shallow, and the copper surface is easily damaged and loses its aesthetic value, which makes it a practical problem to protect such works more effectively. In this paper, a facile method is adopted, wherein a superhydrophobic film is constructed on the surface of copperplate images by straightforward immersion in (heptadecafluoro-1,1,2,2-tetradecyl)trimethoxysilane (FAS-17) solution to achieve the anticorrosive protection of copperplate artworks. The hydrophobicity of the copper surface was analyzed using an instrument that measures contact angles. The superhydrophobic surface morphology and composition were analyzed with a scanning electron microscope coupled with an energy-dispersive spectrometer, and the corrosion resistance was analyzed using an electrochemical workstation. A systematic study is presented on the effect of the immersion time in FAS-17 and the concentration of FAS-17, and the optimal preparation conditions of the superhydrophobic film were determined, which means that the copper substrates were immersed in 0.7 mol L −1 FAS-17 for 40 min. After the treatment of the surface to make it superhydrophobic, the contact angle and the corrosion inhibition efficiency of the copperplate etching surface reached 161.2° and 95.7%, respectively. The results show that the superhydrophobic film was successfully prepared on the surface of the artwork based on copper, which can effectively improve the corrosion resistance and is beneficial for the long-term protection of artwork.",
"conclusion": "4. Conclusion A straightforward immersion method was adopted to prepare a superhydrophobic surface on copper by using a mixed solution of NaOH and (NH 4 ) 2 S 2 O 8 as a surface roughening reagent, and FAS-17 as a low surface energy modifier. The effects of the time of immersion in FAS-17 and the initial concentration of FAS-17 were studied and the results showed that the copper surface had the best superhydrophobicity on immersion in 0.7 mol L −1 FAS-17 for 40 min. Under the optimal conditions, both the contact angle and the corrosion inhibition efficiency reached the maximum value of 161.2° and 95.7%, respectively. Moreover, the stability performance of the superhydrophobic surface on the copper substrate was studied by electrochemical tests, and the results indicated that the superhydrophobic film had good stability. The superhydrophobic film was successfully built on the surface of the copper substrate by using FAS-17, and it can effectively improve the copper's corrosion resistance, which holds promising application prospects in the protection of copper-based artworks.",
"introduction": "1. Introduction Copperplate artworks are also known as copperplate etchings and copperplate paintings. The art of etching is elegant and has been regarded as a rare type of painting. All the masters of past artistic dynasties were deeply focused on the artistic creation of etchings. From the German Durer to the Dutch Rembrandt, from the Spanish Goya to the French impressionists Manet, Monet, Sislan, Degas, and so on, to the modern Picasso and Matisse, artistic masters have left very exquisite etchings. There are many production methods but anticorrosive wax or preservatives (generally made of yellow wax, rosin, asphalt and other acid-resistant materials) are usually used in a coating layout, forming a layer of anticorrosive film. A needle is used for the layout painting and then the image is placed in the corrosive solution (commonly a nitric acid solution). The point at which the needle scratches off the anticorrosive film, that is, where corrosion takes place, a concave line is formed, and the longer the corrosion time, the deeper the concave line after removing the anticorrosive film; this plate is then used on the copperplate machine for gravure printing. Due to the line thickness, depth, and density, the printed works will form a multi-level, complex picture with tonal change. The art of copper etchings enjoys a worldwide reputation for its advanced technology and cultural connotations over the years, which make it quite precious. In contemporary times, copper etchings appear as new and creative methods and have become a new artistic carrier. Combined with the traditional art of new media and digital technology, the ancient etchings form the digital image-based etchings of the new era. Etchings based on digital images are not the same as traditional techniques; they are no longer in the form of direct hand painting but computers are first used for image creation, then the satisfactory image is output into a film, using the film to expose the copper plate coated with photosensitive adhesive or film to obtain the digital copperplate artwork. Although these artworks created with this new technique have strong availability and popularity, there are also some shortcomings. For example, the patterns and textures formed are generally shallow, and beauty and value are easily lost because the copper surface is corroded. How to more effectively protect copper bases from corrosion is a practical problem for the long-term preservation of such artworks. Although copper and its alloys are characterized by relatively good corrosion resistance and fouling resistance, they are prone to corrosion and serious surface discoloration due to long-term exposure to humid air or the presence of corrosive media such as CN − and Cl − . 1 Corrosion inhibition is one of the effective methods for solving the problem of the corrosion of copper and its alloys. At present, the most extensive method is to select suitable organic compounds as corrosion inhibitors, such as benzotriazole and its derivatives, various thiazole derivatives 2 and imidazole. 3 However, most of these inhibitors are expensive and toxic and are not good for the environment, so their use is limited. As such, the development of an environmentally friendly and economical metal protective layer on copper surfaces has gradually become a research hotspot in recent years. The superhydrophobic surface method has been attracting more and more attention 2 and it is based on the concept of “the Lotus effect” proposed by German botanist Wilhelm Barthlott. 1,4 The superhydrophobic surface method that mimics the lotus leaf effect has the advantages of hydrophobicity and self-cleaning. 2–6 At present, the preparation of a superhydrophobic surface can be roughly divided into two categories: the construction of a micron-nano-sized rough structure on the surface of hydrophobic materials with low surface energy or modifications of this rough structure using low-surface energy materials. 7 Preparation methods for superhydrophobic surfaces mainly include the template method, the electrochemical method, the etching method, 8 the gel method, 9 the layered self-assembly method 10 and so on. Active hydrophobic organic molecular groups used to prepare superhydrophobic films on the surface of metal bases include –SiOH, –SH, and –COOH. 9 Common hydrophobic materials include siloxane, fluoroalkyl silane, nutmeg acid, stearic acid, 11,12 etc. The study of superhydrophobic surfaces has generally focused on pure metal or alloys. Recently, the fabrication of superhydrophobic surfaces has been applied to the surfaces of pure copper, 13 magnesium alloy 14 and other metals. Due to the special structural composition of copperplate artistic works, many difficulties remain in the study of superhydrophobic surfaces. At present, there are few studies in this field. In this paper, a (heptadecafluoro-1,1,2,2-tetradecyl)trimethoxysilane (FAS-17) solution was used to prepare a superhydrophobic surface on copperplate artworks. The hydrophobicity and corrosion resistance of superhydrophobic surfaces have been investigated. By measuring contact angles, surface hydrophobicity was analyzed, and the superhydrophobic surface morphology characteristics and composition analysis were determined using a scanning electron microscope coupled with an energy-dispersive spectrometer. An electrochemical workstation was used to perform the corrosion resistance analysis to prove that this method has a good inhibitory effect on copper corrosion. This study can be used as a reference for long-term anti-corrosion protection of copperplate artworks by preparing long-acting and durable super-hydrophobic anti-corrosion surfaces and enhancing the cleanliness and aesthetics of the works' surfaces.",
"discussion": "3. Results and discussion 3.1 Contact angle measurement In order to determine the optimal immersion time and concentration of FAS-17 for obtaining the copperplate artworks with the best superhydrophobicity, experiments were conducted based on the single-factor method, which is shown as Fig. 1 . FAS-17 plays an important role in the change of the copper surface wettability. For the bare copperplate artwork, the water droplets were semi-circular on the surface and the contact angle was 79.8°. As shown in Fig. 1(a) , when immersion time was increased to 40 min, the contact angle between the surface and the water of the test piece superhydrophobic film reached the maximum value of 161.1°. Fig. 1(b) shows that the CA reached the maximum value of 161.2° with the concentration of FAS-17 of 0.7 mol L −1 . With further increase in the immersion time and the concentration of FAS-17, the CA of the copper surface decreased, which can be explained by the damage to the superhydrophobic film. Thus, the optimal preparation process of the superhydrophobic film was immersion in FAS-17 with a concentration of 0.7 mol L −1 for 40 min. Fig. 1 The water contact angles of the samples. (a) The effect of the immersion time with the FAS-17 concentration of 0.5 mol L −1 at 298 K. (b) The effect of the concentration of FAS-17 when the immersion time was 40 min. The superhydrophobicity can be expressed by the Cassie–Baxter equation ( eqn (1) ): 15 1 cos θ c = f 1 cos θ − f 2 where θ c and θ refer to the water contact angles of the copper substrate modified by FAS-17 and the bare copper substrate, respectively. f 1 and f 2 are the area fractions of liquid with solid and air, respectively ( f 1 + f 2 = 1). The calculated values of f 1 and f 2 are given in Tables 1 and 2 . \n f \n 1 and f 2 values of samples with different immersion times Immersion time \n f \n 1 \n \n f \n 2 \n Blank 1.000 0.000 10 min 0.273 0.726 20 min 0.149 0.850 30 min 0.116 0.883 40 min 0.046 0.953 50 min 0.110 0.889 60 min 0.139 0.860 \n f \n 1 and f 2 values of samples immersed in different concentration of FAS-17 Concentration (mol L −1 ) \n f \n 1 \n \n f \n 2 \n Blank 1.000 0.000 0.1 0.140 0.859 0.3 0.064 0.935 0.5 0.031 0.968 0.7 0.026 0.973 0.9 0.040 0.959 \n f \n 2 reached the maximum values of 95.3% and 97.3% under the optimal conditions, which indicated that the “air cushion” hinders the direct contact of water droplets with the copperplate artwork. 16 3.2 Superhydrophobic surface protection of copperplate etchings \n Fig. 2 shows an effect diagram for the superhydrophobic surface anti-corrosion effects of copperplate artworks. The comparison of Fig. 2(a) and (b) shows that the appearance of the copperplate artworks does not change significantly after dipping and coating in the superhydrophobic protective film; i.e. , the film does not affect the aesthetic value. The comparison of Fig. 2(a) and (c) shows that the corrosion of the copperplate artworks is not obvious after 7 days in an ozone aging chamber, due to the superhydrophobic protective film. Fig. 2(a) and (d) show that the corrosion of the copperplate artistic work without superhydrophobic protective film was evidenced after 7 days in the ozone aging chamber. Fig. 2(c) and (d) show that the dip-coating had an obvious protective effect on the corrosion of the copperplate artwork. Fig. 2 Anti-corrosion effect of copperplate digital images with a superhydrophobic surface. (a) Image with a superhydrophobic surface; (b) image without a superhydrophobic surface; (c) image with a superhydrophobic film tested in an ozone aging chamber for 7 days; (d) image without a superhydrophobic film tested in an ozone aging chamber for 7 days. 3.3 Morphology characteristics and surface composition analysis of the superhydrophobic surface \n Fig. 3 shows the scanning electron microscope image of the copperplate specimen without and with the superhydrophobic film. There were some scratches on the surface of the test piece of the copper substrate ( Fig. 3(a) ), and the substrate surface was relatively smooth between scratches. The surface morphology with the superhydrophobic film is shown in Fig. 3(b) , which was covered with a dense network structure. This microstructure produced numerous holes to trap air and increased gas–liquid contact area, so that corrosion inhibition performance was improved. According to the EDS data ( Table 3 ), carbon, oxygen and silicon contents increased after superhydrophobic treatment, indicating that a superhydrophobic film was formed on the surface of the test piece. Fig. 3 SEM-EDS image of a copper specimen surface. (a) Surface of a copper substrate; (b) surface of a copper substrate with superhydrophobic film. Elemental compositions of the copper substrate plate and the copper substrate plate with super-hydrophobic film Elemental compositions Cu/wt% Zn/wt% C/wt% O/wt% F/wt% Si/wt% Blank 58.34 38.63 2.11 0.93 — — Copper substrate with superhydrophobic film 59.55 24.37 2.32 11.00 2.28 0.49 3.4 Electrochemical impedance characteristics and corrosion resistance of copper with superhydrophobic surfaces 3.4.1 Effect of immersion time \n Fig. 4 shows the polarization curve of the copper substrate test sample with superhydrophobic film prepared under different immersion times in a solution simulating atmospheric conditions. The fitting results are shown in Table 4 . E corr is the self-corrosion potential. η is the corrosion inhibition efficiency and it is calculated as the following relationship: 17 η = (1 − J corr / J 0 corr ) × 100% where, J 0 corr and J corr are the self-corrosion current densities of the copper substrate test plates before and after superhydrophobic treatment, respectively. The corrosion potentials of the copper substrate all changed and had a lower corrosion current density after the fabrication of the superhydrophobic surface. It can be seen from Table 4 that E corr reached the maximum value of −0.062 mV and J corr reached the minimum value of 5.07 × 10 −3 mA cm −2 when the immersion time was 40 min. According to the calculations, η reached 91.1%, indicating that the superhydrophobic film formed on the test plate surface could effectively prevent the plasma diffusion of Cl − , HCO 3 − , CO 3 2− and S 2− from the solution to its surface and improve the corrosion resistance of the test plate. Fig. 4 Polarization curves of a copper substrate with superhydrophobic film prepared under different immersion times in a solution simulating atmospheric conditions. Electrochemical parameters of the polarization curve of specimens in a solution simulating atmospheric conditions Immersion time \n E \n corr /mV \n J \n corr /mA cm −2 \n βa /mV \n βc /mV \n η /% 0 min −0.162 5.70 × 10 −2 489.00 75.57 — 10 min −0.077 3.28 × 10 −2 236.29 54.74 42.5 20 min −0.068 8.47 × 10 −3 190.08 68.90 85.1 30 min −0.078 1.24 × 10 −2 240.67 70.97 78.2 40 min −0.062 5.07 × 10 −3 191.46 70.35 91.1 50 min −0.066 7.98 × 10 −3 208.68 56.08 86.0 60 min −0.068 1.03 × 10 −2 230.63 51.88 81.9 \n Fig. 5(a) shows the Nyquist plots and their fitting for the copper substrate samples before and after fabrication of the superhydrophobic surface by immersion in a solution simulating atmospheric conditions. The arc diameter of the capacitive reactance first increased and then decreased with the increasing immersion time. It is worth mentioning that the larger arc diameter represents higher charge transfer resistance. 18 As shown in Fig. 5(a) , when the immersion time was 40 min, the arc radius of the superhydrophobic sample was the largest, which is consistent with the results of the polarization curves. Fig. 5 Nyquist (a) plots and Bode (b) plots of the copper substrate before and after the fabrication of the superhydrophobic surface by immersion in the solution simulating atmospheric conditions; (c) the equivalent circuit used for the copper substrate; (d) the equivalent circuit used for the copper substrate after superhydrophobic surface formation. It can be seen from Fig. 5(b) that there is one time constant for the untreated copper substrate and there are two time constants for the superhydrophobic specimens, so the measured results can be fitted by the electrical equivalent circuit shown in Fig. 5(c) and (d) , respectively. The fitted EIS parameters are listed in Table 5 . Electrochemical impedance parameters obtained from EIS fitting under different immersion times Immersion time \n R \n s /Ω cm 2 \n Y \n CPEf /Ω −1 cm −2 \n n \n f \n \n R \n f /Ω cm 2 \n Y \n CPEdl /Ω −1 cm −2 \n n \n dl \n \n R \n ct /Ω cm 2 0 min 36.13 — — — 0.001106 0.6263 609.6 10 min 20.28 2.156 × 10 −5 0.5981 106.8 1.603 × 10 −4 0.7285 1147 20 min 20.36 2.402 × 10 −5 0.5667 214.8 1.413 × 10 −4 0.6426 1982 30 min 19.41 2.679 × 10 −5 0.5783 218.3 1.119 × 10 −4 0.8211 2190 40 min 17.54 4.496 × 10 −5 0.5493 272.1 7.536 × 10 −5 0.7871 3813 50 min 18.51 6.135 × 10 −5 0.4131 204.0 8.036 × 10 −5 0.8679 2331 60 min 19.01 5.003 × 10 −5 0.5657 204.5 9.260 × 10 −5 0.8250 2169 \n Table 5 shows that the R ct (the charge transfer resistance) tended to increase, especially for the immersion time of 40 min, reaching 3813 Ω cm 2 , which increased by one order of magnitude as compared to the bare copper. The larger the charge transfer resistance, the more difficult it is for the corrosion medium to transfer to the copper surface and to produce an electrochemical corrosion reaction. The R f (the surface film resistance) had a similar tendency to the R ct . The changes in the values of R ct and R f indicate that the corrosion resistance of the copper substrate surface after the fabrication of the superhydrophobic surface with FAS-17 was significantly improved. 3.4.2 The effect of FAS-17 concentration \n Fig. 6 shows the polarization curve of bare copper and superhydrophobic copper surfaces under five different concentrations of FAS-17. The electrochemical parameters obtained from the Tafel curves are given in Table 6 . It can be seen that all of the superhydrophobic samples have a lower corrosion current density as compared to bare copper. When the concentration of FAS-17 was 0.7 mol L −1 , the corrosion current density achieved the minimum value of 2.46 × 10 −3 mA cm −2 and the corrosion inhibition efficiency reached 95.7%, indicating that the corrosion rate of the copper substrate was the lowest at this time, which is consistent with the contact angle test result. Fig. 6 Polarization curves of a copper substrate under different concentrations of FAS-17 in solutions simulating atmospheric conditions. Electrochemical parameters of the polarization curve of specimens in solutions simulating atmospheric conditions Concentration \n E \n corr /mV \n J \n corr /mA cm −2 \n βa /mV \n βc /mV \n η /% 0 mol L −1 −0.162 5.70 × 10 −2 489.00 75.57 — 0.1 mol L −1 −0.049 6.79 × 10 −3 189.79 63.33 88.1 0.3 mol L −1 −0.061 5.85 × 10 −3 193.54 47.54 89.7 0.5 mol L −1 −0.057 3.56 × 10 −3 172.47 56.61 93.8 0.7 mol L −1 −0.051 2.46 × 10 −3 171.94 70.88 95.7 0.9 mol L −1 −0.065 2.95 × 10 −3 173.07 52.05 94.8 The Nyquist plots for samples immersed in different concentrations of FAS-17 are shown in Fig. 7(a) . The sample with 0.7 mol L −1 FAS-17 treatment exhibited the largest diameter of capacitive loop, indicating that the copper substrate plate had the best superhydrophobicity under these conditions. Fig. 7(b) represents the Bode plots of the samples. There are two time constants for superhydrophobic samples, so Fig. 5(d) can be used to fit the electrochemical data; the results are listed in Table 7 . Compared to the bare copper, both R ct and R f increased and peaked at about 1.063 × 10 4 Ω cm 2 and 112.2 Ω cm 2 , respectively, at 0.7 mol L −1 of FAS-17. Combining Tables 5 and 7 , it can be seen that the best process for preparing superhydrophobic films on the copper surface is to immerse in a FAS-17 solution with the concentration of 0.7 mol L −1 for 40 min. Fig. 7 Nyquist (a) plots and Bode (b) plots of the copper substrate in solutions simulating atmospheric conditions. Electrochemical impedance parameters obtained from EIS fitting under different concentrations of FAS-17 Concentration \n R \n s /Ω cm 2 \n Y \n CPEf /Ω −1 cm −2 \n n \n f \n \n R \n f /Ω cm 2 \n Y \n CPEdl /Ω −1 cm −2 \n n \n dl \n \n R \n ct /Ω cm 2 0 mol L −1 36.13 — — — 0.001106 0.6263 609.6 0.1 mol L −1 16.87 1.449 × 10 −5 0.6521 79.35 6.788 × 10 −5 0.6801 5.020 × 10 3 0.3 mol L −1 17.92 1.687 × 10 −5 0.5594 80.73 6.165 × 10 −5 0.6657 5.793 × 10 3 0.5 mol L −1 19.52 2.679 × 10 −5 0.5783 82.67 5.735 × 10 −5 0.6651 2.190 × 10 3 0.7 mol L −1 18.28 1.376 × 10 −5 0.6616 112.2 5.386 × 10 −5 0.7021 1.063 × 10 4 0.9 mol L −1 13.24 1.691 × 10 −5 0.6907 75.83 5.607 × 10 −5 0.6728 8.271 × 10 3 3.5 Stability of the superhydrophobic film To study the stability performance of superhydrophobic films prepared under optimal conditions, the impedance spectrogram and the polarization curves were obtained after immersion for 7 d in a solution simulating atmospheric conditions. As shown in Fig. 8(a) , there was still a capacity loop in the high-frequency area after immersion for 7 d. Although the arc diameter was slightly smaller than that after immersion for 2 h, it was much larger than that of the bare copper substrate. According to Tables 8 and 9 , the superhydrophobic film still had a good corrosion-inhibiting performance after immersion for 7 d, suggesting that the prepared superhydrophobic film has good stability. Fig. 8 Nyquist plots (a) and polarization curves (b) for superhydrophobic copper in a solution simulating atmospheric conditions for different immersion times. EIS fitting results of superhydrophobic copper in a solution simulating atmospheric conditions for different immersion times Time \n R \n s /Ω cm 2 \n Y \n CPEf /Ω −1 cm −2 \n n \n f \n \n R \n f /Ω cm 2 \n Y \n CPEdl /Ω −1 cm −2 \n n \n dl \n \n R \n ct /Ω cm 2 Blank 36.13 — — — 0.001106 0.6263 6.096 × 10 2 Immerse for 2 h 17.54 4.496 × 10 −5 0.5493 272.1 7.536 × 10 −5 0.7871 3.813 × 10 3 Immerse for 7 d 18.24 4.691 × 10 −5 0.6907 156.83 8.407 × 10 −5 0.6728 2.371 × 10 3 Electrochemical parameters of superhydrophobic copper in a solution simulating atmospheric conditions for different immersion times Time \n E \n corr /V \n I \n corr /mA cm −2 \n βa /mV \n βc /mV \n η /% Blank −0.162 5.70 × 10 −2 489.00 75.57 — Immerse for 2 h −0.066 5.07 × 10 −3 191.46 70.35 91.1 Immerse for 7 d −0.086 6.21 × 10 −3 208.33 102.93 89.1"
} | 5,813 |
34143385 | PMC8494672 | pmc | 7,239 | {
"abstract": "The mitigation of metals contamination is currently a crucial issue for the reclamation of mine sites. Indeed, mine wastes are often disposed in open dumps and consequently pollutants are subjected to dispersion in the surrounding areas. In this study, the potential use of Helichrysum microphyllum subsp. tyrrhenicum for phytostabilization was evaluated in ex situ conditions. Ninety specimens were randomly selected and were planted in three substrates (reference substrate, mine waste materials, and mine wastes with compost). Mineralogical compositions of substrates, rhizosphere, and roots were assessed through X-ray diffraction (XRD). Zn, Pb, and Cd concentrations of substrates, rhizosphere, soil pore waters, and plant tissues were determined. The phytostabilization potential was determined through the application of biological accumulation coefficient (BAC), biological concentration factor (BCF), and translocation factor (TF). Moreover, survival and biometric parameters were assessed on plant specimens. The polluted substrates and related rhizosphere materials were mainly composed of dolomite, quartz, pyrite, and phyllosilicate. Zn was the most abundant metal in substrates, rhizosphere, and soil pore waters. XRD analysis on roots showed the presence of amorphous cellulose and quartz and Zn was the most abundant metal in plant tissues. H. microphyllum subsp. tyrrhenicum restricts the accumulation of the metals into roots limiting their translocation in aereal parts, indicating its potential use as phytostabilizer (BCF, BAC, TF < 1). Survival and growth data showed a great adaptability to different substrates, with an evident positive effect of the implementation of compost which increased the plant survival and decreased the metals uptake into roots. Supplementary Information The online version contains supplementary material available at 10.1007/s11356-021-14710-y.",
"conclusion": "Conclusion This ex situ phytoremediation experimental study evaluated the phytostabilization potential of H. tyrrhenicum , suggesting “ ex-situ experiments” as an important source of informations when planning recovery actions. Results showed that H. tyrrhenicum accumulated high concentrations of Zn, Pb, and Cd into roots, mirroring the high contamination levels observed in the polluted substrates. The assessed values of these metals in the considered mine substrate are higher than the established threshold limits of Italian law for industrial use of soil and consequently, hindering the possibility of using these areas. A “gentle remediation,” as also proposed by Jiménez et al. ( 2021 ), can give an important contribution for both remediation and future use of these sites. Taking account of the assessed metal concentrations and BAC, BCF, and TF values, this study emphasizes the phytostabilization potential of H. tyrrhenicum. Moreover, the high plant survival and the assessed growth parameters confirmed the great adaptability of H. tyrrhenicum to the limiting ecological conditions of mine sites, also without amendments. Anyway, the addition of compost created the optimal compromise for plant establishment: on one side, it decreased the metal accumulation into roots; on the other one, it helped in the development of the specimens by increasing the survival. Owing to these reasons, compost can be suggested as an effective amendment in phytostabilization activities of these areas. Future investigation could concern an in situ phytoremediation experiment using seedlings grown naturally in the mine areas, with the aim to shed light about the survival and the metal uptake in this sensitive step of a plant life under field condition.",
"introduction": "Introduction Mine waste dumps are limiting environments for the growth and establishment of most plant species. These unfavorable conditions are related to the properties of mine substrates such as: the absence of a top soil, the lack of nutrients (K, N, and P) and organic matter, the poorly developed structure, and the high concentration of toxic elements like heavy metals (Mendez and Maier 2008 ; Baker et al. 2010 ; Anawar et al. 2011 ; De Agostini et al. 2020 ). Mining activity and in prior way its waste materials are some of the main sources of toxic metal pollution in biogeochemical spheres. Hence, the mitigation of contaminant dispersion in mine sites is currently an environmental and sanitary challenge. Indeed, mine wastes are often disposed in open dumps mainly made of fine particles, which can be subjected to dispersion by wind and water erosion, especially in arid and semiarid environments (Munshower 1994 ; Mendez and Maier 2008 ; Sims et al. 2013 ; Doumas et al. 2018 ; Zine et al. 2021 ). Last but not the least, mining activities can seriously alter ecosystems in terms of soil structure and related functions with loss of biodiversity and land degradation (Ahirwal and Pandey 2021 ). Despite these unfavorable conditions for plant growth, several species have developed different adaptations to survive in these environments, like metal-tolerance capability. Plants who are sufficiently resilient to colonize metals’ polluted areas and have developed metal tolerance can have an unquestionable role in the development of a long-term plant canopy (Freitas et al. 2009 ). Taking advantage of these benefits, plants can give their contribution in the reclamation of these sites by applying phytoremediation, that is a low impact and cost-effective strategy and for the recovery of mine areas (Freitas et al. 2009 ; Favas et al. 2014 ; Bacchetta et al. 2015 ). In order to achieve this purpose, it is necessary to select the best performing plant species with a special focus on native taxa. Owing this issue, several reasons promote the implementation of autochthonous plants over non-native species: (i) they are adapted to local climate/adverse conditions (Pandey and Singh 2012 ; El Hasnaoui et al. 2020 ) and are stronger in terms of growth, survival, and reproduction under environmental stress (Midhat et al. 2017 ); (ii) their use can preserve the plant diversity of the natural territories (Cao et al. 2009 ; Bacchetta et al. 2012 , 2015 ; Concas et al. 2015a ); (iii) they help to start the rehabilitation of vegetational dynamics by the improvement of the physical-chemical properties of the substrates (Vacca 2000 ; Vacca et al. 2017 ), and (iv) they help to the establishment of a long-term plant canopy on mine wastes with a relatively low cost inputs and limited maintenance (Bacchetta et al. 2007a ; Pandey et al. 2015 ; Monaci et al. 2020 ; Zine et al. 2020 ). All these points are pieces of a more holistic and sustainable approach in phytoremediation (Pandey et al. 2015 ; Pandey and Omesh Bajpai 2019 ). An important role in phytoremediation can be played by soil amendments since they can improve the plant growth modifying the physical-chemical properties of mine substrates and the bioavailability of metals (Song and Greenway 2004 ; Bacchetta et al. 2012 , 2015 ). Among the available soil amendments, compost can optimize phytoremediation by lowering the metal availability in substrates and the plant’s tissue uptake and increasing the content of nutrients and organic matter in substrates (Bacchetta et al. 2015 ). Moreover, compost modifies the structure of substrate through the improvement of the bulk’s water retention, density, and ventilation (Fagnano et al. 2011 ). The result of this interaction leads to a better roots penetration and consequently results in a better feeding of the plant. Sardinia (Italy) had an important mining history, which left a burdensome environmental legacy: after the shutdown of mines, few containment activities of mine tailing dumps and mitigation of the related metals impact were developed (Jiménez et al. 2014 ; Bacchetta et al. 2018 ), leaving consequently huge quantities (approximately 70 million of m 3 ) of polluted materials subjected to erosion and dispersion (RAS – Regione Autonoma della Sardegna 2003 ; Bacchetta et al. 2018 ; Boi et al. 2020a , b ). Along the last 15 years, several studies were carried out on the Sardinian mine context through field sampling, in situ and ex situ phytoremediation experiments. In details, these studies highlighted that different autochthonous species are able to grow in these environments and to tolerate extremely high heavy metal concentrations (e.g., Cao et al. 2004 , 2009 ; Jiménez et al. 2005 , 2011 , 2014 ; Bacchetta et al. 2012 , 2015 , 2017 , 2018 ; Concas et al. 2015a ; De Giudici et al. 2015 , 2017 ; Medas et al. 2015 , 2019 ; Fancello et al. 2019 ; Boi et al. 2020a ; De Agostini et al. 2020 ). These abandoned mining areas are anyway rich in terms of plant diversity (Angius et al. 2011 ; Angiolini et al. 2005 ; Bacchetta et al. 2007a , b , c ; Zavattero et al. 2005 , 2006 ) and in particular of endemic taxa: for example, in the Monteponi mine area (Iglesias, South West Sardinia), endemic species are 18.1% of the local flora (Zavattero et al. 2005 ). Indeed, several endemic and/or threatened plant species like Echium anchusoides Bacch., Brullo & Selvi, Brassica insularis Moris, Clinopodium sandalioticum (Bacch. & Brullo) Bacch. & Brullo ex Peruzzi & F.Conti, Staphisagria requienii (DC.) Spach subsp. picta (Willd.) Peruzzi, Euphorbia pithyusa L. subsp. cupanii (Guss. ex Bertol.) Radcl.-Sm., Helichrysum microphyllum Cambess. subsp. tyrrhenicum Bacch., Brullo & Giusso, Iberis integerrima Moris, Linum muelleri L., Limonium merxmulleri Erben, Ptilostemon casabonae (L.) Greuter, Santolina insularis (Gennari ex Fiori) Arrigoni, and Scrophularia canina subsp. bicolor (Sibth. & Sm.) Greuter are recognizable on these sites. For the aim of this study, a lab-scale phytoremediation test was carried out on contaminated matrices from the Sulcis–Iglesiente area (South West Sardinia, Italy) using H. microphyllum subsp. tyrrhenicum (hereafter H. tyrrhenicum ), which is an endemic plant species of Sardinia and Corsica (Bacchetta et al. 2003 ). Previous studies concerning this taxon highlighted its great adaptability to restrictive environments as pioneer of mine wastes (Angiolini et al. 2005 ; Bacchetta et al. 2009 ), the metal tolerance towards Zn, Pb, and Cd (Cao et al. 2004 ; Bacchetta et al. 2017 , 2018 ; Medas et al. 2018 ; Boi et al. 2020a ), and the capability of its seeds to germinate and seedlings to survive under very high concentrations of Zn and Pb (Boi et al. 2020b ). The aims of this experimental study were to evaluate: (i) the capability of H. tyrrhenicum to tolerate high concentration of Zn, Pb, and Cd, assessed in terms of plant survival and growth; (ii) its phytoremediation potential towards these metals by assessing the biological indexes regarding metal accumulation in plant tissue, and (iii) the effectiveness of organic amendments (compost) in the mitigation of heavy metal stress.",
"discussion": "Results and discussion Chemical and mineralogical characterization of the substrates and rhizosphere materials The mineralogical compositions of the substrates and rhizosphere materials collected during the experiment are reported in Table 1 . RS substrate and rhizosphere materials were mainly composed of quartz (SiO 2 ), phyllosilicates, and feldspars (albite, anorthite, orthoclase, microcline, and sanidine) and no minerals linked to polluting metals, such as pyrite (FeS 2 ), goethite (Fe +3 O(OH)), jarosite (KFe 3+ 3 (SO 4 ) 2 (OH) 6 ), and anglesite (PbSO 4 ), were observed. The polluted substrates (CP and CPC) and related rhizosphere materials were mainly composed of dolomite (CaMg(CO 3 ) 2 ), quartz, pyrite and phyllosilicate, and some secondary minerals like goetithe, gypsum (CaSO 4 ·2(H 2 O)), jarosite, and anglesite, according to previous mineralogical characterizations on samples of Campo Pisano mine dump (De Giudici et al. 2015 ; Bacchetta et al. 2017 , 2018 ; Boi et al. 2020a ). In details, quartz and dolomite form the gangue minerals of the ore body and are often recognized together with barite (BaSO 4 ) and Fe-oxy-hidroxydes (De Giudici et al. 2015 ), whereas pyrite is associated with sulfide deposits in SW Sardinia (Boni et al. 1999 , 2003 ) and gypsum and jarosite derive from the dissolution of Ca-carbonates and the oxidation of pyrite, respectively. It is interesting to note that mineral composition of rhizosphere materials did not change during all the trial (Table 1 and Fig. 3 ). Probably, a long-term in situ phytoremediation experiment could be necessary to highlight an evolution in rhizosphere mineralogy.\n Table 1 Mineralogical composition of substrates and rhizosphere in RS, CP, and CPC. The black points indicate the presence of the mineral; S, substrate; Rz, rhizosphere materials; T0, before planting; T1, after 1 month; T2, after 3 months; T3, after 6 months Time Dolomite Quartz Phyllosilicates Albite Anorthite Orthoclase Sanidine Microcline Pyrite Goethite Gypsum Jarosite Anglesite RS S T0 ● ● ● ● ● ● ● Rz T1 ● ● ● ● ● ● Rz T2 ● ● ● ● ● ● Rz T3 ● ● ● ● ● CP S T0 ● ● ● ● ● Rz T1 ● ● ● ● ● ● ● ● Rz T2 ● ● ● ● ● ● Rz T3 ● ● ● ● ● ● ● ● CPC S T0 ● ● ● ● Rz T1 ● ● ● ● ● ● ● ● Rz T2 ● ● ● ● ● ● ● ● ● Rz T3 ● ● ● ● ● ● ● ● ● Fig. 3 Comparison of XRD spectra of selected sample of the substrates and rhizosphere materials in RS ( a ), CP ( b ), and CPC ( c ) among sampling times; rhizo, rhizosphere materials; T1, after 1 month; T2, after 3 months; T3, after 6 months The chemical characterization (Zn, Pb, Cd concentration, and pH values) of the substrates used during the experiment (from T0 to T3) is reported in Table 2 . All the matrices were slightly alkaline (pH 7.60–7.68) and the assessed pH values of CP and CPC are consistent with carbonates lithology of Campo Pisano area (Bechstädt and Boni 1994 ; Bacchetta et al. 2015 , 2018 ). The mine waste (CP) showed the highest values for the three metals, whereas the reference substrate (RS) showed the lowest ones. Metal content in CPC was always lower than in CP, indicating that compost addition to Campo Pisano’s mine waste has produced a diluting effect. Indeed, the concentration of metals in compost was 154 ± 6 mg kg −1 , 38.38 ± 0.01 mg kg −1 , and < DL for Zn, Pb, and Cd, respectively, markedly lower than that in CP. Moreover, the assessed values of metal content in compost were consistent with those assessed in Bacchetta et al. ( 2015 ), though it was provided by different factories. The most abundant metal in each matrix was Zn, followed by Pb and Cd. However, quite variable values in metal concentration were assessed in CP and CPC, reasonably due to the heterogeneity of the mine waste (Boni et al. 1999 ), as also reported by other authors (Cao et al. 2009 ; Bacchetta et al. 2015 , 2018 ) but they also indicate that the homogenization of the substrate prior to planting was not completely effective.\n Table 2 Chemical characterization of substrates during the experiment (mean ± SD, n = 2); RS, reference substrate; CP, Campo Pisano; CPC, Campo Pisano + compost; tot, total concentration; bf , bioavailable fraction; T0, before planting; T1, after 1 month; T2, after 3 months; T3, after 6 months Matrix RS CP CPC Zn tot (mg kg −1 ) T0 152 ± 14 11,442 ± 1018 6726 ± 164 T1 318 ± 10 9227 ± 359 6052 ± 489 T2 231 ± 14 8123 ± 790 7610 ± 215 T3 152 ± 30 7338 ± 496 7874 ± 5 bf (mg kg −1 ) T0 6.21 ± 0.50 45 ± 0.91 135 ± 10 bf (%) T0 4.08 0.39 2 Pb tot (mg kg −1 ) T0 82 ± 13 2478 ± 112 612 ± 96 T1 133 ± 2 1815 ± 188 1046 ±153 T2 102 ± 7 1741 ± 275 1575 ± 210 T3 84 ± 20 1482 ± 153 1204 ± 40 bf (mg kg −1 ) T0 2.95 ± 0.16 9.24 ± 0.18 22.24 ± 1.60 bf (%) T0 3.52 0.37 3.63 Cd tot (mg kg −1 ) T0 0.48 ± 0.13 65 ± 15 34 ± 5 T1 2.60 ± 0.06 49 ± 4 26 ± 3 T2 1.94 ± 0.01 39 ± 1 36 ± 2 T3 1.17 ± 0.01 47 ± 2 52 ± 35 bf (mg kg −1 ) T0 0.05 ± 0.01 0.55 ± 0.02 2.29 ± 0.13 bf (%) T0 10.4 0.84 6.73 pH T0 7.65 7.60 7.68 The high metal content assessed in CP and CPC confirmed the extreme metal pollution of this area (Bacchetta et al. 2012 , 2015 , 2018 ; Concas et al. 2015a ; Lai et al. 2015 ; Boi et al. 2020a ). Bacchetta et al. ( 2015 ) have also shown that this substrate is characterized by a scarce presence of organic matter as organic carbon (C org % = 1.8) and nutrient as nitrogen and phosphorous (N % = 0; P = 360 mg kg −1 ). Hence, carbon was mainly present in inorganic forms as calcite (48.0 ± 4.4 g kg −1 ) and dolomite (430.0 ± 10.1 g kg −1 ), consistently also with mineralogical characterization. As far as the matrix texture is concerned, the above mentioned work reported that it consists of fine-grained particles (< 425 μm, 86.2%), accounting the 27% of lime (< 50 μm) and 3% of silt (< 2 μm) with deleterious consequences for the potential growth and penetration of roots caused by substrate self-compaction. It is noteworthy that metal concentrations assessed in CP and CPC are far greater than the threshold contamination levels set by Italian law (Decreto Legislativo n.152 2006 ) for industrial use of soil (1500, 1000, 15 mg kg −1 for Zn, Pb, and Cd, respectively), with the exception of the Pb in CPC, which was below its threshold. The compost used for this study showed a Zn, Pb, and Cd concentration within the limits established by the European Union (Zn: 2500–4000 mg kg −1 ; Pb: 750–1200 mg kg −1 ; European Communities Council Directive 1986 ) and by the Italian law (Zn ≤ 500 mg kg −1 ; Pb ≤ 140 mg kg −1 ; Decreto legislativo n. 217 2006 ) for agricultural use. Despite the high total content observed in polluted matrices, very low concentrations of bioavailable metals ( bf in Table 2 ) were observed in all substrates (i.e., < 1% in CP) showing that only a little amount is available for plant’s roots. In CPC, a bioavailability higher than in CP tests was measured, despite several studies reported the ability of compost to reduce the bioavailability of metals (Bacchetta et al. 2012 , 2015 ). However, its effect on metal bioavailability can be contradictory (Baldantoni et al. 2010 ; Beesley and Dickinson 2010 ; Fagnano et al. 2011 ) and this ambiguity can depend on different factors, such as the origin of organic material, kind of soil, involved metals, and the formation of soluble metal-organic complexes which can increase or decrease the bioavailability (Baldantoni et al. 2010 ; Fagnano et al. 2011 ; Hattab et al. 2014 ). The increase in metal bioavailability can be ascribable to the formation of metal chelates from fragments of humic acids or low molecular weight organic compounds (Fagnano et al. 2011 ), or from dissolved and colloid organic matter (Kaschl et al. 2002 ). Hence, the choice of the suitable kind of compost is a key point for the design of an efficient phytoremediation project. Anyway, the values here reported were of the same order of magnitude observed by Concas et al. ( 2015a ) on a plot amended with compost of a previous field phytoremediation experiment (Bacchetta et al. 2012 ). The concentrations of Zn, Pb, and Cd in rhizosphere materials from T1 to T3 are reported in Fig. 4 and are consistent with those observed in the related substrates in terms of metal concentration, order of abundance of metals (Zn > Pb > Cd), and involved matrix (CP > CPC > RS). A fairly high heterogeneity was recognized also in the rhizosphere materials of CP and CPC, as already observed in the related substrates. Statistical analysis showed that metal concentration in rhizosphere was significantly different (p < 0.05) between contaminated matrices (CP and CPC) and unpolluted one (RS) for all metals during the experiment (Fig. 4 , part 1). On the contrary, metal concentration into rhizosphere of CP and CPC was statistically similar for each metal tested (except for Cd at T1) showing that the addition of compost in mine waste had no effects in terms of reduction of Zn, Pb, and Cd concentration. Moreover, the concentrations of the three metal remained similar over time in each treatment (except for Zn and Cd in CPC; Fig. 4 , part 2).\n Fig. 4 Concentration (mg kg −1 ) of Zn, Pb, and Cd (mean ± SD, n = 5) in the rhizosphere materials during the trial and related statistical analysis: 1 statistical analysis among treatments considered at a fixed time; 2 statistical analysis among time on the same matrix; different letters indicate statistically significant differences at p < 0.05; T1, after 1 month; T2, after 3 months; T3, after 6 months Soil pore waters analysis The concentrations of Zn, Pb, and Cd in soil pore waters during all the experiment are shown in Fig. 5 . Zn is the most abundant metal, followed by Pb and Cd. It is noteworthy that after the first month, the Pb concentrations were lower than the instrumental detection limit (< 0.02 mg/L). This fact may indicate that only a very smally quantity of Pb is soluble in water and it may have been depleted in the very first period of the experiment. Zinc and Cd were more concentrated in polluted matrices (CP > CPC), than in reference one (RS). The contents of Zn and Cd were significantly different (p < 0.05), with some exceptions, between contaminated matrices (CP and CPC; Fig. 5 , part 1) and the addition of compost in the mine waste seems to be able to reduce the soluble fraction available to roots, decreasing the levels of these metals close the RS values. Some significant differences (p < 0.05) in terms of Zn and Cd concentrations were found in each matrix during the experiment (except for Cd in RS; Fig. 5 , part 2). However, it is important to highlight that pore waters give information only about the soluble fraction but do not consider the solid constituents of soil, which take part of the total bioavailable pool (Adamo et al. 2013 ). The Zn and Cd values here reported are higher (approximately one order of magnitude) than those obtained by Concas et al. ( 2015b ) in the same sampling site, whereas Pb concentrations were higher than those reported by the same authors but of the same order of magnitude. However, it is necessary to clarify that they sampled in field condition wherein the water availability is obviously uncontrollable if compared with a laboratory trial. When the metal availability is addressed, the soil pore waters’ sampling gives some advantages compared with indirect methods (i.e., single or sequential extractions) because it is more rapid and if it is used in field sampling can highlight differences in metals mobility during time (i.e., water seasonality).\n Fig. 5 Zn, Pb, and Cd concentrations (mg/L) in soil pore waters during the experiment (mean ± SD; n = 5) and related statistical analysis: 1 statistical analysis of the metal content among treatments considered at a fixed time; 2 statistical analysis of the metal content at the different times (expressed in months); different letters indicate statistically significant differences at p < 0.05; 1 m, 2 m, 3 m, etc. indicate the number of months from the start of the experiment Mineralogical characterization of roots and metals accumulation in plant tissues XRD analysis showed that roots were mainly composed of amorphous cellulose and quartz and no substantial differences were observed in their mineralogical composition among the different substrates. The formation of biominerals into plant tissue is a well-known phenomenon related to physiological needs and environmental stresses (He et al. 2014 ). The precipitation of mineral phases in plant tissues in presence of heavy metal pollution was observed in several plant species, like Imperata cylindrica (L.) Raeusch. (Rodríguez et al. 2005 ) and Sarcocornia pruinosa Fuente, Rufo & Sánchez Mata (De la Fuente et al. 2018 ) as reaction to environmental stress. Biominerals were detected also in Sardinian autochthonous plant species growing in extreme metal environments like E. pithyusa subsp. cupanii (Medas et al. 2015 ), Juncus acutus L. (Fancello et al. 2019 ; Medas et al. 2019 ), Pistacia lentiscus L. (De Giudici et al. 2015 ), and Phragmites australis (Cav.) Trin. ex Steud (De Giudici et al. 2017 ). A recent study (Boi et al. 2020a ) has depicted the interaction among soil minerals and H. tyrrhenicum showing the presence of some biominerals in plant tissues (among them quartz, dolomite, and weddellite) likely as response to environmental stress. However, weddellite and dolomite were not detected in our work and more time would probably be needed to highlight an evolution in roots mineralogy. The preliminary characterization of plant metal uptake at T0 (Table 3 ) showed that roots and epigean organs mainly accumulated Zn, followed by Pb and Cd, following the same order also after planting and throughout the phytoremediation test (from T0 to T3; Fig. 6 ). The metal concentration assessed at T0 (Table 3 ) was lower than those observed in plant tissues after the planting in the contaminated substrates (CP and CPC; Fig. 6 ). Furthermore, the three metals were generally more abundant in roots than in epigean organs. As far as root accumulation is concerned, the highest concentrations of Zn, Pb, and Cd were assessed in CP followed by CPC and RS. Statistical analysis highlighted, in some cases (see Fig. 6 , part 1), significant differences (p < 0.05) between the three substrates in terms of roots uptake. In details, the accumulation of Zn was different in all substrates, and Pb was different between contaminated matrices (CP and CPC) and RS (Fig. 6 , part 1), while regarding Cd accumulation in CP was different from both RS and CPC. Based on the findings of this study, compost may act differently depending on the involved metal (see Fig. 4 ). It is a well-known fact that organic amendments can reduce the metal uptake operated by roots, as reported for plant species growing in the same kind of polluted substrates of this study (Bacchetta et al. 2012 , 2015 ). Despite the initial increase in terms of root’s metal content in the first month (from T0 to T1) of the trial, no significant differences (p > 0.05) were observed until the end of the trial (from T1 to T3) for each metal (Fig. 6 , part 2). This fact could be reasonably due to the adaption of plant specimens to the new substrate conditions.\n Table 3 Zn, Pb, and Cd concentration (mg kg −1 ) in plant tissues at T0 (mean ± SD n = 5) Zn (mg kg −1 ) Pb (mg kg −1 ) Cd (mg kg −1 ) Roots 57.18 ± 13.41 10.20 ± 4.98 5.78 ± 0.83 Epigean organs 83.77 ± 25.54 4.62 ± 0.74 0.85 ± 0.31 Fig. 6 Zn, Pb, and Cd concentration (mg kg −1 ) in roots and epigean organs (e. org) of H. tyrrhenicum during the experiment (mean ± SD; n = 5) and related statistical analysis: 1 statistical analysis of the metal content among treatments considered at a fixed time; 2 statistical analysis of the metal content in each treatment over time (from T1 to T3); different letters indicate statistically significant differences at p < 0.05; T0, before planting; T1, after 1 month; T2, after 3 months; T3, after 6 months Taking into account the accumulation of metals in epigean organs (Fig. 6 , part 1), no differences were generally observed among substrates (p > 0.05) for Zn and Cd (with the exception of T2 for each metal), whereas the accumulation of Pb was significantly higher (p < 0.05) in CP and CPC, if compared with RS. Though compost can influence the metal uptake by roots, it does not appear to modify the accumulation of metals into epigean organs. Moreover, the accumulation of the three metals did not vary along time (p > 0.05) in CP and CPC; on the contrary, the metals uptake assessed in RS varied during the experiment in epigean organs (Fig. 6 , part 2). It is noteworthy that heavy metals are generally toxic for plants and living organisms even if some of them are essential for metabolism at low concentrations (i.e., Cu, Ni, and Zn; Nagajyoti et al. 2010 ). In particular, Zn is an essential nutrient for all plants and H. tyrrhenicum can catch a certain amount as micronutrient softening the remaining through a tolerance system, whereas Pb and Cd who are toxic and hardly accumulated by plants (Kabata-Pendias 2011 ) are well accumulated in H. tyrrhenicum tissues. Moreover, a recent study has shown that Zn is mainly present in the epidermis of roots of H. tyrrhenicum , indicating an exclusion system for this metal (Boi et al. 2020a ). It is noteworthy that metal-tolerance capability is common of other plant species growing in the same area, like Cistus salviifolius L. (Jiménez et al. 2005 , 2011 , 2021 ), P. lentiscus , and S. canina subsp. bicolor (Bacchetta et al. 2012 , 2015 ; Lai et al. 2015 ; Concas et al. 2015a ). Similar behavior was observed also on other species of the genus Helichrysum , like H. italicum (Roth) G.Don subsp. italicum (Brunetti et al. 2017 ). Phytoremediation potential Biological indexes were calculated in order to evaluate the phytoremediation potential of H. tyrrhenicum . As shown in Table S1 (see Online Resource), the BCFs were calculated using both the total and the bioavailable fraction (BCF bf ) of the three metals. BCFs were < 1, with statistically significant differences (p < 0.05) among substrates. In details, BCFs assessed for Zn and Cd were different between the three substrates (with some exception, at T1 for Zn and at T3 for Cd, where CP and CPC did not show differences), whereas for Pb, the BCF was different between RS and the contaminated substrates (CP and CPC). The highest BCF was assessed for Zn and Cd in the unpolluted substrate and the lowest values assessed in CP and CPC may be due to plants ability to limit the uptake of toxic metals. Significant differences among sampling times (p < 0.05) were observed only in CP and CPC for Zn and Cd, whereas the uptake of Pb did not change (p > 0.05) in each substrate over time. The assessed BCF values showed that H. tyrrhenicum may have a different behavior in terms of roots uptake on the basis of the involved metals and substrates, indeed Zn and Cd were differently uptaken with evident differences among the three substrates and a clear effect operated by compost addition. Taking into account the bioavailable fraction, BCF bf was > 1 for every studied metal and, as observed for BCF, the behavior of the plant species towards metals significantly changed (p < 0.05) according to the involved substrate. In particular, H. tyrrhenicum behaved differently in CP and CPC, for each studied metal, showing again the effect operated by compost addition. Moreover, no significant differences among sampling times (p > 0.05) were highlighted throughout the trial for every substrate, except for Zn in CPC. The assessed BCF and BCF bf are consistent with those assesed in previous studies on this taxon (Cao et al. 2004 ; Bacchetta et al. 2017 , 2018 ). Moreover, the BCF values recorded in this study were similar than those measured in Dittrichia viscosa (L.) Greuter subsp. viscosa (Jiménez et al. 2021 ), showing a similar root’s accumulation capability between this Asteraceae member. BAC values reported in Table S2 (see Online Resource) were always < 1, with the highest values for Zn and Cd, especially in RS. Statistically significant differences among treatments (p < 0.05) were highlighted for Zn and Cd wherein the uptake in the areal organs was different between RS and the contaminated substrates (CP/CPC); on the contrary, no statistically significant differences (p > 0.05) between treatments were observed concerning the Pb’s BAC (except for T3). Significant differences among sampling times were observed (p < 0.05) only in the unpolluted substrate for all metals. As far as the bioavailable fraction is concerned, BAC bf was generally > 1 for each metal, and in particular, Zn showed the highest values, followed by Cd and Pb: RS showed the highest values for each metal and it is generally followed by CP and CPC. Statistical analysis highlighted significant differences (p < 0.05) among treatments for Zn and Cd, while no differences (p > 0.05) were observed for Pb (with the exception at T3). In particular, the uptake of bioavailable Zn and Cd in epigean organs was different between polluted matrices (CP/CPC) and RS. Moreover, significant differences among sampling times (p < 0.05) were generally observed in RS for Zn and Pb, whereas none (p > 0.05) was observed in CP and CPC for each metal. Hence, H. tyrrhenicum is able to uptake only a little concentration of metals into epigean organs as pointed out by BAC and BAC bf and the implementation of compost did not influence the uptake. Moreover, it is evident that the uptake into epigean organs did not vary in CP and CPC during all the trial, highlighting that H. tyrrhenicum catch only a certain amount of metals and showing again the potential presence of such a protective mechanism. The assessed BAC values (< 1) are consistent with other species that grow on polluted substrates as D. viscosa subsp. viscosa , C. salviifolius , and E. pithyusa subsp. cupanii (Jiménez et al. 2021 ). The measured TF index, as reported in Table S3 (see Online Resource), was always < 1. The highest values were observed for Pb and Cd followed by Zn. No significant differences (p > 0.05) were observed among treatments for every studied metal (except for Zn in T2) as well as among sampling times (p > 0.05) in each substrate. Hence, when the translocation of metals from roots to epigean organs is considered, H. tyrrhenicum showed the same behavior in all substrates and during all the trial, showing also in this case the potential presence of a protective mechanism acting to limit metal translocation. Previous study carried out on this plant species (Cao et al. 2004 ; Bacchetta et al. 2017 , 2018 ; Boi et al. 2020b ) indicated TF values > 1. However, it can be noted that plants used in the studies of Bacchetta et al. ( 2017 , 2018 ) were spontaneously grown in the Campo Pisano mine site, while in the study of Cao et al. ( 2004 ), a different substrate was used. As far as a comparison with other species is concerned, previous studies carried out on C. salviifolius , S. canina subsp. bicolor , and D. viscosa subsp. viscosa (Lai et al. 2015 ; Jiménez et al. 2021 ) showed TF higher than H. tyrrhenicum . Otherwise, in the study of Brunetti et al. ( 2017 ), H. italicum subsp. italicum have shown the capability to accumulate metals mainly into roots. Plant growth and survival The highest plants survival percentage was recorded in RS (95%), followed by CPC (90%) and CP (80%), confirming again the great adaptability of H. tyrrhenicum to highly metal-polluted substrates, also without compost. Its addition to mine waste had a positive effect on survival, helping the specimens in their development. Comparing these survival data with the phytoremediation experiment carried out by Bacchetta et al. ( 2015 ), a higher survival percentage of H. tyrrhenicum in the same kind of matrix was recognized. As far as the growth of plant is concerned, Table 4 and Fig. 7 report the values of the biometric parameters measured on five specimens at the beginning of the phytoremediation experiment (T0) and their evolution assessed during the experiment, respectively. Statistical analysis showed no differences between treatments (p > 0.05) in terms of length of roots and epigean organs. In particular, roots length remained constant in the first month (from T0 to T1), but a significant length reduction (p < 0.05) was observed in each treatment (Fig. 7 , part 2) during the rest of the experiment, whereas an elongation of epigean organs was generally observed from T0 to T1, but no statistically significant growth (p > 0.05) was observed from T1 to T3 in each substrate. Taking into account the diameter of the stem, no statistically significant differences (p > 0.05) were assessed between treatments (Fig. 7 , part 1), except for T1 where the diameter was significantly higher in RS than CP/CPC. From T1 to T3, significant differences (p < 0.05) were observed in RS, corresponding with an enlargement, whereas in CP and CPC, the diameter remained constant (Fig. 7 , part 2) during the experiment. Anyway, the diameter seems not to change if compared with measurement at T0 (Table 4 ). As for the roots and epigean biomass, no statistically significant differences (p > 0.05) between treatments were observed (with the exception at T2 for epigean organs biomass; Fig. 7 , part 1). In details, the biomass of the roots did not change after 1 month from the beginning of the experiment (from T0 to T1), but a statistically significant decrease (p < 0.05) in weight was observed from T1 to T3 in all the treatments (Fig. 7 ). Conversely, the epigean organs biomass was constant (p > 0.05) throughout the experiment (p > 0.05) in all substrates. In addition, compost seems not to influence the growth of the roots and epigean organs, diameter of the stem, and biomass in H. tyrrhenicum . The metal pollution seems to operate mainly in terms of enlargement of stems; indeed, a significant enlargement was recorded only in RS, probably due to the lowest metal content: it is a well-known fact that metals can inhibit the water uptake (Kranner and Colville 2011 ) and as a consequence the enlargement of the stem.\n Table 4 Biometric parameters of H. tyrrhenicum at the beginning of the phytoremediation experiment (T0; mean ± SD; n = 5) Biometric parameters Epigean organs length 20.60 ± 6.58 cm Roots length 15.60 ± 6.58 cm Stem diameter 4.14 ± 0.36 mm Roots biomass 4.14 ± 0.36 g Epigean organs biomass 6.62 ± 2.05 g Fig. 7 Biometric parameters of H. tyrrhenicum during the trial (mean ± SD; n = 5) and statistical analysis: 1 statistical analysis of biometric parameters among treatments considered at a fixed time; 2 statistical analysis of biometric parameters in each treatment over time (from T1 to T3); different letters indicate statistically significant differences at p < 0.05; T0, before planting; T1, after 1 month; T2, after 3 months; T3, after 6 months"
} | 9,513 |
34045570 | PMC8160205 | pmc | 7,240 | {
"abstract": "Rhodolith beds built by free-living coralline algae are important ecosystems for marine biodiversity and carbonate production. Yet, our mechanistic understanding regarding rhodolith physiology and its drivers is still limited. Using three rhodolith species with different branching morphologies, we investigated the role of morphology in species’ physiology and the implications for their susceptibility to ocean acidification (OA). For this, we determined the effects of thallus topography on diffusive boundary layer (DBL) thickness, the associated microscale oxygen and pH dynamics and their relationship with species’ metabolic and light and dark calcification rates, as well as species’ responses to short-term OA exposure. Our results show that rhodolith branching creates low-flow microenvironments that exhibit increasing DBL thickness with increasing branch length. This, together with species’ metabolic rates, determined the light-dependent pH dynamics at the algal surface, which in turn dictated species’ calcification rates. While these differences did not translate in species-specific responses to short-term OA exposure, the differences in the magnitude of diurnal pH fluctuations (~ 0.1–1.2 pH units) between species suggest potential differences in phenotypic plasticity to OA that may result in different susceptibilities to long-term OA exposure, supporting the general view that species’ ecomechanical characteristics must be considered for predicting OA responses.",
"introduction": "Introduction Marine coastal ecosystems formed by free-living coralline algae that cover 30–100% of the seafloor, so-called rhodolith or maërl beds, are distributed worldwide and have long been known to be important biodiversity hot-spots and major carbonate-producing ecosystems 1 , 2 . Because of their importance for biodiversity, in several regions of the world they are considered critical habitats for marine conservation and protected by a range of directives, regulations and conventions 2 . Also, more recently it has been pointed out that these habitats may figure more prominently in the global CaCO 3 production than currently recognized 3 – 5 . This seems reasonable, when considering that in the Southwestern Atlantic alone rhodolith beds extend almost continuously for over 4.000 km along the Brazilian coast (2°N to 27°S) 6 , covering an estimated area of 230,000 km 2 7 .\n Like many other marine ecosystems, these habitats are currently impacted by environmental pressures associated with climate change, either due to direct effects on rhodoliths or indirect effects due to altered species interactions 8 – 15 . As calcifying organisms, rhodoliths are especially prone to be negatively impacted by ocean acidification (OA) and the little evidence so far available suggests a large variability in their responses 16 . In combination with our limited mechanistic understanding of their calcification mechanism, its regulation and potential species-specific differences, this makes it difficult to anticipate potential future impacts on these organisms and hence, the ecosystems they build—a point widely stressed as an important tool to improve predictive power of OA studies 17 – 20 . Current evidence shows that coralline algae exhibit species-specific differences in their ability to physiologically control the calcification process, which in turn seems to dictate their resistance to OA 21 – 26 . Previous studies demonstrated that this is related to differences in the species’ capacity to elevate the pH of the calcifying fluid 21 – 27 and within the diffusive boundary layer (DBL) at the algal surface 24 , 28 . The build-up of a pH gradient within the DBL, which exhibits a linear relationship with the pH of the calcifying fluid 24 , and its magnitude are determined by species’ metabolic activity, DBL thickness and seawater pH 24 , 29 – 32 . Generally, increased DBL thickness favors larger pH gradients, with the former being inversely related to water flow velocity 24 , 29 , 30 . Water flow velocity can vary largely in the natural environment of the algae, but at a microscale, it also varies depending on the morphology of the algae (i.e. thallus topography) that can create low-flow microenvironments 24 , 33 . While the effects of water flow velocity on coralline algal calcification and species’ OA responses have been demonstrated in previous studies 24 , 29 , 30 , so far the role of morphology for species’ calcification rates and the implications for the species’ OA responses are not well understood. Rhodoliths are a morphologically highly variable group 34 , 35 and to date, only the study by Hurd et al. 29 on the temperate rhodolith Sporolithon durum has provided evidence of the effects of rhodolith surface topography on DBL thickness and microscale pH dynamics at the thallus surface. It showed that the “bumpy” thallus topography creates microenvironments with thicker DBLs in-between “bumps”, where thallus surface pH is greatly increased, when compared to the tip of the “bumps”. This suggests an important role for morphology in these calcifiers, as the build-up of a pH gradient has a strong effect on the calcification process 24 , 30 . Here we hypothesize that besides inherent physiological differences among rhodolith species, morphology plays an important role in determining species’ capacity to modulate thallus surface pH and hence, calcification rates, which in turn may lead to potential differences in their responses to OA. To test this hypothesis, the goals of this study were to determine (1) whether there is a relationship between rhodolith branching morphology, DBL thickness and associated concentration gradients at the thallus surface, (2) whether the resulting algal surface pH dynamics have an influence in species’ light- and dark calcification rates, and (3) whether morphology-associated diurnal pH fluctuations affect the species’ response to short-term (shock-) exposure to lower seawater pH conditions.",
"discussion": "Discussion This study expands our understanding regarding the factors that regulate rhodolith calcification, showing that morphology and rhodolith species’ metabolism are critical factors determining diurnal fluctuations in thallus surface pH and consequently, species’ light and dark calcification rates (Fig. 7 ). While the species–specific magnitudes of daily pH fluctuations did not result in differences in species’ responses to short-term OA exposure, they suggest potential differences in phenotypic plasticity and hence, susceptibilities to long-term OA exposure. Figure 7 Schematic overview of the relationship between the pH within the DBL (pH DBL ) and coralline species’ calcification rates, as well as the biotic (black) and abiotic factors (grey) affecting directly (thick arrows) and indirectly (thin arrows) the pH DBL and hence, calcification. The influence of species-specific differences in the calcification mechanism cannot be excluded, but further information is required to infer about the effect on species’ calcification rates. The morphological differences of the studied rhodolith species are mainly related to the size of the protuberances (or branches), resulting in distinguished algal thallus surface topographies (see Fig. 1 ). Those were found to exhibit a strong effect on the DBL, showing an increased DBL thickness in-between protuberances, which is consistent with previous studies in the temperate rhodolith Sporolithon durum 29 and in undulated Macrocystis pyrifera blades (undulation apex vs. within undulation) 33 . As shown in the latter study, topographic depressions, like those created by protuberances in rhodoliths, create low-flow microhabitats that lead to an increase of DBL thickness. This explains the species-specific increase in DBL thickness in-between protuberances as a direct result of the species’ protuberance lengths (Fig. 3 ). This effect of thallus topography on DBL thickness and in turn the concentration gradients within is a feature well documented in macroalgae 29 , 33 . However, our data also suggest that DBL thickness was not the sole factor defining the strength of the associated biochemical gradients. Lithophyllum atlanticum , which ranked second in protuberance length and DBL thickness, expressed the largest oxygen gradient and also the strongest pH gradient within the DBL (pH DBL ). This was consistent with the higher metabolic rates of this species, compared to the others (Table 2 ), as the oxygen gradient within the DBL is a result of algal photosynthesis, respiration and DBL thickness. Further, it is linked directly to the pH dynamics, due to the uptake and release of dissolved inorganic carbon in light and dark 32 , 39 , 40 . Altogether, this supports the general notion that both oxygen and pH gradients not only depend on DBL thickness, but also on the species metabolic activity 30 , 40 . Successively, this seemed to create a feedback loop, having a direct influence on the rhodolith calcification process, as shown by the significant direct relationship between the mean pH at the rhodolith thallus surface and the calcification rates under both light and dark conditions (Fig. 5 ). Similarly, a relationship between the pH dynamics at the algal thallus surface and species’ calcification rate had also been reported recently in Sporolithon durum 24 . The assertion appears particularly reasonable when considering that calcification in coralline algae takes place close to the thallus surface 41 and that the pH at the calcification site shows a direct linear relationship with the pH DBL 24 . Interestingly and contrary to what would be expected, there was the slightly higher mean ΔpH DBL in darkness in L. crispatum (0.0009), which resulted from higher ΔpH DBL at the protuberance tips (0.04), compared to the bulk seawater, while at the bases the ΔpH DBL was lower (− 0.036). We do not have an explanation for the higher ΔpH DBL at the tips in darkness, though a recent study on Artic rhodoliths reported similar findings 28 . The authors suggested that it might be due to a reduction in respiratory CO 2 release because of light-independent carbon fixation. The rhodolith species exhibited significant differences in their metabolic rates (Table 2 ), which might be an inherent feature, but it could also be argued that these differences relate, at least partially, to light attenuation caused by the protuberances, increasing with protuberance length and resulting in less light reaching the respective bases, as suggested by Burdett et al. 42 . These authors found higher effective quantum yields and lower non-photochemical quenching at the protuberance bases, compared to the tips, in the rhodolith Lithothamnion glaciale , indicating lower light levels at the base. Yet, in our study, no differences in the light intensities measured with a light microsensor at the protuberance tips and their respective bases of the species were found (see Supplementary Fig. S2 online). A reason for this inconsistency could be the high variability between our measurements and insufficient replication, as M. erubescens showed a high variability and in some measurements the light intensity at the protuberance base was ~ 20% lower compared to the tip, consistent with the found significant differences in gross oxygen flux between base and tip under light conditions in this species. Hence, we cannot exclude potential differences in light availability created by the protuberances that might affect species photosynthesis. Also noteworthy were the differences in oxygen fluxes under light conditions, when comparing protuberance tips and bases. Lithothamnion crispatum and M. erubescens exhibited lower oxygen fluxes at the bases, compared to the respective protuberance tips (Fig. 4 a, c), while those in L. atlanticum were not different between tips and bases, but significantly higher than in the other two species (Table 1 ). This observation, together with the higher daily pH fluctuation at the algal thallus surface expressed by L. atlanticum (~ 0.4–1.2 pH units), compared to the other two species (~ 0.1–0.6 pH units), suggest potential differences between the species regarding inorganic carbon use strategies and/or carbon-concentrating mechanisms. Many studies have focused on the acclimation/adaptation potential of coralline algae to environmental changes, such as OA, in order to predict future trajectories of communities and ecosystems 25 , 43 – 46 . Yet, only a few have considered the specific role of morphology in species’ OA responses 29 , 43 . In this regard, light-dependent changes in pH DBL at the algal thallus surface, reported previously in other coralline algae 28 – 31 , 47 , have led to the ongoing discussion about whether these changes may ameliorate OA impacts during the day and/or increase them during nighttime 22 , 29 , 30 , 40 . Thus, it could be hypothesized that the strong effect of the rhodolith morphological features on algal surface pH dynamics found in the present study might result in species-specific susceptibility to OA. However, the species’ response to the short-term OA exposure did not show a direct relationship with their morphology and associated DBL dynamics. Yet, the clear differences in the magnitude of daily pH fluctuations at the algal surface the rhodolith species are regularly exposed to indicate that their responses under longer-term exposure to lower seawater pH will most likely differ. This is based on previous suggestions and evidence that species which encounter habitually strong daily pH fluctuations may either be better adapted to cope with decreasing mean seawater pH due to higher phenotypic plasticity 29 , 48 , 49 or be at a disadvantage 47 , 50 . Furthermore, the lower pH condition did not have any effects on rhodolith light calcification, but induced a strong decline in dark calcification in all species (Fig. 6 d–f). The former is not surprising as the short-term experiment did not allow accounting for potential species-specific acclimation capacities that might be expressed under longer term exposure 12 , 21 , 23 , 46 . Dark calcification, on the other hand, is considered as a metabolically independent process, resulting from a combination of purely physical CaCO 3 precipitation and belated biological activity after a passage from light to dark 51 – 53 . Thus, it can be assumed that it is independent from acclimation processes and hence, the here found increase in night-time CaCO 3 dissolution in all three species will most likely be sustained also under longer-term OA conditions. This agrees with previous findings, demonstrating that the main OA effect on calcareous macroalgae is the impact on nighttime rather than on light calcification, often resulting in dissolution and decreased daily net calcification 32 , 43 , 45 , 54 – 58 . In this context, other studies have also shown that in some species the increased dissolution under long-term OA exposure is off-set by enhanced light calcification 55 , 56 . Longer-term OA experiments are required to determine if this will be the case in the studied rhodolith species and to further understand the potentially beneficial or detrimental effects of the morphologically related daily pH fluctuations, in order to accurately predict and quantify OA impacts on these organisms and the ecosystems they build. In summary, understanding the importance of physical and biological factors for coralline algal physiology is key for comprehending the large variability of reported species’ metabolic and calcification rates and to accurately predict the effects of environmental changes, i.e. those related to the ongoing climate change, on different species and the habitats they build or are part of. With the present study, we contribute to a greater understanding of the factors that regulate rhodolith calcification, showing that in addition to species’ physiological traits and environmental drivers, rhodolith morphology represents a critical factor determining daily pH dynamics at the algal surface and consequently, species’ light, dark and net calcification rates (Fig. 7 ) and potentially their responses to OA. In fact, morphology might be one of the key factors for the commonly found species-specific differences among free-living coralline algae, not only concerning light calcification rates, but especially CaCO 3 precipitation in the dark (e.g., lower to negative calcification rates in darkness) 8 , 12 , 43 , 53 , 55 , 59 ."
} | 4,126 |
37765573 | PMC10534511 | pmc | 7,241 | {
"abstract": "This study presents, for the first time, a comprehensive investigation of the influence of pre- and post-fabrication parameters for the electroactive properties of electrospun chitosan/PVA-based micro- and nanofibers. Chitosan/PVA fibers were fabricated using electrospinning, characterized, and tested as electroactive materials. Solutions with different acetic acid contents (50, 60, 70, and 80 v/v %) were used, and the rheological properties of the solutions were analyzed. Characterization techniques, such as rheology, conductivity, optical microscopy, a thermogravimetric analysis, differential scanning calorimetry, a tensile test, and FT-IR spectroscopy, were utilized. Fiber mats from the various solutions were thermally treated, and their electroactive behavior was examined under a constant electric potential (10 V) at different pHs (2–13). The results showed that fibers electrospun from 80% acetic acid had a lower electroactive response and dissolved quickly. However, thermal treatment improved the stability and electroactive response of all fiber samples, particularly the ones spun with 80% acetic acid, which exhibited a significant increase in speed displacement from 0 cm −1 (non-thermally treated) to 1.372 cm −1 (thermally treated) at a pH of 3. This study sheds light on the influence of pre- and post-fabrication parameters on the electroactive properties of chitosan/PVA fibers, offering valuable insights for the development of electroactive materials in various applications.",
"conclusion": "4. Conclusions In conclusion, this study successfully electrospun micro- and nano-scale fibers from chitosan/PVA solutions with varying concentrations of acetic acid (50, 60, 70, and 80 v/v%). The rheological properties of the polymeric solutions were notably affected by the acetic acid concentration. An increased acetic acid concentration led to a higher viscosity (from 6157 mPa·s for AA50 to 8747.3 mPa·s for AA80) and reduced conductivity (from 2852.6 µS cm −1 to 999.6 cm −1 for AA50 and AA80, respectively). The fiber morphology correlated with the rheological properties, with the fiber diameter increasing as the acetic acid concentration rose, ranging from 0.482 to 0.793 µm. The thermal treatment at 70 °C enhanced the fiber thermal stability, as shown by the thermogravimetric analysis (TG). The mechanical properties were influenced by the acetic acid concentration. The fiber samples with 50%, 60%, and 70% acetic acid concentrations exhibited curves resembling hard/tough plastic polymers, while the sample with 80% acetic acid demonstrated the characteristics of a brittle polymer. The observed changes in mechanical properties were attributed to the plasticizing effect of the large acetate ion. Electroactive properties were also assessed. The untreated fibers showed a decreased electroactive response in more acidic environments as the acetic acid concentration increased. The fibers electrospun from solutions with 50% acetic acid reached a maximum displacement of 0.7728 mm s −1 at a pH of 3, while those electrospun from solutions with 80% acetic acid dissolved at the same pH. The thermally treated fibers exhibited an improved electroactive response, with a heightened sensitivity to basic mediums (pH > 7). All the samples exhibited two displacement peaks, one at a pH of 3 and another around a pH of 9–10. The FT-IR spectra revealed increased intermolecular hydrogen bonding due to the thermal treatment. This could be related to the higher intermolecular polymer chain interaction as a result of a physical crosslinking effect induced by the thermal treatment. Overall, this study provides insights into how acetic acid concentration and thermal treatment impact the mechanical, thermal, and electroactive properties of chitosan/PVA fibers. These findings advance our understanding of the development of electro-responsive soft actuators. Additionally, comprehending how pre- and post-fabrication parameters impact the electroactive properties of a material allows for the development of soft actuators with tunable electroactive responses without altering the material’s composition. This approach has promising implications, particularly in the field of medical applications, in which the development of adaptive soft grippers could revolutionize minimally invasive procedures.",
"introduction": "1. Introduction Soft actuators are systems inspired by muscles in human and animals, which present characteristics such as agility, environmental adaptability, flexibility, and multifunctionality, among others [ 1 ]. Soft actuators have potential uses in soft robotics, artificial muscles, wearable devices, biomedicine, soft grippers, and so forth [ 2 , 3 , 4 , 5 ]. Mainly, soft actuators are fabricated using soft materials [ 6 ] and can be driven by different physical and chemical stimuli (heat, magnetic fields, humidity, fluid pressure, pH variations, electric fields, and so forth) [ 7 , 8 , 9 , 10 ]. Among these, electrically driven materials, such as electroactive polymers (EAPs), have gained attention in that they resemble natural muscles [ 11 , 12 ]. According to the actuation mechanism, EAPs are grouped into two broad categories. When coulomb forces govern the actuation mechanism, they are referred to as electronic EAPs. Within this group, we can find ferroelectric polymers, electrostrictive graft elastomers, dielectric elastomers, liquid crystal elastomers, and electroviscoelastic elastomers. On the other hand, when the ions’ diffusion or mobility causes the actuation mechanism, they are referred to as ionic EAPs. Herein, we can find conductive polymers, ionic polymer gels, and ionic polymer–metal composites (IPMCs) [ 13 , 14 ]. The electroactive properties of a polymer can be described as a mechanical response to an electrical stimulus, or in other words, the ability of a polymer to deform by changing its shape or size when excited by electrical potentials [ 11 , 12 , 13 , 14 ]. In recent years, extensive scientific research has focused on the potential application of micro- and nanofibers as electroresponsive hydrogels. Despite being a relatively new material, their convergence of favorable properties, such as flexibility, elasticity, substantial porosity, adjustability, and other characteristics, has shown great promise in the context of electroactive polymers (EAPs) [ 15 ]. Nanofibers are highly desirable as electroresponsive materials due to their unique morphology, which offers several advantages. One significant advantage is the shortened response time achieved by reducing the size of the hydrogel, as the actuation rate is inversely proportional to its size. Additionally, nanofibers exhibit a one-dimensional structure with a significant surface-to-volume ratio, leading to the greater ionization of functional groups when interacting with the surrounding medium. This is primarily attributed to the larger surface area of each individual fiber. Furthermore, the high porosity of nanofiber mats effectively facilitates the diffusion of free ions, allowing them to migrate between a material’s interior and exterior. This enhanced ion diffusion contributes to greater material deformation and improved mechanical strength compared to that of hydrogels [ 16 ]. The demand for the development of polymeric micro- and nanofibers has increased over the past decade, due to their potential use in various applications, such as protective clothing, tissue engineering, drug delivery systems, energy storage, filtration, functional materials, sensors, and soft actuators [ 17 ]. For the production of micro- and nanoscale fibers, several techniques have been developed. Based on the fiber formation process, these techniques can be divided into two groups. The first of these, in which the fibers’ production is due to electrostatic forces, are called electrospinning techniques [ 18 ]. The second group of techniques use mechanical forces for fiber formation, such as drawing, template synthesis, phase separation, and so forth [ 19 ]. Electrospinning is a mechanical and electrical technique that allows for the production of submicrometric fibers [ 20 , 21 ]. This technique makes it possible to fabricate hierarchical structures that can emulate the fibrous structure of biological muscles, which is a desired characteristic when designing soft actuators [ 22 ]. Electrospinning involves applying a high voltage to a polymer solution as it is fed through a capillary or needle. This process results in the formation of a conical jet, known as a Taylor cone, due to the accumulation of an electric charge at the tip of the droplet. Subsequently, the thin yarn is gathered onto a grounded electrode plate. Electrospun fiber morphology is influenced by numerous factors, including technical parameters such as flow, voltage, and needle-to-collector distance; solution properties, such as conductivity, viscosity, and polymer concentration; and environmental conditions, such as temperature and humidity [ 23 ]. In recent decades, the use of biopolymers as electroactive materials has been widely explored. Chitosan, cellulose, starch, and alginate, among others, have shown good electroactive properties due to the presence of polar groups that are easily polarizable [ 24 , 25 ]. Chitosan is a copolymer composed of N -acetyl- d -glucosamine and d -glucosamine. The degree of deacetylation (DDA) achieved during the conversion of chitin to chitosan determines the ratio of d -glucosamine in the copolymer. A variety of organic and inorganic acids have been found to be effective solvents for chitosan, such as malic acid, l -glutamic acid, acetic acid, formic acid, and hydrochloric acid, among others [ 26 , 27 ]. Acetic acid and trifluoroacetic acid (TFA) are the most commonly used solvents for electrospinning chitosan, especially for achieving uniform fiber formation. Additionally, acetic acid influences the rheological properties of chitosan solutions, which affect the morphology of electrospun fibers [ 28 , 29 ]. Chitosan possesses amino groups and hydroxyl groups along its backbone, resulting in its polycationic nature. This polycationic property has led to extensive research on chitosan for the advancement of electroresponsive materials [ 30 , 31 ]. Zolfagharian A. et al. [ 32 ] developed a 3D-printed, soft, biodegradable chitosan-based hydrogel actuator. Shamsudeen K. et al. [ 33 ] investigated anionic and cationic polyelectrolyte hydrogels’ electroactive response. Partially hydrolyzed polyacrylamide/PVA was synthesized as an anionic hydrogel; meanwhile, chitosan/PVA was synthesized as a cationic hydrogel. The cationic and anionic gels exhibited contrasting behaviors which have various applications in areas such as medicine and robotics. The formation of pure chitosan fibers by the electrospinning process has proven to be a challenging task due to the high electrostatic repulsion forces between chitosan chains. To overcome this problem, chitosan is often mixed with a highly electrospinnable polymer, such as poly(ethylene oxide) or poly(vinyl alcohol), which neutralizes the repulsive forces [ 34 , 35 , 36 ]. The latter is the most used polymer to blend with chitosan. Polyvinyl alcohol (PVA) is a partially crystalline polymer with a linear structure. This consists of a carbon backbone and —OH groups. PVA possesses several significant characteristics, such as its wide availability, solubility in water, excellent film-forming properties, and thermal stability, among others [ 37 , 38 ]. In our previous study, we made a significant contribution by introducing the utilization of biopolymer nanofibers for the development of a fibrous, electroresponsive soft actuator based on electrospun chitosan/PVA nanofibers [ 39 ]. To the best of the authors’ knowledge, this innovative approach had not been previously reported in the scientific literature. In our previous work, we aimed to study the influence that chitosan has on the electroactive behavior of materials. For this aim, chitosan/PVA fibers were fabricated with different chitosan concentrations and dissolved in a 40% aqueous acetic acid solution. The obtained results showed that the fibers with a higher content of chitosan had faster bending displacement in response to an electric stimulus. Other studies have used nanofibers for the development of electroresponsive soft actuators as well, such as those reported by Riccardo D’Anniballe et al. [ 40 ], Asai H. et al. [ 41 ], Seyed V. et al. [ 42 ], Miranda D. et al. [ 43 ], and Ismail Y. et al. [ 44 ]. Many of these works focus on how the electroactive behavior of fibrous soft actuators can be influenced by the type of polymers utilized, the differential potential, the conductivity of the material, and the electrolyte solution. Nevertheless, as of the writing of this article and to the best of the authors’ knowledge, no previous reports have explored the influence of pre- and post-fabrication parameters on the electroresponsive behavior of fibrous soft actuators based on electrospun fibers. This knowledge gap highlights the novelty and significance of our current study as we aim to address this research gap and shed light on the crucial factors that impact the electroresponse of electrospun fibers.",
"discussion": "3. Results and Discussion 3.1. Rheological Properties and Conductivity of Polymeric Solutions During the electrospinning process, the rheological properties of polymer solutions are critical to fiber formation. Fiber morphology is significantly influenced by parameters such as viscosity and conductivity. Section 3.2 (Morphology and Diameter Distribution of Chitosan/PVA Nanofibers) demonstrates this correlation effectively. All the measured solution exhibited non-Newtonian behavior. Shear thinning or pseudoplastic flow was observed, possibly due to the ionized —NH 2 groups in chitosan [ 47 , 48 , 49 ]. The viscosity is notably increased as a result of the hydrogen bonding between chitosan’s —NH 2 and —OH groups with the —OH groups of PVA, polymer–polymer entanglement, and randomly orientated polymeric chains. As the shear rate goes higher, the viscosity decreases, which can be attributed to the exposed —NH 3 groups affecting the electrostatic and steric repulsion and the realigned polymeric structure [ 50 , 51 ]. Firstly, the viscosity of polymeric solutions obtained by the dissolution of chitosan in different concentrations of acetic acid (50, 60, 70, and 80%) was studied. Figure 2 shows the viscosity dependency in relation to the concentration of acetic acid in chitosan solutions CsAA50, CsAA60, CsAA70, and CsAA80 at room temperature. From the obtained results, it is noticeable that, as the acetic acid concentration increases, the chitosan solution’s viscosity increases. The apparent viscosity measured at γ ˙ = 50 s −1 increased from 986.93 mPa·s (CsAA50) to 2014.2 mPa·s (CsAA70). The protonation of the amino groups may explain this behavior. In an acidic medium, —NH 2 gets protonated (—NH 3 + ), and the increase in charge density throughout the molecular chain results in the unfolding of the chain. The degree of entanglement and intermolecular interaction between the polymer chains increases as the polymer chain unfolds. Therefore, as the concentration of acetic acid is higher, more hydrogen bonds are formed between chitosan and the acetic acid, causing the viscosity of the solutions to increase. Nonetheless, it was observed that the viscosity was drastically reduced at an acetic acid concentration of 80% (1262.8 mPa·s). Afterwards, we evaluated the viscosity of polymeric solutions containing chitosan/PVA dissolved in acetic acid at various concentrations (50%, 60%, 70%, and 80%). The results clearly indicate that the inclusion of poly (vinyl alcohol) causes a significant increase in the polymer solutions’ viscosity, as depicted in Table 1 . Furthermore, as shown in Figure 2 , the viscosity of the solutions increases as the concentration of acetic acid increases. In contrast to the chitosan solutions, which showed varying viscosity trends, the chitosan/PVA solutions exhibited a consistent increase in viscosity as the concentration of acetic acid increased, ranging from 6157 mPa·s (AA50) to 8747.3 mPa·s (AA80) measured at γ ˙ = 50 s −1 . This phenomenon arises due to the entanglement of the polymer chains and the formation of numerous hydrogen bonds between the —H groups present in PVA and the ionized amino groups and —OH groups present in chitosan. The decrease in viscosity for the chitosan solution with 80% acetic acid can be explained as follows. The decrease in viscosity with the increasing electrolyte concentration can be explained by the shielding effect of counterions. Due to ionic dipole forces, acetate ions form a cascade of negatively charged particles over each chitosan molecule, creating Coulomb repulsion forces between them. This leads to a decrease in the flow resistance [ 29 , 52 ]. As it was reported by Kienzle-Sterzer et al. [ 53 ] in a solution with a high acetic acid content, another possible explanation for this behavior is that some of the protonated ions located in the chitosan structure are neutralized by the increase in acetate ions. Due to this neutralization, the flexibility of the polymer chain increases, causing the macroion domain to shrink, thus decreasing the viscosity. Another property that changes depending on the content of acetic acid in the polymeric solution is the conductivity. From the obtained results, it is possible to notice that the conductivity is affected by two factors: the addition of PVA to the chitosan solutions and the acetic acid concentration, as shown in Table 2 . The conductivity decreases with the increasing acetic acid concentration, which may be related to the increase in the density and strength of hydrogen bonds, which reduce the number of free charged groups. In addition, when PVA is combined with chitosan, the formation of both intra- and intermolecular bonds reduces the number of free charged groups in the polymer chain. 3.2. Morphology and Diameter Distribution of Chitosan/PVA Electrospun Fibers The electrospinning process was conducted under environmental conditions at 26 °C and an RH of 23%. The fibers were electrospun from solutions composed of chitosan/PVA dissolved in various concentrations of acetic acid (50%, 60%, 70%, and 80%) in an aqueous solution. In Figure 3 , the electrospun fibers are shown. From the obtained micrographs, it can be observed that the different concentrations of acetic acid used in the solutions do not affect the quality of the fiber formation process. The electrospinning process was shown to be stable for all the solutions, with a uniform fiber formation, absent of any beads or particle formation. As previously mentioned, varying the amount of acetic acid used to produce the fibers does not appear to affect the quality of the electrospun fibers. However, as shown in Table 3 , the fiber diameter was influenced by the concentration of acetic acid used. From the measured diameters, it was noticed that the diameter of the fibers obtained increased as the concentration of acetic acid in the solution increased. Due to the variation in fiber diameter as a function of the acetic acid concentration, it is possible to electrospin fibers from the nano- to the microscale while maintaining a constant polymer content. The variation in the fibers’ diameter can be explained as follows. Chitosan is cationic polymer with amino groups attached to its backbone. In acidic solutions (with a pH of < 6), amino groups (—NH 2 ) are protonated, forming —NH 3 ions. Thus, they generate charge repulsions causing chitosan’s chain to expand. As the acetic acid content increases, the pH in the solution decreases, hence increasing the protonation of amino groups in chitosan. In addition, intermolecular hydrogen bonding allows the protonated amino groups to interact with the —OH groups of PVA [ 54 , 55 ]. The rheological properties of the solutions change as a result of the influence of acetic acid (a more detailed explanation is given in Section 3.1 ). The polymeric solution containing acetic acid at 50% had a lower viscosity and higher conductivity, producing fibers with a thinner diameter (~0.482 µm). On the other hand, fibers obtained from the solution containing acetic acid at 80% were thicker (~0.793 µm), as a result of its higher viscosity and lower conductivity. Similar results have been reported by Cheng T. et al. [ 28 ]. 3.3. Thermogravimetric and Differential Scanning Calorimetry Analysis As was previously mentioned, the aim of this paper is to examine how the electroactive properties of electrospun fibers can be affected by thermal treatments. Nonetheless, thermal treatments can also influence the thermal properties of the fibers. Chitosan/PVA fibers mats were thermally treated, by placing the samples in a drying oven for 24 h at t = 70 °C. Table 4 and Table 5 summarize the thermogravimetric analysis’s results. The thermogravimetric thermographs of non-thermally treated chitosan/PVA fibers exhibited a weight loss profile at three temperature stages, as shown in Figure 4 a. The chitosan and PVA had weight losses at two stages. Regarding the PVA, the initial mass loss was in the temperature range of 51–134 °C, which can be attributed to the evaporation of the moisture content (~4%). Subsequently, a second weight reduction took place between 157 and 450 °C, indicating the thermal degradation of the PVA (~90.52%). In the case of the chitosan, the initial decrease in weight occurred within the temperature range of 35–118 °C, related to the evaporation of the moisture content (~5%). Subsequently, a second weight reduction is observed between 182 and 400 °C, indicating the thermal degradation and deacetylation of the chitosan (~49.68%). Except for sample F-AA50, the mass loss of the non-thermally treated chitosan/PVA fibers happened in three stages [ 56 ]. The derivative thermogravimetric analysis curves showed that the first mass loss happens in a temperature range of 50–160 °C, associated with moisture and residual solvent evaporation. It is possible to observe from the DTG curves that, at the stage of the first mass loss, two peaks are formed. The presence of a second peak in this stage is due to acetic acid residues, which have a boiling point of 118 °C. The second mass loss was observed for F-AA50 in the range of 140–370 °C, and for samples F-AA60, F-AA70, and F-AA80 in the range of 180–370 °C. The second mass loss is related to the chitosan/PVA complex’s thermal destruction. A third mass loss was observable in samples F-AA60, F-AA70, and F-AA80 in the range of 375–500 °C, which could be related to the PVA byproducts and residuals of poly (vinyl acetate) degradation, which has a decomposition temperature of ~400 °C, present in PVA chains [ 57 ]. Regarding the samples that were thermally treated, significant changes were observed in their DTG curves, as shown in Figure 5 b–e for the dried samples F-AA50, F-AA60, F-AA70, and F-AA80, respectively. Unlike the untreated samples, the thermally treated samples exhibited two mass loss stages. In the first mass loss stage, the thermally treated samples F-AA70 and F-AA80 showed the same two peaks (t = 60 °C and t = 125 °C) related to water and acetic acid residue evaporation. In contrast, the thermally treated sample F-AA60 did not display any peaks, and the thermally treated sample F-AA50 showed one peak related to water evaporation (t = 52 °C). Furthermore, for all the thermally treated samples, a peak shifting to a higher temperature in the second mass loss stage was observed. Sample F-AA50 exhibited the most noticeable shift, from 220 °C (for the non-thermally treated sample) to 231 °C (for the thermally treated sample). The obtained results are summarized in Table 5 . From the results obtained, it is noticeable that the thermal stability of the electrospun fibers decreased compared to the thermal stability of the pure chitosan and PVA ( Table 4 ). Moreover, the thermal stability for samples F-AA60, F-AA70, and F-AA80 are relatively similar: 264, 264.8, and 266 °C, respectively. Sample F-AA50 shows a considerable decrease in its thermal stability to 220 °C in comparison to that of the other samples. Firstly, chitosan/PVA fibers have a lower thermal stability as a result of the polymer–polymer interaction. As a result, the structure of amorphous chitosan, which is dispersed along PVA chains, give rise to defects in the crystalline phase of PVA. This hinders the formation of crystalline regions. Therefore, the thermal energy required to break hydrogen bonds and melt free PVA chains is lower, thereby lowering the melting point of the chitosan/PVA. Moreover, the TG results for the thermally treated electrospun fiber samples showed an increase in their thermal stability in comparison to those of the fibers samples that were not thermally treated. The increase in thermal stability can be attributed to the fact that the heat treatment facilitates the mobility of the polymer chains, allowing the amorphous regions of the chains to align and fold into crystalline regions [ 58 , 59 ]; hence, the thermal stability of the material increases as well. In addition, the increase in the thermal stability can be related to a physical crosslinking caused by the remotion of water molecules, allowing for higher intermolecular interactions. The DSC thermographs for the chitosan and PVA are shown in Figure 6 a. The chitosan’s thermograph showed an endothermic followed by an exothermic peak [ 60 ]. The broad endothermic peak found at 180 °C could be related to chitosan’s molecular chains’ arrangement. The exothermic peak found at 286 °C corresponded to chitosan’s thermal decomposition. The PVA DSC thermograph exhibited a shift in the baseline at 71 °C, which is consistent with the glass transition temperature ( tg ) for PVA [ 60 ]. Furthermore, an endothermic peak at 170 °C was observed, followed by a sharp endothermic peak at 220 °C, both of which are attributed to the melting point ( tm ) and PVA crystalline polymer fraction, respectively [ 61 , 62 ]. The DSC thermographs of the thermally and non-thermally treated chitosan/PVA fibers electrospun with different acetic acid concentrations are shown in Figure 6 b–e. The DSC curves for all the samples (the thermally and non-thermally treated fibers) exhibited an endothermic peak followed by an exothermic peak. The endothermic peaks observed at 166, 166, 165, and 164 (F-AA50, F-AA60, F-AA70, and F-AA80, respectively) for the non-thermally treated fibers and 166, 164, 163.8, and 163 (F-AA50, F-AA60, F-AA70, and F-AA80, respectively) for the thermally treated fibers could be related to the melting point ( Tm ) of the chitosan–PVA blend. The exothermic peaks observed at 231, 232, 231, and 231 (F-AA50, F-AA60, F-AA70, and F-AA80, respectively) for the non-thermally treated fibers and 227, 231, 229, and 230 (F-AA50, F-AA60, F-AA70, and F-AA80, respectively) for thermally treated fibers could be associated with a crosslinking (complex formation between polymers) reaction on the chitosan molecules [ 63 ]. Table 6 summarizes the DSC values for the fiber samples. The DSC curves of the non-thermally treated chitosan/PVA indicated that the baseline shift related to the PVA glass transition was no longer observed in any of the fiber samples. This is due to the shear stress caused by the electric field during electrospinning, which rearranges the polymer chains [ 64 , 65 ]. However, the DSC curves of the thermally treated chitosan/PVA fibers exhibited a change in the baseline at 37, 37, 37, and 39 °C (F-AA50, F-AA60, F-AA70, and F-AA80, respectively), as shown in Figure 6 b–e. The decrease in Tg in the fiber composition could be related to the mutual interactions between the components and their partial compatibility [ 66 , 67 ]. However, there have been reports that electrospun fibers have higher crystallinity after the electrospinning process, as is the case with PVA fibers, as reported by Koosha et al. [ 63 ]. Numerous studies have reported that electrospun fibers experience a reduction in their crystalline structures as a result of the rapid solidification process of the stretched polymers. In summary, from the thermal analysis’s results, it is possible to notice that submitting the chitosan/PVA fibers to a thermal treatment at 70 °C for 24 h influenced the thermal properties of the material. The thermogravimetric curves corroborate that the thermal treatment promoted the elimination of water and solvent residues. The mass loss at the first stage, related to the solvent evaporation, decreased in all the thermally treated samples from 11.6 to 2.42%, 11.2 to 3.1%, 10.52 to 3.26%, and 9.58 to 6.47% for F-AA50, F-AA60, F-AA70, and F-AA80, respectively. Moreover, the DTG thermographs showed that the thermal stability of the electrospun fibers increased after been thermally treated. This can be explained as the result of a physical crosslinking effect. Another plausible explanation is that thermal treatments that use temperatures above the glass transition temperature of the polymers could provide enough energy to the polymer chains to move freely and rearrange, relaxing the stretched stress molecules after the electrospinning process, which is a prerequisite for sufficient crystallinity [ 58 ]. This was confirmed by the DSC analysis results, from which it was possible to calculate the degree of crystallinity (%) from the enthalpy of the melting point ( ∆ H ) . As shown in Table 6 , the degree of crystallinity of the electrospun fibers increased after being subjected to the thermal treatment, from 1.43 to 1.87%, 1.77 to 1.96%, 1.86 to 2.25%, and 1.30 to 1.90% for F-AA50, F-AA60, F-AA70, and F-AA80, respectively. 3.4. Tensile Properties of Electrospun Fibers Table 7 displays the measured tensile strength, Young’s modulus, and elongation at break of the chitosan/PVA fiber mats, both before and after the thermal treatment. Based on the obtained results, a notable observation can be made regarding the significant alteration in the tensile properties of the electrospun polymeric fibers. These changes are evident not only in relation to the concentration of acetic acid but also as a direct outcome of the applied thermal treatment. Derived from the obtained results, it is noteworthy that fiber samples F-AA-50, F-AA-60, and F-AA-70 demonstrate distinctive curves that resemble those of hard/tough plastic polymers. These curves exhibit a distinct peak stress at the point of transition from elastic to plastic deformation, which is a characteristic feature of such materials. On the other hand, fiber sample F-AA80 shows the characteristic curve of a brittle polymer, which elastically deforms and fractures before deforming plastically, as shown in Figure 7 . Other studies have reported the change in the mechanical properties of electrospun chitosan/PVA fibers in relation to the polymeric concentration. Zhou Y. et al. [ 67 ] evaluate the mechanical properties of electrospun chitosan/PVA fibers with different polymer concentrations. Their study showed that as the mass ratio of PVA/chitosan increased from 10/90 to 50/50, the tensile strength also increased from 2.78 to 5.55 MPa. Similar results were reported by Charernsriwilaiwat N. et al. [ 68 ], who investigated the influence of the polymer concentration on the mechanical properties of electrospun chitosan/PVA fibers. The tensile strength of chitosan/PVA fibers with mass ratios ranging from 10/90 to 50/50 decreased from 8.9 to 1.5 MPa, respectively. The tensile strength of pure PVA fibers was reported to be 12.8 MPa. Nonetheless, and to the best of the authors’ knowledge, there are not many works that report the influence of acetic acid concentration on the mechanical properties of electrospun fibers. Mohammad Z. et al. [ 69 ] reported the fabrication of chitosan fibers via extrusion and a coagulation bath. The fibers were obtained from solution containing 1.5 wt.% chitosan dissolved in different acetic acid concentrations (1, 2, 3, and 6 v.% ). The mechanical test showed an improvement in the tensile strength in fibers from 0.7 to 1.2 MPa as the acetic acid increased from 1 to 2 v.% . Nevertheless, above 2 v.% , the tensile strength decreases to 0.3 MPa at 6 v.% . The change in the tensile properties of the electrospun chitosan/PVA fibers can be attributed to the plasticizing effect of the water and acetic acid molecules. The plasticizing effect of acetic acid has been reported by Carmiña et al. [ 70 ] and Zhang Y. et al. [ 71 ]. In their studies, they attributed the plasticizing effect to the large acetate ion which has a delocalized charge, capable of plasticizing the chitosan structure and facilitating a favorable long-range molecular arrangement, hence enhancing the mechanical properties. Similar results were reported by Vu T. et al. [ 72 ]. The authors investigated the mechanical properties of nanofibers prepared from an 8% PVA solution using electrospinning with varying concentrations of acetic acid (0%, 20%, 35%, and 50% w/w ). The results indicated a significant rise in both the elongation at break and Young’s modulus with an increasing acetic acid concentration. Specifically, the elongation at break increased from 37% (acetic acid, 0% w/w ) to 69.6% (acetic acid, 50% w/w ), while the Young’s modulus exhibited an enhancement from 213 MPa (acetic acid, 0% w/w ) to 449 MPa (acetic acid, 50% w/w ). Figure 7 shows the stress–strain diagrams of the thermally treated fiber mats. A decrease in both the tensile strength and elongation at break was observed for all the samples. The sample F-AA50 displayed a significant change in both the tensile strength and elongation at break, with values decreasing from 13.1 to 10.34 MPa and from 12.13 to 7.5%, respectively. In contrast, sample F-AA80 exhibited a lower change in both the tensile strength and elongation at break, with values decreasing from 11.43 to 9.56 MPa and from 5 to 4.3%, respectively. The change in the mechanical properties of the thermally treated fibers could be related to the evaporation of the acetic acid residues and water molecules. This could induce a non-covalent polymer–polymer interaction, primarily by the formation of hydrogen bonding [ 73 ], altering the tensile properties of the fiber mats. Furthermore, the thermal energy provided by the heat treatment can facilitate the mobility of polymer chains, allowing the amorphous regions to align and fold to form crystallites [ 53 ]. Many other works have reported the influence of thermal treatments on PVA fibers. Wong K. et al. [ 58 ] reported the change in mechanical properties measured for individual PVA fibers before and after 4 h of annealing at 135 °C. Their results shows that the mean elastic modulus of the individual fibers increased from 4.4 ± 1.4 GPa to 7.6 ± 2.3 GPa, exhibiting an increase of 80% over the fibers before the thermal treatment. Furthermore, Mahir Es-saheb et al. [ 74 ] assess the change in the mechanical properties of PVA nanofiber sheets after heating. In their work, the PVA nanofiber sheets were thermally treated at 85 and 140 °C. Their work shows that the yield stress changed from 6.981 MPa (for the non-heated fibers), 9.63 Mpa (the fibers heated at 85 °C), and 6.298 Mpa (the fibers heated at 140 °C). Meanwhile, the reported elongation at break changed from 59.81% (for the non-heated fibers), 31.26% (the fibers heated at 85 °C), and 29.71% (the fibers heated at 140 °C). 3.5. Electroactive Test Response As can be seen from the results obtained in the previous sections of this manuscript, the physical properties of the electrospun fibers vary based on the parameters used before (the concentration of acetic acid for the polymer’s dissolution) and after (the thermal treatment) the fabrication of the fibers. Taking this into consideration, the aim of this section was to study the influence the thermal treatment and acetic acid concentration have on chitosan/PVA fibers’ electroactive response. The electroactive response of the electrospun fibers was evaluated with an electrochemical cell via the measurement of the speed displacement of the fiber samples as a function of time under a cyclic potential between −10 and 10 V. The samples used in this experiment were obtained from the non-thermally and thermally treated electrospun fibers mats. The size of the samples used was 20 × 4 mm (length × width). The deformation response of the chitosan/PVA fiber sample under a cycling differential potential (−10 to 10 V) in an electrolytic solution (HCl with a pH of 3) is shown in Figure 8 . From the obtained results, it was observed that the different acetic acid concentrations used for the fabrication of the fibers had a strong influence on the electroactive response of the material, as shown in Figure 9 a. Even though all the electrospun fibers had the same polymeric concentration, they should exhibit a similar deformation to an electric stimulus, as was observed in our previous work [ 39 ]. Nonetheless, it is possible to observe that as the concentration of acetic acid utilized for the production of the fibers rises, two interesting behaviors were exhibited by the fibers. Firstly, the samples that were electrospun from solutions with higher concentrations of acetic acid became less electrically responsive in acidic media (a pH of <7). Sample F-AA50 exhibited the largest and fastest electrical response of 0.77 mm s −1 at a pH of 3. Meanwhile, sample F-AA60 showed the fastest displacement of 0.546 mm s −1 at a pH of 4. On the other hand, F-AA80 was not suitable for use in acidic media below a pH of 5, due to the fact that all the samples dissolved after their immersion in the solutions. The second behavior worth mentioning is that all the samples exhibited a notable electrical sensitivity between a pH of 7 and 8, with a speed displacement peak in this region. In this case, the fibers that were electrospun from solutions with higher concentrations of acetic acid became more electrically responsive at a pH of 7–8. The highest speed displacement peak was found for samples F-AA60 and F-AA70, exhibiting a maximal speed displacement of 1.06 mm s −1 and 1.14 mm s −1 at a pH of 7, respectively. Meanwhile, samples F-AA50 and F-AA80 had the lowest speed displacement at a pH of 7: 0.75 mm s −1 and 0.51 mm s −1 , respectively. From these results, it is possible to conclude that the electroactive response sensitivity of the fibers in acidic and basic media changes depending on the acetic acid concentration used. It was observed that the fibers that were electrospun from solutions with higher acid concentrations tend to have a lower electroactive response at a pH < 6, due to the high swelling and fast deterioration of the samples. These behaviors can be related to two factors. Firstly, the acetic acid remnants found in the fibers can promote the fast swelling and degradation of the samples fabricated with higher concentrations of acetic acid. Furthermore, as it was discussed in Section 3.2 (Study on the solutions’ rheological properties), the amount of —NH 2 that can get ionized (NH + 3 ) in the chitosan chain is directly proportional to the acid concentration used for its solubility. Therefore, the possible amount of free —NH 2 groups and NH + 3 cations could be higher in the fibers electrospun from highly concentrated acetic acid solutions. Thus, due to the increase in the amount of free —NH 2 groups and cations, the chitosan/PVA fibers became more electroactive responsive at neutral and basic pHs. To the best of the authors’ knowledge, no similar results have been reported prior to the completion of this paper. Figure 9 b shows the speed displacement measured from the thermally treated samples. It is possible to notice that the thermal treatment has a strong influence on the fibers’ electroactive response. In comparison to the non-thermally treated samples, which had a fast deterioration in acidic media, the thermally treated samples showed a better stability and a higher deformation at an acidic pH (2–3). F-AA80 was the most remarkable case, as it went from not been appliable at a pH of 3 to exhibiting a speed displacement of 1.37 mm s −1 at a pH of 3. This improvement was also observable for all the samples, which increased their speed displacement from 0.77 mm s −1 , 0.117 mm s −1 , and 0.308 mm s −1 to 2.16 mm s −1 , 1.56 mm s −1 , and 1.38 mm s −1 at a pH of 3 for F-AA50, F-AA60, and F-AA70 (thermally treated), respectively. Furthermore, the fibers’ electroactive response that was exhibited in all the samples showed a similar trend. Two speed displacement peaks were exhibited, the first at a pH of 3 and the second at around a pH of 9–10. The fibers electrospun with higher concentrations of acetic acid showed a greater electrical response, having a maximal speed displacement of 1.56 mm s −1 , 1.54 mm s −1 , and 1.31 mm s −1 at a pH of 9 for F-AA60, F-AA70, and F-AA80 (thermally treated), respectively. The thermally treated fibers showed an improved electroactive response in comparison to the non-thermally treated fibers. The partial elimination of residual acetic acid and water, as well as the increase in fiber crystallinity, as observed in Section 3.4 , may account for this improvement, such as their increased stability in acidic media. To the best of the authors’ knowledge, no similar results have been reported prior to the completion of this paper. Briefly, we will explain the bending mechanism as a reaction to an electric stimulus exhibited by the fibers. The amino groups contained in chitosan are responsible for its electroactive properties. These —NH 2 groups undergo protonation (NH + 3 ) in an acidic environment with a pH below 7, thus causing the chitosan to function as a cationic polyelectrolyte [ 75 ]. As described in the literature, the bending deformation of the fibers under an electric field can be described using Flory’s theory of osmotic pressure [ 33 , 76 ]. By applying an electric field, free ions are attracted to their counter electrodes, creating a gradient concentration of mobile ions in the solution. Hence, this leads to a difference in osmotic pressure (∆π) between the anode side (π1) and the cathode side (π2), causing the fibers to bend [ 77 , 78 ]. It can be concluded that several factors significantly influence the electroactive response of materials based on the results obtained. Specifically, in the case of electrospun-fiber-based electroactive materials, certain parameters such as the solvent concentration used prior to electrospinning exhibit a substantial influence over the material’s electroactive properties. Additionally, physical treatments, particularly thermal treatments, have demonstrated a remarkable ability to profoundly alter the material’s electroactive response. Remarkably, it has also been observed that thermal treatments enhance the stability of the fibers in aqueous environments compared to non-thermally treated fibers. Considering the combined impact of these findings, it is evident that these parameters play a significant role in tailoring electroactive properties during the development of responsive materials. 3.6. Fourier-Transform Infrared Spectroscopy FTIR Analysis The FTIR spectra of the chitosan, PVA powder, and chitosan/PVA nanofiber mats were analyzed to investigate the molecular interaction in the chitosan/PVA fibers. Figure 10 shows the FTIR spectra of the chitosan, PVA, and chitosan/PVA fibers (F-AA50, F-AA60, F-AA70, and F-AA80). Specific absorption peaks indicative of the molecular composition of chitosan were identified via a Fourier-transform infrared (FTIR) analysis. The peak observed at 3354 cm −1 could be related to the combined stretching vibrations of the O—H and N—H groups. The aliphatic C—H bonds that can be attributed to the peak at 2926 cm −1 indicate the stretching vibrations of aliphatic C—H bonds. Furthermore, the stretching vibration of the amino group is attributed to the peak at 1561 cm −1 . Additionally, the saccharide structure of chitosan is associated with characteristic peaks at 892 and 1150 cm −1 . These findings align with those of previous studies reported in [ 63 , 79 ]. The PVA’s FTIR spectrum shows distinctive absorption peaks that provide insights into its molecular characteristics. The hydroxyl group (—OH)’s stretching vibrations are associated with the absorption peak observed at approximately 3290 cm −1 . The peak at 2937 cm −1 is attributed to antisymmetric stretching vibrations of the CH2 groups. The peaks observed at 1709 cm −1 indicate C=O bonds’ stretching vibrations present in the PVA acetate units. An absorption peak related to the vibration of the C—H bonds in the methyl group was observed at 1420 cm −1 . The stretching of the C—O bonds associated with the crystalline portion of the polymer chain is observed at the absorption peak at 1141 cm −1 . The spectrum of the C—O bond’s asymmetric stretching vibration in the acetate group was observed around 1087 cm −1 . These findings are consistent with the existing literature in the field [ 80 ]. It is evident that the FTIR spectra are very similar to those of the PVA for all the chitosan/PVA fiber samples. A broad and intense band from 3000 to 3600 cm −1 was observed to be associated with O−H and N−H stretching vibrations. As a result of the dehydration process, this area is narrower and slightly sharper in thermally treated fiber mats [ 81 ]. The spectra reveal the formation of hydrogen bonds between PVA and chitosan, evident from the shift toward lower wavenumbers in the O−H and N−H stretching vibration peaks. Specifically, the peak at 3354 cm −1 for chitosan shifts to approximately 3300 cm −1 for all the chitosan/PVA fiber samples. The stretching vibrations of the C=O and C−O bonds of the acetate units in the PVA can be attributed to the peaks at 1712 cm −1 and 1640 cm −1 . The thermally treated fibers showed a peak shifting from 1640 to 1652 cm −1 (associated with the C=O stretching in amide and amide I vibration), which could be due to the formation of an amide group from the reaction of carboxylic with amine groups, as a result of the heat treatment [ 81 , 82 ]. Additionally, a shift peak was observed from around 1590 cm −1 to 1562 cm −1 , attributed to the hydrogen bonding between the PVA’s —OH groups and chitosan’s —NH group. The peaks at around 1410 (for the non-thermally treated fibers) and 1415 cm −1 (for the thermally treated fibers) are related to the methyl group (—CH 3 ) C—H bond vibrations. The peak around 1075 cm −1 could be related to the C−O bond asymmetric stretching vibration of the acetate group. A peak associated with C—H bending vibrations in the molecule was found around 842 cm −1 . These results are in good agreement with those of previous reports [ 63 , 83 ]. Spectra Deconvolution for the Determination of Intermolecular Hydrogen Bonding and Free Amine (—NH 2 ) Variation Post-Thermal Treatment It was found that the mechanical properties, the thermal properties, and the electroactive properties of the electrospun chitosan/PVA fibers were altered after being subjected to the thermal treatment. The change in the thermal and tensile properties, as was established previously, could be related to two possible effects: the increase in the crystallinity and/or the physical crosslinking between polymers chains. Nonetheless, the FT-IR spectra ( Figure 10 ) showed that the characteristic peak associated with the crystalline part of the PVA (1141 cm −1 ) observed in the PVA spectrum overlapped the peak at ~1075 cm −1 for all the fiber mat samples. Moreover, after being thermally treated, the fiber samples did not exhibit any noticeable change at this range. Thus, the change in the mechanical and thermal properties could be strongly related to a physical crosslinking effect, as a result of the removal of acetic acid and water residues from the microstructure of the electrospun fibers. The remotion of the acetic acid and water residues from the microstructure could lead to an increase in the intermolecular hydrogen bonding between the —NH 2 and —OH groups of the chitosan with the —OH groups of the PVA. On the other hand, from the electroactive response test, it was reported that the fibers electrospun from the solutions with high acetic acid concentrations at a pH below 5 exhibited a low to non-tip displacement. Moreover, these samples were dissolved after being immersed in the acidic medium. However, after being thermally treated, not only were their electroactive properties drastically enhanced, as shown in Figure 9 , but the solubility of fiber mats was also reduced, up to the point that all the samples showed a high stability in acidic media. The higher stability of the fibers in the aqueous media could also due to the physical crosslinking effect caused by the thermal treatment. Meanwhile, the improvement in their electroactive response can be associated with an increase in the proportion of the free amino groups anchored to the molecular structure of the fibers. As a consequence of the thermal treatment, the fibers present a change in their physical properties, which could be due to the variation in the intermolecular hydrogen bonding between the polymers and the variation in the free amino groups. In order to elucidate the variation in the proportion of hydrogen bonding interactions and free amine, after the thermal treatment, a deconvolution analysis employing Gaussian line shapes was employed on the peak of the Fourier-transform infrared (FTIR) spectrum in the range of 3000–3700 cm −1 . The —OH and the —NH region were studied in order to analyze the types of hydrogen bonds. In the —NH region, the free amine absorption peak is 3408 cm −1 ; the intermolecular association (N 2— H 1 …O 5 /N 2 —H 2 …O 1 ) peak is around 3335 cm −1 ; the intramolecular association (O 3 H…O 5 /O 3 H…O 6 ) absorption peak is around 3366 cm −1 ; the amide group (—CONH-) absorption peak is around 3240 cm −1 ; and the primary ammonium (—NH + 3 ) absorption peak is around 3100 cm −1 . In the −OH region, the free hydroxyl (—OH) absorption peak is around 3580 cm −1 , and the multimer intermolecular association (O 6 H…N 2 ) peak is around 3462 cm −1 [ 84 , 85 , 86 , 87 , 88 ]. Table 8 shows the results obtained from the spectra deconvolution. From Table 8 , it is noticeable that the fibers that were thermally treated showed a variation in relative strength (%) at the peak around 3408 cm −1 , associated with the free amine. The proportion of —NH 2 for all the samples increased after the thermal treatment, and sample F-AA50 had the highest increment from 6.39% to 6.9%. This variation can explain the change in the electroactive properties of the fibers. Furthermore, the peaks associated with the intermolecular hydrogen bonding exhibited an increase in all the samples that were thermally treated. The peak at 3335 cm −1 associated with intermolecular association (N 2— H 1 …O 5 /N 2 —H 2 …O 1 ) increased from 27.5%, 27.46%, 27.7%, and 27.6% to 28%, 29.15%, 27.9%, and 28.9% for F-AA50, F-AA60, F-AA70, and F-AA80, respectively. The peak at 3462 cm −1 associated with multimer-intermolecular association (O 6 H…N 2 ) increased from 22.5%, 20%, 23.3%, and 21.77% to 22.5%, 21.2%, 23.35%, and 21.77%, for F-AA50, F-AA60, F-AA70, and F-AA80, respectively. Additionally, it was observed that the proportion in the peak related to primary ammonium (—NH + 3 ) at 3100 cm −1 decreased in the samples that were thermally treated. From these results, it can be deduced that the thermal treatment induces a physical crosslinking effect on the chitosan/PVA fibers."
