pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
28624614 | PMC5496983 | pmc | 9,072 | {
"abstract": "Summary The Bartonella gene transfer agent (BaGTA) is an archetypical example for domestication of a phage-derived element to permit high-frequency genetic exchange in bacterial populations. Here we used multiplexed transposon sequencing (TnSeq) and single-cell reporters to globally define the core components and transfer dynamics of BaGTA. Our systems-level analysis has identified inner- and outer-circle components of the BaGTA system, including 55 regulatory components, as well as an additional 74 and 107 components mediating donor transfer and recipient uptake functions. We show that the stringent response signal guanosine-tetraphosphate (ppGpp) restricts BaGTA induction to a subset of fast-growing cells, whereas BaGTA particle uptake depends on a functional Tol-Pal trans -envelope complex that mediates outer-membrane invagination upon cell division. Our findings suggest that Bartonella evolved an efficient strategy to promote genetic exchange within the fittest subpopulation while disfavoring exchange of deleterious genetic information, thereby facilitating genome integrity and rapid host adaptation.",
"introduction": "Introduction Clonal populations typically deteriorate due to gradual loss of fitness and eventual extinction caused by accumulation of slightly deleterious mutations via genetic drift ( Charlesworth et al., 1993 , Muller, 1964 ). This evolutionary mechanism, known as Muller's ratchet, appears to be the typical fate of pathogens that thrive on complex infection cycles, including intracellular parasitism ( Iranzo et al., 2016 , Moran, 1996 ), that experience frequent population bottlenecks. One route of actual escape from Muller's ratchet appears to be gene acquisition via horizontal gene transfer (HGT), resulting in either repair of a deleteriously mutated gene by a functional copy or acquisition of new genes that offsets the deleterious effects of accumulating mutations ( Takeuchi et al., 2014 ). A prominent source of HGT in bacterial communities is gene transfer agents (GTAs). GTAs are chromosomally encoded DNA transfer systems that promote genetic exchanges between individual cells through cell lysis and release of bacteriophage-like intermediates. The current view is that a fraction of a bacterial population sacrifices itself to serve as GTA donors, producing phage-like particles that are packed with a random portion of their chromosomal DNA. The remaining part of the population may act as recipient, taking up the particles and incorporating the contained DNA into their chromosome ( Figure 1 A and for review, see Lang et al., 2012 ). Considering the limited DNA packing size of GTAs (less than the size of the locus encoding the GTA function) and the apparent lack of sequence specificity, GTAs are considered to be strictly vertically inherited. This is in contrast to bacteriophages, which largely promote exchange of phage-specific genes and are therefore self-transmissible. In this respect, GTAs can be considered as a form of “domesticated” prophage—that is, ancestrally derived from a bacteriophage genome but altered by the host to confer an adaptive benefit—and thus represent one of many phage-derived adaptive functions observed in bacterial genomes ( Bobay et al., 2014 , Hynes et al., 2016 ). It is, however, unclear how maintenance of a system that can induce the lysis of individual cells may still be beneficial to the entire population. Figure 1 Chromosomal Marker Exchanges during B. henselae Co-culture (A) Schematic model for BaGTA-mediated genetic exchange in Bartonella . In a three-stage process, induction of BaGTA in donor cells (blue) leads to host cell lysis (purple) and release of BaGTA particles that are taken up by recipient cells (green). (B) Induction dynamics of the BaGTA locus as detected by FACS analysis of gfp - bgtC reporter strain upon cultivation of Bartonella in M199. A subpopulation of wild-type cells synchronously induces BaGTA locus (open circles) leading to early and late peaks in GTA induction. BaGTA induction is absent in cells expressing the constitutive active ppGpp synthase gene RelA 1–455 from E. coli , suggesting that elevated levels of ppGpp repress induction of BaGTA. The mean values from a biological triplicate and their associated SEs are displayed. (C) Cell density upon cultivation in M199 is shown as a spline fit (dotted line, purple) to the experimental growth data as determined by CFU counts. A temporal decrease in viable cells around 10 and 30 hr indicates BaGTA-induced cell lysis (early and late lysis). The mean values and associated SD from a biological triplicate are displayed. (D) Relative transduction dynamics as a function of the co-culturing time are shown. The experimentally determined DNA uptake rate (transduction frequency, open circles) follows a time-delayed function (black line) proportional to the integrated GFP-expression rate of the first BaGTA induction peak (blue dashed line), while the integrated GFP-expression rate of the second BaGTA induction peak (cyan dashed line) does not contribute to detectable DNA uptake. The mean values from a biological triplicate and their associated SEs are displayed. The alpha-proteobacterial genus Bartonella comprises an expanding number of arthropod-borne pathogens (e.g., lice, fleas, or keds) that share the ability to cause long-lasting intraerythrocytic infections in their mammalian reservoir hosts ( Harms and Dehio, 2012 ). Genomic sequence analysis has revealed that all bacteria of the genus Bartonella are characterized by the presence of a Bartonella -specific GTA (BaGTA), which shares no homologies to previously described GTA systems. Interestingly, BaGTA is encoded upstream from an origin of run-off replication (ROR), another conserved feature of Bartonella genomes which has been suggested to be linked to the activity of BaGTA. Phylogenetic analyses have identified BaGTA as a key innovation associated with the spectacular adaptive radiation that characterizes these zoonotic bacterial pathogens ( Guy et al., 2013 ). Although BaGTA has not been directly linked to Bartonella pathogenicity, it has been proposed to drive the exchange and the diversification of host-interaction factors within Bartonella communities such as the well-characterized VirB type IV secretion system (T4SS) and its cognate Bartonella effector proteins ( Engel et al., 2012 , Guy et al., 2012 ). Maintenance of BaGTA is likely driven by selection to increase the likelihood of genetic exchange and facilitates rapid adaptation to host-specific defense systems during infection ( Guy et al., 2013 ). In addition, BaGTA may help Bartonella avoid Muller's ratchet. As a facultative intracellular pathogen Bartonella experiences major transmission bottlenecks throughout its infection cycle, both within the arthropod and while breaching the mammalian host barriers during invasion of erythrocytes. Exposure to host defense mechanisms such as reactive oxygen species likely results in the accumulation of mutations within the pathogen’s chromosome. Most acquired mutations are neutral, only very few evolve new functions and possibly show beneficial effects, while a sizable fraction of mutations will have deleterious effects resulting in fitness losses. Despite clear genomics-based arguments pointing to a central role for BaGTA in Bartonella biology, direct experimental evidence for its activity are scarce and the molecular mechanisms underlying its activity and regulation have remained fully elusive. Here, we have taken an experimental systems biology approach based on multiplexed transposon sequencing (TnSeq) measurements and single-cell reporter assays to globally define the core components and transfer dynamics of the Bartonella GTA system. Our analysis has identified inner- and outer-circle components of BaGTA mediating donor transfer and recipient uptake functions. Moreover, rather than being a random process, we found that BaGTA specifically promotes genetic exchange between the fittest subpopulation of cells, as the stringent response signal guanosine-tetraphosphate (ppGpp) restricts BaGTA induction to a subset of fast-growing cells.",
"discussion": "Discussion We combined quantitative single-cell FACS reporter studies with multiplexed TnSeq analyses to define the regulatory networks and molecular determinants of BaGTA that mediate DNA transfer from donor to recipient cells. Using a GFP-based FACS reporter assay, we found that BaGTA is heterogeneously induced during a short time window within a subpopulation of rapidly proliferating cells, possibly at the onset of exponential growth. Our systems-genetic experiments derived from TnSeq measurements support two further key features for BaGTA ( Figure 5 ). First, because the alarmone ppGpp is a key repressor of BaGTA induction, only healthy cells can serve as DNA donors within the population. Second, uptake of donor DNA is likely restricted to actively dividing cells, as a plausible consequence of BaGTA tropism toward components of the divisome. Considering population dynamics, these findings clearly support the logical argument that an individual bacterium with a high fitness cost mutation is less likely to take up transduced DNA as the result of its dilution in a competing set-up. While our multiplexed TnSeq analysis reveals a broad panel of inner- and outer-circle components of BaGTA, additional essential genes that contribute to the functioning of BaGTA likely exist, and further studies will be required to decipher the precise involvement of the individual components identified to the global processes described. Taken together, our findings suggest that both the induction of BaGTA particle formation and uptake preferentially occur within a cell subpopulation that experience rapid cell proliferation. In evolutionary terms, this suggests that BaGTA is tuned toward achieving high rates of DNA exchange between the fittest bacteria within a population. This would maintain the species' overall genetic stability and promote the propagation of beneficial traits among the fittest subpopulations of cells ( Figure 5 A). Combination of ppGpp-dependent BaGTA induction with cell division-dependent particle adsorption ( Figure 5 B) provides an excellent strategy to promote evolvability and to circumvent Muller's ratchet in isolated bacterial communities. Furthermore, such genome-wide HGT would preferentially occur when the bacteria reach a favorable environment, such as their replicative niche in the mammalian host or the midgut of their arthropod vector. Thus, the wide spread of BaGTA among Bartonella species appears to present strong evidence of evolvability within these ubiquitous zoonotic pathogens, which increases their capacity to generate descendants with increased fitness and acquire new adaptive characteristics for efficient host infection. Moreover, the strict conservation of BaGTA observed across the bartonellae suggests that it supports an essential function during the infection cycle of these bacteria, where a high selection pressure could avoid the emergence of cheaters. Further studies should determine whether GTA-mediated HGT takes place in the mammalian host, where it would provide an excellent strategy to circumvent Muller's ratchet in isolated bacterial communities. An alternative, but not exclusive, possibility is that this process happens in the arthropod vector. Despite constituting a separate transmission bottleneck, the lumen of the arthropod gut also constitutes a potential melting pot for genetically diverse genotypes with distinct successful infection history. Besides the RcGTA system, only few other GTA systems have been characterized to date. Our study indicates that despite a distinct origin, the later steps of DNA uptake and integration into recipient cells seem conserved between BaGTA and RcGTA ( Brimacombe et al., 2014 , Brimacombe et al., 2015 ), suggesting a shared uptake mechanism between all GTA systems. This also supports the emerging view that GTA-mediated DNA transfer may generally be considered as a form of transformation rather than transduction or conjugation ( Redfield, 2001 , Brimacombe et al., 2015 , Takeuchi et al., 2014 , Hynes et al., 2016 ). It is very likely that similar examples of phage-derived systems will emerge as functional analysis of sequenced genomes progress. In this context, our study also provides a methodological framework that can be applied to analyze and decipher the functionality of any gene transfer machinery at the systems level and should contribute toward extending our understanding of the molecular mechanisms underlying HGT in bacterial communities."
} | 3,169 |
37138608 | PMC10150012 | pmc | 9,073 | {
"abstract": "Numerous microorganisms and other invertebrates that are able to degrade polyethylene (PE) have been reported. However, studies on PE biodegradation are still limited due to its extreme stability and the lack of explicit insights into the mechanisms and efficient enzymes involved in its metabolism by microorganisms. In this review, current studies of PE biodegradation, including the fundamental stages, important microorganisms and enzymes, and functional microbial consortia, were examined. Considering the bottlenecks in the construction of PE-degrading consortia, a combination of top-down and bottom-up approaches is proposed to identify the mechanisms and metabolites of PE degradation, related enzymes, and efficient synthetic microbial consortia. In addition, the exploration of the plastisphere based on omics tools is proposed as a future principal research direction for the construction of synthetic microbial consortia for PE degradation. Combining chemical and biological upcycling processes for PE waste could be widely applied in various fields to promote a sustainable environment.",
"conclusion": "6. Conclusion The prevalence of plastic waste and micro/nanoplastic pollution has always been a great challenge facing environmental remediation. Traditional plastic disposal methods have not been effective in solving the global plastic waste pollution problem, which is becoming increasingly serious. However, recent studies on plastic biodegradation have led to new perspectives on the sustainable disposal of plastic waste and the discovery of an increasing number of microbes and other invertebrates that are involved in plastic degradation. PE is the most widespread and recalcitrant petroleum-based plastic and the study of its biodegradation has received considerable interest over the years. The microorganisms and other organisms involved in the degradation of PE have been extensively explored in the literature; however, there have been no breakthroughs in the study of PE biodegradation, particularly regarding the mechanisms of PE degradation and the identification of specific PE depolymerases. Recent studies have examined the constraints that single strains encounter when trying to degrade PE, and progressive efforts have been made to combine several strains to develop more effective consortia or co-cultures. Compared to single strains, functional microbial consortia offer many advantages in degrading plastic waste; however, the lack of an understanding of the byproducts and mechanisms of PE degradation has limited the development of artificial functional consortia for PE degradation, which was the focus of this review. Considering the existing bottlenecks in the construction of PE-degrading consortia and the pros and cons of each strategy, we propose combining top-down and bottom-up approaches for the synthesis of efficient PE-degrading consortia. In this way, the selective, top-down enrichment of natural microbial communities can be applied to optimize community functions, and the enriched communities can undergo omics analyses of their functions and community structures to help us better understand the microbial PE -degradation mechanisms and microbial interactions at play within the communities. The insights into microbial communities achieved using top-down strategies could contribute to a more reasonable and appropriate bottom-up synthesis of functional PE -degrading consortia. The plastisphere is a new ecological niche that spontaneously forms on plastic debris and is closely associated with plastic polymers. Investigating the microorganisms or microbial communities that grow in the plastisphere could uncover principal sources of novel PE-degrading microorganisms/microbiota. The combination of genomic, transcriptomic, metagenomic, and other omics/meta-omics analyses could help to unravel the microbial diversity and community composition of the plastispheres in different ecosystems and the metabolic relationships between microbes, which could also enable the discovery of core microbiota and key pathways/genes for PE degradation and provide a foundation for the subsequent construction of functional consortia. We close with a discussion of concerns and bottlenecks in the existing studies on PE biodegradation as well as some insights into the topics that should be considered in future research. Due to insufficient study of the mechanisms of PE biodegradation, the construction of effective microbial consortia currently faces numerous obstacles. In addition, the safety and stability of synthetic microbial consortia and the regulation of coexistence between microorganisms could represent challenges for future research due to the dynamic nature of microbial communities, biological variability, and various other uncertainties.",
"introduction": "1. Introduction Plastic pollution remediation is always a global environmental protection issue. As the most commonly used material, plastic products are, however, increasingly in demand and used in all fields globally along with socio-economic development due to their excellent characteristics such as durability, low cost and convenience, and it is predicted that the global production of plastic products will be over 26 billion tons by 2050 ( Plastics Europe, 2021 ; Zeenat et al., 2021 ; Liu et al., 2022b ). Approximately 80% of the world’s 400 million tons of plastic waste is dumped in landfills or discarded directly into the environment ( Tejaswini et al., 2022 ). And the most common types of plastic waste are divided into two categories according to their thermal properties: thermoplastics and thermosetting plastics ( Canopoli et al., 2018 ). Thermoplastics are types of linear-chain polymer compounds that have plasticity at a certain temperature, the most common of which are polyethylene (PE), polyvinyl chloride (PVC), polypropylene (PP), and polystyrene (PS). On the other hand, thermosetting plastics, such as polyurethane (PUR), cannot be melted due to irreversible thermal chemical changes ( Amobonye et al., 2021 ). The backbones of thermosetting polymers are highly crosslinked by heteroatoms, making it easy for fracture to occur at ester or amide bonds. In contrast, the junctions of the primary chains in thermoplastics, which mostly comprise carbon atoms, render them more resistant to deterioration ( Zheng et al., 2005 ). The majority of plastic waste is sent to landfill or discharged directly into the environment where it undergoes a very sluggish natural degradation process, which for PE specifically, the most inert polyolefin plastic, the half-lives were estimated to vary from decades to centuries ( Chamas et al., 2020 ). As the primary C–C-chain polymer ( Figure 1 ), PE is currently the most extensively used plastic type, accounting for approximately 38% of the market share ( Danso et al., 2019 ). PE is a thermoplastic made of ethylene that has been through high-pressure polymerization ( Kumar Sen and Raut, 2015 ). According to different densities, branching degrees, and the availability of surface functional groups, PE can be divided into low-density polyethylene (LDPE), high-density polyethylene (HDPE), linear low-density polyethylene (LLDPE), etc. ( Restrepo-Flórez et al., 2014 ). PE is widely utilized in cling film, commercial plastic bags, pharmaceutical and food packaging films, and other consumer and manufacturing industries, because of its desirable properties, such as non-toxicity, tastelessness, high tensile strength, low permeability, and durability ( Shah et al., 2008 ; Liu et al., 2021 ). Due to its large molecular weight, stable chemical structure, high hydrophobicity, crystallinity, and limited number of functional groups required for biodegradation, PE is the plastic polymer most resistant to degradation, this means that polyethylene debris has lingered in the marine and other ecosystems for two decades and is extremely resistant to degradation ( Kyaw et al., 2012 ; Dey et al., 2020 ; Krause et al., 2020 ). In addition, PE is the most prevalent municipal solid waste (MSW), and accounts for a high proportion of plastic waste in the environment ( Zhou et al., 2014 ). Therefore, understanding the mechanisms of PE degradation and devising effective and environmentally friendly methods for PE degradation could provide ideas and data to help to mitigate our ever-worsening plastic pollution on a massive scale. FIGURE 1 (A) Market share distribution of plastic types in 2021 ( Plastics Europe, 2021 ); (B) common types of polyethylene (PE) and their molecular structure. Current research indicates that the decomposition of plastic waste, such as PE, in landfills and other ecosystems is predominantly accomplished via physicochemical (abiotic) degradation and biodegradation ( Dimassi et al., 2022 ). Physical degradation, such as cracking, embrittlement, and spalling, typically modifies the primary structures of polymers; whereas chemical degradation primarily modifies the molecular structures of polymers, such as breaking bonds or the oxidation of long polymer chains to produce compounds with low molecular weights ( Andrady, 2011 ; Gewert et al., 2015 ; Ali et al., 2021 ). Mechanical degradation (including tidal forces, waves, and erosion), photo-oxidation (such as ultraviolet rays), thermal oxidation (such as incineration, pyrolysis, and gasification), and chemical hydrolysis processes (such as those involving acids, alkalis, and other organic solvents) are the primary physicochemical degradation mechanisms for PE and other plastic waste in the environment ( Kyrikou and Briassoulis, 2007 ). Among them, photodegradation and thermo-oxidative degradation are the most common mechanisms of PE degradation in the environment ( Canopoli et al., 2020 ). Through a variety of physical and chemical processes, large-sized plastic waste is broken down into plastic debris and micro/nanoplastics, which are then more easily digested and consumed by microorganisms ( Taghavi et al., 2021b ; Singh Jadaun et al., 2022 ). Researchers have become increasingly interested in the biodegradation of plastic in recent years due to the high efficiency, availability, and eco-friendliness of this approach ( Khatoon et al., 2017 ). Deepening our understanding of the natural microbial communities that arise in the plastisphere may lead to novel approaches for the development of PE-degrading consortia. The key to the construction of efficient, stable, and controllable consortia is the design, which needs to ensure that the microorganisms in these communities interact synergistically. Certain technologies, such as next-generation sequencing technologies like metagenomic sequencing, can be used to determine the microbial community structure of the plastisphere ( Gilmore et al., 2019 ; San León and Nogales, 2022 ). This study focused on PE, the most prevalent plastic polymer. We briefly examined its characteristics and biodegradation mechanisms, as well as the current research gaps and recent advances in PE biodegradation. Then, we explored the advantages of using microbial functional consortia and their potential for PE pollution remediation compared to single strains. Lastly, we emphatically investigated the prospect of combining the plastisphere concept and metagenomics to inform the construction of functional microbial consortia."
} | 2,834 |
33265470 | PMC7512898 | pmc | 9,074 | {
"abstract": "Training deep learning networks is a difficult task due to computational complexity, and this is traditionally handled by simplifying network topology to enable parallel computation on graphical processing units (GPUs). However, the emergence of quantum devices allows reconsideration of complex topologies. We illustrate a particular network topology that can be trained to classify MNIST data (an image dataset of handwritten digits) and neutrino detection data using a restricted form of adiabatic quantum computation known as quantum annealing performed by a D-Wave processor. We provide a brief description of the hardware and how it solves Ising models, how we translate our data into the corresponding Ising models, and how we use available expanded topology options to explore potential performance improvements. Although we focus on the application of quantum annealing in this article, the work discussed here is just one of three approaches we explored as part of a larger project that considers alternative means for training deep learning networks. The other approaches involve using a high performance computing (HPC) environment to automatically find network topologies with good performance and using neuromorphic computing to find a low-power solution for training deep learning networks. Our results show that our quantum approach can find good network parameters in a reasonable time despite increased network topology complexity; that HPC can find good parameters for traditional, simplified network topologies; and that neuromorphic computers can use low power memristive hardware to represent complex topologies and parameters derived from other architecture choices.",
"conclusion": "7. Conclusions Though inspired by biological neural models, deep learning networks make many simplifications to their connectivity topologies to enable efficient training algorithms and parallelization on GPUs. CNNs in particular have emerged as a standard high performance architecture on tasks such as object or facial recognition. While they are powerful tools, deep learning still has several limitations. First, we are restricted to relatively simple topologies; second, a significant portion of network tuning is done by hand; and third, we are still investigating how to implement low power, complex topologies in native hardware. We chose three different computing environments to begin to address the issues respectively: quantum adiabatic computing, high performance computing clusters, and neuromorphic hardware. Because these environments are quite different, we chose to use different deep learning models for each. This includes Boltzmann machines in the quantum environment, CNNs in the HPC environment, and SNNs in the neuromorphic environment. We chose to use the well-understood MNIST hand-written digit dataset and a neutrino particle detection dataset. Our results suggest these different architectures have the potential to address the identified deficiencies in complex deep learning networks that are inherent to the von Neumann CPU/memory architecture that is ubiquitous in computing. The quantum annealing experiment showed that a complex neural network, namely one with intralayer connections, can be successfully trained on the MNIST digit recognition and neutrino particle detection tasks. The ability to train complex networks is a key advantage for a quantum annealing approach and opens the possibility of training networks with greater representational power than those currently used in deep learning trained on classical machines. High performance computing clusters can use such complex networks as building blocks to compare thousands of models to find the best performing networks for a given problem. Finally, the best performing neural network and its parameters can be implemented on a complex network of memristors to produce a low-power hardware device capable of solving difficult problems. This is a capability that is not feasible with a von Neumann architecture and holds the potential to solve much more complicated problems than can currently be solved with deep learning on classical machines. We proposed a new deep learning architecture based on the unique capabilities of the quantum annealing, high performance computing, and neuromorphic approaches presented in this paper. This new architecture addresses three major limitations we see in current deep learning methods and holds the promise of higher classification accuracy, faster network creation times, and low power, native implementation in hardware.",
"introduction": "1. Introduction A neural network is a machine learning concept originally inspired by studies of the visual cortex of the brain. In biology, neural networks are the neurons of the brain connected to each other via synapses; accordingly, in machine learning, they are graphical models where variables are connected to each other with certain weights. Both are highly useful in analyzing image data, but practical considerations regarding network topology limit the potential of simulating neural networks on computers. Simulated networks tend to divide neurons into different layers and prohibit intralayer connections. Many-layered networks are called deep learning networks, and the restriction of intralayer connections allows rapid training on graphical processing units (GPUs). We explain some current limitations of deep learning networks and offer approaches to help mitigate them. For this article we focus on a quantum adiabatic computing approach, which is one of a trio in a larger project to survey machine learning in non-traditional computing environments, though we also describe the other approaches at a high level to offer comparison and context for experiment designs. The second approach uses a high performance computing environment to automatically discover good network topologies, albeit they remain restricted from using intralayer connections. The third approach uses neuromorphic computing as a low-power alternative for representing neural networks. Rather than explicitly choosing one solution or another, these approaches are meant to augment each other. Describing these different approaches necessitates a brief description of various machine learning models and networks including Boltzmann machines (BMs), convolutional neural networks (CNNs), and spiking neural networks (SNNs). Results obtained from CNNs and SNNs, while important to our project, are not the focus of this article and are presented in the appendix. 1.1. Boltzmann Machines A Boltzmann machine is an energy-based generative model of data. BMs contain binary units, and each possible configuration of units is assigned a certain energy based on edge weights. The goal of training is to find edge weights that result in low energy configurations for patterns more likely to occur in data. Since BMs can be represented as Ising models, and because the D-Wave processor is designed to natively solve Ising models, BMs are particularly attractive for our purposes. We tend to view BMs as probabilistic neural networks with symmetrically connected units [ 1 ]. BMs are well suited to solving constraint satisfaction tasks with many weak constraints, including digit and object recognition, compression/coding, and natural language processing. A common algorithm for training BMs exposes a BM to input data and updates the weights in order to maximize the likelihood that the underlying model of the BM reproduces the data set. This method requires computing certain quantities which, due to the specific form of the BM, turn out to be the values of certain correlation functions in thermal equilibrium. However, training is a slow and arduous task if we allow models with unrestricted topology. Connectivity loops slow down the convergence of many algorithms used to estimate thermal equilibrium properties. Simulated annealing is a generic and widely used algorithm to reach this thermal equilibrium, but this remains a slow and expensive process for large networks. This forces us to either use tiny networks or to give up complex topologies, with the latter option leading to the popular choice of using restricted Boltzmann machines (RBMs) [ 2 ]. Units in RBMs are categorized as “visible” or “hidden.” During training, the visible units of a RBM represent the input dataset whereas the hidden units represent latent factors that control the data distribution. After undergoing the above training process, an RBM will produce a distribution of visible unit states that should closely match the input dataset. Additionally, only bipartite connectivity between the two types is allowed, which makes parallel computation feasible. Figure 1 shows an example of this bipartite connectivity. Approximation algorithms make training tractable in practice, and RBMs can be stacked together to form deep belief networks (DBNs) [ 3 ]. 1.2. Convolutional Neural Networks Of the many designs for deep learning networks, CNNs have become the most widely used for analyzing image data [ 4 ]. As with other deep learning networks, CNNs contain many layers of neural units with many connections between different layers but no connections between units of a particular layer. They also use standard stochastic gradient descent and back-propagation combined with labeled data to train. What separates a CNN from other networks are its unique connectivity arrangement and different types of layers. See Figure 2 for a high-level diagram of the CNN architecture. One type of layer in CNNs is the convolutional layer. Unlike in other neural networks, a convolutional layer uses a kernel, or small set of shared weights, to produce a feature map of the input to the layer, and many convolutional layers operate in succession. Other networks would typically have every input unit connected to every processing unit in a layer whereas a CNN is satisfied with using convolution to produce sparse connections between layers—see Figure 1 for the dense connectivity of a BM and compare it against the sparse CNN connectivity shown in Figure 3 . A kernel captures a certain feature from the input, and convolving a kernel with the data finds this feature across the whole input. For example, a kernel that detects diagonal lines can be convolved with an image to produce a feature map that can be interpreted as identifying all areas of an image that contain diagonal lines. The second type of layer is the pooling layer. Pooling layers use the many feature maps produced by convolutional layers as input and subsample them to produce smaller feature maps to help take advantage of data locality within images. CNNs use alternating layers of convolutional and pooling layers to extract and abstract image features. Pooling operations makes feature detection in CNNs resilient to position shifts in images [ 5 ]. 1.3. Spiking Neural Networks SNNs differ from both BMs and CNNs by incorporating the extra dimension of time into how information is processed. BMs and CNNs do not have a sense of time built into their architectures—neural unit activity is iteratively calculated on a layer-by-layer basis. SNNs instead use integrate-and-fire neurons, units that collect activation potential over time and fire or “spike” upon reaching a threshold, after which they cannot fire during what is known as a refractory period. Additionally, synapses in a SNN can include programmable delay components, where larger delay values on the synapse correspond to longer propagation time of signals along that synapse. Additionally, there is not necessarily a division of units into well-organized layers in a SNN, and input is fed to the network over time. SNNs have great potential in moving away from the traditional implementation of machine learning algorithms on the CPU/memory von Neumann architecture. For example, the CPU/memory model, while useful on many diverse applications, has the drawback of high power requirements. Nature’s biological neural networks have extremely low power requirements by comparison. There are many different ways to implement neuromorphic systems, but one of the more promising device types to include in neuromorphic systems is memristors. Development of memristive technology opens the potential of running spiking neural networks using low power consumption on neuromorphic architectures. A key challenge associated with SNNs in general and SNNs for neuromorphic systems in particular is determining the correct training or learning algorithm with which to build the SNN. Though there have been efforts to map existing architectures like CNNs to equivalent spiking neuromorphic systems [ 6 , 7 ], there is also potential to develop independent deep learning architectures that exploit the temporal processing power of SNNs. 1.4. Challenges Complex networks pose enormous problems for deep learning, three of which we identify. How we tackle each of these challenges is the basis of our project, where we seek relief from these issues through quantum adiabatic computing, high performance computing, and neuromorphic computing. The first of these challenges comes from complex network topology in neural networks. By complex network topology we mean bidirectional connections and looping connectivity between neural units, which slow training to a crawl. The training algorithms we know for such complex networks have greater than polynomial runtime, making them effectively intractable and untenable for practical purposes. Deep networks deployed on real-world problems, like the previously discussed CNN architecture, instead impose limitations on network topology. Removing intralayer connections or enforcing strict rules for network topology allows faster and tractable training algorithms to run. However, doing so takes away some of the representational power of the network [ 8 ], and these restricted or limited networks do not reflect models found in nature. While tractable models perform remarkably well on specialized classification tasks, we speculate that other more complex and generalized tasks may benefit from the additional representational power offered by complex networks. We believe quantum adiabatic computing offers part of a potential solution through its ability to sample from complex probability distributions such as those generated by neural networks containing intralayer connections. The second challenge is automatically discovering optimal or near-optimal network hyperparameters and topologies. Hyperparameters in deep learning refer to the model parameters, i.e., the activation function used, the number of hidden units in a layer, the kernel size of a convolutional layer, and the learning rate of the solver. Currently the best deep learning models are discovered by creating, training, testing, and tuning many models on some well-known reference dataset and reporting the best model in the literature. However, if the dataset has not been examined before, it is difficult to know how to tune networks for optimal performance. GPU-based high performance computing provides an opportunity to automate much of this process—to train, test, and evolve thousands of deep learning networks to find optimally-performing network hyperparameters and network topologies. The last challenge is power consumption, which we can help address through neuromorphic computing. Machine learning’s computational needs have so far been met with power-hungry CPUs and more recently GPUs. The switch from CPUs to GPUs has significantly sped up computation and lowered computation costs, but GPU efficiency in training networks still pales in comparison to the efficiency of biological brains. For an image recognition task, it might take many server farms and a hydroelectric dam to compete with a mundane human brain running on a bit of glucose. Neuromorphic computing offers a potential solution by developing specialized low-power hardware that can implement SNNs approximating trained networks derived from more orthodox architectures. This article focuses on deep learning’s challenges related to quantum adiabatic computing. Though high performance and neuromorphic computing are an integral part of our project, we move discussions of these topics to the appendix to better fit our focus for this journal, though mentions of both appear as necessary through the rest of the article. Our experiments use the MNIST dataset [ 9 ], an image dataset of handwritten digits, and a neutrino particle detection dataset produced by Fermi National Accelerator Laboratory. Next, we will review works related to quantum computing; then we provide our experimental approach, results, and future research.",
"discussion": "6. Discussion We compared a standard benchmark problem, MNIST digit recognition, on three different platforms: quantum adiabatic optimization, HPC, and neuromorphic. Our results show each option offers a unique benefit. Quantum adiabatic computation opens up complex topologies for use in deep learning models that would normally prove intractable for classical machines. HPC allows us to optimize CNNs on a large scale to find an optimal topology with its associated parameters. Neuromorphic lets us implement low power neural network solutions derived from other platforms. Figure 12 provides a summary of these platforms and their associated qualities. However, it is also clear that the MNIST problem is not ideally suited to showcase the capabilities of either the quantum or neuromorphic systems because it has been essentially solved using CNNs. For example, the greater representational power of the quantum LBM approach is likely better utilized on a more complex dataset. Similarly, spiking neuromorphic systems may be better suited for use on datasets that include temporal components. In Figure 13 we propose an architecture we believe provides the ability to leverage the strengths of each of these computing platforms for future, more complex data sets. The goal of this study is to explore how to address some of the current limitations of deep learning, namely networks containing intralayer connections, automatically configuring the hyperparameters of a network, and natively implementing a deep learning model using energy efficient neuron and synapse hardware. We used quantum computing, high performance computing, and neuromorphic computing to address these issues using three different deep learning models (LBM, CNN, and SNN). The quantum adiabatic computing approach allows deep learning network topologies to be much more complex than what is feasible with conventional von Neumann architecture computers. The results show training convergence with a high number of intralayer connections, thus opening the possibility of using much more complex topologies that can be trained on a quantum computer. There is no time-based performance penalty due to the addition of intralayer connections, though there may be a need to sample more often in order to reduce potential errors. HPC allows us to automatically develop an optimal network topology and create a high performing network. Many popular topologies used today are developed through trial and error methods. This approach works well with standard research datasets because the research community can learn and publish the topologies that produce the highest accuracy networks for these data. However, when the dataset is relatively unknown or not well studied, the trial-and-error approach loses its effectiveness. The HPC approach provides a way to optimize the hyper-parameters of a CNN, saving significant amounts of time when working on new datasets, perhaps even bootstrapping under-studied datasets into the regular publish-and-review iterative process. Memristor-based hardware provides an opportunity to natively implement a low-power SNN as part of a neuromorphic computing environment. Such a network has the potential to feature broader connectivity than a CNN and the ability to dynamically reconfigure itself over time. Neuromorphic computers’ benefits, including robustness, low energy usage, and small device footprint, can prove useful in a real-world environment today if we develop a mechanism for finding good network solutions for deployment on memristor-based devices that do not rely on conversions from non-spiking neural network types. We can use the three different architectures together to create powerful deep learning systems to go beyond our current capabilities. For example, current quantum annealing hardware is limited in the size and scope of problems it can solve but does allow us to use more complex networks. We can turn this into an opportunity by using a complex network as a higher level layer in a CNN as seen in Figure 13 . Higher layers typically combine rich features and can benefit from increase intralayer connectivity; they also have smaller-sized inputs than lower layers, easing the limited-scope issue of current quantum annealing hardware. Such an augmented CNN may improve overall accuracy. The HPC approach of automatically finding optimal deep learning topologies is a fairly robust and scalable capability, though quite expensive in development and computer costs. The ability to use deep learning methods on new or under-studied datasets (such as the neutrino particle detection dataset) can provide huge time savings and analytical benefit to the scientific community. The neuromorphic approach is limited by the lack of robust neuromorphic hardware and algorithms, but it holds the potential of analyzing complex data using temporal analysis using very low power hardware. One of the most compelling aspects of this approach is the combination of a SNN and neuromorphic hardware that can analyze the temporal aspects of data. The MNIST problem does not have a temporal component, but one can imagine a dataset that has both image and temporal aspects such as a video or our neutrino detection dataset. A CNN approach has been shown to perform well on the image side, so perhaps a SNN can provide increased accuracy by analyzing the temporal aspects as well. For example, a CNN could analyze an image to detect objects within the image and output the location and/or orientation of those objects. This output can be used as input for an SNN. As each video frame is processed independently by the CNN, the output can be fed into the SNN, which can aggregate information over time and make conclusions about what is occurring in the video or detect particular events that occur over time, all in an online fashion. In this example the CNN could be trained independently using the labeled frames of the video as input images while the SNN could be trained independently utilizing different objects with their locations and orientations as input. These experiments provide valuable insights into deep learning by exploring the combination of three novel approaches to challenging deep learning problems. We believe that these three architectures can be combined to gain greater accuracy, flexibility, and insight into a deep learning approach. Figure 13 shows a possible configuration of the three approaches that addresses the three deep learning challenges we discussed above. The high performance computer is used to create a high performing CNN on image type data. The final layer or two is then processed by the quantum computer using an LBM network that contains greater complexity than a CNN. The temporal aspects of the data are modeled using an SNN, and the ensemble models are then merged and an output produced. Our belief is that this approach has the potential to yield greater accuracy than existing CNN models. Future Work We will test the proposed architecture to determine if it provides improved accuracy, flexibility, and insight into a dataset over methods derived from a traditional CNN approach. We will apply this to neutrino particle detection data and compare the proposed architecture against other contemporary methods. We will also investigate how qubit mapping affects LBM results. Our experiment used a simple 1:1 mapping of hidden units to qubits by placing qubits in chimera cells in the order we defined them. However, this does not take advantage of locality within data; we will examine which methods of qubit mapping produce better results and see how they reveal patterns within our datasets."
} | 6,084 |
29299295 | PMC5743483 | pmc | 9,075 | {
"abstract": "Abstract Invasive plants are often associated with greater productivity and soil nutrient availabilities, but whether invasive plants with dissimilar traits change decomposer communities and decomposition rates in consistent ways is little known. We compared decomposition rates and the fungal and bacterial communities associated with the litter of three problematic invaders in intermountain grasslands; cheatgrass ( Bromus tectorum ), spotted knapweed ( Centaurea stoebe ) and leafy spurge ( Euphorbia esula ), as well as the native bluebunch wheatgrass ( Pseudoroegneria spicata ). Shoot and root litter from each plant was placed in cheatgrass, spotted knapweed, and leafy spurge invasions as well as remnant native communities in a fully reciprocal design for 6 months to see whether decomposer communities were species‐specific, and whether litter decomposed fastest when placed in a community composed of its own species (referred to hereafter as home‐field advantage–HFA). Overall, litter from the two invasive forbs, spotted knapweed and leafy spurge, decomposed faster than the native and invasive grasses, regardless of the plant community of incubation. Thus, we found no evidence of HFA. T‐RFLP profiles indicated that both fungal and bacterial communities differed between roots and shoots and among plant species, and that fungal communities also differed among plant community types. Synthesis . These results show that litter from three common invaders to intermountain grasslands decomposes at different rates and cultures microbial communities that are species‐specific, widespread, and persistent through the dramatic shifts in plant communities associated with invasions.",
"conclusion": "5 CONCLUSIONS By incubating native and invasive plant litter in the field, we determined that the high nutrient availability often observed in plant invasions may be driven in part by rapid decomposition of exotic plant litter. However, the substantial differences in decomposition of roots between cheatgrass and the two invasive forbs also indicate that generalizations do not apply to all invaders. Even though litter can culture a plant‐specific microbial community and fungi can persist when a novel litter is introduced, decomposition rates often did not differ based on whether the litter was placed in home or away soils. Overall, we show that exotic plants that are common to the same ecosystem may use different strategies toward creating successful invasions.",
"introduction": "1 INTRODUCTION Invasion by exotic plants is often associated with higher net primary productivity (NPP) and greater nutrient availability in the soil (Ehrenfeld, 2003 ; Liao 2008). Many of the mechanisms responsible for these changes occur belowground and can include lack of natural pathogens (Reinhart & Callaway, 2006 ), increased abundance and activity of symbiotic microbes (Hawkes, Wren, Herman, & Firestone, 2005 ; Lekberg, Gibbons, Rosendahl, & Ramsey, 2013 ), and higher mineralization rates of nitrogen (Ehrenfeld, Kourtev, & Huang, 2001 ; McLeod et al., 2016 ). Greater nutrient availability in soils largely depends on organic inputs that decomposer communities deliver from litter (Wardle et al., 2004 ). While decomposer communities are often considered to be functionally redundant (Wardle et al., 2004 ), potential differences in litter quality among native and invasive plants (Liao et al., 2008 ) may result in altered decomposition rates and possible shifts in decomposer communities (van der Putten, Klironomos, & Wardle, 2007 ). The extent to which decomposition rates and the composition of decomposer communities depends on specific invaders is unclear. The consequences of plant invasion on ecosystem processes are often generalized from meta‐analyses and review articles that combine all invaders into one homogenous group (Ehrenfeld, 2003 ; Liao et al., 2008 ). While informative, these approaches may be biased by findings from heavily studied ecosystems. Also, even though invaders can share many attributes (e.g. high NPP), they often differ substantially in life histories, and those species‐specific differences may not be captured in meta‐analyses. In the intermountain west, for example, cheatgrass ( Bromus tectorum ), spotted knapweed ( Centaurea stoebe ), and leafy spurge ( Euphorbia esula ) invade grasslands and create persistent invasions (Figure 1 ). Yet cheatgrass is an annual grass that senesces early in the growing season (Mack & Pyke, 1983 ), whereas spotted knapweed and leafy spurge are perennial forbs that are active throughout the growing season (Messersmith, Lym, & Galitz, 1985 ; Sheley, Jacobs, & Carpinelli, 1998 ). Leafy spurge differs from spotted knapweed in that it has deeper roots, spreads through rhizomes, and exudes latex to defend against herbivores (Lym & Kirby, 1987 ; McLeod et al., 2016 ). The three species are highly invasive, produce more biomass, and are associated with higher soil nitrogen availability than native plants (McLeod et al., 2016 ). One possible reason for this is a faster turnover of litter (Fierer, Craine, McLauchlan, & Schimel, 2005 ; Wardle 2006 ), but to what extent litter decomposition rates and decomposer communities differ among these dissimilar invaders is unknown. Figure 1 Photographs of a native plant community and three invasive species common to grasslands in the Intermountain West, USA. Photos courtesy of A. Ramsey and C. Spencer‐Bower Many plants change the belowground microbial community in a way that increases decomposition rates, termed home‐field advantage (HFA) (Austin, Vivanco, González‐Arzac, & Pérez, 2014 ; Ayres et al., 2009 ; Elgersma, Yu, Vor, & Ehrenfeld, 2012 ). Home‐field advantages occur worldwide (Austin et al., 2014 ), but their role in plant invasion is not well known. It might be expected that if invasive plants, as a group, produce higher quality litter than native plants (Liao et al., 2008 ), decomposer communities may shift from oligotroph‐dominated to copiotroph‐dominated, which are organisms that thrive in low‐ and high‐nutrient environments, respectively (Fierer, Bradford, & Jackson, 2007 ). Indeed, one study found that litter from an invasive plant fostered a microbial community capable of faster decomposition of litter from both the invasive host and other plants (Elgersma et al., 2012 ). This could increase nutrient availability for the invader and generate a positive feedback, although more rigorous investigations of HFA in the field for multiple invaders are required to assess general patterns. We compared decomposition rates and microbial communities associated with root and shoot litter of cheatgrass, spotted knapweed, and leafy spurge as well as bluebunch wheatgrass ( Pseudoroegneria spicata ), which is a native grass common to grasslands in the northern Rocky Mountains. We placed shoot litter on the surface and buried root litter from each plant species into replicated plant community types in a factorial design. Three research questions were addressed as follows: (1) Do decomposition rates differ among plant species? (2) Do microbial communities associated with shoot and root litter differ among plant species? (3) Is decomposition faster when litter is placed in a “home” community, that is, are invaders generating a HFA, which may further increase their capacity to invade?",
"discussion": "4 DISCUSSION Previous work has shown that cheatgrass, spotted knapweed, and leafy spurge invasions associate with different bacterial and fungal communities (Gibbons et al., 2017 ; Lekberg et al., 2013 ). We show here that these species‐specific effects extend to litter, because microbial communities colonizing litter from these invaders differed significantly from each other and from a native bunchgrass. However, unlike earlier work showing that invasive plants often culture soil biota that promote their own growth (Callaway, Thelen, Rodriguez, & Holben, 2004 ; Klironomos, 2002 ), we found no evidence for a HFA when it came to decomposition rates; that is, litter did not decompose faster when placed in its home community. This may be because the effect of litter exceeded the effect of where the litter was placed, suggesting that bacterial and fungal decomposers can be widespread and that chemical and/or physical attributes of litter exert a strong habitat filter. Even so, some generalities based on plant functional group identity were apparent, because the two grasses harbored more similar microbial communities and decomposition rates than the forbs. This reiterates recent pleas to better incorporate a trait‐based approach in invasion biology (Bunn, Ramsey, & Lekberg, 2015 ; Meisner et al., 2014 ). 4.1 Decomposition rates differed among invasive plants Litter decomposition is tightly driven by litter quality, which is often characterized by C:N ratios and lignin content (Silver & Miya, 2001 ; Zhang, Hui, Luo, & Zhou, 2008 ). A meta‐analysis of previously published data showed that exotic plants tend to decompose faster than native plants likely due to the high nutrient quality of the exotic plants (Liao et al., 2008 ). We found partial support for this because roots from spotted knapweed and leafy spurge decomposed faster than bluebunch wheatgrass roots (Table 1 ; Figure 2 b). However, the slow decomposition rate of cheatgrass roots shows that generalizations about invasive plants do not always apply. Further, the differences in decomposition rates of roots depended not on whether they were native or exotic, but whether they were a forb or a grass. Roots from the two invasive forbs, spotted knapweed and leafy spurge, tended to have higher phosphorus content than roots from cheatgrass and bluebunch wheatgrass (Table S2 ), which may explain the disparity in decomposition rates because higher quality root litter tends to decompose faster (Silver & Miya, 2001 ; Zhang et al., 2008 ). The limited number of invaders included in the study clearly limits broad generalizations, but different decomposition rates have been shown depending on plant traits (Cornwell et al., 2008 ) and reinforce recent suggestions that plant life form should be included when analyzing the impacts of specific exotic plants (Meisner et al., 2014 ). Shoots from leafy spurge decomposed faster than shoots from all other plants at 3 months, but at 6 months, there were no significant differences in decomposition (Table 1 ; Figure 2 a). Like the decomposition of roots, the decomposition of shoots largely depends on the nitrogen and phosphorus content (Cornwell et al., 2008 ; Parton et al., 2007 ). Leafy spurge had more nitrogen content in its shoots than all other species, suggesting that at 3 months, the nutrient quality of litter may have influenced the different decomposition rates among species (Table S2 ). Two important abiotic conditions that influence decomposition are soil moisture and temperature. Differences in these conditions were kept minimal between plant communities because we chose sites that were close in proximity to each other and shared similar aspect, slope, and elevation. However, plants can shade soils, which changes soil temperature and moisture, both of which influence decomposition rates (Köchy & Wilson, 1997 ). Cheatgrass communities had warmer (Fig. S1 ), and potentially drier soils, whereas leafy spurge communities had cool, and possibly wetter soils, yet these differences in abiotic conditions among plant communities did not lead to different decomposition rates. Although ultraviolet radiation contributes to the decomposition of shoot litter in grasslands as well (Austin & Vivanco, 2006 ; Parton et al., 2007 ), it likely did not drive differences observed here, because the plant community where litter was placed did not influence decomposition rates of shoots. 4.2 Bacterial and fungal communities differed among plant species The nutritional makeup of litter can drive the structure of decomposer communities (Bray, Kitajima, & Mack, 2012 ; Cline & Zak, 2015 ; Purahong et al., 2016 ; Voříšková & Baldrian, 2013 ). We found that the forbs generally had greater nitrogen and phosphorus content than grasses (Table S2 ) while also harboring different decomposer communities (Table S3 ). As litter decomposes, r strategists (copiotrophs) use labile matter, and when only recalcitrant matter remains, k strategists (oligotrophs) become dominant (Dilly, Bloem, Vos, & Munch, 2004 ). Gibbons et al. ( 2017 ) found that spotted knapweed and leafy spurge shifted the bacterial communities in soils toward copiotrophs. This suggests that certain phyla of bacteria can become enriched in response to the litter used in this study. One interesting finding was that decomposer communities depended more on the species of plant litter than the plant community of incubation (Table 2 ). Two different processes could explain this. First, all plant communities may harbor a diversity of fungal and bacterial species where a subset of the community colonizes litter depending on nutrient quality. This suggests that the characteristics of litter provide a strong habitat filter for these decomposers. An alternative explanation is that these bacteria and fungi occurred as endophytes that changed to saprophytic strategies upon senescence or harvest of plant tissues (Kembel & Mueller, 2014 ; Omacini, Chaneton, Ghersa, & Otero, 2004 ; Voříšková & Baldrian, 2013 ). While this could partly explain the strong effect of plant species, the dependence of fungal communities on the plant community of incubation (Table 2 ) suggests that the litter was at least partially colonized at the site of incubation. Bacterial endophytes also could have been present, however, the amount of bacterial DNA on litter at the onset of decomposition can be small (Dilly et al., 2004 ), suggesting that endophytic bacteria may not strongly influence initial decomposition. Fungal communities did not restructure based on the nutritional quality of litter alone because they differed based on the plant community of incubation, showing that these persistent plant invasions created legacies of fungal communities that resisted change upon the introduction of new plant species. This supports previous work where microbial communities persisted even after the host litter had been removed and new litter was introduced (Elgersma, Ehrenfeld, Yu, & Vor, 2011 ; Elgersma et al., 2012 ). However, bacterial communities on litter did not depend on the plant community of incubation (Table 2 ), suggesting that they may rapidly restructure to the litter they encountered in an absence of a strong legacy effect. The relative importance of legacy effects and nutrient quality on decomposition rates and decomposers of invasive plants is unresolved but likely depends on plant species and duration of the invasion (Elgersma et al., 2011 ). 4.3 Home‐field advantage The term HFA implies a positive effect where litter decomposes faster in its local environment, but effects are not always positive, suggesting HFA is context‐dependent (Freschet et al., 2012 ; Veen et al., 2015 ). Freschet et al. ( 2012 ) proposed that HFA is just a facet of a more comprehensive hypothesis called the substrate quality–matrix quality interaction (SMI) where HFA effects are greatest when litter quality is strongly dissimilar from the litter matrix associated with a site. For example, high‐quality litter placed in a matrix of high‐quality litter would decompose rapidly, but in contrast, it would decompose slower when placed in a matrix of lower quality litter (Freschet et al., 2012 ). The influence of HFA or the SMI in the decomposition of exotic plants is not well studied, although one study found that an invasive shrub changed the microbial communities in a way that increased decomposition rates (Elgersma et al., 2012 ). Here, we expected the strong dissimilarities in the nutrient quality of litter and long established plant communities (≥10 years) to change the decomposer community in a way that created HFAs based on the SMI hypothesis (Elgersma et al., 2012 ; Freschet et al., 2012 ; Strickland et al., 2009 ). For instance, decomposers within cheatgrass and native plant communities may specialize on recalcitrant matter, whereas decomposers in spotted knapweed and leafy spurge communities may specialize on labile matter. However, we did not observe a HFA even though each plant species reshaped the microbial communities and forbs tended to decompose faster than grasses. The lack of an HFA effect may be explained by an insufficient disparity between the quality of litter (e.g. N:P) commonly found in each plant community where litter was incubated. In a meta‐analysis by Veen et al. ( 2015 ), they observed stronger HFA effects for forest–grassland transplants than for grassland–grassland transplants, which were likely driven by the SMI hypothesis. That finding, taken together with this study, further suggests that HFA effects in grasslands may be minimal. It is important to note that HFA effects can be subtle and often influence decomposition rates by <10% (Veen et al., 2015 ), so our small sample size may have precluded the detection of a minor change in decomposition rates."
} | 4,304 |
19939938 | PMC2817479 | pmc | 9,076 | {
"abstract": "Geobacter species play important roles in bioremediation of contaminated environments and in electricity production from waste organic matter in microbial fuel cells. To better understand physiology of Geobacter species, expression and function of citrate synthase, a key enzyme in the TCA cycle that is important for organic acid oxidation in Geobacter species, was investigated. Geobacter sulfurreducens did not require citrate synthase for growth with hydrogen as the electron donor and fumarate as the electron acceptor. Expression of the citrate synthase gene, gltA, was repressed by a transcription factor under this growth condition. Functional and comparative genomics approaches, coupled with genetic and biochemical assays, identified a novel transcription factor termed HgtR that acts as a repressor for gltA . Further analysis revealed that HgtR is a global regulator for genes involved in biosynthesis and energy generation in Geobacter species. The hgtR gene was essential for growth with hydrogen, during which hgtR expression was induced. These findings provide important new insights into the mechanisms by which Geobacter species regulate their central metabolism under different environmental conditions.",
"introduction": "INTRODUCTION Geobacter species can play an important role in the bioremediation of groundwater contaminated with organics or metals ( 1–7 ) and are one of the most effective microorganisms in converting organic compounds to electricity in microbial fuel cells ( 8–11 ). Studies on the physiology of Geobacter species have primarily focused on Geobacter sulfurreducens because it has the key hall mark physiological characteristics of Geobacter species ( 12 ), including the ability to completely oxidize organic acids to carbon dioxide with electron transfer to extracellular electron acceptors such as Fe(III) oxides ( 13–15 ), toxic metals ( 16 ), humic substances ( 17 ) and electrodes ( 18 , 19 ). In addition to organic compounds, Geobacter species can utilize hydrogen as an electron donor to generate energy for growth ( 12 , 20 , 21 ). The tricarboxylic acid (TCA) cycle is the main pathway for oxidation of organic compounds for energy conservation in G. sulfurreducens and serves to synthesize a diversity of precursor metabolites for biosynthetic reactions ( 22 , 23 ). Citrate synthase is a key TCA cycle enzyme. Analysis of the G. sulfurreducens genome revealed only one homologue of the citrate synthase gene, termed gltA ( 24 ), which encodes the protein responsible for citrate synthase activity ( 25 ). Surprisingly, the citrate synthases of G. sulfurreducens as well as other members of Geobacter species show higher sequence similarity to eukaryotic citrate synthases than to the majority of prokaryotic citrate synthases ( 24–26 ). The production of citrate synthase in Geobacter species appears to be highly regulated. For example, cells grown with hydrogen as the electron donor had much lower citrate synthase activities than cells grown on acetate ( 25 ). Transcript abundance of gltA directly correlated with the rates of Fe(III) reduction in chemostats or the rates of electron transfer to electrodes in microbial fuel cells ( 26 ). Here we report on one of the mechanisms by which the expression of gltA and other genes encoding proteins important for central metabolism is regulated in Geobacter species. The results suggest that a novel transcriptional repressor plays an important role in controlling the expression of these genes.",
"discussion": "DISCUSSION These results suggest that HgtR is a global transcriptional regulator that represses a diversity of genes involved in biosynthesis and energy generation in Geobacter species during hydrogen-dependent growth. Proper differential expression of this suite of genes appears to be important for growth with acetate or hydrogen as the electron donor. In addition to gltA , many of the other genes that were predicted to have an HgtR-binding site, such as malate dehydrogenase, fumarate reductase, succinate dehydrogenase and sfrAB, are required for growth with acetate as the electron donor, but not for growth on hydrogen ( 42 , 43 , 47 ). In contrast, mutants deficient in these genes can grow with hydrogen as the electron donor. Previous studies have demonstrated that levels of expression of genes such as citrate synthase and malate dehydrogenase that are under the transcriptional control by HgtR can be related to the rates of electron transfer to Fe(III) and electrodes in cultures of G. sulfurreducens as well as the in situ metabolic rates of subsurface Geobacter species during the bioremediation of uranium-contaminated groundwater ( 26 , 48 ). Therefore, HgtR is likely to play an important role in optimizing the growth of Geobacter species in a range of environments. Regulation of the expression of citrate synthase via transcriptional repression has been noted previously in other microorganisms. For example, the ArcA two-component response regulator controls the expression of gltA in E. coli in response to anaerobiosis and carbon supply ( 49 , 50 ). The expression of citZ encoding the major citrate synthase in Bacillus subtilis is under the control of catabolite repression by CcpA and CcpC, a member of the LacI/GalR family and the LysR family of transcriptional regulators, respectively ( 51 , 52 ). As with HgtR, ArcA ( 53 ), CcpA and CcpC ( 54 , 55 ) are global transcriptional regulators in these bacteria and also control other genes involved in biosynthesis and energy generation. The amino acid sequences of the HgtR homologues showed no homology to known proteins. The only predicted domain from the structure prediction of the HgtR homologues is a DNA-binding domain, which was predicted to form a structure similar to those of the ribbon–helix–helix (RHH) transcription factor superfamily (Aklujkar, personal communication). The RHH transcription factors share a conserved three dimensional structure, but their amino acid sequences are diversified ( 56 ). The molecular mechanism of hydrogen-dependent gene regulation was previously characterized in the aerobic hydrogen oxidizer, Ralstonia eutropha , in which a regulatory [NiFe]-hydrogenase acts as a hydrogen receptor and regulates the two-component system consisting of HoxJ and HoxA that controls the gene expression ( 57 ). The increase in hgtR transcripts during growth with hydrogen suggests that the expression of hgtR is regulated at the level of transcription and the transcription of hgtR appears to be mediated by the sigma factor RpoN ( Figure 6 ). Unlike most of other bacterial rpoN genes, the G. sulfurreducens rpoN is essential under the all conditions tested ( 58 ). In G. sulfurreducens , RpoN controls the expression of genes involved in a wide range of cellular processes such as fumarate respiration, Fe(III) reduction, nitrogen fixation, and pili and flagella biosyntheses ( 58 ). The genome sequence analysis identified the gene located at the upstream of hgtR , which encodes a putative enhancer-binding protein (EBP) that has a domain similar to the iron only hydrogenase large subunit (data not shown). HgtR does not have an apparent regulatory (sensor) domain. Therefore, it is possible that the EBP senses hydrogen and activates the transcription of hgtR . This type of EBP has not been identified in other bacteria. A gene encoding a homologue of the transcription factor GntR appears to be a target of HgtR ( Figure 5 ), and thus the expression of other genes is likely to be affected during growth with hydrogen. Taken together, it is likely that metabolic genes are controlled by novel transcriptional regulatory cascades in Geobacter species. It is likely that transcriptional regulation by HgtR represents only one of several levels of regulation of citrate synthase. For example, the promoter regions of gltA, ato, frdCAB and icd-mdh contain a long 5′ untranslated region, which was predicted to form a secondary structure that appears to play a role in translation of their mRNA (data not shown), providing the possibility for further regulation at the post-transcriptional level. Furthermore, western blot analysis showed that the migration of the G. sulfurreducens citrate synthase in SDS–PAGE gel was affected by heat-denaturation, suggesting that heat-sensitive modulation occurred at the post-translational level (Yun, personal communication). Phosphorylation is a means of regulating citrate synthase in Tetrahymena ( 59 ) and isocitrate dehydrogenase, another enzyme in the TCA cycle, in some bacteria ( 60–62 ). In summary, our findings provide important new insights into the mechanisms by which Geobacter species regulate, and presumably optimize, their central metabolism under different environmental conditions. Further research into the environmental cue(s) controlling the expression of HgtR and additional mechanisms for regulation of central metabolism is warranted and is underway."
} | 2,250 |
30894690 | null | s2 | 9,077 | {
"abstract": "Recent work has shown that subsurface microbial communities assemble by selective survival of surface community members during sediment burial, but it remains unclear to what extent the compositions of the subsurface communities are a product of their founding population at the sediment surface or of the changing geochemical conditions during burial. Here we investigate this question for communities of sulfate-reducing microorganisms (SRMs). We collected marine sediment samples from the upper 3-5 m at four geochemically contrasting sites in the Skagerrak and Baltic Sea and measured SRM abundance (quantitative PCR of dsrB), metabolic activity (radiotracer rate measurements), and community composition (Illumina sequencing of dsrB amplicons). These data showed that SRM abundance, richness, and phylogenetic clustering as determined by the nearest taxon index peaked below the bioturbation zone and above the depth of sulfate depletion. Minimum cell-specific rates of sulfate reduction did not vary substantially between sites. SRM communities at different sites were best distinguished based on their composition of amplicon sequence variants (ASVs), while communities in different geochemical zones were best distinguished based on their composition of SRM families. This demonstrates environmental filtering of SRM communities in sediment while a site-specific fingerprint of the founding community is retained."
} | 355 |
29025976 | null | s2 | 9,079 | {
"abstract": "In the interest of decreasing dependence on fossil fuels, microbial hydrocarbon biosynthesis pathways are being studied for renewable, tailored production of specialty chemicals and biofuels. One candidate is long-chain olefin biosynthesis, a widespread bacterial pathway that produces waxy hydrocarbons. Found in three- and four-gene clusters, "
} | 86 |
39807368 | PMC11728893 | pmc | 9,081 | {
"abstract": "Highlights • Nitrification-partial DNRA coupled anammox, PNA, and PdNA were compared firstly. • Anammox bacteria-mediated partial DNRA is the key in this nitrogen removal system. • Partial DNRA-anammox has thermodynamic advantages in oxidizing NH 4 + than that of anammox alone. • Partial DNRA-anammox appears to be feasible for energy-positive wastewater treatment."
} | 91 |
36254846 | PMC9788401 | pmc | 9,082 | {
"abstract": "Biology demonstrates meticulous ways to control biomaterials self-assemble into ordered and disordered structures to carry out necessary bioprocesses. Empowering the synthetic polymers to self-assemble like biomaterials is a hallmark of polymer physics studies. Unlike protein engineering, polymer science demystifies self-assembly by purposely embedding particular functional groups into the backbone of the polymer while isolating others. The polymer field has now entered an era of advancing materials design by mimicking nature to a very large extend. For example, we can make sequence-specific polymers to study highly ordered mesostructures similar to studying proteins, and use charged polymers to study liquid–liquid phase separation as in membraneless organelles. This mini-review summarizes recent advances in studying self-assembly using bio-inspired strategies on single-component and multi-component systems. Sequence-defined techniques are used to make on-demand hybrid materials to isolate the effects of chirality and chemistry in synthetic block copolymer self-assembly. In the meantime, sequence patterning leads to more hierarchical assemblies comprised of only hydrophobic and hydrophilic comonomers. The second half of the review discusses complex coacervates formed as a result of the associative charge interactions of oppositely charged polyelectrolytes. The tunable phase behavior and viscoelasticity are unique in studying liquid macrophase separation because the slow polymer relaxation comes primarily from charge interactions. Studies of bio-inspired polymer self-assembly significantly impact how we optimize user-defined materials on a molecular level.",
"conclusion": "Conclusion Nature demonstrated excellent skillsets in making spontaneous self-assembly happen in water. This review paper discusses how sequence and charge, two of the most impactful drivers in biomaterial self-assembly can be used to build fundamental understandings of synthetic polymer self-assembly. In particular, bio-inspired sequence-specific polymers provide an extremely well-controlled method to make user-defined molecules, offering a handle to directly and precisely correlate polymer chemistry and performance on a molecular level. Charge-driven self-assembly in multi-polymer network opens opportunities to modulate polymer interactions in water, allowing for a fully water-based material platform in applied sciences. Building on the well-established science of bio-inspired polymers, engineering applied materials will be enlightened in fields, such as energy conversion, nanolithography, and biomedicine.",
"introduction": "Introduction Biomaterials have a lot to teach us how to control polymer self-assembly. Biological macromolecules are encoded with information to form intrinsically ordered and disordered regions to carry out a plethora of functionalities necessary for bioprocesses [ 1 , 2 ]. Encoding synthetic materials with information opens opportunities to guide polymer self-assembly for hierarchical organization and compartmentalization that only biopolymers were capable of before [ 3 ]. Multiple intra- and inter-molecular interactions (e.g. hydrophobicity, electrostatics, sterics, and hydrogen bonding) contribute simultaneously to the ultimate shape of the polymer. Isolating these intertwined factors is meaningful but challenging for the polymer community when developing fundamental understandings of controlling polymer self-assembly. In return, wisdom gained from artificial polymer systems could potentially be applied to explain malfunctions of biomaterials due to loss of desired structures. Where biology and polymers meet inspires tremendous innovations for materials to have nature's ability to form structured complexes with advanced performance. Seeking molecular control over bulk polymer assembly has been a longstanding quest in polymer and biology studies. How much power do we need in our synthetic polymer systems to make them work like biological macromolecules? What control handles have been given to us to tune artificial soft materials to make more meaningful polymers [ 4 ]? The increasing number of publications on ‘bio-inspired polymers self-assembly’ (111k as of August 2022, Web of Science!) indeed tracks the vigorous activities of embedding nature's capability into synthetic polymers. In this mini-review article, challenges and recent advances in controlling self-assembly will be elaborated. Self-assembly will be discussed in (1) single-component systems to study the phase junction in block copolymers and to investigate hierarchical mesostructures formation [ 5 , 6 ], and (2) multi-component systems in the context of complex coacervation, which is a charge-driven liquid–liquid phase separation phenomenon [ 7 ] ( Figure 1 ). Figure 1. A schematic overview of the topics covered in this review, including (a) single-component and (b) the multi-component polymer systems. In the single-component systems, hybrid polymers will be discussed to demonstrate the steric effects between domains in self-assembled block copolymers. The role of sequences will be elucidated in controlling hierarchical structures formation. In the multi-component system, complex coacervation as a classic macro-ionic phase separation phenomenon will be discussed in terms of phase behavior, viscoelasticity, and hierarchical structures formation."
} | 1,339 |
29272410 | PMC5765554 | pmc | 9,083 | {
"abstract": "Abstract Prokaryote genomes are the result of a dynamic flux of genes, with increases achieved via horizontal gene transfer and reductions occurring through gene loss. The ecological and selective forces that drive this genomic flexibility vary across species. Bacillus subtilis is a naturally competent bacterium that occupies various environments, including plant-associated, soil, and marine niches, and the gut of both invertebrates and vertebrates. Here, we quantify the genomic diversity of B. subtilis and infer the genome dynamics that explain the high genetic and phenotypic diversity observed. Phylogenomic and comparative genomic analyses of 42 B. subtilis genomes uncover a remarkable genome diversity that translates into a core genome of 1,659 genes and an asymptotic pangenome growth rate of 57 new genes per new genome added. This diversity is due to a large proportion of low-frequency genes that are acquired from closely related species. We find no gene-loss bias among wild isolates, which explains why the cloud genome, 43% of the species pangenome, represents only a small proportion of each genome. We show that B. subtilis can acquire xenologous copies of core genes that propagate laterally among strains within a niche. While not excluding the contributions of other mechanisms, our results strongly suggest a process of gene acquisition that is largely driven by competence, where the long-term maintenance of acquired genes depends on local and global fitness effects. This competence-driven genomic diversity provides B. subtilis with its generalist character, enabling it to occupy a wide range of ecological niches and cycle through them.",
"introduction": "Introduction Prokaryotes genomes are highly dynamic, with rates of gene gain and gene loss comparable to mutation rates ( Kolstø 1997 ; Doolittle 1999 ; Koonin and Wolf 2008 ). The major promoter of gene gain is lateral gene transfer (LGT), which dominates the bacterial world at varying levels depending on the species biology and the ecological interactions established in local environments ( Kolstø 1997 ; Doolittle 1999 ; Ochman et al. 2000 ; Koonin and Wolf 2008 ; Puigbò et al. 2014 ). Contradicting initial concerns ( Bapteste et al. 2005 ; Doolittle and Bapteste 2007 ), the dominance of LGT has not precluded the inference of vertical phylogenies that describe the diversification process and the relationships among species ( Daubin et al. 2003 ; Lerat et al. 2005 ; Puigbò et al. 2009 ; Hug et al. 2016 ). It is on the top of these phylogenies that we reconstruct the evolutionary processes that shape the microbial world and where estimates of the rates of gene gain, gene loss, and gene family expansion and regression are made. These efforts have produced the body of knowledge currently available on the diversity and dynamics of bacterial genomes in their natural habitats. Studies carried out with widely distinct bacterial groups, encompassing diverse phylogenetic depths and data set completeness, led to the view that persistent gene loss intersected by episodic events of massive gene gains through LGT dominate prokaryote genome dynamics (reviewed by Wolf and Koonin 2013 ). Overall, these processes are expected to lead to genome streamlining, that is, to the reduction of genome size due to relaxed selection and gene loss. Genome streamlining is well documented among obligate intracellular parasites, which typically have reduced genomes ( McCutcheon and Moran 2011 ; Merhej and Raoult 2011 ), and among free-living bacteria ( Luo et al. 2013 ; Swan et al. 2013 ). It is also believed that if genomes are unable to laterally acquire genes, they will face extinction, similar to asexual species that are thought to be doomed to extinction due to a lack of recombination ( Moran 1996 ; Baltrus et al. 2008 ; Naito and Pawlowska 2016 ). The theoretical predictions that underlie many expectations of how prokaryote genomes evolve are based on population genetic processes. Yet, much of the data that have been analyzed covers billions of years of evolution, frequently encompassing evolutionary trends across major taxonomic groups, such as different phyla ( Snel et al. 2002 ; Kunin 2003 ; Charlebois and Doolittle 2004 ; Kettler et al. 2007 ; Richards et al. 2014 ). Eased by the increased facility of sequencing new bacterial genomes, recent efforts have been made to deeply sample many genomes within individual species with the goal of uncovering intraspecific evolutionary processes. These developments have been fundamental to our understanding of intraspecific pangenome dynamics as it has become clear that bacterial genomes are dynamic containers of essential and accessory elements, where multiple isolates are required to understand the global complexity of a single species ( Tettelin et al. 2005 ; Touchon et al. 2009 ; Lefébure et al. 2010 ; Ahmed et al. 2012 ; Kaas et al. 2012 ; McNally et al. 2013 ; Bolotin & Hershberg 2015 ). A recent mathematical model of microbial evolution that uses intraspecific data on gene acquisition and protein-level selection proposes that the number of genes in a genome reflects an equilibrium between the benefit accrued by acquiring new genes and the cost of maintaining a larger genome ( Sela et al. 2016 ). Puigbò et al. (2014 ) reconstructed the genome dynamics across many groups of closely related organisms and corroborated the rapid and variable flux of genes due to extensive gene loss and LGT. These authors, however, still group under the same name genomes belonging to different species that albeit close relatives, are not sister taxa, as for instance B. subtilis and B. amyloliquefaciens ( Priest and Goodfellow 1987 ; Rooney et al. 2009 ; Connor et al. 2010 ). Hence, a detailed analysis of such an important organism as B. subtilis is still lacking. \n B. subtilis is an endospore-forming gram-positive bacterium that belongs to the deeply rooted phylum Firmicutes. It is a model organism for many molecular processes as well as an industrial workhorse ( Harwood 1992 ; Schallmey et al. 2004 ). It has received considerable interest as a promoter of plant growth and as a plant–disease control organism ( Schisler et al. 2004 ; Deng et al. 2011 ; Falardeau et al. 2013 ). Its status as a GRAS (generally regarded as safe) organism and its ability to form endospores (hereafter, spores) have prompted several applications in biomedicine and biotechnology; these include the use of spores in probiotic formulations and as efficient platforms for surface display and vaccine delivery ( Hong et al. 2005 ; Tavares Batista et al. 2014 ; Wu et al. 2015 ). Commonly described as a ‘soil bacterium’, B. subtilis can be sampled from a diverse set of environments that includes the invertebrate and vertebrate gut as well as several plant-associated, soil, and marine niches ( Tam et al. 2006 ; Fan et al. 2011 ; Um et al. 2013 ). This diversity of niches and the associated diversity of social interactions raise the question of how these are achieved within a single species. In this study, we analyze the genomic dynamics within B. subtilis and estimate the size of its pangenome by following a strictly intraspecific approach. We aim at identifying the processes that can explain its pangenome size. The first description of this species’ genomic diversity, which was obtained by performing a microarray-based comparative genomic hybridization analysis with the genome of strain 168 as reference, suggested substantial gene content diversity among B. subtilis isolates ( Earl et al. 2007 ). Earl et al. (2007) queried for the presence of each coding sequence in the genome of strain 168 in a collection of closely related strains and identified presence/absence polymorphisms among genes involved in antibiotic production, cell wall synthesis, sporulation, and germination. An unexpectedly high-genomic diversity in gene content was found, but this diversity was likely underestimated as genes present in other strains but absent from the reference genome would have been missed. This finding led to the view that B. subtilis is a highly versatile organism, which is consistent with its presence in diverse natural settings ( Earl et al. 2007 ). The processes leading to this diversity have, however, not been addressed. A key feature of B. subtilis is its ability to initiate a developmental program that leads to the production of competent cells that are able to efficiently internalize and recombine exogenous DNA with no apparent sequence specificity ( Haijema et al. 2001 ; Maamar and Dubnau 2005 ; Smits et al. 2005 ). The lack of self-specificity provides opportunities for genomic diversification through the acquisition of novel genes. However, the consequences of natural competence on the genomic composition and diversity of this species remain unclear. Competence in B. subtilis is induced transiently as cells enter the stationary phase of growth. This is a stochastic process driven by noise in the expression of an auto-regulatory transcription factor, ComK, and results in competence development in only a small fraction, typically ∼1%, of the cells in populations of natural isolates ( Maamar and Dubnau 2005 ; Smits et al. 2005 ; Claverys et al. 2006 ; Yüksel et al. 2016 ). Competence may provide templates for DNA repair, defense against genomic parasites, a source of nucleotides, or a source of genetic variation ( Finkel and Kolter 2001 ; Claverys et al. 2006 ; Engelmoer and Rozen 2011 ; Johnston et al. 2014 ; Ambur et al. 2016 ; Croucher et al. 2016 ). Competence can also endow B. subtilis with tolerance to certain antibiotics, as cells can enter a non-dividing state that is similar to persistence ( Hahn et al. 2015 ; Yüksel et al. 2016 ). The bi-stable switch governing competence is interpreted as a bet-hedging strategy to improve fitness under adverse or variable environments ( Johnsen et al. 2009 ). Competence raises important questions concerning its contribution to the diversity and evolution of the species’ genome and the influence of local ecological conditions on genome diversity and cohesion. It is expected that the capacity to enter a natural competence state can be beneficial as it allows the input of new and locally adapted genomic diversity into a species’ gene repertoire ( Wylie et al. 2010 ). This capacity is of particular relevance to species such as B. subtilis that are found in a wide range of different niches and have the ability to cycle among them. In this study, we found that B. subtilis has a large and open pangenome that results from the continuous acquisition of new genes through LGT balanced by an equivalent proportion of gene loss. We found a preponderant role for gene acquisition through competence, although other modes of acquisition, such as transduction, also carry new genes into the species pangenome. We posit that by using natural competence for genetic transformation in a wide diversity of niches, this species can potentially generate countless individual niche adaptation paths.",
"discussion": "Discussion We show that B. subtilis has a large, open and dynamic pangenome that largely results from the continuous acquisition of genes from closely related species along with extensive gene loss. This pattern of genome dynamics suggests a key role for natural competence coupled with a wide niche breadth in the creation and maintenance of the species’ pangenome diversity. We corroborate previous results on the evolution of bacterial genomes, showing that this evolution is a highly dynamic process. Importantly, we show that this dynamicity does not lead to major differences among strains. On average, 77.1% of all genes in each genome are shared among all strains, and 43% of the species pangenome that exists in only one or two strains represents no more than 3% of the genome of natural strains and 0.5% of that of laboratory strains. Consistent with the results of previous studies (reviewed by Wolf and Koonin 2013 ), we infer that the main determinants in the evolution of bacterial genomes are gene gain through LGT and gene loss. The net result of these processes shape and maintain a species-specific genome cohesion by streamlining the genomes along with the acquisition of a new and diverse genomic repertoire. At the terminals of the core genome tree, we do not recover the gene loss bias that has been widely documented in the literature (Kunin 2003; Makarova et al. 2006 ; Csurös and Miklós 2009 ; Wolf et al. 2012 ; Puigbò et al. 2014 ). Instead, we found high and ubiquitous gene gain in the evolution of wild strains, suggesting this to be a fairly frequent rather than episodic mechanism of genome diversity acquisition. Similar dynamics were observed in the genome evolution of Escherichia coli but were attributed to phage-related genes or transposable elements ( Touchon et al. 2009 ). Bolotin and Hershberg (2015 , 2016 ) suggest that a gene-loss bias in intraspecific genome dynamics is mainly a feature of highly clonal species. However, in species with higher rates of recombination, such as B. subtilis , the two rates are expected to balance out. More generally, Puigbò et al. (2014 ) suggest that gene loss bias may not manifest at short evolutionary scales. Losses of strain-specific genes cannot be detected, imposing an underestimation of the gene loss at these branches. It is also possible that stochastically acquired genes could temporarily accumulate in the genomes if neutral or slightly deleterious, similar to the inflated number of non-synonymous to synonymous substitutions estimated in data sets of intraspecific genomes ( Rocha et al. 2006 ). The high rates of gene gain and gene loss that we recover for the tips of the tree coupled with the ability of B. subtilis to uptake exogenous DNA without self-specificity of sequence identification are highly suggestive of an evolutionary model in which cloud genes are the result of stochastic events of gene acquisitions. The long-term maintenance of these genes in the genome repertoire of the species should thus depend on their fitness effects given the genomic context (epistatic interactions) and the specificities of the biotic and abiotic environment ( Graham and Istock 1979 ; Berg and Kurland 2002 ; Johnsen et al. 2009 ). Several studies have addressed the fate of acquired genes, and the results suggest that gene loss of recently acquired genes is pervasive in many bacterial groups ( van Passel et al. 2008 ; Lo et al. 2015 ; Kuo and Ochman 2009 ). Simulations under a birth-and-death model of prokaryote genome evolution suggest that neutral or nearly neutral gene acquisitions in microbial populations are expected to generate a large diversity of transient gene content, where only sequences that are under strong selection, globally or in individual patches, are expected to persist ( Berg and Kurland 2002 ). An example of strong local episodic selection promoting genome diversity was previously proposed for B. subtilis . In this species, competence develops in non-dividing cells in an otherwise growing population, imposing a short-term fitness cost on the competent cells. Johnsen et al. (2009) show that this impairment can be overcome if episodic stresses, such as antimicrobials, preferentially affect the dividing cells, an advantage that is highly augmented if selection favors the competent cells that have acquired new DNA ( Johnsen et al. 2009 ). It is tempting to speculate that through competence, B. subtilis stochastically surveys the environment for new genes, potentiating a dynamic process of niche adaptation in which each organism can have its own evolutionary trajectory, as proposed by Gogarten et al. (2002) . A population could simultaneously express diverse phenotypes in an extension of bet-hedging strategies, as documented for genetically identical cells ( Ackermann et al. 2008 ; Leisner et al. 2008 ; Veening et al. 2008 ; Beaumont et al. 2009 ). Bet-hedging is important for survival during the rapid expansion of sub-populations in a rapidly changing environment or when an organism frequently transits between niches, as it is likely the case in B. subtilis ( Kussell and Leibler 2005 ; Tam et al. 2006 ; Wolf et al. 2005 ; Wylie et al. 2010 ). The maintenance of different genomes in a local population might also foster strategies for the division of labor and the evolution of cooperative behaviors ( Morris et al. 2012 ; Mas et al. 2016 ), both of which are well documented in B. subtilis ( Lopez et al. 2009 ; Shank et al. 2011 ; van Gestel et al. 2015 ). Nevertheless, transduction by the integration of phage DNA into the bacterial chromosome certainly plays a role in the acquisition of new genes, but it is not the prevailing mode of LGT in B. subtilis . Only a few strains show evidence of having prophages enriched in cloud genes, and in those strains, a large proportion of the recently acquired cloud genome is distributed throughout the entire genome and does not map to prophages ( fig. 2 C , supplementary fig. S5 and table S5, Supplementary Material online). We have not analyzed the diversity present in plasmids. We note, however, that although genes in replicative plasmids would likely further increase the pangenome size of B. subtilis , the integration of plasmids or other integrative and conjugative elements into the chromosome is expected to generate a clumped distribution of laterally transferred genes ( Wozniak and Waldor 2010 ), which is clearly not the dominant pattern observed here ( fig. 2 C , supplementary fig. S5 , Supplementary Material online). Natural competence is not expected to play a similar role in every organism. For instance, both Streptococcus pneumoniae and Haemophilus influenzae are naturally competent species, but both have pangenome sizes that are much smaller than the one estimated here for B. subtilis ( Hiller et al. 2007 ; Hogg et al. 2007 ). These species have a narrower niche breadth that is typically restricted to the human respiratory tract, in which they can become highly pathogenic. Streptococcus pneumonia , unlike B. subtilis , lacks a specific DNA damage repair mechanism ( Charpentier et al. 2012 ), and it might rely on competence to overcome DNA damage caused by the host immune system ( Claverys et al. 2006 ). In contrast, in H. influenza , the diversity of the genes acquired through transformation is clearly constrained by the need for a sequence-specific identifier that limits DNA binding and uptake to intraspecific DNA ( Mell and Redfield 2014 ). The diversity of the B. subtilis pangenome reflects both the vertical inheritance of genes (the genealogical process) and convergent evolution related to the occupancy of similar environments. Overall, there is not a strong pangenome signature for niche occupancy, as we did not recover niche-specific genes. We note, however, that strains from cluster II ( fig. 5A ) can be traced to an environment in which host–microbe interactions are likely. This cluster contains all Plant- and Gut-associated strains as well as two strains sampled from fermented food, one strain sampled from a chicken feather, and two strains isolated from the abdomen and nest of fungus-growing termites. Microbes that cycle through environments could benefit from common (in addition to specific) adaptation strategies in such cross-kingdom host colonizations ( Wiedemann and Virlogeux-Payant 2015 ). In addition, fecal contamination of the soil by animals can expose plants to gut bacteria, creating a cycle of transmission that closes when animals are fed with plants and fermented (probiotic) foods, exposing the animal gut to plant-adapted and fermented bacteria ( Tam et al. 2006 ; Barak and Schroeder 2012 ; Melotto et al. 2014 ; Serra et al. 2014 ). The lack of a strong signature for niche adaptation could reflect a sampling problem if the strains were isolated as spores, which can be easily dispersed, and not as growing cells. It might also be explained by the influences of the several environments through which the strains cycled or by the ability of bacteria to adopt a multitude of different strategies to adapt to the same environment. Two genomes can be similar if they share a common evolutionary history of gene gains and losses or if they evolve convergently. However, when there is no tendency for closely related strains to occupy similar niches, as appears to be the case for B. subtilis , a common history in the genomic dynamics of vertically inherited genes would be shared, at least transiently, between strains sampled in different environments. For this reason, it is not always easy to differentiate common ancestry from convergent evolution in a pangenome-wide analysis. A lack of genes with a distribution that closely follows the niche classification most likely reflects a species that has not specialized into living in a particular habitat, that is, a species that is a generalist microorganism, of which B. subtilis might be a paradigm."
} | 5,275 |
29295915 | PMC5750404 | pmc | 9,084 | {
"abstract": "ABSTRACT Microbial ecology has been transformed by the advent of high-throughput marker gene and metagenomic sequencing methods. These tools provide expansive descriptions of microbial communities, but the descriptions are framed in terms of molecular objects, such as 97% ribosomal operational taxonomic units (OTUs), rather than biological objects, such as species. A recent study by A. B. Chase and colleagues (mBio 8:e01809-17, 2017, https://doi.org/10.1128/mBio.01809-17 ) explores the so-called microdiversity within the Curtobacterium OTU, the most abundant OTU in a leaf litter community. Perhaps unsurprisingly, they find that some important ecologic traits, such as drought response, are coherent within the OTU, but that others vary significantly. Here we discuss their findings in relation to the more general issue of how molecular tools can be effectively used to study microbial ecology. We specifically note the need for investigators to choose the right molecular methods for their biological problem, as nature does not respect the limitations and conventions associated with our methods."
} | 277 |
28123379 | PMC5226446 | pmc | 9,085 | {
"abstract": "Community assembly processes generate shifts in species abundances that influence ecosystem cycling of carbon and nutrients, yet our understanding of assembly remains largely separate from ecosystem-level functioning. Here, we investigate relationships between assembly and changes in microbial metabolism across space and time in hyporheic microbial communities. We pair sampling of two habitat types (i.e., attached and planktonic) through seasonal and sub-hourly hydrologic fluctuation with null modeling and temporally explicit multivariate statistics. We demonstrate that multiple selective pressures—imposed by sediment and porewater physicochemistry—integrate to generate changes in microbial community composition at distinct timescales among habitat types. These changes in composition are reflective of contrasting associations of Betaproteobacteria and Thaumarchaeota with ecological selection and with seasonal changes in microbial metabolism. We present a conceptual model based on our results in which metabolism increases when oscillating selective pressures oppose temporally stable selective pressures. Our conceptual model is pertinent to both macrobial and microbial systems experiencing multiple selective pressures and presents an avenue for assimilating community assembly processes into predictions of ecosystem-level functioning.",
"conclusion": "Conclusion While future work is needed to validate our conceptual model, our research represents a key step forward in spatiotemporal ecological research by assimilating shifts in community composition, assembly processes, and microbial metabolism. We develop this model by repeatedly sampling attached and planktonic microbial communities within an environmental transition zone over a 9-month time period. Our results indicate a rise in specialist autotrophs associated with increased rates of microbial metabolism when the direction of selection imposed by an oscillating hydrologic environment opposes that of stable selective pressures imposed by sediments. Further, we show that assembly processes operate at different temporal scales in planktonic and attached communities and have taxon-specific effects within each community. Our conceptual model can be applied in a range of ecological contexts beyond microbiology-for example, as a framework for understanding relationships between environmental change and competition under the Stress-Dominance hypothesis in macroecology ( Brown et al., 1996 ; Normand et al., 2009 ) and for developing predictive models of species distributions under novel climate scenarios. As a whole, our work is an advancement in the integration of individual and community-level ecology theory with ecosystem function and develops a conceptual framework for coordinating assembly processes, changes in species abundance, and predictions of ecosystem-level functioning in response to environmental change.",
"introduction": "Introduction The collective effects of community assembly processes (e.g., dispersal, drift, and selection) on microbial metabolism of carbon and nutrients in the environment are poorly understood, and they constitute a key knowledge gap in process-based modeling of biogeochemical cycles. Selection and dispersal both have the potential to impact rates of microbial metabolism. For example, selection can enhance metabolism via species sorting mechanisms that optimize the microbiome for a given environment ( Van der Gucht et al., 2007 ; Lindström and Langenheder, 2012 ), while dispersal limitation can inhibit immigration of metabolic diversity, and in some cases, lead to a maladapted and poorly functioning community ( Telford et al., 2006 ; Lindström and Östman, 2011 ; Hanson et al., 2012 ; Peres et al., 2016 ). The extent to which community assembly processes regulate metabolism is contingent on myriad spatiotemporal dynamics including the geographic distance separating communities, the rate of environmental change, and historical abiotic conditions ( Fukami et al., 2010 ; Graham et al., 2014 ; Nemergut et al., 2014 ; Hawkes and Keitt, 2015 ; Graham et al., 2016 ). Yet, we lack a conceptual basis for how multiple community assembly processes jointly influence microbial metabolism ( Prosser et al., 2007 ; Gonzalez et al., 2012 ; Shade et al., 2013 ; Graham et al., 2016 ). Community assembly processes act through space and time to impact community membership, which then impacts microbial metabolism ( Vellend, 2010 ; Nemergut et al., 2013 ). For instance, communities experiencing a history of strong and consistent selection may contain taxa that are well-adapted to their environment and exhibit high metabolic rates. Alternatively, intense, unyielding selection may eliminate microbial taxa that metabolize scarce resource pools and impair community metabolic functioning ( Grime, 1998 ; Loreau and Hector, 2001 ; Knelman and Nemergut, 2014 ). In this case, more diverse communities (e.g., those experiencing higher rates of dispersal or counteracting selective pressures) would be expected to exhibit higher and more consistent rates of metabolism than those structured by one dominant selective pressure. Dispersal limitation can inhibit the ability of organisms to reach their optimal environment, resulting in lower rates of community metabolism, while high rates of dispersal may either reduce or enhance microbial metabolism, respectively, by allowing for immigration of maladapted organisms or by increasing biodiversity ( Hooper et al., 2012 ; Nemergut et al., 2014 ). Additionally, individual taxa are differentially impacted by community assembly processes. Only some taxa contain traits that are under selection in given environmental conditions, and changes in the environment may affect some taxa to a greater extent than others ( Poff, 1997 ; Lebrija-Trejos et al., 2010 ; Knelman and Nemergut, 2014 ; Krause et al., 2014 ). For example, a change in a certain nutrient should have a larger influence on taxa that directly metabolize it than those that utilize it as a secondary resource. Similarly, traits that facilitate dispersal (e.g., spore formation) are preferentially contained within certain taxa ( Martiny et al., 2006 ; Tremlová and Münzbergová, 2007 ) such that changes in abiotic transport mechanisms should have contrasting effects on taxa with or without traits that facilitate dispersal. Thus, taxon-specific relationships between assembly processes and microbial communities may inform our understanding of the ecological processes influencing changes in environmental microbiomes beyond trends we observe at the community-level. Environmental transition zones present a unique opportunity for examining interactions between microbial metabolism and both long- and short-term assembly processes, as they experience extreme spatiotemporal variation in physicochemical characteristics and microbial community composition across tractable spatial and temporal scales. Here, we leverage inherent variation in hydrology, habitat heterogeneity, and aerobic respiration in a zone of subsurface groundwater-surface water mixing (hereafter termed “hyporheic zone”) to examine the interplay of community assembly processes and microbial metabolism through time. Hyporheic zones are subsurface regions below and adjacent to rivers and streams that experience mixing between surface water and groundwater. The groundwater-surface water mixing within the hyporheic zone can result in blending of complementary resources and, in turn, elevated rates of microbial metabolism relative to other systems ( Hancock et al., 2005 ; Boulton et al., 2010 ). We employ null modeling in conjunction with temporally explicit multivariate statistics to characterize assembly processes driving shifts in microbial communities and microbial metabolism in the Columbia River hyporheic zone. We examine two co-occurring, yet ecologically distinct, habitat types—attached and planktonic communities. Further, we extend the analysis of community assembly processes to individual taxa that likely contribute to observed shifts in microbial metabolism. Our results culminate in a broadly applicable conceptual model coupling changes in selective environments, trait abundance, and ecosystem-level functioning through time.",
"discussion": "Discussion Our results show pronounced seasonal changes in hydrologic and microbial characteristics within the Columbia River hyporheic zone, as well as variation in the importance of selection and dispersal in structuring attached vs. planktonic communities. These community assembly processes operated at distinct timescales in each habitat and were also associated with changes in the relative abundance of certain taxa. In particular, changes in selection exhibited contrasting relationships with putative heterotrophic and autotrophic taxa in attached communities. Further, changes in community-level microbial metabolism correlated with the abundance of these same taxa in attached communities. Based on our findings, we present a conceptual model applicable within both macrobial and microbial systems that links trait selection, organismal fitness, and ecosystem-level functioning in habitats that are characterized by opposing selective pressures across multiple timescales. Microbial Responses to Environmental Change Hyporheic microbial community composition shifted in conjunction with seasonal changes in organic carbon concentration, temperature, and groundwater-surface water mixing conditions. These variables each explained some variation in temporal community dissimilarity within attached and planktonic communities, indicating potential influences of selection (e.g., mediated by environmental change) and dispersal (e.g., mediated by hydrologic transport) over microbial community composition (Table S8). Both selection by the geochemical environment and dispersal from local sediment communities have been demonstrated within the groundwater aquifer in our system ( Stegen et al., 2012 ); however, the balance of these two processes in structuring hyporheic microbial communities remains unclear. Our results indicate that homogeneous selection (i.e., consistently imposed selection for a given set of traits, βNTI < -2) was the dominant assembly process in attached communities, while planktonic communities were influenced by a combination of homogeneous selection, variable selection (i.e., selective pressures that change through space or time, βNTI > 2), and spatial processes (i.e., dispersal, -2 < βNTI < 2 and |RC bray | > 0.95). Because planktonic communities consistently displayed higher βNTI values in comparison to attached communities, often with -2 < βNTI < 2, dispersal processes may play a greater role in structuring planktonic vs. attached communities. In contrast, attached communities always yielded large negative βNTI values suggesting that features of the physical sediment environment may impose stronger selective pressures than aqueous physicochemistry. Because we observed pronounced seasonal trends in species richness associated with changes in the environment, we used null modeling to assess the extent to which selection versus dispersal influenced community composition across changes in species richness. Homogenous selection prevailed in attached communities as differences in species richness increased ( Figure 2B ), indicating strong and consistent selective pressures imposed by a relatively stable environment across temporal changes in microbial diversity. Strong consistent selection through time in attached communities suggests that the physical substrate may inherently contain a limited number of ecological niches—potentially related to mineralogy or physical structure—with slow changes in available niche space. In this case, temporal increases in richness were likely due to the addition of taxa that were ecologically similar to existing taxa and occupied similar niche space. In contrast, differences in planktonic species richness were positively correlated with βNTI, with βNTI values supporting a role for more variable selection as differences in richness increased ( Figure 2C ). In this case, increases in species richness were likely due to the addition of taxa occupying newly available niche space generated by changes in the selective environment, yielding new taxa that were ecologically dissimilar to existing taxa. Variation in assembly processes between attached and planktonic communities may be due to inherent differences among these environments, such as influences of mineralogy ( Carson et al., 2007 ; Jorgensen et al., 2012 ), physical matrix composition ( Vos et al., 2013 ; Breulmann et al., 2014 ), and/or relative rates of change in environment characteristics (discussed below). Further, differences in assembly processes and niche dynamics between the attached and planktonic communities may also be reflective of differing rates of organismal response to fluctuations in the hyporheic environment. Timescales of Selection The timescales at which selection imposes constraints on microbial community composition are poorly understood ( Shade et al., 2012 ; Nemergut et al., 2013 ). Here, we provide new insights into these timescales within the hyporheic zone, showing that selection on planktonic communities operates at the timescale of shifting porewater conditions (sub-hourly to seasonal), while selection on attached communities operates primarily at seasonal timescales ( Figure 3 ). Both communities experienced a seasonal change in composition; however, variation in planktonic communities correlated with porewater physicochemistry at both the sub-hourly and seasonal timescales ( Figure 3A , red and blue arrows, respectively). In our system, planktonic organisms may therefore be influenced by short-term fluctuations in groundwater-surface mixing, either through rapid changes in the selective environment or dispersal via hydrologic transport. In contrast, selection on attached communities was detectable only at the seasonal timescale and exhibited resistance to short-term hydrologic fluctuations ( Figure 3B ). This short-term stability could be facilitated by a number of potentially complementary mechanisms. For example, attached communities may reside within biofilms, whereby microbial cells are imbedded in a matrix of extracellular polymeric substances. Biofilms are prevalent in aquatic systems and buffer communities against fluctuations in the hydrologic environment ( Battin et al., 2016 ). Attached microbial communities may also have adhesion mechanisms ( Hori and Matsumoto, 2010 ) that confer stability. Community assembly processes, such as priority effects, may also contribute to relatively slow rates of community turnover ( Fukami, 2004 ; Fukami et al., 2010 ). Attached communities contained more species (on average) than planktonic communities ( Figure 1D ), further suggesting that temporal stability in attached communities may be enhanced by high species richness reducing susceptibility to invasion ( Stachowicz et al., 2002 ). Taxon-Specific Assembly Processes Taxon-specific assembly processes are masked when relating βNTI and RC bray to environmental variables. To elucidate taxon-specific selection and dispersal mechanisms, we compared the relative abundance of taxa identified by SIMPER to βNTI and RC bray values. In light of rapid hydrologic fluctuations in the porewater environment, relationships between RC bray and taxa abundances in planktonic communities provide evidence for a role of taxon-specific dispersal mechanisms (Table S6). In particular, positive relationships of Thaumarchaeota ( Figure 4A ) and a class of Acidobacteria ( Figure 4B ) with RC bray and negative relationships of Actinobacteria ( Figure 4C ) and Alphaproteobacteria ( Figure 4D ) with RC bray were among the strongest correlations (Table S6). These positive or negative relationships indicate higher or lower relative abundances, respectively, under higher levels of dispersal limitation. No relationships existed between planktonic taxa abundances and βNTI. Although we cannot be certain of the mechanisms responsible for RC bray -abundance relationships in planktonic communities, the trends we observed help elucidate ecological dynamics impacting the abundance of microbial taxa within hyporheic zones. For example, taxa showing positive relationships with RC bray — Acidobacteria and Thaumarchaeota —are widely distributed globally ( Francis et al., 2005 ; Fierer and Jackson, 2006 ; Jones et al., 2009 ; Pester et al., 2011 ), suggesting these organisms may be able to disperse under community-level dispersal limitation. Conversely, taxa exhibiting negative relationships with RC bray have physiologies that may diminish dispersal ability. Alphaproteobacteria can produce filaments that aid in attachment ( Kragelund et al., 2006 ; Jones et al., 2007 ), and dispersal limitation has been demonstrated in soil Actinobacteria ( Eisenlord et al., 2012 ). Thus, negative relationships between these taxa and RC bray may reflect an enhanced ability of these organisms to persist locally relative to other community members. Additionally, when we examined assembly processes governing differences between attached and planktonic communities, we observed selection for microbial taxa in planktonic communities with unique ecological properties (Table S6). No correlations with RC bray were found in these comparisons; however, we identified positive relationships between βNTI and the average relative abundance of two classes of organisms—a candidate class of archaea ( Parvarchaeota , Figure 4D ) and a class of the candidate phylum OP3 ( koll11 , Figure 4E ). Positive correlations between a particular taxon and βNTI across habitat types imply that the taxon increases in abundance as the habitats diverge in selective environments (i.e., selection becomes more variable). In our system, Parvarchaeota and koll11 were almost exclusively found in planktonic communities suggesting that selection in the porewater environment favors these organisms. Although the specific selective pressures regulating the abundance of these organisms are unknown, archaea and members of the PVC superphyla to which OP3 belongs have a cell membrane lacking peptidoglycan that conveys resistance to common antibiotics and have the genetic potential to metabolize C1 compounds such as methane ( Fuerst and Sagulenko, 2011 ). The distinctive features of these organisms and abundance within our system merits future investigating into their role in carbon cycling in hyporheic environments. Finally, we observed changes in the abundance of two major taxa— Betaproteobacteria and Thaumarchaeota —within attached communities that correlated with changes in βNTI. Members of Betaproteobacteria increased in relative abundance with increases in the strength of homogeneous selection ( Figure 4H ), which occurred during times of low groundwater intrusion ( Figures 5A,B ). In contrast, members of Thaumarchaeota ( Figure 4G ) increased as homogeneous selection waned during times of high groundwater intrusion ( Figures 5A,B ). In this scenario positive correlations between a taxon and βNTI indicate that the taxon becomes more abundant as homogeneous selection wanes, and thus, that selection targets traits outside the taxon. In contrast, negative correlations between a taxon and βNTI should indicate that the primary selective pressure is for traits contained within that taxon. Because organisms with wider niche breadths are favored by selection in a greater variety of environmental conditions, positive relationships should be more probable for organisms occupying niches defined by a narrow subset of environmental attributes, whereas negative relationships should be more probable for organisms occupying broader niches. Indeed, Betaproteobacteria and Thaumarchaeota , respectively, contain organisms with diverse and limited metabolic capabilities ( Amakata et al., 2005 ; Yang et al., 2005 ; Sato et al., 2009 ; Pester et al., 2011 ; Beam et al., 2014 ; Weber et al., 2015 ). Functional Effects Through Time Shifts in the relative abundances of Betaproteobacteria and Thaumarchaeota also correlate with changes in groundwater-surface water mixing and in rates of microbial metabolism, denoting a link between community assembly processes and seasonal trends in autotrophic vs. heterotrophic metabolism. Betaproteobacteria is a metabolically diverse taxon, exhibiting a range of aerobic and facultative metabolisms including methylotrophy ( Kalyuzhnaya et al., 2006 ), ammonia-oxidation ( Freitag et al., 2006 ), nitrogen fixation ( Rees et al., 2009 ), phototrophy ( Gifford et al., 2013 ), and a variety of heterotrophic metabolisms ( Amakata et al., 2005 ; Yang et al., 2005 ; Sato et al., 2009 ). Although we cannot be certain of the primary metabolic role(s) of these organisms, a positive correlation between their abundance and NPOC concentration supports their important contribution to heterotrophy in this system (Table S7). Betaproteobacteria remained in high abundance relative to other organisms throughout our study period, despite a seasonal decline, possibly indicating a broad metabolic role for these organisms. In contrast, metabolic activity of Thaumarchaeota is primarily constrained to ammonia-oxidation ( Pester et al., 2011 ; Beam et al., 2014 ; Weber et al., 2015 ). Thaumarchaeota abundance was negatively correlated with NPOC, consistent with their involvement in ammonia-oxidation (Table S7). The relative abundance of Thaumarchaeota also correlated with physicochemical properties associated with increased groundwater-surface water mixing (SO 4 2 , NO 3 , IC, Table S7), suggesting a heightened importance of Thaumarchaeota -mediated nitrification when enhanced groundwater discharge into the hyporheic zone leads to organic carbon limitation ( Taylor and Townsend, 2010 ). Taken together, we hypothesize that selective pressures, both from sediment composition and from porewater physicochemistry, favor diverse heterotrophs within Betaproteobacteria . However, when the porewater environment changes due to a seasonal change in groundwater-surface water mixing, selective pressures shift to facilitate ammonia-oxidizing Thaumarchaeota . More broadly, we propose that community composition in dynamic environments is often the product of multiple selective pressures that operate across different timescales, resulting in an increase of specialist organisms during periods in which selection by an oscillating environment (e.g., hydrologic change) opposes that of a temporally stable environment (e.g., consistent sediment chemistry). Importantly, changes in community composition in our system are associated with a seasonal increase in microbial metabolism, consistent with work in both micro- and macroecology demonstrating that productivity increases with niche diversification ( Hooper et al., 2005 ; Cardinale et al., 2007 ; Cardinale, 2011 ; Gravel et al., 2011 ; Hunting et al., 2015 ). Specifically, we observed positive correlations between Raz:ATP and Thaumarchaeota abundance coincident with negative correlations between Raz:ATP and Betaproteobacteria ( Figure 5D ). We therefore infer that seasonally fluctuating selective pressures from the porewater environment impact microbial metabolism via their influences on niche dynamics. We propose that community-level niche diversification generated by a seasonal rise of specialized autotrophs leads to an increase in community metabolic activity despite selection for heterotrophs imposed by the consistent sediment environment. Ecological Implications Our findings lead to a conceptual model describing relationships between trait selection, organismal fitness, and ecosystem functioning for communities experiencing multiple selective pressures ( Figure 6 ). The conceptual model focuses on the combined influences of stable and oscillating selective pressures, which should be prevalent across ecosystems. For example, in terrestrial ecosystems, physical soil properties and soil water content are relatively stable and oscillating, respectively, from the perspective of associated plant communities. Likewise, in subsurface systems (such as presented here), sediment geochemistry is relatively stable over monthly timescales and hydrologic conditions are continuously fluctuating generating dual selective pressures for benthic organisms. FIGURE 6 Figure depicts a conceptual model describing relationships between trait selection, organismal fitness, and microbial metabolism for communities experiencing dual selective pressures. (A) Selection for a trait follows a continuous gradient within a stable environment (light blue) and oscillating environment (blue to red gradient). Organisms that contain opposing traits (dashed vs. solid lines) are favored at each end of the spectrum, delineated here as to the left (selection against trait 1 and for trait 2) or right (selection for trait 1 and against trait 2) of the vertical gray line. Given selection in a stable environment denoted by the black dot in (A) , variation in homogeneous selection (B) is driven by the magnitude and direction of selection in the oscillating environment. When selection in the oscillating environment opposes selection in the stable environment, homogeneous selection decreases (B) and microbial metabolism increases (C) due to an increase in realized niche space and biodiversity. Blue and red triangles in (B,C) correspond to oscillating selection locations on the fitness landscape in (A) . We specifically hypothesize that selection imposed by a stable feature of the environment can favor organisms possessing traits that oppose traits selected for by the oscillating environment ( Figure 6A ). Because the stable environment applies consistent selective pressures, shifts in the strength of homogeneous selection are driven by changes in the oscillating environment. For instance, given a stable environment represented by the black dot in Figure 6A , the strength of homogeneous selection in Figure 6B is influenced by selection imparted by the oscillating environment, denoted as a gradient from blue to red. In this scenario, increased selection from the oscillating environment causes a decrease in homogeneous selection that results in niche diversification and enhanced microbial metabolism ( Figures 6B,C )."