} | 12,953 |
29740597 | PMC5938282 | pmc | 7,244 | {
"abstract": "We present a genome-level understanding of how cellulose is metabolized by Thermoanaerobacterium for biobutanol production.",
"introduction": "INTRODUCTION Biofuels produced from renewable lignocellulosic biomass are expected to meet growing energy demands without increasing greenhouse gas emissions as fossil fuels ( 1 , 2 ). Among all biofuels, butanol is one of the most promising biofuels because of its high energy density (29.2 MJ/liter for butanol versus 19.6 MJ/liter for ethanol and 32 MJ/liter for gasoline) and is more similar to gasoline ( 3 , 4 ). As a natural reservoir for biomass-based carbon and the most abundant biomass on Earth ( 5 ), cellulose has become a major feedstock for butanol production in biorefinery processes. Currently, bioconversion of cellulose to biofuels in industries requires several steps including pretreatment, enzymatic saccharification, detoxification, and fermentation ( 6 – 8 ). Therefore, it is desirable to develop a bioconversion technology for the direct conversion of cellulosic biomass into biofuels without entailing any of the pretreatment steps mentioned above ( 9 ). However, this effort has been hampered by recalcitrance of cellulose and the lack of potent microbes. A key step to address the issue is to discover novel microorganisms having unique genetic cassettes to convert cellulosic materials into biofuels ( 10 , 11 ). As reported previously, a number of solventogenic strains from the genus Clostridium have been exploited to generate butanol from monosaccharides (for example, glucose and xylose) and starch (for example, corn and cassava). However, these strains cannot use polysaccharides, such as cellulose, for butanol generation; they can do so only for ethanol or hydrogen generation ( 12 ). Recent research attempts have been directed toward engineering microorganisms that can directly convert cellulose into biofuels in consolidated bioprocesses ( 9 , 13 , 14 ). For example, genes encoding for biosynthetic pathways of biofuels have been engineered into natural cellulolytic microorganisms such as Caldicellulosiruptor bescii , Clostridium cellulolyticum , and Clostridium thermocellum , which were capable of producing ethanol [0.64 g/liter ( 15 ) and 22.4 g/liter ( 16 )] and isobutanol [0.66 g/liter ( 17 ) and 5.4 g/liter ( 18 )] from cellulose. Similar attempts on Clostridium acetobutylicum have also been made by introducing cellulosome genes; however, the engineered strain was unable to use cellulose because of the complex assembly and the expressional stability of functional minicellulosomes ( 19 ). Until now, few wild-type species can produce biofuels, particularly butanol, directly from cellulosic biomass at high concentrations and yields. Thus, there is still a need to develop a consolidated bioprocessing technology for butanol production. Note that sustainable butanol production is also impeded by (i) product inhibition and carbon catabolite repression (CCR) ( 20 ) and (ii) complexities of downstream purification of butanol from other by-products such as acetone and ethanol ( 21 ). Hence, to achieve and simplify the consolidated bioprocessing, it is desirable to discover bacterial strains that can directly ferment cellulose and hemicellulose to butanol as the main product. Here, we report the discovery of the first cellulolytic wild-type bacterium ( Thermoanaerobacterium thermosaccharolyticum strain TG57) that can directly convert cellulose and xylan to butanol. Studies on the genomic characteristics of cellulose-assimilating butanologenic lifestyle are limited thus far ( 22 ). In Clostridium , the genes responsible for the formation of butyrate and acetone encode phosphotransbutyrylase and butyrate kinase. However, similar genes are absent in the genome of T. thermosaccharolyticum strain TG57. Instead, the genome of strain TG57 contains novel genes encoding butanol dehydrogenase (Bdh), endocellulase, and cellobiohydrolase, which were not found in T. thermosaccharolyticum DSM 571 ( 22 ), a strain closely related to TG57. Thus, to reveal the genetic repertoire of T. thermosaccharolyticum strain TG57, we carried out genomic, transcriptomic, functional, and biochemical characterization of this new isolate to explore the relationship among genetics, metabolism, and molecular regulation.",
"discussion": "DISCUSSION The newly discovered wild-type cellulolytic T. thermosaccharolyticum TG57 reveals novel characteristics to convert microcrystalline cellulose to butanol: (i) generating butanol up to 1.93 g/liter with a yield of 0.20 g/g; (ii) simultaneous utilization of hexose and pentose to produce butanol (7.33 g/liter) from glucose, xylose, and arabinose mixtures ( Figs. 1 and 3 ); and (iii) lack of by-product acetone from using cellulose. Among the few available reports on biofuel (for example, butanol and ethanol) production from lignocelluloses, strain TG57 represents the first wild-type strain that can ferment microcrystalline cellulose to butanol (1.93 g/liter), comparable to previously reported wild-type [ethanol (0.46 g/liter)] or gene-modified [butanol (1.42 g/liter)] microbes ( Table 1 ). Collectively, the metabolic characteristic of strain TG57 strengthens the economic feasibility of butanol generation on the basis of lowering substrate costs and simplifying downstream product extraction complexities. Both genomic and transcriptomic characterizations of T. thermosaccharolyticum strain TG57 provide a deep insight into the following: (i) the unusual metabolic characteristics of its genome; (ii) silencing in a CCR-encoding ccp gene to eliminate catabolite repression (for example, negligible transcription of the ccp gene coding a regulator that mediates high catabolite repression, leading to higher expressions of the xylm and xylk genes responsible for xylose utilization); (iii) a streamlined genome with a compact butonate pathway, a shortage of transposons, and a lack of genes encoding acetone bypass synthesis, while having plenty of determinants responsible for central metabolic functions with wide carbohydrate assimilation and efficient butanol production capabilities; and (iv) active transcription variations in diverse aspects of metabolism including up-regulated tryptophan synthesis as well as efflux and secretory systems, but down-regulated fatty acid synthesis, suggesting multiple ways of optimizing cellulosic fermentation by dosing growth supplements (for example, l -tryptophan). The genome of T. thermosaccharolyticum strain TG57 is of potential biotechnological interest because it harbors many of the genes responsible for metabolic processes such as 7 haloacid dehalogenase genes; 11 genes for biotin, thiamine, cobalamin, and riboflavin biosynthesis; 10 various cytochrome genes that function as carriers of electrons; and 37 genes for oxidoreductases such as the NADH/flavin family, the Fe-S family, and related transcriptional regulators and transporters with clustered forms similar to those described previously ( 54 ). Collectively, T. thermosaccharolyticum TG57 extends a biotechnological potential beyond lignocellulosic biofuel. Production of butanol from cellulose exhibits a foundational milestone to realize the ultimate goal of cost-effective production of renewable biofuels and chemicals from lignocellulosic biomass in a consolidated bioprocess."
} | 1,836 |
30881355 | PMC6406033 | pmc | 7,246 | {
"abstract": "Anaerobic biodegradation of aromatic compounds under sulfate-reducing conditions is important to marine sediments. Sulfate respiration by a single bacterial strain and syntrophic metabolism by a syntrophic bacterial consortium are primary strategies for sulfate-dependent biodegradation of aromatic compounds. The objective of this study was to investigate the potential of conductive iron oxides to facilitate the degradation of aromatic compounds under sulfate-reducing conditions in marine sediments, using benzoate as a model aromatic compound. Here, in anaerobic incubations of sediments from the Pearl River Estuary, the addition of hematite or magnetite (20 mM as Fe atom) enhanced the rates of sulfate-dependent benzoate degradation by 81.8 and 91.5%, respectively, compared with control incubations without iron oxides. Further experiments demonstrated that the rate of sulfate-dependent benzoate degradation accelerated with increased magnetite concentration (5, 10, and 20 mM). The detection of acetate as an intermediate product implied syntrophic benzoate degradation pathway, which was also supported by the abundance of putative acetate- or/and H 2 -utilizing sulfate reducers from microbial community analysis. Microbial reduction of iron oxides under sulfate-reducing conditions only accounted for 2–11% of electrons produced by benzoate oxidation, thus the stimulatory effect of conductive iron oxides on sulfate-dependent benzoate degradation was not mainly due to an increased pool of terminal electron acceptors. The enhanced rates of syntrophic benzoate degradation by the presence of conductive iron oxides probably resulted from the establishment of a direct interspecies electron transfer (DIET) between syntrophic partners. In the presence of magnetite, Bacteroidetes and Desulfobulbaceae with potential function of extracellular electron transfer might be involved in syntrophic benzoate degradation. Results from this study will contribute to the development of new strategies for in situ bioremediation of anaerobic sediments contaminated with aromatic compounds, and provide a new perspective for the natural attenuation of aromatic compounds in iron-rich marine sediments.",
"conclusion": "Conclusion In summary, using sediment from Pearl River Estuary as microbial inocula, the present study demonstrated that the supplementation of (semi)conductive iron oxides, magnetite or hematite, accelerated the rate of benzoate degradation under sulfate-reducing conditions. The detection of acetate, along with microbial analysis implied that syntrophic dependence was essential to benzoate degradation under sulfate-reducing conditions. The stimulatory effect of (semi)conductive iron oxides on sulfate-dependent benzoate degradation could be a result of stimulating DIET within syntrophic microorganisms. This hypothesis, however, warrants further investigation. An increasing number of studies have demonstrated the capability of conductive iron oxides to promote DIET, which highlights the potential application of iron oxides in industrial and environmental biotechnologies such as bioremediation. Considering the ubiquity of conductive iron oxides in soils and sediments, DIET-facilitated syntrophic metabolisms could be responsible for natural attenuation of aromatic compounds under sulfate-reducing and methanogenic conditions in anaerobic environments.",
"introduction": "Introduction Aromatic compounds comprise numerous environmental pollutants and their removal often relies on microbial degradation ( Cao et al., 2009 ). Most contaminated subsurface environments are anaerobic, microbial degradation of aromatic compounds, such as BTEX (benzene, toluene, ethylbenzene, and xylene) or PAH (polycyclic aromatic hydrocarbons), have been observed under nitrate-reducing, Fe(III)-reducing, sulfate-reducing, and methanogenic conditions ( Fuchs et al., 2011 ; Ghattas et al., 2017 ; Muller et al., 2017 ; Varjani et al., 2017 ). This might be the main mechanism in the natural attenuation of aromatic compounds in natural environments, and the supplementation of alternative electron acceptors has been considered as an attractive strategy for bioremediation of sites contaminated with aromatic compounds ( Perelo, 2010 ; Ghattas et al., 2017 ; Hou et al., 2018 ; Guo et al., 2019 ). Marine sediments are a major final sink and reservoir for aromatic compounds in aquatic environments which often derived from oil spills, industrial waste, sewage effluent, and surface runoff. Due to the high sulfate concentrations in marine sediments, biodegradation of aromatic compounds under sulfate-reducing conditions might be the dominant biodegradation pathway in marine environments. In previous studies, the degradation of aromatic hydrocarbons has been inhibited by suppressing the activity of sulfate-reducing microorganisms or depleting the presence of sulfate in sediments ( Lovley et al., 1995 ). Rothermich et al. (2002) demonstrated biodegradation of in situ pools of aromatic hydrocarbons in petroleum-contaminated marine sediments under sulfate-reducing conditions, and suggested that microbial sulfate reduction was the main driving force responsible for the self-purification capacity of contaminated harbor sediments. In marine ecosystems, many isolated sulfate reducers are capable of mineralizing aromatics with sulfate as the final electron acceptor ( Davidova et al., 2007 ; Musat and Widdel, 2008 ; Ahn et al., 2009 ; Musat et al., 2009 ; Meckenstock and Mouttaki, 2011 ). As a niche adaption, syntrophic aromatics degradation under sulfate-reducing conditions is also widespread in subsurface environments ( Jackson et al., 1999 ; Kleinsteuber et al., 2008 , 2012 ; Herrmann et al., 2010 ; Rakoczy et al., 2011 ; Gieg et al., 2014 ). In a syntrophic metabolism, aromatics-degrading bacteria cooperate with sulfate-reducing bacteria that can promptly consume end products generated from the breakdown of aromatic compounds ( Gibson and Harwood, 2002 ). Recently, (semi)conductive iron oxides, such as magnetite or hematite, have been demonstrated to be capable of stimulating anaerobic syntrophic metabolism ( Kato et al., 2012a , b ; Viggi et al., 2014 ; Baek et al., 2015 ; Li et al., 2015 ; Liu et al., 2015 ; Yamada et al., 2015 ; Zhuang et al., 2015 ; Tang et al., 2016 ; Zhang and Lu, 2016 ; Wang et al., 2018 ). In these either laboratory cultures or complex environments, (semi)conductive iron oxides have been proposed to stimulate direct interspecies electron transfer (DIET) in syntrophic interactions. To date, iron oxides-mediated DIET has been mostly observed in syntrophic methanogenesis, the potential of iron oxides to facilitate sulfate-dependent syntrophic metabolism remains unexplored, especially for syntrophic oxidation of aromatic compounds. Benzoate is most frequently used as a model compound for studying anaerobic metabolism of aromatic compounds ( Young and Frazer, 1987 ). Benzoate biodegradation under sulfate-reducing conditions may occur through direct sulfate reduction by a single culture ( Drzyzga et al., 1993 ; Madan and Bernhard, 2015 ) or, syntrophically, by a consortium of bacteria ( Hopkins et al., 1995 ). The aim of the present study was to investigate the potential of (semi)conductive iron oxides to enhance the rate of benzoate degradation under sulfate-reducing conditions, which might provide effective bioremediation technology for marine sediments contaminated by aromatic compounds. Here, in the absence and presence of magnetite or hematite, the coupling of benzoate degradation and sulfate reduction was analyzed in anaerobic incubations of sediment from the Pearl River Estuary, China. Using 16S rRNA sequencing, microbial communities from sulfate-dependent benzoate degradation in the absence and presence of conductive iron oxides were characterized to further elucidate the microbial syntrophic mechanisms involved.",
"discussion": "Discussion Syntrophic Benzoate Degradation Under Sulfate-Reducing Conditions Under sulfate-reducing conditions, benzoate can be oxidized completely to CO 2 (Equation 1); free hydrogen or acetate is not expected in this metabolism ( Drzyzga et al., 1993 ). However, acetate was detected during sulfate-dependent benzoate degradation in the present study, which suggests the occurrence of incomplete oxidation of benzoate. There are two pathways of sulfate-dependent benzoate degradation with the appearance of acetate as an intermediate product: (i) using sulfate as electron acceptor, sulfate reducers oxidize benzoate incompletely to produce acetate (Equation 3), which might be further utilized by other acetate-utilizing sulfate reducers (Equation 4) ( Kamagata et al., 1992 ); (ii) benzoate is syntrophically degraded by acetate/H 2 -producing benzoate-degraders and acetate/H 2 -using sulfate reducers (Equations 5–7) ( Fang et al., 1997 ). (3) C 7 H 5 O 2 − + 0.75 S O 4 2 − + 4 H 2 O → 3 C H 3 C O O − + H C O 3 − + 0.75 H S − + 2.25 H + Δ G ′ 0 = − 81.4 k J / r e a c t i o n (4) C H 3 C O O − + S O 4 2 − → 2 H C O 3 − + H S − Δ G ′ 0 = − 47.6 k J / r e a c t i o n (5) C 7 H 5 O 2 − + 7 H 2 O → 3 H 2 + H C O 3 − + 3 C H 3 C O O − + 3 H + Δ G ′ 0 = + 70.5 k J / r e a c t i o n (6) C H 3 C O O − + S O 4 2 − → 2 H C O 3 − + H S − Δ G ′ 0 = − 47.6 k J / r e a c t i o n (7) 4 H 2 + S O 4 2 − + H + → H S − + 4 H 2 OΔ G ′ 0 = − 151.9 k J / r e a c t i o n Although the incomplete oxidization of benzoate to acetate coupling sulfate reduction is thermodynamically favorable (Equation 3), to date, not a single sulfate-reducing microorganism has been reported to be responsible for this incomplete oxidation of benzoate under sulfate-reducing conditions. Kamagata et al. (1992) found that an anaerobic consortium degraded benzoate producing acetate with a sulfate reducer that did not oxidize benzoate, which implied that syntrophic interaction was essential for benzoate degradation. For the second pathway, the need for syntrophic metabolism is determined by the energetically unfavorable reaction of benzoate oxidation to acetate (Equation 5), which can become exergonic if sulfate reducers keep acetate or H 2 at very low concentrations by active consumption (Equations 6–7). In the present study, based on the presence of acetate during sulfate-dependent benzoate degradation, syntrophic metabolism was suggested to be highly involved. In all anaerobic incubations of sediment, the most predominant bacterial sequences were related to δ-Proteobacteria which include all gram-negative mesophilic sulfate-reducing bacteria. The known sulfate-reducing bacteria observed in the present study were Desulfobulbaceae (15.0–31.7%), Syntrophobacteraceae (10.0–28.1%), Desulfobacteraceae (0.21–2.5%), and Desulfomicrobiaceae (0.13–1.0%). These sulfate-reducing taxa implied that they were potentially responsible for sulfate reduction during benzoate degradation in the present study. The family Desulfobulbaceae comprises physiologically versatile sulfate reducers, and the cultured representatives are capable of utilizing H 2 , acetate, propionate, lactate, pyruvate, and alcohols ( Kuever, 2014a ; Rabus et al., 2015 ). Recently, a species of Desulfoprunum benzoelyticum within the family Desulfobulbaceae has been identified to be capable of utilizing benzoate with sulfate reduction ( Madan and Bernhard, 2015 ), however, the metabolism is complete oxidation without acetate production. Syntrophobacteraceae are the second most important sulfate reducers detected in this experiment. Majority of them are able to use H 2 or acetate to reduce sulfate, but their capability of utilizing benzoate for sulfate reduction has not been reported. Syntrophobacteraceae affiliated species have been identified as the major acetate-degrading sulfate reducers in Italian paddy soil ( Liu et al., 2018 ). Members within the family Desulfobacteraceae are either mesophilic or psychrophilic sulfate-reducing bacteria, and some of them can completely oxidize benzoate under sulfate-reducing conditions. Although, all members within the Desulfomicrobiaceae family can oxidize organic substrates incompletely to acetate with sulfate reduction, benzoate is not included ( Kuever and Galushko, 2014 ). In summary, the phylotypes affiliated to sulfate-reducing δ-Proteobacteria in this experiment were mainly putative acetate- or/and H 2 -utilizing sulfate reducers. Bacteria capable of syntrophic metabolism largely belong to δ-Proteobacteria ( Mcinerney et al., 2008 ), here we detected the well-known syntrophic bacteria including Syntrophobacteraceae, Syntrophaceae, and Geobacteraceae with a relative abundance of > 1%. Among them, species of Syntrophaceae generally oxidize benzoate incompletely with H 2 -utilizing partners ( Kuever, 2014b ). For example, Syntrophus species have been reported to grow syntrophically on benzoate with H 2 -using sulfate reducers ( Auburger and Winter, 1995 ; Hopkins et al., 1995 ; Elshahed et al., 2001 ). All Geobacteraceae can oxidize acetate ( Röling, 2014 ), and several species are capable of degrading benzoate ( Coates et al., 2001 ). Species affiliated to the family Geobacteraceae do not use sulfate as an electron acceptor ( Röling, 2014 ), but can act as syntrophic acetate oxidizer in the presence of hydrogenotrophic partner ( Kato et al., 2012b ). Syntrophobacteraceae are capable of fermentative metabolism or growing in syntrophic association with H 2 /formate-utilizing partner ( Kuever, 2014c ). Thus, the presence of various syntrophic bacteria suggested their role to oxidize benzoate or acetate with the production of H 2 , and potentially live associated with hydrogenotrophic sulfate-reducing bacteria. In the present study, the detection of acetate during sulfate-dependent benzoate degradation, as well as microbial communities that characterized with the abundant presence of syntrophic benzoate/acetate oxidizer and acetate/H 2 -utilizing sulfate reducers demonstrated that benzoate was not degraded by a single sulfate reducer, but by syntrophic metabolism of several different microorganisms. Conductive Iron Oxides Accelerated Syntrophic Benzoate Degradation In the absence of sulfate, microbial reduction of magnetite or hematite was insignificant, and they were not serving as electron acceptors for benzoate oxidation ( Figure 1A ). In the presence of sulfate, both magnetite and hematite reduction were significantly enhanced, which was consistent with previous findings that iron reduction can be enhanced during bacterial sulfate reduction ( Li et al., 2006 ). Fe(II) production from magnetite or hematite reduction in the presence of sulfate was close to (first feeding cycle) or much lower than (second feeding cycle) that from NTA-Fe(III) reduction. Thus, the very comparable rates of sulfate-dependent benzoate degradation in the absence and presence of NTA-Fe(III) ( Figure 4 ) might help to eliminate two possibilities for the stimulatory effect of conductive iron oxides on benzoate degradation under sulfate-reducing conditions: (i) the presence of iron oxides increasing the pool of terminal electron acceptors; (ii) decreasing H 2 S toxicity by FeS formation via Fe(II) and sulfide (caused by sulfate reduction). Conductive iron oxides have been documented to accelerate a range of anaerobic syntrophic interactions, including methanogenesis ( Kato et al., 2012b ; Viggi et al., 2014 ; Li et al., 2015 ; Yamada et al., 2015 ; Zhuang et al., 2015 ), dechlorination ( Aulenta et al., 2013 , 2014 ) and nitrate reduction ( Kato et al., 2012a ). It has been proposed that conductive minerals have stimulatory effect by promoting DIET between electron-donating microorganisms and electron-accepting microorganisms responsible for syntrophic metabolism. The function of conductive minerals has been suggested as an electron conduit for DIET ( Kato et al., 2012b ) or a substitute for pilin-associated c-type cytochrome ( Liu et al., 2015 ). As an alternative to traditional interspecies H 2 /formate transfer, the interspecies electron transfer occurred via electrical conduction is faster than Fick’s law-based H 2 diffusion, making DIET advantageous over interspecies H 2 /formate transfer. In the present study, since syntrophic interactions were necessary for benzoate degradation under sulfate-reducing conditions, the facilitated benzoate degradation by the addition of conductive iron oxides was likely due to the establishment of DIET between syntrophic benzoate oxidizers and sulfate reducers. As the rates of sulfate-dependent benzoate degradation increased with increasing addition of magnetite, the relative abundances of Bacteroidetes and Desulfobulbaceae also showed an increasing trend with the increasing dose of magnetite. Although, the physiological function of Bacteroidales in oxidizing benzoate has not been reported so far, species of Bacteroidales are known to be capable of anaerobic degradation of complex substrates to simple sugars and other products ( Xu et al., 2003 ). Dawson et al. (2012) observed dense aggregates of a methanogenic community and a bacterial community consisting of Acetobacterium spp., Bacteroidales, and SRB385-hybridizing Firmicutes, which were responsible for syntrophic interactions in a biogenic gas field containing aromatic compounds. In a benzene-contaminated aquifer with ongoing sulfate reduction, Bacteroidales has been identified as the dominant order of bacteria ( Herrmann et al., 2008 ). These findings provide evidence for the involvement of Bacteroidales in aromatic compounds degradation under anaerobic conditions. To participate in DIET, microorganisms must be capable of extracellular electron transfer. Bacteroidetes have been detected as the predominant microorganisms in the anode biofilm in microbial fuel cells (MFC) fed with complex organic compounds ( Kim et al., 2006 ; Zhang et al., 2009 ) and their function of extracellular electron transfer were evidenced by Fe(III)-reducing activity and electrochemical activities ( Kim et al., 2006 ). The family Desulfobulbaceae was found to consistently be enriched on the surface of electrodes generating electricity from marine sediment fuel cells ( Tender et al., 2002 ; Holmes et al., 2004b ). For example, Desulfobulbus propionicus was the first sulfate-reducing bacteria capable of conserving energy to support growth via electron transfer to insoluble electron acceptors (iron oxides and electrodes) ( Holmes et al., 2004a ). Considering that both Bacteroidetes and Desulfobulbaceae have the potential capability for extracellular electron transfer, as well as their enrichment with magnetite supplementation, they might be involved in magnetite-mediated DIET during the course of benzoate degradation coupled with sulfate reduction. Most probably, in the presence of magnetite, electrons generated from the metabolic oxidation of benzoate by Bacteroidetes were transferred via electrical conduction to Desulfobulbaceae for reducing sulfate ( Supplementary Figure S2 ). However, this hypothesis warrants further investigations. In the present study, sulfate-dependent benzoate degradation produced acetate as the primary intermediate that was further oxidized. Though acetate oxidation coupled to sulfate reduction can be performed by a single microorganism, the relative abundances of Syntrophobacteraceae that are capable of acetate-consuming sulfate reduction decreased with the increasing concentration of magnetite. In comparison, the abundances of Desulfobulbaceae that comprises many genera capable of utilizing H 2 as an electron donor were increased with the increasing loading of magnetite. As well as being performed by a single organism, sulfate-dependent acetate oxidation might also occur through syntrophic associations according to the following equations (Equations 8 and 9). (8) C H 3 C O O H + 2 H 2 O → 4 H 2 + 2 C O 2 Δ G ′ 0 = + 104.6 k J / r e a c t i o n (9) 4 H 2 + S O 4 2 − + H + → H S − + 4 H 2 OΔ G ′ 0 = − 151.9 k J / r e a c t i o n Magnetite-facilitated syntrophic acetate oxidation has been reported under methanogenic conditions, in which magnetite stimulated DIET between Geobacter and Methanosarcina species ( Kato et al., 2012b ; Zhou et al., 2014 ; Rotaru et al., 2018 ). The possibility of magnetite-stimulated DIET between acetate-oxidizing bacteria and H 2 -using sulfate reducer ( Supplementary Figure S3 ) can be further evidenced by the co-culture of Geobacter sulfurreducens (oxidizes acetate but cannot use sulfate) and Desulfovibrio sp. (reduces sulfate but cannot use acetate)."
} | 5,140 |
24846283 | PMC4028902 | pmc | 7,247 | {
"abstract": "Biochar, a charcoal-like product of the incomplete combustion of organic materials, is an increasingly popular soil amendment designed to improve soil fertility. We investigated the possibility that biochar could promote direct interspecies electron transfer (DIET) in a manner similar to that previously reported for granular activated carbon (GAC). Although the biochars investigated were 1000 times less conductive than GAC, they stimulated DIET in co-cultures of Geobacter metallireducens with Geobacter sulfurreducens or Methanosarcina barkeri in which ethanol was the electron donor. Cells were attached to the biochar, yet not in close contact, suggesting that electrons were likely conducted through the biochar, rather than biological electrical connections. The finding that biochar can stimulate DIET may be an important consideration when amending soils with biochar and can help explain why biochar may enhance methane production from organic wastes under anaerobic conditions.",
"discussion": "Discussion The results demonstrate that biochar has sufficient conductivity to promote direct electron transfer between syntrophic partners in co-cultures based on DIET. This provides a potential explanation for observations that some biochar amendments can enhance methane production in soils 33 or in small-scale digesters converting organic waste to methane 34 35 . The results suggest that biochar promotes interspecies electron exchange via a conduction-based mechanism, in which electrons migrate through the biochar from electron-donating to electron-accepting cells. This is similar to the mechanism proposed for interspecies electron transfer through GAC 9 , but differs significantly from extracellular electron exchange with electron shuttles such as humic substances 25 . In the absence of conductive materials, microorganisms growing together, required a long adaption time and numerous transfers 13 , to get to the same substrate consumption rates as those observed with biochar or GAC 9 . This suggests that cells required time to express cellular components required for extracellular electron transfer 14 . The ability of biochar to promote DIET with similar rates and stoichiometries as those observed in co-cultures amended with GAC 9 , might be surprising considering the conductivity of biochar is 1000-fold less than that of GAC. The higher conductivity of activated carbon is likely due to increased surface area and porosity and increased aromatization, which happens during the conversion of biochars into activated carbon at higher temperatures 22 23 . Aromaticity, a consequence of electron delocalization between aromatic rings localized on distinct neighboring planes, gives conductive properties to graphite, charcoals or other organic polymers 24 , and as discovered recently even to the pili of Geobacter species 21 . The conductivity of the biochars evaluated here was comparable to that of G. sulfurreducens pili preparations 20 21 ,which are sufficient to effectively promote DIET. The ability of biochar to stimulate DIET appears to overcome the adaption period that cells require to begin expressing high levels of the components that are required for pili-based DIET 13 14 16 . Materials with increased aromaticity are doped by reduction or oxidation reactions 36 . If the acceptor microorganism reduces sections of the biochar and the donor microorganism oxidizes sections of the biochar, there will be intrinsic charge differences between sections of biochar, promoting electron flow. This has been also noted on activated carbon, which accepted electrons from microorganisms and then released the electrons to Fe(III) citrate under acidic conditions 38 . However, biochar is a complex material and can modify environments to which it is added with properties other than conductivity. For example, biochar was speculated to act as a “ shuttle ” to mitigate N 2 O emissions during denitrification in soils 38 . Whereas other studies suggested that N 2 O formation in soils is due to abiotic processes happening on biochars surface enriched in surface charged groups, like quinones, metal ions or radicals 39 40 . Our observations that biochar increases methane production in defined co-culture systems, in which partners were capable of direct electron transfer, changes the present understanding that biochar could mitigate methane gas emissions. Considering the potential significant impact of methane production on global warming, and the persistence of biochar in soil, warrants further long-term studies on how soil methanogenic communities are affected by biochar amendments and the impact of biochar on the global carbon cycle."
} | 1,168 |
26906501 | null | s2 | 7,249 | {
"abstract": "Most of the networks used by computer scientists and many of those studied by modelers in neuroscience represent unit activities as continuous variables. Neurons, however, communicate primarily through discontinuous spiking. We review methods for transferring our ability to construct interesting networks that perform relevant tasks from the artificial continuous domain to more realistic spiking network models. These methods raise a number of issues that warrant further theoretical and experimental study."
} | 127 |
35946347 | PMC9435057 | pmc | 7,250 | {
"abstract": "Abstract Biological nitrogen fixation (BNF) by cyanobacteria is of significant importance for the Earth’s biogeochemical nitrogen cycle but is restricted to a few genera that do not form monophyletic group. To explore the evolutionary trajectory of BNF and investigate the driving forces of its evolution, we analyze 650 cyanobacterial genomes and compile the database of diazotrophic cyanobacteria based on the presence of nitrogen fixation gene clusters (NFGCs). We report that 266 of 650 examined genomes are NFGC-carrying members, and these potentially diazotrophic cyanobacteria are unevenly distributed across the phylogeny of Cyanobacteria , that multiple independent losses shaped the scattered distribution. Among the diazotrophic cyanobacteria, two types of NFGC exist, with one being ancestral and abundant, which have descended from diazotrophic ancestors, and the other being anaerobe-like and sparse, possibly being acquired from anaerobic microbes through horizontal gene transfer. Interestingly, we illustrate that the origin of BNF in Cyanobacteria coincide with two major evolutionary events. One is the origin of multicellularity of cyanobacteria, and the other is concurrent genetic innovations with massive gene gains and expansions, implicating their key roles in triggering the evolutionary transition from nondiazotrophic to diazotrophic cyanobacteria. Additionally, we reveal that genes involved in accelerating respiratory electron transport ( coxABC ), anoxygenic photosynthetic electron transport ( sqr ), as well as anaerobic metabolisms ( pfor , hemN , nrdG , adhE ) are enriched in diazotrophic cyanobacteria, representing adaptive genetic signatures that underpin the diazotrophic lifestyle. Collectively, our study suggests that multicellularity, together with concurrent genetic adaptations contribute to the evolution of diazotrophic cyanobacteria.",
"introduction": "Introduction Biological nitrogen fixation (BNF) is a critical process in the nitrogen biogeochemical cycle that impacts primary productivity in both marine and terrestrial ecosystems ( LeBauer and Treseder 2008 ; Canfield et al. 2010 ). Although the process is meditated by various archaeal and bacterial lineages, diazotrophic cyanobacteria are considered to be of paramount importance in the modern nitrogen cycle. Nitrogen fixation in the oceans is largely attributed to a small group of marine cyanobacteria (e.g., Trichodesmium , Richelia , UCYN-A (unicellular cyanobacterium Candidatus Atelocyanobacterium thalassa), and Crocosphaera ) ( Sohm et al. 2011 ; Martínez-Pérez et al. 2016 ; Zehr and Capone 2020 ), and cyanobacteria in cryptogamic covers are recognized as major players that contribute to nearly half of the total BNF on land (e.g., Microcoleus , Leptolyngbya , and Pseudanabaenaceae ) ( Elbert et al. 2012 ; Pietrasiak et al. 2013 ). Therefore, elucidating the evolutionary history of this important trait within Cyanobacteria sets the ground for understanding the evolution of Earth’s biogeochemical nitrogen cycle ( Stüeken et al. 2016 ). BNF is solely catalyzed by nitrogenase enzymes ( Canfield et al. 2010 ). There are three forms of nitrogenase, characterized by different metal contents of the active-site cofactor: molybdenum-iron nitrogenase (Mo-nitrogenase), vanadium-iron nitrogenase (V-nitrogenase), and iron-iron nitrogenase (Fe-nitrogenase) ( Eady 1996 ; Raymond et al. 2004 ). Among these, Mo-nitrogenase is the most effective and dominant enzyme complex. The Mo-nitrogenase enzyme consists of two components: the electron-transfer Fe protein (dinitrogenase reductase) encoded by nifH , and the MoFe protein (dinitrogenase) encoded by nifD and nifK genes. The Fe protein provides the driving force for electron transfer, which is a homodimer that contains two adenosine triphosphate (ATP)-binding sites and a 4Fe-4S cluster. The MoFe protein is a α 2 β 2 tetramer that contains two metal clusters: the FeMo-co in the active site, and the P cluster for transferring electrons from the Fe protein to the FeMo-co. In addition, the biosynthesis of the FeMo-co in the MoFe protein depends on three additional genes, nifENB ( Raymond et al. 2004 ; Esteves-Ferreira et al. 2017 ). The nitrogenase is an oxygen-sensitive enzyme. Only under microanaerobic or anaerobic conditions, the synthesis of nitrogenase and its catalytic reaction occur ( Fay 1992 ). During the BNF process, the Fe protein is reduced by electron donor, such as ferredoxin. Then, with hydrolysis of two ATPs, electrons are further transferred from the Fe protein to the MoFe, resulting in N 2 being reduced to NH 3 ( Duval et al. 2013 ). Hence, cyanobacteria that are capable of BNF usually satisfy the following conditions: 1) possessing a genetic toolbox ( nifHDKENB ) that encodes for nitrogen-fixing enzyme nitrogenase; 2) having strategies that protect oxygen-sensitive nitrogenase from atmospheric oxygen and oxygen produced in oxygenic photosynthesis process ( Berman-Frank et al. 2001 , 2003 ; Zehr et al. 2008 ; Bandyopadhyay et al. 2013 ; Bergman et al. 2013 ; Cornejo-Castillo and Zehr 2019 ; Inomura et al. 2019b ); and 3) generating sufficient ATP and reductant required for BNF ( Scherer et al. 1988 ). Despite the advances in understanding structural and functional profiles of BNF in Cyanobacteria , the evolution of BNF across cyanobacterial lineages remain poorly characterized. A preliminary study which utilized public genomic data of 49 cyanobacteria strains has provided an insight into this question ( Latysheva et al. 2012 ). Through the ancestral state reconstruction analysis for these 49 cyanobacteria genomes, researchers proposed that the BNF capacities of extant cyanobacteria are all derived from a single gain of BNF in the last common ancestor of Cyanobacteria (LCAC). Nevertheless, the limited number of strains and biased genome sampling might influence the resulting inference, as a comprehensive coverage of diazotrophic and nondiazotrophic cyanobacteria is crucial to mapping BNF capacity across extant cyanobacteria and accurately infer the evolutionary events (gains and losses of BNF capacity) ( Raymond et al. 2004 ; Shi and Falkowski 2008 ; Yan et al. 2008 ; Falcón et al. 2010 ; Latysheva et al. 2012 ). The recent explosive growth of genomic data of cyanobacteria offers the opportunity to uncover the evolutionary history of BNF in Cyanobacteria ( Latysheva et al. 2012 ; Harel et al. 2015 ; Esteves-Ferreira et al. 2017 ; Chen et al. 2021 ). Furthermore, diazotrophic cyanobacteria exhibit distinct morphologies and phenotypes, and some of these features have been reported to be related BNF capacity. For example, some diazotrophic cyanobacteria, such as strains from Nostoc and Aphanizomenon genera, form nitrogen-fixing heterocyst to compartmentalize oxygenic photosynthesis and BNF ( Flores and Herrero 2010 ). While some marine nonheterocyst-forming diazotrophic cyanobacteria synthesize hopanoid-derivatives to reduce membrane permeability to extracellular oxygen ( Cornejo-Castillo and Zehr 2019 ). Thus, it could be assumed that diazotrophic cyanobacteria might exhibit unique adaptive features that correlated with the evolution of BNF ( Berman-Frank et al. 2001 , 2003 ; Zehr et al. 2008 ; Bandyopadhyay et al. 2013 ; Bergman et al. 2013 ; Cornejo-Castillo and Zehr 2019 ; Inomura, Deutsch, et al. 2019 ; Inomura, Wilson, et al. 2019 ). However, the genetic factors that are evolutionarily correlated with BNF remain poorly characterized. Here, on the basis of the representative genomic data set and the backbone phylogeny of Cyanobacteria that we recently published ( Chen et al. 2021 ), we address the knowledge gaps in understanding the origin and evolutionary history of BNF in Cyanobacteria and explore the underlying genetic players affecting the evolutionary dynamics of BNF. We carried out comparative genomic analysis of the Mo-nitrogenase gene family ( nifHDKENB ) across 650 cyanobacterial genomes and determined the phylogenetic distribution of diazotrophic cyanobacteria. Then plausible evolutionary scenarios of BNF were evaluated according to the phylogenetic patterns of Mo-nitrogenase genes. By inferring the ancestral gene repertoire, ancestral states of BN,F and multicellularity, we explored the potential evolutionary forces that shaped the evolution of BNF. Using gene enrichment analysis, we further uncovered genes correlated with BNF trait, which are likely to be involved in evolutionary adaptation to diazotrophic lifestyle.",
"discussion": "Discussion In this study, we asked what evolutionary pattern of BNF can be found in Cyanobacteria . Our dataset, which includes genomes from Cyanobacteria with extensive taxon sampling, enabling us to perform a detailed analysis of the evolutionary history of BNF in Cyanobacteria . As a result of our analyses, we propose the following evolutionary trajectory of BNF in Cyanobacteria . First, our results evidenced that BNF was not an ancestral feature of the LCAC ( Shi and Falkowski 2008 ; Falcón et al. 2010 ), the capacity to fix nitrogen evolved in filamentous cyanobacteria ( fig. 2 ), which recently estimated to appeared around 2.2 billion years ago, following the great oxidation event (GOE) ( Boden et al. 2021 ). In the presence of oxygen arose in GOE, ammonium would be biologically converted to nitrite and nitrate, promoting the diversification of organisms which could utilize the newly available oxidants ( Ren et al. 2019 ). On the other hand, organic nitrogen returned to atmosphere via enhanced denitrification and anaerobic ammonium oxidation processes, thus a modern-style aerobic nitrogen cycle dominated by nitrogen loss initially prevail globally ( Zerkle et al. 2017 ). In this scenario, the cyanobacteria with the capacity to produce oxygen and fix both CO 2 and N 2 thus would be fundamental players in the global C and N cycles since Proterozoic eon. Subsequently, massive independent losses of BNF occurred at broad phylogenetic scales, as it is costly to maintain BNF, including the high energy cost of BNF and the sensitivity of nitrogenase to oxygen poisoning, resulting in the uneven distribution of BNF among cyanobacterial lineages ( fig. 1 ). The maintenance of BNF in specific lineages is often accompanied by evolving more sophisticated mechanisms to coordinate BNF and oxygenic photosynthesis, such as forming specialized heterocysts to fix nitrogen ( Nostocales ; fig. 1 ) ( Herrero et al. 2016 ), exhibiting a circadian rhythm of BNF in unicellular cyanobacteria ( Bandyopadhyay et al. 2013 ). It therefore seems that the differential trait losses reflected the trade-off between photosynthesis and BNF in some cyanobacteria ( Albalat and Cañestro 2016 ). Meanwhile, we found evidence for several HGT events generating diversified nitrogenases through the evolution of Cyanobacteria , including interphylum HGT events that contributed to the nitrogenase reported in obligate anaerobes ( fig. 3 ) and interclade HGT events that provided additional nitrogenase ( supplementary fig. S5, Supplementary Material online). Altogether, the findings here provided insight into the potential roles of vertical inheritance, differential loss and HGT in the evolution of BNF in Cyanobacteria . On the basis of a wealth of genomic data of diazotrophic and nondiazotrophic cyanobacteria, we further explored the genetic and mechanistic basis underlying the evolutionary dynamics of BNF. Our results indicate that the two episodes of extensive gene gains and expansions (Node 8 and its preceding node, fig. 2 ), likely enlarged functional capabilities in ancestral cyanobacteria, contributing to the emergence of BNF and multicellularity. Indeed, gene families which facilitated BNF were equipped during the events. Since BNF evolved in cyanobacteria, adaptation to diazotrophic lifestyle drove changes in genomic contents. As a result, a small set of genes was found to be universally enriched in diazotrophic cyanobacteria, which remodeled the metabolic core to provide fitness advantages to diazotrophic cyanobacteria under anaerobic conditions ( fig. 4 ). In subsequent evolutionary steps, further genomic adaptation appeared in response to different morphologies and phenotypes of diazotrophic cyanobacteria (e.g., nonheterocyst-forming diazotrophic cyanobacteria and heterocyst-forming diazotrophic cyanobacteria; fig. 4 ). The findings here further improved our understanding of the evolutionary adaptation of Cyanobacteria from nondiazotrophic to diazotrophic."
} | 3,140 |
33910898 | PMC8081358 | pmc | 7,254 | {
"abstract": "Quantitative assessments of microbial activity in crustal fluids yield potential strategies to supplement recalcitrant DOC.",
"introduction": "INTRODUCTION In the deep ocean, seawater is entrained into the rocky crust, chemically altered by abiotic and microbial processes, and discharged from the seafloor as hydrothermal fluid with a global flux that rivals riverine inputs ( 1 , 2 ). More than 90% of this hydrothermal fluid discharge is from low-temperature fluids (5° to 20°C) circulating on the flanks of mid-ocean ridges ( 1 ), where these fluids are generally inaccessible and their microbial assemblages are largely unexplored ( 3 ). The biogeochemical influence of this cool, ridge-flank microbiome on net chemical fluxes, and particularly on the enormous, climate-sensitive reservoir of deep-ocean dissolved organic carbon (DOC), is potentially substantial, but poorly constrained. For this study, pristine, cool, basaltic subseafloor fluids ( 4 ) from 8–million year-old crust were recovered from ocean drilling borehole observatories of the North Pond site located at 22°N on the western flank of the Mid-Atlantic Ridge. Here, oxygenated crustal fluids are largely indistinguishable from bottom seawater and concentrations of ammonium, methane, hydrogen sulfide, and iron(II) are below detection, indicating an overall low redox energy potential in these subseafloor crustal fluids ( 5 ). While DOC sourced from the deep ocean is thought to be unreactive and resistant to microbial degradation ( 6 , 7 ), previous isotopic data from these fluids suggest that selective removal of DOC via microbial oxidation does occur ( 8 , 9 ) and DOC may, therefore, be the most abundant reduced substrate available for microbes to oxidize in cool crustal fluids ( 9 ). While the abundance and diversity of microorganisms in the subseafloor have been explored for decades via scientific drilling programs ( 10 ), slow growth and often low biomass present challenges for demonstrating microbial activity under environmentally relevant conditions. Microbial activity via uptake of labeled substrates has been successfully observed in sedimentary ( 11 – 13 ) and diffuse flow hydrothermal vent fluids ( 14 – 16 ) using stable isotope probing (SIP) incubations coupled to single-cell measurements with nanoscale secondary ion mass spectrometry (NanoSIMS), providing constraints on the potential microbial contribution to primary production and organotrophy in these habitats. However, no such data exist from the crustal biome. Here, we determine single-cell and bulk estimates of microbial carbon and nitrogen incorporation from the ridge flank crustal habitat that represents the majority of fluid flux between the subsurface and the overlying ocean. Our extensive and quantitative assessments highlight a microbial population poised to incorporate fresh sources of labile organic carbon and a consistent, wide range of intercell incorporation rates across fluids and conditions. We also report significant bicarbonate incorporation, despite the absence of abundant inorganic sources of redox energy that could fuel chemolithotrophy, and suggest that this may be part of a metabolic strategy of supplementing anabolic carbon needs with bicarbonate to reduce the reliance on aged and recalcitrant deep-ocean DOC. North Pond hosts two CORK (Circulation Obviation Retrofit Kit) seafloor borehole observatories installed in 2011 at Integrated Ocean Drilling Program (IODP) Sites U1382A and U1383C ( Fig. 1A ). The CORK at site U1382A accesses circulating fluids from one depth interval below the sediment in the rocky subseafloor [90 to 210 meters below seafloor (mbsf)], and the U1383C CORK accesses three depth ranges: Shallow (70 to 146 mbsf), Middle (146 to 200 mbsf), and Deep (200 to 332 mbsf). Although the geochemistry of fluids recovered from both CORK observatories is largely similar to overlying seawater, radiocarbon measurements and larger differences in dissolved oxygen and DOC concentrations indicate that fluids recovered from U1383C are more isolated from bottom water recharge than U1382A ( 5 , 8 , 9 ). Recent numerical simulations suggest that there is convective and oscillatory fluid movement through the rocky crust ( 17 ) rather than the simple linear flow along the north-south axis as had been hypothesized in earlier studies of North Pond ( 4 ). Fig. 1 North Pond is located in young, cool crust (8 Ma) with a “sediment pond” roughly 13 km long by 7 km wide and up to 200 m deep. ( A ) Diagram of CORK observatories at North Pond with different sampling locations (U1382A, bottom seawater, and U1383C) and depths in meters below surface (m) or meters below seafloor (mbsf). U1383C is located about 6 km to the NE of U1382A. Bulk 13 C-carbon isotope incorporation rates calculated from ( B ) 13 C-acetate, ( C ) 13 C-bicarbonate, and ( D ) 13 C-methylamine SIP incubations of each fluid sample for 12-hour, 2-day, and 6-day incubation periods in order shown in (A). Asterisk indicates the value computed from SIP-NanoSIMS single-cell measurements ( Table 3 ) instead of bulk elemental analysis. No bar indicates that the data are below detection (see Materials and Methods).",
"discussion": "RESULTS AND DISCUSSION Fluids described in this study were collected using a mobile pumping system (MPS) ( 18 ) in October 2017 as a part of the third sampling expedition to North Pond. Bottom seawater was also collected by Niskin bottles on a conductivity, temperature, depth (CTD) water sampling rosette. 16 S ribosomal RNA (rRNA) gene sequencing of all samples (North Pond crustal fluids and bottom seawater) indicates that the microbial community in crustal fluids is distinct from those communities in bottom seawater in 2017 (figs. S1 and S2), as has been seen in previous years ( 8 , 19 ). Within the crustal fluids, the microbial community at U1382A is more similar to bottom water than the U1383C fluid horizons, which is also consistent with geochemical data ( 5 ). Moreover, in a separate study of a diverse range of mineral chips incubated in the CORKs for 4 to 6 years (retrieved in 2017), the authors recovered similar taxa across all sample types (CORK fluids, bottom water, and mineral chips) and determined that the incubation fluid explained more of the microbial community composition colonizing minerals than the type of mineral surface ( 20 ). Together, these results indicate that the fluids collected in 2017, and used in the current study, capture the dominant microbial communities in the North Pond crustal fluids. Cell counts from CORK fluids in 2017 ranged from 2.1 × 10 3 to 5.1 × 10 3 cells ml −1 ( Table 1 ). These 2017 counts are lower than historical cell count data for previously collected fluids, which have ranged from high 10 3 to low 10 4 cells ml −1 of fluid ( 8 , 19 ). Decreasing cell concentrations after drilling has also been observed in other CORKs [Juan de Fuca ridge flank; ( 21 )] and groundwater well systems ( 22 ). Together with geochemical data [ Table 1 ; ( 5 )], low cell counts likely indicate that the North Pond system had recovered from drilling and that the 2017 fluids (and resulting data) are the best representation of microbial activity in the cold, oxic crustal subseafloor aquifer to date. Table 1 North Pond CORK fluids and bottom water values for cell enumeration, dissolved inorganic carbon (DIC), and dissolved organic carbon (DOC) from 2017 samples collected for this study. Additional geochemistry (oxygen, nitrate, and pH) reproduced from ( 5 ) also collected in 2017. Ammonium concentrations were all below detection (<0.1 μmol/kg) from 2017 ( 5 ). Sample Depth (m) Cells ×10 3 ml −1 ±95% CI pH DIC (μmol/kg) O 2 (μM) Nitrate (μmol/kg) DOC (μmol/kg) Bottom Water 4397* 9.0 0.4 7.92 2188 250 21.8 39 U1382A 90–210 5.1 0.2 7.91 2164 228 22.3 31 U1383C Shallow 70–146 3.7 0.3 8.07 2167 198 22.8 20 U1383C Middle 146–200 2.1 0.2 8.13 2189 205 22.9 22 U1383C Deep 200–332 3.7 0.3 8.06 2156 173 22.7 22 *Depth in meters below surface, remaining depths are in meters below seafloor (mbsf). Bicarbonate concentrations are ±5 μmol/kg. Incubations of bottom seawater and crustal fluids were all amended with deuterated ( 2 H 2 O) water, which can be used as a general tracer of microbial anabolic activity ( 23 ). Select combinations of 13 C carbon (bicarbonate, acetate, methylamine, and diatom lysate) and 15 N nitrogen (ammonium, methylamine, and diatom lysate) were provided as substrate-specific tracers of anabolic activity ( Table 2 ). No substrates were added to CN controls. For diatom lysate, diatoms were grown in the presence of isotopically enriched 13 C-bicarbonate and 15 N-nitrate and then lysed before addition as a proxy for environmentally relevant complex organic matter. Incubations were conducted at temperatures bracketing the expected range in the North Pond aquifer (4° to 20°C), and cells were harvested from separate incubations prepared for three time points (12 hours, 2 days, and 6 days; Table 2 ). Isotope incorporation rates were calculated from bulk elemental analysis ( Fig. 1, B to D ) and single-cell NanoSIMS measurements ( Figs. 2 and 3 ). Table 2 Experimental conditions of stable isotope probing incubations for the five fluid sources collected at North Pond (bottom water, U1382A, U1383C Shallow, U1383C Middle, and U1383C Deep). All incubations were conducted in triplicate separate bottles for each time point at 12 hours, 2 days, and 6 days. Condition identifiers used in figures referenced here, where C is control, D is diatom lysate, M is methylamine, A is acetate, and B is bicarbonate or bicarbonate, and the temperature is in subscript. Condition 2 H label 13 C label 15 N label Temperature (°C) C 20 Water None None 20 D 20 Water Diatom lysate Diatom lysate 20 M 20 Water Methylamine Methylamine 20 A 20 Water Acetate Ammonium 20 B 20 Water Bicarbonate Ammonium 20 C 4 Water None None 4 D 4 Water Diatom lysate Diatom lysate 4 M 4 Water Methylamine Methylamine 4 A 4 Water Acetate Ammonium 4 B 4 Water Bicarbonate Ammonium 4 Fig. 2 Selected NanoSIMS ratio images for each stable isotope ( 2 H, 15 N, and 13 C). ( A ) The most active Shallow sample amended with 13 C-DIC and 15 N-ammonium (B 20 U1383C Shallow), ( B ) U1383C Deep 13 C 15 N–diatom lysate amendment, where the remaining diatom lysate is visible in 15 N and 13 C images, but 2 H uptake is only seen in associated filaments (D 20 U1383C Deep), ( C ) 13 C 15 N-methylamine uptake in 2 H and 15 N without visible 13 C incorporation (M 20 U1383C Deep), and ( D ) most active Deep sample (A 20 U1383C Deep) amended with 13 C-acetate and 15 N-ammonium. All incubations were amended with 2 H 2 O. Scale bars, 3 μm. Ratio range scale differs between samples to show maximum values per image. Fig. 3 Single-cell isotope incorporation rates calculated from U1383C Shallow and Deep incubations. Computed fmol ( A ) C and ( B ) N cell −1 day −1 rates are based on NanoSIMS data and plotted as the kernel density distribution with true values overlain. Microbial populations in North Pond fluids appear poised for dynamic and heterogeneous conditions. Cell counts from organic carbon–amended incubations (acetate and diatom lysate) had the highest cell densities in earlier time points (fig. S3). By 2 days, all incubations with acetate (1383C Middle) or diatom lysate (all other samples) had the highest cell densities. By the final time point of 6 days, most of these organic carbon incubations had a lower cell density than earlier time points or CN controls, suggesting an initial period of cell growth followed by death. This pattern of cell increase in the first few days followed by decrease has also been observed in groundwater wells with no amendments ( 22 ), which the authors attribute to necromass-induced growth from cells that die off at the onset of the incubation. In bulk analysis performed on 13 C-acetate–, 13 C-methylamine–, and 13 C-bicarbonate–amended incubations, acetate incorporation was widespread across fluids and conditions ( Fig. 1B ). Bicarbonate incorporation was not detected as consistently as acetate incorporation but was on par with acetate incorporation in some fluids (e.g., U1383C Middle and Deep; Fig. 1C ). Methylamine incorporation rates were lower overall, compared to acetate and bicarbonate, but uniquely elevated in U1383C Middle at both 4° and 20°C ( Fig. 1D ). Bulk rates of carbon incorporation from acetate were highest at 6 days for all 20°C fluids except U1383C Shallow, where the rate decreased between days 2 and 6 ( Fig. 1B ). Acetate 4°C incubations were generally lower than their 20°C counterparts (except for U1383C Shallow where the 12-hour rate was highest at 4°C; Fig. 1B ). Examination of the single-cell rates for carbon incorporation ( Fig. 3A ) showed a bimodal distribution of rates for U1383C Deep (20°C) and Shallow (4°C) at their respective in situ temperatures. These data suggest that microbes in deeper crustal fluids may persist on recalcitrant carbon sources, with select members poised to quickly respond to an influx of more labile organic carbon. Previous genomic studies of North Pond crustal fluid microbial communities collected in 2012 recovered genes for autotrophic CO 2 fixation ( 8 , 19 ), and much higher incorporation of 13 C-bicarbonate than 13 C-acetate (800 to 4300 pmol ml −1 day −1 versus no more than 104 pmol ml −1 day −1 ; table S4) from SIP incubations ( 8 ). However, the fluids collected in 2012, only 6 months after drilling, were particle-laden and geochemical data suggested that they represented a mixture of crustal fluids, bottom seawater, and surface seawater ( 5 ). 13 C-bicarbonate incorporation rates from 2017 CORK fluids were much lower and ranged from below detection to 94 pmol ml −1 day −1 . 2017 bottom seawater rates (0.95 pmol ml −1 day −1 ) were comparable to a study from the western branch of North Atlantic Deep Water, which overlays the North Pond site (0.24 pmol ml −1 day −1 from 14 C-bicarbonate, 2537-m water depth; table S4) ( 24 ). Thus, it appears that bicarbonate incorporation may have been artificially elevated in these early studies. Still, bicarbonate incorporation rates on par with acetate incorporation in some 2017 fluids is surprising because of the lack of abundant electron donors typically used for chemolithotrophy. Though oxygen was abundant, methane, hydrogen sulfide, ammonium, and reduced iron are all below detection limits ( 5 , 8 ) and unlikely to fuel extensive autotrophic carbon fixation in North Pond crustal fluids. Limited carbon fixation driven by the oxidation of reduced iron and sulfur minerals on basalt surfaces is still possible, although 13 C-bicarbonate stable isotope incubations conducted with basalt samples and metagenomics investigations of these same rocks showed no conclusive evidence for carbon fixation ( 25 , 26 ). However, global estimates indicate its importance in the first ~10 million-years (Ma) of crustal evolution ( 27 ), and chemolithotrophy in biofilms on rock surfaces may not have been captured by observations of bulk rates. Given the lack of electron donors that could be paired with chemolithotrophy in our incubations, it is unlikely that rock-driven autotrophy is occurring in our experiments. Metagenomic and metatranscriptomic analysis of the same 2017 North Pond fluids used in our experiments showed more transcription of organotrophic genes than autotrophic genes across all sampling horizons, suggesting a microbial community utilizing organic carbon ( 28 ). Carbon fixation transcripts were most abundant in 1383C Middle and Deep where we also detected the highest rates of bicarbonate incorporation. Furthermore, metagenomic-assembled genomes (MAGs) from the fluids revealed that a number of MAGs contained carbon fixation pathways linked to oxidation of sulfide and thiosulfate, but these same MAGs also contained numerous extracellular protease and carbohydrate catabolism genes, consistent with a mixotrophic lifestyle. These North Pond results are in accordance with the predominance of heterotrophic bacteria reported in ~33- and 104-Ma basalt crust cored beneath the South Pacific Gyre ( 29 ), as well as in subseafloor ultramafic and gabbroic rocks cored from the Atlantic Ocean ( 30 , 31 ), where organotrophic microbial processes also dominate. Because DOC is the reduced substrate with the highest concentration in the North Pond crustal fluids (20 to 31 μmol/kg DOC; Table 1 ), organotrophy coupled with anabolism of both organic and inorganic carbon is the more likely explanation for observed uptake of bicarbonate. While anabolic incorporation of bicarbonate is underexplored in environmental settings, it has been demonstrated in laboratory conditions with pure cultures. For example, Pseudomonas AM1, Hyphomicrobium vulgare , and Methylobacterium extorquens are able to use a combination of the ethylmalonyl–coenzyme A pathway and serine cycle, resulting in as much as 50% of biomass carbon derived from bicarbonate ( 32 , 33 ). We hypothesize that organotrophy coupled to anabolism of both organic and inorganic carbon reflects microbial communities optimizing the low potential redox energy conditions in the rocky subseafloor. Crustal fluid DOC likely represents the fraction of deep-ocean DOC that remains after the more bioavailable components are removed on short time scales after fluids are entrained in the crust ( 9 ). Our results indicate that this degraded DOC alone supports an active microbial population because cell growth was observed even when no carbon or nitrogen or other reduced substrates were added to incubations. DOC may be largely oxidized for energy by organotrophic pathways and bicarbonate may provide a supplementary, and less metabolically expensive, anabolic source of carbon. Bicarbonate incorporation rates from 2017 were consistently high across time points in U1383C Deep fluids and highest with U1383C Middle fluids in 2-day incubations at 4° and 20°C ( Fig. 1C ). U1383C Middle and Deep fluids are also the fluids with the most “aged” or 14 C-depleted DOC ( 9 ). DOC in fluids from U1383C has a radiocarbon age of 7300 to 9200 years and a higher aromaticity index and percent carboxyl-rich alicyclic molecules (CRAM) than bottom water or fluids from U1382A ( 9 ). Because DOC concentrations are lower at all subseafloor depths at U1383C compared to U1382A ( Table 1 ), and compositional evidence indicates that the DOC is the most aged and degraded at U1383C ( 9 ), microbial communities at this site may have the most to gain from supplementing their anabolic needs with bicarbonate incorporation. Our results indicate that there is not a concentration limit below which natural DOC is unavailable to microbes ( 34 ), but rather, there could be a different strategy for accessing the aged and recalcitrant fraction of deep-ocean DOC by combining biomass production from both inorganic and organic carbon sources (fig. S8). This could be supplemented by other oligotrophic strategies to use diverse sources of organic carbon in low-DOC (<4 μM) conditions ( 35 ). While dissolved nitrogen has been less studied at North Pond than DOC, ammonium loss through nitrification has been hypothesized in North Pond fluids based on geochemical fluxes ( 5 ). All of the provided nitrogen sources analyzed from U1383C Deep ( 15 N-ammonium, 15 N-methylamine, and 15 N–diatom lysate) and U1383C Shallow fluids ( 15 N-ammonium) were incorporated by cells based on NanoSIMS analysis ( Fig. 3B ). It was generally observed that when 15 N-ammonium was paired with 13 C-acetate, the nitrogen incorporation rate was higher than when paired with 13 C-bicarbonate. It was also observed that 13 C– and 15 N–diatom lysate C and N incorporation rates were similar to the 15 N-ammonium and 13 C-acetate C and N rates for comparable 12-hour 20°C U1383C Deep incubations. Therefore, organic nitrogen anabolic rates may have been higher when organic carbon was also provided ( Table 3 and fig. S8). Table 3 Average per-cell and per-milliliter rates of carbon and nitrogen uptake from NanoSIMS data of 12-hour incubations. Incubations that were conducted near in situ temperatures for each fluid source are shaded in gray. Condition U1383C fluid fmol C cell −1 day −1 fmol N cell −1 day −1 pmol C ml −1 day −1 pmol N ml −1 day −1 C:N D 20 Deep 1.3 0.096 86 6.2 14 M 20 Deep 0.0085 0.0087 0.16 0.17 0.98 A 20 Deep 1.15 0.097 15 1.3 12 B 20 Deep 0.042 0.021 0.4 0.17 2.0 A 4 Deep 0.21 0.12 2 1.5 1.7 B 4 Deep 0.46 0.0029 3 0.020 155 A 20 Shallow 0.30 0.054 1.1 0.2 5.5 B 20 Shallow 3.0 0.42 5 0.6 7.0 A 4 Shallow 0.33 0.025 10 0.8 13 B 4 Shallow 0.13 0.0085 0.4 0.023 16 Methylamine incorporation was observed in all fluids. The highest methylamine incorporation rate (72 pmol C ml −1 day −1 ) was from the same condition that had the highest bicarbonate incorporation rate, U1383C Middle 20°C at 2 days. This suggests that Middle fluids were the most amenable to different carbon sources, with the highest rates of bicarbonate and methylamine carbon incorporation and the third highest acetate incorporation. Bulk carbon incorporation of methylamine was comparable to carbon uptake rates of other simple organic N compounds from oxygen-deficient zones (ODZs) with similar oxygen concentrations to North Pond crustal fluids ( 36 ), such as urea and cyanate (5 to 12 pmol C ml −1 day −1 and 2 to 14 pmol C ml −1 day −1 , respectively; table S4). However, nitrogen incorporation rates from the same compounds in ODZ incubations ranged from 2 to 265 pmol N ml −1 day −1 , which were much higher than average nitrogen incorporation rates derived from single-cell analysis of North Pond microbial communities ( Table 3 ), where the highest average rate from diatoms was 6 pmol N ml −1 day −1 and ammonium was 1 pmol ml −1 day −1 . Overall, this suggests that both C and N can be incorporated from organic N sources in North Pond crustal fluids and, from these organic N sources, North Pond C incorporation rates were more similar to ODZ rates than the N incorporation rates. Cool, crustal aquifer fluids appear to support a wider range of microbial activities than more stable subsurface sedimentary and diffuse flow hydrothermal vent counterparts. The average observed single-cell uptake rates (10 −3 to 10 0 fmol C or N cell −1 day −1 ; Table 3 ) fall between SIP-NanoSIMS rates derived from deep, organic-rich marine sediments incubated with bicarbonate and ammonium (10 −2 fmol C or N cell −1 day −1 ; table S5) and diffuse flow hydrothermal vent fluids of the East Pacific Rise incubated with bicarbonate (10 0 to 10 1 fmol C cell −1 day −1 ; table S5). The highest average C and N North Pond incorporation rates were observed in 20°C incubations (U1383C Deep with acetate and ammonium at 1 fmol C cell −1 day −1 and 0.1 fmol N cell −1 day −1 ; U1383C Shallow with bicarbonate and ammonium at 3 fmol C cell −1 day −1 and 0.4 fmol N cell −1 day −1 ; Table 3 ). While the upper end of this range may only be observed under incubation conditions, the highest average rates were from incubations where no additional DOC was provided (bicarbonate amendments), illustrating the potential for North Pond crustal microbes to exhibit high anabolic rates using aged DOC and little or no inorganic redox-active substrates for energy conservation. Higher bicarbonate uptake observed in the shallower fluids at 20°C may also be the result of supplemental bicarbonate anabolism, which allowed the colder in situ fluid community to more quickly take advantage of growth under warmer incubation temperatures, similar to deeper communities using bicarbonate to supplement available DOC. Therefore, in addition to implications for marine carbon and nitrogen cycling, the wide range in potential incorporation rates may have additional import for understanding evolutionary constraints in subseafloor ecosystems, which have been heretofore been based on sedimentary and diffuse flow “end-member” subseafloor biomes. A global first-order estimate approximated that 10 11 to 10 12 mol organic carbon could be incorporated per year in the cool subseafloor crustal aquifer fluids. This is derived from an estimated habitable pore volume for North Pond–age crust (<10 Ma) and the average single-cell uptake rates from U1383C Shallow and Deep fluid incubations amended with organic carbon at the temperatures most relevant to in situ conditions ( Table 4 ). This average estimate for fluid biomass production from organic carbon agrees with the estimated annual removal of DOC from cool crust at 10 11 mol DOC year −1 ( 9 ) and suggests that the majority of DOC loss could be associated with organotrophic and anabolic processes rather than catabolism. We also find that organic nitrogen may be anabolized in this system at average rates of 10 10 to 10 11 mol N year −1 . Table 4 Estimated average, maximum, and minimum moles of C or N year −1 produced in young (<10 Ma) crustal fluids for incubations that were conducted near in situ temperatures for each fluid. Estimates are based on single-cell average, maximum, and minimum carbon and nitrogen isotope incorporation rates from NanoSIMS analysis for U1383C Shallow and Deep samples. Fluid Condition mol C year −1 mol N year −1 Avg Max Min Avg Max Min U1383C Deep D 20 2 × 10 12 1 × 10 12 8 × 10 8 1 × 10 11 8 × 10 11 2 × 10 8 M 20 1 × 10 10 2 × 10 11 2 × 10 7 1 × 10 10 7 × 10 10 2 × 10 7 A 20 2 × 10 12 2 × 10 13 2 × 10 8 1 × 10 11 5 × 10 11 5 × 10 7 B 20 6 × 10 10 7 × 10 11 2 × 10 8 3 × 10 10 2 × 10 10 9 × 10 7 U1383C Shallow A 4 4 × 10 11 3 × 10 12 1 × 10 8 3 × 10 10 7 × 10 11 1 × 10 7 B 4 2 × 10 11 5 × 10 11 9 × 10 9 1 × 10 10 3 × 10 10 6 × 10 7 Estimates for bicarbonate incorporation were approximately 10 10 to 10 11 mol C year −1 ( Table 4 ) and provide a potential mechanism for adding more labile, organic carbon to the crustal aquifer (fig. S8). This estimate from young crustal fluids is higher than estimates of annual primary productivity from hydrothermal systems at 10 9 mol C year −1 ( 24 ), suggesting that cool crustal fluid bicarbonate utilization may be equally or more significant than on-axis hydrothermal vent fluids on a global scale."
} | 6,570 |
35422414 | PMC9199093 | pmc | 7,255 | {
"abstract": "Despite major environmental and genetic differences, microbial metabolic networks are known to generate consistent physiological outcomes across vastly different organisms. This remarkable robustness suggests that, at least in bacteria, metabolic activity may be guided by universal principles. The constrained optimization of evolutionarily motivated objective functions, such as the growth rate, has emerged as the key theoretical assumption for the study of bacterial metabolism. While conceptually and practically useful in many situations, the idea that certain functions are optimized is hard to validate in data. Moreover, it is not always clear how optimality can be reconciled with the high degree of single-cell variability observed in experiments within microbial populations. To shed light on these issues, we develop an inverse modeling framework that connects the fitness of a population of cells (represented by the mean single-cell growth rate) to the underlying metabolic variability through the maximum entropy inference of the distribution of metabolic phenotypes from data. While no clear objective function emerges, we find that, as the medium gets richer, the fitness and inferred variability for Escherichia coli populations follow and slowly approach the theoretically optimal bound defined by minimal reduction of variability at given fitness. These results suggest that bacterial metabolism may be crucially shaped by a population-level trade-off between growth and heterogeneity.",
"conclusion": "Conclusion We have shown here that, as the growth medium gets richer, phenotype distributions inferred for E. coli populations appear to follow and slowly approach a theoretical limit that quantitatively relates the mean biomass production rate to the cell-to-cell variability in metabolic phenotypes. Specifically, the mean biomass gets closer to the maximum allowed by the inferred heterogeneity of the population. Despite the fact that large fluctuations affect the activity of metabolic pathways, the scenario we obtain reproduces some of the well-known trade-offs that characterize E. coli growth under carbon limitation, including downregulation of glycolysis, upregulation of respiration and the TCA cycle, and the transition to acetate overflow. This suggests that E. coli populations trade some of their fitness to maintain their metabolic heterogeneity nearly as large as possible in all growth conditions considered.",
"introduction": "Introduction A standard assumption of theoretical models of microbial metabolism is that cells regulate the fluxes through metabolic reactions to maximize their growth rate (i.e., their biomass output) ( 1 , 2 , 3 , 4 ). While intuitive and highly successful in many applications ( 5 ), this idea is not easy to validate in exponentially growing microbial populations. Quantitative studies of the interplay between metabolism and gene expression suggest, for instance, that microbial fitness is strongly gauged by regulatory constraints ( 6 , 7 ), biosynthetic costs ( 8 , 9 ), and the ability to respond to changing environments ( 10 , 11 , 12 , 13 , 14 ). As a consequence, the trade-offs arising from a complex multiobjective optimization often give a more accurate description of microbial growth than straightforward biomass maximization ( 15 , 16 , 17 ). Moreover, experiments characterizing bacterial growth at single-cell resolution have shown tight links between fitness and cell-to-cell variability ( 18 , 19 , 20 ). Growth rate distributions and metabolic fluxes indeed appear to be best captured by modeling such variability rather than assuming growth rate maximization ( 21 , 22 ), with the implication that trade-offs between metabolism and gene expression may affect not only bulk (average) properties but also the overall structure of a microbial population. While genome-scale models of metabolic networks can account for some of these facts ( 23 , 24 , 25 , 26 ), the maximization of biomass output remains a key conceptual premise. Addressing the question of “what cells actually want” ( 27 ) requires in essence to reverse the usual theoretical pipeline and infer from empirical data (reaction fluxes, growth rates, nutrient intake rates, etc.) 1) how the flow of metabolites through the metabolic network is organized and 2) whether some objective function is optimized. In this work we develop a framework to learn the probability distribution of metabolic phenotypes (namely, whole network flux configurations) using mass-spectrometry data ( 28 ) to inform a constraint-based model of E. coli ’s metabolism. This approach differs significantly from previous inference studies, specifically those of Refs ( 21 , 22 ), where the probability to observe a certain phenotype was effectively assumed to be a Boltzmann-like exponential function of its growth rate. No such assumption is made here. Rather, for each experimental sample, we compute the most likely distribution of phenotypes compatible with flux data resorting to the maximum entropy (MaxEnt) principle. In a nutshell, this approach prescribes that, if one has to infer a probability distribution subject to a given set of constraints, the distribution having the largest entropy provides the best guess, in the sense of being closest to uniform (i.e., minimizing its divergence from the uniform distribution), thus avoiding the introduction of biases that are not needed to accommodate constraints (see ( 29 ) for a simple introduction to this idea). Each inferred distribution is then characterized via 1) its mean biomass output (a proxy for the fitness) and 2) its “information content,” a global measure of cell-to-cell variability introduced in ( 21 ) that in essence quantifies the “volume” of the space of allowed phenotypes over which the inferred distribution is spread, with high information content corresponding to small volume. Fitness and information values inferred at different glucose levels appear to draw a well-defined curve in the fitness-information plane, supporting the idea of a tight link between growth and (inferred) variability. As a benchmark, we compare this curve against a purely theoretical bound obtained by maximizing, in each condition, the mean biomass output at fixed information content (similar to a “rate-distortion curve” in information theory ( 30 )). We found that empirical populations qualitatively follow and slowly approach the theoretical limit as the growth medium gets richer. In other words, as the fitness (mean biomass output) increases, the inferred phenotypic variability tends to remain as large as possible. This quantitatively supports the idea that heterogeneity plays a key role in shaping the fitness of a microbial population.",
"discussion": "Discussion Biological significance of the information content While the idea that fitness and information content of flux distributions are interrelated seems rather natural, the physiological meaning of the latter is not obvious. Technically, I quantifies the deviation of the inferred distribution p ( v ; c ) from uniformity in a given medium. Small values of I imply that experiment-derived constraints do not significantly modify our previous knowledge of the flux distribution, corresponding to all flux vectors in the feasible space defined by the given uptake rates being equally likely. On the other hand, large values of I imply that the inferred likelihood has a small overlap with the uniform distribution in the same medium. In this sense, one can say that I provides a proxy for the amount of metabolic regulation required to grow in a given medium. We have seen that the information content per degree of freedom more than doubles as the growth rate goes from 0.05/h to about 1/h ( Fig. 1 \n D ). Such a gain is mainly due to a systemic fine-tuning of fluxes and correlations rather than to the tightened control of a few pathways, in agreement with evidence suggesting that system-wide rearrangements underlie response to changing carbon levels in E. coli ( 7 ). Our analysis also shows that inferred distributions exceed the minimum required information content by roughly 1 bit per degree of freedom in all growth conditions ( Fig. 1 \n D ). This suggests that the biochemical constraints used to define the feasible space (flux reversibility, ranges of variability, etc.) might be too conservative. Further ingredients affecting the metabolism of single cells, such as biosynthetic costs ( 25 ), might also reduce the feasible space and bring data closer to the theoretical bound. However, the gap may also indicate that population growth requires a minimum amount of regulatory information, in line with the idea that minimal complexity (as opposed to minimal number of components) is the defining characteristic of cells ( 42 ). That regulatory interactions and mechanical effects are crucial in determining E. coli ’s overall metabolic capabilities is indicated by the fact that they remain substantially unchanged after a large-scale removal of unnecessary genes ( 43 , 44 , 45 ). By contrast, they are significantly affected by the selective knockout of a small number of genes through which specific cellular tasks are optimized ( 46 , 47 , 48 ). In this sense, constraint-based models may be missing a substantial amount of regulatory interactions that would effectively reduce the size of the feasible space F . Identifying these constraints could bring empirical populations closer to the F-I bound and provide crucial hints about the nature of optimality in bacterial growth. It is finally important to remark that the F-I bound we define is fundamentally different from the fitness-information relationship derived in ( 49 ). In that case, one quantifies the information about nutrient availability that has to be encoded in the level of a nutrient-processing enzyme to achieve a given fitness. In our case, information is a measure of the high-dimensional space of flux configurations that is effectively accessible to the system. Limitations of the study Besides the information encoded in the network structure, the key physical assumption made in our inference is that metabolic networks are at a NESS described by the mass balance conditions alone. This means that we do not account for factors such as biosynthetic costs, molecular crowding, membrane occupancy, etc. All of these are likely essential for the metabolic behavior single cells. However, including them in an inverse model defined on F would necessarily require additional assumptions about how they are linked to metabolic fluxes. On the technical side, our study is limited by two not-easily avoidable facts. 1) The data sets we used are not homogeneous, so it is a priori difficult to consider one as a continuation of the other at different growth rates. Carrying out this study on a broader, unique fluxomic data set covering a large enough range of growth rates would likely yield a more clean-cut picture. That a consistent scenario can emerge despite this limitation is in this respect quite remarkable. 2) In our framework, we implicitly interpret measured flux variances as proxies for the cell-to-cell variability. While this assumption has given consistent results when used in the context of single-cell data ( 21 , 22 ), a more detailed understanding of the sources of variability and error in fluxomics would allow to fine-tune the application of the MaxEnt scheme for the inverse problem considered here. It is, however, important to note that, while our approach is capable of efficiently representing the empirical variability, it cannot point to specific causal factors behind it. For this goal, different types of models (e.g., biochemically detailed dynamical models) are necessary. Relation to other approaches The most immediate comparison for our results is given by standard biomass maximization, which corresponds to the limit β → ∞ in ( 14 ). Previous work has shown that empirical data, including distributions of elongation rates in exponentially growing populations and measured fluxes, are better described using ( 14 ) with finite β rather than its β → ∞ limit ( 21 , 22 ). Here, we have effectively quantified how close flux distributions inferred from data are to ( 14 ) in terms of fitness and information content. Another set of potentially related problems concerns the experiment-guided determination of an objective function for constraint-based models. Different techniques have been proposed in the past to infer objectives or discriminate between various alternatives ( 50 , 51 , 52 , 53 , 54 , 55 ). While the vector c of Lagrange multipliers does partially align with the biomass output, our analysis does not highlight a clear objective function for constraint-based models. On the contrary, our results support the idea that the growth of bacterial populations is governed by a trade-off between mean single-cell biomass and heterogeneity. Notice that optimizing the mean biomass over time provides individual cells with an effective way to cope with multiple sources of variability ( 56 ). For instance, bacteria in fluctuating environments may be unable to adjust fluxes to the distribution that maximizes the instantaneous biomass synthetic rate due to the biosynthetic cost of the regulatory machinery implementing the adjustments. Regulatory programs selected over longer timescales would essentially optimize the frequency with which metabolism is adjusted in varying conditions. Finally, we note that MaxEnt-based models of metabolic networks have been employed in the past for a variety of purposes: to guide the decomposition of flux configurations into physiologically significant modes ( 57 , 58 ); explain the variability observed in bacterial populations ( 21 , 59 ) and continuous cell cultures ( 60 , 61 ) (note: the latter article appeared while the present paper was under review); reproduce empirical data on fluxes ( 22 ); derive dynamic strategies of cellular resource allocation ( 62 ); or predict response times to changing environments ( 63 ). While also based on the MaxEnt principle, the work presented here faces the question of heterogeneity from a rather different viewpoint, aiming essentially at bridging the gap between optimization-based methods and empirical results by building an efficient representation of metabolic data using constraint-based models. Our hope is that such an approach will lead to new theoretical insights into the nature and optimality of bacterial growth."