} | 6,626 |
35479792 | PMC9036576 | pmc | 9,087 | {
"abstract": "Nanopore detection is a label-free detection method designed to analyze single molecules by comparing specific translocation events with high signal-to-noise ratios. However, it is still challenging to understand the influences of structural flexibility of 100 nm DNA origami on nanopore translocations. Here, we used solid-state nanopores to characterize the translocation of “nunchaku” origami structures, the flexibility of which can be regulated by introducing specific DNA strands and streptavidin protein. The structural changes can result in significant variations in the translocation signals and distributions. It is anticipated that such a method of the flexible DNA origami translocation through a solid-state nanopore will find further applications in molecular detection as well as biosensing.",
"conclusion": "Conclusions In this work, we reported a method that used translocation through nanopores to characterize the flexible DNA origami structures with 100 nm dimensions. The flexibility of the origami structures was designed to be controlled via DNA strand hybridizations and protein interactions. The experimental results revealed that small conformational changes in the structures, as verified by TEM, resulted in significant variations of the translocation signals and their distributions. Hence, a SS-nanopore is able to monitor conformational changes in DNA origami structures and could be used for broader applications. Overall, this approach provides a label-free and rapid tool for conformational characterization of DNA origami, as well as nanopore diagnostics and biosensing.",
"introduction": "Introduction DNA origami is a versatile assembly method to construct various programmable and addressable nanostructures. It has attracted significant research interest with applications in molecular engineering, diagnostics, biosensing, molecular sensors, 1,2 drug delivery, 3–8 and enzyme cascade reactions. 9–13 In particular, origami enables an accurate control of nanostructure conformations via DNA base-pairing. 14–20 Nanopore-based devices are ultra-sensitive with regard to conformational changes and label-free modifications at the single-molecule level. 21–24 Nanopore detection extracts characteristics of target molecules by monitoring changes in current signals as the target moves through the pore. Compared with biological nanopores, solid-state (SS) nanopores have controllable shapes and sizes, enabling the detection of various molecules, 25–35 such as nucleotides, 25–29 proteins, 30–32 and reaction products. 33–35 DNA origami translocation through a SS-nanopore has been well-developed by using simple, linear, double DNA duplexes and nanostructures with fixed geometries. 36–38 Although DNA origamis are detected adequately with SS-nanopores, the focus has been on monitoring the translocation of fixed geometries. It is still a challenge to analyze the flexibility of origami structures via translocation through a SS-nanopore. Ideally, it would be advantageous to monitor small nanostructural changes via nanopore detections. In this study, we designed a 100 nm three-dimensional DNA origami with a “nunchaku” structure resembling two sticks connected at one end by a short chain, and examined translocation through 20 nm-diameter silicon-nitride nanopores. Moreover, the flexibility of the origami structures can be regulated by DNA hybridization and streptavidin (SA) protein binding. Here, various kinds of origami structures are detected using the SS-nanopores to produce different nanopore signals and distributions. The experimental results demonstrated that SS-nanopores can monitor the origami flexibility induced by the binding of DNA and protein. This method has potential applications in biomolecular detection and biosensing.",
"discussion": "Result and discussion A DNA origami structure (origami-1) was designed as a nunchaku with a total length of 90 nm ( Fig. 1a ), assembled with a m13mp18 scaffold and 145 short staples (Table S1 † ). 39 The origami-1 structures can be flexibly and dynamically controlled, and are divided into three parts. The two ends are nano-cylinders with 30 nm lengths and 14 nm diameters. A group of connectors consisting of six single-stranded (ss) DNAs form a 30 nm link between the two nano-cylinders. Control of the origami conformation was established by introducing specific DNA helper strands to pull the cylinders ends closer to each other (Fig. S1 † ). Fig. 1 (a) Origami-1 structure and dimensions. (b) 1% agarose Gel results for origami-1. (c) TEM images of origami-1. Scale bar: 200 nm. (d) Schematic of nanopore detection of origami translocation. The structures to be detected and the corresponding specific signals are shown. The nunchaku origamis were assembled via one-step annealing and characterized with agarose gel electrophoresis and transmission electron microscopy (TEM). As shown in Fig. 1b , a sharp gel band was observed in the gel lanes indicating the target products. In the TEM images in Fig. 1c , various origami-1 structures were observed as straight and bent shapes, possibly due to the softness in the middle section. In the TEM results, most of the origami-1 structures were formed as designed. In the nanopore experiments, origami-1 was detected with 20 nm-diameter SS-nanopores fabricated in 30 nm thick SiNx membranes ( Fig. 1d ). To avoid misfolded products, the assembled DNA origami structures were first purified by electrophoresis, and then detected with the SS-nanopore in buffer of 1 M KCl. Accordingly, the specific nanopore signals representing the target origami translocation can be obtained as shown in Fig. 1d . In the design, the origami structures can be regulated by adding DNA helper strands (H1⋯H6) that specifically complement the six ssDNA connectors. Origami-1 is constructed without adding DNA helper strands, thus with flexible structures of straight and bent shapes as indicated in Fig. 2 and S8. † By selectively adding specific DNA helper strands, two other structures of origami-2 and -3 can be constructed. When only one pair of helper strands (H1 and H2) are introduced, the origami-2 is designed to form with a bent structure ( Fig. 2a and S1b † ). In this state, only one point of the cylindrical cross section was tightly connected, while two other points freely swayed. TEM images indicated that most of the origami-2 structures were bent with a statistical range of angles over 100–180 degrees (Fig. S9 † ). Interestingly, when all three pairs of helper strands (H1⋯H6) were introduced to produce origami-3, the cross-sections of the two origami cylinders were tightly connected, the structures were designed to become straight with a relative compact state. The TEM images indicated most of the origami-3 structures were straight with angles of 180 degrees ( Fig. 2a and S10 † ). Meanwhile, the gel electrophoresis results verified that three distinct origami structures were assembled before the purifications ( Fig. 2b ). However, there were no significant differences can be found in the gel migration speeds of the three origami product bands. Fig. 2 Dynamic control of origami structures. (a) Schematic of structural changes in origami-1, TEM images and characteristic nanopore signals (more signals are shown in Fig. S4–S6 † ) when only one pair of the complementary strands (H1, H2) was introduced to produce origami-2. When three pairs of the complementary strands (H1⋯H6) were introduced, origami-1 became origami-3. Scale bar: 200 nm. (b) 1% agarose gel results for the three origamis. The red arrow is the origami structures that was formed. To demonstrate the nanopore signals are indeed induced by the translocations of DNA origami, the varied voltages of 400 mV, 500 mV and 600 mV were applied. The specific translocation events of origami-3 can be observed under the different voltages in Fig. 3a . The nanopore results indicated that the frequency of translocation events increased when higher bias voltages were applied in Fig. 3b and S2. † For origami-3, the mean blockage currents were about 896 pA, 1126 pA, 1568 pA for applied biases of 400 mV, 500 mV, and 600 mV, respectively ( Fig. 3c ). The statistical results demonstrated that the maximum blockage current increased with applied bias. In addition, the mean dwell time (0.53 ms, 0.38 ms, and 0.35 ms) decreased with increasing biases (400 mV, 500 mV, and 600 mV, respectively). These results were consistent with previous reports on assembled DNA complex ( e.g. , tetrahedra DNA) translocation through SS-nanopores. 40 Fig. 3 Translocation event characteristics of origami-3 through a 20 nm-diameter nanopore in 1 M KCl at pH 8.0, for positive applied biases of 400 mV, 500 mV, and 600 mV. (a) Scatter plots of origami-3 events at three applied biases. (b) Current traces at the same time scale for the three applied biases. (c) Mean maximum current blockage vs. applied bias. (d) Mean dwell time as a function of applied bias. The results for translocations of origami-1 structure ( Fig. 2 ) at applied positive biases of 300 mV, 400 mV, and 500 mV were similar to those for origami-3 (Fig. S3 † ). The number of translocation events of origami-1 was significantly reduced at +300 mV, possibly due to an insufficient force to pull the structures through the channel (Fig. S3 † ). Because the flexible origami cylinder structures collided with nanopore edges during the translocation, the flexibility of origami structures would have an influence on the nanopore signals. Here, three origami structures have different conformational flexibilities as indicated in Fig. 4a . Clearly, the structure of origami-1 has a maximum flexibility for its softness in the middle section (Fig. S7 † ). While the flexibility of origami-2 become less, it is because one point of the cylindrical cross section was tightly connected Fig. 4a . Nevertheless, the structure of origami-3 almost lost the flexibility due to the tight connection between the cross-sections of the two origami cylinders Fig. 4a . Fig. 4 Translocation event characteristics of three different DNA origamis through a 20 nm-diameter nanopore in 1 M KCl at pH 8.0 and a +500 mV bias. (a) Schematic of flexibility of three origami. (b) Scatter plot of maximum current blockage versus dwell time. (c) Comparison of mean variances in blockage current. (d) Comparison of mean variances in dwell time. (e) Comparison of dispersions. The feasibility of using nanopores to characterize DNA origami with different structural flexibilities was examined by individually translocating the origami-1, origami-2, and origami-3 structures through the 20 nm-diameter SS-nanopore. All the experiments were performed in 1 M KCl at a +500 mV bias, and the origami samples were added to the cis reservoir at a final concentration of 1 nM. To compare differences in the translocation signals of the three samples, statistical analyses of the translocation events were focused on current blockage and dwell time. Fig. 4b plots the distributions of the translocation event signals from the three origami samples. The mean values in blockage current of origami-1, origami-2, and origami-3 were 1 nA, 0.59 nA, and 0.86 nA, respectively ( Fig. 4c ). Origami-3 translocations induced much higher mean current blockages than those of origami-2, possibly because of the protruding loop structures in the middle section created by the extensive hybridization of the three helper strands. However, origami-3 had the fastest mean translocation time at 0.37 ms ( Fig. 4d ). But the mean translocation time for origami-1 and origami-2 is 0.47 ms and 0.46 ms, respectively. This can be understood in terms of the rigid and compact origami-3 structure relative to those of origami-2 and origami-1. Therefore, the results indicate the flexibilities of origami structures do affect translocation through the nanopore. It was observed in Fig. 4e that the signal distributions for origami-1 had the largest dispersion in translocation events, with much larger current blockages and longer translocation (dwell) time (details of the calculations for the signal distributions are in the Fig. S12b † ). In contrast, the origami-2 and origami-3 signal distributions exhibited much less scattering ( Fig. 4e and S11 † ). A possible reason is that the free state of origami-1 has more flexible structures, which may greatly increase collisions with the nanopore channel during translocations, thus inducing larger blockages and longer translocations. The relatively rigid structures of origami-2 and origami-3 reduced the chances for collisions with the nanopore, and thus had fewer interactions during translocation. To verify whether the signal distributions were affected by structural flexibility, origami-4 was designed with 15 nm-long link connectors in the middle of the origami structure, instead of the 30 nm-long link connectors used for origami-1 (Fig. S13a † ). Accordingly, there was less scattering in the signal distribution for origami-4 relative to that of origami-1, as the shorter link connector in origami-4 allows less structural flexibility (Fig. S13b–d † ). Taking advantages of molecular interactions influencing the origami conformational flexibility, the streptavidin (SA) binding method was used to regulate the origami nanopore translocations. As shown in Fig. 5a , two biotin molecules were designed to attach on the opposite cross sections of the two cylinders in origami-2. Because one SA protein has four biotin connection sites, origami-2 can form an origami/SA complex. After the SA/biotin binding, the structure of origami-2/SA became more rigid because the binding limited the swaying flexibility of the two cylinders. In agarose gel results, a slower migrating band in the gel was observed upon introducing SA to the biotin-labelled origami-2 ( Fig. 5a ). The best origami-2/SA yield was obtained at an origami-2 to SA concentration ratio of 1 : 2. The SA binding-induced origami-2 conformational change was confirmed by TEM images, where most of the origami structures were bent ( Fig. 5b ). By statistical analysis, the origami-2 angle distribution becomes narrow after SA binding, indicating an effective regulation on the origami structures (Fig. S14 † ). Fig. 5 Nanopore characterization of an origami/protein complex. (a) Binding schematic and gel results for a complex of origami-2 and streptavidin (SA). (b) TEM images of the complex. Scale bar: 200 nm. (c) The green current trace was produced by SA translocation through the nanopore; the dark blue current trace was produced by origami-2 translocation; and the yellow trace was produced by translocation of the complex. (d) Scatter plots of the complex translocations in 1 M KCl at pH 8.0 and a +500 mV bias. Histogram of blockage current and dwell time distribution. In the SA-origami binding influenced nanopore experiments, three kinds of samples were used to produce nanopore translocation signals: SA protein, origami-2, and the origami-2/SA complex ( Fig. 5c ). The origami-2/SA complex was purified by agarose electrophoresis. No translocation signal was generated when only the SA protein was present. While significant translocation signals were generated by origami-2 and the complex, using the same SS-nanopore ( Fig. 5c ). Interestingly, the signal distribution range of the SA/origami complex was much wider, with relative larger current blockages and longer times, than that of origami-2 only. Thus, the results indicate the SA binding induced origami-2 conformational changes affected translocation. The mean value of the current blockages for the complex was about 2000 pA, almost twice that of origami-2 ( Fig. 5d ), whereas the mean dwell time were almost the same for origami-2 and the complex. The nanopore results demonstrates that protein-binding-induced conformational changes have more impacts on the current blockage than the dwell time."
} | 3,973 |
34943161 | PMC8698972 | pmc | 9,089 | {
"abstract": "Simple Summary The richness (number of species) of the fungi kingdom is estimated at 1.5 million species, but the vast majority remains unknown. Many of them inhabit plants—the so-called fungal endophytes—and may establish different types of interactions with their host plant. Fungal endophytes have been traditionally studied by letting them grow in appropriate culturing media in petri dishes, but novel massive DNA sequencing techniques which do not require a cultivation step (metabarcoding) are gaining ground. Both techniques were applied and compared to characterize the mycobiome of plants of a tall grass ( Brachypodium rupestre ) growing in high-mountain grasslands with different plant diversity (low and high). The two methods showed similar trends comparing endophyte richness between plant tissue types (root > rhizome > shoot) and between grasslands (low-diversity > high-diversity). However, the metabarcoding identified almost six times more endophyte species than the culturing although the most isolated fungal species via culturing, Omnidemptus graminis, was not recognized via metabarcoding. We conclude that the complementation of both techniques is still the best option to obtain a complete characterization of the fungal endophytic assemblage of the plant species. Abstract Fungal endophytes develop inside plants without visible external signs, and they may confer adaptive advantages to their hosts. Culturing methods have been traditionally used to recognize the fungal endophytic assemblage, but novel metabarcoding techniques are being increasingly applied. This study aims to characterize the fungal endophytic assemblage in shoots, rhizomes and roots of the tall grass Brachypodium rupestre growing in a large area of natural grasslands with a continuum of anthropized disturbance regimes. Seven out of 88 taxa identified via metabarcoding accounted for 81.2% of the reads (Helotiaceae, Lachnum sp. A, Albotricha sp. A, Helotiales A, Agaricales A, Mycena sp. and Mollisiaceae C), revealing a small group of abundant endophytes and a large group of rare species. Although both methods detected the same trends in richness and fungal diversity among the tissues (root > rhizome > shoot) and grasslands (low-diversity > high-diversity grasslands), the metabarcoding tool identified 5.8 times more taxa than the traditional culturing method (15 taxa) but, surprisingly, failed to sequence the most isolated endophyte on plates, Omnidemptus graminis . Since both methods are still subject to important constraints, both are required to obtain a complete characterization of the fungal endophytic assemblage of the plant species.",
"conclusion": "5. Conclusions The endophytic mycobiome of B. rupestre is composed of a few abundant and many rare species, the identification of which depends on the sampling effort. Despite the restricted sampling effort, the two methodologies produced consistent results and detected the same trends in endophytic richness and diversity among tissues (roots > rhizomes > shoot) and grassland types (low-diversity > high-diversity). Comparatively, the metabarcoding method allowed the identification of a much larger number of taxa than the culturing method and revealed differences in richness and diversity that were not apparent with the culturing method (even when a larger number of samples was collected [ 39 ]). Despite the promising results of the metabarcoding technique, the data indicate that a combination of the two methodologies is the best current option to obtain an adequate characterization of the plant fungal assemblage. In this study, metabarcoding did not identify Omnidemptus graminis , the most abundant fungal endophyte isolated in shoots via culturing; this recently described species is included in a family where there have been repeated taxonomic restructurings as a result of molecular advances [ 65 ].",
"introduction": "1. Introduction The study of microorganisms in their natural environment is a recent branch of research compared to microbial investigations undertaken in disciplines such as medicine and agronomy, with high impact on human health and development [ 1 , 2 ]. Nowadays, microbial ecology, i.e., their diversity in nature, their response to prevailing and future environmental conditions, the associations they establish with plants and the complex network of interactions and functions they are involved in, are gaining ground in ecological research [ 3 , 4 , 5 ]. One example involves examining the associations that endophytic fungi establish with plants. These associations were first studied in agronomic grasses [ 6 , 7 , 8 ] and the research has extended to natural plant communities in recent decades [ 9 , 10 , 11 ]. Scientific literature has shown that these hidden associations are ubiquitous in nature and that all plants harbor an endophyte assemblage that delivers different functions and constitutes a collective and complex holobiont [ 12 ]. Nowadays, two techniques, culturing and metabarcoding, are used for the determination of fungal endophyte assemblages [ 13 ]. The protocols of culturable techniques have a longer record and have been implemented in many laboratories [ 14 ]. In this method, important constraints include the possibility that some fungal species are unculturable on artificial medium and the accumulation of inaccuracies and errors due to different sterilization times, diverse species growth rates and the presence of surface contaminants [ 15 ]. Metabarcoding techniques (culture-independent) [ 16 ], despite appearing very promising, still remain costly and lack a complete repository of sequences with taxonomic identification, a task which is under way [ 17 , 18 ]. In the latter, the potential for providing quantitative data based on the proportion of read sequences makes it a very powerful ecological tool [ 19 , 20 ]. The genus Brachypodium encompasses several perennial tall grasses, native to European calcareous grasslands, which have been expanding aggressively in the last decades due to the global change conditions ( B. pinnatum , B. genuense and B. rupestre) [ 21 , 22 , 23 , 24 , 25 ]. This tall grass expansion causes a decline of the biodiversity of the natural grasslands and also has an impact on the ecosystem service of provisioning [ 26 ]. The competitive strategies of this group of species that explain the expansive process is a matter of interest [ 27 , 28 , 29 , 30 , 31 , 32 , 33 ], as it is the study of the mycobiome that may help to understand these advantages. To date, the research in the Brachypodium genus has focused on the systemic fungi of the Clavicipitaceae family hosted by B. sylvaticum [ 34 , 35 ], B. phoenicoides [ 36 , 37 ] and B. pinnatum [ 38 ]. Only a previous study of our research team has characterized the systemic and non-systemic mycobiome of Brachypodium rupestre under a gradient of grazing and fire disturbances using culturable techniques [ 39 ]. The aim of this research is to provide a characterization of the endophytic mycobiome of the tall grass species Brachypodium rupestre and to compare culture and metabarcoding techniques applied to conditions with restricted sampling effort due to the high cost of the novel technique. The comparison includes the aboveground (shoot) and the underground (rhizome and root) component of a set of B. rupestre individuals growing in the same region but subjected to different levels of anthropic disturbance (grasslands with different regimes of grazing and prescribed burning and, consequently, encompassing a different plant community composition). Through this range of regional variation, and considering different tissues and different environmental drivers, we are interested in determining the capacity of the two methods to identify and characterize the fungal endophyte assemblage of B. rupestre .",
"discussion": "4. Discussion 4.1. The Mycobiome of B. rupestre According to the Metabarcoding Data The results of the metabarcoding showed that 88 taxa constituted the mycobiome of B. rupestre and that only seven taxa sequenced from the belowground tissues accounted for 81.2% of the total reads (Helotiaceae, Lachnum sp. A, Albotricha sp. A, Helotiales A, Agaricales A, Mycena sp. A and Mollisiaceae C), while the other 81 taxa were responsible for the remaining 18.8%, and 25 of them were only sequenced in a single sample. Therefore, a restricted sampling effort using the metabarcoding method was able to identify a small group of abundant fungal endophytes and a large group of rare species. The accumulation curves also supported the idea that extension of the sampling effort would enrich the group of rare species but not the most common species. This pattern of fungal endophyte distribution seems common to grasses [ 52 ] and indicates that a limited sampling effort is enough to provide good characterization of the dominant fungal species in plants, which is important considering the high cost of metabarcoding. However, when addressing studies on fungal richness and diversity, more extended sampling appears necessary to avoid an underestimation of the values. The results of the study also highlight the importance of sampling the different tissues of plants to obtain a reliable characterization of its mycobiome [ 53 , 54 ]. Aboveground fungal assemblages were much poorer in species, less diverse and taxonomically different from those of rhizomes and roots, and this pattern was consistent between the grassland types, as observed by other authors in different plant species and different habitats [ 55 , 56 , 57 ]. The soil rhizosphere is the main route of fungal transmission to plants [ 58 , 59 ], and the high biomass of rhizome and roots developed by B. rupestre offers a large surface in contact with the soil microbiome. The majority of taxa identified were specific to a tissue, or exhibited a strong preference for it, and only five taxa appeared in all tissues (Helotiaceae, Lachnum sp. A, Ophiosphaerella sp., Microdochium sp. and Epicoccum sp.). As expected, the relative abundances of taxonomic orders and families also varied between tissues, with Pleosporales and Phaeosphaeriaceae more abundant in shoots and Helotiales and Hyaloscyphaceae more abundant in rhizomes and roots. When comparing these results with previous characterizations of fungal endophyte assemblages in perennial temperate grasses based on culture techniques and extensive surveys, we realize the power of the metabarcoding tool, which is capable of identifying a large set of taxa with much less sampling effort. In Dactylis glomerata , 22 and 48 taxa were identified using culturing methods from the leaves and the roots of 120 samples [ 60 ], and in Holcus lanatus , 77 and 79 were identified in the same tissues of 77 samples [ 61 ]. The results of our survey of the leaves and roots of B. rupestre (2 and 11 taxa identified using the culturing method and 12 and 82 taxa identified using metabarcoding) obtained from a small number of samples in a regional sampling suggest that the real diversity and richness of the endophytic fungal assemblages of the previously studied grass species have probably been underestimated and would increase greatly if the novel metabarcoding techniques were used. 4.2. Culturing vs. Metabarcoding Methods Modern massive sequencing techniques are gaining ground over traditional culturing methods due to the quantitative power of data that they are able to generate. With equal sampling effort, metabarcoding identified 13, 32 and 71 more taxa than culturing methods in shoots, rhizomes and roots, respectively, which means around ×5.8 times more species identified by the novel technique consistently in the three tissues. In similar studies comparing both methods, the metabarcoding identified ×5.2 and ×4.3 times more OTUs in roots of Elymus repens and Deschampsia flexuosa respectively than the culturing technique [ 62 , 63 ]. A parallel study using 240 plants of B. rupestre recognized 45 fungal endophytic taxa using the culturing method [ 39 ], in contrast to the 88 taxa sequenced using metabarcoding from 10 plants in the current survey. In this parallel study, the singletons isolated accounted for 48.9% of the taxa identified via culturing methods and 28.4% of the taxa identified via metabarcoding (with OTUs clustered with a 97% of similarity threshold). Regarding belowground tissues, four fungal species with high incidence in root tissues were identified via both methodologies: Albotricha sp., Helotiaceae, Lachnum sp. and Mollisiaceae. In shoots, surprisingly, the most frequent shoot endophyte identified via the culturing method, Omnidemptus graminis , was not identified using the metabarcoding technique. O. graminis is a recently described taxon, included in a family associated with ongoing taxonomic changes due to molecular advances [ 64 , 65 ]. Its fast mycelial growth observed on culture plates may suggest the encrypting of other endophytes, but how O. graminis escaped the sequencing process of the metabarcoding is a matter that needs further study. At this point, some issues need to be discussed when comparing the technical procedures of sequencing in both techniques. The ITS region is a universal and commonly used DNA barcode marker for fungi [ 66 ], and in the metabarcoding study undertaken by an external company, only the ITS2 region was amplified to identify the fungal sequences [ 67 , 68 ]. In the culturing method undertaken in the UPNA’s lab, the fungal mycelium was collected and the complete ITS region was amplified (ITS1-5.8S-ITS2), generating longer DNA sequences. We suggest that, since the ITS2 region is more restrictive, taxonomic inconsistencies may occur when short sequences are compared in the databases, thus affecting taxon identification [ 18 ]. The percentages of taxa identified for the metabarcoding were in the range 78.1–100%, and 97.6–100% for the culture sequencing, evidencing this restriction and indicating the value of sequencing the complete ITS region to achieve better fungal taxa identification. As a particular example, the taxon proposed as Codinaea sp. reached a match of 99.74% with the complete ITS region sequenced, while this percentage decreased to 97.52% when considering only the ITS2 region. As a consequence, the species was identified as Chaetosphaeriaceae in the metabarcoding, following a more conservative approach, although it was probably the same taxon. Similar situations may occur in other closely related taxa, when there is no reference specimen in the database [ 43 , 69 ]. Taxa identified as Mollisiaceae in our study probably belong to the genera Mollisia and/or Phialocephala [ 70 , 71 ] and the family Helotiaceae to the genera Glarea and/or Hymenoscyphus [ 72 ]. Both families were abundant in our samples. Other highly inclusive taxa, such as Pleosporales, raised similar doubts in the identification due to the still high uncertainty in the genetic characterization of the type specimens. Despite the remarkable differences between the quantitative data generated using the two methods, the characteristics of the fungal assemblages in the different plant communities and tissues types are consistent between methods. Root tissues display the most diverse and rich fungal assemblages, and the endophytic community in plants collected in more disrupted, LD grasslands had the highest diversity and richness. Similar patterns have been reported in previous research in the area, conducted with a much greater sampling effort and using the culturing method [ 39 ], that analyzed the fungal assemblages in terms of the ecological mechanisms favored by the different disturbance regimes."
} | 3,930 |
26442074 | PMC4585140 | pmc | 9,090 | {
"abstract": "Properties encompassed by host-pathogen interaction networks have potential to give valuable insight into the evolution of specialization and coevolutionary dynamics in host-pathogen interactions. However, network approaches have been rarely utilized in previous studies of host and pathogen phenotypic variation. Here we applied quantitative analyses to eight networks derived from spatially and temporally segregated host ( Linum marginale ) and pathogen ( Melampsora lini ) populations. First, we found that resistance strategies are highly variable within and among networks, corresponding to a spectrum of specialist and generalist resistance types being maintained within all networks. At the individual level, specialization was strongly linked to partial resistance, such that partial resistance was effective against a greater number of pathogens compared to full resistance. Second, we found that all networks were significantly nested. There was little support for the hypothesis that temporal evolutionary dynamics may lead to the development of nestedness in host-pathogen infection networks. Rather, the common patterns observed in terms of nestedness suggests a universal driver (or multiple drivers) that may be independent of spatial and temporal structure. Third, we found that resistance networks were significantly modular in two spatial networks, clearly reflecting spatial and ecological structure within one of the networks. We conclude that (1) overall patterns of specialization in the networks we studied mirror evolutionary trade-offs with the strength of resistance; (2) that specific network architecture can emerge under different evolutionary scenarios; and (3) network approaches offer great utility as a tool for probing the evolutionary and ecological genetics of host-pathogen interactions.",
"conclusion": "Conclusions We conclude that network approaches offer great utility as a tool for probing the ecological and evolutionary genetics of host-pathogen interactions. Prior analyses of the datasets used in this study did not attempt to estimate nor dissect the network characteristics examined here, and the network meta-analysis has revealed several novel results that complement and extend findings revealed in previous studies. First, we found that host resistance (and by extension pathogen infectivity) strategies are consistently variable within and among networks, corresponding to a spectrum of specialists and generalists being maintained within all networks. While the maintenance of diversity for resistance and infectivity is well known in this system, the network approaches we have utilized reveal novel information regarding the distribution and statistical structure of these specificities. Relationships between patterns of specialization and partial resistance further suggest evolutionary trade-offs between specialization and the strength of resistance. Second, we found that all networks were significantly nested. While we hypothesized that temporal evolutionary dynamics might be important for the development of nestedness in host-pathogen infection networks, there was little evidence to suggest that this was so. Rather, the common patterns observed in terms of nestedness suggest the existence of general determinants across networks (e.g., trade-offs or underlying genetics). Third, we found that resistance networks were significantly modular in two spatial networks, clearly reflecting spatial and ecological dynamics within one of the networks, and perhaps reflecting genetic structure within networks more generally. Together, these results demonstrate that analysis of the topology of bipartite interaction networks has the potential to provide important information regarding the genetic interactions that underpin disease outcomes and coevolutionary dynamics within host-pathogen associations. In particular, our results demonstrate that network approaches have potential to complement more commonly used approaches for analysing population and temporal-based sampling by placing a more explicit focus on individual variation in patterns of specificity, and revealing structure (i.e., nestedness and modularity) in the data that are independent of common a priori hypotheses (e.g., local adaption). Finally, in terms of new directions, we suggest that network approaches may offer great utility for dissecting the genetic nature of host-pathogen interactions from population level data, perhaps in combination with other approaches (Heath and Nuismer, 2014 ). In particular, we suggest that in systems where performing classical genetics is problematic, network structure has potential to reveal the genetic architecture underlying antagonistic host-pathogen interactions (e.g., GFG, MA). In particular, these genetic models are predicted to generate contrasting patterns in terms of modularity (high for MA) and nestedness (high for GFG) (Moury et al., 2014 ). We suggest that a useful goal for future theoretical studies is to examine the role of life-history (e.g., dispersal) and contrasting genetics in driving patterns of nestedness and modularity in host-pathogen networks. Conflict of interest statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.",
"introduction": "Introduction Interactions between wild hosts and their pathogens are typically characterized by high levels of genetic diversity for partner specificity, such that pathogens vary in their capacity to infect individual hosts, and hosts are likewise variable in their ability to resist attack by individual pathogens (Laine et al., 2011 ; Tack et al., 2012 ). Furthermore, hosts and pathogens vary not only in terms of specificity for individual partners, but also in terms of partner range (i.e., breadth of resistance; pathogen host-range) (Barrett et al., 2009a ; Barrett and Heil, 2012 ). Understanding the factors that influence the genetic specificity of host-parasite interactions is important as patterns of association underlie susceptibility to disease, and thus many aspects of disease dynamics and epidemiology (see Thrall et al. this issue for review). However, the ecological and evolutionary factors that determine variation in individual specialization and partner breadth in host-pathogen interactions are generally not well understood. Polymorphisms for resistance and infectivity in wild host–pathogen interactions are typically measured by performing pair-wise infections of host lines by pathogens isolated from natural populations. The results of such pair-wise infections can be represented as a matrix or bipartite network (see Box 1 for description), where the rows indicate host genotypes, the columns indicate pathogen genotypes, and the cells within the matrix indicate whether a given combination results in resistance or infectivity. Bipartite-network analytical methods and approaches have been used extensively in observational studies of mutualistic (e.g., Vázquez et al., 2009 ; Guimarães et al., 2011 ), and host-parasite (e.g., Vázquez et al., 2005 ; Morris et al., 2014 ) networks, but have been only rarely used to analyse patterns of phenotypic variation in host resistance and pathogen infectivity (but see Flores et al., 2012 ; Poisot et al., 2013 ). Network approaches are of potential utility for analysing host-pathogen interactions because the statistical structure of such networks offers a standardized framework for describing and quantifying patterns of specialization within host-pathogen interactions (Blüthgen et al., 2006 ; Vacher et al., 2008 ; Weitz et al., 2013 ). Specificity can be simply estimated for individuals, or the network as a whole, based on the number of links (i.e., resistance or infectivity interactions) an individual has with its antagonistic partners (Poisot et al., 2013 ). In addition, patterns of specialization can be characterized by estimating two key network properties, nestedness and modularity, within the infection or resistance matrix. Nestedness is a network property describing the extent to which specialists interact with a subset of partners that also interact with generalists, whereas modularity indicates the extent to which resistance or infectivity interactions can be partitioned into distinct groups, each of which has many internal interactions but few with other groups (see Box 1 for a detailed description of these key statistical properties). Box 1 The structure of host-pathogen coevolutionary networks. A network can be defined as a set of items, called nodes, connected by links if they interact. In host pathogen interactions, networks have two sets of nodes (each set representing individual hosts or pathogens) and are therefore termed “bipartite networks.” Links between nodes represent resistance or infection phenotypes (depending on the focus of the analysis). In such a network, specificity can be estimated for individuals or the network as a whole simply based on the number of links an individual has with its partners. In addition, patterns of specialization are often characterized by estimating two key network properties commonly known as nestedness and modularity (Figure 1 ). In host-pathogen networks, nestedness relates to the differentiation of resistance or infectivity specificities along a contained gradient, within which specialists (individuals with few links) interact with subsets of the partners interacting with generalists (individuals with many links) (Flores et al., 2012 ). For example, in a maximally nested infection network, the most specialized pathogen can infect only the hosts most susceptible to infection. The next most specialized pathogen could infect the host most susceptible to infection as well as one additional host, and so on (Figure 1A ). Nestedness is a commonly encountered property in mutualistic networks (Jordano et al., 2003 ), and significant nested structures have been also found in studies of host-parasite (Vázquez et al., 2005 ; Vacher et al., 2008 ) and bacteria-phage (Flores et al., 2012 ; Poisot et al., 2013 ) networks. Modularity indicates the extent to which resistance or infectivity interactions can be partitioned into groups (referred to as modules: Figure 1C ) with many interactions within groups but few among them (Blüthgen et al., 2008 ). In a maximally modular network there would be no cross-infections between pathogens in one module and hosts in another. Modularity differs from nestedness in that specificities cannot be simply ranked by increasing range. Rather, interactions take place among distinct clusters of host and pathogen individuals, within which distinct patterns of specificity (including nestedness) may be evident (Flores et al., 2012 ). Like nestedness, modularity is commonly detected in species interaction networks (Olesen et al., 2007 ; Fortuna et al., 2010 ). Figure 1 Network structure properties of host-pathogen interactions . Two important properties of ecological networks are nestedness and modularity. Here we show four cartoons representing different host-pathogen genetic interaction matrices, each with different levels of nestedness (A, B) and modularity (C, D) . For each matrix we show hosts in columns and pathogens in rows. Black squares in each matrix represent resistance between a plant and a pathogen genotype and gray squares represent host susceptibility. In (C) , red solid lines define host-pathogen interaction modules. Key Network Properties: Nodes : bipartite networks have two sets of nodes with each set representing individual hosts or pathogens. Links : resistance or infection phenotypes “linking” the nodes Nestedness : is a network structure describing a scenario where specialist individuals or genotypes (e.g., hosts with a very narrow resistance spectrum) interact with a subset of partners with which generalist individuals or genotypes (e.g., hosts with a very broad resistance spectrum) also interact (Bascompte et al., 2006 ). Modularity : A module (sometimes called a compartment) in a network is formed by a group of genotypes or individuals, which are more connected to one another than to individuals in other groups (Olesen et al., 2007 ). A modular network consists of a series of interconnected modules. In a host-pathogen network, modularity indicates the extent to which resistance or infectivity interactions can be partitioned into groups (referred to as modules: Figure 1C ) with many links within groups but few among them. Connectance : is the proportion of links or realized interactions (p) over all possible interactions (P × H) ( C = p/(P × H)). Patterns of specialization can in turn be employed to make inferences regarding evolutionary dynamics, in that specific structures in host-pathogen networks may be predicted given different evolutionary scenarios. For example, modular patterns of specialization might be expected to emerge in networks where there is spatial or temporal segregation of host and pathogen genotypes (e.g., local adaptation; ecotypic divergence; cryptic speciation). Simply put, spatially or temporally co-occurring individuals within a network may be predicted to be more likely to fall within the same module, and the greater the evolutionary, spatial or temporal separation, the stronger the forces generating modularity should be. Nested patterns of specialization may likewise emerge under specific evolutionary conditions. For example, within a network constructed from a temporal sequence of interacting hosts and pathogen populations, patterns of nestedness emerging over time may be consistent with the classically envisaged stepwise coevolutionary arms race (Beckett and Williams, 2013 ) occurring via selective sweeps of novel resistance and infectivity alleles. Under this simple model in which pathogens evolve over time to expand their range of infectivity, and hosts counter-evolve to become more resistant, the host range of past pathogens should be sequentially nested within the host range of future pathogens. In other cases, spatial variation in the distribution of resistance and infectivity genotypes and fitness trade-offs (Thrall and Burdon, 2003 ) may maintain variation in specificity within or among populations, which may in turn result in promote the nesting of resistance and infectivity specificities. The underlying genetic architecture of the interaction has also been predicted to be important in generating patterns of specificity and hence network structure (Flores et al., 2012 ; Moury et al., 2014 ). Several different models that describe the genetic nature of host-pathogen interactions have been proposed, the most important of which are the gene-for-gene (GFG) and matching allele (MA) models (see Thrall et al. this issue for review). The GFG model is commonly described in plant-pathogen associations and is generally well supported by phenotypic and genetic data as well as by a detailed mechanistic understanding of genes governing plant immune responses and pathogen infection (Dodds and Rathjen, 2010 ). In GFG systems, resistance is dependent upon plants producing a specific resistance gene product that recognizes specific pathogen elicitors (known as Avr genes). Pathogens that do not produce an elicitor recognized by a given R gene are able to infect. In contrast, the MA model describes an interaction where infection is dependent upon the pathogen carrying a gene or allele that is a specific match to a corresponding host genotype. Hosts without the recognized allele are resistant (Thrall et al. this issue). All else being equal, the topology of resistance and infectivity matrices (i.e., patterns of nestedness and modularity) is likely to more or less reflect underlying genetic interaction. Under a GFG model with additive variation in the number of genes or alleles conferring additional resistance or infectivity specificities, cross-infectivity, where a parasite with a given genotype is able to infect hosts with different genotypes, is common (Thrall and Burdon, 2002 ). Pathogen mutations conferring new infectivity do not necessarily result in recognition by an alternative R gene. The same is true for host resistance. Hence, there is strong potential for the maintenance of variation in partner breadth within GFG interactions, and the genetic interaction matrix is seemingly naturally nested (Thrall et al. this issue). In contrast, under the MA model, pathogen mutations conferring new infectivity simultaneously result in the emergence of resistance in existing host genotypes. Hence, cross-infectivity is rare, and the interaction matrix is inherently modular (Moury et al., 2014 ). The main aim of this paper is to apply network based approaches to investigate patterns of specialization within spatial and temporal plant-host fungal-pathogen interaction (i.e., resistance/infection) networks. Linum marginale is a perennial herb endemic to southern, temperate areas in Australia. Melampsora lini is an autoecious, macrocyclic foliar rust pathogen that is restricted in Australia to L. marginale . Utilizing existing data from the M. lini - L. marginale interaction, network structure in resistance and infectivity networks can be investigated in a relatively extensive, balanced and rigorous way. Disease outcomes in this system are governed by a GFG interaction and extensive polymorphisms for host resistance and pathogen infectivity form the basis for selective changes in host and pathogen populations (Burdon, 1994 ; Barrett et al., 2009b ; Thrall et al., 2012 ). Furthermore, sampling can readily be done across a range of defined temporal and spatial scales, and the system is well characterized from various ecological and evolutionary perspectives. The pathogen has substantial fitness effects on the host, with 60–80% reductions in population size documented during severe epidemics (Jarosz and Burdon, 1992 ). Plants recruit via predominantly self-fertilized seed (Burdon et al., 1999 ) and the pathogen is almost exclusively clonal within the region from which all pathogens used in this study were collected (Barrett et al., 2008 ). Previous work has established roles for local adaptation (Thrall et al., 2002 ), gene-flow (Barrett et al., 2008 ), trade-offs (Thrall and Burdon, 2003 ), diversifying selection (van der Merwe et al., 2009 ), and spatio-temporal dynamics (Thrall et al., 2012 ) in driving evolutionary change in interacting host and pathogen populations. The results of these studies provide clear evidence: (i) of a genetic basis for race-specific resistance and infectivity; (ii) that there is variation among individuals for these traits within and among populations; and (iii) that pathogen attack impairs host fitness. To examine how patterns of specialization in infectivity and resistance develop in the L. marginale - M. lini interaction we analyzed data networks collected as part of previous investigations of spatial and temporal evolutionary dynamics. Specifically, we analyzed data from two spatial (Thrall et al., 2002 ; Laine et al., 2014 ) and six temporal interaction networks (Thrall et al., 2012 ). We asked: Do patterns of specificity vary across different networks within the same interaction? Are there significant modules and nested structures within different spatial and temporal networks? Are these related to spatial and temporal segregation of interacting genotypes? How do patterns of specificity change across different networks and can drivers of change be identified?",
"discussion": "Discussion Summary of major results Genetic variation for resistance and infectivity are ubiquitous in wild populations of plants and their associated pathogens (Laine et al., 2011 ; Tack et al., 2012 ). However, many questions remain regarding the ecological and evolutionary processes that generate and maintain such variation. In this study we used a network analytical approach to examine the architecture of host-pathogen interactions through space and time using several extensive datasets from a longstanding wild plant-pathogen system. We found consistent, and in some cases, strong patterns in the data. First, we found that a spectrum of specialist and generalist resistance (or infectivity) types are consistently maintained within these networks and that partially resistant host are more likely to have a broad resistance spectrum. Second, we found that all networks were significantly nested. Third, we found that resistance networks were significantly modular in both spatial networks, but in only one of the six temporal networks. These results demonstrate that network approaches have potential to complement more commonly used approaches to analysing population and temporal-based sampling by placing a more explicit focus on individual variation in patterns of specificity, and revealing structure (i.e., nestedness and modularity) in the data that are independent of common a priori hypotheses (e.g., local adaption). Below, we discuss how the topology of L. marginale-M. lini interaction networks has the potential to provide novel insight into the evolution of genetic interactions that underpin disease outcomes within this host-pathogen association. Spatial dynamics and network structure Spatial evolutionary processes (e.g., local adaptation of pathogens to their hosts) have been demonstrated to be important drivers of genetic structure and specialization in several plant host–pathogen interactions (see Barrett et al., 2009a for review). Here we investigated the idea that patterns of modularity within networks have potential to reveal the strength, scale, and direction of spatial (co)evolutionary processes. In particular, given small, geographically, and genetically differentiated L. marginale and M. lini populations (e.g., Barrett et al., 2008 ; Nemri et al., 2012 ; Thrall et al., 2012 ), and previously reported patterns of local adaptation in this system (Thrall et al., 2002 ; Laine et al., 2014 ) we hypothesized that modularity should be apparent in the two spatial networks (i.e., LA and BH) and that there should be clear links between population structure and patterns of modularity. For the BH network, ecotype emerged as a strong predictor of modularity. In particular, for the four modules detected, one module almost exclusively comprised bog hosts resistant to hill pathogens, while the three remaining modules comprised mostly hill hosts resistant to bog pathogens. At the population level (i.e., within each ecotype), the locality from which individuals were collected was also a significant predictor of modularity, although divisions were not as clear as for ecotype. These results are consistent with Laine et al. ( 2014 ) who demonstrated using the same dataset that hosts were more likely to be resistant to pathogens from a different host-ecotype, and conclude that habitat type is a strong driver of evolutionary divergence among both hosts and pathogens. The results also support those of Flores et al. ( 2012 ) found that modularity in a bacteria-phage network was driven in part by geographic structure, suggesting that geography may be an important determinant of modularity generally. However, it should be noted that resistance across habitat types was also observed frequently (e.g., generalist hosts with wide ranging resistance), and modules typically contained hosts and pathogens from both habitats. In contrast to the BH network, there was no obvious geographic signal in the LA network. Rather, modules were mostly composed of individuals sourced from a mix of different populations (geographical locations) despite the strong evidence for local adaptation (Thrall et al., 2002 ). Hence, while our results are consistent with the idea that spatial evolutionary dynamics may result in the emergence of modularity, with regards to our data it seems that spatial structure may only be sufficient under some conditions. For example, the ecotypic differentiation among hill and bog environments may well result in stronger barriers to gene flow and stronger local spatial selection gradients than in the LA network which included populations within the hill ecotype only. Local geographic separation (as is the case for the LA network) may not be enough to drive the evolution (or maintenance) of significant modularity at the local metapopulation scale where ecological barriers to gene flow are weak or non-existent and other selective forces are at play (e.g., Thrall and Burdon, 2003 ). Temporal dynamics and network structure Temporal coevolutionary processes have been widely hypothesized and in some cases demonstrated to be important drivers of genetic structure and specialization in host–pathogen interactions (Decaestecker et al., 2007 ). We investigated whether networks constructed from temporal sequences of interacting host and pathogen populations displayed patterns of change in nestedness or modularity over time (Beckett and Williams, 2013 ). We found little evidence for consistent temporal changes in either metric, despite findings supporting reciprocal coevolution in a previous study (Thrall et al., 2012 ). Significant modularity was evident in only one of the six temporal networks, and different modules in the Kiandra network were composed of various individuals collected from a mixture of different time-points. In addition, while we found significant patterns of nestedness in all temporal networks, there was little evidence to support the hypothesis that temporal patterns of genetic change were responsible for generating them. In particular, no patterns of increasing nestedness or even generality over time were evident in the data (e.g., with respect to host resistance to pathogens sampled from previous time points). This suggests that temporal evolutionary processes, at least over the time-frames, within which we sampled, are insufficient for generating clear patterns of either nestedness or modularity within the host-pathogen networks sampled in this study. However, it should be noted that L. marginale is a short-lived perennial host (approximately 5 years) and while pathogen can have substantial fitness effects on the host, disease epidemics do not predictably occur in all years or populations. As for the spatial evolutionary scenarios discussed above, ecological and evolutionary patterns are strongly dependent on the scale of inquiry as well as host and pathogen life-history, and in this case, it is possible that patterns may only become evident for comparisons that encompass longer time periods. Furthermore, these results should also be interpreted in light of the high levels of partial resistance that we found in 5 of the 6 temporal networks, in that selection pressures may be more diffuse in these situations given the lower levels of resistance specificity maintained (Antonovics et al., 2011 ). Mechanisms determining nestedness Considering the variation in sampling scale, the repeated finding of nested patterns of specialization across networks suggests this is a universal pattern for L. marginale-M. lini interaction networks. Because utilization of network approaches in studies of patterns of infectivity and resistance within host-pathogen interactions is still in its infancy, it is difficult to speculate on the comparative significance of these results. However, in one of the few other large scale studies of network structure in host-pathogen interactions, Flores et al. ( 2011 ) report consistent, significant patterns of nestedness across 38 bacteria-phage networks, suggesting that nestedness may be a common property of antagonistic interaction networks. However, the processes generating nested structures in our system (or the phage networks) are not obvious and may be attributable to several non-exclusive mechanisms. One parsimonious explanation is that the underlying genetic architecture of the interaction may constrain networks to a nested shape (Flores et al., 2012 ; Moury et al., 2014 ). In GFG systems such as the L. marginale-M. lini interaction, qualitative patterns of resistance such as those reported in this study are dependent upon plants producing a specific resistance gene product that recognizes specific pathogen elicitors. Assuming that loss of the elicitor does not enable activation of another R gene or otherwise compromise infectivity, pathogens that do not produce a recognized elicitor are able to infect. Assuming additive variation in the number of genes or alleles conferring additional resistance or infectivity specificities, cross-infectivity, where a pathogen with a given genotype is able to infect hosts with different genotypes, should be common (Thrall and Burdon, 2002 ). Hence GFG systems generate strong potential for variation in partner specificity, and may naturally generate nested interaction networks (Moury et al., 2014 ). A related and non-exclusive mechanism involves fitness trade-offs. In particular, pleiotropic costs of maintaining multiple R genes, or alleles that confer multiple specificities, may also be important in explaining the consistent maintenance of variation in resistance breadth (Barrett and Heil, 2012 ). Costs of resistance (e.g., Karasov et al., 2014 ) and infectivity (e.g., Barrett et al., 2011 ) have been demonstrated in several plant-pathogen interactions, including for M. lini , where trade-offs between pathogen host-range and spore production have been demonstrated (Thrall and Burdon, 2003 ). Certainly trade-offs provide a general explanation for the maintenance of specialist resistance types in the face of pathogen induced morbidity and mortality. Finally, it has also been proposed that nestedness may be a reflection of ongoing and dynamic evolutionary processes that typify antagonistic species interactions, such that hosts generally adapt to new infectivity genes or alleles without losing their specificity for older forms of infectivity (Flores et al., 2011 ). However, while nested patterns may potentially emerge over long-term time scales, results from our temporal networks (as discussed above) are not consistent with such dynamics generating these patterns on shorter time scales. Mechanisms determining modularity While our results for the BH network are consistent with the idea that spatial evolutionary dynamics can drive the emergence of modularity (as discussed above), results from the LA and K (temporal) networks demonstrate that modularity can transcend spatial (and temporal) structure. One likely common determinant of modularity within host-pathogen interaction networks is genetic divergence among groups of individuals comprising one or both nodes in the network (Flores et al., 2011 ; Weitz et al., 2013 ). This also serves as a potentially general explanation for our findings of modularity across the LA, BH, and K networks. Certainly, in the bog-hill populations, ecotypic structure also reflects strong patterns of genetic divergence between hosts (Thrall et al., 2001 ) and pathogens (Laine et al., 2014 ; LG Barrett unpublished data). In addition, previous population genetic work demonstrates the maintenance of multiple clonal lineages of M. lini throughout the region where sampling for all of these studies was conducted (Barrett et al., 2008 ). Importantly, although there was some signature of population-level differentiation, different lineages were not confined to individual populations (Barrett et al., 2008 ). Therefore, it is possible that modularity in the LA and K networks may also reflect underlying genetic heterogeneities within and among the pathogen populations (and potentially hosts) from which the individuals comprising these networks were sampled. We suggest that for future studies, the generation of complementary population genetic data when examining resistance and infection networks could likely help reveal the proximate source of modular resistance structure. Partial resistance One of the novel results emerging from this study was the finding that broad resistance specificities were more likely to be conferred via partial resistance. Importantly, the expression of partial resistance was still dependent on host-pathogen genotype interactions (i.e., partial resistance is race-specific), consistent with previous studies suggesting that partial resistance is under GFG control (Burdon, 1994 ). This result has interesting implications for our understanding of the factors that drive the evolution of specificity and resistance strategies and can perhaps most obviously be explained via a trade-off between host range and the strength of resistance. Negative genetic correlations among resistance strategies have been demonstrated in some previous studies (e.g., between induced and constitutive resistance), suggesting that pleiotropy or the otherwise costly expression of linked resistance traits may be common (Agrawal et al., 2010 ; Rasmann et al., 2015 ). However, a mechanistic understanding of how such trade-offs arise is generally lacking. While we do not have any data that speaks directly to the mechanisms that may underlie a trade-off between specificity and the strength of the resistance response, a recent study using the interaction between Melampsora lini and Linum usitatissimum shows that the recognition of pathogen elicitors and the strength of the subsequent response are intimately related at the molecular level (Bernoux et al., submitted). This work suggests the potential existence of relatively simple (in a genetic sense) functional constraints associated with the breadth of pathogen elicitors that are recognized by resistance proteins."
} | 8,436 |
35145022 | PMC8851463 | pmc | 9,091 | {
"abstract": "Significance Mixotrophy is a ubiquitous nutritional strategy in marine ecosystems. Although our understanding of the distribution and abundance of mixotrophic plankton has improved significantly, the functional roles of mixotrophs are difficult to pinpoint, as mixotroph nutritional strategies are flexible and form a continuum between heterotrophy and phototrophy. We developed a machine learning–based method to assess the nutritional strategies of in situ planktonic populations based on metatranscriptomic profiles. We demonstrate that mixotrophic populations play varying functional roles along physicochemical gradients in the North Pacific Ocean, revealing a degree of physiological plasticity unique to aquatic mixotrophs. Our results highlight mechanisms that may dictate the flow of biogeochemical elements and ecology of the North Pacific Ocean, one of Earth's largest biogeographical provinces.",
"discussion": "Discussion Our study introduces a machine learning approach that leverages transcriptional profiles to predict the in situ trophic mode (heterotrophy, phototrophy, or mixotrophy) of protists in the natural environment. This method avoids potential artifacts associated with traditional, labor-intensive, incubation-based estimates and opens the door for large-scale studies of how carbon flows through specific members of microbial communities. The model was trained with transcriptomes derived from protists grown under controlled laboratory conditions and was challenged with a variety of validation transcriptomes. This approach identified a subset of gene family transcriptional profiles that, given sufficient transcriptome replicates, resulted in trophic mode predictions consistent with observed species-level nutritional strategies and broad phylogenetic patterns. The model accurately and consistently predicted the trophic mode of heterotrophic or phototrophic specialists, whereas mixotrophy predictions were often coupled with those of either heterotrophy or phototrophy, likely reflecting both overlapping and distinctive mixotrophic attributes. In addition, distinctive subsets of gene family clusters were identified that are associated with different trophic modes and can serve as starting points for understanding the molecular underpinnings of different metabolic strategies. A large public collection of microeukaryote transcriptomes (MMETSP) was used as the foundation for this work. However, the MMETSP was not designed for this particular application, and with more targeted molecular studies on mixotrophic organisms to include in the training set, the predictive power of this model will be improved. We predicted significant shifts in the trophic mode of natural communities under different environmental conditions both at a community level and for taxonomic bins corresponding to individual populations. Within the oligotrophic gyre, our model predicts that protists rely primarily on phagotrophy to acquire sufficient carbon and nitrogen for growth, a result consistent with OFT, a resource allocation model ( 24 ), and a global ecosystem model that incorporates mixotrophy ( 29 ). Observed biomass C:N ratios greater than the Redfield ratio and results from our on-deck transcriptomics experiments suggest that, under nitrogen limitation, mixotrophs prioritize phagotrophy to acquire nitrogen from their prey ( Fig. 5 ). The lower intracellular C:N and C:P ratios of marine prokaryotes compared with protists ( 45 ) makes phagotrophy even better suited to fulfill the nutrient requirements of cells in the nitrogen-limited waters of the gyre. Given the presumed optimality of heterotrophic metabolism, why do mixotrophic organisms with both phagotrophic and photosynthetic capabilities dominate protist communities within the oligotrophic gyre? We propose that, similar to what was observed in the amendment experiments, ephemeral injections of nitrogen into surface waters ( 52 , 53 ) rapidly shift the mixotrophic community to more energetically favorable photosynthesis-based growth, thus preventing their displacement by heterotrophic specialists. During the dominant low-nutrient conditions, the ability of mixotrophs to consume prey (including photosynthetic prey) to fulfill their carbon and nitrogen requirements prevents their displacement by protist phototrophic specialists. The variable conditions of the gyre thus appear to mitigate potential costs associated with maintaining the complicated cellular machinery required for both photosynthesis and phagotrophy. In the nutrient-rich North Pacific transition zone, a different scenario likely regulates the trophic modes of the protist community. As nutrient concentrations increase, our model predicts that the community shifts from a primary reliance on phagotrophy in the gyre toward an increasing reliance on phototrophy and mixotrophy by smaller protists (0.2 to 3 µm) and mixotrophy by larger protists (3 to 200 µm). A similar shift with latitude is predicted for individual species, with increased predictions of phototrophy for the small chlorophyte Micromonas and mixotrophy for the larger Chrysochromulina sp. The differing proportions of phototrophy predictions between size classes likely reflects the relationship between size-dependent uptake kinetics and nutrient availability, as smaller cells with a greater surface area:volume can more efficiently capitalize on lower dissolved nutrient concentrations ( 54 ). However, the MTE predicts rates of heterotrophic metabolism will decrease disproportionately with decreasing temperature ( 55 , 56 ). Why, then, do the larger mixotrophs continue to carry out phagotrophy in the cooler, nutrient-rich waters of the transition zone? Recent studies suggest that mixotrophic cells use the reductants generated via photosynthesis for organic carbon metabolism rather than carbon fixation ( 57 ). Moreover, the mixotrophs within our training set were distinguished by the transcription of gene families encoding distinctive varieties of carbohydrate active enzymes. Thus, we propose that, at the highest latitudes reached on the Gradients cruise, these larger mixotrophs fulfill their nutrient requirements via prey engulfment despite reduced sea surface temperatures with the necessary reductants for organic carbon metabolism supplemented through photosynthesis. We further propose that, in accordance with the results from the Edwards model ( 24 ), the reduced light levels we observed at the higher latitudes remain sufficient to have little impact on mixotroph abundance in surface waters. Deciphering the functional role of mixotrophic protists in the marine carbon cycle has been a longstanding challenge, made difficult by the many caveats involved with current methods. Here, we introduced a machine learning approach, made possible by the recent explosion of available transcriptomic data, and demonstrated that the model is skilled at inferring the trophic mode of natural populations based on the transcriptional patterns of select gene families. When we applied the model to metatranscriptomic data from the open ocean, the predicted patterns in trophic mode compare favorably with other results based on resource allocation models ( 24 ), global ecosystem models ( 29 ), the MTE ( 58 ), and OFT ( 49 ). By combining model predictions with the bioinformatic analysis of metatranscriptomes obtained during on-board incubation experiments, we were able to develop intuitive explanations for observed functional differences between organisms and highlight the potential drivers of mixotroph ecosystem function. As the ubiquity of mixotrophy in the marine environment becomes increasingly apparent, so does the need to incorporate mixotrophs into our understanding of the ocean’s carbon cycle and the microbial ecology of the marine water column. The future coupling of our machine learning technique with targeted field experiments and numerical modeling will enable detailed dissection of the role that mixotrophs play in marine ecosystem processes."
} | 2,004 |
28050162 | PMC5165138 | pmc | 9,092 | {
"abstract": "The archaeal ancestor scenario (AAS) for the origin of eukaryotes implies the emergence of a new kind of organism from the fusion of ancestral archaeal and bacterial cells. Equipped with this “chimeric” molecular arsenal, the resulting cell would gradually accumulate unique genes and develop the complex molecular machineries and cellular compartments that are hallmarks of modern eukaryotes. In this regard, proteins related to phagocytosis and cell movement should be present in the archaeal ancestor, thus identifying the recently described candidate archaeal phylum “Lokiarchaeota” as resembling a possible candidate ancestor of eukaryotes. Despite its appeal, AAS seems incompatible with the genomic, molecular, and biochemical differences that exist between Archaea and Eukarya. In particular, the distribution of conserved protein domain structures in the proteomes of cellular organisms and viruses appears hard to reconcile with the AAS. In addition, concerns related to taxon and character sampling, presupposing bacterial outgroups in phylogenies, and nonuniform effects of protein domain structure rearrangement and gain/loss in concatenated alignments of protein sequences cast further doubt on AAS-supporting phylogenies. Here, we evaluate AAS against the traditional “three-domain” world of cellular organisms and propose that the discovery of Lokiarchaeota could be better reconciled under the latter view, especially in light of several additional biological and technical considerations.",
"conclusion": "7. Conclusions Metagenomic explorations, development of single-cell sequencing technologies, and improvements in in silico reconstruction of (meta)genomes are yielding novel insights into our understanding of the evolutionary history of cellular organisms. The recent sequencing of Lokiarchaeota composite genomes and resulting phylogenetic analysis suggested an archaeal origin for the eukaryotic cell. The discovery has been widely publicized and the debate surrounding the origin of eukaryotes now considered by many to be settled. However, history inferred from protein structure data reveals a more global picture of the genetic composition of eukaryotic proteomes. Specifically, it takes into account the shared genes with Archaea, Bacteria, and viruses and challenges the purported eukaryotic genomic chimerism that is at the root of AAS models. While some interpret genomic chimerism in eukaryotes by invoking a fusion event at the root of eukaryote evolution, inferences redrawn from phylogenomic analyses performed after balanced taxon and character sampling, removal of fast-evolving species, and comparative analysis of protein structure distribution contradict that interpretation. Moreover, several biological and technical considerations are at odds with the proposed Lokiarchaeota-Eukarya phylogenetic affiliation and suggest that the 3D ToL may still be the more reasonable evolutionary scenario considering biological plausibility and support from molecular data.",
"introduction": "1. Introduction The discovery of the novel candidate archaeal phylum “Lokiarchaeota” from metagenomic samples taken from sites near Loki's Castle hydrothermal vents of the Arctic Ocean was recently reported [ 1 ]. There are two interesting aspects to this discovery: (i) several eukaryotic signature proteins (ESPs) related to membrane remodeling, cell division, and the cytoskeleton, previously thought to be either absent or rare in akaryotes (Archaea and Bacteria; sensu [ 2 ]), were detected in the composite Lokiarchaeota genomes (Loki 1, Loki 2, and Loki 3), and (ii) phylogenomic analyses of concatenated alignment of 36 conserved proteins revealed that eukaryotes and Lokiarchaeota grouped together within Archaea, suggesting an archaeal ancestor scenario (AAS) for the origin of eukaryotes [ 3 ]. The AAS thus favors a two-domain (2D) view of the tree of life (ToL) where eukaryotes emerge from within Archaea, specifically as sister group to the proposed TACKL (including Thaumarchaeota, Aigarchaeota, Crenarchaeota, Korarchaeota, and Lokiarchaeota) superphylum [ 4 , 5 ], after a likely merger of archaeal microbes (resembling Lokiarchaeota) and the mitochondrial ancestors [ 6 ]. AAS is fast becoming an accepted scenario to explain deep evolutionary history (e.g., [ 7 – 9 ]) and the origin of eukaryotic cells [ 10 , 11 ]. Except for some dissenting opinions [ 12 ], Lokiarchaeota is now commonly viewed as the “missing link” in the transition from “simple” to “complex” life [ 1 ]. However, several key differences in the membrane biology, biochemistry, and virospheres of Archaea and Eukarya seem at odds with AAS (see [ 13 ] for a recent review). Simultaneous ToL reconstructions from concatenated ribosomal proteins and the small-subunit ribosomal RNA (SSU rRNA) gene produced conflicting topologies with the former supporting the AAS while the latter recovering the “Woesian” three-domain (3D) ToL [ 14 ] of cellular diversification into domains Archaea, Bacteria, and Eukarya [ 15 ]. Because protein sequences are generally more conserved than nucleic acid sequences, SSU rRNA genes possess relatively lower number of informative sites and a higher rate of evolution compared to concatenated ribosomal protein sets. SSU rRNA genes are therefore likely more sensitive to known issues such as the notorious long-branch-attraction (LBA) artifact [ 16 ]. In turn, ribosomal proteins exhibit strong compositional biases among the cellular domains of life that need to be better understood [ 15 ]. While the study provided an “updated” view of the ToL incorporating hundreds of uncultivated representatives of archaeal and bacterial genera (the so-called “microbial dark matter” [ 17 ]) into ToL reconstructions, the authors remained indecisive in picking either the 2D (from concatenated ribosomal proteins) or the 3D (from SSU rRNA) ToL to explain the origin of eukaryotes beyond any doubt [ 15 ]. The AAS is also in conflict with several historical phylogenetic and phylogenomic frameworks such as phylogenies built from SSU rRNA sequences [ 14 ], single-gene alignments of ancient paralogous genes [ 18 , 19 ], gene content and order [ 20 , 21 ], concatenated gene [ 22 ] and protein domain [ 23 , 24 ] sets, and abundance combination and architecture of protein structural domains in modern genomes [ 23 , 25 , 26 ] that have consistently supported the 3D ToL despite disagreements on the location of the root of the ToL and the fact that most generated trees are unrooted [ 27 – 30 ]. It has been argued however that the use of “advanced” models of sequence evolution with relaxed assumptions of homogenous amino acid compositions of gene products across sites and branches is necessary to recover the origin of Eukarya from within Archaea (see [ 31 ] for a recent review). However, the presence of distant outgroups (e.g., bacterial ribosomal proteins that are quite divergent from archaeal-eukaryotic counterparts but are used to root the ToLs) and fast-evolving species (e.g., Nanoarchaeota [ 32 ] and Methanopyrus kandleri [ 33 ]) in datasets can make even these sophisticated methods prone to LBA, as shown by recent simulations [ 29 ] (see also [ 34 ]). Moreover, a concatenated (i.e., supermatrix) approach to phylogenetics, as applied by Spang et al. [ 1 ] to support AAS, could be problematic especially when member genes have independent evolutionary histories. Simulations have shown that concatenated gene sets can produce aberrant trees with high bootstrap (BS) support [ 35 ]. The approach is also susceptible to heterotachy (i.e., unequal evolutionary rates among genes in a concatenated set) [ 35 , 36 ], which can complicate inferring deep evolutionary relationships and can introduce distortions to interdomain calculations, among other issues (see Section 5 ). In light of these considerations, here we examine the evidence supporting the 2D scenario for the diversification of cellular life, perform taxa and character manipulations to reanalyze the dataset of Spang et al. [ 1 ] that supported the Lokiarchaeota-Eukarya sisterhood, and consider several biological and technical issues that weaken the 2D in favor of the 3D ToL."
} | 2,041 |
35519072 | PMC9056752 | pmc | 9,093 | {
"abstract": "Inspired by biology, underwater self-healing polymer composites with damage-healing visible agents were successfully designed and prepared. The healing agents, same as epoxy resin matrices, were encapsulated and embedded into a matrix that contained fluorescent latent curing agents. The results of investigation on healing properties revealed that the fluorescent latent curing agents and the microcapsules in the matrix play two roles. First, the matrix could be self-healed via a crosslinking reaction between the amine group and epoxy resin, in which the amine group could be released from the fluorescent latent curing agents (FLCAs) after exposure to water. Second, the fluorescent dyes released under water could indicate the scratches and healing area visually. Embedding 15 mass% microcapsules and 6 mass% FLCAs in self-healing materials yielded a healing efficiency of 85.6% and the most efficient fluorescence detection. Self-healing materials can be repaired underwater and they show the location of damage, which is of great significance in applications such as water conservation engineering, environmental treatment engineering, ship engineering and ocean engineering.",
"conclusion": "4 Conclusions In this work, the microcapsule-type self-healing system successfully used changes in visible colours and fluorescence colours to detect cracks and healed regions. In addition, the material can self-repair from 25 °C to 100 °C due to the hydrolysis of FLCAs into a fluorescent agent and a curing agent. The optimum content of epoxy resin microcapsules was determined to be 15%, and the optimal FLCA content was 6%. When the material was damaged by an external force, it could be repaired in water at 60 °C for 4 h, and the healing efficiency reached 85.6%. Additionally, the fluorescence colour changed to yellow after repairing the crack region, and the degree of self-repair could be judged by the change in colour.",
"introduction": "1 Introduction The property of self-healing plays an important role in improving the safety of structures and extending their service life, especially in applications such as underground pipeline transportation and marine engineering. 1,2 Moreover, cracks tend to occur in dark environments and are difficult to view. Therefore, it is crucial to detect and repair any damage in advance to protect materials. 3–5 Inspired by the self-healing function of natural organisms, researchers have designed materials that can retain functionalities and restore their structure automatically after damage to improve the service and safety life of materials. 6,7 Thus, self-healing materials have attracted great attention and resulted in a certain degree of progress in the past decade. 8–10 Considering the long-term use of materials in high humidity environment or underwater is inevitable. Therefore, the stability of self-healing materials underwater has been recognized as necessary. 11–13 Among external self-healing materials, microcapsule-embedded self-healing materials are much more popular for investigation due to their outstanding mechanical properties and simple operation. 14 White et al. 15 first applied microcapsule technology to material design and developed an intelligent self-healing material, which contained embedded encapsulated dicyclopentadiene (DCPD) and Grubbs catalyst in an epoxy resin matrix. In addition, fracture experiments yielded as much as 75% recovery in toughness. Since then, researchers have developed various microcapsule-type self-healing materials such as epoxy resin, 16 polydimethylsiloxane 17 glycidyl methacrylate 18,19 and tung oil. 20 Moreover, most self-healing materials are repaired usually when triggered by force, 21 thermal initiation, 16,22 photoinitiation 23–25 or pH. 26 Currently, reported underwater self-healing materials are mostly based on boronic acid derivatives 27,28 and catechol groups. 29,30 Relatively, there are few studies on the ability of microcapsule-type self-healing materials to self-heal underwater. In principle, amine-based curing agents or metal catalysts easily dissolve or get deactivated in the presence of water, and is a huge challenge that self-healing materials automatically heal underwater. Li et al. 31 microencapsulated isophorone diisocyanate (IPDI) and dispersed it in a coating that can bond the crack by immersion in water. Cho et al. 32 presented a self-healing material system based on polydimethylsiloxane to repair underwater. Among the materials, the fracture toughness repair rate is not more than 46% due to the difference between the core material and the matrix material, and the fracture toughness repair rate of healing is not more than 31% in water. In general, some cracks and the extent of self-repair are invisible to the naked eye during the self-healing process. Therefore, it is meaningful that the healing process of polymer materials be detected by observing the colour and fluorescent changes in the damaged area. Many researchers have developed various methods to detect the damage. Hamilton et al. 33 created a visual system to detect any mechanical damage by embedding two different colours into the self-healing materials. When the material was damaged, the two dyes mixed together to create a new colour. The average recovery of fracture toughness was 86% of the virgin sample fracture toughness. Song et al. 34 applied aggregation-induced emission (AIE), which is a unique phenomenon in which the emission intensity of a solid-state fluorescent dye is higher than that of the solution state, to detect the region of a coating undergoing self-healing. The degree of repair was determined by the degree of crack filling, and the damage was indicated by AIE. However, in many applications, simply filling or sealing crack does not meet the requirements of the overall material performance, and the recovery of fracture toughness strength is also crucial. In contrast to previous microcapsule-based materials, the damage indication we report here does not require any additional chromogenic agent. Instead, it is achieved by the fluorescent latent curing agents (FLCAs) in the matrix, which has both the damage indication function and the repair function. In this work, a new kind of microcapsule-based self-healing material was designed. The principle of damage indication and self-healing is illustrated in Fig. 1 . We used the same repair agent as the substrate to improve the repair rate. In order to avoid the failure of amine curing agents in humid environments, we blended FLCAs with the matrix. The latent curing agent was a Schiff base. In the presence of water, it hydrolyzes and releases amine groups to enable the cured epoxy repair agent self-repair. At the same time, a light-emitting substance with an aldehyde group is released to indicate the damage and damage repair. Therefore, the material has the ability to self-heal underwater and visualize the damage. In addition, it can not only repair the damaged region in water at room temperature but also detect the damage and the state of healing of the self-healing materials through two highly distinguishable colours under UV irradiation, which are visible to the naked eye. The influence of the addition of microcapsules and fluorescent latent curing agent on the self-healing properties and mechanical performance was also studied. Accordingly, self-healing materials that can be repaired underwater and have indications of damage are important application prospects for hydraulic engineering, environmental treatment engineering, underground concrete facilities, pipeline projects, and marine engineering. Fig. 1 Schematic of a self-healing material embedded with fluorescent latent curing agents and microcapsules.",
"discussion": "3 Results and discussion 3.1. Characterization of the latent curing agent As shown in Fig. 3a , the FLCA was the latent curing agent that can decompose into a fluorescent dye and a curing agent (MXDA) by reacting with water. The formation of a fluorescent dye can cause a colour change in the material and produce a fluorescence colour change under ultraviolet light ( λ = 365 nm). The functional groups of FLCAs and the fluorescent dyes were determined by infrared spectroscopy, as shown in Fig. 3b . For the FLCA (green curve), the wide absorption peaks of the curves centered at 3352 cm −1 corresponded to the O–H stretching vibration. The weak absorption peak at 2954 cm −1 was attributed to the C–H tensile vibration, the strong absorption peak at 1658 cm −1 was attributed to the deformation vibration of the C \n \n\n<svg xmlns=\"http://www.w3.org/2000/svg\" version=\"1.0\" width=\"13.200000pt\" height=\"16.000000pt\" viewBox=\"0 0 13.200000 16.000000\" preserveAspectRatio=\"xMidYMid meet\"><metadata>\nCreated by potrace 1.16, written by Peter Selinger 2001-2019\n</metadata><g transform=\"translate(1.000000,15.000000) scale(0.017500,-0.017500)\" fill=\"currentColor\" stroke=\"none\"><path d=\"M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z\"/></g></svg>\n\n N group, and the absorption peak at 1581 cm −1 was assigned to the C C stretching vibration. For the fluorescent dye (yellow curve), two distinct absorption peaks at 3352 cm −1 and 1710 cm −1 were ascribed to the OH and C O tensile vibration of the fluorescent dye. Fig. 3 (a) Structure and reversible reaction of the FLCA in the presence of water producing fluorescent dyes and MDXA. (b) FT-IR spectra of the FLCA and fluorescent dyes. (c) UV-vis spectrum of the FLCA (1 × 10 −5 mol L −1 ) in THF–H 2 O (volume ratio 4 : 1) at different time points and the pictures under ultraviolet light ( λ = 365 nm). The UV-vis spectra of FLCAs at different reaction times with water are shown in Fig. 3c . As the response time increased, the absorption peak of the FLCA at 386 nm decreased, but the absorption peak at 296 nm gradually increased, indicating that the FLCA underwent a hydrolysis reaction under the condition of water. Moreover, it was revealed that the FLCA overall emits blue fluorescence under ultraviolet light before the reaction. After contact with water, the fluorescent dye concentration increased and the fluorescence colour changed from blue to green and eventually turned yellow over time. When the material is damaged, the crack extension causes the microcapsules to rupture. The FLCA in the matrix contacts the epoxy resin flowing out of the capsule. Under the action of water, the FLCA is hydrolysed to produce a fluorescent dye and a curing agent to achieve damage indication and damage. Therefore, to test whether the colour change of FLCAs in the epoxy resin is obvious, the epoxy resin and FLCA were mixed in a mass ratio of 2 : 1 and then dropped into a plastic crucible. As illustrated in ESI Fig. 6, † the colour change could be clearly observed in water with changes in time and temperature. The colour of the mixture changed from green to yellow, and the fluorescence colour changed from blue to yellow. 3.2. Preparation of microcapsules \n Fig. 4a illustrates the structure of the microcapsule; epoxy resin was the core material and polyurea formaldehyde (PUF) was the shell material. FTIR was used to qualitatively characterize whether the epoxy polymer was successfully encapsulated in the microcapsules. Fig. 4b displays the FTIR spectra of the pristine epoxy resin, PUF-epoxy microcapsules and neat PU shell material. Specifically, the wide absorption peak exhibited at approximately 3100–3700 cm −1 corresponded to the N–H and O–H stretching vibration. The absorption peaks at 1669 cm −1 and 1572 cm −1 were attributed to the C O and C–N stretching vibrations, respectively. The above-mentioned peaks confirmed that the PUF shell was successfully constructed by in situ polymerization. For the core material, the characteristic absorption peak at 3013 cm −1 denoted the stretching vibration of CH 2 , and the appearance of the absorption peaks at 2923 cm −1 indicated the stretching vibration of –CH 3 . The absorption peak observed at 1239 cm −1 denoted the stretching vibration of C–O–C. The characteristic absorption peaks observed at 915 cm −1 and 831 cm −1 were attributed to the epoxy group. Finally, all of the above characteristic peaks were matched in the spectrum of the microcapsule, which strongly verified that epoxy as a healing agent was successfully encapsulated in the PUF shell. As Fig. 4c shows, it was clearly observed that the shape of the microcapsules was a regular sphere, while the surface was rough, indicating that the shell of the microcapsules was produced by in situ deposition of PUF particles between the water and oil phase interfaces. The particle size distribution of the microcapsules is shown in Fig. 4d . The size of the microcapsules followed a normal distribution, and the average diameter was approximately 140 μm. In summary, these results indicated that the spherical microcapsules were synthesized successfully and did not leak out through the capsule wall. Fig. 4 (a) Structure of the microcapsule. (b) FTIR spectra of the PUF shell, core and microcapsule. (c) SEM micrograph of the microcapsule. (d) Histogram of the microcapsules' size distribution. (e) TGA curves of the PUF shell, epoxy resin and microcapsules. The thermal stability of microcapsules is critical to their preservation and practical application. Consequently, the thermal performance of the PUF shell, epoxy resin and microcapsules was characterized by TGA. As shown in Fig. 4e , the PUF shell had a 5 wt% weight loss from 120 °C to 236 °C due to the decomposition of free formaldehyde absorbed on the surface of the microcapsules. In addition, the PUF shell began to decompose at 220 °C and completely decomposed at 600 °C. For the epoxy resin, almost no mass loss was observed at temperatures below 298 °C. Through TG analysis of microcapsules, it was proven that the microcapsules have good thermal stability. In addition, the mass of the shell and solvent was 10.8%, the remaining mass was 11.0%, and the encapsulation percentage was 78.2% by calculation. 3.3. Optimization of the ratio of microcapsules and FLCAs in the self-healing material To investigate the optimal self-healing performance, different mass percentages of FLCAs (2 wt%, 4 wt%, 6 wt%, and 8 wt%) and microcapsules (5 wt%, 10 wt%, 15 wt%, and 20 wt%) were dispersed in the epoxy matrix. The microcapsules and FLCAs were tested by lap-shear tests, and the average shear strength was 9.80 MPa (ESI Table 2 † ) in normal water for seven days. It was proved that the combination of FLCAs and epoxy resin microcapsule made the material self-healing underwater. As shown in Fig. 5a , when the content of the epoxy resin microcapsules was 5 wt%, the repair efficiency was much lower within the whole addition range of the latent curing agent. This is precisely because the epoxy resin released from the microcapsules could not sufficiently fill the crack, and the resin solidified to form a polymer with a small molecular weight. Furthermore, the self-healing efficiency increased with the increase in epoxy resin microcapsule content, suggesting that the issued epoxy resin might be adequately cured with FLCAs and led to a higher repair efficiency. In the case of 15 wt% microcapsules, the self-repair efficiency with a content of 6 wt% was almost the same as that with 8 wt% FLCA, indicating that too much latent hardener did not ensure the best degree of crosslinking. While the amount of microcapsules reached 20 wt%, the growth rate of the self-healing efficiency was very low, suggesting that 15 wt% microcapsules was able to effectively repair cracks and that the excessive content of microcapsules might lead to additional defect generation. In summary, considering that the content of epoxy resin microcapsules and FLCAs would dramatically affect the mechanical properties of the material, the optimal ratio was confirmed as 15% epoxy resin microcapsules and 6% FLCAs. Fig. 5 (a) Influence of the content of FLCAs on the healing efficiency at different contents of FLCAs. (b) Load-displacement curves of the specimen with 15 wt% epoxy resin microcapsules and 6 wt% FLCAs. 3.4. Healing properties of the self-healing material at different temperatures The healing efficiency of samples with 6 wt% latent curing agents and 15 wt% microcapsules repaired at different temperatures under water was discussed. As shown in Fig. 6 , the materials could self-repair from 25 °C to 100 °C. As expected, the self-healing process at lower temperatures required more time, and the self-repair efficiency was accordingly reduced due to the high viscosity of the epoxy resin and the low solubility of FLCAs. With the increase in temperature, the solubility of FLCAs in the core material increased and accelerated the decomposition rate of FLCAs, thereby improving the self-repair efficiency. However, further increasing the temperature beyond 60 °C caused gradual evaporation of ethyl acetate in the microcapsule and made FLCAs difficult to dissolve. Moreover, it decreased the amount of FLCAs in contact with the epoxy polymer, thereby reducing the degree of curing and the self-repair efficiency. Over sufficient time, the healing efficiency at 25 °C and 100 °C is lower than that at 60 °C, mainly due to the poor fluidity of the epoxy resin in the microcapsules at normal temperatures, resulting in defects in the healing area. According to Fig. 7 , the healing efficiency was 83.5% ± 9.4 after healing at 60 °C for 4 h, which was similar to the healing efficiency (85.6% ± 10.2%) of the material healed at 60 °C, confirming that the epoxy resin in the microcapsules healed at 60 °C for 4 h was substantially cured. Fig. 6 Healing efficiency of the samples healed at different temperatures. Fig. 7 Healing efficiency of the samples healed at 60 °C for different time periods in water. 3.5. Detection of different fluorescence colours and visible colours from cracked and healed regions In general, the ability to detect a change in fluorescence colours with regard to damage and healing in the self-healing field could allow the monitoring of surface conditions and prevent material damage. The epoxy sample with 15% microcapsules and 8% FLCAs embedded was fractured using a universal testing machine (Z100, Zwick/Roell) and imaged using a digital camera (D7000, NIKON), an optical microscope and a scanning electron microscope (ISM-7610F, JEOL). As shown in Fig. 8 , the crack could be clearly observed in the image before and after the repair of the self-healing material, and the width was approximately 20 μm. In Fig. 8a-1 and a-2 , the colour of the cracked area before healing under visible light was green. After repairing in water at 60 °C, the epoxy filled the crack, and the damaged region turned yellow. To confirm that the scratching and healing region yielded different fluorescence colours, the cracks were observed under UV irradiation. Before healing, the fluorescence colour of the cracked area did not change in Fig. 8b-1 . After healing, the crack region produced obvious yellow fluorescence in Fig. 8b-2 . As shown in Fig. 8c-1 and c-2 , the SEM image shows that the crack healing area was completely healed and the interface was smooth. Therefore, not only the designed material can detect the position of crack generation under ultraviolet light, but the change in the crack can also be observed under visible light. Fig. 8 (a-1) OM image of the material before healing. (a-2) OM image of the material after repair. (b-1) OM image of material under UV light before healing ( λ = 365 nm). (b-2) OM image of the material under UV light ( λ = 365 nm) after healing. (c-1) SEM image of the material before healing. (c-2) SEM image of the material after healing."
} | 4,944 |
26943622 | PMC5029178 | pmc | 9,096 | {
"abstract": "Predicting the dynamics of community composition and functional attributes responding to environmental changes is an essential goal in community ecology but remains a major challenge, particularly in microbial ecology. Here, by targeting a model system with low species richness, we explore the spatial distribution of taxonomic and functional structure of 40 acid mine drainage (AMD) microbial communities across Southeast China profiled by 16S ribosomal RNA pyrosequencing and a comprehensive microarray (GeoChip). Similar environmentally dependent patterns of dominant microbial lineages and key functional genes were observed regardless of the large-scale geographical isolation. Functional and phylogenetic β-diversities were significantly correlated, whereas functional metabolic potentials were strongly influenced by environmental conditions and community taxonomic structure. Using advanced modeling approaches based on artificial neural networks, we successfully predicted the taxonomic and functional dynamics with significantly higher prediction accuracies of metabolic potentials (average Bray–Curtis similarity 87.8) as compared with relative microbial abundances (similarity 66.8), implying that natural AMD microbial assemblages may be better predicted at the functional genes level rather than at taxonomic level. Furthermore, relative metabolic potentials of genes involved in many key ecological functions (for example, nitrogen and phosphate utilization, metals resistance and stress response) were extrapolated to increase under more acidic and metal-rich conditions, indicating a critical strategy of stress adaptation in these extraordinary communities. Collectively, our findings indicate that natural selection rather than geographic distance has a more crucial role in shaping the taxonomic and functional patterns of AMD microbial community that readily predicted by modeling methods and suggest that the model-based approach is essential to better understand natural acidophilic microbial communities.",
"introduction": "Introduction Given the critical importance of species biogeography for biological conservation and climate change management, the development and application of statistic models for predicting the species distribution are an essential issue in community ecology ( Elith and Leathwick, 2009 ). In the past two decades, the number of studies involved in species distribution models of plants and animals has increased markedly, providing ecological insights into the assessment of impacts and consequences of environmental changes on natural communities and ecosystems ( Guisan and Thuiller, 2005 ; Guisan et al. , 2006 ; Austin, 2007 ; Pearman et al. , 2008 ). Microorganisms are arguably the most diverse and abundant group of organisms on Earth ( Fierer and Jackson, 2006 ), driving the bulk of biogeochemical cycles on the planet and influencing the functioning of virtually all ecosystems. During the last few years, a large number of phylogeny/taxonomy-based surveys have focused on the spatio-temporal dynamics and biogeographic patterns of microbial communities, revealing environmental variations (that is, contemporary environmental conditions) ( Lozupone and Knight, 2007 ; Lauber et al. , 2009 ) or spatial isolation (that is, historical events and disturbances) ( Whitaker et al. , 2003 ; Martiny et al. , 2011 ) are the major factors shaping the large-scale ecological breadth of microbes. However, these studies are mainly limited to descriptive approaches rather than predictive model-based analyses ( Gonzalez et al. , 2012 ). With the recent development of high-throughput molecular technologies and advanced bioinformatics tools, there have been increasing attempts to predict the biogeographic distributions of microbes across diverse ecosystems ( King et al. , 2010 ; Larsen et al. , 2012 ; Bokulich et al. , 2013 ; Ladau et al. , 2013 ; Szabo et al. , 2013 ). These pioneering studies demonstrate that it is now possible to obtain more comprehensive understanding of microbial communities and their connections with climate change and biogeochemical cycling using vastly increased data sets. Although the novel predictive strategies based on phylogenetic/taxonomic profiles have significantly advanced the study of microbial communities from the descriptive nature to the predictive science, the underlying mechanisms of how changes in these spatio-temporal variations of biogeographic pattern affect the processes of ecosystem functioning remain largely unknown, especially in a predictive scheme. As a broad range of functional variation may occur among closely related organisms, taxonomic distributions are assumed to be ambiguous in assessing the response of microbial communities to environmental changes ( Green et al. , 2008 ) and may be of little value in predicting the functional dynamics in ecosystems. Thus, the functional traits (for example, gene content and metabolic potential), which determine the habitat-related attributes of a specific microbial species, have recently received a great deal of attention. Recent studies have highlighted the critical importance of trait-based approaches for studying microbial biogeography ( Green et al. , 2008 ; Raes et al. , 2011 ; Barberán et al. , 2014 ). The investigation of functional traits distribution across spatial/temporal scales and along geochemical gradients will help elucidate how natural communities and their ecological functions respond to environmental changes ( Green et al. , 2008 ; Bryant et al. , 2012 ; Fierer et al. , 2012 ) and subsequently identify the interaction of ecological processes affecting biogeographic patterns ( Hanson et al. , 2012 ). Consequently, by combining the advanced modeling strategy and trait-based approaches, never has there been a greater opportunity for investigating the dynamics of functional community structure in space and time. Community assembly is previously suggested to be deterministic in trait-based functional structure but historically contingent in taxonomic composition, indicating that environmental conditions would determine the types of ecological niches available for specific functional groups, whereas species compositions with similar physiological fitness are stochastically influenced by the history ( Fukami et al. , 2005 ). Accordingly, the responses of functional traits specifically associated with the habitat-related attributes of microbial taxa may be more deterministic to environmental changes compared with those of taxonomic community composition. Thus, we hypothesized that natural microbial assemblages may be better predicted at the functional genes level rather than species. Here, we use the acid mine drainage (AMD) model system to test this hypothesis. These acidic, metal-rich drainages arise largely from the microbially mediated oxidative dissolution of sulfide minerals (for example, pyrite) and represent a major environmental problem worldwide ( Baker and Banfield, 2003 ; Johnson and Hallberg, 2003 ). The microbial and geochemical simplicity of AMD systems makes them ideal targets for a quantitative, genomic-based test of our assumption. We applied a comprehensive functional gene array (GeoChip 4.0) ( Tu et al., 2014 ) and a recently developed modeling approach ( Larsen et al. , 2012 ) to 40 environmental samples that were previously collected from diverse AMD sites across Southeast China with detailed microbial community composition and associated geochemical properties ( Kuang et al. , 2013 ). Our results demonstrated that the patterns of taxonomic and functional community structure were environmentally dependent and readily predictable with significant higher prediction accuracies of metabolic potentials compared with relative microbial abundances. These findings provide ecological important insights into the adaptive strategies of how these microorganisms can survive and thrive in the extreme AMD environment.",
"discussion": "Discussion Similar to the large-scale taxonomic composition patterns of AMD communites ( Kuang et al. , 2013 ), the spatial variation of functional community structure resolved by GeoChip was remarkably environment-dependent. Likewise, a previous study has demonstrated significant correlations between proteogenomics and geochemical and physical attributes in shaping communities of AMD biofilm ( Mueller et al. , 2010 ). These findings highlight the importance of natural selection in this extreme environment. Such severe environmental filtering may lead to a smaller available pool of species/genes that can persist under the harsh conditions, making their structure more niche-assembled ( Chase, 2007 ). These patterns of taxonomic and functional biogeography shaped by the measurable environmental variables rather than geographic distance are highly consistent with the assumption of the modeling method ( Larsen et al. , 2012 ), enabling a successful application of the predictive strategy in this study. According to the ANN-based predictive framework, we demonstrated that functional traits were more predictable by environmental variations and provided more useful explanation than taxonomic diversity based on phylogenetic markers in assessing the relationship between microbial communities and ecological processes. Although several recent metagenomic studies have revealed a significant correlation between phylogenetic diversity and functional diversity ( Bryant et al. , 2012 ; Fierer et al. , 2012 ), specific functional traits and microbial species may not always have a the definite relationship, as functional interchange may occur across different taxa ( Green et al. , 2008 ), resulting in the conspicuous decoupling of ecological attributes from phylogeny ( Raes et al. , 2011 ; Barberán et al. , 2014 ). In supporting this, a recent study has documented that specific functions could be widely detected across a variety of taxa or phylogenetic groups ( Burke et al. , 2011 ). Importantly, previous research has revealed that lateral gene transfer is prevailing mechanisms for AMD microbes to rapidly acquire and possess new genes involved in survival and habitat-specific functions (for example, heavy metals resistance) ( Baker and Banfield, 2003 ; Tyson et al. , 2004 ). Indeed, it was recently suggested that functional traits are valuable ecological markers to understand bacterial community assembly ( Barberán et al. , 2012 ) and to explain shifts in microbial community composition across environmental gradients ( Edwards et al. , 2013 ). As such, it is reasonable to obtain more accurate prediction of the metabolic potentials of key functional genes in response to environmental change, as these specific functional capabilities may directly impact how microbial communities interact with their environments. Application of the predictive models allowed an accurate estimation of the dynamics of taxonomic and functional community structure along a pH range typically reported for AMD environments. As nitrogen resources are very limited in natural AMD systems ( Baker and Banfield, 2003 ), their bioavailability and biogeochemical processes are vital to the acidophilic communities and essential in understanding of how these extraordinary assemblages respond and adapt to the harsh conditions. Diverse genes involved in nitrogen cycling were detected and predicted to show clear patterns of relative metabolic potentials ( Figure 5a ). In a recent transcriptional analysis of several AMD communities ( Chen et al. , 2015 ), nitrogen-fixation transcripts such as nifH were commonly found and associated with Leptospirillum ferrooxidans , Leptospirillum ferrodiazotrophum , Acidithiobacillus ferrivorans and Acidithiobacillus sp. GGI-221. In our predictive models, the relative metabolic potential of nifH exhibited a notable increase with the decrease of solution pH, which was possibly attributed to the dominance of Leptospirillum spp. and Acidithiobacillus spp. under more acidic conditions ( Figures 4 and 5a ). In addition, an increase of relative metabolic potential of glutamate dehydrogenase ( gdh ) mostly derived from Thermoplasma was predicted, indicating an alternative strategy of ammonium acquisition from organic N conducted by this dominant population in low pH conditions ( Ruepp et al. , 2000 ). Similar patterns of increased predictive potential activities were also found for genes encoding the enzymes for nitrite utilization (for example, nirA , nirB and nrfA ). This accumulation of ammonia/ammonium might indicate a high requirement of nitrogen resources for microbial protein synthesis, which further supported by the higher relative metabolic potential of glutamine synthetase ( glnA ) that associated with incorporation of ammonium into glutamine ( Leigh and Dodsworth, 2007 ) ( Supplementary Figure S5b ). These findings indicated the nitrogen-limited adaptation and the prosperity of these extremely acidophilic populations. Phosphate represents another key nutrient limited in the extreme AMD environment. With the increase of acidity, the high amounts of Fe 3+ and Al 3+ ions might favor phosphate precipitation ( Moreno-Paz et al. , 2010 ), resulting in further phosphate starvation. As predicted in our models, multiple strategies of phosphate uptake and utilization were used by enhancing the relative metabolic potentials of genes involved in polyphosphate metabolism ( ppk and ppx ) ( Vera et al. , 2003 ), phosphate regulon ( Pho ) ( Lamarche et al. , 2008 ) and specific phosphate ABC transporters ( pstSCAB ) ( Parro et al. , 2007 ) ( Figure 5b ), reflecting a positive response to the phosphate deficiency in the AMD systems. Various protective mechanisms were identified to compensate for the deleterious effects of the extreme acidity. Diverse genes encoding proteins of heavy metal resistance and cation efflux systems were widely detected, and their relative metabolic potentials were predicted to be remarkably higher in lower pH conditions ( Figure 5c ). This was possibly attributed to and stimulated by the increased concentrations of dissolved heavy metals. Likewise, the relative metabolic potentials of genes involved in the defense against oxidative and osmotic stress (for example, oxyR , proV and ABC transporters) were predicted to be highly increased as well ( Supplementary Figures S5b and c ). The membrane-binding ABC transporters are identified to function as pumps to exclude toxins and drugs from the cell ( Higgins, 2001 ), and these transport systems such as potassium transporters ( kdpBAC ) are suggested to be an effective strategy to maintain pH homeostasis and cellular osmotic pressure ( Baker-Austin and Dopson, 2007 ; Parro et al. , 2007 ; Moreno-Paz et al. , 2010 ). Collectively, these stress-resistant mechanisms may provide the populated microbes important strategy for surviving and thriving in the extreme environment. Our predictive models also revealed some clues about microbial interaction in the AMD communities. It was suggested that more extreme conditions are less conducive to microbial growth, making survival capacities more important than the abilities for enhancing microbial competition ( Fierer et al. , 2012 ). However, the indigenous AMD populations are assumed to be well adapted to the extremely acidic conditions, whereas the decrease of energy sources such as pyrite and ferrous iron in lower pH environments might largely increase the importance of competition between sulfur/iron oxidizers. This assumption is partly supported by the higher relative metabolic potentials of genes involved in antibiotic resistance in our predictive models ( Supplementary Figure S5d ), as elevated microbial competition would select for increased antibiotic resistance ( Fierer et al. , 2012 ). In summary, our analyses of the dynamics of taxonomic and functional community structure in response to the environmental changes by modeling strategy represents a crucial step toward a predictive model-based understanding of the distribution mechanisms of acidophilic microorganisms in the extreme AMD system. Our results showed that the environmentally dependent patterns of taxonomy and traits (functional genes) are readily predictable, whereas the notable enhancement of relative metabolic potentials of a suite of key functional genes under more acidic and metal-rich conditions may reflect an important adaptation strategy of these extraordinary assemblages. More importantly, we demonstrated that natural microbial communities in the AMD model system are better predicted at the functional genes level rather than species, at least by the set of functional genes considered in the current study. It should be pointed out that although the microbial taxonomic composition was resolved by pyrosequencing of universal 16S ribosomal RNA gene, the functional structure of AMD assemblages was profiled by GeoChip, which is a high-throughput microarray-based genomic technology designed for detecting ‘known' genes specifically involved in biogeochemical processes and stress toleration and adaptation ( Zhou et al. , 2015 ). Alternatively, metagenomic sequencing represents another way to study the microbial community and its traits by simultaneously generating information on functional and taxonomic data sets. Such approaches could be adopted to verify our findings for its universality in diverse habitats."