} | 3,630 |
37092565 | PMC10323669 | pmc | 7,259 | {
"abstract": "Abstract Ionogels prepared from ionic liquid (IL) have the characteristics of nonevaporation and stable performance relative to traditional hydrogels. However, the conductivities of commonly used ionogels are at very low relative to traditional hydrogels because the large sizes of the cation and anion in an IL impedes ion migration in polymer networks. In this study, ultradurable ionogels with suitable mechanical properties and high conductivities are prepared by impregnating IL into a safe, environmentally friendly water‐based polyurethane (WPU) network by mimicking the ion transport channels in the phospholipid bilayer of the cell membrane. The increase in electrical conductivity is attributed to the introduction of carboxylic acid in the hard segment of WPU; this phenomenon regularly arranges hard segment structural domains by hydrogen bonding, forming ionic conduction channels. The conductivities of their ionogels are >28–39 mS cm −1 . These ionogels have adjustable mechanical properties that make the Young's modulus value (0.1–0.6 MPa) similar to that of natural skin. The strain sensor has an ultrahigh sensitivity that ranges from 0.99 to 1.35, with a wide sensing range of 0.1%–200%. The findings are promising for various ionotronics requiring environmental stability and high conductivity characteristics.",
"conclusion": "3 Conclusion In summary, we have designed and efficiently prepared conductive ionogels. We introduce carboxyl groups into the hard segment domain of WPU to reduce the obstruction of the WPU hard segment domain on ionic migration by mimicking the principle of the cell membrane ion channel, increasing the ion conductivity (28–39 mS cm −1 ). Moreover, we find that the increase in IL content reduces the contents of hydrogen bonds and hard segment domains in WPU. As a result, for WPU ionogels, the ion conductivities increase and the moduli decrease (0.6–0.1 MPa), exhibiting mechanical properties similar to those of human skin. Using IL as solvents enable ionogels to remain stable over a long period relative to conventional hydrogels. The ionogels have an ultrahigh sensitivity ranging from 0.99 to 1.35, with a wide sensing range from 0.1 to 200%. In addition, strain testing of the ionogels shows that the ionogels have almost no hysteresis and creep, allowing the accurate monitoring of various human movements. Moreover, there is no solvent volatilization after long‐term exposure to air, and it works for a long period. Due to the ultrahigh conductivities of ionogels, their potential applications as conductive layers and flexible electrodes for capacitive sensors is demonstrated. This work provides new insights and practical methods for the molecular design and fabrication of ionogel polymer matrices, with broad application prospects in wearable electronics, biomedicine, and artificial intelligence.",
"introduction": "1 Introduction In the last decade, the increasing interest in human–computer interactions and health monitoring has accelerated the development of flexible electronics. [ \n \n 1 \n , \n 2 \n , \n 3 \n \n ] In the early days of this field, metal, and graphite conductive electrodes were used to prepare stretchable flexible electronic materials with flexible polymer elastomers, ensuring relatively high conductivity; however, their flexibility, sensing sensitivity, and range are relatively low, which is not suitable for monitoring human motion behaviors over a wide range. [ \n \n 4 \n , \n 5 \n , \n 6 \n , \n 7 \n \n ] Afterward, hydrogels became a new research topic for scientists as a new generation of flexible conductive materials. [ \n \n 8 \n , \n 9 \n , \n 10 \n , \n 11 \n , \n 12 \n \n ] This kind of material is soft and has a wide detection range. For organisms, mechanoreceptors release transmembrane Donnan potential through mechanical gating; ionogels change their physical properties through pressure drive. Hydrogels take ions as charge carriers; this phenomenon is similar to human nerve signal conduction. [ \n \n 13 \n , \n 14 \n \n ] However, the solvent of hydrogels is easy to volatilize, causing unstable shortcomings, such as electrical signal baseline that drifts over time. [ \n \n 15 \n \n ] \n Recently, there have been gels (or ionogels) made with ionic liquids (IL) in polymer networks without aqueous solvents. [ \n \n 16 \n , \n 17 \n , \n 18 \n , \n 19 \n \n ] An IL is a salt completely composed of cations and anions in liquid form at or near room temperature. IL substituents have high steric hindrance, preventing ions from stacking regularly. Relative to hydrogels, ionogels have nonvolatile properties and chemical and heat stability properties. [ \n \n 6 \n , \n 11 \n , \n 12 \n , \n 13 \n , \n 14 \n \n ] Ionogels have low conductivities because various charged carriers have large volumes and are easily hindered by the polymer matrix. [ \n \n 20 \n \n ] A large amount of evidence has shown that in typical polymer ionic conductor systems, the dominant ion motion occurs in the amorphous region through local sectional motion. [ \n \n 21 \n , \n 22 \n \n ] Carrier mobility is strongly linked to the glass transition of the polymer in the amorphous state. A low glass transition temperature ( T g \n ) usually results in conductive enhancement. [ \n \n 22 \n \n ] The polymer must have an amorphous structure at room temperature (or working temperature), or it must operate around its crystalline melting point to achieve a liquid‐like microscopic environment. However, it is difficult to use a single polymer with a low T g \n as a substrate for ionogels. These low T g \n polymers make the whole system appear as a viscous liquid when they swell at room temperature; the ionogels in this state do not maintain their shape or have stretchability. To date, physical blending, cross‐linking, or copolymerization are used to prepare the polymer matrices of ionogels to improve mechanical properties and conductivities. [ \n \n 23 \n , \n 24 \n \n ] Polyurethane is composed of a soft and hard segment. The soft segment refers to the polyol part in which the T g \n is lower than room temperature, giving the polyurethane better flexibility. The hard segment formed by the reaction of isocyanate and small molecule chain extender has high rigidity, strong polarity and high T g \n , endowing the polyurethane with certain strength and hardness characteristics. [ \n \n 25 \n , \n 26 \n , \n 27 \n \n ] Due to the thermodynamic incompatibility between soft and hard segments, polyurethane materials tend to aggregate and form independent microregions; this phenomenon presents a unique microphase separation structure. In this special structure, the hard segment region acts as a physical cross‐linking point to connect the polyurethane chain segments together. According to dissipation‐induced toughening theory, [ \n \n 17 \n , \n 28 \n , \n 29 \n , \n 30 \n , \n 31 \n , \n 32 \n \n ] the hard segment of polyurethane connecting the polyurethane polymer chains dissipates energy during deformation and promotes toughness. However, due to the high T g \n and ion incompatibility, the hard segment structural domain of polyurethane hinders ion migration and reduces the conductivity. [ \n \n 33 \n \n ] According to the theory proposed previously, the conductivity and mechanical strength of the gel are in conflict with each other. this conflict appears even more pronounced for IL with large steric hindrance. A type of mechanically robust ionogels have reinforced conductivity by introducing lithium ions that can migrate in the hard segment structural domain. [ \n \n 18 \n \n ] In this article, by introducing ion exchange sites into the forbidden zone of ion movement (the hard domain of polyurethane), the conductivity of the gel significantly improves, and solid formation of the gel is ensured. In the principle of the ion pathway on biological cell membranes, a pure phospholipid bilayer does not allow ions to pass through. Most of the amino acid side groups of sodium ion channel proteins contain carboxyl groups. These groups negatively charge the surfaces of their channels; thus, sodium ions are repeatedly combined and released, and sodium ions enter cells under an ion concentration gradient ( Figure \n \n 1 b ). Inspired by ion channel proteins, we introduce carboxyl groups into the hard segment domain of polyurethane to promote conductivity; the artificial ion exchange sites of carboxyl groups endow the smooth passage of ions through the hard segment of polyurethane under an electric field (Figure 1a,c ). The prepared polyurethane ionogels have high ionic conductivities of 37 mS cm −1 ; the tensile stress is 0.2 MPa and the strain is 200%, generating fast and repeatable sensing signals over a wide strain range. The high ionic conductivity opens more potential applications for ionogels in the future. Figure 1 Schematic illustration of the principle of conductive enhanced ion gels inspired by cell membrane ion channels. a) Hard domain of WPU without dihydroxy‐methylpropionic acid (DMPA) doping severely affects ion migration in the gel. b) We were inspired by the fact that ion channels in the cell membrane allow for ion migration. c) A carboxyl group was introduced into the hard domain of WPU to provide an ion migration site and effectively enhance ion migration in the gel. d) Ionogel physical picture (ionogels are attached to the back of the hand, and they stay attached with skin movement).",
"discussion": "2 Results and Discussion In this work, transparent ionogels with high ionic conductivities are achieved by uniformly distributing an IL of 1‐ethyl‐3‐methyl‐imidazold‐dicyandiamide ([EMIm][DCA]) in a polymerization network formed by green and nontoxic waterborne polyurethane (WPU). The hydrogen bond formed between WPU are adjusted by changing the content of [EMIm][DCA] to obtain ionogels with modulus values similar to those of human skin. WPU is synthesized in three steps (Scheme S1 , Supporting Information). First, a linear WPU prepolymer is synthesized by a condensation reaction of polytetrahydrofuran (PTMG) diol, polydimethylsiloxane (PDMS) diol, isoflurone diisocyanate and dihydroxy‐methylpropionic acid (DMPA). Second, 1,4‐butandiol (BDO) as a chain extender and 2‐hydroxy‐ethyl acrylate (HEMA) as an end‐capping reagent are added to the WPU prepolymer to form a WPU polymer with a double‐bonded end. Finally, the polymer with a double bond closure is subjected to radical polymerization to form a WPU solution (30 wt.%) with a cross‐linked structure. A WPU network consists of PTMG and PDMS soft and hard segments, including diisocyanate and a chain extender. The structure of the WPU is verified by Fourier transform infrared spectroscopy (Figure S2 , Supporting Information). [EMIm][DCA] has high electrochemical stability and is intersoluble with water in various proportions. The aqueous solution of WPU is mixed with the IL [EMIm][DCA] solution (30 wt.%) and cast on a polytetrafluoroethylene (PTFE) substrate; this process is followed by solvent evaporation at 25 °C and a humidity of 60% (Scheme S2 , Supporting Information). As the solvent evaporates, the nonvolatile IL and WPU gradually and slowly form ionogels, named WPU/IL \n x \n ( x is the mass ratio of IL:WPU). Figures 1d and S1a , Supporting Information, show photographs of WPU/IL 2.0 ionogels. To systematically discuss the correlations between the structures and properties of the WPU/IL \n x \n ionogels, we characterized the chemical structures of the WPU, IL, and ionogels by Fourier transform infrared (FTIR) spectroscopy (Figure S3 , Supporting Information). The absorption peak at 1026 cm −1 corresponds to the vibration absorption of Si—O–Si, the absorption peak at 1700 cm −1 represents the vibration absorption of C≐O, and the peak at 2126 cm −1 corresponds to the anion symmetric stretching vibration of [DCA] − . These peaks appear in the infrared absorption spectra of WPU/IL 2.0 simultaneously, indicating that WPU and IL successfully mix. To explore the strong interactions of hydrogen bonding between WPU and IL, carbonyl (CO) vibration absorptions of FTIR between 1580–1775 cm −1 on the WPU molecules are conducted for peak splitting ( Figure \n \n 2 d,e , and Figure S3 , Supporting Information). [ \n \n 25 \n \n ] Through deconvolution peak splitting, we obtain a C≐O ordered hydrogen bond vibration absorption peak at 1600 cm −1 , a C≐O disordered hydrogen bond vibration absorption peak at 1700 cm −1 and a C≐O nonhydrogen bond vibration absorption peak at 1722 cm −1 . The area of each absorption peak corresponds to the proportion of different vibration absorption modes. With increasing IL content, the ratio of the vibration absorption peak of the C≐O nonhydrogen bond gradually increases, potentially reducing the hard segment domain content. [ \n \n 24 \n \n ] \n Figure 2 a) Conductivities of WPU/IL \n x \n with various IL contents (x represents the mass ratio of PU and IL, x = 0.5 to 3.0). b) Influences of different DMPA contents on electrical conductivities. c) Comparison of conductivity and modulus values with other ionogels. d) Peak deconvolution of WPU/IL 2.0 . e) Hydrogen bond contents in the WPU/IL \n x \n values of samples. f) DSC curves of WPU and WPU/IL 2.0 ionogels. The ionogel conductivities with different IL contents are measured (Figure 2a ). With increasing IL content, the conductivities of the ionogels first increase from 2 to 39 mS cm −1 . To investigate whether the introduced carboxyl groups in the hard segment domain of WPU have the same effects as ion channels in the cell membrane. By reducing the DMPA contents to different values, a series of WPUs are synthesized (Figure 2b ), and the conductivities gradually increase with increasing DMPA hard segments. This result suggests that IL ions pass through the incompatible hard segment domain to improve the ionic conductivity. The conductivity of the enhanced ionogel is much higher than that of all of the ionic gel diagrams that have been previously reported (Figure 2c ). We have tested the alternating current (AC) impedance of the ionogels (Figure S12 , Supporting Information), and its Nyquist plot conforms to the electrochemical behaviors of traditional gels. The semicircular arc in the high‐frequency region is small, indicating that the impedances between the ionogels and the contact faces are low. To explore the ionic interactions between WPU and IL, the mixed emulsions of WPU and IL are measured by a dynamic light scattering (DLS) particle size analyzer (Figure S4 , Supporting Information). With increasing IL content, the particle size of the WPU colloid gradually increases, and the state of the WPU colloid changes from stable to unstable. In DLOV theory, the interactions between two latex particles are considered the sum of the van der Waals attraction and electrostatic repulsion characteristics. Ion binding between the IL liquid and carboxyl group on WPU reduces the electrostatic repulsion of WPU and the net forces between latex particles. X‐ray diffraction (XRD) data of WPU/IL 2.0 and WPU only show broad peaks at 20°; these findings are consistent with the XRD data of amorphous polymers, indicating that WPU/IL 2.0 ionogels are amorphous and IL is uniformly dispersed in WPU without crystallization (Figure S8 , Supporting Information). [ \n \n 34 \n \n ] Relative to the XRD pattern of pure WPU, the peak of WPU/IL 2.0 is wider, indicating that the content of the hard segment domain is lower. Due to the amorphous structure, WPU/IL 2.0 ionogels are optically transparent (Figure S1a , Supporting Information). By measuring the optical losses of WPU/IL \n x \n ionogels in different proportions (Figure S9 , Supporting Information), the gel transparency increases with increasing IL, which is evidence that IL opens the hydrogen bond structure of the WPU hard segment. The T g \n values of WPU and WPU/IL 2.0 ionogels are measured by differential scanning calorimetry (DSC) to further study the interaction between WPU and IL. Both WPU and WPU/IL 2.0 have T g \n values of 20—60 °C; this phenomenon is attributed to the hard segment of WPU (Figure 2f ). When the concentration of IL increases, T g \n shifts to a low value, indicating that the presence of IL impedes the formation of hydrogen bonds between WPU hard segments and regularly stacks the hard segments. [ \n \n 31 \n \n ] The low shift in the T g \n of the WPU hard segment is consistent with the results obtained by peak splitting in the FTIR data. To characterize the temperature stability and nonvolatilization characteristics of ionogels, the influences of temperature on the conductivities of ionogels are investigated using an oven and freezer layer (Figure S5 , Supporting Information). The resistance value of WPU/IL 2.0 is almost stable at ≈22 kΩ from −25 °C to 80 °C; however, the resistance value increases to 34 kΩ near −20 °C because IL is stable over a wide temperature range. Nevertheless, the melting points of [EMIm][DCA] at −5 °C and −20 °C limit the migration of [EMIm][DCA] inside the ionogel. The influence of humidity on ionogels is also explored, we find that humidity affects the stability of the resistance of ionogels (Figure S6 , Supporting Information). With the increase of humidity, the resistance value of the ionogels gradually decreases, which means that its conductivity is gradually increasing. To investigate the stability levels of ionogels in air, we compare the ionogels with some of the common conductive hydrogels by weightlessness tests (Figure S7 , Supporting Information). After 10 days of exposure to an external atmosphere environment (25 °C, 65 RH%), WPU/IL 2.0 does not absorb volatile or water components; polypyrrole (PPy), silk fibroin, and polyacrylonitrile (PAN) hydrogels lose all their moisture and conductive functions after one day. The mechanical properties of WPU/IL \n x \n are measured, and the tensile fracture toughness strength and Young's modulus values of the sample are extracted from the stress‒strain curves ( Figure \n \n 3 a , Figure S10 , Supporting Information). With increasing IL concentration, the elongation at break and Young's modulus decrease gradually (Figure 3b ). This phenomenon occurs because the IL reduces the content of the WPU hard segment domain. The WPU/IL 2.0 ionogels have Young's modulus values between 0.075 and 0.586 MPa; this range is similar to that of human skin. Additionally, the ionogels have elongation at break values of 200%. Mechanical relaxation experiments (Figure 3c ), stress recovery curves (Figure 3d ), and stress cycle curves at 30% strain (Figure 3e, f ) are performed on WPU/IL 2.0 under 100% strain to simulate various states of the ionogel working on the human surface. From Figure 3c , the ionogel retains 80% of its stress over a long period (1000 s) of tension; this phenomenon indicates that the ionogel maintains its shape for a long time without relaxation when deformed by human movement. Figure 3 a) Stress–strain curves. b) Young's modulus and stretchability of WPU/IL \n x \n . c) Mechanical relaxation profile of WPU/IL 2.0 after 100% strain; the inset is the tensile displacement‒time curve during stress loading and relaxation. d) Stress recovery curve of WPU/IL 2.0 . e) Typical tensile stress–strain curves of WPU/IL 2.0 . f) Residual strain and recovery ratio plotted versus the loading/unloading cycles. The stress recovery curve is completely restored to the original curve after a relatively short period after a large deformation of 100% (Figure 3d ). WPU/IL 2.0 ionogels’ cyclic tensile tests were carried out at high strain (100%) in 50‐cycle curve (Figure S11 , Supporting Information), it's ≈50% strain lag after 50th cycle, and the ionogels can restore to its original state by taking a short break. And the loading and unloading curves of WPU/IL 2.0 under low strain (30%) overlap and show a negligible lag (Figure 3e ). The excellent mechanical reversibility of WPU/IL 2.0 is quantitatively represented by the low residual strain and high recovery rate from the curves of Figure 3e; a low residual strain level (15%) corresponding to a recovery rate of 75% is observed in 50 consecutive cycles (Figure 3f ). Rapid deformation and recovery at low strain (30%) are ideal and important properties for sensory and drive applications, such as artificial muscle skin‐like sensors and tissue–electronic interfaces. Recently, soft and flexible skin‐like electronic devices have received increasing attention as a novel platform for integration with soft tissues for human–computer interactions, health monitoring, and medical treatment. [ \n \n 36 \n , \n 37 \n \n ] Ionic skin (I‐skin) is an ionotronic device in which ions replace electrons as the carrier of electrical signal transmission to simulate the sensing function of natural skin. From the mechanical properties of WPU/IL x , the modulus of WPU/IL 2.0 is similar to that of the human epidermis, and the strain elongation rate of 200% is sufficient for monitoring all parts of the human body, [ \n \n 1 \n \n ] while its conductivity is also at a high level. Figure \n \n 4 a shows the relative change in the resistance (∆ R / R \n 0 ) values of WPU/IL 2.0 ionogels as an I‐skin function of strain. The gauge factor (GF) values of strain sensors are the slopes of the strain sensors (∆ R / R \n 0 ) relative to the strain curve; this phenomenon is an important parameter for measuring the sensitivity of a strain sensor. Figure 4 Relative resistance changes of WPU/IL 2.0 I‐skins (strain sensor) as a function of a) strain, b) under small strains (3, 5, and 10%), and c) under large strains (30, 50, and 100%). d) I‐skin fixed on a wrist that rotates back and forth. e) Facial movement during breathing. f) Repeated bending/unbending movements of a finger. g) Heartbeats. h) I‐skin fixed on a leg joint bent at different angles. i) Cyclic stability tests of I‐skin under 30% strain for 200 cycles. The detection range of our ionogels is from 1% to 200%, and the detection range monitors all parts of human motion. In the small strain range (1% to 30%), the GF of the ionogel is 0.53. GF is 0.83 in the medium strain range of 30% to 100%. In the large strain range (100% to 200%), the GF of the ionogel is 1.38. Sensitivity is the most important sensor indicator. The sensitivity of the simplest piezoresistive sensor is described by Equation ( 1 ),\n \n (1) \n d R R = 1 + 2 μ ε − d σ σ \n where R is the resistance, σ is the ionic conductivity, µ is Poisson's ratio, and ε is the strain. From the equation, the sensitivity is related to the Poisson's ratio and the ionic conductivity of the material. We calculated the conductivities of the ionogels under dynamic conditions (Figure S13 , Supporting Information) and Poisson's ratios (Table S1 , Supporting Information). The µ of WPU/IL 2.0 ionogels equals 0.97. During stretching, the conductivities of the ionogels gradually increase. Ionogels may generate orientation during stretching to form high‐speed channels for ion transfer. [ \n \n 34 \n \n ] Therefore, the GFs of the ionogels are affected by both the deformation and conductivity. From the equation, we can find that the resistivity change rate ( dR / R ) is negatively correlated with the conductivity change rate ( dσ ), and the conductivity of WPU/IL 2.0 increases with the strain increase (Figure S13 , Supporting Information), which will reduce the GF of the sensor. To obtain higher GF, the effect of the conductivity change rate can be reduced by increasing the initial conductivity ( σ ). WPU/IL 0.5–2.0 ionogels were fabricated as strain sensors for the measurement of GF (Figure S14 , Supporting Information). It can be found that their GF decreases continuously with the decrease of conductivity, while the GF of WPU/IL 0.5 ionogels shows a negative value, which indicates that the change rate of conductivity is much higher than that of strain. This shows that increasing the conductivity of the polyurethane ionogel helps to increase its GF. In addition, the ionogels have a low detection limit (0.68%), even when monitoring the human heartbeat (Figure 4g ). This phenomenon suggests that WPU/IL 2.0 ionogels have the potential to be used as I‐skin. The WPU/IL 2.0 ionogels produce a stable ∆ R / R \n 0 signal during cyclic stretching as the mechanical strain gradually increases from 1% to 200%, indicating their high monitoring sensitivity and reliability (Figure 4b,c ). Figure 4i shows the ∆ R / R \n 0 signal of the I‐skin at 30% strain over 200 uninterrupted extension–release cycles. During expansion, the intensity of the ∆ R / R \n 0 signal is 38%, and the reduction is negligible after 200 extension–release cycles. This finding indicates that WPU/IL 2.0 ionogels have ultradurable and sensitive sensing performance levels as I‐skin. Next, WPU/IL 2.0 ionogels are used as wearable ionotronic devices to monitor various human movements. Figure 4d shows the signal of the wrist flip motion. Figure 4e shows the agitation of the cheeks during breathing. Figure 4f shows the change in the R / R \n 0 signal while gradually bending a straight finger. Figure 4h shows the changes in the R / R \n 0 signal when the knee is bent. The R / R \n 0 signal increases with increasing knee bending angle. When the knee bending angle is constant, the R / R \n 0 signal remains unchanged. Due to the good resilience characteristics of the ionogels, the R / R \n 0 signal returns to its original value when the knee straightens again. As shown in Figure 4h , I‐Skin monitors multiple cycles of wrist back‐and‐forth motion in real time; there is no lag in R / R \n 0 changes whether the ionogels are stretched or compressed. In addition, I‐Skin monitors very small human activities, such as heartbeats. When the I‐skin is attached to the left chest with commercial tape, the I‐skin generates a repeatable and reversible R / R \n 0 signal (Figure 4g ). In addition, the WPU/IL 2.0 ionogels were used as conductive gels to obtain electrophysiological signals. Figure \n \n 5 a shows the applications of the WPU/IL 2.0 ionogels in electromyography (EMG) and electrocardiograph (ECG). EMG and ECG signals were collected by BiosiganlsPULX Researcher, and the volunteer is fully voluntary and aware of any risks involved. Low‐modulus and high‐conductivity electrodes fit well to the skin surface and conduct electrophysiological signals, these characteristics are key to achieving a high signal‐to‐noise ratio (SNR) in electrophysiological measurements. To compare commercial and WPU/IL \n x \n ionogel electrodes, we collect the EMG signals of the forearm at 5 kg grip strength (Figure 5b ). Through calculation, ionogels electrode has higher power value. The ionogel electrode of WPU/IL 2 also has the highest SNR which is 8.2. Figure 5 WPU/IL \n x \n ionogels are used for EMG and ECG monitoring. a) EMG and ECG signal acquisition diagram. b) EMG and d) ECG acquisition contrast diagram of the WPU/IL \n x \n ionogels and commercial electrode. c) SNR and power values of commercial and WPU/IL \n x \n ionogels electrodes in EMG. e) Signal amplitude of commercial and WPU/IL \n x \n ionogels electrodes in ECG. With the decrease of electrical conductivity, the SNR gradually decreases to 5.3. The SNR of a commercial electrode is 6.3. Commercial and WPU/IL \n x \n ionogel electrodes were used to collect the ECG signals (Figure 5d ). As the IL content decreases, the collected ECG signals become more ambiguous. The difference between the highest value and the lowest value of the ECG signal is taken as the signal amplitude (Figure 5e ). WPU/IL 2.0 electrode can collect the strongest ECG signal. This is due to the high conductivity of WPU/IL 2.0 ionogels. To show the multifunctional characteristics of our ionogels, a capacitive pressure sensor is prepared, and the principle diagram describes the design principle of the piezoelectric capacitive pressure sensor. The sensor features a sandwich structure of WPU/IL 2.0 ionogels, silk fibroin microneedle, and ionogels. [ \n \n 38 \n \n ] The thickness of the whole sensor is ≈2 mm ( Figure \n \n 6 a ). The optical microscopic image of the dielectric layer shows that the pyramidal array microstructures on the surface of the silk fibroin film are 600 µm high, 200 µm wide, and 500 µm apart (Figure 6b ). When pressure is applied, the two gel electrodes move close to each other to change the capacitance value. When the capacitive pressure sensor is pressed with a 200 g weight (≈40 kPa), the response time is 1.7 s and the recovery time is 0.8 s. This result shows that the capacitive sensor has a fast response time and responds to pressure in a timely and sensitive manner (Figure 6c ). This capacitive sensor is applied to monitor a range of intermittent stimuli; this sensor has a quick deformation response, a quick recovery after force release, a short response time, and a short recovery time (Figure 6d ). Figure 6 a) Model diagram of a capacitive pressure sensor. b) Micrograph of the silk fibroin‐based dielectric layer. c) Sensor signal when a 200 g weight (40 kPa) is pressed. d) Sensor signal diagram of frequent pressing by hand."
} | 7,277 |
35795346 | PMC9251461 | pmc | 7,260 | {
"abstract": "With a growing world population and increasing frequency of climate disturbance events, we are in dire need of methods to improve plant productivity, resilience, and resistance to both abiotic and biotic stressors, both for agriculture and conservation efforts. Microorganisms play an essential role in supporting plant growth, environmental response, and susceptibility to disease. However, understanding the specific mechanisms by which microbes interact with each other and with plants to influence plant phenotypes is a major challenge due to the complexity of natural communities, simultaneous competition and cooperation effects, signalling interactions, and environmental impacts. Synthetic communities are a major asset in reducing the complexity of these systems by simplifying to dominant components and isolating specific variables for controlled experiments, yet there still remains a large gap in our understanding of plant microbiome interactions. This perspectives article presents a brief review discussing ways in which metabolic modelling can be used in combination with synthetic communities to continue progress toward understanding the complexity of plant-microbe-environment interactions. We highlight the utility of metabolic models as applied to a community setting, identify different applications for both flux balance and elementary flux mode simulation approaches, emphasize the importance of ecological theory in guiding data interpretation, and provide ideas for how the integration of metabolic modelling techniques with big data may bridge the gap between simplified synthetic communities and the complexity of natural plant-microbe systems.",
"conclusion": "Conclusion and Outlook The above narrative addresses the current state of metabolic modelling as applied specifically to plant-microbe-environment interactions, detailing some of the key challenges in expanding modelling applications to these complex systems. Merging computational and experimental approaches will improve our limited understanding of multi-faceted plant-microbe-environment relationships ( Figure 1 ). Advances will enable agricultural improvements, such as development of microbial inoculants to promote plant growth or resilience in the face of increasing global climate events. With these efforts, maintaining quality standards for modelling across the scientific community ( Carey et al., 2020 ; Lieven et al., 2020 ) will continue to be essential when extending models and algorithms to more complex plant-microbe systems. Envisioned future applications ( Figure 1 ) involve starting from a relatively small synthetic microbial community (e.g., less than 10 members), developing a genome-enabled model for each member, validating individual models with experimental data (as much as possible), and investigating in silico pairwise interactions between species. Successively larger community models can be constructed by including additional member(s) and comparing with the previous pairwise models to observe how predictions change as more members are added. Laboratory and greenhouse experiments testing different community compositions, environmental factors, or other variables of interest can be used to verify and refine the community model, with the ultimate goal of moving to more complex systems (e.g., field-scale validation in agricultural settings). With continued effort and progress, the general principles and pattern of interactions uncovered through this systematic process is anticipated to be transferable to other plant systems; the microbes involved in the interactions may be different but likely follow similar governing interaction rules and can be used to engineer agricultural solutions in many different crops.",
"introduction": "Introduction Plants have evolved intricate signalling networks sensing and responding to environmental stimuli. In recent years, the complex network of interactions existing between plants, microorganisms, and their environment has been recognized as an important factor impacting plant health and productivity ( Timm et al., 2018 ; Kumar and Verma, 2019 ; Harman et al., 2021 ). Plants rely on microbes for biologically available forms of essential nutrients ( Richardson et al., 2009 ); beneficial microbes also aid in protecting against pathogens and pests and are involved in plant response to environmental conditions such as heat and drought ( Bhattacharyya et al., 2016 ; Liu et al., 2020 ). The numerous multi-directional interactions that occur among plants, colonizing microorganisms, and environmental variables create difficulty in determining cause-and-effect relationships. For example, soil contains thousands of different microbial taxa ( Lennon et al., 2012 ), a subset of which are recruited for colonization by plant secretion of root exudates (e.g., Lebeis et al., 2015 ); in addition to chemical signalling factors from roots, environmental factors such as soil chemical properties, pH, moisture content, and temperature also affect microbial colonization ( Andrew et al., 2012 ; Islam et al., 2020b ), influencing both microbe-microbe and plant-microbe interactions. Synthetic communities provide a reductionist approach to help constrain biological factors, examining a limited number of microbial species with key roles in the community to better understand driving forces and their overall effect on the productivity and resilience of plant-microbe systems ( Vorholt et al., 2017 ). In the last decade, synthetic communities have become widely applied in a variety of contexts including agriculture ( De Roy et al., 2014 ; De Souza et al., 2020 ; Sgobba and Wendisch, 2020 ), recognizing their value in experimental hypothesis testing. Even with the tools provided by synthetic communities, an accurate understanding of multi-species, cross-domain interactions is still complicated by the complexity of each individual member’s metabolic network. Metabolic modelling provides mathematical predictions of metabolic routes used by an organism under different biological and environmental constraints and can thus help quantify the function of individual members and identify potential roles in community interaction on a metabolic level ( Bosi et al., 2017 ). This technique complements experimental findings with a computational aspect that can aid in elucidating interactions that may elude solely experimental approaches. This perspectives article provides insight into how metabolic modelling can be integrated with experiments, using ecological theories to mine the complexities of plant-microbe-environment interactions and illuminate underpinning rules that lead to increased productivity and robustness in these systems."
} | 1,672 |
26167106 | null | s2 | 7,261 | {
"abstract": "The spatial organization of cellular communities plays a fundamental role in determining intercellular communication and emergent behavior. However, few tools exist to modulate tissue organization at the scale of individual cells, particularly in the case of dynamic manipulation. Micromechanical reconfigurable culture achieves dynamic control of tissue organization by culturing adherent cells on microfabricated plates that can be shifted to reorganize the arrangement of the cells. While biological studies utilizing this approach have been previously reported, this paper focuses on the engineering of the device, including the mechanism for translating manual manipulation to precise microscale position control, fault-tolerant design for manufacture, and the synthetic-to-living interface."
} | 199 |
24492676 | PMC4041232 | pmc | 7,264 | {
"abstract": "This paper describes the photosynthetic response of a Roseobacter strain of marine aerobic anoxygenic phototrophic bacteria to an organic substrate limitation. In batch cultures, higher values of the spheroidenone/bacteriochlorophyll a ratio were observed under substrate-deficient conditions. Interestingly, the maximum photochemical quantum efficiencies of the photosystem under substrate-deficient conditions using blue or green excitation were significantly higher than those under substrate-replete conditions. These results indicate that spheroidenone, which can absorb green light, may play an important role in their photosynthesis as a light-harvesting antenna pigment, and the photosynthetic competence of the Roseobacter strain can increase in an organic substrate-deficient environment."
} | 201 |
36132283 | PMC9417820 | pmc | 7,265 | {
"abstract": "Magnetotactic bacteria Magnetospirillum gryphiswaldense MSR-1 biosynthesise chains of cube–octahedral magnetosomes, which are 40 nm magnetite high quality (Fe 3 O 4 ) nanoparticles. The magnetic properties of these crystalline magnetite nanoparticles, which can be modified by the addition of other elements into the magnetosome structure (doping), are of prime interest in a plethora of applications, those related to cancer therapy being some of the most promising ones. Although previous studies have focused on transition metal elements, rare earth (RE) elements are very interesting as doping agents, both from a fundamental point of view ( e.g. significant differences in ionic sizes) and for the potential applications, especially in biomedicine ( e.g. magnetic resonance imaging and luminescence). In this work, we have investigated the impact of Gd and Tb on the magnetic properties of magnetosomes by using different complementary techniques. X-ray diffraction, transmission electron microscopy, and X-ray absorption near edge spectroscopy analyses have revealed that a small amount of RE ions, ∼3–4%, incorporate into the Fe 3 O 4 structure as Gd 3+ and Tb 3+ ions. The experimental magnetic characterisation has shown a clear Verwey transition for the RE-doped bacteria, located at T ∼ 100 K, which is slightly below the one corresponding to the undoped ones (106 K). However, we report a decrease in the coercivity and remanence of the RE-doped bacteria. Simulations based on the Stoner–Wohlfarth model have allowed us to associate these changes in the magnetic response with a reduction of the magnetocrystalline ( K C ) and, especially, the uniaxial ( K uni ) anisotropies below the Verwey transition. In this way, K uni reaches a value of 23 and 26 kJ m −3 for the Gd- and Tb-doped bacteria, respectively, whilst a value of 37 kJ m −3 is obtained for the undoped bacteria.",
"conclusion": "4 Conclusions In this work, we have been able to modify the magnetic properties of magnetotactic bacteria M. gryphiswaldense by doping them with magnetic rare earth ions Gd 3+ and Tb 3+ . These rare earth ions are incorporated in a proportion of around 3–4% by substituting Fe 3+ in octahedral positions, a substitution that barely modifies the magnetosome shape, size and chain morphology. The magnetite Fd 3 m crystalline structure is either almost unaffected by the RE-incorporation, despite the large difference in the atomic radii between RE 3+ and Fe 3+ . However, the incorporation of Gd 3+ and Tb 3+ , even if low in content, clearly modifies the magnetic response of the bacterial magnetosomes. This alteration is especially noticeable in the thermal evolution of the magnetic anisotropy contributions (magnetocrystalline, shape, and interactions) of the doped bacteria, which differs from the undoped ones, especially below the Verwey transition (smooth K C reduction, smaller K uni ). The decrease in uniaxial anisotropy can be related to small differences in the size distribution and morphology of RE-doped magnetosomes, whereas the low temperature changes in magnetocrystalline anisotropy indicate a modification of the final monoclinic crystalline structure at low temperatures due to the substitution of Fe 3+ ions by Gd 3+ and Tb 3+ . All in all, this work paves the way towards a better understanding of how the magnetic response of magnetite nanoparticles in general, and bacterial magnetosomes in particular, can be modified by the incorporation of RE 3+ ions into the magnetite structure. The results presented in this study are not only of fundamental interest, but they also open the door to expanding the biomedical applications of magnetosomes by the incorporation of relevant RE ions.",
"introduction": "1 Introduction Magnetosomes are membrane-enclosed single-domain magnetic nanoparticles made of magnetite (Fe 3 O 4 ) or greigite (Fe 3 S 4 ) synthesised by magnetotactic bacteria (MTB). These magnetosomes are arranged in one or several chains inside the MTB, which allow the MTB to orient in water by means of the torque exerted by the Earth’s magnetic field on the chain. 1 The size and shape of magnetosomes strongly depend on the MTB species, their sizes generally being comprised between 35 and 120 nm, and they can be shaped as cube–octahedral, hexagonal prisms, arrows, etc. 2 The potential transfer of magnetosomes towards biomedical applications has boosted the interest of these biosynthesised magnetic nanoparticles in the recent years. Their high purity and crystallinity (referred to as high quality from hereunder), narrow size distribution, good biocompatibility, and relatively easy functionalization have made magnetosomes promising candidates as theranostic agents for magnetic hyperthermia, drug delivery, and magnetic resonance imaging (MRI), among other applications. 3–10 Furthermore, magnetosomes have also been considered as reliable models to investigate the relationship between the structural and magnetic properties of magnetite at the nanoscale, 11–14 a matter of debate that has attracted great interest for many years in the scientific community. To this respect, magnetosomes can be employed to investigate different relevant issues, including the survival of the Verwey transition of magnetite at the nanoscale or the role of shape anisotropy in faceted nanoparticles. 15–17 Nevertheless, despite these very promising features, magnetosomes present some drawbacks, especially when compared with their chemically synthesised counterparts. These include, for example, the restricted tunability of their shape, size and chemical composition, as these features are strictly genetically determined. 18,19 These restrictions constitute a nuisance when trying to modify the magnetic response of magnetosomes for different applications. 20–23 However, alternative routes have been devised to overcome some of these limitations. MTB exhibit a high affinity and specificity towards iron, which they extract from the medium in order to synthesise magnetosomes. In the same way, it has been demonstrated that MTB can also synthesise magnetosomes doped with some transition metals such as manganese, titanium, copper, or cobalt, 23–28 by adding limited amounts of these metals to the growth medium. There are however very few studies describing the incorporation of other elements, 29 which underlines the inherent complexity associated with the doping process. Among all the possible doping candidates, the incorporation of Rare Earth (RE) ions into magnetosomes would be considerably appealing. RE doping opens the door to modifying the internal structure and the magnetic properties of the nanoparticles, both being accomplished at the same time. Moreover, RE elements are currently used in several top-notch fields, such as biomedicine, catalysis, and/or solar cells. 30,31 The fascination towards RE does not stop at their potential biomedical and technological transfer, yet there is also room for the emergence of new magnetic phenomena. In this way, from a fundamental point of view, the large unquenched orbital angular momentum and high spin–orbit coupling of the 4f electrons in some RE ions can give rise to more pronounced magnetic features in comparison to transition metal ions. 32,33 Therefore, there is also a great potential for investigation on RE-doped magnetosomes, apart from the ones doped with transition metals. To the best of our knowledge, the only work that has been published in this area is the one by Shimoshige et al. 34 Specifically, they doped Magnetospirillum magneticum RSS-1 with Sm ions, obtaining core/shell magnetosomes made of magnetite in the core and samarium oxide in the shell. In our work, we have been able to incorporate, for the first time, Gd 3+ and Tb 3+ ions into magnetosomes from the Magnetospirillum gryphiswaldense strain MSR-1. Gd 3+ is a S-state ion ( L = 0) with seven unpaired electrons, which has been investigated, among other things, to develop gadolinium-doped iron oxide nanoparticles exhibiting a T 1 –T 2 dual-model MRI contrast. 35,36 On the other hand, Tb 3+ is an ion with six unpaired electrons, which has attracted attention for the possibility of providing magnetite nanoparticles with luminescence properties, which can be useful for monitoring the nanoparticles within the context of several biomedical applications. 36,37 Furthermore, the incorporation of Gd and Tb ions into the magnetite structure has also attracted attention due to the modulation of the magnetic properties of magnetite when the larger Gd 3+ and Tb 3+ ions are incorporated into its inverse spinel structure. 38 Bearing all these considerations in mind, we present here a combination of experimental and theoretical results to investigate the role of Gd 3+ and Tb 3+ cations in the magnetic response of magnetosomes. The morphological and structural properties of these RE-doped magnetosomes have been studied by transmission electron microscopy (TEM) and X-ray diffraction (XRD). The incorporation of the RE ions into the magnetosome structure has been investigated by X-ray absorption near edge spectroscopy (XANES) experiments, carried out in large scale Synchrotron facilities. In addition, the magnetic response of these doped magnetosomes has been thoroughly analysed, and compared with undoped magnetosomes, by using different experimental magnetic measurements, including zero-field cooling/field-cooling (ZFC/FC) curves and hysteresis loops ( M vs. H ). Finally, a modified Stoner–Wohlfarth model has been employed to simulate the experimental M vs. H loops. This has allowed us to pinpoint the specific magnetic changes taking place, and to relate these changes to the intrinsic modification of the effective anisotropies of these Gd- and Tb-doped magnetosomes.",
"discussion": "3 Results and discussion 3.1 Structural characterisation Transmission electron microscopy (TEM) was employed to study the size, shape, and arrangement of the RE doped magnetosomes. Fig. 1(a)–(c) show representative images of the magnetosomes extracted from the bacteria corresponding to the undoped, Gd-, and Tb-doped bacteria, respectively. We have included in Fig. S1 and S2 of the ESI † additional TEM images of the MTB and their magnetosome chains. In both Gd- and Tb-doped bacteria, the magnetosomes clearly exhibit the faceted cube-octahedral morphology typical of M. gryphiswaldense (see Fig. 1(a) ). However, some of these RE doped magnetosomes seem to present a less faceted morphology compared to their undoped counterparts (see Fig. S1 in the ESI † ). Along these lines, it has been reported that the presence of doping salts in the culture medium and the incorporation of the doping elements into the magnetosome structure can impose stress in the biomineralisation process. 25,27,28 In fact, similar shape irregularities have also been reported, for example, in Mn-doped magnetosomes. 27 Indeed, high-resolution transmission electron microscopy (HRTEM) or similar high resolution imaging techniques would be needed for quantitative analyses. We have also observed that the chains of magnetosomes inside the RE-doped MTB occasionally present minor irregularities and deformations, as depicted in Fig. S1 and S2 in the ESI. † Moreover, the RE doped bacteria tend to form larger chains (∼27 magnetosomes/chain) compared to the undoped bacteria (∼20) (see Table 1 ). Histograms accounting for the size-distribution of the magnetosomes are shown in Fig. 1(d)–(f) , together with the corresponding Gaussian fits. For the undoped magnetosomes, two size distributions can be observed, one centered around 47(8) nm and the other one centered at 22(8) nm. This double size distribution is typical of these M. gryphiswaldense bacteria, and accounts for the difference in size between the magnetosomes located at the ends of the chain (smaller) and those located at the inner positions (larger). Just in the same way, two size distributions are also observed for the Tb-doped magnetosomes, centered at 42(6) nm and 29(2) nm. However, only a single size distribution centered at 33(9) nm is obtained for the Gd-doped magnetosomes. What is clear according to these TEM analyses, is that the Gd- and Tb-doped bacteria tend to synthesise longer chains with smaller magnetosomes. A similar size reduction was also found for M. gryphiswaldense bacteria doped with other elements, such as Mn and Co. 26–28 A possible explanation for the presence of longer chains in RE doped bacteria could be that an increase in the magnetosomes/chain ratio would compensate for the reduction of the magnetic moment per magnetosome, given the smaller average size of the RE doped magnetosomes compared to the undoped ones. As a result, the net magnetic moment per chain would remain similar in both cases. However, further work will be needed to confirm this. Fig. 1 Representative TEM images (a)–(c) of the magnetosomes (extracted from the bacteria), size-distribution histograms (d)–(f) and XRD patterns (g)–(i), together with Rietveld refinements, corresponding to the undoped, Gd- and Tb-doped bacteria, respectively. The size-distributions are fitted with Gaussian distribution. In (g)–(i), the position of the hkl reflections are marked below the XRD patterns in green lines. In all cases, the Fe 3 O 4 phase gives rise to the most intense peaks. XRD refinements for the undoped bacteria are reproduced from ref. 9 with permission. Insets in XRD show a representative TEM image of M. gryphiswaldense bacteria. Average (TEM) diameter, 〈 D 〉, number of magnetosomes per chain, N , and lattice parameter, a , for the undoped, Gd- and Tb-doped samples. The error in the average diameter corresponds to the standard deviation, σ Sample Undoped Gd-doped Tb-doped 〈 D 〉 (nm) 47(8), 22(8) 33(9) 42(6), 29(2) \n N \n 20 27 27 \n a (Å) 8.3985(2) 8.3598(3) 8.3815(1.1) X-ray diffraction (XRD) analyses were performed to detect the possible presence of internal structural changes in the RE-doped magnetosomes. Fig. 1(g)–(i) show the XRD patterns for the undoped, Gd- and Tb-doped bacterial samples, respectively, together with the corresponding Rietveld refinements 41 (background was effectively subtracted during the analysis). The obtained Bragg errors R B are always below 4% for Gd-doped, and 15% for Tb-doped samples, ensuring the reliability of the performed analysis. The peak identification of the XRD patterns, as shown by the vertical green bars below them, has confirmed the presence of magnetite (Fe 3 O 4 ) in both RE-doped bacteria (25.2(3)% content for Gd-, and 12.31(1)% for Tb-doped bacteria). Apart from the Fe 3 O 4 phase, the XRD patterns also present some reflections corresponding to NaCl (66.0(2)% for Gd-, and 68.66(1)% Tb-doped bacteria) and KCl salts (8.5(1)% for Tb-doped bacteria). These contributions come from the PBS medium employed for washing the harvested bacteria. Besides, a poorly crystallised contribution related to the GdCl 3 salt (8.8%) has also been shown in Fig. 1(h) . We must clarify that the XRD contributions of these additional salts are well differentiated from the one corresponding to the magnetosomes, and therefore they do not affect the analysis of the Fe 3 O 4 phase. Rietveld refinements shown in Fig. 1(g)–(i) (black colour) corroborate the presence of well-formed crystalline magnetosomes in the undoped and RE doped bacteria. The obtained lattice parameters for each ensemble are a = 8.3598(3) Å for Gd-doped, and a = 8.3815(1.1) Å for Tb-doped samples. These values are slightly reduced (<0.4%) with respect to the one typically reported for bulk Fe 3 O 4 ( a = 8.397 Å) 42 and undoped magnetosomes ( a = 8.3985(2) Å). 9 This slight contraction of the unit cell parameter could in principle seem counterintuitive, since the ionic radius of Gd 3+ (1.08 Å) and Tb 3+ (1.06 Å) is larger than that of Fe 3+ (0.63–0.78 Å) or Fe 2+ (0.92 Å). 36,43 Nevertheless, similar reductions in the lattice parameter have been reported in other RE-doped Fe 3 O 4 nanoparticles, and understood in terms of the RE-mediated strain 44 and/or surface stress. 45 Rietveld refinements also provide information on the mean diameter and microstrain. The obtained values of the mean diameter 〈 D 〉 of the magnetosomes are 34.8(2) nm for the Gd-doped, and 32.7(3) nm for Tb-doped samples, values that are in good agreement with the results obtained by TEM (see above), confirming the single crystalline nature of the magnetosomes. On the other hand, microstrain values of η = 1.92(9)% (Gd) and η = 3.9(1)% (Tb) have been obtained. These strain values indicate that the presence of both doping ions distorts the crystalline structure of the Fe 3 O 4 magnetosomes. Similar results have been reported for other doped magnetosomes. 22,25,26 At this point, TEM and XRD results have revealed that the crystalline structure of the magnetosome is mostly maintained despite the presence of Gd and Tb ions inside the Fe 3 O 4 lattice. This structural characterisation has been completed by investigating the incorporation of the Gd and Tb ions into the magnetosomes using XANES. XANES is a very powerful element-sensitive synchrotron technique that has provided us accurate information on the oxidation state and site occupancy of the Gd and Tb ions in the spinel structure of magnetite. 26,46 XANES experiments were carried out on Gd- and Tb-doped magnetosomes, extracted from the bacteria, both at the Fe–K and RE-L 3 edges. Since the XANES signal of the RE salts attached to the bacterial body is so large that it masks any signal due to the RE doped magnetosomes, this time we have worked with isolated magnetosomes instead of the whole bacteria in order to avoid this effect. Fig. 2(a) and (b) show the XANES spectra for the Gd- and Tb-doped magnetosomes at the Gd-L 3 (7243 eV) and Tb-L 3 (7514 eV) edges, respectively. The presence of a clear absorption edge for both samples is an indicator of the incorporation of both Tb and Gd into the magnetosome structure. Nevertheless, we cannot completely discard the possibility of the presence of some Gd/Tb salts attached to the membrane of the magnetosomes, despite the multiple washings to remove any remaining salts after extraction. It should also be noted that, regardless the low Tb-content would, in principle, have led to measuring the Tb-L 3 edge in fluorescence mode, the emission lines of Tb-L 3 overlap with the Fe–K ones, imposing the use of transmission measuring mode. Hence, the normalised transmission spectrum shown in Fig. 2(b) corresponds to an extremely low absorption jump. It is also noticeable that the high absorption white line for the RE-doped magnetosomes, associated with the number of holes in the 5d band (valence) and the location of the 5d states. The overall shapes of the XANES spectra resemble those of the reference compounds shown in Fig. 2 , i.e. , GdCl 3 and Tb(NO 3 ) 3 . The overlapping edge position is a clear-cut indicator of the oxidation state of the absorbing atom, 47 indicating that the oxidation state of the RE ions inside magnetosomes is that of RE 3+ . Fig. 2 (a) and (b) normalised Gd- and Tb-L 3 -edge XANES spectra for Gd- and Tb-doped magnetosomes, respectively. The corresponding XANES spectrum for the GdCl 3 and Tb(NO 3 ) 3 reference samples have been included for comparison. (c) and (d) normalised Fe–K edge XANES spectra of magnetosomes from Gd- and Tb-doped magnetosomes, respectively. The control spectrum ( i.e. , undoped magnetosomes) has been included as the reference. The insets show the pre-edge and edge regions in more detail. The incorporation of the RE 3+ ions into the Fe 3 O 4 structure of the magnetosomes is further confirmed by absorption measurements on the Fe K-edge. Fig. 2(c) and (d) show the Fe K-edge XANES spectra of the Gd- and Tb-doped magnetosomes, together with undoped magnetosomes. As shown, the Fe K-edge spectra for both Gd- and Tb-doped magnetosomes are very alike, and they qualitatively reproduce the shape of the spectrum recorded for the control magnetosomes. However, for both the Gd- and Tb- doped magnetosomes, there is a minor damping of the oscillations of the XANES spectra, that can be traced for instance, in the white line (∼7131 eV) and the valley (∼7160 eV) amplitudes. A similar damping has been observed in XANES of magnetite nanoparticles when the purity and/or crystallinity of magnetite is slightly reduced. 47,48 On top of that, a displacement towards lower energies of the absorption edge can be observed, while for the undoped magnetosomes, the absorption edge is located at 7123.1 eV, being its position shifted towards 7122.5 eV for both RE-doped magnetosomes. Therefore, this negative shift of 0.6 eV could indicate a reduction of the average valence state of Fe in the RE-doped magnetosomes. Considering that the difference between the edge position for Fe 2+ and Fe 3+ is ∼7 eV, we can estimate the valence of the RE-doped magnetosomes to be ≈2.55–2.57, while in the undoped MTB, the valence is 2.66. This estimation has been made taking into account the fact that the edge position depends linearly on the valence, therefore, a substitution of approximately 3–4% of Fe 3+ ions by Gd 3+ or Tb 3+ can account for the aforementioned valence reduction. In order to obtain a more accurate estimate, additional techniques such as X-ray magnetic circular dichroism may prove useful. At the same time, an increase in the pre-edge peak amplitude can also be observed for the RE-doped magnetosomes. This modification reflects a change of the symmetry around the Fe atoms, towards a more non-centrosymmetric site. Therefore, these results suggest a reduction in the number of the centrosymmetric octahedral sites occupied by Fe 3+ ions, as a consequence of their substitution by Gd 3+ and Tb 3+ ions. 3.2 Magnetic characterisation The magnetic response of the doped-MTB has been analysed by tracing the M vs. T and M vs. H dependence. All the measurements were performed in whole bacteria in order to minimise the effect of the interchain interactions, but also to allow a better comparison with the results obtained for undoped bacteria. 26,49 Fig. 3(a) shows the ZFC–FC curves of the RE-doped and undoped bacteria, shifted in the Y -axis for clarity purposes. Starting with the undoped sample, the M vs. T curves present a strong irreversibility in the whole temperature range studied, and a sharp transition in the ZFC curve around T V ∼ 105 K, which is also accompanied by a smaller peak in the FC curve. This transition corresponds to the well-known Verwey transition, constituting a fingerprint of the presence of stoichiometric magnetite. 13,50 Concerning the Gd- and Tb-doped bacteria, their overall M vs. T evolution is very similar to the one of the undoped bacteria. The ZFC–FC curves evidence clear irreversibility, and the presence of the Verwey transition is also evident, although it seems to be now slightly displaced towards lower T values, ∼95 K for the Gd-doped and ∼99 K for Tb-doped bacteria. This displacement becomes more evident by comparing the derivatives of the ZFC curves for the three samples, as shown in Fig. 3(b) . There, we can observe that the T V , marked by the point at which the derivative becomes null, is slightly shifted towards lower values for both the Gd- and Tb-doped bacteria. In addition, the peak of the derivative, which marks the onset of the transition, is broader, less intense, and also displaced towards lower temperatures for the RE-doped bacteria. It must be noted that the survival of the Verwey transition in magnetite nanoparticles is strongly dependent on the crystallinity and stoichiometry. Small changes in the magnetite structure, for example by doping with other elements or by creating defects/vacancies, 12,26,51 can quickly lead to the displacement and disappearance of this transition. Finally, on the low-temperature side, a strong paramagnetic contribution appears in both RE doped bacteria below T ∼ 25 K. This is caused by the presence of the Gd and Tb salts attached to the bacterial body, as shown by the TEM images (see Fig. S1 and S2 in the ESI † ). Fig. 3 (a) M – T curves measured following the ZFC–FC protocol for the undoped (cyan), Gd- (red) and Tb-doped (blue) MTB. Note that the curves are displaced in the Y -axis for clarity purposes. (b) Derivatives of the ZFC magnetisation curves (d M d T −1 ) of these three samples. In both (a) and (b), the measurements for undoped MTB are included for comparison purposes, and the position of the Verwey transition corresponding to the undoped bacteria is marked with a gray line. To further explore the magnetic behaviour of the Gd- and Tb-doped magnetosomes, we have also analysed their magnetic response as a function of the applied magnetic field. Hysteresis loops, M vs. H , have been measured at different temperatures, from 5 to 300 K, after a cooling process with either no applied magnetic field (zero-field-cooling, ZFC) or with an applied field of 1 T (field-cooling, FC). We have included several of these M vs. H loops in Fig. S3 of the ESI. † Here it can be seen that at 300 K, the M vs. H loops of the 3 samples (undoped, Gd-doped, and Tb-doped) are very similar, whereas clear differences emerge when decreasing the temperature, especially, below the Verwey temperature (∼100 K). Fig. 4 shows the thermal evolution of the most relevant hysteresis parameters, i . e ., the coercive field, μ 0 H C (left panels), and the magnetisation remanence, normalised by the saturation magnetisation, M r / M s (right panels). These have been measured under ZFC (top), and FC (middle) protocols, to finally compare them by plotting the difference (in absolute value) between the FC and ZFC values (bottom). There, in all cases (ZFC, FC and difference), it can be seen that either the coercive field or the remanent magnetisation corresponding to the doped and undoped bacteria no longer overlap below T V , getting more and more differentiated with decreasing temperature, all the way down to 5 K. The same happens for the M vs. H loops shown in Fig. S3 of the ESI † . Fig. 4 Evolution with T of the coercive field, μ 0 H C , ((a), (c) and (e)), and the normalised remanence, M r / M s ((b), (d) and (f)) for the undoped, Gd-, and Tb-doped bacteria. Samples were cooled under no field (ZFC, (a) and (b)), and under a field of 1 T (FC, (c) and (d)). (e) and (f) depict the difference between FC and ZFC measurements, in absolute value, of the μ 0 H C and the M r / M s values. We will now analyse the coercive field and remanence magnetisation in greater detail. Concerning the coercive field, both the ZFC and FC coercive field curves for the doped bacteria [ Fig. 4 (a) and (c) ] remain nearly constant down to 100 K, with a lower μ 0 H C value (∼0.017 T) than the undoped bacteria (∼0.023 T). Then, the μ 0 H C slowly increases up to ∼0.024 T at 50 K, and finally rises more steeply reaching a value of ∼0.045 T at 5 K, again smaller than the one obtained for the undoped bacteria. The differences in coercivity between RE-doped and undoped magnetosomes can be seen more clearly if we focus on Fig. 4(e) , where the difference between ZFC and FC values, |Δ μ 0 H C | curves, is shown. The Verwey transition, delimited by the non-zero value of |Δ μ 0 H C |, is clearly defined around 107 K for the undoped bacteria, while in the case of the Gd- and Tb-doped samples, this transition is less abrupt and smoother. This result agrees well with the magnetic behaviour observed in the ZFC/FC M vs. T curves, indicating that the RE ions inside the magnetosomes, on one hand, reduce the effective anisotropy, and, on the other hand, slightly modify the Verwey transition due to minor structural changes. These results are further supported by the M r / M s curves shown in Fig. 4(b), (d), and (f) . There, it can be seen how the shoulders found at T ∼107 and 50 K for the undoped bacteria become smooth and broadened in the case of the RE-doped bacteria. Although some small differences can be observed between the values of the M r / M s curves of the Gd- and Tb-doped samples, when plotting the change of remanence |Δ M r / M s | in Fig. 4(f) , both curves overlap, as for |Δ μ 0 H C |. Following a similar evolution to that of the |Δ μ 0 H C | curves shown in (e), |Δ M r / M s | for the RE-doped magnetosomes remains very small (<0.005) down to 90 K. Then, it slowly starts increasing up to ∼0.01 at 50 K, and below that temperature, the increase becomes more abrupt, although the maximum values reached (∼0.18) are again smaller than those obtained for the undoped magnetosomes (∼0.24). A comparison between the thermal evolution of |Δ μ 0 H C | and |Δ M r / M s | of the undoped, Gd-doped, Tb-doped, and Mn-doped bacterial samples is presented in Fig. S4 of the ESI † . In order to shed light on the specific changes that are taking place in the intrinsic magnetic properties of the Gd- and Tb-doped samples, we have carried out magnetic simulations of the M vs. H loops measured at different temperatures. For this, we have employed a modified Stoner–Wohlfarth approach, which has been extensively described in our previous studies. 4,27 Briefly, the equilibrium configuration of the magnetic moment of each magnetosome is calculated as the sum of three contributions: (i) the magnetocrystalline anisotropy energy, E C ; (ii) the effective uniaxial anisotropy energy, E uni , arising from the competition between the magnetosome shape anisotropy and the dipolar interactions between magnetosomes inside the chain; and (iii) the Zeeman energy term, E Z . 26,52 In spherical coordinates, considering the 〈100〉 crystallographic directions of magnetite as the reference system, the total energy density is given by: 1 E ( θ , ϕ ) = E C ( θ , ϕ ) + E uni ( θ , ϕ ) + E Z ( θ , ϕ ) being 2 where θ and ϕ account for the polar and azimuthal angles of the magnetic moment of each magnetosome, respectively. K C and K uni stand for the magnetocrystalline and uniaxial anisotropy constants, respectively. The û i represents the unitary vector along the magnetic moment ( û m ), the uniaxial anisotropy vector ( û uni ) and the external magnetic field ( û H ) directions, respectively. As proved in previous studies, by SANS and electron cryotomography imaging, among other techniques, the û m forms an angle of ∼20° with the chain axis direction, 〈111〉. 15,52 Based on these considerations, the ZFC M vs. H loops at different temperatures have been simulated employing the dynamical approach already described in ref. 52 and 53 . The anisotropy terms, K C and K uni , have been adjusted to attain the best match between experimental M vs. H loops and the corresponding simulations. As shown in Fig. 5(a)–(f) , the calculated loops closely follow the experimental ones. The thermal evolution of K C and K uni for the undoped, Gd- and Tb-doped samples is shown in Fig. 5(g) and (h) . At room temperature, the values of K C for the three samples are similar: −11.0 kJ m −3 for the undoped and Gd-doped samples, and −12.0 kJ m −3 for the Tb-doped sample. These values are close to the theoretical K C value for bulk magnetite, −10.8 kJ m −3 . With decreasing temperature, | K C | (in absolute value) slightly increases for the undoped and Gd-doped bacteria, while it instead decreases for the Tb-doped sample, but overall, the change is small, remaining around | K C | ≈ 11–12 kJ m −3 . However, below 180 K, | K C | tends to decrease for the undoped sample, becoming null at 110 K, around the Verwey transition. This indicates that the role of the cubic magnetocrystalline anisotropy becomes negligible below T V for the undoped magnetosomes, as reported before. 26,27 A similar behaviour is also observed for the RE doped bacteria, but the drop in | K C | is displaced towards lower temperatures for both the Gd- and Tb-doped bacteria, becoming practically null at ∼90 and ∼100 K, respectively. This follows the same trend observed in the ZFC–FC curves, which indicates, again, that the incorporation of RE ions into the magnetite structure is modifying the Verwey transition. The particular differences in the K C values and evolution between Gd- and Tb-doped bacteria may be associated with differences in the incorporation of the Gd 3+ and Tb 3+ -ions into the magnetosome structure. Fig. 5 (a)–(f) Experimental (dots) and simulated (continuous line) M vs. H loops measured at T = 300, 90 and 30 K, for Gd- ((a), (c) and (e)) and Tb-doped bacteria ((b), (d) and (f)), respectively. (g) and (h) depict the thermal evolution of K C and K uni extracted from the simulated M vs. H loops included in (a)–(f). For comparison purposes, values from the undoped MTB have also been calculated and inserted. Concerning the uniaxial anisotropy term, K uni , it remains almost constant for the three samples down to T V : ∼11.5 kJ m −3 for the undoped and Gd-doped, and ∼10 kJ m −3 for the Tb-doped bacteria. In both cases, the K uni value is smaller than the one obtained for the undoped magnetosomes (∼12 kJ m −3 ). Being above the Verwey transition, K uni is mainly related to the effect of shape anisotropy and dipolar interactions, and this decrease could be ascribed to differences in the size distribution and/or morphology of the magnetosomes, as already observed in the TEM images. These differences are more appreciable in the case of the Tb-doped bacteria. Below T V , K uni increases substantially for the undoped bacteria, up to 37 kJ m −3 . However, for the Gd- and Tb-doped bacteria, the increase is slower, and the onset is not so well defined (90–100 K). Besides, the change in the slope obtained for K uni of the undoped magnetosomes below 50 K is also present in the Gd- and Tb-doped bacteria, but is less obvious, especially in the case of the Gd-doped bacteria. In the end, at 10 K, a maximum K uni value of 23 and 26 kJ m −3 is reached for the Gd- and Tb-doped bacteria, respectively. Qualitatively similar results were obtained in the case of Mn-doped bacteria. 27 All these results clearly indicate that there is a modification of both the magnetocrystalline and uniaxial anisotropies in RE-doped magnetosomes. The changes above the Verwey transition can be most likely associated to modifications in the shape/size of the RE-doped magnetosomes in comparison to the undoped ones. On the other hand, at low temperatures, the observed changes in the evolution of magnetocrystalline anisotropy, together with the less abrupt increase of the uniaxial anisotropy, lead us to conclude that the substitution of Fe 3+ ions by Gd 3+ and Tb 3+ is effectively modifying both the Verwey transition and the final monoclinic crystalline structure of the RE-doped magnetosomes. To the best of our knowledge, this is the first time that such a modification has been reported in the literature for magnetite nanoparticles doped with Gd or Tb."
} | 8,761 |
26418631 | PMC4796938 | pmc | 7,266 | {
"abstract": "Chemosynthetic symbiosis is one of the successful systems for adapting to a wide range of habitats including extreme environments, and the metabolic capabilities of symbionts enable host organisms to expand their habitat ranges. However, our understanding of the adaptive strategies that enable symbiotic organisms to expand their habitats is still fragmentary. Here, we report that a single-ribotype endosymbiont population in an individual of the host vent mussel, Bathymodiolus septemdierum has heterogeneous genomes with regard to the composition of key metabolic gene clusters for hydrogen oxidation and nitrate reduction. The host individual harbours heterogeneous symbiont subpopulations that either possess or lack the gene clusters encoding hydrogenase or nitrate reductase. The proportions of the different symbiont subpopulations in a host appeared to vary with the environment or with the host's development. Furthermore, the symbiont subpopulations were distributed in patches to form a mosaic pattern in the gill. Genomic heterogeneity in an endosymbiont population may enable differential utilization of diverse substrates and confer metabolic flexibility. Our findings open a new chapter in our understanding of how symbiotic organisms alter their metabolic capabilities and expand their range of habitats.",
"conclusion": "Conclusions Here, we have shown that a symbiont population with a single ribotype in an individual B. septemdierum host is composed of several heterogeneous subpopulations that differ in gene sets for key metabolic enzymes. Recently, genomic diversity in single nucleotide polymorphisms, short insertions/deletions and structural variants within microbial species has been also discovered in free-living populations and human gut microbes ( Gonzaga et al. , 2012 ; Grote et al. , 2012 ; Schloissnig et al. , 2013 ; Kashtan et al. , 2014 ). Our findings shed light on the ecological significance of the genomic diversity not only within symbiotic bacterial species, but also natural free-living or other animal-associated microbial communities.",
"introduction": "Introduction A wide variety of animals have acquired the ability to live on inorganic carbon sources by establishing symbioses with chemoautotrophic bacteria in various different habitats from shallow water to deep sea, such as hydrothermal vents. In the deep-sea vent symbioses, the symbiotic bacteria use the chemical energy of reduced sulphur, methane or hydrogen from vent fluid ( Dubilier et al. , 2008 ; Petersen et al. , 2011 ). These bacteria also require sources for assimilation such as carbon dioxide, methane, ammonia, nitrate and so on. The metabolic ability of the symbionts is largely affected by the geochemical conditions, whereas the physicochemical features of the hydrothermal vent environment are highly variable ( Zielinski et al. , 2011 ), and the availability of metabolic substrates is unpredictable. Due to the variable nature of hydrothermal vent environments, some host animals have multiple phylogenetically and physiologically distinct endosymbiont species allowing them to use a variety of energy sources, that is, conferring metabolic flexibility, and to occupy various habitats ( Fisher et al. , 1993 ; Distel et al. , 1995 ; Woyke et al. , 2006 ; Dubilier et al. , 2008 ; Robidart et al. , 2008 ; Beinart et al. , 2012 ). On the other hand, intra-species subtype variations with single nucleotide substitutions have been reported in some bacterial endosymbionts ( DeChaine et al. , 2006 ; Vrijenhoek et al. , 2007 ; Dubilier et al. , 2008 ). However, until recently, genomic diversity on the level of gene composition within a single symbiont species has been considerably underestimated, and we have known little about the ecological importance of such genomic variations. Here, we report a new and unique genome heterogeneity in a single thioautotrophic endosymbiont population in the deep-sea vent mussel, Bathymodiolus septemdierum . Whole-genome sequencing showed metabolically heterogeneous genomes in a B. septemdierum endosymbiont (BSEPE) population with a single type of 16S ribosomal gene in a single host individual. The symbiont population was composed of several subpopulations that possessed or lacked gene clusters for key metabolic enzymes such as hydrogenase and nitrate reductase. To confirm the generality of the heterogeneity in the BSEPE genome and to examine the proportions of the symbiont subpopulations harbouring each gene cluster within an individual host, we performed quantitative polymerase chain reaction (qPCR) in multiple hosts from two different vent sites. To investigate the distributions of the symbiont subpopulations in the host gill, we conducted in situ hybridization (ISH) analyses. In addition, to test whether the genomic variants corresponding to the symbiont subpopulations exist in the habitat, we performed PCR using DNA extracted from seawater near the Bathymodiolus mussel colony as a template. On the basis of our findings, we discuss possible scenarios for producing the heterogeneous symbiont subpopulations.",
"discussion": "Discussion Possible scenarios for producing the heterogeneous subpopulations Here we have shown that a symbiont population was composed of several subpopulations possessing one of the heterogeneous genomes, each of which had or lacked gene clusters for key metabolic enzymes such as hydrogenase and nitrate reductase. Petersen et al. (2011) reported that hydrogen could be used as an energy source by thiotrophic symbionts of Bathymodiolus mussels from the Mid-Atlantic Ridge. Nitrate respiration has been reported in some symbiotic bacteria ( Hentschel and Felbeck, 1993 ; Hentschel et al. , 1996 ), and it has been recently shown that nitrate reductase is also used for nitrate assimilation by chemoautotrophic symbioses ( Liao et al. , 2014 ). The different symbiont variants might have been produced by the gain or loss of gene clusters in specific symbiont populations. Recently, Kleiner et al. (2012) proposed that the ancestor of Bathymodiolus symbionts acquired hupL through lateral gene transfer. Our phylogenetic analysis of hupL gene products supports this idea. The hupL gene might have been laterally transferred into the lineage of the clade that includes Bathymodiolus symbionts and SUP05 ( Supplementary Figure S5A ). As the genomic synteny of the hup gene clusters was relatively well conserved in this clade ( Supplementary Figure S6A ), hup genes probably moved intergenomically as a cluster. The most parsimonious explanation for the phylogenetic relationship between the core genomes of chemoautotrophic sulphur-oxidizing symbiotic and free-living bacteria and their hydrogen oxidation cluster ( Supplementary Figure S6B ) is that the shared ancestor of the Bathymodiolus symbionts and free-living SUP05, as well as the Calyptogena symbiont, already harboured the hup gene cluster, which might have been gained by lateral gene transfer into the ancestral lineage. The hup gene cluster was not found in one metagenomic datum of SUP05 ( Walsh et al. , 2009 ) or Calyptogena symbiont lineages ( Kuwahara et al. , 2007 ; Newton et al. , 2007 ). These situations can be accounted for by loss of hup genes from these lineages. Similarly, the hup gene cluster was probably lost from some BSEPE populations. Thus, at least for hup genes, the genomic variations in BSEPE could have been produced by the loss of gene clusters in specific symbiont populations. In contrast to HupL, NarG from BSEPE, a SUP05 and a Calyptogena symbiont did not form a monophyletic cluster ( Supplementary Figure S5B ). However, because only three NarG sequences are currently available in these taxa and their close relatives, the evolutionary history of nar genes is difficult to estimate in the lineage into BSEPE from this limited data set. Our PCR data indicate that genomic variants of the symbiont also exist in the environment ( Figure 2 ). Then, the genomic variants of the symbiont in the environment might be horizontally transmitted in a host individual likely being unable to discriminate among the multiple metabolic subtypes ( Figure 5 ). In our qPCR, the proportion of the symbiont subpopulation harbouring hup or nar genes in juveniles (from Suiyo Seamount) was higher than that in adult specimens (from Myojin Knoll and Suiyo Seamount) ( Figure 3 and Supplementary Table S1 ). Also, the qPCR results indicated that the proportion of the symbiont subpopulation harbouring nar genes in specimens from Suiyo Seamount (including adults and juveniles) was higher than that in specimens from Myojin Knoll (adults only) ( Figure 3 and Supplementary Table S1 ). These results suggest that the composition of symbiont subpopulations varies depending on the environment (difference within a vent site over time or between two different vent sites) or host developmental stage. Symbiont localization in Bathymodiolus mussels has been reported to shift from a wide range of epithelia in early life stages (⩽9 mm) to only the gills in later life stages ( Wentrup et al. , 2013 ). Therefore, certain subpopulations may be selected in the process of establishing a symbiotic relationship in the gill epithelium during development. Alternatively, multiple infections of the symbiont over time may affect the composition of symbiont subpopulations. The cause of the variations in the composition of symbiont subpopulations remains to be determined. It is also not clear how the patchy distribution patterns of subpopulations are produced, and each bacteriocyte contains one or more symbiont subpopulations. Further investigations into the infection and distribution of symbiont subpopulations throughout host ontogeny, the proliferation of bacteriocytes and symbionts, and single-cell genome sequencing will improve our understanding of these questions. Genome heterogeneity may be beneficial for utilizing diverse energy substrates The lack of hydrogenase genes may be associated with the geochemical and geographical characteristics of the studied fields. Three decades of investigation into global hydrothermal fields have revealed a clear relationship between H 2 concentration of high-temperature hydrothermal fluid and tectonic background ( Nakamura and Takai, 2014 ). The H 2 concentration of the fluid venting at the Western Pacific Arc/Bac-Arc setting, including the Izu-Ogasawara Arc, is generally one or two orders of magnitude lower than at the Mid-Ocean Ridge setting, including the Mid-Atlantic Ridge and Eastern Pacific Rise, where the ability to use hydrogen as an energy source in symbioses is proposed to be widespread ( Petersen et al. , 2011 ). Moreover, compilation of calculations of bioavailable energy based on fluid chemistry and vent-endemic microbiological investigations ( Nakamura and Takai, 2014 ) indicates that the structure of chemolithotrophic microbial communities in hydrothermal environments is controlled primarily by the concentration of H 2 in the fluid. When the H 2 concentration is low, thiotrophic metabolism is favoured, whereas hydrogenotrophic metabolism is favoured when H 2 is high. Thus, in the Western Pacific region, where sulphur oxidation is more preferable, the BSEPE is likely to have a tendency to lose the hydrogenase. The loss of non-essential genes accompanying a size reduction in the bacterial genome is often attributed to minimization of the material costs of cellular replication, and to the genetic drift ( Mira et al. , 2001 ). In experimental bacterial populations, it has been proposed that loss of specific gene(s) is beneficial and driven by selection ( Lee and Marx, 2012 ). One drawback to the loss of genes that encode non-essential metabolic pathways is reduced flexibility in occupying variable geochemical regimes. In the BSEPE genome, all of the genes necessary for thiotrophy and aerobic respiration occur on loci with average mapping depth, and thus are present in all symbiont subpopulations, whereas genes for hydrogen oxidation and nitrate respiration are not essential for autotrophic growth. However, in the expression analysis, two structural hup genes ( hupL and -S ) and four structural nar genes ( narG, -H, -I, and -J ) were transcribed in the host gill, suggesting that these genes are functional in the symbiosis ( Supplementary Figure S3C and Supplementary Table S1 ), although actual activities of the enzymes need to be assayed to draw conclusion. In the hydrothermal vent environment occupied by Bathymodiolus mussels, sulphide and dissolved oxygen regimes appear to be unstable ( Zielinski et al. , 2011 ); therefore, the retention of variation for energy acquisition potentially increases the opportunities to colonize different and variable geochemical conditions. Thus, hup (and possibly nar ) genes may be conserved in specific subpopulations of BSEPE with functional activity, based on a subtle balance between the tendency to lose unnecessary genes and the tendency to retain flexibility to occupy a variable environment. The genome heterogeneity in BSEPE may permit the host-symbiont association to utilize diverse metabolic substrates. In ocean microorganisms, metabolic capability is deeply correlated with the organism's local acclimation and niche acquisition ( Kashtan et al. , 2014 ). It has been proposed that some host animals have physiologically distinct multiple endosymbiont species allowing them to use a variety of energy sources, which may confer metabolic flexibility and enable the organism to occupy a range of habitats ( Fisher et al. , 1993 ; Distel et al. , 1995 ; Woyke et al. , 2006 ; Dubilier et al. , 2008 ; Robidart et al. , 2008 ; Beinart et al. , 2012 ). The genomic heterogeneity at the sub-species level we found here may also enable differential utilization of diverse substrates, although this model remains to be validated. In horizontally transmitted endosymbioses, the acquisition of symbionts from a (potentially) diverse free-living population is expected to result in multiple symbiont subtypes coexisting within a single host individual ( Stewart and Cavanaugh, 2009 ). In line with this expectation, intra-species subtype variations in a host have been reported in some horizontally transmitted bacterial endosymbionts ( DeChaine et al. , 2006 ; Vrijenhoek et al. , 2007 ). However, until recently, little was known about the ecological significance of the genomic diversity within a symbiotic population. The findings presented here advance our understanding of metabolic acclimation and genomic evolution in symbiotic bacteria."
} | 3,664 |
33444471 | PMC8048573 | pmc | 7,268 | {
"abstract": "Abstract Biological processes rely on transient interactions that govern assembly of biomolecules into higher order, multi‐component systems. A synthetic platform for the dynamic assembly of multicomponent complexes would provide novel entries to study and modulate the assembly of artificial systems into higher order topologies. Here, we establish a hybrid DNA origami‐based approach as an assembly platform that enables dynamic templating of supramolecular architectures. It entails the site‐selective recruitment of supramolecular polymers to the platform with preservation of the intrinsic dynamics and reversibility of the assembly process. The composition of the supramolecular assembly on the platform can be tuned dynamically, allowing for monomer rearrangement and inclusion of molecular cargo. This work should aid the study of supramolecular structures in their native environment in real‐time and incites new strategies for controlled multicomponent self‐assembly of synthetic building blocks."
} | 251 |
35241967 | PMC8864493 | pmc | 7,270 | {
"abstract": "The plant faces different pedological and climatic challenges that influence its growth and enhancement. While, plant-microbes interactions throught the rhizosphere offer several privileges to this hotspot in the service of plant, by attracting multi-beneficial mutualistic and symbiotic microorganisms as plant growth-promoting bacteria (PGPB), archaea, mycorrhizal fungi, endophytic fungi, and others…). Currently, numerous investigations showed the beneficial effects of these microbes on growth and plant health. Indeed, rhizospheric microorganisms offer to host plants the essential assimilable nutrients, stimulate the growth and development of host plants, and induce antibiotics production. They also attributed to host plants numerous phenotypes involved in the increase the resistance to abiotic and biotic stresses. The investigations and the studies on the rhizosphere can offer a way to find a biological and sustainable solution to confront these environmental problems. Therefore, the interactions between microbes and plants may lead to interesting biotechnological applications on plant improvement and the adaptation in different climates to obtain a biological sustainable agricultures without the use of chemical fertilizers.",
"conclusion": "9 Conclusion and future perspectives The relationship between microorganisms and plants in the rhizosphere has been well understood recently. Differ ent microorganisms such as bacteria, fungi and archaea exert direct and/or indirect benefic effects on host plants. These beneficial effects include the improvement of plant growth and development, the enhancement against abiotic stresses, and the increase of resistance to biotic stresses. This positive association between these microorganisms and plants plays an important role in balancing the rhizosphere. However, the molecular mechanisms involved in these beneficial interactions are not well understood. Therefore, further investigations are required to establish molecular pathways by which microorganisms induce phenotypic changes in host plants and the use of these associations for improving the culture of plants as well as their adaptations to different conditions. In addition the study of quorum sensing mediators as molecular communication between plants and microbes could be a key element to understand these interactions and therefore to use them as biotechnological tools.",
"introduction": "1 Introduction The rhizosphere is the area of the soil influenced by plant roots, where plant roots and soil composition interact with each other ( Lynch and de Leij, 2012 ). It represents a dynamic hotspot for interactions between roots and beneficial, as pathogenic soil microbes. However, it is a gathering of several microorganisms such as bacteria, archaea, fungi, nematodes, protozoa, and other organisms that interact with each other, some are beneficial whereas others are harmful ( Pathan et al., 2020 ). May be considered beneficial or neutral, or harmful to the plant, depending on the specific microorganisms and plants involved and on the prevailing environmental conditions ( Jones and Hinsinger, 2008 ). Roots exudates such as sugars, amino acids, organic acids, phenolic compounds, enzymes, phytohormones, and vitamin can attract several microorganisms, and can also act as signal molecules mediating interactions in the rhizosphere ( Olanrewaju et al., 2019 ). The chemical signaling, between plant roots soil organisms, and the neighboring by plant roots may elicit dissimilar responses from different receivers ( Canarini et al., 2019 ). Indeed, the chemical components of root exudates may deter some microorganisms, while attracting another organism, may be classified as either positive associations (mutualistic or symbiotic associations) or negative associations (competition, parasitism among plants or pathogenesis) ( Bais et al., 2006 ). In general, in the rhizosphere, the negative associations express virulence on only a limited number of host species and it is estimated that only about 2% of the known fungal species are able to colonize plants and cause diseases ( Nihorimbere et al., 2011 ). Otherwise, the microbial interactions in the rhizosphere are often of benefit to plants, improve soil fertility, enhance the degradation of toxic chemicals ( Lynch and de Leij, 2012 , Xiong et al., 2020 ) and the secondary metabolites induction of the plant ( Chamkhi et al., 2021 ). However, root-associated microbiota in the rhizosphere play important roles and positively influence the health and the growth of their host plant through various mechanisms. The promotion of plant growth by microorganisms is based on a better acquisition of nutrients, hormonal stimulation and several direct or indirect mechanisms linked to plant growth, and could be involved in the reduction/suppression of plant pathogens ( Velázquez et al., 2005 , Berg, 2009 ). Plant-beneficial microbial interactions can be roughly divided into four categories: (i) the microorganisms in association with plants, are responsible for its nutrition, (ii) the microorganisms that stimulate plant growth indirectly by preventing the growth or activity of pathogens, (iii) the microorganisms responsible for direct growth promotion, for example, by the production of phytohormones ( Nihorimbere et al., 2011 , Okon et al., 2015 ). Indeed, the most root microbial associations in the rhizosphere are bacterial associations or rhizobacteria. They provide benefits to the plant resulting in its growth stimulation and are recognized as plant growth-promoting rhizobacteria (PGPR) ( Bais et al., 2006 , Okon et al., 2015 ). The PGPR can be divided into two groups according to their residing sites: first, symbiotic bacteria, which live inside the plant cells in produced nodules ( Hayat et al., 2010 ) as rhizobia-legume interactions leading to establishment of atmospheric nitrogen fixing symbiose in root nodules as for Sinorhizobium meliloti- alfalfa and Rhizobium leguminosarum- faba bean ( Bais et al., 2006 , Masciarelli et al., 2014 ). On the other hand, the second group is free-living rhizobacteria, which live outside the plant cells and did not produce nodules. But still, prompt plant growth promotors such as Azotobacter, Azospirillum, Bacillus , and Klebsiella sp. are also used as biofertilizers to inoculate a large area of arable land in the world to enhance plant productivity ( Dobbelaere et al., 2001 , Vessey, 2003 , Hayat et al., 2010 ). Indeed, the bacteria in the rhizosphere or rhizobacteria or plant promoting rhizobacteria (PGPR) can play an important role in the growth, the health, and in promoting nutrient acquisition by plants via several beneficial direct and indirect mechanisms ( Singh et al., 2011 ). Contrary to the rhizobacteria, the Archaea is much less in the rhizosphere. They were discovered especially in extreme environments, known to be essential actors in global processes, such as nitrification and ammonification in soils ( Leininger et al., 2006 ). Furthermore, plant-fungal interactions include mycorrhizal fungi (MF) that interact in symbiosis with the roots of the plant and endophytic fungi that live inside living tissue of leaves, stems, or roots ( Zeilinger et al., 2016 ). Likewise rhizobium bacteria, fungi can form symbiotic associations with plants, known as a mycorrhizal association, in which the interaction is between mycelial fungi and plants. Contrary to legume-rhizobia association, the mycorrhizal association is pervasive and can colonize nearly 80% of angiosperms and all gymnosperms plants. This mutualistic association can provide to the plant, phosphorus, water, and other micronutrient acquisitions by increasing the root surface. In return, the fungi receive fixed carbon from the host plant ( Datta et al., 2020 ). On the other hand, endophytic fungi have been detected in hundreds of plants and different studies demonstrated that they produce a large number of interesting secondary metabolites with interesting proprieties which can be used as a natural bioactive source ( Aly et al., 2010 , Chamkhi et al., 2018 ). As mentioned earlier, the rhizosphere is a hotspot gathering of several organisms such as protozoa and nematodes, that can play an important role complementary to the role of bacteria and fungi, as the remobilization of nutrients from consumed bacterial biomass, the nutrient mineralization in soil, and enhanced plant N uptake, and offered a strong stimulation of lateral root growth in presence of protozoa ( Bonkowski and Clarholm, 2012 ). This review focuses on the rhizosphere, particularly microbe and root interactions, the principal interactions that could play a very important role in the growth and the health of the plants and their applications, including some examples of how these interactions can be affected and used to improve crops of sustainable agriculture."
} | 2,216 |
22347878 | PMC3274759 | pmc | 7,272 | {
"abstract": "The Gram-negative bacterium Sideroxydans lithotrophicus ES-1 (ES-1) grows on FeCO 3 or FeS at oxic–anoxic interfaces at circumneutral pH, and the ES-1-mediated Fe(II) oxidation occurs extracellularly. However, the molecular mechanisms underlying ES-1’s ability to oxidize Fe(II) remain unknown. Survey of the ES-1 genome for candidate genes for microbial extracellular Fe(II) oxidation revealed that it contained a three-gene cluster encoding homologs of Shewanella oneidensis MR-1 (MR-1) MtrA, MtrB, and CymA that are involved in extracellular Fe(III) reduction. Homologs of MtrA and MtrB were also previously shown to be involved in extracellular Fe(II) oxidation by Rhodopseudomonas palustris TIE-1. To distinguish them from those found in MR-1, the identified homologs were named MtoAB and CymA ES-1 . Cloned mtoA partially complemented an MR-1 mutant without MtrA with regards to ferrihydrite reduction. Characterization of purified MtoA showed that it was a decaheme c -type cytochrome and oxidized soluble Fe(II). Oxidation of Fe(II) by MtoA was pH- and Fe(II)-complexing ligand-dependent. Under conditions tested, MtoA oxidized Fe(II) from pH 7 to pH 9 with the optimal rate at pH 9. MtoA oxidized Fe(II) complexed with different ligands at different rates. The reaction rates followed the order Fe(II)Cl 2 > Fe(II)–citrate > Fe(II)–NTA > Fe(II)–EDTA with the second-order rate constants ranging from 6.3 × 10 −3 μM −1 s −1 for oxidation of Fe(II)Cl 2 to 1.0 × 10 −3 μM −1 s −1 for oxidation of Fe(II)–EDTA. Thermodynamic modeling showed that redox reaction rates for the different Fe(II)-complexes correlated with their respective estimated reaction-free energies. Collectively, these results demonstrate that MtoA is a functional Fe(II)-oxidizing protein that, by working in concert with MtoB and CymA ES-1 , may oxidize Fe(II) at the bacterial surface and transfer released electrons across the bacterial cell envelope to the quinone pool in the inner membrane during extracellular Fe(II) oxidation by ES-1.",
"introduction": "Introduction The contribution of Fe(II)-oxidizing bacteria (FeOB) to iron cycling in freshwater, groundwater, and marine environments, as well as in most soils and sediments, has been well recognized (Emerson et al., 2010 ). A variety of neutrophilic and acidophilic Fe(II)-oxidizing microorganisms have the ability to derive energy for growth from the oxidation of dissolved or structural Fe(II) under either oxic or anoxic conditions. Unlike aerobic acidophilic or anaerobic neutrophilic FeOB, the geologic importance of aerobic neutrophilic FeOB has long been neglected because of the rapid auto-oxidation of Fe(II) by O 2 at circumneutral pH. However, recent studies indicate that aerobic neutrophilic FeOB would play a key role in microoxic niches with low levels of O 2 concentration, where microbial Fe(II)-oxidation can compete with the chemical oxidation of Fe(II). For example, the Gram-negative bacterium Sideroxydans lithotrophicus ES-1 (ES-1), originally isolated from the ground water with Fe(II) at neutral pH in MI, USA, grows on FeCO 3 or FeS at oxic–anoxic interfaces (Emerson and Moyer, 1997 ; Emerson et al., 2007 ). ES-1 does not grow on Mn(II) oxides, sulfide, or organic carbon sources, such as acetate, pyruvate, and glucose, and does not reduce Fe(III) oxides. Fe(III) (oxy)(hydr)oxide precipitates are closely associated with the ES-1 cells, but do not form sheath- or stalk-like structures (Emerson and Moyer, 1997 ). Recently, the ES-1 genome was sequenced ( http://www.ncbi.nlm.nih.gov/sutils/genom_table.cgi ). However, the molecular mechanism by which ES-1 oxidizes Fe(II) remains unknown. Because Fe(III) oxides produced from biotic Fe(II)-oxidation are usually sparingly soluble at circumneutral pH and in the absence of complexing ligands, bacteria oxidize Fe(II) extracellularly presumably to avoid accumulation of Fe(III) oxides inside their cells. To overcome the physical separation between the bacterial inner membrane where microbial oxidases are located and bacterial cell surface, FeOB have evolved different electron transfer pathways that link the inner membrane to the cell surface. The pathways identified to date include Cyc-2/Rus/Cyc-1 of Acidithiobacillus ferrooxidans , PioABC of Rhodopseudomonas palustris TIE-1 and FoxEYZ of Rhodobacter strain SW2 (Appia-Ayme et al., 1999 ; Yarzabal et al., 2002 ; Croal et al., 2007 ; Jiao and Newman, 2007 ; Castelle et al., 2008 ). Although these systems are phylogenetically unrelated, they all have at least one c -type cytochrome ( c -Cyt) as a key electron transfer protein. These c -Cyts work in concert with other proteins, often in the form of protein–protein complexes that can span the entire microbial cell envelope to facilitate electron conductance between the inner membrane and Fe(II) external to the bacterial cell. Notably, PioAB of R. palustris TIE-1 are homologs of MtrAB of the Fe(III)-reducing bacterium Shewanella oneidensis MR-1 (MR-1; Jiao and Newman, 2007 ). In MR-1, MtrA is a decaheme c -Cyt, while MtrB is a trans-outer membrane (OM), porin-like protein. They form a tight protein complex that transfers electrons across the OM to MtrC and OmcA, two OM decaheme c -Cyts that are localized on bacterial outermost surface (Ross et al., 2007 ; Shi et al., 2008 ; Hartshorne et al., 2009 ; Lower et al., 2009 ; Reardon et al., 2010 ). MtrABC and OmcA are key components of the MR-1 extracellular electron transfer pathway, which also includes a tetraheme c -Cyt CymA in the inner membrane. Together, they facilitate electron transfer from the quinone/quinol pool in the inner membrane across the periplasm, through the OM, to the surface of Fe(III) oxides (Richardson, 2000 ; Shi et al., 2007 , 2009 ; Fredrickson et al., 2008 ). In addition to mediating electron transfer to and from Fe, MtrAB homologs are also involved in extracellular reduction of dimethylsulfoxide by MR-1 and are hypothesized to be the prototype of a model system for electron transfer across the bacterial OM (Gralnick et al., 2006 ; Hartshorne et al., 2009 ). To investigate the molecular mechanism used by ES-1 for oxidizing Fe(II), we searched the ES-1 genome for the homologs of Cyc-2/Rus/Cyc-1 of A. ferrooxidans , FoxEYZ of Rhodobacter strain SW2 and PioAB/MtrAB. This search identified a three-gene cluster that encoded MtrA, MtrB, and CymA homologs. To distinguish them from those found in Fe(III)-reducing bacteria, we named the identified homologs MtoAB and CymA ES-1 . Cloned mtoA partially complemented an MR-1 mutant without MtrA in ferrihydrite (FH) reduction. Recombinant MtoA was purified following overexpression in MR-1 cells and characterized systematically. Purified MtoA was found to be a decaheme c -Cyt and able to oxidize soluble Fe(II) in vitro . Collectively, these results suggest that MtoA is a Fe(II)-oxidizing protein that works in concert with MtoB and CymA ES-1 to mediate electron transfer reactions from the cell surface to the inner membrane during extracellular Fe(II) oxidation by ES-1.",
"discussion": "Discussion The increased oxidation of Fe(II)Cl 2 by MtoA from pH 7 to pH 9 may be attributed mainly to the expected increased concentration of hydroxylated species of Fe(II), such as Fe(OH) + and Fe(OH) 2 . Equilibrium speciation calculations based on Minteqa2 thermodynamic database (Allison et al., 1991 ) as plotted in Figure 7 B show the speciation of Fe(II; 18 μM) in 150 mM NaCl solution as a function of pH. Hexaquo ferrous ion is the dominant species in the pH range used in this study, but the amount of Fe(OH) + increases with increasing pH, especially when pH is greater than 7.5. Both thermodynamic calculations and experimental data indicate that hydroxylated species of Fe(II) have higher reactivity relative to hexaquo Fe(II) (e.g., Wehrli, 1990 ; Sedlak and Chan, 1997 ). OH − ligands in the inner coordination shell of Fe(II) increase its reducing potential and increase its oxidation rate, and therefore could increase reaction to product Fe(III) phases. The concentration change of Fe(II), including hexaquo Fe(II) and Fe(OH) + , at the end of pH-dependent experiments was compared to the initial Fe(OH) + concentration (Figure 7 C). It shows that the amount of Fe(II) oxidized by MtoA appears to correlate positively with the initial concentration of Fe(OH) + from pH 7 to pH 9. Notably, at pH 9, the initial concentration of Fe(OH) + is expected to be significantly higher than that at pH 8.5, but the amount of Fe(II) oxidized by MtoA only increases slightly. This discrepancy is caused, at least in part, by the fact that the average redox potential of MtoA decreases as pH increases. This decrease is particularly accelerated from pH 8.2 to pH 9.2. The decrease of the redox potential of MtoA from pH 8.2 to pH 9.2 could negatively offset the increase of Fe(OH) + concentration in the same pH range, which may have the net result of only a slight increase of the Fe(II) oxidized by MtoA at pH 9. These results imply that change of pH in the environments may also have significant influence on Fe(II) speciation, which in turn will strongly affect microorganism-mediated Fe(II) oxidation in their natural settings. Ligand types impact both reaction rates and equilibrium constants of redox reactions between Fe(II)-complexes and MtoA. The equilibrium speciation calculations showed that the dominant Fe(II) species in the four ligand systems is hexaquo Fe(II), Fe(II)–citrate − , FeOH–NTA − , and Fe–EDTA − , respectively, at a ferrous ion-to-ligand ratio of 1:10. The equilibrium constant for the half electron transfer reactions between Fe(II)–ligand and Fe(III)–ligand (Table 2 ) was calculated using the thermodynamic cycle that involves ligand detachment from Fe(II)–ligand complex, Fe(II) oxidation to Fe(III) and Fe(III)–ligand complexation: (4) Fe(II)–ligand - e - → Fe(III)–ligand ↓ ↑ - ligand + ligand ↓ ↑ Fe(II) - e - → Fe(III) Table 2 Relevant speciation reactions for calculating reaction-free energy at 25°C . Reaction Log K ( I = 0) Reference Fe(III) + e − → Fe(II) 13.00 Martell and Smith ( 1995 ) Fe(III)–3H + + 3H 2 O = Fe(OH) 3 −3.96 Cornell and Schwertmann ( 2003 ) Fe(II) + citrate 3− → Fe(II)–citrate − 5.89 Martell and Smith ( 1995 ) Fe(III) + citrate 3− → Fe(III)–citrate 13.43 Allison et al. ( 1991 ), Timberlake ( 1964 ) Fe(III)OH–NTA − + e − → Fe(II)OH–NTA 2− 0.82 Wang et al. ( 2008 ) Fe(III)–EDTA − + e − → Fe(II)–EDTA 2− 1.35 Wang et al. ( 2008 ) I , ionic strength . Reaction equilibrium constants in this electron transfer pathway are provided in Table 2 . For the case of FeCl 2 at pH 8, it is reasonable to assume that the concentration of produced Fe(III) was controlled by the solubility of FH [Fe(OH) 3 ] because organic ligand was not provided. The equilibrium constant for electron transfer from Fe(II) to FH (log K = 7.04; Table 1 ) was then calculated by combining the redox reaction of Fe(II)/Fe(III) and the formation reaction of FH (Table 2 ). The trend of calculated log K is consistent with that of the log K values fitted from the experimental data except for the Fe–citrate case (Table 1 ). The exception is likely because of incomplete understanding of the speciation in the Fe(III)–citrate system; two Fe(III)–citrate speciation models were assembled previously based on literature data (Liu et al., 2001 ) that involve completely different Fe(III) speciation. In this study, we used the speciation model 1 in Liu et al. ( 2001 ), which is also used in the Minteqa2 database (Allison et al., 1991 ). In this model, the log K for the equation: (5) Fe(III) + citrat e 3 - → Fe(III) - citrate is 13.43 Speciation calculations in our Fe(III)–citrate system using Eq.5, however, suggested that Fe(III)–citrate is not a stable species if Fe(III) is allowed to precipitate as FH. The alternative Fe(III)–citrate speciation model produced the same result. If the FH is used as the end product, then the overall reaction constant (log K ) from Fe(II)–citrate to FH at pH 8 is 1.15. With this value, the trend of the calculated equilibrium reaction constants was consistent with the value estimated from the experimental data. The estimated reaction rate constants ( k ) and equilibrium constants ( K ) were positively correlated, establishing a linear free-energy relationship for this system (Figure 9 ). This result implies that it is the reaction-free energy that mainly determined the observed initial reaction rates when it was far away from equilibrium. It is interesting to note that the rate constant order of Fe(II)-complex oxidation, Fe–citrate > Fe–NTA > Fe–EDTA, is the inverse order observed in the reduction of Fe(III) complexes by MtrC and OmcA, where the Fe(III)–ligand reduction rate was Fe(III)–EDTA > Fe(III)–NTA > Fe(III)–citrate (Wang et al., 2008 ). This is expected from the trend of the activation-free energies for the redox reactions between Fe-complex and proteins (Wang et al., 2008 ). It is also expected from the relationship of K = k f / k b , where k f and k b are the forward [Fe(II) oxidation] and backward [Fe(III) reduction] reaction rate constants, respectively, and K is the equilibrium constant. Using the estimated K and k , which is k f here (Table 1 ), the calculated k b = k f / K , is on the order of Fe–citrate < Fe–NTA < Fe–EDTA, which is consistent with those observed by Wang et al. ( 2008 ). These two studies therefore demonstrate the same effect of ligand complexation on the reaction rates for both reduction and oxidation reactions, indicating that complexing ligands will have a significant impact on the reaction rates of microorganism-mediated Fe(II) oxidation in the environments. Figure 9 Linear free-energy relationship in Fe(II)-complexes oxidation by MtoA . In MR-1, MtrAB form a tight protein complex on the OM where MtrB is proposed to serve as a sheath for embedding MtrA and MtrA mediates electron conductance across the OM, while CymA is located in the inner membrane where it recycles quinol back to quinone during extracellular Fe(III) oxide reduction (Hartshorne et al., 2009 ). In this study, we showed that MtoA was a functional Fe(II)-oxidizing protein with broad redox potential that was more positive than that of MR-1 MtrA, indicating that, like MR-1 MtrA, MtoA is also capable of mediating electron conductance across the OM, but with the direction opposite to that of MR-1 MtrA during metal-reducing conditions. In addition, ES-1 mtoAB–cymA is, to the best of our knowledge, the first example that the genes encoding MtrAB and CymA homologs are clustered together, which suggests that they may belong to the same operon and that their protein products may work together for mediating electron transfer reactions. Based on previous observations and the results from this study, we propose that, similar to MtrAB and CymA in MR-1 cells, MtoAB and CymA ES-1 are also located in the OM and inner membrane, respectively, where MtoA is embedded inside MtoB. However, the direction of MtoAB/CymA ES-1 -mediated electron transfer during Fe(II) oxidation (i.e., outside-in) by ES-1 is opposite to that of MtrABC/CymA-mediated reactions during Fe(III) reduction (i.e., inside-out) by MR-1. We propose that through its heme group(s) exposed to the extracellular environment, MtoA oxidizes Fe(II) directly and then transfers the released electrons across the OM to periplasmic proteins that have yet to be identified, which in turn relay the electrons to CymA ES-1 . CymA ES-1 uses the received electrons to reduce quinone to quinol in the inner membrane (Figure 10 ). Quinol then shuttles the electrons to the redox proteins in the inner membrane for reducing O 2 and/or NAD + . Figure 10 The proposed roles of MtoAB and CymA ES-1 in Sideroxydans lithotrophicus ES-1-mediated extracellular Fe(II) oxidation . Decaheme c -Cyt MtoA, which is inserted into the porin-like, outer membrane (OM) protein MtoB, oxidizes Fe(II) directly on the bacterial surface and transfers the released electrons across the OM to the periplasmic proteins that have yet to be identified. The periplasmic proteins relay the electrons through the periplasm (PS) to the tetraheme c -Cyt CymA ES-1 . CymA ES-1 , a homolog of NapC/NrfH family of quinol dehydrogenase that is located in the cytoplasmic or inner membrane (IM), reduces quinone to quinol. c -Cyts are labeled in red and direction of electron transfer is indicated by a yellow arrow. PioABC homologs are also found in the Fe(II)-oxidizing bacterium Gallionella capsiferriformans (Bonnefoy and Holmes, 2011 ), indicating the broad involvements of MtoAB/PioAB homologs in microbial Fe(II) oxidation. It should be noted that a group of proteins that show no homologous to MtoAB/CymAES-1 have recently been proposed to be involved in Fe(II) oxidation by the bacterium Mariprofundus ferrooxydans PV-1 and homologs of these proteins are also present in ES-1 (Singer et al., 2011 ). These results emphasize the uncertainty that remains regarding the electron transfer pathway(s) utilized by ES-1 for Fe(II) oxidation and the need for additional research. In summary, an mtoAB–cymA gene cluster is found in the genome of the Fe(II)-oxidizing bacterium S. lithotrophicus ES-1. Protein purification and characterization results confirm that MtoA is a decaheme c -type cytochrome and oxidizes soluble Fe(II). MtoA-mediated Fe(II) oxidation is pH- and Fe(II)-complexing ligand-dependent. It is proposed that, together, MtoAB and CymA ES-1 form a pathway for electron conductance from extracellular Fe(II) to the quinone pool in the bacterial inner membrane."
} | 4,422 |
38533191 | PMC10961847 | pmc | 7,273 | {
"abstract": "Microbial ureolysis\noffers the potential to remove metals including\nSr 2+ as carbonate minerals via the generation of alkalinity\ncoupled to NH 4 + and HCO 3 – production. Here, we investigated the potential for bacteria, indigenous\nto sediments representative of the U.K. Sellafield nuclear site where 90 Sr is present as a groundwater contaminant, to utilize urea\nin order to target Sr 2+ -associated (Ca)CO 3 formation\nin sediment microcosm studies. Strontium removal was enhanced in most\nsediments in the presence of urea only, coinciding with a significant\npH increase. Adding the biostimulation agents acetate/lactate, Fe(III),\nand yeast extract to further enhance microbial metabolism, including\nureolysis, enhanced ureolysis and increased Sr and Ca removal. Environmental\nscanning electron microscopy analyses suggested that coprecipitation\nof Ca and Sr occurred, with evidence of Sr associated with calcium\ncarbonate polymorphs. Sr K -edge X-ray absorption\nspectroscopy analysis was conducted on authentic Sellafield sediments\nstimulated with Fe(III) and quarry outcrop sediments amended with\nyeast extract. Spectra from the treated Sellafield and quarry sediments\nshowed Sr 2+ local coordination environments indicative\nof incorporation into calcite and vaterite crystal structures, respectively.\n16S rRNA gene analysis identified ureolytic bacteria of the genus Sporosarcina in these incubations, suggesting they have\na key role in enhancing strontium removal. The onset of ureolysis\nalso appeared to enhance the microbial reduction of Fe(III), potentially\nvia a tight coupling between Fe(III) and NH 4 + as an electron donor for metal reduction. This suggests ureolysis\nmay support the immobilization of 90 Sr via coprecipitation\nwith insoluble calcium carbonate and cofacilitate reductive precipitation\nof certain redox active radionuclides, e.g., uranium.",
"conclusion": "Conclusion and Environmental Significance While MICP\nby urease encoding bacteria have been investigated in sediments from\nother contaminated nuclear sites, this represents the first study\nof urea-facilitated MICP in Sellafield sediments and also quantifies\nthe impact of added electron donors/nutrients (i.e., acetate, lactate,\nand yeast extract) and electron acceptors (i.e., Fe(III)). To the\nbest of our knowledge, this is the first time that enhanced strontium\nremoval by urea-facilitated MICP has been demonstrated by indigenous\nsediment microbial communities. The rate and extent of concomitant\nSr 2+ and Ca 2+ removal at elevated pH varied\nbetween the sediments, and the biostimulation regime applied. Maximal\nremoval of Sr (170 ppm to subppm concentrations over the course of\n1 day) was noted in the PPQ sediment microcosm amended with urea and\nyeast extract. This system reached approximately pH 10 before extensive\nSr removal. Spherical Ca/Sr-bearing precipitates identified in the\nremediated PPQ sediments using ESEM and corresponding EDX analyses\nsuggested that Sr coprecipitated with the calcium carbonate polymorph\nvaterite. EXAFS analysis of the sediments further supported the hypothesis\nthat the end point of Sr remediation was indeed Sr-incorporated vaterite.\nThe RB27 system amended with urea and additional Fe(III) reached a\nfinal pH > 9.5, coinciding with a reduction in strontium concentration\nfrom 93 to 10 ppm over 1 day. Analyzing the remediated sediments in\nthe RB27 system using EXAFS and ESEM suggested that Sr was immobilized\nin calcite. While calcite is a more stable polymorph of calcium carbonate,\nvaterite precipitates more rapidly and is capable of incorporating\nmore Sr. 43 Thus, in addition to the sediment\ncharacteristics, the time scale and aqueous strontium concentrations\nare important considerations for perspective in situ remediation strategies when targeting strontium removal as various\nSr-associated carbonate phases. Anaerobic microcosms using Sellafield\nsediments RB23 and RB27 also showed an unexpected increase in microbial\nFe(III) with urea added. Enhanced Fe(III) bioreduction coupled to\nthe oxidation of NH 4 + formed during ureolysis\nmay have been responsible for enhancing Fe(II) production in these\nsystems. Accumulating sufficient NH 4 + from ureolysis\nmay, therefore, enhance further beneficial microbial activity for\nthe remediation of redox-active radionuclide species such as U, Np,\nand Tc. This work has significant implications for the remediation\nof 90 Sr at Sellafield through MICP, in addition to the\npotential cleanup of redox active radionuclides. The need for further\nwork to define the utility of these processes at a larger scale using\nflow-through systems more representative of the dynamic natural subsurface\nin order to assess the large-scale application and long-term performance\nof this technology is clear.",
"introduction": "Introduction Strontium-90 (half-life = 28.8 years)\ncomprises a significant portion\nof the radioactivity associated with fission products within a variety\nof radioactive wastes. 1 − 6 In addition, 90 Sr is present as a subsurface environmental\ncontaminant at a number of nuclear sites across the world, including\nHanford, USA, 7 , 8 and Mayak, Russia. 9 , 10 For example, at Sellafield, U.K., the unintentional release of 90 Sr has generated radioactive groundwater plumes which originate\nfrom storage units, as well as nitrate-neutralized wastes. 11 In addition, the environmental contamination\nby 90 Sr as a consequence of the accidents at Chernobyl,\nUkraine, and Fukushima Daichii Nuclear Power Plant, Japan, are well\ndocumented to have polluted water bodies and soils at these sites. 12 − 18 In order to remediate 90 Sr subsurface contamination,\nit is necessary to develop a toolkit of effective and sustainable\nremediation technologies in order to limit its migration and minimize\npotential environmental harm. In the hydrosphere, 90 Sr persists as Sr 2+ and exhibits similar biogeochemical\nbehavior to Ca 2+ in\nthe aqueous phase, as they both possess divalent charge and have similar\nionic radii. 19 , 20 At circumneutral pH, the mobility\nof Sr 2+ in soil and groundwater systems is typically controlled\nby outer-sphere adsorption and ion exchange to phyllosilicate clays\nand iron oxide mineral particles, 21 − 23 with the degree of adsorption\ncontrolled by a number of factors including pH and ionic strength. 24 , 25 The pH point of zero charge (pH pzc ) of these naturally\noccurring minerals ranges from approximately 2 to 5 for clay minerals\n(e.g., kaolinite and Illite) 26 − 28 to approximately 5 to 7 for iron\noxides, e.g., goethite (α-FeOOH) and lepidocrocite (γ-FeOOH). 29 , 30 At pH > pH pzc , the negatively charged surface sites\nenhance\nthe adsorption of cations such as Sr 2+ . At the same time,\nincreasing ionic strength often leads to a reduction in Sr 2+ uptake due to the competition for surface sites from other cations. 31 − 11 This can be particularly challenging given radioactive plumes in\nthe subsurface often contain elevated ionic strength attributed to\nreleased liquors. 34 At Sellafield, U.K.,\nthe transport of 90 Sr is hypothesized to be mediated largely\nby an adsorption process given that the groundwater pH is between\n6 and 8. 35 Modeling the partition coefficient\n( K d , a measure of contaminant mobility)\nfor 90 Sr sorption to Sellafield sediments has indicated\nits uptake is significantly reduced in high ionic strength MAGNOX\ntank liquors of pH 9–11.5 ( K d ∼\n40 L/kg) compared with circumneutral groundwaters at pH 6–8\n( K d ∼ 10 3 L/kg). 11 The migration of Sr 2+ in groundwaters\ncan be limited\nvia incorporation into mineral phases that actively form in the subsurface,\nvia either biotic or abiotic processes. The incorporation of Sr 2+ into Ca 2+ -bearing (bio)minerals results in a\nfar more recalcitrant, and less labile, end point compared with Sr 2+ outer-sphere adsorbed to minerals surfaces. However, few\nstudies have investigated Sr 2+ sequestration into Ca 2+ (bio)minerals forming under conditions applicable to contaminated\nland environments. The partitioning of Sr 2+ into calcium\ncarbonate minerals occurs within all environmentally relevant polymorphs\nincluding calcite, 37 aragonite, 38 and vaterite. 39 , 40 Studies investigating strontium uptake within chemically precipitated\ncalcite have reported concentrations ranging from several hundred\nppm 41 , 42 to several thousand ppm. 43 While biogenic aragonite 44 , 45 and calcite 46 , 47 have been shown to contain up to several thousand ppm of Sr, aragonite\nis able to retain far higher Sr concentrations. 48 Mechanistically, this is because the 9-fold coordination\nof divalent cations in aragonite allows for easier accommodation of\nlarger cations, such as Sr 2+ , than the 6-fold coordination\nenvironment in calcite. 49 Furthermore,\naragonite has a lower crystal symmetry compared to calcite, which\nmay contribute to increasing the availability of lattice sites for\nSr incorporation. 50 , 51 Sr K -edge X-ray\nabsorption near edge spectroscopy (XANES) and extended X-ray absorption\nfine structure (EXAFS) are key methods for understanding the Sr atomic\nscale local environment and the mechanism of uptake into calcium carbonate\nminerals. The use of EXAFS analysis to identify Sr–Ca coordination\nat radial distances approximately 3.9–4.1 Å in calcite 52 , 53 and aragonite 54 , 55 and 4.2 Å in vaterite 43 , 53 indicates the substitution of Sr 2+ for Ca 2+ into the crystal structure of calcium carbonate as the uptake mechanism. Microbially induced calcite precipitation (MICP) leads to the incorporation\nof Sr into calcium carbonate, which occurs in a variety of natural\nsettings including marine systems and terrestrial soils. 56 The MICP process is initiated by the metabolic\ngeneration of sufficient elevated alkalinity and high pH conditions\nrequired to precipitate calcium carbonate. Ureolysis is capable of\ninstigating MICP, which involves the degradation of urea [CO(NH 2 ) 2 ] by bacterial urease enzymes to form ammonia\n(NH 3 ) and carbonic acid (H 2 CO 3 ).\nUnder circumneutral groundwater conditions, NH 3 and H 2 CO 3 then equilibrate as ammonium (NH 4 + ) and bicarbonate (HCO 3 – ), respectively [ eqs 1 – 3 57 ]. 1 2 3 NH 4 + formation increases the pH of a solution,\nand the breakdown of urea to bicarbonate increases the alkalinity.\nUnder geochemical conditions commonly found in natural aquatic environments\nwhere Ca 2+ is present, production of HCO 3 – induces conditions conducive to calcium carbonate\nprecipitation [ eq 4 58 ]. 4 Industrial applications for MICP via urea-hydrolyzing (ureolytic)\nmicrobial communities indigenous to soils and groundwater include:\n(i) the sealing of rock permeability and fracture networks in order\nto retard the migration of aqueous contaminants, 59 (ii) the repair and reinforcement of rocks and cement, 60 , 61 (iii) the mitigation of arid soil erosion. 62 Despite the identification of indigenous ureolytic species in various\n(contaminated) natural environments, 63 − 65 few studies have successfully\ndemonstrated ureolytic strontium removal through calcite precipitation\nunder environmentally relevant conditions. Past work including a study\nof Sr removal by MICP used a carbon-rich medium to enhance the activity\nof ureolytic pure cultures during incubation with aqueous Sr 2+ . 58 The study showed that 80% of the initial\naqueous Sr was associated with calcium carbonate only 4 h after the\nonset of ureolysis, with near-total Sr uptake achieved after 24 h.\nOther studies have focused on pure cultures of Bacillus pasteurii to remediate 90 Sr in simulated groundwaters analogous\nto those within the Snake River Plain aquifer, USA, 66 , 67 a radionuclide-impacted water body underlying a licensed nuclear\nsite. Both studies showed the immediate coremoval of Sr and Ca at\nthe start of ureolysis, which was later attributed to calcite precipitation\nby a combination of scanning electron microscopy (SEM) and X-ray absorption\nspectroscopy (XAS). A follow-on study by Fujita et al. 65 evaluated the potential for indigenous urease\nactivity within sediments obtained from the 90 Sr-contaminated\nsubsurface at Hanford to remediate aqueous 90 Sr contamination.\nQuantifying ureC copies (mL –1 ),\nureolysis rates (nmol L –1 h –1 ),\nand ureolytic MPN (cells mL –1 ) enabled a site-specific\ngeochemical model to be generated for strontium remediation via urease-driven\ncalcite precipitation. The distribution of ureolytic activity between\nthe sediments obtained at different depths was heterogeneous. In spite\nof the variability, model results indicated the microbial community\nat the Hanford subsurface is capable of MICP via ureolysis and concurrently\nremediating 90 Sr contamination via calcite coprecipitation.\nThe model also determined that the onset of ureolysis immediately\nprecipitated Sr-containing calcite and that it was possible to sequester\nvirtually all of the 0.25 ppm of Sr 2+ from groundwater\nafter 150 days of ureolysis. To date, this is the only study that\nwe are aware of that has modeled MICP in sediment systems from radionuclide-impacted\nland for urease-facilitated strontium remediation. However, no experiments\nevidencing enhanced Sr 2 removal were performed. Further\nstudies of ureolysis using real sediments and microbial communities\nunder a range of geochemical conditions (e.g., aerobic and anaerobic)\nare required in order to further evaluate the applicability of this\nremediation strategy to a range of contaminated sites. This\ncurrent study aims to demonstrate enhanced Sr 2+ removal\nthrough MICP generated by indigenous ureolytic bacteria\nin sediments representative of the subsurface at the Sellafield, UK\nnuclear licensed site. The potential for native microbial communities\nto hydrolyze urea with associated Sr removal was investigated using\na series of sediment microcosms. In addition, the incubations were\namended with various biostimulants, e.g., acetate, lactate, Fe(III),\nand yeast extract under either aerobic or anaerobic conditions. The\nreaction end-products were characterized by ESEM, XAS, and energy-dispersive\nX-ray spectroscopy (EDS) to determine nature and speciation and fate\nof Sr in the remediated solid phase. Overall, the project assessed\nthe viability of urea addition to groundwater/sediments contaminated\nwith Sr 2+ in the context of future in situ MICP remediation strategies. The results indicated that the native\nmicrobial communities were capable of degrading urea, concomitantly\nincreasing the pH of Sellafield-representative groundwater and facilitating\nincreased levels of Sr removal into biogenic calcium carbonate.",
"discussion": "Results and Discussion Biogeochemical Changes to Sediment Microcosms Within\n6 days of incubation, the urea-free controls (System A) for sediments\nPPQ and RB27 equilibrated at pH 8.3, ( Figure 1 a), whereas sediments CR and RB23 equilibrated\nat pH 6.4 and 7.7, respectively. In contrast, by day 6 in urea-only\nmicrocosms (System B), RB23, RB27, and CR microcosms showed clearly\nelevated pH of 9.4, 8.5, and 9.5, respectively ( Figure 1 b). In these three microcosms, the pH increased\nfrom 6.5 (corresponding to that of Sellafield groundwater) to ≥8.5\nand then either stabilized or continued to increase thereafter. System\nB for PPQ sediment was the only incubation that did not display a\ngreater pH increase compared the respective urea-free control. For\nurea-only incubations, CR and RB23 were the only sediments to generate\na pH increase >9 and display clear evidence of urea degradation.\nRB27\nstabilized at pH 8.5 in System B, likely caused by the presence of\nurea, compared with a decline from pH 8.3 to 7.1 in urea-free controls.\nCR sediments observed the largest increase in pH for urea-only amendments\n(pH 9.8), coinciding with complete urea removal by day 24. Urea degradation\nby ureolysis generates ammonium, responsible for elevating the pH\nof aqueous systems. The decomposition of urea and concomitant increase\nin (and stabilization of) pH thus strongly implies that ureolysis\nwas the degradation mechanism for urea. For urea plus acetate/lactate\namendments (System C), complete urea degradation coincided with an\nincrease in pH above 9.4 by day 6 in sediments RB23, RB27, and CR.\nA similar trend was observed with urea and additional yeast extract\n(System D), where an increase in pH to ≥9.6 was observed in\nall sediments by day 6. In Systems C and D, carbon amendments appeared\nto enhance the rates and extents of ureolysis in all sediments compared\nto urea-only amendments ( Figure 1 c,d). Notably, the addition of acetate and lactate\nfailed to stimulate urea decomposition in the PPQ sediment, while\nrapid ureolysis was observed with additional yeast extract. This observation\nimplies the successful stimulation of ureolytic bacteria native to\nSellafield sediments using carbon sources is sediment and nutrient\ndependent. Urea plus Fe(III) (System E) for sediments RB23, RB27,\nand CR produced a more gradual pH increase and trend in urea degradation.\nTotal urea decomposition in sediments CR and RB27 was accompanied\nby an increase in pH to 9.7. Incomplete ureolysis in sediment RB23\nand PPQ produced an elevated pH of 8.7 and 7.7, respectively ( Figure 1 e). Figure 1 Geochemical changes observed\nin biostimulated microcosms containing\nSellafield-relevant sediments CR (black diamonds), PPQ (blue circles),\nRB23 (red squares), and RB27 (green triangles). Strontium and Calcium Removal The PPQ sediment removed\napproximately 20% Sr after 1 day in the urea-free controls (System\nA). Sr removal in the urea-free controls for RB23, RB27, and CR sediments\nreached 40–60% by the same time point. Approximately 10% additional\nSr was then removed following a spike of 88 ppm of Sr and 80 ppm of\nCa added at day 7. Sr and Ca removal before and after this second\nSr/Ca spike are represented as Phase 1 and Phase 2 in these experiments,\nrespectively ( Figure 2 a). Generally, the urea-bearing incubations (Systems B to E) showed\nenhanced Sr removal by the end of Phase 1 for all sediments aside\nfrom PPQ, typically between 75% and 90% ( Figure 2 b–e). In urea-only incubations, Phase\n1 saw Sr removal in sediments CR and RB23 of 85% and 82%, respectively.\nThe presence of only urea failed to increase Sr removal during Phase\n1 for sediment RB27; however, additional acetate and lactate enhanced\nPhase 1 strontium removal to 84%, which was similar to RB23 (87%)\n( Figure 2 c). Amendments\nusing yeast extract in addition to urea resulted in similar levels\nof Sr removal in RB23, RB27, and CR (78–87%) during Phase 1.\nFor sediments CR and RB27, the presence of 10 mM Fe (III) in addition\nto urea generated the largest Phase 1 decrease in Sr of 88% and 94%,\nrespectively. Figure 2 Strontium concentrations and corresponding Ca removal\nin Sellafield-relevant\nsediment microcosms. CR (black diamonds), PPQ (blue circles), RB23\n(red squares), and RB27 (green triangles). Systems were spiked with\n88 ppm of Sr 2+ + 80 ppm of Ca 2+ at day 7 to\ndisplay clearer evidence of their enhanced concomitant removal with\nurea present. Phase 1 of Sr and Ca removal is represented by the blank\nbackground in individual plots, while Phase 2 is highlighted by the\ngray areas. Sr remediation was more variable\nbetween sediments during Phase\n2 compared to Phase 1 in incubations only containing urea ( Figure 2 b). For this system,\nenhanced Sr and Ca removal continued in Phase 2 in all sediments,\nwith the largest removal of Sr occurring in sediments RB23 and CR\n(82% and 94%, respectively) and lower levels of removal in sediments\nRB27 (54%) and PPQ (17%) ( Figure 2 b). With additional acetate and lactate, Phase 2 displayed\nnear-complete Sr removal in sediments CR (94%), RB27 (93%), and RB23\n(90%) and also enhanced Sr removal in the PPQ sediments (46%) ( Figure 2 c). The presence\nof yeast extract in addition to urea increased Sr removal in sediment\nRB23, RB27, and CR during Phase 2 to 82%, 88%, and 97% respectively.\nMost notably, additional yeast extract in the PPQ incubation produced\nthe most significant removal of Sr observed in this study, achieving\nnear-total Sr removal to subppm levels during Phase 2 ( Figure 2 d). With added 10 mM Fe(III),\nsediments RB23, RB27, and CR displayed a reduction in Sr of 79%, 89%,\nand 94%, respectively, during Phase 2 ( Figure 2 ). For the PPQ sediment, this system produced\nonly a 27% decrease in Sr during Phase 2. It is evident that PPQ sediment\nrequired additional carbon stimulation to enhance Sr removal in the\npresence of urea, which only appeared to occur during Phase 2, contrary\nto the other Sellafield sediments. Strontium and calcium removal\nin the urea-free controls were attributed\nto adsorption ( Figure 2 a). The pH point of zero charge (pH pzc ) of the silicates\nand oxides comprising the dominant mineralogy of the sediments ranges\nfrom approximately 2.5 to 7. At pH values > pH pzc , these\nmineral surfaces are negatively charged and favor the accumulation\nof adsorbed cations, e.g., Sr 2+ . This indicates that increasing\npH above 7 will likely increase Sr adsorption and removal from solution,\nconsistent with previous studies of Sr sorption to Sellafield-relevant\nsediments. 11 Ureolysis by soil bacteria\nis not retarded under anaerobic conditions. 83 Under anoxia, anaerobic bacteria may couple\nthe enzymatic oxidation of simple organic acids (e.g., acetate and\nlactate) to the reduction of electron acceptors (e.g., Fe(III)) in\norder to conserve energy for growth; this has been demonstrated in\nprevious studies for sediments RB23 and RB27. 71 Increased urea degradation and more rapid evolution to high pH with\namendments of acetate and lactate in sediments RB23, RB27, and CR\ncompared to urea-only incubations was likely due to the enhanced growth\nof a ureolytic bacteria under anaerobic conditions ( Figure 1 b,c). In turn, this produced\nfaster and more complete Sr and Ca removal during Phases 1 and 2 compared\nto urea-only systems ( Figure 2 b,c). Additional Fe(III) further enhanced the rate of ureolysis\nand concomitant pH increase in sediment RB27, which did not occur\nwith the other sediments. This resulted in significantly more Sr removal\nrelative to the urea-free control for RB27 and could be due to the\nstimulation of Fe(III)-reducing bacteria that could play a role in\nureolysis. Yeast extract may also serve as an electron donor 84 and is also commonly used as a micronutrient\nto help stimulate bacterial growth. 85 Enhanced\nrates of ureolysis have been displayed in soils amended with yeast\nextract. 62 , 86 A significant pH increase coincided with\ntotal urea degradation in all sediments amended further with yeast\nextract. For example, the RB23, RB27, and CR sediments amended with\nurea plus yeast extract reached pH 10 by day 3 and 6, respectively,\nafter almost all of the urea has been decomposed ( Figure 1 d). The PPQ sediment similarly\nreached pH 10 on day 10 after the urea concentration had decreased\nby 50%, being the only PPQ system to observe both total urea degradation\nand an increase to pH ≥ 8.2 thereafter ( Figure 1 d). As with acetate and lactate amendments,\nit is likely the increases in pH are related to increased rates of\nureolysis after stimulation with yeast extract, resulting in more\nrapid pH increases compared with the urea-only controls. Overall,\namending Sellafield sediment microcosms with urea (and additional\nbiostimulating compounds) correlated with increased and more rapid\nSr and Ca removal, indicating that calcium carbonate precipitation\nmay be occurring and responsible for enhancing Sr uptake. Microbially\nprecipitated carbonates are well-known to simultaneously remove Ca 2+ and Sr 2+ . 87 − 89 Stimulating ureolytic bacteria\nwith urea in pure culture studies has produced an increase in groundwater\npH from circumneutral to ≥9 on the order of hours, under both\noxic and anoxic conditions. 90 , 91 Environmental Scanning\nEmission Spectroscopy (ESEM) Biostimulated sediments from\nRB27 amended with 10 mM Fe(III) and\nPPQ amended with 1 g/L yeast extract were analyzed using ESEM ( Figure 3 ). Samples were predominantly\ncomposed of silicate (e.g., quartz and feldspar) grains coated with\nclay particles (e.g., chlorite and muscovite) approximately 100–200\nμm in size. Backscatter imaging combined with energy dispersive\nspectroscopy (EDS) mapping revealed the presence of approximately\n50 μm crystallites ( Figure 3 a,c) and spheroidal particles ( Figure 3 b,d) on the surface of the silicate grains.\nEDS analysis of these areas also indicates these particles are calcium-rich\nwith strontium present ( Figure 3 e,f). The silicon, iron, and aluminum detected are due to\nthe underlying silicate particles. Figure 3 ESEM analysis of a rhombohedral precipitate\nformed in RB27 microcosm\namended with 10 mM Fe(III) (a, c, and e) and a spheriodal precipitate\nformed in PPQ microcosm amended with 1 g/L yeast extract (b, d, and\nf), 17 days after incubation. (a and b) Images in backscatter mode.\n(c and d) Elemental maps displaying the correlation between Ca and\nSr. (e and f) EDX spectra corresponding to the aggregate circled in\nred in the ESEM image. The morphology and composition\nof these particles indicates they\nare polymorphs of Sr-containing calcium carbonate, likely formed in\nthe microcosms during the decomposition of urea. 92 , 93 The precipitates formed in the RB27 system were more angular in\nappearance compared with the spheriodal particles observed in the\nPPQ sediment. Calcite tends to form rhombohedral crystals with distinct\ncrystal faces, 94 similar in appearance\nto the Ca/Sr-bearing particles present in the RB27 system ( Figures 3 a and S2 ), and other studies investigating Sr removal\nvia urease-facilitated MICP. 95 , 96 The spheriodal particles\nin the PPQ system suggests that vaterite is the calcium carbonate\npolymorph present, responsible for incorporating Sr during precipitation.\nA study by Sheng Han et al. 97 identified\nmicron-scale vaterite spherulites using SEM in a system where ammonia\nwas used to increase pH from 8 to 11. During this process, high calcium\ncarbonate supersaturation was achieved by rapidly dissolving CO 2 , which led to the formation of vaterite due to the sustained\nhigh supersaturation during heterogeneous nucleation. 98 The RB27 system displayed a steady pH increase from\n7.5 to 8.9\non days 1 to 6, coinciding with a near linear urea decomposition rate\nleading to calcite formation ( Figure 1 e). This corresponded to a decrease in strontium concentration\nfrom 30 to 6 ppm and concomitant calcium removal over the 5 day period\n( Figure 2 e). In contrast,\nthe PPQ system produced a much faster pH increase, rising from 7.5\nto 9.6 on days 3 and 6, respectively ( Figure 1 d), leading to vaterite formation. Interestingly,\nin the PPQ system, the strontium concentration remained at ∼80\nppm between days 3 and 6 despite the marked pH increase, but decreased\nfrom 170 ppm on day 7 to below 1 ppm at day 8 ( Figures 1 d and 3 d). The precipitation\nof vaterite has been shown to occur within minutes of supersaturation\nin both abiotic 43 , 97 and pure culture 99 , 100 systems and has the potential to uptake more Sr than calcite. 43 The differing degrees of solution supersaturation\n(and the rate of supersaturation development) with respect to calcium\ncarbonate in the PPQ and RB27 systems likely controlled the differences\nin Sr-containing calcium carbonate polymorphs precipitated. 101 PHREEQC In this study, a geochemical\nmodel simulating\nthe impact of bacterial hydrolysis of urea in Sellafield-representative\ngroundwater containing 100 ppm (∼1.14 mM) strontium was constructed\nusing PHREEQC 79 ( Figures 4 and S3 ). Ureolysis\nand consequent geochemical changes were simulated over 1 day. The\nrate of urea hydrolysis ( k urea ) was assumed\nto follow the first order reaction given the approximately linear trend in\nurea degradation for PPQ with 1 g/L yeast extract (between days 6\nand 15) and RB27 with 10 mM Fe(III) (between days 1 and 25). A previous\nstudy investigating ureolytic MICP by indigenous microbial communities\nused a similar approximation to calculate the ureolysis rate constants. 59 For the RB27 system, k urea = 0.04 day –1 , which was calculated for\nurea-stimulated natural groundwater supplemented with 1 g/L molasses. 59 For the PPQ system, k urea = 0.12 day –1 , which was calculated for urea-stimulated\nartificial groundwater amended with a pure culture of Sporosarcina\npasteurii (∼7.2 × 10 5 cell/mL). 59 There was no significant difference in modeled\nrate of increase in carbonate mineral supersaturation using k urea = 0.04 day –1 ( Figure 4 ) compared with 0.12\nday –1 . Model results showed an immediate pH increase,\ncaused by the stoichiometric hydrolytic breakdown of 1 mol of urea\nto 2 mol of NH 4 + , generating alkalinity ( Figure S4 ). As ureolysis progresses, several\ncalcium carbonate minerals (e.g., calcite and vaterite) became oversaturated\nwithin the pH range observed in this study (from pH 6.5 to 10). Clearly,\nthe simulation suggests that the ureolysis is capable of altering\nSellafield groundwater geochemistry to produce conditions conducive\nto carbonate mineral formation. For the RB27 system, an increase in\npH from 7.5 to 8.1 to 8.9 was observed from days 1, 3, and 6, respectively\n( Figure 1 e). The model\nsuggests that Sellafield AGW is undersaturated with respect to calcium\ncarbonate at pH 7.5 ( Figure 4 ) but shows an increase in the saturation index (SI) for calcite\nfrom −1.04 to 0.44 as pH increases from 7.5 to 8.9. Conversely,\nthe PPQ system displayed a significant increase from pH 7.5 on day\n3 to pH 9.6 on day 6 ( Figure 1 d). The model suggests this rapid evolution of high pH and\nalkalinity leaves the solution oversaturated with respect to vaterite\n(SI = 0.36). Figure 4 Selected PHREEQC output for the microbial hydrolysis of\nurea in\nstrontium-bearing Sellafield groundwater, depicting the supersaturation\nof various Sr/Ca-containing carbonate phases. In the urea-free controls, RB27 removed 50% more Sr than PPQ via\nadsorption ( Figure 2 a). Additionally, the rate of ureolysis and the development of sufficiently\nhigh pH and alkalinity necessary for calcium carbonate precipitation\nwas much slower in the RB27 system with 10 mM Fe(III) than the PPQ\nsystem with 1 g/L yeast extract ( Figures 1 e, 4 , and S4 ). Thus, these factors likely contributed to\nlowering the degree of calcium carbonate supersaturation in the RB27\nsystem, favoring slower and surface-mediated growth of strontium-containing\ncalcite. In contrast, in the PPQ system, rapid pH and alkalinity increases\nlikely pushed the Sellafield AGW much further out of equilibrium with\nrespect to Ca, producing a far higher degree of supersaturation and\nlikely facilitating nucleation-dominated vaterite precipitation ( Figures 1 d, 4 , and S4 ). X-ray Absorption Spectroscopy The speciation of Sr\nwithin RB27 sediments stimulated with urea and 10 mM Fe(III) (E) and\nPPQ sediments stimulated with urea and 1 g/L yeast extract (D) was\nanalyzed via Sr K -edge XAS spectroscopy of the microcosm\nend products ( Figure 5 ). Given that the previous modeling, aqueous geochemical, and microscopic\nanalyses suggested the potential association of Sr 2+ with\nCaCO 3 mineralization, the local environment of various\nSr 2+ -incorporated calcium carbonate phases was used to\ninform the fits to the spectra. 43 , 47 , 52 , 102 , 103 Figure 5 Background-subtracted\nSr K -edge EXAFS data (left-hand\nside) and corresponding Fourier transformations (right-hand side)\nobtained on biostimulated sediments from PPQ amended with 1 g/L yeast\nextract and RB27 amended with 10 mM Fe(III). A description of the\nstrontium-associated calcium carbonate phase used to fit the experimental\ndata is also provided. Experimental data = black line, fit = red line. The untreated PPQ system returned a spectrum that\nwas best fit\nwith 9 O atoms at 2.6 Å indicative of outer-sphere adsorption. 82 , 104 , 105 The best fit to the spectrum\nfrom the PPQ sample indicated a central Sr atom in 8-fold coordination\n(8 O atoms at 2.53 Å) with an addition of 4 C atoms at 2.95 Å.\nThe model also includes 2 distinct Sr–Ca shells at 3.95 and\n4.21 Å. However, substituting the 2 Ca atoms at 4.21 Å with\nSr atoms produced an almost identical fit. Splitting the Sr–Ca\ncoordination path into 2 shells improved the fit with statistical\nsignificance compared with adding 4 Ca atoms at 4.05 Å ( Table S1 ). Overall, this coordination environment\nis indicative of Sr substitution for Ca within vaterite. 43 , 47 , 103 , 106 The best fit achieved here was in closer agreement to that of Littlewood\net al., 43 who studied Sr-incorporated vaterite\nfrom a 0.1% Sr 2+ solution ( Table 2 ). The only notable difference in shell fitting\npresented here compared with that presented in Littlewood et al. 43 is the splitting of the distal Ca shell with\nslightly different Sr–Ca distances. However, we note that the\nDebye–Waller factors for the single Sr–Ca shell in the\nLittlewood et al. 43 study is significantly\nhigher than those of the two individual Sr–Ca shells fitted\nhere (0.014 vs 0.004 and 0.007). This indicates that there may be\nsignificant structural disorder associated with the single Sr–Ca\nshell in the current study, with the Ca atoms spread over the range\nof distances. This is broadly consistent with the split Sr–Ca\nshells in this study, with improvements in data quality and k -range a likely reason for the resolution of 2 Ca shells\nin this study. In the literature, Ca K -edge spectra\nfor vaterite often varies considerably, 107 , 108 referred to without experimental data, 53 , 54 not modeled at R > 3 Å 103 or at all, 109 not accompanied\nby model fitting parameters, 110 or are\nfit poorly/only in conjunction with the “conventional”\n6 Ca atoms at approximately 4.2 Å (in accordance with ref ( 111 )). 55 , 112 Splitting the Ca shell at approximately 3.9–4.3 Å has\nnot been conducted in studies analyzing the structure of vaterite\nusing Ca K -edge. Instead, a Ca shell at 3.95 Å\nhas been reported for calcite 54 and aragonite. 53 In addition, the presence of structural disorder\nof vaterite crystals formed under environmental conditions is widely\naccepted 102 and, in the present study,\nit is likely enhanced by the presence of Sr 2+ ions substituting\nfor smaller Ca 2+ ions (1.18 and 1.00 Å, respectively). 20 This distortion likely explains the presence\nof the split distal Sr–Ca shell in two with similar Sr–Ca\ndistances. Table 2 Details of EXAFS Fitting Parameters\nfor the Sr Biominerals Formed in the Respective PPQ and RB27 Systems a sediment/system;\nfit description coordination\npath coordination\nnumber coordination\nnumber 43 R (Å) R (Å) 43 σ 2 (Å 2 ) σ 2 (Å 2 ) 43 PPQ Sr–O 9 2.60 0.012 Sr sorption standard PPQ + 1 g/L yeast extract Sr–O 8 8 2.53 2.52 0.010 0.011 Sr-incorporated\nvaterite Sr–C 4 4 2.95 3.10 0.006 0.025 Sr–Ca 2 3.95 0.007 Sr–Ca 2 4 4.21 4.19 0.004 0.014 RB27 + 10 mM Fe (III) Sr–O 9 9 2.60 2.57 0.011 0.017 Sr-incorporated\ncalcite Sr–C 5 5 3.03 3.10 0.030 0.020 Sr–Ca 4 4 4.10 4.12 0.019 0.018 a R = atomic distance;\nσ 2 = Debye–Waller factor. The amplitude factor\n(S02) was set to 1 for each sample. As postulated above, spherulitic Ca-bearing precipitates\nevident\nwithin the PPQ system likely signify vaterite formation ( Figures 3 b,d). Results from\nthe PHREEQC calculation of the ureolytic AGW system indicated vaterite\nsupersaturation is achieved at pH ∼ 9.1, which was observed\nin this system during the rapid pH increase. This is consistent with\nbatch experiments investigating Sr bioremediation via urea-induced\nMICP using pure cultures of Sr-resistant Halomonas sp. 113 and Bacillus pasteurii , 58 , 114 which successfully remediated Sr through\ncarbonate-phase formation, including vaterite. Vaterite is generally\nconsidered an unstable intermediate that\nnucleates rapidly from amorphous calcium carbonate, eventually transforming\nto calcite within hours. 43 , 102 , 115 , 116 This study suggests that Sr-incorporated\nvaterite may describe a more stable Sr-incorporated phase than first\nconceived under environmentally relevant conditions. Sustained ammonium\ngeneration, inducing a consistent elevated pH, and the consequent\nretention of a high degree of supersaturation might explain the persistence\nof vaterite. 97 The transformation of vaterite\nto more stable calcium carbonate polymorphs such as calcite could\nreduce the partitioning of Sr. 117 However,\nthe retention of Sr during calcium carbonate recrystallization is\ncomplicated by other factors (especially under environmental conditions)\nincluding the amount of initial vaterite precipitation (mediated by\nthe initial ionic strength of a solution) and the rate of vaterite\ndissolution which, in turn, partially regulates the rate of calcite\nprecipitation. 43 Fitting the EXAFS\nspectrum for RB27 using a crystal structure for\nSr 2+ -incorporated calcite provided an excellent fit, detailing\na central Sr 2+ atom coordinated to 9 O atoms at 2.60 Å,\n5 C atoms at 3.03 Å, and 4 Ca atoms at 4.10 Å ( Figure 5 and Table 2 ). The model contained near\nidentical interatomic distances and shell occupancies to those for\nthe chemical precipitation of calcium carbonate in strontium-rich\nsystems. 43 The lack of a distinct doublet\nat ∼16 115 eV in either XANES spectrum suggested the\nabsence of 6-fold coordination indicative of Sr-incorporated calcite\n( Figure S5 ). 37 , 50 , 118 , 119 The occurrence of\na single peak indicates Sr 2+ in 8/9 fold coordination,\nwhich is consistent with the EXAFS fit. Elevated strontium concentrations\npresent in these systems likely results in Sr 2+ occupying\nlattice sites alternative to the position of 6-fold-coordinated Ca 2+ ions when coprecipitated with calcium carbonate, as reported\nby Littlewood et al. 43 This likely occurs\ndue to the higher concentrations of Sr 2+ within the calcite\ncrystal. Overall, elucidation of the two calcium carbonate polymorphs\nhighlights the differing degrees of supersaturation achieved in these\nsystems. More Sr was removed in the PPQ system, attributed to vaterite\ncoprecipitation at a higher Sr 2+ concentration. Microbial\nCommunity Analysis DNA extractions were performed\non various sediment samples obtained on days 0 and 17, and the 16S\nrRNA genes amplified and sequenced to help identify changes in the\nmicrobial communities within the urea-free controls and urea-stimulated\nsediments. A diverse indigenous community was identified in each sediment\ntype prior to experimental incubations ( Figure 6 ). For each system and sediment, a diverse\nrange of bacteria was maintained after 17 days. Adding solely urea\nto sediments produced a slight shift toward a microbial community\nassociated with ureolysis. Several Bacillus species,\nbelonging to the class Bacilli, are well-documented as being capable\nof hydrolyzing urea. 120 − 123 When compared to urea-free controls after 17 days, urea-only amendments\nenriched the proportion of Bacilli from 14% to 24% in RB23, 1% to\n20% in CR, and 13% to 19% in RB27, while also enhancing Sr and Ca\nremoval. However, microcosms modified with additional biostimulating\ncompounds and urea produced the greatest shifts in microbial community\nstructure. Figure 6 Bacterial phylogenetic diversity within the Sellafield-related\nsediments before and 17 days after biostimualtion with urea and other\ncompounds aimed at promoting ureolytic conditions. The displayed classes\naccount for greater than 1% of the microbial community. Species of the class Bacilli constituted 14% of the PPQ sediment\nprior to incubations. Incubating PPQ with yeast extract in addition\nto urea was the only PPQ system to enrich species belonging to the\nmicrobial class Bacilli (65%) ( Figure 6 ). The other PPQ incubations failed to maintain a relative\nproportion of Bacilli >1%. In the PPQ system amended with yeast\nextract,\nthe closest known relatives assigned to two OTUs (operational taxonomic\nunits), comprising 25% of the microbial population, were strains most\nclosely related to Sporosarcina pasteurii , 124 , 125 previously known as Bacillus pasteurii . The ureolytic\nprecipitation of calcite has been widely demonstrated using pure cultures\nof Sporosarcina pastuerii , 126 , 127 including studies that successfully coprecipitated Sr with calcite. 67 , 96 While biostimulations using yeast extract have been shown to increase\nthe population of Bacillus species in soil from 5%\nto 85–99%, 128 Bacillus urease has also precipitated spheroidal vaterite after increasing\nthe pH of the system from 6.8 to 9. 99 Incubating\nurea-stimulated PPQ sediments with yeast extract yielded the only\nPPQ system with enhanced Sr/Ca removal compared to urea-free controls.\nIt also remained the only PPQ incubation to raise the groundwater\npH above 8.2, consistent with complete urea degradation. Thus, results\nof the DNA extraction support geochemical measurements displaying\nthat yeast extract amendments played a crucial role in enhancing ureolysis\nin PPQ sediment and subsequent Sr remediation. An OTU assigned to\nthe urease-positive species Methylophilus methylotrophus constituted 8.5% of the PPQ urea-free control population, 129 enriched to 29.4% in urea-only incubations\n( Table S2 ). However, amending PPQ solely\nwith urea failed to stimulate urea decomposition or enhance Sr/Ca\nremoval. The dominance of Bacilli species in the yeast-amended PPQ\nsystem, as well as other biostimulated sediments, appears consistent\nwith the development of ureolytic conditions necessary for increasing\nSr/Ca removal compared with urea-free controls. Species of Bacilli\nconstituted 14% of the microbial population\nindigenous to RB27 both at the start and 17 days after urea-free controls\nwere constructed ( Figure 6 ). Incubating RB27 with urea and an additional 10 mM Fe(III)\nfor 17 days led to the prevalence of Bacilli (60%), comparable to\nfurther additions of yeast extracted (65%) and far greater than enrichments\nproduced by urea only (19%) and 5 mM acetate/lactate (30%) amendments\n( Table S3 ). The five most abundant OTUs\nby day 17 all belonged to the class Bacilli. Three of the five OTUs\nwere assigned to ureolytic species of Sporosarcina and constituted around 20% of the total microbial population. The\nclosest known relatives of the three OTUs included a facultatively\nanaerobic strain of Sporosarcina aquimarina (13.2%) 130 , 131 as well as Sporosarcina pasteurii (4.1%) 59 , 124 and Sporosarcina ginsengoli (5%). 132 Fe(II) Generation Facilitated by Bacterial\nUreolysis The reduction of bioavailable Fe(III) to Fe(II)\nwithin Sellafield\nsediments RB23 and RB27 was enhanced with the addition of urea ( Figure 7 ). In both sediments,\nthe percentage of bioavailable Fe as Fe(II) increased from approximately\n20–60% after 25 days of anaerobic incubation with solely urea.\nParallel urea-free controls constructed using only sediments (with\nno amendments) displayed an increase in Fe(II) from approximately\n20% to 30%. Figure 7 Microbial Fe(III) reduction in Sellafield sediment microcosms.\nRB23 urea only (black squares), RB27 urea only (gray triangles), RB23\ncontrol (red squares), and RB27 control (green triangles). Ureolysis-generated ammonium may have served as an electron\ndonor\nin generating microbial reduction of Fe(III). Anammox (anaerobic ammonium\noxidation) processes play a major role in global nitrogen cycling\nwithin aquifers, particularly nitrogen loss. 133 In a separate process, certain organisms are able to couple the\noxidation of ammonium to Fe(III) reduction during anaerobic respiration, 134 a process termed “feammox” that\nhas been observed in Fe(III)-rich anoxic soils [ eq 5 135 ]. 5 Bacteria capable of enzymatically reducing\nFe(III) to Fe(II) are\noften similarly capable of directly reducing U(VI) to U(IV) 136 while microbial Fe(II) generation is capable\nof indirectly reducing U(VI) to U(IV), 137 Tc(VII) to Tc(IV), 82 , 138 and Np(V) to Np(IV) 139 with the reduced forms typically poorly soluble\ncompared to the oxidized species. Thus, results from this study indicate\nthat the ureolytic stimulation of bacteria indigenous to Sellafield\nsediments could be used to immobilize both strontium and uranium/neptunium/technetium\nthrough MICP and feammox processes respectively, in a coupled system. Conclusion and Environmental Significance While MICP\nby urease encoding bacteria have been investigated in sediments from\nother contaminated nuclear sites, this represents the first study\nof urea-facilitated MICP in Sellafield sediments and also quantifies\nthe impact of added electron donors/nutrients (i.e., acetate, lactate,\nand yeast extract) and electron acceptors (i.e., Fe(III)). To the\nbest of our knowledge, this is the first time that enhanced strontium\nremoval by urea-facilitated MICP has been demonstrated by indigenous\nsediment microbial communities. The rate and extent of concomitant\nSr 2+ and Ca 2+ removal at elevated pH varied\nbetween the sediments, and the biostimulation regime applied. Maximal\nremoval of Sr (170 ppm to subppm concentrations over the course of\n1 day) was noted in the PPQ sediment microcosm amended with urea and\nyeast extract. This system reached approximately pH 10 before extensive\nSr removal. Spherical Ca/Sr-bearing precipitates identified in the\nremediated PPQ sediments using ESEM and corresponding EDX analyses\nsuggested that Sr coprecipitated with the calcium carbonate polymorph\nvaterite. EXAFS analysis of the sediments further supported the hypothesis\nthat the end point of Sr remediation was indeed Sr-incorporated vaterite.\nThe RB27 system amended with urea and additional Fe(III) reached a\nfinal pH > 9.5, coinciding with a reduction in strontium concentration\nfrom 93 to 10 ppm over 1 day. Analyzing the remediated sediments in\nthe RB27 system using EXAFS and ESEM suggested that Sr was immobilized\nin calcite. While calcite is a more stable polymorph of calcium carbonate,\nvaterite precipitates more rapidly and is capable of incorporating\nmore Sr. 43 Thus, in addition to the sediment\ncharacteristics, the time scale and aqueous strontium concentrations\nare important considerations for perspective in situ remediation strategies when targeting strontium removal as various\nSr-associated carbonate phases. Anaerobic microcosms using Sellafield\nsediments RB23 and RB27 also showed an unexpected increase in microbial\nFe(III) with urea added. Enhanced Fe(III) bioreduction coupled to\nthe oxidation of NH 4 + formed during ureolysis\nmay have been responsible for enhancing Fe(II) production in these\nsystems. Accumulating sufficient NH 4 + from ureolysis\nmay, therefore, enhance further beneficial microbial activity for\nthe remediation of redox-active radionuclide species such as U, Np,\nand Tc. This work has significant implications for the remediation\nof 90 Sr at Sellafield through MICP, in addition to the\npotential cleanup of redox active radionuclides. The need for further\nwork to define the utility of these processes at a larger scale using\nflow-through systems more representative of the dynamic natural subsurface\nin order to assess the large-scale application and long-term performance\nof this technology is clear."
} | 11,794 |
37755994 | PMC10534446 | pmc | 7,274 | {
"abstract": "Bacterial organisms have undergone homologous recombination (HR) and horizontal gene transfer (HGT) multiple times during their history. These processes could increase fitness to new environments, cause specialization, the emergence of new species, and changes in virulence. Therefore, comprehensive knowledge of the impact and intensity of genetic exchanges and the location of recombination hotspots on the genome is necessary for understanding the dynamics of adaptation to various conditions. To this end, we aimed to characterize the functional impact and genomic context of computationally detected recombination events by analyzing genomic studies of any bacterial species, for which events have been detected in the last 30 years. Genomic loci where the transfer of DNA was detected pertained to mobile genetic elements (MGEs) housing genes that code for proteins engaged in distinct cellular processes, such as secretion systems, toxins, infection effectors, biosynthesis enzymes, etc. We found that all inferences fall into three main lifestyle categories, namely, ecological diversification, pathogenesis, and symbiosis. The latter primarily exhibits ancestral events, thus, possibly indicating that adaptation appears to be governed by similar recombination-dependent mechanisms.",
"conclusion": "5. Conclusions Recombination-driven gene transfer directs prokaryotic evolution as a fundamental mechanism of genetic exchange. Having reviewed current genomic studies, we found that recombination hotspots tend to be located near horizontally acquired genes, mobile elements, and genome repeats ( Figure 1 a). These respective events could be classified according to three major trends, namely (i) ecological diversification, (ii) pathogenesis, and (iii) symbiosis. These exchanges mostly affect genes encoding membrane proteins, toxins, antibiotic resistance factors, polysaccharide biosynthesis enzymes, effectors, secretion systems, and metabolic pathways, with only the latter being involved in all three evolutionary trends mentioned ( Figure 1 b). Presumably, when microorganisms occupy a new environment, genetic exchanges between them and concomitant rearrangements occur more frequently. For instance, recombination occurs at the initial stage of adaptation to new hosts or denotes the continuous process of diversification, as in the case of Cry toxins in B. thuringiensis . Nevertheless, over time, relationships with the host become increasingly specialized, accompanied by an incremental decrease in recombination rate. Therefore, the signals of recombination are primarily detected in loci associated with environmental interaction and can be used as genomic markers displaying the ecological history of the studied strains. In the reviewed studies, predominant methods for detecting recombination events were based on computational predictions in the genomic data. However, despite a large arsenal of tools available, current methods are not free from limitations, e.g., dependence on the models and dubious assumptions that the core genome reflects clonal relationships [ 5 ]. Nevertheless, even with possible artefactual inferences, the data from current genomic studies provided similar functional groups of genes to those reported in laboratory studies. This indicates that computational pipelines seem to correctly display the evolutionary dynamics of bacterial genomes in the context of recombination. It is noteworthy that, when analyzing the functional role of genetic exchanges, we did not select reports with predetermined criteria of grouping, however, we did reveal general trends. Undoubtedly, there is a skew toward pathogenic bacteria due to threats to global health; thus, there is a need for further studies on species that occupy other ecological niches. Observed genomic regions with increased recombination rates might serve as a roadmap for further studies. Possible implications could include targeted analysis of genomic loci housed in mobile genetic elements, such as genomic islands or those flanked by insertions. Another strategy lies in picking particular genes, e.g., toxins of secretion systems with the reconstruction of recombination events, both ancestral and recent. Subsequent matching of these inferences with species’ phylogeny would help to reveal key adaptation steps to novel environments and identify common and/or distinct pathways within different ecological groups. All things considered, there is a strong demand for (i) performing comparative studies of recombination intensity across different bacterial species, (ii) developing new mathematical models and novel bioinformatic tools for recombination detection, and (iii) carrying out experimental validation of computationally derived observations to yield new insights and deepen our understanding of an intricate network of recombination events with their functional ramifications.",
"introduction": "1. Introduction Through shaping the genomic landscape, horizontal gene transfer (HGT) or lateral gene transfer (LGT) and homologous recombination (HR) both serve as principal evolutionary forces in bacteria. These mechanisms provide genetic plasticity and thereby ensure adaptation to ecological niches [ 1 ], regulate virulence [ 2 ] and increase fitness [ 3 ]. The effect of these events is not only constrained to bacteria, but also plays a vital role in orchestrating the evolution and adaptability of archaea, viruses, and even eukaryotes [ 4 ]. The first phenomenon mentioned implies the replacement of DNA sequences when affecting genomic loci with contiguous, highly homologous regions [ 5 ]. HGT, in its turn, could be roughly defined as the acquisition of genetic material from a donor to recipient bacterial cells, predominantly requiring micro-homology sufficient for the incorporation of exogenous DNA into bacterial chromosomes or plasmids [ 6 , 7 ]. Not only do HR and HGT entail speciation [ 8 ] and spark the origin of new strains [ 4 ], but also cause antibiotic resistance [ 9 ], enhanced virulence [ 10 ], and reduced efficiency of vaccines [ 11 ]. However, the outcomes of these processes, in some cases, are beneficial for industries insofar as they result in the modulation of symbiotic relationships with agriculturally important plants [ 12 ] or lead to the emergence of strains capable of metabolizing pollutants, thus exhibiting promising biotechnological potential [ 13 ]. In terms of the parts of the bacterial genome affected, HR was reported to exert an effect on core genes [ 14 ], i.e., those shared by the majority of isolates within a certain population, whereas the accessory component is commonly embedded into genomic regions via HGT [ 15 ]. Therefore, the former alters allelic diversity, and the latter modifies gene composition. The preliminary step preceding the import or insertion of loci is DNA acquisition. Foreign DNA could enter bacterial cells via three fundamental mechanisms, namely, transformation, transduction, and conjugation [ 16 ]. Transformation involves a direct influx of DNA from the environment and is observed in diverse pathogens, including the genera Neisseria , Helicobacter , Streptococcus , etc. [ 7 ]. Being a complex multi-stage procedure, it is comprised of multiple steps, such as the activation of competence in the early stages of bacterial growth, triggering a two-component system ComD/ComE, and the action of the type IV pilus apparatus, through which foreign DNA transfer is performed [ 7 ]. The import is accompanied by DNA processing into 6 kb fragments by surface endonuclease [ 7 ]. Transduction is a phage-mediated DNA transfer between cells, whereby the phage carrying genetic material can integrate into a host genome [ 7 ]. The embedded loci could represent virulence determinants, e.g., toxin-encoding genes [ 17 ]. Conjugation, first discovered in E. coli when characterizing self-replicating F-plasmids, is carried out through direct cell-to-cell contact [ 16 ]. This necessity for maintaining cell contacts does not allow the transmission of genetic material other than small genetic elements, e.g., plasmids. It may also regulate the transfer of integrative and conjugative elements (ICEs), which lack a self-replication system and are capable of copying only during conjugation [ 7 , 16 ]. Upon obtaining DNA, either homologous or site-specific recombination occurs, with the latter governing HGT. HR is accompanied by the formation and consequent resolution of the Holliday junction [ 18 ]. It requires the RecA protein with polymerase activity attached to single-stranded DNA stabilized with SSB (single-stranded DNA-binding protein) [ 18 ]. Two alternative protein complexes, RecBCD and RecFOR, participate in the downstream stages [ 7 , 19 ]. Noteworthy, bacterial genomes demonstrate a non-uniform distribution of recombination systems with the prevalence of RecFOR over RecBCD in some organisms, presumably indicating the redundancy of the molecular mechanisms governing recombination [ 19 ]. Alternatively, obligate endosymbionts lack the RecA protein, while in their genomes, RecA-independent recombination has been reported [ 19 , 20 , 21 ], which can probably be explained by tandem repeats inducing genetic exchange [ 22 ]. HGT, understood as the incorporation of mobile genetic elements (MGEs) including plasmids, prophages, ICEs, and pathogenicity islands, is mainly mediated by site-specific recombination [ 7 ]. The respective integration is carried out by tyrosine recombinases, also known as integrases, and serine recombinases called resolvases [ 18 ]. The first family comprises XerC and XerD proteins, which separate chromosomal dimers during bacterial replication occurring at the dif site [ 18 ], whereas serine recombinases operate by making four breaks in double-stranded DNA molecules with their subsequent ligation [ 7 ]. HR and HGT are tightly interconnected, given that horizontally acquired genes are often flanked with genomic regions with an excessive HR frequency. This possibly serves to regulate genome size through excising obtained genes [ 23 , 24 ]. Additionally, transmitted MGEs could further engage in HR, as shown for prophages in cases of co-infection [ 25 ]. Moreover, pathogenic islands, insertion sequences, and other imported loci are characterized by HR signals predicted through bioinformatics tools [ 9 , 26 , 27 , 28 ]. On this account, in the current review, we would not distinguish between site-specific recombination-driven HGT and HR, using an umbrella term recombination instead. Comparative studies have examined the intensity of HGT and HR to be unequal within bacterial populations that belong to different groups according to the ecological niches they occupy. As an example, obligate pathogens and symbionts are characterized by their lower HR rates than free-living organisms, commensals, and opportunistic pathogens [ 29 , 30 ]. Similar results held when considering horizontally transferred genes being presented by infection effectors and antibiotic resistance factors, thus, serving as modulators of bacterial invasion and promoting adaptation to certain hosts [ 3 ]. Taking into consideration the above-mentioned information, an in-depth understanding of the functional effect that recombination exerts on bacterial populations is needed concerning fundamental science and practical implications. In the review presented, we analyze aspects of recombination by discussing studies made in the last 30 years. The included studies comprised 91 bacterial species from different taxonomic groups. When choosing the articles, we focused on those in which recombination signals were found through computational predictions in whole genomes and/or individual loci, contiguous regions, and extrachromosomal elements. We summarize which parts of the genome are subjected to recombination and provide a scheme illustrating the biological roles of proteins encoded by the respective chimeric or acquired loci ( Figure 1 ). We reveal that the vast majority of observations ( Table S1 ) fall into three categories related to three extremely dynamic processes: the establishment and development of (i) symbiotic or (ii) pathogenic relationships and (iii) ecological diversification caused by alterations in the environment."
} | 3,064 |
19720520 | null | s2 | 7,275 | {
"abstract": "Constructing novel biological systems that function in a robust and predictable manner requires better methods for discovering new functional molecules and for optimizing their assembly in novel biological contexts. By enabling functional diversification and optimization in the absence of detailed mechanistic understanding, directed evolution is a powerful complement to 'rational' engineering approaches. Aided by clever selection schemes, directed evolution has generated new parts for genetic circuits, cell-cell communication systems, and non-natural metabolic pathways in bacteria."
} | 147 |
39894851 | PMC11788432 | pmc | 7,276 | {
"abstract": "Current artificial systems suffer from catastrophic forgetting during continual learning, a limitation absent in biological systems. Biological mechanisms leverage the dual representation of specific and generalized memories within corticohippocampal circuits to facilitate lifelong learning. Inspired by this, we develop a corticohippocampal circuits-based hybrid neural network (CH-HNN) that emulates these dual representations, significantly mitigating catastrophic forgetting in both task-incremental and class-incremental learning scenarios. Our CH-HNNs incorporate artificial neural networks and spiking neural networks, leveraging prior knowledge to facilitate new concept learning through episode inference, and offering insights into the neural functions of both feedforward and feedback loops within corticohippocampal circuits. Crucially, CH-HNN operates as a task-agnostic system without increasing memory demands, demonstrating adaptability and robustness in real-world applications. Coupled with the low power consumption inherent to SNNs, our model represents the potential for energy-efficient, continual learning in dynamic environments.",
"introduction": "Introduction In recent years, artificial intelligence (AI) has achieved remarkable advances, becoming integral to our daily lives, especially with the development of the generative pre-trained transformer 1 . However, current AI systems still rely on training with entire datasets at once, lacking the ability to incrementally add new data without disrupting the existing model. This limitation presents challenges of catastrophic forgetting in environments that require incremental learning from temporally ordered data. To mitigate this issue, adaptive learning strategies known as continual learning or lifelong learning have garnered increasing attention in research. Despite significant advancements, current methods face persistent challenges, including the need for explicit task identification during inference and the increasing memory demands associated with storing samples or features from previous tasks or classes. These limitations significantly hinder the practical application of continual learning in dynamic, real-world environments. Consequently, the development of task-agnostic approaches to enhance the practical implementation of continual learning in real-world scenarios remains a critical area of research. In contrast, biological systems demonstrate exceptional efficiency in incremental learning with low energy consumption, underscoring the potential for brain-inspired algorithms to enhance the continual learning capabilities of AI by emulating the neural mechanisms underlying lifelong learning. Neuroscientific research has revealed that corticohippocampal circuits play a critical role in the efficacy of episodic learning and generalization, which are fundamental for lifelong learning. Specifically, the medial prefrontal cortex (mPFC) 2 and the CA1 3 region of the hippocampus (HPC) are thought to represent regularities across related episodes, responding to correlated episodes encountered previously. While regions like the dentate gyrus (DG) and CA3 within the HPC are believed to encode specific memories, selectively responding to particular episodes. Together, these interconnected brain regions form a recurrent loop between the mPFC and HPC, hypothesized to drive the integration of episodic information, facilitating both generalization across episodes and the learning of new concepts. Within this loop, the mPFC-CA1 circuits transmit high-order information derived from prior episodes to modulate novel learning in the DG-CA3 circuits, which subsequently relays newly formed associative memories back to the mPFC-CA1 circuits, thus enhancing the encoding of episode-related regularities 4 , as depicted in Fig. 1 . Fig. 1 Widespread corticohippocampal circuits: facilitating and characterizing dual representation for episode learning and generalization. a schematic diagrams illustrating the brain areas involved in episode learning and generalization. The mPFC interact with CA1 or the anterior hippocampus to represent episode-related regularities, as indicated by the area enclosed by a pink dashed line. The interaction between DG and CA3 within hippocampus represents the specific memories, highlighted by an area enclosed by a green dashed line. b A concise schematic depicting the relationship between the DG-CA3 circuits and mPFC-CA1 circuits. In this study, we emulate the dual representation of corticohippocampal recurrent loops and develop a hybrid neural network, termed CH-HNN, for artificial systems. CH-HNN provides a task-agnostic approach to reduce memory overhead and enhance the practical application of continual learning in real-world scenarios. By integrating artificial neural networks (ANNs) and spiking neural networks (SNNs), we replicate the complementary roles of specific and generalized memory representations within these circuits. ANNs, extensively developed in computer vision, excel at processing high spatial complexity and abstracting image regularities 5 , 6 , analogous to the role of mPFC-CA1 circuits that integrate regularities across episodes. In contrast, SNNs, with sparse firing rates and consequently low power consumption 7 , are used to incrementally encode new concepts, simulating the function of DG-CA1 circuits in specific episode memory formation, as illustrated in Fig. 2 a. The regularities abstracted by the ANNs are designed to guide the SNNs to incrementally learn novel concepts via episode inference. CH-HNN overcomes traditional challenge of integrating these distinct network types and reveals new insights into their synergistic potential. Fig. 2 Hybrid neural networks based on corticohippocampus circuits and metaplasticity mechanisms. a The hybrid neural network comprises an ANN (depicted in pink) and a SNN (depicted in green). The ANN is trained on the similarities among image samples to generate episode-related regularities for each task, which modulate the SNN. The SNN is tailored to learn sequential specific tasks, thereby generating specific memories. b The learning process within the SNN is modulated by metaplasticity mechanism. Large synaptic spines, depicted in green, have stored substantial amounts of memory and learn at a slower rate in subsequent learning. Small synaptic spines, colored in pink, store less memories and are capable of learning additional knowledge. The size of synaptic spines varies across different episodes. c The impact of metaplasticity mechanism on learning dynamics across different spine sizes, illustrating the decline in learning capability as the absolute value of neural weights increases. Additionally, we incorporate metaplasticity mechanisms 8 , 9 into the CH-HNN to simulate the dynamic changes in synaptic learning ability as knowledge accumulates. Specifically, episode-related regularities are believed to have side effects, which could increase the incidence of false alarms when recognizing highly similar episodes in the brain 10 . To mitigate this, the interaction between the lateral parietal cortex (LPC) and DG-CA3 circuits, which encode specific memories, is thought to play a crucial role in reducing such errors 11 . Inspired by this, our model hypothesizes that the LPC modulates the metaplasticity of DG-CA3 circuits by preserving synaptic weights across similar episodes. To evaluate the effectiveness of CH-HNN in continual learning, we applied it to the split MNIST (sMNIST), permuted MNIST (pMNIST) 12 , and split CIFAR-100 13 (sCIFAR-100) in task-incremental scenarios, as well as the sMNIST, sCIFAR-100, split Tiny-ImageNet 14 (sTiny-ImageNet), and split DVS Gesture datasets 15 in class-incremental scenarios. Compared to alternative methods, CH-HNN demonstrates superior performance and achieves a more favorable balance between plasticity (the capacity to learn new information) and stability (the ability to retain previously acquired knowledge), a critical duality in the field of continual learning 16 . Furthermore, CH-HNN demonstrates the ability to transfer related-episode information across various datasets, highlighting its capacity for effective knowledge transfer in diverse scenarios. To investigate the role of the feedback loop from DG-CA3 to mPFC-CA1 in facilitating the encoding of episode-related information, we implemented an incremental learning framework for the ANN within CH-HNN instead of an offline learning approach. The results indicate that as the ANN incrementally accumulates knowledge over time, its proficiency in encoding related-episode regularities significantly improves. These findings suggest that this feedback loop plays a crucial role in promoting episodic generalization by relaying novel embedding. Aligned with our goal of enhancing continual learning in real-world scenarios, CH-HNN demonstrates strong adaptability and robustness across diverse applications. When implemented on neuromorphic hardware, the integration of SNNs significantly reduces power consumption, highlighting the model’s potential for energy-efficient deployment in dynamic environments. In summary, the CH-HNN, inspired by corticohippocampal recurrent loops in the brain, effectively mitigates catastrophic forgetting in both task-incremental and class-incremental scenarios. It demonstrates robustness in real-world applications and shows potential for future implementation into neuromorphic hardware. Furthermore, our study provides evidence that enhances the understanding of the neural mechanisms underlying corticohippocampal functions, contributing to a deeper understanding of lifelong learning from a computational neuroscience perspective.\n\nIntroduce metaplasticity mechanism to CH-HNN In the corticohippocampal loops, research indicates that the modulation signals from the mPFC-CA1 circuit may lead to an increase in false alarms among episodes with high similarity 10 , 27 , 28 , potentially due to the highly similar neural synchrony in downstream circuits. To counteract this hypothesis and enhance the performance of our hybrid neural networks, we introduce a metaplasticity mechanism 8 , which allows synapses to exhibit variable learning capabilities. Typically, metaplasticity at each synapse is modulated by chemical neuromodulatory signals, such as dopamine and serotonin 9 , which can manifest as changes in the size of synaptic spines, as illustrated in Fig. 2 b. In this study, we propose that the LPC 11 , particularly the angular gyrus (ANG), and the lateral prefrontal cortex (lPFC) 29 , which are involved in representing recalled content-specific memories, may play a role in modulating synaptic metaplasticity in the DG-CA3 circuit (Fig. 1 a). To implement the metaplasticity mechanism in SNNs, we adopt an exponential meta-function, as proposed in 30 , to simulate plasticity dynamics of biological synapses. As synaptic weights increase in magnitude, the meta-function output decreases from 1 to 0, as illustrated in Fig. 2 c. Integrating the meta-function into the optimization process during SNN training gradually diminishes each synapse’s learning capacity as knowledge accumulates (details in Methods). This approach has proven effective in alleviating catastrophic forgetting in binary neural networks 31 and SNNs 9 , 30 . Thus far, we have outlined the development of the CH-HNN framework. Moving forward, we will assess its performance and adaptability across both task-incremental and class-incremental learning scenarios using a range of datasets.\n\nMetaplasticity Mechanisms introduced to both SNNs and ANNs The metaplasticity mechanisms are used in the optimization process of SNN’s or ANN’s incrementally learning process, with the modulation of the local synaptic plasticity by modifying the optimization process: 9 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$${{{{\\bf{W}}}}}_{i+1}={{{{\\bf{W}}}}}_{i}-\\alpha f(m,{{{{\\bf{W}}}}}_{i}),\\quad i=1,\\ldots,T$$\\end{document} W i + 1 = W i − α f ( m , W i ) , i = 1 , … , T 10 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$f(m,{{{{\\bf{W}}}}}_{i})={e}^{-\\left\\vert m{{{{\\bf{W}}}}}_{i}\\right\\vert }$$\\end{document} f ( m , W i ) = e − m W i where W i represents hidden weights within SNNs or ANNs, and α represents the learning rate. f ( m , W i ) is set as exponential function, such that can decrease learning rate of local synapse with the neural weights accumulate, from 0 to 1 (see Supplementary Fig. 2 a). While the ANN within our CH-HNN model which is responsible for generate modulation signal is incrementally learned, the meta value m is set as 15 in sCIFAR-100 dataset, and set as 10 in sTiny-ImageNet dataset.",
"discussion": "Discussion The challenge of catastrophic forgetting in artificial systems during continual learning has garnered increasing attention. Incorporating brain-inspired learning mechanisms into artificial algorithms has shown promise in addressing this issue. For instance, generative replay 40 emulates the complementary roles of the cortices and hippocampus in managing long-term and short-term memories by storing generative features of old tasks and reusing them during new task learning. Additionally, metaplasticity methods introduce a global modulation mechanism that adjusts synaptic plasticity, offering another brain-inspired strategy to mitigate catastrophic forgetting 9 , 31 . Despite success in specific contexts, continual learning faces challenges, particularly in real-world applications. Methods such as generative replay encounter growing memory demands as tasks accumulate. Additionally, metaplasticity-based approaches, while effective on simpler datasets like MNIST, tend to perform relatively poorly on more complex, real-world data. Furthermore, methods like XdG, which are not task-agnostic, rely on a task oracle, limiting their applicability in real-world scenarios. To address these limitations, we develop a novel method termed CH-HNN, which integrates ANNs and SNNs into a hybrid neural network inspired by recurrent corticohippocampal loops. CH-HNN eliminates the need for a task oracle, exhibiting strong performance and power efficiency in real-world applications. While CH-HNN supports diverse neuron models, selecting the appropriate model involves trade-offs. Complex neuron models like the EIF model enhance biological realism and accuracy but demand more computational resources compared to simpler models like LIF and IF. Although the exponential term in the EIF model can be efficiently managed using a look-up table and results in only a modest power overhead of 8.35% to 8.58% on the “Tianjic” chip, real-world applications must still consider trade-offs among performance, memory cost, biological plausibility, and hardware compatibility to meet specific demands. From the perspective of neural mechanisms in corticohippocampal circuits, CH-HNN provides indirect evidence for potential neural mechanisms. First, modulation of the feedforward loop from mPFC-CA1 to DG-CA3 can be achieved by resetting the neural synchrony state, offering complementary insights to generative replay methods 40 , which propose direct transfer of old knowledge from the cortices to the hippocampus. Second, novel embedding transfers from feedback loops potentially enhance the generalization of related memories. Third, certain regions of the brain, such as the lateral posterior cortex and lateral prefrontal cortex, can modulate metaplasticity in the DG-CA3 circuits through chemical neuromodulatory signals. Furthermore, there is ongoing debate regarding how the brain represents concepts—whether through distinct engrams or episode inference. The success of our CH-HNN method suggests that episodes are not encoded by discrete engrams but are instead processed through guidance based on episode-related information. Although our model simplifies the recurrent loop as mPFC-CA1 and DG-CA3, other research emphasizes the role of the EC in episode-related representations 21 . Furthermore, evidence indicates different functions for the anterior and posterior hippocampus. The anterior hippocampus interacts primarily with regions of the brain associated with generalized knowledge, whereas the posterior hippocampus is involved in specific memory representations. These findings further elucidate an interpretable neural mechanism underlying lifelong learning. While current continual learning algorithms, including our CH-HNN model, effectively leverage prior knowledge to achieve high performance 41 , 51 , comparing models with prior knowledge to those without dedicated prior-learning mechanisms may seem imbalanced. Nevertheless, it is important to highlight that our approach offers an indirect method of prioritizing prior knowledge to facilitate new concept learning, potentially reflecting a neural mechanism for lifelong learning in the brain. Given that humans can sequentially acquire new concepts from only a few examples, therefore, integrating few-shot learning 52 with continual learning could be a promising avenue for future research. This integration may enhance the adaptability and efficiency of online continual learning in dynamic environments. Additionally, our model may encounter challenges in contexts where task correlations are limited, as it relies on the presence of natural or designed correlations among incremental episodes. In conclusion, our study introduces a model that simplifies the simulation of corticohippocampal recurrent circuits, improving the performance and adaptability of continual learning in real-world applications, while emphasizing the potential of integrating neuroscientific insights into artificial intelligence systems."
} | 4,537 |
39475634 | PMC11551383 | pmc | 7,277 | {
"abstract": "Significance We established a strain learning mechanical metamaterial that can not only recover after plastic deformation but also become stronger and stiffer in response to the applied loads. We demonstrate that shape-memory protein–polymer networks can unravel to release their stored length and, after recovery, will reconfigure to accommodate higher loads. These materials are incorporated into lattice frameworks using additive manufacturing which are then shown to become up to 2.5× stiffer after being crushed to 80% strain and recovered. This strain learning behavior enables the creation of metamaterials that are “teachable” and can autonomously remodel in response to applied loads, similar to the processes that occur in natural materials.",
"discussion": "Results and Discussion Strain Learning Behavior via BSA Unfolding. A typical thermoset becomes stiffer as the number of cross-linking junctions increases, but this enhancement also comes with a decrease in the extensibility of the material (i.e., a decrease in toughness) ( 25 ). However, uniaxial tensile experiments on dogbone samples with varying ratios of BSA/PEGDA showed an unusual behavior in these materials: As the concentration of BSA junctions in the network increased, so did the Young’s modulus and the elongation at break ( Fig. 2 A ). We analyzed how the presence of BSA junctions in a polymer network affected the bulk mechanical properties by 3D printing three different compositions of BSA/PEGDA resins with weight ratios of 25/75, 50/50, and 75/25. All BSA/PEGDA samples exhibited improved mechanical properties relative to the neat material network. The BSA/PEGDA 75/25 showed a clear yield point followed by a plateau region representing the plastic deformation of the material. Thus, these protein–polymer thermosets enhance their tensile toughness with increasing amounts of protein, which is attributed to the release of protein stored length. Fig. 2. Characterization of BSA/PEGDA networks before and after mechanical loading shows the release of the stored length in the protein junctions. ( A ) Uniaxial tensile experiments were performed on dogbone samples of PEGDA, BSA/PEGDA compositions of 25/75, 50/50, and 75/25. ( B and C ) Coarse-grained molecular dynamics simulations of the BSA/PEGDA networks with the same composition as the tensile specimens. The number of polymer chains per protein were varied in the simulations from 11 to 25, but this variation did not have a significant effect on the response of the proteins to applied strain. ( D and E ) Dogbone samples of BSA/PEGDA 75/25 were pulled to the different strains shown. The original dogbone length was recovered after submerging in H2O and dehydrating the sample, and then, the uniaxial tensile experiment was repeated. Samples with strains higher than 12% showed a higher modulus in the second cycle. ( F ) SAXS was performed on BSA/PEGDA 75/25 samples to compare the as-printed bioplastic (control), the compressed bioplastic, and the recovered bioplastic. The experimental results were supported with coarse-grained molecular dynamics simulations of the BSA/PEGDA networks comprising the same mass ratios of BSA to PEGDA. Given the huge computational resources of the all-atom model of BSA, we developed a simpler coarse-grained model ( SI Appendix , Fig. S3 ). The Go-like model ( 26 – 28 ) allows for bond breaking and the uniaxial deformation of the material was considered at a constant volume and strain rate of 100% ns −1 . These simulations showed that the strain at the onset of material failure (the point where 20 bonds are broken) was 110, 124, and 147% for BSA/PEGDA compositions of 25/75, 50/50, and 75/25, respectively. The simulations correlate closely with the values from the uniaxial tensile experiments and confirmed that BSA content improves the mechanical properties of these networks. While BSA has approximately 30 surface lysines, it is difficult to quantify the degree of functionalization of the proteins in these resins. Interestingly, our simulations show that the onset of material failure is independent of the number of PEG chains bound to the surface of the protein (between 11 and 25). Understanding the protein’s behavior within the network, we devised a strain learning regimen for these materials. By examining the stress–strain curve of BSA/PEGDA 75/25, we opted to subject them to varying strains: one below the yield point (12.5%) and others above (25, 50, 75, and 100%). Leveraging the hydration-induced shape memory feature of these materials ( 21 , 22 ), we applied the respective strain, restored their shape (99% recovery) via hydration and dehydration of the samples, and then reapplied stress until failure ( Fig. 2 D ). When the load applied to the network is below the yield point of the material, the mechanical properties (elastic modulus, elongation-at-break, and yield point) remain unchanged after shape recovery. On the contrary, if the applied force exceeds the yield point and enters the plastic region, the properties are modified due to the unfolding of the protein. Regardless of the force applied, once the yield point is exceeded, the shape-recovered dogbones demonstrated an elastic modulus and yield strength nearly doubled in value. For example, when a strain of 25% was applied, a modulus of 309 ± 21 MPa and a yield strength of 22 ± 1.1 MPa was obtained. After shape recovery (by immersing the sample in water and then dehydrating under ambient conditions), the elastic modulus and yield strength increased to 492 ± 23 MPa and 35 ± 1.2 MPa, respectively, representing a ~60% increase in the properties. The enhanced mechanical properties are attributed to the reconfiguration of the unfolded BSA after shape recovery. Small-angle X-ray scattering (SAXS) of the networks showed the nanoscale behavior within these protein–polymer networks upon compression and recovery ( Fig. 2 F ). A feature corresponding to the length scale of BSA (at approximately Q = 0.115 Å −1 , corresponding to 5.5 nm in real-space) in the initial sample is no longer seen after compression, exhibiting a reduction in molecular order upon deformation as schematically depicted in ( Fig. 1 , State 2). Interestingly, shape recovery (via hydration and subsequent dehydration) recovers this feature with even greater intensity than initially present ( Fig. 1 , State 3). The peak corresponding to 5.5 nm re-emerging after hydration recovery suggests that the BSA is refolding to a globular form that is comparable in size to its native state. The reduction and re-emergence of the feature are repeatable after an additional cycle of compression and hydration recovery ( SI Appendix , Fig. S10 A ). In contrast, thermal shape recovery (via heating at 120 °C) does not recover this scattering peak, and instead, the scattering profile indicates disordered aggregates of denatured BSA molecules ( SI Appendix , Fig. 10 B ). Thermal shape recovery at this temperature promotes denaturation and protein refolding does not occur. Therefore, the compression samples can only be cycled twice using thermal recovery before catastrophic sample failure occurs, while the samples recovered via hydration and drying could be cycled ten times. The SAXS measurements indicated consistent structural recovery via hydration for both compression and tensile experiments ( SI Appendix , Fig. S11 ). We additionally observed that tensile deformation of the sample caused the material to become oriented along the stretching axis. The degree of orientation, calculated with a tensor approach method (1 is perfectly aligned, 0 is fully isotropic) for a Q-range of 0.089 to 0.150 Å −1 at sample-to-detector distance of 370 mm ( 29 ), depended upon the tensile strain. The first cycle of tensile stretching (75% strain) led to a degree of orientation of 0.170, and the second cycle (50% strain to break) led to a 0.123 degree of orientation. Interestingly, upon hydration recovery and drying of the sample, the network returned to a nearly isotropic state, showing at most 0.005 degree of orientation over the same Q-range after recovery from either deformation cycle and for the as-printed sample. These data further suggest protein unfolding during mechanical deformation ( Fig. 1 , State 2) and protein refolding during the shape recovery (back to Fig. 1 , State 3). Metamaterial Design for Energy Dissipation in Shape Memory Bioplastics. We fabricated lattice-based metamaterials that exhibit varying degrees of stress-induced stiffening depending on their architecture, thereby capitalizing on the strain hardening observed in the constituent material. We used three lattice architectures—octet-truss (OT), truncated octahedron (TO), and twisted truncated octahedron (TTO) ( Fig. 3 ) and printed them using the BSA/PEGDA 75/25 mixture as it demonstrated the best mechanical performance. The lattices were each made with three different relative densities of ρ = 6, 9, and 12%. The OT lattice has been extensively studied in the literature, as has the TO lattice (also known as a Kelvin cell) ( 23 , 30 – 32 ). The TTO lattice was initially developed as a tensegrity architecture capable of sustaining a prestress ( 33 , 34 ), but it should be noted that this and other works on this architecture using additive manufacturing ( 35 , 36 ) are not prestressed and thereby do not constitute a true tensegrity. Fig. 3. Uniaxial compression experiments were performed on 3D printed OT, TO, and TTO lattices. ( A ) The images show that with increasing strain, the energy is dissipated differently based on the lattice design. The front face of the lattices in the optical images was colored with a black marker pen to provide contrast. ( B ) A plot that compares stress vs. strain curves for OT, TO, and TTO lattices with a density of 6%. ( C ) A plot that shows the stress vs. strain curves for TTO lattices with a density of 6, 9, and 12%. We performed uniaxial compression at quasistatic strain rates (έ = 10 −3 ) on each of the lattices to characterize their strength, stiffness, and failure resistance after load-recovery cycles. It is well understood that the strength, stiffness, and energy dissipation depend strongly on the relative density, but these metrics do not capture the uniformity of the deformation in the lattice, which thereby affects the unfolding of the protein under an applied load. Fig. 3 A shows how the energy is dissipated differently based on the lattice design. While the stress-strain responses of the lattices are similar—with a characteristic linear region, plateau region, and densification region—the OT and the TO lattices are not able to dissipate energy uniformly and experience a collapse in their top and bottom layers. This is because the beams of these lattices undergo buckling which causes a softening response and thereby a localization of deformation in a layer. This metamaterial localization causes high stress and corresponding damage at the material level in the collapsed beams. Reducing the density of these lattices reduces the amount of plastic deformation experienced by the slender beams, thereby promoting their resistance, but it does not mitigate the issue of localized failure. The TTO lattice uniquely does not experience any localization and has a uniform deformation throughout its structure even up to 80% strain. This behavior is observed regardless of relative density and only occurs with this lattice geometry. The Young’s modulus of lightweight materials is known to depend strongly on density. For lattices, there is an approximate scaling relationship of E = CE s ρ m , where C is a scaling coefficient, E s is the constituent material modulus, and m is a scaling coefficient that is generally between 1 and 2 depending on the fraction of the lattice that carries load in tension/compression vs. in bending ( 37 ). For lattices with a relative density higher than ~5%, the nodes of the lattice significantly affect the load-carrying pathways such that any “bending” or “stretching” scaling assumptions are incorrect ( 31 ). The modulus-density scaling coefficients for the OT and TO lattices observed here are both m = 1.6 to 1.7, which matches well with previous observations for these lattice architectures in this density range whose properties are largely affected by nodal stress concentrations. Interestingly, the scaling coefficient for the TTO lattice is m = 0.75, but it should be noted that this is a misleading quantity because the mechanical efficiency (i.e., the C value) is low ( SI Appendix , Fig. S12 ). These scaling coefficients are all also highly approximate due to the limited density range of specimens studied here. To understand the differences in recoverability of the lattices at different densities, we must understand the role of the nodes as stress concentration points and of the strain in beams during bending. The sharp intersection between beams in a lattice naturally results in stress concentration. In high-density lattices, the proximity of the nodes amplifies their stress concentration effect, while low-density lattices carry load more optimally in tension or bending of their beams according to the rigidity of their topology ( 31 ). The smaller diameter beams in low-density lattices also experience a lower strain for the same amount of bending. This is directly reflected in the poor cyclic recoverability of all lattices at high relative density, which necessarily experience higher and less uniform local strains for the same global applied strain. The local strains will cause the constituent chains to unfold and harden, but if the strain exceeds the failure strain of the material, it will lead to damage. The differing levels of cyclic recoverability can thereby be understood in the context of a competition between local damage and global strain hardening of the constituent. Although our blends are capable of dissipating energy due to protein unfolding, they do not dissipate the energy uniformly throughout the material, and this can lead to failure. As a comparative control experiment, we printed the TTO lattice using only PEGDA ( SI Appendix , Fig. S13 ). As with BSA/PEGDA 75/25, an increase in modulus was observed with increasing lattice density. However, an important difference is that the PEGDA-only networks did not demonstrate an increase in stiffness when a second cycle of compression was applied. In the case of density of ρ = 12% PEGDA-only sample, the structural integrity was lost during the first compression cycle.\n\nDiscussion The 3D printed lattices showed mechanically induced stiffening after shape deformation and recovery. Strain learning experiments were performed to analyze the degree of shape recovery, the enhancement in modulus due to protein remodeling, and the number of cycles before lattice failure ( Fig. 4 ). All the lattices were compressed to 80% strain and then recovered to 95% of their original size after a 30-min immersion in water followed by dehydration ( Fig. 4 B ). Interestingly, the compression modulus increases from the first cycle to the second cycle as a consequence of the network structure remodeling that occurred as the BSA unfolded under mechanical load ( Fig. 4 A ). Thus, the protein junctions present in the network release their stored length at the molecular scale to enable large deformations of the lattice at the macroscale. The presence of the permanent chemical cross-links in the network from the polymerized PEGDA provides the shape memory. It is possible that refolding of the proteins may also facilitate shape recovery during the hydration step since proteins have been observed to spontaneously refold in solution and in hydrogels ( 38 – 41 ). Our SAXS data ( Fig. 2 F ) suggest that the proteins refold into globular structures with approximately 5.5 nm spacing after shape recovery. However, the enhanced mechanical stiffness that we observe ( Fig. 4 B ) suggests that the proteins are not refolding back into their original state. Instead, the proteins are likely misfolded into globular conformations that ultimately strengthen the material. Of the three lattice designs tested, the TTO lattice with 6 and 12% showed the best performance with ~2.5-fold enhancement in the modulus between the first and second compression cycles ( Fig. 4 and SI Appendix , Table S2 ). For the TTO lattice with density 9%, the modulus was similar between the first and second compression cycles. Across the three lattice designs, we observed the modulus to decrease during the third cycle, which is attributed to reduced ductility and the onset of damage in these materials once the original globular proteins were unfolded ( SI Appendix , Fig. S14 ). Fig. 4. Strain learning in multicycle compression test results for OT, TO, and TTO lattices. ( A ) Compressive moduli after 3 compression-recovery cycles. Each lattice, with a density of 6, 9, or 12% was compressed to 80% strain, recovered via hydration for 30 min, and dried for 48 h, before the next cycle of compression-recovery was performed. ( B ) Percent recovery for OT, TO, and TTO lattices with different densities. The synergy of protein mechanophores with 3D printed lattice architectures enables thermoset materials with unique mechanical responses to deformation. BSA is often a model protein employed in many contexts, and here, it served as a well-folded, single-chain globular protein that was incorporated into polymer networks during a vat photopolymerization process. The excellent solubility of this protein in aqueous solutions and the presence of surface lysines for functionalization contribute to its performance as a resin component for 3D printing. An unexpected feature of these networks was the improvement in mechanical properties as the number of BSA junctions increased. Typically, as the number of junctions in a thermoset increases, there is an overall decrease in the toughness of the material. We observed that as the BSA concentration in the protein–polymer network increased, both the modulus and the strain at failure (and therefore, toughness) increased. Thus, these large globular proteins do not operate as traditional network junctions. We posit that other globular proteins with similar solubility and surface functionality should also demonstrate protein–polymer networks that exhibit unique mechanical responses based on protein primary, secondary, and tertiary structure. While naturally occurring proteins may fit this purpose, engineered proteins could hold the key to truly unlock the potential for these 3D printed protein–polymer networks."