} | 4,366 |
32477317 | PMC7240109 | pmc | 9,097 | {
"abstract": "Food spoilage by certain species of bacteria is reported to be regulated by quorum sensing (QS). Acinetobacter johnsonii and Pseudomonas fluorescens , the major specific spoilage organisms, are found to be limited in their QS and co-culture interactions. The aim of this study was to determine how QS-regulated proteins affect the spoilage potential of co-cultured A. johnsonii and P. fluorescens obtained from spoiled bigeye tuna ( Thunnus obesus ) using a proteomics approach. The A. johnsonii, P. fluorescens , and their co-culture tested the N -acyl-homoserine lactone (AHL) activities using reporter Chromobacterium violaceum CV026 and LC-MS/MS in qualitative and quantitative approaches, respectively. These latter showed that, of the 470 proteins and 444 proteins in A. johnsonii (A) and P. fluorescens (P), respectively, 80 were significantly up-regulated and 97 were significantly down-regulated in A vs. AP, whereas 90 were up-regulated and 65 were down-regulated in P vs. AP. The differentially expressed proteins included the AI-2E family transporter OS, 50S ribosomal protein L3, thioredoxin reductase OS, cysteine synthase CysM OS, DNA-binding response regulator, and amino acid ABC transporter ATPase OS. The cellular process (GO:0009987), metabolic process (GO:0008152), and single-organism process (GO:0044699) were classified into the gene ontology (GO) term. In addition, energy production and conversion, amino acid transport and metabolism, translation, ribosomal structure and biogenesis, post-translational modification, protein turnover, and chaperones were distributed into the clusters of orthologous groups of proteins (COG) terms. The KEGG pathways revealed that 84 and 77 differentially expressed proteins were divided into 20 KEGG pathways in A vs. AP and P vs. AP, respectively, and amino acid metabolism, carbohydrate metabolism, energy metabolism, and translation were significantly enriched. Proteins that correlated with the spoilage-related metabolic pathways, including thioredoxin reductase OS, cysteine synthase OS, and pyridoxal phosphate-dependent enzyme family protein OS, were identified. AI-2E family transporter OS and LuxR family transcriptional regulator OS were identified that related to the QS system. These findings provide a differential proteomic profile of co-culture in A. johnsonii and P. fluorescens , and have potential applications in QS and the regulation of spoilage potential.",
"conclusion": "Conclusion This study was carried out in order to explore cultures of A. johnsonii and P. fluorescens and compare them with their co-cultured state for QS and spoilage potential by means of their proteomic profiles. The results show that the products of AHL production (C4-HSL, C6-HSL, C8-HSL), biofilm production, protease activity, and spoilage potential were at a higher level in the co-culture than those of A. johnsonii and P. fluorescens single cultures alone. The proteomic results revealed that there were differences in the proteins involved in the metabolism of amino acids, carbohydrates, energy, and translation. The differentially expressed proteins that were spoilage-related included thioredoxin reductase OS, cysteine synthase OS, and pyridoxal phosphate-dependent enzyme family protein OS, as well as specific QS system proteins of the AI-2E family transporter OS and the LuxR family transcriptional regulator OS, which could be used as biomarkers in A. johnsonii , P. fluorescens , and their co-cultured state. These results may provide an understanding of how the QS and spoilage mechanisms of A. johnsonii , P. fluorescens , and their co-cultures can be regulated in future.",
"introduction": "Introduction Bigeye tuna ( Thunnus obesus ) is a highly sought after fish species used to prepare sashimi in many countries around the world ( Wang and Xie, 2019 ). However, bigeye tuna is easily spoiled by specific spoilage organisms during refrigerated storage, which leads to a reduced shelf life ( Sun et al., 2013 ; Silbande et al., 2016 ). Currently, more attention has been paid to convenient methods of refrigeration for storing bigeye tuna than to how spoilage bacteria in refrigerated tuna develop through their interactions with each other ( Wang et al., 2017 ). The microbial spoilage of aquatic products is correlated mainly with Gram-negative bacteria, including Acinetobacter spp., Shewanella spp., Pseudomonas spp., Aeromonas spp., lactic acid bacteria, and the Enterobacteriaceae family, when stored under different storage conditions. The main species of bacteria leading to the spoilage of aquatic products during cold storage are Acinetobacter johnsonii and Pseudomonas fluorescens , which are commonly referred to as specific spoilage organisms (SSOs) ( Jia et al., 2018 ; Pang and Yuk, 2019 ; Zhu et al., 2019 ). It is a significant SSOs due to its ability to produce volatile sulfides, amines, trimethylamine extracellular enzymes, trimethylamines, organic acids and some spoilage metabolites. Quorum sensing (QS) is the mechanism by which cell population-dependent signaling and interactions are recognized by bacteria in order to modulate their collective behaviors, including spoilage activity, enzyme secretion, bioluminescence, biofilm formation, virulence, and several signal molecules that mediate this mechanism have now been reported ( Natrah et al., 2012 ; Zhu S. et al., 2015 ). There are various QS molecules, such as CAI-1 (cholera autoinducer 1), DKPs (Diketopiperazines), HAQs (4-hydroxy-2-alkylquinolines), DSFs (Diffusible Signal Factors), AI-2 (Autoinducer-2) and indole, etc ( Monnet and Gardan, 2015 ; Papenfort and Bassler, 2016 ). QS-mediated communication is based on the prevailing interspecies communication in both Gram-negative and Gram-positive bacteria occurring via autoinducer-2 (AI-2) and auto-inducing peptides (AIPs) ( Stephens et al., 2019 ). P fluorescens is a Gram-negative bacteria that has been reported to use N -acyl-homoserine lactone (AHL) signals to monitor its local population through the exchange of extracellular signal molecules ( Li et al., 2018 ). SSOs utilize QS communication of circuits to regulate a diverse array of physiological activities microbial, including eavesdropping, biofilm genesis, and bioluminescence. Recently, several studies have found that QS signal molecules were N -butyryl-DL-homoserine lactone (C4-HSL), N -hexanoyl-DL-homoserine lactone (C6-HSL), octanoyl-L-homoserine lactone (C8-HSL), decanoyl-homoserine lactone (C10-HSL), and N -dodecanoyl-L-homoserine lactone (C12-HSL), which significantly reduced the protease activities and spoilage potential of SSOs. Moreover, QS systems can govern bacterial behavior in food spoilage ecology ( Whiteley et al., 2017 ). Bacterial growth, protease production, spoilage potential, electrochemical tests, and spoilage protein expression were significantly enriched through AHL signal regulation ( Zhu et al., 2017 ). However, few studies have focused on how two species of SSOs regulate their protein function in relation to spoilage potential. Recently, proteomic analysis has become a beneficial and quantifiable technique for providing relative measurements of proteins from different samples by nano-liquid chromatography-tandem mass spectrometry (nano-LC-MS/MS) without any isotope labeling. In a previous study on QS signaling, an absolute quantitative proteomic experiment was designed to investigate how Cyclo (L-phenylalanine-L-proline) effects protein expression in Staphylococcus aureus. Cyclo (L-phenylalanine-L-proline) has focused mainly on LuxR-mediated QS systems in bacteria, while the mechanism of extracellular QS signal molecule remained known ( Ai et al., 2019 ). Wang et al. (2019) identified 1103 acetylated proteins and 2929 acetylation sites in Shewanella baltica from aquatic products to evaluate spoilage activity by LC-MS/MS analysis. A label-free quantitative proteomic approach has been applied to analyze the differential protein expression of Enterococcus faecalis SK460 to determine potential factors related to enterococcal biofilm formation. The results of this study showed that the related protein in the relevance of LuxS QS and pheromone in the biofilm development of E. faecalis was due to the Fsr being lacking in QS ( Suryaletha et al., 2019 ). Based on proteomic analysis, a total of 338 vesicular proteins of Pseudomonas aeruginosa were identified with high confidence by LC-MS/MS analysis. This proteome profile provides a basis for future studies to illustrate the pathological functions of outer membrane vesicles from P. aeruginosa ( Choi et al., 2011 ). However, there are very few studies demonstrating the fact that two SSOs have a possible role in QS for controlling expression of their target proteins and regulating bacterial behavior. The aim of this study was to examine AHL production and AI-2 activity in A. johnsonii , P. fluorescens , and their co-culture, and to examine AHL-based QS systems regulating biofilm formation, protease activity, spoilage potential, and key proteins. Proteomic analysis of A. johnsonii , P. fluorescens , and their co-culture was used to elucidate the potential role of QS systems and their expression of proteins, interspecies communication, and metabolic pathways, which might be useful for developing effective methods for detecting spoilage capability.",
"discussion": "Results and Discussion Identification and Detection of AHLs in A. johnsonii , P. fluorescens , and Co-culture The reporter strain of the cross-feeding plate method is the fastest and most direct method to qualitatively detect AHLs in bacteria. CV026 is one of the commonly used reporter strains. CV026 does not produce AHLs by itself; however, CV026 is able to sense some of the AHLs of CivR protein ( Yu et al., 2019 ). CV026 is fully capable of producing violacein in response to its cognate signal molecule. As shown in Figure 1A , P. fluorescens and the co-culture were capable of producing AHLs induced CV026 color reaction, while A. johnsonii could have low capable of producing AHLs not induce CV026 color reaction. The AHL signals for A. johnsonii , P. fluorescens , and their co-culture are shown in Figure 2 and Supplementary Figure S2 . The three peaks with retention times of 3.26, 4.29, and 4.84 min were identified as C4-HSL, C6-HSL, and C8-HSL, respectively. The AHL (C4-HSL, C6-HSL, and C8-HSL) concentrations of the co-culture samples were significantly higher than those of A. johnsonii and P. fluorescens , indicating that the co-culture had pronounced AHL activities. The cooperative behaviors of the A. johnsonii and P. fluorescens cultures in response to AHLs activities might be to accelerate aquatic products spoilage. These trends are in agreement with the results presented in Table 1 . The HPLC-MS chromatograms of the AHLs of the co-culture in C4-HSL, C6-HSL, and C8-HSL contained 3.145, 7.359, and 1.560 ng/ml, respectively ( Table 1 ). Though several studies have reported that AHL signal molecules were the strongest autoinducers in Gram-negative bacteria from aquatic products ( Zhu J. et al., 2015 ), reports of relative AHL production in A. johnsonii and its co-culture are limited. FIGURE 1 (A) Detection of AHL production in A. johnsonii , P. fluorescens , and their co-culture; changes in: (B) AI-2 production; (C) protease activity; and (D) biofilm production in A. johnsonii , P. fluorescens , and their co-culture after 6 days. FIGURE 2 (A) Chromatograms of three kinds of AHL mixed standards. Chromatograms of C4-HSL produced by: (B) \n A. johnsonii ; (C) \n P. fluorescens ; and (D) their co-culture. TABLE 1 Contents of AHLs in A. johnsonii , P. fluorescens and co-culture. AHLs (ng/ml) A. johnsonii P. fluorescens Co-culture C 4 -HSL 1.204 ± 0.076 c 2.043 ± 0.060 b 3.145 ± 0.014 a C 6 -HSL 7.325 ± 0.299 a 6.802 ± 0.344 b 7.359 ± 0.379 a C 8 -HSL 1.543 ± 0.110 a 1.334 ± 0.308 b 1.560 ± 0.554 a a,b,c Means in the same line with different superscripts are significantly different ( p < 0.05). Detection of AI-2 Activity in A. johnsonii , P. fluorescens , and Co-culture The activities of AI-2 signal molecules of A. johnsonii , P. fluorescens , and their co-culture were detected ( Figure 1B ). There were significant differences in the AI-2 activity of A. johnsonii , P. fluorescens , and their co-culture at different incubation times ( p < 0.05). With increased culture time, the AI-2 activity first increased significantly from 0 to 4 days, then decreased, which was caused by increased bacterial growth density of bacteria, and bacteria in logarithmic phase. At the end of culture time, bacteria were in stationary phase and decline phase. Therefore, AI-2 activity decreased. It indicated that the changes in AI-2 activity were related to the secretion of the spoilage bacteria and environmental changes. Similar studies have shown that the AI-2 of QS is a global regulatory factor in aquatic products and influences the potential for spoilage ( Peng et al., 2018 ; Li S. et al., 2019 ). Protease Activity of A. johnsonii , P. fluorescens , and Co-culture Protease activity plays an essential role in food SSOs and is regulated by QS ( Moradi et al., 2019 ; Li T. et al., 2020 ). Moreover, protease activity decomposes aquatic food proteins into small peptides and amino acids that are metabolized into volatile nitrogenous end products ( Li J. et al., 2019 ). In this study, protease activity exhibited no significant effect after 3 days. The protease activity of A. johnsonii , P. fluorescens , and their co-culture first increased and then decreased ( p < 0.05) during different culture periods ( Figure 1C ). The previous study reported that SSOs growth consumed low-molecular-weight compounds, and then protein was degraded by protease which caused SSOs growth and protease activity increase ( Moradi et al., 2019 ). At the end of culture time, molecular-weight compounds were almost depleted. Therefore, protease activity decreased. The protease activities of the co-cultured samples were also higher than those of the single bacteria groups, in accord with the results reported in the literature ( Li J. et al., 2019 ). This result suggests that protease activity as a key spoilage characteristic of co-cultured samples was regulated significantly, and at least partially, by an AHL-based QS system. Biofilm Formation by A. johnsonii , P. fluorescens , and Co-culture Not only can biofilms influence food spoilage, resulting in reduced shelf life, but they can also adhere to bacteria and colonize surfaces ( Zhou et al., 2019 ). Previous studies have provided some evidence to verify the correlation between biofilm formation and the SSOs of QS systems, by which exogenous bacteria can affect the formation of biofilms ( Cui et al., 2020 ). Figure 1D shows that A. johnsonii , P. fluorescens , and their co-culture had the ability to form biofilms, which were greater in the co-cultured samples than those of the single bacteria. After 4d of culture, biofilm production reached maximum levels of 1.22, 1.38, and 1.39 by A. johnsonii , P. fluorescens , and their co-culture, respectively. Biofilm production increased when the incubation period was extended to 4d, but slowly decreased after 5d. Among the three groups, the presence of the co-cultured samples resulted in a significant increase in biofilm formation. This revealed that the cooperative behaviors of the A. johnsonii and P. fluorescens cultures in response to various signaling molecules helped to accelerate biofilm production. Proteome Analysis of A. johnsonii , P. fluorescens , and Co-culture Acinetobacter johnsonii , P. fluorescens , and their co-cultured samples were prepared using the label-free technique, which showed that a total of 1,176,121 spectra were identified in A. johnsonii and P. fluorescens by Proteome Discoverer TM 2.2 software with a peptide FDR ≤ 0.01. As the fold change was >1.2 or <0.83, a p -value of <0.05 was used as the threshold to define the significance of the difference in protein expression. This enabled the quantification of 470 proteins and 444 proteins in A. johnsonii and P. fluorescens , respectively. As shown in Figure 3A , there were 80 significant up-regulated proteins and 97 down-regulated proteins in the A group, compared with the AP group. In addition, 90 up-regulated and 65 down-regulated proteins were in the P group, compared with the AP group ( Figure 3B ). The differences in protein expression are given in Tables 2 , 3 . Among the remarkably up-regulated proteins, the AI-2E family transporter OS, such as those of A0A2S8XHJ7, A0A166PHR0, A0A423M0X2, A0A2N1DUG2, and A0A165YHM6, showed notable up-regulation, which mediated the regulation of QS system expression ( Quintieri et al., 2019 ; Saipriya et al., 2020 ). The result of AI-2 protein expression obtained above was similar to those shown in Figures 1B , 2 . Moreover, some ribosomal proteins were significantly up-regulated, of which 50S ribosomal protein L3 OS, 50S ribosomal protein L25 OS, 30S ribosomal protein S17 OS, 50S ribosomal protein L23 OS, 30S ribosomal protein S2 OS, 50S ribosomal protein L29 OS, 50S ribosomal protein L10 OS, 50S ribosomal protein L4 OS, 30S ribosomal protein S10 OS, 30S ribosomal protein S4 OS, 50S ribosomal protein L9 OS, 50S ribosomal protein L15 OS, 50S ribosomal protein L11 OS, 50S ribosomal protein L6 OS, 50S ribosomal protein L7/L12 OS, 50S ribosomal protein L2 OS, and 50S ribosomal protein L1 OS were expressed at much higher levels, and their fold changes were all greater than 139.12. A similar result was tested for the progression of the ribosome through a regulatory open reading frame (ORF), which controlled the protein synthesis expression of many genes in P. fluorescens ITEM 17298 and influenced the expression of the downstream gene ( Vazquez-Laslop et al., 2011 ; Gupta et al., 2013 ; Quintieri et al., 2019 ). The results showed that a majority of up-regulated proteins could accelerate protein expression and biological activity. There were some down-regulated proteins, such as thioredoxin reductase OS (A0A2K9M4Z2), cysteine synthase CysM OS (A0A506RJG5), DNA-binding response regulator (A0A423MKX5), amino acid ABC transporter ATPase OS (A0A0F4T6P4), that were expressed at a higher level in A vs. AP and P vs. AP. In the present study, bacteria played a critical role in transporting some molecules, including sugars, amino acids, vitamins, peptides, polysaccharides, lipids, thioredoxin, and ABC transporters ( Rees et al., 2009 ; Zhong et al., 2019 ). Interestingly, more down-regulated proteins of the AI-2 family transporter OS were obtained in A vs. AP than in P vs. AP, which demonstrated more down-regulated proteins affecting AI-2 expression in A. johnsonii . Further investigations of AI-2 activity in A. johnsonii revealed it to be at a lower level than in the co-culture, suggesting that the latter might more easily contribute to AI-2 protein expression. FIGURE 3 Volcanic map of all identified proteins: (A) volcanic map of all identified proteins in A vs. AP; (B) volcanic map of all identified proteins in P vs. AP. Red points: up-regulated proteins (fold change > 1.2, p < 0.05); green points: down-regulated proteins (fold change > 0.83, p < 0.05); black points: unchanged proteins. TABLE 2 The proteins differentially expression and most abundant proteins uniquely identified in A vs AP. Accession Protein name Regulate Fold change P -value A0A0B7DGZ6 60 kDa chaperonin OS Up 1.239 0.01994 A0A3S4MXA1 Chaperone protein HtpG OS Up 1.287 0.007272 A0A3M4FN09 ATP-dependent protease ATPase subunit HslU OS Up 1.223 0.007429 A0A4U3H0Y8 Thioredoxin TrxA OS Up 1.225 0.008509 A0A166XFT9 RpoB (Fragment) OS Up 1.309 0.02291 A0A4V5UF04 Molecular chaperone DnaK OS Up 8.057 0.001808 A0A0D0SI76 Chaperone protein ClpB OS Up 2.421 0.000832 A0A109KWR2 Chaperone protein HtpG OS Up 38.01 0.001028 A0A3S4MHS8 ATP-dependent Clp protease ATP-binding subunit ClpA OS Up 1.205 0.003655 A0A2W5E635 Elongation factor 4 OS Up 1.213 0.01939 J2Y774 Thioredoxin reductase OS Up 1.994 0.000966 A0A109KSE5 Arginine deiminase OS Up 3.039 0.000112 A0A3M4FUT4 Carbamoyl-phosphate synthase large chain OS Up 1.249 0.001582 A0A4R3X4W6 UDP-N-acetylglucosamine 2-epimerase OS Up 1.269 0.019 A0A0D0SKK3 ATP synthase subunit alpha OS Up 1.611 0.02238 A0A3M4FNE4 Elongation factor Tu OS Up 69.56 0.000101 I4KE62 Ornithine carbamoyltransferase OS Up 6.397 0.002375 A0A0D0PLG1 50S ribosomal protein L14 OS Up 1.621 0.000769 Q3KIA0 Chaperone protein DnaK OS Up 1.931 0.002845 A0A379IDI7 ABC transporter ATP-binding protein OS Up 1.4 0.0352 A0A379IHY5 Transcriptional regulator MvaT, P16 subunit OS Up 7.616 0.01267 A0A3M4G573 Dihydrolipoyllysine-residue succinyltransferase component of 2-oxoglutarate dehydrogenase complex OS Up 2.525 0.03714 A0A0D0PNN7 Aldehyde dehydrogenase OS Up 5.861 0.00719 U1TQN4 60 kDa chaperonin OS Up 139.12 0 A0A4V5UFP6 Chaperonin GroEL OS Up 139.12 0 A0A0B7DGY7 Elongation factor Tu OS Up 139.12 0 A0A0W0HIU9 Alkyl hydroperoxide reductase C OS Up 139.12 0 I4K6X1 Putative lipoprotein OS Up 139.12 0 I4KG18 50S ribosomal protein L1 OS Up 139.12 0 A0A109KXR2 Heat-shock protein OS Up 139.12 0 A0A125QDK7 Elongation factor Ts OS Up 139.12 0 A0A075PA24 10 kDa chaperonin OS Up 139.12 0 A0A109KX60 Electron transfer flavoprotein subunit alpha OS Up 139.12 0 A0A0X7K3A3 Dipicolinate synthase OS Up 139.12 0 A0A010RMN0 Succinate–CoA ligase [ADP-forming] subunit alpha OS Up 139.12 0 A0A4U3G4I4 Nucleotide exchange factor GrpE OS Up 139.12 0 A0A0C2A4F3 Nucleoside diphosphate kinase OS Up 139.12 0 A0A0A1YUR6 Membrane protein OS Up 139.12 0 A0A1Q5X417 Cold-shock protein OS Up 139.12 0 A0A0D0TC75 Succinate–CoA ligase [ADP-forming] subunit beta OS Up 139.12 0 A0A0B7DIY4 50S ribosomal protein L2 OS Up 139.12 0 A0A3M3XNL5 Ferritin domain-containing protein OS Up 139.12 0 A0A0A1Z5I2 50S ribosomal protein L7/L12 OS Up 139.12 0 A0A1T2YYC5 Ornithine aminotransferase OS Up 139.12 0 A0A010SEN5 50S ribosomal protein L6 OS Up 139.12 0 A0A010RGK5 Endoribonuclease OS Up 139.12 0 A0A0A1YZ47 Transcriptional regulator HU subunit alpha OS Up 139.12 0 A0A075P8Q2 30S ribosomal protein S4 OS Up 139.12 0 A0A010SQL5 50S ribosomal protein L9 OS Up 139.12 0 A0A0W0HKK7 50S ribosomal protein L15 OS Up 139.12 0 A0A075PC10 50S ribosomal protein L11 OS Up 139.12 0 A0A0C1ZKZ7 Lipoprotein OS Up 139.12 0 A0A010RRM4 30S ribosomal protein S10 OS Up 139.12 0 C3K254 Osmotically inducible protein Y OS Up 139.12 0 A0A0A1Z8J0 50S ribosomal protein L4 OS Up 139.12 0 A0A3M5MJ63 Fatty acid oxidation complex subunit alpha OS Up 139.12 0 A0A0N7H007 Urocanate hydratase OS Up 139.12 0 E2XZ08 50S ribosomal protein L29 OS Up 139.12 0 A0A0W0HLG0 50S ribosomal protein L10 OS Up 139.12 0 A0A3M3Y045 Aspartate ammonia-lyase OS Up 139.12 0 A0A448BWD3 Spermidine/putrescine import ATP-binding protein PotA OS Up 139.12 0 A0A3M3XD24 Nucleoid-associated protein ALQ35_00435 OS Up 139.12 0 A0A0W0HH67 Chromosome partitioning protein ParA OS Up 139.12 0 A0A010RSX6 Adenosylhomocysteinase OS Up 139.12 0 A0A120G5R9 DUF2383 domain-containing protein OS Up 139.12 0 A0A109KMT8 2,3,4,5-tetrahydropyridine-2,6-dicarboxylate N-succinyltransferase OS Up 139.12 0 A0A0G4E5Q2 Single-stranded DNA-binding protein OS Up 139.12 0 A0A0K1QL49 Protein-export protein SecB OS Up 139.12 0 A0A109KQK7 Uncharacterized protein OS Up 139.12 0 A0A3M3XER7 50S ribosomal protein L3 OS Up 139.12 0 A0A387BYY7 50S ribosomal protein L25 OS Up 139.12 0 U1TYF4 30S ribosomal protein S17 OS Up 139.12 0 A0A3M3XF13 50S ribosomal protein L23 OS Up 139.12 0 A0A3M5N9X5 30S ribosomal protein S2 OS Up 139.12 0 A0A3M4G1L6 Aldedh domain-containing protein OS Up 139.12 0 A0A0W0H1J6 Elongation factor P OS Up 139.12 0 A0A3M5N1D0 Superoxide dismutase OS Up 139.12 0 A0A0D0TBX7 OmpW protein OS Up 139.12 0 A0A109LGM0 Cysteine synthase B OS Up 139.12 0 A0A2S8XHJ7 AI-2E family transporter OS Up 139.12 0 A0A109LMS3 Cupin domain protein OS Down 0.7153 0.000434 A0A0B7DGC9 Serine hydroxymethyltransferase OS Down 0.78 0.001373 A0A3S4N2A1 DNA-binding transcriptional dual regulator Crp OS Down 0.7301 0.000463 A0A3M5MWI0 ATP-dependent Clp protease ATP-binding subunit ClpX OS Down 0.8297 0.00003 A0A3M4GP41 Threonine–tRNA ligase OS Down 0.7987 0.001523 A0A0C1X1G2 Ribosomal RNA large subunit methyltransferase J OS Down 0.8124 0.02763 A0A010SIG9 Aconitate hydratase OS Down 0.6839 0.000558 A0A3S4MM74 2-dehydro-3-deoxyphosphooctonate aldolase OS Down 0.7282 0.02122 A0A263S7G8 Elongation factor Tu (Fragment) OS Down 0.2323 0.0226 A0A448BG48 4-hydroxy-3-methylbut-2-en-1-yl diphosphate synthase (flavodoxin) OS Down 0.5716 0.000746 A0A3S4RD08 Glycerol kinase OS Down 0.7652 0.01836 A0A448BUH7 Transcription termination factor Rho OS Down 0.7681 0.04901 A0A109LM37 DUF86 domain-containing protein OS Down 0.7598 0.000255 A0A3S4N8H0 Pyruvate kinase OS Down 0.5626 0.01609 A0A3M3XN23 Adenylosuccinate lyase OS Down 0.8096 0.01257 A0A161Z4S7 Glycerol-3-phosphate dehydrogenase OS Down 0.5523 0.000078 A0A448BM26 Elongation factor G OS Down 0.679 0.003014 A0A4R3X0P0 Integration host factor subunit alpha OS Down 0.5545 0.001699 A0A3S4SXK6 Transcription termination/antitermination protein NusG OS Down 0.7081 0.001008 A0A0K1QXR7 Glycerol kinase OS Down 0.7005 0.0138 A0A345V0J2 dTDP-4-dehydrorhamnose reductase OS Down 0.5598 0.000511 G8Q3T2 Pyruvate dehydrogenase E1 component OS Down 0.781 0.00092 E2XZ43 Transcriptional regulator, Crp/Fnr family OS Down 0.7111 0.001972 A0A109LMK0 Nucleotidyltransferase domain protein OS Down 0.627 0.002719 A0A1T2YJX3 Glycerol kinase OS Down 0.4863 0.006571 A0A4R3XCE2 Uncharacterized protein OS Down 0.7417 0.00254 A0A010RT21 Site-determining protein OS Down 0.5787 0.002727 A0A0D9AUP5 60 kDa chaperonin OS Down 0.2131 0.000078 A0A448BVJ5 ATP synthase subunit alpha OS Down 0.6884 0.003857 A0A0B7D324 Translation initiation factor IF-2 OS Down 0.7328 0.02568 A0A010S2I6 GTP-binding protein TypA OS Down 0.4553 0.0148 A0A3S4MJG8 Succinate–CoA ligase [ADP-forming] subunit beta OS Down 0.717 0.01519 A0A109L0L2 Uncharacterized protein OS Down 0.4868 0.000044 A0A3S4MGB5 NAD-dependent malic enzyme OS Down 0.826 0.002371 A0A0F4VFT8 Quinone oxidoreductase OS Down 0.6851 0.000545 A0A3M5MJB2 DNA topoisomerase 1 OS Down 0.826 0.01052 A0A3M4G5X7 Alanine–tRNA ligase OS Down 0.8054 0.01262 A0A448BUH9 Glycine dehydrogenase (decarboxylating) OS Down 0.6628 0.0003 A0A3M3XUU9 Glucans biosynthesis glucosyltransferase H OS Down 0.43 0.000429 A0A370XHT9 G/U mismatch-specific DNA glycosylase OS Down 0.7478 0.003514 A0A109KNB3 Glutathione peroxidase OS Down 0.6022 0.003997 A0A0F4TQL9 Sulfate adenylyltransferase subunit 2 OS Down 0.5683 0.000132 A0A448BV68 Protease OS Down 0.7601 0.01423 A0A3M3XL87 Carbonic anhydrase OS Down 0.5356 0.000698 A0A3S4MT74 2Fe-2S ferredoxin OS Down 0.6699 0.000918 A0A010SYA6 UDP-N-acetylmuramate–L-alanine ligase OS Down 0.7554 0.03542 A0A0D0TN17 NAD/NADP-dependent betaine aldehyde dehydrogenase OS Down 0.7868 0.02845 A0A3S4PY88 Succinate–CoA ligase [ADP-forming] subunit alpha OS Down 0.753 0.0247 A0A1T2Y0Q8 Oxidoreductase OS Down 0.7698 0.00724 E2XYQ5 UDP-N-acetylmuramate–L-alanyl-gamma-D-glutamyl-meso-2,6-diaminoheptandioate ligase OS Down 0.7967 0.01584 A0A3M4G016 Lipoyl synthase OS Down 0.7383 0.01066 A0A0D9B4C3 NADH-quinone oxidoreductase subunit I OS Down 0.7234 0.01535 A0A0F4TK66 Dienelactone hydrolase OS Down 0.6862 0.003742 A0A010RIC6 Multifunctional fusion protein OS Down 0.4251 0.000001 A0A3S4PCY6 NADH-quinone oxidoreductase subunit F OS Down 0.8283 0.004383 A0A263S827 UDP-glucose 6-dehydrogenase (Fragment) OS Down 0.7358 0.008413 A0A010TH85 Glucarate dehydratase OS Down 0.7664 0.003172 B3PL47 Thioredoxin reductase OS Down 0.3985 0.006945 A0A010T8W6 Succinate dehydrogenase flavoprotein subunit OS Down 0.7114 0.008916 A0A379J8W9 Alkyl hydroperoxide reductase OS Down 0.4814 0.000953 A0A0X7K8K2 Alkene reductase OS Down 0.7332 0.004101 L7H6Z5 1,4-alpha-glucan branching enzyme GlgB OS Down 0.7367 0.005334 A0A0B7DGX9 30S ribosomal protein S7 OS Down 0.3928 0.001471 A0A3M4FP51 Bifunctional protein PutA OS Down 0.5788 0.000048 A0A0A1YRY8 Peroxiredoxin OsmC OS Down 0.6682 0.000699 A0A0D0NJK9 NADH-quinone oxidoreductase subunit C/D OS Down 0.5453 0.009922 A0A0F4T6P4 Amino acid ABC transporter ATPase OS Down 0.2658 0.01775 A0A423M0U6 Microcin ABC transporter ATP-binding protein OS Down 0.6836 0.005146 A0A379IA35 Cysteine synthase OS Down 0.5603 0.00057 A0A120G9E8 DNA-invertase hin OS Down 0.5688 0.0114 A0A3S4RLS6 50S ribosomal protein L17 OS Down 0.5046 0.00556 A0A2A5REX4 AI-2E family transporter Down 0.7659 0.006788 A0A3M3XCL9 Phosphoenolpyruvate synthase OS Down 0.2506 0.000621 A0A010SRH4 Phosphoribosylglycinamide formyltransferase OS Down 0.7039 0.000056 I4K120 Malonate decarboxylase, gamma subunit OS Down 0.7353 0.02236 A0A0C1WLI8 30S ribosomal protein S13 OS Down 0.3449 0.001954 A0A3M3XP35 Aldedh domain-containing protein OS Down 0.3826 0.000729 A0A1B3D5W7 AI-2E family transporter OS Down 0.7473 0.03427 A0A2N1DZF0 Type VI secretion system tube protein Hcp OS Down 0.3845 0.006341 A0A4P7I7D8 AAA family ATPase OS Down 0.4335 0.004957 A0A3M4GSW6 DNA polymerase III subunit alpha OS Down 0.6754 0.004196 A0A0P8X0Z8 Succinate dehydrogenase and fumarate reductase iron-sulfur family protein OS Down 0.2746 0.02616 A0A1T2ZCZ1 Alkyl hydroperoxide reductase C OS Down 0.00001 0 A0A109KFA5 Peptide methionine sulfoxide reductase MsrA OS Down 0.00001 0 A0A0W0HVB3 Electron transfer flavoprotein subunit beta OS Down 0.00001 0 A0A423N111 TetR family transcriptional regulator OS Down 0.00001 0 A0A2W5EAC0 Citrate synthase OS Down 0.00001 0 A0A109FQR8 Ketol-acid reductoisomerase (NADP(+)) OS Down 0.00001 0 A0A423MKX5 DNA-binding response regulator OS Down 0.00001 0 A0A0B7D391 1-(5-phosphoribosyl)-5-[(5-phosphoribosylamino) methylideneamino] imidazole-4-carboxamide isomerase OS Down 0.00001 0 A0A0B7DDH5 Iron-sulfur cluster assembly scaffold protein IscU OS Down 0.00001 0 A0A327NA26 AI-2E family transporter OS Down 0.00001 0 A0A327MWL2 AI-2E family transporter OS Down 0.00001 0 A0A2K9M4Z2 Thioredoxin reductase OS Down 0.00001 0 A0A506RJG5 Cysteine synthase CysM OS Down 0.00001 0 A0A423L5V9 AI-2E family transporter OS Down 0.00001 0 A0A2A5R442 AI-2E family transporter OS Down 0.00001 0 TABLE 3 The proteins differentially expression and most abundant proteins uniquely identified in P vs. AP. Accession Protein name Regulate Fold change p -value A0A448DVQ8 Thioredoxin reductase OS Up 1.226 0.005186 A0A010S5P5 Electron transfer flavoprotein subunit beta OS Up 1.519 0.03911 A0A109KZR3 30S ribosomal protein S1 OS Up 1.22 0.000809 A0A3S4NMX4 Cysteine synthase OS Up 1.447 0.01686 A0A010T165 Cysteine desulfurase IscS OS Up 1.316 0.004729 A0A166PHR0 AI-2E family transporter OS Up 1.441 0.01166 A0A3S4NMW4 Proline–tRNA ligase OS Up 1.213 0.03312 A0A423M0X2 AI-2E family transporter OS Up 1.21 0.03034 A0A423LFX7 Glycine betaine ABC transporter substrate-binding protein OS Up 1.929 0.006274 A0A3M4HEE6 Fn3_like domain-containing protein OS Up 1.361 0.01977 A0A109L3N2 Elongation factor G OS Up 1.288 0.001337 A0A0D0SI76 Chaperone protein ClpB OS Up 2.025 0.0105 A0A4V5UF04 Molecular chaperone DnaK OS Up 6.781 0.000406 A0A0B7DHH8 S-adenosylmethionine:tRNA ribosyltransferase-isomerase OS Up 1.319 0.001532 A0A109KSE5 Arginine deiminase OS Up 2.538 0.000025 A0A109KWR2 Chaperone protein HtpG OS Up 26.25 0.000119 A0A3M4FNE4 Elongation factor Tu OS Up 61.14 0.000005 I4KE62 Ornithine carbamoyltransferase OS Up 5.734 0.002111 A0A0X7K650 Alkyl hydroperoxide reductase OS Up 2.631 0.000265 I4K7P9 ATP synthase subunit beta OS Up 4.626 0.03624 A0A0C1WLI8 30S ribosomal protein S13 OS Up 2.938 0.04752 A0A3M4GZQ4 Ribose-phosphate pyrophosphokinase OS Up 1.725 0.00228 A0A0D0PLG1 50S ribosomal protein L14 OS Up 1.668 0.000929 A0A010RPF8 16S rRNA methyltransferase OS Up 1.548 0.01886 A0A3M4G573 Dihydrolipoyllysine-residue succinyltransferase component of 2-oxoglutarate dehydrogenase complex OS Up 4.699 0.01403 A0A379IDI7 ABC transporter ATP-binding protein OS Up 1.278 0.01057 A0A2K9M4Z2 Thioredoxin reductase OS Up 1.908 0.04652 A0A2N1DUG2 AI-2E family transporter OS Up 2.147 0.000219 A0A379IHY5 Transcriptional regulator MvaT, P16 subunit OS Up 6.47 0.000263 A0A165YHM6 AI-2E family transporter OS Up 1.881 0.00109 A0A0D0PNN7 Aldehyde dehydrogenase OS Up 3.797 0.001845 U1TQN4 60 kDa chaperonin OS Up 122.28 0 A0A4V5UFP6 Chaperonin GroEL OS Up 122.28 0 A0A0W0HIU9 Alkyl hydroperoxide reductase C OS Up 122.28 0 I4K6X1 Putative lipoprotein OS Up 122.28 0 I4KG18 50S ribosomal protein L1 OS Up 122.28 0 A0A109KXR2 Heat-shock protein OS Up 122.28 0 A0A125QDK7 Elongation factor Ts OS Up 122.28 0 A0A075PA24 10 kDa chaperonin OS Up 122.28 0 A0A109KX60 Electron transfer flavoprotein subunit alpha OS Up 122.28 0 A0A0X7K3A3 Dipicolinate synthase OS Up 122.28 0 A0A010RMN0 Succinate–CoA ligase [ADP-forming] subunit alpha OS Up 122.28 0 A0A4U3G4I4 Nucleotide exchange factor GrpE OS Up 122.28 0 A0A0C2A4F3 Nucleoside diphosphate kinase OS Up 122.28 0 A0A0A1YUR6 Membrane protein OS Up 122.28 0 A0A1Q5X417 Cold-shock protein OS Up 122.28 0 A0A0D0TC75 Succinate–CoA ligase [ADP-forming] subunit beta OS Up 122.28 0 A0A0B7DIY4 50S ribosomal protein L2 OS Up 122.28 0 A0A0D0PJB7 Enolase OS Up 122.28 0 A0A3M3XNL5 Ferritin domain-containing protein OS Up 122.28 0 A0A0A1Z5I2 50S ribosomal protein L7/L12 OS Up 122.28 0 A0A1T2YYC5 Ornithine aminotransferase OS Up 122.28 0 A0A010SEN5 50S ribosomal protein L6 OS Up 122.28 0 A0A010RGK5 Endoribonuclease OS Up 122.28 0 A0A0A1YZ47 Transcriptional regulator HU subunit alpha OS Up 122.28 0 A0A075P8Q2 30S ribosomal protein S4 OS Up 122.28 0 A0A010SQL5 50S ribosomal protein L9 OS Up 122.28 0 A0A0W0HKK7 50S ribosomal protein L15 OS Up 122.28 0 A0A075PC10 50S ribosomal protein L11 OS Up 122.28 0 A0A0C1ZKZ7 Lipoprotein OS Up 122.28 0 A0A010RRM4 30S ribosomal protein S10 OS Up 122.28 0 C3K254 Osmotically inducible protein Y OS Up 122.28 0 A0A0A1Z8J0 50S ribosomal protein L4 OS Up 122.28 0 A0A3M5MJ63 Fatty acid oxidation complex subunit alpha OS Up 122.28 0 A0A0N7H007 Urocanate hydratase OS Up 122.28 0 E2XZ08 50S ribosomal protein L29 OS Up 122.28 0 A0A0W0HLG0 50S ribosomal protein L10 OS Up 122.28 0 A0A448BQI1 Phosphoribosylformylglycinamidine cyclo-ligase OS Up 122.28 0 A0A3M3Y045 Aspartate ammonia-lyase OS Up 122.28 0 A0A448BWD3 Spermidine/putrescine import ATP-binding protein PotA OS Up 122.28 0 A0A0W0QVH5 Porin OS Up 122.28 0 A0A3M3XD24 Nucleoid-associated protein ALQ35_00435 OS Up 122.28 0 A0A3M4GIL8 Fumarate hydratase class I OS Up 122.28 0 A0A0W0HH67 Chromosome partitioning protein ParA OS Up 122.28 0 A0A010RSX6 Adenosylhomocysteinase OS Up 122.28 0 A0A0K1QHC1 50S ribosomal protein L18 OS Up 122.28 0 A0A109KMT8 2,3,4,5-tetrahydropyridine-2,6-dicarboxylate N-succinyltransferase OS Up 122.28 0 A0A0G4E5Q2 Single-stranded DNA-binding protein OS Up 122.28 0 A0A109KQK7 Uncharacterized protein OS Up 122.28 0 A0A3M3XER7 50S ribosomal protein L3 OS Up 122.28 0 A0A387BYY7 50S ribosomal protein L25 OS Up 122.28 0 U1TYF4 30S ribosomal protein S17 OS Up 122.28 0 A0A3M3XF13 50S ribosomal protein L23 OS Up 122.28 0 A0A3M5N9X5 30S ribosomal protein S2 OS Up 122.28 0 A0A3S4RR46 Nitrogen regulatory protein PII OS Up 122.28 0 A0A3M5MIC6 S1 motif domain-containing protein OS Up 122.28 0 A0A3M4G1L6 Aldedh domain-containing protein OS Up 122.28 0 A0A0W0H1J6 Elongation factor P OS Up 122.28 0 A0A3M5N1D0 Superoxide dismutase OS Up 122.28 0 A0A2S8XHJ7 AI-2E family transporter OS Up 122.28 0 A0A109LMS3 Cupin domain protein OS Down 0.7207 0.003665 A0A0B7DGC9 Serine hydroxymethyltransferase OS Down 0.706 0.001258 A0A3M5MWI0 ATP-dependent Clp protease ATP-binding subunit ClpX OS Down 0.798 0.00304 A0A010T1X2 50S ribosomal protein L5 OS Down 0.7538 0.005059 A0A448BG48 4-hydroxy-3-methylbut-2-en-1-yl diphosphate synthase (flavodoxin) OS Down 0.6935 0.005875 A0A109LM37 DUF86 domain-containing protein OS Down 0.7715 0.007694 A0A3S4N8H0 Pyruvate kinase OS Down 0.616 0.03967 A0A0A1ZAA7 Glucose-1-phosphate thymidylyltransferase OS Down 0.7914 0.02799 A0A161Z4S7 Glycerol-3-phosphate dehydrogenase OS Down 0.7438 0.003224 A0A3M3XN23 Adenylosuccinate lyase OS Down 0.8242 0.005162 A0A448BGH7 Phosphoenolpyruvate carboxylase OS Down 0.7643 0.01978 A0A3S4SXK6 Transcription termination/antitermination protein NusG OS Down 0.7793 0.03637 A0A345V0J2 dTDP-4-dehydrorhamnose reductase OS Down 0.6827 0.002461 E2XZ43 Transcriptional regulator, Crp/Fnr family OS Down 0.8076 0.01046 A0A010RT21 Site-determining protein OS Down 0.7867 0.01215 A0A109LMK0 Nucleotidyltransferase domain protein OS Down 0.7134 0.003733 A0A4R3XCE2 Uncharacterized protein OS Down 0.8129 0.0114 A0A0N9VVZ3 Phosphoserine aminotransferase OS Down 0.8173 0.04239 A0A3S4N6L6 Isocitrate dehydrogenase [NADP] OS Down 0.78 0.02703 A0A379J9N3 Thioredoxin reductase OS Down 0.7602 0.02114 A0A4Y4JC04 Cysteine synthase B OS Down 0.8055 0.007142 A0A3S4MJG8 Succinate–CoA ligase [ADP-forming] subunit beta OS Down 0.7899 0.0351 A0A3M5MSW3 UvrABC system protein B OS Down 0.8096 0.002896 A0A109L0L2 Uncharacterized protein OS Down 0.6859 0.003862 A0A379J4U0 Cysteine synthase OS Down 0.7646 0.03365 A0A0A1YZP1 Histidine–tRNA ligase OS Down 0.7848 0.00437 A0A3M4G5X7 Alanine–tRNA ligase OS Down 0.7532 0.01493 A0A0K1QSD3 LysR family transcriptional regulator OS Down 0.8259 0.04135 A0A370XHT9 G/U mismatch-specific DNA glycosylase OS Down 0.6979 0.01167 A0A3M3XL87 Carbonic anhydrase OS Down 0.68 0.009262 A0A448BV68 Protease OS Down 0.8028 0.01425 A0A109KNB3 Glutathione peroxidase OS Down 0.4815 0.002483 A0A3S4MT74 2Fe-2S ferredoxin OS Down 0.7981 0.01977 A0A010RGL6 Phosphoglucomutase OS Down 0.7412 0.000758 Q4K9V2 Thioredoxin reductase OS Down 0.7618 0.03801 A0A3M4G016 Lipoyl synthase OS Down 0.7969 0.03549 A0A010RIC6 Multifunctional fusion protein OS Down 0.5799 0.000021 A0A0D9B4C3 NADH-quinone oxidoreductase subunit I OS Down 0.64 0.00074 A0A010RSQ5 L-cystine transporter tcyP OS Down 0.5553 0.01979 A0A0F4T9J8 Diaminopimelate decarboxylase OS Down 0.8255 0.002003 B3PL47 Thioredoxin reductase OS Down 0.5017 0.002077 A0A010T8W6 Succinate dehydrogenase flavoprotein subunit OS Down 0.8224 0.04234 A0A0B7DI73 Isoleucine–tRNA ligase OS Down 0.7257 0.000215 A0A3M4GK93 Phosphoribosylaminoimidazole-succinocarboxamide synthase OS Down 0.686 0.01121 A0A3M5MVH0 Glucans biosynthesis protein D OS Down 0.6468 0.03983 A0A0X7K8K2 Alkene reductase OS Down 0.8053 0.00987 A0A3M4FP51 Bifunctional protein PutA OS Down 0.6509 0.000256 A0A0A1YRY8 Peroxiredoxin OsmC OS Down 0.7261 0.005055 A0A3S4N1Z0 Cytosine deaminase OS Down 0.7876 0.007186 A0A0N9W7I2 Catalase OS Down 0.6428 0.001721 A0A0F4T6P4 Amino acid ABC transporter ATPase OS Down 0.474 0.02104 A0A0B7D1W3 Glycine betaine transport ATP-binding protein OpuAA OS Down 0.7214 0.01965 A0A379IA35 Cysteine synthase OS Down 0.7158 0.004647 I4K120 Malonate decarboxylase, gamma subunit OS Down 0.8073 0.01994 A0A010SRH4 Phosphoribosylglycinamide formyltransferase OS Down 0.7332 0.000756 A0A327MWL2 AI-2E family transporter OS Down 0.6227 0.009033 A0A0N8NY49 Aldo/keto reductase family protein OS Down 0.7843 0.04723 A0A0B7D6X0 Glutathione import ATP-binding protein GsiA OS Down 0.6366 0.003439 A0A0F4TUQ5 Glucose dehydrogenase OS Down 0.454 0.02673 A0A0P9BDW1 Ribose import ATP-binding protein RbsA OS Down 0.6737 0.02709 A0A3S4PXV5 Transcriptional activator CopR OS Down 0.731 0.03974 A0A423N111 TetR family transcriptional regulator OS Down 0.00001 0 A0A0B7DDH5 Iron-sulfur cluster assembly scaffold protein IscU OS Down 0.00001 0 A0A125QFU4 AI-2E family transporter OS Down 0.00001 0 A0A386Y9V3 AI-2E family transporter OS Down 0.00001 0 Bioinformatics Functional Analysis of A. johnsonii , P. fluorescens , and Co-culture GO and COG Enrichment Analysis of A. johnsonii , P. fluorescens , and Co-culture A GO classification and COG enrichment of the 177 and 155 differentially expressed proteins were performed in A vs. AP and P vs. AP, respectively ( Figures 4A,B ). There were three main categories of cellular components, biological processes, and molecular functions in the GO classification. The cellular process (GO:0009987), the metabolic process (GO:0008152), and the single-organism process (GO:0044699) were the three main distributed terms in the biological processes. The 85 and 79 differentially expressed proteins were annotated as belonging to the cell in A vs. AP and P vs. AP, respectively. In addition, A vs. AP (80 out of 177) and P vs. AP (76 out of 155) of the differentially expressed proteins were localized in the cell part. This suggested that A. johnsonii , P. fluorescens , and their co-culture played an essential role in the transmembrane transport function and intracellular and extracellular substance migration, thereby promoting nutrient absorption and excretion of metabolic products. In the molecular functions category, the proteins were related to catalytic activity and binding. FIGURE 4 Gene ontology terms of the differentially expressed proteins in: (A) A vs. AP; (B) P vs. AP. COG terms of the differentially expressed proteins in: (C) A vs. AP; (D) : P vs. AP. KEGG pathway analysis of the differentially expressed proteins in: (E) A vs. AP; (F) P vs. AP. The red bars represented the up-regulated proteins, and the blue bars represent the down-regulated proteins in the KEGG pathway analysis of A. johnsonii , P. fluorescens , and their co-culture. Candida albicans glycolysis/gluconeogenesis pathways (A) and peroxisomal assembly and fatty acid oxidation pathways. Figures 4C,D displays the COG enrichment analysis. A total of 18 categories were classified, in which the top 6 COG terms were (i) energy production and conversion, (ii) amino acid transport and metabolism, (iii) translation, ribosomal structure, and biogenesis, (iv) post-translational modification, (v) protein turnover, and (vi) chaperones. A total of 22 down-regulated proteins were involved in energy production and conversion in A vs. AP, while 23 up-regulated proteins were involved in translation, ribosomal structure, and biogenesis in A vs. AP. Interestingly, translation, ribosomal structure, and biogenesis were significantly up-regulated in P vs. AP, and amino acid transport and metabolism and energy production and conversion were significantly down-regulated in P vs. AP, suggesting that translation, ribosomal structure, and energy production and conversion were valuable targets for co-culturing, and thus deserve further investigation. It has been suggested that under co-culture conditions, energy production is capable of involving translation of the bacteria, resulting in the activation of the ribosomal structure ( Gupta et al., 2013 ). These proteins might also participate in nucleotide catabolism, allowing bacteria to use deoxynucleotides as energy sources. The results of GO analysis and COG enrichment provide a significant view of the proteins differentially expressed in A. johnsonii , P. fluorescens , and their co-culture that can elevate protein functions. Pathway Enrichment Analysis of A. johnsonii , P. fluorescens , and Co-culture Pathway enrichment analysis of the KEGG database was conducted on the differentially expressed proteins to reveal the metabolic and signal transduction pathways of A. johnsonii , P. fluorescens , and their co-culture ( Figures 4E,F ). In our study, taken together, the up/down-regulated differentially expressed proteins of A. johnsonii , P. fluorescens , and their co-culture involved in amino acid metabolism, carbohydrate and energy metabolism, and nucleotide metabolism enabled predictions of the change in the culture of the A. johnsonii , P. fluorescens samples in the co-culture, indicating that protein changes and carbohydrate transformation contributed to bacteria co-culturing. A total of 84 and 77 differentially expressed proteins in A vs. AP and P vs. AP, respectively, were divided into 20 KEGG pathways, and a majority of the metabolic pathways included genetic information processing, environmental information processing, cellular processes, organismal systems, and human diseases. The KEGG pathways of amino acid metabolism, carbohydrate metabolism, energy metabolism, and translation were significantly enriched. A similar result for carbohydrate metabolism and energy metabolism of Vibrio parahaemolyticus revealed that carbohydrate metabolism is the key factor, indicating that carbohydrate can either be converted into glucan and fructose through a glycosyltransferase reaction or transported by sugar transport systems and subsequently metabolized through glycolysis ( Zhong et al., 2019 ). The main significant items relevant to the regulation of biofilm formation of A vs. AP and P vs. AP included the map 02026, map 05111, and map 02025 pathways ( Supplementary Tables S1 , S2 ). Analysis of the proteins of A. johnsonii , P. fluorescens , and their co-culture indicated that all of the pathways were present. In addition, the pathway of map 02024 was that which regulated QS, which is a crucial feature affecting the regulation of biofilm formation by AHLs, bacterial growth, protease activity, and the spoilage potential of bacteria ( Jie et al., 2018 ). Furthermore, as shown in Supplementary Tables S1 , S2 , ribosome was the most significantly enriched pathway, which indicated that protein expression was substantially promoted to achieve a large demand for bacterial growth ( Li J. et al., 2020 ). Ribosome was the most significant pathway, with 43 differentially expressed proteins, of which 22 were up-regulated proteins and 21 were down-regulated proteins. In addition, abundant proteins were associated with the tricarboxylic acid (TCA) cycle and oxidative phosphorylation in response to co-culture conditions ( Supplementary Figure S1 ). The TCA cycle pathway has 7 up-regulated proteins and 9 down-regulated proteins. The most abundant protein detected among the 6 up-regulated and 11 down-regulated proteins was associated with oxidative phosphorylation. These pathways were essential for bacterial growth and cell interactions, which have potential for enhancing bacterial spoilage and QS regulation ( Remenant et al., 2015 ; Otwell et al., 2018 ). Subcellular Localization Prediction of A. johnsonii , P. fluorescens , and Co-culture The subcellular localization prediction of proteins has attracted considerable attention in protein functional annotation. There were two subcellular localization predictions, including for cytoplasm and plasma membranes. Figure 5 shows that subcellular localization in P vs. AP was located mainly in the cytoplasm and the plasma membrane, but mainly in the cytoplasm in A vs. AP. Previous studies have focused mainly on single bacteria producing a few metabolites that migrate from the inner membrane to cell-extracellular membranes during culture ( Ellepola et al., 2019 ), resulting in cell communications. However, few studies have reported explanations for two bacterial subcellular localization predictions. As shown in Figure 5 , expression of cytoplasmic proteins was also regulated in the co-culture. FIGURE 5 Bar plot of subcellular localization prediction in: (A) A vs. AP; (B) P vs. AP. Quorum Sensing and Spoilage-Related Proteins The QS mechanism is the means by which cell population communication can regulate specific proteins to express the physiological characteristics of microorganisms ( Li S. et al., 2019 ). As given in Table 4 , the QS system was related to the protein of the AI-2E family transporter OS and the LuxR family transcriptional regulator OS. The LuxR family transcriptional regulator was found in A vs. AP and P vs. AP. The LuxR protein (AHL receptor) commonly consists of 200–260 amino acids blinding with the key protein, resulting in the expression of the key protein. Five AI-2E family transporter up-regulated proteins and 11 AI-2E family transporter up-regulated proteins were found in A vs. AP and P vs. AP, respectively. This suggests that the AI-2E family transporter played a vital role in regulating QS. Previous studies have shown that P. fluorescens can produce AI-2 proteins and AHLs ( Sharma et al., 2006 ), which is consistent with the results described in Section “Protease Activity of A. johnsonii , P. fluorescens , and Co-culture.” Moreover, there were too many spoilage-related proteins, including thioredoxin reductase OS (6 down-regulated proteins, 5 up-regulated proteins), cysteine synthase OS (7 down-regulated proteins, 1 up-regulated proteins), and pyridoxal phosphate-dependent enzyme family protein OS. Notably, the spoilage-related proteins in A. johnsonii , P. fluorescens , and their co-culture were similar to those of Shewanella baltica and P. fluorescens ITEM 17298 ( Quintieri et al., 2019 ). TABLE 4 Quorum sensing system proteins and spoilage related proteins of A. johnsonii , P. fluorescens and co-culture. Function Protein name Accession number Regulate QS system AI-2E family transporter OS A0A2S8XHJ7, A0A293PZ07, A0A2N1DUG2, A0A165YHM6, A0A2S8XHJ7, Up A0A1T3AM62, A0A161GM51, A0A2K9M819, A0A2A5REX4, A0A386Y9V3, A0A1B3D5W7, A0A327NA26, A0A327MWL2, A0A423L5V9, A0A2A5R442, A0A1T2YCZ7 Down LuxR family transcriptional regulator OS A0A0U3TA42 Down Spoilage Thioredoxin reductase OS A0A0W8H2Q7, A0A4P6V9D1, Q4K9V2, B3PL47, A0A2K9M4Z2, A0A379J9N3 Down A0A379J3Q0, J2Y774, A0A2A5RE67, A0A3S4RHU3, A0A4P7I1W5 Up Cysteine synthase OS A0A380UBL3, G8Q5M5, A0A379IA35, A0A506RJG5, J2MKP0, A0A3S4PFK0, A0A109KX01 Down A0A109LGM0 Up Pyridoxal-phosphate dependent enzyme family protein OS A0A379J407, A0A3S4PDU0 Down Sulfate adenylyltransferase subunit 2 OS A0A0F4TQL9 Down Multifunctional fusion protein OS A0A010RIC6 Down Putative ABC transporter ATP-binding protein OS A0A379JE46 Glycine betaine transport ATP-binding protein OpuAA OS A0A0B7D1W3 Down Microcin ABC transporter ATP-binding protein OS A0A423M0U6 Down ABC transporter ATP-binding protein OS A0A379IDI7 Up Glutathione import ATP-binding protein GsiA OS A0A0B7D6X0 Ribose import ATP-binding protein RbsA OS A0A0P9BDW1 Spermidine/putrescine import ATP-binding protein PotA OS A0A010RIC6 Down Spermidine/putrescine ABC transporter substrate-binding protein OS A0A010RQY0 Spermidine/putrescine import ATP-binding protein PotA OS A0A448BWD3 Up Transcriptional regulator, Crp/Fnr family OS E2XZ43 Down Transcriptional activator CopR OS A0A3S4PXV5 Down Urocanate hydratase OS A0A0N7H007 Up"
} | 12,652 |
24341984 | PMC4344821 | pmc | 9,101 | {
"abstract": "The way species affect one another in ecological communities often depends on the order of species arrival. The magnitude of such historical contingency, known as priority effects, varies across species and environments, but this variation has proven difficult to predict, presenting a major challenge in understanding species interactions and consequences for community structure and function. Here, we argue that improved predictions can be achieved by decomposing species' niches into three components: overlap, impact and requirement. Based on classic theories of community assembly, three hypotheses that emphasise related, but distinct influences of the niche components are proposed: priority effects are stronger among species with higher resource use overlap; species that impact the environment to a greater extent exert stronger priority effects; and species whose growth rate is more sensitive to changes in the environment experience stronger priority effects. Using nectar-inhabiting microorganisms as a model system, we present evidence that these hypotheses complement the conventional hypothesis that focuses on the role of environmental harshness, and show that niches can be twice as predictive when separated into components. Taken together, our hypotheses provide a basis for developing a general framework within which the magnitude of historical contingency in species interactions can be predicted.",
"conclusion": "Conclusion Over the past decade, the controversy surrounding the neutral theory of biodiversity ( Hubbell 2001 ) has led to a resurgence of interest in using classic niche-based theories to explain species interactions and their implications for species diversity (e.g. Chase & Leibold 2003 ; Fargione et al . 2003 ; HilleRisLambers et al . 2012 ). However, despite the intensive research that ensued, the role of niches in species interactions remains elusive. Our experimental findings suggest that substantial progress will come from explicit consideration of multiple niche components, which can greatly improve our ability to explain even the most difficult aspect of species interactions, historical contingency. The relative importance of niche components will likely vary from system to system, but the approach we have taken here provides a basis for developing a general framework within which their relative importance can be determined in different types of communities.",
"introduction": "Introduction One difficulty in predicting how species affect one another in ecological communities is the common occurrence of priority effects, where the order in which species arrive at local sites dictates the effect of species on one another ( Gleason 1926 ; Lewontin 1969 ; MacArthur 1972 ; Gilpin & Case 1976 ; Drake 1991 ; Chase & Leibold 2003 ; Fukami & Morin 2003 ; Petraitis 2013 ). In many cases, species arrival order is highly stochastic and impossible to know, making the outcome of species interactions essentially unpredictable when priority effects are strong. For this reason, understanding factors that determine the magnitude of historical contingency due to priority effects is important not only to improving basic knowledge of how species assemble into communities ( Chase 2003 ; Fukami 2010 ) but also to applying this knowledge to environmental, agricultural, medical and other problems that involve management of ecological systems ( Temperton & Zirr 2004 ; Young et al . 2005 ; Grman & Suding 2010 ; Costello et al . 2012 ; Verbruggen et al . 2013 ). Besides primary productivity, disturbance rate and other environmental factors (e.g. Chase 2003 , 2007 ; Kardol et al . 2013 ), characteristics of potential colonists, including dispersal ability (e.g. Shorrocks & Bingley 1994 ; Porensky et al . 2012 ), organism lifespan (e.g. Munguia et al . 2010 ; Young & Peffer 2010 ) and ecological similarity among species (e.g. Fargione et al . 2003 ; Fukami et al . 2005 ), have been hypothesised to determine the magnitude of historical contingency by priority effects. For example, priority effects have been thought to act strongly when early-arriving species deplete local resources and inhibit colonisation by late-arriving species that have similar resource requirements ( Fox 1987 ; Weiher et al . 1998 ; Wilson 1999 ; Fargione et al . 2003 ; Fukami et al . 2005 ). However, attempts to link the strength of priority effects to resource requirements and other characteristics of species have met with limited success ( Peay et al . 2012 ; Tan et al . 2012 ). Moreover, although consideration of phylogenetic relatedness between species has shown some promise ( Jiang et al . 2010 ; Peay et al . 2012 ; Tan et al . 2012 ), phylogeny is an imperfect proxy for functional traits and lacks a mechanistic basis in predicting priority effects ( Mayfield & Levine 2010 ; Best et al . 2013 ). The goal of this article is to suggest that mechanistic predictions of the strength of priority effects can be accomplished by separating species' niches into basic components. To support this claim, we will demonstrate the utility of the approach with a model experimental system involving nectar-inhabiting yeasts. To stimulate further development in future research, we will then discuss ways to expand the scope of the approach and outline possible applications for management of ecological communities. Niche components and priority effects A species' niche can be viewed as consisting of three components, including niche overlap, impact niche and requirement niche (Fig. 1 ). Derived from Gause's (1932) competitive exclusion principle and MacArthur & Levins's (1967) limiting similarity concept, niche overlap refers to resource use similarity among co-occurring species, independent of their rate of resource consumption ( Pianka 1973 ; Petraitis 1989 ). Based on Elton's (1927) niche concept as a species' role in the environment and Tilman's (1982) theory of resource competition, the impact niche is defined as a species' per capita influence on the environment through resource consumption and other modes of environmental modification ( Leibold 1995 ; Chase & Leibold 2003 ). Finally, originating from Grinnell's (1917) limiting factors and Hutchinson's (1957 ) fundamental niche, the requirement niche describes the environmental conditions that affect a species' survival, growth and reproduction ( Leibold 1995 ; Chase & Leibold 2003 ). The contrast between the impact niche and the requirement niche is similar to that of ‘effects traits’ and ‘response traits’ ( sensu \n Lavorel & Garnier 2002 ; Suding et al . 2008 ), where species with high impact and those with low requirement may be strong ‘effect competitors’ and ‘response competitors’, respectively ( sensu \n Goldberg & Landa 1991 ). Figure 1 Schematic depiction of niche components and environmental harshness (a) and how they are hypothesised to influence the strength of priority effects (b). Recognising that these related, but distinct niche components may differentially determine how species interact, and assuming that interactions among species are generally weaker than interactions within species, one can propose the following hypotheses (Fig. 1 ): priority effects should be strong when (1) species display a high degree of similarity in resource use (high overlap), (2) early-arriving species strongly affect the environment (high impact) and (3) the growth rate of late-arriving species is highly dependent on the environment (high requirement). Emphasising multiple niche components, these hypotheses are firmly founded on classic niche-based theories of community assembly ( Grinnell 1917 ; Elton 1927 ; Gause 1932 ; Hutchinson 1957 ; MacArthur & Levins 1967 ; Pianka 1973 ; Tilman 1982 ; Leibold 1995 ; Chase & Leibold 2003 ). Yet, to our knowledge, niche components have never been distinguished in empirical studies of priority effects. The lack of relevant empirical studies is rather surprising because, as we argue, decomposing niches into components can be essential to explaining priority effects. For example, given the same species that arrives late, the strength of priority effects should depend not only on the requirement niche of that species but also on the impact niche of the species that arrives early. In a similar vein, even with high niche overlap, no strong priority effects are expected when the impact of early-arriving species is low, whereas even with low niche overlap, a species of high impact can cause strong priority effects. Consequently, it seems plausible that the role of niches in priority effects is obscured and left undetected if not decomposed into components. Experimental test of the niche component hypotheses To evaluate the niche-component hypotheses empirically, we conducted a simple experiment using four species of yeast that inhabit floral nectar ( Lachance et al . 2001 ). In this experiment, we assessed the effect of an early-arriving species on late-arriving species in multiple pairs of species in multiple environments that differed in resource richness (either a rich or poor supply of amino acids) and environmental harshness (either benign or harsh osmotic conditions). By measuring population growth and the changes imposed by yeast species on each environment, we quantified the niche components and linked these metrics to the measured strength of priority effects. To place our results in a broad context, data were analysed to determine the amount of variation that our hypotheses could explain over and above one conventional hypothesis that focuses on characteristics of the environment rather than those of the species. According to this hypothesis, priority effects will be stronger when the environment is less harsh, in the sense that species show higher growth rates under benign environmental conditions (e.g. Chase 2003 , 2007 ; Kardol et al . 2013 ).",
"discussion": "Discussion Taken together, these results show that niche components can collectively explain a large fraction of variation in the strength of priority effects. By decomposing niche components, it was possible to predict nearly twice as much variation as explained by phylogenetic relatedness or overall ecological similarity (cf. Peay et al . 2012 ). Furthermore, niche components were fairly robust and consistent predictors of priority effects even in the face of large differences in species growth and resource use among environments. Some species-specific patterns may have existed in our data. For example, priority effects tended to be strong when imposed by C. rancensis and M. reukaufii and when experienced by H. valbyensis , whereas S. bombicola tended to cause strong effects only in benign environments (Fig. 2 ). For the most part, however, species-specific effects appeared too complicated to provide systematic explanation (Fig. 2 ). Our analysis demonstrates that consideration of niche components across species helps to discern patterns in seemingly idiosyncratic variation in the strength of priority effects (Fig. 3 ). The potential scope of the niche-component hypotheses is broader than can be captured by the specific data from our experiment. For example, priority effects were mostly inhibitory in our experiment, but facilitative priority effects, where early-arriving species promote the growth of late colonisers, may also be common in natural communities ( Callaway & Walker 1997 ; Bruno et al . 2003 ). We suggest that both facilitative and inhibitory priority effects can be considered within the same niche-component framework. For instance, just as early-arriving species with high negative impact would cause strong inhibitory priority effects, those with high positive impact should cause strong facilitative priority effects. In addition, the data that we used to analyse priority effects in nectar yeasts mostly concerned interactions via changes in abiotic environmental conditions, such as nectar pH and resource availability. However, the niche-component hypotheses should be equally applicable to interactions via changes in biotic environmental conditions, such as predator and mutualist densities, caused by early-arriving species."
} | 3,048 |
34170119 | PMC8388114 | pmc | 9,102 | {
"abstract": "Molecular traffic\nacross lipid membranes is a vital process in\ncell biology that involves specialized biological pores with a great\nvariety of pore diameters, from fractions of a nanometer to >30\nnm.\nCreating artificial membrane pores covering similar size and complexity\nwill aid the understanding of transmembrane molecular transport in\ncells, while artificial pores are also a necessary ingredient for\nsynthetic cells. Here, we report the construction of DNA origami nanopores\nthat have an inner diameter as large as 30 nm. We developed methods\nto successfully insert these ultrawide pores into the lipid membrane\nof giant unilamellar vesicles (GUVs) by administering the pores concomitantly\nwith vesicle formation in an inverted-emulsion cDICE technique. The\nreconstituted pores permit the transmembrane diffusion of large macromolecules,\nsuch as folded proteins, which demonstrates the formation of large\nmembrane-spanning open pores. The pores are size selective, as dextran\nmolecules with a diameter up to 28 nm can traverse the pores, whereas\nlarger dextran molecules are blocked. By FRAP measurements and modeling\nof the GFP influx rate, we find that up to hundreds of pores can be\nfunctionally reconstituted into a single GUV. Our technique bears\ngreat potential for applications across different fields from biomimetics,\nto synthetic biology, to drug delivery.",
"conclusion": "Conclusions In\nthis work, we achieved functional reconstitution of ultrawide\nDNA origami pores in giant liposomes, obtaining stable transmembrane\npores of 30 nm inner diameter within the lipid bilayer. We successfully\nachieved the pore reconstitution by inserting the origami pore into\na lipid monolayer followed by bilayer formation. This methodology\nlowers the energy barrier for pore insertion, allowing functional\nreconstitution of pores with much larger diameter than previously\npossible. Our results show great potential for a number of biomimetic\napplications.\nFor example, future efforts can be directed toward the functionalization\nof the origami pore with FG-nucleoporins from the nuclear pore complex,\nthe macromolecular complex mediating transport across the nuclear\nenvelope in eukaryotes, 28 as reported in\nprevious works. 14 , 15 This will enable the creation\nof selective biomimetic pores that permit the exclusive transport\nof specific transport receptor proteins. While such biomimetic pores\nconstitute an exciting platform for investigating nucleocytoplasmic\ntransport in vitro , they may also be employed as\na building block for synthetic cells, thus contributing to the in vitro reconstitution of an artificial model of a nucleus.\nThis will, for example, aid the study of emergent properties of liposome-confined\ngenomes 29 by allowing the controlled injection\nof genome-processing proteins. Indeed, our DNA origami pores may be\na great asset for building a synthetic cell as they provide a stable\nportal that is large enough for biological macromolecules like proteins\nand RNAs to enter a liposome. The long-term stability and functionality\nof these liposomes is a particularly useful feature in such synthetic\nbiology applications. Finally, responsive drug delivery systems that\nrelease clinically relevant macromolecules such as antibodies upon\nstimulation by specific biomolecules could be envisioned, as well.\nWe expect that the ultrawide DNA origami nanopores may thus find a\nbroad variety of different applications.",
"introduction": "Introduction In the past decade,\nadvancements in the field of DNA nanotechnology\nhave enabled the fabrication of a great variety of “DNA origami”\nnanostructures, 1 , 2 including transmembrane structures\nthat resemble the biological pores found in cells. 3 Drawing inspiration from protein pores such as α-hemolysin, 4 artificial pores have been engineered with DNA\norigami 1 , 2 that can insert into lipid bilayers and\nallow for transmembrane diffusion of ions 5 , 6 and\nsmall molecules such as fluorophores, 7 , 8 DNA oligomers, 5 , 7 short PEG, 9 and dextran. 8 , 10 To favor partitioning into the lipid bilayer, a popular strategy\nrelies on the chemical modification of the outer nanopore surface\nwith hydrophobic groups, such as cholesterols, that anchor and stabilize\nthe strongly hydrophilic DNA-based nanopores into the lipid membrane. 3 As DNA origami structures can be designed with\nvirtually any shape or size 11 up to the\ndimensions comparable to those of viruses, 12 the approach bears great potential for various applications, from\nrecapitulating the function of complex enzymes like flippase, 13 to mimicking large protein transporters such\nas the nuclear pore complex. 14 , 15 However, so far,\nonly small pores with an inner diameter of a few\nnanometers have been realized, 8 , 10 which is because wider\npores are increasingly difficult to insert into a membrane. A commonly\nused strategy to reconstitute pores into a bilayer relies on spontaneous\ninsertion into a preformed lipid membrane. 3 This method relies on transient membrane instabilities 16 and becomes quickly ineffective for larger size\nobjects. This is because the work required to open up a hole in a\nmembrane grows quadratically with the diameter of the hole, 6 , 8 , 17 and spontaneous creation of such\na hole occurs in a Boltzmann weighted fashion, which exponentially\ndecreases the chances of success for inserting large pores into spontaneously\noccurring membrane defects. Scaling up the size of artificial pores\nthus requires an alternative insertion process that can circumvent\nthe barriers associated with spontaneous insertion. Here, we\novercome these challenges by employing a continuous droplet\ninterface crossing encapsulation (cDICE) technique 18 to incorporate ultrawide (∼30 nm inner diameter,\n∼55 nm outer diameter) DNA origami pores into the membrane\nof giant unilamellar vesicles (GUVs). As we show below, by administering\nthe pores concomitantly with vesicle formation, the pores can be localized\nefficiently at the membrane. To demonstrate the transmembrane transport\nthrough the pores, we measured the influx of multiple fluorescent\nmacromolecules with different sizes using confocal microscopy. We\nfind that the origami pores indeed support the transit of large (up\nto 28 nm) molecular structures, consistent with the 30 nm inner diameter\nof the pores. Using fluorescence recovery after photobleaching (FRAP),\nwe estimate that up to hundreds of pores can be functionally reconstituted\nin a single liposome. These ultrawide pores represent an exciting\ntool for various applications, from the mimicking of large biological\npores such as the nuclear pore complex to applications in synthetic\nbiology and drug delivery.",
"discussion": "Results and Discussion Design and Assembly of\nUltrawide DNA Origami Pores Using DNA origami folding of\na template strand, 1 , 2 we\ndesigned and assembled a rigid octagonal ring-like structure (see Figure 1 ). A multilayer DNA\norigami object was designed in square-lattice helical packing and\nfolded from a 7560 bases long scaffold single strand with 240 individual\noligonucleotide single strands, arranged in a 4 × 4 double-helical\npattern. It was designed to form a closed-loop octagonal shape, formed\nby 8 corner design motifs 19 with single-stranded\npoly-T strands at the corner sites. Deletions were introduced every\n32 bases to correct for global twist. 20 The origami pore was designed with a nominal outer diameter of 55\nnm, an inner diameter of 35 nm, and a height of 10 nm ( Figure 1 a), which represents the channel\nlength. The design includes 48 single-stranded handles that were distributed\nevenly on the interior surface facing the central cavity, as sites\nfor future functionalization purposes ( Figure S15 ). The octagon furthermore included 32 cholesterol modifications\non the outer surface that were arranged evenly with a ∼5.5\nnm horizontal spacing and ∼3 nm vertical distance between neighboring\ncholesterols ( Figure 1 a). An asymmetric feature was added ( Figure 1 b) consisting of four additional double helices\non top of the octagon, arranged in a 2 × 2 double helical pattern\non the top side of the octagon, spanning over 3 corners. The octagon\nfurther features 12 Atto647N dyes for fluorescence imaging, one at\neach corner on the bottom side of the octagon and four more on both\nends of the asymmetric feature ( Figure 1 a). We iteratively refined the octagon design using\ngel electrophoretic mobility analysis as the read-out in order to\nminimize the occurrence of higher-order aggregates and improve the\nfolding quality of the octagon object ( Figures\nS3–S10 ). Figure 1 DNA origami pore design and incorporation approach. (a)\nDNA origami\npore in front (left) and side view (right). Blue cylinders represent\nDNA double helices. The outer diameter is 55 nm, and the inner diameter\nis 35 nm. Green objects along the outer side of the DNA origami pore\nindicate the 32 cholesterol modifications. They vertically span over\na width of ∼3 nm (purposely slightly less than the hydrophobic\nthickness of ∼3.7 nm of a DOPC lipid bilayer 37 ). Black cylinders represent an asymmetric feature that\nwas introduced for class averaging of single particles from TEM. Violet\nspots along the outer surface of the DNA origami pore indicate the\npositions of 12 Atto647N dye modifications. The pore is designed with\na height of 10 nm. (b) Schematic representation of K 10 -PEG 5K coating of the origami pore. (c) Schematic representation\nof the DNA origami pore incorporated into the lipid bilayer of a GUV.\nLeft: GUV with pores incorporated in the lipid bilayer. Middle: zoom-in\non one of the incorporated origami pores resulting in a 30 nm diameter\nhole. Right: further zoom-in showing individual lipid molecules of\nthe bilayer; white spheres represent lipid heads, white strokes represent\nlipid tails, and the green object represents a cholesterol. The origami pore was coated with K 10 -PEG 5K molecules, which consist of 10 positively charged\nlysine amino acids\n(K) linked to a short polyethylene glycol chain (PEG, 5 kDa) ( Figure 1 b) as described in\nPonnuswamy et al. ( 21 ) This\ncoating serves two purposes: (i) it stabilizes the octagon in low-ionic-strength\nenvironments, where uncoated DNA origami structures would otherwise\ndisassemble, and (ii) it prevents the cholesterol-modified origami\npores from aggregating in bulk. Figure 1 c shows a schematic illustration of the incorporation\nof origami pores into the lipid membrane of a liposome (left). The\nzoom-ins highlight the interaction of the cholesterol-modified origami\npores with the lipid molecules. Characterization of the proper\nfolding and verification of the\nintended dimensions of the structure were performed using negative-stain\ntransmission electron microscopy (TEM) imaging and subsequent class\naveraging of single particles ( Figure 2 ). Figure 2 a shows a high-pass-filtered image of a typical field of view\nwith several octagon pores. The overall shape of the majority of the\nparticles matches the intended design, without obvious defects, in\nwhich 8 straight edges and 8 corners form an octagonal ring-like structure\nwith a ∼35 nm inner diameter and an ∼11 nm thickness.\nMost of the structures are oriented in a flat orientation on the surface,\nwhile a few can be seen in their side view. Figure 2 Negative-staining TEM\nand class averaging of the DNA origami pores.\n(a) Typical field of view of a negative-stained DNA origami sample.\nThe pores in this image were prepared as described in the Methods section. The image is high-pass filtered\nwith a radius of 20 nm and autoleveled in Photoshop. Scale bar: 50\nnm. (b) Top: Front view class average of N = 682\nsingle particles ( Figure S12 , left). Bottom:\nSide view class average of N = 80 single particles\n( Figure S12 , right). Scale bar: 25 nm.\n(c) Two exemplary slices of negative-stained TEM tomograms calculated\nfrom EM tilt series, showing octagons (indicated by white arrows)\nembedded in the membrane of liposomes at a z -height\naway from the TEM support surface. See Figure S23 for a z -stack of the TEM tomography data.\nScale bar: 50 nm. We computed 2D class\naverages from single-particle micrographs\n( Figure S12 ), which resulted in two distinct\nimages ( Figure 2 b),\ncorresponding to front and side view transmission projections of the\nparticles. The front view displays four distinct layers of double-stranded\nDNA. The asymmetric feature is visible on the outer two helices on\nthe top side ( Figure 2 b, top). The class-averaged image of the side view shows less detail\nof the expected 4 layers of DNA and the asymmetric feature. This is\nlikely due to the staining with uranyl formate and the tendency of\nthe structures to adhere to the grid surface predominantly with their\nfront side up, resulting in less side-view particles. From these images,\nwe measured an inner diameter of ∼35 nm and a thickness of\n∼11 nm ( Figure S13 ). Accounting\nfor the K 10 -PEG 5K coating by including a layer\nwith a thickness of 2.4 ± 1.3 nm, as measured by Ponnuswamy et al. , 21 would result in an\neffective inner diameter of ∼30 nm and an effective height\nof ∼16 nm including the unstructured PEG brushes. Class-averaged\nimages of K 10 -PEG 5K -coated and bare octagons\nmatch well ( Figure S14 ), which shows that\nthe coating preserves the global shape of the octagon pore. Liposomes with octagons were imaged using negative-stained EM tomography,\nshowing that the octagons can be well incorporated within the lipid\nmembrane ( Movie 4 and Movie 5 ). The tomograms\nwere calculated from the EM tilt series. Slices through the tomograms\nshow octagons located at different heights on the surface of the liposomes\n( Figure S23 ). The top slices of the tomograms\nshow octagons embedded in the lipid membrane on top of liposomes ( Figure 2 c). The global shape\nand the inner diameter of these octagons remained intact. DNA Origami\nPore Reconstitution in Liposomes by cDICE We first tested\nspontaneous insertions observed for our 55 nm (outer\ndiameter) large DNA origami pores into free-standing planar lipid\nbilayers or into preformed GUVs, under various conditions and with\ndifferent nanopore versions with either 32 or 64 cholesterols. As\nexpected, these attempts were unsuccessful. While membrane interactions\ndrove the DNA origami objects to adhere on the GUVs ( Figure S16 ), there was a total lack of influx of molecules\n( Figure S17 ), indicating that no actual\npores were created with a functional transmembrane opening. We therefore used an alternative approach where we incorporate the\nDNA origami pores during the process of the formation of the GUVs.\nWe employed an inverted emulsion technique called cDICE, 18 which yields unilamellar liposomes with good\nencapsulation of a large variety of macromolecules. 22 Briefly, in cDICE, layers of buffer solution and lipid-in-oil\nsuspension are deposited subsequently in a rotating chamber. As illustrated\nin Figure 3 a, an inner\nbuffer solution containing the origami pores is then injected through\na capillary into the rotating lipid-in-oil suspension at a constant\nflow rate. Shear forces due to the rotation result in detachment of\ndroplets from the capillary orifice. While traveling through the oil\nphase, these droplets acquire a first monolayer of lipids where the\norigami pores can insert. A second monolayer is subsequently formed\nwhen the droplets cross the oil–water interface, resulting\nin GUVs that are eventually collected from the outer buffer solution.\nIn this cDICE process, we thus aimed to reconstitute the pore within\nthe lipid monolayer that is formed in the water-in-oil droplets during\nthe first step of cDICE. Whereas the origami rings with 64 cholesterols\nmainly resulted in formation of large aggregates at the droplet interface\n( Figure S18 ), the rings with 32 cholesterols\nappeared to interact with the lipid monolayer without inducing significant\naggregation, suggesting a more promising route. Hereafter, we report\nthe results for the 32 cholesterol version of the DNA origami rings. Figure 3 Reconstitution\nof wide DNA origami pores in GUVs by cDICE. (a)\nSchematics of the cDICE workflow. A section of the rotating chamber\nis depicted, including the oil and outer buffer phases. Origami pores\nare indicated in yellow. (b) Comparison of the efficiency of DNA origami\npore localization at the membrane by addition of 2 MDa dextran in\nthe inner buffer. Without dextran, origami pores were observed to\nremain dispersed in bulk, whereas inclusion of dextran drove the origami\npores to the lipid surface. (c) FRAP analysis of origami pores reconstituted\nin GUV that shows that the pores are mobile within the lipid bilayer.\nThe dashed rectangle indicates the bleached area, which recovers within\na few minutes. (d) Example of a large field of GUVs with reconstituted\norigami pores. (e) Example of a GUV retaining the 2 MDa dextran in\nthe lumen after 24 h incubation in buffer. Scale bars: 10 μm. In a standard cDICE protocol, 22 the\ninner buffer solution that is injected contains either sucrose or\nOptiprep 22 to obtain an optimal density\nof the inner GUV solution. We found, however, that in these conditions\nthe reconstitution of 32 cholesterol origami rings was inefficient,\nwith most of the origami pores remaining in solution after vesicle\nformation ( Figure 3 b). When we included high molecular weight dextran (∼2 MDa)\nin the inner buffer, we found that 32 cholesterol origami rings could\nbe fully localized at the membrane ( Figure 3 b). Note that the choice of such a large\ndextran molecule (2 MDa) was also motivated by the high diameter of\ngyration (∼76 nm; ref ( 23 )), which makes it large enough to stay in the vesicle despite\nthe presence of the 30 nm wide DNA origami pores. The origami\npores appeared to be evenly distributed on the GUV\nsurface, only occasionally forming clusters, which then were also\nenriched in lipids ( Movie 1 and Movie 2 : z -stack and 3D rendering). Moreover, fluorescence\nrecovery after photobleaching (FRAP) analysis revealed that DNA origami\npores could freely diffuse within lipid bilayer of the vesicle membrane\n( Figure 3 c). Hundreds\nof GUVs could be obtained from a single preparation ( Figure 3 d). The resulting GUVs retained\nthe 2 MDa dextran over prolonged periods of time (>24 h, Figure 3 e) and were remarkably\nstable, remaining spherical in spite of osmotic changes, suggesting\nthat the origami pores were very well able to equilibrate osmotic\ndifferences by allowing osmolytes to rapidly cross the membrane. All\nin all, the data indicate the successful insertion of wide DNA origami\npores in the membrane of GUVs. Protein Influx through\nthe Pores Demonstrates Their Transmembrane\nTransport Capabilities To assess whether the DNA origami\npores were functional, that is, inserted in the intended orientation\nthat allows for transmembrane transport through the central cavity\nof the octagon, we tested the influx of IBB-mEGFP-Cys, a 34.6 kDa\nvariant of the soluble green fluorescent protein that includes an\nimportin β-binding domain 24 (see Supporting Information S22 ), henceforth referred\nto as “GFP” for simplicity. We measured the normalized\nfluorescence intensity difference I n,diff = ( I out – I in )/ I out of the vesicles, that\nis, the difference between in-vesicle intensity ( I in ) and the outer bulk ( I out ) divided by I out ( Figure 4 a) after overnight incubation of the GUVs\nin an outer solution with 2.3 μM of GFP. While poreless vesicles\n(control, Figure 4 b)\nremained empty, yielding a finite value I n,diff = 0.13 ± 0.05 (errors are standard deviation, N = 34), the pore-containing liposomes showed an influx of GFP for\nabout 50% of the vesicles ( Figure 4 c): half of the vesicles exhibited an almost complete\nsaturation to I n1,diff = 0.01 ± 0.03\n( N = 45), while the other half yielded I n2,diff = 0.09 ± 0.03 ( N = 45),\nthat is, no significant change compared to the control. We note that\nheterogeneity in GUV samples is a widely reported phenomenon 25 that apparently is associated with variations\nin the efficiency of pore insertion during vesicle formation in cDICE.\nMost importantly, a clear fraction of the liposomes unambiguously\nshowed transmembrane transport of the folded proteins, indicating\ntransport through the ultrawide DNA origami pores. Figure 4 Test of GFP influx through\norigami pores. (a) Schematic of the\ninflux experiment. Left: Sketch of a vesicle (red) containing pores\n(blue) that is incubated with GFP (green). Right: Fluorescence intensities I in and I out that\nare expected to be measured in a time-lapse experiment on the dashed\npurple and yellow areas, respectively. The right part of the graph\nillustrates the end-point intensities after ∼24 h, after an\ninitial transient. (b) Influx experiment with vesicles that were not\nsubjected to DNA pore insertion. Top right: sketch of a poreless vesicle\nthat excludes GFP. Top left: example of fluorescence image for a poreless\nvesicle (cyan) that prevents GFP (green) from entering the vesicle,\nshowing up as a darker area signifying the absence of GFP. Bottom:\nHistogram of the normalized end-point intensity difference ( I out – I in )/ I out for the poreless vesicles, verifying\nthat no vesicles showed GFP influx. (c) Influx experiment with vesicles\nthat were functionalized with origami pores. Top right: sketch of\na pore-containing vesicle filled with GFP. Top left: example of fluorescence\nimage for a pore-containing vesicle (cyan) that allows GFP to diffuse\n(green) into the vescicle, resulting in I out – I in ≈ 0. Bottom: Histogram\nshowing normalized end-point intensity difference ( I out – I in )/ I out for the poreless vesicles. We find that\n∼50% of the vesicle population manifested GFP influx. Scale\nbars: 10 μm. Next, we characterized\nthe GFP influx rate quantitatively using\na FRAP assay ( Figure 5 a). In this experiment, we selected pore-containing vesicles that\nshowed that influx occurred during the overnight incubation ( i.e. , those with I n,diff ≈\n0), then photobleached the inner part, and subsequently measured the\nrecovery of the fluorescence signal. An example of the FRAP experiment\nis shown in Figure 5 b (top) and Movie 3 . Starting from an\nequilibrium state where I in ≈ I out ( Figure 5 b, left), the vesicle fluorescence drops upon photobleaching\nto a level comparable to the one of an empty vesicle, to subsequently\nfully recover within a few minutes ( Figure 5 b, right). Notably, this time scale is much\nfaster compared to those of previously reported fluorescence recovery\nwith DNA origami pores. 7 , 8 A control with vesicles lacking\npores ( Figure 5 b, bottom)\nresulted in no appreciable recovery upon photobleaching. Figure 5 c displays I in and I out for\na pore-containing GUV as a function of time where, after initial transient\nbehavior (gray area), I out is constant,\nwhereas I in continues to slowly increase\nover time, which represents the influx of GFP through the ultrawide\nDNA origami pores. Figure 5 FRAP assay, modeling, and extraction of number of functional\npores.\n(a) Schematic of the FRAP assay: (1) the vesicle is left to equilibrate\nwith bulk solution containing GFP; (2) GFP molecules inside the vesicle\nare photobleached; (3) new GFP molecules are driven through the ultrawide\norigami pores from the outer solution, causing a recovery of the fluorescence\nsignal. (b) Example of a FRAP experiment. Top left: image showing\nthe pore-containing vesicle (origami pores in cyan) at equilibrium\nwith the outer GFP (green) solution. Top right: series of frames showing\nrecovery (within minutes) of the GFP signal after photobleaching.\nBottom left: image showing a pore-less vesicle (cyan) that excludes\nGFP (green) present in the outer solution. Bottom right: series of\nframes showing the absence of a signal increase after photobleaching.\n(c) Fluorescence intensities I in (red)\nand I out (purple) that reduce and recover\nafter photobleaching for the vesicle shown in (b). Gray area denotes\nthe initial transient, after which I out can be considered constant. Inset: Highlight of the regions where I in (red) and I out (purple) were measured. (d) Normalized intensity difference I n,diff over time for the vesicle illustrated\nin (b). Data are fitted by eq 4 . Gray area was excluded from the fit as the model assumes\na constant I out . For such a 13.3 μm\ndiameter vesicle, we estimated a number of functional pores N p = 250. (e) Number of origami pores extracted\nfrom fits for different vesicles, as a function of vesicle diameter\n(gray circles). Averaged data over 2.5 μm bins (red squares,\nerror bars are standard deviation) show an increase of the number\nof pores as a function of vesicle diameter. Black line represents\na cubic fit to the data. Scale bars: 10 μm. To obtain the number of functional pores, we model the influx rate\nusing a diffusion model derived from Fick’s first law: 26 1 which describes the flux ϕ\nof GFP molecules\nwith diffusion constant D , as a result of the concentration\ngradient d c /d x between the outer\n( c out ) and inner ( c in ) environment of the GUV. The flux ϕ is defined as\nnumber of molecules d N crossing a membrane area d A in a time d t . Since the vesicle volume V is constant over time, the in-vesicle concentration c in ( t ) is a time-dependent variable\nthat reflects the fact that the vesicle is filling up over time: . The outside concentration c out instead is assumed to be constant over time, which\nis reasonable given the large volume (∼200 μL) of the\nbulk solution as compared to the ∼picoliter volume of the vesicle.\nAssuming that the diffusion occurs through a number N p of DNA origami pores that each have a length L p and an area A p , yielding a total integrated area A = A p N p , we can rewrite eq 1 as the first-order differential\nequation 2 which, with c in ( t = 0) = c in, start , has a solution 3 Note, however, that\nfor the\ntransport through the pores, we need to employ an effective diffusion\nconstant D eff which is reduced compared\nto the bulk diffusion constant D to account for the\nconfined transport through the pores, as reported by Dechadilok and\nDeen. 27 For our pore dimensions, this results\nin almost a factor of 2 decrease of the GFP diffusivity, that is, D eff = 0.54 D bulk .\nUpon rearranging eq 3 , we then finally obtain 4 As the measured fluorescence\nintensities are a good measure for the protein concentrations of the\nfluorescent GFP, we can use eq 4 to fit the I n,diff ( t ) data, with c in,start / c out and N p as fit parameters. Figure 5 d provides an example,\nshowing an excellent fit to the FRAP data. From fitting FRAP curves\nacquired from different vesicles, we find that the estimated number\nof pores N p varies between a few to a\nfew hundred per GUV. Despite the substantial scatter in the extracted N p (gray circles in Figure 5 e), which we attribute to the stochasticity\ninherent to the vesicle production and origami insertion during cDICE,\nwe do observe an overall increase of N p with the vesicle diameter. Assuming that all origami pores would\nmove from the volume into the membrane during GUV formation, N p should scale with the cube of the vesicle\ndiameter, which is consistent with the behavior observed (black line\nin Figure 5 e). Ultrawide\nDNA Origami Pores Act As a Molecular Sieve with ∼30\nnm Cutoff To provide evidence that our DNA origami pores\nindeed form open channels with an effective size of the order of the\ninner pore diameter of 30 nm, we tested the influx of dextran-FITC\nmolecules with a variety of sizes, viz. , D g ≈ 15 nm (70 kDa), 22 nm (150 kDa),\n28 nm (250 kDa), 39 nm (500 kDa), and 76 nm (2 MDa) (see Figure 6 a), where D g is the diameter of gyration as measured by\nHanselmann and Burchard. 23 As in the GFP\ninflux experiment, we incubated the pore-containing vesicles with\n2 μM of dextran-FITC and measured the normalized intensity difference I n,diff after ∼24 h. This was done independently\nfor each of the 5 molecules, keeping the same molar concentration\nof dextran-FITC molecules in bulk solution. Similar to the GFP experiments\n( Figure 4 c), I n,diff of pore-containing vesicles yielded two\npopulations when incubated with 70 kDa dextran-FITC ( Figure 6 b, left), where 58% manifested\ninflux ( I n1,diff,70k = 0.02 ± 0.03,\nerrors are standard deviation, N = 47), while 42%\nremained empty ( I n2,diff,70k = 0.13 ±\n0.02, N = 34). Incubation with 150 kDa dextran-FITC\n( Figure 6 b, center-left)\nled to a similar result, with 47% GUVs that were filled ( I n1,diff,150k = 0.03 ± 0.03, N =\n24) against 53% of empty vesicles ( I n2,diff,150k = 0.11 ± 0.03, N = 27). Influx of the larger\n250 kDa dextran-FITC ( Figure 6 b, center) resulted in a reduced (19%), yet present, population\nof filled vesicles ( I n1,diff,250k = 0.01\n± 0.02, N = 8), while 79% ( I n2,diff,250k = 0.10 ± 0.02, N =\n34) showed no sign of influx. Importantly, this demonstrates that\nvery large (250 kDa; 28 nm) macromolecules can be transported across\nthe ultrawide DNA origami nanopores. By contrast, incubation with\nthe even larger 500 kDa and 2 MDa dextran-FITC ( Figure 6 b, center-right and right) resulted in only\none population of empty vesicles, with I n,diff,500k = 0.13 ± 0.03 ( N = 49) and I n,diff,2M = 0.09 ± 0.04 ( N = 98),\nrespectively, indicating a lack of any transmembrane transport during\n24 h. This is fully consistent with expectations as their size clearly\nexceeds the 30 nm size of the origami pore. It is also consistent\nwith the use of 2 MDa dextran as a macromolecular crowder (cf. Figure 3 e). Figure 6 Size selectivity of DNA\norigami nanopores. (a) Diameter of gyration\nof dextran D g vs molecular\nweight M w . Black line is a fit of D g = 0.072 × M w 0.48 (ref ( 23 )). Dextran sizes of 70 kDa, 150 kDa, 250 kDa, 500 kDa, and 2 MDa\nemployed in our study are highlighted in blue, yellow, magenta, and\nblack, respectively. Gray area underlies dextran sizes larger than\nthe origami pore (red) which are not expected to translocate. (b)\nHistograms showing normalized intensity difference of pore-containing\nvesicles after overnight incubation the dextran molecules. Blue indicate\nvesicles that showed influx of 70 kDa, yellow bars for 150 kDa, magenta\nbars for 250 kDa, whereas gray bars represent empty vesicles. Dashed\nlines represent hypothetical populations of vesicles that would have\nbeen present if influx of 500 kDa or 2 MDa through the origami pores\nhad occurred."