} | 4,652 |
28097922 | PMC5560479 | pmc | 7,278 | {
"abstract": "When surveying the trends and criteria for the design for recycling (DfR) of bio-based polymers, priorities appear to lie in energy recovery at the end of the product life of durable products, such as bio-based thermosets. Non-durable products made of thermoplastic polymers exhibit good properties for material recycling. The latter commonly enjoy growing material recycling quotas in countries that enforce a landfill ban. Quantitative and qualitative indicators are needed for characterizing progress in the development towards more recycling friendly bio-based polymers. This would enable the deficits in recycling bio-based plastics to be tracked and improved. The aim of this paper is to analyse the trends in the DfR of bio-based polymers and the constraints posed by the recycling infrastructure on plastic polymers from a systems perspective. This analysis produces recommendations on how life cycle assessment indicators can be introduced into the dialogue between designers and recyclers in order to promote DfR principles to enhance the cascading use of bio-based polymers within the bioeconomy, and to meet circular economy goals.",
"conclusion": "Conclusion From a methodological point of view, analysing the incorporation of DfR criteria in actual polymer design activities and examining the most common recycling paths of these polymers through a series of cascading stages helps us to establish the qualitative prerequisites and the goal and indicator systems in order to carrying out a systems analysis of the techno-environmental performance of recycling strategies for bio-based polymers. The most extensive demands for research lie in forecasting future production capacities of bio-based polymers and in surveying the trends for introducing durable bio-based polymer products onto the market. A more accurate projection of future use of long-life durable goods is needed in order to estimate the future release of bio-based waste flows from materials stocks, such as buildings, vehicles and other long-life items, back into the recycling system. In addition to bio-based packaging, one major issue is to establish when a critical mass of bio-based postconsumer waste flows is reached so that material recycling will become a real business case and the recycling infrastructure can adapt to high annual mass streams of bio-based plastic waste. By assessing the existing solutions for DfR of polymer products, the following conclusions and recommendations should be taken into account in order to promote the right design for enhanced cascading use. Recommendations from the design perspective Promoting polymers with the most flexible EOL options and customizing them with the least harmful additives for the use phase and cascade use: Recognizing that the range of additives is almost more important in determining the product properties and recyclability than the intrinsic properties of the polymers themselves. It appears that only bioplastic polymers with the broadest possible range of EOL options should be promoted in the long-term. Promoting expandable thermoplastics such as expanded PLA for extended cascade use options: Recycled polymers should be used for durable foam-based products, such as bio-based furniture and insulation materials, since they can serve as a platform for prolonging material use without conflicting with the first premise of food safety. Promoting fully certified, biodegradable and bio-based plastics such as polyhydroxyalkanoates and starch-based polymers for applications with low collection rates or low energy recovery rates: As the future materials of choice for non-durable products with low collection rates, only biodegradable polymers with a full range of biodegradability (compost, soil, water environments) that are harmonized to the process conditions and retention times of technical facilities and to environmental conditions should be introduced in the long-term. Recommendations from the perspective of pre-standardization and innovation management Rethinking the product design of consumer and durable goods under the food safety premises: Under the first food safety premise, and based on the observation that there will be new durable products made from foamable bio-based polypropylene and PLA foam, food-safe applications should be prioritized for virgin materials in consumer goods that have high collection quotas and good sortability in NIR sorting processes and durable products should be made from recycled materials enjoying a second product life since additives, such as flame retardants, are mandatory and would compromise food safety in the first place. Pre-standardizing novel consumer-centric manufacturing paths and extended polymer resin coding information: An enhanced coding system explicitly connecting the information chain between future application options and resin codes is required for bio-based polymers and for novel manufacturing processes such as 3D printing. It would be a valuable solution for linking DfR with cascade use while ensuring resource efficiency for filament use in distributed manufacturing. Recommendations from a monitoring and circular economy perspective: Establishing statistical databases for the recycling processes of bio-based industries and novel consumer-centric fabrication labs: In order to monitor polymer cascades, it would be helpful to have a pan-European statistical database for the average number of extrusion cycles and extended lists of qualitative criteria for additive compatibility and compatibilizer options. For emerging technologies, such as 3D printers, these statistics should be automatically submitted electronically to central assessment databases.",
"introduction": "Introduction Using bioplastics to replace fossil resources and to harness novel polymer properties has received much attention in recent decades. At approximately 7.848 million t a −1 (ca. 0.7%), the global production of bioplastics currently makes up only a fraction of the world’s overall plastic production ( European Bioplastics et al., 2015 ; PlasticsEurope, 2015 ). While this share is expected to increase significantly in the coming decades, there are major uncertainties with regard to calculating this increase; projections vary between 6% and 30% of the total production volume by 2020 and 2030, respectively ( IfBB, 2014 ; NNFCC and Williams, 2011 ). Because bioplastics are being introduced gradually to the market, establishing the right infrastructure and proper standards to govern the recycling and treatment at their end of life (EOL) will also become increasingly more relevant. In this regard, successfully implementing optimized recycling cascades for bioplastics is expected to play a key role in decreasing material intensity and increasing material efficiency of limited biomass resources ( Carus et al., 2014 ). The initial experience of recycling plant operators indicates that there are basic needs and problems in treating bio-based polymers. In sorting systems for lightweight packaging, drop-in polymers, such as bio-based polyethylene terephthalate (PET) and polyethylene (PE), are fully compatible with near-infrared (NIR) sorting. In contrast, starch blends and PLA require NIR spectrometers to be more finely calibrated ( Hollstein and Wohlebe, 2015 ). In bio-waste treatment plants, biodegradable and bio-based plastic bags may be increasing the collection quota of organic waste; however, they are not contributing to compost quality. As a result, they are most commonly sorted out and used energetically in thermal treatment plants ( Association of German Municipal Utilities, 2013 ; BGK, 2014 ). In order to clarify the situation with regard to EOL paths and options, improvements are needed, starting with the product design. This includes eco-design initiatives and the improvement of product compatibility with different national recycling systems ( European Commission, 2015 ), that is, by formulating advanced criteria for design for recycling (DfR). DfR offers guiding principles for addressing the qualitative requirements of the recycling sector through eco-design. It integrates suitable rules, such as ‘use of recyclable material’, ‘minimizing the number of parts involved in product components’, ‘minimizing the number of different types of material’, ‘marking parts for easier identification’ or ‘making the product easy to disassemble’ ( BIO Intelligence Service, 2011 ). The first initiatives and guidelines for DfR were launched at the end of the 1970s. In the last 15 years there have been more specific engineering standards and procedures, such as VDI 2243, the aim of which is to incorporate DfR principles into product design ( VDI, 2002 ). A broad adoption of these criteria is expected to contribute to a better revalorization of polymer waste streams as part of the circular economy strategy, despite other strategies such as the reuse or remanufacturing of products. The impact and benefits of applying DfR principles for better sustainability and performance of polymer products can be evaluated through life cycle assessments. Even though studies on sustainability metrics exist, they only consider cradle-to-gate system boundaries ( Tabone et al., 2010 ) and not extended EOL scenarios. In order to select life cycle sustainability indicators capable of assessing the influences of DfR on the subsequent cascading use of polymer materials, the authors designed the assessment framework presented in Figure 1 . It covers the following qualitative and quantitative aspects of polymer recycling. At the product level , the decomposition or recycling of individual bio-based polymer waste flows at their EOL can occur in natural and technical environments. The producers and product designers of bio-based polymer products who use virgin materials can influence these EOL treatment paths by meeting and improving recyclability standards and by applying DfR criteria. The various EOL and recycling paths must first be specified from a system’s perspective in order to specify material flow parameters for comparing intended recovery performance, recovery in the most prevalent final treatment paths and the potentials for increasing material recovery. Therefore, the assessment framework focuses on identifying, characterizing and quantifying the options for enhancing the recyclability of individual bio-based polymer products during the design and manufacturing stages. At the circular economy level , quality standards have to be harmonized between key sectors (e.g. composite manufacturing, packaging and construction industries) in order to guarantee the quality needed when using secondary raw materials derived from bio-based polymers. Thus, these standards can support a more wide-spread use of recycled plastics in polymer processing and thereby help to potentially increase the volume of innovative bio-based plastics that can be recycled through multiple cascade use stages. Therefore, the assessment framework also focuses on identifying relevant polymer innovations and areas where standards can be harmonized. This will enable recycling paths to be identified, which will increase the throughput of secondary raw materials in the future. At the recycling sector level , plastic recyclers have to guarantee the quality of recycled polymer materials despite being increasingly faced with challenges posed by novel polymers, additives, etc. This requires ongoing polymer developments and capabilities to be assessed in current and future recycling systems. The extent to which standards for EOL management have to be adapted and harmonized also needs to be identified. Therefore, the framework focuses on identifying and characterizing product standards, recycling processes and quality requirements for secondary raw materials in order to set up qualitative benchmarks for interpreting product-specific recovery rates along the recycling paths. Figure 1. Methodological procedure proposed for assessing cascade use performance. DfR: design for recycling. The aim of this paper is to explore and compile the most crucial aspects (‘trigger points’) of where the design of bio-based polymers conflicts with and where it reinforces circular economy goals for extending further material life cycles of recycled polymers. It also aims to establish a preliminary system of objectives, which would allow quantitative and qualitative criteria to be broken down in order to assess the performance of the cascade use of bio-based plastic waste.",
"discussion": "Results and discussion Analysis of design options and obstacles for DfR The environmental strategies of product design are mainly concerned with extending the useful life of the product and with the type of recovery after their EOL. To provide insights on selected DfR criteria as compiled, for example, by Dwek and Zwolinski (2015) , the following common design conflicts can occur when designing polymer products: the design of reinforced composites can conflict with the DfR criterion ‘increasing ease of disassembly’; the use of fillers and the blending of polymers can conflict with the DfR criterion ‘reducing material variety’; the use of flame retardants and plasticizers can conflict with the DfR criterion ‘reducing hazardous material’; improper disposal routes can conflict with the DfR criterion ‘increasing recyclability’. In addition, future trends and drivers for the following design strategies have been identified that extend useful product life ( Giudice et al., 2006 ): prototyping for three-dimensional (3D) printing can enhance the DfR criterion ‘designing multifunctional materials’. One important step in the design process is to choose suitable materials for creating the envisioned affordances, such as affording recyclability ( Maier and Jonathan, 2007 ). Selecting the material used for a certain product purpose determines whether further augmentation steps are needed to meet the design requirements, such as the need for additives in polymer processing and the stabilization of product properties ( Giudice et al., 2006 ). In this sense, the additives are especially important for facilitating polymerization, processing, stability against degradation and improving mechanical, structural and surface properties ( Bart, 2006 ). Table 2 presents the main fields of application for a selection of the most frequently used additives, as well as the potential negative side-effects. Table 2. Application of standard additives for selected bio-based polymers (adapted from Bart, 2006 ; bioplastics magazine, 2015 ; EC Health and Consumers Directorate-General, 2014 ; Imre and Pukánszky, 2013 ; Villanueva and Eder, 2014 ; Weil and Levchik, 2016 ). Polymer Additives Exemplary types of additives Product application Possible negative effects on second use phase Legislative coverage Bio-based PU and PP Flame retardants Metal hydroxides, brominated or phosphorus additives Wires, cables, insulation panels and furniture Can compromise food safety in the 2nd product life Brominated flame retardants should be removed according to WEEE (2012/19/EU) Bio-based PET and PE UV-protection, stabilizers and absorbers Organic and organo-metallic stabilizers Plastic bottles, outdoor equipment Maximum sum content of heavy metals 0.001% w/w according to packaging waste directive Bio-based PET and PE Plasticizers/ softening agents Phthalates, citrate esters Plastic bottles Can compromise e.g. recycling of PET in textile industries Phthalates are regulated under REACH, food contact legislation, and RoHS Unsaturated polyester resins and acrylic resins, Fillers and modifiers Mineral fillers such calcium carbonate Thermally conductive plastics, such as marbles and panels Can be beneficial for stabilizing recyclate PE-PLA blends Compatibilizers Co-polymers, maleic anhydrid Mixed polymer recyclate Blends in the 1st use phase affect blendability in 2nd use phase PU: polyurethane; PP: polypropylene; PET: polyethylene terephthalate; PE: polyethylene; PLA: polylactic acid; UV: ultraviolet WEEE: Waste Electrical and Electronic Equipment Directive; ROHS: restriction of hazardous substances directive; REACH: European regulation on registration, evaluation, authorisation and restriction of chemicals. Achieving DfR criteria is often compromised by design conflicts caused by the intended and unintended side-effects of additives. Unintended side-effects caused by additives include human and eco-toxicity effects, especially through materials that come into contact with food, altered biodegradability and deterioration in further extrusion processes, to name a few of the most important concerns ( Bart, 2006 ; EC Health and Consumers Directorate-General, 2014 ). The recycling codes and the food safety symbol currently offer a reliable system for sorting out the additives’ ingredients and the types of polymers found in post-consumer recycling waste. However, they cannot compensate for the lack of a prioritization strategy for appropriate recycling sequences under the conditions of additive restrictions. The most critical additives and their required removal during the respective recycling phase are covered by the Packaging Waste Directive, food contact legislation, the Waste Electrical and Electronic Equipment Directive (WEEE), etc. ( Villanueva and Eder, 2014 ). Moreover, the choice of material and its compatibility with existing polymer types predetermines the specific EOL treatment of bio-based polymers and how they interact with the other polymers in cascade use with regard to blendability, separability and applicable recycling processes (as shown in Table 3 ). Table 3. Common recycling processes and deterioration rates for selected polymers (adapted from Hamad et al., 2013 ; Hollstein and Wohlebe, 2015 ; La Mantia 1998 ). Types of emerging solutions Polymer type Common separation efficiency and Impurities Recycling process Most prominent secondary products Relevant criteria for recycling Most relevant properties Recycling processes Drop-in solution Bio-based PET NIR sorting applicable, no extra separation needed Extrusion with virgin material, polyester fibres production Polyester fibres, non-food bottles Decreasing melt viscosity <3 Extrusion cycles Bio-based PE NIR sorting applicable, no extra separation needed Extrusion melting Downcycling (non-opaque materials), refuse derived fuel Deteriorated melt flow index (MFI) <4 Extrusion cycles Novel polymer solution PLA Small amount blendable with PE or PET, extra separation or collection with organic waste a. Home and industrial composting or b. Screw extrusion or injection moulding Refuse derived fuel, thermal recycling, mineralization Decreasing intrinsic viscosity <2 Extrusion cycles Blends of novel polymers and fossil-based polymers PLA-PMMA blend Bulky waste/ scrap collection Thermal treatment or pyrolysis Waste2Energy or MMA as recovered feedstock Mineral contents, heavy metal contents content of halogenated compounds Recovery efficiency of pyrolysis PET: polyethylene terephthalate; PMMA: polymethylmethacrylate; PE: polyethylene; PLA: polylactic acid; NIR: near-infrared; MMA: methylmethacrylate. As a drop-in solution, the virgin material of bio-based PET can be processed just like its fossil-based competitors in the first product use phase. For recycled PET (both fossil and bio-based), the content of recycled secondary raw materials in the second product use phase is limited by the proportion and molecular weight of the virgin material and by its contribution to the deterioration of the mechanical properties. A content of 30% recycled PET appears to be possible without significantly affecting mechanical properties ( La Mantia, 2002 ). However, the material recycling of polyolefin and PET waste flows remains low compared to energetic recycling. A major portion of PE (67%) and PET (50%) is energetically recycled in MSW incineration plants and a further 15% is used energetically as solid recovered fuels (SRF) in the cement industry and SRF-based power plants ( Wagner et al., 2012 ). The miscibility of polylactic acid (PLA) with polyolefin and PET for blending is limited, whereas poly (methyl methacrylate) blends allow for a miscibility of up to 50% PLA content ( Imre and Pukánszky, 2013 ). Analysis of trends and innovations in manufacturing, recycling and end-of-life treatment Emerging trends in recycling, material science and manufacturing are surveyed and analysed qualitatively in the following analytical section. The analysis provides insights into the links between these emerging trends and their resulting options and risks for meeting specific DfR criteria in future product design processes. The analysis focuses in particular on the following four trends, which were considered because of their wide-spread influence on the options for material recovery alongside the life cycle stages of bio-based polymers: introduction of novel bio-based polymers to broaden design options and recycling paths; diffusion of novel manufacturing technologies to influence polymer flows and recycling paths; introduction of novel additives that comply with more restrictive legal thresholds in recycling; harmonizing standards to improve biodegradability or overall recyclability. Do emerging materials meet the DfR criterion ‘increasing recyclability’? PLA is an ideal prototype of bio-based polymers as it is polymerized from a biogenic compound (lactic acid), is a mouldable thermoplastic, deteriorates only moderately through a subsequent series of extrusion cycles, can potentially be biodegradable even in home composting, can be blended with PE and PET and can also be a basis for the production of expanded foams ( Detzel et al., 2013 ; Madival et al., 2009 ). As of today, however, estimated decomposition times of 0.5–1 year in mechanical biological treatment (MBT) facilities and home composting respectively remains questionable ( Association of German Municipal Utilities, 2013 ) and mainly non-durable products are made out of PLA. Therefore, the future development of novel material recipes and foaming additives for PLA may play an important role if durable, biodegradable, bio-based foams are developed and upscaled. Currently there are several research and development (R&D) activities and medium-scale industrial production capacities that are researching options for manufacturing novel products based on durable poly(d-lactic) acid (PDLA) ( Synbra, 2015 ). The possible applications for expanded PDLA range from food packaging applications to insulation materials with a high potential of substituting conventional expanded polystyrene ( Dean, 2013 ; Synbra, 2015 ). Furthermore, the developments of PLA-based blends and the testing of suitable compatibilizers may open up further routes for the production of polymer products with enhanced durability and varying recycling routes, such as the pyrolysis of PLA-polymethylmethacrylat (PMMA) blends ( Groot and Borén, 2010 ; Imre and Pukánszky, 2013 ; Lopez et al., 2010 ). In summary, the trend analysis shows that (a) manufacturing technologies are a key factor in achieving product performance for durable products made from bioplastics, for example, in flexural strength or high insulation values and (b) for non-durable products, the standards for biodegradability have to be harmonized with the actual retention times of biodegradable polymer waste in MBT and composting facilities. Do manufacturing trends, such as 3D printing, meet the DfR criterion ‘reducing material variety’? In polymer manufacturing industries several disruptive technologies, such as 3D printing applications and new social initiatives that involve consumers in product design and production processes, such as FabLabs, are on the rise and directly or indirectly affect the associated disposal paths and their related resource use efficiencies ( MGI et al., 2013 ; Riverra and Van der Meulen, 2014 ). The major advantage of 3D printing technology is its capability of manufacturing almost any kind of consumer good while simultaneously allowing a wide variety of bio-based polymers to be used ( van Wijk et al., 2015 ). Recent developments have already induced the need for rapid pre-standardization of environmental and material standards in additive manufacturing and have also led to promising prototypes in the 3D printing application of mono-compound furniture (AM Platform and Feenstra, 2014 ; Van Daal, 2015 ). However, recycling open-source 3D printing prototypes may introduce risks to recyclability by breaking the information chains and therefore require consistent resin coding standards for 3D printer filaments in order to prevent rebound effects in resource use efficiency ( Hunt et al., 2015 ). In summary, the trends for novel manufacturing technologies indicate that material variety will be reduced at the product level but, at the same time, the variety and number of polymer types in post-consumer waste is likely to increase. Do manufacturing trends such as supercomputing, digital prototyping and biomimicry meet the DfR criteria ‘multifunctional materials and reducing material variety’? In the coming decades, the options for programming surface and structural functionality of plastic polymers will offer a whole new range of opportunities for product designers, especially in the manufacturing of nanostructures and nanomaterials (AM Platform and Feenstra, 2014 ). This implies a variety of chances and risks for the recycling system introduced by novel materials, as well as by new consumer-centric and bio-inspired designs. The emerging options offered by novel moulding technologies, data processing and bio-inspired designs provide an entirely new range of options for product designers when programming material functionality. The manufacturing of ‘lightweight designs’ that mimic natural lattice structures may promote the reduction of energy and material consumption to a significant extent, while simultaneously reducing the variety of materials ( Gebler et al., 2014 ; Sendel et al., 2015 ; Van Daal, 2015 ). In summary, digital prototyping to mimic natural lattice structures will be more widely adopted and is a promising way to reduce material variety and manufacture multi-functional materials in the future. Do trends such as degradability additives and in-situ biodegradability contribute to meeting the DfR criteria ‘reducing hazardous material and increasing recyclability’? The debates on degradability additives (e.g. for previously non-degradable polyolefins) show that proper EOL management cannot be tackled alone through actions such as changing polymer recipes ( Selke et al., 2015 ). In fact, new standards for biodegradation in landfills, bodies of water and marine environments will also spread into business practices and may improve the overall metabolic consistency of bioplastics ( Mortier, 2014 ; Vincotte, 2015 ). It can therefore be maintained that by-pass solutions for non-biodegradable plastics neither support the path change away from landfill-oriented waste management nor do they contribute to improving the overall recovery rates and recyclability of polymer waste. Furthermore, the analysis suggests that the standardization of bio-based polymers cannot be static. Instead, it has to be imbedded into a system of learning between designers and recyclers and has to be more closely oriented to the needs and process conditions of recycling plant operators. If this is not facilitated, the material recovery rates in material recycling will not increase; instead, the energetic recycling of bio-based polymers will be the more dominant EOL treatment path. Characterizing sustainability goals for a circular economy A classification system for assessing the recyclability performance of bio-based polymers was derived from analysing the most prevalent recycling paths and the qualitative constraints for recyclability. The qualitative classification system serves as a way to aggregate the assessment metrics along cascading chains for bio-based and fossil-based polymer products. A key factor in characterizing the recyclability of specific polymer products is the concept of metabolic consistency ( Huber, 2000 , 2003 ). It determines the best path for polymer materials in terms of natural material cycles and recycling paths of a circular economy during post-consumer recycling. Furthermore, the system boundaries for business-as-usual recycling chains and best practice recycling paths are set according to the definition of the material flow system; in other words, whether it can be regarded as a single-stage or multi-stage cascade ( Essel et al., 2014 ). Several qualitative and quantitative aspects (as suggested by Sirkin and ten Houten, 1994 ), such as the material’s inherent and intrinsic properties, its energy content and its purity, have to be taken into account when defining and aggregating the cascade use indicators. The conceptualized system of objectives (as depicted in Figure 3 ), derived from the proposed analysis, enables us to classify the material flows from industrial by-products and post-consumer recycling in terms of the treatment routes they take when returning to the production chains of new goods and services in the industrial and energy sectors. Moreover, it allows us to derive suitable indicators for monitoring the consistency of polymer waste flows within technical or natural environments. Figure 3. Sustainability rules for cascading use of bio-based polymers within a circular economy. EOL: end of life; PET: polyethylene terephthalate; PE: polyethylene; PP: polypropylene; PLA: polylactic acid; PHA: polyhydroxyalkanoat; MFI: melt flow index. The main material flows are currently induced through bio-based PET and bio-based PE ( European Bioplastics et al., 2015 ), which can be considered a technical nutrient and can be recycled in a multi-stage cascade depending on the product applications and the thresholds for extrusion cycles. This system is well described and often used for quantitative assessment studies and for the qualitative assessment of system design opportunities. In order to aggregate the proposed indicators, the calculation has to be performed throughout a series of material life cycles for the selected polymer products, as presented in Figure 4 . Figure 4. Procedure for calculating performance metrics along cascade use chains for polymers. The accounting procedure includes the following quantitative indicators for cumulating the performance along the paths for energy recovery and material recovery: the cumulated energy demand ( CED polym.recov. ) (in kWh kg −1 ) of recycled polymer required for each waste treatment and material recovery stage; the efficiency of material use ( ƞ mat. use ) (in kg kg −1 ) in kg of product recovered per kg of raw material input for individual material life cycles and cumulated along a series of life cycles; the substitution factor of virgin materials f subst. (in kg kg −1 ) for each stage where secondary raw materials are recovered in kg of virgin material substituted per kg of input of secondary raw material; the energy yields y energy recyc (in kWh kg −1 ) for each fraction obtained at EOL through energetic recycling; the net calorific value (in kWh kg −1 ) and the non-substitutable fractions (in kg kg −1 ) of the fossil-based additives that have to be subtracted. The second indicator category has to account for qualitative criteria, which allows us to explore the potentials for future increases in energy recovery yields and recyclate recovery. It also has to take into account future improvements in the quality of secondary raw materials that have fewer impurities and a lower deterioration of the polymer’s rheological properties. Once these aspects are identified and/or defined, the resulting classification system can rank, for example, the cost of energy and resource demands in the production and recycling phases, the effects of product utility in the use phases produced, which EOL scenario is the most common and which reference products are consistently comparable in this matrix. An overview is presented in Table 4 . Table 4. Qualitative classification with related quantitative metrics. Polymer types Performance in maximizing resource efficiency & minimizing deterioration over multiple use stages Energy demand in processing and recycling Compatibility in end-of-life recovery environments EOL scenarios Metabolic consistency at most common EOL paths Bio-PA, expanded PLA High benefits & low polymer deterioration at first EOL Medium energy demand (LCI) Easy collection of bulky waste and recycling via extrusion Material recycling in technical environment Bio-PU, bio-phenol resin, PLA-biofoam High benefits & high polymer deterioration at first EOL Medium energy demand (LCI) Thermal recovery in waste incinerators Treatment and inertization in technical environment PLA-blends e.g. with PMMA High benefits & medium polymer deterioration at first EOL Medium to high energy demand (LCI) Easy collection, but advanced chemical recycling necessary Partially recyclable in technical environment Starch blends Low benefits and high polymer deterioration at first EOL Low energy demand (LCI) Separable partly bio-degradable, partly useful in mechanical recycling Not given in natural environments, partly given in technical environments PLA foils and bags Low benefits and high polymer deterioration at first EOL Low energy demand (LCI) Separation and energetic recovery in waste incineration Inertization via incineration and under defined industrial composting conditions EOL: end-of-life; LCI: life cycle inventory; Bio-PU: bio-based polyurethane; PLA: polylactic acid; MFA: material flow accounting; PMMA: polymethylmethacrylate. In order to achieve a highly resource-efficient recycling of polymers throughout the technical cycles of the circular economy, it is necessary to obtain (a) the highest possible quality and yields in material recovery, (b) the highest possible substitution of virgin materials along a series of material life cycles and (c) the highest possible yields in energy recovery. In this regard, the proposed set of six indicators for quantifying the fractions recovered in energetic and material recycling proves to be suitable for aggregating the assessment metrics required for assessing the actual cascade use performance against future development potentials towards more recycling-friendly polymer products. In particular, the indicators also ensure in the assessment of life cycle metrics that recycling routes where energy recovery is obtainable are outranking EOL routes without energy recovery, such as landfilling or biodegradation in the environment."
} | 8,682 |
39853017 | PMC11768817 | pmc | 7,285 | {
"abstract": "The development of the non-ferrous metal industry is generating increasingly large quantities of wastewater containing heavy metals (e.g., Sb). The precipitation of heavy metals by microorganisms involves complex mechanisms that require further investigation to optimize bioremediation technologies. In this study, we employed a sulfate-reducing bacteria (SRB) strain Desulfovibrio desulfuricans CSU_dl to treat the antimony (Sb)-containing wastewater; the behavior of Sb and mechanisms underlying precipitation were investigated by characterizing the precipitates. The results showed that the abiotic factors constraining SRB bacterial growth greatly affect Sb forms and precipitation. For instance, Sb precipitation maximumly occurred at pH 6 and 7, or C:N ratio of 10:1 and 40:3 for Sb(III) and Sb(V), respectively, resulting in a maximum Sb removal rate of 94%. Interestingly, we found that substantial antimonate and antimonite were adsorbed on the SRB cell surface, indicating that cell surface is a critical reaction site of Sb transformation and precipitation. Sb was adsorbed to the cell surface by C-C and C=O groups, and was further precipitated by forming Sb 2 S 3 and Sb 2 S 5 or was coprecipitated with the P-containing group. Partial Sb(V) reduction was also observed on the SRB cell surface. These results provided a deep insight into the Sb bio-transformation and were an advancement with respect to understanding bioremediation of Sb-contaminated wastewater.",
"conclusion": "4. Conclusions Due to the massive mining and utilization of antimony-mining resources, the surrounding water and soils are seriously polluted. The sulfate-reducing bacteria (SRB) method has the advantages of high efficiency, low cost, and no secondary pollution, and is highly advantageous in the treatment of antimony-contaminated water and soil. In this study, we find that under anoxic conditions, the precipitation rate of Sb(III) wastewater treated by SRB reached more than 91.02% when the Sb(III) concentration was 30 mg/L, temperature was 30 °C, pH was 7.13, C/N ratio was 10:1, and SO 4 2− concentration was 1600 mg/L. When the Sb(V) concentration was 45 mg/L, the temperature was 35 °C, the pH was 7.24, the C/N ratio was 40:3, and the SO 4 2− concentration was 2000 mg/L, the precipitation rate of Sb(V) from the wastewater by SRB was more than 94%. The morphology and precipitation of antimony are largely influenced by abiotic factors that restrict the growth of SRB bacteria. However, these factors do not work exactly the same for Sb(III) and Sb(V). The cell surface is the main activate site for Sb adsorption and precipitation. Sb was adsorbed to the SRB cell surface by the C-C and C=O group; the P-containing group also played a role in adsorption and precipitation of SRB. The adsorbed Sb was then precipitated by forming the sulfide precipitation. In this process, S 2 2− and S n 2− and their surface functional groups acted as electron donors, reducing a part of Sb(V) to Sb(III) and combining with functional groups. Mineralization occurred on the surface of SRB cells, generating Sb 2 S 3 and Sb 2 S 5 solid precipitates, thus achieving the precipitation of Sb(III)/Sb(V) from wastewater by SRB. However, this study used artificially generated wastewater, which may be different to real conditions. A further study using real wastewater with complex pollutants is also needed to promote the SRB remediation technology.",
"introduction": "1. Introduction In addition to being used in a wide variety of industries, such as electronics and metallurgy, antimony (Sb) is also known to cause water contamination due to improper handling [ 1 ]. Antimony-induced water pollution can be both natural and anthropogenic. The former sources include the weathering of rocks and minerals containing antimony, while the latter sources involve industrial activities, mining, and the discharge of wastewater containing antimony compounds. The non-ferrous metal industry, in particular, accounts for a significant portion in generating a substantial amount of heavy metal-containing wastewater, including antimony (Sb). Some studies have found that nearly 100 years of mining activities have caused severe soil contamination and water pollution in mining areas and these contaminations may alter the stability and functioning of ecosystems [ 2 , 3 ]. The unmitigated release of antimony mine drainage and wastewater has resulted in a persistent escalation of antimony pollutants [ 4 ]. It is known to all that the presence of antimony in water poses significant risks to both humans and ecosystems. Studies have shown that antimony (Sb) is listed as one of the major pollutants by the US EPA and the EU EPA due to its potential and proven carcinogenicity, immunotoxicity, genotoxicity, and reproductive toxicity like arsenic, and its toxicity Sb(metal) > Sb(III) > Sb(V), and the soluble compounds of antimony are more toxic than the insoluble compounds [ 5 , 6 ]. When the solubility of antimony in water reaches 3.5 mg/L, it becomes toxic to algae, and when it reaches 12 mg/L, it becomes toxic to fish and shrimps. Long-term ingestion of antimony-contaminated water may cause gastrointestinal disorders, skin irritation, respiratory problems, and may even lead to carcinogenic effects in humans. Antimony and its compounds have been consistently identified as priority contaminants due to their toxicity and biohazardous nature [ 1 , 4 ]. Antimony exhibits toxicity not only to higher organisms such as humans, algae, fish, and shrimps but also to microorganisms. The toxicity of Sb to bacteria, including sulfate-reducing bacteria (SRB), can significantly affect their metabolic activity and heavy metal-precipitation efficiency. Studies have shown that microbial toxicity assays, such as dehydrogenase activity tests and other redox reaction-based methods, provide valuable insights into the impact of toxic substances on microbial communities [ 7 , 8 ]). These methods could help elucidate the potential inhibitory effects of Sb on Desulfovibrio species and optimize bioremediation strategies. The chemical speciation of antimony in water is highly dependent on the surrounding pH. Sb(III) undergoes significant changes in its chemical forms across a pH range of 1 to 12. At pH 1 to 4, the content of compounds in the form of positively charged Sb(OH) 2 + decreases, while the content of compounds in the form of electrically neutral Sb(OH) 3 increases. Between pH 5 and 9, compounds mainly exist in the form of electrically neutral Sb(OH) 3 . At pH 10 to 12, the content of electrically neutral Sb(OH) 3 decreases, and compounds in the form of negatively charged Sb(OH) 4 − become more prevalent [ 9 ]. In order to mitigate the negative impacts of antimony pollution in water, it is essential to seek effective treatment methods. Numerous approaches have been explored in the field of antimony-removal technology, including physical processes (e.g., precipitation, filtration), chemical methods (e.g., flocculation, precipitation), and advanced techniques utilizing adsorption with materials like activated carbon or other specialized materials. Additionally, the use of sulfate-reducing bacteria (SRB) for mediating metal sulfide precipitation is seen as a promising approach for antimony removal [ 10 , 11 ]. The SRB bacteria put into application contain many species such as Metallobacterium [ 12 ], Bacillus thiophilus , and Bacillus citriodora . Extensive research has been carried out on the use of sulfate-reducing bacteria (SRB) for heavy metal remediation, and significant achievements have been made [ 13 , 14 ]. However, there are relatively few studies using SRB to treat Sb-containing wastewater. Wang demonstrated for the first time that SRB can convert SO 4 2− to S 2− in mine drainage while converting Sb(V) to Sb(III), and verified that the ratio of Sb(V)/SO 4 2− was an important parameter affecting the efficiency of antimony removal [ 11 ]. As previously reported, the addition of Fe 2+ to the SRB system significantly increased the metabolic activity of SRB [ 15 ]. Complementarily, the addition of iron scrap and iron oxidizing bacteria (IOB) to the SRB system resulted in 99.98% Sb(V) removal for the Fe + IOB + SRB system [ 16 ]. Differences in carbon sources affected the efficiency of antimony removal in desulfurization processes, with the SRB system utilizing ethanol demonstrating a higher removal efficiency of 97.8% [ 17 ]. Temperature was also identified as a primary factor controlling microbial Sb reduction [ 17 ]. Characterization of the SRB antimony-removal system’s precipitate indicated that Sb(V) was reduced to Sb 2 S 3 [ 18 , 19 ]. Proteomic analysis showed that the extracellular protein functional groups of SRB were capable of adsorbing and immobilizing Sb(III), with a significant increase in extracellular proteins involved in electron transfer [ 20 ]. Despite many studies on Sb precipitation from wastewater by SRB, the effects of SRB on Sb(III)/Sb(V) precipitation in various habitats under anoxic conditions remain unclear, as do the magnitudes of the impacts of different environmental elements on precipitation efficiency. Furthermore, the precipitation mechanism of SRB on Sb(III)/Sb(V) under anoxic conditions requires further elucidation. In this work, we analyzed the effect of different conditions (temperature, pH, C/N ratio, and SO 4 2− concentration) on the precipitation of Sb(III)/Sb(V) by SRB, and used scanning electron microscopy–energy dispersive X-ray (SEM-EDS), and X-ray photoelectron spectroscopy (XPS) to study the morphology, components, and surface chemistry of SRB before and after the immobilization of Sb(III)/Sb(V) by adsorption and the precipitation mechanism of Sb(III)/Sb(V) from water under oxygen conditions. The aim of this work is to reveal the mechanism related to the fixation of Sb(III) and Sb(V) by SRB. The work provided theoretical and technological foundations for the application of SRB in the remediation of antimony-polluted wastewater.",
"discussion": "3. Results and Discussion 3.1. Morphology and Growth Characteristics of SRB The bacterial liquid was black in color, accompanied by hydrogen sulphide gas. Scanning electron microscope (SEM) observation showed that the SRB bacteria were in the shape of an arc ( Figure A1 ). During the SRB growth, the pH value fluctuated between 7.2 and 7.5, the Eh redox potential was maintained between −55 mV and −70 mV, and the number of bacteria reached the maximum on the 4th day, and then began to decrease on the 5th day due to the accumulation of the product H 2 S. After 7–9 days of incubation, the precipitation effect of SRB on SO 4 2− in the medium was higher than 91%, and a large desulphurization rate was achieved. It was shown that the sulphate-reduction rate of SRB reached a maximum in the pH range of 7.0–7.5, which justified the better desulphurization efficiency of the SRB bacteria used in this experiment [ 21 , 22 ]. 3.2. Sb(III)/Sb(V) Precipitation by SRB Under Different Conditions The pH, temperature, carbon to nitrogen ratio, and SO 4 2− concentration exert different impacts on the precipitation of Sb(III) and Sb(V) ( Table 1 ). pH : For Sb(III), sulphate-reducing bacteria started to precipitate Sb(III) at partial neutrality (pH 6–8). The precipitating efficiencies of Sb(III) at pH 6, 7, and 8 on the seventh day were 98.09%, 91.57%, and 89.89%, while it was 25.89% at the weakly acidic condition (pH 4). Similar findings were observed for Sb(V). At pH 4, Sb(V) precipitated at a rate of 62.06%; at pH 7, it reached a maximum fixation rate of 94.89%; at pH 8, it decreased slightly to 91.21%. The pH values of 6 and 7, respectively, were optimal for the precipitation of Sb(III) and Sb(V) by SRB. The result is similar to a previous study [ 23 ], which showed that SRB were less active under acidic conditions and had a higher solution redox potential during growth, which reduced the efficiency of desulphurization and antimony precipitation. The inhibitory effects of low pH on the desulphurization and antimony precipitation by SRB are varying and complicated. The effect of pH on SRB can be explained in two ways. Firstly, pH has a direct impact on SRB metabolism, disrupting its cellular homeostasis, destroying the pH gradient from the inside and outside of the cell, and ultimately causing energy loss. More protons diffuse through the cell membrane at lower pH levels than at neutral pH due to diffusion pressure across the membrane. To cope with low pH stress, SRB employ both active and passive mechanisms to maintain pH homeostasis. Active mechanisms include proton pumps, which actively expel excess H⁺ ions from the cell using ATP, and amino acid decarboxylases, which produce neutralizing amines through decarboxylation reactions. Passive mechanisms involve changes in membrane lipid composition to reduce proton permeability and the expression of positively charged surface proteins that buffer external H⁺ ions. These strategies help stabilize intracellular pH but divert energy from growth and metabolism, thereby impacting SRB performance in acidic environments. More energy from redox processes was needed to maintain pH homeostasis if the pH gradient in the extracellular and intracellular environments was too great [ 24 ]. Second, the pH changed the forms of some elements found in the environment, including organic acids, heavy metals, and sulfides. The forms of sulphur metabolites (H 2 S, HS − and S 2− ) present in the solution of sulphate-reducing bacteria are dependent on the pH of the solution, and in acidic solutions (pH 4.5–5.27), the main form of sulfide substances present was molecular H 2 S (~99%), which tended to volatilize easily [ 25 ], and therefore desulphurization for Sb(III) precipitation at a pH of 4.57 was less efficient. Under the condition of a neutral environment, the aqueous solution had a high concentration of sulfide, which increased the probability of Sb(III) binding with sulfide [ 26 ]. The SRB metabolism was stronger, so the neutral environment was more conducive to the desulfurization and antimony precipitation from wastewater by SRB. As shown in Figure 1 a, as the initial pH increased, so did the removal of SO4 2− and Sb(V). This might be the result of a higher pH, a drop in H + concentration, less competition, and a larger amount of heavy metal ion adsorption on the active sites, which would increase the removal of heavy metal ions [ 27 ]. Temperature : Figure 1 b shows the effect of different temperature conditions on antimony precipitation. On the 7th day, the precipitation efficiency of SRB for both Sb(III) and Sb(V) exceeded 90% at 25–35 °C. The precipitation efficiency of SRB for Sb(III) was 91.56%, 88.97%, and 94.60%, respectively. The precipitation efficiencies were 91.56%, 88.97%, and 94.60% for Sb(III) and 91.56%, 94.89%, and 97.53% for Sb(V). The combined findings demonstrated that 35 °C was the ideal temperature for both Sb(III) and Sb(V) precipitation by SRB in this investigation. A previous study concluded that SRB reduced thiosulfate at a growth temperature of 40–50 °C. However, when the growth temperature reached 30°C, virtually no H 2 S was produced [ 28 ]. And the results showed that the ideal growth temperature range for SRB was 25 °C to 35 °C, which was consistent with this study. There are two main ways that temperature affects Sb precipitation by SRB. On the one hand, temperature is an important influencing factor for the growth of SRB. Temperature modifies the microorganism’s own enzyme activity, which in turn affects the efficiency of Sb(III) fixation. Certain enzymes in these cells will be inhibited by too high or too low a temperature, which will negatively impact cell growth and product synthesis, as well as alter cell morphology, metabolic function, and microbial toxicity, or even cause cell death [ 29 ]. Additionally, low temperature caused the lipids in the cell membrane to wax and the activity of membrane proteins to decrease, which limited the cell membrane’s ability to transport electron donors and electron acceptors. This was the primary mechanism by which low temperature affected SRB metabolism [ 30 ]. On the other hand, the solubility of H 2 S in wastewater was influenced by temperature; high temperatures make H 2 S less soluble in wastewater, which lessens H 2 S’s inhibitory effect on SRB [ 31 ]. C/N ratio : The C/N ratio represents the relative availability of carbon and nitrogen sources in the medium, which is a critical factor affecting bacterial growth and metabolism. Different C/N ratios in the substrate have certain effects on microbial growth. Reasonable adjustment of the C/N ratio in the substrate was one of the efficient measures to accelerate microbial growth and promote microbial action. The C/N ratio of Sb(III) and Sb(V) were shown in Figure 1 c. Both high (20:1) and low (20:3) C/N ratio affected the growth of SRBs and their efficiency of desulphurization and antimony precipitation, and the immobilization rates of Sb(III) were 90.26%, 91.57%, and 86.66% for C/N ratios of 40:3, 10:1, and 8:1, respectively. When the C/N ratio was 40:3, the Sb(V) precipitation rate reached the highest value of 95.28%, which was the optimal C/N ratio. At a low C/N ratio (e.g., 20:3), Sb(V)-removal efficiency decreased significantly due to limitations in microbial growth and metabolism. Microbial stoichiometry and metabolic theory state that when a substrate’s low C/N ratio (or high N availability) satisfies microbial requirements or when the compounds are easily broken down by the microorganisms, increasing the production of enzymes and the breakdown of organic carbon, microbial activity is higher [ 32 ]. Therefore, the selection of an appropriate carbon-to-nitrogen ratio in microbial growth media can alleviate the metabolic constraints of microorganisms in heavy metal environments and enhance the efficiency of heavy metal fixation. Research has demonstrated that the growth and metabolism of SRB were restricted by a low C/N ratio [ 33 ]. In contrast, a high C/N ratio, insufficient nitrogen, low buffering capacity of the digestive solution, and easy decrease in pH caused the effectiveness of SRB’s removal of Sb to decline. The selection of an appropriate carbon-to-nitrogen ratio is critical to alleviating metabolic constraints in microorganisms under heavy metal stress. While a low C/N ratio can inhibit Sb(V) removal by limiting SRB activity, excessively high C/N ratios can lead to nitrogen insufficiency, low buffering capacity, and pH reductions, all of which also reduce Sb(V) removal efficiency. SO 4 2− concentration : Increase in SO 4 2− concentration had little impact on the Sb(V) and Sb(III) immobilization efficiency of SRB; moreover, the Sb(V) immobilization efficiencies were higher than 94% in the range of sulphate concentration of 800–2000 mg/L. The results indicated that the immobilization efficiency of Sb(V) in the range of 800–2000 mg/L was higher than 94%. Meanwhile, the precipitation rate of Sb(III) under each SO 4 2− concentration condition was above 91.02%. The SO 4 2− concentration in the solution directly reflects the equilibrium relationship between the substrate and microorganisms, and is also the main indicator of the system’s ability to reduce sulfate ions. The production of hydrogen sulfide through heterogeneous sulfate reduction, or sulfide generation, is the fundamental metabolic characteristic of SRB. Sulfate and sulfite can be the electron acceptors for SRB during heterogeneous sulfate reduction [ 34 ]. When sulfate levels are sufficient, sulfate serves as the main electron acceptor for SRB metabolism; in situations where sulfate levels are insufficient, sulfite serves as the primary electron acceptor. It was demonstrated that SRBs can grow normally despite very low sulfate concentrations, and that microorganisms in the environment can propel this low-sulfate lake’s sulfur recovery to high levels [ 35 ]. In this study, the SRB strain could grow well under the Sb(III) concentration of 30 mg/L. Under anaerobic conditions, when the Sb(III) concentration is 30 mg/L, the temperature is 30 °C, the pH value is 7.13, the C/N ratio is 10:1, and the SO 4 2− concentration is 1600 mg/L; the removal rate of Sb(III) by SRB could reach up to 91.02%. For Sb(V), the SRB strain could grow well under 60 mg/L of Sb(V). A lower toxicity of Sb(V) compared to Sb(III) would lead to the tolerance concentration of Sb(V) being higher than that of Sb(III). Under anaerobic conditions, when the Sb(V) concentration is 45 mg/L, the temperature is 35 °C, the pH value is 7.24, the C/N ratio is 40:3, the SO 4 2− concentration is 2000 mg/L, and the removal rate of Sb(V) by the SRB reaches 94%. Even though the precipitation rate was a bit lower than some previous studies which reached up to 98.7% removal rate [ 10 ], the initial Sb concentration was much higher in our study than other studies (45 mg/L vs. 20 mg/L). Much more Sb was removed in our study than previous studies. The results indicate that the Desulfovibrio desulfuricans CSU_dl strain had excellent Sb tolerance, as well as bioremediation potentials. 3.3. Characterization of Precipitates Generated by SRB The morphology and chemical composition of Sb(III)/Sb(V) precipitates on the SRB cell surface were investigated using SEM-EDS ( Figure 2 ). The SRB cells exhibited a smooth surface with significant pore cavities and thin pore walls. Chemical composition analysis of the SRB cell surface indicated that C, P, O, S, and N were the primary elements. After immobilization, Sb(III) adsorption on the SRB cell surface resulted in morphological changes, including a rougher surface, blocked pore cavities, thicker pore walls, and the formation of ellipsoidal insoluble material. The corresponding surface chemical analysis revealed that C, S, P, Sb, O, and N were the main elements in the reaction products. These observations suggest that Sb(III) precipitation occurred on the SRB cell surface, producing antimony trisulphide (Sb 2 S 3 ) as the primary product. The mechanism driving Sb(III) precipitation is hypothesized to involve mineralization mediated by cell wall surface materials, as represented by the following reaction (Equation (1)): 3S 2− + 2Sb(OH) 3 + 6H + → Sb 2 S 3 (s) + 6H 2 O (1) The mechanism of Sb(V) precipitation by SRB organisms (Equations (1) and (2)) is inferred from the above results, where part of Sb(V) were reduced to Sb(III), and mineralization occurred under the action of cell wall surface substances to produce antimony trisulphide and antimony pentasulphide, resulting in the precipitation of Sb(V) from the wastewater by SRB.\n 5S 2− + 2Sb(OH) 6 − + 12H + → Sb 2 S 5 (s) + 12H 2 O (2) Cell morphology analysis by SEM indicated that the precipitation of Sb occurred mainly on the cell surface. In particular, abundant P elements were observed on the cell surface as indicated by EDS analysis. This could be due to P being the essential elements for the cell membrane. P could co-precipitate with heavy metals, which further promoted the precipitation of Sb. The results were further supported by the XPS and FTIR analysis. 3.4. Analysis of Surface Functional Groups Involved in the Adsorption and Immobilization of Sb(III)/Sb(V) by SRB The chemical composition, elemental valence, and binding forms of SRB before and after immobilization of Sb(III)/Sb(V) by adsorption were investigated by XPS analysis. As shown in Figure 3 , the SRB adsorption-immobilized Sb(III)/Sb(V) products clearly identified the peaks of P 2p, S 2p, C 1s, N 1s, O 1s, and Sb 3d 5/2 in the scanning spectra, with the peaks of Sb 3d 5/2 and O 1s being very close to each other at the binding energy of 528 eV–530.3 eV. The characteristic peak of Sb 3d 3/2 was also identified in the product of SRB adsorption of immobilized Sb(III). The S 2p peak corresponds to a binding energy of 164.3 eV, and its increase indicated the production of sulphides in the product. This suggests that sulphate-reducing bacteria react with Sb(III) or Sb(V) in solution, allowing antimony to be adsorbed onto the product surface, which remains consistent with our previous predictions. On the basis of the shape of the XPS absorption peak in the N 1s orbital fractionation spectrum ( Figure 3 b), the binding energy of the N 1s peak is 399.08 eV in the SRB treatment, 399.10 eV in the SRB-Sb(III) treatment, and 399.38 eV in the SRB-Sb(V) treatment, suggesting that the binding energy of the N 1s peak in SRB to Sb(V) biosorption immobilization is shifted to the right, indicating a possible loss of electrons. This may be due to the metal ion chelating with the N atom in the amino and imine groups and the pair of lone electrons in the N atom being shared with the metal ion, resulting in a lower electron cloud density and a higher binding energy peak for the N atom [ 36 ]. Three peaks originally made up the C 1s orbital fractionation of the XPS absorption peaks ( Figure 4 a), which corresponded to C-C (284.08 eV), C-O (285.58 eV), and C=O (287.28 eV). The amount of C-O increases and the amount of C-C and C=O decreases following the adsorption immobilization of Sb(III) and Sb(V), as can be observed, suggesting the participation of carbonaceous groups in the Sb adsorption immobilization process. The P 2p orbital spectra of the XPS absorption peaks ( Figure 4 b) initially consisted of two peaks corresponding to O=P(OR) 3 (133.18 eV) and P-C (132.28 eV), respectively. It can be seen that after the adsorption and immobilization of Sb(III)/Sb(V), the O=P(OR) 3 content decreased and the P-C content increased, and the P-C content exceeded the O=P(OR) 3 content, indicating that there is a relevant chemical reaction occurring between Sb(III)/Sb(V) and phosphorus-containing groups, and that phosphorus-containing functional groups play an important role in the adsorption and immobilization of Sb, which is consistent with the results of previous research [ 37 ]. The S 2p channel splitting spectrum of the XPS absorption peaks ( Figure 4 c) initially consisted of four peaks corresponding to S 2− (160.68 eV), S 2 2− (162.38 eV), S n 2− (163.98 eV), and SO 4 2− (167.88 eV). Significant increases in S 2− and decreases in S 2 2− and S n 2− were seen in Sb(III) and Sb(V); after Sb(III) was immobilized, SO 4 2− increased, and after Sb(V) was immobilized, SO 4 2− dropped. Sb 3d 3/2 , which does not overlap with the O 1s, was employed as a guide for the fitting of the Sb 3d 5/2 peak because the Sb 3d 5/2 peak overlaps with the O 1s. Therefore, the spin-orbit splitting and ratio of the Sb 3d 3/2 peak dictate the strength and binding energy of the Sb 3d 5/2 peak. The two characteristic peaks in Figure 4 d for Sb(III) and Sb(V) at the binding site during Sb(V) adsorption show that Sb(V) was partially decreased during adsorption. The reduction reaction, however, is thought to occur during the adsorption of Sb(V) by SRB, with S 2 2− and S n 2− in solution and their surface functional groups acting as electron donors, reducing Sb(V) to Sb(III) and immobilizing it by binding to functional groups to form mineralized substances. This is supported by the presence of the Sb(III) peak at 530.88 eV."
} | 6,816 |
31619728 | PMC6795859 | pmc | 7,292 | {
"abstract": "The positive effects of arbuscular mycorrhizal fungi (AMF) have been demonstrated for plant biomass, and zinc (Zn) and phosphorus (P) uptake, under soil nutrient deficiency. Additionally, a number of Zn and P transporter genes are affected by mycorrhizal colonisation or implicated in the mycorrhizal pathway of uptake. However, a comprehensive study of plant physiology and gene expression simultaneously, remains to be undertaken. Medicago truncatula was grown at different soil P and Zn availabilities, with or without inoculation of Rhizophagus irregularis . Measures of biomass, shoot elemental concentrations, mycorrhizal colonisation, and expression of Zn transporter ( ZIP ) and phosphate transporter ( PT ) genes in the roots, were taken. Mycorrhizal plants had a greater tolerance of both P and Zn soil deficiency; there was also evidence of AMF protecting plants against excessive Zn accumulation at high soil Zn. The expression of all PT genes was interactive with both P availability and mycorrhizal colonisation. MtZIP5 expression was induced both by AMF and soil Zn deficiency, while MtZIP2 was down-regulated in mycorrhizal plants, and up-regulated with increasing soil Zn concentration. These findings provide the first comprehensive physiological and molecular picture of plant-mycorrhizal fungal symbiosis with regard to soil P and Zn availability. Mycorrhizal fungi conferred tolerance to soil Zn and P deficiency and this could be linked to the induction of the ZIP transporter gene MtZIP5 , and the PT gene MtPT4 .",
"conclusion": "Conclusions and future research This experiment presents, to our knowledge, the first attempt to link physiological and molecular markers of mycorrhizal associations pertaining to both Zn and P nutrition, in Medicago. The expression of MtZIP5 was induced both by mycorrhizal colonisation and low soil Zn availability. In contrast, MtZIP2 expression was up-regulated in non-mycorrhizal roots, and increased with soil Zn availability. In examining shoot biomass and Zn concentration, there was evidence of a ‘protective’ role of mycorrhizal fungi at high levels of soil Zn. Regarding PTs, MtPT4 and MtPT8 were up-regulated, and MtPT1 and MtMT4 down-regulated in mycorrhizal plants; the expression of all PTs was interactive with available soil P. The expression of both MPU- and DPU-related PTs likely conferred greater P uptake in the mycorrhizal plants when soil P was limiting. Further studies are necessary to understand the potential of mycorrhizal fungi and the role of ZIP transporters to improve the Zn nutrient uptake via the MPU.",
"introduction": "Introduction Zinc (Zn) is important both for agricultural production and for human development. A deficiency of Zn can seriously affect plant and human development because Zn is a regulatory co-factor and structural constituent in proteins and enzymes involved in many biochemical pathways 1 , 2 . Soil Zn deficiency affects millions of hectares of cropland worldwide, and is particularly prevalent in developing countries, and in the cereal-growing regions of Australia 2 , 3 . In addition, approximately one third of the global population suffers from an inadequate dietary intake of Zn 4 – 6 . Therefore, increasing the level of Zn uptake by crops - known as biofortification - is a subject of considerable international interest 7 – 9 . Phosphorus (P) is one of the most important macronutrients for plant growth. Soil P deficiency leads to reduced plant nutrient uptake, prolonged maturity stage 10 , 11 , and affects enzyme activity and many signal transduction cascades 12 . However, the primary source of P fertilizer (phosphate rock) for plant growth is a finite resource, and thus is becoming depleted over time because of demand from the agricultural sector 13 , 14 . To effectively manage the available levels of soil Zn and P, an advanced understanding of plant root uptake capacity of nutrients is required, in particular, of plant membrane transporters related to P and Zn transport 15 . Additionally, it is important to understand factors that affect P and Zn transport gene regulation, such as associations with arbuscular mycorrhizal fungi (AMF), which can enhance plant nutrient uptake 16 . AMF form associations with more than 80% of flowering plant species, and form part of the function of plant root systems 17 , 18 . AMF colonise the root cortex, and can extend their hyphal network into the surrounding soil environment 19 , 20 . These external hyphae contribute to plant uptake of P and Zn, as well as other mineral nutrient including iron (Fe), calcium (Ca), and copper (Cu) 18 . Previous research has demonstrated that AMF can enhance plant Zn uptake, sometimes leading to a boost in plant growth and Zn concentration in plant tissue 21 , 22 . Furthermore, P concentration in the shoots and roots of mycorrhizal plants can be significantly higher than in non-mycorrhizal plants grown under soil P deficiency 23 – 26 . Additionally, AMF can reduce plant heavy metal uptake (Zn, copper (Cu), lead (Pb), arsenic (As)) in contaminated soils, thereby protecting plants from toxic effects 27 , 28 . Studies on red clover and tomato, respectively, have established that Zn content uptake in shoots and roots decreased substantially under high soil Zn concentration when plants were colonised by AMF 29 , 30 . There are two pathways for plant uptake of nutrients from soil: via root epidermal cells (direct pathway; DPU), and via associations with arbuscular mycorrhizal fungi (mycorrhizal pathway; MPU). In plants, Zn is taken up from the rhizosphere as Zn 2+ via ZIP (Zrt, Irt-like Protein) membrane transporters 31 . Some ZIP transporters have the potential to also transport Fe 2+ and Mn 2+ 31 , 32 . Several studies have focused on characterising the ZIP transporter family in different plant species, including barley 32 – 35 , rice 36 , potato, and Arabidopsis thaliana 31 . In Medicago truncatula (Medicago), four ZIP transporters - MtZIP1, MtZIP2, MtZIP5 and MtZIP6 - are able to transport Zn 2+ , as confirmed by yeast complementation assays, have been identified 37 , 38 . As such, Medicago provide a model plant species for studies of mycorrhizal impacts on plant Zn (and P, see below) nutrition. Additionally, MtZIP2 has been localised to the plasma membrane in onion epidermal cells 39 . Further work is required to discover whether any of these ZIP transporters are implicated in the mycorrhizal transport of Zn into plants (via the MPU). In terms of plant P uptake, the phosphate transporters (PTs) MtPT1, MtPT2, and MtPT3 are involved in the DPU for P uptake, and are closely related to low-affinity P transporters, belonging to the Pht1 family 40 , 41 . The genes encoding these PTs are generally highly expressed when plants are growing under low soil P conditions. However, in roots colonised by AMF, MtPT1 and MtPT2 are down-regulated significantly as the symbiosis develops 26 , 42 , 43 . Furthermore, one of the PTs in M . truncatula (MtPT4) has been demonstrated to be a mycorrhiza-induced phosphate transporter 44 . MtPT4 is expressed exclusively in mycorrhizal roots, and specifically, in cells containing arbuscules 44 . Loss of MtPT4 function in plants leads to impairment of the mycorrhizal symbiosis, because it causes mature arbuscules to degenerate, resulting in premature arbuscular death 45 , 46 . In addition, MtPT8 from the Pht1 family has also been identified to be induced upon formation of the mycorrhizal symbiosis, and contribute to the uptake of P ions released by the fungal membranes 47 – 49 . A phosphate starvation-induced (PSI) gene in M . truncatula , MT4 26 , 50 , has been shown to be involved in P accumulation in plants, and is down-regulated in response to both P fertilisation and to mycorrhizal colonization 51 . More recently, another gene ( LysoPhosphatidylCholine AcylTransferase 1; AtLPCAT1 ) has been demonstrated to be involved in the accumulation of P in the shoot under soil Zn deficiency in A . thaliana (a non-mycorrhizal species) 52 . Therefore, these genes are also of interest with regards to plant responses to P and Zn deficiency in the present study. In M . truncatula , there is evidence that mycorrhizal inoculation affects plant P and Zn uptake at different soil P concentrations, and at a range of soil Zn concentrations, ranging from deficient to toxic 30 , 53 , 54 . However, there presently exists a lack of research focusing on the three-way interaction between soil P and Zn availabilities and AMF inoculation, that also addresses potential underlying molecular mechanisms. Therefore, the current study aimed to link the mycorrhizal effects under variable P and Zn soil conditions, on whole plant physiology, with expression of molecular markers for Zn and P uptake. Specifically, there were several hypotheses related to the experiment: That the effects of soil Zn and P availability were interactive with AMF function, and That the effects of AMF on plant nutrition and gene expression could be linked: In relation to the dual role of AMF at low and high soil Zn availability, and In relation to the role of AMF in improving P uptake at low soil P availability",
"discussion": "Discussion Mycorrhizal plants are more tolerant of soil P and Zn deficiencies The positive effects of forming mycorrhizal associations on plant growth are considered one of the main benefits of the symbiosis, and these effects have been studied extensively in various plant species 57 – 62 . In this study, the benefits of mycorrhizal fungal inoculation in terms of increased Medicago biomass were the greatest under low soil P concentration, which confirms earlier studies in this and other plant species 13 , 30 , 63 . In the two soil P addition treatments (P20, P50), there were smaller, albeit still positive, plant biomass responses to mycorrhizal inoculation. These smaller responses may be due to the sufficient levels of P in soil, which allowed plants to increase their biomass and nutrient uptake without the aid of mycorrhizal fungi. In an earlier study, it was demonstrated using radioisotope tracing that even where the biomass of mycorrhizal flax ( Linum usitatissium ) plants were not larger than mock-inoculated plants, the MPU was still active and transporting substantial amounts of P to the plant 64 . Thus, the growth response of a plant to mycorrhizal fungi, whether it is positive, neutral or negative, cannot be used to estimate the activity of the mycorrhizal pathway of P uptake. In the present study, the reduction in growth response was in line with the decreased percentage of mycorrhizal colonisation in the roots, when P was added to the soil, as demonstrated in previous studies in both the same plant and fungi species 26 and in other species 14 , 65 – 68 . Consequently, the reduction in mycorrhizal colonisation and MGR with soil P addition points to a reduced contribution of P by the MPU; however, the actual activity of the MPU would need to be confirmed by radioisotope labelling. With regards to soil Zn, colonisation by mycorrhizal fungi resulted in increased Medicago biomass under both soil Zn deficiency (Zn0) and soil Zn toxicity (Zn20); this result highlights the dual role of mycorrhizal fungi in ameliorating Zn stress for the host plant 53 , 69 , 70 . While the mycorrhizal plants thrived in soil with high levels of Zn addition (Zn20), the non-mycorrhizal plants experienced reduced growth and visual symptoms of Zn toxicity in the leaves, as in earlier study 53 . Additionally, at high soil P supply (P50), the non-mycorrhizal M . truncatula plants were not as affected by the high soil Zn concentration, and this ‘alleviation’ of Zn toxicity at high P availability is possibly a result of dilution of Zn in the plant tissues due to the greater P uptake and, thus, biomass of the plants 63 , 71 – 74 . Medicago plants are more tolerant of Zn stress as deficiency or toxicity when colonised by AMF In this study, there was evidence for the ‘dual roles’ of mycorrhizal fungi at low and at high soil Zn availabilities. When Zn was limiting to the plant, the mycorrhizal plants generally had higher shoot Zn contents; colonisation by R . irregularis also conferred a benefit to plants growing in toxic soil Zn conditions (Zn20), as indicated by the reduced shoot Zn concentration in the mycorrhizal plants at high soil P supply. Along with the increased biomass at Zn20 in the mycorrhizal plants at low soil P availability, these results support the hypothesis that mycorrhizal inoculation provides a ‘protective’ effect against high soil Zn availability. These results are in accordance with earlier studies exploring the benefits of mycorrhizal fungi in other plant species (red clover and white clover) in Zn-contaminated soil 29 , 75 . Additionally, other studies have indicated that under soil heavy metal contamination with high levels of plant-available soil P, mycorrhizal plants can increase their biomass, thus resulting in reduced uptake of heavy metals such as Zn, copper (Cu), and lead (Pb) into their tissues 28 , 70 , 72 , 76 . Expression of ZIP genes may be linked to mycorrhizal Zn tolerance mechanisms The ZIP membrane transporter gene family can transport Zn 2+ from the rhizosphere into plants via the roots 31 . There are four characterised ZIP transporters (MtZIP1, MtZIP2, MtZIP5 and MtZIP6) that are able to transport Zn 2+ in M . truncatula , as confirmed by yeast complementation assays 37 , 38 . Inoculation with AMF induced the expression of MtZIP5 in Medicago roots when grown in Zn-deficient soil; this result has not to our knowledge been reported before. Given that the Medicago plants were growth-limited by Zn availability at Zn0, the induction of MtZIP5 in mycorrhizal roots at all Zn0 treatments may have helped to overcome the Zn limitation somewhat, as supported by the greater biomass and Zn concentration. Therefore, as a Zn transporter, MtZIP5 may have a direct or indirect association with the MPU of Zn uptake, and this novel result is deserving of further study. The expression of MtZIP2 was down-regulated in the roots when inoculated with AMF, which is in line with an earlier study 39 . These results together suggest that MtZIP2 may not be directly involved in Zn uptake via the MPU. Similarly, the expression of the MtZIP6 gene was also found to be down-regulated by AMF. By contrast, in a previous study, MtZIP6 was highly up-regulated using the same plant and AMF species 53 ; this difference may be due to differences in Zn and/or P availability in the two studies, or the different soil type used. The contrasting MtZIP6 results with previous work, highlight that the role of the MtZIP6 protein may not be one directly involved in the MPU. While the present study has shown that the expression of a number of ZIP genes are modified by AMF inoculation, further studies are required that use knock-out mutant plants in conjunction with radioisotope tracing to elicit the function of ZIP transporters, and their potential role in Zn uptake via the mycorrhizal pathway. Plant P nutrition was improved under P-deficiency and buffered under varying Zn-stress in mycorrhizal plants The results of this study clearly confirm that mycorrhizal colonisation benefits Medicago plant tissue P concentration and contents at low soil P addition 13 , 54 , 71 , 77 . In addition, shoot P contents in mycorrhizal plants were maintained across the range of soil Zn availabilities (from deficient to toxic) at high soil P treatments (P20, P50). By contrast, the shoot P content in non-mycorrhizal plants was negatively influenced by soil Zn deficiency at high soil P concentration treatments, and is in agreement with a previous Zn-AMF interaction study in Medicago, up to 20 mg Zn kg −1 53 . However, once soil Zn reached a highly toxic concentration (40 mg Zn kg −1 ) in Watts-Williams, et al . 53 , mycorrhizal colonisation no longer had the capability to maintain plant shoot P contents. This ‘buffering’ capability of mycorrhizal plants at varying soil Zn concentrations when P is not deficient could be considered a benefit of AMF beyond simply increased plant P concentration and/or contents at P-deficiency. PT gene expression patterns are highly influenced by mycorrhizal colonisation This study investigated the different roles of the phosphate (Pi) transport-related genes in M . truncatula plants and the effect of soil P availability and mycorrhizal inoculation. When plants are colonised by AMF, the expression of a number of membrane PT genes is modified. Specifically, two P transporters (MtPT4 and MtPT8) have been localised to the peri-arbuscular membrane (PAM), the site of plant-fungus nutrient exchange, and are directly implicated in the transport of P from fungus to the host plant 44 , 48 . In this study, MtPT4 was exclusively expressed in mycorrhizal plants and was down-regulated with increasing soil P availability, with no influence of soil Zn availability. The lack of effect of Zn availability on MtPT4 expression is in contrast with a previous study that reported that an increased expression of MtPT4 corresponded with an increase in soil Zn concentration 53 . The present study suggests that MtPT4 expression interacts with soil Zn availability under certain conditions only. Furthermore, in this study, the expression of MtPT4 was highly correlated with the expression of R . irregularis \n α-tubulin (a mycorrhizal fungal biomass marker gene), as demonstrated in earlier studies 53 , 78 . Therefore, the induced expression of MtPT4 at low soil P concentration may be due to the increased requirement for the MPU, and thus transport of P across the PAM, which in turn contributed to the higher plant P concentrations and contents in the mycorrhizal plants. Conversely, MtPT8 , a second AMF-induced Pi transporter gene 47 , 48 , was highly expressed in the P20 treatment, where MtPT4 had reduced expression. This interplay between the expression of MtPT4 and MtPT8 has been discussed in previous studies 48 , 49 in which MtPT4 was mutated, and this led to the compensation by MtPT8 expression, presumably to balance the P homeostasis in plants colonised by AMF. This compensation by MtPT8 in previous studies may also explain why MtPT4 and MtPT8 expression was dominant at different soil P availabilities in this study. The expression of PT gene MtPT1 is a representative of the direct pathway of P uptake (DPU), and here was down-regulated in the mycorrhizal plants across different soil P and Zn availabilities, which is consistent with previous studies 26 , 42 . The expression of MtPT1 in non-mycorrhizal plants was particularly up-regulated when soil P was the most limiting to plants. This highlights again the important role that MtPT1 and other PTs may have in P uptake via the DPU in non-mycorrhizal plants, as they are relying on a single pathway of P uptake from the soil. Furthermore, the P starvation-induced gene MT4 50 was also down-regulated in the mycorrhizal plants at low soil P availability; at any higher levels of soil P addition the expression of MT4 was similar between mycorrhizal and non-mycorrhizal plants, presumably because the plant was not P-starved 26 , 51 , 79 . In summary, whereas MtPT4 and MtPT8 expression interacted with soil P availability in the mycorrhizal plants, MtPT1 and MtMT4 were likely important for P uptake and regulation in the non-mycorrhizal plants at low soil P availability. Soil P and Zn availabilities strongly influence responses to mycorrhiza and gene expression This study demonstrated that, aside from inoculation with mycorrhizal fungi, soil Zn and P availability also have a powerful impact on plant physiology and gene expression. For example, in addition to the down-regulation effects of AMF on the expression of MtZIP2 , this gene was also up-regulated by increased soil Zn addition. This finding has also been documented in Burleigh, et al . 39 , in which the authors observed that MtZIP2 was up-regulated within roots by high Zn fertilisation, and was highest in roots exposed to a toxic level of soil Zn. Thus, it is possible that MtZIP2 is either expressed directly in response to high concentrations of soil Zn, or is stimulated by internal plant Zn concentrations. Given that both shoot Zn concentrations and the expression of MtZIP2 were generally lower in the mycorrhizal plants, the results presented here suggest that the second hypothesis is more plausible. By contrast, expression of MtZIP5 in roots appeared to be soil Zn deficiency-induced, which has not been shown previously; this result suggests that a threshold in soil or plant Zn concentration exists between the Zn0 and Zn5 treatments, and thus that ZIP5 expression was reduced to a baseline level when the plant was not considered Zn-limited. A similar trend was also identified in barley whereby the expression of six HvZIP genes was increased by at least three-fold in Zn-deficient roots compared to the expression in Zn-sufficient plant roots 34 . In a previous study, researchers observed that loss-of-function of LPCAT1 in Arabidopsis thaliana at soil Zn-deficient conditions lead to increased shoot P accumulation 52 . Therefore, this recent study hypothesised that the expression of the nearest orthologue of LPCAT1 in Medicago may be highly interactive with soil P and Zn concentration. However, the expression of this gene was not affected by the interaction between soil P and Zn availability, although expression was higher in mycorrhizal plants than in non-mycorrhizal plants across all soil Zn addition treatments. Therefore, the role of this gene in Medicago, and other species with the ability to form arbuscular mycorrhizal associations remains unclear, but we have in this study uncovered a potential interaction between the gene and mycorrhizal inoculation. Conclusions and future research This experiment presents, to our knowledge, the first attempt to link physiological and molecular markers of mycorrhizal associations pertaining to both Zn and P nutrition, in Medicago. The expression of MtZIP5 was induced both by mycorrhizal colonisation and low soil Zn availability. In contrast, MtZIP2 expression was up-regulated in non-mycorrhizal roots, and increased with soil Zn availability. In examining shoot biomass and Zn concentration, there was evidence of a ‘protective’ role of mycorrhizal fungi at high levels of soil Zn. Regarding PTs, MtPT4 and MtPT8 were up-regulated, and MtPT1 and MtMT4 down-regulated in mycorrhizal plants; the expression of all PTs was interactive with available soil P. The expression of both MPU- and DPU-related PTs likely conferred greater P uptake in the mycorrhizal plants when soil P was limiting. Further studies are necessary to understand the potential of mycorrhizal fungi and the role of ZIP transporters to improve the Zn nutrient uptake via the MPU."
} | 5,741 |
25707972 | null | s2 | 7,295 | {
"abstract": "No abstract available"
} | 5 |
35519064 | PMC9056714 | pmc | 7,298 | {
"abstract": "Light-weight, mechanically flexible, transparent thermoelectric devices are promising as portable, and easy-to-integrate energy sources. Poly(3,4-ethylenedioxythiophene) nanowires (PEDOT NWs) possessing high electrical conductivity were synthesized by a facile self-assembled micellar soft-template method. And then, Te nanowires (Te NWs) with high Seebeck coefficient were easily synthesized by the solution process and then added as an inorganic filler to form the PEDOT NW/Te NW nanocomposite films via a simple and convenient vacuum filtration method. The thermoelectric (TE) properties of the nanocomposites were characterized in this research. A maximum power factor of 58.03 μW m −1 K −2 is obtained from the film containing 90 wt% Te NWs at room temperature, which is dozens of times that of the pure PEDOT NW film. This work uses the as-prepared PEDOT NWs/Te NW (90 wt%) nanocomposite film to fabricate a flexible thermoelectric generator and an output voltage of 2.8 mV was generated at a temperature difference of 13.5 K between the environment and human body.",
"conclusion": "4. Conclusions The water soluble PEDOT NWs were synthesized successfully without insulated PSS. Te NWs with high Seebeck coefficient were prepared by solution process. An optimized power factor of 58.03 μW m −1 K −2 has been obtained for the PEDOT NWs/Te NWs (90 wt%) nanocomposite film. The flexible TE generator exhibits an acceptable output voltage of 2.8 mV at an about 13.5 K temperature difference between environment and human body. It promotes research advances in organic materials in the field of thermoelectric and provides great inspiration for the development of flexible and wearable devices. The obtained films are flexible and are ideal for wearable electronic devices. This study provides a new approach to facilitate practical application of flexible TE generator for low-grade energy harvesting.",
"introduction": "1. Introduction In recent years, thermoelectric generator (TEG) devices have gradually been recognized as a viable alternative to power generating applications. 1–3 Thermoelectric generator devices can directly convert thermal energy that passes through them into electricity, which is a pollution-free, convenient and safe conversion technology, and plays an irreplaceable role in environmental friendliness and sustainable development. With the continuous development of portable/wearable and self-powered electronic devices, flexible thermoelectric generators have become a promising power source. 4,5 The advantage of thermoelectric generators is that they have no moving parts and can run continuously without any need for supplementary materials. 6 And flexible thermoelectric generators benefit wearable applications due to their good fit with human skin, thus helping to collect human heat with minimal energy loss. 7 The efficiency of a TE generator is governed by the thermoelectric materials' figure of merit ZT , ZT = S 2 σ / κ , where T is the absolute temperature, S is the Seebeck coefficient, σ is the electrical conductivity, κ is the thermal conductivity and the power factor (PF) is S 2 σ . Applied to the field of thermoelectric, materials should exhibit a large power factor and low thermal conductivity. 8 However, for a given material, the S , σ and κ of thermoelectric materials are highly interdependent and conflict with each other, which causes the difficulty in obtaining high ZT . 9,10 Conventional inorganic materials such as Bi 2 Te 3 , 11 PbTe 12 and Sb 2 Te 3 ( ref. 13 ) have attracted wide attention due to their outstanding thermoelectric properties. However, the expensive raw materials, poor processability and the toxicity restrict the application of inorganic materials in the field of thermoelectrics. 14–16 Conducting polymers are now believed to be potential candidates for TE materials. 17–19 Conducting polymers possess many advantages of light-weight, low cost, mechanical flexibility and enabling the development of portable and wearable thermoelectric devices. 20–22 The development of portable/wearable TE devices has become an area of increasing interest because this device has the potential to use the temperature difference between the human body and its environment to continuously convert human body heat into electrical energy. 23 Moreover, their thermal conductivity is also much lower than that of inorganic materials, which is beneficial to the enhancement of the ZT value. Poly(3,4-ethylenedioxythiophene) (PEDOT) is one of the most promising polymer materials for practical applications in thermoelectric device due to the stability and high conductivity of the p-doped state. 24,25 In order to improve the solubility, PEDOT needs to be emulsified with PSS in water to form PEDOT:PSS. But the PEDOT:PSS film prepared from its aqueous solution has the electrical conductivity of 0.15 ± 0.01 S cm −1 , 26 because in the mixture of PEDOT and PSS, PEDOT is responsible for conducting, whereas PSS is insulated. In general, there are two strategies to optimize the thermoelectric properties of the PEDOT:PSS. In one strategy, it can be secondly doped by a range of chemicals, including ethylene glycol (EG), 27 dimethyl sulfoxide (DMSO) 28,29 or sulfuric acid (H 2 SO 4 ). 30–32 The excess insulating PSS could be removed by treatments process and the PEDOT component transforms from a coil to a expanded-coil or linear conformation, leading to an enhancement of electrical conductivity while the Seebeck coefficient remained essentially constant. 33 In the another strategy, PEDOT:PSS can be mixed with inorganic components which possess large Seebeck coefficients to form PEDOT:PSS/inorganic composites, such as PEDOT:PSS/Bi 2 Te 3 , 34 PEDOT:PSS/PbTe 35 and PEDOT:PSS/SnSe. 36 One-dimensional (1D) nanostructures of inorganic materials were typically used as filers to improve the TE properties. One-dimensional Te nanostructures are widely used as inorganic filers to mix with polymer because of the high Seebeck coefficients, the interface phonon scattering effect and the quantum size effect. 37,38 In addition to the “energy-filtering” process, 39 low-energy carriers are preferentially scattered by the potential barrier at the more interface between polymer matrix and inorganic nanostructures, whereas high-energy carriers easily through cross these barriers, resulting in an increase in S . See et al. 40 reported the hybrid materials of PEDOT:PSS/Te nanorods exhibit a large room temperature ZT of 0.1. Song et al. 41 prepared PEDOT:PSS/PF-Te composites and a maximum power factor of 51.4 μW m −1 K −2 was obtained for the composite film containing 70 wt% PF-Te. In this work, we have firstly prepared one-dimensional PEDOT NWs, which does not need to be mixed with PSS but have good water solubility, and then, we choose Te NWs as inorganic filler to mixed with PEDOT NWs to optimized the TE properties of materials by taking the advantages of the high S of the Te NWs and the low κ of the polymer. The nanocomposite film has excellent room temperature power factor. The Seebeck coefficient and electrical conductivity of the PEDOT NWs/Te NWs (90 wt%) from 100 to 300 K are also reported. The prepared thermoelectric film device shows excellent power generation capacity and the novel method for manufacturing thermoelectric device with PEDOT NWs/Te NWs (90 wt%) composite film provides a good application prospect of the polymer/inorganic flexible materials in the field of portable/wearable thermoelectric devices.",
"discussion": "3. Results and discussion The morphology of the prepared PEDOT NWs and Te NWs are characterized by TEM. Fig. 1a shows the clearly microscopic morphology of the prepared PEDOT NWs. A nanowire bundle consists of many individual nanowires. The average diameter of a single PEDOT NWs is about 100 nm and the length of nanowires can reach several micrometers. As the Fig. 1b shown, it can be observed that the surface of Te NWs is relatively smooth and uniform in size with a diameter of about Te NWs 400 nm and a length of up to 10 μm. Fig. 1 TEM images of the PEDOT NWs (a) and Te NWs (b). The XRD peaks are shown in Fig. 2a , all diffraction peaks of the synthesized Te NWs are consistent with the crystal structure of pure Te phase, and it can be indexed to a trigonal structure Te phase with lattice parameters of a = 4.4328 Å and c = 5.9466 Å, which are consistent with the standard PDF card (JCPDS card no. 36-1452). No diffraction peaks of other phases were detected, indicating that we obtained the pure Te NWs by solution process. As showed in Fig. 2b , the broad peak at 2 θ = 25.8° corresponds to the structure of PEDOT chain. 42,43 The diffraction peaks of both PEDOT NWs and Te NWs can be observed in the XRD spectrum of the nanocomposite films. When the content of Te NWs content in the composite film is low, the diffraction peaks are not easily observed. As the content of Te NWs increases, the diffraction peaks of Te NWs become more obvious and they are very strong for the nanocomposite with 90 wt% Te NWs. Fig. 2 XRD patterns of Te NWs (a), XRD patterns of different Te NWs contents (b), PEDOT NWs (i), and PEDOT NWs/Te NWs nanocomposite films with 10 wt% (ii), 50 wt% (iii) and 90 wt% (iv) Te NWs. \n Fig. 3a shows SEM images of pure PEDOT NWs film. PEDOT NWs in the film formed by vacuum filtration is interwoven together, which may be for this reason that our PEDOT NWs film exhibit high electric conductivity. Fig. 3b and c show SEM images of composite films with different Te NWs contents. Randomly distributed Te NWs can be observed, indicating that Te NWs are well dispersed in PEDOT matrix and form a homogeneous film structure. Clearly, due to the higher electron density of inorganic Te NWs than PEDOT NWs, the Te NWs of the composite films appear bright while the PEDOT NWs matrix appear black or grey in the SEM images. In the Fig. 3d , a higher magnification SEM image clearly shows that Te NWs and PEDOT NWs interweave to form a network microstructure and many polymer/inorganic interfaces. Fig. 3 SEM image of pure PEDOT NWs film (a), PEDOT NWs/Te NWs thin film with 50 wt% Te NWs (b) and 90 wt% Te NWs (c), the high magnification SEM shown in (d) is the red squared portion of the (c). The room-temperature TE properties of the PEDOT NWs/Te NWs nanocomposite films with different contents of Te NWs are shown in Fig. 4 . As observed, pure PEDOT NWs film exhibited power factor of 2.54 μW m −1 K −2 which is 3 orders of magnitude higher than that of the PEDOT:PSS (0.0027 μW m −1 K −2 ), 26 due to the high σ value of 249.5 S cm −1 , and the Seebeck coefficient of 10.08 μV K −1 just a slight drop, which may be due to the absence of PSS and the effect of our PEDOT NWs special 1D conductive network structure. As the content of Te NWs increases, the electrical conductivity of the composite films decreases gradually while the Seebeck coefficient shows an opposite tendency. The enhanced Seebeck coefficient is mainly due to the inherent high Seebeck coefficient of resultant Te NWs (330.04 μV K −1 ) and “energy filtering” effect 44–47 at the interface between PEDOT NWs and Te NWs. Because in the “energy filtering” process, 48–50 low-energy carriers are scattered by the potential barrier formed by the organic/inorganic interfaces of PEDOT NWs and Te NWs, only high-energy carriers can pass through the barrier. High-energy carriers can transfer more heat than low-energy carriers, which increase the Seebeck coefficient. 51 Since Te NWs has a relatively low electrical conductivity compared to PEDOT NWs, the electrical conductivity of the nanocomposites decreased with increasing Te NWs content. However, the electric conductivity of the nanocomposite film decreased slowly, especially before Te NWs content was 70 wt%. This is explained by the fact that the conductive link between the PEDOT NWs was not completely broken when the content of PEDOT NWs is low. In addition, many Te NWs–PEDOT NWs–Te NWs junctions are established in PEDOT NWs/Te NWs nanocomposite films. Therefore carriers can also transport through the Te–PEDOT–Te junctions with low energy barrier (in Fig. 3d ), so that the decrease of electric conductivity caused by the increase of Te content is hindered. 52 It is worth noting that carrier filtering and other mechanisms of improving the thermoelectric properties are mainly related to the interface and surface area, rather than weight fraction of each component. 53 An optimized power factor (PF = S 2 σ) of 58.03 μV m −1 K −2 is obtained for PEDOT NWs/Te NWs composites with 90 wt% Te NWs content at room temperature, which is about dozens of times that of the pure PEDOT NWs film and pure Te NWs film. Moreover, the its thermal conductivity was measured and the value of thermal conductivity for the hybrid ranges from 0.423 to 0.502 W (m K) −1 . Thus, the optimized ZT value is 0.041 at room temperature. All the performance values are comparable with the previous works as showed in Table 1 . Fig. 4 Thermoelectric properties of PEDOT NWs/Te NWs nanocomposite films with different contents of Te NWs. Comparison between the TE properties at room temperature of reported PEDOT:PSS/Te composites System \n σ (cm −1 ) \n S (μV K −1 ) \n S \n 2 \n σ (μW m −1 K −2 ) Reference PEDOT:PSS/Te nanorods treated with H 2 SO 4 — — 42.1 \n 33 \n PEDOT:PSS/Te nanorod ∼700 ∼27 51.4 \n 41 \n PEDOT:PSS/Te NWs — — 28.5 \n 52 \n PEDOT:PSS/Te NWs 163(±4) 19.3(±2.3) 70.9 \n 54 \n PEDOT:PSS/Te NWs doping with EG or DMSO — — 100 \n 55 \n PEDOT:PSS/Te NWs ∼150 ∼0.2 ∼4.5 \n 56 \n PEDOT NWs/Te NWs 72.41 89.52 58.03 This work Furthermore, it is insightful to consider the temperature dependent thermoelectric properties of the PEDOT NWs/Te NWs (90 wt%) nanocomposite film (as showed in Fig. 5 ). At a temperature between 100 and 300 K, the Seebeck coefficient of the nanocomposite exhibit p-type semiconductor characteristics and gradually increases strongly with increasing temperature. The electric conductivity shows a negative TCR where the electric conductivity increases with increasing temperature below 300 K. As described by the Mott's variable-range hopping model, the electrical conductivity and Seebeck coefficient would simultaneously increase with increasing temperature, 57 which is consistent with the results we tested. This characteristic is different from the band-like conduction theory in which the Seebeck coefficient and electrical conductivity show anti-correlated with temperature changes. Fig. 5 Temperature-dependent thermoelectric properties of PEDOT NWs/Te NWs (90 wt%) nanocomposite film. \n Fig. 6a shows a prototype of a flexible power generator with PEDOT NWs/Te NWs nanocomposite film as thermoelectric element. Three single-leg of PEDOT NWs/Te NWs (90 wt%) nanocomposite film (length 36 mm, width 6 mm) were adhered to the polyimide tape and the silver wires were used as electrode. The film is fixed to the silver wires with conductive silver paste to ensure good electrical conductivity. One side of the device is in contact with the bubble film. The output performance of the power generator at a temperature difference of 13.5 K is shown in Fig. 6b . The temperature difference is obtained by contacting one end of the device with the wrist and the other end separated from the skin by a bubble film as a thermal insulator. The output voltage of the module is 2.8 mV at a temperature difference of approximately 13.5 K. The output voltage V 0 of the power generator is defined as: V 0 = NS Δ T , where S is the Seebeck coefficient, N is the number of TE elements and the Δ T is temperature difference. For this flexible device, N = 3, S = 89.52 mV K −1 , Δ T = 13.5 K, the V 0 calculated based on these parameters is 3.63 mV, which is higher than the value obtained by the experiment, probably because the actual temperature of the device's hot end is lower than that of the wrist and the inevitable contact resistance in the device. We firstly used PEDOT NWs/Te NWs composite film as raw material to produce flexible wear-resistant thermoelectric devices with high power factors. The preparation process is simple without PSS, which saves cost and reduces environmental pollution. This experimental result provides attempt for the application of flexible TE generators based on PEDOT NWs/Te NWs films in the wearable field. Fig. 6 (a) Photograph of a flexible device prepared using PEDOT NWs/Te NWs (90 wt%) nanocomposite film as a thermoelectric element and (b) output performance of the deviced produced by a temperature difference between wrist (∼309.65 K) and the ambient (∼296.15 K)."
} | 4,161 |
30366368 | PMC6315946 | pmc | 7,299 | {
"abstract": "The anode of a microbial fuel cell (MFC) was formed on a graphite electrode and immobilized Gluconobacter oxydans VKM-1280 bacterial cells. Immobilization was performed in chitosan, poly(vinyl alcohol) or N -vinylpyrrolidone-modified poly(vinyl alcohol). Ethanol was used as substrate. The anode was modified using multiwalled carbon nanotubes. The aim of the modification was to create a conductive network between cell lipid membranes, containing exposed pyrroloquinoline quinone (PQQ)-dependent alcoholdehydrogenases, and the electrode to facilitate electron transfer in the system. The bioelectrochemical characteristics of modified anodes at various cell/polymer ratios were assessed via current density, power density, polarization curves and impedance spectres. Microbial fuel cells based on chitosan at a matrix/cell volume ratio of 5:1 produced maximal power characteristics of the system (8.3 μW/cm 2 ) at a minimal resistance (1111 Ohm cm 2 ). Modification of the anode by multiwalled carbon nanotubes (MWCNT) led to a slight decrease of internal resistance (down to 1078 Ohm cm 2 ) and to an increase of generated power density up to 10.6 μW/cm 2 . We explored the possibility of accumulating electric energy from an MFC on a 6800-μF capacitor via a boost converter. Generated voltage was increased from 0.3 V up to 3.2 V. Accumulated energy was used to power a Clark-type biosensor and a Bluetooth transmitter with three sensors, a miniature electric motor and a light-emitting diode.",
"conclusion": "4. Conclusions In the course of the research, we investigated the effects of polymers for immobilizing bacterial cells on the anode surface of a microbial fuel cell on its bioelectrochemical characteristics. An MFC with chitosan gel used for immobilization was shown to have the lowest internal resistance (1111 Ohm cm 2 ) and to develop the maximal power (7.6 μW/cm 2 ). Additional introduction of carbon nanotubes into chitosan gel enabled decreasing the MFC internal resistance to 977 Ohm cm 2 and increasing the developed power to 10.6 μW/cm 2 . Modification of PVA and mPVA polymers by carbon nanotubes also reduced the MFC internal resistance by 2.5 and 12.2% and increased the developed MFC power by 20.7 and 18.5%, respectively. Thus, a promising polymer for immobilization of bacterial cells on the graphite anode surface is chitosan modified by carbon nanotubes. Addition of nanotubes leads to an increase in the MFC power due both to the increased biocatalyst/electrode-surface contact area and to the facilitated charge transfer along the formed conductive network of nanotubes on the surface of the electrode. Microbial fuel cells are a new and promising technology, and its cost-effectiveness is still being estimated. There are some reports [ 46 ] that inexpensive materials are capable of achieving more cost-effective energy generation than high-performing materials despite generating lower power. In this work low-cost materials such as graphite rods and chitosan were used. Bacterial cells are also more cost-effective than purified enzymes for energy generation. In this work we present a solution which allows MFCs to power several IoT devices due to the accumulation of energy via a boost converter unit, which increased MFC voltage from 0.3 V to a value of 3.2 V. We are hoping that this solution could be useful for such applications as waste water management and sewage analysis. The obtained results form the basis for application of microbial fuel cells in smart-home and smart-city technologies.",
"introduction": "1. Introduction Efficiency of microbial fuel cells (MFCs) directly depends on the efficiency and sustainability of biocatalyst operation. In most MFCs developed to date, cells are attached to the electrode surface by adsorption. Adsorption is a simple and efficient immobilization technique not affecting the physiological properties of microorganisms, but its drawbacks are insecure attachment of biocatalysts and poor protection from external factors. Entrapment of microorganisms in polymer gels contributes to the preservation of their physiological activity, stimulates metabolism, and protects cells from adverse agents [ 1 , 2 , 3 , 4 ]. Polymers used to immobilize biocatalysts on conductive electrodes can be divided into two groups—natural and synthetic. Natural polymers include polysaccharides, alginic acid salts, carrageenans, chitin and chitosan, gelatin, agar, etc., as well as protein hydrogels. Synthetic matrices for immobilization comprise poly(vinyl alcohol), polyacrylamide, and polyethylene glycol. Smart polymers, subject to strong conformational changes at minor changes in the environment, can be both natural and synthetic. Polymer matrices (e.g., polyacetylene, polypyrrol, polyaniline, etc., in which carbon, graphite, nanoparticles of silver and other metals are used as conductive components) are also used. Various combinations of polymers with nanostructures, such as nanoparticles, nanotubes and nanoengineered smart polymers, enable matrices with novel properties providing for higher stability, sensitivity, biocompatibility, etc. [ 5 ]. Among natural polymer gels, chitosan gel is one of the most widespread agents to immobilize microbial cells [ 6 , 7 ]. Immobilization in chitosan gel is used in MFCs because chitosan, due to its large pores, enables the development of biofilms on the electrode without preventing the flow of nutrient substrates to cells [ 8 ]. Besides, carbon nanotube/chitosan nanocomposites are used to increase the power of microbial fuel cells [ 9 , 10 ]. On the whole, the use of chitosan, owing to its mechanical strength and structure, has a positive effect on operational stability and analytical characteristics of MFCs [ 11 ]. Advantages of synthetic polymers include the possibility of developing carriers with preset properties. Gels based on poly(vinyl alcohol) (PVA) possess a high micro- and macroporosity, which ensures the mass transfer of substrates. Moreover, they exhibit great thermal stability, high resistance to biological degradation and, in practice, insensitivity to the composition of the medium (dissolved substances, buffer, pH) [ 12 ]. Gel based on poly(vinyl alcohol) is a biologically compatible [ 13 ], nontoxic [ 14 ], readily available and cheap polymer. In reference [ 15 ], PVA-based gel has been used for immobilization of Photobacterium phosphoreum to study effects of various toxic substances (phenol, pentachlorophenol, some metal ions, 2,4-dichlorophenoxyacetic acid and 2,4,5-trichlorophenoxyacetic acid) on bacterial luminescence. It has been shown to ensure a long-time stability and intensity of luminescence in biosensor operation. Modification of PVA by, e.g., N -vinylpyrrolidone (mPVA) enables a more stable operation of biosensor receptor elements and MFCs [ 16 , 17 ]. N -vinylpyrrolidone not only is nontoxic but also enhances the activity of the enzyme systems in some microorganisms [ 18 ]. The high resistance of polymer gels necessitates the use of additional materials in bioelectrode constructions, e.g., carbon nanotubes. In reference [ 19 ], the authors describe a mechanism according to which covalent binding forces can arise between lipid membranes and carbon nanotubes. In reference [ 20 ], an assumption is hypothetically considered that, as the distance between the enzyme’s active centre and the graphite electrode decreases, at a critical distance between them conditions can arise for a facilitated transfer of charge from the electrode to the enzyme. The same interaction pattern, not necessarily leading to mediator-free transfer but facilitating charge transfer, can be assumed to exist for whole bacterial cells, where PQQ (pyrroloquinoline quinone)-dependent dehydrogenases incorporated into the lipid envelope of cell membranes will be an analogue of the enzyme. Gluconobacter oxydans are frequently used for biotechnological applications, because these bacteria have high growth rates when cultivated on growth media, possess a high metabolic activity, are relatively stable in immobilization and require no external cofactor for their PQQ-dependent dehydrogenases to function [ 21 , 22 , 23 ]. Besides, they have a significant number of membrane-localized enzyme complexes (aldose and alcohol dehydrogenases capable of oxidizing quite a number of carbohydrates and alcohols), which simplifies electron transfer in the system and enables their use in biosensors and as an MFC biocatalyst [ 22 , 24 , 25 , 26 , 27 , 28 , 29 ]. In adsorption contact of bacterial cells with carbon nanotubes, a quite possible situation is, according to the principle of random interaction, to form covalent bonds between lipids, that enclose PQQ-dependent dehydrogenases, and nanotubes. Due to these properties, Gluconobacter cells are ideal for test studies. The aim of the work was to assess the electrochemical characteristics of MFCs based on G. oxydans bacterial cells immobilized into the following polymers: chitosan, poly(vinyl alcohol) and N -vinylpyrrolidone-modified poly(vinyl alcohol). The developed MFCs were modified by multiwalled carbon nanotubes (MWCNT) and used to accumulate electric energy via a boost converter to power a biosensor electrode, Bluetooth transmitter, miniature electric motor and light-emitting diode.",
"discussion": "3. Results and Discussion 3.1. Applied Potential Selection The amplitude of current generated by the MFC at the oxidation of ethanol by bacterial cells depends on the applied potential ( Figure 2 , curve 1). The maximal generated current of the electrode in the presence of mediator was observed at an applied potential of 200 mV. These results correlate with the literature data (the standard redox potential of DCPIP mediator is +0.217 V [ 34 ]). Also, we obtained the values of charge transfer resistance ( R ct ) at various applied potentials by electrochemical impedance spectroscopy ( Figure 2 , curve 2). The minimal values of R ct were obtained at an applied potential of −150 to −180 mV; maximal, at 200 mV. These dependences were also obtained for MFCs with bacterial cells immobilized into PVA and mPVA (data not shown). For this reason, further amperometric measurements were carried out at an applied potential of 200 mV, and the impedance measurements were conducted at an applied potential of −150 mV. 3.2. Behaviour of Bacterial Cells in Polymer Gels The effect of the immobilization matrix on cells was investigated by light microscopy. Cells were mixed with various polymers ( Figure 3 ) under the same conditions and their behaviour was compared. The micrographs show that cells mixed with chitosan are uniformly distributed along the matrix, and this state does not change with time. In PVA gel, cells form conglomerates immediately after mixing; the size of the conglomerates changes with time. This can be due to the unfavourable impact of this matrix on bacterial cells, as the result of which bacteria tend to minimize the contact with the matrix. This effect is also observed for N -vinylpyrrolidone-modified PVA; however, conglomerates are smaller. This can be due to a decrease of agglomeration of PVA in the case of its modification by various compounds similar to those described in references [ 14 , 35 ]. The presented data suggest that, when using PVA and mPVA, only those cells that are on the outer boundary of conglomerates are in contact with the electrode surface. This means that the total electric signal from the same number of cells should be theoretically higher in the case of using chitosan gel. 3.3. Effect of the Concentration of Polymer and Bacterial Cells on the Electrochemical Characteristics of MFC Bioanode Figure 4 a shows the dependences of MFC power and internal resistance on the ratio of cells and chitosan polymer on the anode surface. The maximal values of MFC power and the minimal values of MFC internal resistance were obtained at a cell/chitosan volume ratio of 1:5. Figure 4 b presents the same characteristics for the electrode based on mPVA polymer. The maximal values of MFC power and the minimal values of MFC internal resistance were obtained at a cell/mPVA volume ratio of 2:1. A similar dependence was also obtained for PVA polymer (data not shown). The obtained data for synthetic polymers can be due to the self-organization of Gluconobacter cells in the PVA and mPVA matrices ( Figure 3 ), the extent of which increases at a prolonged contact with polymer [ 36 ] and, possibly, at an increase of its amount. We obtained dependences of MFC power and internal resistance for these polymers at various concentrations of bacterial cells on the anode surface mixed at ratios of 1:5 (cells/chitosan, v / v ) and 2:1 (cells/PVA or mPVA, v / v ). Variation of the concentration of cells on the anode surface within the range of 0.012–0.35 mg/mm 2 for all polymers had no significant effect on the electrochemical parameters of the MFCs. Inset of Figure 5 shows typical MFC signals in response to the introduction of ethanol at an applied potential of 200 mV. The maximal amplitude of the signal was observed for an MFC in which chitosan was used as an immobilization agent. Figure 4 presents typical polarization curves for an MFC modified by various polymers, and MFC power characteristics calculated from them. Correspondingly, the minimal values of MFC internal resistance and the maximal values of developed power are also observed for an MFC modified by chitosan ( Table 1 ). The table compares the characteristics of the main bioelectrochemical parameters for MFCs modified by various polymers as measured by voltammetry and impedance spectroscopy. All values of resistances obtained by impedance spectroscopy were measured at an applied potential of −150 mV, because at this potential the MFC internal resistance is minimal, as seen in Figure 2 , curve 2. Table 1 presents both the electrode charge transfer resistances for three types of bioanodes ( R ct ) and the total internal resistances of the systems ( R in ). The value of anode resistance is up to 75% of the total internal resistance of the system, which once again emphasizes the importance of modifying bioanodes with the view of reducing their resistances. The contribution of chitosan to the bioanode resistance is the least of the three investigated polymers; therefore, chitosan ensures a reliable attachment of material and does not prevent vital activities of cells. Thus, the use of chitosan for immobilization of microorganisms makes it possible to achieve the highest MFC power. 3.4. Modification of the Anode by Carbon Nanotubes To improve the conductance and increase the active area of the surface, the bioanode is modified by various nanomaterials, including nanotubes [ 10 , 37 , 38 ]. In this work we used multiwalled carbon nanotubes, because earlier they have been shown to be less toxic for bacterial cells than single-walled nanotubes [ 39 ]. Cells of Gluconobacter oxydans differ from many other bacteria by the arrangement of their enzyme complexes [ 26 ]. Pyrroloquinoline quinone-dependent dehydrogenases are exposed to the outer side of the cell lipid membrane, which simplifies the access of substrates to them. Modification of the bioanodes by carbon nanotubes may create conductive channels constructed from nanotubes that are distinguished by minimal resistance. Nanotubes, being in immediate contact with cell lipid membranes provide for facilitated transfer of electrons from enzyme complexes of bacteria to the surface of graphite electrodes via the layer of the immobilization matrix. Such a network should theoretically reduce the bioanode impedance and MFC resistance, which would ultimately increase the generated power of the MFC [ 40 ]. Table 2 shows the MFC characteristics at the modification of the electrode by carbon nanotubes immobilized with cells in three polymers. As seen from Table 1 and Table 2 , modification of a PVA/ G. oxydans electrode by carbon nanotubes increases the MFC power by 20.7% (from 6.56 to 7.92 μW/cm 2 ); thereby, the internal resistance decreases by only 2.5% (from 1251 to 1219 Ohm cm 2 ). Modification of an mPVA/ G. oxydans electrode by carbon nanotubes also led to a rise in the MFC power by 18.5% (from 6.10 to 7.23 μW/cm 2 ) and a decrease of the internal resistance by 12.2% (from 1455 to 1277 Ohm cm 2 ). Modification of a chitosan/ G. oxydans electrode by nanotubes caused a power increase of 38.4% (from 7.63 to 10.56 μW/cm 2 ) and a decrease of internal resistance by 12.1% (from 1111 to 977 Ohm cm 2 ). Figure 6 shows the impedance spectres of MWCNT-modified electrodes depending on the used polymer. The lowest charge transfer resistance was demonstrated by an electrode with bacterial cells immobilized using chitosan. Thus, modification of an MFC anode by carbon nanotubes leads to an increase of MFC power and a decrease of internal resistance as compared with the nonmodified anode when using any of the three investigated polymers. 3.5. Applications of MWCNT/Chitosan/G. oxydans MFC Microbial fuel cells are low-power supply sources (10 −7 –10 −3 W/cm 2 ). Still, MFC applications are limited by not only their low power but also the low value of generated voltage. In this context, a topical problem for their broader application is to increase the MFC output voltage. To increase MFC energy by increasing voltage, this work used a boost converter based on a bq 25504 integrated circuit that performs direct current transformation [ 41 ]. In the general case, the converter accumulates and stores electric energy generated by unstable sources. As applied to MFCs, the method of increasing voltage by means of a converter has not been reported earlier. We developed a system connecting a bq 25504-based converter to an MFC [ 42 ] and worked out various modes of increasing and storing output voltage, which in practice corresponded to accumulation of electric energy. This system increased MFC voltage from open-circuit potential (0.3 V) to a preset value of 3.2 V. The time of accumulating the preset voltage across a 6800-μF capacitor depends on the number of MFCs connected to the converter and their modifications ( Table 3 ). It is seen from the data presented that in a single MFC the generated voltage is too low for an efficient start of the converter and capacitor charge. This charging feature is due to the characteristic of the used integrated circuit bq 25504, which requires an input voltage of 300 mV and higher. Series connection of two MFCs makes it possible to sum up their voltages to ensure the converter start voltage. Therefore, the internal resistance of the system does not decrease, but the total voltage output of the MFC increases. Figure 7 plots the charging of a 6800-μF capacitor by two MWCNT-modified MFCs connected in series. The low-efficiency slow charge phase (up to 1.7 V) was 28 min; the high-efficiency fast charge phase, 4 min; the total capacitor charge time was 32 min. The dependence of capacitor charge time on MWCNT-modified MFC functioning time as compared with nonmodified MFCs is given on Figure 8 . Two nonmodified microbial fuel cells charged a capacitor of 6800 μF in 80 min; therefore, the larger part of the time (60 min) was the slow charge phase. At a modification of the MFC by nanotubes the total charge time decreased to 60 min, the slow charge time being 52 min. The time occupied by the slow charge phase decreased by 23%; the time of the fast charge phase, by 34%. It should be noted that two series-connected nanotubes-modified MFCs efficiently charged a capacitor of 6800 μF in 42 min after a 24-h continuous operation. After a 48-h continuous operation of the MFCs, the capacitor charge time decreased more than two times as compared with two nonmodified cells to make 30 min. It can be assumed that this improvement of MFC parameters is due to the phenomenon of adaptation presented, e.g., in references [ 43 , 44 ]. The charged capacitor contained a charge of 21 × 10 −3 C, having accumulated energy of 32.7 mJ, which made it possible to maintain the luminescence of a light-emitting diode (L-1154SURDK, Kingbright; 2.0 V, 20 mA) or to rotate an electric motor rotor (M25E-4L, MITSUMI; 3.0 V, 100 mA) in a short-time regime. Accumulated charge was used to power a low-energy amperometric biosensor based on a Clark-type electrode. The value of consumed current for an ~0.3 mm-diameter platinum electrode was of the order of 10 –9 А. A charged capacitor of 6800 μF provided the operation of a biosensor based on a Clark-type oxygen electrode for 27 h (biosensor connections and operation scheme not shown). The charged capacitor of 30,000 μF (charging time, 80 min) was used to power a device with a Bluetooth transmitter and three sensors (for temperature, humidity, light intensity) for 2 min net time (preliminary data presented at the IEEE 4th World Forum on the Internet of Things [ 45 ]). Such devices can be used as part of the Internet of Things technology to monitor various parameters of the environment with sensors’ polling period of approximately once per 2 h."
} | 5,269 |
36844455 | PMC9947323 | pmc | 7,301 | {
"abstract": "Summary Regenerative agriculture (RA) is gaining traction globally as an approach for meeting growing food demands while avoiding, or even remediating, the detrimental environmental consequences associated with conventional farming. Momentum is building for science to provide evidence for, or against, the putative ecosystem benefits of RA practices relative to conventional farming. In this perspective article, we advance the argument that consideration of the soil microbiome in RA research is crucial for disentangling the varied and complex relationships RA practices have with the biotic and abiotic environment, outline the expected changes in soil microbiomes under RA, and make recommendations for designing research that will answer the outstanding questions on the soil microbiome under RA. Ultimately, deeper insights into the role of microbial communities in RA soils will allow the development of biologically relevant monitoring tools which will support land managers in addressing the key environmental issues associated with agriculture.",
"conclusion": "Conclusion RA systems are complex and can be context specific; understanding these systems requires multidisciplinary investigations at a range of spatiotemporal scales and across geographic areas and farming systems. Soil microbial communities should not be overlooked in this research as we add more scientific knowledge to the field. Understanding the role of soil microbial communities in RA will not only allow the development of biologically relevant monitoring tools but potentially could also help us understand how microbial communities can be manipulated or supplemented to better support the goals of RA. Implementation of such practices supports meeting increasing global food demands while reducing the environmental burden of agricultural production.",
"introduction": "Introduction One of the greatest challenges that humanity is currently facing is meeting growing food demands while reducing environmental damage from agriculture. To guarantee food security into the future, it is estimated that food production needs to roughly double. 1 Most of the current demand is met by conventional agriculture. Conventional agriculture contributes heavily to the deterioration of our soil and water environments, the accumulation of greenhouse gasses, and biodiversity losses, in part due to high stock densities, heavy pesticide and fertilizer use, monocultures, and soil tilling. 2 , 3 , 4 Drastic changes are required, not only to address the extent of soil degradation and related productivity and biodiversity loss but also to ensure food security into the future. Regenerative agriculture (RA) is touted as a “back to basics” solution to improve soil quality and biodiversity, while maintaining or improving productivity and profitability. 5 While RA lacks an accepted definition, it is commonly described as a set of on-farm practices that generally differ from conventional farming in terms of the types and intensities of human disturbances and/or inputs within the farm system 6 ( Table 1 ). These practices vary, but most often include a lack of tillage (mechanical disturbance), excluding synthetic fertilizer and pesticide use, inclusion of increased plant diversity, and the integration and altered management of crops and livestock. 5 , 6 Limited evidence shows that RA can improve soil nutrient content and physical structure, 7 , 8 , 9 while increasing profit and reducing pests. 5 Perhaps most importantly, RA is purported to increase atmospheric carbon sequestration and is, therefore, touted as a strategy to help address global climate change issues. 10 While the RA “movement” is being increasingly adopted by farmers at the grassroots level, comprehensive scientific research on the ecosystem-level changes induced by RA, relative to conventional farming systems, has only recently gained momentum; thus, questions remain about the extent to which regenerative approaches are able to achieve better environmental outcomes, such as soil quality improvements and climate change mitigation, compared to conventional practices. Table 1 Descriptions of practices commonly associated with regenerative agriculture, the rationale for their implementation, and examples of research relevant to each Practice Description Rationale for implementation Examples Reduced tillage Soil disturbances are minimized by adopting a no-till or conservation-tillage approach, the latter meaning plant residues are maintained on at least 30% of the soil surface. Direct-drill cropping and pasture-sowing methods are adopted. Can decrease energy consumption and reduce CO 2 emissions while potentially increasing carbon sequestration (but see Cai et al. 11 ). Reduces soil erosion, improves soil fertility, and increases biodiversity and water retention Holland, 12 ; Blanco-Canqui and Ruis, 13 Cover crops and crop rotation Using close-growing crops to cover the soil between normal crop production or between trees/vines in orchards and vineyards. Production crops alternate sequentially on the same land. Improves soil nutrients, while reducing erosion and weed growth. Legume cover crops can increase soil N content and reduce the need for fertilizers. Crop rotation improves production and disrupts insect/pest reproduction cycles Blanco-Canqui and Ruis, 13 ; Adetunji et al. 14 ; Shah et al. 15 High-diversity pasture Replacing traditional low-diversity pastures (usually made up of one or two species) with diverse mixes, often selected specifically to suit the farm purpose. Better pasture utilization, growing season extension, increased nutrition for grazing animals, increased milk production, and improved animal welfare. Thought to increase overall biodiversity and improve ecosystem functioning. Pembleton et al. 16 ; Distel et al. 17 No synthetic fertilizers or pesticides Reduce or eliminate the need for synthetic fertilizers, instead using things like organic compost and bio-supplements (e.g., compost, seaweed extracts, fish hydrolysates, and vermicast). Avoidance of pesticide use. Avoids the negative environmental impacts of synthetic chemicals, including increased greenhouse gas emissions and eutrophication of aquatic ecosystems. Paungfoo-Lonhienne et al. 18 Grazing management Rotating livestock across smaller paddocks or delineating grazing areas for short periods and only returning to a previously grazed paddock when pasture has recovered and regrown. Manipulating the amount of time that grass spends in the active growth phase by managing the duration and timing of grazing is thought to enhance soil carbon sequestration. Furthermore, preventing overgrazing means rootstocks are less impacted, allowing for quicker recovery. Teague and Barnes, 19 Because soil microbial communities are crucial for maintaining soil quality and are a fundamental component of the soil ecosystem, 20 , 21 the soil microbiome is likely to serve as the belowground “engine” for delivering many of the key benefits of RA practices. A difference in microbial community composition in RA sites compared to conventional systems has been confirmed, 22 and the activity of microbes under RA may increase. 9 Microbial biomass is also known to increase in RA sites compared to conventional systems. 8 This is not surprising, given that there is clear evidence that aboveground activity impacts both the bacterial 23 , 24 and fungal 25 belowground ecosystem components. However, the limited scientific research on RA that has been conducted to date has largely overlooked relationships between RA practices and soil microbial communities. We need to investigate how both the composition and functional profiles of the microbiome change. For example, genes that can be used to define substrate utilization and the capacity for nutrient cycling, toxin or heavy metal degradation, and microbial stress responses can be quantified."
} | 1,971 |
30218023 | PMC6138649 | pmc | 7,302 | {
"abstract": "The uniformity of crop yield is extremely important for consumers and of as much relevance to the grower as overall yield. However, size inequality within a plant population is rarely measured and has never before been considered in relation to the use of beneficial microbes for yield enhancement. For the first time, we show that addition of soil bacteria to calabrese plants significantly increased size inequality. These effects were usually more apparent in above-ground biomass. This was caused by some (but not all) plants growing very large when inoculated with bacteria, while control plants were mostly small. We suggest that the main reason is the incompatibility of the inoculated bacteria with those already present in the rhizosphere. In some cases the inoculum matched the indigenous community, providing a benefit to plant growth, while often it did not and plants remained relatively small. We conclude that analyses of size inequality should be an integral part of experiments using microbial soil amendments. These analyses can help to inform the production of more effective microbial products and to ensure that the integration of beneficial microbes into sustainable production systems does not impair uniformity in yield.",
"conclusion": "Conclusions We have presented the first demonstration that application of PGPR may promote size inequality within crop yield, even if there are no absolute changes in yield. These results have important implications for consumers as well as farmers and growers and those who produce microbial inoculants. Uniformity within a crop strongly influences consumer preference and if PGPR are to be integrated into organic production systems, there is a need to ensure that overall quality of the product is not reduced, even if yield is enhanced. Any attempt to reduce the yield gap between conventional and organic systems should involve an analysis of size inequality, to ensure marketability is maintained. Analyses of size inequality can also tell us a lot about the compatibility of products with the rhizosphere. Our results are most likely caused by inconsistent establishment of the inoculated species in soil, meaning that some plants benefited from inoculation and grew large, while many did not and remained small. This situation is the opposite of the uniform distribution that growers aim to produce. Future PGPR products need to be more tailored to specific crop situations or soil types, and a universal inoculant is therefore unlikely to be successful at present.",
"introduction": "Introduction Conventional agriculture, through the use of selected varieties and inputs of fertilizers and pesticides, seeks to maximise not only total yield, but also the uniformity of the crop. Much research has shown that uniformity of size and appearance of a fruit or vegetable influences consumer choice 1 . Indeed, uniformity of size is a more important factor than knowledge of pesticide application to the crop, and even scent and flavour 2 , 3 . Furthermore, in field crops such as calabrese, a lack of uniformity in plant size makes harvesting difficult and prolonged, thereby increasing production costs 4 . Organic production systems, designed to minimise the impact of agriculture on humans and the environment, have seen a rise in popularity over the last 20 years. While the environmental benefits cannot be disputed, the main disadvantage is the difference in total yield, which may be 20–30% less than that of conventional systems 5 , 6 . However, an unrecognised aspect of organic agriculture may be an increase in size inequality (i.e. a decrease in uniformity) of the crop. Levels of beneficial soil microbes, such as arbuscular mycorrhizal (AM) fungi, and foliar-feeding insects and pathogenic fungi may all be higher in organic systems and all have the potential to increase size inequality 7 – 9 . This is because not all plants in a population are colonized to the same degree by AM fungi, or attacked equally by insects or pathogens, resulting in plant size distributions that become ‘stretched’ at the lower end (many small individuals) and/or at the upper end (fewer very large individuals that escape attack or are mycorrhizal). An important aspect of any agricultural system is the quality of the soil, with much recent interest focused on improving quality, and ultimately yield, through the addition of microbial inoculants 10 . Prominent amongst these inoculants are plant growth-promoting rhizobacteria (PGPR), including species in genera such as Azotobacter , Bacillus , and Pseudomonas 11 . These PGPR can enhance plant growth through nutrient recycling, nitrogen fixation, phytohormone production, solubilisation of nutrients such as P, K and Fe, and enhancing plant resistance to pests and diseases 11 . Therefore, the primary aim of PGPR addition is to increase overall yield. However, whether their addition has any effects on crop uniformity is unknown. Thus, the aim of this paper is to examine the effects of PGPR addition on size inequality within an important field crop, Brassica oleracea var. italica (calabrese). Our holistic approach to studying the effects of ubiquitous PGPR ( Bacillus spp.) on calabrese growth, endophytic bacterial community and plant biotic stress in previous studies led us to investigate Bacillus -mediated effects on plant size inequality. In a series of experiments in controlled and field conditions, with single and multiple Bacillus spp. additions, we showed that these PGPR affected plant growth 12 , altered endophytic bacterial community diversity, evenness and composition 13 , and suppressed cabbage aphid growth 14 and field incidence 15 in a context specific manner. Calabrese is sensitive to variation in soil N and water availability, which can often lead to a lack of uniformity in the crop 16 . Therefore, it is relevant that nutrient delivery to roots and amelioration of drought stress are just two of the benefits that PGPR can provide to plants 11 . However, much attention has been focused on the inconsistent efficacy of PGPR inoculants in field conditions, due mainly to compatibility issues of species in the inoculant with those in the rhizosphere and the heterogeneous distribution of nutrients in soils 17 , 18 . We therefore hypothesized that addition of PGPR to the roots of calabrese plants would increase plant size inequality, due to the differential ability of bacterial species to establish in the rhizosphere 19 . Furthermore, we hypothesized that plant size inequality would be further amplified in field conditions and with multispecies bacterial inoculants, because of nutrient heterogeneity and competitive interactions between the bacterial species 13 , 18 .",
"discussion": "Results and Discussion Addition of the bacterial mixture only had an effect on root biomass in the controlled experiment, which was reduced when PGPR were applied (Fig. 1 ). However, bacterial addition caused a significant increase in the inequality of total biomass, raising the CV from 18% to 44% ( Z = 4.1, p < 0.001) (Table 1 ). Similar significant results were seen with both root and shoot biomass and were reflected in the Gini coefficient and the Gini Mean of Differences (Table 1 ). A notable feature of the Lorenz asymmetry coefficients was that those for control plants were always less than one, while those for treated plants were all greater than one (Fig. 2 and Table 1 ). The former indicates that the majority of control plants were small, while the latter indicates that some large individuals occurred when bacteria were applied. PGPR addition therefore increased plant size variation, upholding our original hypothesis. Figure 1 Box plots showing the range in size distributions for ( a ) total biomass, ( b ) root biomass and ( c ) shoot biomass of calabrese plants grown in the controlled experiment, with and without the addition of a mixture of plant growth-promoting rhizobacteria. The horizontal line within the box is the median, while edges of the box represent the inter-quartile ranges. The whiskers depict 1.5x the inter-quartile ranges, while points depict outliers beyond the whiskers. Table 1 Measures of inequality for calabrese plants grown in controlled conditions, with and without the addition of PGPR. Total biomass Root biomass Shoot biomass CV control 18.03 18.7 18.1 CV bacteria 44 . 01 45 . 09 46 . 52 Z test 4.13, p < 0.001 4.05, p < 0.001 4.31, p < 0.001 Gini control (95% CI) 0.0952 (0.0645–0.1432) 0.101 (0.0699–0.1466) 0.0968 (0.0659–0.1422) Gini bacteria (95% CI) 0 . 231 (0.1582–0.3468) 0 . 247 (0.1777–0.3565) 0 . 251 (0.1784–0.3646) Gini MD control (95% CI) 8.789 (6.097–12.603) 1.728 (1.221–2.376) 7.252 (5.084–10.203) Gini MD bacteria (95% CI) 21 . 699 (14.74–30.55) 3.223 (2.36–4.283) 20 . 906 (14.62–28.29) Lorenz AC control 0.8774 0.7703 0.8537 Lorenz AC bacteria 1.0918 1.0146 1.0587 CV is Coefficient of Variation, Gini is Gini Coefficient, Gini MD is Gini Mean of Differences and Lorenz AC is Lorenz Asymmetry Coefficient. Differences between bacterial treatments and the control are indicated in bold text. Differences in Gini Coefficients and Gini MD determined by non-overlap of confidence intervals at p = 0.05. Figure 2 Lorenz curves for ( a ) total biomass, ( b ) root biomass and ( c ) shoot biomass of calabrese plants grown in the controlled experiment. The diagonal straight line represents the line of equality. Control plants depicted in black, PGPR addition in red. There are many studies in which bacterial addition has been shown to increase plant biomass 11 , but all previous authors report changes in mean plant size, rather than inequality. Patchy effects of inoculation have been recorded before 17 , but only at the overall treatment or plot level and never at the within-treatment level, as described here. The most likely reason for the effects seen is variable colonization and establishment of the PGPR in the rhizosphere 19 . The addition of just one bacterial species can have major effects on the structure of the rhizosphere microbial community, depending on whether the added species was already a member of that community or not 13 . Such effects can be magnified if a mixture of PGPR is added 18 . Thus, the most plausible explanation is that the mixed inoculum of bacteria did not establish to the same degree in each pot. When successful rhizosphere colonization occurred, plants benefited from the addition and grew large, but in some individuals, growth promotion did not occur, leading to the increase in inequality and decrease in uniformity of plant size. Commercial potting media may vary from bag to bag in their physical and chemical characteristics 20 and it is highly likely that they also vary in soil microbial community composition too. Field populations of soil bacteria are notoriously heterogeneous at all spatial scales 21 and so it is not surprising that the CVs for field grown plants in the UK were considerably larger than those for the pot-grown plants (Table 2 ). It should be noted that these plants were also subject to insect attack, which can affect size inequality 7 . Here, addition of either a single inoculation or a mixture of species increased the CV of total biomass from 25.5% in controls up to 71% in plants inoculated with B . amyloliquefaciens ( Z = 3.4, p < 0.001). It is probably no coincidence that addition of this bacterium also caused the greatest changes in the indigenous rhizosphere microbial community in this experiment 13 . There was a much greater range in biomass in all treated plants, relative to controls (Fig. 3 ), and the Lorenz curves for total and shoot biomass showed clear increases in inequality when any PGPR were applied (Fig. 4 ). However, addition of B . subtilis or B . amyloliquefaciens had no effect on the inequality of root biomass (Table 2 and Fig. 4b ), showing that effects of PGPR addition are not just species-specific, but also plant-organ specific. Furthermore, the addition of inoculants did not result in increases in the Lorenz asymmetry coefficients (Table 2 ) and in all cases, populations were composed of a majority of small individuals. This is likely due to the incompatibility of the added species with those already in the rhizosphere, suggesting that relatively few plants benefited from the addition of PGPR. Indeed, no significant increases in mean plant size were found in this study, due in part to the large inequality seen in some treatments 15 . Table 2 Measures of inequality for calabrese plants grown in UK and Indian field conditions, with and without the addition of PGPR. Abbreviations as in Table 1 . UK field experiment Indian field experiment Total biomass Root biomass Shoot biomass Total biomass Root biomass Shoot biomass CV control 25.51 41.25 29.14 23.36 38.22 21.36 CV B . amyloliquefaciens Z test 71 . 41 3.41, p < 0.001 48.83 0.619, NS 77 . 61 3.16, p < 0.01 42 . 45 2.29, p < 0.05 41.86 0.34, NS 42 . 78 2.65, p < 0.01 CV B . cereus Z test 48 . 19 2.37, p < 0.05 70 . 79 2.82, p < 0.05 53 . 33 2.21, p < 0.05 38.17 1.91, NS 41.53 0.31, NS 37 . 91 2.24, p < 0.05 CV B . subtilis Z test 67 . 42 3.29, p < 0.001 46.09 0.41, NS 66 . 83 2.83, p < 0.01 41 . 79 2.24, p < 0.05 43.69 0.51, NS 41 . 79 2.57, p < 0.05 CV mixture Z test 70 . 01 3.36, p < 0.001 90 . 15 2.37, p < 0.05 82 . 68 3.27, p < 0.01 39 . 57 2.06, p < 0.05 35.93 0.24, NS 40 . 11 2.43, p < 0.05 Gini control (95% CI) 0.150 (0.123–0.189) 0.237 (0.187–0.312) 0.167 (0.140–0.215) 0.128 (0.092–0.188) 0.221 (0.176–0.286) 0.122 (0.092–0.168) Gini B . amyloliquefaciens (95% CI) 0 . 391 (0.317–0.490) 0.282 (0.219–0.331) 0 . 413 (0.332–0.507) 0 . 249 (0.194–0.301) 0.244 (0.186–0.311) 0 . 251 (0.195–0.313) Gini B . cereus (95% CI) 0 . 277 (0.234–0.352) 0.367 (0.290–0.471) 0 . 299 (0.254–0.374) 0.214 (0.167–0.286) 0.233 (0.181–0.328) 0.212 (0.168–0.279) Gini B . subtilis (95% CI) 0 . 379 (0.307–0.476) 0.263 (0.209–0.331) 0 . 375 (0.323–0.444) 0 . 243 (0.195–0.318) 0.253 (0.205–0.319) 0 . 243 (0.185–0.318) Gini mixture (95% CI) 0 . 372 (0.287–0.493) 0 . 437 (0.317–0.567) 0 . 404 (0.293–0.523) 0.230 (0.170–0.313) 0.206 (0.143–0.295) 0 . 233 (0.174–0.316) Gini MD control (95% CI) 48.25 (39.59–60.11) 4.86 (3.86–6.94) 35.04 (27.92–45.3) 34.42 (24.66–55.71) 4.83 (3.75–6.44) 30.16 (22.60–42.44) Gini MD B . amyloliquefaciens (95% CI) 204 . 45 (148.3–319.7) 9 . 35 (7.39–12.86) 145 . 22 (95.2–215.9) 99 . 69 (84.56–115.38) 8 . 13 (6.52–10.19) 92 . 05 (78.17–106.30) Gini MD B . cereus (95% CI) 133 . 07 (105.1–166.9) 11 . 49 (7.37–18.01) 98 . 53 (72.65–132.52) 56.11 (40.15–74.32) 5.37 (3.84–7.48) 50.83 (35.86–67.33) Gini MD B . subtilis (95% CI) 168 . 24 (133.52–252.8) 6 . 92 (5.22–9.28) 98 . 97 (77.32–130.97) 90 . 29 (72.82–116.56) 7.40 (6.18–9.66) 83 . 13 (66.75–107.81) Gini MD mixture (95% CI) 234 . 01 (157.3–370.8) 13 . 17 (7.36–23.66) 156 . 80 (87.6–268.2) 105 . 14 (83.60–134.20) 7.50 (5.63–9.89) 98 . 05 (77.80–125.09) LAC control 0.998 0.928 1.073 1.208 0.923 1.161 LAC B . amyloliquefaciens 0.880 0.859 0.976 0.802 0.867 0.804 LAC B . cereus 0.976 1.084 1.049 1.256 1.121 1.254 LAC B . subtilis 0.927 0.945 0.969 0.738 0.853 0.717 LAC mixture 1.078 0.913 1.218 0.986 0.734 1.015 Differences between bacterial treatments and the control are indicated in bold text. Differences in Gini Coefficients and Gini MD determined by non-overlap of confidence intervals at p = 0.05; NS = no significant difference. Figure 3 Box plots showing the range in size distributions for ( a ) total biomass, ( b ) root biomass and ( c ) shoot biomass of calabrese plants grown in UK field soil, with and without the addition of plant growth-promoting rhizobacteria. Addition of PGPR indicated by: B. amy: Bacillus amyloliquefaciens ; B. cer: B . cereus ; B. sub: B . subtilis ; mix: mixture of all three species. Figure 4 Lorenz curves for ( a ) total biomass, ( b ) root biomass and ( c ) shoot biomass of calabrese plants grown in UK field soil. The diagonal straight line represents the line of equality. Control plants depicted in black, B . amyloliquefaciens addition in green, B . cereus addition in red, B . subtilis addition in blue and the mixed inoculum in brown. Plants grew more rapidly in the warmer climate of India, but were smaller in size at harvest than those from the UK. Addition of the mixture increased plant biomass (Fig. 5 ) and the CV for total biomass of controls was similar to that in the UK, at 23.4% (Table 2 ). Most PGPR treatments increased size inequality, although B . cereus had far fewer effects than in the UK (Fig. 5 ). Addition of B . amyloliquefaciens again produced the greatest increase in the CV of total biomass, raising this to 42.4% ( Z = 2.29, p < 0.05). As with UK plants, addition of single species and the mixture caused increases in the Gini coefficient and Gini Mean of Differences, with greatest increases seen when B . amyloliquefaciens or the mixture were applied (Table 2 ). As with the UK experiment, inequality of root biomass was much less affected than shoot biomass (Table 2 ). Lorenz asymmetry coefficients also showed a similar trend to the UK, in that the majority of plants in the bacterial treatments were small. Coefficients in B . cereus treatments were very similar to controls, reinforcing the conclusion that addition of this species had no measurable effects on the plants (Fig. 6 ). Figure 5 Box plots showing the range in size distributions for ( a ) total biomass, ( b ) root biomass and ( c ) shoot biomass of calabrese plants grown in Indian field soil, with and without the addition of plant growth-promoting rhizobacteria. Figure 6 Lorenz curves for ( a ) total biomass, ( b ) root biomass and ( c ) shoot biomass of calabrese plants grown in Indian field soil. Legend as in Fig. 4 . Despite the differences in calabrese cultivars and the strains of Bacillus spp. used in the two field experiments, the results were consistent in that PGPR addition increased inequality, but also showed context-specificity, as inoculation with the same bacteria did not produce the same effect in each place. The most likely explanation is that the natural rhizosphere bacterial communities in the UK and India differed 21 , resulting in inconsistent effects. It has been suggested that inoculation of seed with PGPR can help to overcome such problems and enable the added bacteria to establish better, in the face of antagonism from indigenous species 22 . However, we inoculated seeds in our field experiments and still found patchy effects of inoculation, even though the inoculated species successfully colonized the rhizosphere 13 . Perhaps of more promise is the encapsulation of bacterial cells within products such as sodium alginate 23 , though these authors also acknowledge that the most appropriate encapsulation method likely varies from one bacterial species to another. Intriguingly, in a study that involved one of the species used here ( B . subtilis ), addition of encapsulated bacteria reduced inequality of lettuce growth, relative to controls 24 . The analysis of heterogeneity within crops is of much interest from the economic point of view, as consumers are willing to pay a premium for attributes that include quality and uniformity 25 . Harvesting of crops such as calabrese is more efficient and economical when uniformity of the product is high 4 . The size of an individual calabrese plant may be limited by the availability of N and water, both of which PGPR could potentially help to ameliorate. However, our results clearly demonstrate the need for a better understanding of how individual plants respond to inoculation, perhaps with tailoring of products to specific soil types to account for differences in indigenous microbes and soil nutrients 10 . Furthermore, they also have critical importance for identification of the limiting abiotic factors that may restrict yield 26 . In any manipulative experiment, secondary statistics of size hierarchies, or size inequality, are useful as measures of plant competitive ability and for understanding how abiotic factors such as water or fertilization affect plant growth 26 , 27 . If agriculture is to become more sustainable, with less reliance on synthetic pesticides and inorganic fertilizers, then microbes such as PGPR will play a crucial role 11 . In particular, PGPR offer the potential to reduce the yield gap between conventional and organic systems 5 . We suggest that in future studies of organic systems, an analysis of size inequality should accompany experiments involving PGPR and plant growth, to inform inoculation methods, and ensure that overall yield enhancements are not at the expense of crop quality."
} | 5,192 |
30298258 | null | s2 | 7,303 | {
"abstract": "Improvements in sequencing technologies and reduced experimental costs have resulted in a vast number of studies generating high-throughput data. Although the number of methods to analyze these \"omics\" data has also increased, computational complexity and lack of documentation hinder researchers from analyzing their high-throughput data to its true potential. In this chapter we detail our data-driven, transkingdom network (TransNet) analysis protocol to integrate and interrogate multi-omics data. This systems biology approach has allowed us to successfully identify important causal relationships between different taxonomic kingdoms (e.g., mammals and microbes) using diverse types of data."
} | 174 |
33995895 | PMC8079826 | pmc | 7,306 | {
"abstract": "Graphical abstract",
"conclusion": "4 Conclusions Distinct bacterial communities can be formed from a more complex mixture of microbial isolates. Diverse bacterial species, without previously known or expected obligate relationships, were combined and put through dilution cycles. After approximately five cycles, select members prevail and form fairly stable community structures that persist through successive cycles. Further, the resulting community structures depend on the media environment used for the dilution cycles. The described approach to discovering stable, media-dependent emergent communities takes advantage of genome-defined isolates to allow for effective implementation of systems biology tools. Growth curve analyses, metaproteomics and FBA analyses were employed to identify key factors that contribute to the resulting community structure. Growth rate was identified as providing an advantage for a microbial member to maintain presence in the community. Interestingly, some of the fastest growing organisms remain in the final community structure but other rapid growing organisms do not. Further, under minimal medium conditions, a relatively slow growing organism was found to persist. Pairwise interaction measurements highlight that selective antagonistic relationships may contribute to the final structuring of the community. In order to gain a molecular-level understanding of the resulting microbial organization, FBA analyses were performed. Metabolic exchanges between organisms can be identified and likely underpin the shaping of community membership. Metaproteomic results support general findings of the FBA models. However, presently accessible FBA tools primarily account for central metabolite fluxes with cell growth as the final objective. Understanding dynamic processes at the molecular and cell population levels will be required to understand community formation, dynamics, and structure. Improved modeling capabilities, coupled with time dependent measurements and the described, scalable approach to identifying stable communities will facilitate definition of the molecular events that result in microbial community structure and dynamics.",
"introduction": "1 Introduction Bacterial communities exert significant influence over a wide range of biological processes, such as human disease, plant interactions, biogeochemical cycles, and food fermentation [1] , [2] , [3] . An important challenge exists in understanding the underlying mechanisms that contribute to how bacteria shape their community and how the resulting structure depends on distinct environmental niches [2] , [4] , [5] . The interactions between species are dynamic and community membership depends on possessing the metabolic capabilities needed to survive in a particular environment [6] . Determining these trophic exchanges and interdependent metabolic processes is difficult in natural microbial communities comprised of hundreds of members. For example, phylogenetic marker and metagenomics analyses have revealed the extreme diversity of rhizosphere bacterial communities and the complex interplay among them [7] , [8] , [9] , [10] , [11] . It has been established that the composition and activity of root bacterial communities is spatially and temporally dynamic and can be influenced by both abiotic (e.g., soil nutrients, O 2 , pH, etc.) and biotic (e.g., host and microbe-microbe) factors [12] , [13] , [14] , [15] . This complexity prevents tracking of metabolic fluxes from specific donor to acceptor strains or identifying competitive and cooperative relationships that leads to community structure [16] , [17] . Simplified synthetic microbial communities are being considered as comprehensible systems for uncovering an in-depth view of community assembly principles. These systems are able to circumvent the complexity of natural ecosystems and allow the capture of community behaviors [17] , [18] , [19] , [20] , [21] . One approach is to reduce the complexity of natural communities by selection of microbial consortia under laboratory conditions from environmental samples [22] . This top-down approach can provide an overall co-occurrence correlation network but does not assess metabolic interactions in detail as individual genome and metabolic profiles are lacking [23] . A second approach is to construct synthetic bacterial communities from the bottom-up [24] . In the bottom-up approach, individual bacterial isolates are combined to give rise to a more complex microbial system where the original strains serve as sub-systems in an emergent community [25] , [26] , [27] . These easily manipulated bottom-up assemblies contribute to a promising approach for understanding interactions in natural communities [26] , [28] , [29] . Definition and characterization of each individual strain facilitates the study of potential synergistic effects in the synthetic community [25] , [30] . The metabolites driving interspecies interactions can be determined and modeling of the metabolite exchange is possible [16] , [19] , [29] . This bottom-up approach can be used to experimentally select and investigate stable microbial community assembly [22] . The interplay among bacterial members in a consortium can be reconstructed using community-wide genome-scale metabolic models [31] . Recent applications of computational biology and genome-scale modeling approaches to the analysis of bottom-up assembled communities is providing mechanistic insights into the dynamic interactions occurring in defined bacterial communities [17] , [23] , [25] . For example, modeling studies have been applied for understanding biodegradation and bioproductivity [32] . In these models, it is often assumed that species interact in a pairwise manner [33] , [34] . Two-species metabolic models assess cross-feeding networks and usually capture the positive interactions between the microorganisms [34] , [35] , [36] . Currently, metabolic exchanges among greater numbers of microbes are being modeled and found important for shaping community distribution [22] , [37] . Modeling these higher-order interactions will be helpful for addressing questions regarding how and why a stable bacterial community forms. In this study, we describe the formation, characterization, and modeling of synthetic bacterial communities assembled from a highly characterized, phylogenetically diverse set of selected isolates in different media environments. The aim of these efforts is to define an approach to discovering simple, reproducible microbial communities, without predefined relationships, for detailed experimental studies that allow molecular and cellular level investigations into community structure. Using this discovery-based approach, we hypothesize that different community structures will result and depend on the media environment and the cooperative and competitive characteristics of the emergent community members. Ten bacterial strains ( Table 1 ), isolated from Populus deltoides rhizosphere and with defined genome sequences, were co-cultured in either complex or minimal glucose media and serially transferred until a stable community structure formed. The resulting, reproducible system allows for understanding community assembly processes and investigation of causative molecular and cellular level events. To this end, a combination of marker gene profiling and metaproteomics characterization was carried out for tracking community structure and for gaining mechanistic insights into interactions between isolates ( Fig. 1 ). These data are complemented by growth curve analyses and pairwise interaction screens. Different stable communities assemble in these environments and the higher-order interactions among community members are investigated. To unravel potential metabolic interactions among the surviving community members, genome-scale, community-level metabolic models were constructed for predicting potential metabolic processes involved in shaping the bacterial communities. The approach of discovering new microbial community structures under laboratory-defined conditions will facilitate understanding of the formation and dynamics of natural communities and the rational design of synthetic consortia with desired biological functions. Table 1 General features of the bacterial isolates utilized for community experiments. Strain Phylogeny Genome size (bp) Number of coding sequence (CDS) G + C content (%) Reference Pantoea sp. YR343 γ-Proteobacteria 5,391,843 4985 54.5 [7] Pseudomonas sp. GM17 γ-Proteobacteria 6,866,808 6199 62.8 [7] Sphingobium sp. AP49 α-Proteobacteria 4,506,188 4280 64.1 [7] Rhizobium sp. CF142 α-Proteobacteria 6,068,985 5714 66.8 [7] Variovorax sp. CF313 β -Proteobacteria 7,510,066 7608 60.1 [7] Bacillus sp. BC15 Firmicutes 6,240,445 5413 62.9 [40] Caulobacter sp. AP07 α-Proteobacteria 5,615,958 4915 68.9 [7] Duganella sp. CF402 β -Proteobacteria 11,048,459 9632 61.9 [39] Streptomyces mirabilis YR139 Actinobacteria 5,742,731 5635 34.8 [39] Paraburkholderia sp. BT03 β -Proteobacteria 11,452,267 11,227 70.3 [38] Fig. 1 Overview of experimental design for the bottom-up assembly of stable communities utilizing defined bacterial isolates.",
"discussion": "3 Results and discussion 3.1 Stable community structure in minimal and complex media Stable microbial communities were formed by serial transfer of batch cultures containing a mixture of ten, phylogenetically diverse bacterial strains derived from the Populus rhizosphere ( Table 1 ). These strains represent phyla that are abundant in the rhizosphere of plants [43] , [60] . The use of batch cultures allows for effective exchange of metabolites and the preparation of samples for analytical measurements. The component strains’ genomes are sequenced and comprise three α-Proteobacteria, three β-Proteobacteria, two γ-Proteobacteria, one Firmicute and one Actinobacterium. These batch cultures were subsequently transferred to fresh medium every 48 h and the relative proportion of each member was analyzed by 16S rRNA gene amplicon Illumina sequencing and quantitative metaproteomics. They are in general agreement regarding the trends in membership of the stable microbial communities that are formed in the two different media environments ( Fig. 2 ). In both environments, the bacterial diversity decreases from the initial inoculation, and both measurement approaches show changes in the relative proportions of the bacterial strains that settle into a similar distribution beginning with approximately the fifth dilution cycle (see Supplemental Fig. S1 ). This observation is consistent with the expectation that competition for local resources will reduce the genotypic diversity within a bacterial community [61] . In the MOPS minimal medium, four strains consistently dominate in abundance and stably coexist starting with passage No. 4. Organismal abundance trends in R2A show a fluctuation in organism relative abundances occurring between passages two and five that substantially alters the abundance of several members of the community until a community stably coexists starting with passage No. 10. Fig. 2 Analysis of the relative abundances of the 10 bacterial strains after sequential passages in MOPS minimal and R2A complex media. The relative abundances of each bacterial strain in the community are based on A) 16S rRNA gene amplicon sequencing results and B) metaproteomic results. Passage 0 represents the end of the first growth cycle after the initial inoculation. Each bar is a replicate, with three replicates per passage. The numbers shown on the bottom represent the passage number for those samples. In general, 16S rRNA gene amplicon sequencing and metaproteomics results provide similar estimates of the relative microbial abundance distribution for each measured passage in the minimal MOPS medium but differ in the rich R2A medium. In MOPS medium, Pseudomonas sp. GM17 is the most abundant member along with three other strains, Variovorax sp. CF313, Rhizobium sp. CF142 and Sphingobium sp. AP49, that persist in consistent proportions ( Fig. 2 ). Estimates of organism proportion in the R2A medium, on the other hand, are quite different between the two approaches. For 16S rRNA gene amplicon sequencing, three members dominate in abundance when grown in R2A complex medium, and Pantoea sp. YR343 is the member with highest content after 15 passages ( Fig. 2 A). The less abundant strains in R2A medium are Pseudomonas sp. GM17 and Sphingobium sp. AP49. Metaproteome analysis reveals the same dominant members, albeit the relative abundance of Pantoea sp. YR343 is strikingly different ( Fig. 2 B). Additionally, there is a notable abundance of Bacillus sp. BC15 not seen in 16S rRNA gene amplicon sequencing data. These differences can be attributed to measurement distinctions that result from using either a DNA- or protein-based approach to assess organism proportions in microbial communities. For example, examination of the proteins expressed by Bacillus sp. across the passages reveals an abundance of sporulation-related processes. Spores are notoriously challenging to measure by 16S rRNA gene amplicon sequencing because they are difficult to lyse, which negatively impacts DNA extraction efficiency [62] . Selective PCR amplification before amplicon sequencing may be another potential source of bias, since the designation of a perfectly matching universal primer is not possible [63] , [64] . Nevertheless, there is clearly a benefit in using both approaches. For instance, the observed differences in the relative abundance of Pantoea sp. YR343 is likely because 16S rRNA gene amplicon sequencing measures DNA from viable cells as well as ‘relic’ DNA (i.e., DNA from dead cells), whereas metaproteomics is a more accurate estimate of biomass from viable, functioning cells [59] , [65] , [66] . Assimilating the results between these two measurements suggests that Pantoea sp. YR343 may experience a population dieback event prior to the timepoint of sampling. 3.2 Experimentally determined growth rates for the individual bacterial isolates Individual microbial growth rates can impact interactions within a community [33] , experimentally determined growth rates of the 10 individual strains were obtained for both media and show substantial differences. In the MOPS medium, strain Pseudomonas sp. GM17 has the highest growth rate (0.463 h −1 ) among the group ( Fig. 3 A). It is also the dominant strain in the community growth experiment ( Fig. 2 ). In the same medium, Sphingobium sp. AP49 (0.430 h −1 ) has a maximum growth rate that is similar to that observed with the Pseudomonas sp. GM17 and also maintains a presence in the community. Compared to these two organisms, Paraburkholderia sp. BT03 (0.379 h −1 ), Variovorax sp. CF313 (0.362 h −1 ), Pantoea sp. YR343 (0.346 h −1 ), Duganella sp. CF402 (0.333 h −1 ) and Caulobacter sp. AP07 (0.296 h −1 ) have slightly lower maximum growth rates, but only Variovorax sp. CF313 persists in the community. Three organisms, Rhizobium sp. CF142 (0.082 h −1 ), Bacillus sp. BC15 (0.055 h −1 ) and Streptomyces mirabilis YR139 (0.039 h −1 ), have considerably lower growth rates when compared to the other organisms. Interestingly, despite having a relatively slow growth rate in the MOPS medium, Rhizobium sp. CF142 prevails as a dominant community member. Fig. 3 Comparison of experimentally determined growth parameters with FBA model predicted growth. A) Relative growth rates and lag times for the individual strains in MOPS medium; B) relative growth rates and lag times for the strains in R2A medium. Each data column of experimental results represents the mean and error bars are the standard deviation over three parallel experiments. In R2A medium, there is a strong correlation between the growth rates of individual strains and community membership. Not surprisingly, the growth rates of all of the selected microbes are higher in this rich medium when compared to growth rates observed in MOPS medium. Overall, Pantoea sp. YR343 (0.656 h −1 ), Pseudomonas sp. GM17 (0.563 h −1 ) and Sphingobium sp. AP49 (0.479 h −1 ) are relatively fast growers in R2A and are significant components of the emergent community. In contrast, Paraburkholderia sp. BT03 (0.472 h −1 ), Bacillus sp. BC15 (0.442 h −1 ), and Duganella sp. CF402 (0.411 h −1 ), despite showing growth rates only slightly lower than Sphingobium sp. AP49, do not maintain a significant presence in successive community growth cycles ( Fig. 3 B). Comparison of the individual member growth rates to the observed community composition indicates complex relationships among the community members. On the one hand, the presence of fast-growing microbes in the emerging community composition is expected and consistent with observations in other systems. The competitive exclusion principle predicts that when the bacterial members in a community compete for the same resources, the fitter strain will outcompete the other members [6] and has a greater opportunity for recurrent colonization that can allow for persistence in the community [67] . Yet, several fast-growing microbes are absent in the final passages of the community and slow growers are present. Often, slower growing microbes persist in communities and allow for maintenance of diversity. Here, the well-mixed conditions promote exchange of metabolites during bacterial growth and prevents spatial structuring that often allows for maintenance of slower growing organisms [22] , [68] . Therefore, bacterial strains must collectively adjust their behavior and selectively cooperate in order to emerge into a community with stable proportions. The resulting supportive associations among community members likely proceeds through metabolic interactions such as the cross-feeding of essential nutrients [61] . One particularly interesting observation to this point is the persistence of Rhizobium sp. CF142 in the MOPS minimal medium community. When grown in monoculture in MOPS minimal medium, this strain has a slower growth rate and lower final OD when compared to the other strains. It could be speculated that metabolic cooperation emerges in these mixed microbial communities increasing the fitness of strain CF142, potentially by providing some missing nutrients made available by other community members [69] , [70] . This is consistent with the existence of both cooperative and competitive associations between the component members that lead to the formation of a community network structure [71] . 3.3 FBA-based growth prediction models for individual microbes To gain a better understanding of individual strain metabolism and the potential variety of exchanged metabolites, FBA models of each of the component microbes were generated. By estimating the reaction fluxes to generate biomass constituents, the growth rate of the microorganism can be predicted [72] . The maximum relative growth rates of the ten bacterial isolates in R2A and MOPS media are displayed in Fig. 3 . In these models, the objective value of growth is determined by setting a maximum glucose uptake flux of 100 mmol/g DCW/h. Among the 10 Populus bacterial isolates in MOPS medium, Pseudomonas sp. GM17 has the highest predicted and actual growth rate and was used for normalization. Similarly, among the 10 Populus bacterial isolates grown in R2A medium, Pantoea sp. YR343 has the maximal predicted and actual growth rate and was used as the reference standard. In MOPS medium, the hierarchy of relative predicted growth rates generally matches the experimentally observed growth rates. In this medium, the maximal growth rate predictions for the slowest growing organisms, Rhizobium sp. CF142, Bacillus sp. BC15, and Streptomyces mirabilis YR139 are significantly overestimated and likely reflect imperfect understanding of metabolism in these organisms. When compared to the others, these organisms all have long lag phases ( Fig. 3 A) which likely reflects unknown adaptations to environmental conditions [73] . Further, these organisms may adopt different growth strategies that do not prioritize the conversion of glucose to biomass. In the case of Streptomyces mirabilis YR139, the unusual growth and morphological characteristics of this genus can contribute to experimental and predictive errors. In R2A media, growth rate predictions show a different trend. In general, the predicted relative growth rates of the slower growing organisms match the experimentally observed maximal growth rates. In contrast, growth rate predictions are poor for several of the faster growing microbes ( Fig. 3 B). In particular, growth rate predictions are significantly underestimated for Sphingobium sp. AP49 and Bacillus sp. BC15 in this complex medium. Again, this may reflect unknown limits on metabolism for these species. The FBA predicted growth rates assume ideal conditions; all nutrients in the medium are made available at the maximum uptake flux. In the growth experiments of those individual strains, the growth rate could not be ideally as high as the FBA models. In the FBA model using R2A medium, the dominant strain Pantoea sp. YR343 has the highest number of exchange reactions of nutrients ( Table S4 ), and it has more transporters based on genome annotation compared with the other community members. For Pantoea sp. YR343, the higher number of transporter genes may be related to its stronger metabolic interaction potential encoded in the genome. In the nutrient rich complex medium, Pantoea sp. YR343 is capable of utilizing a variety of nutrients which is consistent with the observation that this organism has the highest simulated growth rate. In contrast, the number of exchange reactions identified in individual FBA models carried out in MOPS medium is similar ( Table S5 ). This likely results from the minimal medium environment where nutrition is relatively limited compared with the complex R2A medium. In MOPS medium models, the medium-specific metabolic processes may override the strain-specific metabolisms, leading to much less difference in simulated growth rates among the 10 isolates when compared to models employing R2A medium [74] . The discrepancy of simulated growth rates of the same strain between MOPS and R2A media also corresponds to the experimental observation that bacteria utilize their metabolism for survival in the minimal medium, in contrast they tend to have more active growth in the complex medium [75] . 3.4 Evaluation of microbial community models 3.4.1 FBA-based growth prediction for community models FBA-based community models were assembled in order to identify microbial features that account for the observed stability of the community and for assessing the suitability of these models and their use in understanding the molecular genetic bases for the resulting community structure. Flux changes of the metabolites involved in exchange may explain the interaction mechanisms between the component members. Compartmentalized models were created using the KBase platform to allow for the community members to secrete and take up metabolites from a shared environment [17] , [51] . To create these models, the genome-scale metabolic models of the primary constituents of the final community were combined using the relative ratio of the community members determined from the 16S rRNA gene marker data for the final passage. These final, persistent strains are considered the best-performing species and as community drivers that affect dependent species and community organization [76] , [77] . The FBA-based community models predict altered growth rates for the constituent members when compared to their individual growth rates. In MOPS medium, the dominant strain Pseudomonas sp. GM17 is predicted to have the fastest growth among the consortium members. However, this growth rate is lower when compared with its individual FBA model ( Fig. 3 A). This is most likely due to competition with other community members in this limited nutrient environment and thus a lower growth rate is not surprising. In comparison, the three other major strains in this consortium have significantly lower, but similar, predicted growth rates when compared to Pseudomonas sp. GM17 as well as to their individual models ( Fig. 3 A). Altered growth rates can result from competitive interactions and have been observed in other studies using minimal media for microbial community assembly [78] , [79] , [80] . These prior studies indicate that the metabolic capability associated with each strain can influence community composition and lead to the survival of the strongest competitors [22] , [81] . When using R2A medium in the community metabolic model, the dominant strain Pantoea sp. YR343 has a much higher theoretical growth rate than that of its individual model. This suggests that it is beneficial for Pantoea sp. YR343 to grow in the presence of the other two members in the community. The higher predicted growth rate for Pantoea sp. YR343 in the community model may relate to the organism’s broad spectrum of transport reactions [82] . In a community, the bacterial members that are metabolic generalists have a better chance of survival compared to those that are adapted to specific substrates [22] , [83] . For both media, the community-FBA models support the dominant presence of a fast-growing member and suggest that members of the community influence each other’s growth rate. 3.4.2 FBA-based predictions of metabolic exchanges Limiting the uptake flux of carbon is the foundation of a constraint-based FBA model and allows prediction of the distribution of metabolic fluxes that depend on the medium [84] . Without a carbon source uptake limitation, metabolite exchanges between community members will be overestimated. Therefore, a limitation of carbon uptake flux was added to the FBA models for the communities modeled in either MOPS or R2A medium. Calibration curves relating carbon source uptake flux and growth rate objective value were established for each community member, and the individual carbon uptake flux limitation of each strain was calculated based on the experimental growth rate. The carbon uptake limit for a community FBA model was calculated as the sum of the relative ratios of the individual microbial components multiplied by their individual carbon uptake rates. The extracellular metabolites involved in interspecies exchanges can be predicted by the community FBA models [51] . Fig. 4 illustrates the predicted metabolite exchanges among the four component members in MOPS medium. The dominant strain, Pseudomonas sp. GM17 is predicted to supply more metabolites to the other community members than it receives. Key among these predicted metabolites are amino acids, sugars and purine derivatives. In the minimal medium environment, cellular building materials must be wholly prepared from the glucose carbon source, or, in a community environment, scavenged from the excretions of other microbes. Accessing excreted metabolites may be vital for the maintenance of those members with minor content in the community. In turn, Pseudomonas sp. GM17 may rely on the production of metabolites from these minor community members as evidenced by its lower growth rate in the community FBA model when compared with its individual FBA model. The metaproteomics results provide an opportunity to assess whether the appropriate pathways are represented. The percentage of the detected enzymes in the metabolite supplier is related to the total protein involved in the KEGG pathways for the exchanged metabolites. Considering the dominant microbial component Pseudomonas sp. GM17, there is high representation of the enzymes related to the metabolites supplied by this organism ( Fig. 4 ). In contrast, enzymes involved in the preparation of metabolites shared from the other organisms are low. This lower representation is a consequence of the relatively lower number of proteins detected from these organisms relative to Pseudomonas sp. GM17. Fig. 4 Predicted metabolite exchange among the four dominant members of the microbial community formed in MOPS medium as proposed by the community FBA model. The percentage of detected enzymes by metaproteomics analyses is shown in parentheses. In the community model using R2A medium, metabolite exchange can also be predicted. Here, the dominant strain Pantoea sp. YR343, based on the 16S rRNA marker data, is expected to export a greater range of metabolites than it receives from the other two members of the community ( Fig. 5 ). Organic acids, purine derivatives and biogenic amines are predicted to be excreted and support the growth of the other community members. The community FBA model predicts a faster growth rate for Pantoea sp. YR343 in the community model when compared to the individual model and this may result from access to metabolites excreted by the other community members. Again, the community metaproteomics results can be used to assess the presence of relevant pathways. The high representation of both Pantoea sp. YR343 and Pseudomonas sp. GM17 in the metaproteomics data allows detection of most of the enzymes expected to participate in the FBA-based predictions. As in the case of the minimal medium environment, representation of the minor microbial components is poor, and confirmation of relevant components of metabolism is much lower when compared to the major microbial components. Fig. 5 Predicted metabolites exchange among the three dominant members of the microbial community formed in R2A medium as proposed by the community FBA model. The percentage of detected enzymes by metaproteomics analyses is shown in parentheses. The combination of omics data with FBA modeling can aid with understanding community function and stability. Here, the metaproteomics results help support predictions of the FBA model by providing evidence for the metabolic capabilities of each community member and making them more faithful representations of the biological system being interrogated [23] . While the coverage of enzymes implicated in the exchanged metabolites is relatively high for prominent members of the community, as communities grow in membership diversity, metaproteomic sequencing depth will need to improve for the more minor members, which may disclose their survival mechanisms. Additionally, the integration of metabolomics data will be valuable for constraining FBA models and for confirming predictions of interdependencies between community members. The inherent complexity of metabolic interactions is a challenge in the modeling of microbial communities [85] . The genome-scale, mechanistic modeling provided by the FBA approach is presently insufficient to account for a large fraction of intracellular networks or assess dynamic, population level changes that likely lead to community structuring [72] . Integration with population-level dynamic models, such Lotka-Volterra (LV) [86] , [87] , [88] , or r/K selection strategies [89] may help to describe the temporal progress of species abundances and community formation processes. Dynamic FBA, which simulates the dynamics of community growth and substrate consumption in time-dependent processes, can also extend current FBA approaches to temporal changes [90] , [91] , [92] . Effective application of these tools will require new, time dependent global sampling and measurement strategies for verifying the efficacy of dynamic models. 3.5 Pairwise interaction screens While the present metabolic models help to understand interactions that support growth of a stable community, they provide only partial insight into the selection process that leads to community formation. In both tested environments, a fast-growing microbe emerges as a dominant component that is supportive of other community members. However, other relatively fast-growing microbes are out competed in early growth cycles, and their abundance fades from the composition of the community. Antagonistic interactions may be present that facilitate the community selection process. To assess this possibility, community members were screened for mutualistic, commensal, or antagonistic colony phenotypes in a pairwise interaction screen. While a majority of the interactions appear to be commensal, with no obvious phenotypes, the screen did identify both positive and negative interactions ( Table 2 ). Pseudomonas sp. GM17 cells were antagonistic to the growth of the majority of the other community members, while Pantoea sp. YR343 demonstrated positive interactions with several strains indicating an obvious role for competition, antimicrobial production and or beneficial interactions in microbial community selection, structure and stability. Table 2 Pairwise interaction screen results. Strain designations across top of table indicate lawn of microbes spread on R2A agar plate and designations on left indicate cells spotted on center of lawn. + indicates a positive interaction while - indicates an antagonistic interaction. Empty cells indicate no obvious colony phenotype change. Genus Strain YR343 GM17 AP49 CF142 CF313 BC15 AP07 CF402 YR139* BT03 Pantoea YR343 + + ND Pseudomonas GM17 – – – – – – ND – Sphingobium AP49 – + ND Rhizobium CF142 + ND Variovorax CF313 – ND – Bacillus BC15 – ND Caulobacter AP07 – ND Duganella CF402 + + ND Streptomyces YR139 – – – ND Paraburkholderia BT03 – ND * Due to the growth characteristics of Streptomyces sp. YR139, a lawn of bacteria could not be prepared and resulted in no data (ND). Interestingly, although the growth of strain AP49, CF313 and CF142 were inhibited by strain GM17 in pairwise interaction screens, these strains still co-existed with strain GM17 during the 10-member community cultivation. This may be attributed to intertwined metabolic interactions among these four strains, in which the beneficial effect from the metabolites in a shared extracellular environment for growth overwhelms antagonistic effects by strain GM17. Alternatively, different experimental conditions may account for the unexpected result. The multi-member community was grown in a well-mixed liquid medium condition which is different from the static agar plate conditions of the pairwise interaction screen and may account for the observed discrepancy [93] . The integration of temporal modeling of microbial communities with time course of community composition within the growth period in a passage can be promising to accommodate these antagonistic interactions and the dynamic processes that shape community structure."
} | 8,702 |
34210981 | PMC8249394 | pmc | 7,307 | {
"abstract": "Biofilm and nitrogen fixation are two competitive strategies used by many plant-associated bacteria; however, the mechanisms underlying the formation of nitrogen-fixing biofilms remain largely unknown. Here, we examined the roles of multiple signalling systems in the regulation of biofilm formation by root-associated diazotrophic P. stutzeri A1501. Physiological analysis, construction of mutant strains and microscale thermophoresis experiments showed that RpoN is a regulatory hub coupling nitrogen fixation and biofilm formation by directly activating the transcription of pslA , a major gene involved in the synthesis of the Psl exopolysaccharide component of the biofilm matrix and nifA , the transcriptional activator of nif gene expression. Genetic complementation studies and determination of the copy number of transcripts by droplet digital PCR confirmed that the regulatory ncRNA RsmZ serves as a signal amplifier to trigger biofilm formation by sequestering the translational repressor protein RsmA away from pslA and sadC mRNAs, the latter of which encodes a diguanylate cyclase that synthesises c-di-GMP. Moreover, RpoS exerts a braking effect on biofilm formation by transcriptionally downregulating RsmZ expression, while RpoS expression is repressed posttranscriptionally by RsmA. These findings provide mechanistic insights into how the Rpo/Gac/Rsm regulatory networks fine-tune nitrogen-fixing biofilm formation in response to the availability of nutrients.",
"introduction": "Introduction The term ‘biofilm’ can be defined as a community of microbes adhering to biotic or abiotic surfaces that is protected from environmental stresses by a self-produced extracellular matrix 1 , 2 . The extracellular matrix, often referred to as extracellular polymeric substances, is composed of exopolysaccharides, proteins and extracellular DNA present in various concentrations depending on the bacterial species 3 , 4 . The biofilm state provides potential advantages over the planktonic state, including increased resistance to antimicrobial agents, protection from environmental stresses, and improved adaptation to nutrient deprivation 5 . Numerous investigations in recent decades have demonstrated that bacterial biofilm formation is a sequential process governed by complex regulatory networks that differ from one bacterial species to another 1 , 6 . It is now well accepted that microbial biofilms are the most widely distributed and predominant mode of life on Earth, influencing our lives tremendously in both positive and negative ways 6 – 9 . In general, as established in the model bacterium Pseudomonas aeruginosa , biofilm development usually begins with attachment to a surface, followed by microcolony formation and production of the extracellular matrix responsible for the biofilm architecture 10 – 14 . Biofilm formation has been studied intensively in the genus Pseudomonas , with an emphasis on genetic elements and molecular mechanisms; Gac/Rsm, c-di-GMP signalling and quorum-sensing (QS) pathways were reported as the main mechanisms leading to biofilm formation 15 , 16 . The Gac/Rsm signalling pathway involves the GacS/GacA two-component regulatory system, the RNA-binding protein RmsA, and its cognate regulatory non-coding RNAs (ncRNAs) 17 , 18 . The GacS/GacA two-component system activates the transcription of one or several genes for Rsm ncRNAs, which contain multiple GGA motifs in exposed stem loops of their predicted secondary structures 19 . The GGA motifs allow Rsm ncRNAs to bind the RNA-binding proteins that act as global posttranscriptional repressors, e.g., CsrA (in Escherichia coli ) and RsmA (in P. aeruginosa ), controlling important cellular processes, such as secondary metabolism (e.g., metabolism of pyocyanine or the QS signal N-butyryl-homoserine lactone in P. aeruginosa ), motility, and biofilm formation 17 , 20 . RsmA specifically recognises and binds to conserved GGA motifs in the 5′-untranslated region (5′-UTR) of target mRNAs, thereby preventing ribosome access and protein translation 17 , 21 . RsmA controls biofilm formation through direct repression of various target genes, such as pslA (involved in the synthesis of the exopolysaccharide Psl) and sadC (involved in c-di-GMP synthesis) 22 , 23 . As a key biofilm regulatory molecule, the second messenger c-di-GMP is synthesised by diguanylate cyclases (DGCs) that bear a GGDEF domain and is degraded by phosphodiesterases (PDEs) that harbour EAL or HD-GYP domains. P. aeruginosa encodes several DGCs and PDEs; for example, WspR/SadC/RoeA (DGC) and RocR/BifA (PDE), are absent in the P. stutzeri A1501 genome, except for SadC and BifA, which modulate the level of c-di-GMP and influence ‘surface-associated behaviours’ by controlling polysaccharide syntheses 16 , 24 – 27 . The P. aeruginosa biofilm matrix contains several polysaccharide components, including alginate, pellicle (Pel) and Psl exopolysaccharides 28 . It has been shown that pslA is the first gene in the psl operon, which comprises 15 cotranscribed genes that are involved in the synthesis of Psl 29 . Although current data relating to the roles of Psl are limited, Psl is a critical component of the P. aeruginosa biofilm matrix, which functions as a scaffold, holding biofilm cells together to initiate biofilm development 30 . In addition, evidence demonstrates that biofilm formation is controlled positively by RpoN but negatively by RpoS, suggesting global antagonism between RpoN and RpoS, although there are contradictory reports 31 – 35 . Microbial biofilms are common on plant surfaces and have been associated with phytopathogenic infections and colonisation by nitrogen-fixing rhizobacteria 36 , 37 . Because of dynamically fluctuating conditions in the rhizosphere, the ability of diazotrophic bacteria to form nitrogen-fixing biofilms may confer many ecological advantages and thereby facilitate their physiological and metabolic adaptation to successfully survive in the rhizosphere, a nitrogen-limited environment. An early study compared biofilm formation by a nitrogen-fixing strain of Klebsiella pneumoniae with that of two other members of Enterobacteriaceae, Salmonella enteritidis and E. coli , and showed that the nitrogen-fixing strain formed the densest and most metabolically active biofilms 38 . Many nitrogen-fixing bacteria, such as those of the genera Rhizobium , Gluconacetobacter and Azospirillum , produce biofilms containing various exopolysaccharides 39 – 42 . For instance, Sinorhizobium meliloti produces two symbiosis-promoting exopolysaccharides, succinoglycan and galactoglucan, which function in host specificity and participate in early stages of a host plant infection, biofilm formation, and, most importantly, protection from environmental stresses 43 – 45 . Azospirillum cells are also capable of forming biofilms on both abiotic surfaces and in association with host plants 46 . Previous studies have demonstrated that two response regulator proteins, TyrR and FlcA, were found to be involved in the transcriptional regulation of biofilm formation by A. brasilense Sp7 via the production of capsular polysaccharides 42 , 47 . The root-associated bacterium P. stutzeri A1501 is a rare example of a Pseudomonas strain with nitrogen fixation ability 48 . P. stutzeri A1501 can survive in the soil, colonise the root surface, and endophytically invade the root tissues of host plants. During evolution, A1501 acquired a nitrogen fixation island with a nif -specific regulatory system from a diazotrophic common ancestor 48 . Similar to many other Pseudomonas species, the nitrogen regulatory cascade in A1501 comprises the AmtB–GlnK–NtrBC-RpoN global nitrogen regulation proteins and a set of regulatory ncRNAs that control the expression of nif genes and the consequent optimal nitrogen fixation in response to nutrient stress 49 – 52 . Comparative genomics analysis showed that A1501 does not possess the well-known QS systems and does not produce alginate, but it contains genes possibly involved in cellulose biosynthesis and an incomplete psl operon 4 , 48 . It was previously shown that a nonpolar mutation of the fleQ gene, encoding FleQ (the main regulator of flagella synthesis), impaired motility and root colonisation but enhanced biofilm formation by P. stutzeri A1501 53 . Additionally, Wang et al. investigated the effect of physiological conditions on the formation and architecture of nitrogen-fixing biofilms by P. stutzeri A1501 41 . However, the composition of the polysaccharide matrix remains unknown. To date, studies on biofilm formation by nitrogen-fixing rhizobacteria have focused on ecological, physiological and architectural analyses. Despite its importance to microbial adaptation and survival, there is surprisingly little information about the genetics of nitrogen-fixing biofilm formation. In this work, physiological conditions leading to nitrogen-fixing biofilm formation by the root-associated bacterium P. stutzeri A1501 were further investigated. We found that conditions favouring biofilm formation differ between diazotrophic and non-diazotrophic P. stutzeri strains, although both strains contain the same set of regulatory genes involved in biofilm formation in other systems. Thus, we systematically characterised genetic elements and molecular mechanisms involved in nitrogen-fixing biofilm formation. Genome-wide identification of putative genes involved in biofilm formation and mutant construction led to the identification of a complex regulatory circuitry involving the alternative sigma factors RpoN and RpoS and the Gac/Rsm regulators, and to the proposal of a model that integrates multiple levels of positive and negative regulation.",
"discussion": "Discussion Numerous studies have established that regulatory circuits governing the transition from planktonic to biofilm lifestyles are very complex and differ between Pseudomonas species, although common regulatory mechanisms such as the c-di-GMP signalling and Gac/Rsm pathways exist. On the other hand, the available literature on the regulatory mechanisms underlying biofilm formation by nitrogen-fixing bacteria is still very scarce. Here, we aim to fill this knowledge gap by elucidating the complex mechanisms for fine-tuning nitrogen-fixing biofilm formation. In view of the data reported, we propose that multiple signalling systems regulate nitrogen-fixing biofilm formation in the rhizosphere bacterium P. stutzeri A1501 (as depicted in Fig. 6 ), including the well-studied Gac/Rsm pathway at the posttranscriptional level, RpoN-driven positive regulation at the transcriptional level, and a RpoS-mediated repression circuit at both levels. The Gac/Rsm pathway is generally considered the main mechanism controlling biofilm formation in non-diazotrophic Pseudomonas 15 , 17 . Indeed, we have shown that deletions of each of the A1501 gac/rsm genes can positively or negatively affect biofilm formation, but rpoN , by controlling the transcription of nifA and plsA , is the only gene whose inactivation resulted in the poorest biofilm and a Nif-minus phenotype. These results suggest that RpoN-driven positive regulation at the transcriptional level is one of the key mechanisms underlying nitrogen-fixing biofilm formation, which may override the effect of the Gac/Rsm pathway in diazotrophic P. stutzeri . Fig. 6 Proposed regulatory model for the P. stutzeri RpoN/RpoS/Gac/Rsm signal transduction systems controlling nitrogen-fixing biofilm formation at multiple levels in response to the availability of nutrients. In this model, RpoN plays a central role and may be considered a ‘hub’ to bridge nitrogen fixation and biofilm formation by activating the transcription of the pslA and nifA genes under nitrogen-deficient and carbon-sufficient conditions. RsmZ was upregulated rapidly at the early stage of biofilm formation and then downregulated remarkably during biofilm maturation, thereby acting as a potent trigger for the initiation of biofilm formation. During biofilm development, RpoS exerts a braking effect on biofilm formation by transcriptionally downregulating RsmZ expression at the mature stage; this effect is restrained by RsmA at the early stage, thereby resulting in a novel repression circuit. Additionally, RpoN likely acts as a repressor of the Gac/Rsm pathway, markedly increasing the complexity of the regulatory circuitry. Arrows and T-shaped bars indicate positive and negative regulation, respectively. Broken lines indicate direct or indirect regulations for which evidence exists but that need to be studied in further detail. The black tail arrow indicates biochemical conversion reactions. The involvement of c-di-GMP in the biosynthesis of various polysaccharides has not been demonstrated experimentally, as marked by the punctuated tail arrow. For details, refer to the text. An additional level of complexity is added to this regulatory system by the presence of two structurally and functionally similar ncRNAs, RsmY and RsmZ. The presence of multiple ncRNAs with structural similarity was reported in other systems, e.g., RsmX, RsmY, and RsmZ in P. fluorescens 60 and RsmW, RsmY, and RsmZ in P. aeruginosa 58 . These regulatory ncRNAs show similar secondary structures with numerous unpaired GGA motifs that act to sequester RsmA proteins from their targets, suggesting possible functional redundancy 20 . Since the effectiveness of ncRNA regulation is directly related to ncRNA abundance relative to their mRNA targets, this redundancy has been proposed to permit a more efficient and precise regulatory response by providing additional possibilities for integrating various signals into complex networks 18 . In the case of P. stutzeri A1501, a double rsmY/rsmZ mutation caused the same phenotypic effects on biofilm formation as those observed in the gacA mutant (Supplementary Table 1 ), suggesting that no additional Rsm ncRNAs participate in the activation of biofilm formation via the Gac/Rsm cascade in A1501. Moreover, the transcription rates of the rsmY and rsmZ genes in A1501 are clearly distinct; rsmZ is expressed at ~100-fold higher levels than rsmY under biofilm growth conditions, and the level of RsmZ is very high during biofilm growth compared to planktonic growth. In addition, a quantitative assessment by ddPCR demonstrated that RsmZ showed a biofilm stage-dependent pattern of expression with a significant increase during early stages of biofilm formation caused by transcriptional activation by GacA, which was followed by a decrease in mature biofilms. We thus propose that RsmZ rather than RsmY acts as a signal amplifier to trigger the phenotypic switch from the planktonic mode to the biofilm mode of growth. Although the biological role of RsmY is unclear at this stage, a very strong induction of rsmY expression by glucose was observed (Fig. 4o ), implying that this ncRNA may be required for glucose-related metabolism in A1501. We also observed that single mutations of rsmY and rsmZ limit biofilm formation by A1501 but decrease and increase nitrogenase activities, respectively. These results suggest that both ncRNAs have overlapping functions in the regulation of biofilm formation but distinctive roles in the regulation of nitrogenase activity. Our results from ddPCR experiments quantitatively show that RpoS is a mature stage-induced protein whose expression is downregulated by RsmA at the early stage of nitrogen-fixing biofilm formation. Similarly, Huertas-Rosales et al. identified rpoS as a target of Rsm proteins in RIP-seq experiments as an indication that RpoS regulation by Rsm proteins is direct 61 . In addition, RpoS was also found to be negatively regulated by RsmA in P. protegens CHA0 62 . This led us to speculate that RpoS contributes to the significant reduction in RsmZ levels when RsmZ is not needed at a high level in mature biofilm cells, while RsmA posttranscriptionally decreases RpoS expression and prevents the repression of RsmZ exerted by RpoS when RsmZ is most needed at a high level in early-stage biofilm cells. In addition to RpoS, we also found that RpoN monitors global changes in gene expression that may lead to more complex effects on nitrogen-fixing biofilm formation. For example, rpoN mutation significantly increased the expression of rsmZ in nitrogen-fixing biofilm cells, implying that RpoN likely acts as a repressor in the regulation of the RsmZ level. At least part of this effect might be mediated by GacA, as described previously in P. aeruginosa 55 . This appears to contradict the enhanced expression of RsmZ in nitrogen-fixing biofilm cells. However, the stronger effect of the RsmZ mutation on biofilm formation under NH 4 + -rich conditions than under NH 4 + -deficient conditions suggests that additional repression of RpoN ensures accurate and economical but not consistently high expression of RsmZ since the Gac/Rsm pathway is not the dominant player in nitrogen-fixing biofilm formation. The initiation of biofilm formation in P. aeruginosa has been correlated with high intracellular levels of c-di-GMP 16 . In general, high internal levels of c-di-GMP induce the production of adhesins and extracellular matrix components, which enable bacteria to form biofilms, whereas low c-di-GMP levels lead biofilm bacteria into dispersal to shift to a planktonic mode of growth 24 . The Gac/Rsm cascade in P. aeruginosa is genetically linked to c-di-GMP through SadC, whose production is repressed by RsmA. We also observed a similar connection, but deletion of the sadC gene resulted in a strain that is partially defective in biofilm formation and c-di-GMP synthesis. This means that SadC likely contributes some but not all of the c-di-GMP under the conditions tested. Therefore, we can further infer that at least one other DGC in the A1501 genome can produce c-di-GMP. Indeed, the exact mechanism underlying c-di-GMP synthesis and biofilm formation in A1501 remains to be elucidated. Phylogenetically close members of the Pseudomonas genus produce a wide diversity of exopolysaccharides, such as cellulose, Psl, and Pel 3 . The Psl polysaccharide, which is composed of mannose, glucose and rhamnose, was first described in P. aeruginosa 63 . Although research on Psl polysaccharides has been mostly conducted in P. aeruginosa , a number of psl gene clusters have been identified in several Pseudomonas strains 4 and recently the existence of a psl -like gene cluster has been reported in some environmental non-aeruginosa Pseudomonas species 64 . Furthermore, two P. fluorescens strains isolated from rotted bell pepper, were previously described to produce an exopolysaccharide composed of mannose, rhamnose, and glucose substituted with pyruvate and acetate 65 . Although the exact composition of the PlsA-dependent polysaccharide, tentatively referred to as the Psl-like exopolysaccharide, is not yet established, the analysis of the glycosyl residues present in a pslA mutant suggests that the A1501 Psl-like exopolysaccharide contains mannose and galactose since both sugars were not found in the mutant (Table 2 ). From this analysis, it can be deduced that A1501 Psl differs from P. aeruginosa Psl, which does not contain galactose 28 . As glucose is the main sugar produced by the pslA mutant, it is likely that A1501 produces cellulose, in agreement with the presence of a cluster of genes in its genome that are similar to the cellulose biosynthesis genes of P. putida KT2440 48 . In the most recent review, Herredia-Ponce et al. 63 stress the fact that the differences in polysaccharide composition depending on growth conditions may reflect better adaptation to specific environments due to the differential evolution that occurs in different niches. In the case of P. aeruginosa , the Psl exopolysaccharide is known to be a key element at the early stage of biofilm formation and is regulated transcriptionally by RpoS 22 , 30 . Unlike what was observed in P. aeruginosa , we found that in A1501, the PlsA-dependent exopolysaccharide is essential for biofilm formation under nitrogen fixation conditions but not under nitrogen-sufficient conditions, in agreement with the fact that pslA transcription is RpoN-dependent. In addition to playing a major structural role in biofilms, Psl was further shown to have a signalling role in stimulating two DGCs, SiaD and SadC, to produce more of the intracellular second messenger molecule c-di-GMP 66 . A Psl-mediated increase in c-di-GMP was observed to result in two- to threefold higher levels of pslA transcripts, ultimately increasing the production of Psl itself and forming a unique positive feedback regulatory circuit 66 . These observations led us to speculate that PslA may be a rate-limiting enzyme of Psl synthesis. To experimentally address this possibility, pL pslA was introduced into A1501, generating a strain overexpressing PslA. As predicted, this overexpression strain produces much more Psl than the wild-type strain (Fig. 3c ). Bacteria in biofilms are surrounded by an extracellular matrix, which can account for up to 90% of the biofilm biomass and create a microenvironment favourable for protecting cells against various stresses 3 , 67 . Biofilms may provide especially suitable conditions for nitrogen fixation, as this process is extremely sensitive to oxygen and rapidly inhibited by ammonia. An early study reported that the production of exopolysaccharides under N-limiting conditions may be a survival mechanism favouring the exclusion of oxygen and increasing nitrogenase activity 68 . In addition, biofilm formation enables A1501 to fix nitrogen under aerobic conditions by forming EPS-encased cysts to protect nitrogenase from oxygen 41 . In accordance with these previous results, we found that biofilm formation was enhanced under nitrogen-deficient and carbon-sufficient conditions, which favour nitrogen fixation. Interestingly, we also observed that biofilms displayed significant nitrogenase activity at a concentration of NH 4 + that completely abolished the nitrogenase activity of planktonic cells. Nitrogen-fixing bacteria occur predominately in the rhizosphere, where carbon-rich root exudates can support the energy demands of the nitrogen fixation process, while microbial cell densities and microbial activities are the greatest, making nitrogen a key modulator of survival and competitiveness. The colonisation of the root rhizosphere is an essential step in the establishment of efficient nitrogen-fixing associations, and thus, understanding the mechanism of biofilm formation is of major interest. In the present work, we found that conditions favouring biofilm formation differ between the diazotrophic and non-diazotrophic P. stutzeri strains, although both strains contain the same set of Gac/Rsm and c-di-GMP signalling pathway genes, reflecting the differential evolution of their regulatory networks due to different physiologies and niches. We hypothesised that variations in biofilm phenotypes could be due to differences in transcriptional regulation. However, we found no significant differences in the putative promoter sequences of the genes listed in Fig. 2b between the two strains, suggesting that the mechanism that causes the biofilm phenotypes of the two strains to differ is much more complex than we initially believed. In addition, it is not surprising that with evolutionary optimisation in the rice rhizosphere, A1501 has evolved sophisticated regulatory networks to respond to multiple environmental cues and adapt to the environmental conditions of the rhizosphere. Of particular note is that RpoN, an alternative sigma factor typically associated with general nitrogen responses in bacteria, was found to act as a critical regulatory hub to activate the transcription of pslA and nifA , consequently forming a novel regulatory link between nitrogen fixation and biofilm formation. This regulation is probably more direct and efficient than the Gac/Rsm regulatory cascades widely found in Pseudomonas , and is likely advantageous, especially when diazotrophs face competition from other species in nitrogen-limited environments, such as the rhizosphere. To our knowledge, this is the unique example of multiple regulatory networks governing the transition from the planktonic mode to the nitrogen-fixing biofilm mode, which may contribute to diazotrophic P. stutzeri being highly adaptable to nitrogen-poor environments and have implications for the control of biofilm-related interactions between diazotrophs and host plants. Our results provide a basis for understanding a regulatory mechanism including RpoN, RpoS, Gac and Rsm regulators that underlies nitrogen-fixing biofilm development and may be applicable to various diazotrophic species. As nitrogen-fixing bacteria are found ubiquitously in most ecosystems and widely used as biofertilizers worldwide, our systematic study of nitrogen-fixing biofilms will be of both ecological and biotechnological importance."
} | 6,282 |
27197212 | PMC4889971 | pmc | 7,308 | {
"abstract": "Evolutionary innovation must occur in the context of some genomic background, which limits available evolutionary paths. For example, protein evolution by sequence substitution is constrained by epistasis between residues. In prokaryotes, evolutionary innovation frequently happens by macrogenomic events such as horizontal gene transfer (HGT). Previous work has suggested that HGT can be influenced by ancestral genomic content, yet the extent of such gene-level constraints has not yet been systematically characterized. Here, we evaluated the evolutionary impact of such constraints in prokaryotes, using probabilistic ancestral reconstructions from 634 extant prokaryotic genomes and a novel framework for detecting evolutionary constraints on HGT events. We identified 8228 directional dependencies between genes and demonstrated that many such dependencies reflect known functional relationships, including for example, evolutionary dependencies of the photosynthetic enzyme RuBisCO. Modeling all dependencies as a network, we adapted an approach from graph theory to establish chronological precedence in the acquisition of different genomic functions. Specifically, we demonstrated that specific functions tend to be gained sequentially, suggesting that evolution in prokaryotes is governed by functional assembly patterns. Finally, we showed that these dependencies are universal rather than clade-specific and are often sufficient for predicting whether or not a given ancestral genome will acquire specific genes. Combined, our results indicate that evolutionary innovation via HGT is profoundly constrained by epistasis and historical contingency, similar to the evolution of proteins and phenotypic characters, and suggest that the emergence of specific metabolic and pathological phenotypes in prokaryotes can be predictable from current genomes.",
"discussion": "Discussion Combined, our findings provide substantial evidence to suggest that gene acquisitions in bacteria are governed by genome content through numerous gene-level dependencies. Our ability to detect these underlying dependencies is clearly imperfect, owing to various data and methodological limitations ( Supplemental Text ; Supplemental Fig. S3 ). Therefore, in reality, the complete dependency network is likely much denser than that described above and includes numerous dependencies and constraints that our approach may not be able to detect. Consequently, our estimates should be considered as a lower bound on the extent of gene–gene interactions, and accordingly, the predictability of HGT. Notably, even considering such caveats, our observations dramatically expand our knowledge of the constraints on HGT. Previous studies of such constraints demonstrated that genes frequently acquired by HGT tend to occupy peripheral positions in biological networks, are often associated with specific cellular functions, and are phylogenetically clustered ( Jain et al. 1999 ; Cohen et al. 2011 ). These observations suggested that properties of transferred genes are also important determinants of HGT regardless of recipient genome content ( Jain et al. 1999 ; Cohen et al. 2011 ; Gophna and Ofran 2011 ) and that the acquisition of certain genes is clade-specific ( Andam and Gogarten 2011 ; Popa et al. 2011 ). In contrast, our analysis demonstrates the importance of recipient genome content in influencing the propensity of a new gene to be acquired. In fact, to some extent, properties previously reported as determining the general “acquirability” of genes across all species may reflect an average constraint across genomes. By also considering variation in genomes acquiring genes, our analysis focused on specific biological effects, whose strengths may vary from genome to genome. Importantly, our model that gene acquisition is affected by recipient genome content is consistent with the observed enrichment of HGT among close relatives, which presumably have similar genome content ( Gogarten et al. 2002 ; Andam and Gogarten 2011 ; Popa and Dagan 2011 ; Popa et al. 2011 ). This taxonomic clustering of innovation by HGT is also in agreement with previous studies that demonstrated that phenotypic and genetic parallel evolution is more common than convergent evolution, potentially due to the effects of historical contingency ( Gould and Lewontin 1979 ; Conte et al. 2012 ; Christin et al. 2015 ; Ord and Summers 2015 ). However, in contrast to other studies, we present direct evidence that the mechanism by which contingency controls evolution is epistasis. Furthermore, the universality of PGCEs shows that the constraints underlying the effect of contingency operate outside the context of parallel evolution. Put differently, since each phylum-level clade is subject to an independent evolutionary trajectory, it is unlikely that the same dependency patterns would repeat solely due to parallel evolution. Moreover, our ability to predict where exactly along the tree gains of a specific gene are likely to occur ( Fig. 5 B) suggests that PGCEs successfully capture how variation in the genomic content (even among closely related species) affects future gain events. Such PGCE specificity therefore indicates that observed dependencies are not a trivial byproduct of prevalent gene transfer events among taxonomically closely related genomes (e.g., due to homologous recombination constraints) ( Popa et al. 2011 ). Nonetheless, the relative contributions of each of these various processes governing the assembly of prokaryotic genomes (and the evolution of complex systems in general) clearly deserve future study. Although our analysis revealed several intriguing patterns, the precise interpretation of some of these patterns remains unclear. For instance, the observed correspondence of topological ranks of genes to chronology suggests that evolutionary age is a potential contributor to such ranking, especially considering that our reconstructions likely lack many genes that have not been retained in any extant genomes. However, the biological plausibility and statistical robustness of PGCEs demonstrated above strongly argue that the observed evolutionary patterns are the result of constraint-inducing dependencies. Future work may therefore aim to quantify the trade-off between functional and chronological determinants in apparent evolutionary constraints. Finally, we demonstrate the predictability of genomic evolution by horizontal transfer from current genomic content. As stated above, this finding also suggests that such dependencies are fairly universal across the prokaryotic tree. Our approach was designed specifically to understand the PGCE network's significance and universality, rather than predict gene acquisition. It is likely that an approach specifically engineered for gene acquisition prediction would substantially outperform our approach. The estimates of predictability of genomic evolution presented here are accordingly quite conservative. The determinism and predictability of evolutionary patterns therefore appear to be an outcome not only of intramolecular epistasis in proteins or phylogenetic constraints, but also of genome-wide interactions between genes. This suggests that the evolution of medically, economically, and ecologically important traits in prokaryotes depends on ancestral genome content and is hence at least partly predictable, potentially informing research in the epidemiology of infectious diseases, bioengineering, and biotechnology."
} | 1,876 |
35744257 | PMC9228329 | pmc | 7,310 | {
"abstract": "The wettability, surface energy, structure, and morphology of a material’s surface will affect the interaction process between the material and the organism. Moreover, these factors are not independent of each other, but will affect each other, which together determine the biological surface of the material. Although two classic theories of surface wettability control have been established, including the Wenzel model and the Cassie–Baxter model, the mechanism of the microstructure parameters on the surface wettability has not been considered. This paper established a two-dimensional mathematical model of the composite wetting pattern based on microstructure parameters, revealed the mechanism of the microstructure parameters on the surface wettability, and then used ultra-precision cutting and molding composite preparation methods to quickly and efficiently prepare bionic structures, and the hydrophobic character of the microstructure was characterized by the contact angle meter, which provides theoretical support and preparation technology for the modification of the hydrophobic character of the material.",
"conclusion": "4. Conclusions In this paper, a two-dimensional mathematical model of the composite infiltration mode considering the hydrophobic properties of the microstructure parameters was established, the influence of the microstructure width and inclination angle on the hydrophobic properties of materials was discussed, and it was prepared by ultra-precision cutting and molding. The superhydrophobic structure (CA = 153.2°) was fabricated by the method, which improved the hydrophobic properties of the material.",
"introduction": "1. Introduction Due to hundreds of millions of years of natural selection and biological evolution, many animal and plant surfaces exhibit different excellent functional properties, such as self-cleaning [ 1 ], drag reduction and wear resistance [ 2 ], corrosion resistance [ 3 ], and low adhesion [ 4 ]. As people continue to research and discover, these peculiar functions have a huge connection with biological surface structure and microscopic morphology. The surface layer of the lotus leaf exhibits excellent self-cleaning properties, that is, the phenomenon of “lotus leaf effect” [ 5 ]. This is because there are a large number of micron papillae on the surface layer of the lotus leaf, and there are nano-villi structures and low-energy waxy layers on the papillary structure. The effect leads to the super-hydrophobicity of the lotus leaf; the surface of shark skin [ 6 ] has a special groove structure, which can make it swim quickly in the water and achieve excellent drag reduction function. Imitating the special wettability surface in nature, researchers designed and prepared a variety of special wettability surfaces, among which the superhydrophobic surface attracted the most attention of researchers [ 7 , 8 ]. The super-hydrophobic surface has super waterproof and self-cleaning properties, and has important application value in industrial production and daily life. However, the superhydrophobic surface also has some defects and shortcomings [ 9 , 10 ]; for example, it only exhibits superphobic characteristics for water with high surface tension, and is easily contaminated by organic oil stains with low surface tension. The “Cassie” of water on superhydrophobic surfaces, the contact state, is unstable, and external forces such as high pressure and impact can easily cause the “Cassie” contact state to change to the “Wenzel” contact state and lose its waterproof properties. Therefore, the study of superhydrophobic models based on the surface microstructure becomes particularly important. In recent years, scholars all over the world have conducted a large number of experiments and studies on superhydrophobic functional surfaces. There are many preparation methods, including sol-gel method [ 11 ], template method [ 12 ], chemical vapor deposition [ 13 ], laser-processing method [ 14 ] and so on. Many micro-nano manufacturing methods have been applied to prepare super-slip surfaces [ 15 , 16 , 17 , 18 ], but most of these methods are to build an additional layer of heterogeneous porous structure on the substrate material, and then refill lubricating oil to further obtain super-slip properties. Therefore, the prepared super-smooth surface substrate material and the liquid-infused porous structure layer have different physical, chemical, thermodynamic, and mechanical properties. When heated, bent, impacted, or other external forces are applied, it is easy to cause damage or damage to the super-slip surface layer. This paper established a microstructure-based superhydrophobic model to study the mechanism of the influence of microstructure parameters on the superhydrophobicity of the material, and then prepared a microstructure with superhydrophobic characteristics through a composite processing method of ultra-precision cutting and molding. The microstructure was verified by experiments. The superhydrophobicity of the structure has important reference value for the further development and application of superhydrophobic materials."
} | 1,277 |
27066306 | PMC4802760 | pmc | 7,311 | {
"abstract": "ABSTRACT Researchers contest the importance of gene flow in bacterial core genomes, as traditionalists view microbes as predominantly clonal, asexually reproducing organisms. Contrary to the traditional perspective, Escherichia coli core genes vary greatly in their levels of synonymous genetic diversity. This observation indicates that the relative importance of evolutionary forces such as mutation, selection, and recombination varies from gene to gene. In this paper, I highlight why the synonymous diversity observation is broadly relevant to researchers interested in the evolutionary dynamics of microbial populations and communities. I explain how a model of evolution called the coalescent relates neutral diversity (i.e. mutations with negligible fitness effects) to mutation rates, evolutionary time, and a parameter called effective population size. I then describe the possible ways in which mutation, selection, and recombination can explain observed patterns of synonymous diversity in E. coli. Finally, I describe a model for E. coli genome evolution in which different loci are subject to varying levels of gene flow among co-occurring microbes and viruses in the environment. Researchers can falsify the gene flow hypothesis by sequencing genes and strains isolated from stable microbiomes or by carrying out evolution experiments that trace gene genealogies in real-time.",
"conclusion": "Conclusion Synonymous genetic diversity depends on both the mutation rate and effective population size. In neutral models of evolution, effective population size has a second interpretation as the average time for 2 lineages to coalesce. Many evolutionary forces, including mutation, selection, and recombination can affect genome-wide variation in synonymous genetic diversity. While researchers recognize the importance of gene flow in structuring the flexible genome of microbes, gene flow may also affect the core genome of microbes. If so, gene flow could explain why highly important E. coli core genes have less synonymous genetic diversity than other core genes. While the importance of gene flow in microbial genome evolution depends strongly on ecological context, many important microbiomes, such as the animal gut, might be effectively described as metapopulations of genes that interact within and across genomes over multiple spatial and temporal scales."
} | 591 |
34086906 | PMC8214833 | pmc | 7,312 | {
"abstract": "ABSTRACT Learning allows animals to respond to changes in their environment within their lifespan. However, many responses to the environment are innate, and need not be learned. Depending on the level of cognitive flexibility an animal shows, such responses can either be modified by learning or not. Many ants deposit pheromone trails to resources, and innately follow such trails. Here, we investigated cognitive flexibility in the ant Lasius niger by asking whether ants can overcome their innate tendency and learn to avoid conspecific pheromone trails when these predict a negative stimulus. Ants were allowed to repeatedly visit a Y -maze, one arm of which was marked with a strong but realistic pheromone trail and led to a punishment (electric shock and/or quinine solution), and the other arm of which was unmarked and led to a 1 mol l −1 sucrose reward. After ca. 10 trials, ants stopped relying on the pheromone trail, but even after 25 exposures they failed to improve beyond chance levels. However, the ants did not choose randomly: rather, most ants began to favour just one side of the Y -maze, a strategy which resulted in more efficient food retrieval over time, when compared with the first visits. Even when trained in a go/no-go paradigm which precludes side bias development, ants failed to learn to avoid a pheromone trail. These results show rapid learning flexibility towards an innate social signal, but also demonstrate a rarely seen hard limit to this flexibility.",
"introduction": "INTRODUCTION Organisms can respond to predictable changes in the world in different ways, and over many time scales. Evolution can shape hard-wired adaptations, including behaviours, in response to environmental change occurring over many generations. Pleiotropic effects can allow animals to respond to changes in the world more rapidly, between single generations. However, these adaptations cannot help organisms cope with predictable events that change within their own lifetime. Learning offers such flexibility ( Shettleworth, 2009 ). However, learning comes at a cost: it requires not only neural architecture to be grown and maintained but also a period of information collection, before which it is ineffective ( Dukas, 1999 ; Dunlap and Stephens, 2016 ). This leaves animals exposed to making poor decisions until learning has taken place. Thus, learning is not predicted to occur for responses that are always appropriate. Studying where learning does or does not occur can thus inform us about the evolutionary pressures on an organism. In addition to learned responses, animals are equipped with a range of innate responses, which do not require learning. However, although these do not require learning, they are often open to it – these responses can be modified. For example, flies innately explore novel environments incessantly, but if repeatedly punished for moving, reduce their movement ( Sun et al., 2020 ; ‘operant conditioning’: Skinner, 1963 ). Similarly, crayfish express an innate positive taxis towards blue light but learned to avoid it for at least 48 h after three pairings with electric shocks ( Okada et al., 2021 ). Pheromones are chemical signalling substances, triggering innate responses ( Karlson and Lüscher, 1959 ; Sudd, 1959 ; Wyatt, 2014 ; but see Baracchi et al., 2020 ). However, studies on pheromone-mediated reproductive behaviours (for reviews, see Beny and Kimchi, 2014 ; Pfaus et al., 2001 ) have demonstrated the role of past experiences and learning in the modification of those responses. For example, male mice reduced the production of ultrasonic vocalizations in response to female urine presentation when urine presentation was repeatedly not followed by the female herself ( Nyby et al., 1978 ). Male golden hamsters were shown to suppress pheromone-mediated sexual behaviours towards females after vaginal secretion presentation was followed by lithium chloride poisoning ( Johnston and Zahorik, 1975 ; Zahorik and Johnston, 1976 ). However, not only does it seem possible to modify the frequency of innate behaviours after punishment via operant conditioning but also animals can learn to act against their innate response by changing the representation of the valence associated with certain stimuli (‘anti-instinctive learning’): rats innately react aversively towards peppermint odour and cadaverine but when associating the odours with a positive stimulus (e.g. a tactile stimulation that mimics suckling in pups: Yuan et al., 2013 ; or sexual behaviour in adult males: Pfaus et al., 2012 ), the valence of the odours can positively shift and even become attractive (‘classical conditioning’: Pavlov, 1927 ). Honeybees, as another example, are innately attracted to geraniol and citral, two major components of the Nasanov pheromone ( Pickett et al., 1980 ). However, when these substances are paired with electric shocks, bees eventually begin to show aversive responses (sting extension) when exposed to these pheromones alone and efficiently retrieve the learned association 1 h after the initial test ( Roussel et al., 2012 ). The last example is one of the very few demonstrations of anti-instinctive learning in an insect. However, while the valence response was reversed (something positive became negative), the behavioural response modality was different: a shift from a locomotion response (attraction) to a defensive response (sting extension). To date, only very few examples of anti-instinctive learning in the same behavioural response modality have been demonstrated in mammals ( Pfaus et al., 2012 ; Yuan et al., 2013 ), and it is unclear whether insects would be capable of such a complete behavioural reversal (opposite-instinctive learning). Many ants, as well as termites, bees and wasps, deploy pheromone trails to guide nestmates to important resources ( Bordereau and Pasteels, 2010 ; Czaczkes et al., 2015 ; Jeanne, 1981 ; Lindauer and Kerr, 1958 ). Trail following is overwhelmingly considered an innate behaviour which does not require learning (but see Cammaerts, 2013 ; Reichle et al., 2011 ). However, the response to pheromone trails is not always full and absolute: other information sources affect how organisms respond to such trails ( Czaczkes et al., 2015 ). For example, ants may ignore pheromone trails if they conflict with memory ( Cronin, 2013 ; Grüter et al., 2011 ; Harrison et al., 1989 ), or use orientation cues to decide in which direction to follow a trail if they join it in the middle ( Czaczkes and Ratnieks, 2012 ; Minoura et al., 2016 ). Ants are also good learners, and well-able to rapidly form associations between odours and rewards or punishments ( Desmedt et al., 2017 ; Dupuy et al., 2006 ; Oberhauser and Czaczkes, 2018 ; Turner, 1907 ). In order to examine opposite-instinctive learning in Lasius niger (henceforth ‘ants’), four experiments were conducted. First, we attempted to teach ants that a pheromone trail predicted a quinine punishment on one arm of a Y -maze while the unmarked arm led to a sucrose reward (experiment 1). As ants in experiment 1 quickly learned to avoid punishment by carefully probing the quinine drop with their antennae without tasting it, we developed an apparatus for delivering inescapable electric shock punishments to free-walking ants and used this in addition to quinine in experiment 2. Then, to exclude the possibility that side-bias learning was preventing opposite-instinctive learning, we examined whether ants could learn that the pheromone predicted a punishment using a go/no-go paradigm on a linear runway (experiment 3). Finally, we demonstrated that ants can learn to avoid an odour when it predicts a negative stimulus (experiment 4).",
"discussion": "DISCUSSION Ants were adept learners, quickly learning to ignore pheromone trails ( Fig. 2 ). However, we discovered a hard limit to their learning – while they could learn to ignore trails, they could not learn to actively avoid them. After 5–10 visits of rapid performance improvement, improvements stopped completely, and ants never increased choice accuracy above chance level. However, ants learned to probe the drop before attempting to drink it, and thus avoided the quinine punishment, which potentially restricted their learning success. We therefore conducted an experiment adding an unavoidable punishment by introducing a shocker (using a relatively high voltage, comparable to that used for harnessed honeybees, e.g. Roussel et al., 2009 ) but still ants did not improve above chance level in the given learning task. We can rule out a general inability of ants to learn to avoid chemical signals, as ants demonstrated one-trial learning with a consequent 100% accuracy in a comparable setup when the predictor for punishment was a lemon odour instead of a pheromone trail. These results indicate that odours could easily acquire negative valence through associative learning, while with pheromones, ants were unable to reliably perform opposite-valence responses. However, ants did not choose randomly when confronted with the Y -maze: within the first 10 trials, most subjects started applying an alternative strategy and developed a side bias. This allowed them to improve their choice accuracy from ∼6.5% (initial choice accuracy) to ∼50% and shortened the latency to reach the sucrose reward ( Fig. 3 ) over successive trials. The display of repetitive behaviours and simple navigation rules, such as the formation of a side bias, has previously been described when ants are confronted with complex tasks ( Macquart et al., 2008 ; Oberhauser et al., 2020 ). However, the ability to form side bias might have blocked the ants’ ability to learn to avoid pheromone trails in the presented task. We therefore confronted subjects with a go/no-go setup where the formation of a side bias was impossible, but ants still showed hard limits to their learning flexibility: subjects took increasingly longer to approach the presented quinine drop in punishment trials but remained faster in comparison to the reward trials ( Fig. 4 ). While increasing their performance in the unrewarded tests in the go/no-go experiment, ants failed to improve beyond chance level. Overall, our results show rapid learning flexibility towards an innate social signal, but also demonstrate a rarely seen hard limit to this flexibility. Retarded learning due to biological constraints is a well-known phenomenon in animal learning. It appears specifically where certain responses are very difficult or impossible for animals to learn – usually those that lie outside or are in conflict with the animals’ natural responses ( Krause, 2015 ; LoLordo, 1979 ; Rozin and Kalat, 1971 ; Seligman and Hager, 1972 ; Shettleworth, 2009 ). Rats, for example, can easily form an association between a taste and a subsequent gastric illness (taste aversion) while they fail to associate audio-visual cues with gastric illness or a taste with subsequent electric shocks ( Garcia and Koelling, 1966 ; Wilcoxon et al., 1971 ). In contrast, vampire bats, a species that only feed on a single kind of food, do not demonstrate taste aversion learning at all ( Ratcliffe et al., 2003 ). However, to date, demonstrations of constraints on learning have only shown a simple inability to apply learning in a particular domain ( Domjan, 2005 ; Krause, 2015 ; Mineka and Cook, 1988 ). Here, we provide a very rare demonstration of a case in which a domain is very amenable to learning, but only up to a very well-defined point. In the present study, we not only investigated the ants’ ability to learn a switch in valence of social signals (compare Bos et al., 2010 ; Roussel et al., 2012 ) but also asked our subjects to express the exact opposite of their innate behaviour, making the present experiments, to our knowledge, one of the first assessments of such learning ability in insects. An equivalent in honeybees might be, for example, to attempt to train workers to enter the hive via an unmarked entrance, while avoiding an entrance marked with Nasanov pheromone ( Sladen, 1901 ; von Frisch, 1923 ). To conclude, while ants immediately learned to avoid odours when associated with a punishment by assigning negative valence to the stimulus, they failed to do so within 25 trials when the stimulus was a pheromone trail. However, the response to the pheromone was very open to manipulation through experience: ants quickly learned to ignore trails. By developing a simple rule (‘always choose one side and correct if wrong’), they were able to dramatically increase their foraging efficiency. Individual ants can thus develop effective solutions to problems that are beyond their cognitive limitations, by relying on simple rules ( Oberhauser et al., 2020 )."