} | 7,625 |
31182799 | PMC6708440 | pmc | 9,103 | {
"abstract": "How bacteria colonise surfaces and how they distinguish the individuals around them are fundamental biological questions. Type IV pili are a widespread and multi-purpose class of cell surface polymers. Here we directly visualise the DNA-uptake pilus of Vibrio cholerae , which is produced specifically during growth upon its natural habitat - chitinous surfaces. As predicted, these pili are highly dynamic and retract prior to DNA-uptake during competence for natural transformation. Interestingly, DNA-uptake pili can also self-interact to mediate auto-aggregation. This capability is conserved in disease-causing pandemic strains, which typically encode the same major pilin subunit, PilA. Unexpectedly, however, we discovered that extensive strain-to-strain variability in PilA, present in environmental isolates, creates a set of highly specific interactions, enabling cells producing pili composed of different PilA subunits to distinguish between one another. We go on to show that DNA-uptake pili bind to chitinous surfaces, are required for chitin colonisation under flow, and that pili capable of self-interaction connect cells on chitin within dense pili networks. Our results suggest a model whereby DNA-uptake pili function to promote inter-bacterial interactions during surface colonisation. Moreover, they provide evidence that type IV pili could offer a simple and potentially widespread mechanism for bacterial kin recognition.",
"introduction": "Introduction How bacteria physically sense and interact with their environment is a fundamental problem in biology. Type IV pili (T4P) are cell surface polymers ideally suited to this task 1 . Composed of a single major pilin and assembled by widely distributed and conserved machinery, T4P exhibit extensive functional versatility, with roles in motility, DNA-uptake, surface sensing and adhesion 2 , 3 . Consequently, T4P are critical virulence factors for numerous important human pathogens including Vibrio cholerae , which causes the pandemic diarrhoeal disease cholera. In Gram-negative bacteria pilins are processed at the inner-membrane, extracted by the assembly machinery and polymerised into a helical pilus fibre that exits the cell surface through a gated outer-membrane pore; the secretin 4 – 7 . A key feature of T4P is their ability to undergo dynamic cycles of extension and retraction 8 , 9 , powered by the action of dedicated extension ( e.g. PilB) and retraction ( e.g. PilT) ATPases, which either add or liberate pilin subunits at the base 10 , 11 . These dynamics are essential for many T4P functions e.g. twitching-motility 8 , 9 . Thus, understanding how T4P function may yield insights valuable for understanding mechanisms of environmental survival and pathogenesis. Despite their multifunctional potential, pandemic V. cholerae strains typically encode three distinct T4P systems – two of which are well characterised. First, toxin co-regulated pili (TCP) serve a dual role as both a receptor for CTXφ bacteriophage 12 , which carries the cholera toxin genes, and as the primary human colonisation factor with multiple essential roles in infection involving adhesion and auto-aggregation on the intestinal cell surface 13 – 15 . Second, Mannose-sensitive haemagglutinin (MSHA) pili are involved in surface sensing and attachment and thus, are important in the initiation of biofilm formation 16 – 20 . Third, in its natural aquatic environment V. cholerae often associates with chitinous surfaces 21 , which are nutritious, foster biofilm formation and likely play a role in environmental dissemination and transmission to humans in cholera endemic regions 22 – 24 . Chitin utilisation triggers competence for natural transformation 25 , a widely used mode of horizontal gene transfer that allows bacteria to take up DNA from their environment, and which can thus, foster rapid bacterial evolution 26 . This requires the production of the Chitin-Regulated (ChiRP) or DNA-uptake pilus 24 , 25 . Importantly, in strains representative of the on-going 7th cholera pandemic such as A1552, the O1 El Tor clinical isolate used throughout this work, only MSHA pili are produced constitutively under laboratory conditions. We previously showed DNA-uptake pili form bona fide pili composed of the major subunit PilA and that transformation was dependent on the presumed retraction ATPase PilT 27 . However, the pilus itself is not sufficient for transformation and requires the concerted action of a periplasmic DNA-binding protein, ComEA 27 , 28 . Upon receipt of transforming DNA ComEA switches from a diffuse to focal localisation 28 , 29 . These findings, together with work in other organisms, led to a model in which pilus retraction facilitates DNA entry into the periplasm 30 , wherein ComEA acts as ‘ratchet’ to pull in the remaining DNA 28 . Subsequently, DNA transport across the inner-membrane occurs via a spatially coupled channel, ComEC 29 . Though this model is well supported by genetic experiments 27 and the similarly combined action of T4P and ComEA in other organisms 31 , 32 , direct evidence remained lacking. Here, using a recently validated cysteine-labelling approach 33 , we show that, as predicted, DNA-uptake pili are highly dynamic and that these dynamics are PilT-dependent. Unexpectedly, DNA-uptake pili can self-interact, resulting in auto-aggregation. Remarkably, specific interactions between divergent PilA subunits allow pili composed of different PilA subunits to distinguish between one-another, enabling a simple mechanism for kin recognition.",
"discussion": "Discussion Here we demonstrate that DNA-uptake pili are highly dynamic, that these dynamics are PilT-dependent, and that cells lacking pilT are minimally transformable, providing direct evidence for the longstanding model, whereby pilus retraction facilitates DNA-uptake. Indeed, our results on pilus dynamics are in close agreement with those recently reported by Ellison et al. , who notably, went on to demonstrate that the pilus binds directly to DNA 47 . The major finding of this work, however, is that DNA-uptake pili are able to interact and distinguish between one another in a sequence-specific manner. In liquid culture when retraction is deficient via the deletion of pilT this manifests as an exaggerated auto-aggregation phenotype, which we have used as a convenient tool to reveal and then investigate the ability of pili to self-interact. Since only a subpopulation of cells is piliated at any one time, and these pili are dynamic, eliminating retraction likely facilitates auto-aggregation by producing a homogenous population of hyper-piliated cells, thereby increasing the chances of interactions between pili. Work in Neisseria meningitidis , which auto-aggregates naturally at low levels but is dramatically enhanced by the deletion of pilT , supports this idea 48 . Importantly, however, the ability of pili to self-interact in artificial liquid culture conditions reflects a natural ability to interact on chitin surfaces. Indeed, under natural induction conditions on chitin surfaces, cells producing pili capable of aggregation elaborate multiple pili and form dense pili networks in an otherwise unmodified background, indicating that the chitin surface likely promotes interactions between pili. The proximity of cells to each other in the crowded surface environment might inherently foster these interactions. Alternatively, the altered physiology of cells growing on chitin might also impact pilus assembly via effects on the extension/retraction motors. Our discovery that DNA-uptake pili bind chitin and are required for chitin colonisation under flow suggests an important role in the aquatic environment. Consequently, we propose that the primary function of DNA-uptake pili in the environment is likely not for natural transformation but rather for chitin colonisation. Indeed, as hypothesised elsewhere 24 , 49 , since pilus production is dependent on an intact chitin-utilisation pathway, colonisation mechanisms using DNA-uptake pili would be inherently selective for (i) nutritious chitinous surfaces and (ii) favour the recruitment and retention of productive cells while excluding non-productive cells unable to make pili. In contrast, MSHA pili are produced constitutively and bind biotic and abiotic surfaces similarly 16 , 19 , 20 . Furthermore, the ability to interact not only with the surface but to mediate selective interactions with other cells would confer additional benefits that could act at multiple stages of colonisation. This might be particularly advantageous during early stages to bring smaller chitin particles together, and thus provide resistance to protozoan grazers, as well as at later stages to keep cells together during biofilm dispersal and thus, by allowing cells to arrive at a new niche in greater number, aid in persistence and dissemination. However, further work will be necessary to rule out the possibility that these interactions represent an ancient colonisation mechanism that has since been replaced. Nevertheless, the energetic cost of producing the pili networks we observe on chitin, the existence of a set of highly specific interactions, and the fact that in contrast to environmental isolates, pandemic strains all retain the same interaction-proficient PilA, argue for an on-going and important role. Interestingly, auto-aggregation by TCP during virulence is essential for host colonisation 13 – 15 , 43 , 44 . Furthermore, TCP networks encasing cells on the intestinal cell surface have been suggested to protect cells from host defences 50 . Given the similarities between the two systems, especially our observation of dense pili networks on colonised chitin surfaces, we propose that the DNA-uptake pilus might play an analogous role in the aquatic environment. Indeed, ingestion of colonised chitin particles is thought to facilitate transmission to humans 22 , 23 in cholera endemic areas and recovering cholera patients exhibit a strong immune response to PilA 51 . Thus, future work should investigate whether the networks of DNA-uptake pili we observed on chitin surfaces protect cells during this process. In N. gonorrhoeae , artificially varying the density or post-translational modification state of pili leads to a form of cell-sorting based on differential interaction forces between pili 52 . This effect is likely related to aggregate dispersal during its infective lifestyle but does not permit specific recognition per se 53 . In contrast, the discovery here that natural PilA variability controls the ability of pili to self-interact, and creates highly specific interactions, provides a direct mechanism for kin recognition. The best-studied examples of kin recognition in microorganisms all involve adhesins and some form of aggregation ( e.g. Flo1; Saccharomyces cerevisiae 54 , TgrB1-TgrC1; Dictyostelium discoideum 55 , and TraA; Myxobacteria 56 ). In evolutionary terms, these recognition mechanisms are classified as ‘greenbeards’ because the cue, recognition of the cue and the resulting cooperative activity are all encoded by the same gene 57 , 58 . The ability of DNA-uptake pili to recognise and interact with pili composed of the same kind of PilA fits this classification and is therefore a specific form of greenbeard recognition 57 , 58 . However, this form of recognition implies close identity only at the greenbeard locus and so is better referred to as kind recognition 57 , 58 . An important question going forward will be to understand what drives PilA diversity and how this is related to the type VI secretion system, which acts to kill non-kin bacteria 59 . Similarly, do those strains that lack the ability to self-interact employ alternative mechanisms ( e.g. up-regulation of biofilm production) to colonise chitin surfaces? Indeed, the apparent acquisition of an inhibitor of pilus interactions by some PilA ( e.g. ATCC25872/V52) hints that the ability to interact may not always be beneficial or else may reflect an adaptation to a specific niche. Therefore, future work should focus on how the pili networks we observed on chitin surfaces contribute to the ecology of V. cholerae , especially under environmental conditions. Indeed, we still know relatively little about its natural lifestyle on chitin, partly due to the inherent technical difficulties associated with manipulating these surfaces. Nevertheless, the demonstration that in liquid media, specific interactions between pili composed of different major pilins is sufficient to enable segregation, provides a robust proof-of-concept that T4P have the capacity to function as a recognition mechanism. Finally, the fact that (i) T4P are widespread, (ii) auto-aggregation via T4P has been reported in multiple species 60 and (iii) the major pilin subunit often varies 3 , 39 , raises the possibility that specific interactions between T4P could be relatively common and therefore represent an important contribution to bacterial kin recognition worthy of continued investigation."
} | 3,263 |
35541067 | PMC9083020 | pmc | 9,104 | {
"abstract": "In this work, a wide-range operating synaptic device based on organic ferroelectricity has been demonstrated. The device possesses a simple two-terminal structure by using a ferroelectric phase-separated polymer blend as the active layer and gold/indium tin oxide (ITO) as the top/bottom electrodes, and exhibits a distinctive history-dependent resistive switching behavior at room temperature. And the device with low energy consumption (∼50 fJ μm −2 per synaptic event) can provide a reliable synaptic function of potentiation, depression and the complex memory behavior simulation of differential responses to diverse stimulations. In addition, using simulations, the accuracy of 32 × 32 pixel image recognition is improved from 76.21% to 85.06% in the classical model Cifar-10 with 1024 levels of the device, which is an important step towards the higher performance goal in image recognition based on memristive neuromorphic networks.",
"conclusion": "Conclusions In conclusion, we have demonstrated a wide-range operating synaptic device with low energy consumption based on organic ferroelectricity for image recognition. The device possesses a simple structure of Au/CuPc/P(VDF-TrFE) + PFO/ITO, which exhibits a distinctive history-dependent resistive switching behavior at room temperature. And a minimum energy consumption as low as ∼50 nJ mm −2 per synaptic event was achieved. We estimate that a device with an area of 0.1 μm 2 may use as little as 5 fJ per operation, which is close to a natural synapse. The synaptic connection controlled by the applied pulse's features (number, width and height) has been implemented with the device, which closely emulates the memory behavior and depends heavily on the way in which the information is learned. This device can provide 1024 levels under continuous pulses, which are expected to improve the accuracy of 32 × 32 pixel image recognition to 85.06% from 76.21% of 256 levels by using simulations in the classical model Cifar-10. This is an important step towards a higher performance goal in image recognition based on memristive neuromorphic networks. In further study, more efforts can be focused on optimization of the device thickness and structure for lower energy consumption and wider-range operating performance. With the emerge of synaptic devices with more levels under continuous pulses in the future, the recognition accuracy would be further improved and more complex patterns could be recognized.",
"introduction": "Introduction With approximately 10 14 synapses, the extremely complex human cerebral cortex is difficult to implement on the hardware based on traditional semiconductor integrated circuit technology. 1 To provide comparable complexity and manageable power dissipation, hybrid neuromorphic networks composed of complementary metal-oxide semiconductor (CMOS) circuits and adjustable two-terminal resistive devices (synaptic devices) have been proposed. 2,3 Unlike conventional von Neumann type computers, these networks enable adaptable and high-efficiency computing by implementing massive parallelism at a physical level. 4,5 In biological neural networks, the realization of learning and memory behavior is essentially based on the adjustment of relative synaptic connection strengths ( i.e. , the synaptic weights) between pre- and postsynaptic neurons. 6,7 Similarly in this hybrid neuromorphic network, the synaptic devices which mimic biological synapses can provide a series of weight levels under continuous pulse stimulations. And each value of these weight levels can represent a certain synaptic weight reflecting the relative synaptic connection strength. In this way, the function of the human brain, such as image recognition, can be well realized. 1,4 For learning more complex input patterns and reducing the error rate of image recognition, wide-range operating ( i.e. more levels under continuous pulses) synaptic devices are desperately desired, which have the ability to provide more training epochs. 1,8 In a typical training epoch of an artificial neural network, patterns from the training set were applied in the form of voltage pulse, one by one, to the network's input, and its outputs in the form of current value were used to calculate the change of weight (conductance) through a series of training algorithms. 1 According to the result, a certain pulse sequence would be applied to the corresponding synaptic device to update weight. The more complex input patterns often need more training epochs in order to meet a higher performance goal (indicated by a smaller error). 9 So synaptic device with more weight levels under continuous pulses is becoming a focus of research, attracting a large number of researchers. To achieve these goals, several types of electronic devices based on organic materials with different mechanisms – including redox reaction, 10–13 charge trapping and de-trapping, 14,15 filamentary switching, 16,17 ferroelectric switching 18 and ion migration 19–22 – have been proposed as a synapse for neuromorphic computing application. In comparison to the inorganic counterparts, the organic devices distinguish themselves with low-cost, easy solution processability, large-area implementation, mechanical flexibility and ductility, and more importantly, tunable electronic properties by designing molecular structure and the possibility of forming self-assembled three dimensional (3D) structures. 23–27 In addition, current synaptic device still consume energy that is orders of magnitude greater than biological synapses, and reduction of energy consumption of artificial synapses remains a difficult challenge. 20 In this study, a wide-range operating synaptic device based on organic ferroelectricity with low energy consumption is demonstrated. As the fundamental requirements for mimicking biological synapse, 28 history-dependent synaptic behaviors of the device were observed in the experiment. In addition to the potentiation and depression of a biological synapse, the complex memory behavior of differential responses to diverse stimulations can also be well simulated with the device. This device can provide 1024 levels under continuous pulses, which are expected to improve the accuracy of 32 × 32 pixel image recognition to 85.06% from 76.21% of 256 levels by using simulations in the classical model Cifar-10. These demonstrations show the possibility of achieving a higher performance goal in image recognition based on memristive neuromorphic networks.",
"discussion": "Results and discussion Organic ferroelectric materials, represented by P(VDF-TrFE), distinguish themselves with its relatively low crystallization temperature (∼140 °C), a relatively large remanent polarization and a short switching time. 29 When blending the insulating ferroelectric P(VDF-TrFE) and a semiconducting polymer, a conducting composite with distinct ferroelectric and semiconducting areas can be obtained, which enable an independent tuning of the ferroelectric and conductive properties of the composite film. 29 Besides, it was demonstrated that the maximum barrier that can be switched by the P(VDF-TrFE) ferroelectric polarization is 1.6 eV. 30 For the PFO/ITO interface, the injection barrier, which is the difference between the ITO work function (∼4.7 eV) and PFO ionization potential (∼6.1 eV), is about 1.4 eV. This relatively high barrier is expected to generate more conductance levels due to a broader adjusting range of the injection barrier progressively polarized by P(VDF-TrFE) component under continuous pulses. On these foundations, a wide-range operating synaptic device with a simple two-terminal structure, by using a ferroelectric phase-separated polymer blend [P(VDF-TrFE) and PFO] as the active layer and Au/ITO as top/bottom electrodes, have been fabricated, which is illustrated schematically in Fig. 1a . More importantly, a layer of CuPc, which is a very robust material, is first used to fill up the rough interface of Au and the polymer blend. The film of CuPc can effectively prevent the previously reported formation of the metallic conducting channels, making the device more stable. 31 Fig. 1 (a) Schematic illustration of the Au/CuPc/P(VDF-TrFE) + PFO/ITO synaptic device. (b) Sketch of the operation mechanism of a polymeric ferroelectric interpenetrating network. Polarization of the ferroelectric phase leads to accumulation of charges in the organic semiconductor domain, which modulates the injection barrier of the semiconductor/electrode interface leading to the change of the device conductance. The positive (c) and negative (d) current–voltage characteristics of the synaptic device showing a distinctive history-dependent resistive switching behavior. The step width of voltage sweeps are 0.15 V and 0.06 V, respectively. Inset: an equivalent circuit model of the device. The device exhibits a distinctive history-dependent resistive switching behavior at room temperature, as plotted in the current–voltage ( I – V ) characteristics of Fig. 1c with the Au electrode grounded and the ITO electrode applied with a scanning voltage and of Fig. 1d with the opposite direction. The operation mechanism of a polymeric ferroelectric interpenetrating network is shown in Fig. 1b , the left of which corresponds to Fig. 1c while the right corresponds to Fig. 1d . As for Fig. 1c , the P(VDF-TrFE) component was initially polarized completely in the opposite direction by applying a voltage of −20 V with a period of time. Upon being subjected to each consecutive positive voltage sweep circle of 0 V → 10 V → 0 V at the ITO electrode, the polarization of P(VDF-TrFE) component was enhanced bit by bit. Owing to the poling, a negative polarization charge is built up in the ferroelectric P(VDF-TrFE). To neutralize the polarization charges, holes will accumulate in the semiconducting phase PFO and in the ITO electrode. The band diagram at the ITO electrode along the cross section A-B in Fig. 1b is shown in Fig. S1b of ESI. † The accumulated charge gives rise to strong band bending in the PFO semiconductor, which effectively reduces the injection barrier at the interface between ITO and PFO. The charge injection as indicated by the arrows in Fig. S1b † is then strongly enhanced and the contact becomes ohmic, resulting in the increase of the device conductance. 30–32 The same mechanism happens in the CuPc/PFO interface when applied the opposite direction voltage, the result of which is shown in Fig. 1d . And the retention time (over 24 h) of the device based on ferroelectric phase-separated blend have been demonstrated in the ref. 32 . That is to say, the main active interface is different under the different direction of applied voltage, which is ITO/PFO interface in Fig. 1c or CuPc/PFO interface in Fig. 1d . So the device can be equivalent to two memristor in series with a respective diode connected in parallel. An equivalent circuit model of the device is shown in Fig. 1c and d inset. As seen from Fig. 1c and d , both directions of voltage sweeps tends to increase the current. While in the actual training system, each synapse is often implemented with two synaptic devices in order to speed up the convergence process of training algorithm in the system. 1 In this case, synaptic weight ( W ) is characterized by the difference between the effective conductances of these two devices represented as G + and G − ( W = G + − G − ). The increase/decrease of synaptic weight ( W ) is characterized by the increase of G + / G − which is the conductance of each device. That is to say, as for a single device, only positive I – V characteristics will be used in the training system. In our device, there are two memristive mode M1 and M2 shown in Fig. 1c and d inset, which can be used separately or jointly such as the emulation of a biological synapse in the form of M1–M1, M2–M2 or M1–M2. The injection barrier between CuPc (∼5.3 eV) and PFO (∼6.1 eV) is 0.8 eV, lower than the interface ITO/PFO of 1.4 eV. Thus, M2 (main active interface is CuPc/PFO) has lower operating voltage 6 V and energy consumption (∼50 nJ mm −2 per synaptic event) than M1, while M1 has more levels under continuous pulses. These features make it possible to switch two modes by changing the direction of applied voltage according to needs. Further optimization of device performance can be done from two aspects. One is to reduce the size of device fabricated by lithography technology, and we estimate that a device with area of 0.1 μm 2 may use as little as 5 fJ per operation, which is close to a natural synapse (∼10 fJ per operation). 20 Another is to optimize the device thickness and structure, which can decrease both energy consumption and operating voltage. Besides, for high performance in image recognition, the positive mode M1 is used in the following experiment. The potentiation and depression of biological synapse, which are considered as the neurobiological basis of the human brain memory functions, 11 are well simulated with this device. As mentioned before, synaptic weight ( W ) is characterized by the difference between the effective conductances of two devices represented as G + and G − ( W = G + − G − ). At first, synaptic device G + was subjected to 200 consecutive positive pulses with the amplitude of 10 V, duration of 50 ms. And then synaptic device G − was immediately subjected to 200 consecutive positive pulses with the same amplitude and duration. The change trend of W in the whole process is shown in Fig. 2a , which is extremely smooth and stable without significant disturbance. The main reason is due to the relatively good polarization stability of P(VDF-TrFE), and meanwhile it can be polarized progressively under continuous pulses. Obviously, the synaptic connection has been well potentiated or depressed with the consecutive pulses, which lay a good foundation for the emulation of more complex learning and memory behavior. Fig. 2 (a) The potentiation and depression of synaptic weight W ( W = G + − G − ). The relative changes of conductance versus pulse numbers under (b) pulse voltage of 10 to 20 V, pulse width of 500 ms and (c) pulse voltage of 15 V, pulse width of 100 to 900 ms, respectively. In the human brain, memory strength depends on the way in which the information was learned. This can also be well simulated with the device. Here, the conductance of the device is analogous to the memory state, while the “learning” conditions ( i.e. , the stimulation conditions) can be varied by changing the pulse number, the pulse voltage and the pulse width. 33,34 The influence of the pulse number on the synaptic weight can be seen obviously in the first 200 pulses of Fig. 2a , which resembles the phenomenon that the more stimulations lead to the stronger synaptic connection. In order to demonstrate the effect of the pulse height and pulse width on the synaptic weight, the number of the applied pulses is fixed at a constant of 10 while changing the pulse voltage from 10 to 20 V with constant pulse width of 500 ms and the pulse width from 100 to 900 ms with constant pulse voltage of 15 V, respectively. The relative changes of conductance versus pulse numbers under different pulse voltage or pulse width were plotted in Fig. 2b or c . With the increase of the pulse height or the pulse width, the synaptic weight characterized by conductance of the device increase dramatically. The main reason is that the polarization of P(VDF-TrFE) component is enhanced more greatly under each pulse, leading to a sharply slump in injection barrier of ITO/PFO interface. These demonstrations indicate that synaptic connection depends heavily on the stimulation number and the stimulation strength ( i.e. , the pulse height and width), similar to the memory behavior of the human brain. To quantitatively demonstrate the extremely wide-range operating performance, the device was programmed by a series of identical positive pulses with the amplitude of 15 V, duration of 50 ms, and period of 200 ms. Fig. 3a shows continuous response of a single device to the 1024 spike pulses, which has good reproducibility among different devices with the same process parameters. The whole response curve is relatively smooth. During the first 400 pulses, the conductance increases sublinearly and the conductance difference between contiguous levels decreases gradually, the partial detail of which is shown in Fig. 3b . After then the increase of conductance is close to linear while the conductance difference between contiguous levels maintains at around 0.06 μS, which is shown in Fig. 3c . Essentially, the conductance change is attributed to the modulation of the injection barrier polarized by P(VDF-TrFE) component under applied pulse. 30–32 When the device was applied pulses continuously, the polarization of P(VDF-TrFE) component can be enhanced easily at first, and then become slowly, which lead to the trend of conductance change. Fig. 3 (a) The device was programmed by 1024 identical positive pulses with the amplitude of 15 V, duration of 50 ms, and period of 200 ms (b) and (c) the detail of the dashed frames in (a) shows the conductance difference between contiguous levels (Δ G c ). Compared with other three terminals synaptic device based on ferroelectric switching, our device has simpler structure and is more close to biological synapse in morphology. These features make it possible to form higher-density neuromorphic computing networks. Besides, in comparison to other synaptic device with similar structure, this device with comparable energy consumption can provide 1024 levels under continuous pulses, more than all results reported so far. In order to prove the great benefit of the device with 1024 levels to the image recognition, simulation experiment was carried out with a classical image dataset called Cifar-10 and its neural network model. Cifar-10 has 60 000 colorful images of 32 × 32 pixel in 10 classes, 50 000 images of which for training and 10 000 images for testing. 35,36 An example of image recognition with neural network model of 13 layers including five convolution layers is sketched in Fig. 4a . In this simulation experiments, a weight array composed of 145 578 weights is obtained at first after a series of training epochs, which brings almost the best recognition accuracy of 89.98% remaining basically unchanged with more training epochs. The weight distribution is shown in Fig. 4b . Then the weight array has been separately density-quantized in different bit (from 6 bit to 14 bit). Density quantization is a type of quantizing method, which is to divide the amount of weight values into equal parts and replace the value in each part with the intermediate. 37–39 The typical distributions of 10 bit and 8 bit are shown in Fig. 4c and d respectively. With the decrease of quantization bit, the weight distribution becomes more discrete and dispersed, improving system efficiency at the expense of the recognition accuracy. Finally, we import the processed weight array to the Cifar-10 model and take image recognition test, respectively. Fig. 4e shows the result of recognition accuracy with quantized weight array of different bits. As the number of bits increases from 6 to 14, the recognition accuracy is correspondingly 74.5%, 76.21%, 85.06%, 89.9% and 90.11%, rising rapidly at first and then tending to an upper limit that is almost the best recognition accuracy of the original. 10 bit (corresponding to 1024 levels) with the accuracy of 85.06% is much better than 8 bit (corresponding to 256 levels) with that of 76.21%, while 12 bit is almost close to the original. Obviously, a certain amount of weight levels is indispensable for improving the recognition accuracy. Fig. 4 (a) An example of image recognition process with neural network model of Cifar-10. The weight distribution of (b) the original, (c) 10 bit quantization and (d) 8 bit quantization. (e) The recognition accuracy during the testing process with the original weight array and quantized weight array of 6 bit to 14 bit."
} | 5,033 |
36419772 | PMC9681136 | pmc | 9,106 | {
"abstract": "Understanding the temporal effects of organic matter input and water influx on metal lability and translocation is critical to evaluate the success of the phytostabilization of metalliferous mine tailings. Trends of metal lability, e.g., V, Cr, Mn, Co, Ni, Cu, Zn, and Pb, were investigated for three years following a direct-planting phytostabilization trial at a Superfund mine tailings site in semi-arid central Arizona, USA. Unamended tailings were characterized by high concentrations (mmol kg −1 ) of Fe (2100), S (3100), As (41), Zn (39), and Pb (11), where As and Pb greatly exceeded non-residential soil remediation levels established by Arizona. Phytostabilization treatments included a no-compost control, 100 g kg −1 compost with seed, and 200 g kg −1 compost with and without seed to the top 20 cm of the tailings profile. All plots received supplemental irrigation, effectively doubling the mean annual precipitation. Tailings cores up to 90 cm were Collected at the time of planting and every summer for 3 years. The cores were sub-sectioned at 20 cm increments and analyzed via total digestion and an operationally defined sequential extraction for elemental analysis and the calculation of a mass transfer coefficient normalized to Ti as an assigned immobile element. The results indicate that Pb was recalcitrant and relatively immobile in the tailings environment for both the uncomposted control and composted treatments with a maximum variation in the total concentration of 9–14 mmol kg −1 among all samples. Metal lability and translocation above the redox boundary ( ca . 30 cm depth) was governed by acid generation, where surficial pH was measured as low as 2.7 ± 0.1 in year three and strongly correlated with the increased lability of Mn, Co, Ni, Cu, and Zn. There was no significant pH effect on the lability of V, Cr, or Pb. Translocation to depths was greatest for Mn and Co; however, Zn, Ni, Cr, and Cu were also mobilized. The addition of organic matter enhanced the mobilization of Cr from the near surface to 40–60 cm depth (pH > 6) over the three-year phytostabilization study compared to the control. The increased enrichment of some metals at 60–90 cm indicates that the long-term monitoring of elemental translocation is necessary to assess the efficacy of phytostabilization to contain subsurface metal contaminants and thereby protect the surrounding community from exposure.",
"conclusion": "5. Conclusions Although compost-assisted phytostabilization initially raised the tailings’ pH, oxidative weathering and acid generation potential in surface tailings produced progressive acidification which was the primary mechanism for the increased lability and translocation of metals over time. The mobilization of some metals across short distances (<1 m) from the surface to depths occurred in all treatments, even in phytostabilized plots that had an initial pH neutralization and buffering provided by OM addition. The establishment of a vegetation cap or the addition of compost may have contributed to an enhanced rate of metal translocation, especially for Mn and Co, relative to the irrigation-only control. However, sequential extraction analysis did not indicate that organic matter or vegetation influenced a change in the specific sequential extraction-assigned speciation pools for any of the metals investigated. Further, the magnitude of 1D translocation to depth was low, generally less than 20 cm over 3 y. While Pb and V remained largely recalcitrant, Mn, Cu, Cr, and Cu fate and mobility were governed by the stability of poorly crystalline (oxyhydr)oxide secondary mineral precipitates forming under oxidative conditions. Although Zn and Ni were shown to translocate to depths over time, there was not a significant relationship between mobilization and the labile pools considered. These results highlight the importance of considering both depth and time in assessing of the utility and success of phytostabilization for sequestering metal contaminants in near-surface mine tailings.",
"introduction": "1. Introduction Abandoned, metalliferous, sulfide mine tailings pose a major human health and environmental hazard due to anthropogenic disturbance and the associated introduction of contaminant-bearing minerals into the environment [ 1 – 3 ]. Legacy tailings exhibit nutrient deficiency (C, N, and P), acidic pH, elevated salt levels, and high concentrations of hazardous metals and metalloids such as lead, zinc, copper, mercury, cadmium, and arsenic [ 4 ], Together, these characteristics make tailings an inhospitable environment for plant and heterotrophic microbial growth [ 5 – 8 ], which keeps them devoid of vegetation establishment for decades to centuries following mining cessation, particularly in (semi-) arid regions. As a result, tailings are highly susceptible to fugitive dust emissions and rainwater runoff erosion, leading to the exposure of wildlife and humans to toxic elements [ 5 , 7 , 9 – 11 ]. The ever-increasing demand for mining commodities requires corresponding research into improved long-term and affordable management strategies for mine waste containment to avoid the degradation of the proximal environment; the development of such techniques is an essential component of environmentally sustainable mining strategies. Phytostabilization is such a strategy, offering long-term, vegetation-driven containment to sequester and immobilize contaminants in the subsurface and rhizosphere under an established vegetation cap [ 3 , 12 – 14 ]. The growth of vegetation cover is key to providing the long-term physical stabilization of the tailings. The cover decreases wind shear at the surface and provides stabilization of the tailings, reducing fugitive emissions of particulate matter. Ferric (sulf)hydroxide crusts, which have high affinity for metal(loid) sorption, can form on root surfaces in mine tailings and sequester labile metal(loids) [ 12 ]. Additionally, roots provide a supportive scaffold that reduces wind and water erosion, and transpiration from vegetation diminishes contaminant leaching [ 4 ], Many areas experiencing heavy mining activity are located in (semi-) arid regions where water throughflux is limited, and therefore, so is the production of acid mine drainage (AMD). Dry conditions lead to the persistence of secondary weathering minerals, metal contaminants, and acidity in tailings media, preventing natural plant establishment [ 3 , 15 , 16 ]. Secondary minerals in mine tailings play important roles in contaminant retention and stabilization [ 17 ], For example, sulfates and (oxyhydr)oxides have been shown to attenuate solubilized metals in acid mine drainage [ 18 – 22 ]. Successful phytostabilization causes persistent organic root exudates and leaf litter to serve as a carbon source to the tailing ecosystem, with positive feedback on developing soil characteristics. The aim of phytostabilization is to achieve a self-sustaining ecosystem that eschews additional (and costly) inputs. The key to the success of phytostabilization is the negligible accumulation of toxic elements into the food-web in the above-ground biomass to reduce contaminant exposure to local wildlife [ 8 , 13 , 14 ], Given the low pH, low nutrient capacity, and poor soil structure of sulfide-ore-derived mine tailings, amendments such as irrigation, biosolids, fertilizers, lime, and chemical adsorbents are frequently employed to enhance plant establishment and growth [ 14 , 23 – 27 ], Despite an interest in phytostabilization as a remediation approach for mine tailings, the availability of published, long-term (≥2 years) field studies that assess the impacts of phytostabilization of metals contained in mine waste in (semi-) arid climates is lacking [ 28 – 32 ], Among long-term field-scale phytostabilization studies, this is the first to assess metal lability and translocation across both temporal and spatial variances. Elements of interest for this study include those in high concentrations and trace levels. The Iron King Mine Humboldt Smelter site (IKMHSS) tailings contain high zinc and manganese levels, which are both essential micronutrients but are toxic at high activities, and high lead levels, a toxic metal with no safe concentration level. Nickel, copper, chromium, vanadium, and cobalt, which are also micronutrients with toxic potential, are present in the tailings, but in much lower concentrations (<4 mmol kg −1 ). The comparative analysis of enriched and trace element lability and translocation provided a comprehensive dataset to assess geochemical processes taking place in the subsurface. The principal aims of this investigation were to (i) evaluate changes in transition metal (V-Zn) and Pb lability and translocation at the field-scale in response to the phytostabilization of sulfidic tailings from the surface to 90 cm deep and (ii) assess phytostabilization as an effective long-term stabilization strategy for these metals in tailings in a (semi-) arid climate over three years of tailings remediation. The IKMHSS acid sulfide tailings [ 33 ], listed due to high arsenic and lead concentrations, are dominated by iron and sulfur (13 wt% and 12 wt% in the top 0.5 m, respectively) [ 17 , 34 ]. The oxidative dissolution of initially deposited pyrite (FeS 2 ) releases iron, sulfate, acid, and toxic elements, which drives the precipitation of secondary ferric minerals such as jarosite (KFe 3 +3 (OH) 6 (SO 4 ) 2 ), ferrihydrite (Fe 2 3+ O 3 ·0.5(H 2 O), and schwertmannite (Fe 16 3+ O 16 (OH) 12 (SO 4 ) 2 ). The wind-driven erosion of the IKMHSS surface tailings is known to transport toxic metal(loid)-containing particulates to the surrounding community [ 5 , 8 , 10 , 35 , 36 ], Toxic metal(loid) lability and transport are of particular concern at this site due to arsenic concentrations being measured above the allowed state limit for residential soils (10 mg kg −1 ) [ 37 ], The field-scale (appx. 1 ha) phytostabilization of the IKMHSS tailings, initiated in 2010, established the long-term feasibility and efficacy of the compost-assisted, direct planting of a vegetation cap to reduce off-site dust emissions and sequester toxic metalloids in place [ 4 , 38 ], Previous studies at this site showed that phytoremediation was especially effective in reducing emissions of fine particulate matter (e.g., PM 2.5 ), which have the greatest health risks and the highest potential for long-distance transport [ 38 ], The IKMHSS tailings remedial-strategy investigation of the subsurface fate and mobility of hazardous metals is a case study of the fate and mobility of metal contaminants in acidic sulfide mine waste in (semi-) arid regions worldwide. The tailings at IKMSS were sourced from regional (mostly Pb, Zn) mines (<100 km) with the mineralogy dominated by pyrite (FeS 2 ), ankerite (Ca[Fe, Mg](CO 3 ) 2 ), and quartz (SiO 2 ). Valuable ores in the area included chalcopyrite (CuFeS 2 ), galena (PbS), sphalerite (ZnS), and tennantite ([Cu, Fe] 12 As 4 S 13 ), as sources of zinc and lead, while gold and silver, occluded in pyrite and tennantite, respectively, were mined as secondary commodities [ 39 ]. The dominant non-sulfide “gangue” minerals (those having no commercial value) were ankerite, quartz, pyrite (up to 75% of the original vein), and arsenopyrite (FeAsS) [ 39 ]. The tailings have been weathering in place, unamended for half a century following the mine’s closure. A redox gradient with a boundary at ca. 30 cm depth with a transition zone from 25 to 54 cm depth was identified and associated with distinct depth-dependent geochemical and morphological changes. Surficial tailings were characterized by (oxyhydr)oxide and sulfate minerals ( ex . jarosite, ferrihydrite, schwertmannite, zincosite [ZnSO 4 ], and gypsum [CaSO 4 ·2H 2 O]), and metal(loid)s adsorbed to jarosite and ferrihydrite. Tailings deeper than 55 cm were characterized by carbonates and sulfides [ 17 , 34 ]. We hypothesized that the addition of labile carbon and a vegetative cap would help to buffer a more neutral pH, relative to unamended tailings, and decrease metal solubility. It was established that near-surface ferric (sulf)oxyhydroxide minerals with high adsorption capacity for metals sequestered arsenic and could also decrease metal mobility [ 4 ]. Therefore, it was expected that compost-assisted vegetation plots would show greater metal retention in the rhizosphere compared to unamended control plots due to plant-promoted pH buffering, sorption on existing ferric hydroxide surface sites, and organo-metallic complexation.",
"discussion": "4. Discussion This experiment investigated field-scale, direct-planting, compost-assisted phytostabilization in legacy acid sulfide tailings undergoing surficial oxidative weathering in a (semi-) arid environment. The temporal changes in metal lability and translocation to depth investigated at four time points over three years with four depth increments from 0 to 90 cm showed dynamic responses to remediation. Despite the variable solubilities and geochemical properties of the metals considered (V, Cr, Mn, Co, Ni, Cu, Zn, and Pb), it was hypothesized that secondary minerals formed during oxidative weathering (e.g., (oxyhydr)oxides and sulfates) would contribute to the overall attenuation of metals. It was expected that the abundance of secondary minerals in the surficial tailings above 20 cm and the near-neutral pH below 20 cm depth would lead to metal sequestration in the near surface with little translocation to the mostly unweathered 60–90 cm depth increment (Hayes et al., 2014; Root et al., 2015). The results show that 3 y phytostabilization treatments affect metal lability and mobility greatest at the surface 0–20 cm. Continued metal enrichment at intermediate depths (20–40 cm) after the establishment of vegetation indicates that evapotranspiration in the (semi-) arid climate does not prevent the downward flux of solution-phase-dissolved or colloidal metals, as has been shown to occur in hyper-arid climates [ 44 ], Although metals migrated from the top 20 cm to the horizon below, this extended field study demonstrated that metals are retained in the rhizosphere, as hypothesized, under OM and planting treatments. However, the final-year development of enrichment at 60–90 cm for Cr and Mn and depletion at shallow depths for V, Cu, Zn, and Pb indicate that further metal translocation with increased hydrologie throughflux may be a long-term concern. 4.1. Acidification The progressive acidification of amended tailings over the three-year study indicated that the initial pH buffering via OM (manure compost pH = 10.0 ± 0.1) was temporary, and tailings acidification progressed despite the establishment of vegetation. The temporary neutralization in the tailings via compost addition was not sufficient to prevent acid-mediated metal dissolution [ 45 , 46 ]. A Pearson’s correlation coefficient matrix of pH compared to the extractable metals showed the relationships between the highly labile water-soluble pools and acidity ( Table 2 ). Correlation constants at the 95% confidence interval from averaged quadruplicate samples of 0% compost control, 10% compost and seed, 20% compost, and 20% compost and seed for all depths and years were compared ( n = 64 averaged values) ( Table 2 ). The correlation coefficients revealed a strong (negative) relationship between tailings pH and lability. Water-soluble metals that form carbonates and sulfate salts (Mn, Co, Ni, Cu, and Zn) showed higher solubility at lower pH ( Table 2 ). It is well-documented that increasing pH plays an important role in decreasing the solubility of metal cations and adsorption to (oxyhydr)oxide and humic surfaces through surface functional group deprotonation [ 47 , 48 ]. Acidity showed very little correlation to the water solubility of V, Cr, or Pb, although the water-soluble pools of V and Cr were strongly correlated to each other ( Table 2 ). Metal lability in other systems of weathering mine tailings have shown relatively high solubility for Mn and Zn and low solubility of Pb under similar, but unamended, field conditions [ 9 , 45 , 49 ]. The stabilization of lead in acid mine tailings via incorporation into jarosite likely prevents lability, even under changing pH or organic matter additions [ 50 ]. 4.2. Metal Lability Enrichment was observed for Mn, Co, Ni, Cu, and Zn ( Figures S12 and S13 ) at >20 cm compared to the near surface. This is attributed to surficial mobilization (and depletion) with deposition (and enrichment) at depths due to downward translocation in response to irrigation. The release of these metals in the first steps of the sequential extraction showed higher surface lability that progressively decreased with depth. Temporal enrichment analysis (Δτ) confirmed that downward mobilization was taking place and was enhanced by phytostabilization ( Figure 3 ). Mass transfer from the surface to depths in the tailings was especially well-defined for Mn, Co, and Zn enrichment/depletion (τ) ( Figure S13 ) and dynamic mass transfer (Δτ) calculations ( Figure 3 ). Regression analysis was employed to investigate a statistically significant ( p < 0.05) linear relationship between Δτ and Δ lability at the 95% confidence interval for a two-tailed test. The relationship between lability and mobility over time was examined first by comparing the change in lability measured as the change in the labile pool concentration, versus the change in enrichment during the three-year study ( Figure 4 ). Second, the change in the recalcitrant but extractable pool of metals associated with poorly crystalline (oxyhydr)oxide secondary minerals versus the change in enrichment during the three-year study was analyzed ( Figure 5 ). The change over 3y of averaged field quadruplicates for all four depth increments of 0% control, 10% compost and seed, and 20% compost and seed ( n = 12) showed a significant relationship for the easily mobilized pools of Cr and Mn ( Figure 4 ). Among the metals mobilized in the labile fraction and the recalcitrant fraction, only Mn and Co exhibited a statistically significant linear relationship between temporal changes in lability and enrichment ( Figure 4 ). These results support the observation that Mn and Co exhibit the greatest mobilization compared to other metals and suggests that in the oxidative weathering environment, Mn and Co are the most susceptible to water solubility. Interestingly, there was no significant relationship between Δτ and Δ lability for Zn, Pb, Ni, Cu, Cr, or V. The relation between change in τ and lability over time was positive (slope) for Mn, Pb, Ni, and Co, but only significant at the 95% confidence interval for Mn and Co. A negative relationship was observed for Cu, Cr, and V. While Zn lability was strongly correlated with acidity, it showed no temporal relationship with enrichment or depletion. Evidence of a significant relationship between the change in enrichment and changes in the poorly crystalline (oxyhydr)oxide-associated Mn, Cu, Cr, and Co supports the hypothesis of secondary precipitates attenuating metals ( Figure 5 ). This observation is consistent with metal sequestration in other tailing systems [ 45 , 51 ], Comparing dynamic elemental enrichment/depletion and extractability over 3 y reveals that Mn and Co were most susceptible to mobilization from the tailings surface. Significant trends ( p < 0.05) in surficial depletion (and enrichment at depths) correlating to a change in extractability show that Mn and Co are leached from the near surface and deposited at greater depths. Copper in the near-surface oxic tailings showed increased lability and translocation to depths after phytostabilization, indicating that weathering and OM enhance copper mobility The zinc mobilized from the surface tailings was likely the divalent cation, released from sulfate salts and sorption sites on (oxyhydr)oxides at the start of this experiment. These sulfates and (oxyhydr)oxides are weathering products of the original zinc mineral sphalerite (ZnS) and secondary mineral smithsonite (ZnCO 3 ) [ 17 , 39 ]. Zinc lability correlated with decreasing pH. The greatest enrichment of zinc occurred at 40–60 cm in the control plots, indicating that phytostabilization decreased zinc mobility in the oxic zone of the tailings. Although Zn was observed to translocate from surface tailings to depths, a relationship with the operationally defined sequential extraction pools was not observed. 4.3. Recalcitrant Metals Lead, vanadium, and chromium were generally recalcitrant in the tailings and were only extracted at appreciable fractions from the pool associated with poorly crystalline (oxyhydr)oxide minerals. In a similar fashion, all three elements exhibited a lack of correlation between mobilization and pH ( Table 2 ). Vanadium and chromium likely originated as minor components in silicates, sulfides, and carbonates. Lead in deep tailings was predominantly in the form of the sulfide galena and carbonate cerussite (PbCO 3 ) [ 17 , 39 ]. Lead in the surface tailings mainly existed as plumbojarosite [ 17 ]. Although the lead mineralogy of the surface tailings differed significantly from the deep tailings, there was little difference in lead lability or mobility throughout the depths considered. This indicates that in the IKMHSS tailings, weathering and phytostabilization do not increase lead mobility. Similarly, vanadium exhibits both low lability and low mobilization with respect to treatment, time, and depth, generally being unaffected by oxidative weathering and phytostabilization. Although chromium exhibited high recalcitrance for all treatments and depths throughout the three-year study, translocation to depths was observed for composted treatments during phytostabilization and showed a statistically significant linear relationship with the pool associated with poorly crystalline (oxyhydr)oxides. Unlike in lead and vanadium, chromium release to porewaters and the promotion of translocation was enhanced during phytostabilization."
} | 5,513 |
36419772 | PMC9681136 | pmc | 9,106 | {
"abstract": "Understanding the temporal effects of organic matter input and water influx on metal lability and translocation is critical to evaluate the success of the phytostabilization of metalliferous mine tailings. Trends of metal lability, e.g., V, Cr, Mn, Co, Ni, Cu, Zn, and Pb, were investigated for three years following a direct-planting phytostabilization trial at a Superfund mine tailings site in semi-arid central Arizona, USA. Unamended tailings were characterized by high concentrations (mmol kg −1 ) of Fe (2100), S (3100), As (41), Zn (39), and Pb (11), where As and Pb greatly exceeded non-residential soil remediation levels established by Arizona. Phytostabilization treatments included a no-compost control, 100 g kg −1 compost with seed, and 200 g kg −1 compost with and without seed to the top 20 cm of the tailings profile. All plots received supplemental irrigation, effectively doubling the mean annual precipitation. Tailings cores up to 90 cm were Collected at the time of planting and every summer for 3 years. The cores were sub-sectioned at 20 cm increments and analyzed via total digestion and an operationally defined sequential extraction for elemental analysis and the calculation of a mass transfer coefficient normalized to Ti as an assigned immobile element. The results indicate that Pb was recalcitrant and relatively immobile in the tailings environment for both the uncomposted control and composted treatments with a maximum variation in the total concentration of 9–14 mmol kg −1 among all samples. Metal lability and translocation above the redox boundary ( ca . 30 cm depth) was governed by acid generation, where surficial pH was measured as low as 2.7 ± 0.1 in year three and strongly correlated with the increased lability of Mn, Co, Ni, Cu, and Zn. There was no significant pH effect on the lability of V, Cr, or Pb. Translocation to depths was greatest for Mn and Co; however, Zn, Ni, Cr, and Cu were also mobilized. The addition of organic matter enhanced the mobilization of Cr from the near surface to 40–60 cm depth (pH > 6) over the three-year phytostabilization study compared to the control. The increased enrichment of some metals at 60–90 cm indicates that the long-term monitoring of elemental translocation is necessary to assess the efficacy of phytostabilization to contain subsurface metal contaminants and thereby protect the surrounding community from exposure.",
"conclusion": "5. Conclusions Although compost-assisted phytostabilization initially raised the tailings’ pH, oxidative weathering and acid generation potential in surface tailings produced progressive acidification which was the primary mechanism for the increased lability and translocation of metals over time. The mobilization of some metals across short distances (<1 m) from the surface to depths occurred in all treatments, even in phytostabilized plots that had an initial pH neutralization and buffering provided by OM addition. The establishment of a vegetation cap or the addition of compost may have contributed to an enhanced rate of metal translocation, especially for Mn and Co, relative to the irrigation-only control. However, sequential extraction analysis did not indicate that organic matter or vegetation influenced a change in the specific sequential extraction-assigned speciation pools for any of the metals investigated. Further, the magnitude of 1D translocation to depth was low, generally less than 20 cm over 3 y. While Pb and V remained largely recalcitrant, Mn, Cu, Cr, and Cu fate and mobility were governed by the stability of poorly crystalline (oxyhydr)oxide secondary mineral precipitates forming under oxidative conditions. Although Zn and Ni were shown to translocate to depths over time, there was not a significant relationship between mobilization and the labile pools considered. These results highlight the importance of considering both depth and time in assessing of the utility and success of phytostabilization for sequestering metal contaminants in near-surface mine tailings.",
"introduction": "1. Introduction Abandoned, metalliferous, sulfide mine tailings pose a major human health and environmental hazard due to anthropogenic disturbance and the associated introduction of contaminant-bearing minerals into the environment [ 1 – 3 ]. Legacy tailings exhibit nutrient deficiency (C, N, and P), acidic pH, elevated salt levels, and high concentrations of hazardous metals and metalloids such as lead, zinc, copper, mercury, cadmium, and arsenic [ 4 ], Together, these characteristics make tailings an inhospitable environment for plant and heterotrophic microbial growth [ 5 – 8 ], which keeps them devoid of vegetation establishment for decades to centuries following mining cessation, particularly in (semi-) arid regions. As a result, tailings are highly susceptible to fugitive dust emissions and rainwater runoff erosion, leading to the exposure of wildlife and humans to toxic elements [ 5 , 7 , 9 – 11 ]. The ever-increasing demand for mining commodities requires corresponding research into improved long-term and affordable management strategies for mine waste containment to avoid the degradation of the proximal environment; the development of such techniques is an essential component of environmentally sustainable mining strategies. Phytostabilization is such a strategy, offering long-term, vegetation-driven containment to sequester and immobilize contaminants in the subsurface and rhizosphere under an established vegetation cap [ 3 , 12 – 14 ]. The growth of vegetation cover is key to providing the long-term physical stabilization of the tailings. The cover decreases wind shear at the surface and provides stabilization of the tailings, reducing fugitive emissions of particulate matter. Ferric (sulf)hydroxide crusts, which have high affinity for metal(loid) sorption, can form on root surfaces in mine tailings and sequester labile metal(loids) [ 12 ]. Additionally, roots provide a supportive scaffold that reduces wind and water erosion, and transpiration from vegetation diminishes contaminant leaching [ 4 ], Many areas experiencing heavy mining activity are located in (semi-) arid regions where water throughflux is limited, and therefore, so is the production of acid mine drainage (AMD). Dry conditions lead to the persistence of secondary weathering minerals, metal contaminants, and acidity in tailings media, preventing natural plant establishment [ 3 , 15 , 16 ]. Secondary minerals in mine tailings play important roles in contaminant retention and stabilization [ 17 ], For example, sulfates and (oxyhydr)oxides have been shown to attenuate solubilized metals in acid mine drainage [ 18 – 22 ]. Successful phytostabilization causes persistent organic root exudates and leaf litter to serve as a carbon source to the tailing ecosystem, with positive feedback on developing soil characteristics. The aim of phytostabilization is to achieve a self-sustaining ecosystem that eschews additional (and costly) inputs. The key to the success of phytostabilization is the negligible accumulation of toxic elements into the food-web in the above-ground biomass to reduce contaminant exposure to local wildlife [ 8 , 13 , 14 ], Given the low pH, low nutrient capacity, and poor soil structure of sulfide-ore-derived mine tailings, amendments such as irrigation, biosolids, fertilizers, lime, and chemical adsorbents are frequently employed to enhance plant establishment and growth [ 14 , 23 – 27 ], Despite an interest in phytostabilization as a remediation approach for mine tailings, the availability of published, long-term (≥2 years) field studies that assess the impacts of phytostabilization of metals contained in mine waste in (semi-) arid climates is lacking [ 28 – 32 ], Among long-term field-scale phytostabilization studies, this is the first to assess metal lability and translocation across both temporal and spatial variances. Elements of interest for this study include those in high concentrations and trace levels. The Iron King Mine Humboldt Smelter site (IKMHSS) tailings contain high zinc and manganese levels, which are both essential micronutrients but are toxic at high activities, and high lead levels, a toxic metal with no safe concentration level. Nickel, copper, chromium, vanadium, and cobalt, which are also micronutrients with toxic potential, are present in the tailings, but in much lower concentrations (<4 mmol kg −1 ). The comparative analysis of enriched and trace element lability and translocation provided a comprehensive dataset to assess geochemical processes taking place in the subsurface. The principal aims of this investigation were to (i) evaluate changes in transition metal (V-Zn) and Pb lability and translocation at the field-scale in response to the phytostabilization of sulfidic tailings from the surface to 90 cm deep and (ii) assess phytostabilization as an effective long-term stabilization strategy for these metals in tailings in a (semi-) arid climate over three years of tailings remediation. The IKMHSS acid sulfide tailings [ 33 ], listed due to high arsenic and lead concentrations, are dominated by iron and sulfur (13 wt% and 12 wt% in the top 0.5 m, respectively) [ 17 , 34 ]. The oxidative dissolution of initially deposited pyrite (FeS 2 ) releases iron, sulfate, acid, and toxic elements, which drives the precipitation of secondary ferric minerals such as jarosite (KFe 3 +3 (OH) 6 (SO 4 ) 2 ), ferrihydrite (Fe 2 3+ O 3 ·0.5(H 2 O), and schwertmannite (Fe 16 3+ O 16 (OH) 12 (SO 4 ) 2 ). The wind-driven erosion of the IKMHSS surface tailings is known to transport toxic metal(loid)-containing particulates to the surrounding community [ 5 , 8 , 10 , 35 , 36 ], Toxic metal(loid) lability and transport are of particular concern at this site due to arsenic concentrations being measured above the allowed state limit for residential soils (10 mg kg −1 ) [ 37 ], The field-scale (appx. 1 ha) phytostabilization of the IKMHSS tailings, initiated in 2010, established the long-term feasibility and efficacy of the compost-assisted, direct planting of a vegetation cap to reduce off-site dust emissions and sequester toxic metalloids in place [ 4 , 38 ], Previous studies at this site showed that phytoremediation was especially effective in reducing emissions of fine particulate matter (e.g., PM 2.5 ), which have the greatest health risks and the highest potential for long-distance transport [ 38 ], The IKMHSS tailings remedial-strategy investigation of the subsurface fate and mobility of hazardous metals is a case study of the fate and mobility of metal contaminants in acidic sulfide mine waste in (semi-) arid regions worldwide. The tailings at IKMSS were sourced from regional (mostly Pb, Zn) mines (<100 km) with the mineralogy dominated by pyrite (FeS 2 ), ankerite (Ca[Fe, Mg](CO 3 ) 2 ), and quartz (SiO 2 ). Valuable ores in the area included chalcopyrite (CuFeS 2 ), galena (PbS), sphalerite (ZnS), and tennantite ([Cu, Fe] 12 As 4 S 13 ), as sources of zinc and lead, while gold and silver, occluded in pyrite and tennantite, respectively, were mined as secondary commodities [ 39 ]. The dominant non-sulfide “gangue” minerals (those having no commercial value) were ankerite, quartz, pyrite (up to 75% of the original vein), and arsenopyrite (FeAsS) [ 39 ]. The tailings have been weathering in place, unamended for half a century following the mine’s closure. A redox gradient with a boundary at ca. 30 cm depth with a transition zone from 25 to 54 cm depth was identified and associated with distinct depth-dependent geochemical and morphological changes. Surficial tailings were characterized by (oxyhydr)oxide and sulfate minerals ( ex . jarosite, ferrihydrite, schwertmannite, zincosite [ZnSO 4 ], and gypsum [CaSO 4 ·2H 2 O]), and metal(loid)s adsorbed to jarosite and ferrihydrite. Tailings deeper than 55 cm were characterized by carbonates and sulfides [ 17 , 34 ]. We hypothesized that the addition of labile carbon and a vegetative cap would help to buffer a more neutral pH, relative to unamended tailings, and decrease metal solubility. It was established that near-surface ferric (sulf)oxyhydroxide minerals with high adsorption capacity for metals sequestered arsenic and could also decrease metal mobility [ 4 ]. Therefore, it was expected that compost-assisted vegetation plots would show greater metal retention in the rhizosphere compared to unamended control plots due to plant-promoted pH buffering, sorption on existing ferric hydroxide surface sites, and organo-metallic complexation.",
"discussion": "4. Discussion This experiment investigated field-scale, direct-planting, compost-assisted phytostabilization in legacy acid sulfide tailings undergoing surficial oxidative weathering in a (semi-) arid environment. The temporal changes in metal lability and translocation to depth investigated at four time points over three years with four depth increments from 0 to 90 cm showed dynamic responses to remediation. Despite the variable solubilities and geochemical properties of the metals considered (V, Cr, Mn, Co, Ni, Cu, Zn, and Pb), it was hypothesized that secondary minerals formed during oxidative weathering (e.g., (oxyhydr)oxides and sulfates) would contribute to the overall attenuation of metals. It was expected that the abundance of secondary minerals in the surficial tailings above 20 cm and the near-neutral pH below 20 cm depth would lead to metal sequestration in the near surface with little translocation to the mostly unweathered 60–90 cm depth increment (Hayes et al., 2014; Root et al., 2015). The results show that 3 y phytostabilization treatments affect metal lability and mobility greatest at the surface 0–20 cm. Continued metal enrichment at intermediate depths (20–40 cm) after the establishment of vegetation indicates that evapotranspiration in the (semi-) arid climate does not prevent the downward flux of solution-phase-dissolved or colloidal metals, as has been shown to occur in hyper-arid climates [ 44 ], Although metals migrated from the top 20 cm to the horizon below, this extended field study demonstrated that metals are retained in the rhizosphere, as hypothesized, under OM and planting treatments. However, the final-year development of enrichment at 60–90 cm for Cr and Mn and depletion at shallow depths for V, Cu, Zn, and Pb indicate that further metal translocation with increased hydrologie throughflux may be a long-term concern. 4.1. Acidification The progressive acidification of amended tailings over the three-year study indicated that the initial pH buffering via OM (manure compost pH = 10.0 ± 0.1) was temporary, and tailings acidification progressed despite the establishment of vegetation. The temporary neutralization in the tailings via compost addition was not sufficient to prevent acid-mediated metal dissolution [ 45 , 46 ]. A Pearson’s correlation coefficient matrix of pH compared to the extractable metals showed the relationships between the highly labile water-soluble pools and acidity ( Table 2 ). Correlation constants at the 95% confidence interval from averaged quadruplicate samples of 0% compost control, 10% compost and seed, 20% compost, and 20% compost and seed for all depths and years were compared ( n = 64 averaged values) ( Table 2 ). The correlation coefficients revealed a strong (negative) relationship between tailings pH and lability. Water-soluble metals that form carbonates and sulfate salts (Mn, Co, Ni, Cu, and Zn) showed higher solubility at lower pH ( Table 2 ). It is well-documented that increasing pH plays an important role in decreasing the solubility of metal cations and adsorption to (oxyhydr)oxide and humic surfaces through surface functional group deprotonation [ 47 , 48 ]. Acidity showed very little correlation to the water solubility of V, Cr, or Pb, although the water-soluble pools of V and Cr were strongly correlated to each other ( Table 2 ). Metal lability in other systems of weathering mine tailings have shown relatively high solubility for Mn and Zn and low solubility of Pb under similar, but unamended, field conditions [ 9 , 45 , 49 ]. The stabilization of lead in acid mine tailings via incorporation into jarosite likely prevents lability, even under changing pH or organic matter additions [ 50 ]. 4.2. Metal Lability Enrichment was observed for Mn, Co, Ni, Cu, and Zn ( Figures S12 and S13 ) at >20 cm compared to the near surface. This is attributed to surficial mobilization (and depletion) with deposition (and enrichment) at depths due to downward translocation in response to irrigation. The release of these metals in the first steps of the sequential extraction showed higher surface lability that progressively decreased with depth. Temporal enrichment analysis (Δτ) confirmed that downward mobilization was taking place and was enhanced by phytostabilization ( Figure 3 ). Mass transfer from the surface to depths in the tailings was especially well-defined for Mn, Co, and Zn enrichment/depletion (τ) ( Figure S13 ) and dynamic mass transfer (Δτ) calculations ( Figure 3 ). Regression analysis was employed to investigate a statistically significant ( p < 0.05) linear relationship between Δτ and Δ lability at the 95% confidence interval for a two-tailed test. The relationship between lability and mobility over time was examined first by comparing the change in lability measured as the change in the labile pool concentration, versus the change in enrichment during the three-year study ( Figure 4 ). Second, the change in the recalcitrant but extractable pool of metals associated with poorly crystalline (oxyhydr)oxide secondary minerals versus the change in enrichment during the three-year study was analyzed ( Figure 5 ). The change over 3y of averaged field quadruplicates for all four depth increments of 0% control, 10% compost and seed, and 20% compost and seed ( n = 12) showed a significant relationship for the easily mobilized pools of Cr and Mn ( Figure 4 ). Among the metals mobilized in the labile fraction and the recalcitrant fraction, only Mn and Co exhibited a statistically significant linear relationship between temporal changes in lability and enrichment ( Figure 4 ). These results support the observation that Mn and Co exhibit the greatest mobilization compared to other metals and suggests that in the oxidative weathering environment, Mn and Co are the most susceptible to water solubility. Interestingly, there was no significant relationship between Δτ and Δ lability for Zn, Pb, Ni, Cu, Cr, or V. The relation between change in τ and lability over time was positive (slope) for Mn, Pb, Ni, and Co, but only significant at the 95% confidence interval for Mn and Co. A negative relationship was observed for Cu, Cr, and V. While Zn lability was strongly correlated with acidity, it showed no temporal relationship with enrichment or depletion. Evidence of a significant relationship between the change in enrichment and changes in the poorly crystalline (oxyhydr)oxide-associated Mn, Cu, Cr, and Co supports the hypothesis of secondary precipitates attenuating metals ( Figure 5 ). This observation is consistent with metal sequestration in other tailing systems [ 45 , 51 ], Comparing dynamic elemental enrichment/depletion and extractability over 3 y reveals that Mn and Co were most susceptible to mobilization from the tailings surface. Significant trends ( p < 0.05) in surficial depletion (and enrichment at depths) correlating to a change in extractability show that Mn and Co are leached from the near surface and deposited at greater depths. Copper in the near-surface oxic tailings showed increased lability and translocation to depths after phytostabilization, indicating that weathering and OM enhance copper mobility The zinc mobilized from the surface tailings was likely the divalent cation, released from sulfate salts and sorption sites on (oxyhydr)oxides at the start of this experiment. These sulfates and (oxyhydr)oxides are weathering products of the original zinc mineral sphalerite (ZnS) and secondary mineral smithsonite (ZnCO 3 ) [ 17 , 39 ]. Zinc lability correlated with decreasing pH. The greatest enrichment of zinc occurred at 40–60 cm in the control plots, indicating that phytostabilization decreased zinc mobility in the oxic zone of the tailings. Although Zn was observed to translocate from surface tailings to depths, a relationship with the operationally defined sequential extraction pools was not observed. 4.3. Recalcitrant Metals Lead, vanadium, and chromium were generally recalcitrant in the tailings and were only extracted at appreciable fractions from the pool associated with poorly crystalline (oxyhydr)oxide minerals. In a similar fashion, all three elements exhibited a lack of correlation between mobilization and pH ( Table 2 ). Vanadium and chromium likely originated as minor components in silicates, sulfides, and carbonates. Lead in deep tailings was predominantly in the form of the sulfide galena and carbonate cerussite (PbCO 3 ) [ 17 , 39 ]. Lead in the surface tailings mainly existed as plumbojarosite [ 17 ]. Although the lead mineralogy of the surface tailings differed significantly from the deep tailings, there was little difference in lead lability or mobility throughout the depths considered. This indicates that in the IKMHSS tailings, weathering and phytostabilization do not increase lead mobility. Similarly, vanadium exhibits both low lability and low mobilization with respect to treatment, time, and depth, generally being unaffected by oxidative weathering and phytostabilization. Although chromium exhibited high recalcitrance for all treatments and depths throughout the three-year study, translocation to depths was observed for composted treatments during phytostabilization and showed a statistically significant linear relationship with the pool associated with poorly crystalline (oxyhydr)oxides. Unlike in lead and vanadium, chromium release to porewaters and the promotion of translocation was enhanced during phytostabilization."
} | 5,513 |
36419772 | PMC9681136 | pmc | 9,107 | {
"abstract": "Understanding the temporal effects of organic matter input and water influx on metal lability and translocation is critical to evaluate the success of the phytostabilization of metalliferous mine tailings. Trends of metal lability, e.g., V, Cr, Mn, Co, Ni, Cu, Zn, and Pb, were investigated for three years following a direct-planting phytostabilization trial at a Superfund mine tailings site in semi-arid central Arizona, USA. Unamended tailings were characterized by high concentrations (mmol kg −1 ) of Fe (2100), S (3100), As (41), Zn (39), and Pb (11), where As and Pb greatly exceeded non-residential soil remediation levels established by Arizona. Phytostabilization treatments included a no-compost control, 100 g kg −1 compost with seed, and 200 g kg −1 compost with and without seed to the top 20 cm of the tailings profile. All plots received supplemental irrigation, effectively doubling the mean annual precipitation. Tailings cores up to 90 cm were Collected at the time of planting and every summer for 3 years. The cores were sub-sectioned at 20 cm increments and analyzed via total digestion and an operationally defined sequential extraction for elemental analysis and the calculation of a mass transfer coefficient normalized to Ti as an assigned immobile element. The results indicate that Pb was recalcitrant and relatively immobile in the tailings environment for both the uncomposted control and composted treatments with a maximum variation in the total concentration of 9–14 mmol kg −1 among all samples. Metal lability and translocation above the redox boundary ( ca . 30 cm depth) was governed by acid generation, where surficial pH was measured as low as 2.7 ± 0.1 in year three and strongly correlated with the increased lability of Mn, Co, Ni, Cu, and Zn. There was no significant pH effect on the lability of V, Cr, or Pb. Translocation to depths was greatest for Mn and Co; however, Zn, Ni, Cr, and Cu were also mobilized. The addition of organic matter enhanced the mobilization of Cr from the near surface to 40–60 cm depth (pH > 6) over the three-year phytostabilization study compared to the control. The increased enrichment of some metals at 60–90 cm indicates that the long-term monitoring of elemental translocation is necessary to assess the efficacy of phytostabilization to contain subsurface metal contaminants and thereby protect the surrounding community from exposure.",
"conclusion": "5. Conclusions Although compost-assisted phytostabilization initially raised the tailings’ pH, oxidative weathering and acid generation potential in surface tailings produced progressive acidification which was the primary mechanism for the increased lability and translocation of metals over time. The mobilization of some metals across short distances (<1 m) from the surface to depths occurred in all treatments, even in phytostabilized plots that had an initial pH neutralization and buffering provided by OM addition. The establishment of a vegetation cap or the addition of compost may have contributed to an enhanced rate of metal translocation, especially for Mn and Co, relative to the irrigation-only control. However, sequential extraction analysis did not indicate that organic matter or vegetation influenced a change in the specific sequential extraction-assigned speciation pools for any of the metals investigated. Further, the magnitude of 1D translocation to depth was low, generally less than 20 cm over 3 y. While Pb and V remained largely recalcitrant, Mn, Cu, Cr, and Cu fate and mobility were governed by the stability of poorly crystalline (oxyhydr)oxide secondary mineral precipitates forming under oxidative conditions. Although Zn and Ni were shown to translocate to depths over time, there was not a significant relationship between mobilization and the labile pools considered. These results highlight the importance of considering both depth and time in assessing of the utility and success of phytostabilization for sequestering metal contaminants in near-surface mine tailings.",
"introduction": "1. Introduction Abandoned, metalliferous, sulfide mine tailings pose a major human health and environmental hazard due to anthropogenic disturbance and the associated introduction of contaminant-bearing minerals into the environment [ 1 – 3 ]. Legacy tailings exhibit nutrient deficiency (C, N, and P), acidic pH, elevated salt levels, and high concentrations of hazardous metals and metalloids such as lead, zinc, copper, mercury, cadmium, and arsenic [ 4 ], Together, these characteristics make tailings an inhospitable environment for plant and heterotrophic microbial growth [ 5 – 8 ], which keeps them devoid of vegetation establishment for decades to centuries following mining cessation, particularly in (semi-) arid regions. As a result, tailings are highly susceptible to fugitive dust emissions and rainwater runoff erosion, leading to the exposure of wildlife and humans to toxic elements [ 5 , 7 , 9 – 11 ]. The ever-increasing demand for mining commodities requires corresponding research into improved long-term and affordable management strategies for mine waste containment to avoid the degradation of the proximal environment; the development of such techniques is an essential component of environmentally sustainable mining strategies. Phytostabilization is such a strategy, offering long-term, vegetation-driven containment to sequester and immobilize contaminants in the subsurface and rhizosphere under an established vegetation cap [ 3 , 12 – 14 ]. The growth of vegetation cover is key to providing the long-term physical stabilization of the tailings. The cover decreases wind shear at the surface and provides stabilization of the tailings, reducing fugitive emissions of particulate matter. Ferric (sulf)hydroxide crusts, which have high affinity for metal(loid) sorption, can form on root surfaces in mine tailings and sequester labile metal(loids) [ 12 ]. Additionally, roots provide a supportive scaffold that reduces wind and water erosion, and transpiration from vegetation diminishes contaminant leaching [ 4 ], Many areas experiencing heavy mining activity are located in (semi-) arid regions where water throughflux is limited, and therefore, so is the production of acid mine drainage (AMD). Dry conditions lead to the persistence of secondary weathering minerals, metal contaminants, and acidity in tailings media, preventing natural plant establishment [ 3 , 15 , 16 ]. Secondary minerals in mine tailings play important roles in contaminant retention and stabilization [ 17 ], For example, sulfates and (oxyhydr)oxides have been shown to attenuate solubilized metals in acid mine drainage [ 18 – 22 ]. Successful phytostabilization causes persistent organic root exudates and leaf litter to serve as a carbon source to the tailing ecosystem, with positive feedback on developing soil characteristics. The aim of phytostabilization is to achieve a self-sustaining ecosystem that eschews additional (and costly) inputs. The key to the success of phytostabilization is the negligible accumulation of toxic elements into the food-web in the above-ground biomass to reduce contaminant exposure to local wildlife [ 8 , 13 , 14 ], Given the low pH, low nutrient capacity, and poor soil structure of sulfide-ore-derived mine tailings, amendments such as irrigation, biosolids, fertilizers, lime, and chemical adsorbents are frequently employed to enhance plant establishment and growth [ 14 , 23 – 27 ], Despite an interest in phytostabilization as a remediation approach for mine tailings, the availability of published, long-term (≥2 years) field studies that assess the impacts of phytostabilization of metals contained in mine waste in (semi-) arid climates is lacking [ 28 – 32 ], Among long-term field-scale phytostabilization studies, this is the first to assess metal lability and translocation across both temporal and spatial variances. Elements of interest for this study include those in high concentrations and trace levels. The Iron King Mine Humboldt Smelter site (IKMHSS) tailings contain high zinc and manganese levels, which are both essential micronutrients but are toxic at high activities, and high lead levels, a toxic metal with no safe concentration level. Nickel, copper, chromium, vanadium, and cobalt, which are also micronutrients with toxic potential, are present in the tailings, but in much lower concentrations (<4 mmol kg −1 ). The comparative analysis of enriched and trace element lability and translocation provided a comprehensive dataset to assess geochemical processes taking place in the subsurface. The principal aims of this investigation were to (i) evaluate changes in transition metal (V-Zn) and Pb lability and translocation at the field-scale in response to the phytostabilization of sulfidic tailings from the surface to 90 cm deep and (ii) assess phytostabilization as an effective long-term stabilization strategy for these metals in tailings in a (semi-) arid climate over three years of tailings remediation. The IKMHSS acid sulfide tailings [ 33 ], listed due to high arsenic and lead concentrations, are dominated by iron and sulfur (13 wt% and 12 wt% in the top 0.5 m, respectively) [ 17 , 34 ]. The oxidative dissolution of initially deposited pyrite (FeS 2 ) releases iron, sulfate, acid, and toxic elements, which drives the precipitation of secondary ferric minerals such as jarosite (KFe 3 +3 (OH) 6 (SO 4 ) 2 ), ferrihydrite (Fe 2 3+ O 3 ·0.5(H 2 O), and schwertmannite (Fe 16 3+ O 16 (OH) 12 (SO 4 ) 2 ). The wind-driven erosion of the IKMHSS surface tailings is known to transport toxic metal(loid)-containing particulates to the surrounding community [ 5 , 8 , 10 , 35 , 36 ], Toxic metal(loid) lability and transport are of particular concern at this site due to arsenic concentrations being measured above the allowed state limit for residential soils (10 mg kg −1 ) [ 37 ], The field-scale (appx. 1 ha) phytostabilization of the IKMHSS tailings, initiated in 2010, established the long-term feasibility and efficacy of the compost-assisted, direct planting of a vegetation cap to reduce off-site dust emissions and sequester toxic metalloids in place [ 4 , 38 ], Previous studies at this site showed that phytoremediation was especially effective in reducing emissions of fine particulate matter (e.g., PM 2.5 ), which have the greatest health risks and the highest potential for long-distance transport [ 38 ], The IKMHSS tailings remedial-strategy investigation of the subsurface fate and mobility of hazardous metals is a case study of the fate and mobility of metal contaminants in acidic sulfide mine waste in (semi-) arid regions worldwide. The tailings at IKMSS were sourced from regional (mostly Pb, Zn) mines (<100 km) with the mineralogy dominated by pyrite (FeS 2 ), ankerite (Ca[Fe, Mg](CO 3 ) 2 ), and quartz (SiO 2 ). Valuable ores in the area included chalcopyrite (CuFeS 2 ), galena (PbS), sphalerite (ZnS), and tennantite ([Cu, Fe] 12 As 4 S 13 ), as sources of zinc and lead, while gold and silver, occluded in pyrite and tennantite, respectively, were mined as secondary commodities [ 39 ]. The dominant non-sulfide “gangue” minerals (those having no commercial value) were ankerite, quartz, pyrite (up to 75% of the original vein), and arsenopyrite (FeAsS) [ 39 ]. The tailings have been weathering in place, unamended for half a century following the mine’s closure. A redox gradient with a boundary at ca. 30 cm depth with a transition zone from 25 to 54 cm depth was identified and associated with distinct depth-dependent geochemical and morphological changes. Surficial tailings were characterized by (oxyhydr)oxide and sulfate minerals ( ex . jarosite, ferrihydrite, schwertmannite, zincosite [ZnSO 4 ], and gypsum [CaSO 4 ·2H 2 O]), and metal(loid)s adsorbed to jarosite and ferrihydrite. Tailings deeper than 55 cm were characterized by carbonates and sulfides [ 17 , 34 ]. We hypothesized that the addition of labile carbon and a vegetative cap would help to buffer a more neutral pH, relative to unamended tailings, and decrease metal solubility. It was established that near-surface ferric (sulf)oxyhydroxide minerals with high adsorption capacity for metals sequestered arsenic and could also decrease metal mobility [ 4 ]. Therefore, it was expected that compost-assisted vegetation plots would show greater metal retention in the rhizosphere compared to unamended control plots due to plant-promoted pH buffering, sorption on existing ferric hydroxide surface sites, and organo-metallic complexation.",
"discussion": "4. Discussion This experiment investigated field-scale, direct-planting, compost-assisted phytostabilization in legacy acid sulfide tailings undergoing surficial oxidative weathering in a (semi-) arid environment. The temporal changes in metal lability and translocation to depth investigated at four time points over three years with four depth increments from 0 to 90 cm showed dynamic responses to remediation. Despite the variable solubilities and geochemical properties of the metals considered (V, Cr, Mn, Co, Ni, Cu, Zn, and Pb), it was hypothesized that secondary minerals formed during oxidative weathering (e.g., (oxyhydr)oxides and sulfates) would contribute to the overall attenuation of metals. It was expected that the abundance of secondary minerals in the surficial tailings above 20 cm and the near-neutral pH below 20 cm depth would lead to metal sequestration in the near surface with little translocation to the mostly unweathered 60–90 cm depth increment (Hayes et al., 2014; Root et al., 2015). The results show that 3 y phytostabilization treatments affect metal lability and mobility greatest at the surface 0–20 cm. Continued metal enrichment at intermediate depths (20–40 cm) after the establishment of vegetation indicates that evapotranspiration in the (semi-) arid climate does not prevent the downward flux of solution-phase-dissolved or colloidal metals, as has been shown to occur in hyper-arid climates [ 44 ], Although metals migrated from the top 20 cm to the horizon below, this extended field study demonstrated that metals are retained in the rhizosphere, as hypothesized, under OM and planting treatments. However, the final-year development of enrichment at 60–90 cm for Cr and Mn and depletion at shallow depths for V, Cu, Zn, and Pb indicate that further metal translocation with increased hydrologie throughflux may be a long-term concern. 4.1. Acidification The progressive acidification of amended tailings over the three-year study indicated that the initial pH buffering via OM (manure compost pH = 10.0 ± 0.1) was temporary, and tailings acidification progressed despite the establishment of vegetation. The temporary neutralization in the tailings via compost addition was not sufficient to prevent acid-mediated metal dissolution [ 45 , 46 ]. A Pearson’s correlation coefficient matrix of pH compared to the extractable metals showed the relationships between the highly labile water-soluble pools and acidity ( Table 2 ). Correlation constants at the 95% confidence interval from averaged quadruplicate samples of 0% compost control, 10% compost and seed, 20% compost, and 20% compost and seed for all depths and years were compared ( n = 64 averaged values) ( Table 2 ). The correlation coefficients revealed a strong (negative) relationship between tailings pH and lability. Water-soluble metals that form carbonates and sulfate salts (Mn, Co, Ni, Cu, and Zn) showed higher solubility at lower pH ( Table 2 ). It is well-documented that increasing pH plays an important role in decreasing the solubility of metal cations and adsorption to (oxyhydr)oxide and humic surfaces through surface functional group deprotonation [ 47 , 48 ]. Acidity showed very little correlation to the water solubility of V, Cr, or Pb, although the water-soluble pools of V and Cr were strongly correlated to each other ( Table 2 ). Metal lability in other systems of weathering mine tailings have shown relatively high solubility for Mn and Zn and low solubility of Pb under similar, but unamended, field conditions [ 9 , 45 , 49 ]. The stabilization of lead in acid mine tailings via incorporation into jarosite likely prevents lability, even under changing pH or organic matter additions [ 50 ]. 4.2. Metal Lability Enrichment was observed for Mn, Co, Ni, Cu, and Zn ( Figures S12 and S13 ) at >20 cm compared to the near surface. This is attributed to surficial mobilization (and depletion) with deposition (and enrichment) at depths due to downward translocation in response to irrigation. The release of these metals in the first steps of the sequential extraction showed higher surface lability that progressively decreased with depth. Temporal enrichment analysis (Δτ) confirmed that downward mobilization was taking place and was enhanced by phytostabilization ( Figure 3 ). Mass transfer from the surface to depths in the tailings was especially well-defined for Mn, Co, and Zn enrichment/depletion (τ) ( Figure S13 ) and dynamic mass transfer (Δτ) calculations ( Figure 3 ). Regression analysis was employed to investigate a statistically significant ( p < 0.05) linear relationship between Δτ and Δ lability at the 95% confidence interval for a two-tailed test. The relationship between lability and mobility over time was examined first by comparing the change in lability measured as the change in the labile pool concentration, versus the change in enrichment during the three-year study ( Figure 4 ). Second, the change in the recalcitrant but extractable pool of metals associated with poorly crystalline (oxyhydr)oxide secondary minerals versus the change in enrichment during the three-year study was analyzed ( Figure 5 ). The change over 3y of averaged field quadruplicates for all four depth increments of 0% control, 10% compost and seed, and 20% compost and seed ( n = 12) showed a significant relationship for the easily mobilized pools of Cr and Mn ( Figure 4 ). Among the metals mobilized in the labile fraction and the recalcitrant fraction, only Mn and Co exhibited a statistically significant linear relationship between temporal changes in lability and enrichment ( Figure 4 ). These results support the observation that Mn and Co exhibit the greatest mobilization compared to other metals and suggests that in the oxidative weathering environment, Mn and Co are the most susceptible to water solubility. Interestingly, there was no significant relationship between Δτ and Δ lability for Zn, Pb, Ni, Cu, Cr, or V. The relation between change in τ and lability over time was positive (slope) for Mn, Pb, Ni, and Co, but only significant at the 95% confidence interval for Mn and Co. A negative relationship was observed for Cu, Cr, and V. While Zn lability was strongly correlated with acidity, it showed no temporal relationship with enrichment or depletion. Evidence of a significant relationship between the change in enrichment and changes in the poorly crystalline (oxyhydr)oxide-associated Mn, Cu, Cr, and Co supports the hypothesis of secondary precipitates attenuating metals ( Figure 5 ). This observation is consistent with metal sequestration in other tailing systems [ 45 , 51 ], Comparing dynamic elemental enrichment/depletion and extractability over 3 y reveals that Mn and Co were most susceptible to mobilization from the tailings surface. Significant trends ( p < 0.05) in surficial depletion (and enrichment at depths) correlating to a change in extractability show that Mn and Co are leached from the near surface and deposited at greater depths. Copper in the near-surface oxic tailings showed increased lability and translocation to depths after phytostabilization, indicating that weathering and OM enhance copper mobility The zinc mobilized from the surface tailings was likely the divalent cation, released from sulfate salts and sorption sites on (oxyhydr)oxides at the start of this experiment. These sulfates and (oxyhydr)oxides are weathering products of the original zinc mineral sphalerite (ZnS) and secondary mineral smithsonite (ZnCO 3 ) [ 17 , 39 ]. Zinc lability correlated with decreasing pH. The greatest enrichment of zinc occurred at 40–60 cm in the control plots, indicating that phytostabilization decreased zinc mobility in the oxic zone of the tailings. Although Zn was observed to translocate from surface tailings to depths, a relationship with the operationally defined sequential extraction pools was not observed. 4.3. Recalcitrant Metals Lead, vanadium, and chromium were generally recalcitrant in the tailings and were only extracted at appreciable fractions from the pool associated with poorly crystalline (oxyhydr)oxide minerals. In a similar fashion, all three elements exhibited a lack of correlation between mobilization and pH ( Table 2 ). Vanadium and chromium likely originated as minor components in silicates, sulfides, and carbonates. Lead in deep tailings was predominantly in the form of the sulfide galena and carbonate cerussite (PbCO 3 ) [ 17 , 39 ]. Lead in the surface tailings mainly existed as plumbojarosite [ 17 ]. Although the lead mineralogy of the surface tailings differed significantly from the deep tailings, there was little difference in lead lability or mobility throughout the depths considered. This indicates that in the IKMHSS tailings, weathering and phytostabilization do not increase lead mobility. Similarly, vanadium exhibits both low lability and low mobilization with respect to treatment, time, and depth, generally being unaffected by oxidative weathering and phytostabilization. Although chromium exhibited high recalcitrance for all treatments and depths throughout the three-year study, translocation to depths was observed for composted treatments during phytostabilization and showed a statistically significant linear relationship with the pool associated with poorly crystalline (oxyhydr)oxides. Unlike in lead and vanadium, chromium release to porewaters and the promotion of translocation was enhanced during phytostabilization."
} | 5,513 |
36419772 | PMC9681136 | pmc | 9,107 | {
"abstract": "Understanding the temporal effects of organic matter input and water influx on metal lability and translocation is critical to evaluate the success of the phytostabilization of metalliferous mine tailings. Trends of metal lability, e.g., V, Cr, Mn, Co, Ni, Cu, Zn, and Pb, were investigated for three years following a direct-planting phytostabilization trial at a Superfund mine tailings site in semi-arid central Arizona, USA. Unamended tailings were characterized by high concentrations (mmol kg −1 ) of Fe (2100), S (3100), As (41), Zn (39), and Pb (11), where As and Pb greatly exceeded non-residential soil remediation levels established by Arizona. Phytostabilization treatments included a no-compost control, 100 g kg −1 compost with seed, and 200 g kg −1 compost with and without seed to the top 20 cm of the tailings profile. All plots received supplemental irrigation, effectively doubling the mean annual precipitation. Tailings cores up to 90 cm were Collected at the time of planting and every summer for 3 years. The cores were sub-sectioned at 20 cm increments and analyzed via total digestion and an operationally defined sequential extraction for elemental analysis and the calculation of a mass transfer coefficient normalized to Ti as an assigned immobile element. The results indicate that Pb was recalcitrant and relatively immobile in the tailings environment for both the uncomposted control and composted treatments with a maximum variation in the total concentration of 9–14 mmol kg −1 among all samples. Metal lability and translocation above the redox boundary ( ca . 30 cm depth) was governed by acid generation, where surficial pH was measured as low as 2.7 ± 0.1 in year three and strongly correlated with the increased lability of Mn, Co, Ni, Cu, and Zn. There was no significant pH effect on the lability of V, Cr, or Pb. Translocation to depths was greatest for Mn and Co; however, Zn, Ni, Cr, and Cu were also mobilized. The addition of organic matter enhanced the mobilization of Cr from the near surface to 40–60 cm depth (pH > 6) over the three-year phytostabilization study compared to the control. The increased enrichment of some metals at 60–90 cm indicates that the long-term monitoring of elemental translocation is necessary to assess the efficacy of phytostabilization to contain subsurface metal contaminants and thereby protect the surrounding community from exposure.",
"conclusion": "5. Conclusions Although compost-assisted phytostabilization initially raised the tailings’ pH, oxidative weathering and acid generation potential in surface tailings produced progressive acidification which was the primary mechanism for the increased lability and translocation of metals over time. The mobilization of some metals across short distances (<1 m) from the surface to depths occurred in all treatments, even in phytostabilized plots that had an initial pH neutralization and buffering provided by OM addition. The establishment of a vegetation cap or the addition of compost may have contributed to an enhanced rate of metal translocation, especially for Mn and Co, relative to the irrigation-only control. However, sequential extraction analysis did not indicate that organic matter or vegetation influenced a change in the specific sequential extraction-assigned speciation pools for any of the metals investigated. Further, the magnitude of 1D translocation to depth was low, generally less than 20 cm over 3 y. While Pb and V remained largely recalcitrant, Mn, Cu, Cr, and Cu fate and mobility were governed by the stability of poorly crystalline (oxyhydr)oxide secondary mineral precipitates forming under oxidative conditions. Although Zn and Ni were shown to translocate to depths over time, there was not a significant relationship between mobilization and the labile pools considered. These results highlight the importance of considering both depth and time in assessing of the utility and success of phytostabilization for sequestering metal contaminants in near-surface mine tailings.",
"introduction": "1. Introduction Abandoned, metalliferous, sulfide mine tailings pose a major human health and environmental hazard due to anthropogenic disturbance and the associated introduction of contaminant-bearing minerals into the environment [ 1 – 3 ]. Legacy tailings exhibit nutrient deficiency (C, N, and P), acidic pH, elevated salt levels, and high concentrations of hazardous metals and metalloids such as lead, zinc, copper, mercury, cadmium, and arsenic [ 4 ], Together, these characteristics make tailings an inhospitable environment for plant and heterotrophic microbial growth [ 5 – 8 ], which keeps them devoid of vegetation establishment for decades to centuries following mining cessation, particularly in (semi-) arid regions. As a result, tailings are highly susceptible to fugitive dust emissions and rainwater runoff erosion, leading to the exposure of wildlife and humans to toxic elements [ 5 , 7 , 9 – 11 ]. The ever-increasing demand for mining commodities requires corresponding research into improved long-term and affordable management strategies for mine waste containment to avoid the degradation of the proximal environment; the development of such techniques is an essential component of environmentally sustainable mining strategies. Phytostabilization is such a strategy, offering long-term, vegetation-driven containment to sequester and immobilize contaminants in the subsurface and rhizosphere under an established vegetation cap [ 3 , 12 – 14 ]. The growth of vegetation cover is key to providing the long-term physical stabilization of the tailings. The cover decreases wind shear at the surface and provides stabilization of the tailings, reducing fugitive emissions of particulate matter. Ferric (sulf)hydroxide crusts, which have high affinity for metal(loid) sorption, can form on root surfaces in mine tailings and sequester labile metal(loids) [ 12 ]. Additionally, roots provide a supportive scaffold that reduces wind and water erosion, and transpiration from vegetation diminishes contaminant leaching [ 4 ], Many areas experiencing heavy mining activity are located in (semi-) arid regions where water throughflux is limited, and therefore, so is the production of acid mine drainage (AMD). Dry conditions lead to the persistence of secondary weathering minerals, metal contaminants, and acidity in tailings media, preventing natural plant establishment [ 3 , 15 , 16 ]. Secondary minerals in mine tailings play important roles in contaminant retention and stabilization [ 17 ], For example, sulfates and (oxyhydr)oxides have been shown to attenuate solubilized metals in acid mine drainage [ 18 – 22 ]. Successful phytostabilization causes persistent organic root exudates and leaf litter to serve as a carbon source to the tailing ecosystem, with positive feedback on developing soil characteristics. The aim of phytostabilization is to achieve a self-sustaining ecosystem that eschews additional (and costly) inputs. The key to the success of phytostabilization is the negligible accumulation of toxic elements into the food-web in the above-ground biomass to reduce contaminant exposure to local wildlife [ 8 , 13 , 14 ], Given the low pH, low nutrient capacity, and poor soil structure of sulfide-ore-derived mine tailings, amendments such as irrigation, biosolids, fertilizers, lime, and chemical adsorbents are frequently employed to enhance plant establishment and growth [ 14 , 23 – 27 ], Despite an interest in phytostabilization as a remediation approach for mine tailings, the availability of published, long-term (≥2 years) field studies that assess the impacts of phytostabilization of metals contained in mine waste in (semi-) arid climates is lacking [ 28 – 32 ], Among long-term field-scale phytostabilization studies, this is the first to assess metal lability and translocation across both temporal and spatial variances. Elements of interest for this study include those in high concentrations and trace levels. The Iron King Mine Humboldt Smelter site (IKMHSS) tailings contain high zinc and manganese levels, which are both essential micronutrients but are toxic at high activities, and high lead levels, a toxic metal with no safe concentration level. Nickel, copper, chromium, vanadium, and cobalt, which are also micronutrients with toxic potential, are present in the tailings, but in much lower concentrations (<4 mmol kg −1 ). The comparative analysis of enriched and trace element lability and translocation provided a comprehensive dataset to assess geochemical processes taking place in the subsurface. The principal aims of this investigation were to (i) evaluate changes in transition metal (V-Zn) and Pb lability and translocation at the field-scale in response to the phytostabilization of sulfidic tailings from the surface to 90 cm deep and (ii) assess phytostabilization as an effective long-term stabilization strategy for these metals in tailings in a (semi-) arid climate over three years of tailings remediation. The IKMHSS acid sulfide tailings [ 33 ], listed due to high arsenic and lead concentrations, are dominated by iron and sulfur (13 wt% and 12 wt% in the top 0.5 m, respectively) [ 17 , 34 ]. The oxidative dissolution of initially deposited pyrite (FeS 2 ) releases iron, sulfate, acid, and toxic elements, which drives the precipitation of secondary ferric minerals such as jarosite (KFe 3 +3 (OH) 6 (SO 4 ) 2 ), ferrihydrite (Fe 2 3+ O 3 ·0.5(H 2 O), and schwertmannite (Fe 16 3+ O 16 (OH) 12 (SO 4 ) 2 ). The wind-driven erosion of the IKMHSS surface tailings is known to transport toxic metal(loid)-containing particulates to the surrounding community [ 5 , 8 , 10 , 35 , 36 ], Toxic metal(loid) lability and transport are of particular concern at this site due to arsenic concentrations being measured above the allowed state limit for residential soils (10 mg kg −1 ) [ 37 ], The field-scale (appx. 1 ha) phytostabilization of the IKMHSS tailings, initiated in 2010, established the long-term feasibility and efficacy of the compost-assisted, direct planting of a vegetation cap to reduce off-site dust emissions and sequester toxic metalloids in place [ 4 , 38 ], Previous studies at this site showed that phytoremediation was especially effective in reducing emissions of fine particulate matter (e.g., PM 2.5 ), which have the greatest health risks and the highest potential for long-distance transport [ 38 ], The IKMHSS tailings remedial-strategy investigation of the subsurface fate and mobility of hazardous metals is a case study of the fate and mobility of metal contaminants in acidic sulfide mine waste in (semi-) arid regions worldwide. The tailings at IKMSS were sourced from regional (mostly Pb, Zn) mines (<100 km) with the mineralogy dominated by pyrite (FeS 2 ), ankerite (Ca[Fe, Mg](CO 3 ) 2 ), and quartz (SiO 2 ). Valuable ores in the area included chalcopyrite (CuFeS 2 ), galena (PbS), sphalerite (ZnS), and tennantite ([Cu, Fe] 12 As 4 S 13 ), as sources of zinc and lead, while gold and silver, occluded in pyrite and tennantite, respectively, were mined as secondary commodities [ 39 ]. The dominant non-sulfide “gangue” minerals (those having no commercial value) were ankerite, quartz, pyrite (up to 75% of the original vein), and arsenopyrite (FeAsS) [ 39 ]. The tailings have been weathering in place, unamended for half a century following the mine’s closure. A redox gradient with a boundary at ca. 30 cm depth with a transition zone from 25 to 54 cm depth was identified and associated with distinct depth-dependent geochemical and morphological changes. Surficial tailings were characterized by (oxyhydr)oxide and sulfate minerals ( ex . jarosite, ferrihydrite, schwertmannite, zincosite [ZnSO 4 ], and gypsum [CaSO 4 ·2H 2 O]), and metal(loid)s adsorbed to jarosite and ferrihydrite. Tailings deeper than 55 cm were characterized by carbonates and sulfides [ 17 , 34 ]. We hypothesized that the addition of labile carbon and a vegetative cap would help to buffer a more neutral pH, relative to unamended tailings, and decrease metal solubility. It was established that near-surface ferric (sulf)oxyhydroxide minerals with high adsorption capacity for metals sequestered arsenic and could also decrease metal mobility [ 4 ]. Therefore, it was expected that compost-assisted vegetation plots would show greater metal retention in the rhizosphere compared to unamended control plots due to plant-promoted pH buffering, sorption on existing ferric hydroxide surface sites, and organo-metallic complexation.",
"discussion": "4. Discussion This experiment investigated field-scale, direct-planting, compost-assisted phytostabilization in legacy acid sulfide tailings undergoing surficial oxidative weathering in a (semi-) arid environment. The temporal changes in metal lability and translocation to depth investigated at four time points over three years with four depth increments from 0 to 90 cm showed dynamic responses to remediation. Despite the variable solubilities and geochemical properties of the metals considered (V, Cr, Mn, Co, Ni, Cu, Zn, and Pb), it was hypothesized that secondary minerals formed during oxidative weathering (e.g., (oxyhydr)oxides and sulfates) would contribute to the overall attenuation of metals. It was expected that the abundance of secondary minerals in the surficial tailings above 20 cm and the near-neutral pH below 20 cm depth would lead to metal sequestration in the near surface with little translocation to the mostly unweathered 60–90 cm depth increment (Hayes et al., 2014; Root et al., 2015). The results show that 3 y phytostabilization treatments affect metal lability and mobility greatest at the surface 0–20 cm. Continued metal enrichment at intermediate depths (20–40 cm) after the establishment of vegetation indicates that evapotranspiration in the (semi-) arid climate does not prevent the downward flux of solution-phase-dissolved or colloidal metals, as has been shown to occur in hyper-arid climates [ 44 ], Although metals migrated from the top 20 cm to the horizon below, this extended field study demonstrated that metals are retained in the rhizosphere, as hypothesized, under OM and planting treatments. However, the final-year development of enrichment at 60–90 cm for Cr and Mn and depletion at shallow depths for V, Cu, Zn, and Pb indicate that further metal translocation with increased hydrologie throughflux may be a long-term concern. 4.1. Acidification The progressive acidification of amended tailings over the three-year study indicated that the initial pH buffering via OM (manure compost pH = 10.0 ± 0.1) was temporary, and tailings acidification progressed despite the establishment of vegetation. The temporary neutralization in the tailings via compost addition was not sufficient to prevent acid-mediated metal dissolution [ 45 , 46 ]. A Pearson’s correlation coefficient matrix of pH compared to the extractable metals showed the relationships between the highly labile water-soluble pools and acidity ( Table 2 ). Correlation constants at the 95% confidence interval from averaged quadruplicate samples of 0% compost control, 10% compost and seed, 20% compost, and 20% compost and seed for all depths and years were compared ( n = 64 averaged values) ( Table 2 ). The correlation coefficients revealed a strong (negative) relationship between tailings pH and lability. Water-soluble metals that form carbonates and sulfate salts (Mn, Co, Ni, Cu, and Zn) showed higher solubility at lower pH ( Table 2 ). It is well-documented that increasing pH plays an important role in decreasing the solubility of metal cations and adsorption to (oxyhydr)oxide and humic surfaces through surface functional group deprotonation [ 47 , 48 ]. Acidity showed very little correlation to the water solubility of V, Cr, or Pb, although the water-soluble pools of V and Cr were strongly correlated to each other ( Table 2 ). Metal lability in other systems of weathering mine tailings have shown relatively high solubility for Mn and Zn and low solubility of Pb under similar, but unamended, field conditions [ 9 , 45 , 49 ]. The stabilization of lead in acid mine tailings via incorporation into jarosite likely prevents lability, even under changing pH or organic matter additions [ 50 ]. 4.2. Metal Lability Enrichment was observed for Mn, Co, Ni, Cu, and Zn ( Figures S12 and S13 ) at >20 cm compared to the near surface. This is attributed to surficial mobilization (and depletion) with deposition (and enrichment) at depths due to downward translocation in response to irrigation. The release of these metals in the first steps of the sequential extraction showed higher surface lability that progressively decreased with depth. Temporal enrichment analysis (Δτ) confirmed that downward mobilization was taking place and was enhanced by phytostabilization ( Figure 3 ). Mass transfer from the surface to depths in the tailings was especially well-defined for Mn, Co, and Zn enrichment/depletion (τ) ( Figure S13 ) and dynamic mass transfer (Δτ) calculations ( Figure 3 ). Regression analysis was employed to investigate a statistically significant ( p < 0.05) linear relationship between Δτ and Δ lability at the 95% confidence interval for a two-tailed test. The relationship between lability and mobility over time was examined first by comparing the change in lability measured as the change in the labile pool concentration, versus the change in enrichment during the three-year study ( Figure 4 ). Second, the change in the recalcitrant but extractable pool of metals associated with poorly crystalline (oxyhydr)oxide secondary minerals versus the change in enrichment during the three-year study was analyzed ( Figure 5 ). The change over 3y of averaged field quadruplicates for all four depth increments of 0% control, 10% compost and seed, and 20% compost and seed ( n = 12) showed a significant relationship for the easily mobilized pools of Cr and Mn ( Figure 4 ). Among the metals mobilized in the labile fraction and the recalcitrant fraction, only Mn and Co exhibited a statistically significant linear relationship between temporal changes in lability and enrichment ( Figure 4 ). These results support the observation that Mn and Co exhibit the greatest mobilization compared to other metals and suggests that in the oxidative weathering environment, Mn and Co are the most susceptible to water solubility. Interestingly, there was no significant relationship between Δτ and Δ lability for Zn, Pb, Ni, Cu, Cr, or V. The relation between change in τ and lability over time was positive (slope) for Mn, Pb, Ni, and Co, but only significant at the 95% confidence interval for Mn and Co. A negative relationship was observed for Cu, Cr, and V. While Zn lability was strongly correlated with acidity, it showed no temporal relationship with enrichment or depletion. Evidence of a significant relationship between the change in enrichment and changes in the poorly crystalline (oxyhydr)oxide-associated Mn, Cu, Cr, and Co supports the hypothesis of secondary precipitates attenuating metals ( Figure 5 ). This observation is consistent with metal sequestration in other tailing systems [ 45 , 51 ], Comparing dynamic elemental enrichment/depletion and extractability over 3 y reveals that Mn and Co were most susceptible to mobilization from the tailings surface. Significant trends ( p < 0.05) in surficial depletion (and enrichment at depths) correlating to a change in extractability show that Mn and Co are leached from the near surface and deposited at greater depths. Copper in the near-surface oxic tailings showed increased lability and translocation to depths after phytostabilization, indicating that weathering and OM enhance copper mobility The zinc mobilized from the surface tailings was likely the divalent cation, released from sulfate salts and sorption sites on (oxyhydr)oxides at the start of this experiment. These sulfates and (oxyhydr)oxides are weathering products of the original zinc mineral sphalerite (ZnS) and secondary mineral smithsonite (ZnCO 3 ) [ 17 , 39 ]. Zinc lability correlated with decreasing pH. The greatest enrichment of zinc occurred at 40–60 cm in the control plots, indicating that phytostabilization decreased zinc mobility in the oxic zone of the tailings. Although Zn was observed to translocate from surface tailings to depths, a relationship with the operationally defined sequential extraction pools was not observed. 4.3. Recalcitrant Metals Lead, vanadium, and chromium were generally recalcitrant in the tailings and were only extracted at appreciable fractions from the pool associated with poorly crystalline (oxyhydr)oxide minerals. In a similar fashion, all three elements exhibited a lack of correlation between mobilization and pH ( Table 2 ). Vanadium and chromium likely originated as minor components in silicates, sulfides, and carbonates. Lead in deep tailings was predominantly in the form of the sulfide galena and carbonate cerussite (PbCO 3 ) [ 17 , 39 ]. Lead in the surface tailings mainly existed as plumbojarosite [ 17 ]. Although the lead mineralogy of the surface tailings differed significantly from the deep tailings, there was little difference in lead lability or mobility throughout the depths considered. This indicates that in the IKMHSS tailings, weathering and phytostabilization do not increase lead mobility. Similarly, vanadium exhibits both low lability and low mobilization with respect to treatment, time, and depth, generally being unaffected by oxidative weathering and phytostabilization. Although chromium exhibited high recalcitrance for all treatments and depths throughout the three-year study, translocation to depths was observed for composted treatments during phytostabilization and showed a statistically significant linear relationship with the pool associated with poorly crystalline (oxyhydr)oxides. Unlike in lead and vanadium, chromium release to porewaters and the promotion of translocation was enhanced during phytostabilization."