} | 3,192 |
29483940 | PMC5820786 | pmc | 7,313 | {
"abstract": "Background Part-stream low-frequency ultrasound (LFUS) was one of the common practices for sludge disintegration in full-scale anaerobic digestion (AD) facilities. However, the effectiveness of part-stream LFUS treatment and its effect on AD microbiome have not been fully elucidated. Methods Here we testified the effectiveness of part-stream LFUS pretreatment by treating only a fraction of feed sludge (23% and 33% total solid of the feed sludge) with 20 Hz LFUS for 70 s. State-of-the-art metagenomic and metatranscriptomic analysis was used to investigate the microbial process underpinning the enhanced AD performance by part-stream LFUS pretreatment. Results By pretreating 33% total solid of the feed sludge, methane yield was increased by 36.5%, while the volatile solid reduction ratio remained unchanged. RNA-seq of the microbiome at stable stage showed that the continuous dosage of easy-degradable LFUS-pretreated feed sludge had gradually altered the microbial community by selecting Bacteroidales hydrolyzer with greater metabolic capability to hydrolyze cellulosic biomass without substrate attachment. Meanwhile, Thermotogales with excellent cell mobility for nutrient capturing was highly active within the community. Foremost proportion of the methanogenesis was contributed by the dominant Methanomicrobiales via carbon dioxide reduction. More interestingly, a perceivable proportion of the reverse electron flow of the community was input from Methanoculleus species other than syntrophic acetate-oxidizing bacteria. In addition, metagenomic binning retrieved several interesting novel metagenomic-assembled genomes (MAGs): MAG-bin6 of Alistipes shahii showed exceptional transcriptional activities towards protein degradation and MAG-bin11 of Candidatus Cloacimonetes with active cellulolytic GH74 gene detected. Conclusions In summary, despite the unchanged sludge digestibility, the applied part-stream LFUS pretreatment strategy was robust in adjusting the microbial pathways towards more effective substrate conversion enabled by free-living hydrolyser and beta-oxidation-capable methanogens. Electronic supplementary material The online version of this article (10.1186/s13068-018-1042-y) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusion Using state-of-the-art HTS-based metagenomics and metatranscriptomics, microbial mechanisms underpinning the enhanced bioenergy recovery by part-stream LFUS pretreatment were investigated. Results showed that the continuous dosage of LFUS-treated sludge was robust to enrich an effective hydrolyzer community adept of hydrolyzing recalcitrant substrate without attachment. In addition, the vigorous contribution of beta-oxidizing capable methanogens of Methanoculleus may play an important role in the promoted methane productivity by LFUS pretreatment.",
"discussion": "Results and discussion Effectiveness of part-stream LFUS pretreatment on AD performance All the digesters reached steady methane generation after 4 weeks (day 25) of inoculation (Fig. 1 ). Except for bioreactor M3 whose operational pH shifted from 9.4 before day 41 to 7.3; after that, the other reactors all showed stable operational pH around 7.0 (Additional file 1 : Figure S2). During the 3-month operation period, the control and ultrasound treatment group (M3 and M4, respectively, fed with 23 and 33% of LFUS-treated TSAS) had significant differences in some key parameters such as daily biogas production and methane (CH 4 ) yield (Fig. 1 ). Complete test results are shown in Additional file 1 : Table S3. During steady operation, the biogas production of the two LFUS treatment groups (M3 and M4) was, respectively, 8.7 and 36.5% higher ( p value < 0.05) than that of the two control reactors (treating regular combined sludge). In the literature, methane yield of digesters treating LFUS-pretreated sludge with full-stream strategy was roughly 40–50% higher than that of the untreated sludge [ 3 , 18 , 61 , 66 ]. Therefore, the degree of methane yield increase (36.5% of digester M4 in which 33% of feed sludge was LFUS-pretreated) in our part-stream method was slightly lower than digesters applying a full-stream pretreatment strategy in which all the feed sludge was pretreated. In addition, the average increase on VSR of around 40% by full-stream LFUS pretreatment [ 3 , 18 , 24 , 46 , 61 , 72 ] was not observed in our digesters applying part-stream strategy. In our digesters, the average VSR of the LFUS treatment groups (61.9% for M3 and 60.0% for M4) did not show significant difference from that of control without LFUS-treated TSAS feeding (averagely 59.6%) (Additional file 1 : Figure S2). Such unaffected VSR was expectable as only a fraction of feed sludge (33% feed sludge of digester M4) was pretreated, indicating the applied feed loading (LFUS-treated TS taking 9% of the total TS of M4 digester, Table 1 ) did not decidedly alter the VS digestibility of sludge. However, the evident increased (36.5% of digester M4) methane production with unchanged VSR was noteworthy, suggesting that comparing to full-stream method, part-stream LFUS pretreatment showed more obvious effect on promoting methanogenesis than sludge hydrolysis of AD process. Despite the unaffected sludge solubility, the applied part-stream LFUS pretreatment on feed sludge was enough to induce evident change on chemical composition of the solubilized VS which may facilitate the methanogenesis process and alter the overall microbial pathway within the digester. These observations roused our interests to investigate to what extent the feeding of LFUS-treated sludge could affect the structure of microbial community and what are the active functional populations involves in digestion of LFUS-treated sludge. One more thing to point out was that the sludge used in this study was generated from chemically enhanced primary settling (CEPT process). Such CEPT sludge had shown better digestibility than regular sludge [ 29 ]. Consequently, the degree of performance promotion reported here may be higher than other studies apply similar part-stream strategy [ 12 , 52 ]. Fig. 1 3-month reactor performance in terms of daily methane production (top figure) and methane yield (bottom figure). M3 and M4 group was, respectively, fed with 23 and 33% of LFUS-treated TSAS. Sampling points for metagenomic and metatranscriptomic sequencing were indicated by red arrow \n Metagenomic and metatranscriptomic data sets To reveal the core active populations within the digestion process, three sludge samples were collected from M4 digesters, respectively, at 41, 57, and 77 days of operation. Metagenomic DNA and total RNA were freshly extracted from these biological triplicates for sequencing on Illumina Hiseq platform. Illumina sequencing resulted in totally 9.9 and 10.9 Gb of metagenomic and metatranscriptomic reads (Additional file 1 : Table S2). rRNA sequences took 0.2% of the metagenomes constructed. De novo assembly of the metagenomes recovered 401,646 genes (open reading frames) with 42.2% got transcriptional activities detected (Additional file 1 : Table S4). Rarefaction analysis based on assembled genes and 16S rRNA sequences of the metagenome data sets both suggested a sufficient coverage of the core populations of the digestive community (Additional file 1 : Figure S3). The metagenome data sets of the three biological triplicates showed high reproducibility in term of RPKM-DNA of the assembled genes (correlation coefficient between replicates > 0.8) (Additional file 1 : Figure S4). The highly reproducible gene abundance (in term of RPKM-DNA) in the metagenomic data sets at three sampling times indicated a very stable community structure during steady operation. The transcriptional variation (RPKM-RNA values of genes) among biological duplicates (Pearson’s correlation coefficient of around 0.8) was consistent with the previous evaluation on the reproducibility between metatranscriptomic replicates [ 62 ], suggesting reliable metatranscriptome construction in this study. 10,790 genes showed > 4 times variation among metatranscriptomic triplicates. These genes took 6% of the total transcribed genes; among them, genes encoding translation of ribosomal protein showed highest transcriptional variation, suggesting high susceptibility of ribosomal protein synthesis towards environmental change [ 23 , 41 ]. Active populations in AD with LFUS pretreatment Similar to other municipal AD systems [ 9 , 30 ], Proteobacteria, Bacteroidetes, and Firmicutes, respectively, taking 24.7, 22.2, and 18.0% of the community were the dominant populations of our AD with LFUS-treated feeding (Fig. 2 ). Consistent with their prevalence was their active transcriptional activities detected within the community that these dominant phyla contributed 64.5% of the total transcripts. Noteworthy, there were three particularly active populations: Euryarchaeota , Cloacimonetes, and Thermotogae (respectively, taking 3.0, 2.7, and 1.2% of the community). Despite being less prevalence, their transcripts took 25.3% of the whole community transcription, suggesting their important roles within AD fed with LFUS-treated sludge. Fig. 2 Transcriptional activities of active populations in the CEPT sludge digestion system. The scatter plot on the left shows the active orders within the community. Since active members of Cloacimonetes phylum do not have order level taxonomic classification but could be classified at species level (by either PhyloPythiaS + or metagenomic binning analysis), classified species of this phylum are included in the scatter plot. Points of different orders are sized according to their relative abundance within the community; major orders showing > 0.6% abundance are colored in red, while other minor populations are colored in blue. Labels of different orders are colored according to their phylum affiliation as in the right figure. RPKM-DNA, RPKM-RNA, and MRPKM values are respectively summarized for the major and minor populations in the table below scatter plot. The right figure shows the active phyla within the system. Phyla are sorted decreasingly accord to their relative abundance within the community (blue bar), while their transcriptional activities (RPKM-RNA) were shown in the right (pink bar) \n The active involvement of Cloacimonetes 18 (a Candidatus phylum formerly known as candidatus phylum OP5 and WWE1) was noteworthy. This newly defined population has been widely found in prevalence in AD systems [ 15 , 57 , 59 ]. Genome reconstruction has revealed a putative syntrophic propionate-metabolizing lifestyle of some members of Cloacimonetes [ 44 , 48 , 57 ]; however, the available genomic information (one complete genome of Cloacamonas acidaminovorans str. Evry [ 50 ] and several draft genomes) of this phylum could only covered 1/3 of the Cloacimonetes population in our digestion system, suggesting the presence of novel active players within our community. Using multi-dimensional binning strategy, we retrieved the genome (named as MAG-bin11 with 90% completeness and 1.0% contamination) of this novel active member of Cloacimonetes . MAG-bin11 was phylogenetically divergent from previously identified lineages of Cloacimonetes phylum (Additional file 1 : Figure S5). The role of this population will be discussed in detail in subsequent sections. To gain higher resolution into the roles of different populations, we investigated the major players within the community at Order level based on the trade-off between phylogenetic classification ration and functional interpretability of genes. When interesting expression pattern was discovered in an order, additional effort had been put to retrieve the genomes [more precisely named as the metagenomic-assembled genomes (MAGs)] of the active members within the lineage. Methanomicrobiales , Bacteroidales ; Clostridiales ; Cytophagales, as well as Thermotogales [all showed RPKM-RNA (sqrt) > 300, Additional file 1 : Figure S6] were the most active orders identified in the LFUS-treated sludge digestion community. Interesting transcriptional patterns were observed in these active populations. High transcription of genes empowering mobility and chemotaxis was observed in Thermotogales (Additional file 1 : Figure S6), consistent with our previous finding in the metatranscriptome of thermophilic AD system treating cellulolytic biomass [ 70 ]. Such strong transcription was principally (81.8%) contributed by the flagellin protein FlaA which showed 60% amino acid similarity to flagellin protein of Fervidobacterium changbaicum and Fervidobacterium nodosum [ 8 , 49 ]. MAG of the active member of Thermotogales was retrieved as MAG-bin3. Marker gene-based phylogenetic analysis confirmed MAG-bin3 affiliated with the Fervidobacterium genus which contained a variety of hyperthermophilic species that could utilize a wide spectrum of carbohydrate substrates for growth [ 8 , 49 ]. However, MAG-bin3 showed very low ANI (< 80%) to representative strains of Fervidobacterium suggesting its genotype novelty within the genus. The dosage of LFUS-treated sludge might had facilitated its wide spread within our AD community as the increased concentration of soluble substrate by LFUS pretreatment would selectively enrich free-living microbes with higher cellular mobility for effective nutrient capturing. Such metabolic advantage could shed light on the wide spread of this population in AD systems, especially when high content of easy-degradable substrates was available. One thing to point out is that as only roughly 40% of the genes could be assigned to an order-level taxonomy by the annotation methods used (Additional file 1 : Table S4), biases caused by the uneven classification ratio of different populations may alter the structure of main active orders revealed in Fig. 2 . In addition, there may exist novel active populations that could not be assigned into a given order or could not be recovered by our metagenomic binning protocol due to assembly difficulties. Active populations in the hydrolysis pathway Hydrolysis of cellulolytic substrate Hydrolysis of complex polysaccharides, especially the recalcitrant cellulosic component, was regarded as the rate-limiting step for AD process [ 32 ]; consequently, the cellulolytic activities of different populations within the LFUS-treated sludge digestion community were studied by comparing the transcriptional activities (in terms of RPKM-RNA) of glycoside hydrolases families (GH families defined by CAZy database). By contributing 24.9% of all the active transcriptions of GH families within the community, Bacteroidales played the most active role in carbohydrate hydrolysis within the community. As shown in Fig. 3 , active expression of a complete set of hydrolases associated with hydrolysis of cellulosic substrate (including: endoglucanase of GH5 and GH74, hemicellulose of GH43, as well as beta-glycosidase of GH2, GH3, and GH92) were identified in this population. Noteworthy, genes enabling cellulose degradation were more active in Bacteroidales than Clostridiales in our community (Fig. 3 ). Bacteroidales and Clostridiales are both well-known cellulose degraders [ 37 , 38 ]. However, Bacteroidales and Clostridiales hydrolyze cellulose with different mechanisms: Clostridiales hydrolyze cellulose using cellulosomes [ 37 , 42 ]. However, Bacteroidales do not produce cellulosome; instead, their cellulose hydrolysis is associated with the production of very versatile polysaccharide utilization locis (PULs) [ 38 , 39 , 60 ]. Both cellulosome and PULs are attachment-based cellulose hydrolysis mechanism. However, in our digestion community, only one cellulolytic PULs (PULs that had cellulase-encoding genes located next to or in close proximity to a SusC and SusD gene pair) were identified in an undefined Bacteroidetes population with limited level of expression detected (sum of RPKM-RNA of surrounding genes of 36.3). In addition, we observed very low activities of the key building blocks of cellulosomes in Clostridiales (RPKM-RNA of Dockerin and Cohesin, respectively, of 0.4 and 110.1). Such reluctant PULs and cellulosome activity indicated a general lack of attachment initiated cellulose conversion in the LFUS-treated sludge digestion system, which was in sharp contradiction to systems digesting rigid cellulosic substrate like raw sludge, grass, or microcrystalline cellulose [ 20 , 70 ]. These results suggested that despite the physically unchanged digestibility of the treated biomass (as revealed by unaffected VSR% in Additional file 1 : Figure S2), the continuous dosage of LFUS-treated sludge (at 33% TS of feed sludge) has gradually altered the hydrolysis pathway by selecting microbes more capable of hydrolyzing cellulosic substrate without attachment. Fig. 3 Transcriptional activities (top figure) and relative abundance (bottom figure) of various glycoside hydrolase (GH) families attributed by the major orders of the LFUS-treated sludge digestion community \n Protein hydrolysis Since the biodegradation of protein content of the dead cells (residue populations) is associated with the release of ammonia, the major inhibitory compound to both hydrolysis and methanogenesis in AD metabolism, it is indispensable to identify the key protein degraders whose metabolism may cause the ammonia accumulation in the AD systems. As a result, the active populations of protein hydrolysis in the LFUS-treated sludge digestion community were investigated. Protein metabolism (RPKM-RNA of 42708.3) was as active as carbohydrate metabolism (RPKM-RNA of 41789.2) in the community. By comparing the transcriptional activities of key genes associated with SEED subsystem of “Protein degradation”, we observed four populations of Bacteroidales , Clostridiales , Methanomicrobiales, and Cloacimonas acidaminovorans actively involved in protein degradation in our digestion community (Additional file 1 : Figure S7). As indicated by the different protein degradation-related genes (Additional file 1 : Figure S7), polypeptides were broken-down by these populations via miscellaneous metabolic pathways for different motives in the system. Aminopeptidase C (pepC, KO1372, EC 3.4.22.40), taking 55.2% of total transcriptions of protein degradation-related genes, was the most active peptidase within the community. Though the majority (53.3%) of the activities of pepC were contributed by microbes that cannot be phylogenetically assigned, Alistipes shahii of Bacteroidales, taking 19.3% of pepC activities, was the most active protein degrading species identified in the community. Alistipes shahii species is one of the key members of Bacteroidales resident in human gut [ 2 , 75 ]. Its remarkable metabolic capacity towards polypeptides degradation showed here had not been reported elsewhere before, including in isolated strains [ 58 ]. Such metabolic advantage in protein degradation might play a vital role in facilitating its widespread in human gut especially in elder adults with more protein-rich diet [ 33 , 74 ]. A high-quality MAG (named as MAG-bin6 with estimated completeness of 90.7% and contamination of 3.7%) was recovered for Alistipes shahii , adding up contextual genomic information to this important species. Also noteworthy was that genus of Methanoculleus and Candidatus Cloacimonas acidaminovorans species were another two important protein hydrolyzers within the digestion community. Protein hydrolysis by Methanoculleus and C. acidaminovorans was, respectively, empowered by the active transcription of AAA-ATPase (PAN) and Proteasome subunit alpha (EC 3.4.25.1). It was the first time that these populations were observed as major protein degraders in AD communities. In addition, energy-dependent proteolysis by Clp protease of an undefined genus of Peptococcaceae , taking 37.1% of all the protein degradation activities of Clostridiales , was particularly active within our digestion system. Proteolysis by Clp protease was regulatory important for its ability to effectively turnover terminally damaged polypeptides under adverse conditions [ 13 , 22 , 34 ]. The active transcription of Clp protease in Peptococcaceae suggested its stressful living condition which was probably imposed by the temporal temperature increase caused by dosage of LFUS-treated feed sludge. Active populations in the methanogenesis pathway Methanomicrobiales and Methanosarcinales, respectively, taking 1.04 and 0.14% of the community (Additional file 1 : Table S5) were the prevalent methanogenic populations within our LFUS-treated sludge digestion system. Among them, strains of Methanomicrobiales were exclusively hydrogenotrophic, reducing CO 2 into methane with H 2 as electron donor [ 47 , 63 ]; while Methanosarcinales strains could also produce methane through acetate cleavage [ 63 ]. These two archaeal populations often found separately or together as the dominant methanogens in anaerobic digesters [ 28 , 29 , 64 ]. Within our digestion system, the dominant proportion (96.1%) of methanogenic activities of the digestion system was endorsed by Methanomicrobiales that the overall transcriptional activity of Methanomicrobiales was 14.7-fold higher than that of Methanosarcinales (Fig. 4 ). Correspondingly, we observed the hydrogenotrophic pathway was 10.8-fold more active than aceticlastic pathway in the community (Fig. 4 ), indicating methane produced in the digestion system was mainly via hydrogenotrophic pathway. Fig. 4 Transcriptional activities (right) and genomic prevalence (left) of genes in the methanogenesis pathway. Genes were classified into various phylogenetic orders, as shown in the top figure, and colored according to their functions in the methanogenesis process (as adopted from the KEGG Methane Metabolism pathway), as shown in the flowchart (bottom figure) \n Syntrophic associations between fermentative bacteria and methanogens were important for process stability of hydrogenotrophic methanogenic systems [ 54 , 56 , 67 ]. In anaerobic digesters, syntrophic bacteria produce hydrogen and formate (CO 2 ) from its growth substrate (e.g., acetate, propionate, and butyrate). The hydrogenotrophic methanogens consume these products, keeping them at concentration low enough for the overall degradative reaction is thermodynamically favorable [ 56 ]. The syntrophic beta-oxidizing population of LFUS-treated sludge digestion system was inferred by the phylogenetic affiliation of the key genes (carbon-monoxide dehydrogenase, EC1.2.7.4, K00198) [ 56 ]. Typical syntrophic acetate-oxidizing bacteria (SAOB) were identified by this method. These SAOB included: Anaerolinea thermophila of Anaerolineale [ 55 ], Candidatus Cloacimonas acidaminovorans [ 50 ], Cloacimonetes bacterium JGI 0000039-G13 [ 53 ], and some unknown species of Peptococcaceae family of Clostridiales order (Fig. 4 ). It was interesting to notice that the majority (68.9%) of beta-oxidation activities were not attributed by these SAOBs. Instead, Methanoculleus genus (consist of two species of Methanoculleus marisnigri and Methanoculleus bourgensis ), taking 44.3% of the methanogenesis activities of Methanomicrobiales , were the most active syntrophic oxidizer within the community. In our system, active involvement of Methanoculleus species in beta-oxidation was evidenced by the transcription of carbon-monoxide dehydrogenase (RPKM-RNA equal to 122.8). A unique metabolic feature of Methanoculleus species is their capability to utilize a variety of secondary alcohols as electron donors to reduce CO 2 to methane [ 1 , 5 , 73 ]. Meanwhile, strains in closely related genus Methanospirillum could oxidize ethanol for CO 2 reduction [ 68 ]. The active transcription of beta-oxidation pathway in Methanoculleus indicated an important, at least perceptible, proportion of reverse electron transfer within the community took place within the cells of these beta-oxidization-capable methanogens (BOM) other than between their SAOB partners. In the literature, Methanoculleus and SAOB were often got concurrently enriched in AD systems with elevated level of ammonia, volatile fatty acids, or other inhibitory intermediate metabolites [ 14 , 54 ], implying that BOM shared very similar ecological niches with SAOB in AD systems. The functional redundancy on beta-oxidation by BOM and SAOB could help to ensure the stability of hydrogenotrophic methanogenic performance. Our results also suggested that BOM may contribute more in the reverse electron flow of hydrogenotrophic methanogenesis than previously reported. In summary, foremost proportion of the methanogenic activities of the LFUS-treated sludge digestion system was contributed by the dominant Methanomicrobiales via carbon dioxide reduction. More interestingly, the input from Methanoculleus species in beta-oxidation was larger than SAOBs of the community. Since BOM and SAOB were concurrently enriched in a variety of AD systems, such major influence of BOM in beta-oxidation revealed here may necessitate a re-evaluation of the syntrophic beta-oxidation pathway in hydrogenotrophic methanogenesis."
} | 6,339 |
28555001 | PMC5484101 | pmc | 7,315 | {
"abstract": "Water surface-floating microalgae have great potential for biofuel applications due to the ease of the harvesting process, which is one of the most problematic steps in conventional microalgal biofuel production. We have collected promising water surface-floating microalgae and characterized their capacity for biomass and lipid production. In this study, we performed chemical mutagenesis of two water surface-floating microalgae to elevate productivity. Floating microalgal strains AVFF007 and FFG039 (tentatively identified as Botryosphaerella sp. and Chlorococcum sp., respectively) were exposed to ethyl methane sulfonate (EMS) or 1-methyl-3-nitro-1-nitrosoguanidine (MNNG), and pale green mutants (PMs) were obtained. The most promising FFG039 PM formed robust biofilms on the surface of the culture medium, similar to those formed by wild type strains, and it exhibited 1.7-fold and 1.9-fold higher biomass and lipid productivities than those of the wild type. This study indicates that the chemical mutation strategy improves the lipid productivity of water surface-floating microalgae without inhibiting biofilm formation and floating ability.",
"conclusion": "5. Conclusions Chemical mutagenesis of two strains of water surface-floating microalgae, Botryosphaerella sp. AVFF007 and Chlorococcum sp. FFG039, was performed. FFG039 mutants with a pale green color (58 clones) were isolated, 46 of which showed higher biomass productivity. The most productive clone, PM11, showed biomass and lipid productivities of 5.4 ± 0.2 g/(m 2 day) and 1.9 ± 0.1 g/(m 2 day), respectively, which were 1.7-fold and 1.9-fold higher than those of the wild type. Even after mutation, biofilm formation and water surface-floating abilities were not lost. These results suggest that chemical mutagenesis is a promising way to enhance the potential of water surface-floating microalgae in biofuel production.",
"introduction": "1. Introduction Microalgal biofuels have been recognized as a promising alternative to petroleum owing to their high productivity of biomass and lipids compared with that of terrestrial crop plants, no competition with food and feed productions that require extensive arable land, and the advantage of CO 2 fixation through photosynthesis [ 1 , 2 , 3 , 4 ]. Although high energy consumption during the production of microalgal lipids has long hampered the industrial production of microalgal biofuels, researchers have recently explored routes to circumvent this issue [ 5 ]. To minimize energy consumption, the methods for harvesting and drying microalgal biomass must be improved; these processes have been estimated to consume more than 69% of the entire energy input for biodiesel production [ 6 ]. Several alternative methods such as filtration [ 7 , 8 , 9 ], gravity sedimentation [ 10 , 11 ], affinity sedimentation [ 12 ], flocculation [ 13 ], and flotation [ 14 , 15 ] have been proposed to replace centrifugation as the harvesting technique. We have previously proposed a novel approach for efficient biomass harvesting in which the water surface-floating microalgal Botryosphaerella sp. AVFF007 was employed for lipid production [ 16 ]. The strain AVFF007 forms robust biofilms that float on the surface of the culture medium. The floating biofilms can be harvested by adsorption onto poly(ethylene) (PE) films. This approach is promising because of the ease of the harvesting process and lower moisture content of the floating biofilm, which is largely correlated to the amount of energy input required for the drying process. However, lipid productivity of this strain was not higher than that of other microalgae. Because the water surface-floating microalgae found in natural habitats have exhibited a moderate oleaginous phenotype [ 16 ], improvement of the biomass and lipid productivities are needed. Chemical mutagenesis has long been employed to improve the biomass and/or lipid productivities of microalgae such as Nannochloropsis sp. [ 17 ], Chlorella vulgaris [ 18 ], Desmodesmus sp. [ 19 ], and Chlamydomonas perigranulata [ 20 ]. In particular, it has been demonstrated that the attenuation of the light-harvesting property by mutation could improve the photosynthetic efficiency of microalgae by decreasing the cell-shading effect, allowing an increase in biomass and lipid productivities [ 21 ]. However, technical challenges still exist in the mutagenesis of biofilm-forming microalgae because the microalgal biofilms contain an expansive exopolymeric matrix that might decrease the effect of mutagens. In addition, tight cell aggregation in a biofilm is problematic in the screening step following mutagenesis. In a traditional chemical mutagenesis protocol, the target cells are treated with mutagens and subjected to colony formation on agar media. Subsequently, the clones that exhibit the desired phenotypes are screened. Recently, alternative high-throughput screening methods based on flow cytometry were developed [ 17 , 22 ]. However, in any of these methods, the mutated cells need to be adequately dispersed, without which it is difficult to isolate mutated cells with the desired phenotype from the adhesive and surrounding non-mutated cells. In this study, we established an effective method to generate chemical mutants of two water surface-floating microalgae strains, Botryosphaerella sp. AVFF007 and Chlorococcum sp. FFG039. Prior to chemical mutagenesis, the cells were dispersed by sonication. After the sonicated cells were treated with mutagens, the resulting colonies containing less pigment were isolated. Finally, the biomass and lipid productivities of the isolated mutant clones were investigated. On the basis of these results, the potential of the mutated microalgae for oil production is discussed.",
"discussion": "3. Discussion It has long been known that some algal strains (including cyanobacteria and eukaryotic micro- and macroalgae) occasionally form floating clusters (biofilms) on the water surface and play a substantial role in a variety of ecosystems [ 23 , 24 , 25 , 26 ]. However, little is known about the potential of floating algae for biotechnological applications. Recently, biofuel production research using floating microalgae was launched, and critical advantages were proposed, such as ease of harvesting [ 16 ] and concomitant in situ bioremediation of metals (e.g., Mn, As, Ni, Cr, and Cu) [ 25 ]. In our previous study [ 16 ], we isolated 168 microalgal strains with water surface-floating ability. Of these, Botryosphaerella sp. AVFF007 was selected as one of the most promising candidates for mutagenesis. However, the lipid content of AVFF007 was moderate, and thus this strain may not be the best candidate. In this study, we investigated another water surface-floating microalga, Chlorococcum sp. FFG039, as a candidate for mutagenesis, owing to its higher lipid content, compared to that that of AVFF007 ( Table 1 ). Molecular phylogenetic analysis of the 18S rRNA gene sequence revealed that FFG039 is closely related to Chlorococcum aquaticum UTEX2222. It was reported that Chlorococcum aquaticum UTEX2222 contained a large number of oil bodies, as does FFG039 ( Figure S3 ) [ 27 ]. Marine strains belonging to the genus Chlorococcum (e.g., Chlorococcum littorale ) were also studied for the production of biodiesel [ 28 , 29 ], ethanol [ 30 ], and hydrogen gas [ 31 ]. However, none of these studies have reported the generation of robust biofilms along with gas bubbles and the resulting water surface-floating ability. Therefore, this phenotype is likely uncommon in microalgae belonging to the Chlorococcum species and could be a unique characteristic of FFG039. The biofilm of Botryosphaerella sp. AVFF007 also captures gas bubbles, which could, in part, contribute to its water surface-floating ability [ 16 ], suggesting that these strains might share similar floating mechanisms. In this study, chemical mutagenesis of water surface-floating microalgae was demonstrated to improve their biomass and lipid productivities. The biofilms on the surface of the culture medium may attenuate light penetration and inhibit photosynthesis of the cells beneath the biofilms. Therefore, it is reasonable to select the mutants with less photosynthetic pigment for better light penetration. This strategy has frequently been employed to improve the productivity of microalgal biomass in large-scale photobioreactors [ 21 ], whereas this study, to the best of our knowledge, is the first to apply this strategy to water surface-floating microalgae. The microalgal cells were dispersed by sonication and then exposed to mutagens, following which the PMs were selected. As expected, the selected PMs exhibited higher biomass productivity than the wild type, and their lipid content was comparable to that of the wild type, leading to higher lipid productivity than the wild type. PM11, which is the most promising clone among the PMs, exhibited a biomass productivity of 5.4 ± 0.2 g/(m 2 day), and a lipid productivity of 1.9 ± 0.1 g/(m 2 day), which were 1.7-fold and 1.9-fold higher than those of the wild type. These productivities were also higher than those of AVFF039, which had productivities comparable to those of other green microalgae ( Table S2 ) [ 16 ]. PM11 retained the biofilm formation and floating phenotypes, suggesting that the chemical mutation did not impair the advantage of the water surface-floating microalgae in biofuel production. In this study, we selected gravimetry as a method of quantification of dry biomass and lipids, because a number of studies have employed the same method [ 17 , 18 , 19 , 32 , 33 , 34 ] and confirmed that the relative standard deviation (RSD) is usually less than 10%. We should note that other methods are also worthwhile for better quantification, such as thin layer chromatography (TLC), in which stained lipid spots are analyzed by imaging tools [ 35 ], and gas-chromatography mass spectrometry (GC-MS), in which respective target ion responses are measured [ 36 ]. Although the gravimetry data might be still preliminary, this study highlights the promise of the mutant strains for improved biofuel production. In future studies, further investigations will be performed to add more evidence."
} | 2,560 |
34695974 | PMC8539789 | pmc | 7,316 | {
"abstract": "This paper deals with analytical modelling of piezoelectric energy harvesting systems for generating useful electricity from ambient vibrations and comparing the usefulness of materials commonly used in designing such harvesters for energy harvesting applications. The kinetic energy harvesters have the potential to be used as an autonomous source of energy for wireless applications. Here in this paper, the considered energy harvesting device is designed as a piezoelectric cantilever beam with different piezoelectric materials in both bimorph and unimorph configurations. For both these configurations a single degree-of-freedom model of a kinematically excited cantilever with a full and partial electrode length respecting the dimensions of added tip mass is derived. The analytical model is based on Euler-Bernoulli beam theory and its output is successfully verified with available experimental results of piezoelectric energy harvesters in three different configurations. The electrical output of the derived model for the three different materials (PZT-5A, PZZN-PLZT and PVDF) and design configurations is in accordance with lab measurements which are presented in the paper. Therefore, this model can be used for predicting the amount of harvested power in a particular vibratory environment. Finally, the derived analytical model was used to compare the energy harvesting effectiveness of the three considered materials for both simple harmonic excitation and random vibrations of the corresponding harvesters. The comparison revealed that both PZT-5A and PZZN-PLZT are an excellent choice for energy harvesting purposes thanks to high electrical power output, whereas PVDF should be used only for sensing applications due to low harvested electrical power output.",
"conclusion": "5. Conclusions The main aim of this paper was to compare the effectiveness of materials commonly used in energy harvesting operations using a single DOF model and at the same time analyze the effect of used mode shape function on simulation results. The single DOF model of a cantilever piezoelectric harvester in both bimorph and unimorph configurations was derived based on Euler-Bernoulli beam theory. Output of the model was confronted with available experimental data obtained from three different piezoelectric harvesters (PZT-5A bimorph, PZZN-PLZT bimorph and PVDF unimorph) and showed a good degree of accuracy. It is obvious that the presented model of an energy harvester can be used for various piezoelectric materials. Therefore, the developed single DOF analytical model represents a simple and very helpful tool for designing piezoceramic vibration energy harvesters. Moreover, it could easily be employed to check if a particular kinetic energy harvester provides sufficient output power for the intended application. Or inversely, the model could be used to design a piezoceramic harvester with optimized operational parameters and dimensions due to the model’s ability to predict the amount of harvested energy in particular operational conditions. Moreover, the developed model can easily be extended to support calculations of strain and stress levels within harvester’s layers for further assessments concerning strength and fatigue limits. The model is primarily intended for operations of the harvester at frequencies where vibrations consist mostly of the first mode shape, since it offers the best operational conditions for energy harvesting (no strain nodes). If higher vibrational modes are of interest, the developed model can easily be extended by supplementing their respective shape functions and using the superposition principle. It was found that the quality of the used approximative function for the first mode shape affects the beam model’s stiffness in such a way so that simpler (less accurate) approximative functions force the beam model to behave stiffer, i.e., its resonant frequency being shifted to higher values. Also, the way how device layers and electrodes are assembled can also affect the stiffness of the system and could potentially result in a weak nonlinearity, which was observed in one of our experiments. Nevertheless, a typical assembly of layers and the approximative shape function used in this work (the shape of the beam’s first vibrational mode without a tip mass) still shows a good degree of accuracy. The single DOF model itself poses as a very effective tool whose main advantages are low computer resources usage and the ability to calculate transient responses for arbitrary time-dependent loads. Both these features were employed in an energy harvesting effectivity comparison of the three materials used in the scope of this work. The materials comprised of PZT-5A, which is known nowadays to be one of the best materials for energy harvesting purposes, and PZZN-PLZT and PVDF which are used in our laboratory for designing vibration energy harvesting devices. The comparison was split into two parts with respect to forcing: case of simple harmonic vibrations and case of random vibrations. The case of simple harmonic vibrations was carried out so that the harvesters’ dimensions were tuned in order to achieve common value of seismic mass and resonant frequency for all three harvesters and at the same time the harvesters had same volume of polarized piezoelectric material. In case of random vibrations the harvesters were subjected to a non-harmonic and non-periodic vibrations typical for wearables applications. Results from both comparison cases showed that the piezoceramic harvesters (PZT-5A and PZZN-PLZT) are a perfect choice for energy harvesting applications, though geometry and electrical load must be optimized. On the contrary, the PVDF harvester is not suitable for energy harvesting purposes due to very low values of harvested energy despite many recent papers reporting the otherwise, and its potential lies in sensing applications.",
"introduction": "1. Introduction Energy harvesting is more than 20 years a hot topic in the field of wireless sensing [ 1 ] since it allows for converting various energy types from ambient sources into an electrical one. Although the amount of such harvested energy is usually small (tens of μW up to several mW), it can be used as a source of electrical power for modern, low power-consuming sensors that are typically used in wearable electronics and industrial applications [ 2 ] where powering using cables is not feasible (either due to a hazardous environment or complex setup). Piezoelectric kinetic energy harvesters in the form of a vibrating multilayer structure with piezoelectric layers [ 3 ] are commonly used in vibration energy harvesting applications, where the structure is excited by an ambient source of vibrations. The main task of kinetic energy harvesters is then to transform the mechanical energy of ambient vibrations, mainly those of machine frames or human body movement, into useful electrical energy by means of the direct piezoelectric phenomenon. The main goal in the field of energy harvesting is to design a kinetic energy harvester which is capable to generate a sufficient amount of electrical energy in a particular vibratory environment [ 4 ] in order to power some other electronic equipment. However, each application has its requirements or limits for dimensions and weight of the harvester; the principle of energy harvesting can be used practically everywhere, for example in the field of medicine [ 5 ], wearables [ 6 ], portables [ 7 ], aircrafts [ 8 ], structural health monitoring of railways [ 9 ] or bridges [ 10 ]. It has been proved many times that for harvesting energy from ambient vibrations the kinematically excited cantilever beam is one of the most effective designs of a piezoelectric energy harvester. The fundamental and also the most important issue of this solution is the choice of a suitable piezoelectric material for effective electromechanical conversion. The review of commonly used piezoelectric materials and structures for energy harvesting purposes is summarized in publication [ 11 ], where it is shown that not only the material itself but also the intended operational mode significantly affects the amount of harvested power due to a great variation in piezoelectric coefficients. The highest piezoelectric coefficients (generally, the higher the coefficients, the higher the amount of harvested power) are provided by piezoceramic materials [ 12 ], especially those based on lead (PZT). As a non-toxic alternative, new lead-free piezoceramic materials have been developed which are based on multifunctional Perovskite [ 13 ] or structured layers made of Barium and Titanate [ 14 ]. Besides these piezoceramic materials which are inherently very brittle and stiff, there are also more flexible materials such as macro-fiber composites which are very promising in the area of strain energy harvesting [ 15 ] and piezopolymers which are summarized in review paper [ 16 ]. An example of a cantilever harvesting device based on a piezoelectric polymer (PVDF) is presented in paper [ 17 ] and the effectivity of PVDF in energy harvesting applications is nowadays widely discussed [ 6 ]. In conclusion, the two most important factors that determine the effectiveness of a vibrational energy harvesting device are the used piezoelectric material and the harvester’s geometry. Many recent works were concerned about the optimal harvester’s geometry for selected piezoelectric material, e.g., [ 18 ], but the effectivity of various piezoelectric materials has not been widely discussed yet. Both the selection of efficient piezoelectric material and suitable geometry of the harvester can be solved with an appropriate model of the piezoelectric resonator. For this reason, the presented paper is organized as follows. First, derivation of an analytical beam model of a kinematically excited piezoelectric cantilever in both bimorph/unimorph configurations which also respects the dimensions of used tip mass. This beam model is subsequently reduced to a single degree-of-freedom (DOF) system using the first mode shape function. Although, the derivation of a coupled electromechanical model was published several times, e.g., [ 19 , 20 , 21 , 22 , 23 ], here, we also show the effect of chosen mode shape function which is used in reducing the beam model into single DOF model. Then, the model is verified with 3 different experimental results. Finally, the main aim of this paper is to provide a methodology based on a verified model that can be used to compare the effectivity of materials commonly used in energy harvesting applications."
} | 2,647 |
39475286 | PMC11577766 | pmc | 7,319 | {
"abstract": "ABSTRACT Wildfires are unpredictable disturbances with profound effects on soil properties and microbial communities within forest ecosystems. However, knowledge of post-fire microbial communities in karst forests remains limited. In this study, microbial amplicon sequencing techniques were employed to investigate the impact of wildfires on the composition, diversity, function, and co-occurrence network of soil microbial communities in karst forest landscapes and to identify the key soil physicochemical factors affecting the post-fire microbial communities. The wildfire affected the fungal community to a greater extent than the bacterial community, with the former shifting from a dominance of Basidiomycota to Ascomycota at the phylum level, while the relative abundance of Actinobacteria increased significantly in the bacterial community. Moreover, the wildfire increased the α-diversity of the microbial community and changed the β-diversity. Network analysis indicated significant reductions in the complexity of microbial community networks and the hub microbiome in burned soils compared to those of unburned soils. Functional predictions indicated an increase in the highly abundant functional taxa of chemoheterotrophic and aerobic chemoheterotrophic bacteria, along with a significant rise in saprotrophic functional fungal taxa following the fire. In addition, soil organic matter, total nitrogen, total phosphorus, and soil water content emerged as key soil physicochemical factors affecting post-fire soil microbial communities in the karst forest. Overall, this study revealed the structural and functional characteristics of soil microbial communities and their key influencing factors after a fire in a karst forest, which will provide a valuable theoretical basis for ecosystem restoration after a wildfire. IMPORTANCE Despite the significant impacts of wildfires on forest ecosystems, most existing studies have largely focused on boreal and Mediterranean coniferous forest types, with limited research on the impacts of coniferous and broadleaf forest types in subtropical karst regions. This study reveals the effects of wildfires on soil microbial communities of coniferous and broadleaf forest types in a karst forest. The results of this study not only improve the understanding of the effects of wildfires on the composition, diversity, function, and network of soil microbial communities but also provide a meaningful theoretical basis for post-fire ecosystem restoration in the karst forest.",
"conclusion": "Conclusions This study revealed the composition, diversity, function, and network of microbial communities after a wildfire in different forest types of the karst ecosystem, demonstrating that soil physicochemical factors can affect the microbial communities after fire to varying degrees. SOM, TN, TP, and SWC have significant impacts on the composition, function, and network of post-fire microbial communities and were identified as the key soil physicochemical factors influencing the short-term recovery of microbial communities in the woodland after a fire in the karst region. These findings can provide certain guidance for the restoration and management of post-fire soil ecosystems in the karst forest region. However, this study only provides characteristics of the microbial communities in the short-term post-fire stage in the karst forest region. Future studies should focus on long-term site-based monitoring to gain a deeper understanding of the succession patterns and recovery mechanisms of the post-fire microbial community.",
"introduction": "INTRODUCTION The increasing warming of the global climate and extreme weather conditions have led to an increased frequency of wildfires. From 2000 to 2021, China experienced an average of 5,688 wildfires per year, resulting in the burning of up to 73,964 ha of forest ( http://www.stats.gov.cn/tjsj/ndsj/ ) ( 1 ), making it one of the countries with the most serious forest fires. Wildfires not only destroy valuable forest resources but also cause severe disturbance to the microbial communities of forest ecosystems ( 2 ). These communities play important roles in nutrient cycling, energy flow, and information transfer and thus have a substantial functional impact on the recovery process after wildfires ( 3 – 5 ). Therefore, understanding how wildfires affect the soil microbial communities in forest ecosystems is of great significance for post-fire ecosystem restoration. In general, the effects of forest fires on soil microbial communities are categorized into direct and indirect effects. Direct effects include cell death due to high-temperature stress, leading to a reduction in microbial relative abundance and diversity ( 6 , 7 ), whereas indirect effects mainly involve changes in soil physicochemical properties, nutrients, and the microenvironment after fire. These indirect effects impact microbial communities mainly through changes in soil organic matter (SOM), soil water content (SWC), soil pH, and extracellular enzyme activity, as well as the accessible nutrient supply ( 8 , 9 ). All of these effects have implications for carbon and nitrogen cycling, mineral nutrient transformations, and other processes that regulate the forest ecosystem during restoration ( 10 ). During succession of forest ecosystems after a fire, the direct effects of the fire on microbial communities may become less important, while the indirect effects exert a major influence on microbial communities and play a positive role in the recovery of post-fire forests ( 11 ). Microorganisms not only are involved in nutrient cycling and organic matter transformation in the soil but also modify the soil environment; this microbial-mediated modification of soil properties can in turn have an impact on the composition of the microbial community and thus has important ecosystem implications. There is a close interaction between microbial communities and soil properties ( 12 ). Fire leads to the burning of numerous organic materials such as vegetation and litter, thereby increasing available nutrients and altering soil physicochemical properties ( 13 , 14 ), ultimately causing changes in soil microbial communities. Previous studies showed that post-fire pyrogenic organic matter (PyOM) formation results in significant changes in microbial compositional abundance, with a significant enrichment in Actinobacteria and a significant decline in Ascomycetes abundance in the post-fire soil ( 15 , 16 ). In addition, post-fire changes in soil pH result in significant changes in microbial diversity and were identified as key determinants in post-fire microbial recovery ( 17 ). With respect to functional impacts, Sun et al. ( 18 ) showed that the relative contribution of ectomycorrhizal fungi to the community increased during post-fire recovery, which was mainly driven by the available SOM after the fire. Furthermore, post-fire soil microbial networks and the hub microbiome were found to be affected by the changes in SWC, pH, and nitrogenous compounds occurring during post-fire recovery ( 13 , 14 , 19 ). Despite numerous studies on the interactions between soil microbial communities and soil physicochemical factors after fire, these studies have mainly focused on ecosystems of coniferous forest types in boreal and Mediterranean regions, whereas short-term post-fire recovery studies for tropical and subtropical ecosystems of both coniferous and broadleaf forest types remain limited ( 13 , 19 ). In particular, such studies are notably lacking for karst forest areas. The karst forest ecosystems in southern China have unique geomorphic structures, characterized by high habitat heterogeneity, shallow and discontinuous soils, multilayered ecological space, and extreme fragility; however, these forests play an important role in preventing soil erosion ( 20 ). Guizhou Province exhibits a typical karst landscape development with a widespread distribution area. Between 2001 and 2020, a total of 15,754 fires were recorded in the province, covering an area exceeding 175,000 hm 2 . These fires have had a significant impact on the forest ecosystem, resulting in substantial direct and indirect economic losses. Therefore, determining the best strategy to quickly renew and restore forest land after a fire is not only an important ecological issue but also an important issue for sustainable development. In this study, we used 16S rRNA and internal transcribed spacer (ITS) gene amplicon sequencing to investigate the characteristics of soil microbial communities in burned and unburned coniferous and broadleaf forests in karst regions of China, and we further explored the impact of the post-fire soil physicochemical properties on microbial community construction. The aims of this study were therefore to (1) evaluate the immediate impact of forest fires on the composition, diversity, function, and symbiotic network of soil bacterial and fungal communities in karst forests (2); compare the similarities and differences in soil bacterial and fungal communities between coniferous and broadleaf forests following fire; and (3) identify the key soil physicochemical factors influencing the soil microbial communities in karst forests after fires. These findings will provide information for understanding the short-term recovery of soil microbial communities in burned areas of karst forest ecosystems.",
"discussion": "DISCUSSION Composition and diversity of microbial communities after a wildfire At the class taxonomic level, the same dominant microbial groups were found in the karst forest before and after burning. Bacteria were mainly composed of Alphaproteobacteria, Gammaproteobacteria, Vicinamibacteria, and Acidobacteriae, and fungi were mainly composed of Agaricomycetes. These groups have also been identified as the common dominant microbial groups in other studies related to burned areas ( 19 , 21 ). Notably, we found that the relative abundances of Proteobacteria and Acidobacteriota increased after burning in the broadleaf forest, which is consistent with the findings of Yang et al. ( 22 ), whereas the relative abundances of these taxa decreased after burning in the coniferous forest, which is also consistent with previous findings ( 6 , 21 ). The reason for this reduction may be due to the differences in above-ground vegetation types and soil physicochemical properties after a fire in different forest types. In both coniferous and broadleaf forest types, Actinobacteria significantly increased after burning. Actinobacteria can utilize PyOM following a fire to ensure their survival ( 23 ). Therefore, our results suggest that bacterial communities undergo different changes according to the forest type in the post-fire karst region. We found that, after the burning of two types of forests, the relative abundance of Basidiomycota decreased while that of Ascomycota increased, which is consistent with numerous previous studies ( 4 , 22 , 24 ). This phenomenon may be due to the burning off of litter and the stress imposed on mycorrhizal fungi by fire, as well as the chemical inhibition of compounds following a fire ( 19 , 25 , 26 ), which significantly affect the survival of Basidiomycota fungi. Another reason might be that certain classes of Ascomycota fungi such as Leotiomycetes possess thermophilic and heat-resistant properties ( 4 , 27 ), making them more likely to survive and adapt to post-fire conditions. This study revealed a slight increase in the α-diversity of microbial communities in the post-fire forest, which is in contrast to previous studies reporting a decrease in α-diversity following forest fires ( 4 , 11 ). This discrepancy may be attributed to differences in study areas or sampling periods. This study focused on short-term changes in microbial communities in the forest land within the karst rocky desertification area after burning, whereas prior studies primarily conducted long-term monitoring in other regions. Moreover, our β-diversity analysis demonstrated a significant community separation between burned and unburned forest soil microbial communities ( P < 0.001), indicating that microbial communities in the karst rocky desertification forest region are unlikely to recover to pre-fire levels within 260 days. Functions of microbial communities after a wildfire Our study found that wildfires can cause significant changes in soil microbial community networks. The complexity of the co-occurrence network of microorganisms decreased after the forest fire, which is consistent with the findings of Dai et al. ( 14 ) and Yang et al. ( 19 ). After a forest fire, the number of microorganisms in the hub microbiome drastically decreased, along with a notable change in the composition of the taxa. This result is attributed to the fact that the hub microbiome in unburned forests is primarily composed of fungi that perform ectomycorrhizal functions. These fungi are a crucial component in maintaining the ecological functions of forest ecosystems, but they are also highly susceptible to the impact of wildfires, including Scleroderma , Rhodanobacter , and Nordella ( 28 – 30 ). Another reason might be that the hub microbiome after a forest fire has specific characteristics such as the ability to survive under extreme environmental conditions and heat resistance. Therefore, the hub microbiome comprises taxa that can not only quickly adapt to the soil environment after a forest fire but can also effectively utilize the nutrients released by the fire to promote their growth and reproduction. Indeed, previous studies confirmed that members of the hub microbiome, including Occallatibacter , Aspergillus , and Oidiodendron , identified after the forest fire in this study have the above characteristics ( 31 – 33 ). In addition, the proportion of all fungi nodes increased after the forest fire, indicating that fungi may play an important role in the short-term succession process in the woodland after a fire. Our results also showed that the complexity of the microbial network and the hub microbiome in the broadleaf forest after a wildfire in the karst region were significantly higher than those in the coniferous forest, indicating that the recovery speed of the broadleaf forest may be faster than that of the coniferous forest in the short-term recovery process after a fire. The soil microbial functional groups in the forest lands were affected to varying degrees after forest fires. For bacterial communities, the low-abundance functional groups in the two types of forests were more susceptible to the impact of the fire than the high-abundance groups. The low-abundance functional groups underwent a notable shift from being predominantly parasitic before the fire to primarily phototrophic and photoautotrophic after the fire. The high-abundance functional groups were relatively less affected and remained predominantly chemoheterotrophic. This result is consistent with the findings of Cheng et al. ( 7 ). This phenomenon may be due to the large-scale destruction of vegetation after the fire, resulting in exposed soil and direct sunlight. For fungal communities, both high- and low-abundance functional groups were greatly affected after the forest fire possibly because fungal communities are more sensitive to forest fires, as highlighted above. Several studies have also confirmed this effect ( 15 , 18 , 19 ). Additionally, forest fires led to a shift of the fungal symbiotic community toward saprotrophic functional groups. For instance, the relative abundance of ectomycorrhizal fungi decreased after the fire, whereas the relative abundance of saprotrophic fungi significantly increased following the fire. The reason for this difference might be due to the damage or death caused by the burned above-ground vegetation to the underground root systems, thus greatly reducing the symbiotic fungi associated with plant roots. In addition, the dead root systems may recruit or enrich more microorganisms with saprotrophic characteristics. For example, the hub microbiome comprising Udeniomyces , Aspergillus , and Chloridium identified after the forest fire in this study has saprophytic properties, and related studies have confirmed that the hub microbiome not only survives in harsh environments but also has the ability to produce high levels of lignocellulosic enzymes ( 34 – 36 ). Additionally, we found that the functional groups for aromatic compound degradation significantly increased after the forest fire, which may be attributed to the additional input of aromatic compounds into the soil after a fire. Microbial groups capable of utilizing such substances are more likely to adapt to this environment and grow rapidly, such as the hub microbiome comprising Mycobacterium and Aspergillus identified in this study. Previous research has confirmed the ability of these microorganisms to decompose and transform aromatic compounds ( 36 , 37 ). Effect of karst forest soil physicochemical factors on microbial communities after a wildfire Wildfires burn a substantial proportion of the vegetation within the woodland and return nutrients to the soil, leading to changes in soil physicochemical factors such as SOM, TN, TP, pH, and β-glucosidase in the forest. We found that SOM, TN, TP, and SWC increased significantly in the coniferous woodland soils after fire, whereas these factors decreased significantly in the burned broadleaf woodland. This may be due to the differences in nutrient mineralization rates between the two forest types following exposure to high-temperature and high-heat conditions, along with the different mechanisms of post-fire woodland restoration ( 38 – 40 ). In addition, there were no significant differences in the activities of the three extracellular enzymes urease, Mn-peroxidase, and laccase between burned and unburned forests. This result differs from previous studies that reported a significant decrease in extracellular enzyme activities after burning ( 41 ). This difference may be related to the sampling sites after the fire, as Rasmussen et al. ( 41 ) reported significant differences in extracellular enzyme activities at different locations within the same habitat in Mississippi, USA. This difference could also reflect the specific life activities of functional microbial groups in the burned forest, as some members of the identified hub microbiome are efficient producers of these enzymes. Numerous studies have shown that soil physicochemistry significantly affects the α-diversity of microbial communities after a fire; however, we found no significant correlation between the α-diversity of microbial communities after the fire with the soil physicochemical factors for both forest types. This result may be attributed to the unique climatic environment of the karst region and the greater influence of surface vegetation than soil physicochemical properties during the restoration process ( 42 , 43 ). Wildfires lead to varying degrees of changes in microbial functions, which can be attributed to changes in the soil physicochemical properties such as the soil moisture, carbon, and nitrogen levels caused by wildfires. This study found that SOM, TN, and SWC significantly affected the function of soil microbial communities. Cheng et al. ( 44 ) found that SWC and TN were the main influencing factors of the functional diversity of soil microbes in burn sites of a Larix gmelinii forest in a cold temperate zone. Our results thus support these conclusions. Additionally, soil TN and SWC are essential for maintaining various functions and material conversions in soil microorganisms, and significant changes in their contents play a key role in the functional diversity of microbial communities ( 44 – 46 ). We found that the activities of extracellular carbohydrate hydrolases (β-glucosidase and laccase) also significantly influence the function of soil microorganisms in burnt coniferous forests. This phenomenon may be due to microorganisms secreting more hydrolases to break down unstable SOM to help adapt to the post-fire environment and maintain their survival, thereby leading to functional shifts in specific microorganisms. SOM, TN, TP, and SWC were simultaneously and significantly associated with the hub microbiome, functions, and networks and were defined as the key soil physiochemical factors impacting post-fire microbial communities in karst regional woodlands. SOM is an important carbon reservoir and nutrient for post-fire microbial metabolism, which plays an important role in soil carbon and nutrient cycling ( 47 , 48 ). VanderRoest et al. ( 16 ) simulated severe wildfires through a controlled “pyrocosm” approach and identified that post-fire biodegradable SOM was enriched with heterotrophic microbes and also contained many substrates that support microbial metabolism. In addition, the decomposition and mineralization of SOM by post-fire microorganisms are also inseparable from TN and TP ( 14 , 17 ), whereas SWC is an important predictor of the post-fire soil microbial community composition ( 11 , 13 , 33 ). Our study also found a significant positive correlation among these four key soil physicochemical factors (Fig. S6B). This indicates that the recovery of post-fire microorganisms in the karst forests is a result of the synergistic effects of multiple soil physicochemical factors. Conclusions This study revealed the composition, diversity, function, and network of microbial communities after a wildfire in different forest types of the karst ecosystem, demonstrating that soil physicochemical factors can affect the microbial communities after fire to varying degrees. SOM, TN, TP, and SWC have significant impacts on the composition, function, and network of post-fire microbial communities and were identified as the key soil physicochemical factors influencing the short-term recovery of microbial communities in the woodland after a fire in the karst region. These findings can provide certain guidance for the restoration and management of post-fire soil ecosystems in the karst forest region. However, this study only provides characteristics of the microbial communities in the short-term post-fire stage in the karst forest region. Future studies should focus on long-term site-based monitoring to gain a deeper understanding of the succession patterns and recovery mechanisms of the post-fire microbial community."
} | 5,611 |
28773929 | PMC5456629 | pmc | 7,321 | {
"abstract": "Graphene-modified materials have captured increasing attention for energy applications due to their superior physical and chemical properties, which can significantly enhance the electricity generation performance of microbial fuel cells (MFC). In this review, several typical synthesis methods of graphene-modified electrodes, such as graphite oxide reduction methods, self-assembly methods, and chemical vapor deposition, are summarized. According to the different functions of the graphene-modified materials in the MFC anode and cathode chambers, a series of design concepts for MFC electrodes are assembled, e.g., enhancing the biocompatibility and improving the extracellular electron transfer efficiency for anode electrodes and increasing the active sites and strengthening the reduction pathway for cathode electrodes. In spite of the challenges of MFC electrodes, graphene-modified electrodes are promising for MFC development to address the reduction in efficiency brought about by organic waste by converting it into electrical energy.",
"conclusion": "4. Conclusions and Outlook Although a series of challenges faces the practical application of MFCs, the development of MFCs for wastewater treatment has concerned many researchers for a long time. There has been undeniably great progress in enhancing the performance of MFCs, and various reactor configurations have been designed to explore the operating principle of MFCs, e.g., H-shaped MFCs [ 78 ], air-cathode single-chamber MFCs [ 20 , 79 ] and single-chamber membrane-free MFCs [ 18 ]. The reactions all occur on the surface of electrode materials, whether the microorganism catalytic degradation of organic matter in the anode chamber or the ORR of the electron acceptor in the cathode chamber. Hence, the excellent properties of electrode materials are the essential factor in determining the electricity generation performance of MFCs, which is significant for their practical application. Due to the excellent physical, chemical, and biological performance of graphene and its compounds, as mentioned earlier, they have gradually become one of the most popular materials in MFC research. In this review, we have shown that 3D porous graphene-based materials have a higher specific surface area, an improved electrical conductivity, and a more outstanding catalytic performance than traditional materials such as carbon paper, carbon cloth, and graphite particles, which indicates that graphene and its compounds are the ideal electrode materials in MFCs. With the use of modified graphene in the anode chamber, large defective sites are produced on these materials, enhancing their catalytic performance and electron transfer ability and reducing the polarization phenomenon. In the cathode chamber, the superior active sites and the hydrophilicity of graphene-modified materials are beneficial for the interconnection between the catalyst and the electrolyte; this interconnection can improve the ORR rate. The major areas for future studies are to develop the affordability, superior electrical conductivity, high catalytic activity, and outstanding biocompatibility of 3D graphene materials. In the operation of MFCs, multiple purified exoelectrogens were separated from anaerobic sludge, such as Escherichia coli , Pseudomonas aeruginosa , and Shewanella oneidensis MR-1 leading to the situation that the mechanism of action among these purified exoelectrogens was not clarified (i.e., whether they act by synergistic effect or inhibiting effect), which is important to enhance the performance of MFCs. Partial exoelectrogens have a good electricity generation ability in alkaline condition that enables their use in some conductive compounds, such as PANI, which could then be developed as a new technology to address the alkaline industrial wastewater. In addition, the use of expensive membranes, such as Nafion 117, severely limits the practical application of some MFCs. Therefore, developing a membrane-free MFC or seeking alternative mediator materials, such as salt bridges, instead of these costly membranes could significantly reduce the construction cost of MFCs. To improve the utilization efficiency of oxygen, the cathode can be designed as a biodegradation reactor, such as an aerobic biological reactor in sewage treatment plants, which can effectively accept the electrons coming from the anode chamber while degrading contaminants in the wastewater. The study of MFCs is an interlaced subject, built on the basis of physics, chemistry, and biology. MFCs can be combined with other technologies, such as membrane bioreactor (MBR) technology that can not only enhance the electrical generation performance but also increase the removal rate of contaminants [ 145 ]. In addition, power management systems (PMS) that harvest energy are crucial for the scale-up and practical application of MFCs [ 146 ]. However, there are many factors that impact the electrical generation performance of MFCs, for instance, the external resistance, nutrient solution system, microorganism species, and electrode materials. Hence, it is difficult to compare systems to determine which one is better than the others. With the further development of graphene-modified MFCs, a steady operating MFC system can be developed to evaluate the performance of MFCs. Overall, graphene and its compounds, as the ideal electrode materials, increase the performance of MFCs, and they can serve as the core technology needed to address organic wastewater in the future.",
"introduction": "1. Introduction Although fossil fuels are essential for economic development, the increasing consumption of fossil fuels has come with significant negative drawbacks, such as air pollution and global warming, which have consequently accelerated the exploration of renewable energy technologies by scientists [ 1 , 2 ]. The microbial fuel cell (MFC) is a promising recently developed device that can convert the chemical energy stored in organic fuels as nutritional substrates into electrical energy though the metabolism of microorganisms, while degrading the organic contaminant to an extent [ 3 , 4 , 5 ]. Compared with traditional chemical fuel cells [ 6 ], large-scale organic substrates, such as municipal treatment plants [ 7 , 8 ], agriculture wastes, solid wastes from dairy farms [ 9 , 10 , 11 , 12 ], and even human waste [ 13 , 14 ], can be used as fuels in MFCs. However, many factors affect the performance of MFCs, including the chemical substrate, ionic concentration, proton exchange material, catalyst, internal resistance, electrode spacing, and electrode materials [ 15 , 16 , 17 , 18 , 19 , 20 ]. The low extracellular electron transfer (EET) efficiency between the microorganism and the electrode is still the main bottleneck limiting the practical applications of MFCs, resulting in poor energy conversion efficiency and low power density [ 21 ]. Generally, there are two main approaches used to cope with these problems: one is to improve the electrode properties by surface treatment [ 22 ], for example, by means of microbial reduction [ 23 ], electrostatic incorporation or ionic liquid functionalization [ 21 ]; the other is to fabricate new electrode materials to enhance the EET at the anode or the catalytic activity at the cathode [ 22 ]. Because microbial growth on metal surfaces can accelerate metallic corrosion in an aqueous environment [ 24 ], carbon-based materials, such as carbon cloth [ 3 ], carbon paper [ 25 ], carbon felt [ 26 ], carbon fiber [ 27 ] and graphite particles [ 4 , 28 ], are used as electrode materials for MFCs. With the rapid developments in materials research, graphene, a new and renowned member of the carbon family, has been adopted in MFC electrodes because of its excellent physical and chemical properties, for instance, its high specific surface area (2630 m 2 ·g −1 ) [ 29 , 30 , 31 ], outstanding electrical conductivity [ 32 ], and extraordinary biocompatibility [ 33 ]. Additionally, it has been widely used in Li-ion batteries [ 34 ], supercapacitors [ 35 , 36 ], sensors [ 37 ], electrochemical catalysts [ 38 ], and oil sorbents [ 39 ]. Graphene, discovered in 2004 by Geim and Novoselov [ 40 ], is a two-dimensional (2D) single-atom-thick flat material consisting of sp 2 hybridized carbon atoms arranged in a honeycomb lattice [ 41 , 42 , 43 ]; at present, graphene is the thinnest material in the world. Its theoretical thickness value, band length, and bond angle are 0.335 nm, 0.142 nm, and 120°, respectively [ 44 ]. However, the monolayer graphene sheet is known to irreversibly agglomerate or form multilayer graphite through strong π-π stacking and van der Waals interactions [ 45 ]. Graphene oxide (GO) is an important derivative of graphene that contains heavy epoxy and hydroxyl functional groups on the basal planes, and carbonyl and carboxyl groups on the sheet edges, which make it possible to fabricate graphene based materials on a large scale [ 42 ]. Although these functional groups increase its hydrophilic character, the conjugated sp 2 network of the individual graphene basal planes is disrupted and the electrical properties of GO are decreased. Therefore, the removal of oxygen functional groups can enhance the electrical conductivity of graphene modified materials. Compared with 2D graphene, three-dimensional (3D) graphene structures have outstanding characteristics, e.g., a large accessible surface, excellent mechanical strength, and remarkable flexibility [ 46 ], which can enhance the number of microorganisms on their surface and are the ideal electrode materials in MFCs. Recently, different reduction methods have been developed to obtain macroscopic 3D graphene structures from GO sheets, e.g., hydrothermal reduction, chemical vapor deposition, chemical reduction, electrochemical reduction, and microbial reduction [ 47 ]. In this review, the methodologies on the synthesis of graphene-based electrodes, and the design principles of a desirable MFC electrode are covered. The influence of graphene-modified electrodes (anodes and cathodes) on the electricity generation of MFCs is analyzed and discussed."
} | 2,534 |
31493408 | null | s2 | 7,322 | {
"abstract": "Bacteria have developed numerous protection strategies to ensure survival in harsh environments, with perhaps the most robust method being the formation of a protective biofilm. In biofilms, bacterial cells are embedded within a matrix that is composed of a complex mixture of polysaccharides, proteins, and DNA. The gram-positive bacterium Bacillus subtilis has become a model organism for studying regulatory networks directing biofilm formation. The phenotypic transition from a planktonic to biofilm state is regulated by the activity of the transcriptional repressor, SinR, and its inactivation by its primary antagonist, SinI. In this work, we present the first full-length structural model of tetrameric SinR using a hybrid approach combining high-resolution solution nuclear magnetic resonance (NMR), chemical cross-linking, mass spectrometry, and molecular docking. We also present the solution NMR structure of the antagonist SinI dimer and probe the mechanism behind the SinR-SinI interaction using a combination of biochemical and biophysical techniques. As a result of these findings, we propose that SinI utilizes a residue replacement mechanism to block SinR multimerization, resulting in diminished DNA binding and concomitant decreased repressor activity. Finally, we provide an evidence-based mechanism that confirms how disruption of the SinR tetramer by SinI regulates gene expression."
} | 351 |
31493408 | null | s2 | 7,323 | {
"abstract": "Bacteria have developed numerous protection strategies to ensure survival in harsh environments, with perhaps the most robust method being the formation of a protective biofilm. In biofilms, bacterial cells are embedded within a matrix that is composed of a complex mixture of polysaccharides, proteins, and DNA. The gram-positive bacterium Bacillus subtilis has become a model organism for studying regulatory networks directing biofilm formation. The phenotypic transition from a planktonic to biofilm state is regulated by the activity of the transcriptional repressor, SinR, and its inactivation by its primary antagonist, SinI. In this work, we present the first full-length structural model of tetrameric SinR using a hybrid approach combining high-resolution solution nuclear magnetic resonance (NMR), chemical cross-linking, mass spectrometry, and molecular docking. We also present the solution NMR structure of the antagonist SinI dimer and probe the mechanism behind the SinR-SinI interaction using a combination of biochemical and biophysical techniques. As a result of these findings, we propose that SinI utilizes a residue replacement mechanism to block SinR multimerization, resulting in diminished DNA binding and concomitant decreased repressor activity. Finally, we provide an evidence-based mechanism that confirms how disruption of the SinR tetramer by SinI regulates gene expression."
} | 351 |
38363846 | PMC10871528 | pmc | 7,324 | {
"abstract": "Reservoir computing is a powerful neural network–based computing paradigm for spatiotemporal signal processing. Recently, physical reservoirs have been explored based on various electronic devices with outstanding efficiency. However, the inflexible temporal dynamics of these reservoirs have posed fundamental restrictions in processing spatiotemporal signals with various timescales. Here, we fabricated thin-film transistors with controllable temporal dynamics, which can be easily tuned with electrical operation signals and showed excellent cycle-to-cycle uniformity. Based on this, we constructed a temporal adaptive reservoir capable of extracting temporal information of multiple timescales, thereby achieving improved accuracy in the human-activity-recognition task. Moreover, by leveraging the former computing output to modify the hyperparameters, we constructed a closed-loop architecture that equips the reservoir computing system with temporal self-adaptability according to the current input. The adaptability is demonstrated by accurate real-time recognition of objects moving at diverse speed levels. This work provides an approach for reservoir computing systems to achieve real-time processing of spatiotemporal signals with compound temporal characteristics.",
"introduction": "INTRODUCTION The neural network–based computing paradigm offers the possibility of processing information with extremely high energy efficiency and handling the avalanche of spatiotemporal data generated at an ever-increasing rate ( 1 – 4 ). Recently, recurrent neural network (RNN) has shown remarkable strength in dealing with spatiotemporal data ( 5 , 6 ). Nonetheless, the complicated internal connections of the conventional RNN require massive memory for storage. Coupled with the vanishing and exploding gradient issues, the training and hardware implementation of RNN are complex and costly ( 7 , 8 ). As a variant of RNN, reservoir computing (RC) is proposed by replacing the complex recursive network with relatively fixed nonlinear reservoirs ( 9 , 10 ). A typical RC system consists of reservoirs for mapping input into high-dimensional space and a readout network for analyzing the high-dimensional reservoir states. In general, only the readout network requires training, which can be accomplished with simple methods such as the back-propagation algorithm and linear regression. Therefore, the training cost is substantially reduced ( 11 ). Another crucial characteristic of reservoirs is the fading memory (FM), which means that the reservoir state is influenced by input from the recent past while independent of that from the far past ( 12 – 14 ). With this feature, RC can reveal the implicit correlation of temporal patterns. In recent years, RC has succeeded in a broad range of applications, such as temporal pattern classification and prediction ( 15 – 17 ). Various physical systems have been proposed for hardware implementation of the reservoir, including the coupled mechanical oscillator ( 18 , 19 ) and photonic system ( 20 , 21 ). However, these systems are still relatively restricted in processing speed, power efficiency, and scalability, and are mostly not compatible with the modern integrated circuit ( 22 ). Benefitted by the theory developed for RC with a single nonlinear node ( 14 , 23 , 24 ), constructing RC systems with electronic devices, including memristor devices ( 25 – 29 ), ferroelectric devices ( 30 – 32 ), nanowire network ( 33 – 35 ), and spintronic oscillator ( 36 – 38 ), is successfully demonstrated with the advantages of small area, simple structure, and high power efficiency. The intrinsic nonlinearity and randomness of the device can be naturally used to promote the richness and separability of reservoir states and thus can further improve the system performance. By far, these hardware RC systems have shown strong capability in diverse temporal processing tasks, such as recognizing spoken digits ( 26 , 31 , 39 ) and medical signals ( 40 , 41 ), as well as forecasting chaotic systems ( 42 – 44 ). However, a fundamental problem still lies in the way of its further application. The temporal dynamics of the reservoir, as key hyperparameters, would notably affect the system performance in a highly task-specific manner ( 14 , 41 , 45 , 46 ), which means that for the optimization of the system, deliberate adjustment of the reservoir is generally required. In contrast, the electronic devices implementing physical reservoirs are used as a “black box,” resulting in the characteristics of reservoirs being determined by the fabrication process, and therefore are normally not adaptive. This conflict has rendered most precedent systems only applicable for tasks with a timescale matching specific physical devices. A compromising solution addressing this challenge is to rescale the recorded temporal data to adapt to the temporal dynamics of the hardware RC system ( 26 ). However, this will cost excessive latency and storage overhead and is especially not in favor of real-time processing. A superior solution would be the development of electronic devices capable of reservoir adaptation, which has been experimentally shown to be effective in extracting information from multiple temporal levels ( 41 , 45 ). However, it still requires the knowledge of the optimal reservoir characteristics for specific tasks in advance, which places a severe obstacle in the actual applications because the timescale is usually an unknown and constantly varying quantity. In this work, we fabricated an indium gallium zinc oxide (IGZO) thin-film transistor (TFT) with controllable temporal dynamics, which shows excellent cycle-to-cycle uniformity as well as the four-order-tunable timescale. On the basis of the device, we experimentally demonstrated a temporal adaptive reservoir where the operation voltage can easily adjust the hyperparameter timescale (τ). By matching the temporal dynamics and applying a multi-timescale strategy, this RC system outperforms its temporal invariant counterpart in classifying six human activities, improving the accuracy from 84.2 to 96.7%. Moreover, we proposed a closed-loop architecture that can dynamically adjust the reservoir hyperparameters by self-feedback to flexibly adapt the system to present input signals. We experimentally verified this design in a real-time motion classifying task containing vastly different speed conditions. This work shows a back-end-of-line (BEOL) compatible approach for implementing the temporally controllable hardware RC system based on a simple physical structure. It further presents a promising solution to self-adaptively optimize the hyperparameters in the RC systems toward a broad input spectrum.",
"discussion": "DISCUSSION In this work, we proposed a closed-loop architecture to address the limitation in self-adaptation of C, and experimentally performed real-time optimization in a hardware RC system. The demonstration is based on IGZO-channel TFT with ZnO interlayer. This device shows excellent cycle-to-cycle uniformity and endurance of 10 8 and, the controllable FM characteristics. With this device, we constructed the temporal adaptive reservoir, where the temporal dynamics of the reservoir are changed by varying operation voltage. We proved its advantage over the trivial temporal invariant reservoir by applying this reservoir to a HAR task. We tested the reservoirs with various temporal constant τ and found that by choosing the optimal condition, accuracy is improved from 84.2 to 93.3%. On the basis of this, we applied a mixed-τ scheme to capture multi-timescale information and obtained a higher accuracy of 96.7%. Furthermore, we proposed a closed-loop architecture that can perform self-adaptive optimization of the reservoir hyperparameters. The closed-loop RC system is experimentally demonstrated in the task of real-time recognition of objects moving at speeds varying within a wide range. For example, the self-adaptive transition is performed as the moving speed changes from 0.8 to 8 m/s. This work proves that the performance of hardware RC systems can be prominently improved by optimizing the hyperparameters. It also shows a method for the self-adaptive optimization of the RC system without additional external information. We mainly focused on the temporal characteristics of the reservoir by far, while precedent researches have also revealed the importance of other hyperparameters including the input gain, the scaling factor, and the feedback density ( 20 , 21 ). Inspired by this, the effects of other hyperparameters remain to be elucidated in the context of closed-loop architecture. Future works may also involve more refined closed-loop architecture for more sophisticated scenarios. Besides, integrating the RC systems with closed-loop structures on chips will be a valuable and appealing topic."
} | 2,215 |
35362974 | null | s2 | 7,327 | {
"abstract": "Defining chemical properties of intracellular organelles is necessary to determine their function(s) as well as understand and mimic the reactions they host. However, the small size of bacterial and archaeal microorganisms often prevents defining local intracellular chemical conditions in a similar way to what has been established for eukaryotic organelles. This work proposes to use magnetite (Fe"
} | 99 |
27175224 | null | s2 | 7,328 | {
"abstract": "Many fish exhibit rheotaxis, a behavior in which fish orient themselves relative to flow. Rheotaxis confers many benefits, including energetic cost savings and interception of drifting prey. Despite the fact that most species of fish school during at least some portion of their life, little is known about the importance of rheotactic behavior to schooling fish and, conversely, how the presence of nearby conspecifics affects rheotactic behavior. Understanding how rheotaxis is modified by social factors is thus of ecological importance. Here we present a mathematical model in the form of an all-to-all, coupled-oscillator framework over the non-Euclidean space of fish orientations to model group rheotactic behavior. Individuals in the model measure the orientation of their neighbors and the flow direction relative to their own orientation. These measures are corrupted by sensory noise. We study the effect of sensory noise and group size on internal (i.e., within the school) and external (i.e., with the flow) disagreement in orientation. We find that under noisy environmental conditions, increased group size improves rheotaxis. Results of this study have implications for understanding animal behavior, as well as for potential applications in bio-inspired engineering."
} | 320 |
31698841 | PMC6918236 | pmc | 7,332 | {
"abstract": "Plant attributes have direct and indirect effects on soil microbes via plant inputs and plant-mediated soil changes. However, whether plant taxonomic and functional diversities can explain the soil microbial diversity of restored forest ecosystems remains elusive. Here, we tested the linkage between plant attributes and soil microbial communities in four restored forests ( Acacia species, Eucalyptus species, mixed coniferous species, mixed native species). The trait-based approaches were applied for plant properties and high-throughput Illumina sequencing was applied for fungal and bacterial diversity. The total number of soil microbial operational taxonomic units (OTUs) varied among the four forests. The highest richness of fungal OTUs was found in the Acacia forest. However, bacterial OTUs were highest in the Eucalyptus forest. Species richness was positively and significantly related to fungal and bacterial richness. Plant taxonomic diversity (species richness and species diversity) explained more of the soil microbial diversity than the functional diversity and soil properties. Prediction of fungal richness was better than that of bacterial richness. In addition, root traits explained more variation than the leaf traits. Overall, plant taxonomic diversity played a more important role than plant functional diversity and soil properties in shaping the soil microbial diversity of the four forests.",
"conclusion": "5. Conclusions Disentangling of plant-microbe interactions during restoration can improve our understanding of successful restoration trajectories. This article reports the understanding of plant microbial interaction in four different plantation forests in southern China. Plant taxonomic diversity explained better the soil microbial diversity implies the importance of maintaining high species diversity (richness and diversity) in order to maintain high microbial richness. High microbial richness can have a synergistic impact on the success of forest restoration. Forest restoration with leguminous species was associated with fungal richness and diversity. This result indicated that restoration with productive species might have significant impact on the diversity soil microbes and outcomes of the restoration. In addition, the plant functional traits (aboveground and Belowground) contributed to soil microbial community, belowground traits exhibited a stronger effect than aboveground traits. Such information suggests that restored species through their plant mediated inputs can assist in the sustainable management of restored forest ecosystems. Indeed, microbial communities have an important role in ecosystems functioning and the knowledge of plant-microbe interactions of restored forest ecosystem might trigger restoration success around the globe. Future research on the links between plant attributes and soil microbial communities should focus on a wider range scale, including other microbial groups (e.g., archaea, protists, microeukaryotes and other small metazoans), exogenous nutrient deposition, and climate change regimes.",
"introduction": "1. Introduction Biodiversity conservation and protection of land from degradation are the key strategies behind ecosystem restoration [ 1 ]. Half of the world’s degraded tropical forests are restored through reforestation or are converted to secondary plantation forests [ 2 ]. These degraded forests are restored with monoculture or mixed plantations as a process of the restoration strategy [ 3 ]. However, the success of ecological restoration largely depends upon the interactions between above and belowground communities driving ecosystem processes [ 1 , 4 ]. We lack empirical evidence on plant-microbe linkage from restored forest landscapes, especially those exploring the relative contribution of plant attributes and soil properties to explain soil microbial diversity. Plantation types, preference of habitats and quality of plant inputs are the key factors that influence soil microbial communities during restoration [ 5 , 6 ]. Vegetation types having contrasting diversity of plant communities will have a distinct effect on soil properties through plant inputs, which in turn has diverse effects on soil microbial communities [ 3 , 4 ]. Moreover, the individual species alter soil chemical properties through the species-specific chemistry of litter inputs, which inevitably affects soil microbial communities [ 7 , 8 ]. The presence of a higher diversity of productive species can also have a strong influence on soil functions and microbial diversity [ 9 , 10 ]. For example, long-term restoration with Pinus massoniana and Eucalyptus spp. causes soil degradation, which profoundly influences soil microbial communities [ 4 ]. In Pinus elliottii plantations, fungal biomass increased, but bacterial biomass decreased due to microecological imbalance and a gradual decrease in the quality of inputs during long-term restoration in subtropical China [ 3 ]. Specific species were also found to have a contrasting impact on specific taxa of bacteria during forest conversion from native to teak plantations [ 11 ]. Therefore, how soil fungal and bacterial communities respond to vegetation restoration with respect to different plantations needs to be explored. We hypothesized that different plantations in our study restored with contrasting species will vary in terms of plant richness and diversity and the distinct effect of plant-mediated soil changes will influence soil fungal and bacterial communities. Plant taxonomic and functional diversity affect soil microbial diversity during restoration [ 5 , 12 ] by altering the available resources [ 13 , 14 ], niche differentiation and resource partitioning [ 15 , 16 ]. In natural ecosystems, soil fungal diversity is inevitably dependent on species richness indicating the importance of individual species that generate complementary belowground niches by their inputs [ 17 , 18 , 19 ]. Roy-bolduc et al. [ 20 ] found a strong positive relationship between plant diversity and soil fungal diversity. In contrast, Shi et al. [ 21 ] reported an inverse relationship between tree diversity and fungal diversity. Bacterial diversity was also found to decrease with the species richness along latitudinal gradients where environmental conditions act as the predominant driver [ 13 ]. Distinct bacterial groups were found in broadleaf forest and coniferous forest during restoration in the mountain region of China [ 5 ]. Therefore, plant taxonomic diversity will have an impact on soil microbial diversity, but to better understand the effect of plant attributes on the soil microbial community, it is necessary to determine the relative contribution of plant functional diversity along with plant taxonomic diversity under the vegetation change regime [ 12 ].Community-weighted mean (CWM) traits of dominant species and multi-trait functional dispersion (FD) are the two important hypotheses by which we can find the effect of plant functional diversity on soil microbial diversity. Ecosystem functions and properties are intensely influenced by the traits of the dominant species [ 22 , 23 ]. Based on the biomass ratio hypothesis (CWM traits) [ 24 ], attributes of dominant species in a community regulate ecosystem properties, and the quality and quantity of litter traits of dominant species play a key role in regulating soil microbial richness [ 25 ]. Moreover, based on the niche complementarity hypothesis, FD also plays a pivotal role in several ecosystem functions [ 26 ]. FD can facilitate niche partitioning, which creates more available resources for niche spaces and is beneficial for microbial communities [ 16 ]. Therefore, it is imperative to include both CWM and FD to better understand the effect of plant functional trait diversity on microbial communities. We hypothesized that under this circumstance, the proportional contribution of plant taxonomic diversity will be greater than that of the functional diversity in explaining soil microbial diversity. In recent years, trait-based approaches have been applied for the prediction of fungal and bacterial diversities at the individual plant [ 9 ], community [ 27 ], and regional scales [ 28 ]. The richness of some groups of soil microbes, such as mycorrhizal fungi and archaeal ammonia oxidizers, is influenced by plant functional traits [ 29 , 30 ]. Plant functional traits such as specific leaf area (SLA) and leaf nitrogen concentrations (LN) can alter soil properties via the input of litter and detritus, which can affect the richness of soil microbes [ 31 , 32 ]. Moreover, the diversity and richness of soil microbial communities are affected by interspecific variation in both the quality and quantity of resource inputs [ 33 , 34 , 35 ]. Microbial properties were found to have a closer linkage with belowground root traits than leaf traits in the grasslands of Europe [ 27 ]. Root nutrients such as root nitrogen concentration (RN) and the root carbon: nitrogen ratio (C: N) were found to have a direct effect, whereas leaf traits (SLA, shoot N, and C: N) indirectly influence soil fungal and bacterial diversity [ 31 ]. Previous studies have mostly included only aboveground plant functional traits [ 25 , 28 , 32 ] or, in some instances, a few belowground traits to describe the plant-microbe interaction [ 27 ]. However, the combination of both aboveground and belowground traits and their proportional contribution in explaining soil microbial diversity would be a more viable option. Therefore, we hypothesized that the root traits might contribute more than the leaf traits in explaining belowground soil microbial diversity. Here, we explored the role of plant diversity (taxonomic and functional diversity) and plant functional traits (aboveground and belowground) as predictors of belowground soil fungal and bacterial diversities in four subtropical plantation forests of southern China. The four types of forests included an Acacia mangium (AM) forest, a mixed Eucalyptus species (EE) forest, a mixed coniferous species forest (MC) of Cunninghamia lanceolata and Pinus massoniana, and a mixed Schima species (NS) forest. Specifically, we aimed to address three questions: (1) How does the soil microbial (fungi and bacteria) community structure differ among the four plantations? (2) What is the relative importance of plant functional diversity vs. taxonomic diversity for microbial diversity? (3) Which plant traits (aboveground or belowground) contribute more to determining the soil microbial community composition? We hypothesized that (1) restoration with diverse species through plant-mediated soil changes might alter soil microbial diversity [ 5 , 6 , 7 ]; (2) taxonomic diversity might contribute more than functional diversity due to species-specific linkage between microbes and individual plants [ 17 , 18 , 19 ]; and (3) belowground traits might contribute more due to their close association with soil microbes [ 27 ].",
"discussion": "3. Discussion Forest restoration with different plant species influenced soil microbial diversity. Fungal richness and diversity were highest in the AM plantation. AM is the leguminous forest with abundant Acacia mangium species. Restoration with leguminous plants ( Acacia ) fixes atmospheric nitrogen, which is added to the soil and might influence higher fungal communities in AM forests [ 36 , 37 ]. Moreover, SOC and TN ( Table A5 ) were significantly higher in the AM forest than in the other forests, which perhaps influences the soil fungal community. Among the edaphic factors, soil fertility (for example, SOC, TN) is the key edaphic factor that influences soil microbial richness [ 13 ] because SOC and TN provide energy to soil fungi [ 6 ], which increases their activity and subsequently increases fungal diversity [ 7 , 8 ]. The CWM traits of the leaves (SLA, LC, and LP) and roots (RDMC, RC, and RN) were higher in AM forest ( Table 1 ) suggesting the presence of more exploitative species that stimulate rapid acquisition and turnover, thus facilitating fungal composition [ 31 , 38 ]. Moreover, the multi-trait FD and single trait functional diversity measures were highest in the AM forest, which indicates that the higher resource availability (nutrients entering the soil via the plant parts) was present in this site, leading to more availability of niche space for fungi [ 39 , 40 , 41 , 42 ]. Therefore, vegetation restoration with leguminous plants compared to other plants might facilitate the activity of the fungal community due to increased resource availability [ 43 ]. Bacterial richness and diversity were lower in AM forest. The quality of nutrients and lower plant diversity in the AM forest compared to the other forests might be attributed to the lower bacterial richness and diversity. In contrast, bacterial richness was highest in EE plantation where the plant diversity was also found highest among the studied forests. The higher soil bacterial richness in the EE forest might be influenced by the quality of the substrate entering the soil [ 44 ]. Leaf and root C: N ratios were significantly lower in the EE forest than in the other forests ( Figure A3 ); therefore, the quality of the inputs was higher in the EE forest, which might facilitate higher bacterial richness [ 13 ]. Moreover, bacterial richness may be more associated with the abundance of specific plants (e.g., Melicope pteleifolia , Gardenia jasminoides ) in the EE forest. Plant taxonomic diversity explained more of the soil microbial diversity than the functional diversity and soil properties, which supports our second hypothesis. Soil microbial (fungi and bacteria) richness increased with species richness. This might be due to the greater diversity of organic substrates, resources and carbon compounds for soil microbes [ 40 , 43 , 45 ]. The VPA analysis also indicated that plant taxonomic diversity better explained fungal diversity than bacterial diversity ( Figure 5 ). This result indicates that the individual tree effect is much stronger in shaping fungal and bacterial diversity [ 5 , 6 , 12 , 40 ]. Fungal richness was found to increase with the increase of the abundance of some specific species ( Illex asprella and Clerodendrum fortunatum ) those were predominant (after pioneer species) across the four forests. In the case of plant functional diversity, it better explained fungal than bacterial diversity. Complex litter biopolymers are decomposed by fungi, and thus, the properties of litter should be reflected in fungi, which might be one of the reasons that plant functional diversity better predicts fungi than bacteria [ 46 ]. Again, the greater dependency of fungi on plant products [ 47 ] and fungal mycelia from the plant rhizosphere extending to the bulk soil might be the other probable reasons for the better prediction of fungi [ 46 ]. In the present study, there was a 3% contribution from soil properties in explaining fungal and bacterial diversity. Recent article reported 2-4% contribution of soil properties in explaining variations of fungal and bacterial diversity in a species-rich grassland [ 48 ]. Greater contributions of plant attributes than soil properties in explaining soil microbial diversity were also reported by several empirical studies [ 6 , 46 , 49 ]. Usually, soil chemical properties, latitudinal distances, or climatic factors (MAP and MAT) are the dominant drivers of soil microbial communities [ 35 , 50 , 51 ]. However, the climatic factors and soil types of the four forests were nearly uniform in this study, with the only difference being the species that were introduced or planted. This might be the probable reason of plant attributes being the dominant drivers of soil microbial communities other than the environmental drivers. Increase in soil fertility (SOC and TN)through plant-mediated inputs increased soil microbial richness by providing more resources and available niches [ 13 ]. Functional traits have shown significant effects on soil microbial populations. The aboveground traits (e.g., SLA, LDMC, VD, LC, LN, LP) and belowground traits (SRL, RDMC, RC, RN, RP) were important predictors of soil microbial communities. Overall, belowground traits better explained soil microbial diversity than aboveground traits. Plant functional traits related to photosynthesis, carbon chemistry of litter and roots, hydraulic conductance and nutrient acquisition can profoundly influence soil microbes [ 31 , 32 , 38 , 52 ]. Moreover, the functional traits found to influence the soil microbial communities were considered as the fundamental indicators controlling the quality and quantity of inputs that stimulate soil fertility [ 38 , 53 ]. These functional traits promote niche partitioning and rhizodeposition via the diversity of resources, which in turn influence fungal and bacterial richness. Studies that reported the link between fungi and bacteria richness and diversity with plant functional traits (SLA, LDMC, RDMC, Shoot C, N, root C, N) from grassland and forest ecosystems [ 31 , 32 , 52 ] were consistent with our findings. A recent article reported functional traits related to nutrient acquisition can better predict fungal and bacterial diversity [ 31 ]. In this study, belowground traits were the ones that mostly influenced the variation observed, due to a closer association of soil microbes with roots [ 27 ]. Additionally, as root traits determine the quality and quantity of plant carbon and nitrogen supply for the activity of soil microbial communities. At last, plant responses to soil properties that directly or indirectly influence soil microbial communities can be reflected through the root traits. In summary, the present study showed that plant attributes are fundamental in driving microbial diversity of restored forests. Among the plant attributes, plant taxonomic diversity explained more variation of the fungal and bacterial diversity than plant functional diversity. Specific species were also found to influence fungal and bacterial richness. Plant functional diversity alone (individual effect) only explained of fungal not bacterial diversity. Indeed, our results suggest that experiments studying the influence of plant functional diversity on soil microbial communities should include both above and below ground plant functional traits. Furthermore, fora better understanding of the effect of plant functional diversity on soil microbial communities, plant functional traits of understorey species need to be incorporated."
} | 4,636 |
32946129 | null | s2 | 7,334 | {
"abstract": "Cooperation can be favoured through the green-beard mechanism, where a set of linked genes encodes both a cooperative trait and a phenotypic marker (green beard), which allows carriers of the trait to selectively direct cooperative acts to other carriers. In theory, the green-beard mechanism should favour cooperation even when interacting partners are totally unrelated at the genome level. Here, we explore such an extreme green-beard scenario between two unrelated bacterial species-Pseudomonas aeruginosa and Burkholderia cenocepacia, which share a cooperative locus encoding the public good pyochelin (an iron-scavenging siderophore) and its cognate receptor (green beard) required for iron-pyochelin uptake. We show that pyochelin, when provided in cell-free supernatants, can be mutually exchanged between species and provide fitness benefits under iron limitation. However, in co-culture we observed that these cooperative benefits vanished and communities were dominated by P. aeruginosa, regardless of strain background and species starting frequencies. Our results further suggest that P. aeruginosa engages in interference competition to suppress B. cenocepacia, indicating that inter-species conflict arising from dissimilarities at the genome level overrule the aligned cooperative interests at the pyochelin locus. Thus, green-beard cooperation is subdued by competition, indicating that interspecific siderophore cooperation is difficult to evolve and to be maintained."