} | 5,513 |
35479885 | PMC9033653 | pmc | 9,109 | {
"abstract": "Calculations of chemical oxygen demand (COD) degradation in sewage by a microbial fuel cell (MFC) were used to estimate the total energy required for treatment of the sewage. Mono-exponential regression (MER) and the Michaelis–Menten equation (MME) were used to describe the MFC's COD removal rate (CRR). The tubular MFC used in this study ( ϕ 5.0 × 100 cm) consisted of an air core surrounding a carbon-based cathode, an anion exchange membrane, and graphite non-woven fabric immersed in sewage. The MFC generated 0.26 A m −2 of the electrode area and 0.32 W m −3 of the sewage water, and 3.9 W h m −3 in a chemostat reactor supplemented continuously with sewage containing 180 mg L −1 of COD with a hydraulic retention time (HRT) of 12 h. The COD removal and coulombic efficiency (CE) were 46% and 19%, respectively, and the energy generation efficiency (EGE) was 0.054 kW h kg −1 -COD. The CRR and current in the MFC were strongly dependent on the COD, which could be controlled by varying the HRT. The MER model predicted first-order rate constants of 0.054 and 0.034 for reactors with and without MFC, respectively. The difference in these values indicated that using MFC significantly increased the COD removal. The results of fitting the experimental data to the MME suggested that the total COD can be separated into nondegradable CODs ( C n ) and degradable CODs ( C d ) via MFC. The values of CRR for C d and CE suggest that MFC pretreatment for 12 hours prior to aeration results in a 75% decrease in net energy consumption while reducing sewage COD from 180 to 20 mg L −1 .",
"conclusion": "5. Conclusion Both the mono-exponential regression and the Michaelis–Menten calculations of COD removal indicated a strong dependence on COD in an MFC fed with sewage wastewater. The data suggest that increasing the anode-specific surface area will yield improvements in MFC performance. The experimental data and calculations both demonstrate that coulombic efficiency is almost constant. When this MFC was combined with post-aeration for COD reduction to meet discharge standards, a 75% reduction of total energy could be anticipated by feeding the MFC at an HRT of 12 h.",
"introduction": "1. Introduction The recovery of current by electrodes from wastewater has received great attention as a novel treatment used in microbial fuel cells (MFC) 1 and microbial electrochemical systems (MES). 2 Extensive studies on MFCs have demonstrated the considerable potential of these devices, as well as the difficulties involved in upscaling them for practical application. 3–5 Studies on the direct recovery of electricity from sanitary wastewater 6,7 and urine 8 have also shown the efficient removal of organic matter, and have proven that the recovered electricity can be used to power lighting devices. 9 Onsite installation of floating MFC vessels in sewage wastewater treatment plants can enable current recovery at relatively low concentrations of chemical oxygen demand (COD), i.e. , 400–20 mg L −1 . 10–12 These results suggest that MFCs can be employed in municipal wastewater treatment for a wide range of COD values. Large-scale MFCs are effective in removing organic compounds from real wastewater with relatively low energy consumption, 13–15 or even net production of energy in some cases. 16,17 Studies on sewage treatment using MFCs with capacities greater than five liters and hydraulic retention times (HRTs) between 12 hours and six days have observed 54–79% COD removal efficiency (COD-RE) with 70–210 mg L −1 of effluent COD (COD EF ). 12,15,17,18 In Japan, the standard for sewage treatment plant effluent is less than 15 mg L −1 of BOD, which corresponds to a COD range of 20–35 mg L −1 . 19 The difference between the COD EF and the discharge standards demonstrates the need for improvements in MFC design or the integration of MFCs with post-treatment processes such as aeration, 15 anaerobic membrane filtration, 20 and flocculation. 21 Conventionally, the performance of an MFC is characterized in terms of output variable values such as COD-RE, electric power density, and coulombic efficiency (CE). 1 Recently, parameters such as current density, 16 electrical energy, 22 normalized energy recovery, 23 and energy generation efficiency (EGE) 12,15 have been used to shift the focus towards practical application. In most cases, these values are independently determined by averaging measurements under different operational conditions despite reports that the operational parameters themselves affect the performance. Specifically, influents with higher COD resulted in higher COD-RE 24,25 and power density 25,26 but lower CE. 15,27 Increasing the HRT caused further reductions in the COD and current and power densities. 15,17 The mutual interactions between operational and output parameters make the evaluation of MFC performance difficult. A judicious combination of these parameters based on analysis of the governing equations and experimental data may provide a more concise and comprehensive basis for evaluating MFC performance. To determine the optimal operation, it is crucial to consider the balance between energy production and consumption for the wastewater treatment process as a whole. 15–17 In some cases, a trade-off is required between power production and COD-RE. Power production can be increased at higher cell voltages by lowering the anode potential and oppositely recovering current from anode to cathode. Reducing the anode potential theoretically decreases the generation of Gibbs free energy, which decreases COD-RE. 28 This increases the energy requirement for the post-treatment of the remaining COD. The total energy (TE), rather than its inversely related individual components of COD-RE and electricity production, should therefore be optimized. During the sewage treatment process, the COD removed by the MFC is converted to electricity while that remaining in the effluent is assumed to be removed to meet a given discharge standard by a standard activated sludge process at an energy cost of 0.6 kW h kg −1 -COD. 29 In this study, the rate of COD degradation in an MFC was calculated using two equations: the mono-exponential regression equation (MER) and the Michaelis–Menten equation (MME). The Michaelis–Menten equation is originally a concept whereby the number of complexes of a single substrate and single catalyst determine the reaction rate. Consequently, the use of this model in wastewater treatment is logically incorrect because wastewater includes a variety of redox reactions, and cannot be modeled by a defined calculation method. However, only MME can integrate the MFC performance for different influents and HRTs in a single line, which has been applied to the COD as a bundle of organic matter and complex microbial communities. 30–32 The CE was used as a constant and was calculated by fitting the experimentally observed COD degradation and current. This result was then used to calculate the TE for a combined MFC and post-aeration process that met the discharge standard. A tubular 33 and air-core MFC unit 34 with a depth of 1.0 m was evaluated for sewage wastewater treatment. 15 To the best of our knowledge, this unit is comparable to the previously studied deepest air-cathode MFCs. 35",
"discussion": "4. Discussion We estimated COD consumption using the MER and MME equations. Both calculations reproduced COD-RE and current effectively for the air-cathode-AEM-MFC treating sewage wastewater. The calculations include uncertainty owing to the lack of replicates running for multiple MFCs, although the equivalent current and COD reductions were confirmed by using MFCs with the same configurations but with a length of 30 cm. 15 The consistency between the results of these independently performed evaluations involving similar MFCs indicates the reproducibility of the performance of the proposed type of MFC. The goodness-of-fit suggests a strong dependence of CRR and current on COD in a steady-state MFC reactor, with HRT as the only variable parameter and a relatively low COD, and was polarized using a low external resistance. The operating conditions were such that the MFC was less affected by external factors such as cathode potential, inhibition of substrates in the influent, 42 and electron mediators. 43 Comparing the MER and MME, the former provides a convenient and straightforward rough estimate of the MFC performance. The latter requires improvement, but is advantageous for integrating all data of COD degradation and electricity with influent at different CODs as a single line, and to determine two critical factors, V max and K s , that are important for further improvement of MFC performance. In the MME calculation, two different COD parameters were used: C and C d . These resulted in different values for V max and K s . The C-MME overestimated COD consumption in the MFC, especially at lower COD. This difference is likely caused by the heterogeneity of organic matter in the MFC 44 and the overestimation of V max as a result of using the data at COD including abundant readily biodegradable matter. This value could not be used in calculations for the low COD dominated by unfavorable residues of organic matter. 45 The C d -MME calculation yielded more realistic performance predictions for our MFC. Evaluations with longer HRT are required to validate the underlying assumption, but tube clogging prevented these experiments. The C d -MME calculation is contradictory; the current production at time C d equals zero. We believe that the current was produced from endogenous substrates in the anode biofilm 46 and estimated current as the sum of the current from C d ( I C d ) and a constant current from the endogenous substrate ( I E ) for HRTs until 18 h. Strictly, the value of I E can vary depending on the balance of CODs inside and outside the biofilm. The HRT in this case determines these CODs in the MFC. Treating a mixture of organic matter as a unit is a limitation of the approximation that also occurs in other calculations of sewage treatment by MFCs, for example, the Monod equation 47 which is an empirical equation of biomass growth rate. 48 Notwithstanding the limitation, approximations of COD-RE and current still provide useful insights into the applicability and limitations of MFCs, and these insights can be used to improve MFC performance. The C d -MME calculation predicted 13 mg L −1 h −1 of V max and 14 mg L −1 of K s for C d , corresponding to 73 mg L −1 as C . According to the predicted values, our MFC maintains CRR at more than 80% V max until HRT = 6 h, but the CRR decreases to half at HRT = 11 h, and approximates zero at HRT = 18 h. The CRR is the overall consumption, including fermentation in the liquid phase and current recovery at the anode. Minor contribution of current recovery in COD-RE and low CE in MFC was noticeably observed in the MFC when treating municipal wastewater, whereas the MFC with the same configuration showed high COD-RE and CE for synthetic wastewater. 49 The other electron scavengers in the MFC treating municipal wastewater have been reported as sulfate in wastewater and oxygen derived from IEM, 40 although the sulfate-reduction products, H 2 S and HS − , can also act as electron donors in MFCs. 50 As the COD-RE in the liquid are virtually uncontrollable, current recovery must be increased for further COD-RE. The observed COD-RE was lower than the values reported in studies on other MFCs (ESI Table 1 † ). The comparison with other MFCs suggests two possible methods to improve the performance—increasing the specific surface area of the separator (SSSA) and anode (ASSA). The former is ideal but not practical when scaling up because the percentage of IEM in the initial cost of an air-IEM-MFC is the highest among all parts of the MFC. 16 The use of a cost-effective separator such as a ceramic separator 6–8 is a promising approach to increase SSSA; however, wastewater treatment performance at a range similar to COD was not reported in previous studies. The latter is more effective for the proposed MFC by eliminating the rate-determination on anodes due to the lowest ASSA, 16 m 2 m − 3, compared to the other large MFCs. To employ an anode with a high ASSA, carbon brush 51 or other three-dimensional anodes 52,53 would be effective. In addition to increasing ASSA, designing fluid to enhance the COD supplement to anodes is also effective to eliminate the rate-determination on anodes at low COD. 54 In this present study, the COD-RE and current parameters were combined as the energy requirement per volume of wastewater to achieve a specified effluent COD. For the calculation, a value of 0.6 W h g −1 -COD was taken as the energy cost to remove COD by aeration. 29 In this study, the COD-RE and current parameters were combined as the energy requirement per volume of wastewater to achieve a specified effluent COD. For the calculation, a value of 0.6 W h g −1 -COD was taken as the energy cost to remove the COD via aeration. 29 The value is an indicator to determine whether the MFC's operational mode should be focused on power generation or COD removal, but it varies depending on the installation process and post-COD-removal treatments such as intermittent aeration 55 and usage of anaerobic membrane bioreactors. 20 Specifically, if the energy requirement is less than the EGE of the MFC, the MFC can be optimized for power generation. Although the value is greater than the EGE of the MFC, the operation focusing on COD removal is preferable considering overall energy requirement. A voltage booster has been used to boost the MFC power, although the EGE value has rarely been calculated and an MFC treating sewage wastewater was calculated to have 0.146 of EGE 56 in an MFC. Surpassing the value by system control is extremely challenging, the organic loading rate and a long HRT would likely become key parameters, as well as biogas production. 57 The use of MFC as power for some electric devices has been demonstrated. MFCs have been fed with human urine to power a mobile phone 58 and a microcomputer. 59 MFCs treating municipal wastewater have also been used to power an intermittent pump 16 and an aerator. 60 For the practical application of MFC in treating wastewater, thorough feasibility studies are still required to match the advantage of MFC and the social demand. However, in all matters, further improvement of the MFC unit is most important. Carbon brush is an excellent anode 51 but is expensive to process from fiber to anode; carbonization, forming brush, and surface oxidization for the hydrophilizing. 61 Reducing the steps is required for mass production to decrease the processing cost. The AEM rather than CEM has become a good ion-selective separator for MFC. 12,62–64 However, all membrane needing to separate liquid and gas have the problem to be scaled-up to several meters in depth and the ceramic separator is promising in terms of strength. 6–8 The air-cathode also needs improvement to maintain a high potential for the removal of nitrogenous compounds from wastewater. 65 We believe that solving these individual problems will enable the realization of practical use of MFC in wastewater treatment."
} | 3,825 |
25810302 | null | s2 | 9,110 | {
"abstract": "DNA nanostructured tiles play an active role in their own self-assembly in the system described herein whereby they initiate a binding event that produces a cascading assembly process. We present DNA tiles that have a simple but powerful property: they respond to a binding event at one end of the tile by passing a signal across the tile to activate a binding site at the other end. This action allows sequential, virtually irreversible self-assembly of tiles and enables local communication during the self-assembly process. This localized signal-passing mechanism provides a new element of control for autonomous self-assembly of DNA nanostructures."
} | 163 |
36630512 | PMC9833652 | pmc | 9,111 | {
"abstract": "Tough natural materials such as nacre, bone, and silk exhibit multiscale hierarchical structures where distinct toughening mechanisms occur at each level of the hierarchy, ranging from molecular uncoiling to microscale fibrillar sliding to macroscale crack deflection. An open question is whether and how the multiscale design motifs of natural materials can be translated to the development of next-generation biomimetic hydrogels. To address this challenge, we fabricate strong and tough hydrogel with architected multiscale hierarchical structures using a freeze-casting–assisted solution substitution strategy. The underlying multiscale multimechanisms are attributed to the gel’s hierarchical structures, including microscale anisotropic honeycomb–structured fiber walls and matrix, with a modulus of 8.96 and 0.73 MPa, respectively; hydrogen bond–enhanced fibers with nanocrystalline domains; and cross-linked strong polyvinyl alcohol chains with chain-connecting ionic bonds. This study establishes a blueprint of structure-performance mechanisms in tough hierarchically structured hydrogels and can inspire advanced design strategies for other promising hierarchical materials.",
"introduction": "INTRODUCTION Hydrogels have excellent potential as advanced engineering materials for wearable electronics ( 1 , 2 ), tissue engineering ( 3 ), soft robotics ( 4 , 5 ), and biomedical engineering ( 6 ). However, conventional hydrogels are generally weak and fragile ( 7 ), which substantially limits their applications. In the last decade, many efforts have been devoted to developing enhanced hydrogels with excellent mechanical properties, such as topological hydrogels ( 8 ), nanocomposite hydrogels ( 9 ), double-network (DN) hydrogels ( 10 ), dual cross-linked hydrogels ( 11 ), and nanocrystalline hydrogels ( 12 ). However, these studies mainly focus on molecular engineering and composition, and the involved structural changes are limited to molecular scale or nanoscale. For example, the rupture of the weak network and associated cross-links in a DN hydrogel can dissipate mechanical energy at the molecular level, which can effectively enhance toughness ( 13 ). At present, the toughness of even well-designed DN hydrogels hardly exceeds 10 MJ/m 3 . To further enhance the strength and toughness of hydrogels, it will be meaningful to construct more advanced designs by exploring toughening mechanisms over a wider range of length scales. Natural hydrogels, which typically exhibit superior strength and toughness, are abundant in various plant and animal tissues, including xylems, phloems, muscles, and cartilages ( 14 ). This is attributed to their unique hierarchical structures, which range from microscopic anisotropic alignments to distinctive crystalline units at the molecular scale, resulting in synergistic strengthening and toughening of the overall materials. Inspired by this, biomimetic hydrogels with anisotropic multiscale hierarchical structures have been proven to have excellent mechanical performances ( 15 ). Moreover, several kinds of anisotropic hydrogels with improved mechanical properties have been developed by mechanical training ( 16 , 17 ), additional fillers ( 18 , 19 ), and freeze-casting ( 20 , 21 ). For example, muscle-like fatigue-resistant hydrogels with directionally aligned nanofibrillar structures were obtained via mechanical training ( 17 ). Although the mechanically trained hydrogels showed enhanced strength and fracture toughness compared to untrained hydrogels, their water content and stretchability were substantially reduced. As for freeze-casting, it is a general method for fabricating anisotropic hydrogels. However, once freeze-thawed hydrogels with initial microscopic alignment exhibit limited mechanical properties and generally need to be further enhanced by postprocessing ( 22 ). These hydrogels exhibited somewhat improved mechanical properties with anisotropically aligned structures at a single scale but showed a conflict between strength and stretchability due to the lack of multiscale hierarchical structures. At this point, the fabrication of stretchable and tough hydrogels with hierarchical structures across multiple length scales from micro-, nano-, and molecular levels remains challenging. Recently, a design principle of simultaneously enhancing stretch, strength, and toughness has been proposed and demonstrated for hierarchical fiber-reinforced hydrogels, including microlevel fiber anisotropic alignment, nanoscale fiber aggregation, and molecular-scale fiber reinforcement ( 23 ). For example, combining unidirectional freeze-casting and salting-out ( 24 ), a single-composition fibrous hydrogel exhibited unique stepwise fracture behavior and crack propagation blocking ability, resulting in superior mechanical properties coupled with high strength and toughness. Furthermore, a number of tough and functional fibrous hydrogels have been fabricated through a combination of freeze-casting with other procedures, such as ion enhancement ( 25 ), solution replacement ( 26 ), and annealing ( 15 , 27 ). A universal design strategy of ice-templating and subsequent thermal annealing has been proposed with impressive enhancement in fiber crystallinity, fracture energy, and fatigue threshold ( 27 ). These fibrous hydrogels combined the advantages of hierarchical structural engineering and molecular engineering, achieving simultaneously improved stretchability, strength, and toughness, as well as functionalities such as conductivity and freezing tolerance. While these successes of introducing hierarchical structures in fibrous hydrogels have been characterized at different length scales, the current development of fibrous hydrogels has mainly depended on trial-and-error and empirical methods. The intrinsic mechanisms at micro-, nano-, and molecular levels for improving modulus, stretchability, and toughness in fibrous hydrogels have not been systematically explored and discussed. This greatly restricts the advanced design strategies and engineering application of biomimetic hydrogels. Here, we propose a freeze-casting–assisted solution substitution strategy to fabricate strong and tough fibrous hydrogels with hierarchical structures from molecular to micrometer scales. This is achieved by immersing the directional frozen polyvinyl alcohol (PVA) ice blocks into an ethanol solution with the addition of ferric chloride. Tensile and pure shear tests are performed to experimentally investigate their mechanical and fracture performance. We further combine experimental characterization with theoretical simulations to understand the underlying strengthening and toughening mechanisms at each length scale. At the microscopic level, representative volume element (RVE) analysis coupled with periodic boundary conditions (PBCs) ( 28 , 29 ) is carried out to characterize the mechanical behavior and elastic properties of the anisotropic fibrous hydrogel. Atomistic-level simulations are conducted using Materials Studio ( 30 ) to explore the molecular mechanical behavior of PVA hydrogels and the effect of ethanol substitution and Fe 3+ on intramolecular interactions between the PVA chains and water molecules. This approach allows the multiscale multimechanisms of fibrous hydrogels to be systematically characterized through an integrated experiment-simulation approach. Our study demonstrates an effective design and analysis strategy for strong and tough fibrous hydrogels with hierarchical structures and multiple enhancement mechanisms. In addition, the developed theoretical and simulation modeling framework can be applied to other artificial hydrogel systems and even natural materials.",
"discussion": "DISCUSSION This study has demonstrated a versatile strategy to introduce multiple strengthening and toughening mechanisms across multiple length scales into micro-, nano-, molecular-level structures of hydrogels. First, PVA chains were concentrated during freeze-casting to form initial hydrogen bonds and anisotropically aligned micro-honeycomb structures with relatively low crystallinity (fig. S3 and Fig. 2, E to G ). Next, ethanol substitution induced the PVA polymers to aggregate and form more hydrogen bonds, stabilizing and strengthening their micromorphology of hard honeycomb walls and soft matrix. Last, Fe 3+ ion enhancement provides chain-connecting bonds and further promotes the formation of additional hydrogen bonds between PVA chains. Moreover, more crystalline domains appear as high-functionality cross-linkers that improved the elasticity, energy dissipation, and strength of hydrogels at the nanoscale. As a result, the multimechanisms across multiple length scales of fibrous hydrogels result in tremendous enhancement of the material properties including flaw tolerance, strength, stretchability, and toughness. In summary, we have proposed a nature-inspired synergistic strategy for the fabrication of strong and tough fibrous hydrogels with hierarchical structures across multiple length scales at micro-, nano-, and molecular levels. Mechanical tests demonstrated that the fibrous hydrogels exhibited exceptional mechanical properties including strength, stretchability, toughness, and flaw tolerance. The multimechanisms of strengthening and energy dissipation were validated by various characterization and simulations. The anisotropically hierarchical structures of fibrous hydrogels were observed by scanning electron microscopy (SEM), based on which FEM analysis characterized the strengthening effects and elastic properties of the reinforced honeycomb structural units. The interaction effects between microfiber walls and matrix led to high strength and high resistance to deformation, which becomes more evident with the deepening of modulus-contrasting and Poisson’s ratio contrasting values. The high crystallinity and bonding of the fibrous hydrogels were characterized by material component measurements, and molecular dynamics simulations showed that the solvent enhancement and ion enhancement on the hydrogels were dominated by the formation of chain-connecting Fe bonds and multiply promoted hydrogen bonds. Therefore, the involved strengthening and toughening mechanisms were demonstrated to be attributed to the hierarchical structures, including microscale anisotropic honeycomb–structured three-dimensional fiber walls with matrix-, hydrogen bond–, and coordination bond–enhanced fibers with nanocrystalline domains, as well as entangled and cross-linked strong PVA chains at the molecular level. This study characterizes the multiscale multimechanisms of fibrous hydrogels in detail and establish the relationship between structure-performance mechanisms, which provide insight into the designing of hierarchical tough hydrogels. The systematic mechanistic analysis model can also be extended to other hierarchical material systems as a novel yet generic approach to the analysis of natural materials."
} | 2,737 |
34759436 | PMC8570341 | pmc | 9,112 | {
"abstract": "Soil organic nitrogen (N) is a critical resource for plants and microbes, but the processes that govern its cycle are not well-described. To promote a holistic understanding of soil N dynamics, we need an integrated model that links soil organic matter (SOM) cycling to bioavailable N in both unmanaged and managed landscapes, including agroecosystems. We present a framework that unifies recent conceptual advances in our understanding of three critical steps in bioavailable N cycling: organic N (ON) depolymerization and solubilization; bioavailable N sorption and desorption on mineral surfaces; and microbial ON turnover including assimilation, mineralization, and the recycling of microbial products. Consideration of the balance between these processes provides insight into the sources, sinks, and flux rates of bioavailable N. By accounting for interactions among the biological, physical, and chemical controls over ON and its availability to plants and microbes, our conceptual model unifies complex mechanisms of ON transformation in a concrete conceptual framework that is amenable to experimental testing and translates into ideas for new management practices. This framework will allow researchers and practitioners to use common measurements of particulate organic matter (POM) and mineral-associated organic matter (MAOM) to design strategic organic N-cycle interventions that optimize ecosystem productivity and minimize environmental N loss. Supplementary Information The online version contains supplementary material available at 10.1007/s10533-021-00793-9.",
"conclusion": "Conclusion We present a new framework of bioavailable N cycling based on the interactions between organic N depolymerization, mineral sorption-desorption dynamics, and the actions of plants and microbes. New research, enabled by methodological advances of the last decade, has revealed depolymerization to be a dynamic process that drives substantial fluxes of bioavailable N from POM; this organic N subsequently associates with soil minerals to form MAOM, a large and heterogeneous pool of SOM enriched in nutrients that roots and microbes can actively mine. Our framework suggests that the flow of bioavailable N from MAOM is based on the relative balance between POM-N inputs and the soil’s mineral sorption potential, further shaped by plant-microbe interactions and environmental conditions. Microbial physiological traits substantially impact the entire bioavailable N cycle. By accounting for MAOM-N dynamics, we can develop agricultural management strategies that better minimize N pollution while reaching crop yield goals. As the SON paradigm is reshaped—the way SOC paradigm has been reshaped over the last two decades—new avenues will open to understanding the cycling of bioavailable N.",
"introduction": "Introduction Nitrogen (N) is essential for life as a key constituent of biomolecules including DNA, RNA, chlorophyll, and enzymes. In soil, bioavailable N is comprised of dissolved inorganic and organic N—including small polymers and monomers—that can be assimilated by plants and/or microbes. Supplies of bioavailable soil N sometimes exceed plant requirements, but often fail to meet them, resulting in N asynchrony that constrains ecosystem productivity and exacerbates environmental nutrient losses, which are expected to intensify under climate change (Sinha et al. 2017 ; Bowles et al. 2018 ; Houlton et al. 2019 ; Dai et al. 2020 ). This “N problem” arises in part because of nitrogen's changeable nature: as a reactive element found in multiple forms and seven oxidation states, N is difficult to track and manage. Unresolved issues in intensively managed agroecosystems epitomize our incomplete understanding of bioavailable N. In these systems, the persistent challenge of minimizing N losses and improving the spatial and temporal match between N availability and plant N demand (i.e. N synchrony) derives in part from a historical focus on the inorganic N pool. Even with high synthetic N inputs, however, a substantial fraction of inorganic N is derived from the soil organic matter pool (Yan et al. 2020 ). Yet, we remain without a universal and accurate assay or model that can predict organic N (ON) conversion to plant-available inorganic N, despite the long-acknowledged need for one (e.g. Vitousek 1982 ; Schimel and Bennett 2004 ) and continuing efforts to develop a suitable N availability index (Ros 2012 ; Curtin et al. 2017 ; Clivot et al. 2017 ; McDaniel et al. 2020 ). A focus on inorganic N pools overlooks the important mechanisms occurring in soil that determine how much ON feeds into and supplies the inorganic N pool. Moreover, the ON component of the bioavailable N pool is itself a critical N source to plants and microbes. Estimates of bioavailable N that do include ON usually represent it as the short-term potentially mineralizable N pool. However, this pool is operationally defined; in measuring net changes in inorganic N under optimized conditions and in the absence of live plant roots, potentially mineralizable N often poorly explains the variability in outcomes such as crop yields, estimated or actual crop N availability, and fertilizer needs (Fox and Piekielek 1984 ; Thicke et al. 1993 ; Curtin and McCallum 2004 ; Dessureault-Rompré et al. 2014 ; McDaniel et al. 2020 ). Agricultural practitioners currently rely on N-credit calculators that do not explicitly consider soil processes and interactions (Lory et al. 1995 ) and are prone to uncertainty, bias, and error (Sharma and Bali 2018 ). The struggle to quantify the pool of plant- and microbe-accessible N arises from conceptual gaps in current explanations about the fundamental mechanisms that drive N bioavailability; these stem in large part from failing to accurately account for the organic component of the soil N cycle and its biogeochemical drivers. The need to emphasize organic N is reminiscent of the impetus that led to developments in how the soil organic carbon (SOC) cycle is conceptualized. In the twentieth century, researchers theorized that the inherent chemical recalcitrance of carbon (C) to decomposition controlled SOC turnover, but evidence from the last two or more decades reveals that microbes can degrade even the most complex molecules (Gleixner et al. 2001 , 2002 ; Rasse et al. 2006 ) and that, in the context of overall soil organic matter (SOM) dynamics, recalcitrance only temporarily controls microbial SOC processing rates. Instead, SOC persistence largely emerges from constraints that the soil mineral matrix imposes on microbial access to substrates (Kleber et al. 2011 ; Schimel and Schaeffer 2012 ) and SOC dynamics are better predicted by biological and physical controls on C transfer between different SOC pools (Six et al. 2006 ; Grandy and Neff 2008 ), motivating several recent soil C cycling models to explicitly incorporate soil physical fractions (Sulman et al. 2014 ; Wieder et al. 2015 ; Abramoff et al. 2018 ; Kyker-Snowman et al. 2019 ). The fate of ON similarly relies on how associations with minerals regulate access to N-containing molecules (Lavallee et al. 2020 ) which are in turn regulated by biologically mediated chemical and physical processes that have yet to be integrated into the soil N paradigm (Darrouzet-Nardi and Weintraub 2014 ). Here, we aim to unify advances in the understanding of N transformations by developing a new, testable conceptual model of organic bioavailable N in soil. We ground our model in two commonly measured SOM pools: particulate organic matter (POM) and mineral-associated organic matter (MAOM), capturing the importance of both the depolymerization of N-containing molecules (Schimel and Bennett 2004 ) and mineral sorption-desorption (Sollins et al. 1996 ; Jilling et al. 2018 ). We highlight how microbial physiological traits shape the fate of N once it is taken up by microbes. Finally, consistent with Drinkwater and Snapp’s ( 2007 ) agroecosystem N model and insights into priming mechanisms (e.g. Cheng and Coleman 1990 ; Dijkstra and Cheng 2007 ; Phillips et al. 2012 ; Zhu et al. 2014 ), we explicitly address the role of plants and their interactions with minerals and microbes in mobilizing N. Below we outline our new model, synthesize relevant new data, and examine some implications of our model in fertilized agroecosystems, aggrading and degrading soils, and under a changing global climate."
} | 2,106 |
32182360 | PMC7192604 | pmc | 9,113 | {
"abstract": "Abstract The evolution of regulatory networks in Bacteria has largely been explained at macroevolutionary scales through lateral gene transfer and gene duplication. Transcription factors (TF) have been found to be less conserved across species than their target genes (TG). This would be expected if TFs accumulate mutations faster than TGs. This hypothesis is supported by several lab evolution studies which found TFs, especially global regulators, to be frequently mutated. Despite these studies, the contribution of point mutations in TFs to the evolution of regulatory network is poorly understood. We tested if TFs show greater genetic variation than their TGs using whole-genome sequencing data from a large collection of Escherichia coli isolates. TFs were less diverse than their TGs across natural isolates, with TFs of large regulons being more conserved. In contrast, TFs showed higher mutation frequency in adaptive laboratory evolution experiments. However, over long-term laboratory evolution spanning 60 000 generations, mutation frequency in TFs gradually declined after a rapid initial burst. Extrapolating the dynamics of genetic variation from long-term laboratory evolution to natural populations, we propose that point mutations, conferring large-scale gene expression changes, may drive the early stages of adaptation but gene regulation is subjected to stronger purifying selection post adaptation.",
"introduction": "INTRODUCTION The dynamic environments colonized by bacteria demand optimal regulation of gene expression ( 1 ). Transcription initiation, the primary checkpoint in this regulation, is influenced by the activity of a set of DNA-binding proteins called transcription factors (TFs). TFs sense the cellular environment and respond by activating or suppressing the expression of their target genes (TG). Different species of bacteria occupying diverse niches, thus differ more in the set of their TFs than that of their TGs ( 2 ). The set of transcriptional regulatory interactions in an organism is usually represented as a transcriptional regulatory network (TRN). TRNs have been found to evolve faster than other biological networks ( 3 ), based on detection of orthologs across species. Their evolution has been explained largely by duplication ( 4 ) and horizontal gene transfer (HGT) ( 5 ). Even though both of these processes are accompanied/followed by DNA sequence level changes in the TFs ( 6 , 7 ), the contribution of point mutations to TRN evolution is poorly understood ( 8 ). The significance of point mutations can be realized by the fact that, even where both a TF and its TG are present, the regulatory interaction is often not conserved ( 9 ). Macroevolutionary changes in TRN can be explained, in principle, through mutations at microevolutionary scales, i.e. mutations may accumulate faster in TFs than in TGs within species, and this would be reflected as lower conservation of TFs across species. If selection drives TFs evolution, populations adapting to different environments may select for different mutations in TFs at a higher frequency than in TGs. Some of these may be loss-of-function mutations, leading to complete loss of the TF over a long period of time. In contrast, if TFs evolve through neutral processes, a population may have more standing genetic variation in TFs than in TGs, presumably due to weaker selective constraints. For the same reason, the loss of a TF at large evolutionary distance would be more likely than that of a TG. The adaptive evolution hypothesis seems to be supported by multiple lab evolution experiments, which found many beneficial mutations to occur in TFs ( 10 ). However, this cannot be concluded in the absence of a statistical analysis, of enrichment of beneficial mutations in TFs over TGs, across multiple such studies. Often, these mutations were found in the hubs of the TRN ( 11 ), generally referred to as ‘global’ regulators (GR). As TRN follows a power-law distribution of edges per nodes ( 12 ), only a few TFs act as GRs, and influence gene expression on a global scale. In this regard, majority of the TFs are considered as ‘fine-tuners’, and it is not evident if mutations in these TFs should also be more adaptive than in TGs. The role of adaptation in shaping gene regulation across wild strains also has been demonstrated by several studies ( 13–15 ). However, it is still not clear to what extent mutations in TFs drive regulatory diversification. Some studies suggest that few mutations in TFs may be sufficient for changes in regulation ( 7 , 16 , 17 ). Therefore, it is possible that even if the evolution of TFs is shaped by adaptation, we may find TFs to be less diverse than their TGs within a species. Additionally, lower diversity of TFs would be indicative of stronger selective constraints. Recent advances in genome-scale sequencing and analysis, and readily available large-scale WGS datasets offer an exciting opportunity to investigate the evolution of regulatory networks over short time-scales, i.e. across strains or within species. Equipped with tens of thousands of sequencing runs on Escherichia coli from various hosts and geographical regions, we set out to estimate the sequence diversity of a thousand genes. Using these datasets, we tested if transcription factors of a bacterial species are indeed more diverse than its target genes.",
"discussion": "DISCUSSION We showed that bacterial TFs are less diverse in sequence than their TGs within species, i.e. across short time-scales, and that their diversity is a function of their regulon size. It has been reported previously that global regulators (GR) are more conserved across species than other TFs ( 5 , 31 ). However, even after excluding GRs, we found that TFs were more conserved - in sequence - than their TGs within species. This was contrary to the conservation of these ‘local’ regulators (LR)—in terms of presence/absence—across species. If two bacterial species have widely different environments, then their set of TFs are also expected to be different. However, within species, the niche differences may not be drastic enough to warrant diverse TF alleles. Under this scenario, the low sequence diversity of TFs within species is indicative of stronger selective constraints, imposed by the requirement of their optimal activity in a given environment. As a corollary, adaptation to a new environment may demand a new optimum of gene expression which is conferred through mutations in TFs. Indeed, multiple adaptive lab evolution (ALE) studies were found to be enriched with regulatory mutations ( 10 ). We performed a statistical analysis on many experimental evolution studies to verify this observation. Indeed, we found that TF mutations were enriched in ALEs under those selection pressures which can be satisfied by changes in multiple pathways. In contrast, none of the mutation accumulation (MA) experiments showed an excess of mutations in TFs. To observe the long-term dynamics of the above trend, we analyzed whole-population data from an evolution experiment spanning 60 000 generations ( 54 ). We found that the frequency of mutations in TFs rapidly rose above that of TGs in first 10 000 generations and then declined over time. This trend was stronger for GRs and the decline was faster. However, only mutations in a few specific GRs conferred an advantage whereas multiple LRs were found to have beneficial mutations. In mutator populations, TFs and TGs accumulated mutations at similar rates and towards the end, any difference in trends seemed to be in accordance with selective constraints. By synthesizing these findings along with the existing body of literature, we propose the following model of TRN evolution in prokaryotes (Figure 8 ). As a population first encounters an environment, it experiences global expression changes, brought about by mutations to regulatory hubs ( 11 ). Since these changes may also have adverse pleiotropic effects, as evolution proceeds, more mutations accumulate in LRs ( 56 ). As a consequence of these mutations, specific pathways of TRN, which are irrelevant to the present selection pressure, are inactivated ( 59 ). However, the fitness benefit of these TF mutations decline over time, likely as a consequence of diminishing returns epistasis ( 60 ), and as adaptation decelerates, selective constraints play a bigger role in the mutation frequency. In a well adapted population, the optimal variants of TFs are maintained by purifying selection. Across environments, different segments of the TRN are targeted and inactivated, such that over millions of years, species adapted to different environments have few TFs in common ( 2 ). Figure 8. A graphical representation of the proposed model of TRN evolution in Bacteria. A population exposed to an environment X rapidly accumulates mutations in TFs, especially GRs. Some of these mutants reach fixation while also accumulating mutations in other TFs and TGs. In long-term, mutations in local TFs are more beneficial than in global TFs. Farther out in time, when the population is well adapted, mutants of greater fitness rarely appear and hence, TFs show low sequence diversity. Also, irrelevant TRN modules are eventually lost. In a different environment Y, adaptation may proceed through integration of or substitution by xenologs in the native network. Thus, across species comparisons show low conservation of TFs relative to their TGs. This model does not underestimate the significance of HGT and duplication in the growth of regulatory networks. However, it emphasizes the role of small-scale changes, observed over short time-scales, in its modification. Specifically, in the early stages of adaptation, these changes set the path for long-term evolution of the network and facilitates pruning of branches irrelevant to the new environment. A major concern with this proposition might be the extrapolation of results observed in LTEE, which is unrealistically simple as opposed to natural environments. However, similar dynamics were observed during in situ evolution of Pseudomonas aeruginosa in cystic fibrosis patients over 200 000 bacterial generations ( 61 ). As the population approached the fitness peak, the relative frequency of regulatory mutations declined and these dynamics were subsequently governed by negative selection. Given this consistency across two diverse species adapting to two drastically different environments, we feel confident in generalizing our proposed model to other Bacteria. A common assumption in comparative genomics of regulatory networks is the regulog concept, i.e. a regulatory interaction is transferable if orthologs of both TF and TG are present. Despite some counter-evidence that regulatory interactions change even in the presence of both partners ( 9 ), this is a reasonable assumption for within species comparison. We note that regulatory interactions can be modified by sequence changes in TFs ( 17 ). Adaptive transcriptomic polymorphism has been observed across E. coli strains ( 14 ), and regulatory divergence seems to be connected to divergence in the coding genome ( 15 ). This might seem contrary to our observation of low sequence diversity of TFs. However, the adaptive portion of these regulatory differences is often attributable to a few key mutations in TFs ( 7 , 16 ), that might have appeared very early-on in the course of adaptation. Another way by which regulatory evolution can occur is by mutations in cis-regulatory regions. Besides the high false-positive rate associated with the identification of these sites, the problem of estimation of variation across homologous sites and its comparison with variation in the coding genome is beyond the scope of the present work. However, if the variation in TFs and their binding sites(TFBS) could be compared, we suspect TFBS to be more variable than their corresponding TFs, if at all. Since local regulators, which regulate very few operons, were also more conserved than their TGs, variation in TFBS can allow for modification in the regulation of individual operons without perturbing the regulation of others. Another issue relevant to our study is that of differential impact of non-synonymous mutations on a protein's function. This difference arises not only as a consequence of differential physico-chemical properties of amino acids but also due to the differences in functional significance of the domain or site affected by the mutation. Through our analysis, we essentially rejected the null hypothesis of no difference in the distribution of fitness effects (DFE) between TFs and their TGs. In other words, fewer non-synonymous changes were observed in TFs, in the well-adapted natural populations, presumably due to a larger proportion of mutations being deleterious in TFs relative to their TGs. Further understanding on the effect of these mutations can be derived from a domain-specific estimation of variation. It is well established, in the metazoan gene regulation, that cis -regulatory regions are significant for phenotypic evolution ( 62 ). The contribution of mutations in trans-acting factors was assumed to be limited, due to the pleiotropic effect of such changes on gene regulation. However, a growing body of literature suggests that the adaptive changes in TFs are more common than imagined. Across species, TFs, along with other regulatory proteins, were more diverse as compared to other proteins ( 63 ). At least some TFs in human show high variation across populations, hinting at their role in local adaptation ( 64 ), although TFs may still be more conserved than their TGs on average. However, analysis of this sequence variation in our framework is complicated by the presence of long signaling cascades in eukaryotes, since many TGs also play crucial roles in gene regulation. In prokaryotes, previous studies have established the flexibility of TRN across species ( 3 ). Either two bacteria have vastly different set of TFs, as compared to TGs ( 2 ), or the gene-expression of only a small set of regulons is conserved ( 9 ). With our study, we highlight the presence of strong selective constraints on gene regulation, at least over small evolutionary distance. A large proportion of mutations in TFs seems to be deleterious. However, a small portion that can be beneficial in a novel environment, re-adjusts and optimizes the metabolic flux to the prevalent condition, and thereby both directs and constraints the future course of adaptation."
} | 3,628 |
40316720 | PMC12048519 | pmc | 9,115 | {
"abstract": "Studies on electron-transfer pathways in certain bacterial strains have revealed that the degree of coupling of electron transfer to proton translocation along the respiratory chain can be regulated according to metabolic demands. This first line of metabolic response, based on the existence of energy dissipation mechanisms, has not been demonstrated to be a general pattern across the bacterial kingdom, let alone to be operative in electro-active bacteria. In this study, we hypothesized that electro-active cells should respond to over-polarization by also triggering energy decoupling mechanisms to prevent metabolic overloads. Based on electrochemical analyses, we propose that the recently discovered inner-membrane cytochrome CbcBA - used by electro-active Geobacter sulfurreducens bacteria for cellular respiration near the thermodynamic energetic limit - can also act as an energy dissipation gate when the metabolism is demanded, contributing to regulate the energy balance of the cell by decoupling carbon assimilation from electrode respiration.",
"conclusion": "Concluding remarks The results of the present study contribute to expanding the current knowledge on energy decoupling mechanisms in bacteria, unveiling the functional activity of CbcBA in G. sulfurreducens as a dissipative electron-transfer path triggered when the metabolism is pushed to the limit. In context, this mechanism contributes to impart respiratory flexibility and energy management control, inducing low carbon assimilation when the respiration is forced beyond maximum levels. The recruitment of CbcBA during highly efficient respiratory activities may have relevant practical consequences on the area of bio-electrochemical current generation, counting on just the necessary number of cells to produce maximum amounts of charge per unit time without reaching the stratification limit of the biofilm; at the time that electrode clogging, efficiency losses and stop dead times might be minimized by lowering biomass-to-current ratios. As electro-activity appears to be widely distributed in the bacterial kingdom, further insights into the regulation of cellular electron flux as a function of the electrode potential - at the level of gene expression - should provide new hints to unveil other or similar mechanisms of energy control in bacteria.",
"introduction": "Introduction Living organisms are open thermodynamic systems that exchange matter and energy with the environment 1 . At cellular level, life depends on the regular external supply of electron/carbon sources to get a significant driving force to maintain the living system out of equilibrium (steady state) 2 . Microbial existence is a typical and well-known example of unicellular life where the structural and functional order arises under non-equilibrium conditions; however, due to the complexity of cell functioning, the metabolic mechanisms behind energy management control are not fully understood yet. Electron-transfer reactions play a fundamental role in biological respiration and energy conservation processes, taking place between molecular components in a chain of thermodynamically and kinetically favorable events used to generate transmembrane proton gradients, which in turn lead to the establishment of a proton motive force ( pmf ) that the cell uses to synthesize ATP as the primary form of conserved energy 3 . However, this is the case only if the catabolic energy-producing path is efficiently coupled to the energy storage process. Studies on electron-transfer pathways in certain bacterial strains indicate that the degree of coupling of electron transfer to proton translocation can be finely tuned by the microbial machinery according to metabolic demands, this occurring regardless of the thermodynamic hierarchy of the electron gates involved along the respiratory chain 4 , 5 . This mechanism seems to have been designed so as to function as a metabolic drain, through which the microorganisms can free themselves from all of the excess internal entropy they cannot help producing as a consequence of the irreversible metabolic activities. This gives rise to energy dissipation in the form of heat and/or chemical entropy 6 , due to an incomplete energy coupling between the catabolic process and the ATP synthesis. This first line of metabolic response, that seems to regulate the energy balance in some microbial strains, has not been demonstrated to be a general pattern across the bacterial kingdom, let alone to be operative in electro-active bacteria, the latter being capable of coupling intracellular organic matter oxidation with extracellular reduction processes 7 . Due to the fact that dissipation mechanisms play a central role in the metabolism of all living forms, and building on the existence of an upper limit on Gibbs energy dissipation for sustaining life 8 , we hypothesized that electro-active bacteria should respond to external polarization beyond that supporting maximal energy production by also triggering decoupling mechanisms, so as to prevent metabolic overloads induced by this non-natural condition. In this study, to experimentally test our hypothesis, we electrochemically grew electro-active G. sulfurreducens biofilms at increasing polarization potentials to study their voltammetric response, in the search for redox centers participating of the electron-transfer process out of the cell with the expectancy of finding clues leading to metabolic decoupling mechanisms. As a result, we here provide evidence on the recruitment of CbcBA - a recently discovered low-potential inner-membrane bc -type cytochrome in G. sulfurreducens that barely contributes to the generation of the pmf - as an electron-transfer path operating also at high potential beyond that which maximizes respiration, leading to a steep drop in cell energy conservation. This counteracting response, by using a low-potential electron gate when the cell is exposed to extreme oxidative conditions, contributes to regulate the metabolic energy balance of the electro-active cell, with CbcBA functioning as a decoupling gate.",
"discussion": "Results and discussion G. sulfurreducens inner-membrane cytochromes display performance flexibility as a function of the external potential Studies on electron-transfer pathways in G. sulfurreducens have demonstrated that at least three different potential-dependent electron gateways - the inner-membrane cytochromes CbcBA, CbcL and ImcH - are functional in channeling electrons out of the cell 9 – 11 . In accordance with the effective midpoint potentials associated to the metabolic pathways involving each protein (around −0.21 V vs SHE for CbcBA, −0.15 V for CbcL, and −0.05 V for ImcH) 10 , 12 , 13 , the latter are employed during respiration to harvest different amounts of free energy (ΔG) depending on the associated H + /e − stoichiometry. On this basis, we tested our hypothesis that electro-active microorganisms should count on dissipative mechanisms when taken to an overloaded respiration. To do this, we grew G. sulfurreducens biofilms on graphite electrodes polarized at different potentials (−0.23 V; −0.15 V; 0 V; +0.2 V; +0.4 V and +0.5 V vs SHE), and performed cyclic voltammetry analyses once stationary growth was reached to confirm the occurrence of the typical S-shaped oxidative-catalysis process for all of the biofilms at all potentials (Supplementary Material, Supplementary Section SM-1 , Supplementary Fig. SM-1 ). The first derivatives of the obtained voltammograms indeed revealed the participation of the three cytochromes in generating the catalytic signal, but showed potential-dependent differences in their contribution to the overall current (Supplementary Material, Supplementary Section SM-2 , Supplementary Fig. SM-2 ). To calculate the individual cytochrome participation to the total electron flux out of G. sulfurreducens we applied the three-site limiting current approach described by Zacharoff et al. and Howley et al. 12 , 14 . Our strategy here was focused on analyzing any differences arose in the voltammetry curves, evaluating the participation of each cytochrome as electron-transfer path out of the cell by fitting the entire voltammetry profile. The confirmation of the potential-dependent response by the shifting of the catalytic curve was taken as an indication of the influence of the growth conditions on the involved paths. The results are summarized in Fig. 1 . Fig. 1 Electron-transfer path splitting along the inner-membrane of G. sulfurreducens as a function of the external electrode potential. The three cytochromes are active regardless of the external potential at which the biofilms were grown (except for the lowest potential tested, at which only CbcBA and CbcL were detected), thus partitioning the catabolic energy along the inner membrane. The electron path splitting through each cytochrome (represented as % of participation) depends on both the amount of each cytochrome along the electron-transport chain (given by genetic control) and their biochemical activity (metabolic control). For comparative purposes, the results are displayed together with those reported in previous studies: red crosses (CbcL activity) 14 , green crosses (CbcBA activity) 14 , black cross (CbcBA activity) 16 . n = 2 (independent experiments for each potential tested, represented as dots). The gradual changes in contribution to the total current for each of the three cytochromes as the growth potential was varied were considered as a proxy of the polarization-dependent variations in the electron-transfer paths experienced during growth, suggesting somehow the branching of the electron flow out of the cell while growing under different polarization conditions. In agreement with previous studies 10 , our results indicate that CbcBA is used by G. sulfurreducens as the main electron gate when the cell faces energy limitations as a consequence of the low potential of the electron acceptor (−0.23 V vs SHE), this leading to an energetic balance that is only enough to meet cellular maintenance requirements, a condition fulfilled when the pmf approaches zero. As the electrode potential is raised, and the total current reaches to the absolute maximum, the participation of CbcBA markedly drops to a minimum level of activity (which was sustained up to +0.4 V) (Fig. 1 ), being replaced by the transporter associated to the second proton pump (CbcL), which takes the role of the main electron gate in the −0.15/+0.4 potential window. The inverse correlation between the two low-potential proteins is complemented by a continuous increase of the activity of ImcH along the potential window (Fig. 1 ), the third coupling site in the transport chain. Set in a thermodynamic frame, in which the inner-membrane cytochromes are organized in a redox sequence that would allow the efficient switching from low- to high-potential electron-transfer pathways as the external potential is raised, the electrochemical activity detected here for the three cytochromes as a function of the electrode potential deserves a first comment. Whereas a mechanism through which the activation of a high-potential gate would be coupled to the shutting off of the preceding lower-potential path is consistent with experimental evidence collected on single/multiple mutant strains (deletions of genes encoding CbcBA, CbcL or ImcH), the underlying potential/electron-gate mechanism in wild-type (WT) strains could a priori pose a different scenario, leading to a non-identical behavior compared to that of mutant microorganisms 15 . Although the electrochemical activity of ImcH as an electron gate could have been anticipated to take center stage as the electrode potential was raised, the obtained results are not unexpected, lining with previous studies that have already revealed a higher participation of CbcL (compared to that of ImcH) at potentials at which a higher involvement of ImcH would have been beforehand awaited 14 . Additional support to the activity not only of CbcL, but also of CbcBA, above −0.1 V was reported by Marsili et al. 16 . In line with these studies, Zacharoff et al. demonstrated that in fact CbcL contributes to over 60% of the total current at +0.24 V in WT strains 12 , in excellent agreement with the results depicted in Fig. 1 . In this scenario, and according to current literature on metabolic gradients identified throughout biofilms as they approach the stationary state under continuous cultures, we should also be cautious and point initially on the possibility that not all the cells along the biofilm would be experiencing the same effective potential, which could also contribute to the observed potential-dependent effect. This point will be further addressed in sections below. Taking into consideration that the electron flow through each of the three inner-membrane cytochromes is thermodynamically (and kinetically) feasible - added to the fact that electro-active bacteria rely on branched respiratory chains associated to multiple metabolic electron-flux patterns - it would not be surprising that in WT strains the three electron gateways might take part simultaneously (although in different proportions) of the electron-transfer process in a range of external potentials. In the present study we intend to contribute to the current knowledge on the electrochemical behavior of the inner-membrane cytochromes in G. sulfurreducens , demonstrating their dynamic flexibility as a function of the electrode potential. As to this flexibility, when the potential was poised beyond that supporting maximum current generation (+0.4 V), an unexpected contribution of the lowest-potential CbcBA cytochrome was noticeably detected (Fig. 1 , +0.5 V), working together with the other two paths, but becoming the main route for transferring electrons out of the cell. In accordance with the above mentioned, two scenarios initially arise: (i) or CbcBA takes a center role in far-from-the-electrode bacteria exposed to lower effective potentials, under a condition in which the bacteria near the electrode would be prevented from functioning at full metabolism maybe caused by local stress at high potential (+0.5 V); or (ii) as it bypasses the thermodynamic hierarchy of the electron transporters, the result obtained would be a consequence of a mechanism of proton pump (energy) regulation along the inner-membrane, compatible with the existence of an overall physiological flexibility. The respiratory flexibility proposed here for G. sulfurreducens is not unknown in cellular dynamics, representing a strategy used not only by bacteria (like E. coli and/or P. pantotrophus ) 4 , 5 , but also by eukaryotes through modulation of the activity of the complexes I, III and IV along the mitochondrial respiratory chain 17 – 19 . This mechanism supports electron-transfer regulation through inner-membrane proteins (e.g., heterodimeric reductases, terminal oxidases) that display flexible activity to counteract extreme metabolic changes. In bacteria, respiration flexibility has been nicely demonstrated for example in P. pantotrophus , where the presence of nitrate reductases alongside multiple oxidases in the respiratory chain during the aerobic growth of the microorganism seems to respond to a tight metabolic control. In analogy to our results, P. pantotrophus up-regulates - strikingly under excess oxygen - an inner-membrane system (the NapABC complex) which catalyzes the reduction of nitrate without contributing to proton translocation 20 . This system is used as an electron gate that prevents the building-up of an excessive transmembrane proton gradient (acidification of the periplasm). Similarly, E. coli has also been demonstrated to exhibit strong respiratory flexibility, by decoupling catabolism from ATP synthesis through the use of an inner-membrane cytochrome (cytochrome bd -II oxidase) that plays the role of an electron carrier that does not contribute to the pmf \n 4 . Likewise, a number of studies have also reported on the dropping of the microbial growth yield when the metabolism is demanded, suggesting the existence of an active energy-wasting mechanism at the core of the microbial respiratory chain 21 , 22 . This leads to wide variation in microbial energetic efficiencies, as an adaptive response to changes in the electron influx into the respiratory chain and/or to the potential of the available electron acceptors 23 . According to this background, we anticipated that the flexibility evidenced in the participation of the three inner-membrane cytochromes in G. sulfurreducens might play a central role in the energy partitioning mechanisms inside the electro-active cell. We then speculated that the proton translocation activity of each electron-transfer path might offer clues on possible dissipation gates. The two constitutively expressed cytochromes ImcH and CbcL have been demonstrated to effectively contribute to the pmf when operating as electron paths, although supporting different growth yields; with ImcH being essential for respiration under energy-plentiful conditions, a situation encountered when G. sulfurreducens faces high-potential acceptors and/or electrodes poised above −0.1 V (SHE) 13 . As a complement, CbcL has been demonstrated to couple electron transfer to proton translocation when the redox potential falls below −0.1 V 24 ; whereas the recently discovered bc -type cytochrome CbcBA has been shown to be essential for respiration when the cell accesses only to limited energy, acting as a path that barely contributes to proton translocation 10 . By considering the Gibbs energy change associated to the electron-transfer processes involved in proton translocation during respiration ( \\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}$${\\Delta {{\\rm{G}}}}_{{{{\\rm{e}}}}^{-}}$$\\end{document} Δ G e − ), and regarding that the driving redox reaction is the only energy source feeding the translocation activity ( \\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}$${\\Delta {{\\rm{G}}}}_{{{{\\rm{H}}}}^{+}}$$\\end{document} Δ G H + ), it is possible to estimate the maximum amount of protons that could be translocated per electron through each protein (H + /e − ratio) on the basis of upper limits imposed by energy conservation 25 , 26 . In stark contrast to linear electron-transport chains in eukaryotes, electro-active bacteria rely on branched respiratory chains associated to multiple metabolic electron-flux patterns. This poses certain difficulties when it comes to identifying the redox counterparts involved in proton translocation through CbcBA, CbcL and ImcH; especially when the electron-transfer out of the cell requires multiple soluble carriers (not yet univocally identified) across the periplasm. Recognizing that energy conservation takes place at the inner-membrane level, it is still possible to make an estimation of the maximum energy available to translocate charges along the respiratory chain of G. sulfurreducens . According to reported physiological data, respiratory membranes for different strains and under a variety of conditions exhibit pmf values spanning 130–200 mV 27 – 30 . Due to the lack of experimental pmf data for G. sulfurreducens , we approached this value to the one experimentally measured for E. coli growing under anaerobic conditions (140 mV) 27 , on the basis of conserved structural and functional respiratory patterns across the bacterial domain 31 . Thus, an energy of 13.5 kJ would be required in G. sulfurreducens to translocate a minimum of one mol of charge against its electrochemical potential. By considering the maximum energy available along the inner membrane up to the location of each cytochrome, maximum H + /e - ratios of ∼ 0.8, 1.2 and 1.9 would be accessible when CbcBA \\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}$$({\\Delta {{\\rm{G}}}}_{{{\\rm{CbcBA}}}}=-10.6\\frac{{{\\rm{kJ}}}}{{{\\rm{mol}}}\\, {{{\\rm{e}}}}^{-}})$$\\end{document} ( Δ G CbcBA = − 10.6 kJ mol e − ) , CbcL \\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}$$({\\Delta {{\\rm{G}}}}_{{{\\rm{CbcL}}}}=-16.4\\frac{{{\\rm{kJ}}}}{{{\\rm{mol}}}\\, {{{\\rm{e}}}}^{-}})$$\\end{document} ( Δ G CbcL = − 16.4 kJ mol e − ) and ImcH \\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}$$({\\Delta {{\\rm{G}}}}_{{{\\rm{ImcH}}}}=-26.0\\frac{{{\\rm{kJ}}}}{{{\\rm{mol}}}\\, {{{\\rm{e}}}}^{-}})$$\\end{document} ( Δ G ImcH = − 26.0 kJ mol e − ) take part of the electron transfer process, respectively. These values only set upper limits given by energetic constrains, the physiological number of charges effectively translocated depending on the periplasmic redox counterparts for each cytochrome and the coupling efficiency of each gate, the latter finally dictated by external boundary conditions. According to its thermodynamic hierarchy, and in agreement with recently published results 10 , CbcBA would be effectively involved in a path contributing to a much lesser extent to proton translocation compared to the other two inner-membrane cytochromes, marginally providing energy to translocate one proton per electron even at a hypothetical case of 100% coupling efficiency. Interestingly, H + /e − ratios lower than unity have been reported to be operative, for example, in the complex I of E. coli and other bacteria, when the Gibbs energy of the driving redox reaction is not enough to meet the corresponding proton translocation stoichiometry. In these cases, additional energy sources which partially dissipate the membrane electrochemical potential (e.g., Na + /H + antiporter mechanisms) are needed to build up the remaining energy for net translocation according to the cell needs 32 , 33 . In this scenario, the results shown in Fig. 1 can be interpreted in the context of the existence of parallel electron routes out of G. sulfurreducens depending on the external conditions, with the three cytochromes playing a simultaneous active role and CbcBA acting as a metabolic way out far from equilibrium marginally contributing to the pmf . Energy transduction mechanisms associated to coupled processes, like redox-coupled proton translocation phenomena, are typically addressed within the formalism of linear non-equilibrium thermodynamics, as long as the linearity of the flux-force functions of the coupled phenomena can be guaranteed 19 , 34 . This can only be ensured in regions about the equilibrium - at moderate growing conditions - where Onsager symmetry relations hold; however, at critical distances far from equilibrium the linear behavior cannot longer be assumed. In this sense, under a demanded metabolism - like the one induced under polarization at extreme external potentials - it would be in principle expected the charge-translocating energy converters to move from the linear regime, leading to H + /e − stoichiometries deviating even more from the maximum values dictated by the energetic upper limits. This probably represents a challenge for cell homeostasis in terms of, for example, local pH gradients, supporting the role of the molecular complex involved in the path of lowest H + /e − ratio (CbcBA) as an emergency gate. In a scenario in which coupled linear processes cannot be ensured, and due to the lack of both mechanistic and molecular information on proton translocation activities for each cytochrome in G. sulfurreducens , the H + /e − stoichiometric description above was complemented with calculations of energy balances associated to the experimentally determined production of microbial biomass as a function of the imposed external potential. CbcBA activation at high potentials lessens carbon assimilation To further explore energetic balances in response to the potential of the electron acceptor, biomass yields were calculated on the basis of acetate consumption (Y X,Acetate ). For this, experimental bacterial counting data were combined with the amount of electron/carbon donor consumed at each growing potential. The relation between the amount of acetate consumed and the corresponding biomass produced, the latter expressed as C-mol of G. sulfurreducens , was determined by identifying the catabolic and anabolic reactions involved in the production of G. sulfurreducens cells, establishing for this the macrochemical equation for the heterotrophic growth on acetate. From a physicochemical point of view, and in stark contrast to isolated or closed systems, living organisms are open thermodynamic systems that exchange matter and energy with the environment. Chemotrophic life thus depends on the regular external supply of electron/carbon sources to get a significant driving force to maintain the living system in steady state. In this frame, to calculate the relation between the total amount of acetate consumed by G. sulfurreducens and the corresponding biomass produced, we first identified the catabolic and anabolic reactions that take place in the heterotrophic production of G. sulfurreducens cells growing on acetate. The microbial metabolism is typically simplified by considering the overall catabolic and anabolic reactions coupled by an energy carrier (ATP). The stoichiometry of these coupled reactions further allows to define mass, charge and Gibbs free energy balances, from which the biomass yield on the corresponding electron/carbon source can be finally determined 2 . After being taken up by the cell, part of the acetate is oxidized in the catabolic reaction to produce energy, while the remaining is employed in the anabolic reaction to produce new cells. The catabolic reaction corresponding to the oxidation of acetate is (per mol of electron donor): 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}$${{{\\rm{CH}}}}_{3}{{\\rm{CO}}}{{{\\rm{O}}}}^{-}+4{{{\\rm{H}}}}_{2}{{\\rm{O}}}\\to {2{{\\rm{HCO}}}}_{{3}^{-}}+9{{{\\rm{H}}}}^{+}+8\\, {{{\\rm{e}}}}^{-}$$\\end{document} CH 3 CO O − + 4 H 2 O → 2 HCO 3 − + 9 H + + 8 e − Acetate molecules not used in the catabolic reaction are diverted to the anabolic process. Electron transfer in the anabolism depends on the oxidation state of the carbon source compared to that of biomass. When the degree of reduction of the carbon source is lower than that of biomass, the substrate has to act not only as a source of carbon, but to the electrons as well. According to the degree of reduction of acetate (4 mol e − /C-mol acetate) and the elemental composition of one C-mol of G. sulfurreducens (CH 1.84 O 0.42 N 0.17 , 4.49 mol e − /C-mol G. sulfurreducens cell) 35 , part of the acetate diverted to the anabolic reaction is first oxidized to supply the electrons necessary to reach the degree of reduction of biomass, while the remaining is used as carbon source. These reactions can be represented as: 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}$$0.06125\\,{{{\\rm{CH}}}}_{3}{{\\rm{CO}}}{{{\\rm{O}}}}^{-}+0.245\\,{{{\\rm{H}}}}_{2}{{\\rm{O}}}\\to 0.1225\\,{{{\\rm{HCO}}}}_{{3}^{-}}+0.55125\\,{{{\\rm{H}}}}^{+}+0.49 \\, {{{\\rm{e}}}}^{-}$$\\end{document} 0.06125 CH 3 CO O − + 0.245 H 2 O → 0.1225 HCO 3 − + 0.55125 H + + 0.49 e − 3 \\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}$$0.5\\,{{{\\rm{CH}}}}_{3}{{\\rm{CO}}}{{{\\rm{O}}}}^{-}+0.17\\,{{\\rm{N}}}{{{\\rm{H}}}}_{{4}^{+}}+0.82{{{\\rm{H}}}}^{+}+0.49{{{\\rm{e}}}}^{-}\\to {{{\\rm{CH}}}}_{1.84}{{{\\rm{O}}}}_{0.42}{{{\\rm{N}}}}_{0.17}+0.58{{{\\rm{H}}}}_{2}{{\\rm{O}}}$$\\end{document} 0.5 CH 3 CO O − + 0.17 N H 4 + + 0.82 H + + 0.49 e − → CH 1.84 O 0.42 N 0.17 + 0.58 H 2 O where the reaction (2) embodies the oxidation of acetate to supply the electrons needed in the production of one C-mol biomass (reaction (3)). According to (2) and (3), the overall anabolic reaction for the production of one C-mol of G. sulfurreducens is given by: 4 \\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}$$0.56125\\,{{{\\rm{CH}}}}_{3}{{\\rm{CO}}}{{{\\rm{O}}}}^{-}+0.17\\,{{\\rm{N}}}{{{\\rm{H}}}}_{{4}^{+}}+0.26875{{{\\rm{H}}}}^{+}\\to \t{{{\\rm{CH}}}}_{1.84}{{{\\rm{O}}}}_{0.42}{{{\\rm{N}}}}_{0.17}\\\\ \t+0.1225{{{\\rm{HCO}}}}_{{3}^{-}}+0.335{{{\\rm{H}}}}_{2}{{\\rm{O}}}$$\\end{document} 0.56125 CH 3 CO O − + 0.17 N H 4 + + 0.26875 H + → CH 1.84 O 0.42 N 0.17 + 0.1225 HCO 3 − + 0.335 H 2 O The stoichiometries of the reactions (1) and (4) were finally used to link the total amount of acetate consumed - both that consumed in the oxidation process (calculated from the electrical charge experimentally determined from the chronoamperometries once steady state was reached) plus that used in cell synthesis - and the C-mol of G. sulfurreducens cells grown (experimentally measured) for each of the conditions evaluated. The Table shown in the Supplementary Material (Supplementary Section SM-3 ) depicts the calculated biomass yields (Y X,Acetate ) expressed as moles of carbon of biomass (C-mol) per mol acetate consumed. Figure 2 below shows the obtained biomass yields as a function of the electrode potential. Fig. 2 G. sulfurreducens cell yield based on acetate oxidation as a function of the electrode potential. Microbial yield revealed the effect of two opposing mechanisms: at low potentials (low available energy) proton translocation barely contributes to ATP synthesis; at extreme high potentials (demanded respiration) dissipating mechanisms lead to low carbon assimilation efficiencies. n = 2 (independent experiments for each potential tested, represented as dots). The biomass yield showed a positive exponential trend as a function of the external potential in the −0.23 V/+0.4 V window, increasing the biomass production per unit acetate consumed as the energy became increasingly available. When the biofilm was grown at +0.5 V, the yield abruptly dropped to values well comparable to those for biofilms growing under limiting energy at the lowest potentials, indicating that most of the energy was not directed to growth. Analyzed in context with the increased contribution of CbcBA as the main electron-transfer path at +0.5 V (Fig. 1 ), the result is compatible with a considerable fraction of the available electrons flowing through a route of low proton translocation activity, thus not leading to significant biomass production. From an energetic point of view, this supports the occurrence of an energy dissipation mechanism to maintain cell viability. This is consistent with thermodynamic efficiency calculations (η, Supplementary Material, Supplementary Section SM-4 , Supplementary Fig. SM-4 ), the latter representing the fraction of the catabolic energy that is effectively conserved in the anabolic reaction. Interestingly, the thermodynamic efficiency calculated from the catabolic energy partitioning can in turn be linked to the overall thermodynamic efficiency of the energy converters along the respiratory chain; in particular, to the redox-driven proton translocation processes and the associated output/input powers (forces and fluxes) of the two coupled phenomena 19 : 5 \\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}$${{\\rm{\\eta }}}=-\\frac{{J}_{H\\,.\\, {pmf}}}{{J}_{e\\,.\\,}\\Delta {{\\rm{E}}}}$$\\end{document} η = − J H . p m f J e . Δ E where \\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}$${J}_{H}$$\\end{document} J H is the net proton flux, \\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}$${J}_{{e}}$$\\end{document} J e the electron flux, pmf the proton motive force and ΔE the corresponding redox span. The negative sign implies that the proton flux occurs uphill; that is, against its conjugate force ( pmf ). Although relations of the form shown in Eq. ( 5 ) have been typically used for the description of linear energy converters 19 , 34 , 36 , the dissipation function from which energy efficiency correlations are derived is valid unrestrictedly, regardless the formalism is addressed in the Onsager region or not 37 . Thus - within the phenomenological description of the coupled energy transduction processes and their relation to the thermodynamic efficiency calculated - at maximum redox input, a lower thermodynamic efficiency can only indicate a lower net proton flux and/or pmf , or a higher electron flux. The recruitment at +0.5 V of the electron gate involved in the pathway of lowest H + /e − stoichiometry (CbcBA), along with a maximal overall electron flux out of the cell (1.78 pA/cell at +0.5 V compared to 0.46 pA/cell at +0.4 V, determined from bacterial counting and current density measurements at steady state), support the lower thermodynamic efficiency at the highest potential, in a scheme in which CbcBA acts as a dissipative path contributing to decouple the anabolic process from the electrode respiration. In essence, as the potential exceeds that which maximizes respiration (+0.4 V), an increased portion of the energy derived from the catabolic reaction is wasted, requiring metabolic decoupling mechanisms for this to occur. The recruitment of the lowest-midpoint potential inner-membrane cytochrome CbcBA, matching the point at which the cells were taken to an overloaded respiration, reveals the response of the electro-active cells against the pressure exerted by a metabolism pushed to the limit. In this context it is important to highlight that the energy balance, together with voltammetry results (addressed in the next section), indicates that the CbcBA activity at the highest potential cannot be attributed to a distance-dependent drop in redox potential along the biofilm. If that were the case, the higher bacterial counting measured for the biofilm grown at +0.4 V \\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}$$(\\frac{1.39x{10}^{9}{{\\rm{bacteria}}}}{{{{\\rm{cm}}}}^{2}})$$\\end{document} ( 1.39 x 10 9 bacteria cm 2 ) should have displayed a concomitant higher participation of CbcBA compared to that at +0.5 V \\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}$$(\\frac{3.94x{10}^{8}{{\\rm{bacteria}}}}{{{{\\rm{cm}}}}^{2}})$$\\end{document} ( 3.94 x 10 8 bacteria cm 2 ) , rather than the opposite trend (Fig. 1 ). In order to rule out possible drops in biomass counting due to cell detachment associated to harmful conditions (in the event they had occurred, triggered by the high potential), planktonic cell counts were also performed in the reactor media. The results showed that - both at +0.4 and +0.5 V - around 90% of the total biomass was present in the biofilms. Revisiting the literature on the area, our results were pretty much the same as those previously reported for G. sulfurreducens 38 , demonstrating similar biofilm dynamics regardless of the external potential. This fact gives additional support to the sudden drop in cell replication on the electrode surface as a consequence of the activation of CbcBA at the highest potential tested. Out of the relevant practical consequences that the presented results could have in the area of bio-electrochemical current generation it is to be highlighted that, although there were noticeable differences in biomass production, the electrical current density reached by the biofilms grown at +0.4 V and +0.5 V was well comparable (Supplementary Material, Section SM-5 , Supplementary Fig. SM-5 ). This would bring about a profound impact on power-generating bio-electrochemical systems, counting on active cells working at their most while avoiding at the same time futile biomass accumulation occurring in typical physiological stratification events that lead to the drop of the mean electrochemical performance 39 . In other words, polarization beyond that required for maximal current production may prevent accumulation of cells that do not contribute to current production, thus leading to an optimized technological performance. CbcBA is recruited in the late exponential/early stationary phase at +0.5 V (SHE) It is known that both cbcL and imcH are constitutively expressed in G. sulfurreducens ; however, unlike its inner-membrane partners, cbcBA transcription is metabolically regulated, requiring the σ 54 -dependent BccR regulator to be expressed 10 . To gain insight into the recruitment of CbcBA over time during the evolution of the microbial biofilm grown at +0.5 V, cyclic voltammetry scans were performed at different growing stages along the chronoamperometry evaluation. Figure 3 shows the chronoamperometry of the biofilm at +0.5 V, along with the first derivatives of the voltammograms collected at different evolution times. Fig. 3 Electrochemical behavior of the inner-membrane cytochromes in G. sulfurreducens as a function of biofilm evolution at + 0.5 V (SHE). a Normalized chronoamperometry under catalytic conditions; b first derivatives of the voltammograms collected at different growing stages along the chronoamperometry evolution. The broken line in ( b ) indicates the mid-point potential associated to the metabolic path involving mainly the participation of CbcBA. Colors are related in each figure to a particular growing stage. A progressive shift to higher potentials was evidenced in the first 72 h (right-pointing arrow in ( b )), reversing back to lower potentials as CbcBA takes over (left-pointing arrow). No catalytic signals of CbcBA were detected during the first days of evolution (up to 72 h, included), indicating that neither the early biofilm nor the exponentially-growing microorganisms seem to have needed this cytochrome to channel electrons out of the cell along the first growing phase. This, as expected, agrees with the non-constitutive nature of CbcBA. However, a strong shift of the midpoint towards lower potentials - matching the slowing down of growth - was clearly observed in the late exponential/early stationary phase (beyond 72 h), continuously evolving and becoming more negative well into the stable stationary phase (168 h). This result is in sharp agreement with those reported by Joshi et al. 10 , who demonstrated that the expression of the cbcBA operon in G. sulfurreducens growing on Fe(III)-citrate is strongly up-regulated at the late exponential/early stationary stage, but not before. The results shown in Fig. 3b reveal overall physiological shifts during the biofilm evolution, calling again the attention to the causes triggering CbcBA appearance. As mentioned along the discussion of the results shown in Fig. 1 , CbcBA could take part of the electron transfer process at extreme potentials as a consequence of a potential drop along the developing biofilm (with far-from-the-electrode bacteria exposed to lower effective potentials, counting on CbcBA for channeling electrons out of the cell), or due to a process triggered in response to metabolic needs. On the basis of bacterial counting at +0.4 V and +0.5 V, as addressed in the previous section, we hypothesize that in this case the recruitment of CbcBA at the highest potential would not be attributed to a distance-dependent drop in redox potential throughout the biofilm, but to a metabolic response contributing to the redistribution of energy along the inner membrane by the ´re-allocation´ of the participation of each of the inner-membrane cytochromes. These findings give way to future studies, e.g. conducted on mutant strains, studies on ATP counting as a function of evolution stage, among others; so as to complement (and eventually support) the proposed hypothesis. In accordance to our results, we conclude that the CbcBA activity triggered in the biofilm grown at +0.5 V cannot be either attributed to voltammetry scan-induced effects, since, in that case, CbcBA would have been detected in all the stages during growth. Particularly interesting, the overall physiological shifts shown here indicate a certain plasticity in performance of the three inner-membrane cytochromes along the biofilm evolution, with an initial displacement of the catalytic envelope towards higher potentials, then followed by a counteracting shift to lower potentials just before the stationary phase sets in. The dynamic modulation of the inner-membrane behavior suggests a synergistic coupling of the three electron gates over the biofilm evolution, which also gives support to their potential-dependent activities, as depicted in Fig. 1 . Revised electron-transfer path model in G. sulfurreducens On the basis of the simultaneous involvement of the three inner-membrane cytochromes in the respiration process, and considering the role of CbcBA as a dissipative path at overloaded respiration, we propose a revised model of the electron-transfer paths in G. sulfurreducens as a function of the electrode potential (Fig. 4 ). The schematic representation, far from stablishing a definitive discussion on the topic, is deliberately represented in a simplistic manner, just as a way of expanding the current knowledge on the electron-transfer paths in this microorganism, leaving the door open to new results to come, mainly in the unexplored area of metabolic overloads. Fig. 4 Simplified scheme for the electron-transfer path splitting in G. sulfurreducens through CbcBA, CbcL and ImcH as a function of the electrode potential. The metabolic scenarios corresponding to ( a – d ) are discussed in the text. The amount of periplasmic protons depicted does not mirror the exact H + /e - stoichiometry, but just emphasizes the building-up of protons as the recruitment of the cytochromes progresses in response to increasing electrode potentials. The basis of the sketch was taken from the representation depicted in the Fig. 1 of ref. 24 , so as to expand the interpretation by including the results of the present study. According to its active role as an electron gate near the thermodynamic limit of respiration, CbcBA is used as the main path (with a minor contribution of CbcL) at the lowest potential, consistent with the participation detected for each cytochrome in the biofilms grown at −0.23 V (Fig. 1 ). This pathway represents a gate that barely conserves energy and that only helps maintaining cell viability supporting microbial life. This scenario is illustrated in the panel (a) of Fig. 4 . At higher external potentials, greater transmembrane proton pumping stoichiometries are expected as a consequence of the increasing participation of CbcL to the electron flow, accompanied by the attenuation of CbcBA, as revealed by the inverse relationship observed for both proteins (Fig. 1 ). This is reinforced when the external potential goes beyond −0.1 V, from which the activity of ImcH becomes increasingly relevant (Fig. 1 ) leading to higher proton translocation. This is represented in the Fig. 4 by progressively going from panel (b) to (c). This situation leads to an increase of both the pmf and the ATP synthesis, being mirrored in an increasingly higher biomass yield (up to +0.4 V in Fig. 2 ) and an efficient use of the available catabolic energy at each electrode potential (higher thermodynamic efficiency, Supplementary Material, Section SM-4 ). At potentials beyond +0.4 V, the building-up of an excessive transmembrane proton gradient would lead to both the acidification of the periplasm and the maximization of the electrochemical potential, inducing the partial decoupling of the electron-transfer process from the oxidative phosphorylation, this being boosted by either a raise of the electron influx in the respiratory chain and/or a high ATP/ADP ratio. To this respect, two possible decoupling mechanisms have been identified in energy-conserving processes along the inner-membrane of bacteria, also shared by mitochondria during the aerobic ATP synthesis in eukaryotes 40 . One of these mechanisms has been associated to proton gradient degradation - termed ´proton leak´ - via decouplers that backflow protons through the plasma membrane (a mechanism coined ´extrinsic uncoupling´). The other mechanism has been related to a decrease of the proton translocation efficiency itself - termed ´proton slip´ - via a decreased H + /e - ratio through the use of less thermodynamically-efficient routes with decreased proton pumping stoichiometries (´intrinsic uncoupling´), bypassing the redox hierarchy of the available electron paths 19 , 40 . On this background, we propose that intrinsic decoupling mechanisms take part of the electron-transfer process out of G. sulfurreducens when the external electrode potential exceeds certain levels (beyond +0.4 V vs SHE). In this scenario, electron paths involving low H + /e − stoichiometries must be used as energy dissipation gates, different from those depending on CbcL and ImcH. According to our results, CbcBA - the only inner-membrane cytochrome in G. sulfurreducens demonstrated to barely contribute to the pmf 10 - is reactivated at +0.5 V (Fig. 1 ), becoming a non-energy conserving (dissipative) electron gate when the metabolism is demanded (Fig. 2 and Supplementary Fig. SM-4 ). The situation of metabolic overload triggering CbcBA activation (at the expense of CbcL attenuation) is represented in the panel (d) of Fig. 4 . Regarding the attenuated activity of CbcL at +0.5 V, this fact must have also promoted a concomitant fall in ATP production, thus contributing as well to the overall bacterial biomass drop. In this scenario, alternative mechanisms (yet to be tested) providing additional decoupling routes cannot either be a priori ruled out. As to this, possible partial inhibitions of ImcH and/or porin-cytochromes at extreme potentials - likely inducing the electro-active cells to use different lower- midpoint potential structures - should be also bear in mind, probably complementing the activation of CbcBA. Once the electrons are transferred through each of the inner-membrane cytochromes, they have to reach out the outer membrane across the periplasm. For this, G. sulfurreducens counts on a broad periplasmic electron transfer network, which has been reported to be less responsive to the external potential compared to the inner-membrane cytochromes 41 . This opens up the possibility for CbcBA to loosely interact with a diversity of electron carriers to transport the electrical charge to the outer membrane. Particularly, G. sulfurreducens has been demonstrated to make use of a family of five tri-heme redox-mediating periplasmic cytochromes (PpcABCDE) with a variety of characteristics (expression levels, mid-point potentials, Bohr-redox effects), which could take part in the periplasm electron transfer network. Whereas it had been first hypothesized that each Ppc was involved in specific respiratory pathways depending on the external potential and energy needs, they were finally demonstrated to display not only lack of specificity as redox mediators (which would contribute to the respiratory flexibility) but being not completely essential either when an electrode is used as the external acceptor 42 . This supports, on the one hand, the existence of a flexible path bridging the periplasm that might contribute to the electron transfer; and suggests on the other hand that a yet unidentified electron transfer network is employed in G. sulfurreducens to channel electrons out of the cell. In this context it is noteworthy that other bacteria, including another model microorganism regarding extracellular reduction of electron acceptors - like Shewanella oneidensis 43 - and other Gram-negative bacteria ( Paracoccus denitrificans ) 44 have demonstrated to count on a dynamic periplasmic electron transfer network to impart respiratory flexibility, not sticking to any redox thermodynamic hierarchy, regardless of the growing conditions. This background would give support to the possibility for G. sulfurreducens to conduct electrons out of the cell through different pathways bridging the periplasm regardless of the external potential, with CbcBA being activated at high potentials as part of the electron-transfer machinery that provides respiratory flexibility, contributing to prevent the microorganisms from becoming metabolically compromised. Concluding remarks The results of the present study contribute to expanding the current knowledge on energy decoupling mechanisms in bacteria, unveiling the functional activity of CbcBA in G. sulfurreducens as a dissipative electron-transfer path triggered when the metabolism is pushed to the limit. In context, this mechanism contributes to impart respiratory flexibility and energy management control, inducing low carbon assimilation when the respiration is forced beyond maximum levels. The recruitment of CbcBA during highly efficient respiratory activities may have relevant practical consequences on the area of bio-electrochemical current generation, counting on just the necessary number of cells to produce maximum amounts of charge per unit time without reaching the stratification limit of the biofilm; at the time that electrode clogging, efficiency losses and stop dead times might be minimized by lowering biomass-to-current ratios. As electro-activity appears to be widely distributed in the bacterial kingdom, further insights into the regulation of cellular electron flux as a function of the electrode potential - at the level of gene expression - should provide new hints to unveil other or similar mechanisms of energy control in bacteria."
} | 13,034 |
39516921 | PMC11549841 | pmc | 9,116 | {
"abstract": "Background Modulating the microbiome with nanomaterials has been proposed to improve plant growth, and reduce reliance on external inputs. Carbon Nanosol (CNS) was attracted for its potential to improve plant productivity. However, the mechanism between CNS and rhizosphere microorganisms remained largely elusive. Results Here, we tried to systematically explore the effects of CNS (600 and 1200 mg/L by concentration) on tobacco growth, soil physical properties, and root-associated microbiome. The influence of CNS on soil physicochemical properties and plant growth was significant and dose-dependent, leading to a 28.82% increase in biomass accumulation by 600 mg/L CNS. Comparison between the CNS-treated and control plants revealed significant differences in microbiome composition, including 1148 distinct ASVs (923 bacteria and 225 fungi), microbiome interactions, and metabolic function of root-associated microbiomes. Fungal and bacterial communities had different response patterns for CNS treatment, with phased and dose-dependent effects, with the most significant changes in microbial community structure observed at 1200 mg/L after 10 days of treatment. Microbial networks of CNS-treated plants had more nodes and edges, higher connectivity, and more hub microorganisms than those of control plants. Compared with control, CNS significantly elevated abundances of various bacterial biomarkers (such as Sphingomonas and Burkholderia ) and fungi biomarkers (including Penicillium , Myceliophthora , and Talaromyces ), which were potential plant-beneficial organisms. Functional prediction based on metagenomic data demonstrated pathways related to nutrient cycling being greatly enriched under CNS treatment. Furthermore, 391 culturable bacteria and 44 culturable fungi were isolated from soil and root samples. Among them, six bacteria and two fungi strains enriched upon CNS treatment were validated to have plant growth promotion effect, and two fungi ( Cladosporium spp. and Talaromyces spp.) played their roles by mediating volatile organic compounds (VOCs). To some extent, the driving and shaping of the microbiome by CNS contributed to its impact on plant growth and development. Conclusion Our results revealed the key role of root-associated microbiota in mediating the interaction between CNS and plants, thus providing valuable insights and strategies for harnessing CNS to enhance plant growth. Graphical Abstract \n Supplementary Information The online version contains supplementary material available at 10.1186/s12951-024-02971-x.",
"conclusion": "Conclusions This work extended previous findings on the impact of CNS addition on the tobacco root community, as well as the alterations in plant growth and soil properties. Our results revealed that CNS treatment significantly influenced plant growth and soil nutrient availability in a dose-dependent manner. Specifically, to some extent, the relative lower concentration (600 mg/L) of CNS shortened vegetative growth duration and phosphorus availability. A substantial shift in the root-associated microbiome assembly in CNS-treated plants compared to control plants was observed. Moreover, the diversity and structure of fungal and bacterial communities exhibited phase and dose-dependent responses to CNS. Remarkably, the microbial networks in CNS-treated plants exhibited greater complexity, increased connectivity, and the presence of more hub microorganisms than those in control plants. Our analysis revealed that some potential plant-beneficial microorganisms were highly enriched under CNS treatment, including Sphingomonas , Burkholderia , Penicillium , Myceliophthora , and Talaromyces . Furthermore, 391 culturable bacteria and 44 culturable fungi were isolated from soil and root samples. Finally, 6 bacterial and 2 fungi were validated to have microbial-mediated growth effects on tobacco, and these two fungi performed their functions by volatile organic compounds (VOCs). Our investigation broadened the understanding of CNS impacts on plant–microbe interactions and underscored the promising practice of utilizing nanomaterials to modulate microbiome.",
"discussion": "Discussion CNS strongly influences the assembly of root-associated microbiomes Similar with previously studies, incorporating nanomaterials (NMs) into the soil would affect rhizospheric microbes, and enhance plant roots and crop growth, which offered novel strategies for optimizing agricultural practices and addressing the challenges of sustainable food production in the future. Previous research had demonstrated that relative concentrations of CNS could enhance plant growth in controlled environments, such as suspension cultures, hydroponics, and laboratory pot experiments [ 32 ]. For instance, CNS promoted the absorption of potassium ions by tobacco roots in hydroponic [ 53 ]. Despite the promising capacity of CNS, the cascading effects of CNS, via the roots, on plant physiology and plant–microbiota interaction remained elusive. Here, we tried to use high throughput sequencing analysis and culture collection to investigate the effects of CNS on plant growth and root-associated microbiota in field experiments. Overall, our findings revealed distinct differences in the composition, microbiome interactions, and metabolic capacity of root-associated microbiome between CNS-treated plants and control plants. Further, multiple microbial attributes revealed that microbiomes in the endosphere compartment were more sensitive to CNS than soil microbiomes, such as alpha-diversity, microbiome composition, and interkingdom co-occurrence networks. Plants exposed to CNS reshaped root-associated microbiome with high abundance of beneficial bacteria, forming a relative stable network enriched with beneficial hub microorganisms that favored plant growth. Based on cultivated bacteria and fungi, we found that CNS treatment led to the recruitment of specific beneficial microbiota, such as Sphingomonas, Variovorax , and Burkholderia , which in turn enhanced plant growth. In summary, these discoveries provided valuable insights into the recruitment of specific root-associated microorganisms in plants exposed to CNS. In this study, CNS was introduced into the plant through root irrigation. However, it was worth noting that alternative application methods might need further investigation. For instance, the delivery of CNS via spray onto leaf surfaces might represent a promising avenue for future research, which might reduce soil fixation and ensure more efficient absorption by the plant. One recent study has demonstrated that the foliar application of carbon nanosol at concentration of 40–70 μg/mL could also promote the growth of tobacco seedlings [ 54 ]. The dose-dependent and phase-dependent effects of CNS on the microbiome composition Numerous studies had highlighted the various impact of ENMs on plant growth, yield, and quality, pointing out various factors might affect their pivotal roles, such as type, concentration, size, application mode, exposure duration, and plant species [ 55 , 56 ]. Although great progress had been derived in engineering the plant-associated microbial communities using ENMs, many of these studies only focused on short-term investigations or the effects of a single nanomaterial concentration within controlled laboratory conditions, like potted plants in the greenhouse. In contrast, our study aimed to explore various concentrations of CNS and their dynamic interactions with the microbial community under natural field conditions. Our findings revealed a dose-dependent and time-dependent influence of CNS on both plant performance and the associated soil microbiome. 25 days after CNS treatment, microbial communities still exhibited significant differences, suggesting that changes in microbial community composition resulting from CNS might not be transient. The impacts of CNS on microbiota were observed more pronounced at 10 days, and over time, the disturbance caused by CNS gradually recovered (Fig. 2 D). We also observed the remarkable dose-dependent influence of CNS on plant growth and development (Fig. 1 C). Low and high-concentration CNS caused earlier or later flowering times, respectively. Notably, lower concentrations of CNS led to earlier flowering by shortening the vegetative development phase, consistent with previous studies [ 57 , 58 ]. These dose-dependent responses were likely due to intricate mechanisms encompassing direct toxicity, dynamics in microbial community, and resource competition among microorganisms [ 59 ]. High ENM concentrations have been shown to induce cellular stress [ 60 ], disrupt metabolism [ 61 ], and suppressed growth for both plants and microorganisms [ 62 ]. Moreover, elevated ENM levels could alter microbial community composition of different compartment niches, potentially fostering resource competition among microorganisms. For example, the structure and metabolism of the endophytic bacterial community in roots underwent remarkable changes under CNS treatment. These complex interactions, along with stress-induced responses and metabolic changes, might collectively contribute to the dose-dependent effects of ENM on plant-microorganism interactions. In summary, our research indicated the importance of ENM concentration and exposure duration when assessing their impact on plants and associated microbiomes. These insights expanded our understanding of the intricate interplay between ENMs and biological systems. In future practical applications, our results indicated that CNS could be strategically applied to enhance plant growth and improve soil health, potentially reducing the need of fertilizers and pesticides. This offered valuable insights for developing sustainable agricultural practices, where careful management of ENM application could lead to more resilient crops and healthier ecosystems. The alteration in ecological functions of microbiome communities under CNS addition Soil microbiome could play important roles in determining soil environmental properties including mineralization, pH, and nutrient availability [ 63 ]. The surface of ENMs could affect their hydrophobicity, and soil environments, provoking mobilization and nutrient mineralization and through rhizospheric microorganisms [ 64 ]. Moreover, root-associated microbiota could produce abundant bioactive secondary metabolites, including phytohormones and siderophores. We applied metagenome to explore relevant functions for rhizosphere community associated with plant growth from both the soil and root, aiming to establish a link between these functions and the microbiome composition within tobacco roots. In summary, the relative abundances of some specific genes were much higher among the CNS-treated samples compared to the control, which might account for the observed enhancement in plant growth (Fig. 5 C and S7). In our study, the abundance of genes related to the sulfur cycle pathways, nitrogen metabolism, and phosphate solubilization within the tobacco rhizosphere and endosphere were increased under CNS treatment. Similar to our findings, other studies have also shown that microorganisms could promote plant growth by improving the availability of nutrients, thereby enhancing the uptake of plants. Recent rhizospheres functional studies of Arabidopsis and soybean have revealed the roles of iron acquisition and mineral nutrient metabolism for plant growth [ 65 , 66 ]. Insufficient available nitrogen and soluble phosphate could inhibit plant growth [ 67 ]. Another noteworthy enhancement was observed in rhizosphere functions involved in the production of plant growth regulators, including IAA, 2,3-butanediol, and acetoin, all of which could promote plant growth [ 68 ]. In our study, the enhanced plant growth-promoting effect, coupled with the differences among various microbial communities, suggested that the microbial community played a crucial mediating role between soil conditions, CNS application, and plant performance. Additionally, the observed increase in soil phosphorus availability, together with the elevated abundances in genes related to the solubilization of phosphorus, further supported the role of the root-associated microbial community in linking CNS application with soil quality improvement. CNS-enriched microbial strains and their effects on plant growth Our microbiome analysis unveiled a notable enrichment of plant-beneficial microorganisms following CNS treatment, implying an increased colonization of potential plant growth-promoting microorganisms. Furthermore, we isolated the bacteria and fungi enriched under CNS treatment, and confirmed their plant growth-promoting effects. These findings highlighted the potential of CNS applied to plant roots for targeted and precise manipulation of microbiomes. Further research was needed to explore the potential advantages of utilizing complex synthetic microbial communities, comprising a mix of plant beneficial microbes. These communities might offer synergistic benefits that enhanced plant growth, resilience to environmental stressors, and nutrient uptake in ways that might be difficult or impossible to single-strain inoculants [ 69 ]. The study of such interactions between different strains could deepen our understanding of the optimal conditions for using microbial consortia in sustainable agriculture. Except promoting plant growth by the production of phytohormones, these beneficial microbiomes could also accelerate nutrient availability and suppress disease. Commonly, these beneficial microbes could strengthen the nutrient status of plants by P solubilization or N fixation, enhancing surface area of roots, and increasing the available nutrients in the rhizosphere [ 70 ]. Many previous studies have demonstrated that Variovorax sp . was regarded as potential plant growth-promoting bacteria for many plants [ 71 ]. Genomic analysis of the Variovorax sp. B9, a prominent PGPR enriched under CNS treatment, revealed the presence of functional genes associated with IAA production, ACC deaminase activity, as well as phosphorus solubilization, and nitrogen fixation capabilities (Figure S10). It was noteworthy that several studies have reported fungi and bacteria might influence plant growth through VOC production [ 72 ]. These VOCs were often composed of simple hydrocarbons, alcohols, esters, and other small molecules that could directly impact plant physiology. For instance, 2,3-butanediol and acetoin have demonstrated a strong capacity to promote plant growth and trigger induced resistance against fungal pathogens, exhibiting antimicrobial activity [ 73 ]. Current research suggested that VOCs functioned as plant growth boosters by activating various signaling pathways, often in correlation with plant growth hormones [ 74 ]. These VOCs were produced through specific metabolic pathways in microbe, as secondary metabolites during their interactions with plants. In particular, Trichoderma spp . had the capability to produce these volatile secondary metabolites, which played important function in the interaction between Trichoderma and plants [ 75 ]. Additionally, our partitioned plate experiments indicated that two PGPR fungi might exert their growth-promoting effects through the production of VOCs (Fig. 7 D and 7E). However, further investigations using mutant strains of these VOCs were needed to identify specific compounds responsible for the plant–microbe interactions. It would be useful to identify suitable VOCs, which would provide promising commercial products in crop grown."
} | 3,902 |
25370493 | PMC4222105 | pmc | 9,117 | {
"abstract": "ABSTRACT Marine sponges are the most primitive metazoan and host symbiotic microorganisms. They are crucial components of the marine ecological system and play an essential role in pelagic processes. Copper pollution is currently a widespread problem and poses a threat to marine organisms. Here, we examined the effects of copper treatment on the composition of the sponge-associated bacterial community and the genetic features that facilitate the survival of enriched bacteria under copper stress. The 16S rRNA gene sequencing results showed that the sponge Haliclona cymaeformis harbored symbiotic sulfur-oxidizing Ectothiorhodospiraceae and photosynthetic Cyanobacteria as dominant species. However, these autotrophic bacteria decreased substantially after treatment with a high copper concentration, which enriched for a heterotrophic-bacterium-dominated community. Metagenomic comparison revealed a varied profile of functional genes and enriched functions, including bacterial motility and chemotaxis, extracellular polysaccharide and capsule synthesis, virulence-associated genes, and genes involved in cell signaling and regulation, suggesting short-period mechanisms of the enriched bacterial community for surviving copper stress in the microenvironment of the sponge. Microscopic observation and comparison revealed dynamic bacterial aggregation within the matrix and lysis of sponge cells. The bacteriophage community was also enriched, and the complete genome of a dominant phage was determined, implying that a lytic phage cycle was stimulated by the high copper concentration. This study demonstrated a copper-induced shift in the composition of functional genes of the sponge-associated bacterial community, revealing the selective effect of copper treatment on the functions of the bacterial community in the microenvironment of the sponge.",
"introduction": "INTRODUCTION Copper, an essential trace metal for organisms, can be highly toxic to organisms at excessive concentrations ( 1 ). The toxicity of copper is attributed to its harmful effects on cell membranes and nucleic acid structure together with its ability to alter enzyme specificity and disrupt cellular functions ( 2 ). Currently, the frequent use of copper as antifouling paint has led to marine organisms being exposed to high concentrations of copper ( 3 ). Thus, the effect of copper pollution on marine ecology is an important area of study. Marine sponges serve as an important component of the benthic environment, with a considerable biomass and an essential influence on pelagic processes such as the food chain and elemental cycling ( 4 , 5 ). Sponges establish close relationships with microbes, and this process occurs in the body of sponges as a result of symbiosis ( 6 ). Sponge-specific microbial clusters have been defined and constitute a 16S rRNA gene database for search purposes ( 7 ). The biomass of microbes in sponges is very high, constituting up to 60% of the total biomass of the sponge ( 8 ). Until now, more than 35 microbial phyla have been detected in sponges by denaturing gradient gel electrophoresis, 16S rRNA gene sequencing, and clone library sequencing. The majority of these bacteria were found to belong to the phyla Proteobacteria , Cyanobacteria , Firmicutes , Bacteroidetes , Acidobacteria , Actinobacteria , and Chloroflexi ( 9 ). Symbiotic microorganisms interact with the host by providing carbohydrates ( 10 ), vitamins ( 11 ), and antibiotics ( 12 ); oxidizing toxic metabolites such as ammonia ( 13 ) and sulfide ( 14 ); and participating in carbon, sulfur, and nitrogen cycles, all of which result in a relatively independent ecosystem in the sponge body. In return, the microorganisms acquire from the host a living shield and some metabolic substrates. Sessile and filter-feeding characteristics have made sponges efficient predators of food bacteria (a marine sponge of 1 kg may filter up to 24,000 liters of seawater per hour in the field) ( 15 ) but also susceptible to the bioconcentration and bioaccumulation of pollutants ( 16 ). The harmful effects of heavy metal on sponges have been studied, and the results showed that the survival, growth, shape, water motion, and reproduction of sponges are all affected by copper pollution ( 17 ). At the cellular level, cadmium, copper, and mercury may have adverse effects on sponges in terms of the shape, motility, and aggregation of isolated cells ( 18 ). Moreover, high concentrations due to long-term copper and cadmium treatments also can lead to settler mortality and larval settlement failure ( 17 ). In contrast, knowledge regarding the effect of heavy metals on the sponge-associated microbial community is limited. A previous study using copper, cadmium, and mercury to treat bacteria isolated from the sponge Fasciospongia cavernosa defined a group of heavy-metal-resistant bacteria in sponges as indicators of aquatic pollution ( 19 ). Specifically, copper exposure shifted the microbial community of the sponge Rhopaloeides odorabile with respect to morphology and taxonomy ( 20 ). In the present study, we investigated the effect of copper treatment on the sponge-associated bacterial community in terms of community structure and functional gene composition to determine how copper affects the symbiotic bacterial community (and even bacteriophages). We also examined the genetic features that facilitated the survival of the bacterial community that was enriched in response to copper stress.",
"discussion": "DISCUSSION In the present study, we investigated the shift of the sponge-associated bacterial community in response to copper treatment and the mechanisms underlying the functional selection of the treatment of the bacterial community. The results revealed that the dominant species of the bacterial community harbored by the sponge belonged to the symbiotic sulfur-oxidizing Ectothiorhodospiraceae and photosynthetic Cyanobacteria . However, the dominant symbiotic bacteria decreased substantially in response to the high concentration of copper, which enriched for a heterotrophic-bacterium-dominated community. A metagenomic comparison revealed the functional selective effect of the treatment on the enriched bacterial community. Additionally, the bacteriophage community was also enriched, suggesting that a lytic cycle was stimulated by copper. Potential roles of autotrophic bacteria in the sponge. In the present study, the natural sponge was dominated by an unclassified member of the family Ectothiorhodospiraceae and an unclassified cyanobacterium, which are sulfur-oxidizing and photosynthetic bacteria, respectively. Ectothiorhodospiraceae is a family that includes numerous anoxygenic photosynthetic SOB and nonphotosynthetic SOB. Previous studies have shown that symbiotic SOB in the sponge can oxidize sulfide (produced by anaerobic sulfur-reducing bacteria [SRB]) in the sponge body as an energy source and detoxify sulfide, which is toxic to sponges ( 9 , 14 , 24 ). Sponge-specific Ectothiorhodospiraceae also have been shown to be dominant in the sponge Axinella corrugata , accounting for more than 34.5% of the entire microbial community ( 25 ). In the present study, the unclassified member of the family Ectothiorhodospiraceae (accounting for 54.6% of the bacterial community) was phylogenetically sponge specific and was not detected in the seawater samples. Our analyses of the draft genome of the unclassified Ectothiorhodospiraceae revealed a complete sulfur oxidation pathway (reverse dissimilatory sulfate reduction) and Calvin cycle for CO 2 fixation ( 26 ). The symbiotic role of the sponge-specific Ectothiorhodospiraceae could be fixation of CO 2 and detoxification of sulfide generated by anaerobic SRB in the sponge body. The dominance of bacteria in the sponge, sulfide oxidation, and symbiotic features, including the streamlining of virulence-associated genes and enrichment for symbiosis-related genes ( 26 ), suggested an important role for Ectothiorhodospiraceae in the sponge. Symbiotic Cyanobacteria have been found in many sponges that perform photosynthesis ( 27 – 29 ). They are known to play beneficial roles, providing carbohydrates and oxygen for the host via photosynthesis. The roles of the Cyanobacteria herein could be fixation of CO 2 and provision of carbohydrates to the host. Unique functional selective effect of copper treatment on enriched bacteria in the sponge. Copper may damage cells beyond their physiological limits, inducing the release of reactive oxygen species ( 30 , 31 ) and destabilization of the iron-sulfur cluster and thiol group proteins. Previous studies examining individual cultured bacteria have revealed mechanisms responsible for resistance to high copper concentrations and for maintenance of cellular homeostasis. The mechanism of resistance is conserved in most Gram-negative bacteria and even in Gram-positive bacteria ( 30 ). Escherichia coli contains a P-type ATPase that efficiently pumps Cu ( 32 ). The efflux of periplasmic Cu is due to huge multicomponent protein complexes such as CusCBA ( 33 ). In Salmonella , the Cus system is replaced by another periplasmic Cu defense protein termed CueP ( 34 ). Eleven metal-translocating P-type ATPases of class IB have been described in Mycobacterium tuberculosis , and only CtpV was directly associated with Cu homeostasis ( 35 ). These mechanisms are utilized by individual bacteria for resistance to a high concentration of copper ions. However, in the present study, no copper efflux pump genes were significantly enriched in the bacterial community by copper treatment. The enriched bacteria might employ a quick and efficient mechanism for protection against copper in the unique environment of the sponge body. The functional genes that were enriched by copper treatment, including those that play a role in bacterial motility and chemotaxis, bacterial capsule synthesis, virulence, and bacterial signaling and regulation, reflected the functions required by the enriched bacteria for surviving the stress induced by copper treatment in the sponge. Functional genes responsible for bacterial motility and chemotaxis (including the flagellum and chemotactic proteins) together with signaling and regulation (a two-component regulatory system, orphan regulatory proteins, and cAMP signaling) might facilitate the transport of bacteria to areas (such as bacterial aggregates) with a low copper concentration in the matrix of the sponge body. The bacterial capsule and extracellular polysaccharides might act as a barrier, preventing copper from entering bacterial cells. Morphological observation of the bacterial aggregates in the treated sponge (see Fig. S7 in the supplemental material) demonstrated a directional migration of the enriched bacteria to the lysed sponge cells (which may require the functions of bacterial motility, chemotaxis, signal regulation, and polysaccharide synthesis, which were enriched in the treated sponges according to the metagenomic analysis) and suggested the bacterial strategy of forming aggregates to escape the stress of the copper treatment. The enrichment for functions of directional bacterial movement and supplementary cellular components, together with the observation of bacterial aggregation, are novel in sponge studies. Additionally, functional genes responsible for bacterial virulence (type IV pilus, Ton and Tol transport system, invasion and intracellular resistance, and adhesion genes, among others) might facilitate the adherence and invasion of host cells by potentially pathogenic bacteria, as suggested by the remarkably reduced numbers of sponge cells in the treatment (postulated to be caused by the attack of pathogenic bacteria). The enriched bacteria were not detected in the reference seawater samples. Thus, copper treatment might have enriched for rare species in the sponge rather than causing invasion of the sponge by seawater bacteria. The change in sponge health status also may have affected the bacterial community composition and functional gene composition. As shown in Fig. 5 (see also Fig. S7 in the supplemental material), the sponge cell number was reduced by the treatment (time and concentration related), suggesting damage to the health of the sponge (although no obvious necrosis was observed). The dynamics of the reduction in sponge cell numbers (see Fig. S7 in the supplemental material) indicated a loss of control of the sponge cells (potential copper intoxication or virus invasion, followed by consumption by the enriched bacteria) over the bacterial community. The normal interaction between the bacterial community and sponge host may have been disrupted, and exposure to copper may have been the major factor shaping the bacterial community. FIG 5 Microscopic observation of bacteria and sponge cells in control (N4, a to c) and treated (H4, d to f) samples following DAPI staining. Sponge cells (indicated by the arrow in panel b) and bacteria were stained blue. Bacteria in the treated samples aggregated (indicated by the arrow in panel d) and were distributed in the matrix (i). Novel bacterial morphotypes (vibrios) appeared in the bacterial aggregates (i). Panels g and h show bright-field images corresponding to those in panels d and e, respectively. The lack of related functions and the subsequent sharp decrease in Ectothiorhodospiraceae and Cyanobacteria confirmed the functional selection induced by copper treatment. Genomic analysis of the unclassified Ectothiorhodospiraceae by the binning method revealed a substantial lack of genes encoding the flagellum, chemotactic protein, and virulence-associated functions compared with its closest free-living relative, Thioalkalivibrio ( 26 ). Although the genome of the unclassified cyanobacterium was not determined, it is well known that most cyanobacteria lack a flagellum and are unable to move freely. The weakness in core functions selected by copper treatment may contribute to the failure of the symbiotic bacteria to survive following exposure to copper. In conclusion, copper treatment could shift the bacterial community of the sponge, decreasing the number of symbiotic autotrophs and enriching for heterotrophs. The treatment selected for the following functions: bacterial motility, chemotaxis, capsule and polysaccharide synthesis, signal transduction and regulation, and virulence, which would facilitate the survival of bacteria under copper stress. The damage to sponge cells and shift of the symbiotic bacterial community reflected the harmful effects of copper on the symbiont system."
} | 3,658 |
27239785 | null | s2 | 9,118 | {
"abstract": "A laser-based hydrogel degradation technique is developed that allows for local control over hydrogel porosity, fabrication of 3D vascular-derived, biomimetic, hydrogel-embedded microfluidic networks, and generation of two intertwining, yet independent, microfluidic networks in a single construct."
} | 74 |
37660814 | PMC10646785 | pmc | 9,119 | {
"abstract": "Removal of recalcitrant lignin from wastewater remains a critical bottleneck in multiple aspects relating to microbial carbon cycling ranging from incomplete treatment of biosolids during wastewater treatment to limited conversion of biomass feedstock to biofuels. Based on previous studies showing that the white rot fungus Phanerochaete chrysosporium and Fenton chemistry synergistically degrade lignin, we sought to determine optimum levels of Fenton addition and the mechanisms underlying this synergy. We tested the extent of degradation of lignin under different ratios of Fenton reagents and found that relatively low levels of H 2 O 2 and Fe(II) enhanced fungal lignin degradation, achieving 80.4 ± 1.61 % lignin degradation at 1.5 mM H 2 O 2 and 0.3 mM Fe(II). Using a combination of whole-transcriptome sequencing and iron speciation assays, we determined that at these concentrations, Fenton chemistry induced the upregulation of 80 differentially expressed genes in P. ch including several oxidative enzymes. This study underlines the importance of non-canonical, auxiliary lignin-degrading pathways in the synergy between white rot fungi and Fenton chemistry in lignin degradation. We also found that, relative to the abiotic control, P. ch. increases the availability of Fe(II) for the production of hydroxyl radicals in the Fenton reaction by recycling Fe(III) (p < 0.001), decreasing the Fe(II) inputs necessary for lignin degradation via the Fenton reaction.",
"conclusion": "4 Conclusions The addition of Fenton chemistry synergistically enhanced lignin degradation in P. ch. cultures from 58.8 % to 80.2 %. Notably, this study provides novel insight into the underlying mechanisms driving this synergy. Our results suggest that P. ch. stimulates Fenton chemistry by cycling Fe(II)/Fe(III) and upregulating enzymatic pathways that prevent lignin repolymerization, stabilize ligninolytic enzymes, and catabolize lignin-derived aromatics . These findings improve our understanding of the transformation of lignin by P. ch. and highlight the relevance of non-canonical auxiliary enzymes and metabolites for the efficient degradation of lignin in conjunction with Fenton chemistry. Overall, the integration of Fenton chemistry presents an exciting opportunity to advance the technology-readiness white-rot fungal bioprocesses for lignocellulosic biomass pretreatment. The insights gained from our study lay the foundation for further exploration and optimization of engineered fungal systems that efficiently degrade lignin in a sustainable and cost-effective manner.",
"introduction": "1 Introduction The limited biotransformation of recalcitrant organic compounds is a technological bottleneck in water treatment and waste resource recovery. A large portion of the heterogeneous mix of organic compounds that make up wastewater and excess sludge consists of lignocellulose ( Hu et al., 2016 ; Liu and Smith, 2022 ). Lignin hampers carbon recovery from lignocellulosic material by encrusting and preventing the valorization of hemicellulose and cellulose ( Hu et al., 2016 ; Liu and Smith, 2022 ). Recently-developed technologies to delignify biomass or remove aromatic pollutants include thermal pre-treatment and advanced oxidation processes such as Fenton-based treatments. Although these technologies have promising performance, they require costly energy and/or chemical inputs and can release inhibitory byproducts from lignin or additional waste streams ( Del Álamo et al., 2022 ; Teixeira et al., 2014 ). For example, conventional Fenton processes require the input of unstable and expensive homogeneous solutions of ferrous iron and can produce iron sludge due to the precipitation of ferric iron, which is difficult to separate and recover ( Zhang et al., 2019 ; Bello et al., 2019 ). One especially promising but unrealized avenue for the engineered biotransformation of recalcitrant organic compounds is the inclusion of white rot fungi (WRF) or their oxidative enzymes ( Kameshwar and Qin, 2017 ; Janusz et al., 2017 ). These enzymes, including manganese peroxidase, lignin peroxidase, and laccase, have been shown to non-specifically transform a variety of recalcitrant organic compounds including lignin ( Del Álamo et al., 2022 ; Barber et al., 2020 ). Additionally, since WRF can fully mineralize lignin, there is comparatively minimal formation of byproducts inhibitory to downstream processes, such as anaerobic digestion ( Teixeira et al., 2014 ). However, the biological potential of WRF has not been translated to engineered bioprocess reactors due to the operational challenges that fungi present ( Sankaran et al., 2010 ; More et al., 2010 ; Mir-Tutusaus et al., 2018 ). On one hand, fungal populations can be difficult to maintain. They can require nutrient supplementation and are sensitive to shear stress and microbial competition ( Mir-Tutusaus et al., 2018 ; Mir-Tutusaus et al., 2016 ). On the other hand, excessive growth of fungal biomass in bioreactors can be problematic; excess mycelia can wrap around impellers, cause blockages in influent and effluent lines, and increase viscosity, thereby limiting mass transfer ( Moreira et al., 2003 ). Even if fungal biomass is maintained and managed in an engineered process, its ligninolytic activity can be unstable ( Singh and Chen, 2008 ; Moreira et al., 2003 ). As part of its secondary metabolism, ligninolytic activity in WRF is governed by a complex set of nutritional, physiological, and environmental conditions and involves the production of a battery of intracellular and extracellular enzymes and redox-mediating metabolites which are not fully understood in terms of their mechanisms or their interactions ( Singh and Chen, 2008 ; Moreira et al., 2003 ; Mir-Tutusaus et al., 2018 ). To date, such operational challenges and knowledge gaps have prevented the adaptation of fungal bioprocesses in all but a few small-scale cases dealing with industry effluents. In order to realize the promise and potential of white-rot fungal metabolism into engineered bioprocess technologies, as a first step, improved characterization of their metabolic capabilities and biokinetics is needed. White rot fungal enzymes and Fenton chemistry have been shown to synergistically degrade lignin, and previous studies have attributed this synergy to direct lignin oxidation and modification by Fenton chemistry, induction of fungal ligninolytic enzymes LiP, MnP, and laccase, and an increase in H 2 O 2 , which is a required co-substrate for certain lignin-degrading enzymes ( Hou et al., 2020 ; Merino et al., 2020 ). In this study, we further elucidate the mechanisms underpinning the synergy between Fenton chemistry and WRF and propose that these two technologies may be complementary in more dimensions than previously understood ( Fig. 1 ), utilizing the model WRF Phanerochaete chrysosporium. Fig. 1 Conceptual diagram of synergistic relationship between white rot fungi and Fenton chemistry. Green arrows indicate promotion. For example, we propose that Fenton chemistry (top left) promotes ligninolytic enzyme induction and directly contributes to the degradation of lignin. Fig. 1 We hypothesized two important mechanisms for synergistic increases in lignin degradation which had not been previously investigated. First, although previous studies utilized qPCR to target the canonical lignin degrading enzymes in P. ch : LiP and MnP ( Hou et al., 2020 ), we expected a broader transcriptional response to Fenton than just the upregulation of LiP and MnP, given that under ligninolytic conditions P. ch. has been shown to upregulate a wide variety of genes including several other oxidative and auxiliary enzymes ( Wymelenberg et al., 2009 ) Second, we hypothesized that P. ch. could reduce the quantity of Fenton reagents needed for effective compound degradation because certain wood-rotting fungi have been shown to produce various low molecular weight Fe(III)-reducing compounds and iron-solubilizing carboxylic acids, such as phenolate-derivative compounds and oxalic acid ( Gómez-Toribio et al., 2009 ; Arantes et al., 2011 ). Thus, P. ch. could solubilize Fe(III) and reduce it to Fe(II) ( Fig. 1 ). This would be a significant advantage to combining these technologies since the cost of chemical inputs and the generation of iron sludge are major limitations to Fenton processes ( Zhang et al., 2019 ; Bello et al., 2019 ). In this study, we utilized soluble alkali lignin as a model organic compound to represent soluble and insoluble lignin ( Su et al., 2018 ). We measured its degradation across a range of Fenton compound concentrations both with and without fungi in order to confirm the synergistic interaction between the two technologies and identify the concentrations of Fenton reagents that optimize lignin degradation in the combined system. Using whole-transcriptome sequencing and iron speciation assays, we identified potential mechanisms underlying this synergy. Additionally, we tested the effects of Fenton compounds on the biomass formation of P. ch . and investigated the relationship between biomass and lignin degradation. Taken together, our results elucidate mechanisms underlying the enzymatic degradation of lignin by P. ch. in synergy with Fenton chemistry and suggest a framework for how Fenton chemistry might be incorporated into ligninolytic bioprocesses utilizing WRF .",
"discussion": "3 Results and discussion 3.1 Confirming and optimizing synergy between P. ch. and Fenton chemistry We found that P. ch. alone degraded 58.8 ± 3.67 % of lignin after 10 days, and that Fenton chemistry alone degraded 92.3 ± 6.69 %, but only at the highest inputs of Fenton reagents ([H 2 O 2 ] = 20 mM and [Fe(II)] = 1 mM). The dosing concentration of H 2 O 2 was a significant driver of lignin degradation ( Fig. 2 ) both in the P. ch. + Fenton treatments (F(4) = 63, p < 0.001) and in the Fenton alone treatments (F(4) = 29, p < 0.001). Lignin degradation in the combined treatment was highest at [H 2 O 2 ] = 1.5 mM, declining at higher concentrations of H 2 O 2 , and significantly different than any other H 2 O 2 -dosing concentration (Tukey p < 0.001, see supplementary information for full table). Fig. 2 Lignin degradation is enhanced by combining P. ch. and Fenton chemistry, with an H 2 O 2 optimum of 1.5 mM. In the absence of P. ch. , lignin is only degraded at the highest concentrations of Fenton reagents. A) Lignin degradation in the combined treatment B) Lignin degradation with only Fenton chemistry C) Same as A and B, but as box plots showing interquartile ranges and median values. Alkali lignin concentrations were measured after 10-day incubations, and for each treatment N = 5. Fig. 2 At this optimal H 2 O 2 concentration of 1.5 mM, a synergistic effect between fungal activity and Fenton chemistry toward lignin degradation was observed. To illustrate, at [H 2 O 2 ] = 1.5 mM and [Fe(II)] = 0.3 mM, there was negligible lignin degradation (−0.95 ± 9.89 %) by Fenton activity alone, and P. ch. degraded 58.8 ± 3.67 %. Together, at these same concentrations of Fenton reagents, 80.4 ± 1.61 % of the lignin was degraded ( Fig. 3 A). Fig. 3 Fungal biomass declines for H2O2 concentrations at or above 1.5 mM (the degradation optimum). A) Fungal biomass across all five iron treatments for each concentration of H2O2 tested (F(4) = 76.3, ANOVA p < 0.001); N = 5 [Fe(II)] conditions and N = 5 replicates, so each box and whisker plot represents N = 25 samples, showing interquartile ranges and median values. Labels represent significant differences between groups (Tukey's P each <0.001 between significant comparisons). B) Fungal biomass across all five H2O2 treatments for each concentration of Fe(II) tested. N = 5 [H2O2] conditions and N = 5 replicates, so each box and whisker plot represents N = 25 samples, showing interquartile ranges and median values). Iron did not significantly affect fungal biomass (F(4) = 1.56, p = 0.19). Fig. 3 This set of experiments also pointed to the range over which P. ch. produces biomass under different amounts of Fenton reagents, with P. ch. biomass production declining at 1.5 mM of H 2 O 2 (Tukey p < 0.001) and producing no biomass at 20 mM ( Fig. 4 A). Iron dosing concentrations did not significantly affect fungal biomass (F(4) = 1.56, p = 0.19) ( Fig. 4 B). Interestingly, fungal biomass had a negative correlation with lignin degradation (r(121) = −0.42, p < 0.001), implying that over the range of Fenton reagent concentrations tested, there was a tradeoff between production of biomass and degradation of lignin. This is supported by theoretical assumptions and by experimental evidence ( Zheng et al., 2020 ) which posit that because oxidative enzymes are metabolically expensive, fungi trade off growth with enzyme production. The finding that fungal biomass and lignin degradation do not have the same optimum is promising for the inclusion of WRF in bioreactors since excess fungal biomass can cause operational difficulties in a bioreactor setting ( Couto and Toca-Herrera, 2007 ; Espinosa-Ortiz et al., 2015 ). Fig. 4 Fe(III)/Fe(II) cycling activity observed in P. ch. and induction of lignin-degrading genes by Fenton chemistry could explain synergistic lignin degradation. A) Comparison of measured combined lignin degradation activity of P. ch. and Fenton chemistry to the theoretical additive activity of P. ch. and Fenton chemistry alone. B) Fe(II) and hydroxyl radicals produced in the presence of 0.3 mM Fe(III) in 30 min in P. ch. cultures versus uninoculated media. C) Individual genes which are differentially expressed according to a Fisher's exact test. Red lines indicate cutoffs for significant genes discussed in the text (log2-fold changes>2 and p < 0.01). Fig. 4 3.2 Mechanisms underlying synergy between P. ch. and Fenton chemistry 3.2.1 P. ch. increases the availability of Fe(II) The dosing concentration of iron was highly significant in the Fenton alone treatments (F(4) = 80, p < 0.001) and statistically not significant in the combined P. ch. + Fenton treatments (F(4) = 40, p = 0.062). Notably, the calculations did not include the [H 2 O 2 ] = 20 mM for the P. ch. + Fenton treatments, because P. ch. did not persist at this concentration of H 2 O 2 . When [H 2 O 2 ] = 20 mM and P. ch. was no longer present, Fe(II) became a significant driver of lignin degradation in the P. ch. + Fenton treatment (F(4) = 42, p < 0.001). Together, these results suggest that iron availability is not a limiting factor for lignin degradation in the presence of P. ch. We hypothesized that P. ch. activity could be increasing the availability of Fe(II) by solubilizing iron and recycling Fe(III) for the production of hydroxyl radicals in the Fenton reaction. When Fe(III)-reducing activity was measured in 5-day-old cultures of P. ch, the fungal cultures produced 79.5 ± 8.32 μM Fe(II) in the span of 30 min, significantly (t(8) = 21.04, p < 0.001) more than the negligible amounts of Fe(II) (0.56 ± 0.42 μM) that were produced in the uninoculated samples ( Fig. 4 B). The inoculated samples also had significantly (t(8) = 19.87, p < 0.001) higher levels of hydroxyl radical production than the uninoculated samples ( Fig. 4 B). These data suggest that P. ch. aids hydroxyl radical production in the Fenton reaction by increasing Fe(II) availability through reduction and solubilization of iron. In the conventional homogeneous Fenton reaction, the regeneration of Fe(II) from Fe(III) is much slower than its consumption ( Zhang et al., 2019 ). Additionally, Fe(III) precipitates at pH values higher than 3 or in the absence of suitable chelators, leading to iron sludge formation ( Bello et al., 2019 ; Zhang et al., 2019 ). These challenges necessitate the usage of higher Fe(II) concentrations in engineered Fenton processes to achieve desired degradation rates ( Bello et al., 2019 ; Zhang et al., 2019 ). However, our data suggest that Fe(II)/Fe(III) cycling by P. ch. promotes efficient hydroxyl radical production even at low total iron concentrations, which could relieve the requirement for high iron inputs in Fenton processes. 3.2.2 Fenton chemistry induces auxiliary pathways to lignin degradation in P. ch. Another mechanism that could explain synergistic lignin degradation is the induction of fungal ligninolytic enzyme activity in the presence of reactive oxygen species. Hou et al. (2020) used RT-qPCR to measure the expression of the target enzymes lignin peroxidase (LiP) and manganese peroxidase (MnP) in P. ch. and found that expression was increased in the presence of electro-Fenton chemistry, presumably as a response to oxidative stress. Given that lignin degradation and modification is known to involve a suite of accessory enzymes beyond LiP and MnP ( Kameshwar and Qin, 2017 ; Janusz et al., 2017 ), we utilized an untargeted approach to gain a broader understanding of changes in gene expression in P. ch. after sustained exposure to Fenton chemistry. We utilized whole-transcriptome sequencing to identify differentially expressed genes (DEGs) in cultures extracted RNA from 5-day-old P. ch. cultures with and without exposure to Fenton reagents ([H 2 O 2 ] = 1.5 mM and [Fe(II)] = 0.3 mM). The goal was to get a global understanding of the transcriptomic response in P. ch. in response to Fenton chemistry to further elucidate the mechanistic underpinnings of the observed synergy. Due to the close concordance between RNA-seq and qPCR generally ( Coenye, 2021 ) and the stringent criteria used in the study for classifying a gene as differentially expressed, qPCR validation of the RNA-seq results was not conducted. Based on these results, however, specific genes can be targeted with qPCR in future studies. Contrary to our expectations, we did not find DEGs known to be directly involved in lignin-degrading enzyme systems (MnP or LiP). This does not exclude the possibility that at other timepoints, these enzymes may be differentially expressed in the presence of Fenton chemistry. We did, however, find 83 genes which were significantly (log2 fold-change >2, FDR-corrected p < 0.01) over-expressed in the Fenton treatment and 14 genes which were over-expressed in the control treatment ( Fig. 3 C). Additionally, we found 35 enriched molecular function GO terms, 20 enriched KEGG pathways, and 10 enriched KEGG modules (see supplementary materials for tables of enriched GO and KEGG terms). Among the GO terms, those correlating with oxidoreductases, iron ion binding and catalytic activity were the most over-represented. The enrichment of these particular GO categories generally supports our hypothesis that the induction of fungal enzymes is an important mechanism for the Fenton- P. ch. synergy, and also points to the importance of non-canonical oxidative enzymes besides MnP and LiP in lignin degradation. The most differentially expressed individual gene was annotated as an FAD-dependent oxidoreductase (log2 fold-change +7.2 in the combined treatment, p < 0.001). FAD-dependent oxidoreductases have been shown to inhibit lignin re-polymerization by reducing and therefore stabilizing lignin-derived phenoxy radicals produced during oxidative degradation ( Marzullo et al., 1995 ; Samejima and Eriksson, 1992 ; Ai et al., 2014 ). Without stabilization, these soluble phenoxy radicals can both re-polymerize and inactivate fungal LiP, decreasing lignin-degradation efficiency. A protein annotated as caffeoyl-coA O -methyltransferase was part of the core enrichment in the phenylpropanoid biosynthesis pathway (NES = 1.627, FDR-corrected p = 0.005). In P. ch ., this protein converts toxic phenoxy by-products into non-toxic, methylated phenolic groups, thus preventing phenoxy radical repolymerization and protecting cells from toxic byproducts ( Le and Kim, 2016 ). In the combined treatment, we found evidence for upregulation of genes involved in the stabilization of the LiP enzyme: phenylalanine ammonia lyase (log2 fold-change +3.4, p < 0.001) and tryptophan biosynthesis from the chorismate module (NES = 2.04, FDR-corrected p = 0.0018). Phenylalanine ammonia lyase initiates the production of veratryl alcohol (VA) from phenylalanine ( Kameshwar and Qin, 2017 ). VA plays a well-established role in the stabilization of lignin derivatives as well as the LiP enzyme ( Harper et al., 1996 ; Marzullo et al., 1995 ). Tryptophan and its indole derivative have been shown to increase LiP activity in WRF by protecting LiP from inactivation in a similar fashion to VA ( Collins et al., 1997 ). Other pathways and genes associated with catabolism of lignin-derived aromatic compounds were enriched in the Fenton treatment, including styrene degradation (NES = 1.75, FDR-corrected p = 0.015). Although styrene was not present in our study, the enzymes in the core enrichment of this pathway participate in the degradation of low-molecular weight lignin fragments through phenylacetate, a known intermediate in the degradation of lignin ( Zhu et al., 2017 ; Kameshwar and Qin, 2017 ) and a precursor to 4-hydroxy-phenylacetic acid, which is an Fe(III)-reducing compound produced by certain wood decay fungi ( Arantes et al., 2011 ). Notably, four genes annotated as cytochrome P450 oxidoreductases were also significantly (log2 fold-change >2, p < 0.01; see supplementary materials for exact P values) upregulated. These intracellular enzymes are thought to be involved in demethylation and hydroxylation of lignin-derived aromatics ( Wolf et al., 2022 ; Del Cerro et al., 2021 ). Our analysis found differential expression of a suite of accessory enzymes involved in lignin degradation, but did not show upregulation of the canonical lignin-degrading enzymes LiP and MnP. Hou et al. (2020) previously found that LiP and MnP steadily increased over the course of 96 h after beginning electro-Fenton addition. Our experiments differed in that we exposed fungi to chemical Fenton reagents for 5 days via once-daily dosing of H 2 O 2 ,whereas electro-Fenton supplies H 2 O 2 continuously. It is likely that the addition of Fenton chemistry quickly increases the initial depolymerization of lignin (and the expression of LiP and MnP), producing higher quantities of lignin-derived aromatic compounds. Our results reinforce the role of non-canonical lignin-degrading machinery involving the stabilization of LiP, prevention of lignin re-polymerization, and catabolism of low-molecular weight lignin fragments. The differential expression of these accessory enzymes may have other applications beyond lignin degradation. Cytochrome P450s in particular, which were up-regulated under Fenton addition, are well-documented as degraders of several micropollutants in wastewater and display wide substrate versatility ( Mir-Tutusaus et al., 2018 ). In P. ch., cytochrome P450s have been shown to play a role in the degradation of industrial pollutants such as polycyclic aromatic hydrocarbons ( Syed et al., 2010 ), chlorinated dioxins ( Chigu et al., 2010 ), and neonicotinoid insecticides ( Wang et al., 2019 ), to name a few. For fungal processes aiming to remove micropollutants, Fenton chemistry may be useful for inducing the expression of these versatile fungal enzymes. Engineered processes utilizing white-rot fungi hold immense potential for sustainable solutions in lignocellulosic biomass pre-treatment, enabling efficient conversion of biomass into value-added products. With the exciting prospect for increased applications of engineered fungal processes, recent studies have explored the combination of white-rot fungi with Fenton processes to enhance degradation. However, despite promising results, the underlying mechanisms driving this synergy and strategies for optimizing it are still relatively unexplored. In this study, we have addressed these gaps by demonstrating the iron cycling capabilities of P. ch. without the need for externally-added mediators and by investigating the comprehensive molecular response of P. ch. to Fenton chemistry. Through transcriptomic analysis, we have uncovered the upregulation of various oxidative and auxiliary enzymes under Fenton conditions. Notably, our findings differed from a previous study by Hou et al. (2020) , as we did not observe increased expression of the canonical lignin-degrading enzymes MnP and LiP in the presence of Fenton chemistry. Instead, we highlight upregulated auxiliary pathways involved in stabilizing lignin-degrading metabolites and enzymes, preventing lignin re-polymerization, and breaking down low-molecular weight lignin fragments. These novel insights expand our understanding of non-canonical lignin degradation machinery of P. ch. in synergy with Fenton chemistry. By providing fundamental knowledge and insights, our study lays the groundwork for the development of engineered bioprocesses that integrate fungal-chemical transformations."
} | 6,236 |
28679374 | PMC5499019 | pmc | 9,120 | {
"abstract": "Background Lactic acid bacteria (LAB) are receiving more attention to act as cell factories for the production of high-value metabolites. However, the molecular tools for genetic modifying these strains are mainly vector-based double-crossover strategies, which are laborious and inefficient. To address this problem, several counterselectable markers have been developed, while few of them could be used in the wild-type host cells without pretreatment. Results The pheS gene encoding phenylalanyl-tRNA synthetase alpha subunit was identified in Lactococcus lactis NZ9000 genome. When mutant pheS gene ( pheS* ) under the control of the Lc. lactis NZ9000 l -lactate dehydrogenase promoter (P ldh ) was expressed from a plasmid, the resulted PheS* with an A312G substitution rendered cells sensitive to the phenylalanine analog p -chloro-phenylalanine ( p -Cl-Phe). This result suggested pheS* was suitable to be used as a counterselectable marker in Lc. lactis . However, the expression level of pheS* from a chromosomal copy was too low to confer p -Cl-Phe sensitivity. Therefore, a strategy of cascading promoters was attempted for strengthening the expression level of pheS* . Expectedly, a cassette 5Pldh- pheS* with five tandem repetitive promoters P ldh resulted in a sensitivity to 15 mM p -Cl-Phe. Subsequently, a counterselectable seamless mutagenesis system PheS*/pG + host9 based on a temperature-sensitive plasmid pG + host9 harboring a 5Pldh- pheS* cassette was developed in Lc. lactis . We also demonstrated the possibility of applying pheS* to be a counterselectable marker in Lactobacillus casei BL23. Conclusions As reported in E. coli , pheS* as a counterselectable marker has been demonstrated to be functional in targeted gene(s) deletion in Lc. lactis as well as in L. casei . Moreover, the efficiency and timesaving counterselectable seamless mutagenesis system PheS*/pG + host9 could be used in the wild-type host cells without pretreatment.",
"conclusion": "Conclusions A seamless mutagenesis system PheS*/pG + host9 based on a counterselectable marker pheS* and a temperature sensitive plasmid pG + host9 was developed in Lc. lactis . Moreover, this system can be used for rapidly constructing a seamless mutagenesis (deleted or inserted) strain. We also extended the counterselectable marker pheS* to L. casei . Although the feasibility of pheS* as a counterselectable marker used in other LAB remains to be demonstrated, we speculated that this conterselectable marker will accelerate the analysis of genes with unknown function and metabolic engineering research in LAB.",
"discussion": "Discussion In consideration of the increasing use in industrial and medical area, LAB are intensively studied on their genetics and metabolism [ 9 , 10 ]. Therefore, efficient genome engineering tools are necessary for target gene(s) deletion or insertion for functional analysis or rerouting the metabolic flux [ 38 ]. In this study, a seamless negative selectable mutagenesis system PheS*/pG + host9 was developed. We also demonstrated its feasibility by constructing strains bearing the targeting seamless deletion of a 709 bp fragment in lactococcal galactose operon and aldB gene. Expectedly, the ratio of the double-crossover event was 100% after counterselection by p -Cl-Phe. To our knowledge, this is the first report that the mutated pheS allele can be used as a counterselection marker for efficient and rapid genomic engineering in Lc. lactis . Previously, the development of a pheS based counterselection system in Streptococcus mutans , which is a close relative to Lc. lactis , has been reported [ 28 ]. However, S. mutans is a pathogenic bacterium distributed in the dental caries and could not be applied in the food field and used as a cell factory [ 39 ]. We expected that combining the counterselectable marker pheS* with the traditional genetic tool pG + host9 [ 16 ] would overcome the bottleneck of laboriously screening of the double-crossover recombinants, and this system has greatly potential in genome engineering in LAB. Protein sequence analysis suggested that the alanine residue of PheS protein is highly conserved in LAB (Fig. 1 ). Here we have demonstrated the feasibility of pheS* as a counterselectable marker in Lc. lactic and L. casei , these results were consistent with the previously results in S. mutans and Enterococcus faecalis [ 28 , 32 ]. Therefore, we speculated that the dominant-negative mutant gene pheS* might be widely used as a counterselectable marker in a variety of lactic acid bacterial species. However, the sensitivity of the cells to p -Cl-Phe was depended on strain specific manner, such as 15 mM p -Cl-Phe for Lc. lactis NZ9000, 20 mM p -Cl-Phe for S. mutans [ 28 ]. Hence, optimization of the PheS* expression is needed when employing pheS* as a counterselectable marker in other LAB strains [ 25 , 28 ]. In this study, the PheS* protein under the control of a promoter P ldh has the ability of completely inhibiting the growth of Lc. lactis NZ9000 at 15 mM p -Cl-Phe, suggesting it is possible to use P ldh - pheS* cassette as a counterselectable marker in Lc. lactis . However, the growth of the recombinants with P ldh - pheS* inserted into the chromosomal locus was not completely inhibited by even higher concentration of p -Cl-Phe. This unexpected result means that the ratio of screening double-crossover recombinants would not be 100% after p -Cl-Phe counterselection. We speculated that this phenomenon was caused by low expression level of PheS* [ 28 ], because the copy number of P ldh - pheS* from the chromosomal locus was lower than that in the plasmid pleiss-P-pheS*. Lower yield of PheS* was insufficient to compete with the background expression of wild-type PheS to form complexes with phenylalanyl-tRNA synthetase beta subunit (PheT) [ 40 ]. In these cases, a strategy of cascading promoters [ 41 ] was employed to improve the expression level of protein PheS*. Surprisingly, when protein PheS* was driven simultaneously by five copies of the P ldh , the generating integrant Lc. lactis IG5 was substantially inhibited in the presence of 15 mM p -Cl-Phe and the ratio of screening double-crossover recombinants was 100%, suggesting recombination among the promoters was not occurred and the use of repeated P ldh promoters could not confer genetic instability [ 41 ]. This strategy provides a new idea to address the issue of the low expression of the exogenous protein(s) in LAB. Several strategies have been employed to fulfill the genome engineering in LAB by homologous double-crossover using a solely conditional replication plasmid [ 38 ] or combining with other counterselectable system, such as upp [ 22 ] or oroP [ 24 ] based cassettes. Compared with those methods, the negative selectable system PheS*/pG + host9 has several advantages. (1) It greatly simplifies the procedure for screening double-crossover recombinants. For example, taking only 2 days to screen double-crossover variants after the single-crossover integrants were subcultured at 28 °C. The ratio of the double-crossover recombinants was 100% after p -Cl-Phe counterselection. However, the ratio between the deletion and wild-type strains may not be the theoretical value (1:1), it can vary considerably depending on the function of gene(s) to be deleted. (2) To our knowledge, among all the reported counterselectable markers, only pheS* has the greatly potential to be widely utilized in wild-type LAB without pretreatment. In contrast to other counterselectable system, the variants required the counterselectable marker deficient strains, as in the case of upp [ 21 – 23 ] and oroP [ 24 ]. Recently, a new counterselection method for wild-type Lc. lactis genome engineering based on class IIa bacteriocin sensitivity was reported [ 42 ]. However, the li006Dlitation of this method to be widely used in LAB was the sensitivity to bacteriocins which would depend on the interaction between the listerial MpnC and the native PtnD [ 42 ]. (3) Strains without pheS* can naturally grow on GM9 medium with 15 mM p -Cl-Phe. This means 15 mM p -Cl-Phe has no side-effect on the growth of the expected mutants. Moreover, this mutagenesis system PheS*/pG + host9 allowed gene deletion without any genomic scarring [ 15 ] in Lc. lactis , as the case of the aldB gene. The generating genetically modified microorganisms (GMOs) [ 14 ] were seamless mutagenesis which means only leaving self-DNA in its native genome location [ 15 ]. Therefore, this system is useful in seamless gene deletions in industrial strains. However, this seamless mutagenesis system PheS*/pG + host9 remains challenging in large DNA fragment deletions or insertions. In this study, the limited length of the targeted DNA fragment was mostly from the low efficient homologous recombination mediated by RecA [ 17 ]. In consideration of the ratio of the double-crossover recombinants was 100% after p -Cl-Phe counterselection, the ideal goal for deletion or insertion of large DNA fragment is the new genome engineering tools responsible for targeted fragments replacement by selection and the 5P ldh - pheS* cassette responsible for selectable marker excision by counterselection [ 15 ]."
} | 2,325 |
29456857 | PMC5795293 | pmc | 9,121 | {
"abstract": "In this work graphene-based aerogel anodes and graphene/stainless steel cathodes have been optimised as platinum-free electrodes in Rhodopseudomonas palustris microbial fuel cells, achieving a maximum power output of ∼3.5 W m –3 .",
"conclusion": "Conclusions In this study we demonstrated the three-fold advantage of graphene based aerogels towards enhancing the efficiency of MFC electrodes. First, we demonstrated the effectiveness of unmodified, pristine graphene composite aerogels. Second, we showed that a pristine graphene coating can enhance EET compared to standard carbon anodes. Third, we reveal that pristine graphene can catalyse the cathodic ORR as well as anodic EET. We provide a direct demonstration of graphene enabled MFCs powering a commercial digital clock, illustrating the possibility of MFC in power demanding applications such as micro-electronics, beyond the already existing application ranges of sustainable electricity production, treatment of municipal waste water streams and biosensors. 26 , 80 Finally, we demonstrate for the first time an entirely graphene catalysed MFC.",
"introduction": "Introduction Global population expansion and economic development result in increasing demand for energy and clean water, leading to a pressing need for innovative renewable energy sources and more efficient and sustainable waste treatment technologies. Microbial fuel cell (MFC) technology may satisfy both requirements by tapping into the significant chemical energy in wastewater, exploiting the electrogenic nature of various microorganisms that oxidize organic substrates and donate electrons to an external electron acceptor. Although examples of carbon-based anodes and platinum (Pt) cathodes exist, there remains major scope for improving the performance of electrodes for MFCs. Enhanced understanding of the parameters determining electrode performance will help in the development of environmentally-friendly, abundant catalytic cathode materials, and highly electron-accepting anode materials. Here we identify key parameters of a range of MFC electrodes and characterise the performance of a set of novel, environmentally-friendly low-cost graphene-based anodes and cathodes. Power output from a single chamber MFC is primarily limited by the efficiency of extracellular electron transfer (EET) from the cell to the anode, 1 – 4 mass transport of protons to the cathode, 4 – 6 and the catalytic efficiency of the oxygen reduction reaction (ORR) at the air cathode. 7 , 8 Therefore, ideal anode materials for a single chamber MFC should maximize conductivity 1 and surface area 9 to facilitate current generation via direct extracellular electron transfer (DEET) from anodic biofilms 10 and efficiently catalyse H 2 O formation and evaporation via ORR at the air exposed cathode. 11 Cost-effective carbon-based materials such as carbon felt, 2 carbon fibre, 12 carbon paper, 9 and graphite 13 have been used extensively as anode materials due to their chemical stability ( i.e. their resistance to corrosion in an aqueous environment), surface area (∼0.5 m 2 g –1 ) and electrical conductivity. 1 Pt is often incorporated with materials such as carbon paper as an optimal air cathode catalyst for laboratory scale MFCs. However, high costs prohibit scale up using Pt-based catalysts. 14 Given the need to develop low-cost, environmentally-friendly applications in MFCs there is a growing interest in graphene-based electrodes. Carbon nanotubes 15 and graphene are at the forefront of research in electronics, 16 energy 17 and photonics. 18 Graphene has a theoretical surface area of 2630 m 2 g –1 (∼5000 times higher than traditional anode materials), 19 potential for cost-effective mass production, 20 unique electrical conductivity, 21 catalytic activity, 22 and mechanical strength. 23 These properties, combined with the ease of functionalization 24 and biocompatibility, 25 promise to widen the potential range of applications of graphene based MFCs including incorporation in wastewater treatment plants for pathogen reduction, 26 biological oxygen demand biosensors 27 and powering implantable medical devices. 28 Additionally, oxygen and nitrogen-containing functional groups (present in graphene oxide – GO-produced by the modified Hummers method 29 and chemical vapour deposition 22 ) have been reported to impart a catalytic effect to graphene oxide, improving EET efficiency via an electron shuttling processes, and potentially providing an alternative and cheaper cathodic catalyst to commonly used Pt 30 , 31 (cost ∼ £26 g –1 ). MFC anodes using chemically modified GO have been shown to enhance power densities to 2.67 W m –2 (∼18-fold) for stainless steel mesh, 32 to 0.0525 W m –3 (∼3-fold) from carbon cloth, 33 and to 661 W m –3 (∼19-fold) for nickel oxide foams. 3 Recently, an MFC operated with a modified GO-based aerogel anode achieved the highest volumetric power density reported to date, 34 750 W m –3 (normalised to anode volume). Their low density and high surface area, together with high conductivity, establish aerogels based on GO as high performance MFC anodes. 13 , 35 , 36 However, GO suffers from defects induced into graphene's basal plane from chemical oxidation, significantly impairing its mechanical and electrical properties. Chemical or thermal reduction to reduced GO (RGO) 31 , 37 only partially recovers the mechanical and electrical properties of graphene. Pristine graphene with an unaltered basal plane has been grown by chemical vapour deposition (CVD) on a nickel mesh template to create conductive and porous (∼850 m 2 g –1 ) structures. 38 However, the mesh template, usually copper or nickel, often requires intensive procedures including an acid etching step for removal (which can create chemical residuals), and gas precursors ( e.g. methane) for CVD, 39 substantially increasing costs of electrode fabrication. Pristine graphene flake based aerogels created by freeze gelation of solvent/graphene solutions offer a simple alternative, with superior electrical properties to GO/RGO aerogels. 40 Despite being labelled as electrochemically inert 41 and lacking the density of functional groups present on GO/RGO, pristine graphene can catalyse the reduction of oxygen. 42 Molecular oxygen (i) binds ionically to graphene followed by (ii) endothermic formation of two covalent bonds in an intermediate metastable configuration, energetically favourable (iii) separation of the oxygen atoms to form two epoxy groups on the graphene lattice, and (iv) formation of hydroxyl groups and release of H 2 O. 43 Additionally, the incorporation of conductive polymers into the backbone scaffold of the aerogel may help bridge the graphene flakes and help to maintain mechanical integrity. Poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) (PD) is an interesting MFC electrode material thanks to its high conductivity (∼5 × 10 4 S m –1 ), and positively charged backbone that may interact electrostatically with negatively charged cells to facilitate cell–anode interactions and biofilm formation. 44 Pristine graphene can be produced sustainably by liquid phase exfoliation (LPE) or cracking of methane biogas (derived from food waste and other renewable sources) both of which we used to produce low-cost, scalable, and environmentally friendly electrodes for MFCs. In the anodic chamber we use the metabolically diverse, purple non-sulphur α-proteobacterium, Rhodopseudomonas palustris ( R. palustris ) CGA009 ( ref. 45, 46 and 47 ). R. palustris has been shown to express electrically conductive type IV pili, or ‘nanowires’ 48 that facilitate DEET and allow long range charge transfer through an established biofilm of cells attached to a surface. Our work describes the role of surface area, conductivity, and catalytic effect in MFC anodes. Volumetric power density ( P V ) from MFCs using carbon foam anodes was doubled to 265 ± 12.1 mW m –3 by coating with a pristine graphene/PD (Gr–PD) based ink, and the enhanced surface from composite Gr–PD aerogel anodes increased P V 13-fold to 3.51 ± 0.50 W m –3 , closer to our benchmark provided by carbon fibre as an anode material (5.37 ± 1.16 W m –3 ). As a practical application, we show that a circuit of 10 single chamber MFC devices operated with Gr–PD aerogel anodes with a total volume of 1.32 cm 3 generated 4.19 μW of power, sufficient to run a digital clock. In addition, we show that a Gr–PD ink coating is able to impart catalytic activity onto a standard marine grade stainless steel mesh (SS), showing feasibility as a cost-effective Pt-free air cathode. Finally, we demonstrate a fully pristine graphene-enabled MFC by integrating our pristine graphene-based anode aerogel and air cathode in a single chamber MFC, paving the way to a cost-effective, environmentally friendly energy source. Fabrication and characterisation of aerogel anodes We prepared low density, highly porous aerogel anodes using a biocompatible and biodegradable non-conductive polymer, carboxymethylcellulose sodium salt (CMC) as a scaffold material. 49 To establish the effect of conductivity and catalysis on MFC performance we created four aerogels by freeze drying (see Experimental for more details): a control CMC aerogel (A-CMC), a CMC–graphene aerogel (A-CMC–Gr), a CMC–PD aerogel (A-CMC–PD), and a CMC–Gr–PD aerogel (A-CMC–Gr–PD). For A-CMC, a CMC–water precursor solution was prepared, to which graphene flakes (Gr flakes, Cambridge Nanosystems, thickness ∼5 nm and lateral size ∼1 μm, ESI Fig. 1a † ) were added to make A-CMC–Gr. PD (10% v/v) was added to make A-CMC–PD, and both Gr flakes and PD were added to make A-CMC–Gr–PD. The aerogels were then characterised by Raman spectroscopy, electrochemical impedance spectroscopy (EIS), mercury porosimetry and scanning electron microscopy (SEM). \n Fig. 1a plots the Raman spectrum of the Gr powder (green curve), A-CMC–Gr–PD (blue curve), A-CMC–Gr (red curve), and A-CMC–PD (black curve). The Raman spectrum, taken at 514 nm of the A-CMC–PD (black curve) aerogel exhibits several peaks which are typically assigned to PD's carbon stretching vibrations. 50 , 51 The two more prominent peaks, found at ∼1435 cm –1 (PD1) and ∼1508 cm –1 (PD2), are assigned to the asymmetric C α = C β stretching and symmetric C α = C β (–O) stretching vibrations respectively. 50 , 51 The red, blue and green curves present a G peak which corresponds to the E 2g phonon at the Brillouin zone centre in graphene, while the D peak (red, blue and green curves) is due to the breathing modes of carbon sp 2 atoms and requires a defect for its activation. 52 – 54 The 2D peak (red, blue and green curves) is the D peak overtone and is usually composed of a single Lorentzian in single layer graphene. 55 A single Lorentzian fit of the 2D peak indicated that the graphene in our aerogels was comprised of electronically decoupled graphene layers. The analysis of the dispersion of the G peak (Disp(G)) (see Experimental) allows one to distinguish between in-plane defects and edge defects in graphene. The Disp(G) (0.07 ± 0.03 cm –1 nm –1 ) for each of the aerogels with graphene flakes (red, blue and green curves) indicates that the D peak originated from defects in the basal plane of the graphene in addition to defects along the flake edges. 52 , 56 The PD1 and PD2 peaks were also found alongside the G and D peaks in the spectrum of the A-CMC–Gr–PD aerogel indicating the presence of both PD and graphene. Fig. 1 (a) Raman spectra of Gr flakes (green), A-CMC–PD aerogel (black), A-CMC–Gr aerogel (red), and A-CMC–Gr–PD aerogel (blue). (b) Nyquist curves of the electrochemical impedance spectra for each aerogel anode material. (c) Pore size distribution for A-CMC (orange), A-CMC–Gr (red), A-CMC–PD (black) and A-CMC–Gr–PD (blue). (d) SEM showing micrometre sized pores in the A-CMC, (e) A-CMC–Gr, (f) A-CMC–PD and (g) A-CMC–Gr–PD aerogels. The Nyquist plots of the electrochemical impedance spectra (EIS) before the addition of R. palustris cells were used to determine the charge transfer resistance ( R ct ) of each aerogel. Nyquist plots are generated by plotting the imaginary impedance Im( Z ) versus the real impedance Re( Z ) for each aerogel, and show a semicircle at high frequencies, where the system is under kinetic control ( i.e. when the electrochemical reaction is slow and local concentration gradients of electroactive species, such as ions and molecules, are negligible), followed by a straight line at low frequencies, where the system is under a diffusive controlled regime ( i.e. when the electrochemical reaction is limited by the mass transport of the electroactive species that enter or leave the electrode surface). 40 This can be described by an equivalent circuit model (ECM) (see ESI Fig. 2 † ). The series resistance ( R s ) combines the ionic resistance of the electrolyte, the intrinsic substrate resistance and the contact resistance, and it is defined as the value where the semicircle intercepts the real impedance (Re( Z )) axis. 57 The high frequency semicircle can be described by the double layer capacitance ( C d ), and the charge transfer resistance ( R ct ). The diffusive regime is modelled by the Warburg resistance ( Z W ), which describes the frequency dependence of the ion transport to the electrode. In cases where the R ct is sufficiently high and the diffusive regime is not reached, Z W is set to zero 40 (ESI Table 1 † ). Fig. 1b shows the Nyquist plots for each aerogel sample. In the case of A-CMC (orange curve) and A-CMC–PD (black curve) the R ct is ∼101 kΩ and ∼176 kΩ respectively, with the single semicircle indicating that the electrochemical reaction is kinetically controlled, hindering the process of electron transfer and indicating a resistive behaviour of the materials. 36 The lower R ct for A-CMC–Gr (∼46.2 Ω) (red curve) and A-CMC–Gr–PD (∼21.0 Ω) (blue curve) suggests that the addition of Gr flakes is primarily responsible for decrease in R ct , while PD likely helps to bridge between the conductive graphene flakes, improving R ct further. On the other hand, we noticed the absence of a diffusion element in the A-CMC–PD electrode, which indicates a slower ion transfer process. Therefore we can conclude that the addition of graphene flakes can improve electron transfer in the anode aerogels. The pore size distribution of the aerogels was estimated by mercury porosimetry (see Experimental for more details). Fig. 1c shows the differential intrusion as a function of the pore size diameter for all the aerogels. In the case of A-CMC aerogel (orange line) we noticed a predominant peak in pore size at ∼17 μm, while the A-CMC–PD aerogel (red line) showed a broader pore size distribution between ∼10 to 90 μm. In the case of the A-CMC–Gr (red line) and A-CMC–Gr–PD (blue line) aerogels, the pore distribution shifted down to a 0.1–1 μm range, which might be attributed to graphene flakes blocking the pores which are >10 μm and thus creating smaller cavities throughout the aerogel which increases the resulting surface area. The corresponding calculated surface area ( S a ) was 3.9 m 2 g –1 for the A-CMC–PD aerogel and 7.1 m 2 g –1 for A-CMC, while S a increased to 20.2 m 2 g –1 and 8.2 m 2 g –1 for A-CMC–Gr and A-CMC–Gr–PD respectively, with the addition of graphene flakes. We suspect that the surface areas are indeed much higher ( i.e. 50–100 m 2 g –1 ) than those calculated. However, a collapse of the aerogels due to increasingly high mercury pressure (∼400 psi) is known to affect soft foams analysed with mercury porosimetry 58 by altering the statistics of the smallest pores (∼10 nm) 58 , 59 due to their collapse. Thus their surface area contribution is masked. Scanning electron microscopy (SEM) of the aerogels was used to corroborate the results on the pore size distribution of the aerogels. Arrays of pores >1 μm in diameter were visible in the SEM images of the A-CMC ( Fig. 1d ), while in A-CMC–Gr ( Fig. 1e ) graphene flakes were entwined in fine porous structures with <1 μm in diameter. SEM images of A-CMC–PD ( Fig. 1f ) showed a smooth structure, with larger pores ∼20 μm in size comparable to the pore distribution (∼10 to 90 μm) determined by mercury porosimetry. Fig. 1g shows the graphene flakes blocking the majority of macropores (>1 μm in diameter) in A-CMC–Gr–PD, while pores <1 μm in diameter are still observable, thus confirming the role of the large flakes (∼1 μm) of graphene as a bridging material across the porous CMC scaffold. Fabrication and characterization of stainless steel cathodes and carbon foam anodes In order to establish the effect of enhancing anode conductivity and surface area on MFC performance, we compared the aerogel anodes with a conductive graphene coated carbon foam (CF-Gr–PD) (see Experimental). This was prepared using a low surface area (∼3 m 2 g –1 ) carbon foam (CF) coated with a graphene–PD ink (Gr–PD–IPA) formulated by liquid phase exfoliation (LPE) (see Experimental). Isopropyl alcohol (IPA) was used as the solvent for this ink as the low surface tension (∼29 mN m –1 ) helps to transport the graphene flakes by capillary action around the porous structure. Additionally, to investigate the catalytic activity of a graphene–PD coating as a Pt-free air cathode we coated a standard SS mesh with a graphene–PD ink (Gr–PD–W), by vacuum filtration (see Experimental), to make a graphene coated stainless steel (SS–Gr–PD) cathode. Water was chosen as the solvent for this ink as organic solvents such as IPA will dissolve the nitrocellulose membrane used in the vacuum filtration. The optical absorption spectra of the Gr–PD–W and Gr–PD–IPA inks (ESI Fig. 3 † ) were used to estimate the flake concentration 60 , 61 \n c , obtaining c Gr–PD–W ∼ 0.18 mg ml –1 and c Gr–PD–IPA ∼ 0.08 mg ml –1 . Atomic force microscopy (AFM) statistics showed a thickness of ∼6 nm (ESI Fig. 1a † ) and lateral size ∼135 nm (ESI Fig. 1b † ) for the Gr–PD flakes. Rheological measurements determining viscosity ( η ), surface tension ( γ ), and density ( ρ ) for the two inks showed that η Gr–PD–W ∼ 0.89 mPa s, γ Gr–PD–W ∼ 70 mN m –1 , ρ Gr–PD–W ∼ 1.02 g cm –3 ; η Gr–PD–IPA ∼ 2.5 mPa s, γ Gr–PD–IPA ∼ 27 mN m –1 , ρ Gr–PD–IPA ∼ 0.785 g cm –3 , consistent with previous reports. 62 – 64 \n Raman spectroscopy was also used to characterize the quality of the cathode and anodes. Fig. 2a shows the Raman spectra (acquired at 514 nm) of the Gr–PD flakes (green curve) (which show the Raman fingerprint of the Gr–PD–W flakes as discussed in ESI Fig. 4 † ), the PD (pink curve), the SS–Gr–PD (blue curve), the CF-Gr–PD anode (red curve) and the CF anode (black curve). Besides the PD1 and PD2 peaks at 1435 cm –1 and 1508 cm –1 , the SS–Gr–PD cathode (blue curve) and CF-Gr–PD anode (red curve) have the typical D, G and 2D peaks of graphene as described in the previous section which are in line with the spectra of Gr–PD–IPA flakes (green curve). The blue curve showed a combination of both Gr–PD and PD spectra, while the red curve brings additional features to the CF anode (black curve) where the absence of a distinct 2D peak and the G peak position Pos(G) ∼ 1600 cm –1 indicated the more defective nature of the CF. 52 , 65 , 66 For disordered carbons Pos(G) increases linearly as the excitation wavelength decreases from infrared to ultraviolet, therefore Disp(G) increases with disorder. 45 For carbon systems which have a large number of structural defects Disp(G) > 0.1 cm –1 nm –1 . 55 We attribute the D peak intensity predominantly to the edges of our submicrometer flakes, rather than to structural defects within the flake, given a Disp(G) (0.011 ± 0.003 cm –1 nm –1 ) lower than that expected for disordered carbon. 52 , 56 Therefore, there was a lack of large structural disorder within our flakes and scattering only occurred at the edges of the flakes in an otherwise defect-free sample. 49 Fig. 2 (a) Raman spectra of Gr–PD–IPA flakes (green), PD (pink), CF (black), CF-Gr–PD (red), and SS–Gr–PD (blue). (b) Nyquist curves of the electrochemical impedance spectra for the CF and CF-Gr–PD anodes. (c) Pore size distribution for the CF (0.0140 g) (black) and CF-Gr–PD (0.0120 g) (red) anodes. SEM images of the (d) CF, (e) CF-Gr–PD, and (f) SS anode, (g) bright field optical microscopy image of the SS–Gr–PD. Nyquist plots of EIS ( Fig. 2b ) showed that the R ct of CF-Gr–PD anode decreased with respect to that of the CF anode from 41.4 kΩ to 0.930 kΩ. (ECM, ESI Fig. 2 † ), indicating that addition of the Gr–PD–IPA flakes results in a decrease in R ct . The pore size distribution and specific surface area of the CF-Gr–PD and CF anodes were determined using mercury porosimetry. Fig. 2c shows a broad pore size distribution for the CF and CF-Gr–PD anodes between ∼1 and 100 μm. The specific surface area was calculated (see Experimental) and was similar for both CF (3.7 m 2 g –1 ) and CF-Gr–PD (2 m 2 g –1 ) anodes. SEM of the CF and CF-Gr–PD anodes showed average pore sizes of ∼2 μm ( Fig. 2d ) and ∼5 μm ( Fig. 2e ) in diameter respectively, matching with the porosimetry results. Fig. 2f , acquired by SEM, shows the microstructure of the SS mesh, and Fig. 2g shows the SS–Gr–PD cathode, by optical microscopy, confirms the presence of a Gr–PD continuous film in between the SS mesh wires. Bioelectrochemical characterization of aerogel anodes To compare the bioelectrochemical performance of the aerogel and CF based anodes we designed a single chamber MFC ( Fig. 3a ) for repeated and reliable experimental use. The MFC electrode components were assembled in a stack made in descending order of the anodic aerogel ( Fig. 3b–e ), an SS anode connector, a dielectric dialysis membrane layer with pore size sufficient to block bacterial cells, a Nafion® proton exchange membrane (PEM), with the lower side coated with a conductive and catalytic carbon–Pt surface, and an SS cathode connector. The electrode materials stack was clamped between two Teflon® blocks each with 4 ml cylindrical chambers drilled through them (see Experimental for dimensions), and a rubber gasket seal. The upper chamber was inoculated with R. palustris at an optical density (measured at 600 nm) OD 660 = 3.0 without stirring, to encourage cells to form an electroactive biofilm on the anode surface (as shown by the SEM image in Fig. 3f ). Fig. 3g shows a basic schematic of MFC function. Fig. 3 (a) Diagram of the MFC chamber components. Photographs of the aerogel disks: (b) a A-CMC, (c) PD A-CMC–PD, (d) A-CMC–Gr and (e) A-CMC–Gr–PD. (f) SEM micrograph of R. palustris biofilm on the anode. (g) A simplified diagram illustrating the oxidation of substrate metabolites to provide reducing power that is transferred to the anode through a circuit to the cathode via a potentiostat. To balance charges, mass transport of protons also generated from metabolism occurs through a proton exchange membrane to the air exposed catalytic side of the cathode membrane, where they combine with electrons and atmospheric oxygen to form water. We used linear sweep voltammetry (LSV) to calculate polarization and power curves of the aerogel anodes by applying a linear sweep potential from the open circuit voltage (OCV) to 0 V. The OCV ( Fig. 4a ), surface resistance (calculated by the gradient of the I – V polarization curve) R sur ( Fig. 4b ), maximum current density, I D ( Fig. 4c ) (normalized to projected cathode surface area 67 ), and maximum volumetric power output, P V ( Fig. 4d ) (normalised to anode volume 68 and calculated via Ohm's law 69 ) for the A-CMC, A-CMC–PD, A-CMC–Gr, and A-CMC–Gr–PD aerogels were determined and compared with the benchmark anode material, carbon fibre (CFi). After inoculation for 12 hours with R. palustris , the stable OCV of A-CMC, A-CMC–PD, A-CMC–Gr and A-CMC–Gr–PD reached 522 ± 47.7 mV, 476 ± 35.0 mV, 391 ± 38.9 mV, and 456 ± 38.0 mV respectively. The MFC using A-CMC aerogel gave I D and P V of 4.04 ± 0.618 A m –2 and 0.648 ± 0.178 W m –3 respectively ( Fig. 4c and d ). Considering that CMC is an electrical insulating polymer which gives A-CMC a R sur of 146 ± 17.3 Ω m –2 (the highest of the aerogel anodes, Fig. 4b ), A-CMC may favour the formation of a conductive biofilm in contact with the SS anode connector which could result in giving the A-CMC some conductive properties. Using A-CMC–PD as an anode reduced R sur to 63.5 ± 5.09 Ω m –2 , while I D and P V were marginally increased over A-CMC to 7.91 ± 0.923 A m –2 and 1.01 ± 0.178 W m –3 respectively ( Fig. 4c and d ). Using A-CMC–Gr further reduced R sur to 26.3 ± 3.24 Ω m –2 and increased I D and P V to 17.9 ± 3.09 A m –2 and 2.59 ± 0.514 W m –3 , which is a 4-fold increase over the A-CMC and A-CMC–PD aerogels. We noticed that R sur was lowest in A-CMC–Gr–PD at 16.7 ± 2.86 Ω m –2 , and I D and P V increased to 34.61 ± 5.84 A m –2 and 3.51 ± 0.504 W m –3 . These results are consistent with the trend of R ct shown previously by EIS indicating once more that the addition of graphene flakes helps to improve electron transfer between the anode and biofilm, resulting in improved P V and I D in our MFC. Notably, whilst P V increased by 36% using the A-CMC–Gr–PD over the A-CMC–Gr aerogel, inclusion of PD doubled I D ; this is likely to be due to the highly conductive nature of the PD which allows current to flow easily though the aerogel matrix to the SS anode connector. This may provide conductive bridges between the graphene flakes, 70 thereby reducing R sur and improving charge transfer. Moreover, while it was observed that the fragile A-CMC–Gr aerogels partially disintegrated in the aqueous cell culture medium, the presence of PD in A-CMC–Gr–PD improved the structural robustness of the aerogels. A-CMC–Gr–PD comes closer to the CFi, in terms of power output (5.37 ± 1.16 W m –3 ), and has similar surface resistance (15.3 ± 1.21 Ω m –2 ), however the higher OCV of CFi (605 ± 70.3 mV) may have facilitated higher current (39.7 ± 4.86 A m –2 ) due to the unbroken conductive CFi connection, as opposed to discontinuous pristine graphene flakes. Fig. 4 (a) Comparison of open circuit potentials, (b) surface resistances R sur , (c) maximum current densities, and (d) maximum volumetric power outputs from stainless steel mesh anodes mounting A-CMC, A-CMC–PD, A-CMC–Gr, A-CMC–Gr–PD aerogels, and carbon fibre (CFi), respectively. Error bars show the standard error, for the aerogels of n = 10, and for CFi n = 3. Cyclic voltammograms of the stainless steel anode with (e) A-CMC, (f) A-CMC–PD, (g) A-CMC–Gr, and (h) A-CMC–Gr–PD aerogels on top of stainless steel mesh. Scan speed was 1 mV s –1 between –0.9 and +0.9 V. Each CV trace is a single representative sample, arrows show discernible oxidation and reduction peaks. Sem images of the aerogel anodes show R. palustris cells embedded in the (i) A-CMC, (j) A-CMC–PD, (k) A-CMC–Gr, and (l) A-CMC–Gr–PD aerogels. Cyclic voltammetry (CV) can be used to gain a qualitative insight into the redox mechanisms used to transfer electrons between the cell and the anode and thus give information on pristine graphene's catalytic properties ( i.e. efficiency of EET or charge transfer). 71 Fig. 4e–h shows representative cyclic voltammograms of MFC devices operated with each aerogel colonized by R. palustris taken at 1 mV s –1 between –900 mV and 900 mV after 72 hours in the devices. A broad oxidation–reduction peak pair at ∼360 mV and ∼–200 mV from A-CMC are most likely a result of the interaction between cells and the stainless steel anode connector, 72 and confirm the establishment of an electroactive biofilm on the anode. 73 The addition of PD to CMC (A-CMC–PD) did not change the CV profile significantly ( Fig. 4f ), other than the emergence of a small reduction peak at –640 mV. The addition of Gr flakes to A-CMC–Gr ( Fig. 4g ) increased the current range in response to the voltage scan, and revealed two pairs of oxidation–reduction peaks at –230 mV and –420 mV, and –100 mV and –160 mV. These peaks are similar to those reported for R. palustris on carbon paper, 46 and are in accordance with previous studies that suggest graphene has a more significant effect on EET to enhance MFC current generation rather than via interaction with excreted mediators. 33 , 74 The emergence of these peaks indicates a favourable interaction between at least two extracellular redox mechanisms with graphene, with a low degree of separation between oxidation and reduction peaks being characteristic of an easily reversible reaction with enhanced charge transfer. 75 When both graphene and PD were incorporated into the A-CMC–PD–Gr aerogel anode ( Fig. 4h ), the oxidative peak at –230 mV was more prominent, and the neighbouring oxidation peak at –100 mV was no longer visible, suggesting that one redox mechanism with a lower activation energy for charge transfer is being favoured. Our results show similar profiles to other organisms with more extensively characterized metal reducing outer membrane cytochromes (Omc) such as OmcA and the Mtr pathway from Shewanella oneidensis . 75 However R. palustris homologs to OmcA have low genetic and structural similarity 48 and R. palustris is known to have other important mechanisms of both oxidising and reducing its surroundings such as the phototrophic iron oxidation (Pio) pathway. 76 CV data show that graphene enhances DEET from R. palustris in a MFC. However, further work is required to elucidate the molecular basis of the precise redox active mechanisms acting in synergy with graphene. SEM of colonized aerogel anodes SEM imaging of the aerogel anodes after their use in the MFC ( Fig. 4i–l ) showed cells embedded in the aerogel material with an extremely high level of cell to anode contact. SEM of the A-CMC and A-CMC–PD showed networks of anode aerogel material surrounding and in contact with cells ( Fig. 4i and j ). SEM of the A-CMC–Gr and A-CMC–Gr–PD aerogels ( Fig. 4k and l ) also showed an interconnected structure of graphene and polymer matrix. We noticed that unlike some silica based aerogels that maintain their structure after re-hydration, 77 our CMC based aerogels visibly contract upon contact with cell media. Capillary action and contraction of the super-dehydrated material may help incorporate and immobilize cells within the microstructure of the aerogel and maximize anode to cell contact, which could potentially improve EET. Bioelectrochemical characterization of graphene coated carbon foam anodes We assessed the influence of surface area on the MFC anodes by acquiring the polarisation and power curves from linear sweep voltammetry from MFC devices equipped with CF anodes and the SS anode connector alone (used as control), as done for the aerogels. The I D and P V can be identified from the polarization curves ( Fig. 5a ) and power curves ( Fig. 5b ) of the CF (inset red), CF-Gr–PD anodes (inset black) and SS (inset blue), with the A-CMC–Gr–PD curves shown to illustrate the order of magnitude difference in power output most likely due to surface area. OCV for the CF-Gr–PD anode was measured as 669 ± 6.92 mV, compared to 480 ± 30.9 mV with CF and 392 ± 21.4 mV with just SS at the anode. We also obtained similar values of R sur for CF and CF-Gr–PD at 30.8 ± 5.30 Ω m –2 and 23.7 ± 3.48 Ω m –2 respectively, and lower than SS at 86.4 ± 25.5 Ω m –2 . The SS alone at the anode yielded I D and P V of 4.47 ± 1.06 mA m –2 and 43.6 ± 10.1 mW m –3 respectively. I D and P V of CF-Gr–PD were 25.7 ± 1.113 mA m –2 and 265 ± 12.1 mW m –3 , both nearly 2-fold higher than I D and P V of CF at 15.9 ± 2.80 mA m –2 ( p = 0.020) and 138 ± 28.2 mW m –3 ( p = 0.007) respectively (ESI Table 2 † ). Since the CF and CF-Gr–PD have similar surface area, these results indicate that the conductive graphene coating improves the MFC performance resulting in higher P V , I D , and lower R sur . Furthermore the positive effect of the anode surface area on MFC performance can be inferred as P V increased 13-fold from 0.265 ± 0.0121 W m –3 from the low surface area (∼3 m 2 g –1 ) CF-Gr–PD anodes to 3.51 ± 0.504 W m –3 from the A-CMC–Gr–PD aerogel anodes with a surface area of ∼10 to 20 m 2 g –1 . This clearly indicates that increasing anodic surface area is a key factor to improve MFC performances. Fig. 5 (a) Polarization curves and (b) power curves for MFC devices operated with SS (blue), CF (red), CF-Gr–PD (black), and A-CMC–Gr–PD (purple) anodes respectively. Inset to (b) are the power curves of SS, CF, and CF-Gr–PD for clarity. Platinum coated carbon paper (CP-Pt) was used as open air cathode. The volumetric current and power output is expressed based on the geometrical size of the anode chamber. For SS n = 6, for CF and CF-Gr–PD n = 9, and for A-CMC–Gr–PD n = 10. SEM images of (c) SS with R. palustris cells and (d) SS with a graphene coating and cells. SEM images of (e) CF with R. palustris cells and (f) CF with R. palustris cells with graphene, inset: further magnification showing graphene flakes. SEM of biofilms on carbon foam and steel anodes after use in MFC In order to visualize the distribution of the R. palustris biofilms on the SS, CF, and CF-Gr–PD anodes after operations, the anodes were removed from the MFC and prepared for SEM imaging after performing the bioelectrochemical measurements. The SS anode connector ( Fig. 5c and d ) showed a very sparse presence of microbial cells, suggesting that these were more weakly attached to the smooth, convex surface than to the rougher surface of CF ( Fig. 5e ) and CF-Gr–PD anodes ( Fig. 5f ). Graphene flakes were visible on the surface of CF-Gr–PD ( Fig. 5f inset). Whereas mainly single cells were visible attached to the smooth surface the SS wires, more extensive networks of cells with many cell to cell interactions were visible in the CF biofilm ( Fig. 5f ). Conductive type IV pili are thought to be an essential element of electrogenic and conductive bacterial biofilms, 10 , 12 and R. palustris has previously been shown to produce conductive filamentous structures such as pili or ‘nanowires’. 48 Here at least some of the filamentous structures visibly connecting cells to each other and to the surface of the anode were likely to be conductive pili that may play an important role in DEET. Graphene ink modified stainless steel cathode To establish the effect of PD and graphene flakes as a cathode we compared the MFC performance of devices configured with plain SS, SS–Gr–PD, and an industry standard platinum coated carbon paper (CP-Pt) at the cathode. 11 A carbon fibre anode was used to verify the catalytic function ( i.e. in this case to facilitate ORR) of pristine graphene with the SS–Gr–PD electrode. Fig. 6a and b show the polarization and power curves for each of the cathodes (see Experimental). SS–Gr–PD (blue curve) as a cathode yielded I D and P V of 0.172 ± 0.0378 A m –2 and 1.04 ± 0.252 W m –3 respectively, which are 500-fold greater than those achieved with SS (black curve) at the cathode (0.000373 ± 0.000103 A m –2 and 0.00228 ± 0.000552 W m –3 ) ( Fig. 6a and b , insert, p = 0.011) and is only slightly lower than what was achieved when using a CP-Pt cathode (red curve) ( I D ∼ 0.672 ± 0.152 A m –2 and P V ∼ 5.01 ± 0.302 W m –3 ). We attribute the improved performance of the SS–Gr–PD to the catalytic properties of pristine graphene flakes on the SS mesh facilitating the ORR. 7 , 8 Finally, the combination of a A-CMC–Gr–PD aerogel anode with a SS–Gr–PD cathode yielded an I D of 0.0753 ± 0.739 A m –2 ( Fig. 6c ) and P V of 0.390 W m –3 ( Fig. 6d ), demonstrating the feasibility of Pt-free all-graphene catalysed MFCs. Fig. 6 (a) Polarization and (b) power curves using SS (black) (inset), SS–Gr–PD (blue), and CP-Pt (red) air cathodes. Carbon fibre (2 g) was used in the anodic chamber, n = 6 for each sample, and error bars show the standard error. (c) Polarization curve and (d) power curve for MFCs using graphene at both electrodes, with A-CMC–Gr–PD in the anodic chamber and SS–Gr–PD as an air cathode, n = 3 and error bars show the standard error. A digital clock powered by graphene based MFCs We tested the feasibility of graphene based MFCs to power small electrical devices, such as a digital clock. A circuit of 10 MFC chambers using A-CMC–Gr–PD anodes and Nafion® carbon–Pt cathodes was connected, with two series of five MFCs in parallel ( Fig. 7a ). This configuration was chosen to reach sufficient voltage for the digital clock to operate correctly. The MFCs produced 4.19 μW at 1.29 V and were able to power a clock successfully. Chronovoltammetry ( Fig. 7b ) showed a steady potential drop from 1.38 to 1.10 V in the circuit when the clock was connected. This was potentially due to an effective internal anode–cathode short circuit caused by faster oxidation–reduction kinetics at the anode than the cathode. 78 , 79 Polarization ( Fig. 7c ) and power curves ( Fig. 7d ) from data before (red) and after (blue) the clock was connected indicate I D and P V decreasing from 8.3 to 6.5 μA and from 4.19 to 2.76 μW, respectively. This application demonstrates how graphene based MFCs are able to produce sufficient power to run commercial electronic devices such as those that may be found in wearable technology or low powered sensors. 28 Our results demonstrate the viability of graphene based electrodes as efficient, cost-effective, biocompatible, environmentally-friendly and platinum-free anodes and cathodes for MFCs. This represents a disruptive step change in the manufacturing, cost, accessibility and sustainability of MFCs, paving the way, for example, to more accessible energy sources enabling democratisation of energy supply with important impacts in many aspect of our society from medicine to energy and consumer electronics. Fig. 7 (a) Photograph 10 MFCs configuration powering clock, (b) chronopotentiometry showing effect of clock being connected. (c) Polarization and (d) power curves of MFCs with the clock absent (red) and connected (blue)."
} | 9,571 |
26482260 | null | s2 | 9,122 | {
"abstract": "How should evolution and development build the brain to be capable of flexible and generative cognition? I wish to put forth a 'power-of-two'-based wiring logic that provides the basic computational principle in organizing the microarchitecture of cell assemblies that would readily enable knowledge and adaptive behaviors to emerge upon learning."
} | 86 |
27128993 | PMC5148189 | pmc | 9,123 | {
"abstract": "The evolution of complex traits is hypothesized to occur incrementally. Identifying the transitions that lead to extant complex traits may provide a better understanding of the genetic nature of the observed phenotype. A keystone functional group in wastewater treatment processes are polyphosphate accumulating organisms (PAOs), however the evolution of the PAO phenotype has yet to be explicitly investigated and the specific metabolic traits that discriminate non-PAO from PAO are currently unknown. Here we perform the first comprehensive investigation on the evolution of the PAO phenotype using the model uncultured organism Candidatus Accumulibacter phosphatis (Accumulibacter) through ancestral genome reconstruction, identification of horizontal gene transfer, and a kinetic/stoichiometric characterization of Accumulibacter Clade IIA. The analysis of Accumulibacter's last common ancestor identified 135 laterally derived genes, including genes involved in glycogen, polyhydroxyalkanoate, pyruvate and NADH/NADPH metabolisms, as well as inorganic ion transport and regulatory mechanisms. In contrast, pathways such as the TCA cycle and polyphosphate metabolism displayed minimal horizontal gene transfer. We show that the transition from non-PAO to PAO coincided with horizontal gene transfer within Accumulibacter's core metabolism; likely alleviating key kinetic and stoichiometric bottlenecks, such as anaerobically linking glycogen degradation to polyhydroxyalkanoate synthesis. These results demonstrate the utility of investigating the derived genome of a lineage to identify key transitions leading to an extant complex phenotype.",
"conclusion": "Conclusion Here we report the first evolutionary study on the PAO phenotype through ancestral genome reconstructions, identification of HGT and chemical characterization. Through this analysis, we identified important metabolic transformations that occurred in the Accumulibacter LCA, where the transition from non-PAO to PAO is hypothesized to have occurred. Prominent lateral acquisitions include numerous genes involved in glycogen degradation, glycolysis, pyruvate metabolism and PHB pathways, as well as regulatory and sensory mechanisms involved in redox and P metabolism. In contrast, the TCA cycle and polyP metabolism are composed almost entirely of ancestral genes present before the Accumulibacter LCA. The molecular evolution that occurred in these pathways was likely necessary to overcome key stoichiometric and kinetic bottlenecks identified in PAO metabolism; specifically anaerobic carbon flux from glycogen to PHA via PFOR, P and counter cation transporters to maintain polyP synthesis, and anaerobic NADPH production from NADH via PntAB. Convergent evolution often occurs when non-related organisms under similar selective pressures independently evolve similar adaptations. Based on this assumption, the molecular evolution that occurred at the Accumulibacter LCA is likely representative of the general adaptations necessary for the Accumulibacter-type PAO phenotype to emerge. This analysis demonstrates the significance of differentiating the core genome of a lineage into ancestral and derived states when investigating a complex and phylogenetically cohesive phenotype.",
"introduction": "Introduction The ability of some microbes to store large quantities of intracellular polyphosphate (polyP) is an important trait that differentiates them from other closely related taxa. Engineers exploit this trait to increase the efficacy of phosphorus (P) removal from wastewater by designing treatment systems that select for a combination of physiological capabilities that comprise a distinct and complex phenotype ( Seviour et al. , 2003 ). The term ‘polyphosphate accumulating organism' (PAO) is typically used to distinguish these organisms from others that may not display the complete phenotype needed for successful enrichment in wastewater treatment systems. One of the most abundant PAO in wastewater treatment systems is Candidatus Accumulibacter phosphatis (henceforth Accumulibacter) of the family Rhodocyclaceae ( He et al. , 2008 ; Mielczarek et al. , 2013 ). Accumulibacter's ability to sequester P during cyclic ‘anaerobic feast' and ‘aerobic famine' conditions characteristic of enhanced biological P removal (EBPR) wastewater treatment processes ( Figure 1 ) is the result of a complex metabolism that cycles three storage polymers: polyP, glycogen and polyhydroxyalkanoate (PHA) ( Seviour et al. , 2003 ). Despite the broad phylogenetic distribution of these polymers across all domains of life ( Wilkinson, 1963 ; Kornberg et al. , 1999 ; Jendrossek, 2009 ), the hallmark anaerobic/aerobic cycling phenotype displayed by Accumulibacter is uncommon, and the specific metabolic traits discriminating the PAO phenotype from non-PAO phenotype have yet to be explicitly defined. Identifying these traits and the boundaries of the PAO phenotype within the Rhodocyclaceae will provide a better understanding of the evolution of the PAO phenotype and hence a metabolic framework for the further identification and monitoring of additional bacterial groups responsible for key EBPR functions. Ultimately, this knowledge may be harnessed for more strategic design of wastewater treatment systems that implement the PAO phenotype. Several challenges exist in identifying the traits that differentiate PAO from non-PAO. First, the term ‘PAO' is sometimes used indiscriminately, to discuss organisms that store polyP without displaying the full phenotype described above, encompassing many diverse metabolisms that result in polyP synthesis ( Grillo, 1979 ; Rao et al. , 1998 ; García Martín et al. , 2006 ; Kristiansen et al. , 2013 ; Zhang et al. , 2015 ). Second, there are inherent complications in determining which organisms in mixed communities are PAOs because high-throughput techniques to identify organisms that cycle polyP are currently unavailable. Finally, once a putative PAO is identified, the metabolism that allows polyP storage must be determined; an arduous assignment in mixed communities. To date, only two major PAOs have been identified and functionally characterized; organisms belonging to the genera Accumulibacter ( Crocetti et al. , 2000 ; Zilles et al. , 2002 ) and Tetrasphaera ( Maszenan et al. , 2000 ; Kong et al. , 2005 ). Although the hunt to identify, characterize and differentiate the phenotypes of PAOs continues, a recent proliferation of available Accumulibacter genomes ( García Martín et al. , 2006 ; Flowers et al. , 2013 ; Skennerton et al. , 2015 ) has made it possible to investigate the genomic evolution of the Accumulibacter-type PAO phenotype for the first time. Many simultaneous evolutionary processes such as horizontal gene transfer (HGT), point mutations, re-arrangements, recombination, expansions and contractions contribute to genome evolution ( Ochman et al. , 2005 ; Hao and Golding, 2006 ; Touchon et al. , 2009 ; Zaremba-Niedzwiedzka et al. , 2013 ; Nowell et al. , 2014 ). The result of these concomitant evolutionary processes is that the pan-genome of related bacteria may be categorized into lineage-specific (gene families unique to specific genotypes), flexible (gene families with irregular occurrence across numerous genotypes) and core genomes (gene families present in all genotypes) ( Ochman et al. , 2000 ; Hacker and Carniel, 2001 ). The core genome represents the uniting genomic features, while the flexible and lineage-specific genome provides insight into the evolution of population structure and the speciation process of closely related strains ( Ochman et al. , 2000 ; Kettler et al. , 2007 ; Polz et al. , 2013 ; Chan et al. , 2015 ). The categorization of genome content in this way may provide insight on the boundaries of the PAO phenotype within the Rhodocyclaceae and bring us closer to a more high-throughput way to identify other lineages with similar polyP cycling traits. To investigate ancient evolutionary events, such as the emergence of the PAO phenotype in Accumulibacter, it is common to use ancestral state reconstructions ( Schluter et al. , 1997 ; Larsson et al. , 2011 ; Latysheva et al. , 2012 ). Ancestral state reconstructions allow the division of a genome into ancestral and derived traits. Traits present before the last common ancestor (LCA) of a lineage are considered ancestral, whereas those that have transitioned to new states, such as genes acquired through HGT, are considered derived traits. Previous studies on the evolution of bacterial metabolic networks using gene gain and loss analysis have demonstrated the importance of derived traits from HGT in contributing to the expansion of metabolic network capabilities ( Pál et al. , 2005 ). Thus, by inferring the laterally acquired derived traits of a lineage using ancestral genome reconstructions, the molecular evolution resulting in the emergence of novel phenotypes may be studied. Although ancestral reconstructions provide evidence of the changes that occurred in the past, extant phenotypes may be used to deduce the evolutionary pressures which were selected for these changes ( Connell, 1980 ). In this investigation we merged these lines of evidence; using kinetic and stoichiometric values for Accumulibacter Clade IIA, coupled with the reconstructed ancestral states of 26 genomes in the family Rhodocyclaceae (10 Accumulibacter, 4 Dechloromonas , 8 Thauera , 3 Azoarcus , 1 Zooglea ,). Using the resulting inferred ancestral states, the Accumulibacter Clade IIA genome, CAP2UW1 ( García Martín et al. , 2006 ), was parsed into an ancestral, derived, flexible and lineage-specific genome. A phylogenetic analysis on derived genes within KEGG pathways/COG Inorganic Ion Transport and Metabolism was conducted to determine which were laterally derived. Using these discrete inferred categories, an evolutionary model of Accumulibacter was constructed and integrated with measured phenotypic data, providing the first comprehensive analysis of the molecular evolution of the polyP accumulating phenotype in Accumulibacter. Our results reveal that the laterally derived genes of Accumulibacter's LCA contain numerous adaptations important to the PAO phenotype, demonstrating the utility of investigations into the derived genome of an organism for identifying key adaptations that lead to its present phenotype.",
"discussion": "Discussion The transition from non-PAO to PAO, hypothesized to have occurred at the Accumulibacter LCA, was accompanied by significant molecular evolution in key carbon pathways, transporters, energy metabolism and regulatory elements. The changes in these pathways ranged from considerable, such as in glycolysis, to nearly no change at all such as in the TCA cycle ( Figures 6a and b ). Below we provide a detailed discussion of key laterally derived genes in the context of known aspects of PAO metabolism and the measured stoichiometry/kinetics of Accumulibacter Clade IIA identified in this study. In addition, we incorporate previous metatranscriptomic analyses ( Oyserman et al. , 2015 ) to postulate the relative importance of these derived genes in optimizing and linking key pathways in the Accumulibacter-type PAO phenotype. Finally, we discuss the broader implications of how these findings will change the search for additional PAO. Acetate activation The primary route for carbon acquisition in Accumulibacter is through the anaerobic uptake of volatile fatty acids, such as acetate, and the subsequent synthesis of the storage polymer PHA. After anaerobic acetate contact, acetate is transported into the cell via both passive and active transport ( Saunders et al. , 2007 ; Burow et al. , 2008 ) and activated to acetyl-CoA ( Figures 4 and 5 ). The activation of acetyl-CoA occurs either through acetyl-P or acetyl-AMP intermediates. The primary route for the activation of acetate is currently unknown, however higher relative expression of genes involved in acetyl-CoA synthetase suggest that the primary route is via acetyl-AMP ( Oyserman et al. , 2015 ). Although no laterally derived acetate transporters were identified, both routes for acetate activation contain laterally derived genes (CAP2UW1_1515 and CAP2UW1_2035) ( Figure 7 ). Numerous copies of acetyl-CoA synthetase are found in the CAP2UW1 genome, including flexible (CAP2UW1_1069, CAP2UW1_2247, CAP2UW1_3266) and an ancestral gene (CAP2UW1_3755). Of these, the laterally derived gene had the lowest transcription rates while the ancestral copy (CAP2UW1_3755) was one of the most highly expressed genes in the CAP2UW1 genome ( Oyserman et al. , 2015 ). In contrast, no redundant copies for acylphosphatase are annotated in the CAP2UW1 genome aside from the laterally derived gene (CAP2UW1_1515) and this gene is also not highly expressed ( Oyserman et al. , 2015 ). This analysis suggests that despite containing laterally derived genes, the evolution of acetate activation at the Accumulibacter LCA may not have contributed substantially to transitioning from non-PAO to PAO. PHB synthesis Once acetate has been transported into the cell and activated to acetyl-CoA, it enters the PHB synthesis pathway. The synthesis of PHB (7 C-mmol (gVSS-h) −1 ) in Accumulibacter Clade IIA occurs at twice the rate of the degradation (3.4 C-mmol (gVSS-h) −1 ) and is also greater than the acetate uptake rate (4.8 C-mol (gVSS-h) −1 ) ( Figure 5 and Supplementary Figure 5 ). The kinetic disparity between PHA synthesis, degradation and acetate uptake is due to the additional intracellular flux of carbon from anaerobic glycogen degradation via pyruvate, acetyl-CoA and finally to PHB. Together, these kinetic parameters suggest that a strong evolutionary pressure for rapid PHB synthesis exists. Of the three enzymes in the PHA synthesis pathway (PhaA, PhaB and PhaC), only PhaC contains laterally derived genes. Of the four copies of the PhaC gene in the CAP2UW1 genome, three of these are laterally derived (CAP2UW1_0143, CAP2UW1_3191 and CAP2UW1_3185) and two are among the most highly transcribed genes in CAP2UW1 (CAP2UW1_3191 and CAP2UW1_3185). In addition, CAP2UW1_3191 is co-expressed with a predicted PHA modulon controlled by the ancestral core regulatory protein phaR (CAP2UW1_3918) ( Oyserman et al. , 2015 ). Thus, in contrast to the activation of the acetate to acetyl-CoA, the polymerization of 3-hydroxybutyryl-CoA to PHB is likely to occur primarily through laterally derived genes, suggesting that evolution of PHB metabolism in Accumulibacter was significant in transitioning from non-PAO to PAO. It is noteworthy that the laterally derived PhaC genes represent both class I and III PHA synthase (CAP2UW1_3191 and CAP2UW1_3185, respectively) ( Yuan et al. , 2001 ; Rehm, 2003 ) and that these genes were highly expressed and showed dissimilar expression profiles from each other ( Oyserman et al. , 2015 ). The dissimilar expression profiles of related but functionally divergent PhaC suggests these genes contribute differentially to the PAO metabolism of Accumulibacter, however more research is required to make such a conclusion. Regardless, dose effect (for example, numerous copies of PhaC) has been shown to increase PHA synthesis capabilities ( Maehara et al. , 1998 ). Anaerobic reducing equivalents: glycolysis, glycogen degradation and PntAB Anaerobic PHB synthesis requires both ATP and reducing equivalents. One strategy used by Accumulibacter to meet this demand is to use stored glycogen ( Schuler and Jenkins, 1994 ). As noted earlier, a striking number of genes involved in glycogen degradation (starch/sucrose metabolism) and glycolysis are laterally derived genes ( Figures 6 and 7 ). These include glycogen degradation via glucose phosphorylase (CAP2UW1_0255, CAP2UW1_2663), glucose-6-phosphate isomerase (CAP2UW1_2124), fructose-bisphosphate aldolase (CAP2UW1_2669, CAP2UW1_3196), phosphoglycerate kinase (CAP2UW1_0487), phosphopyruvate hydratase (CAP2UW1_2666) and pyruvate kinase (CAP2UW1_1890) ( Figure 7 ). Although glycolysis produces reducing equivalents in the form of NADH, NADPH is generally required for PHB synthesis ( Peoples and Sinskey, 1989 ; Steinbüchel et al. , 1993 ; Madison and Huisman, 1999 ; Kim et al. , 2014 ). A recent investigation demonstrating hydrogen gas production during anaerobic acetate contact in Accumulibacter enriched bioreactors suggests the regeneration of NAD+ may represent a bottleneck in PAO metabolism that is alleviated through hydrogenase activity ( Oyserman et al. , 2015 ). Furthermore, metatranscriptomic evidence from this same study suggests that the demand for the conversion of NADH to NADPH is met by the NADPH/NADH transhydrogenase PntAB (CAP2UW1_4179, CAP2UW1_4180; Oyserman et al. , 2015 ). Although the hydrogenases are ancestral (CAP2UW1_0999, CAP2UW1_2286), interestingly, both complexes of PntAB are laterally derived. Furthermore, these complexes are highly expressed, as well as many of the laterally derived genes involved in glycogen degradation and glycolysis (CAP2UW1_2124, CAP2UW1_2126, CAP2UW1_2127, CAP2UW1_2662, CAP2UW1_2663, CAP2UW1_2666, CAP2UW1_0255, CAP2UW1_3196, CAP2UW1_0487, CAP2UW1_1890, CAP2UW1_4179, CAP2UW1_4180; Oyserman et al. , 2015 ). Together, this evidence suggests that considerable selective pressures to optimize the production of reducing equivalents in the form of NADPH via glycogen degradation, glycolysis and the activity of NADPH/NADH transhydrogenase existed at the LCA of Accumulibacter and is an important adaptation for the storing PHA anaerobically. Pyruvate metabolism Anaerobic glycogen degradation provides both ATP and NADH, but also produces abundant pyruvate that must be converted to PHB via acetyl-CoA. In general, two complexes exist that may convert pyruvate to acetyl-CoA, pyruvate-ferredoxin oxidoreductase (PFOR) and pyruvate dehydrogenase (PDH). These multi-enzyme complexes differ in that PFOR uses ferredoxin and is often coupled with hydrogen production ( Chabrière et al. , 1999 ), while PDH uses NAD+ and is inhibited by high levels of NADH ( Snoep et al. , 1993 ). Both of these complexes in CAP2UW1 are highly expressed and form separate operons (PFOR, CAP2UW1_2510-CAP2UW1_2512; pyruvate dehydrogenase CAP2UW1_1838-CAP2UW1_1840). However, because PFOR is the primary route from pyruvate to acetyl-CoA under NADH rich conditions ( Patel and Roche, 1990 ; Blamey and Adams, 1993 ; Townson et al. , 1996 ), it likely fills this role in Accumulibacter PAO metabolism, contributing to the hydrogen gas production recently reported ( Oyserman et al. , 2015 ). Interestingly, the PFOR operon in Accumulibacter is composed of laterally derived genes ( Figure 7 ). Thus, the kinetic, evolutionary and transcriptional data all suggest that the ability to efficiently shunt pyruvate to PHB via acetyl-CoA anaerobically is an essential adaptation for the Accumulibacter-type PAO phenotype, without which a build-up of pyruvate would likely inhibit glycogen degradation and stall the anaerobic metabolism of Accumulibacter. Phosphorus and counter cation transport PolyP is a source of ATP in anaerobic PAO metabolism ( Comeau et al. , 1986 ). Thus, one of the key metabolic processes in Accumulibacter is the degradation and synthesis of polyP. Transport of P into and out of the cell must accompany the degradation and synthesis of polyP, as well as the transport of counter cations that are used to balance the negative charge of phosphate. Indeed, the stoichiometric analysis in this investigation demonstrates that P transport of Accumulibacter is linked to the counter cations magnesium and potassium at a nearly 1:1 molar equivalent ratio ( Figure 5 and Supplementary Figure 5 ). Despite the obvious linkage between polyP metabolism and the transport of P, Mg and K, the evolutionary histories of these genes differ significantly. The polyP metabolism of Accumulibacter is ancestral, whereas many of the transporters involved in P (Pit CAP2UW1_3785, CAP2UW1_3788; PstS, CAP2UW1_1747 PstB, CAP2UW1_1751–1752 PstC CAP2UW1_1749) and magnesium transport (corA CAP2UW1_3581, CAP2UW1_2797) are laterally derived genes. The kinetic/stoichiometric and evolutionary data presented here suggests that an increased capability to transport P and counter cations such as Mg was an important adaptation at the Accumulibacter LCA, supporting and expanding upon previous hypotheses that inorganic P transporters may be absolutely required for the Accumulibacter-PAO phenotype ( Saunders et al. , 2007 ; Kristiansen et al. , 2013 ; Nobu et al. , 2014 ). Ferrous iron transport Iron is an essential co-factor in many enzymes, and bacteria have evolved many diverse strategies for the transport and acquisition of iron from the environment ( Andrews et al. , 2003 ; Wandersman and Delepelaire, 2004 ). When reducing (that is, anaerobic) environmental conditions prevail, ferrous iron predominates over ferric iron. Under these conditions, ferrous iron transport using the Feo pathway is favored over alternative ferric transporter mechanisms, such as siderophores ( Cartron et al. , 2006 ). The Feo system was laterally acquired at the Accumulibacter LCA suggesting that anaerobic demand for iron-containing enzymes, such as by the highly expressed PFOR and hydrogenases, is an important adaptation for the Accumulibacter-type PAO phenotype. Signaling and regulation It has been demonstrated that Accumulibacter transcriptionally regulates genes correlating with carbon, P and oxygen availability ( Oyserman et al. , 2015 ). In order to accurately respond to such environmental cues, bacteria rely primarily upon two-component systems ( Chang and Stewart, 1998 ). Furthermore, HGT of two-component systems is an important mechanisms for niche adaptation, reflecting the selective pressures of the environment ( Alm et al. , 2006 ). In Accumulibacter, both phosphate limitation (PhoR CAP2UW1_1995, PhoB CAP2UW1_1996, PhoR-PhoB CAP2UW1_1997) and redox signaling (RegB CAP2UW1_0008, RegA CAP2UW1_0009) two-component systems are laterally derived at the LCA. Although it is difficult to surmise what specific genes may be under control of these two-component systems without additional molecular evidence, metatranscriptomic analysis identified many co-expressed genes responding to aerobic (1844) and low P (438) conditions ( Supplementary Spreadsheet 5 ; Oyserman et al. , 2015 ), which may be good candidates for further study in this regard. In addition to the evolution of novel regulatory mechanisms in Accumulibacter, it is also possible for genes to integrate into existing regulatory networks; albeit this process often occurs slowly, with both recent and ancient laterally acquired genes generally showing lower degrees of co-expression than non-laterally transferred counterparts ( Lercher and Pál, 2008 ). Currently, one of the most well-examined aspects of the Accumulibacter regulatory network is a putative PHA regulon likely controlled by the ancestral core regulatory protein (CAP2UW1_3918) ( Oyserman et al. , 2015 ). A key gene proposed to be in this regulon, a type III PhaC, is laterally derived providing evidence that laterally derived core genes integrated into existing ancestral regulatory networks. Thus, evolution of the regulatory networks through novel P and redox signaling, as well as through the integration of novel genes into existing regulatory networks such as the PHA regulon, likely contributed to the evolution of the PAO phenotype in Accumulibacter. Uncertainty in reconstructions and future work The analysis on Accumulibacter evolution was conducted within the constraints of our current knowledge into the phenotypic and genotypic diversity within the Rhodocyclaceae . We included all closely related, publically available, completed genomes (aside from the Accumulibacter genomes) at the time of the start of this analysis. Our understanding of the evolutionary and genomic capabilities of many lineages is continuously being re-written as the available data on a lineage increases. For example, recent investigations have expanded upon the definition of the Cyanobacteria phylum is based on new genomic information ( Soo et al. , 2014 ). One of the key uncertainties in our analysis is a lack of closely related non-Accumulibacter Rhodocyclaceae genomes that have been reconstructed from EBPR systems (for example, from Dechloromonas spp.). In addition, it remains difficult to distinguish ancient HGT events, especially if they are obfuscated by multiple gains and losses. Future discoveries may expand the diversity of Rhodocyclaceae involved in EBPR, either blurring or clarifying the delineation between PAO and non-PAO."
} | 6,186 |
37729206 | PMC10523604 | pmc | 9,124 | {
"abstract": "Significance Different neural circuits in the brain are involved in the development of different cognitive functions, and the adaptive synergy of the circuits promotes human perception, learning, and decision-making. The current structure design of the spiking neural network still draws on the architecture of deep learning and does not fully reflect the characteristics of spiking neural networks. In this paper, biologically plausible neural circuit structures are evolved by introducing excitatory inhibitory neurons and feedforward feedback connections in the brain through the local unsupervised learning rules. By combining the global error signal, the evolved spiking neural network greatly enhances capability in both perception and reinforcement learning.",
"discussion": "3. Discussion Biological nervous systems utilize multiple circuits and synaptic plasticity mechanisms to accomplish diverse cognitive functions. The prevailing SNN structures lean heavily on a single connection and a single neuron type, reducing their capacity to process complex information. This study integrates feedforward and feedback connections with excitatory and inhibitory neurons to create diverse, biologically plausible neural circuits. We take inspiration from brain region circuit evolution, using an unsupervised local optimization strategy based on STDP for automatic network structure construction. This method enhances circuits with consistent pre- and postsynaptic activities while others gradually weaken and disappear. Coupled with a global error signal, our method allows adaptive network structure and synaptic weight adjustment. We validate our adaptive network structure on classification and reinforcement learning tasks. Experiments demonstrate the superior performance of SNNs constructed by NeuEvo. The ablation analysis of NeuEvo’s search space and circuit evolution strategy proves that introducing neuron types and feedback circuits enhances our search space. The unsupervised STDP structural search strategy based on local neural activities plays a crucial role in adaptive circuit evolution. Diverse biological circuits improve SNNs’ information processing efficiency and complex task performance. In the future, we aim to investigate the effects of different neuron types and synaptic plasticity rules on neural circuits. We hope to achieve more biologically plausible neural networks through inspiration from neural system evolution."