} | 371 |
37091415 | PMC10116497 | pmc | 7,335 | {
"abstract": "A series of highly\nflexible and environmentally friendly\ncomposites\nbased on polydimethylsiloxane (PDMS) filled with 200 nm size ferroelectric\nBaTiO 3 (BTO) particles at different concentrations (from\n7 to 23 vol %) have been fabricated by a simple dispersion method.\nThe dielectric, piezoelectric, and ultrasonic properties have been\nstudied. The ferroelectric state of BTO was confirmed by differential\nscanning calorimetry and ultrasonic spectroscopy. The addition of\nBTO into PDMS strongly affects the dielectric properties of the composites.\nAt low temperatures close to 160 K, the PDMS matrix exhibits a dielectric\nanomaly related to a dynamic glass transition, which shifts to higher\ntemperatures as the BTO content increases due to the strong interaction\nbetween polymer chains and nanoparticles. Ultrasonic measurements\ndemonstrate the appearance of a piezoelectric voltage signal on a\nthin plate of the composite with the highest available filler concentration\n(23 vol %) under longitudinal stress applied by a 10 MHz ultrasonic\nwave. As a result, at room temperature, the detected signal is characterized\nby output voltage and specific stored energy values of 10 mV and 367.3\nMeV/m 2 , respectively, followed by a further increase with\ncooling to 35 mV at 150 K. The proposed BTO/PDMS composite system\nis thus a potential candidate for nanogenerators, namely, a simple,\nflexible, and lead-free device converting high-frequency (10 MHz)\nmechanical vibrations into electrical voltage.",
"conclusion": "Conclusions In\nsummary, using a simple dispersion method,\na series of flexible\ncomposites with ferroelectric BTO nanoparticles at different concentrations\n(7–23 vol %) in a PDMS matrix have been prepared. The dielectric,\npiezoelectric, and ultrasonic properties of the obtained composites\nhave been studied. At high temperatures, the dielectric permittivity\nspectra of the BTO/PDMS composites are frequency-independent over\na wide range, while at low temperatures a pronounced anomaly appears\nthat is associated with the dynamic glass transition in PDMS. Moreover,\ndue to the strong interaction between PDMS chains and BTO nanoparticles,\nthe glass transition temperature derived from the Vogel–Fulcher\nlaw shifts to higher temperatures from 121.8 to 130 K with the increasing\nBTO content (from 7 to 23 vol %). Ultrasonic measurements at 10 MHz\nalso confirm the presence of the glass transition of the polymer,\nwhich is expressed by the steepest wave velocity dispersion and attenuation\nmaximum near the transition temperature. Additionally, the composites\nhave been checked for piezoelectric response under longitudinal stress\napplied by a 10 MHz ultrasonic wave: below 400 K, the piezoelectric\nvoltage signal appears only for a sample with the highest concentration\n(23 vol %), while for the others the signal is absent. The output\nvoltages are 10 and 35 mV above the glass transition and in the glass\nstate, respectively. Having the dielectric permittivity data of the\nmaterial, the value of the specific stored energy has been estimated\nand, in the particular case of room temperature, it is 367.3 MeV/m 2 . The output characteristics are not very high due to the\nlow power of ultrasonic excitation and the decrease in the piezoelectric\ncoefficient as the frequency increases. It is expected that the use\nof more powerful vibration sources will lead to a significant increase\nin the output voltage. The BTO/PDMS composite system is a prospective\ncandidate for high-frequency nanogenerator and energy harvester applications.",
"introduction": "Introduction The rapid decline of fossil fuels, as\nwell as the resulting environmental\npollution, has led to an active search for alternative “green”\nenergy sources. Since piezoelectric (in 2007) 1 and triboelectric (in 2012) 2 nanogenerators\nwere first mentioned, they have continued to receive attention due\nto the attractive prospect of creating efficient energy-harvesting\ndevices by converting a ubiquitous energy source such as mechanical\nmovement (acoustic waves, fluid flow, body deflection, compression,\nvibration, etc.) into useful electricity. 3 − 5 Polymers,\nwhether synthetic or organic, that are sufficiently elastic\nto respond to any mechanical pressure are ideal mediums for holding\npiezoelectric particles. Particularly, flexible polydimethylsiloxane\n(PDMS), which is chemically stable and exhibits a low glass transition\ntemperature, simple processing, and durability in the cured state,\nhas already been identified as one of the most popular polymers for\nnanogenerator design. 6 − 10 In turn, piezoelectric lead-free perovskite-type barium titanate\nBaTiO 3 is a balanced option in terms of environmental safety\nand a high piezoelectric coefficient. 6 , 9 , 11 Thus, for more than a decade already there\nhas been an active fight\nfor every volt of output power, preferably with the simultaneous presence\nof other benefits, such as lightness, ease of large-scale fabrication,\nand low cost of the final product. The output performance of a nanogenerator\nbased on polymer composite is determined, and therefore can be controlled,\nby many factors: 12 qualitative and quantitative\ncompositions, homogeneity of filler distribution, 13 , 14 chemical doping, 15 − 19 morphology of the particles, 13 , 14 , 20 , 21 and 3D spatial ordering of particles\nin the matrix. 22 − 25 Regarding morphology, Jian et al. 14 obtained\nan output power value of 260 V for a PDMS-based system with hierarchical\nBTO flowers due to the high local stress at the petals, while for\nBTO in the form of nanoparticles, 20 nanowires, 21 and nanocubes 13 the\nvalues were 3.2, 7, and 126.3 V, respectively. Chemical doping was\nfound to be an effective tool to adjust the piezoelectric coefficients\nof the materials and increase their ability to harvest energy. 15 − 19 Besides, the load transfer efficiency can also be improved by forming\nthree-dimensional (3D) skeletons, such as foams, 22 porous aerogels, 23 , 24 and sponge-like structures. 25 Finally, hybrid nanogenerators, which operate\nusing both tribo- and piezoelectric effects, demonstrate enhanced\noutput characteristics. 26 − 28 However, most research\nis devoted to the quasi-static frequency\nrange (i.e., less than 100 Hz). The latter corresponds to wasted tiny\nbiomechanical motions and wind and water vibrations, while for many\nexisting devices nanogenerators could be beneficial for converting\nhigh-frequency vibrations (e.g., from cars, aircrafts, and electric\nengines) into useful electrical energy. Besides, the majority of studies\non mechanical energy harvesting mainly cover the region of room temperature\nwithout data on a parameter as important for defining the properties\nof any material as the dielectric permittivity, and its complex investigation\nas a function of temperature and frequency is extremely rare. 28 Therefore, to fill the gaps, the current study\naims to perform complex investigations (dielectric and ultrasonic)\non composite systems containing ferroelectric barium titanate (BaTiO 3 ) nanoparticles in a polydimethylsiloxane (PDMS) polymer matrix\nin order to determine their suitability as nanogenerators for high-frequency\n(10 MHz) applications.",
"discussion": "Results and Discussion Characterizations\nof BTO/PDMS Composites The volume\ndistribution of BTO nanoparticles within the PDMS matrix is quite\nhomogeneous, as can be seen in the scanning electron microscopy (SEM)\nimage ( Figure 2 a) for\nthe composite with the maximum available filler concentration of 23\nvol %. Differential scanning calorimetry (DSC) analysis was performed\nfor the BTO nanopowder and the 23 vol % BTO/PDMS composite (see Figure 2 b). The peak near\n400 K for both curves is associated with a ferroelectric–paraelectric\nsecond-order phase transition, which reveals the ferroelectric ordering\nof BaTiO 3 . The latter is also confirmed by the XRD pattern\nobtained at room temperature for the 23 vol % BTO/PDMS composite (see Figure 2 c). All peaks are\nidentified as the P 4 mm BaTiO 3 tetragonal (PDF-2 no. 00-081-2201) phase ( Figure 2 c, black symbols). The peak\nnear 45° is characterized by a splitting, which corresponds to\nthe reflections from the (002) and (200) planes of BaTiO 3 (see the inset in Figure 2 c). This indicates the tetragonal distortion of the BaTiO 3 crystal lattice typical of the ferroelectric state. 31 , 32 Although the considered BaTiO 3 grain size of 200 nm is\nclose to the critical size, which means that presumably at least some\nof the smaller grains are cubic at room temperature, the observed\nsplitting proves the presence of the ferroelectric phase as well. Figure 2 (a) Scanning\nelectron microscopy (SEM) image of the 23 vol % BTO/PDMS\ncomposite. (b) Differential scanning calorimetry curves of BTO nanopowder\nand the 23 vol % BTO/PDMS composite. (c) X-ray diffraction pattern\nof the 23 vol % BTO/PDMS composite at room temperature. Dielectric Properties below Room Temperature The temperature\ndependencies (below room temperature) of the real part of the complex\ndielectric permittivity at different frequencies for all studied BTO/PDMS\ncomposites are presented in Figure 3 a, while the imaginary parts as a function of temperature\nat 75 kHz are presented in Figure 3 b. Figure 3 Temperature dependencies (below RT) of (a) the real part\nof the\ndielectric permittivity at different frequencies, and (b) the imaginary\npart at the 75 kHz frequency for 7, 11, 15, and 23 vol % BTO/PDMS\ncomposites. The permittivity at room temperature\n(300 K) is\ncharacterized by\ntwo features: on the one hand, an increase in the value of the real\npart with the loading of BTO particles, and on the other hand a frequency\nindependence. However, at low temperatures close to 160 K, an anomalous\nbehavior is detected, which is associated with α-relaxation,\nalso called the dynamic glass transition of PDMS. 33 , 34 Considering the thermal behavior of pure PDMS, 35 the increase in BTO content is manifested by a shift of\nthe dielectric loss to higher temperatures. Besides, the position\nof the peak considered for each particular sample is frequency-dependent\n(see Figure 4 a for\na 15 vol % BTO/PDMS composite). Namely, as frequency ν increases,\nthe maximum shifts to higher temperatures according to the Vogel–Fulcher\nlaw (see Figure 4 b): 36 1 where T ref is\nthe glass transition temperature of the polymer, T max is the temperature at which the dielectric losses\nare the highest, k B is the Boltzmann constant,\nν 0 is the frequency at very high T max , and E VF is the activation\nenergy. Figure 4 (a) Typical temperature dependencies of dielectric losses at different\nfrequencies, illustrated by that of the BTO/PDMS composite at 15 vol\n% taken as an example. (b) Frequency at maximum dielectric losses\nmeasured as a function of temperature for BTO/PDMS composites at 7,\n11, 15, and 23 vol % . The solid lines are the best fits according\nto the Vogel–Fulcher equation ( eq 1 ). The parameters of the\nbest fits to the experimental\ndata for all\nBTO/PDMS composites are presented in Table 1 (marked by superscript (1)). In the composites,\naccording to Table 1 , the glass transition temperature increases with the concentration\nof BTO particles, while the activation energy decreases. This increase\nin T ref is related to the slowing of local\nsegmental dynamics, 37 i.e., to the reduction\nof the mobility of polymer chains located near the filler due to strong\ninterfacial interactions. Similar behavior has been frequently observed\nin various PDMS-based composite systems. 35 , 38 − 40 Table 1 Glass Transition Temperatures and\nVogel–Fulcher Activation Energies of BTO/PDMS Composites BTO, vol % ν 0 , THz E VF (1) / k B , K T ref (1) , K τ 0 , ns E VF (2) / k B , K T ref (2) , K 7 4.8 642.7 123.9 0.06 433.9 121.8 11 1.18 575.6 125.3 0.1 399.3 124.8 15 0.88 512.4 127.7 0.17 347.0 125.5 23 0.18 437.7 132.5 0.27 328.7 130.0 The frequency spectra at a temperature close to the\nglass transition\nalso exhibit relaxation behavior, namely, a maximum of the imaginary\npart of the dielectric permittivity is observed. Figure 5 a shows the frequency dependencies\nof ε″ for a 15 vol % BTO/PDMS composite at different\ntemperatures. Figure 5 (a) Frequency spectra of the imaginary part of the dielectric\npermittivity\nat different temperatures for the 15 vol % BTO/PDMS composite. The\nsolid lines are the best fits according to the Cole–Cole equation\n( eq 2 ). (b) Temperature\ndependencies of the relaxation time obtained from Cole–Cole\nfits for the 7, 11, 15, and 23 vol % BTO/PDMS composites. The solid\nlines are the best fits according to the Vogel–Fulcher equation\n( eq 1 ). To analyze the frequency spectra of ε″,\nthe Cole–Cole\nequation 41 can be used (see solid lines\non Figure 5 a): 2 where\nε* = ε′ – iε ″\nis the complex dielectric permittivity,\nε ∞ is the dielectric permittivity at infinite\nfrequency, Δε is the dielectric strength\n(i.e., the difference between the static dielectric permittivity and\nε ∞ ), τ is the relaxation time, ω\nis the angular frequency defined as ω = 2 πν , and α is the parameter responsible for the width of the relaxation\ntime distribution. The fitted parameters α and ε ∞ were\nfound to be weakly dependent on temperature in the considered range\nfrom 150 to 170 K, while the relaxation time increased with cooling.\nThe obtained τ values can be plotted as a function of temperature\nand then fitted according to the Vogel–Fulcher equation ( eq 1 ), taking into account\nthe reciprocal relationship between τ and the relaxation frequency\nν r , i.e., τ = 1/ν r . The temperature\ndependencies of the relaxation time, as well as the best Vogel–Fulcher\nfits for all the composites investigated, are presented in Figure 5 b. The relaxation\ntime fitting parameters are collected in Table 1 (marked by superscript (2)) and are in agreement\nwith the data obtained using the ε″( T ) dependencies because the dielectric dispersion is very narrow. As a result, due to the strong interactions between BTO nanoparticles\nand PDMS polymer chains, the properties of the composites can be adjusted\nby controlling the BTO concentration. In particular, the glass transition\ntemperature monotonously shifts to higher temperatures as the BTO\ncontent increases. Dielectric Properties above Room Temperature Dielectric\nmeasurements for higher temperatures (above room temperature) have\nalso been performed, and the corresponding temperature dependencies\nof the real part of the complex dielectric permittivity at the frequency\nof 1 MHz for all composites studied are presented in Figure 6 a. In this case, there is no\nneed to show the ε′( T ) dependencies\nat different frequencies as in Figure 3 a, since the frequency dispersion of the dielectric\nspectra is absent far from the glass transition. This fact is also\nconfirmed by the broadband spectra shown in Figure 6 b for the 23 vol % BTO/PDMS composite at\ndifferent temperatures in a wide range including low (20 Hz to 1 MHz)\nand high (10–300 MHz) frequencies. Thus, for all concentrations\nconsidered, ε′ is characterized by a monotonic decrease\nwith heating. Figure 6 (a) Temperature dependence (above RT) of the real part\nof the dielectric\npermittivity at the frequency of 1 MHz for the 7, 11, 15, and 23 vol\n% BTO/PDMS composites. (b) Frequency dependence (from 20 Hz to 300\nMHz) of the real part of the dielectric permittivity for the 23 vol\n% BTO/PDMS composite at different temperatures. Piezoelectric Properties To investigate the influence\nof BTO nanoparticles on the glass transition and freezing–melting\ndynamics of the PDMS-based composite, ultrasonic measurements have\nbeen performed. The longitudinal ultrasonic wave velocity V and attenuation values Δα were measured at a frequency of 10 MHz upon cooling from room temperature\nto 150 K and upon subsequent heating to 430 K. The results for the\nhighest available concentration of 23% of BTO in the PDMS-based composite\nare presented in Figure 7 a and b. Figure 7 Temperature dependencies of (a) the ultrasonic wave velocity and\n(b) the ultrasonic wave attenuation measured at a frequency of 10\nMHz for the 23 vol % BTO/PDMS composite upon cooling (blue curve)\nand heating (red curve). Upon cooling, the anomalous\ntemperature behavior\nof both V and Δα is\ncentered around\n190 K. This anomaly has a characteristic shape usually observed in\nthe crossing of the glass transition, where the steepest velocity\ndispersion region corresponds to the attenuation maximum. 42 In glass formation and other systems with strong\ndynamic relaxation, the sound velocity and attenuation are also frequency-dependent.\nSuch systems, when probed ultrasonically, are less rigid on time scales\ngreater than the stress relaxation time (in contrast to much shorter\ntime scales), and the energy dissipation is the highest when the probing\nfrequency is close to the inverse value of the relaxation time. Upon\nheating, the Δα maximum splits into two\npeaks, indicating a sequence of glassy and semicrystalline/amorphous\ntransitions, which have also been observed in other PDMS-based systems,\nparticularly with ZnO 39 and OLC (onion-like\ncarbon) 43 nanoinclusions. The smaller PDMS\ncrystallites begin to melt, while the larger ones remain large enough\nto preserve the stiff structure of the glassy elastomer. 44 When this structure finally collapses, the longitudinal\nvelocity ( Figure 7 a)\nbegins to decrease further again, resulting in a step-like temperature\ndependence. Between the two mentioned attenuation peaks of the\nglass transition\nand semicrystalline melting in the PDMS matrix, a third one at 188\nK has been observed. This temperature value is close to the rhombohedral–orthorhombic\nphase transition in the BTO bulk material. However, this anomaly cannot\nbe attributed to the BTO phase transition based solely on V and Δα measurements, as no\ncorresponding velocity anomalies were observed in this or higher regions.\nIt is only at high temperatures around 380 K that a small and broad Δα maximum, with an associated anomalous ultrasonic\nvelocity slow down (see Figure 7 a(inset)), suggests a possible ferroelectric to paraelectric\nphase transition in BTO nanoparticles. To obtain a more stringent\nevidence for the formation of a ferroelectric state, as well as to\nevaluate the suitability of the 23 vol % BTO/PDMS composite as a high-frequency\npiezo-nanogenerator, some additional ultrasonic measurements of the\npiezoelectric properties at 10 MHz have been performed. The\ntemperature dependence of the piezoelectric voltage signal U p arising on a piezoelectric plate under longitudinal\nstress applied by the ultrasonic wave is presented in Figure 8 a. The U p signal decreases upon heating, and distinct anomalies appear\nat the temperatures of the phase transitions in the BTO bulk: the\nsignal drops to a value of 10 V at 190 K, followed by a small kink\naround 273 K, and disappears completely in the paraelectric phase\nabove 390 K. Samples at lower concentration showed no piezoelectric\nsignal (so no data are presented). This means that the BTO grains\ndo not percolate. Figure 8 Temperature dependencies of (a) the piezoelectric voltage\nsignal\nand (b) the specific stored energy at a frequency of 10 MHz for a\n23 vol % BTO/PDMS composite of 0.2 mm thickness. By having data on the voltage across the capacitor\nplates ( U ), as well as the dielectric permittivity\nof the capacitor\nspace-filling material (ε), the stored energy E of such a system can be estimated. For this purpose, the following\nsimple formula can be applied. Here, the capacitance C in the case of a flat capacitor is defined as C = εε 0 S / d , where ε 0 is the permittivity\nof vacuum, S is the area of the capacitor plates,\nand d is the distance between them. The temperature\ndependence of the specific energy stored by a flat capacitor with\na potential difference U p filled with\na 0.2 mm-thick 23 vol % BTO/PDMS composite is presented in Figure 8 b. Due to the noisy\nresults of the temperature measurements in the 10–300 MHz frequency\nrange, for the energy calculations, the data at 1 MHz (blue curve\nin Figure 6 a) were\nlegitimately taken, since the frequency dispersion is absent far from\nthe glass transition (see Figure 6 b). Thus, a 0.2 mm-thick piezoelectric plate\nof the 23 vol % BTO/PDMS\ncomposite converts high-frequency mechanical vibrations in the form\nof a 10 MHz ultrasonic wave into an electrical signal with an output\nvoltage of 10 mV and a specific stored energy of 367.3 MeV/m 2 at room temperature, which reaches 35 mV after the glass transition\nof PDMS. The measured value of the output voltage is lower 13 , 14 , 20 , 21 due to two factors: (i) the increase in frequency from quasi-static\nto MHz, which reduces the piezoelectric coefficient, and (ii) the\nlow power of ultrasonic excitation."
} | 5,194 |
35281742 | PMC8904616 | pmc | 7,336 | {
"abstract": "Summary Two-dimensional van der Waals materials offer various possibilities for synaptic devices, matching the requirements of intelligent and energy-efficient computation. However, very few studies on robust flexible synaptic transistors have been reported, which hold great potential for soft robotics and wearable applications. Here a floating gate synaptic device based on ambipolar black phosphorus (BP) on a flexible substrate has been demonstrated with two working modes. The three-terminal mode, where the carriers are injected into the floating gate, shows a nonvolatile memory effect, whereas the two-terminal mode shows a quasi-nonvolatile memory effect. Remarkably, the synaptic device working on the three-terminal mode shows an excellent performance in the energy-efficient computation of sub-fJ/spike with a fast gate voltage response down to ∼10 ns. Furthermore, the flexible synaptic device exhibits good endurance under 5,000 bending cycles with a strain of ∼0.63%, suggesting great potential in flexible neuromorphic applications with low energy consumption.",
"introduction": "Introduction The recent rise of interest in neuromorphic computing that mimics the ability of the brain in combining computation and storage with energy efficiency into a compact space focus on new computing architectures that distinguish from traditional von Neumann architectures ( Boybat et al., 2018 ; van de Burgt et al., 2017 ; von Neumann, 2012 ; Wang et al., 2017 ; Zidan et al., 2018 ). In neuroscience, biological neurons with synapses form a living electrochemical system. The excitatory or inhibitory effect of a biological synapse is determined by the specific neurotransmitter released from the presynaptic neuron to the postsynaptic neuron ( Gerstner et al., 2014 ). As a result, synaptic devices designed by imitating biological synapses play a crucial role in brain-inspired computing architecture. Many kinds of artificial synapses have been demonstrated in the past few decades. Synapses based on complementary metal-oxide-semiconductor transistor (CMOS) circuit are compatible with existing industrial processes but require many transistors to realize one synapse with complex interconnection and high energy consumption ( Arthur and Boahen, 2006 ; Indiveri et al., 2006 ; Merolla et al., 2014 ; Qiao et al., 2015 ). Synaptic functions with a single device can be realized by conventional memristors and transistor-based neuromorphic devices but with unsatisfactory energy consumption when compared with that of the human brain of ∼10 fJ level ( Erokhin et al., 2011 ; Jo et al., 2010 ; Kuzum et al., 2013 ; Yang et al., 2017 ). Two-dimensional (2D) van der Waals heterojunctions offer various possibilities for new classes of functional devices with low power and flexible applications in a compact space because of their multifarious energy band design and ultrathin body ( Gao et al., 2019 ; Geim and Grigorieva, 2013 ; Gong et al., 2013 ; Huang et al., 2017 ; Shim et al., 2016 ; Xiong et al., 2020a , 2020b ). Various heterostructures can be designed and fabricated without considering the lattice constant mismatching by van der Waals forces between different 2D semiconductors ( Geim and Grigorieva, 2013 ; Huang et al., 2017 ). Synaptic devices based on 2D semiconductors or 2D heterostructures with low energy consumption ( Wang et al., 2020 ) were reported but most of them suffered from a slow operation speed around millisecond-level ( Huh et al., 2018 ; Qin et al., 2017 ; Sharbati et al., 2018 ; Tian et al., 2016 , 2017a , 2017b ; Zhu et al., 2018 ), preventing the further reduction in power consumption. Moreover, very few studies have been performed on the mechanically bending robustness of flexible synaptic devices. As a result, high performance bendable artificial synaptic devices with low energy consumption and fast operation time are strongly desired and remain to be explored. In this work, we demonstrate floating-gate synaptic devices based on black phosphorus (BP) transistors working at dual modes with pre-patterned local-gate structures using another BP film as a charge trapping layer on a flexible substrate. As expected, the devices exhibit a significant current hysteresis owing to the superior charge-trapping capacity of the BP floating gate, which is essential for synaptic operation. Different from traditional RRAM devices with abrupt switching ( Ge et al., 2018 ; Zhao et al., 2017 ), the postsynaptic current (PSC) of the floating gate synaptic devices can be gradually depressed or potentiated by continuous pulse stimulation. It is interesting to note that the two modes correspond to the nonvolatile and the quasi-nonvolatile memory effects by the three-terminal and two-terminal operations, respectively. The devices can be operated using an ultrafast pulse voltage of ∼10 ns at three-terminal mode with a low energy consumption of sub-fJ/spike. Furthermore, the bending test for the synaptic functions has been carried out, suggesting a great potential in flexible neuromorphic applications with low energy consumption.",
"discussion": "Discussion In summary, we have realized the BP-based floating gate synaptic devices with two operating modes in one structure. By using the ambipolar materials BP as the channel layer, both electrons and holes can be injected into the floating gate layer and imitate biological synapses excitatory and inhibitory behaviors. Ultrafast pulse operation of of ∼10 ns and ultralow power consumption are demonstrated by V g driven mode. In addition, the quasi-long-term memory effect for mode II realized by V d pulses and the long-term memory effect for mode I realized by V g pulse indicate potential for versatile applications. Finally, the synaptic devices show a good endurance against mechanical bending of over 5,000 times, suggesting possible flexible and ultralow power consumption electronic applications. Limitations of the study The excitatory synaptic response by the gate voltage of the device is at 10-ns level, which is limited by the measurement equipment. Although the flexible synaptic devices based on BP show remarkable performance and diverse operational functions, array demonstration for artificial neural networks based on the BP devices is always a challenge. This is because of the high-quality synthesis of BP materials and nondestructive device fabrication methods have not been solved."
} | 1,600 |
32774862 | PMC7362743 | pmc | 7,337 | {
"abstract": "Bio-inspired chiral nematic cellulose films with periodic surface grating structures exhibit optically programmable photonic–photonic coupling.",
"conclusion": "Conclusions In conclusion, here we have studied the chiral nematic organization and surface topography relations governing dual structural colorations in a CNPF sample, showing the possibility to engineer photonic colour in a custom-made cellulose paper by patterning the liquid crystalline CNCs. Our observations shed light on the colour appearance of CNPF arising from the selective Bragg reflection as well as further scattering of LCP light which related to the hierarchical structure, presenting programmable photonic–photonic coupling similar to the exocuticle of Chalcothea smaragdina . In addition, because of the 3D sinusoidal fluctuation on the film surface and tuneable chiral anisotropic bulk phase in the matrix, the reflected light underwent a polarization rotation along with a directional sensitive constructive interference. We believe that the present finding of nanoimprinting a self-assembled CNC helicoid with a precise spatial structure (micro- and nanoscopic scales) will provide knowledge regarding the designing and fabrication of future metamaterials and metasurfaces with cost-effective and eco-friendly procedures, which can also be introduced into other self-organized systems, e.g. , block- co -polymers, liquid crystal colloids and functional nanoparticles.",
"introduction": "Introduction Colours play essential roles in the evolution and survival of plants and animals. 1 Depending on the basic optical mechanism, colours and colour appearance in living creatures can be classified into three classes: absorption, bioluminescence and interference. These mechanisms, alone or combined, generate the palette of the colourful biological world. 2 While the first two mechanisms strongly rely on the chemical composition of the molecules that compose the organisms, structural colour is mainly the result of light interference in approximately micro- or nanometre-scaled periodic structures. 3 Some of the photonic structures in nature are chiral, namely, the organization of bio-derived building blocks in plants or insects is spatially assembled to form a helicoid architecture, known as a Bouligand structure or a twisted plywood structure. 4 , 5 The attractive feature of this chiral structure is its circular polarized light–matter interaction, which selectively reflects the circularly polarized light with the same handedness, whereas light with opposite handedness is transmitted and experiences no phase change. 6 Apart from helicoid ordering, the origin of such brilliant chiral structural colours in natural materials is quite complex: they are often organised into hierarchical architectures and combined with other optical materials such as pigments to enhance or modulate their response, leading to vivid colours which perform signalling functions within or between species. 7 – 9 For example, in the fruit of Pollia condensata , the colour is originated by the helical organization of cellulose fibrils with a spherical geometry on top of a layer of brown tannin pigments, which exhibits intense blue colour reflection. 10 In other cases such as scarab beetle Chalcothea smaragdina , the metallic green colour appearance results from the combination of helical fibrous chitin with periodic blazed diffraction grating-like surface structures, yielding morphology-induced light scattering as well as helical matrix derived structural colour, named as multiple photonic–photonic colour mixing. 11 Knowledge of the interplay between the morphology, composition and optical appearance of biological photonic systems can provide inspiration for novel artificial photonic materials with a wide range of applications. 12 – 14 \n Cellulose is an inexhaustible biopolymer on earth that can be easily found in plants and bacteria. 15 , 16 When the bulk cellulose material is subjected to controlled sulfuric acid assisted hydrolysis, it transforms into a well-defined rigid rod-like morphology with excellent dispersibility in water. 17 , 18 The residues are termed as cellulose nanocrystals (CNCs). It is well known that CNCs can self-assemble into a chiral nematic liquid crystal phase with its ordering preserved in a solid film after evaporation, producing vivid structural colour as observed in plants and certain animals when its helical pitch is on the order of the visible light wavelength. 19 – 21 Up to now, copious studies on CNC liquid crystals have been presented from the perspective of their templating, 22 , 23 colloidal assembly, 24 – 27 drying dynamics 28 – 30 and optical coupling with guest nanoparticles, 31 – 34 however, their potential uses as building blocks to fabricate periodic grating-like surface patterned films which have programmable photonic–photonic couplings still remain relatively unexplored, i.e. , two spatial structures with different scales (micro and nano) coexist and modulate the light independently, showing hierarchical photonic structures. The mimicry of biological species to produce photonic materials has raised increasing interest in the scientific community. However, the edge of knowledge in this research topic is today settled by the limitation that scientists face in the fabrication of materials with similar structural complexity to biological nanostructures. In fact, the fabrication of nano-scale structures on large area with additional control on their micro-scale morphology is very demanding. Drawing inspiration from Chalcothea smaragdina , here we develop a hierarchical structured photonic paper that contains both chiral nematic organization of a CNC matrix and periodic grating lines on its surface, showing dual photonic colours. The patterned photonic film is prepared by confining the CNC suspension into highly aligned microgrooves through soft nanoimprinting lithography followed by evaporation induced self-assembly processes. After drying, these films are peeled off from the template with highly ordered grating lines imprinted onto the lower surface of the films, showing precisely controlled microstructures not only with a surface morphology but also with a bulk phase with deformed helical organization. 35 In particular, we describe that tuning the composition of initial CNC ink can be used to manipulate the photonic band-gap in the helical matrix which couples with diffraction derived from surface grating lines, showing programmable photonic–photonic colour mixing in reflection mode and affecting the diffuse appearance of the dual photonic paper. These bio-inspired films offer a convenient opportunity for the production of coupled photonic structures with complex polarization control, which may be of value in advanced optical devices.",
"discussion": "Results and discussion The CNC used in this study was produced by sulfuric acid hydrolysis of bleached cotton pulp, which exhibited rod-like morphology with an average length and diameter of 300 nm and 16 nm, respectively (Fig. S1, ESI † ). Before drying, the as prepared CNC colloid suspension was diluted to the desired concentration (5.0 wt%, zeta potential –53 mV) and mixed with varying amounts of water-soluble polymers (polyvinyl alcohol, PVA) to generate homogeneous mixtures without disrupting the ability of the CNC to form the chiral nematic phase. Then, these mixtures were cast onto a hydrophilic polydimethylsiloxane (PDMS) template with periodic surface microgrooves ( Fig. 1a ). After evaporation, iridescent composite films were obtained with vivid structural colour, which were tunable from blue to green and red, due to the increase of helical pitch ( Fig. 1b ). Finally, a centimetre-scaled composite film (3 × 5 cm 2 ) was carefully peeled off from the template with a Janus layer, i.e. , a film with an upper smooth surface and a bottom patterned surface. This chiral nematic patterned film was denoted as CNPF which revealed rainbow-like bistructural colour and high mechanical strength (Fig. S2 and S3, ESI † ). Fig. 1 (a) Schematic of a soft lithography fabrication method to generate freestanding CNPF by using patterned PDMS sheets as the template. (b) Photographs of CNPF1–3 with varying structural colour from blue to red. (c) Reflection POM image of CNPF3 where the surface concavity edge appears red while the central region appears dark. (d) Reflection optical image of CNPF3 in the same area but without polarizers, revealing a dark surface concavity edge and a bright central region. (e) Experimental image analysis of the polarized and nonpolarized optical image for CNPF3, respectively, showing the fluctuation in the stripe area that confirms the optical switching between the surface concavity edge and the centre. (f) Schematic of the polarization rotation caused by a double reflection inside the concavity, which rotates the polarized light by an angle of 2 φ . In order to test the influence of surface topography on CNPF, the optical details were first obtained from polarized optical microscopy (POM) images in reflection mode. For normal light incidence, when the sample was placed between crossed polarizers, it exhibited alternately dark and bright stripes with the width of dark stripes similar to the grating distance. However, removing the polarizers led to a reverse of the optical pattern compared with crossed polarizers, namely, the center of the grating line concavity turned bright while the edge of the stripes became dark ( Fig. 1c–e ). This phenomenon is a result of the geometry induced double reflection and polarization rotation. 36 In a regular reflection mode without polarizers, light from the cavity center is directly reflected with the strongest reflection intensity, while at the grating line edge retro-reflection of incident light occurs by double reflection off the cavity multilayer. On the other hand, if the incident light is linearly polarized with angle φ to the initial incident plane, it will be retro-reflected by the double bounce and undergoes a polarization rotation of 2 φ at each interface ( Fig. 1f ). Therefore, under crossed polarization state light reflected off the center of the grating line is suppressed, whereas only retro-reflected light from the grating line edges passes through the analyzer and is detected. 37 Besides, for the transmission mode the optical images of the same sample showed no sign of reversing dark and bright stripes (Fig. S4, ESI † ), indicative of lacking double reflection and polarization rotation. Visually, as well as by large area optical microscopy and small area scanning electron microscopy (SEM) tests, the chiral nematic patterned film showed periodic grating lines that fully replicate the microstructure of the template ( Fig. 2a and b ). Fig. 2c shows the low magnification SEM image of CNPF3 with an oblique view of the film which shows three different parts, i.e. , a periodic surface grating region, the interface between the surface and the bulk and an extensive periodic bulk phase. Looking at the cross sections, a deformed helical layered structure was present near its patterned surface where the helix axis of the CNC remained parallel to the curved surface ( Fig. 2d ), implying the planar anchoring of the CNC at the water–PDMS interface during evaporation. At the nanoscale, CNC nanorods in bulk phase were rotated in a counter-clockwise direction, giving rise to a long-range helical structure ( Fig. 2e ). The distance between two adjacent layers was termed as the helical pitch for a chiral nematic structure, which was 195, 365 and 475 nm for CNPF1–3, as estimated by SEM analysis (Fig. S5, ESI † ). Surface profiles of the CNPF samples demonstrated a groove depth of 275 nm and a grating distance of 2.5 μm as observed in atomic force microscopy (AFM) images (Fig. S6, ESI † ), similar to the surface profile of the PDMS template (260 nm and 2.4 μm, respectively). Besides, tuning the surface profile of the PDMS template by controlled plasma treatment could also manipulate the surface morphology of the corresponding CNPF composites (Tables S1 and S2, ESI † ). All of these data above clearly displayed the coexistence of hierarchical micro- and nanoscale features with the periodic surface grating lines and chiral nematic organization ( Fig. 2f ). Previous reports demonstrated that an individual CNC could be highly aligned into the microgrooves of the PDMS template with periodic line arrangement. 38 Through evaporation, due to the elastic interplays between a liquid crystalline CNC and a PDMS template, the patterned template will deform the interface of liquid crystals into a hilly topography, with each hill accommodating a chiral nematic spiral, generating a nontrivial interfacial curvature with bend-splay orientation distortions along the template surface. 39 , 40 In the end, this deformed helical ordering will be preserved as a solid film, with the distance and amplitude of these grating lines imprinted and determined by the PDMS template. 35 Fig. 2 (a) POM image of CNPF2 in transmission mode which shows strong birefringent colour along with periodic parallel striations. (b) Top view SEM image of CNPF3 with low magnification that shows surface grating. (c) Oblique view of a cracked sample of CNPF3 with three different districts. (d) Side view SEM image of cracked CNPF3 composites which focuses on the CNC–PDMS interface. (e) Side view SEM image of the fracture in CNPF3 at high magnification which reveals the anti-clockwise rotation of CNC organization. The inset arrow shows the rotation direction of CNCs. (f) Sketch of the pitch-splay and layer undulation at the PDMS–CNC interface during an evaporation process. In the chiral nematic film, the position of its photonic band-gap is optically tunable from UV-blue to visible regions by tuning the helical pitch. 20 By changing the CNC–PVA composition in its initial ink, we can easily manipulate the helix derived photonic band-gap in CNPF. 41 Fig. 3 shows the reflection spectra of CNPF1–3 consisting of varied photonic–photonic coupling. A series of photonic band-gaps were obtained with their positions centered at 297, 562 and 722 nm, respectively, which resulted from the increase of helical pitch, typical for the chiral nematic organization. For the sample of CNPF2 and CNPF3, a double-peak spectral feature was observed in the UV-blue range centered at 245 and 300 nm, which resulted from the surface grating line induced photonic structure and almost remained constant without the influence of the surface morphology (also confirmed by a chiral nematic-free patterned CNC film, Fig. S7 and S8, ESI † ). When the photonic band-gap in the host CNC matrix was blue-shifted to smaller wavelength (from 722 to 297 nm), there occurred a strong photonic–photonic coupling between the CNC matrix and surface grating lines, leading to an enhanced coupled optical signal with vivid blue structural colour appearance (see Fig. S2, ESI † ). Hence, both the chiral nematic ordering in the bulk CNC phase and surface grating lines on the film surface can individually be termed as a photonic crystal, showing programmable photonic–photonic coupling. More interestingly, the optical property of the CNPF film could be further tuned by varying the light incident angle (Fig. S9, ESI † ). During the increase of rotation angle, the photonic band-gaps of these two samples were blue-shifted to smaller wavelength which could be attributed to changes in diffraction related to incidence angles. However, the band-gap shifting in CNPF2 showed a monotonic linear relationship whereas the peak shifting in CNPF1 followed a non-linear trend, which implied that the photonic–photonic coupling between the helical matrix and surface gratings exhibited strong influences on the angle-dependence spectra behaviour. Fig. 3 UV-Vis spectra of CNPF1–3 at normal incidence with varying photonic–photonic coupling. The inset shows the magnified UV spectra of CWF2 and 3 at a short wavelength (200–400 nm) which highlights the surface grating induced double-peak spectra. To quantify the diffraction induced photonic property in CNPF, angle-resolved scattering measurements were performed ( Fig. 4a ). The set-up contained a broadband deuterium-halogen light source (mod. DH-2000, Ocean Optics) for a fixed incident light beam (incident angle: 35°), a rotating sample stage and a rotating detector. The output beam was first collimated by a quartz lens, giving rise to a spot size of 3 mm on the sample surface. Then, the light diffused signal was collected by a multimode optical fibre on a rotation stage. The reflected and scattered light was detected at varying detection angles between 0° and 70° in 2.5° steps. Fig. 4b–h exhibit the diffraction results for the sample of CNPF1, 2 and their corresponding grating-free reference (REF) samples. For patterned samples, the presence of surface grating lines generated a series of diffraction peaks in different angular directions, thus, resulting in the first, second and third-order diffractions which located in a wide range of angles, a characteristic of grating-derived iridescence ( Fig. 4b and e ). Compared with a CNPF sample, the scattering orders in REF resulting due to the surface topography were not obvious, and they only demonstrated a broad scattering band at 0° which was ascribed to the zero-order reflection with the highest intensity located at their corresponding photonic band-gaps ( Fig. 4c and f ). As a 1D photonic crystal, the optical property in the chiral nematic structure itself is angle-dependent. 42 During scattering measurement, illumination at an oblique angle leads to iridescence manifested as a blue-shift in the reflected structural colour with the Bragg reflection peak given by λ = n avg P sin θ , where n avg and P are the refractive index and helical pitch, respectively, and θ is the angle between the incident direction and the helical axis. 43 Therefore, the zero-order reflection bands in CNPF and REF showed the strongest scattering intensity at the position of the photonic band-gap and then decreased to further wavelengths due to the increasing incident angle. Fig. 4 (a) Schematic representation of the measurement geometry and optical set-up. Two-dimensional scattering maps of CNPF1 (b), REF1 (c), CNPF2 (e) and REF2 (f) exhibit the angle-resolved scattering intensities which are strongly influenced by programmable photonic–photonic coupling. Bands containing zero-order reflections are marked by stars. (d and g) Comparisons of the angle dependent scattered light intensity between CNPF1 and 2 and their REF samples. Plots are obtained as horizontal profiles of the corresponding maps shown at 280 nm (b) and 480 nm (e), respectively. (h) Calculated ratio between the angular full width at half maximum (FWHM) of the reflected beam from patterned samples (FWHM CNPFs ) and the one from the REF samples (FWHM REF ) for samples CNPF1, CNPF2 and CNPF3, respectively. In addition, we found that the interplay between the helical CNC matrix and imprinted surface gratings could provide dramatic effects on the resulting scattering properties. A significant increase of the reflection and diffraction in CNPF1 was achieved by shifting the band-gap toward the UV-blue spectral band. The specular reflectivity at 280 nm was enhanced by 2 factors in CNPF1 compared to the pattern-free one ( Fig. 4d ), whereas for CNPF2 and CNPF3 the addition of surface gratings did not affect the reflection peaks at 480 and 630 nm, respectively ( Fig. 4g and Fig. S10, ESI † ). Calculations of the first-order diffraction efficiency for CNPF1 and CNPF2 were 0.47 and 0.12, respectively, demonstrating an effective boosting of optical properties in hierarchical photonic structures in CNPF1 through the interplay of photonic–photonic coupling. As a consequence, we concluded that the chiral nematic photonic band-gap could enhance the iridescence response that derived from the periodic surface grating lines. The ensemble structural colour appearance is attributed to the sum of the scattered helicoidal photonic reflections with additional interference resulting from the periodic surface topography. 44 Interestingly, the presence of the photonic–photonic coupling between the grating structure and the helical matrix also decreased the angular width of the reflectance peak in CNPF1, whereas the bandwidth was almost unchanged in samples of CNPF2 and CNPF3 ( Fig. 4h ). The total angular distribution of the light intensity reflected from CNPF samples was the result of the convolution of the angular distribution of the intensity reflected by the bulk CNC cholesteric structures and surface diffraction grating. The latter generated a characteristic angular distribution of the intensity of the reflected light composed by consecutive intensity maxima and minima, whose separation depended on the wavelength of light and the periodic grating period. 45 While for CNPF2 and CNPF3 the first diffraction order was well separated from the zero order (helical induced photonic band-gap) peak, for CNPF1 the first diffraction order was within the angular width of the reflected light beam. This in turn led to a decrease of the intensity of light between the zero and the first diffraction order, in correspondence with the first intensity minimum of the diffracted light pattern, thus determining a decrease in the angular width of the zero order diffracted beam. In nature, the green colour appearance of Chalcothea smaragdina is anticipated to be used as protective coloration and aids in its camouflage against predators. 11 The beetle is visibly vivid green when observed through a left-handed circularly polarized (LCP) filter and totally loses its characteristic bright green reflection under a right-handed circularly polarized (RCP) filter, exhibiting selective reflection of the LCP light exclusively, which is due to the anticlockwise rotation of helicoidal chitin fibrils within the beetle's elytron exocuticle ( Fig. 5a ). The measured average pitch and refractive index of the helicoid beetle cuticle are about 340 nm and 1.647, respectively. Thus, a helical structure derived Bragg reflection occurs at 560 nm in normal incidence which fits well with its saturated green colour appearance. On the other hand, the surface structure of the beetle exocuticle reveals parallel surface striations that possess a fine irregular sawtooth-shaped profile parallel to the surface plane ( Fig. 5b ). And the beetle surface topography is further explored to show a quasi-regular surface periodicity manifested by a distinct acclivity and declivity at the edge of the striation, forming a step structure similar to a blazed grating structure ( Fig. 5b , inset). 11 The measured spacing between neighbouring striations is 5.07 μm with a blaze angle θ b of 3.1°. The blaze angle is optimized with maximum efficiency for all the incident light, which gives rise to the metallic appearance of the beetle exocuticle. The grating equation for the reflective blazed grating can be found as: d (sin α + sin β ) = mλ , where α and β are the incidence and diffraction angles, respectively, d is the periodicity of the grating lines, λ is the diffracted incident light wavelength and m is the diffraction order ( Fig. 5c ). In particular, if the incidence angle and diffraction angle are identical ( α = β = θ b ) which refers to a specific geometry for blazed grating named as the Littrow configuration, the grating equation can be transformed into 2 d sin θ b = mλ . 46 The grating efficiency of the Littrow configuration is mainly dependent on the most intense order ( m = 1), thus, the light wavelength for which scattering is optimized in Chalcothea smaragdina is calculated as 548 nm, coinciding with a green spectral range and overlapping with the reflection peak of an internal helical structure. The dual structural coloration in Chalcothea smaragdina is therefore the result of the combination of helical matrix derived photonic colour and surface blade grating induced interference of reflected light, creating an additive effect on the sum colour appearance and leading to strong photonic–photonic coupling. Fig. 5 (a) Photographs of the beetle Chalcothea smaragdina captured in a LCP (left) and a RCP (right) filter, respectively. (b) SEM image of the periodic surface structure for the beetle. The inset represents the AFM surface profile of the beetle surface, which reveals a blazed grating surface topography. (c) Schematic representation of the polarization-selective surface blazed grating for the beetle exocuticle in which the diffracted LCP light generates constructive interference, whereas for the Littrow configuration (purple lines) the incidence angle and diffraction angle are identical, leading to a special specular reflection. (d) Photographs of CNPF2 captured using a LCP (left) and a RCP (right) filter, respectively. (e) Three-dimensional AFM image of CNPF2 that exhibits a smooth sinusoidal surface profile. (f) Schematic description of the chiral nematic ordered holographic surface grating with selective constructive interference for LCP light. (g and h) LCP and RCP reflectance spectra for the beetle of Chalcothea smaragdina (g) and CNPF2 (h), respectively. (i) Photographs of CNPF1 with strong photonic–photonic coupling captured in a LCP and RCP filter. (a), (b) and (g) were adopted with permission from ref. 11 . To better compare the similarities and differences (structure and optics) between Chalcothea smaragdina and our mimetic patterned sample, we choose CNPF2 with the same green structural colour appearance that has both periodic surface topography and a helicoidal ultrastructure. In agreement with the circularly polarized response associated with the beetle, upon illumination the captured photographs of CNPF2 exhibit bright green appearance under a LCP filter which completely extinguishes through a RCP filter, indicative of the selective reflection of circularly polarized light and optical scattering from the sample ( Fig. 5d ). Closer inspection of the film surface by AFM images reveals a 3D sinusoidal surface pattern with a smooth profile ( Fig. 5e ), which can be classified into holographic grating, responsible for a diffraction-induced rainbow-like appearance in an oblique view (Fig. S11, ESI † ). Blazed gratings can offer extremely high diffraction efficiency at the designed wavelength, whilst they suffer from periodic optical errors and ghosts. 46 When compared with the blazed gratings, the holographic grating can reduce and eliminate these errors, only with the drawback of reduced diffraction efficiency. According to the grating equation d (sin θ i – sin θ d ) = mλ , θ i and θ d are the incidence and diffraction angles, respectively. 47 When the incident light with a wavelength of λ impinges on the grating surface at a given value of θ i , the diffracted light scatters in different angular directions with constructive interference ( Fig. 5f ). The scattering maps of CNPF2 for circularly polarized light diffractions showed that the diffracted signals remained strong under LCP illumination while being almost extinguished under RCP illumination (Fig. S12, ESI † ), which exhibited a close agreement with the circularly polarized optical images. In both cases, Chalcothea smaragdina with blazed grating and CNPF2 with holographic grating, the diffractive scattering resulting due to the patterned chiral nematic structure reflects a half proportion of incident light with LCP handedness away from the grating surface while the remaining with RCP state is transmitted. As a result, the scatterings imparted by surface topography (blade or sinusoidal) can be switched on/off under different circularly polarized illuminations, displaying vivid colour appearance only with LCP light. Furthermore, we also presented the circularly polarized spectral comparison of Chalcothea smaragdina and the sample of CNPF2. Typically, the LCP specular reflection spectrum derived from the beetle surface showed a pronounced double-peak photonic band-gap with the principal reflection peak centred at 565 nm and less intense feature at 585 nm, which was due to the pitch undulations of helicoid structures. Meanwhile, the RCP spectrum exhibited an absence of distinguishable optical features, implying the extinction of RCP reflection ( Fig. 5g ). It should be noted that the surface blazed grating induced reflection feature also vanished during RCP measurement which proved the LCP state of reflected light under Littrow configuration, showing full photonic–photonic coupling inside the dual photonic structures. By contrast, the reflection spectrum of CNPF2 exhibited a distinct peak reflectance at 506 nm for LCP light, which was significantly suppressed under RCP measurement ( Fig. 5h ), which resulted from the left-handed chiral nematic CNC matrix. Unlike Chalcothea smaragdina , the surface grating induced reflection peaks in CNPF2 were located at the UV-blue spectral range (245 and 300 nm, see Fig. 3 ), far away from its corresponding helicoid derived photonic band-gap and lacking photonic–photonic coupling. However, blue shifting the photonic band-gap led to a strong coupling effect between the helicoid organized CNCs and holographic surface grating, showing saturated blue colour appearance (Fig. S2, ESI † ). This metallic blue structural colouration was attributed to the combination of helicoid matrix and surface grating induced photonic coupling, which was sensitive to both the circular polarization state and incident angle of light illumination ( Fig. 5i and Fig. S13, ESI † ). It should be pointed out that the blue structural colouration totally extinguished and became transparent under a RCP filter, implying that the surface grating induced diffuse appearance was fully coupled with helical photonic band-gaps and kept in a circular polarization state, similar to the colouring mechanism of the beetle exocuticle."
} | 7,520 |
28018855 | PMC5176132 | pmc | 7,338 | {
"abstract": "We announce the draft genome sequence of three Gram-negative bacteria isolated from coral tissues affected with the black band disease (BBD), identified with the NCBI's Assembly Database accession numbers: MBQF, MAYB and MBQE. These genome drafts constitute an useful tool for the characterisation of these bacteria and for the understanding of the relationship between the microbial consortia associated with the disease and the onset and progression of the pathology."
} | 117 |
34194405 | PMC8237939 | pmc | 7,339 | {
"abstract": "Metabolically engineered cyanobacteria have the potential to mitigate anthropogenic CO 2 emissions by converting CO 2 into renewable fuels and chemicals. Yet, better understanding of metabolic regulation in cyanobacteria is required to develop more productive strains that can make industrial scale-up economically feasible. The aim of this study was to find the cause for the previously reported inconsistency between oscillating transcription and constant protein levels under day-night growth conditions. To determine whether translational regulation counteracts transcriptional changes, Synechocystis sp. PCC 6803 was cultivated in an artificial day-night setting and the level of transcription, translation and protein was measured across the genome at different time points using mRNA sequencing, ribosome profiling and quantitative proteomics. Furthermore, the effect of protein turnover on the amplitude of protein oscillations was investigated through in silico simulations using a protein mass balance model. Our experimental analysis revealed that protein oscillations were not dampened by translational regulation, as evidenced by high correlation between translational and transcriptional oscillations ( r = 0.88) and unchanged protein levels. Instead, model simulations showed that these observations can be attributed to a slow protein turnover, which reduces the effect of protein synthesis oscillations on the protein level. In conclusion, these results suggest that cyanobacteria have evolved to govern diurnal metabolic shifts through allosteric regulatory mechanisms in order to avoid the energy burden of replacing the proteome on a daily basis. Identification and manipulation of such mechanisms could be part of a metabolic engineering strategy for overproduction of chemicals.",
"introduction": "Introduction Knowledge of cyanobacterial metabolism and its regulation can guide metabolic engineering efforts to create more efficient strains for renewable fuel and chemical production. As their energy source is limited to the light hours of the day, cyanobacteria have evolved to shift between photosynthetic and respiratory metabolism between day and night, respectively. During the day, CO 2 is fixed in the Calvin cycle and converted into biomass, including storage compounds such as glycogen. During the night, CO 2 fixation and most biosynthetic pathways are inactive while glycogen is degraded to support cellular maintenance and a small subset of pathways that prepare the cell for the next light period ( Saha et al., 2016 ; Reimers et al., 2017 ; Welkie et al., 2019 ; Werner et al., 2019 ). Metabolic shifts that occur at specific time points over the day-night cycle are governed by regulating the flux through key enzymes and pathways. The flux through an enzyme is regulated by changing its abundance, product/substrate concentration, or through post-translational effects that alter its apparent kinetic parameters. Several studies have investigated abundance-controlled regulation by tracking changes in the cyanobacterial transcriptome and proteome across the day-night cycle. Transcriptomic data collected from a range of cyanobacteria showed that a large fraction of cyanobacterial transcripts oscillates diurnally (30–87%), with peak expression mostly during the transitions between day and night ( Stöckel et al., 2008 ; Ito et al., 2009 ; Waldbauer et al., 2012 ; Saha et al., 2016 ). Additionally, many transcripts tend to peak just before the time when the gene product’s function is expected to be needed by the cell. For example, transcripts of Calvin cycle and pentose phosphate pathway genes peaked in the beginning of the light and dark period, respectively ( Waldbauer et al., 2012 ). Yet surprisingly, a few proteomics studies have shown that abundance of most proteins remains nearly constant ( Stöckel et al., 2011 ; Waldbauer et al., 2012 ; Guerreiro et al., 2014 ; Angermayr et al., 2016 ). This makes the regulatory purpose of time-dependent transcription seem insignificant for regulating enzyme activity and diurnal metabolic shifts. The underlying cause for a broad discrepancy between transcript and protein dynamics is still not clear, but it could be attributed to post-transcriptional regulation or low daily de novo protein synthesis relative to the protein abundance. One possibility is that translational regulation counteracts changes in mRNA abundance, resulting in reduced variation in protein synthesis rate of genes despite their altered transcript levels. Protein synthesis rates can be measured genome-wide through ribosome profiling (Ribo-Seq), which quantifies the total number of ribosomes translating a gene’s transcripts ( Brar and Weissman, 2015 ; Liu et al., 2017 ). A translationally-regulated gene would show a change in ribosome abundance that is not equal to the change in transcript abundance, or vice versa. Translational regulation was shown to occur in 7% of the genome of Synechocystis sp. PCC 6803 ( Synechocystis ) in response to CO 2 starvation ( Karlsen et al., 2018 ). A second possibility is that protein levels are held relatively constant by active protein degradation. However, rapid degradation of newly synthesized proteins would waste energy and cellular resources and reduce fitness. Lastly, relatively low variation in protein levels could also occur without any post-transcriptional regulation, if the daily variation in protein synthesis rate is low compared to the protein abundance, i.e., if the turnover rate of the proteome is low. Here, we apply a systems biology approach to take a closer look at the discrepancies between transcription and protein abundances during day-night cycles in cyanobacteria. The model cyanobacterium Synechocystis was grown in controlled turbidostat cultures under artificial day-night cycles. To assess the impact of translational regulation on the protein level, the transcriptome, translatome, and proteome was measured at different time points using mRNA sequencing, ribosome profiling, and quantitative proteomics. We found that protein synthesis rates tracked with transcriptional oscillations, while protein abundances remained relatively constant, indicating that translational regulation does not significantly impact the protein-level behavior. We further investigated the effect of protein turnover on protein dynamics in silico . Simulation of protein oscillations using biologically relevant parameter settings, resulted in a protein amplitude similar to experimental observations. The data and model simulations demonstrate that post-translational regulation is not necessary for the proteome to remain stable, even under significant transcriptional oscillations.",
"discussion": "Discussion Post-transcriptional regulation is an intuitive explanation for the discrepancy between cyclic diurnal transcription and relatively constant protein levels in cyanobacteria. Our transcriptomic, translatomic and proteomic data confirmed this discrepancy and showed that it is not caused by translational regulation. In addition, modeling of the protein response to transcriptional oscillations under biologically relevant parameter settings demonstrated that the experimentally observed decrease in protein oscillation amplitude can be attributed to a slow bulk protein turnover, without the requirement of regulated protein degradation that counteracts transcriptional oscillations. Modeling results further suggested that the bulk protein degradation rate was similar to the daily average growth rate. The strong correlation between ribosome and mRNA abundance fold changes indicates that protein synthesis oscillates significantly over the day-night cycle and that translation is not regulated between time points ( Figure 1A ). Synthesis rates were solely based on the ribosome abundance and did not account for within-gene changes in ribosome elongation rate. However, elongation rates were not expected to change significantly on global level between time points, since elongation rates primarily depend on gene-specific properties of the mRNA structure ( Riba et al., 2019 ). Furthermore, variation in elongation rate would more likely result in reduced correlation with mRNA abundance. In contrast, diurnal protein abundance patterns generally did not show a clear cyclic behavior and did not correlate with protein synthesis oscillations ( Figure 1A ). Small cyclic patterns were most likely present, but concealed by technical variation and therefore not detectable. As measurement errors were high relative to diurnal changes in protein abundance, the determined median relative protein amplitude of 1.5 was probably overestimated ( Figure 1B ). The proteome-wide 2.0-fold reduction in amplitude from synthesis to abundance, was comparable to the 2.3-fold reduction determined previously with higher statistical power ( Waldbauer et al., 2012 ). With this approximate ratio taken into account, our model suggests that the bulk degradation rate was in the range of 0.01–0.05 h –1 , i.e., similar to the daily average growth rate, and in line with degradation rates measured in microalgae and plants ( Figure 2A ). This was further supported by bulk degradation rates measured in other organisms which are typically in the same magnitude as the growth rate ( Table 1 ). The positive correlation between growth rate and bulk protein degradation has been attributed to a high energy burden of protein turnover when nutrients are limited ( Lahtvee et al., 2014 ). Our modeling analysis showed that the bulk protein turnover rate (proportional to μ + k D, MEAN ) determines the proteome-wide reduction in amplitude between the synthesis level and the abundance level (mean synthesis:protein amplitude ratio). The model further suggested that gene-specific deviations from the mean synthesis:protein amplitude ratio are determined by deviations in individual protein degradation rates relative to the bulk degradation rate ( Figure 2B ). Waldbauer et al. (2012) reported variation in the synthesis:protein amplitude ratio (synthesis = mRNA level) across the genome of Prochlorococcus MED4 during diurnal growth. While the vast majority of genes in this study also exhibited low amplitude or no oscillations at the protein level, approximately 30 of the 548 analyzed proteins showed an amplitude fold change greater than 2. However, the relatively high amplitude of these proteins was not caused by particularly strong oscillations in protein synthesis relative to other genes. Instead, protein synthesis oscillations of these genes appeared to be less dampened at the level of protein relative to other genes, as indicated by a lower synthesis:protein amplitude ratio. Our model suggests that such outlier proteins are subjected to a high gene-specific degradation rate (i.e., gene-specific protein turnover), which increases the relative amplitude of oscillations by reducing the protein’s daily mean abundance without affecting the absolute amplitude. This was further indicated in our experimental data ( Figure 1C ), where a positive trend between the relative protein amplitude and gene-specific protein turnover (daily mean synthesis rate/daily mean abundance) was detected. A degradation rate for a given protein that is 10-fold higher than the bulk degradation rate is not unrealistic, as gene-specific degradation rates were shown to span two to three orders of magnitude in Lactococcus lactis ( Lahtvee et al., 2014 ). Furthermore, artificially increasing degradation rate, by fusing a ssrA degradation peptide, increased the relative amplitude and decreased the phase shift of a diurnally expressed yellow fluorescent protein in Synechococcus elongatus PCC 7942 ( Chabot et al., 2007 ). The protein oscillation model assumes a constant cellular protein concentration. This assumption was largely satisfied over the day-night cycle, according to measurements of total protein content in cell extracts. The assumption of a constant cellular protein concentration constrains bulk protein synthesis to be proportional to the sum of bulk protein degradation and growth dilution. Consequently, a decreasing protein concentration during night time will lead to an overestimated bulk protein synthesis rate by the model. This will in turn result in an overestimated rate change of each protein’s (J) concentration during night time. However, as the cellular protein concentration increases to its original level during sunrise, the opposite effect will occur. That means bulk protein synthesis will be underestimated and the rate change of each protein’s concentration will be underestimated, which compensates for the overestimated rate change during the night. Thus, small changes in cellular protein concentration will not change the simulated protein amplitude significantly, but rather alter the diurnal pattern of protein abundance. This is analogous to the effect of setting a constant growth rate vs. setting a fluctuating growth rate ( Figure 2D ). Cyclic transcription has been shown to peak near time points of the day-night cycle when the corresponding function is expected to be needed by the cell ( Waldbauer et al., 2012 ; Beck et al., 2014 ; Saha et al., 2016 ; Strenkert et al., 2019 ). However, the regulatory purpose of a diurnally shifting transcriptome appears less meaningful, since the impact on the functional protein level is significantly diminished. It is nonetheless possible that well-timed, yet small, changes in protein abundance results in a growth benefit that increases survival fitness in a natural environment. Furthermore, our model shows that these changes would become increasingly relevant in a condition that permits higher growth rates, such as an eutrophicated lake exposed to intense sunlight ( Figure 2A , right). Indeed, Synechocystis can grow with a growth rate as high as 0.16 h –1 ( van Alphen et al., 2018 ). This growth rate would correspond to a daily average protein turnover (μ + k D, MEAN ) of approximately 0.12 h –1 , considering a diurnal growth pattern and that the bulk degradation rate is typically similar and dependent on the growth rate ( Table 1 ). Protein levels in cyanobacteria do change significantly in response to changes in light intensity, if allowed to adjust to a steady state ( Jahn et al., 2018 ). Yet, during diurnal growth, the co-occurrence of a largely constant proteome and considerable metabolic shifts suggests that allosteric interactions play an important regulatory role. For example, CO 2 fixation is inactivated during the night through an allosteric mechanism where the regulatory protein CP-12 binds and inactivates the Calvin cycle enzymes phosphoribulokinase and glyceraldehyde-3-phosphate dehydrogenase ( Tamoi et al., 2005 ). Glycogen degradation is another potential target of allosteric regulation since it mostly occurs during the night, even though the abundance of glycogen phosphorylase does not change over the day-night cycle ( Supplementary Table 1 ). Our results also have implications for synthetic biology in cyanobacteria. There have been many efforts to control the abundance of heterologous proteins in Synechocystis , at both the level of translation, through alteration of RBS sequence ( Thiel et al., 2018 ), and at the level of degradation, through a synthetic ssrA peptide with a calculated homology to the native sequence ( Landry et al., 2013 ). The perceived ribosome binding site affinity is not an accurate predictor of protein levels, even when comparing ribosome binding sites with the same heterologous protein ( Thiel et al., 2018 ). It is possible that ribosome profiling, which provides a measure of ribosome occupancy across the entire transcript, could provide insight as to how genetic context affects translation of heterologous proteins. The findings in this study suggest that faster changes in a heterologously expressed protein’s abundance can be achieved, if its synthesis rate and degradation rate is high, i.e., if its gene-specific protein turnover is high. In case transcription of the heterologous gene is from a promoter that has an inherent oscillation, then an increased degradation rate, through e.g., a strong degradation tag, could increase oscillations in the protein level. At the same time, a slow bulk protein turnover will extend the time needed for that protein to reach its steady-state abundance, since the cellular protein space is limited. This appears to be the case in cyanobacteria cultures grown at constant light. In a study on the induction kinetics of YFP from various promoters in Synechocystis , the protein accumulated for five days after induction with rhamnose before reaching a steady state ( Behle et al., 2020 ). In day/night cultivations, the change in the target’s protein abundance will be slower still, as total transcription and/or translation is globally downregulated at night, by inactivation of RNA polymerases and/or ribosomes ( Hood et al., 2016 ). Therefore, comparisons of gene expression constructs, such as promoters or ribosome binding sites, should occur only after steady-state has been reached. In conclusion, we show that the relatively constant proteome during diurnal growth can be explained by low protein turnover. A relatively high bulk protein turnover is required to obtain significant diurnal changes at the global proteome level. To minimize protein turnover energy costs and improve fitness under growth limited conditions, cyanobacteria may instead have evolved allosteric mechanisms to regulate metabolic shifts. Such adaptation may be particularly relevant for photosynthetic organisms as their energy supply is limited to times of the day with sunlight exposure. Identifying potential allosteric regulation of key enzymes in cyanobacteria could assist future metabolic engineering attempts to accelerate carbon fixation or divert metabolic flux, as these enzymes could become targets for protein engineering. Incorporating allosteric regulation into metabolic models would also improve their prediction capability when simulating genetic knockouts that result in altered metabolic flux patterns. Furthermore, our results suggest that changes in transcription or translation are not necessarily a good predictor of diurnal changes in enzyme concentration, or metabolic flux."
} | 4,583 |
35416699 | PMC9239051 | pmc | 7,340 | {
"abstract": "ABSTRACT Root nodulating rhizobia are nearly ubiquitous in soils and provide the critical service of nitrogen fixation to thousands of legume species, including staple crops. However, the magnitude of fixed nitrogen provided to hosts varies markedly among rhizobia strains, despite host legumes having mechanisms to selectively reward beneficial strains and to punish ones that do not fix sufficient nitrogen. Variation in the services of microbial mutualists is considered paradoxical given host mechanisms to select beneficial genotypes. Moreover, the recurrent evolution of non-fixing symbiont genotypes is predicted to destabilize symbiosis, but breakdown has rarely been observed. Here, we deconstructed hundreds of genome sequences from genotypically and phenotypically diverse Bradyrhizobium strains and revealed mechanisms that generate variation in symbiotic nitrogen fixation. We show that this trait is conferred by a modular system consisting of many extremely large integrative conjugative elements and few conjugative plasmids. Their transmissibility and propensity to reshuffle genes generate new combinations that lead to uncooperative genotypes and make individual partnerships unstable. We also demonstrate that these same properties extend beneficial associations to diverse host species and transfer symbiotic capacity among diverse strains. Hence, symbiotic nitrogen fixation is underpinned by modularity, which engenders flexibility, a feature that reconciles evolutionary robustness and instability. These results provide new insights into mechanisms driving the evolution of mobile genetic elements. Moreover, they yield a new predictive model on the evolution of rhizobial symbioses, one that informs on the health of organisms and ecosystems that are hosts to symbionts and that helps resolve the long-standing paradox.",
"introduction": "INTRODUCTION A predictive understanding of symbiosis evolution is critical to inform on the health of hosts and ecosystems in which microbial symbionts reside. A central feature of symbiosis is the variation in the magnitude of services that symbionts provide ( 1 ). At one extreme are uncooperative strains, those that abandon intimate association with hosts or that are ineffective in providing benefits ( 2 ). Their recurrent origins are predicted to destabilize symbiosis, but breakdown has rarely been observed, presenting a paradox of instability and robustness ( 2 , 3 ). Discovering mechanisms that generate and maintain variation in microbial symbionts is foundational for building a unifying framework for symbiosis evolution and resolving the paradox ( 1 ). Genetic variation occurs through mutation, recombination, and gene flow, which together underlie the concept of the pangenome, a nonredundant set of genes in organisms related through ancestry and divisible into core and accessory genomes ( 4 ). The core represents genes predicted to encode functions common and essential among related strains. The accessory genome consists of genes polymorphic in presence/absence and confers upon related individuals the ability to adopt diverse lifestyles. In bacteria, mobile genetic elements (MGEs) are important molecules of accessory genomes. MGEs tend to be arranged into functional units, an organization that promotes modularity, a property that preserves functionality while allowing components to separate and recombine, thereby conferring flexibility and robustness to adapt to different conditions ( 5 ). Thus, MGEs are important to bacterial evolution because their exchangeability increases opportunities to recombine, reassort accessory cargo genes, and diversify. Resolving relationships of MGEs is essential for understanding the evolution of traits they encode. Integrative conjugative elements (ICEs) and plasmids are two major classes of MGEs and both carry cargo genes that encode traits, such as virulence and antibiotic resistance, associated with transitions in the evolution of bacteria ( 6 , 7 ). ICEs typically recombine into chromosomes and replicate passively, while plasmids typically replicate independently from the chromosome. ICEs encode integrases that can mediate site-specific recombination between homologous attachment ( att ) sequences located on the ICE and chromosome, which are often in a conserved gene, such as a tRNA gene. ICEs can also excise, circularize, mobilize one strand through a type IV secretion system (T4SS), and recombine into the genome of recipient strains and back into that of the donor strain. The identification of ICEs, and their distinction from nonmobile genomic islands, is confounded by challenges in determining att sites as well as by compositional variation and presence of many repeat sequences that fragment these elements ( 8 ). Consequently, fundamental aspects of the mechanisms that generate diversity and the extent of variation within related ICEs are poorly understood. MGEs are crucial for the ability of many taxa of rhizobia to carry out symbiotic nitrogen fixation (SNF), a service essential to ecosystems ( 9 ). Beneficial rhizobia are defined by two sets of functions, the capacity to nodulate hosts and the ability to fix nitrogen, that are often encoded on MGEs. Symbiosis ICEs and plasmids have clusters of nod , nol , and noe genes (collectively nod genes here) for the synthesis of Nod factors, signaling molecules that initiate interactions and influence host specificity, as well as clusters of nif/fix genes for the catalysis of nitrogen fixation ( 10 ). Common NodABC proteins synthesize the core signaling structure while others, encoded by genes polymorphic in presence/absence, decorate the core with diverse substitutions ( 9 ). Type III secretion system (T3SS)- and effector-encoding genes are frequently linked to symbiosis genes ( 11 ). Effector genes influence host specificity because of their dichotomous abilities in dampening and inciting plant immune responses ( 12 ). Symbiosis ICEs (symICEs) were first characterized in Mesorhizobium ( 10 , 13 , 14 ). Mesorhizobium symICEs adopt monopartite or polypartite structures, with the latter in Mesorhizobium consisting of three elements, each encoding their own integrase, that interact to circularize and mediate genomic rearrangements during integration or excision from the chromosome ( 13 ). In agronomic landscapes – where symICEs have been extensively studied – transfer of entire symICEs promotes diversification of Mesorhizobium strains that nodulate crop hosts, but with mixed results for effective symbiosis. For instance, in settings where legume crops and compatible rhizobia were introduced by growers, transfer of an entire symICE from highly effective inoculant strains to native rhizobia occurred and generated a diversity of novel nodulating strains; however, many of them were ineffective in fixing nitrogen and the bases for loss of SNF on target crops remain unknown ( 15 – 17 ). Moreover, processes that drive symICE variation, e .g., monopartite and polypartite and diverse integration sites, as well as the selective advantages for such variation in nitrogen fixing bacteria are also unknown ( 18 , 19 ). In contrast to other rhizobia, most members of Bradyrhizobium are traditionally thought to have genes necessary for SNF clustered in a genomic island referred to as a symbiosis island (SI) ( 20 ). Bradyrhizobium is cosmopolitan and its members can fix nitrogen in facultative associations with diverse members of the legume family, Fabaceae. ( 21 ). Host species include at least 24 of the 33 legume tribes that can form nodules, spread across the three legume subfamilies, Caesalpinioideae, Mimosoideae, and Papilionoideae. Bradyrhizobium populations have been extensively characterized in native host communities, and shown to exhibit broad variation in symbiotic capacity, providing natural tests to investigate genomic drivers of this variation ( 22 ). We compiled and analyzed a data set of genome sequences of genetically and phenotypically diverse strains of Bradyrhizobium ( Data set S1 ; Extended Data sets S1–S3 available at https://github.com/osuchanglab/BradyrhizobiumPangenomeManuscript/tree/main/Extended_Supplementary_Materials ). Critical to resolving the drivers of symbiosis variation in natural settings, we sequenced genomes of 85 strains (here metapopulation strains) isolated from across an 800 km transect of wild Acmispon strigosus populations in California and phenotyped the strains as beneficial (Nod + /Fix + ), ineffective (Nod + /Fix-), or non-nodulating (Nod-/Fix-) on this host species ( 22 , 23 ). We additionally included 167 publicly available genome sequences of strains beneficial to plants of many legume tribes as well as strains considered non-nodulating, photosynthetic, or not isolated from plants. Here, findings suggested that the SI of Bradyrhizobium represents a diverse set of symICEs. We additionally demonstrate that recombination among symICEs and with nonsymbiosis ICEs as well as nonsymbiosis plasmids generates tremendous structural and functional diversity. Modularity of genes that contribute to SNF and their presence on mobile genetic elements are key to generating variation and conferring robustness to this ecologically important trait. 10.1128/mbio.00074-22.8 DATA SET S1 Strain Metadata. Download Data Set S1, XLSX file, 0.03 MB . Copyright © 2022 Weisberg et al. 2022 Weisberg et al. https://creativecommons.org/licenses/by/4.0/ This content is distributed under the terms of the Creative Commons Attribution 4.0 International license .",
"discussion": "DISCUSSION Here, we showed that modularity and reshuffling of genes by mobile genetic elements generate uncooperative genotypes of rhizobia and make individual partnerships unstable, but these same properties are fundamental for robustness and extending beneficial associations to diverse host species, as well as transferring symbiotic capacity among diverse rhizobia. We sampled native bacteria from an 800 km transect of wild Acmispon strigosus populations and included strains various in symbiosis phenotypes. We applied a strategy, developed for virulence plasmids, to study symbiosis ICEs and plasmids in Bradyrhizobium , and in doing so we were able to infer relationships among many symbiosis MGEs and model interactions among MGEs that led to their diversification ( 31 ). Findings suggested that symICE circularization reorganizes genes and is predicted to promote the shuffling of blocks of SNF genes into different ICE backbones, the generation of new combinations of modules, the acquisition of genes from the chromosome onto the symICE, and transfer of the SNF trait to plasmids. In this regard, despite being integrated in chromosomes for most of their life cycle, symICEs are like plasmids in promoting more rapid evolution of their cargo genes ( 32 ). The symICEs described here are incredibly diverse, and even metapopulation strains isolated from a single plant host species and sampled in one US state can have different symICE types, subtypes, or symbiosis plasmids. No two symICEs isolated from metapopulation strains or in this data set have identical gene content or sequence. This reflects multiple scales of modularity. Plants can be host to diverse species of Bradyrhizobium , strains can host diverse and exchangeable symbiosis MGEs, which themselves exchange and acquire genes, and individual genes within functional units can vary. The patterns that we uncovered in the metapopulation strains, including the role of MGEs in gene reshuffling and diversification, extend broadly across the Bradyrhizobium genus. For example, most analyzed photosynthetic nitrogen fixing Bradyrhizobium strains have only nif/fix islands and are restricted in host range ( 25 , 33 ). But two strains acquired symICEs that expanded their host ranges. Notably, strain ORS285 has an island that we suggest is a remnant of a symICE located at tRNA-Ile that includes nod , T3SS-associated genes, and an integrase gene but no nif/fix genes or conjugation machinery-encoding locus ( 26 ) ( Fig. S1B and S2B ). Moreover, several variants of symICEs have also recurrently gained fixNOQP and fixGHIS , genes necessary for respiration in microoxic root nodules and typically in fix gene cluster III located in the chromosome ( 34 ) (Extended Fig. S4 ). Acquisition of these symICE variants extended SNF to strains that lack fix gene cluster III and would not otherwise be capable of supporting SNF. Thus, acquisition of MGEs, and the traits that they encode, appear to play a major role in the adaptation of Bradyrhizobium to novel lifestyles. We propose that diversification in the Mesorhizobium genus, which exhibits many parallel patterns, is also driven by acquisition and reshuffling of MGEs ( 19 ). Importantly, key aspects of our study differ from those of Mesorhizobium , where diversification has been primarily characterized in managed systems. In Mesorhizobium , entire symICEs originating from inoculum strains were predicted to be transferred into indigenous nonsymbiotic rhizobia, or among strains already naturalized under monoculture crops, a scenario that imposes intense selection for host crop compatibility ( 14 , 16 , 17 , 19 ). Conversely, our findings are based on investigation of phenotypically variable strains of Bradyrhizobium isolated from diverse native plant communities, where multiple legume species overlap and select for differential subsets of rhizobia ( 35 – 37 ). Our study suggested that symICE transfer has recurrently promoted novel host acquisition, and that loss of effectiveness on one host is associated with gains of other hosts, processes that likely require a diverse array of potential hosts. Reconceiving symbiosis as a dynamical system with links that can form and dissolve among symbionts and between symbionts and hosts is essential for revealing emergent properties. Modularity and flexibility of genetic elements, coupled to their mobility, drive diversification, giving rise to variation in symbiosis, such as that revealed in Bradyrhizobium. Modularity and flexibility are central to robustness ( 3 ). A fundamental principle of robustness is that it maintains the function of a system, not a state ( 38 ). Therein lies the source of the paradox where symbiosis functions are maintained at the system level but lost from individual states, such as a symbiont, partnership, or host ( 2 ). Models that reduce symbiosis to bipartite partnerships and ignore symbiont-symbiont interactions unknowingly neglect major sources of variation and overlook robustness ( 2 , 39 ). Additionally, by separating symbiosis into categorical partnerships, these models fail to recognize the effects of multiple and various symbiont-host interactions within the system. Host species select for different combinations of symbiosis genes in their bacterial partners. Pangenome evolution, shaped by individual and gene-level selection, reassorts genes into new combinations that can extend symbiosis to new host species. Hence, alignment of fitness interests between host and symbiont is necessary for persistence of a partnership while interactions diverse in partners are necessary for robustness and evolutionary stability of symbiosis ( 3 ). This alternative framework provides a predictive understanding of symbiosis functions that are encoded on MGEs. SNF is evolutionarily stable despite repeated abandonment by both symbiont and host species ( 22 , 40 ). Conversely, symbioses involving vertically transmitted endosymbiotic bacteria with closed pangenomes are not as robust and are at higher risks of extinction ( 41 ). In agriculture, elite rhizobia strains are often added to monocultures in attempt to establish a highly specific and optimal partnership. Success is difficult to achieve because the system is flexible, and plants can partner with different genotypes of rhizobia ( 9 ). Even if the optimal partnership is attained, the likelihood for it to persist is low because of potential trade-offs between state optimality and system robustness ( 42 ). Strategies that promote interactions between multiple lineages of beneficial nitrogen-fixing rhizobia and diverse crops will have greater success for long-term sustainability."
} | 4,071 |
26872043 | null | s2 | 7,342 | {
"abstract": "Marine sediments are the largest carbon sink on earth. Nearly half of dark carbon fixation in the oceans occurs in coastal sediments, but the microorganisms responsible are largely unknown. By integrating the 16S rRNA approach, single-cell genomics, metagenomics and transcriptomics with (14)C-carbon assimilation experiments, we show that uncultured Gammaproteobacteria account for 70-86% of dark carbon fixation in coastal sediments. First, we surveyed the bacterial 16S rRNA gene diversity of 13 tidal and sublittoral sediments across Europe and Australia to identify ubiquitous core groups of Gammaproteobacteria mainly affiliating with sulfur-oxidizing bacteria. These also accounted for a substantial fraction of the microbial community in anoxic, 490-cm-deep subsurface sediments. We then quantified dark carbon fixation by scintillography of specific microbial populations extracted and flow-sorted from sediments that were short-term incubated with (14)C-bicarbonate. We identified three distinct gammaproteobacterial clades covering diversity ranges on family to order level (the Acidiferrobacter, JTB255 and SSr clades) that made up >50% of dark carbon fixation in a tidal sediment. Consistent with these activity measurements, environmental transcripts of sulfur oxidation and carbon fixation genes mainly affiliated with those of sulfur-oxidizing Gammaproteobacteria. The co-localization of key genes of sulfur and hydrogen oxidation pathways and their expression in genomes of uncultured Gammaproteobacteria illustrates an unknown metabolic plasticity for sulfur oxidizers in marine sediments. Given their global distribution and high abundance, we propose that a stable assemblage of metabolically flexible Gammaproteobacteria drives important parts of marine carbon and sulfur cycles."
} | 450 |
35677905 | PMC9168468 | pmc | 7,343 | {
"abstract": "Salinization poses great threats to soil fungal communities that would cause the losses of ecosystems services. Soil fungal communities are composed of different functional guilds such as saprotrophic, symbiotrophic, and pathotrophic fungi, and each guild includes many rare taxa and a few abundant taxa. Despite of low abundance, rare taxa may be crucial in determining the responses of entire soil fungal communities to salinization. However, it remains poorly understood how rare taxa mediate the impacts of soil salinization on soil fungal community structure. Here, we took advantage of a salinity gradient in a desert ecosystem ranging from 0.60 to 31.09 g kg −1 that was created by a 12-year saline-water irrigation and assessed how the rare vs. abundant taxa of soil saprotrophic, symbiotrophic, and pathotrophic fungi respond to soil salinization through changes in the community biodiversity and composition. We found that the rare taxa of soil saprotrophic, symbiotrophic, and pathographic fungi were more sensitive to changes in soil salinity compared to the abundant taxa. In addition, the community composition of rare taxa of the saprotrophic and pathotrophic fungi not the symbiotrophic fungi was positively associated with soil salinity change. However, the symbiotrophic fungi showed greater variations in the species richness along the salinity gradient. These findings highlight the importance to differentiate rare taxa in predicting how the biodiversity and functional groups of soil fungal communities respond to soil salinization.",
"conclusion": "Conclusion Soil fungal communities are integral to soil carbon cycling, plant nutrition, and pathogenicity. This study indicates that the rare taxa of soil functional fungal guilds (i.e., saprotrophic, pathotrophic, and symbiotrophic fungi) rather than the abundant taxa would be more sensitive to changes in soil salinity, thus the biodiversity responses of fungal guilds to soil salinization should be reflected by rare taxa. Changes in soil salinity increased the diversity of rare saprotrophic and pathotrophic fungal taxa and indirectly affected their community composition. Compared to saprotrophic and pathotrophic fungi, symbiotrophic fungi increased more in alpha diversity but did not significantly change the community composition with increasing soil salinity. These results advance our understanding of the responses of rare taxa of functional fungal guilds to soil salinization and contribute to soil biodiversity conservation and soil health management regarding fugal functionality in the inland saline environments.",
"introduction": "Introduction Salinity affects about 1.1 billion hectares of land surface around the world, accounting for 7.4% of global land area ( Ivushkin et al., 2019 ). Soil salinization is becoming a major environmental challenge specifically in drylands due to climate change and/or poor land management, such as saline-water irrigation ( Rath and Rousk, 2015 ). Soil salinization causes osmotic pressure, nutritional imbalance, and ion toxic effects to plants and microorganisms ( Mansour, 2000 ; Munns and Tester, 2008 ), which threatens ecosystem functions and services ( Rath et al., 2016 ). As one of the most diverse and abundant groups of soil microbiota, soil fungi are important for mitigating the negative impacts of salinization on soil functionality ( Ruppel et al., 2013 ; Frac et al., 2018 ). For instance, soil fungi use low-quality organic matter in infertile soils ( Bardgett, 2005 ) and show higher resistance to environmental stresses over soil bacteria ( Rath et al., 2016 ; Thiem et al., 2018 ). Soil fungi display stable network attributes and can maintain high fungal biomass per unit mass of soil organic matter under saline conditions ( de Vries et al., 2018 ; Rath et al., 2019 ). Therefore, undersatnding the impacts of soil salinization on fungal communities is vital for the conservation of belowground biodiversity in the deteriorating dryland environments. Soil fungal communities are functionally diverse and can be classified into three major functional guilds based on their trophic modes, such as saprotrophic fungi, pathotrophic fungi, and symbiotrophic fungi ( Yang et al., 2016 ; Frac et al., 2018 ), which may adapt to soil salinization differently. Saprotrophic fungi decompose litter and transform nutrients and their activities depend on the quality and quantity of soil organic matter ( Rath et al., 2019 ). In contrast, pathotrophic fungi affect disease, pests, and the growth of other organisms ( Frac et al., 2018 ). Symbiotrophic fungi improve plant nutrition by establishing plant-mycorrhizal associations, and their abundance and composition depend on soil nutrient conditions ( Moore et al., 2021 ). Given various functions of soil fungal guilds, treating soil fungal communities a whole without differentiating functional differences may mislead the prediction regarding how the functionality of soil fungal communities responds to salinization. For instance, the relative abundance of saprotrophic fungi increases and that of pathotrophic fungi decreases after ecosystem restoration in saline-alkaline soils ( Xu et al., 2021 ). Although fungal taxonomic groups increase or decrease significantly with salinity changes ( Kim et al., 2019 ; Yang et al., 2020 ), the responses of functional guilds of soil fungi to salinization are still poorly understood. This hinders our understanding of the mechanisms driving for soil fungal communities in dealing with soil salinization and the cascading effects on ecosystem processes. Communities of soil fungal guilds are comprised of many rare taxa and a few abundant taxa ( Balbuena et al., 2021 ). Current studies mainly focus on the abundant soil fungi, while little is known about how the rare fungal taxa respond to soil salinization. Although a number of studies report that fungal diversity does not change along salinity gradients ( Mohamed and Martiny, 2011 ; Bharti et al., 2015 ), increasing evidences show that rare fungal taxa are more sensitive to salinity or other stresses than abundant taxa ( Wan et al., 2021 ). Compared with the abundant taxa, the rare taxa of soil fungal guilds have fewer niches and less clustered phylotypes and can be influenced more by selective pressure and dispersal effects, thus they would be more likely subject to extinction and local environmental variations ( Galand et al., 2009 ; Pascoal et al., 2021 ; Wan et al., 2021 ). Evidence from forest and wetland soils show bigger changes in the rare microbial taxa with environmental variations compared to the abundant taxa ( Oono et al., 2017 ; Ji et al., 2020 ; Wan et al., 2021 ). In addition, soil functional fungal guilds have different responses to salinity changes ( Xu et al., 2021 ), which might be induced by the rare taxa. Vanegas et al. (2019) have found that saprotrophic fungi have higher sequence abundance than the symbiotrophic fungi and were expected to be more resistant to salinity changes than the latter. But another study observed that saprotrophic fungi displayed greater variations with salinity changes than symbiotrophic fungi ( Bencherif et al., 2015 ), and that the sequenicng abundance of some rare taxa of saproptrophic fungi was higher than that of the abundant taxa of symbiotrophic fungi ( Yang et al., 2019 ). These findings suggest that the rare taxa of saprotrophic fungi may determine the responses of this functional fungal guild to salinity change. Therefore, we need to elucidate how the rare vs. abundant taxa of functional fungal guilds respond to soil salinization in order to better understand the adaptation mechanisms of soil fungal community in dealing with salinity stress. In this study, we examined the spatial variability in soil fungal community along a soil salinity gradient ranging from 0.60 to 31.09 g kg −1 to address two questions: (1) How do the diversity and community composition of three fungal guilds respond to soil salinization? (2) Do the rare fungal taxa determine the response of soil fungal communities to salinization? We predict that the diversity of symbiotrophic fungi increases with soil salinity and that of saprotrophic and pathotrophic fungi decrease along the salinity gradient, because symbiotrophic fungi depends more on plants in high saline soils but the growths of sapraotrophic and pathotrophic fungi are inhibited ( Bencherif et al., 2015 ). In addition, symbiotrophic fungi are expected to occupy more niches and sapraotrophic and pathotrophic fungi occupy less niches with increasing soil salinity ( Figure 1 ). Due to the weaker adaption of microbial rare taxa to environmental changes compared to abundant taxa ( Montoya-Ciriaco et al., 2020 ), we predicted that the rare taxa of soil functional fungal guilds are more sensitive to soil salinization than the abundant taxa. Figure 1 The conceptual framework outlines changes in community structure of three functional guilds of soil fungi along a soil salinity gradient. (A) Alpha diversity. We expect that the alpha-diversity of symbiotrophic fungi would increase with soil salinity because symbiotrophic fungi depend highly on host plant at high-salinity sites ( Ren et al., 2016 ), while that of soil saprotrophic and pathotrophic fungi would decrease with soil salinity because high salinity can hinder the activity of soil fungi that mainly use soil organic matter ( Rath et al., 2016 ). (B) Community composition. Since soil symbiotrophic, sapraotrophic, and pathotrophic fungi have different niches ( Frac et al., 2018 ), we therefore expect that symbiotrophic fungi occur at higher salinity levels, then do pathotrophic and sapraotrophic fungi.",
"discussion": "Discussion Salinity Promotes the Relative Abundance and Species Richness of Symbiotrophic Fungi Over Saprotrophic Fungi The relative abundance of symbiotrophic fungi increased along the salinity gradient, but those of saprotrophic and pathotrophic fungi did not change significantly ( Figure 2 ). Furthermore, the species richness of symbiotrophic fungi showed greater increases to soil salinization than the other two fungal guilds. These results partially supported our prediction that the diversity of symbiotrophic fungi increases with soil salinity and that of saprotrophic and pathotrophic fungi decrease along the salinity gradient. Symbiotrophic fungi can establish mutualistic associations with plants and alleviate the negative impacts of salinization on plants by promoting plant K + uptake and antioxidant enzymes activity ( Akyol et al., 2020 ). In this study, 59.9% symbiotrophic fungi were arbuscular mycorrhizal fungi. In high saline conditions, symbiotrophic fungi could assist plants in maintaining their productivity ( Bencherif et al., 2015 ). Although the functions of soil fungi were not measured, the response of functional guilds to soil salinization could provide the implications for ecosystem functions. For instance, the diversity of symbiotrophic fungi greatly prompts plant biodiversity, nutrient uptake, and productivity in microcosm ( van der Heijden et al., 1998 ). So, increasing relative abundance and species richness of symbiotrophic fungi would improve the plant uptake of water and mineral nutrients at high salt concentrations. The significance of symbiotrophic fungi to plant growth increases remarkably at higher salt concentrations ( Ren et al., 2016 ). Similarly, the spores and biomass of symbiotrophic fungi increase with increasing salinity in natural sites ( Bencherif et al., 2015 ). More symbiotrophic fungal taxa are associated with plants and plants are less selective in choosing symbiotrophic fungi under stress compared to the unstressed condition ( Lin et al., 2021 ). Increased symbiotrophic fungi (i.e., relative abundance and species richness) were likely to meet the demand of plants on nutrients and offset the adverse impacts of salinity on plant productivity at high salt concentrations. In contrast, saprotrophic fungi were the most abundant among three functional fungal guilds and their relative abundance is highly resistant to salinity changes ( Figure 2 ), which has been observed in semi-arid mangroves ( Vanegas et al., 2019 ). It is probably because saprotrophic fungi are actively involved in organic matter decomposition and depend highly on organic inputs. Elmajdoub and Marschner (2015) found that repeated organic matter amendments can alleviate the adverse impacts of salinization on the growth of saprotrophic fungi. In this study, the aboveground plant biomass is similar along the salinity gradient ( Supplementary Figure 2 ), resulting in relatively steady organic matter inputs to soil. Similarly, the saprotrophic fungi to microbial biomass ratio slightly increases with salinity in a plant residue addition experiment ( Wichern et al., 2006 ). The resistance of saprotrophic fungi in terms of relative abundance may be confirmed by the functions, such as litter decomposition and nutrient release in the shrub shelterbelts under soil salinization. Like saprotrophic fungi, the relative abundance of pathotrophic fungi were not correlated with soil salinity, suggesting that pathogenic fungi either adapt or be resistant to salinity changes. Until now, few studies have investigated the response of pathotrophic fungi to soil salinization and little is known about the reasons. Five species of pathotrophic fungi for sugarbeet seedlings exhibited great tolerance to increasing soil salinity ( El-Abyad et al., 1988 ), agreeing with the result of this study. The species richness of three fungal guilds all increased with increasing soil salinity but in different magnitudes, and the saprotrophic and pathotrophic fungi showed smaller changes than the symbiotrophic fungi ( Figure 3A ). We speculate that the increasing species richness of saprotrophic and pathotrophic fungi to salinity changes observed here might be attributed to high plasticity of soil fungi under salinity stress ( Rath et al., 2019 ). The plasticity of soil fungi suggests that fungal individuals, over the course of a lifetime, shift trait expression in dealing with changes in the environment ( Coleine et al., 2022 ). High plasticity of soil fungi may imply great stress tolerance but low growth and death rates ( Plemenitas et al., 2014 ). But we are aware that the plasticity of soil fungi should be measured for active fungal communities by using metagenomics or transcriptomics. Thus, advanced techniques should be used to disentangle the mechanisms that drive different responses of fungal functional guilds to salinity stress in future. Soil organic carbon (SOC) was an important carbon resource for saprotrophic fungi ( Bardgett, 2005 ) and was significantly positively correlated to soil salinity ( Supplementary Figure 4 ). The increment of SOC implies higher soil porosity and water retention capacity ( Pathy et al., 2020 ), providing shelters for soil fungi. Meanwhile, the impacts of geographic distance and soil depth on fungal community dissimilarity imply the stochastic process and dispersal limitation of soil fungi ( Wang et al., 2021 ). The low dispersal rate of soil fungi could be caused by special fungal propagules and the harsh environmental barriers ( Zhou and Ning, 2017 ). Thus, we reason that in studied sites heterogeneous resource at high-salinity soils due to high SOC content could facilitate the migration and occupation of saprotrophic and pathotrophic fungi. In addition, stressful habitats that limit fungal growth may decrease resource competition and increase fungal diversity by suppressing dominating species ( Oono et al., 2020 ). More taxa of saprotrophic fungi would be coexisted with high soil salinity, because the physiological activity of saprotrophic fungi might be inhibited by increasing salinity ( Rath et al., 2019 ). This demonstrates that salinity stress and SOC resource may jointly shape the species richness of saprotrophic fungi under soil salinization. Rare Taxa of Functional Fungi Display Higher Diversity Sensitivity to Soil Salinity This study showed that the diversity of the rare fungal taxa ( q = 0) not the abundant taxa ( q = 1, 2) displayed significant correlations with soil salinity and was more susceptible to other soil properties ( Figure 4 ). The different patterns between the diversities of rare ( q = 0) and abundant fungal taxa ( q = 1, 2) indicate that the rare taxa of functional guilds are more sensitive to increasing soil salinity than the abundant taxa. Previous studies found that abundant fungi display great resistance to drought stress and ecosystem disturbance ( Sun et al., 2017 ; de Vries et al., 2018 ). The activity or growth of abundant fungal taxa might be constricted due to high soil salinity or low-quality carbon resource in the desert soils, which promotes the migration rate and diversity of rare fungal taxa ( Rath et al., 2016 ). The growth of a dominant root fungus (the Inula viscosa ) was strongly inhibited by soil salinity in vitro assays and the fungal diversity increased at sites with high soil salinity ( Macia-Vicente et al., 2012 ). Oono et al. (2020) indicate that abundant fungal taxa were suppressed in stressful conditions, which promoted the population of rare fungal taxa and increased fungal diversity. The population size of rare fungi would be enlarged with increasing SOC along the salinity gradient, because high SOC content provides more niches for rare fungi. Moreover, the great sensitivity of rare taxa of soil functional fungal guilds to salinity changes could attribute to their high variability, as rare taxa could function as a large reservoir of genetic traits and have redundant functions ( Boraks et al., 2020 ; Pascoal et al., 2021 ). Therefore, the diversity of rare fungal taxa in this study increased with salinity changes. This study found that the diversity of rare fungal taxa is more sensitive to soil phosphorus compared to that of abundant fungal taxa. Rare fungal taxa that have phosphorus cycling functions in saline soils can be influenced by soil total phosphorus and available phosphorus. The diversity of rare symbiotrophic fungi ( q = 0) was positively correlated with soil pH as well as soil available phosphorus ( Figure 4C ). Symbiotrophic fungi are associated with plant P and nitrogen nutrition ( Ezawa et al., 2002 ; Lindahl and Tunlid, 2015 ). In P-poor soils, such as the desert soil with high pH, the mutualistic association between plant and symbiotrophic fungi may be stimulated, causing a high diversity of rare symbiotrophic fungi. Phosphorus uptake of symbiotrophic fungi was found to increase at higher soil pH ( Hinsinger, 2001 ). Unlike the diversities of abundant taxa ( q = 1 and 2), the diversity of rare saprotrophic and pathotrophic fungi ( q = 0) was negatively correlated with soil total phosphorus and available phosphorus ( Figures 4A , B ), suggesting that rare taxa may play more important roles in phosphorus cycling than abundant species. The diversity of rare saprotrophic and pathotrophic fungi was positively correlated with soil alkaline phosphatase ( Supplementary Table 16 ). Community Composition of Functional Fungi Indirectly Influenced by Soil Salinity and Mainly Through Rare Taxa The community composition of rare saprotrophic and pathotrophic fungi showed higher positive correlations with salinity changes (distance of soil salinity) compared to those of abundant taxa. This result supported the prediction that rare taxa of the functional guilds of soil fungi are more sensitive to salinity changes than the abundant taxa. However, compared with the diversity, the community composition of rare fungal taxa was not significantly correlated with soil salinity partially, suggesting that soil salinity may have indirect effects on community composition of rare fungal taxa. Kim et al. (2019) reported that fungal community composition in the salinity treatment is mainly driven by soil salinity. The impacts of salinity on fungal community composition might be offset by organic matter, as soil salinity and organic carbon were significantly positively correlated ( Supplementary Figure 4 ) and exhibited opposite effects on fungal community composition in a semi-arid mangrovein salt mash ( Vanegas et al., 2019 ). The community composition of rare saprotrophic and pathotrophic fungi could be indirectly affected by soil salinity through altering soil phosphorus and clay content. Rare taxa of saprotrophic and pathotrophic fungi (i.e., Sorensen index) not the abundant taxa (i.e., Morisita-Horn index) were significantly influenced by changes in soil phosphorus (distance of soil phosphorus; Supplementary Table 11 ). Similar to this study, the community composition of sapraotrophic and pathotrophic fungi in a natural grassland is mainly driven by soil phosphorus ( Delavaux et al., 2021 ). In current study, soil salinity showed significantly positively correlation with soil phosphorus and clay content and was not correlated with soil total nitrogen, which were controlling factors for the community composition of rare taxa (i.e., Sorensen index). These correlations confirmed that the responses of community composition of saprotrophic and pathotrophic fungi to soil salinization might be induced mainly by rare taxa (i.e., Sorensen index), as abundant taxa have wider variability in abundance under changing resource availability, e.g., soil total nitrogen ( Newton and Shade, 2016 ). The community composition of rare and abundant symbiotrophic fungi (i.e., Sorensen, Horn, and Morisita-Horn indices) showed no correlations with salinity changes, indicating that salinization had similar stresses on the rare and abundant symbiotrophic fungi. This result partially supported the prediction that the community composition of symbiotrophic fungi is expected to occupy high-salinity niches compared to that of saprotrophic and pathotrophic fungi. The community composition of symbiotrophic fungi was mainly regulated by soil total nitrogen. Symbiotrophic fungi provide mineral nitrogen and phosphorus nutrients to plants in exchange for carbohydrates ( Kiers et al., 2011 ) and enhance nitrogen accumulation in plants with inreasing soil salinity ( Zhu et al., 2018 ), as the accumulation of leaf non-protein nitrogen increase plant resistance to salinity ( Mansour, 2000 ). Given that the leaf N content and N:P ratio of H. ammodendron at high-salinity sites were significantly higher than those at low-salinity sites (Feng et al., in preparation; Supplementary Figure 5 ), symbiotrophic fungi that provide nitrogen would be promoted with increasing salinity due to plant selection, especially at the high-salinity sites. Furthermore, different symbiotrophic fungi can acquire N from distinctly sources, e.g., ammonium N and nitrate N ( Cox et al., 2010 ) that are influenced by soil total N ( Supplementary Figure 6 ). Therefore, the community composition of symbiotrophic fungi was driven by soil total N under salt stress and resistant to changes in soil salinity. Conclusion Soil fungal communities are integral to soil carbon cycling, plant nutrition, and pathogenicity. This study indicates that the rare taxa of soil functional fungal guilds (i.e., saprotrophic, pathotrophic, and symbiotrophic fungi) rather than the abundant taxa would be more sensitive to changes in soil salinity, thus the biodiversity responses of fungal guilds to soil salinization should be reflected by rare taxa. Changes in soil salinity increased the diversity of rare saprotrophic and pathotrophic fungal taxa and indirectly affected their community composition. Compared to saprotrophic and pathotrophic fungi, symbiotrophic fungi increased more in alpha diversity but did not significantly change the community composition with increasing soil salinity. These results advance our understanding of the responses of rare taxa of functional fungal guilds to soil salinization and contribute to soil biodiversity conservation and soil health management regarding fugal functionality in the inland saline environments."
} | 6,015 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.