} | 608 |
40130040 | PMC11931252 | pmc | 9,127 | {
"abstract": "Nature offers a boundless source of inspiration for designing bio-inspired technologies and advanced materials. Cephalopods, including octopuses, squids, and cuttlefish, exhibit remarkable biological adaptations, such as dynamic camouflage for predator evasion and communication, as well as robust prey-capturing tools, including beaks and sucker-ring teeth that operate under extreme mechanical stresses in aqueous environments. Central to these remarkable traits are structural proteins that serve as versatile polymeric materials. From a materials science perspective, proteins present unique opportunities due to their genetically encoded sequences, enabling access to a diversity of sequences and precise control over polymer composition and properties. This intrinsic programmability allows scalable, environmentally sustainable production through recombinant biotechnology, in contrast to petroleum-derived polymers. This review highlights recent advances in understanding cephalopod-specific proteins, emphasizing their potential for creating next-generation bioengineered materials and driving sustainable innovation in biomaterials science.",
"conclusion": "5 Conclusion and future perspectives Every day we dive deeper into the intricate mechanisms of nature. These discoveries drive innovation, inspiring more efficient, sustainable, and transformative solutions. Cephalopods are part of this phenomenon as their unique skin, beak, and ring teeth compositions have sparked scientific breakthroughs worldwide, leading to novel materials and technologies. At the root of this potential are the unique properties that reflectins, histidine-binding proteins, and suckerins possess, which can be explored to develop novel biomaterials. Since their discovery, reflectins have been studied intensively to understand their in vivo mechanism and how they allow light manipulation in cephalopods cells. Scientists have also tried to use these proteins to mimic the light manipulation effect namely through thin films. These materials showed interesting properties, such as proton conductivity, stimuli responsiveness, and a high refractive index. These allow applications ranging from ionic conductors to responsive coatings and cellular optical engineering. Histidine-binding proteins have been studied due to their ability to partake in liquid-liquid phase separation and to form coacervates. These coacervates can be used for high-efficiency encapsulation and subsequent controlled release of compounds. HBPs and engineered peptides derived from HBPs, which can form coacervates, have been used as very efficient delivery systems that protect cargo from degradation and redox-driven disassembly. This technology rivals existing commercially available vehicles, and using these peptides allows for the introduction of several point modifications that can be made to meet specific requirements. This makes HBP coacervates a very versatile and promising new approach for preventing and treating diverse diseases ranging from diabetes to cancer. Finally, suckerins have been studied due to the elevated degree of mechanical robustness and flexibility they provide to the sucker ring teeth of squids and cuttlefish. They self-assemble into robust supramolecular structures that form a semicrystalline polymer reinforced by nano-confined β-sheets, resulting in a network with mechanical properties rivalling those of the strongest engineered polymers. Suckerins have demonstrated several interesting properties, such as glass-to-rubber transition, thermoplasticity, thermal conductivity, and modulated toughness. These lead to potential applications in various areas, from biomedical applications to conducting materials, sensing devices, and bioplastics. Thanks to the efforts of several research groups over the past decades, the field of protein-based materials has matured to a point in which the design of function can be finely tuned by the encoded genetic sequence, with potential to replace petroleum-based materials. Still, there are several challenges to be addressed in the coming years. One aspect refers to the production by recombinant expression in host cells, which requires carbon sources that may compete with food resources. Thus, low-cost carbon sources as CO 2 from atmosphere or industrial emissions are promising avenues to increase scalability and reduce environmental impact of the manufacturing stage. Other challenges are associated with the implementation of scalable downstream processes that help reduce the final cost of the product with the required purity, as well as water consumption and waste generated. Finally, the processing of proteins into a variety of materials should avoid the use of toxic solvents that perpetuate eternal chemicals in the environmental loop. This can be achieved by fine tuning protein sequence to increase solubility in environmentally friendly solvents. Thus, carbon neutrality deserves special attention for the implementation of protein-based materials, as one must take into account the full loop from production to biodegradation after use.",
"introduction": "1 Introduction From a materials science perspective, proteins – in particular structural proteins - can be seen as extremely interesting biopolymers composed by well-defined amino acid sequences controlled at the genetic level by a bottom-up approach. Thus, the final polymer composition and functional properties are genetically encoded, which represents an enormous advantage in terms of materials design. Host cells translate this genetic code in vivo into a polymeric protein sequence using a carbon source, enabling the production of desired biopolymers through recombinant, scalable biotechnological methods, which offer significant environmental benefits. In addition, protein biopolymers can be processed under mild conditions (usually taking advantage of their intrinsic self-assembling mechanisms) into a variety of structural and functional biodegradable materials. Thus, several biotech companies are currently producing recombinant structural proteins with applications in textiles, medicine and cosmetic applications. A recent in-depth review from Miserez and co-authors covers aspects related to the molecular design and artificial production of protein-based biological materials [ 1 ]. Observing Nature is a source of inspiration to design and engineer advanced functional materials and develop bio-inspired technologies. Cephalopods, including octopuses, squids, and cuttlefish, are fascinating animals with peculiar properties. One of the most astonishing features is their camouflage ability, by modulating their colour and texture as a mechanism to hide from predators and to communicate [ 2 ]. Another interesting observation is that cephalopods are ferocious predators containing extremely strong beaks and sucker ring teeth with unprecedented robustness and flexibility, working under elevated shearing and compressive forces in aqueous environments [ 3 ]. It is fascinating that three cephalopod-specific families of proteins serve as the foundation for these materials, enabling a remarkable range of light-manipulating properties and prey-capturing abilities. Here, we focus our attention on these three protein families: reflectins involved in light manipulation and camouflage found typically in the skin, histidine-binding proteins found in beaks, and suckerins which are the sole components of sucker-ring teeth ( Fig. 1 ). Reflectins are intrinsically disordered proteins (IDPs) that self-assemble into aggregated nanostructures that further organize into platelets and beads. In vitro , reflectin-based materials have shown light modulating properties and proton conductivity, with applications ranging from electronics to tissue engineering. In the beak, histidine-binding proteins (HBPs) form coacervates that infiltrate the chitin-fiber network and provide hardness to the whole structure. HBPs and derived peptides, with their ability to form coacervates, have been explored as efficient delivery systems, rivalling with existing commercially available vehicles. In the sucker ring teeth, suckerin proteins self-assemble into robust supramolecular structures that form a semicrystalline polymer, reinforced by nano-confined β-sheets, resulting in a network with mechanical properties comparable with those of the strongest engineered polymers. Suckerin-based materials present interesting properties, such as glass-to-rubber transition, antibacterial activity, conductive properties, and modulated toughness, giving rise to a wide variety of applications. Fig. 1 Structural and functional insights of cephalopod proteins, their biotechnological production and processing into biomaterials. (A) Cephalopod proteins and their structural features. Reflectins, found in leucophores and iridophores, are important for light reflection and structural coloration. Histidine-rich beak proteins, present in squid beaks, feature Ala-rich, His-rich, and Gly/His-rich motifs that contribute to mechanical strength and rigidity. Squid ring teeth (SRT) proteins, located in the suction cup, contain His-rich (M1) and Gly-rich (M2) motifs, providing toughness and elasticity. (B) Biotechnological production workflow for cephalopod proteins. Genes coding for wild-type or engineered protein sequences are cloned into DNA expression vectors, transformed into bacterial hosts, and expressed in nutrient-rich media. The resulting proteins are purified and processed into functional materials. Created in BioRender. Roque, C. (2025) https://BioRender.com/r11f005 . Fig. 1 This review explores the biological roles of reflectins, HBPs, and suckerins, emphasizing both wild-type and engineered proteins, as well as derived peptides, along with their structure-function relationships. We will discuss how these sequences are processed into advanced functional materials, their applications, and the future potential of engineering cephalopod proteins for materials science."
} | 2,495 |
36130255 | PMC9594778 | pmc | 9,128 | {
"abstract": "Engineered microbes\ncan be used for producing value-added chemicals\nfrom renewable feedstocks, relieving the dependency on nonrenewable\nresources such as petroleum. These microbes often are composed of\nsynthetic metabolic pathways; however, one major problem in establishing\na synthetic pathway is the challenge of precisely controlling competing\nmetabolic routes, some of which could be crucial for fitness and survival.\nWhile traditional gene deletion and/or coarse overexpression approaches\ndo not provide precise regulation, cis -repressors\n(CRs) are RNA-based regulatory elements that can control the production\nlevels of a particular protein in a tunable manner. Here, we describe\na protocol for a generally applicable fluorescence-activated cell\nsorting technique used to isolate eight subpopulations of CRs from\na semidegenerate library in Escherichia coli , followed by deep sequencing that permitted the identification of\n15 individual CRs with a broad range of protein production profiles.\nUsing these new CRs, we demonstrated a change in production levels\nof a fluorescent reporter by over two orders of magnitude and further\nshowed that these CRs are easily ported from E. coli to Pseudomonas putida . We next used\nfour CRs to tune the production of the enzyme PpsA, involved in pyruvate\nto phosphoenolpyruvate (PEP) conversion, to alter the pool of PEP\nthat feeds into the shikimate pathway. In an engineered P. putida strain, where carbon flux in the shikimate\npathway is diverted to the synthesis of the commodity chemical cis , cis -muconate, we found that tuning\nPpsA translation levels increased the overall titer of muconate. Therefore,\nCRs provide an approach to precisely tune protein levels in metabolic\npathways and will be an important tool for other metabolic engineering\nefforts.",
"introduction": "Introduction One of the ultimate\ngoals of synthetic biology is to re-engineer\norganisms to instill or reinforce desirable attributes for biotechnological\napplications such as the production of high-value chemicals. 1 − 3 For example, a great deal of effort has been performed in the field\nof functional metabolic engineering to enhance the production of biofuels\nin microbial hosts. 4 − 7 Balancing the energetic trade-off between microbial growth and bioproduct\nyield is a key challenge in the metabolic engineering field when co-opting\nor designing new pathways. 8 − 11 Although metabolic flux can be controlled through\ngene knockouts 12 and the coarse over- or\nunder production of key enzymes, 13 precisely\ntuning protein production levels can better modulate the host’s\nphysiological needs and provide optimal parameters for bioproduction. Various technologies have been developed as powerful tool sets\nfor manipulating gene expression. While fine-tuning transcriptional\nregulators (i.e., promoters) has cemented their importance in the\nsynthetic toolbox, other metabolic engineering tactics popular for\nits ability to directly tune protein synthesis levels within a pathway\nare ribosomal binding site (RBS) variation and RBS accessibility. 3 , 14 − 25 For example, directly tuning each enzyme in the pathway of interest\nby cis -mediated riboregulation (that act by occluding\nthe RBS) has proven to reduce cellular burden and optimize titer,\nyield, and biomass in various metabolic engineering strategies (reviewed\nby Kent et al. 3 ). In this way, a bioengineer\ncan manipulate protein production within a metabolic engineering pathway\nby directly tuning down specific enzymes at the protein level, rather\nthan at a mRNA transcript level, thus allowing for an almost digitally\nprecise fine-tuning approach. 3 , 16 , 17 , 23 , 24 , 26 Additionally, because of the cis -acting function, there is no risk of off-target effects and no need\nto balance regulatory stoichiometry. Another advantage of RNA-based\ntechnologies such as ‘synthetic\nriboregulators,’ is that they may provide a more versatile\nregulation to a cell compared to those that rely solely on promoter\ncontrol. 17 , 19 , 22 For example,\nuneven translation efficiency within polycistronic transcripts have\nbeen correlated with not only proximity to the promoter 27 but also the structural accessibility of the\nRBS of each cistron. 28 If the RBS accessibility\ncan be hijacked within an operon, specific targets within an operon\ncan be more precisely tuned. However, existing RBS occlusion tools\nmay suffer from an unpredictable dynamic range and lack of design\nrules for achieving optimal performance. 29 Finally, the same RBS sequence in different genetic backgrounds\nhas been shown to lead to large differences in protein production\nlevels. 14 , 30 Thus, there is a need to develop a more\n‘portable’ design across not only genetic contexts \nbut also coding sequence contexts. Here, we describe the development\nof tight protein production control that is evident in two different\npromoter/reporter sequence contexts, as well as different genetic\nbackgrounds. For this study, we utilized our previously published cis -repressors (CRs), which were designed to effectively\nblock the translation\nof mRNA, by inhibiting the access of RBS during initiation of translation,\nand were further optimized to provide minimal leakage of protein production. 31 A key feature is that, unlike with previously\nreported riboregulators, this design achieved a large dynamic range\nof protein activity. 31 Moreover, we found\nthat the regulatory loop sequence was an important design variable\nand fundamental to varying levels of repression. 31 , 32 Therefore, to further improve upon the utility of this design, we\nused this naturally occurring stem-loop CR sequence from our previous\nwork 31 as the parent sequence for creating\na semidegenerate library of CRs containing select, randomly mutated\nbases in the repressive stem-loop structure. We created 128 CR sequences\nwith the potential to control a diverse range of protein production\nlevels. We hypothesized that coupling this library with a highly sensitive\nfluorescence-activated cell sorting (FACS)-based cell isolation strategy\nwould permit the isolation of CRs with distinct phenotypes. To this\nend, we developed an efficient approach to isolate these CRs, including\nlibrary generation, subpopulation isolation based on a fluorescent\nreporter, subpopulation sequencing to identify enriched sequences,\nand selection of individual, new CR sequences. Then, using the new\nCRs at both plasmid and genomic levels, we validated CR activity by\nobserving the control of protein production levels using superfolder\ngreen fluorescence protein ( sfGFP )-based fluorescence 33 and chloramphenicol tolerance-based assays. To demonstrate the potential portable nature of the selected CRs,\nwe tested the function of the CRs across multiple contexts: plasmid-based\nversus genome-based, two bacterial species Escherichia\ncoli and Pseudomonas putida ), two reporter genes ( sfGFP and chloramphenicol\nacetyltransferase ( CAT) ), and one native gene ( ppsA ), as well as under the control of two distinct gene\nexpression systems (Ptac and T7A1, respectively). Finally, as a proof-of-concept\napplication, a specific enzyme in the synthetic metabolic pathway\nleading to muconate production in P. putida ( 5 , 12 ) was regulated. By precisely tuning the protein production\nlevels of the PpsA enzyme to increase PEP, a more balanced metabolic\nflux was achieved, resulting in increased muconate titer. Overall,\nour ‘plug-and-play’ CRs provide an easy means to precisely\ncontrol protein production levels, with the advantages of tuning to\nextremely low levels, potentially allowing for tuning across locations\nwithin an operon and cross-context portability.",
"discussion": "Results and Discussion Overview\nof CR Library Construction and Validation The field of synthetic\nbiology is lacking universal genetic tools\nthat can precisely regulate protein production levels to manipulate\nflux through biosynthetic pathways. Thus, our goal was to create an\nRNA-based design to control cellular networks that are mostly dominated\nby protein-based components. To accomplish this, we first established\na CR DNA library. Second, we used FACS to isolate subpopulations of\nvaried phenotypes, based on the fluorescence of sfGFP. Third, we analyzed\nthe CR sequences of the subpopulations and finally, we selected and\ncharacterized a subset of individual CR sequences. An overview of\nthe workflow is presented in Figure 1 . Figure 1 Workflow of CR library construction and validation. The\nCR design\nincludes a sfGFP reporter gene downstream from a\nsecondary structure that conceals the RBS and start codon. Using a\nsemidegenerate library targeting the hairpin loop structure, CR variants\nwere created, screened, and sorted by sequential FACS based on sfGFP\nfluorescence intensity. Plasmid DNA was extracted from eight sorted\nsubpopulations with different fluorescence intensity levels and deep-sequenced\nusing Miseq Illumina. Data analysis of enriched base pair positions\nrevealed prominent CR sequences that were validated in both E. coli and P. putida . Image created with BioRender. Our previously described CR design, CR-4, was used as a starting\nhairpin structure. 31 CR-4 contains a 34\nbp stem loop that effectively occludes the RBS, thus preventing translation\nof downstream genes. To create a diversity of protein production profiles\nbased on CR-4, a library of CR sequences with the potential for mismatched\nbase-pairs in the stem loop at seven specific positions were chosen\nbased on Krishnamurthy et al. 31 Thus, 128\npermutations were designed and inserted into the pCK vector 34 to control the translation level of the reporter sfGFP gene ( Figure 2 A). These newly constructed pCKCRlib (pCK CR library) vectors\nwere transformed into E. coli for further\ntesting as a single, bulk population. The population of CR variants\nwas observed to have a broad range of sfGFP fluorescence intensities,\ndemonstrating that the library design successfully created a range\nof CR knockdown activities, from full repression to unrepressed, when\ncompared to the negative and positive controls ( Figure 2 B). Figure 2 Cell sorting and deep-sequencing analysis of\nsorted pooled populations.\n(A) Design schematic of CR elements with variable semidegenerate positions\n(SDPs, green text), constrained sequences (black text), and RBS region\n(blue text). (B) Flow cytometry sfGFP fluorescence histograms for\nthe CR library compared to positive (dark gray, sfGFP vector with\nno CR sequence) and negative (light gray, DH5α cells without\nvector) controls. (C) Analytical flow cytometric measurements for\neach sorted population after a 16 h incubation from an inoculation\nat OD 600 0.05. (D) Analytic flow cytometric measurements\nfor each sorted population after a 24 h cultivation directly inoculated\nfrom a glycerol stock. (E) Sequence enrichment analysis of each sorted\npool after deep sequencing. After overnight cultivation, eight distinct fluorescent phenotypes\nfrom the library of CR variants were isolated by FACS using two sequential\nsorting steps. First, the library population was sorted ‘three-ways’,\nresulting in Populations NEG, 1, and 2 ( Figure 2 B and Supplementary Figure S1, Table S1 ). The NEG population represented ‘negative’\n(dark) cells in the CR population, showing fluorescence intensity\nlevels even lower than the negative control ( Figure 2 B). Potential reasons for this phenotype\ninclude: the cells (1) contain a CR variant(s) responsible for complete\nprotein production inhibition, (2) did not grow well, or (3) have\nno plasmid content thus providing no genotypic data. Because we could\nnot unravel these possibilities, we discarded the NEG population (∼10%\nof the population) and moved forward only with Populations 1 and 2\n( Figure 2 B). Then,\nPopulation 1 and Population 2 were each sorted ‘four-ways’\nand labeled A through D, thereby creating eight isolated subpopulations\nof CR variants: 1A, 1B, 1C, 1D, 2A, 2B, 2C, and 2D ( Figure 2 B, Supplementary Figure S1 ). More detailed flow cytometry graphs\nand statistics tables can be viewed in the Supplementary File . The successful isolation of distinct CRs within\na defined subpopulation\nof cells by in vitro fluorescence analysis of the combinatorial library\n(i.e., 128 permutations in this study) depends on the stability of\nthe subpopulation phenotypes. To test the stability of our eight subpopulations,\n1A–1D and 2A–2D, each subpopulation was cultured overnight\ndirectly after FACS. Then, the next day, the cultures were analyzed\nby flow cytometry ( Figure 2 C, Table S2 ), and glycerol stocks\nwere created from these populations. The flow cytometry results showed\nincreasing fluorescence intensity of the eight subpopulations across\nnearly three logs, with 1A being the most repressive and 2D being\nthe least repressive ( Figure 2 C), consistent with the expected fluorescence intensity values\nbased on the gates used for sorting. Furthermore, the populations\ncontinued to demonstrate a >100-fold range of fluorescence profiles\nafter a 24 h cultivation from a glycerol stock inoculum ( Figure 2 D, Supplementary Figure S4, Table S2 ), indicating that the CR\nvariants in each subpopulation consistently regulated sfGFP production\nto varying levels. To identify the CR variant sequences responsible\nfor each subpopulation\nphenotype, the plasmid vectors were purified from each defined pool\nand were PCR-barcoded (Supplementary Table S3 ) for deep sequencing by Illumina MiSeq. Data analyses were performed\nto identify the enriched variant sequence regions ( Figure 2 E). Reads for each pool (∼38,000/pool)\nwere analyzed for sequences whose population was enriched within each\npool. Two distinct CR sequence examples, those showing enrichment\nin a given subpopulation but were represented to a lesser or no degree\nin the other subpopulations, were selected from each subpopulation\nbased on these analyses (totaling 16 sequences). Notably, the sequences\nfor 1C1 and 2A2 were identical; thus, only 1C1 was used in subsequent\nexperiments. Validation of CRs with Discrete Phenotypic\nProfiles in E. coli Utilizing\nthe 15 selected CR elements,\nplasmid-based constructs of the individual CRs directly upstream of sfGFP and CAT genes were tested in E. coli . Analytical flow cytometry measurements quantified\nthe level of CR-sfGFP production ( Figure 3 A). The fluorescence intensity of the defined\nCR sequences based on the deep-sequencing analysis ( Figure 3 A) corresponded to the fluorescence\nintensity results obtained from the eight sorted subpopulations shown\nin Figure 3 A, demonstrating\nthat the ranking order of protein production levels from the individual\nCR sequences is similar to the order of sfGFP activity levels in the\nsubpopulations. Figure 3 Fluorescence intensity and growth profiles of cis -repressed E. coli transformants.\n(A) Analytical flow cytometry results for the 15 unique CR sequences\nin E. coli transformants, each containing\na different cis -repressed sfGFP vector under Ptac\npromoter control. Control strains were E. coli cells lacking transformed CR plasmid (NEG) or those transformed\nwith a vector containing the sfGFP gene with no CR\nregulation (NoCis). (B) Normalized cell growth of E.\ncoli transformants containing 10 different CRs in\nfront of the CAT gene under T7A1 promoter control,\ncultivated in different concentrations of chloramphenicol (Cm). Control\nstrains were E. coli cells lacking transformed\nCR plasmid. (C) E. coli transformants\ncontaining cis -repressed CAT after 6 h cultivation\nin Lysogeny Broth (LB) media supplemented with 30 μg/mL Cm.\nThe error bars represent standard deviations from the mean of biological\ntriplicates. To further confirm that the selected\nCR sequences result in a spectrum\nof protein production levels, a CAT assay in E. coli was performed. The expectation was that the\nrepression activity of the different CRs would differentially confer\nsusceptibility of the cells to chloramphenicol (Cm). Ten CRs were\nanalyzed in the CAT /Cm tolerance assay. As expected,\nthe ranking of chloramphenicol resistance by CR was similar to the\nsfGFP Fl results ( Figure 3 A, Supplementary Figure S5 ). For\nexample, the CRs with higher repression activity, 1A1 and 1B1, demonstrated\nno growth at 30 μg/mL Cm after 6 h cultivation, suggesting the\nweakest protein production levels of CAT compared\nto the other transformants ( Figure 3 C). By contrast, a high Cm tolerance was demonstrated\nby the construct 2D1, with normalized cell growth up to 0.7 OD 600 after 6 h cultivation. Based on the cell growth profile,\nthe 2D1 construct demonstrated a 20% decrease in cell growth (OD 600 ) compared to the control strain that has no CR (NoCis)\ndriving CAT protein production levels ( Figure 3 B). Together, these results\ndemonstrate a large dynamic range of protein production regulation\nfor two distinctly different reporter genes ( sfGFP and CAT ) and promoter systems (Ptac and T7A1, respectively),\nwith clear reproducibility and near digital control of protein production\nfrom a very low level to an “all-on” high level. Establishing\nCRs in P. putida To test the\ncross-strain portability of the new CRs, a subset of\nfive CRs, 1A1, 1C2, 2B1, 2C1, and 2D1 were tested in P. putida . The CR- sfGFP expression\nconstructs were cloned into the pBTL-2 vector with the constitutive\nPtac promoter driving transcription. Plasmids were transformed into P. putida KT2440 resulting in strains NP230, NP231,\nNP232, NP233, NP234, and NP235 for CRs 1A1, 1C2, 2B1, 2C1, 2D1, and\nNoCis, respectively. Fluorescence from each strain was compared to\nthe positive and negative controls, “NoCis” construct\nwithout CR regulation (NP235) and an empty vector construct without\nthe sfGFP gene (NP240), respectively. As expected,\nthe unrepressed sfGFP construct (NoCis) demonstrated the highest fluorescence\nintensity among all strains tested, and the fluorescence intensity\ndecreased with increasing strength of the CRs, as expected ( Figure 4 A). By contrast,\nthe fluorescence intensity of the strongest CR-1A1 showed no difference\nto the negative control (pBLT-2) which lacked sfGFP gene ( Figure 4 A).\nNotably, the ranking of the fluorescence intensity of CRs was similar\nin P. putida ( Figure 4 A) and E. coli ( Figure 3 A), demonstrating\nconsistent performance across both bacterial species and highlighting\na future potential use of CRs as an agnostic bacterial tool. Further\napplications of CRs in diverse hosts will be needed to test the extent\nof how broadly agnostic CRs are with respect to host background, genome\nposition, upstream promoter, and effected gene, especially within\na native operon context. Figure 4 Profiles of repression were consistent between\nplasmid-based and\ngenomic reporters in P. putida KT2440.\n(A) Fluorescence intensity profiles of transformants with pBTL-2-based\nplasmids with various CRs controlling the protein synthesis levels\nof sfGFP. (B) Fluorescence intensity profiles of the transformants\nwith integrated CRs controlling the protein synthesis levels of sfGFP\nat the PP_2684-85 locus. The error bars represent standard deviations\nfrom the mean of biological triplicates. The CR- sfGFP constructs were next evaluated within\nthe P. putida genome to confirm that\ntheir performance is consistent when integrated in the genome versus\nplasmid-based. The same subset of CR-sfGFP constructs (1A1, 1C2, 2B1,\n2C1, 2D1, and NoCis) driven by the Ptac promoter were integrated at\nthe PP_2684-PP_2685 locus of KT2440, resulting in strains NP241, NP242,\nNP243, NP244, NP245, and NP246, respectively. The level of fluorescence\nintensity observed in the genomic-expressed constructs ( Figure 4 B) was markedly lower compared\nto the plasmid-based expression ( Figure 4 A), most likely because of the decreased\ncopy number. 37 However, the fluorescence\nintensity ranking between the constructs for the genomic-expressed\nCR- sfGFP was similar to the plasmid-based system\n( Figure 4 ). For example,\nthe highest fluorescence intensity was observed in the strain without\nany CR present (NoCis), with fluorescence intensity decreasing with\nincreasing strength of the CRs ( Figure 4 ). Again, the strain with the strongest CR element,\nNP241 (CR-1A1), had no significant difference in fluorescence intensity\ncompared to the wildtype KT2440 ( Figure 4 ). Thus, consistent performance from both\nplasmid and integrated CR elements was observed. Using CRs to\nTune a Metabolic Pathway in the Muconate-Producing P. putida CJ442 The CRs were then applied\nto the muconate production pathway, to determine if they can affect\nthe production of a commodity chemical, muconic acid (MA or muconate).\nWe chose Pseudomonas putida strain\nCJ442 as a parent strain because of its successful muconate production 12 , 35 and our previous experience with this system. 36 − 38 P. putida has been identified as a good host for\nthe metabolic conversion of glucose and lignocellulosic biomass (e.g.,\nagricultural byproducts) into the commodity chemical muconate, via\nthe introduction of new and co-opted metabolic pathways 5 , 6 , 12 , 35 , 39 ( Figure 5 ). Johnson et al. 12 demonstrated\nthat the deletion of the P. putida genes pgi-1/2 , pykA / pykF , and ppc in the strain CJ442 diverted carbon flux toward the\nshikimate pathway; carbon is then shunted toward protochatechuate\nusing heterologous enzymes AsbF (a 3-DHS dehydratase) and AroY/EcdB\n(a protocatechuate decarboxylase), which produces catechol and subsequently\nmuconate by the catA gene. At the expense of cell\ngrowth, pyruvate and adenosine triphosphate (ATP) are utilized by\nthe phosphoenolpyruvate synthase enzyme (PpsA) to synthesize phosphoenolpyruvate\n(PEP), an important precursor in muconate biosynthesis. Therefore,\nthis reaction is an important metabolic link between the tricarboxylic\nacid (TCA) cycle and muconate production in P. putida . 40 Ultimately, the phosphoenolpyruvate\nsynthase ( ppsA ) gene was targeted for cis -repression, in order to modulate its protein expression levels,\nbecause the PYR + ATP → PEP reaction is an important metabolic\nlink between the TCA cycle and muconate production in P. putida ( 40 ) ( Figure 5 ). However, initial\nattempts to ‘tune’ the expression of the gene ppsA to balance growth and muconate production failed, due\nto the transcriptional regulation of a PEP synthase regulatory protein\ngene (PP_2081) that is located upstream of ppsA . 40 , 41 Thus, PpsA could theoretically benefit from being precisely controlled\nto retain sufficient pyruvate levels for growth while still shunting\ncarbon toward PEP, analogous to a similar effort in the E. coli shikimate production pathway. 42 Figure 5 Metabolic pathway for muconate production in engineered P. putida CJ442. Abridged metabolic pathway adapted\nfrom the comprehensive metabolic map in Bentley et al. 35 Deleted genes are shown in red; heterologous\ngenes are shown in brown; native genes that are overexpressed are\nshown in green. The riboregulated protein phosphoenolpyruvate synthase\nenzyme (PpsA), which converts PYR into PEP, is shown in blue, and\nits flux is demonstrated with a blue arrow (gradient represents tuning\nof overexpression using CRs in this study). Multiple-step reactions\nare indicated by multiple arrows. See more details in the Materials and Methods section. Image created with BioRender.com . To generate P. putida strains with\nCR-regulated PpsA, an integration strategy was adopted which simultaneously\ndeleted the promoter region of, and integrated individual CRs in front\nof, the psrp gene. Our original intent was to use\nthe same CRs that were tested in Figure 4 . However, attempts at integrating the less-repressive\nCR-2C2 and CR-2D1 elements in front of the PSRP gene were unsuccessful.\nThis result is potentially because our genomic CR system is driven\nby the highly expressing Ptac promoter, rather than psrp ’s native promoter, leading to overexpression of the gene.\nOverexpression of PpsA can cause PEP accumulation and depletion of\nthe pyruvate pool, which has been suggested to lead to poor growth. 43 − 46 Therefore, we chose four CRs with higher repression activity, 1A1,\n1C1, 1D1, and 2B2, still with the goal of testing whether tuning PpsA\nproduction levels can alter growth and muconate production in P. putida . After evaluating the growth, muconate\ntiters, glucose consumption,\nand 2-ketogluconate (2-KG) accumulation of the cis -repressed PpsA strains NP185, NP186, NP187, and NP188 ( Figure 6 ), it was evident\nthat tuning phosphoenolpyruvate synthase production level does affect\nboth growth and muconate production ( Figure 6 ). NP185 (CR-1A1, most repressive) has a\nphenotype similar to the non-CR-regulated strain CJ442, indicating\nthat the remaining, less-repressive CR-regulated strains overexpressed\nPpsA relative to the native system in CJ442. Consistent with this,\nNP186 and NP187, with the 1A1 and 1C1 CRs respectively, demonstrate\nphenotypes similar to each other but different from CJ442. In both\ncases, the cultures grew more slowly, showed slower glucose uptake\nrates, and showed a delayed 2-KG accumulation profile. Muconate accumulation\nrates were also slower; however, the final muconate titers were improved\nfor both of these CR-regulated strains. NP188 (CR-2B2) showed significantly\ndelayed growth, as well as a markedly slower muconate accumulation\nrate relative to the other strains, again possibly because of the\naccumulation of PEP at the expense of pyruvate. Taken together, these\nresults indicate that increasing PEP formation at the expense of growth\nhas the potential to increase final muconate titers. However, a threshold\npotentially exists, whereby significant overexpression of the ppsA gene may prove to critically deplete the pyruvate pool, 45 , 47 , 48 as we observed with NP188 ( Figure 6 ). Importantly, the\nability of CRs to fine-tune the protein synthesis levels of sensitive\nenzymes is indicative of its advantage over the traditional overexpression\nmethod, which may be deleterious toward the product yield and the\ncell’s viability. This novel titration-based method demonstrates\nthe utility of CRs in balancing the metabolic flux without creating\na harmful burden on the cells. Figure 6 Effect of tuning PpsA overexpression in P. putida CJ442\nusing CRs. (A) Growth curves, (B)\nmuconate titers, (C) glucose consumption rates, and (D) build-up of\nthe intermediate 2-ketogluconate (2-KG) of P. putida transformants, namely, CJ442 (black circles, NoCis), NP185 (red\nsquares, CR-1A1), NP186 (yellow triangles, CR-1C1), NP187 (green reverse\ntriangle, CR-1D1), and NP188 (blue diamonds, CR-2B2) measured by a\nshake flask experiment. The error bars represent standard deviations\nfrom the mean of biological triplicates. In the muconate production pathway in P. putida , not all of the ‘ON’ target pathway enzymes need to\nbe expressed at a maximum level and, more often than not, these high-level\nexpressions can cause unnecessary burdens for the host metabolism,\nsuch as an accumulation in intermediates such as 2-KG. 35 In general, when exogenous genes are introduced\ninto cellular pathways/network, a ‘trade-off’ in metabolic\nburden for the cells is observed, that is, limited growth and overall\nproductivity. 12 , 35 Likewise, ‘OFF’\ntarget pathway genes cannot be repressed at too low of levels without\ndetrimental consequences. 12 Furthermore,\nin a balanced cellular network, each enzyme production level is dictated\nby its stability, activity, maturity, and turnover for performing\nits function. 49 Therefore, tuning protein\nsynthesis levels to achieve optimum enzyme levels may be beneficial\nfor cellular metabolism where resources can be redirected for optimal\nproductivity and healthy cellular growth. 19 However, a challenge of fine-tuning the protein production levels\nincludes the potential for extensive testing of protein production\nlevels for each enzyme in the pathway. In addition, initial checks\nare generally not performed to determine if overexpression constructs\nare causing a metabolic burden for cells once the goal of product\nformation is achieved. Importantly, the CR elements described\nhere are agnostic to the\ngene being controlled, only requiring the ATG start codon from the\ncoding sequence to function properly. In contrast to traditional riboregulators\nthat required an additional activation step, 7 , 12 , 35 − 37 the CRs created in this\nstudy do not require activation. The CR activity also does not seem\nto depend upon the upstream or downstream sequence (i.e., from promoter\nto gene), allowing for a seamless incorporation into any design without\nneed for further analysis to ensure proper expression. 54 Other advantages to this design include the\npotential portability feature and reduced labor-intensive screening\ntime usually required for inducers and/or external stimuli. 40 Therefore, this ‘plug-and-play’\nsystem can be incorporated into the design of the CRs whereby multiple\nCRs can be tested to achieve the best protein production level profile. In summary, the riboregulators described here have been applied\nto synthetic biology and metabolic engineering. Additionally, the\nutility of CRs can be expanded beyond a stand-alone expression system,\nwithout any need for an inducer due to the built-in set threshold\nfor protein synthesis allowing for ease of portability. Furthermore,\nthese RNA-based engineering efforts were successfully applied for\na biotechnological application in altering metabolic flux toward the\nproduction of a value-added chemical, cis , cis -muconate, by dialing in the protein levels of PpsA.\nTaken together, this work provides a toolbox of components and validated\nworkflows for implementing CRs where precise expression of specific\nenzymes is required in P. putida . Although\na small library of 128 variants was used in this study, the use of\nFACS for cell sorting opens the possibility of screening much larger\n(10 4 –10 6 ) libraries in future work. More\nbroadly, these efforts establish a framework for further development\nof regulatory tools for tuning translation levels to very low levels\nin industrially promising bacteria."
} | 7,544 |
39437160 | PMC11544928 | pmc | 9,129 | {
"abstract": "The biological membrane is not just a platform for information\nprocessing but also a field of mechanics. The lipid bilayer that constitutes\nthe membrane is an elastic body, generating stress upon deformation,\nwhile the membrane protein embedded therein deforms the bilayer through\nstructural changes. Among membrane-protein interplays, various channel\nspecies act as tension-current converters for signal transduction,\nserving as elementary processes in mechanobiology. However, in situ\nstudies in chaotically complex cell membranes are challenging, and\ncharacterizing the tension dependency of mechanosensitive channels\nremains semiquantitative owing to technical limitations. Here, we\ndeveloped a programmable membrane tension-control apparatus on a lipid\nbilayer system. This synthetic membrane system [contact bubble bilayer\n(CBB)] uses pressure to drive bilayer tension changes via the Young–Laplace\nprinciple, whereas absolute bilayer tension is monitored in real-time\nthrough image analysis of the bubble geometry via the Young principle.\nConsequently, the mechanical nature of the system permits the implementation\nof closed-loop feedback control of bilayer tension (tension-clamp\nCBB), maintaining a constant tension for minutes and allowing stepwise\ntension changes within a hundred milliseconds in the tension range\nof 0.8 to 15 mN·m –1 . We verified the system\nperformance by examining the single-channel behavior of tension-dependent\nKcsA and TREK-1 potassium channels under scheduled tension time courses\nprescribed via visual interfaces. The result revealed steady-state\nactivity and dynamic responses to the step tension changes, which\nare essential to the biophysical characterization of the channels.\nThe apparatus explores a frontier for quantitative mechanobiology\nstudies and promotes the development of a tension-operating experimental\nrobot.",
"conclusion": "Conclusions While the CBB is an open system regarding\nchemical equilibrium,\nit is a straightforward mechanical system driven by bubble pressure\nmanipulations. In contrast, the bilayer tension is governed by the\nforce balance between monolayers and leaflets (Young principle) and\nis not directly driven by the bubble pressure. Separate manipulation\nof each bubble drives monolayer tension directly, leading to leaflet\ntension control via feedback control, which is integrated into controlled\nbilayer tension and constituting the tension-clamp system. The tension-clamp\nCBB has an impact on mechanobiology studies, as does the impact of\nthe voltage-clamp introduced into electrophysiology. The tension-clamp\nCBB extends the frontier for quantitative mechanobiology by characterizing\nthe biophysical properties of membranes and mechanosensitive channels.",
"discussion": "Discussion We established a tension-clamp CBB system\nthat promises to promote\nquantitative mechanobiology experiments. The tension-clamp CBB can\nmaintain steady membrane tension for several tens of minutes over\na tension range of 0.8 to 15 mN·m –1 and execute\nstepwise changes within a hundred milliseconds. Feedback control of\nsuch delicate objects had never been considered before. The stable,\ndynamic, quantitative, and reproducible performance of the tension-clamp\nCBB under single-channel current recordings surpasses any other method\nin mechanobiology studies. Its underlying principles are straightforward,\ninvolving surface chemistry and essential feedback control. Absolute\ntension value evaluation has been reliable since the CBB was first\nused for tension sensitivity experiments in 2018, 53 with more sophisticated tension evaluation methods incorporating\nbubble pressure measurements introduced in 2021. 51 The tension-clamp CBB allows for more sophisticated tension-related\nexperiments than previously possible since asymmetric membranes are\nreadily formed. 52 , 66 In contrast, the conventional\npatch-clamp method 39 and the recently developed\n“freestanding lipid bilayer tensiometer” system 54 face difficulties in maintaining constant tension\ndue to the relaxation characteristics of glass-sealed patch membranes 40 − 43 and the bulged planar lipid bilayers. 54 The tension-clamp CBB was validated using KcsA and TREK-1 potassium\nchannels, demonstrating reproducible channel responses to repeated\ntension changes over a long time course. The tension-clamp CBB\nsystem has significant potential for further\ndevelopment. The tension step speed was set to moderate in the long-run\nexperiments ( Figure 6 ) compared with the short-run experiment ( Figure 5 ). However, it can be speeded up by finely\ntuning the feedback parameters while maintaining bubble integrity.\nPractical applications include loading sinusoidal and other tension\nvariations, enabling the examination of frequency responses of tension-dependent\nchannel activity. The system also serves as a test platform for evaluating\ntension-sensitive dyes. 76 − 78 Additionally, it can form membranes\nwith arbitrary lipid compositions and asymmetric bilayers, 52 allowing distinct tension to be applied to each\nleaflet—a feature reserved for future studies. Unlike other\nstatic systems, the lipid compositions of a bilayer or leaflet are\nimmediately changed through the “membrane perfusion”\nmethod. 48 , 52 The “Chart” software allows\ntension time courses to be designed and operated automatically. The\nsize-clamp system stabilizes experiments by maintaining the bubble\nsize, preventing spontaneous size changes. This method also has critical\nadditional applications, such as replacing surface chemistry experiments\nconducted with the pendant drop method 79 under size-clamp conditions. Moreover, the size clamp also enables\nautomatic CBB formation via the Chart program, such that slow blowing\nof bubbles at the beginning to avoid breakage secures automatic bubble\nformation and maintenance. The entire control system is moving toward\nestablishing a CBB robot for automated lipid bilayer formation and\nmanipulation, simultaneously evaluating membrane tension and single-channel\nbehavior. Study Limitations Various methods for membrane-tension\nexperiments have been developed. For the tension measurements, membrane\ntether pulling from cell membranes has a solid physical basis, evaluating\nin situ membrane tension as low as 0.01 mN·m –1 . 35 This method is, however, confined\nto tension measurements without tension manipulations and channel\nactivity measurements. Patch-clamp methods can evaluate membrane tension\nwhile applying tension during single-channel current measurements.\nHowever, the unknown tension is applied upon giga-seal formation,\nwhich cannot be evaluated precisely, and evaluating the small curvature\nof the patch membrane is technically challenging for precise evaluation\nof absolute membrane tension. 39 The solvent\ninjection assay to droplet lipid bilayer 37 is straightforward, but underlying physicochemical processes upon\nvolume expansion are complex. Freestanding lipid bilayer tensiometer 54 drives tension changes by applying hydrostatic\npressure, and tension can be evaluated from membrane curvature. However,\na freestanding torus containing a bulk organic solvent supports the\nfreestanding lipid bilayer. After bulging of the membrane that includes\nthe torus upon hydrostatic pressure application, equilibration and\nrelaxation between the bilayer and bulk torus phases spontaneously\noccur, which prevents steady maintenance of membrane tension. In contrast, tension-clamp CBB is based on simple physicochemical\nprinciples, and rapid changes in leaflet tension are immediately equilibrated\ncompared to the above systems. Real-time feedback control has never\nbeen considered in tension control experiments. Among others, distinct\ntension control in each leaflet is a specific feature of the tension-clamp\nCBB, which has never been attained. One limitation of the system is\nthe fragile objectives of mechanical control, which are especially\nprominent at very low tension. Below 0.8 mN·m –1 , stable maintenance of bubbles is challenging, which could, however,\nbe overcome in future studies."
} | 2,004 |
36677325 | PMC9862501 | pmc | 9,130 | {
"abstract": "Microalgae are regarded as a promising source of biodiesel. In contrast with conventional crops currently used to produce commercial biodiesel, microalgae can be cultivated on non-arable land, besides having a higher growth rate and productivity. However, microalgal biodiesel is not yet regarded as economically competitive, compared to fossil fuels and crop-based biodiesel; therefore, it is not commercially produced. This review provides an overall perspective on technologies with the potential to increase efficiency and reduce the general costs of biodiesel production from microalgae. Opportunities and challenges for large-scale production are discussed. We present the current scenario of Brazilian research in the field and show a successful case in the research and development of microalgal biodiesel in open ponds by Petrobras. This publicly held Brazilian corporation has been investing in research in this sector for over a decade.",
"introduction": "1. Introduction The world’s energy expenditure is expected to increase by approximately 50% between 2018 and 2050 [ 1 ]. Fossil fuels, a non-renewable energy source, provide around 80% of all energy consumed worldwide [ 1 , 2 ]. Their use, however, leads to large emissions of greenhouse gases (GHGs), mainly CO 2 , which is a major contributor to global warming [ 3 , 4 ]. Since the Kyoto Protocol (1996) and the Paris Agreement (2015) and the last report from the Intergovernmental Panel on Climate Change (IPPC), it is evident that urgent action is necessary to change this scenario. These reports recognize the interdependence of climate, ecosystems, biodiversity, and human societies, and the impact of CO 2 and other toxic gases on the planet. Consequently, the industrial sectors are looking for ecological solutions and green technologies to reduce these emissions, resulting in alternative and innovative solutions [ 5 , 6 ]. Biofuels are one of the main alternatives to fossil fuel exploitation [ 7 , 8 ]. These fuels, produced from biomass or waste feedstocks, have the advantages of renewability and a significantly reduced contribution to global warming. The main biofuels available are biodiesel and bioethanol [ 2 , 9 , 10 ]. Other alternative green solutions are biomethane and biohydrogen [ 11 ]. Biodiesel is produced from lipids mainly by transesterification reactions having oils as the starting material [ 12 , 13 ]. Projections show that in 2040, biodiesel will account for 70% of the growing demand for transport fuel [ 14 ]. An increase in commercial biodiesel is observed. This rise was propelled by the increasing demand for green energy alternatives and public policies in many countries, including the United States, Brazil, and European nations, the leading world biodiesel producers [ 14 , 15 , 16 ]. Commercial biodiesel is currently obtained from different oil crops, such as soybean, corn, sunflower, and oil palm [ 9 ]. One of the main concerns related to these biodiesel sources is the use of arable lands resulting in competition with other segments, such as bioethanol and agriculture/livestock feed/food production [ 17 ]. Moreover, the time of production, climatic dependence, and soil quality, among other factors, introduce significant variability in crop production. In addition, fertilizer application releases nitrous oxide, a potent greenhouse gas [ 18 ]. High lipid contents make microalgae a promising alternative for biodiesel production. Besides this, microalgae are the major source of oxygen on the planet, and their CO 2 biosequestration by photosynthesis point to the biodiesel from microalgae as a promising carbon-neutral fuel. In this context, microalgae biomass is emerging as a source of biodiesel [ 18 ]. These microorganisms show higher growth rates and productivity in comparison with conventional crops. Moreover, they can be cultivated using wastewater, thus avoiding competition for freshwater and increasing sustainability [ 8 , 9 , 19 ]. An essential differential is the concept of biorefinery applied to microalgae cultures. After lipid extraction, microalgal biomass residues contain several bioproducts of high value. This concept decreases the impacts of costs and productivity. Several countries, including Brazil, are investing in the development of algal biotechnology [ 11 ]. However, there are bottlenecks to be overcome, such as expensive and energy-intensive cultivation, microbial contamination, and the biodiesel conversion processes. All these factors lead to a higher production cost, the major challenge for biodiesel production after scaling-up [ 20 ]. In this review, we aim to provide a perspective of biodiesel production from microalgae, with an emphasis on technologies with potential applications to increase efficiency and reduce the overall costs of the process. These strategies include methods for increasing microalgal lipid productivity and production within a biorefinery concept, which enables the exploitation of valuable bioproducts such as carotenoids and fertilizers, amongst others. Moreover, we will present the progress and prospects of microalgae biodiesel production in Brazil."
} | 1,282 |
39550371 | PMC11569254 | pmc | 9,131 | {
"abstract": "Horizontal gene transfer (HGT) mediated diversification is a critical force driving evolutionary and ecological processes. However, how HGT might relate to anthropogenic activity such as nitrogen addition, and its subsequent effect on functional diversity and cooccurrence networks remain unknown. Here we approach this knowledge gap by blending bacterial 16S rRNA gene amplicon and shotgun metagenomes from a platform of cessation of nitrogen additions and continuous nitrogen additions. We found that bacterial HGT events, functional genes, and virus diversities increased whereas bacterial taxonomic diversity decreased by nitrogen additions, resulting in a counterintuitive strong negative association between bacterial taxonomic and functional diversities. Nitrogen additions, especially the ceased one, complexified the cooccurrence network by increasing the contribution of vitamin B12 auxotrophic Acidobacteria, indicating cross-feeding. These findings advance our perceptions of the causes and consequences of the diversification process in community ecology.",
"introduction": "Introduction Diversification, selection, dispersal, and drift are four key processes driving community assembly; among them, the contribution of diversification to community assembly is the least understood 1 – 3 . Horizontal gene transfer (HGT), via which microbes incorporate genes from organisms in their environment, is one of the most essential mechanisms underlying the diversification process of bacterial communities 4 – 9 . Recent research found that HGT occured across a range of ecosystems, mediated by viruses, plasmids, transposons, and integrons 10 – 12 . Through HGT, genetic information can be transferred between microbial taxa, altering fitness differences among co-existing members, with potential cascading effects on community assembly and ecosystem functioning. Ecological frameworks for predicting when and where HGT occurs in nature are lacking 10 , 13 . We propose that HGT is important for the fitness of species when a community experiences a switch of selective forces. This switch of selective force may be induced by excessive nitrogen (N) that alleviates resource limitation but causes stress from increased acidity and heavy metal ions in natural ecosystems 14 – 18 . Although cessation of excessive N addition could restore resource limitation 19 , the stress of increased acidity and heavy metal ions often cannot be alleviated over a short period 17 . Therefore, our first hypothesis (H 1 ) is that both the cessation of N addition and continuous N addition increase the occurrences of HGT. Functional gene diversity does not necessarily obey taxonomic diversity. For example, two recent studies found that bacterial taxonomic diversity increased while functional diversities decreased along an altitudinal or latitudinal gradient 20 , 21 . A body of previous studies has demonstrated that excessive N addition causes a decline in bacterial taxonomic diversity across various ecosystems 22 – 24 ; however, the response of microbial functional diversity remains less understood 25 . Shifts in community-level functional gene diversity may be a joint consequence of taxonomic diversity and HGT 26 – 28 . Recently, a mismatch between increasing taxonomic diversity and decreasing functional diversity of bacterial community was discovered along a natural pH gradient that selects for adaptive strategies that affect functional trait distributions 21 . Although many studies have found that soil acidification could be caused by excessive nitrogen addition (Supplementary Table 1 and Supplementary Table 2 ), its cascading effect on the relationship between bacterial taxonomical and functional diversities remains unresolved. Thus, guided by previous findings that bacterial taxonomical and functional diversities are negatively associated along a pH gradient and that nitrogen addition induces soil acidification 18 , 21 , we hypothesize (H 2 ) that bacterial taxonomic and functional diversities are negatively associated along the nitrogen addition gradient. The development of our hypotheses on biotic interactions in the cessation of N addition and continuous N addition is guided by two essential hypotheses regarding resources and stress. First, the stress gradient hypothesis (SGH) predicts that community members tend to be more facilitative and less competitive under stress, but more competitive and less facilitative under benign conditions 29 – 34 . Second, the hunger game hypothesis (HGH) posits that oligotrophic members tend to be more facilitative and less competitive under resource-poor conditions, while copiotrophic members tend to be more competitive and less facilitative under resource-rich conditions 35 . In our system, we measured soil N, soil pH, and heavy metal (cuprum [Cu] and manganese[Mn]) ion concentrations to depict resource availability and abiotic environment stressors of different treatments. Our results showed that N was limiting (soil available inorganic nitrogen: 7.7 ± 1.2 mg kg −1 ) and the background soil biogeochemistry was benign (soil pH is near neutral) in the control treatment with ambient N (Figs. 1 , 2 and Supplementary Fig. 1 ). By contrast, the continuous N addition treatment (referred to as N cont ) represented an environment where N-limitation was removed, but where soil acidity and metal toxicity increased (Fig. 2 and Supplementary Fig. 1 ). Our third treatment, where N was initially added, but followed by the cessation of N addition (referred to as N cess ), represented an environment where N limitation had been restored, but where the legacy of the shifts in soil biogeochemistry (acidity and heavy metal ions) remained (Fig. 2 and Supplementary Fig. 1 ). By integrating the resource-stress quadrants of various treatments, and considering that oligotrophic members in resource-poor conditions and stress-tolerators in harsh environments are generally more facilitative and less competitive, we hypothesize (H 3 ) that the proportion of positive biotic associations will be higher in N cess compared to N cont and the control. Fig. 1 Experimental design. a Spatial distribution of ten blocks. Each block is comprised of 72 plots (3.5 × 8 m 2 ) representing different treatments of a full factorial combination of two N addition regimes (N add : cessation of N addition or continuous N addition), nine N levels (N level : 0, 1, 2, 3, 5, 10, 15, 20, and 50 g N m −2 yr −1 ), two frequencies (two or twelve times yr −1 ), and two mowing types (mown and unmown). For the two N addition regimes, the continuous N addition (N cont ) received continuous N applications for 11 yr (2008–2019), the cessation of N addition (N cess ) was treated with 6 yr of N addition (2008–2014), followed by 5 yr (2014–2019) of cessation of N addition. Note that ( a ) was captured and provided by our co-author, Jun-Jie Yang. b N addition regime and sampling. In this research, samples were collected in 2 years (2017 and 2019) in 70 plots from 5 blocks comprised of two N addition regimes (N cess and N cont ), seven N levels (0, 2, 5,10, 15, 20, and 50 g N m −2 yr −1 ) in one frequency (twelve times yr −1 ), and one mowing type (unmown). Fig. 2 Framework and hypothesis formulation. a , b Continuous nitrogen addition (N cont ) and cessation of N addition (N cess ) affect soil resource availability and abiotic stressors . a Resource availability as measured by available N was strongly increased by N cont but not by N cess . b Abiotic stress as measured by soil acidity was strongly increased (lower pH) by both N cont and N cess . Thus, N cont increased both resource availability and abiotic stress, and N cess increased abiotic stress but not resource availability. Note we only present high nitrogen levels (N50), and the effects of all nitrogen levels on soil resource availability and abiotic stressors are shown in Supplementary Fig. 1 . c Assignments of control, N cont , and N cess treatments to the quadrants defined by the axes of resource and stress. The orthogonality of the axes resource and stress defined four quadrants by referencing Wang’s research 21 . The control characterized by poor resource and neutral pH is placed in the top-left quadrant, the N cont treatment by rich resource and acidic pH is placed in the bottom-right quadrant, and the N cess treatment by poor resource and acidic pH is placed in the bottom-left quadrant. d , e Hypothesis on horizontal gene transfer (HGT) and biotic interaction. d Considering HGT might be important for the fitness of species when a community experiences a switch of selective forces, and switch of selective force may be induced by N cont and N cess , we hypothesize that (H 1 ) N cont and N cess increase the occurrence of HGT. Guided by Wang’s research 21 that bacterial taxonomical and function diversities are negatively associated along a pH gradient and that nitrogen addition induces soil acidification, we hypothesize that (H 2 ) bacterial taxonomic and functional diversities are negatively associated along the nitrogen addition gradient. e Guided by the stress gradient hypothesis that high abiotic stress causes more biotic facilitation and less competition (vertical axis) and the hunger game hypothesis that high resource availability induces more competition and less facilitation (horizontal axis), we hypothesize that (H 3 ) positive association would be higher in N cess than N cont and the control. The system we used to test these hypotheses consists of a semi-arid grassland field with seven N addition levels (N level ) and three N regimes: control, N cess , and N cont , with five replicate plots per treatment. Soil samples were collected in 2017 and 2019, from which we isolated DNA to characterize bacterial community by 16S rRNA gene sequencing, and functionality by shotgun metagenomics (Fig. 1 ). Our analyses supported the H 1 by finding that N cess and N cont increased the occurrences of HGT events. We also found evidence supporting H 2 , as bacterial taxonomic diversity was negatively correlated with functional diversity. Lastly, in support of H 3 , we found that N cess increased the proportion and strength of positive associations among bacterial taxa, which might be led by vitamin B12 auxotrophic Acidobacteria members.",
"discussion": "Results and discussion The background soil available nitrogen concentration at the experiment site was ~7.7 ± 1.2 mg kg −1 . Continuous nitrogen addition (N cont ) significantly increased soil available nitrogen content (Supplementary Fig. 1a ). For instance, at the highest nitrogen addition level (N cont N50), the soil available nitrogen content reached 460.8 ± 34.1 mg kg −1 (Fig. 2a ). However, following the cessation of nitrogen addition (N cess ), the soil available nitrogen content decreased close to the control level (Fig. 2a ). At the highest nitrogen addition level under the N cess treatment (N cess N50), the soil available nitrogen concentration was ~17.2 ± 3.9 mg kg −1 (Fig. 2a ). The background soil pH was nearly neutral, around 7.4 ± 0.1 (Fig. 2b ). The N cont treatment led to a decrease in soil pH and an increase in soil heavy metals (Supplementary Fig. 1 ). For example, at N cont N50, the soil pH dropped to about 4.5 ± 0.1 (Fig. 2b ). The N cess treatment did not recover soil pH and heavy metals to the control level (Supplementary Fig. 1 ), with the soil pH at N cess N50 measuring around 5.2 ± 0.2 (Fig. 2b ). In summary, the control was characterized by poor resources and neutral pH, the N cont treatment by rich resources and acidic pH, and the N cess treatment by poor resources and acidic pH. By referencing the four quadrants defined by the orthogonality of the axes resource and stress of Wang et al. 21 , based on the soil resource and stress, we placed the control to the top-left quadrant, the N cont treatment to the bottom-right quadrant, and the N cess treatment to the bottom-left quadrant (Fig. 2c ). Testing H 1 : N cess and N cont increase the occurrences of HGT To test the H 1 that N cess and N cont increase the occurrences of HGT, we detected HGT based on contigs using the WAFFLE pipeline ( https://github.com/biobakery/waafle ) from metagenomic data. WAAFLE identifies potential HGT events in metagenomes by aligning metagenomic contigs with microbial reference sequences 36 . Initially, we processed approximately 839 GB of raw metagenomic data from 70 soil samples. After rigorous quality control—removing Illumina adapters and discarding short sequences—we refined the dataset to 820 GB of high-quality metagenomic data. Our WAAFLE analysis of these contigs revealed a total of 3452 HGT events at the contig level. Detailed information on the contigs, HGT events, and normalized HGT events for each sample can be found in Supplementary Table 3 . Overall, the number of HGT events detected from metagenomes increased significantly with increasing N levels (Fig. 3a ). Furthermore, to account for the differences in the number of contigs assembled from different metagenomes, the number of HGT events was normalized by the number of contigs for each sample. We found that the normalized number of HGT events also increased significantly with increasing N levels (Supplementary Fig. 2 ). Thus, our H 1 was supported, as the number of HGT events increased with the increasing N levels in N cess and N cont . Fig. 3 N addition increase horizontal gene transfer (HGT). a The number of HGT events and ( b ) the richness of horizontally transferred genes (S.HGT) were both positively correlated with N addition levels. Shaded areas indicate 95% confidence intervals. The X-axis is nitrogen addition levels (0, 2, 5, 10, 15, 20, 50 g m −2 yr −1 ), which we have taken the natural logarithm of them. Preference analysis of horizontally transferred functional genes between N0 and N50 of either in ( c ) N cess or ( d ) N cont . The diagram of two-dimensional preference (2DP) with false discovery rate (FDR) adjusted P values shows a strong preference of genes toward N50 as judged by P FDR < 0.05 and 2DP > 1.3 in the right-bottom quadrant. e Functional enrichment of genes preferred toward N50 in either N cess or N cont . N50 preferred genes enriched functions of translation and xenobiotics degradation in both N cess and N cont , cell motility, cellular community, signal transduction and membrane transport in N cess , and energy metabolism in N cont . f – s Visualization of horizontal gene transfer among microbial taxa in each treatment. The band width indicates the number of HGTs between two microbial groups. The ring is colored by microbial taxonomy at the phylum level. Note most HGTs occurred within the Actinobacteria, followed by Proteobacteria. Along with the increase of HGT, the richness of horizontally transferred functional genes (S.HGT) was also increased significantly with increasing N addition levels in N cess and N cont (Fig. 3b ) ( R 2 = 0.221, P < 0.001). We then explored the functions of this richness of horizontally transferred genes with preference analysis and functional enrichment. Preference analysis showed that 31 out of the 368 horizontally transferred genes showed significant preferences toward N50 in N cont (Fig. 3c ), and 23 out of the 384 horizontally transferred genes displayed significant preferences toward N50 in N cess (Fig. 3d ). We focused exclusively on the N50 level because we did not detect enough preference Kyoto Encyclopedia of Genes and Genomes Ontology (KEGG) Ontology (KOs) for functional enrichment analysis in other nitrogen addition levels (Supplementary Fig. 3 ). Enrichment analysis showed that horizontally transferred genes preferred in N50 enriched functions of translation and xenobiotics degradation in both N cess and N cont , cell motility, cellular community (quorum sensing & biofilm formation), signal transduction and membrane transport in N cess , and energy metabolism in N cont (Fig. 3e and Supplementary Fig. 4 ). Our results suggest that HGT may play a role for biotic interactions and stress tolerance of bacteria in adaption to N cess and N cont . The occurrences of HGT were then visualized by circlize plots, showing that most of the HGT occurred between members of Actinobacteria (79 to 177 HGTs per treatment), followed by Proteobacteria (74 to 131 HGTs per treatment) (Fig. 3f-s ). This condition might be explained by the filamentous growth mode of Actinobacteria, which might increase the odds of encountering different taxa 37 . Interestingly, despite the increase in the HGT of Actinobacteria members, the relative abundance and richness of Actinobacteria were reduced by N addition (Supplementary Figs. 5 and 6 ). Thus, the increase in HGT might be an important mechanism for Actinobacterial stress tolerance and survival. In addition, we also detected an increase in viral biodiversity and a decrease in plasmid biodiversity with increasing nitrogen addition levels (Supplementary Fig. 10 ). These results together suggest that the viruses might mediate the increased occurrence of HGT that mostly occurred in the monoderm, filamentous Actinobacteria. Testing H 2 : N cess and N cont cause negative correlations between bacterial taxonomic and functional diversities Utilizing Kaiju, a pipeline that classifies individual metagenomic reads by referencing a database containing annotated protein-coding genes from various microbial genomes 38 , we found that the composition of archaeal was significantly affected by nitrogen addition levels, but not by nitrogen addition regimes, or their interaction (Supplementary Fig. 7a ). In contrast, bacterial community composition was significantly affected by nitrogen addition levels, nitrogen addition regimes, and their interactions (Supplementary Fig. 7b ). Consequently, we focused exclusively on bacteria diversity in the subsequent analysis. We tested H 2 by analyzing our 16S rRNA gene amplicon data using the USEARCH pipeline and annotating our metagenome data at contigs level with the KO database, followed by calculation of richness and Shannon diversity index for bacterial taxonomic and functional diversities for each sample. The small subunit (SSU) rRNA gene sequences were also extracted and reconstructed from metagenomic data using the RiboTaxa pipeline 39 . The SSU rRNA gene sequences were composed of 92.4% bacterial, 3.6% archaeal, and 4% eukaryotic reads (Supplementary Fig. 8 ). The level of nitrogen addition did not significantly correlate with bacterial diversity, as determined from the SSU rRNA gene sequences in the metagenome (Supplementary Fig. 9 ). Although searching for rRNA sequences in metagenomes is a promising approach to better depict bacterial communities, its accuracy in depicting diversity can depend on factors such as sequencing depth and length, as well as the bioinformatic algorithms, tools, databases, and pipelines used. For the following analysis, we focused exclusively on bacterial diversity based on 16S rRNA gene amplicon data. We then plotted bacterial taxonomic, virus taxonomic, plasmid taxonomic, and functional diversities against the N level (Fig. 4 , Supplementary Figs. 10 and 11 ). The results showed that bacterial taxonomic diversity decreased with increasing N addition levels in N cess and N cont (Fig. 4a and Supplementary Fig. 11a ), as demonstrated frequently in previous studies (Supplementary Tables 1 and 2 ). However, we found that bacterial functional diversity increased with increasing N addition levels in N cess and N cont (Fig. 4b , and Supplementary Fig. 11b ). Together, having shown that taxonomic and functional diversities respectively decreased and increased with N addition levels, a comparison of these two diversities revealed a strong, significant negative correlation between them (Fig. 4c , and Supplementary Fig. 11c ). Therefore, H 2 is supported because a positive association between taxonomic and functional diversities is not detected. Fig. 4 Bacterial taxonomic diversity mismatched with functional diversity. a , b N additions of either N cess or N cont decreased bacterial taxonomic diversity (S.16S) but increased bacterial functional diversity (S.KO). a Bacterial taxonomic diversity (S.16S) that measured by the richness of operational taxonomic units (OTUs) detected by 16S rRNA amplicon metabarcoding, was decreased with increasing of nitrogen levels in both N cont and N cess . b Bacterial functional diversity (S.KO) measured by the richness of the Kyoto Encyclopedia of Genes and Genomes (KEGG) Ontology (KOs) as annotated from the shotgun metagenome at contigs level, was increased with increasing nitrogen levels in both N cont and N cess . c Bacterial taxonomic diversity (S.16S) and bacterial functional diversity (S.KO) were negatively correlated. d The ratio of functional diversity to taxonomic diversity (S.KO: S.16S) increased with N addition levels in N cess and N cont . e Heatmap showing intercorrelations among N addition levels, soil inorganic N concentration, soil pH, plant biomass, plant richness, the richness of horizontally transferred genes (S.HGT), bacterial average genome size (AGS), and the ratio of functional diversity and taxonomic diversity (S.KO: S.16S) in N cont (top right corner) and N cess (down left corner). f Structural equation modeling (SEM) showing the direct and indirect drivers of the ratio of functional diversity and taxonomic diversity (S.KO: S.16S). The ratio of functional diversity and taxonomic diversity (S.KO: S.16S) were influenced by N addition indirectly through inorganic nitrogen, the average genome size (AGS), and the richness of horizontally transferred genes (S.HGT). Blue and orange solid lines indicate significant ( P < 0.05) positive or negative pathways, respectively. To depict the disparity between taxonomic and functional diversities, we calculated the ratio between them (S.KO: S.16S), followed by plotting this ratio against N level. The results showed that the ratio of functional diversity and taxonomic diversity (S.KO: S.16S or H’.KO : H’.16S) increased with N addition levels in N cess and N cont (Fig. 4d , and Supplementary Fig. 11d ). The ratio of function diversity and taxonomic diversity (S.KO: S.16S) negatively correlated with the richness of horizontally transferred genes in N cont (Fig. 4e ). The results of the structural equation modeling (SEM) showed that the ratio of functional diversity and taxonomic diversity (S. KO: S.16S) were influenced by N addition indirectly through inorganic nitrogen, average genome size, and the richness of horizontally transferred genes (Fig. 4f and Supplementary Fig. 12 ). Thus, the mismatch between taxonomic and functional diversities might be attributed to the increase in average genome size and the richness of horizontally transferred genes induced by N cess and N cont . In partial support of our finding, a recent study found that the changes in average genome size along pH gradient underpin the negative association between taxonomical and functional diversities along a latitudinal gradient 21 . Testing H 3 : N cess increases the proportion and strength of positive associations among bacterial taxa To test our H 3 that the proportion of positive association would be higher in N cess than in control and N cont , we calculated the distribution frequency of all possible Spearman correlations between bacterial taxa by combining amplicon data of 2017 and 2019 in each treatment. Overall, H 3 was supported by our analysis, as the strength of positive correlations was higher in N cess than in control and N cont (Fig. 5a and Supplementary Fig. 13 ). Besides, the results showed that the proportion of positive correlation in N cess was higher than N cont (Fig. 5b ). Fig. 5 More positive associations in N cess than N cont and control. a , b N cess increases the extent and proportion of positive associations between microbial taxa more than N cont . a The frequency distributions included all correlations between microbial taxa as assessed by Spearman’s Rho. b In general, N50 increased the proportion of positive correlations, to a greater extent at N cess than N cont . c The structure of the association network was strikingly complexified by N cess ; and N cont also increased the complexity of the network structure, but only when the drop of richness was accounted for. d As compared to control, all network properties (edges, vertices, positive links, negative links, average degree) were increased (1.5 to 4.5 times) by N cess , and different network properties were either decreased (negative links quartered and vertices halved), increased (average degree doubled), or largely unchanged (edges and positive links) by N cont . e The contribution of Acidobacteria to network degree was greatly increased in N cess . The differences in the extent and proportion of positive associations among treatments might be explained by changes in resources and stress. As measured by N availability, resources were poor in N cess and control but rich in N cont (Fig. 2 ). Regarding soil pH and heavy metal ions, the abiotic environment was found to be harsh in N cess and N cont , but benign in control (Fig. 2 and Supplementary Fig. 1 ). Together, guided by the SGH and HGH, a lack of resources and a harsh environment are two forces that promote positive interactions 34 , 35 , 40 . Thereby, it was unsurprising to observe the highest proportion of positive association in N cess . On the other hand, for N cont treatment, the harsh environment was supposed by SGH to promote positive interactions whereas the rich resource was supposed by HGH to promote negative interactions. Similarly, for the control treatment, the poor resources were supposed by HGH to promote positive interactions whereas the benign environment was supposed by SGH to promote negative interactions. Together, the observed higher proportion of positive association in N cont treatment than in the control indicated that, in our system, the effect of abiotic stress on biotic interactions was stronger than that of resource availability. Our test of H 3 that focused on the proportion of positive associations used all possible correlations, regardless of significance. However, the inclusion of all, significant and nonsignificant correlations in this analysis failed to capture the strength of biotic associations. To determine the influence of different treatments on the strength of biotic associations, we focused on only significant correlations (|Rho| > 0.8, FDR P < 0.05) to construct a co-occurrence network. Most strikingly, the network was much more complex in N cess than N cont or the control (Fig. 5c , Supplementary Figs. 14 – 16 ). For example, as compared to the control, N cess increased the value of all network properties including edges (538 v.s. 2620), vertices (630 v.s. 926), and average degree (1.708 v.s. 5.659) (Fig. 5d ). Compared to the control, N cont also increased the complexity of network structure, but only after accounting for the decline in taxonomic richness caused by N addition. Specifically, as the operational taxonomic units (OTUs) richness of bacterial community at N50 of N cont treatment (richness = 941 ± 178) was less than half of that in control (richness = 2546 ± 35) (Supplementary Fig. 17 ), the network analysis showed that the number of vertices at N50 of N cont treatment (vertices = 363) was also about half of that in control (vertices = 630) (Fig. 5d ). Despite the near halving of the number of vertices (363 v.s. 630), an increase in the complexity of network was depicted by the slight increase in the number of edges (593 v.s. 538) (Fig. 5d ). Together, the value of the average degree at N50 of N cont was about twice (average degree = 3.267) that of the N0 control (average degree = 1.708) (Fig. 5d ). Thus, N addition increased the proportion and strength of positive associations, and these effects were exaggerated in the N cess compared to N cont treatment. The increase in network complexity for N cess corresponded to an increase in the proportion or relative proportion of Acidobacteria vertices (Fig. 5e and Supplementary Fig. 18 ). As N cess was characterized by poor resources and a harsh abiotic environment, the results suggested Acidobacteria might be oligotrophic and stress tolerant. Note that although Acidobacteria dominated the network (Fig. 5e , Supplementary Figs. 18 and 19 ), their relative abundance was lower in N cess than in the control (Supplementary Fig. 5 ). As oligotrophs, Acidobacteria lost their relative abundance when plots received N. However, as slow-growing stress tolerators, they reclaimed relative abundance very slowly, despite seeming high fitness for the resource-poor and harsh abiotic environment of N cess . According to the hunger game hypothesis, in resource-poor conditions, oligotrophic members are thought to facilitate each other through a process known as cross-feeding, where co-factors and vitamins are exchanged among microbial taxa 41 – 43 . To investigate the potential for cross-feeding in our experiment, we compared the functional genes in Acidobacteria with those in other Phyla by assembling genomes from metagenomic data (Supplementary Figs. 20 , 21 , and Supplementary Table 4 ). Using near-complete genomes for vitamin B12 analysis, we found that gene copies encoding vitamin B12 production were lower in Acidobacteria (8) compared to Proteobacteria (15) or Actinobacteria (11) (Supplementary Figs. 22 and 23 ). Vitamin B12, a N-containing compound, is an essential co-factor in DNA synthesis, along with many other critical cellular functions. Thus, our results suggest that the cross-feeding on vitamin B12 of Acidobacteria may be a driver of increased network complexity in the N cess treatment. Patescibacteria, the Candidate Phyla Radiation superphylum characterized by their small cell sizes, the absence of key biosynthetic pathways, and likely obligate intracellular symbionts, are widely distributed across various environments 44 , 45 . Recent studies utilizing imaging techniques and omics approaches have revealed that the lifestyle of Patescibacteria might be symbiotic or parasitic 44 , 46 – 48 . In our study, we observed that the abundance and richness of Patescibacteria increased with elevated levels of nitrogen addition (Supplementary Figs. 5 and 6 ), consistent with earlier findings 49 . However, the underlying cause and consequence of this increase in Patescibacteria relative abundance in response to N addition remains unclear. Nitrogen (N) addition and deposition often result in biological stress due to both the shift in the nutritional environment and the reduction of pH and heavy metal toxicity that subsequently occurs. One novelty of our research is the deconstruction of N addition into axes of resource and stress, resulting in three quadrants defined by ambient N addition, continuous N addition, and cessation of N addition. Focusing on the switch of selective forces across different quadrants, we were able to develop testable hypotheses regarding essential community assembly processes such as diversification and biotic interaction. Our detected increase of HGT helps to bridge the gap in the mismatch between taxonomic and functional diversities induced by N addition. Our analysis challenged the consensus that bacteria lose diversity in response to excessive N addition by considering bacterial functional gene diversity that is related to the variations of HGT. Our results raise serious questions about the contribution of bacterial diversity to ecosystem functioning, as taxonomic and functional diversities exhibited contrasting patterns in response to N addition. Our detection of HGT from metagenomic data is still in its infancy, and its application in microbial ecology will be guaranteed by developing more reliable and robust tools (WAFFLE, DarkHose, MetaCHIP) 36 ."
} | 8,027 |
35514904 | PMC9057943 | pmc | 9,132 | {
"abstract": "In contrast to the commonly present UV light-stimulated synaptic oxide thin-film transistors, this study demonstrates a violet light (wavelength of 405 nm) stimulated zinc–tin oxide (ZTO) photoelectric transistor for potential application in optical neuromorphic computation. Owing to the light-induced oxygen vacancy ionization and persistent photoconductivity effect in ZTO, this device well imitates prominent synaptic functions, including photonic potentiation, electric depression, and short-term memory (STM) to long-term memory (LTM) transition. A highly linear and broad dynamic range of photonic potentiation can be achieved by modulating the light power density, while electric depression is realized by gate voltage pulsing. In addition, the brain-like re-learning experience with extended forgetting time (200 s) is well mimicked by the ZTO photoelectric transistor. As a result, the ZTO photoelectric transistor provides excessive synaptic function with multi-series of synaptic weight levels (90 levels for each given light power density), which makes it prevalent in the neuromorphic computation of massive data as well as in learning-driven artificial intelligence computation.",
"conclusion": "Conclusions In summary, by applying violet light spikes and gate voltage pulses, prominent synaptic functions and memory behaviors, including potentiation, depression, STM-to-LTM transition and learning experience, are well imitated by the ZTO photoelectric neuromorphic transistor. The photonic potentiation process in the ZTO phototransistor is associated with the PPC effect, resulting from the activation energy barrier for the recombination of ionized oxygen vacancies with electrons, while the electric depression process is attained by applying positive gate bias pulses to annihilate the PPC effect. Owing to the light-induced oxygen vacancy generation and ultra-thin ZTO thickness, under sequential light spikes, the excitatory postsynaptic current (EPSC) of the ZTO phototransistor increases continuously and linearly with spike number up to at least 90, and the dynamic range of EPSC can be further modified by light power density. Therefore, the synaptic operation of the ZTO phototransistor provides an extensive series of weight levels and makes it suitable to process large quantities of data for next-generation neuromorphic computing.",
"introduction": "Introduction Computers based on the von Neumann architecture have been successfully developed for solving well-structured mathematical problems during the past several decades. 1 However, physically separated memory modules and processers of von Neumann-based computational systems are not suitable for solving unstructured and large quantities of data. 2 In order to overcome this issue, a new type of neuromorphic computation architecture, namely artificial neural networks (ANN) has been built. 3,4 Inspired by the human brain, the transmission between adjacent neurons in ANN is dependent on the synaptic weights, which are the basis for conveying information, processing data, and memory in neuromorphic systems. 5 ANN-based computational systems can therefore simultaneously process and store information of large quantities of unstructured data. In contrast to purely electrically trigged synaptic devices, 6–9 the photoelectric synaptic devices can be triggered by a series of UV light spikes or electric spikes, 10–13 where In 2 O 3 /ZnO heterojunction, indium gallium zinc oxide (IGZO) 11–15 and indium zinc oxide (IZO) 16 are used as active layers. Although the reported research works are applaudable, these devices are limited to UV light stimuli to intimate synaptic functions. While visible light communication (VLC) is a potential future technology, violet light has been demonstrated to carry a high transmission capacity beyond 25 Gbit s −1 . 17 In addition, human eyes have a relatively low visual sensitivity to violet light. Therefore, using violet light as a source for data transmission will cause fewer disturbances to human daily life. In this study, we have demonstrated a solution-processed zinc–tin oxide (ZTO) thin film transistor that can respond to violet light (405 nm in wavelength) stimulation and mimic the essential synaptic functions and memory behaviors based on the persistent photoconductivity (PPC) effect arising from visible light stimuli and suppression of the PPC effect with positive gate bias pulses. In addition, light-induced oxygen vacancies will replenish the oxygen vacancies for generating photoexcited electron carriers. Therefore, the degrees of photonic potentiation and electric depression are tailored by modulating the light power density and gate voltage, respectively, which provides a wide range of synaptic weight levels and makes it highly providential for the application in the neuromorphic computation of massive data.",
"discussion": "Results and discussion \n Fig. 1a shows the TEM cross-sectional image of the ZTO film deposited on SiO 2 , indicating that the thickness of ZTO is about 5 nm. Fig. 1b shows the transfer characteristic curves of the ZTO transistor measured in dark and under light illumination at different power densities and at a wavelength of 405 nm. The bandgap of ZTO was estimated to be 3.9 eV according to the Tauc plot converted from the UV-Vis transmission spectrum, as shown in Fig. S1 (ESI † ). The XPS O1s spectrum of the ZTO film is shown in Fig. S2 (ESI † ) and is deconvoluted into three components, including lattice oxygen in ZTO (530.1 eV, O I ), oxygen nearby oxygen-deficient ZTO (531.8 eV, O II ), and oxygen absorbed on the ZTO surface (532.6 eV, O III ). 18 The O II subpeak in the XPS O1s spectrum indicates the presence of oxygen vacancies in the ZTO film. When ZTO is illuminated by a sub-bandgap light source, neutral oxygen vacancies will be photo-ionized to positively-charged oxygen vacancies and release electrons to the conduction band of ZTO. 19 As a result, the negative shift of transfer characteristic curves shown in Fig. 1b is dominated by a positively electric field produced by the positively-charged oxygen vacancies mainly located at the ZTO/SiO 2 interface. 20 The photoelectrons released to the conduction band of ZTO will cause an increase in the conductivity of the ZTO phototransistor. Fig. 1 (a) TEM cross-sectional image of the ZTO film deposited on SiO 2 . (b) Transfer characteristic curves of the ZTO phototransistor in dark and under light illumination of 405 nm wavelength light of different power densities ( V D = 10 V). As discussed in the literature, 14 positively-charged oxygen vacancies take a long time to recombine with photoelectrons owing to the existence of the thermal activation energy, thus leading to the persistent photoconductivity (PPC) effect. In previous reports, the thermal activation energy for the neutralization of ionized oxygen vacancies in ZnO-based oxide is mostly found to be in the range from 0.27 eV to 0.37 eV. 14,21,22 By measuring the temperature-dependent photocurrent decay time constant for the ZTO transistor under 405 nm light illumination, the thermal activation energy for the neutralization of ionized oxygen vacancies of ZTO is related to the decay time constant, τ decay , by the Arrhenius equation. 1 where A is the pre-exponential factor, E a is the activation energy, k is the Boltzmann constant, and T is the absolute temperature. From the ln τ vs. 1/ T plot shown in Fig. S3 (ESI † ), the thermal activation energy for the neutralization of the ionized oxygen vacancies of ZTO can be obtained by fitting the Arrhenius plot and is calculated as 0.294 eV, which is consistent with the literature data. A schematic of biological presynaptic neurons, postsynaptic neurons, synapse and operation concept of the ZTO neuromorphic phototransistor are shown in Fig. 2a , where the information will be conveyed from presynaptic neurons to postsynaptic neurons through synapses. To intimate the biological synapses, the light spike, the positive gate bias pulse and the drain current (photo-induced current) are represented as the excitatory presynaptic neuron spike, the inhibitory presynaptic neuron spike, and the postsynaptic current, respectively. The change in the drain current ( i.e. change in conductivity) thus can be represented as synaptic weight. Fig. 2b shows the typical excitatory postsynaptic current (EPSC) of the ZTO phototransistor stimulated by a single light spike (405 nm, 2 mW cm −2 ) with different light spike durations at V D of 10 V and V G of 0 V. EPSC increases linearly with the increase in the spike duration, which is indicative of the proportional ionization of neutral oxygen vacancies to the light spike duration. Fig. 2c shows the EPSC variation of the ZTO phototransistor under 405 nm laser light spikes with different power densities and light spike durations. It is apparent that the light-stimulated ESPC of the ZTO phototransistor can be modulated broadly, which is associated with a wide range of synaptic weights for neuromorphic computing applications. Fig. 2 (a) A schematic of a biological presynaptic neuron, postsynaptic neuron, synapse and operation concept of the ZTO neuromorphic phototransistor. (b) EPSC of the ZTO phototransistor stimulated by a single light spike (405 nm, 2 mW cm −2 ) of different spike durations with V D of 10 V and V G of 0 V. (c) EPSC variation of the ZTO phototransistor triggered by 405 nm light spikes with different power densities and light spike durations. Based on the memory level and retention time, human memory behavior can be categorized into two types: short-term memory (STM) and long-term memory (LTM). 23 STM is recognized as the temporal memory stimulated by external stimulation and will forget immediately (low memory level and short retention time). LTM is recognized as the enduring memory stimulated by external spike and will forget slowly (high memory level and long retention time). With maintaining the rehearsal of stimulation, STM can be transformed into LTM. In order to investigate the memory behaviors, a series of light spikes (spike duration and interval are 0.5 s) are applied to the ZTO phototransistor. The post-synaptic current responses of the ZTO phototransistor to 405 nm light spikes of different power densities (1, 2 and 4 mW cm −2 ) for up to 90 spikes are shown in Fig. S4 (ESI † ), and the data pertaining to the light power of 2 mW cm −2 is plotted in Fig. 3a . As shown in Fig. S4 † and 3a , EPSC increases continuously without saturation under each light spike. The good linearity and wide dynamic synaptic weight range of this device are superior to other reported data and will be discussed later (see Table 1 ). On the other hand, the forgetting behavior (decay of EPSC after the light pulse section is finished) can be described by 24 2 where I ( t ) is the EPSC at time t , I max is the peak EPSC value for the last spike, I S is the steady EPSC after decay ( i.e. , the memory level) and τ is the retention time. Fig. 3 (a) EPSC of the ZTO phototransistor responses to 405 nm light spikes of different spike numbers ( P = 2 mW cm −2 , spike duration = 0.5 s, V D = 10 V and V G = 0 V). The STM-to-LTM transition induced by increasing the number of light spikes (405 nm, P = 2 mW cm −2 , spike duration is 0.5 s, spike interval is 0.5 s) with V D of 10 V and V G of 0 V of the ZTO phototransistor. (b) The variations in the retention time and memory level with the light spike number. (c) EPSC of the ZTO phototransistor responses to light spikes of different frequencies (405 nm, P = 2 mW cm −2 , spike duration = 0.5 s, V D = 10 V and V G = 0 V). (d) The variations in the retention time and memory level with the light spike frequency. A comparison of the synaptic potentiation characteristics of our device with the oxide phototransistors reported in the literature Device structure Light source and light power density (for synaptic functions) \n α \n Multi-level states Dynamic range \n I \n max / I 1 \n I \n max − I 1 (nA) Mo/IGZO/nanogranular SiO 2 /p + -Si 12 395 nm UV \n P = 158.62 mW cm −2 75 20 1.04 0.35 Au/Ti/IGZO/Al 2 O 3 /n + -Si 13 380 nm UV \n P = 0.6 mW cm −2 1.2 20 12.58 0.21 IZO/IGZO/SiO 2 /Si 14 380 nm UV \n P = 0.6 mW cm −2 0.8 30 42.44 121.30 Au/Ti/IGZO/HfO x /Au/SiO 2 /Si 15 254 nm UV \n P = 0.2 mW cm −2 0.7 27 62.84 295.29 Al/IZO/ion gel/Al 16 275 nm UV \n P = 1.82 mW cm −2 5 50 1.82 5.62 Al/ZTO/SiO 2 /p + -Si (this work) 405 nm visible \n P = 1 mW cm −2 1.17 90 11.62 50.13 \n P = 2 mW cm −2 1.07 90 25.64 130.70 \n P = 4 mW cm −2 1.10 90 52.75 527.80 By fitting the experimental data ( Fig. 3a ) with eqn (2) , the retention time ( τ ) and memory level ( I S ) are obtained as shown in Fig. 3b . An increase in the retention time and memory level with light spike number confirms that the ZTO phototransistor can undergo a transition from STM to LTM such as the human brain. Apart from increasing the light spike number, increasing the light spike frequency is also an effective way to mimic the transition from STM to LTM. An increase in EPSC after different spike frequencies are shown in Fig. 3c . During the measurement, the power density of light spikes and spike duration are 2 mW cm −2 and 0.5 s, respectively. The retention time and memory level at different light spike frequencies are shown in Fig. 3d , which clearly exhibits the transition from STM to LTM via increasing light spike frequency. The prolonged retention time of EPSC when the ZTO phototransistor is stimulated with the increase in the light spike number or frequency indicates that the PPC effect becomes stronger with cumulative photon doses. In investigations on the PPC effect of n -GaAs 25 and II–V mixed semiconductors, 26 it has been reported that both PPC build-up level and decay time increased with an increase in the dose of irradiated photons. Theoretical studies suggested that photo-generated electrons and holes were spatially separated and the holes were trapped (or localized), leading to the impediment of photo-generated carrier recombination. As discussed previously, the PPC effect in ZnO-based oxides originates from the energy barrier for the recombination of photo-ionized oxygen vacancies with photoelectrons. The photo-ionized oxygen vacancies are localized in the lattices; therefore, it is reasonable to postulate that the photoelectrons and photo-ionized oxygen vacancies in ZTO are also spatially separated. As a consequence, the PPC effect of the ZTO phototransistor can be enhanced with cumulative photon dose and the STM-to-LTM transition can be emulated by increasing the number or frequency of light spike stimuli. The learning-experience of the human brain has also been intimated by the ZTO phototransistor with three cycles of learning-and-forgetting, as shown in Fig. 4a–f . In the first learning process, EPSC increases linearly after each light spike. In the first forgetting process, EPSC gradually decays to an intermediate state after 200 s. This is analogous to the human brain, where the information will be partially lost after a period of time. Compared to the 18 spikes in the first-learning process, only 8 and 7 spikes are required for the second and third learning processes, respectively, to recover the decayed EPSC. In addition, during the second and third forgetting processes, the decay of EPSC is comparatively less than that for the first forgetting process within the same period (200 s). Similar to the human brain, when the same information is relearnt, we can learn faster and retain it for a longer time (less forgotten). As a result, the ZTO phototransistor imitates the learning-forgetting-relearning behavior of the human brain agreeably. Fig. 4 Learning-experience behavior in the ZTO phototransistor (learning process: 405 nm light, P = 2 mW cm −2 , light spike duration and interval are 0.5 s, V D = 10 V and V G = 0 V; forgetting process: V D = 10 V and V G = 0 V) (a) first learning process, (b) first forgetting process, (c) second learning process, (d) second forgetting process, (e) third learning process, (f) third forgetting process. The photonic potentiation behavior associates with the PPC effect of the ZTO phototransistor indicates the strengthening of the synaptic weight. On the contrary, the depression behavior must be accompanied by the elimination of PPC of the ZTO phototransistor, representing the weakening of the synaptic weight. As suggested in the literature, 19,21 by applying positive gate bias pulses, photoelectrons will accumulate at the front channel to accelerate the recombination rate of photoelectrons and photo-ionized oxygen vacancies. Fig. 5a–c reveal the various degrees of electric depression by modulating the positive gate voltage after photonic potentiation with different light power densities. Accordingly, the ZTO phototransistor can provide a wide range of synaptic weights for neuromorphic computing under different light power densities, spike numbers and positive gate bias voltages. Fig. 5 Photonic potentiation (405 nm, light spike duration and interval are 0.5 s, spike number is 30) under light power density of (a) 1 mW cm −2 (b) 2 mW cm −2 (c) 4 mW cm −2 and the subsequent electrical depression (gate bias is 5 V/10 V/15 V, pulse width and interval are 0.5 s, pulse number is 60) for the ZTO phototransistor. The accuracy of the neuromorphic computation is dependent on the linearity during potentiation, multilevel states ( i.e. number of synaptic weights), and dynamic range ( I max / I 1 and I max − I 1 , where I 1 is the current at the first light spike). 27 To effectively compare the linearity during potentiation, we employed the parameter α defined in the following equation. 28 3 I = (( I max α − I dark α ) × ω + I dark α ) 1/ α if α ≠ 0, I = I dark × ( I max / I dark ) ω if α = 0 where α equals 1 for the ideal case (linear potentiation). I max and I dark represent the maximum photocurrent and initial dark current, respectively. The internal variable ω varies between 0 and 1. The comparisons of these synaptic performances between the ZTO phototransistor and previously reported oxide TFTs are summarized in Table 1 . It is worth noting that our phototransistor is operated in the visible light region (violet, 405 nm) and the previously reported oxide TFT photoelectronic synapses are applied in the UV light region (254–395 nm). The dynamic range of I max / I 1 for our phototransistors may be smaller than that of some other devices operated with UV light of comparable power densities, which can be reasoned from the smaller photon energy of purple light. From Table 1 , it is clear that our ZTO phototransistor has good linearity ( α = ∼1) and a rather wide dynamic range, particularly on the I max − I 1 values, during photonic potentiation, and provides substantially more synaptic weight levels than those for previously reported synaptic phototransistors. The good linearity and wide dynamic synaptic weight range indicate that the photo-induced electrical current of our ZTO phototransistor will not saturate easily. It can be explained by the light-induced oxygen vacancy generation, which has been reported for the interaction of light with the ZnO surface. 29 Because the thickness of our ZTO film is only 5 nm, the light interaction with the ZTO surface will be dominant. The light-induced oxygen vacancies will thus replenish the neutral oxygen vacancies, which have been photo-ionized to generate photoexcited electron carriers. Consequently, a significantly high number of synaptic weight levels can be realized. The light potentiation and gate bias depression synaptic performances of ZTO phototransistors, therefore, are very promising for next-generation neuromorphic computing applications. It is worth noting that in addition to serving as the active layer ZnO can also be the charge trapping medium in the organic–inorganic hybrid transistor, which exhibited quadruple (400%) change in synaptic weight within 10 optical (UV light) pulses. 30"
} | 5,023 |
29323433 | null | s2 | 9,136 | {
"abstract": "To create life-like movements, living muscle actuator technologies have borrowed inspiration from biomimetic concepts in developing bioinspired robots. Here, the development of a bioinspired soft robotics system, with integrated self-actuating cardiac muscles on a hierarchically structured scaffold with flexible gold microelectrodes is reported. Inspired by the movement of living organisms, a batoid-fish-shaped substrate is designed and reported, which is composed of two micropatterned hydrogel layers. The first layer is a poly(ethylene glycol) hydrogel substrate, which provides a mechanically stable structure for the robot, followed by a layer of gelatin methacryloyl embedded with carbon nanotubes, which serves as a cell culture substrate, to create the actuation component for the soft body robot. In addition, flexible Au microelectrodes are embedded into the biomimetic scaffold, which not only enhance the mechanical integrity of the device, but also increase its electrical conductivity. After culturing and maturation of cardiomyocytes on the biomimetic scaffold, they show excellent myofiber organization and provide self-actuating motions aligned with the direction of the contractile force of the cells. The Au microelectrodes placed below the cell layer further provide localized electrical stimulation and control of the beating behavior of the bioinspired soft robot."
} | 347 |
39052827 | PMC11295035 | pmc | 9,137 | {
"abstract": "Significance Microbial communities of ecological and clinical importance are commonly found in interface-associated environments. The physical environment of interfaces may contribute to the development and behavior of surface-associated microbial communities. Here, we find that an actively expanding Bacillus subtilis colony grown at air–solid interface can spontaneously develop interface bulges; the bulging areas tend to enrich immotile cells via a diffusion-trapping mechanism and thus have a higher propensity to transit into a biofilm state. Our findings demonstrate the important role of active interfacial mechanics in bacterial collective behavior, providing a unique approach for colony biofilm patterning and directed self-assembly.",
"discussion": "Discussion In this work, we studied the mechanism and the biological function of active interface bulging in B. subtilis swarms. Combining multimode imaging, single-cell tracking, and numerical simulations, we showed that active interface bulging promotes biofilm formation at air–solid interface presumably via segregation and enrichment of sessile cells in the bulging area. In particular, the diffusivity of passive particles is ~50% lower inside the bulging area than elsewhere, which enables a diffusion-trapping mechanism and may account for the enrichment of sessile cells. We also uncovered a quasilinear relation between cell speed and surface-packing density that underlies the process of active interface bulging. Guided by the speed–density relation, we demonstrated reversible formation of liquid bulges by manipulating the speed and local density of cells with light. Taken together, our findings reveal a unique physical mechanism of biofilm formation at air–solid interface and driven self-assembly in active fluids. The liquid bulges are likely initiated by nonequilibrium density fluctuations ( Fig. 3 A ) ( 58 – 60 ), and they develop due to the increase of surface energy supported by activity-induced pressure. The liquid bulges we report here are distinct from those spatial patterns in bacterial swarms driven by nucleation of immotile cells ( 31 ) or by a mechanism reminiscent of MIPS ( 30 ), although all of these mechanisms yield high-density domains with dynamic and fluctuating boundaries. First of all, liquid bulges consist of almost equally motile cells compared to elsewhere and the liquid bulge formation does not require the presence of pre-existing immotile cells. By contrast, cells in dynamic clusters driven by the nucleation of immotile cells had a much lower speed than those outside the clusters (~80% speed reduction) ( 31 ), and the motility of cells in the MIPS-like clusters is also significantly lower than elsewhere due to density-dependent self-trapping ( 30 ). Second, liquid bulges have a constant (~20%) difference in surface-packing cell density compared to elsewhere, while the surface-packing density in clusters formed by nucleation of immotile cells or by MIPS-like process continuously increased. Finally, the route of further transition to sessile biofilm state is likely different. In our case, the transition from liquid bulges to sessile biofilm state is suggested to be seeded by the enrichment of sessile cells from surroundings via the unique mechanism of diffusion-trapping. Under our laboratory settings, the MIPS-like clusters were reproduced in swarms with higher cell densities and lower cell speeds, which may fall into the “dry” active matter regime where fluid flows and hydrodynamic interactions became irrelevant ( 69 ). However, in our case, the system is a “wet” active matter system, and the hydrodynamic interaction between cells and the deformable air–liquid interface is essential for second-layer formation. This property sets our system apart from the dry active matter systems, including myxobacteria colonies ( 70 ) and bacterial colonies grown in hard confinement (e.g., the interstitial space between an agar pad and a glass slide) ( 66 ). Note that these bacterial systems are dry only from the active matter perspective. In reality, all living cells must be residing in a fluidic milieu, and the fluidic nature of cell’s environment could play important roles even in the absence of hydrodynamic flow. For example, it was recently reported that myxobacteria cells accelerate when meeting the menisci of colloidal particles deposited on an agar surface, suggesting a role of capillary attraction in their gliding motility ( 71 ). Under our experimental conditions (0.6 to 0.8% Luria Broth agar), while the bulge regions transit into biofilm state earlier, other regions of the colony eventually enter the biofilm state as well. We did not see wrinkle structures often observed in B. subtilis colony biofilms grown on hard agar surface (e.g., 1.5% agar infused with MSgg medium like that used in ref. 72 ), presumably due to the difference in surface adhesion ( 73 , 74 ) or in cell death patterns ( 72 ) on substrates with different agar concentrations or different growth media. Nonetheless, we monitored the longer-term evolution of our colony biofilms and found an intriguing role of liquid bulges: A large-scale pattern of parallel ridges with high cell densities emerged in mature colony biofilms, and the position of these ridges lied in between the stripe-shaped bulges that existed earlier in the colony ( SI Appendix , Fig. S12 ). After the striped bulges transited into biofilm state, the striped biofilms expanded laterally; the formation of those high-density ridges followed the lateral expansion of the striped biofilms. We speculate that the mechanical stress during lateral expansion of striped biofilms would compress the cell layer in between, which then buckled and rose upward to form the ridges. This phenomenon may deserve further study, as it links interface bulging to the development of large-scale colony biofilm structures. The phenomenon of active interface bulging and the underlying mechanism reported in this study hold in generic terms, only requiring bacterial motility, a low interface tension, and a high volume density of cells (to enable sufficiently strong active stress and density fluctuations). Given the widespread flagellar motility and surfactant biosynthesis pathways ( 75 ) in the microbial world, we expect that active interface bulging is relevant to surface-associated bacterial colonization in many environmental and clinical settings ( 1 , 2 ), such as on the surface of fresh plant produce ( 76 ) and plant roots ( 54 ). Motile and sessile populations often coexist in bacterial communities. In complex and fluctuating environments, the transition between the two states may confer certain growth advantage to the population: Cells in sessile state are in general more tolerant to unexpected environmental stresses ( 77 , 78 ), while cells in motile state can explore more space and proliferate faster. Thus, the motile-to-sessile transition promoted by active interface bulging could potentially serve as a part of a bet-hedging strategy that enhances the adaptability of both in vitro and in vivo surface-associated bacterial communities ( 24 , 42 , 43 , 79 – 82 ). The ability to control biofilm patterns is essential for engineering bacterial living materials with desired functionalities ( 32 – 34 ). Informed by our study, we envisage that manipulation of cell motility and surface-packing density by either physical stimuli (e.g., light) or genetic control ( 83 ) may serve as a general means for biofilm patterning at air–solid interface via liquid bulge formation. Moreover, our results show that the liquid bulges present a physical environment that tends to segregate passive particles from surroundings; the segregation is driven by unbalanced particle fluxes due to the differential diffusivity inside and outside the bulges. As such, manipulating liquid bulge formation enables phase separation or segregation in binary mixtures that consist of passive particles and an active matter bath, providing a unique approach for directed self-assembly of microscopic structures ( 35 , 36 ) at interfaces."
} | 2,026 |
35755365 | PMC9219074 | pmc | 9,138 | {
"abstract": "Elastomeric surfaces\nand oil-infused elastic surfaces reveal low\nice adhesion, in part because of their deformability. However, these\nsoft surfaces might jeopardize their mechanical durability. In this\nwork, we analyzed the mechanical durability of elastic polydimethylsiloxane\n(PDMS) surfaces with different balances between elasticity and deicing\nperformances. The durability was studied in terms of shear/tensile\nice adhesion strength before and after different wear tests. These\ntests consisted of abrasion/erosion cycles using standard procedures\naimed to reproduce different environmental wearing agents. The main\nobjective is to evaluate if our PDMS surfaces can become long-lasting\nsolutions for ice removal in real conditions. We found that our elastic\nsurfaces show excellent durability. After the wear tests, the ice\nadhesion strength values remained low or even unaltered. Although\nthe oil-infused PDMS surface was the softest one, it presented considerable\ndurability and excellent low ice adhesion, being a promising solution.",
"conclusion": "5 Conclusions We examined the durability properties of\nthree types of PDMS-based\nelastic surfaces under abrasion and erosion, in terms of ice adhesion,\nthickness loss, and roughness modification. We found that, due to\ntheir deformability, these surfaces resist abrasion reasonably well,\nmaintaining a low ice adhesion strength after more than 4000 abrasion\ncycles. The icephobic performance is preserved until the coating thickness\nis low enough to influence ice adhesion. We found that the low ice\nadhesion values of the surfaces fabricated in this study are more\nlikely attributed to the bulk property rather than surface response.\nThe elastic coating preserved its properties although it was partly\ndamaged. For this reason, the interlocking effect seems to be absent\nin elastic surfaces above an elasticity degree. On the other hand,\nwe found that the surfaces become more wettable due to roughness increase. We also evaluated the resistance of the surfaces to erosion, and\nwe found that the erosion was low for most surfaces, especially those\nones with higher elastic modulus. In general, the thickness loss was\nlow, and the ice adhesion strength maintained low values. In our wear\nexperiments, the softest surface presented the fastest decrease in\ncoating thickness, although its ice adhesion strength was unmodified,\nbeing the lowest ice adhesion strength. In conclusion, moderate\nelastic surfaces presented good durability\nalthough their adhesion strength increased upon accumulating wear\nagents. Otherwise, the softest elastic surface presented the best\nresults of durability although their suitability for real applications\nwould require further studies conducted under more realistic conditions.",
"introduction": "1 Introduction For\ndesigning icephobic materials, an extended assumption is that\nwater-repellent surfaces might show a good anti-icing performance.\nDue to their poor affinity to water, superhydrophobic surfaces (SHSs)\nshould avoid or reduce ice accretion. SHSs are able to expel incoming\nwater drops after the impact, so drops may leave the surface before\nfreezing on it. 1 − 4 In addition, SHS could produce freezing delay and retard frost formation\ndue to the reduced contact area with the drops. 5 − 8 Moreover, SHS might reduce the\nice adhesion, but it was shown that in some cases, due to the interlocking\neffect (ice anchoring to the surfaces asperities), SHSs do not reduce\nice adhesion. 9 , 10 This is particularly true when\nice is formed on the surface under humid conditions, due to frost\n(or dew) formed between surface asperities. 10 − 13 This results in a significant\nreduction of the air trapped within the contact area after water freezing. 11 , 14 , 15 More importantly, SHSs show a\nlow durability due to their degradation during ice detachment. 12 , 13 A more recent alternative approach was the use of slippery liquid-infused\nporous surfaces (SLIPS) because they also revealed good liquid repellency,\ndelay frost formation, and reduce ice adhesion. 16 , 17 However, durability is again an issue on SLIPS, whose properties\ndepend on the stability of the top lubricant film, and it could deplete\nunder intensive use. 18 Rykaczewski et al. 19 found that ice formation could displace the\noil; this causes an increase in ice adhesion by mechanical interlocking,\nand surface durability is reduced. Moreover, SLIPS durability can\nalso be compromised by lubricant evaporation. 20 , 21 To address this issue, the stability of the lubricant film has been\nstudied and improved to enlarge durability. 21 − 24 Even if a given surface\nwas able to reduce ice formation, ice would\neventually appear under extreme conditions. Thus, the route can focus\non reducing the ice adhesion strength rather than avoiding the ice\naccretion. A balance between low ice adhesion and durability is getting\nmore attention and needs to be improved, as also stated in a recent\nreview. 25 It is accepted that a surface\nshows low ice adhesion when the detachment pressure is lower than\n100 kPa and super-low ice adhesion for values lower than 10–20\nkPa. 26 , 27 With these former values, ice is spontaneously\ndetached under natural forces such as wind, gravity, or ambient vibrations. More recently, new surfaces with promising anti-icing performance\nhave emerged. These surfaces are elastomeric, with low ice adhesion\nstrength due to their deformability. 28 − 30 In addition, some types\nof elastomeric surfaces combine the properties of SLIPS, such as oil-infused\npolymer matrix surfaces. 31 , 32 This combination improves\ndurability due to their self-repairing properties. 33 , 34 Moreover, certain elastic surfaces have shown resistance to ice\nformation–detachment cycles without noticeable increase in\nice adhesion strength. 29 , 33 , 35 However, surface deformability might rule mechanical durability.\nIn this work, we analyzed the mechanical durability of three elastic\npolydimethylsiloxane (PDMS) surfaces with very different elastic moduli\nand low ice adhesion strength. The ice adhesion strength of the surfaces\nwas evaluated in tensile and shear modes, before and after durability\ntests consisting of abrasion/erosion cycles using standard procedures\naimed to emulate different wearing conditions.",
"discussion": "4 Discussion In this study, we explored the durability\nproperties of several\nsoft coatings with low ice adhesion properties. The durability was\nstudied by analyzing how the wear tests modified the ice adhesion\nproperties of the coatings. Our results are in overall satisfactory,\nin comparison with previous studies. For example, Beemer et al. 43 presented PDMS gels that maintain their properties\nafter more than 1000 abrasion cycles by using a setup similar to this\nstudy, with similar grit number of the abrader (sandpaper grit 400)\nbut lower pressure (6.8 kPa). Similarly, Zhuo et al. 45 studied durability of their anti-icing materials by applying\n1.5 kPa and grit number 400. In the present work, the used pressure\nwas higher (20.5 kPa), and as shown in Figure 7 , the wear tests conducted with lower pressure\n(about 5.7 kPa) provided much lower thickness loss. In addition, we\nfound that the surface 1:2.50% maintained low ice adhesion, showing\nno evidence of the interlocking effect. This phenomenon is usually\nassumed as the origin of the ice adhesion increase observed after\nconducting the wearing tests. 43 , 45 In conclusion, our\nsurfaces showed better durability than other proposed solutions because\nthey resisted more cycles and under higher pressure. We further estimated\nthe resistance of our coatings through the thickness loss, resulting\nin large durability. On the other hand, we found that erosion\ndid not have a strong\nimpact on the samples. Nevertheless, we found that the dirt that was\naccumulated on surfaces during the erosion tests was hardly cleanable.\nIndeed, we observed that some dirt remained attached after the cleaning\nprocess (see Supporting Information Section\nS6). This might be a problem in real applications because the fabricated\nsurfaces may accumulate environmental dirt. However, we found that\nthe ice adhesion strength remains low, almost unaltered, for the three\nkinds of surfaces. Thus, the anti-icing performance of these surfaces\nis not much affected by erosion. In comparison to other harder\ncoatings proposed in the literature\n(such as aluminum coated with a thin film of fluoropolymer 60 ), which lose their wettability properties and\nconsequently the icephobic properties after few wear cycles, 42 the PDMS elastic materials proposed in this\nwork are able to maintain the icephobic properties after highly aggressive\nwear tests. This is a proof of the advantage that is added when using\ncoatings whose bulk properties instead of surface properties are relevant. 48 However, in our opinion, to compare meaningfully\nsurface resistance\nand durability the conditions should be harsher, such as higher pressure\nor more abrasive agents (like P60 sandpaper). Other studies have examined\ndurability under “presumably” higher pressures, 44 , 46 but none of these studies specified the pressure value, only the\nload applied. We explored the durability under higher pressure, and\nwe found that the surface 1:2.50% was destroyed after 120 cycles at\n110 kPa, while the other two surfaces resisted the abrasion (see Supporting Information Figure S8). Thus, the\nsurface 1:2.50% does not tolerate high pressure in abrasion. It would\nbe important to estimate the actual magnitude of abrasion that working\nsurfaces suffer under real conditions. We evaluated the resistance\nunder abrasion and erosion, but there\nare other types of damage in real world. For example, less cross-linked\nPDMS (far from the 1:10 ratio) reveals lower E and lower resistance\nto break under tension (ultimate tensile strength) but higher maximum\nelongation. 61 In consequence, although\nin our study, we report a great mechanical durability of PDMS surfaces,\nthey could be weak under other stresses. For this reason, a more complete\nevaluation would be necessary to establish the practical durability."
} | 2,517 |
25709712 | PMC4337199 | pmc | 9,139 | {
"abstract": "Background The microbial community in a biogas reactor greatly influences the process performance. However, only the effects of deterministic factors (such as temperature and hydraulic retention time (HRT)) on the microbial community and performance have been investigated in biogas reactors. Little is known about the manner in which stochastic factors (for example, stochastic birth, death, colonization, and extinction) and disturbance affect the stable-state microbial community and reactor performances. Results In the present study, three replicate biogas reactors treating cattle manure were run to examine the role of stochastic factors and disturbance in shaping microbial communities. In the triplicate biogas reactors with the same inoculum and operational conditions, similar process performances and microbial community profiles were observed under steady-state conditions. This indicated that stochastic factors had a minor role in shaping the profile of the microbial community composition and activity in biogas reactors. On the contrary, temperature disturbance was found to play an important role in the microbial community composition as well as process performance for biogas reactors. Although three different temperature disturbances were applied to each biogas reactor, the increased methane yields (around 10% higher) and decreased volatile fatty acids (VFAs) concentrations at steady state were found in all three reactors after the temperature disturbances. After the temperature disturbance, the biogas reactors were brought back to the original operational conditions; however, new steady-state microbial community profiles were observed in all the biogas reactors. Conclusions The present study demonstrated that temperature disturbance, but not stochastic factors, played an important role in shaping the profile of the microbial community composition and activity in biogas reactors. New steady-state microbial community profiles and reactor performances were observed in all the biogas reactors after the temperature disturbance. Electronic supplementary material The online version of this article (doi:10.1186/s13068-014-0182-y) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions The present study showed that similar steady-state process performances and microbial community profiles were achieved in three biogas reactors with the same inoculum and operational conditions, which suggested a minor role of stochastic factors in shaping the profile of the microbial community composition and activity in biogas reactor. Instead, temperature disturbance played an important role in the microbial community composition as well as process performance in biogas reactors. Increased methane yields (around 10% higher) and decreased VFAs concentrations at steady states were found in all three reactors after the temperature disturbances, although three different temperature disturbances were applied to each biogas reactor. New steady-state microbial community profiles were also observed in all the biogas reactors after the temperature disturbance.",
"discussion": "Results and discussion Reactor performance The monitoring profiles for methane yield, pH, and total VFAs in the three reactors are shown in Figure 1 , and the overall performances of the reactors at steady state are summarized in Table 1 . The initial higher methane yield was due to the methane production from the inoculum, since there are still organics in the inoculum which can be digested. After around 30 days’ operation, the methane yields were relatively stable. The steady-state methane yields for the three reactors were not significantly different ( P < 0.05): 194 ± 7.3, 189 ± 14.5, and 195 ± 6.9 mL/g volatile solids (VS) for reactors A, B, and C, respectively. Both the pH values (around 7.5) and the total VFAs concentrations were also similar for all three reactors. Acetate was the dominant VFAs, as seen in Figure 2 . The above results indicated that the replicate reactors (A, B, and C) did not present obvious differences in their performances. Figure 1 \n Reactor performances for the whole operational period. \n Table 1 \n Summary of the reactor performances at steady state before (phase I, 0 to 50 days) and after (phase II, 61 to 112 days) temperature disturbance \n \n Parameter \n \n A \n b \n \n A \n a \n \n B \n b \n \n B \n a \n \n C \n b \n \n C \n a \n Methane yield (mLCH 4 /gVS) 194 ± 7.3 220 ± 17.5 189 ± 14.5 213 ± 11.7 195 ± 6.9 214 ± 13.1 pH 7.50 ± 0.06 7.64 ± 0.02 7.51 ± 0.04 7.66 ± 0.01 7.52 ± 0.04 7.60 ± 0.02 Total VFAs(mM) 25.5 ± 3.9 3.3 ± 0.9 24.9 ± 1.7 5.3 ± 0.6 21.1 ± 1.9 7.8 ± 1.2 Subscript b means before temperature disturbance, subscript a means after temperature disturbance. Figure 2 \n Individual VFAs changes for the whole operational period. \n From day 50 (phase II), the temperatures of the reactors were changed (A 25°C, B 45°C, C 55°C) from the original temperature of 37°C. A sharp decrease in methane yields was observed in all the reactors, together with a decrease in pH and an increase in total VFAs. After 10 days at the new temperatures, the total VFAs increased to around 60 mM for reactor A, and to around 90 mM for both reactors B and C, which clearly indicates that the increase of temperature had a more profound effect on the stability of the reactors. There are several reasons leading to the higher VFAs accumulation when the temperatures were increased. It could be due to faster adaptation of acidogenic bacteria or to a greater temperature span of acidogens at higher temperatures compared to methanogens, as they generally grow more slowly and have a narrower temperature span [ 20 , 21 ]. Among the total VFAs, acetate was still the dominant component, although propionate was also accumulated, which is consistent with previous reports that propionate is easy to accumulate when the biogas reactor is disturbed [ 22 , 23 ]. From day 60 (phase III), the operating temperatures of all the reactors were changed back to 37°C. The methane yield increased immediately for reactor A. However, a slower recovery of the methane yield was observed for reactors B and C, which might indicate that the higher temperature disturbances (45°C and 55°C) had a more negative impact on the stability of the biogas reactors. In particular, reactor C took around 10 days before the methane yield increased to a similar level to that of the steady-state level in phase I. The fast increase of methane yield in reactor A was in good agreement with the fast decrease of total VFAs. During the steady states of phase III, the methane yields for reactors A, B, and C were 220 ± 17.5, 213 ± 11.7, and 214 ± 13.1 mL/gVS, respectively, and there were no significant differences ( P < 0.05) among the reactors. However, the methane yields were all significantly higher (around 10%) than those in phase I. The results indicated that the temperature disturbances affected the performances of all the reactors, although the operational conditions were the same for all the reactors in both phases I and III. The lower VFAs concentrations in phase III compared with those in phase I could explain the increased methane yields. However, the decreased VFAs concentration only accounted for 10% or less of the increased methane yield. This indicates that the increased methane yield could also be related with the increased hydrolysis of the solid part in cattle manure, which would lead to higher methane production. The lower VFAs concentrations in phase III for all the three reactors also resulted in the relatively higher pH (around 7.6). Microbial community analysis The numbers of sequences after quality filtration from different samples are shown in Additional file 1 : Table S1. The average sequence lengths were around 273 bp for all the samples. The high-quality sequences were assigned to taxonomic classifications by the Ribosomal Database Project (RDP) classifier. Since the primers used in the present study were universal primers, sequences belonging to both bacteria and archaea were obtained at the same time [ 24 , 25 ]. The phylogenetic classification of sequences assigned to bacteria from all the samples is summarized in Figure 3 . Samples A1, B1, and C1 had similar distributions of the sequences at the phylum, class, and genus levels. Firmicutes , Bacteroidetes , and Proteobacteria were dominant at the phylum level, and their dominance in biogas reactors was in accordance with other studies [ 26 , 27 ]. Clostridia , Bacteroidia , and Gammaproteobacteria were dominant at the class level. However, a considerable amount of the sequences (around 50%) were unclassified at the genus level, which could be due to some new microorganisms that have not yet been identified. The high percentages of unclassified sequences at the genus level were also found in previous studies [ 1 , 28 ]. The temperature disturbances had different effects on the shift of bacterial communities in reactors A, B, and C. The decrease of temperature from 37°C to 25°C in reactor A resulted in an increased abundance of Proteobacteria (A2), and the reason might be that some of the classes (for example, Gammaproteobacteria ) belonging to Proteobacteria can grow well at lower temperatures [ 29 ]. In reactor B, the increase of temperature from 37°C to 45°C led to an increased relative abundance of the unclassified sequences at the phylum level (B2). The increased relative abundance of Firmicutes (C2) was observed with a further increase of temperature from 37°C to 55°C in reactor C. The relative abundance of class Clostridia , belonging to Firmicutes , was enriched in sample C2, which could be due to their spore-forming character and gradual adaptation to thermophilic conditions [ 26 , 30 ]. The bacterial communities (A3, B3, and C3, sampled on day 60) continued to change after the temperatures in all the reactors returned to 37°C, which was consistent with the unstable reactor performances (Figure 1 , day 60). Samples A4, B4, and C4 were obtained during the steady states of phase III, and they had similar distributions at the phylum, class, and genus levels. However, compared with A1, B1, and C1, the relative abundance of Bacteroidetes in A4, B4, and C4 increased and that of Proteobacteria decreased. Differences at the class and genus levels were also observed between A1, B1, C1 and A4, B4, C4. The above results showed that although the biogas reactors, before and after the temperature disturbance, were run under exactly the same operational conditions, the bacterial communities did not return to the original bacterial composition. New steady-state bacterial community compositions, distinct from the original, were established after the temperature disturbances. Figure 3 \n Taxonomic classification of the bacteria communities. Phyla, classes, and genera making up less than 1% of total composition in all the samples were classified as Others. The phylogenetic classification of sequences assigned to archaea from all the samples is summarized in Figure 4 . The archaea mediating hydrogenotrophic and aceticlastic methanogenesis were found mainly within four orders ( Methanobacteriales , Methanococcales , Methanomicrobiales , and Methanosarcinales ) [ 31 ]. Therefore, only order- and genus-level classifications are shown in Figure 4 . Samples A1, B1, and C1 were all dominated by Methanomicrobiales and Methanobacteriales in phase I, which belonged to hydrogenotrophic methanogens. Similar distributions of samples A1, B1, and C1 at the genus level were also observed, and the dominant genera were Methanoculleus , Methanocorpusculum , Methanobrevibacter , and Methanobacterium . An increase of Methanobacterial es was found in all the reactors after temperature disturbance, which may indicate that the archaea belonging to this order were more resistant to the temperature disturbance (both downwards and upwards). It has been reported that the most frequently observed hydrogen utilizers are members of Methanobacteriales , present in both manure and sewage sludge digesters [ 31 ]. Further study is needed in order to understand why Methanobacteriales were more resistant to temperature changes than other methanogens. In reactor C, the temperature increase resulted in the increased relative abundance of order Methanosarcinales , which are mainly aceticlastic methanogens. Since the methane production during temperature shock was significantly reduced, the changes of archaeal communities during the temperature disturbance might be due to the different decay rates of the archaea rather than the different growth rates of the archaea. After the temperature was changed back to 37°C, Methanosarcinales became dominant in all the reactors (A3, B3, and C3). At the steady states of phase III, Methanomicrobiales , Methanosarcinales , and Methanobacteriales were all dominant in reactors A, B, and C. The genus Methanosarcina , belonging to Methanosarcinales , mainly mediates aceticlastic methanogenesis, and therefore it is expected that the methanogenic pathway was changed before and after temperature disturbance. The dominance of Methanosarcina might be related to the better reactor performances in phase III compared with phase I. It has been reported that the dominance of hydrogenotrophic methanogenesis is always related to extreme conditions such as high ammonia or acetate concentration [ 32 - 34 ]. It is possible that the higher acetate concentration in phase I in all the reactors induced the dominance of archaea belonging to hydrogenotrophic methanogens. Figure 4 \n Taxonomic classification of the archaea communities. \n The differences between the microbial communities of different samples were further assessed by principal coordinates analysis (PCoA) and hierarchical cluster analysis. The results from PCoA are shown in Figure 5 . Principal components 1 and 2 explained 31.6% and 26.3% of the total community variations, respectively. A1, B1, and C1 were clustered together, and were well separated from the group of A4, B4, and C4. The results were consistent with the taxonomic analysis that steady-state microbial community compositions were changed after temperature disturbances in reactors A, B, and C, which could also be used to explain the different steady-state reactor performances in phase I and phase III, as discussed in the preceding section. A2 was close to A1, while B2 and C2 were far away from B1 and C1, which indicated that the increase of temperature to 45°C or 55°C in the biogas reactor induced significant changes in the microbial communities. A3, B3, and C3 were closer to A4, B4, and C4, which suggested that the microbial communities gradually changed to the new steady states after the temperature changed back to 37°C. The hierarchical cluster analysis (Additional file 1 : Figure S1) also clustered A1, B1, and C1 as one group and A4, B4, and C4 as one group, which further supported the cluster results from PCoA. Figure 5 \n PCoA of all the samples. \n Microbial community assembly and function in anaerobic digestion of cattle manure The results in the present study clearly showed that replicate biogas reactors treating cattle manure had similar steady-state reactor performances under the same environmental conditions. The replicate biogas reactors also had similar microbial community compositions based on RDP classification, PCoA, and hierarchical cluster analysis. However, Zhou et al. found that, under the same environmental conditions, both bacterial community compositions and functions in replicate microbial electrolysis cells were different, and they proposed that stochastic assembly played a dominant role in determining not only community structure but also ecosystem functions [ 35 ]. Based on our results, the microbial community compositions and functions in anaerobic digestion of cattle manure were not obviously affected by stochastic assembly. It seems that stochastic assembly played different roles in different ecosystems. A recent publication, which found that community history affects the predictability of microbial ecosystem development, might explain the differences between our results and Zhou’s results. Pagaling et al. [ 36 ] demonstrated that the final community composition and function are unpredictable when the source communities (inoculum) colonize a novel environment, while the community development is more reproducible when source communities are pre-conditioned to their new habitat. In our study, the inoculum was obtained from a biogas reactor, which was adapted to the anaerobic condition for methane production. Therefore, the steady-state microbial community composition and reactor performance were reproducible in the replicate biogas reactors. However, all the reactors in Zhou’s study were inoculated with wastewater not pre-conditioned in the microbial electrolysis cell reactors, which might have been the reason for the variation of microbial community compositions and functions observed. Further investigation is imperative to elucidate whether other inocula sources (sewage sludge, soil, and others), which are not derived from anaerobic reactors, would lead to different stable-state microbial communities and reactor performances in biogas reactors under the same operational conditions. Although the effect of temperature disturbance on biogas production has been studied before [ 13 , 16 ], the previous studies only focused on the reactor performances. The rapid recovery of biogas production after the temperature disturbance was also previously observed [ 13 ], but in the cited study the biogas production reached the same level as that prior to the disturbance. We are the first to report the increased biogas production after temperature disturbance. The difference in biogas production recovery levels could be due to the fact that the substrate was already efficiently degraded before temperature shock in Chae’s study [ 13 ]. Our results suggested that temperature disturbance could be used as a strategy to stimulate biogas production in the biogas reactor, where the biogas production was not so efficient (possibly due to lower HRT). In a previous study, Nielsen et al. investigated the effects of disturbance of oleate on the performance of a biogas reactor treating cattle and pig manure, and they also found that a lower VFAs concentration along with a higher methane production were achieved in the biogas reactor after the disturbance of oleate [ 15 ]. However, they could not link the microbial community composition with the reactor performance due to the lack of microbial analysis. Our results combined with the above literature suggest that the idea that different types (temperature, oleate) of disturbances might have similar stimulation effects on the biogas process, and the effects of other disturbances (ammonia, organic loading shock, and others) on the community assembly and functioning of the biogas reactor deserve to be further investigated. Although in the present study, three different temperature disturbances were investigated, the establishment of new steady-state microbial community compositions in all the reactors after temperature disturbance was observed. More importantly, the three new steady-state microbial community compositions were clustered together and were clearly distinguished from the steady-state microbial community compositions before the temperature disturbances. A comparison of steady-state microbial community compositions before and after temperature disturbance showed that the temperature disturbance played an important role in the microbial community assembly and ecosystem function during the anaerobic digestion of cattle manure."
} | 4,941 |
22169400 | null | s2 | 9,143 | {
"abstract": "Chemotaxis allows bacteria to follow gradients of nutrients and other environmental stimuli. The bacterium Escherichia coli performs chemotaxis via a run-and-tumble strategy in which sensitive temporal comparisons lead to a biased random walk, with longer runs in the preferred gradient direction. The chemotaxis network of E. coli has developed over the years into one of the most thoroughly studied model systems for signal transduction and behavior, yielding general insights into such properties of cellular networks as signal amplification, signal integration, and robustness. Despite its relative simplicity, the operation of the E. coli chemotaxis network is highly refined and evolutionarily optimized at many levels. For example, recent studies revealed that the network adjusts its signaling properties dependent on the extracellular environment, apparently to optimize chemotaxis under particular conditions. The network can even utilize potentially detrimental stochastic fluctuations in protein levels and reaction rates to maximize the chemotactic performance of the population."
} | 273 |
35832889 | PMC9272384 | pmc | 9,144 | {
"abstract": "Summary Plant-derived biomass is the most abundant biogenic carbon source on Earth. Despite this, only a small clade of organisms known as white-rot fungi (WRF) can efficiently break down both the polysaccharide and lignin components of plant cell walls. This unique ability imparts a key role for WRF in global carbon cycling and highlights their potential utilization in diverse biotechnological applications. To date, research on WRF has primarily focused on their extracellular ‘digestive enzymes’ whereas knowledge of their intracellular metabolism remains underexplored. Systems biology is a powerful approach to elucidate biological processes in numerous organisms, including WRF. Thus, here we review systems biology methods applied to WRF to date, highlight observations related to their intracellular metabolism, and conduct comparative extracellular proteomic analyses to establish further correlations between WRF species, enzymes, and cultivation conditions. Lastly, we discuss biotechnological opportunities of WRF as well as challenges and future research directions.",
"conclusion": "Conclusions and future perspectives In nature, WRF are continuously adapting to nutritional availability over their life cycles, which demands that optimal enzyme cocktails are employed to obtain carbon for energy and growth in a variety of environmental conditions, as well as coping with environmental and systemic toxicity. As shown in a variety of examples in this review, the processes used to address each of these issues cannot be investigated as individual modules. To understand the mechanisms that WRF employ to efficiently degrade lignocellulose and xenobiotics, or to produce fruiting bodies, among other applications, requires a holistic systems biology approach and more fundamental knowledge of triggered intracellular mechanisms from regulatory processes to specific pathways and enzymes. Nonetheless, a genetic toolkit for WRF is essential to validate systems biology observations, gene-function relationships, and further improve their performance. Progress on the development of efficient genetic tools (i.e., CRISPR/Cas9) has been recently published for few species of WRF such as P. ostreatus ( Boontawon et al., 2021 ), G. lucidum ( Wang et al., 2020 ), and D. squalens ( Kowalczyk et al., 2021 ), which is a significant advance in fungal biology and proves the suitability of WRF for further metabolic engineering efforts. Despite the vast benefits of systems biology methods to understand biological processes, there are still some limitations that remain: • First, gene functional annotation is, in many cases, inaccurate (i.e., prediction of introns in eukaryotic systems), and between 40 and 50% of the proteins are hypothetical and lack functional predictions ( Meyer et al., 2020 ). Developing high-throughput platforms for enzyme production and evaluation on a diversity of substrates would be essential to improve these annotations. • Second, to build a robust model for degradation and bioconversion scenarios by WRF, numerous datasets from a higher diversity of systems biology methods would be required due to the nature of some non-specific and non-enzymatic driven reactions. Even though the throughput of -omic analyses has significantly improved over the past few years, the generation of large datasets is still a challenge due to the strong fragmentation in the community studying particular species of filamentous fungi ( Meyer et al., 2020 ). This fragmentation is not only notable on the selection of WRF but also on the substrates (in applications that involve lignocellulose), which are many times selected based on the main biomass sources in each region. Despite these challenges, there is currently a higher awareness on the need of generating detailed metadata with experimental information and sharing raw data with the community, which will be key to continue conducting integrative and comparative -omic analyses. • Third, metabolomic studies are an essential component in multi-omic studies and have allowed proposing aromatic catabolic pathways in WRF ( del Cerro et al., 2021 ), but as described here, these studies are emerging in research with WRF ( Figure 2 ). Furthermore, metabolomic analyses become particularly complex in laboratory cultivation conditions that mimic those found in nature (i.e., solid-state cultivations, Figures 1 B and 1E). Namely, WRF growing on solid materials could not be easily isolated from the material (e.g., lignocellulose, lignin, plastics). Thus, sample collection protocols need to be developed to distinguish intracellular from extracellular metabolites, especially in research studies that aim to elucidate metabolic pathways. • Lastly, due to the compartmentalization in eukaryotic cells, certain biochemical steps can occur in multiple subcellular locations ( Figure 3 ), which is another constraint to be considered for metabolic modeling. Multi-omic approaches in individual organelles and the compartmentalization of model metabolic networks would further advance the understanding of carbon flux in WRF. To select robust WRF for biotechnological applications, bioprocess development and scale-up would also need to be investigated in parallel toward building relevant industrial processes and identifying shortcomings both in the process and the fungal species. Each application will need specific cultivation conditions; for instance, if the application involves polymer degradation processes, WRF are more efficient breaking down polymers in the solid-state ( Figure 1 E) than submerged-state ( Figure 1 D) cultivation mode. Solid-state cultivation technologies have not been as well developed as submerged cultivations at large scales. Some operational parameters such as substrate and particle size, inoculum, nutrient supplementation, aeration, temperature, moisture content, pH, and mixing are common in both solid-state and submerged cultivations. Nevertheless, solid-state cultivations are highly exothermic, and aeration is essential to not only provide oxygen to the organisms, but also dissipate the heat and moisture generated during growth ( Mansour et al., 2016 ). Despite the aeration, it is likely that temperature gradients will be observed in the cultivation. Thus, WRF able to tolerate different temperature gradients without significant performance variations will be preferred. Solid-state cultivations may be conducted in trays or rotating drums, in the presence or the absence of mixing ( Mansour et al., 2016 ). The incorporation of mixing will also depend on the robustness of the fungal mycelium. An optimal bioreactor design will also need to allow easy diffusion and extraction of metabolites for consolidated bioprocessing ( Olson et al., 2012 ), for which WRF are promising biocatalysts. The slow lignocellulose degradation rates by WRF may be considered detrimental for industrial processes. However, it is worth noting that the volumetric productivity in solid-state cultivation can be significantly higher (∼10 times for enzyme production, for example) compared to submerged fermentations ( Durand, 2003 ). Techno-economic analyses and life cycle assessments will be required to assess the feasibility of these processes and inform about cost-effective strategies to scale-up WRF cultivations. In conclusion, this review aims to encourage the global research community to address the open questions presented here to ultimately harness WRF as future biocatalysts.",
"introduction": "Introduction Carbon from terrestrial plants accounts for 450 gigatons of the total 550 gigatons of bio-based carbon on the planet, with the latter including all kingdoms of life ( Bar-On et al., 2018 ). Most of the plant-derived mass resides in their cell walls, which primarily comprise two polysaccharides, cellulose and hemicellulose, and an aromatic polymer, lignin, with minor components such as pectin, suberin, and cutin. The dry weight concentration of each component varies between plant types but, in general, cellulose is the most abundant biopolymer (∼50%), followed by lignin (∼15–40%) and hemicellulose (∼10–35%) ( Chundawat et al., 2011 ; Martínez et al., 2005 ; Ragauskas et al., 2014 ). Cellulose is a homopolysaccharide consisting of β-1,4 linked D-glucose monomers whereas hemicellulose is a heteropolymer composed of pentose and hexose sugars, which are often acetylated. Lignin is a heterogeneous aromatic polymer which includes phenylpropanoid units that differ in their extent of ring methoxylation and are linked by a variety of C–C and C–O bonds, along with other phenolic compounds (e.g., tricin) that have been also recently described as lignin monomers ( del Río et al., 2020 ). The structural complexity of lignin confers defense against pathogens and mechanical strength to the plant cell walls ( Martínez et al., 2005 ). Although many bacteria and fungi are able to depolymerize cellulose and hemicellulose in nature, white-rot fungi (WRF) are the only clade of known organisms that have evolved to efficiently depolymerize and mineralize the lignin polymer to CO 2 and H 2 O ( Floudas et al., 2012 ). WRF are saprotrophic organisms that are classified in the fungal kingdom, Dikarya subkingdom, Basidiomycota phylum, and Agaricomycotina subphylum ( Figure 1 A). This subphylum is divided into four classes and 18 orders ( Grigoriev et al., 2011 , 2014 ). WRF can be found within the orders Agaricales, Auriculariales, Hymenochaetales, and Russulales, among others ( Figure 1 A). However, most WRF belong to the order Polyporales ( Martínez et al., 2018 ). All of these orders include a variety of fungi apart from WRF, some of which are also involved in the degradation of plant-derived biomass (lignocellulose), namely brown-rot fungi (BRF). WRF and BRF utilize different mechanisms to degrade lignocellulose ( Riley et al., 2014 ; Schilling et al., 2020 ). WRF degrade all plant cell wall components via extracellular, enzymatic, and non-enzymatic (e.g., reactive oxygen species) mechanisms. Subsequently, WRF obtain carbon from sugars and, as recently suggested, from lignin-derived aromatic compounds ( del Cerro et al., 2021 ). This type of degradation usually causes an enrichment in cellulose and generates a white decay residue ( Martínez et al., 2005 ). Conversely, BRF modify, but do not degrade, the lignin polymer via extracellular, non-enzymatic chemistries to increase the exposure of polysaccharides for successful enzymatic depolymerization and utilization of sugars as carbon sources. The enrichment of lignin results in a brown decay residue ( Martínez et al., 2018 ; Schilling et al., 2020 ). Regarding the enzymatic mechanisms, carbohydrate active enzymes (CAZymes) are typically secreted by both WRF and BRF to cleave glycosidic bonds ( Drula et al., 2022 ). However, WRF produce additional enzymes that belong to the class of ‘auxiliary activities (AA)’, which act in conjunction with CAZymes and are involved in redox processes and lignin breakdown, such as laccases (subfamily AA1_1) and class II heme-containing peroxidases (a subset of the AA2 subfamily) ( Drula et al., 2022 ). WRF can be further divided into two groups based on the degradation patterns – those that selectively degrade lignin and those that degrade all lignocellulosic components simultaneously ( Fernández-Fueyo et al., 2012 ). Figure 1 White-rot fungi (WRF) distribution in the fungal kingdom and examples of a WRF growing in different environments and experimental conditions (A) Simplified diagram of the fungal kingdom with branches that include WRF with the number of published genomes in parentheses at the time of writing (the species are detailed in Table S1 ). (B) Fruiting bodies of Trametes versicolor (commonly known as turkey tail) growing on woody biomass. (C–E) Mycelia of T. versicolor growing (C) on a yeast extract-peptone-dextrose (YPD) agar plate, (D) in YPD broth under submerged-state and agitation cultivation conditions, and (E) on milled poplar in solid-state cultivation conditions (defined as cultivations in the absence of free water). Based on the unique features of WRF, these organisms and their extracellular enzymes have long been realized to exhibit promise in biotechnological applications involving, among others, conversion of both lignocellulose and lignin-rich substrates ( Camarero et al., 2014 ). However, the full potential of WRF as biocatalysts for the conversion of carbohydrates and lignin-derived aromatic compounds to value-added fuel and chemical precursors – as commonly demonstrated in the metabolic engineering field for bacteria, yeasts, and filamentous ascomycetes ( Lee et al., 2019 ) – is still uncertain. This is, in part, due to the lack of efficient genetic tools in WRF and the limited information about their intracellular metabolism. Indeed, neither a comprehensive metabolic map or carbon flux data have been described yet, in either native environments or in varied laboratory conditions, where fungi can display a diversity of morphologies ( Figures 1 B–1E). To that end, systems biology methods can substantially accelerate the development of biocatalysts via (1) the elucidation of metabolic pathways, regulatory mechanisms, and metabolic fluxes, (2) the discovery of enzymes, (3) the construction of metabolic networks, and (4) the generation of computational models to predict biological outputs. Modern systems biology comprises a large number of methods (e.g., genomics, transcriptomics, proteomics, metabolomics, lipidomics, fluxomics, glycomics, epigenomics, etc.) ( Veenstra, 2021 ), but to date, the main approaches utilized to study WRF have been primarily limited to genomics, transcriptomics, and extracellular proteomics. Very few systems biology studies of WRF have incorporated intracellular proteomics or metabolomics ( Figure 2 ). Reviews or comparative-omic analyses that include WRF have specifically focused on genomics in basidiomycetes ( Lundell et al., 2014 ), genomics in the order Agaricales ( Ruiz-Dueñas et al., 2020 ), genomics in the order Polyporales ( Hage et al., 2021a ), transcriptomics in basidiomycetes ( Peng et al., 2018 ), extracellular proteomics in basidiomycetes ( Alfaro et al., 2014 ), and extracellular proteomics in lignocellulose-degrading bacteria and fungi ( Guo et al., 2018 ). Here, we aim to review the breadth of systems biology studies conducted exclusively in WRF to date. First, we highlight observations that relate to both the extracellular and intracellular metabolism of WRF from a genomic, transcriptomic, and proteomic perspective. We note that most of the -omic studies in WRF utilize lignocellulose or lignocellulose-derived compounds in the cultivation media and thus, we will focus on these publications. However, it is worth mentioning that several-omic studies have been also conducted for other purposes in WRF (e.g., to describe the degradation of xenobiotic dyes ( Sun et al., 2017 ) or interspecific fungal interactions ( Luo et al., 2017 )). This review does not include a metabolomics section because metabolomic analyses are emerging and are scarce in research with WRF ( Daly et al., 2018 ; del Cerro et al., 2021 ; Matsuzaki et al., 2008 ; Miura et al., 2004 ). Second, within the proteomics section, we show a comparative analysis from over 90 independent extracellular proteomic datasets to establish correlations among fungal species, secreted enzymes, and cultivation conditions. Third, we provide an overview of the biotechnological opportunities of WRF in addition to lignocellulose bioconversion and lastly, we present a series of conclusions and perspectives related to future research in WRF. Figure 2 Cumulative number of WRF genomes published over time and transcriptomic, proteomic, and metabolomic studies of WRF conducted on lignocellulose or lignocellulosic-derived substrates The references used to generate these graphs are shown in Tables S1 and S2 ."
} | 3,985 |
35515442 | PMC9054060 | pmc | 9,147 | {
"abstract": "Stretchable superhydrophobic film was fabricated by casting silicone rubber polydimethylsiloxane (PDMS) on a SiO 2 nanoparticle-decorated template and subsequent stripping. PDMS endowed the resulting surface with excellent flexibility and stretchability. The use of nanoparticles contributed to the sustained roughening of the surface, even under large strain, offering mechanically durable superhydrophobicity. The resulting composite film could maintain its superhydrophobicity (water contact angle ≈ 161° and sliding angle close to 0°) under a large stretching strain of up to 100% and could withstand 500 stretching–releasing cycles without losing its superhydrophobic properties. Furthermore, the obtained film was resistant to long term exposure to different pH solutions and ultraviolet light irradiation, as well as to manual destruction, sandpaper abrasion, and weight pressing.",
"conclusion": "4 Conclusions The template method usually requires complex technological processes and specialized equipment. Here, we have adopted a modified template to prepare stretchable and durable PDMS/SiO 2 superhydrophobic composite film. It was shown that the composite film, consisting of a PDMS elastomer, created superhydrophobic surfaces that were resistant to sandpaper abrasion, UV irradiation, chemical etching, and man-made friction. Importantly, the film with a groove-bulge structure has strong resistance to long term pressure. The modified template method might pave a new way to the fabrication of stretchable, anti-wetting materials.",
"introduction": "1 Introduction Superhydrophobic surfaces with a water contact angle (CA) greater than 150° and a sliding angle (SA) smaller than 10° have attracted considerable attention in the past few decades, 1–5 with the SA being analyzed by the Wolfram approach. 6 Inspiration from natural non-wetting structures, particularly the leaves of lotus plants 7,8 , has led to numerous applications in self-cleaning, 9–11 dragging reduction, 12,13 oil–water separation 14–16 and corrosion resistance. 17–19 Representative examples of non-wetting prototypes in nature not only include the leaves of lotus plants, but also pond skater legs 20,21 and butterfly wings. 22 Research shows that such surfaces largely depend on hierarchical roughness with micro- and nanoscale structures and low-surface-energy components. 23–26 Over the last few decades, a large number of studies relating to durable superhydrophobic materials have been reported. 27–32 Many methods have been dedicated to the production of stable and durable superhydrophobic material, which can be divided into two categories. One method involves integrating a self-healing ability into superhydrophobic materials, including relocating of hydrophobic components or recombining topographic structures, 33–40 resulting in the restoration of their superhydrophobicity. Due to the limited number of hydrophobic components, the repairing cycles in this approach are limited. Furthermore, the self-healing process requires certain specific conditions in terms of light and temperature, and time (from tens of minutes to hours). 37,38,40 Hence, the application of this approach is limited. The second method is to use expensive low-surface-free energy compounds. Fluorine-containing low surface-energy materials are used to improve the stability of superhydrophobic materials. 41,42 However, fluorinated compounds have been identified as a source of toxic pollutants, which are harmful to the environment and increase the chances of many intractable diseases ( e.g. , cancer). Therefore, it is necessary to develop a new method to fabricate durable superhydrophobic materials. Most such reports are on rigid substrates or flexible substrates with low deformation ability. The micro-nanostructures on the surface are mostly fragile and easily lose their anti-wetting characteristics under large mechanical deformation (>50%). 43,44 Hence, maintaining superhydrophobicity under large mechanical deformation still poses considerable challenges. Hard substrates have been replaced by highly stretchable substrates to fabricate durable superhydrophobic materials. 45–47 Furthermore, due to rapid developments in the stretchable electronic industry, superhydrophobic surfaces based on flexible substrates have become candidates to meet the demands of modern industry. Mates et al. achieved a conductive and stretchable superhydrophobic material by spraying carbon nanotubes-Parafil-M (a commercial paraffin wax–polyolefin thermoplastic blend) composite on natural rubber. 47 Cho et al. prepared a gas-breathable superhydrophobic membrane with stretchability via three steps, including electrospinning, construction of micro-nano roughness, and hydrophobization. 44 Although these results were encouraging, and the superhydrophobic characteristic could be maintained during deformation, these coatings are easily exfoliated from the substrate during relaxation because of weak adhesion between the substrate and the coating. Possible strategies to resolve these issues might include increasing cross-linking sites of coatings, and maintaining the substrate and roughening structures as a monolithic entirety. The template method is a simple method to create roughened structures for superhydrophobic surfaces. 48,49 However, natural templates with rough structures from nature, such as lotus leaves, are not usually uniform due to the stems on the leaves, and do not easily allow the fabrication of materials with a large area. Furthermore, man-made master templates usually need special equipment to create roughened structures. In this study, we adopted a modified template to prepare a superhydrophobic composite film from a polymer-based shrink film which is easy to process using heat treatment. We prepared polydimethylsiloxane (PDMS)/SiO 2 composite films by spraying SiO 2 nanoparticles on a shrunken template with a micron hierarchical structure, and casting and curing the PDMS on the template, followed by stripping the composite film, as shown in Fig. 1 . The film retained superhydrophobicity after tolerating ultraviolet (UV) irradiation for 150 h, sandpaper abrasion over 6 m with 100 g loading weight, weight pressing of 2000 g for 80 h, as well as immersion in acid (pH = 1), alkali (pH = 13) or organic solvents for 60 h. In addition, the as-prepared superhydrophobic composite films retained their anti-wettability under different strains, repeated stretching, and manual friction. Fig. 1 A schematic illustration of the procedure used to prepare superhydrophobic PDMS/SiO 2 composite film.",
"discussion": "3 Results and discussion 3.1 Characterization of the PDMS/SiO 2 composite film Surface microtopography has large influence on the wetting properties of a superhydrophobic surface. From Fig. 2a , it can be seen that the side-facing shrunk film became superhydrophobic with a micro-nano structure. The hierarchical surface displayed a typical Cassie state with a water CA up to 161° and an SA close to 0°. In this experiment, the dosage of nanoparticle dispersion sprayed on shrunk film is the most important factor relating to the obtained samples. The shrink film transformed from plain to highly wrinkled after heating (Fig. S1 † ). The corresponding superhydrophobic PDMS film is metastable, and the CA of the film reduced to less than 150° after 5 min (Fig. S2 † ). As shown in Fig. 2b , the hierarchical roughness consisted of a groove-bulge structure. On the microscale, a micro-texture was formed by the shrunk film with micron hierarchical structure. On the nanoscale, densely distributed nanoparticles appeared on each micro-groove and bulge ( Fig. 2c and d ). The high micro/nanoscale roughness created by well-designed micro-texture and the abundant nanoparticles provided sufficient room for “air pockets” between the patterned solid and water layer to form a Cassie–Baxter state. Fig. 2 (a and b) SEM images of superhydrophobic PDMS/SiO 2 composite film at different magnifications, showing nano-textured grooves (c) and bulges (d) at high magnification. 3.2 Mechanical stability It is well known that the robustness and stability of superhydrophobic surfaces under the rubbing processes play a significant role in practical applications. To assess the robustness of our superhydrophobic surface, manual destruction and sandpaper abrasion were implemented to demonstrate the anti-wear properties of the superhydrophobic composite film. In the manual destruction, the surface was rubbed with two fingers back and forth (see the Video S1 in the ESI † ). Fig. 3a shows screenshots of the Video S1. † After friction by hands, the superhydrophobic surface still retained water rolling properties. The water droplet flattened at the moment of contacting the film, then it bounced on the composite film without wetting or contaminating the surface (see the Video S2 in the ESI † ), suggesting stable Cassie–Baxter state on the superhydrophobic surface. Here, resistance against mechanical damage was evaluated by abrading the sample using sandpaper to quantitatively evaluate the mechanical durability of the superhydrophobic surface. 50 The sandpaper test was carried out by loading a weight of 100 g onto the sample ( Fig. 3b and c ) and moving it across the sandpaper along a ruler. Fig. 3d demonstrates the changes in CA and SA as a function of the number of abrasion cycles. The superhydrophobic surface showed no obvious decay in CA and SA after the first three cycles. The CA decreased from 161.3 ± 1.2° to 153 ± 1.4°, and the SA of superhydrophobic surface was maintained at less than 5°, even after 20 abrasion cycles. As the number of abrasion cycles increased, the roughness of this superhydrophobic surface decreased, which may cause the loss of superhydrophobicity. Comparison of the results of the anti-abrasion test with those reported previously further confirm the excellent mechanical properties of the as-obtained superhydrophobic surfaces (Table S1 † ). Fig. 3 (a) The process of producing man-made friction. (b and c) One cycle of sandpaper abrasion for superhydrophobic PDMS/SiO 2 composite film. (d) CAs and SAs of the superhydrophobic PDMS/SiO 2 film over 0–20 abrasion cycles on 240# sandpaper. In addition to surface abrasion damage, the mechanical stability of the superhydrophobic film under stretching strain is extremely crucial. As shown in Fig. 4a , all CAs were still larger than 160° and SAs remained below 1° with increasing stretching strain to 100% from 0%. Fig. 4b show that a water droplet can maintain its spherical shape under different strains, indicating that the Cassie–Baxter state between the water and film was retained. Furthermore, the superhydrophobicity of the composite film could remain when it was returned to its initial position, and the film showed no deformation or damage with strain of 50% (see Video S3 in the ESI † ). Fig. 4 shows the CAs and SAs of the superhydrophobic film under the stretching–relaxing procedure with a strain of 100%. It was found that the CAs of the film were still larger than 160°, whereas the SAs were still below 1° after the 500 cycles of stretching–releasing, indicating excellent adaptability to withstand mechanical cycles. The inset images of the water droplets before and after 500 cycles of the stretching–relaxing test demonstrate little difference. One cycle of tensile test is shown as Fig. 4d . Fig. 4 (a) CAs and SAs of the superhydrophobic PDMS/SiO 2 film under different strains. (b) A water droplet can maintain its spherical shape under different strains. (c) CAs and SAs are over 155° and below 1° over numerous cycles with a strain of 100%. (d) Optical photographs of one cycle, the end cycle of tensile testing. Generally, the loss of superhydrophobicity under large mechanical deformation is due to the transition of the Cassie–Baxter state to the Wenzel state. Thus, it is necessary to understand the state of our composite films under various strains. Here, we designed a simple model to analyze the wetting mechanisms. Fig. 5a depicts the structural deformation of hierarchical structure under stretching. We assume that the space distance between two micromastoids are grooves. When the superhydrophobic film was stretched, the grooves on the film became closely aligned parallel to the strain direction from random directions ( Fig. 5b ). Interestingly, the reorientation of the texture under stretching occurred only in the grooves ( Fig. 5c ). However, the structures on the microbulges were preserved under stretching ( Fig. 5d ). Microscale open air pockets remained between microbulges, and nanoscale sealed air pockets between nanomastoids on the microbulges. The open air pockets are connected to the atmosphere, contributing little to surface adhesive force and causing a high CA on the rough solid surfaces. All these changes still met the requirements of the Cassie state, and the composite film remained superhydrophobic. Fig. 5 A partial hydrophobic model of structure deformation under 50% tensile strain. (b–d) SEM images of stretched film (50% tensile strain); (b) and (d) showed zoomed in images of grooves and bulges from (c). In addition to the static parameters providing some information about the wetting properties, superhydrophobic surfaces show remarkable dynamic properties under the impact of droplets, which is important for the stability of Cassie–Baxter state. Herein, the behavior of a droplet impinging onto the surface of the relaxed film was recorded from side views (see the Video S4 in the ESI † ). After contact with the surface, the droplet deformed first and flattened into the shape of a pancake, but then recoiled and finally rebounded off the surface without penetrating the nanostructure. The droplet rebounded completely, which suggests that a stable Cassie–Baxter state existed at the moment of impact. Superhydrophobic bouncing (see Video S5 in the ESI † ) was maintained, even after mechanical stretching under 50% uniaxial strain. The almost complete rebound occurred on the stretched superhydrophobic surface when the dropping height of the water droplet increased (see Video S6 in the ESI † ). We concluded that the behavior of water droplets dropping on stretched surfaces was nearly identical to that on hierarchical surfaces before stretching. The similar behavior of a droplet falling on the films was due to the stable Cassie–Baxter state of superhydrophobicity. In practical applications, the pressure resistance of the film should be taken into consideration. Therefore, we evaluated the pressure resistance by placing a weight of 2000 g onto the sample, with the hierarchical side touching the weight. 51 The pressing test model is shown in the inset of Fig. 6a . From Fig. 6a , we can see that CA and SA only changed a little after the first 10 h of pressing. The CA of the PDMS/SiO 2 composite film decreased slightly, from 161.6 ± 1° to 159.2 ± 1.6°, and the SA increased to 1° after pressing for 40 h. When the pressing time reached 80 h, the CA decreased to about 150°, with SA still below 10°, showing the persistent superhydrophobicity of the composite film. The micromastoids on the surface became flat with increasing time, which lead to a reduction of roughness. Nevertheless, the nanostructures inside the grooves still remained ( Fig. 6b ). The composite film remained water-repellent, and the hierarchical structure of the composite film still retained the Cassie state. Fig. 6 (a) CAs and SAs of the film after pressing for different times under a 2000 g weight. (b) An SEM image of the superhydrophobic PDMS/SiO 2 film after pressing for 80 h under a 2000 g weight (the inset in (a) is a photograph of the pressing test model). 3.3 Chemical durability and self-cleaning properties We evaluated the chemical durability and robustness of the superhydrophobic film by dipping the samples in solution with different pH values (pH = 1–13) and organic solvents for 60 h. Fig. 7a shows that the CA of the superhydrophobic film changed little, indicating strong resistance to different pH solutions. This phenomenon may be due to the air layer trapped on the surface, which can prevent acid or alkali from contacting samples. Additionally, after immersing in various organic solvents (toluene, acetone, ethanol, and dimethyl formamide [DMF]) for 60 h and washing with ethanol and water, then drying at 60 °C, the CAs of the samples remained above 155°, which indicated that all samples still retained superhydrophobicity. We compared the acidic/alkaline and solvent resistance of the PDMS/SiO 2 film with that of the reported superhydrophobic surfaces (Tables S2 and S3 † ). The as-obtained surfaces show longer stability than previous examples. Fig. 7 (a) CAs of films treated by immersion in different pH solutions and various organic solvents for 60 h. (b) Changes in the CA and SA of film with UV irradiation time. (c) A self-cleaning test. (d) Water droplets sitting on the superhydrophobic PDMS/SiO 2 film under oil (left part) and on a glass slide (right part). (e) Methylene blue dissolved in water droplets, demonstrating spherical shapes under oil. For the UV irradiation stability test, the sample was placed horizontally under the UV lamp at a distance of 20 cm. Fig. 7b shows that the CA and SA showed no obvious change after 150 h of irradiation, implying excellent resistance of the superhydrophobic film to UV light. The UV resistance was higher compared with samples reported in previous studies (Table S4 † ). It is well known that the superhydrophobic surfaces have a self-cleaning property. When dirt falls on the surface, the rolling water droplets can remove the dirt. Here, methylene blue was applied as a marker of self-cleaning properties. From Fig. 7c , we can see that when the dye was washed with water droplets, the dry was easily removed by water droplet. In this work, the obtained superhydrophobic film also demonstrated self-cleaning properties in hexane. After being immersed in oil, the water droplets sat on the film demonstrating spherical shapes, and the water droplets spread and wet the hydrophilic glass surface ( Fig. 7d ). Methylene blue dissolved in the water droplet sitting on the superhydrophobic surface with a spherical shape, and the waste could easily be removed from the surface ( Fig. 7e )."
} | 4,588 |
24428220 | PMC4257568 | pmc | 9,148 | {
"abstract": "Euryarchaea from the genus H alorhabdus have been found in hypersaline habitats worldwide, yet are represented by only two isolates: H alorhabdus utahensis AX-2 T from the shallow Great Salt Lake of Utah, and H alorhabdus tiamatea SARL4B T from the Shaban deep-sea hypersaline anoxic lake (DHAL) in the Red Sea. We sequenced the H . tiamatea genome to elucidate its niche adaptations. Among sequenced archaea, H . tiamatea features the highest number of glycoside hydrolases, the majority of which were expressed in proteome experiments. Annotations and glycosidase activity measurements suggested an adaptation towards recalcitrant algal and plant-derived hemicelluloses. Glycosidase activities were higher at 2% than at 0% or 5% oxygen, supporting a preference for low-oxygen conditions. Likewise, proteomics indicated quinone-mediated electron transport at 2% oxygen, but a notable stress response at 5% oxygen. H alorhabdus tiamatea furthermore encodes proteins characteristic for thermophiles and light-dependent enzymes (e.g. bacteriorhodopsin), suggesting that H . tiamatea evolution was mostly not governed by a cold, dark, anoxic deep-sea habitat. Using enrichment and metagenomics, we could demonstrate presence of similar glycoside hydrolase-rich H alorhabdus members in the Mediterranean DHAL Medee, which supports that H alorhabdus species can occupy a distinct niche as polysaccharide degraders in hypersaline environments.",
"conclusion": "Concluding remarks Hypersaline habitats in general and DHALs in particular seem geographically isolated. Nonetheless, strains of Haloquadratum walsbyi with highly similar genomes have been isolated from salterns in Spain and Australia, indicating mechanisms for exchange ( Dyall-Smith et al. , 2011 ). Such exchange would also explain the highly similar genomes of H. tiamatea and H. utahensis . It is therefore possible that H. tiamatea has been transported to the Shaban Deep, but does not belong to its autochthonous microbial community. While H. tiamatea has a gene repertoire that is distinct from that of H. utahensis and allows survival in DHALs, its preference for micro-oxic conditions, its light-dependent enzymes as well as its relatively high optimum growth temperature are much more in line with life in the upper sediment of terrestrial shallow warm hypersaline lakes. Likewise, specialization of H. tiamatea on plant-derived polysaccharides contradicts preference for DHALs because these habitats receive only little such material. This would also explain why we found only barely detectable abundances of Hrd. species in Medee brine samples by catalysed reporter deposition fluorescence in situ hybridization with a novel Hrd. -specific probe (Supporting Information Fig. S4 and Supporting Information Text). Genomic and functional data from the as yet investigated Hrd. members point towards an ecological niche that involves the capability to degrade complex polysaccharides. Recent identifications of Hrd. members in various other hypersaline habitats (see Supporting Information Text) might even suggest that Hrd. species occupy such a niche in many hypersaline habitats worldwide.",
"introduction": "Introduction Hypersaline habitats are found worldwide, for example in the form of terrestrial and deep-sea brine lakes or man-made solar salterns. The salinities of these habitats range from just above seawater to salt saturation, and their salt compositions range from concentrated seawater with sodium chloride as major salt (thalassohaline habitats) to compositions where other salts such as magnesium chloride dominate (athalassohaline habitats). Despite harsh conditions, microorganisms inhabit hypersaline habitats with a spectrum from species that merely tolerate hypersalinity to true halophiles that require 0.5–2.5 M of salt for growth ( Andrei et al. , 2012 ). Two major strategies have evolved to cope with high salinities and prevent enzymes from denaturing and salt-out precipitation ( Galinski, 1998 ). The first, the organic osmolyte strategy, consists of countering high osmolarities by intracytoplasmic accumulation of compatible solutes like quaternary amines or sugars such as trehalose. The second, the salt-in strategy, relies on accumulation of high levels of internal potassium (and to lesser extents sodium) chloride. Among the most peculiar hypersaline habitats are deep-sea brine lakes, like the Orca Basin in the Northern Gulf of Mexico ( Pilcher and Blumstein, 2007 ), the ice-sealed Antarctic Vida lake ( Murray et al. , 2011 ), the numerous deep-sea hypersaline anoxic lakes (DHALs) in the Eastern Mediterranean Sea ( Bortoluzzi et al. , 2011 ) and the Red Sea ( Antunes et al. , 2011a ). The thalassohaline DHAL Shaban Deep in the Red Sea was discovered in 1984 ( Pautot et al. , 1984 ), and since several novel species were isolated from this location ( Antunes et al ., 2003 ; 2007 ; 2008a , b ). Halorhabdus tiamatea SARL4B T stems from the brine–sediment interface of the Shaban Deep's Eastern basin (26°13.9′ N, 35°21.3′ E, −1447 m depth, pH 6.0, salinity: 244) and features pleomorphic, non-pigmented cells that grow chemoorganoheterotrophically under anoxic to micro-oxic conditions [optimum: 45°C; pH 5.6–7.0; 27% NaCl (w/v)], but poorly under oxic conditions ( Antunes et al. , 2008a ). The only other Halorhabdus ( Hrd. ) ( Oren et al. , 2007 ) species with a validly published name so far is Halorhabdus utahensis AX-2 T (DSM 12940 T ), a sediment isolate from the southern arm of the shallow thalassohaline Great Salt Lake in Utah, USA ( Wainø et al ., 2000 ). Halorhabdus utahensis also features pleomorphic but pigmented cells that can grow under anoxic and oxic conditions (Table 1 ). Both Hrd. species exhibit a 16S ribosomal RNA (rRNA) sequence identity of 99.3% (Fig. 1 ). The genome of H. utahensis has been completely sequenced ( Anderson et al. , 2009 ), whereas until now only a draft sequence was available for H. tiamatea ( Antunes et al. , 2011b ). Both genomes share a large proportion of genes, but also exhibit notable niche differentiations, such as an increased number of genes for membrane transport and utilization of maltose, maltodextrin, phosphonate, and di- and oligopeptides in H. tiamatea ( Antunes et al. , 2011b ). Halorhabdus tiamatea and H. utahensis belong to those Halobacteriaceae species that can degrade plant polysaccharides ( Anderson et al. , 2011 ). For instance, H. utahensis has proven β-xylanase, β-xylosidase ( Wainø and Ingvorsen, 2003 ) and cellulase activities ( Zhang et al. , 2011 ). Fig 1 Maximum likelihood tree of the family H alobacteriaceae. The tree was calculated with RAxML v. 7.0.3 ( Stamatakis et al. , 2005 ) with M ethanospirillum hungatei JF-1 as outgroup. The scale bar represents 10% estimated sequence divergence. Table 1 General characteristics of the H . tiamatea and H . utahensis genomes H. tiamatea H. utahensis Contigs 1 chromosome, 1 plasmid 1 chromosome Chromosome size (G + C content) 2 815 791 bp (63.4%) 3 116 795 bp (62.9%) Plasmid size (G + C content) 330 369 bp (57.4%) – Total genes (coding density) 3023 (83.2%) 2998 (86.2%) Genes with annotated functions 1974 (65.3%) 2243 (74.8%) rRNAs 3 (one rRNA operon) 3 (one rRNA operon) tRNAs 46 (all 20 amino acids) 45 (all 20 amino acids) We sequenced and closed the H. tiamatea genome de novo , annotated it manually and analysed its niche adaptations with an emphasis on polysaccharide degradation and response to oxygen. This included in-depth phylogeny-based annotations of its carbohydrate-active enzymes (CAZymes), corresponding glycosidase activity measurements and proteome analyses of H. tiamatea cultures grown in liquid media under 0%, 2% and 5% of oxygen-containing headspace.",
"discussion": "Results and discussion Genome features The genome of H. tiamatea (Fig. 2 A and B) consists of a 2.8 Mbp chromosome and a putative 330 kbp plasmid with 2743 and 280 predicted genes respectively (Table 1 ). The chromosome exhibits high overall collinearity with the H. utahensis chromosome, but differs by multiple larger inversions and minor genome rearrangements (Supporting Information Fig. S1). Conversely, the putative H. tiamatea plasmid exhibits no such collinearity, and mostly harbours hypothetical and conserved hypothetical genes as well as genes for transposases, DNA-associated proteins and restriction enzymes. The transposase density of the putative plasmid is 12.5% whereas it is only ∼ 2% for the H. tiamatea chromosome. The chromosome furthermore features three regions with putative phage genes that are characterized by dissimilar tetranucleotide usage patterns (Fig. 2 A). Halorhabdus tiamatea , like H. utahensis , contains one clustered regularly interspaced short palindromic repeats (CRISPR) element ( H. tiamatea : 4703 bp with 71 spacers; H. utahensis : 3381 bp with 51 spacers). Fig 2 Circular representation of the (A) chromosome and (B) plasmid of H . tiamatea . From inside to outside: GC content, GC skew, DNA curvature, DNA bending, deviation from the average tetranucleotide composition, CAZymes (blue: glycoside hydrolase, red: glycosyl transferase, green: carbohydrate esterase, orange: polysaccharide lyase, cyan: carbohydrate binding module), RNAs (red: rRNA, green: tRNA, orange: other RNA), genes in reverse direction and genes in forward direction. GC content and GC skew were calculated with a self-written PERL script (sliding windows: 5 kbp for chromosome; 0.5 kbp for plasmid). DNA curvature and bending were calculated with the program banana from the EMBOSS package ( Rice et al. , 2000 ). TETRA ( Teeling et al. , 2005 ) was used for the calculation of the deviation from the average tetranucleotide composition (sliding windows: 5 kbp for chromosome; 1 kbp for plasmid). Monosaccharide utilization Halorhabdus tiamatea can degrade hexoses via the semi-phosphorylated Entner–Doudoroff (ED) pathway. Resulting D-glyceraldehyde-3-phosphate can be further oxidized to pyruvate via the lower part of the Embden–Meyerhof–Parnas (EMP) pathway. The upper part lacks 6-phosphofructokinase. Instead the genome encodes a 1-phosphofructokinase that is co-located with a fructose-1,6-bisphosphate aldolase gene. Such incompleteness or variations of the EMP and gluconeogenesis pathways are common in Archaea . Gluconeogenesis has been deemed non-operational in H. utahensis because of a lack of pyruvate phosphate dikinase ( Anderson et al. , 2011 ), which in H. tiamatea is also not present. Halorhabdus tiamatea has the potential to use fructose by conversion to fructose-1,6-bisphosphate (1-phosphofructokinase), and galactose via the Leloir pathway. The pentose-5-phosphate (PP) pathway in H. tiamatea is missing the oxidative branch, but the non-oxidative branch is present. The latter is likely used to convert pentoses to fructose-6-phosphate. Without the PP pathway's oxidative branch, glucose cannot be converted to ribulose-5-phosphate. However, both Hrd. genomes harbour genes to convert xylose and arabinose to ribulose-5-phosphate. Halorhabdus tiamatea has a ribokinase that H. utahensis lacks, which implies that H. tiamatea in contrast to H. utahensis can also utilize ribose. Polysaccharide utilization Both Hrd. species contain high numbers of CAZyme genes, i.e. genes for enzymes that synthesize, modify or breakup glycosides ( Henrissat and Coutinho, 2001 ). Halorhabdus tiamatea SARLB T has in total 50 glycoside hydrolase genes (15.9 GHs Mbp −1 ), 42 on its chromosome and eight on its putative plasmid (Supporting Information Table S1) – the highest numbers so far observed in Archaea (Supporting Information Fig. S2). Halorhabdus utahensis has 44 GH genes (14.1 GHs Mbp −1 ) according to the CAZy database as of 10 December 2013 ( Cantarel et al. , 2009 ; Lombard et al. , 2012 ). Halorhabdus tiamatea features genes for the degradation of xylan, arabinan, arabinoxylan and galactan-containing hemicelluloses, pectin and likely cellulose. These polysaccharides occur in land plant cell walls, algae ( Popper et al. , 2011 ) and seagrass [e.g. pectin in Zostera marina ( Zaporozhets, 2003 ; Khotimchenko et al. , 2012 )]. Halorhabdus tiamatea furthermore has the genomic potential to degrade exogenous storage carbohydrates such as sucrose or α-1,4-glucans [e.g. starch ( Antunes et al. , 2008a ) or glycogen]. Xylans can be hydrolysed to xylose by concerted action of seven GH10 endo-β-1,4-xylanases and three GH43 β-xylosidases. A dedicated transporter can subsequently import the xylose monomers. Arabinans can be cleaved to L-arabinose by concerted action of a GH43 endo-α-1,5-L-arabinosidase and six GH51 exo-acting α-L-arabinofuranosidases. The latter can also remove decorating L-arabinose side chains from arabinoxylans and arabinogalactans. Arabinoxylans are likely degraded by concerted action of GH10 xylanases and GH51 arabinosidases. The resulting L-arabinose is then likely taken up, isomerized to L-ribulose and subsequently funnelled into the PP pathway. Halorhabdus tiamatea lacks any obvious galactanase and thus probably cannot use the backbones of galactans and arabinogalactans. However, its genome codes for a GH4 α-galactosidase and a GH42 β-galactosidase, which likely enable H. tiamatea to cleave galactose side chains from hemicelluloses ( Popper et al. , 2011 ). The H. tiamatea genome encodes a single PL1 family polysaccharide lyase with pectate lyase function, and a GH88 enzyme. The latter resembles the d-4,5-unsaturated β-glucuronyl hydrolase from Bacillus sp. GL1, which participates in the hydrolysis of unsaturated glycosaminoglycan oligosaccharides released by glycosaminoglycan lyases ( Itoh et al. , 2009 ). Similarly, the H. tiamatea GH88 likely cleaves unsaturated oligopectins released by its PL. Like H. utahensis , H. tiamatea has a modular GH9 with a C-terminal family 3 carbohydrate-binding module. This architecture resembles the endo-processive cellulase E4 from Thermomonospora fusca ( Sakon et al. , 1997 ). Two GH5 enzymes with possible glucanase functions could provide complementary cellulolytic activity. Released oligoglucans may be further degraded by GH3 β-glucosidases. Alternatively, resulting cellobiose dimers might be processed by two GH94 cellobiose phosphorylases with inorganic phosphate to glucose and glucose-1-phosphate ( Yernool et al. , 2000 ). A third GH94 gene (HTIA_1257) is highly similar to laminaribiose phosphorylase from Paenibacillus sp. YM1 ( Kitaoka et al. , 2012 ). Thus, even though H. tiamatea lacks obvious laminarinases, it still may be able to use exogenous laminaribioses. Halorhabdus tiamatea has a putative maltose transporter and is known to grow on maltose ( Antunes et al. , 2008a ). This disaccharide results from degradation of exogenous starch or glycogen by action of maltogenic GH13 enzymes (Supporting Information Fig. S3) and can be subsequently hydrolysed into two α-D-glucose units. Sucrose is another disaccharide that H. tiamatea can potentially use because of presence of a GH32 β-fructosidase, which cleaves sucrose to glucose and fructose. Both sequenced Hrd. species are particularly rich in GH10 xylanases and GH43 β-xylosidases ( H. utahensis : 4× GH10, 4× GH43; H. tiamatea : 7× GH10, 3× GH43). Other polysaccharide-degrading enzymes are abundant in both species as well (Table 2 ), for instance GH2 (e.g. β-mannosidase and β-glucuronidase) and GH3 (β-glucosidase). However, their CAZyme repertoires also exhibit differences, as for example GH32 (β-fructofuranosidase) was only found in H. tiamatea . Likewise, GH13 genes are notably more frequent in H. tiamatea than in H. utahensis (7 vs 1). Conversely, H. utahensis has more GH5 genes (7 vs 2), one with proven cellulase activity ( Zhang et al. , 2011 ). Table 2 Glycoside hydrolases in the Hrd . genomes H. tiamatea H. utahensis GH2 4 (1.27), 3 4 (1.28) GH3 6 (1.91), 4 7 (2.25) GH4 1 (0.32), 1 2 (0.64) GH5 2 (0.64) 7 (2.25) GH9 1 (0.32) 1 (0.32) GH10 7 (2.22), 4 4 (1.28) GH11 0 (0.00) 2 (0.64) GH13 7 (2.22), 6 1 (0.32) GH31 1 (0.32), 1 0 (0.00) GH32 2 (0.64) 0 (0.00) GH42 1 (0.32) 0 (0.00) GH43 3 (0.95), 2 4 (1.28) GH51 6 (1.91), 1 1 (0.32) GH67 1 (0.32), 1 1 (0.32) GH77 1 (0.32), 1 1 (0.32) GH88 1 (0.32) 0 (0.00) GH93 1 (0.32) 0 (0.00) GH94 3 (0.95) 2 (0.64) GH95 1 (0.32) 1 (0.32) GH97 1 (0.32) 1 (0.32) GH abundances in the genomes of H. tiamatea and H. utahensis (according to the CAZy database as of 10 December 2013). The first number represents absolute counts, the number in parentheses counts per Mbp and numbers in boldface GHs detected in proteome data. A peculiarity of H. tiamatea is that most of its arabinan-degradation genes are encoded on its putative plasmid as a single cluster of four GH51 exo-acting α-N-arabinofuranosidases and a L-arabinose isomerase, whereas the complementing GH43 endo-α-1,5-L-arabinosidase is encoded by its chromosome. The GH cluster on the putative plasmid as well as one large GH cluster on the chromosome have dissimilar tetranucleotide usage patterns that stand out even above those of the three putative phage-infected regions (Fig. 2 A and B). This indicates that the capacity for the degradation of some polysaccharides might have been laterally acquired. Extensive lateral acquisition of genes is common in Halobacteria and likely played a major role in their evolution from anaerobic methanogens ( Nelson-Sathi et al. , 2012 ). Fermentations Halorhabdus tiamatea relies on fermentations under anoxic conditions (see Supporting Information Text). It has a four-subunit pyruvate : ferredoxin oxidoreductase for pyruvate oxidation, allowing disposal of reducing equivalents by hydrogen release. Indeed, H. tiamatea encodes a cytoplasmic heterotetrameric [Ni-Fe] hydrogenase, which agrees with the finding that H. tiamatea produces gas from sugars ( Antunes et al. , 2008a ). Sulphur stimulates growth of H. utahensis and is reduced to hydrogen sulphide. It has been suggested that this is facilitated fermentation rather than respiration that serves as hydrogen sink without producing energy ( Wainø et al ., 2000 ). Conversely, sulphur reduction has not been reported for H. tiamatea ( Antunes et al. , 2008a ), and its genome does not seem to contain any respective genes. The bidirectional tetrameric hydrogenase (I) might be able to reduce sulphur as shown in Pyrococcus furiosus ( Ma et al ., 1993 ; 2000 ), but this has not been observed in H. tiamatea . Halorhabdus tiamatea produces acids when grown on maltose ( Antunes et al. , 2008a ) and is known to possess an L-lactate dehydrogenase ( Antunes et al. , 2011b ) and an L-lactate permease, likely for lactate export. Besides, H. tiamatea also features a D-lactate dehydrogenase. Lactate fermentation seems to be the sole mechanism for recycling of reduced pyridine and flavin adenine dinucleotides under anoxic conditions. Acetate is a second likely fermentation product as the H. tiamatea genome encodes an AMP-forming acetyl-CoA synthetase, whose reverse reaction releases acetate from acetyl-CoA while conserving energy as ATP. Halorhabdus tiamatea has all genes for anaerobic glycerol degradation: a glycerol kinase and an anaerobic glycerol-3-phosphate dehydrogenase ( glpABC ) ( Rawls et al. , 2011 ). In Escherichia coli , the anaerobic oxidation of glycerol-3-phosphate to dihydroxyacetone phosphate by GlpABC is coupled to the reduction of fumarate to succinate ( Schryvers and Weiner, 1981 ), but other halophilic archaea such as representatives of the genera Haloferax and Haloarcula have been shown to metabolize glycerol under anoxic conditions to D-lactate, acetate and pyruvate ( Oren and Gurevich, 1991 ). Krebs cycle Halorhabdus tiamatea has a complete Krebs cycle (without glyoxylate shunt). The only anaplerotic reaction seems to be the carboxylation of phosphoenolpyruvate (PEP) by PEP carboxylase. Halorhabdus tiamatea features a malate : quinone-oxidoreductase that funnels electrons directly in the quinone pool and a ferredoxin-dependent 2-oxoglutarate oxidoreductase. Both enzymes might facilitate to run the Krebs cycle in reverse from oxaloacetate to the precursor 2-oxoglutarate, as in some methanogenic archaea ( Sakai et al. , 2000 ). The malate : quinone-oxidoreductase could operate reversely in terms of thermodynamics, as has been discussed for Helicobacter pylori ( Kather et al. , 2000 ). Halorhabdus tiamatea lacks a distinct fumarate reductase such as the membrane-bound type found in E. coli or the coenzyme M-reducing cytoplasmic type found in many methanogenic archaea; hence, a reverse Krebs cycle would involve the regular succinate dehydrogenase. Respiration Halorhabdus tiamatea grows under hypoxic, but poorly under oxic conditions ( Antunes et al. , 2008a ). Its genome encodes all genes for the archaeal membrane-bound NADH : ubiquinone oxidoreductase, as well as cytochrome bd and bc ubiquinol oxidase subunits, a complete 3-subunit copper-containing cytochrome oxidase, together with cytochrome c biogenesis and copper transport genes, and a V-type ATPase. It is known that H. tiamatea reduces nitrate and nitrite ( Antunes et al. , 2008a ). However, its genome does not encode any membrane-associated (Nar) or periplasmic (Nap) respiratory nitrate reductase. It also lacks a membrane-bound Nrf-type cytochrome c nitrite reductase as it is typically found in anaerobes employing dissimilatory nitrate reduction to ammonium. Halorhabdus tiamatea does possess genes for a Nir-type cytoplasmic nitrite reductase, which, however, is non-respiratory as it typically acts only as an electron sink to cope with excess reductants under anoxic conditions. Nir activity is strictly anaerobic and if constitutively required might contribute to the oxygen sensitivity of H. tiamatea . It is not clear whether respiration of endogenous fumarate produced by carboxylation of PEP to oxaloacetate via a reversely operating Krebs cycle is functional. However, exogenous fumarate can likely be reduced to succinate as in other halophilic archaea ( Oren, 1991 ). Response to oxygen Proteomics on extracts of H. tiamatea cultures grown under three oxygen conditions (0%, 2% and 5% in the culture headspace) revealed an increasing stress response with increasing oxygen exposure. Of 699 identified proteins (∼ 24% of the cytosolic proteome), 435 were quantified using a label-dependent method, with at least two peptides in triplicate experiments (Supporting Information Table S2). Compared with the anoxic condition, expression ratios (O 2 /anoxic) varied from 0.36 to 5.22-fold at 2% oxygen, and from below 0.01 to 10.19-fold at 5% oxygen. The 2% oxygen condition caused a mild response with induction of six (including a catalase) and repression of two proteins with at least ± twofold changes. The 5% oxygen condition caused a more pronounced shift in the protein expression pattern, with higher abundances of 32 (7.0%) and lower abundances of 38 (8.4%) proteins. Notably expressed were a catalase (7.5-fold), a superoxide dismutase (6.8-fold), a thiosulfate-sulfurtransferase-like rhodanese (4.9-fold) likely for scavenging oxidative thiyl-radicals ( Remelli et al. , 2012 ), a thioredoxin reductase (2.7-fold) probably with antioxidant functions and a chlorite dismutase (2.4-fold) counteracting oxidative hypochlorite. In contrast, expression of the chaperonine Hsp20 was more than 100-fold lower at 5% oxygen than at the anoxic condition. The energy metabolism was down-regulated at the 5% oxygen condition, most notably subunits of the pyruvate : ferredoxin oxidoreductase (> 100-fold), which seemed to act as major regulation unit for controlling the intracellular carbon flux, but also enzymes from the semi-phosphorylated ED (2.0-fold) and lower EMP pathways (up to 2.3-fold), the Krebs cycle (up to 4.0-fold) and a putative hydrogenase component (3.6-fold). Likewise, most subunits of the NADH-quinone dehydrogenase were down regulated (up to 2.5-fold). Almost half (24/50) of the glycoside hydrolases (3× GH2, 4× GH3, 1× GH4, 4× GH10, 6× GH13, 1× GH31, 2× GH43, 1× GH51, 1× GH67 and 1× GH77) as well as three polysaccharide deacetylases were identified, which stresses their relevance. The 5% oxygen condition led to a down-regulation of six GHs (2× GH10, 3× GH13 and 1× GH31 ranging from 2.2 to 3.9-fold), including the trehalose synthase. Additional GHs seemed to be down-regulated, albeit with a lower than twofold change in expression ratio. Likewise, a subunit of the maltose transporter was down regulated (> 100-fold). Glucosidase activities We tested hydrolysis of 18 p -nitrophenol ( p NP) glycosides with H. tiamatea protein extracts from the above-mentioned three oxygen conditions with and without 3 M KCl. Glucosidase activities were not detected for three p NP glycosides (α-maltose, α-xylose and β-arabinose) and positive for the remaining 15 (Fig. 3 ). The latter support that H. tiamatea can utilize α-glucans (starch, amylose and amylopectin), β-glucans (cellulose), β-xylans, arabinans and galactose-containing hemicelluloses, as suggested by the genome analysis. Activity was also positive for p NP glycosides of the typical plant saccharides α-fucose, α-rhamnose and α-mannose, as well as for β-lactose. Fig 3 Glucosidase activity measurements in H . tiamatea cell extracts with p NP sugar derivatives: (A) total glycosidase activity (units/mg total protein), (B) relative proportion of glycosidase activities in reaction mixtures containing 3 M KCl and (C) relative proportion of glycosidase activities in reaction mixtures without 3 M KCl. Note: In B, sugars marked by an asterisk are shown in the inset and α-glucose in a dedicated panel on the right, each at separate scales. Total activity was highest for the 2% and lowest for the 5% oxygen condition, which agrees with the proteome experiments and supports an adaptation towards micro-oxic conditions (Fig. 3 A). Individual activities varied with salt and oxygen concentrations, indicating that H. tiamatea actively regulates its sugar metabolism. Alpha-arabinose activity was positive, and a GH51 α-L-arabinofuranosidase (located on the putative plasmid) was expressed in the proteome experiments. This conflicts with previous findings that H. tiamatea does not grow on arabinose ( Antunes et al. , 2008a ). Such discrepancies among genomic potential, functional assays and growth experiments are common. For instance, it is known that sugar oligomers rather than monomers induce genes for polysaccharide degradation in Flavobacteria ( Martens et al. , 2011 ). Hence, growth experiments with monomers might constitute an artificial situation for polymer degraders under which they do not exhibit their normal in situ physiology. Light-dependent enzymes Some halobacteria use light-driven ion pumps: bacteriorhodopsin as a proton pump for chemiosmotic ATP generation and halorhodopsin for importing chloride against the concentration gradient. Halorhabdus tiamatea has two genes with rhodopsin domains, one bacteriorhodopsin-like gene and 10 genes containing the bacterio-opsin activator domain (including one located on the putative plasmid). Both rhodopsin domain proteins have the R-X-X-D proton acceptor and D-X-X-X-K retinal Schiff-base binding sites ( Jiao et al. , 2006 ). One of these genes likely has a sensory function, as it is co-located with a methyl-accepting chemotaxis signal transducer gene (see Supporting Information Text for motility). The second rhodopsin gene is co-located with genes encoding the bacteriorhodopsin-like protein, one of the bacterio-opsin activator domain proteins, two isoprenoid biosynthesis enzymes (likely for the retinal cofactor) and the exinuclease subunit UvrA. It is noteworthy that the genome encodes not only the complete UvrABC DNA repair system, but also the blue light-dependent deoxyribodipyrimidine photolyase ( Park et al. , 1995 ), both of which repair ultraviolet-induced DNA damages. Traits associated with a thermophilic lifestyle As many other halophilic archaea, both Hrd. species share traits that are typically associated with a thermophilic lifestyle. Both feature a four-subunit pyruvate : ferredoxin oxidoreductase, which is usually found in hyperthermophilic anaerobes ( Mai and Adams, 1996 ) and in methanogenic archaea, many of which are also moderately thermophilic. In addition, the known β-xylanases, β-xylosidase and cellulase of H. utahensis feature remarkably high temperature optima of 55/70°C, 65°C and 70°C (3 M NaCl) respectively ( Wainø and Ingvorsen, 2003 ; Zhang et al. , 2011 ). Both species also feature surprisingly high optimum growth temperatures [ H. utahensis : 50°C ( Wainø et al ., 2000 ); H. tiamatea : 45°C ( Antunes et al. , 2008a )]. Likewise, both species harbour a gene for LysW, an enzyme that bypasses thermolabile diaminopimelate in lysine biosynthesis in Thermus thermophilus and hyperthermophilic Archaea ( Horie et al. , 2009 ). Storage compounds and osmoregulation Halorhabdus tiamatea is known to produce poly-β-hydroxy-alkanoates (PHA) ( Antunes et al. , 2008a ). A class III PHA synthase is encoded by clustered phaE and phaC genes, and likely forms short-chained polymers with up to five carbon moieties in the hydroxyacyl backbone. Some bacteria are known to build a membrane complex from poly-β-hydroxy-butyrate, polyphosphate and calcium ions. This complex is believed to function as a calcium/phosphate symporter ( Reusch and Sadoff, 1988 ) and if present in H. tiamatea might take part in osmoregulation. Halorhabdus tiamatea also encodes a family 35 glycosyltransferase (GT) glycogen phosphorylase and a GH77 α-1,4-glucanotransferase. Both are essential for endogenous glycogen usage. However, H. tiamatea lacks known enzymes for de novo glycogen biosynthesis, i.e. a glycogen synthase (GT3 or GT5) or glycogen branching enzyme (GH13). This suggests that H. tiamatea either uses unknown isofunctional enzymes or employs a novel pathway for internal α-1,4-glucan biosynthesis. A possible candidate is a GH13 gene (HTIA_0925) with high similarity to the amylosucrase from Neisseria polysaccharea (Supporting Information Fig. S3). The latter can build an internal storage α-1,4 glucan by simultaneously adding glucose units of maltose and sucrose [maltose + sucrose + α-1,4 glucan n = fructose + α-1,4 glucan n + 3 ] ( Okada and Hehre, 1974 ; De Montalk et al. , 1999 ). Such a storage polysaccharide would probably remain linear because H. tiamatea lacks any obvious glycogen branching and GH13 debranching enzyme (isoamylase). Recycling of the α-1,4-glucan could be mediated by concerted action of a GT35 glycogen phosphorylase and a GH77 α-1,4-glucanotransferase via glucose-1-phosphate to glucose-6-phosphate. Almost all respective enzymes are co-located in a single cluster in the H. tiamatea genome, which supports that they act together. Halorhabdus tiamatea also possesses a trehalose synthase-like GH13 (HTIA_0926). Trehalose synthase (TreS) converts maltose to the non-reducing disaccharide trehalose, which can serve as additional storage compound, but is mainly known for its function as compatible solute. Presence of trehalose in H. tiamatea has already been hypothesized before ( Antunes et al. , 2011b ) and corroborates recent findings that Halobacteriales not only use the salt-in, but also the organic osmolyte strategy, e.g. by internal accumulation of trehalose and glycine-betaine ( Youssef et al. , 2013 ). It is noteworthy that in H. tiamatea storage compounds and osmoregulation are possibly interconnected, similarly as in species that use the alternate TreY/TreZ trehalose biosynthesis pathway such as Sulfolobus spp. ( Avonce et al. , 2006 ). In addition to trehalose, also maltose might play a dual role as precursor for both trehalose and a possible storage glucan. Thus, H. tiamatea could use storage glucan biosynthesis to regulate its pool of osmoregulating trehalose. Habitat-specific adaptations Halorhabdus tiamatea and H. utahensis exhibit some notable differences that are likely linked to their isolation sites. For example, the lack of pigmentation of H. tiamatea has been interpreted as adaptation to a light-deprived environment ( Antunes et al. , 2008a ). Halorhabdus tiamatea features mercuric ion and arsenate reductases, which mirrors the specific ion compositions in Red Sea anoxic brine pools ( Antunes et al. , 2011a ). Halorhabdus tiamatea also contains genes for the degradation of methylphosphonate that H. utahensis lacks (see Supporting Information Text). The Shaban Deep is a phosphate-limited environment ( Antunes et al. , 2011b ), which favours species with alternative phosphorus acquisition strategies. Halorhabdus tiamatea has sugar transporters with annotated specificities (ribose, xylulose, maltose and maltodextrin), whereas H. utahensis has been reported not to contain any known sugar transporter ( Anderson et al. , 2011 ). Likewise, H. tiamatea has the capacity to target a broader spectrum of polysaccharides than H. utahensis . Deep-sea habitats typically receive only little land plant, algal and seagrass carbohydrate substrates. Detritus sinking in from the photic zone will be largely consumed when it reaches the deep sea, and mostly the more recalcitrant components such as xylan and arabinan hemicelluloses, cellulose and connecting pectin will prevail. In case of DHALs, such compounds accumulate at the boundary layer of the seawater and the denser brine, and only little ultimately reaches the sediment. Hence, H. tiamatea is likely carbon limited in the Shaban Deep. In this context, the ability to store PHA and possibly α-1,4-glucans might be crucial for its survival. Halorhabdus utahensis in contrast stems from the sediment of the shallow Great Salt Lake of Utah. This lake has influxes from the Bear, Webber and Jordan Rivers, and thus features higher availabilities of suitable substrates. One distinction of H. tiamatea from H. utahensis is its higher GH13 gene number (Table 2 ). Halorhabdus tiamatea has seven such genes (six of which were expressed) with functions that indicate (i) the degradation of exogenous α-glucans, (ii) the possible synthesis/turnover of a storage α-glucan and (iii) synthesis/turnover of trehalose. We found similarly high GH13 gene frequencies in Hrd. species that we enriched from the brine of another DHAL, the Mediterranean thalassohaline DHAL Medee (SW off the Western coast of Crete). Shotgun metagenomics of the enrichment (termed ANR26) yielded a 184 kbp contig carrying a 16S rRNA gene with 99.7% sequence identity to H. tiamatea (Fig. 1 and Supporting Information Text). Metagenome analyses demonstrated that the enrichment contained 80–95% Hrd. species and that its CAZyme profile correlated much better to H. tiamatea than to H. utahensis (Supporting Information Table S4 and Supporting Information Text)."
} | 8,774 |
28181400 | PMC5363340 | pmc | 9,150 | {
"abstract": "Abstract Motion in plants often relies on dynamic helical systems as seen in coiling tendrils, spasmoneme springs, and the opening of chiral seedpods. Developing nanotechnology that would allow molecular‐level phenomena to drive such movements in artificial systems remains a scientific challenge. Herein, we describe a soft device that uses nanoscale information to mimic seedpod opening. The system exploits a fundamental mechanism of stimuli‐responsive deformation in plants, namely that inflexible elements with specific orientations are integrated into a stimuli‐responsive matrix. The device is operated by isomerization of a light‐responsive molecular switch that drives the twisting of strips of liquid‐crystal elastomers. The strips twist in opposite directions and work against each other until the pod pops open from stress. This mechanism allows the photoisomerization of molecular switches to stimulate rapid shape changes at the macroscale and thus to maximize actuation power."
} | 247 |
22427488 | null | s2 | 9,151 | {
"abstract": "Recurrent neural networks (RNNs) are useful tools for learning nonlinear relationships in time series data with complex temporal dependences. In this paper, we explore the ability of a simplified type of RNN, one with limited modifications to the internal weights called an echostate network (ESN), to effectively and continuously decode monkey reaches during a standard center-out reach task using a cortical brain-machine interface (BMI) in a closed loop. We demonstrate that the RNN, an ESN implementation termed a FORCE decoder (from first order reduced and controlled error learning), learns the task quickly and significantly outperforms the current state-of-the-art method, the velocity Kalman filter (VKF), using the measure of target acquire time. We also demonstrate that the FORCE decoder generalizes to a more difficult task by successfully operating the BMI in a randomized point-to-point task. The FORCE decoder is also robust as measured by the success rate over extended sessions. Finally, we show that decoded cursor dynamics are more like naturalistic hand movements than those of the VKF. Taken together, these results suggest that RNNs in general, and the FORCE decoder in particular, are powerful tools for BMI decoder applications."
} | 313 |
36531386 | PMC9757167 | pmc | 9,153 | {
"abstract": "The potential benefits of adding raw, non-food, lignocellulosic plant material as a carbon source for mixotrophic growth of microalgae have previously been demonstrated. This approach has advantages over using traditional carbon sources like glucose or acetate due to wide-spread plant biomass availability and substrate recalcitrance to bacterial contamination. Here, we report the overall growth characteristics and explore the metabolic patterns of Scenedesmus obliquus cultured in the presence raw plant substrate. An initial screen of plant substrate candidates showed an increase in specific growth rate and biomass accumulation when S. obliquus was cultured in the presence of switchgrass or yard waste compared to media alone. We observed a near doubling of microalgal dry weight when S. obliquus was grown with 0.2% ( w/v ) switchgrass under ambient CO 2 . Scanning electron microscopy (SEM) of corn stem after S. obliquus cultivation exhibited substantial phloem degradation. Transcriptomic analyses of S. obliquus during mid- and late-log phase growth revealed a dynamic metabolic landscape within many KEGG pathways. Notably, differential expression was observed for several potential glycosyl hydrolases. We also investigated the influence of switchgrass on the growth of S. obliquus at 50 L volume in mini raceway ponds to determine the scalability of this approach.",
"conclusion": "5 Conclusion We grew S. obliquus in the presence of corn stover, yard waste, sugarcane bagasse, and switchgrass and observed increases in overall growth and dry weight with yard waste and switchgrass. Fluorescence microscopy and SEM of corn stem cross sections in culture with S. obliquus showed degradation of phloem and provides evidence of plant degradation by the microalga. When scaled up to 50 L in mini-raceway ponds we did not see the expected increase in growth nor a substantial difference in the culture microbiome with switchgrass present in the culture. Growth and dry weight of S. obliquus was also increased when we added raw switchgrass to the flask under both ambient and 1% CO 2 . We observed differential expression of potential glycosyl hydrolases that could suggest S. obliquus is degrading the hemicellulose fraction of switchgrass under ambient CO 2 . However, under 1% CO 2 expression of potential glycosyl hydrolases was largely inversed in the presence of switchgrass when compared to that in ambient CO 2 suggesting molecular mechanisms of plant substrate utilization change when inorganic carbon is more available. Upregulation of transcripts encoding PSI and PSII subunits as well as FNR were only observed under ambient CO 2 which indicate either that plant substrate utilization is more reliant on photosynthate or that photosynthesis is stimulated because of plant consumption. The transcriptomic analysis employed here only provides a snapshot of the potential metabolic activity and interaction between S. obliquus and switchgrass during mid-photoperiod and how increased CO 2 concentration may influence that. Future work, like targeted transcriptomics or proteomics of the potential glycosyl hydrolases found differentially expressed in this study, or tracing stable isotopically labeled plant material into S. obliquus biomass, is necessary to gain a more comprehensive view of the molecular mechanisms of plant substrate utilization and to characterize the energy fluxes and metabolic interactions of photosynthesis and respiration that promote increased overall growth of S. obliquus in the presence of switchgrass.",
"introduction": "1 Introduction Select unicellular microalgae can utilize different trophic modes for metabolism and can switch between these modes in response to changing environmental conditions. Growth characteristics of microalgae have been described under photoautotrophic, chemoheterotrophic, and mixotrophic modes, particularly in the context of trying to improve growth rates in industrial applications. Mixotrophy in microalgae has often been described as a synergistic combination of autotrophy and heterotrophy and has been reported to enhance growth rates and biomass accumulation when compared to either trophic mode ( Smith et al., 2015 ; Zhang et al., 2017 ; Abiusi et al., 2020 ). The hypothesis for this is that common hinderances experienced in strictly heterotrophic or photoautotrophic cultivations can be overcome by leveraging two energy sources within one cell ( Wang et al., 2014 ; Abiusi et al., 2020 ). For example, in photoautotrophic cultivation, light availability can be the main growth limiting factor, but with supplementation of an organic carbon source this can be alleviated ( Wang et al., 2014 ; Abiusi et al., 2020 ). We have previously investigated mixotrophy of microalgal strains by adding raw, non-food, plant material to microalgal cultures as carbon source ( Vogler et al., 2018 ; Schambach et al., 2020 ). Not only is this lignocellulosic biomass the most abundant biomaterial in the biosphere ( Malhi, 2002 ), it is inexpensive compared to substrates like glucose, and recalcitrant to bacterial contamination in microalgal cultures; important aspects when considering industrial applicability. Plant substrate utilization has been observed in the freshwater green alga Auxenochlorella protothecoides and in two species of Nannochloropsis ( Vogler et al., 2018 ; Schambach et al., 2020 ). These strains exhibited increased specific growth rates and biomass accumulation in the presence of switchgrass or corn stover. Glycome profiling of switchgrass after addition to A. protothecoides cultures compared to switchgrass in media alone demonstrated the utilization of the hemicellulose xyloglucan by the alga ( Vogler et al., 2018 ). Moreover, differential expression analysis revealed enzymes likely involved in degradation of lignocellulose, including family 5 and 9 glycosyl hydrolases which catalyze hydrolysis of glycosidic linkages in cellulose ( Vogler et al., 2018 ). SEM analysis and sugar analysis of corn stover revealed signs of phloem degradation and glucan utilization, respectively, by Nannochloropsis gaditana ( Schambach et al., 2020 ). Here, we examined potential mixotrophic growth of S. obliquus in the presence of plant substrate. Scenedesmus obliquus , a freshwater chlorophyte known for its robust overall growth, has been reported to grow mixotrophically using commonly investigated carbon sources like glucose or acetate ( Mandal and Mallick, 2009 ; Shen et al., 2018 ; Song et al., 2021 ). This strain’s high tolerance to a wide range of environmental conditions suggests it is metabolically flexible ( Msanne et al., 2020 ), making it an ideal strain to investigate utilization of lignocellulosic biomass. Through transcriptomic analysis we aimed to describe the molecular mechanisms, particularly regarding glycosyl hydrolase expression, employed by this strain to potentially degrade and utilize switchgrass. We also examine the overarching energy metabolism of S. obliquus based on differential expression of enzymes involved in pathways like oxidative phosphorylation, carbon fixation, and photosynthesis. In addition, we investigated the effects of increased CO 2 concentration and scaling up to 50 L volume in mini-raceway ponds on plant substrate utilization.",
"discussion": "4 Discussion We grew Scenedesmus obliquus in the presence of raw plant substrates: corn stover, switchgrass, sugarcane bagasse, and yard waste. Compared to photoautotrophic growth of the alga, we found that with switchgrass or yard waste present, the average specific growth rate increased 13% and 33%, respectively, although not statistically significant (p ≤ 0.05). We did observe a statistically significant increase in specific growth rate in the presence of switchgrass under ambient CO 2 (p ≤ 0.05), with a 73% increase in dry weight. This difference in results is likely due to the greater concentration of plant substrate and greater inoculation cell density used in the latter experiment. As expected, under an increased CO 2 concentration the dry weight of S. obliquus exceeded that of both conditions tested in air, regardless of switchgrass addition. However, with switchgrass present in the culture we observed an additional 7% increase in dry weight, although this was not statistically significant. We have previously reported increased growth rates and biomass accumulation of Auxenochlorella protothecoides and Nannochloropsis sp. when cultured with raw switchgrass or corn stover, respectively, with increases ranging from 13-40% in flask cultures. As with these other species, this overall increase in growth in the presence of plant substrate indicated the potential of S. obliquus to use the raw plant as a carbon source for mixotrophic growth. When switchgrass utilization was tested at 50 L in mini raceway ponds, we did not see the expected increase in growth of S. obliquus with switchgrass present with no difference observed between treatments. In contrast, when Nannochloropsis gaditana was grown at 50 L in mini raceway ponds, specific growth rate was increased 30% with corn stover present when compared to cultures of the alga alone ( Schambach et al., 2020 ). The discrepancy between laboratory scale results and pond scale results is an ongoing hurdle within applied microalgal research ( da Silva and Reis, 2015 ). Many factors, such as culture depth and mixing, light intensity, temperature range, aeration, and on-demand CO 2 for pH balance, can be markedly different from the laboratory experiments and likely play a major role in the outcome. Gaining a better understanding of how these factors play a role in plant substrate utilization and optimizing conditions based on that understanding is necessary to improve scalability of this strategy in the raceway pond setting. Since we did not see the expected growth increase at 50 L, we hypothesized that we would not see a large difference in microbiome composition, and this was generally the case. We did see more representation from members with low abundance during late log phase. However, Paracoccus sp., Allorhizobium - Neorhizodium - Pararhizobium - Rhizobium sp., and Rudanella sp. were present in all cultures regardless of plant substrate or timepoint. This could indicate that bacteria from these genera could be key members of the microbiome that support growth of S. obliquus . Paracoccus is a genus whose members are ubiquitous to both natural and anthropogenic environments and play important roles in many types of microbiomes ( Lasek et al., 2018 ). This genus was abundant in Nannochloropsis gaditana cultures both with and without plant substrate ( Schambach et al., 2020 ). Allorhizobium-Neorhizodium-Pararhizobium-Rhizobium include diazotrophic bacteria commonly found in association with plant roots as well as some green algae including Scenedesmus ( Kim et al., 2014 ). Rudanella is a genus in the family Spirosomaceae, however there is very little literature about the habitats and ecological functions of members of this genus. Characterizing the microbiome of S. obliquus under favorable conditions for plant substrate utilization is necessary for understanding how the microbiome may be playing a role in utilization and how to leverage it for increased microalgal productivity and culture resilience. Increases in biomass production of S. obliquus under mixotrophic conditions have been reported with addition of low molecular weight, soluble organic substrates like glucose or acetate to the culture ( Mandal and Mallick, 2009 ; Yang et al., 2014 ; Shen et al., 2018 ; Song et al., 2021 ). In addition, Yang et al., 2014 reported S. obliquus could utilize xylose, and when added at a 4 g L -1 concentration (0.4% w/v ), cell density was increased 2.9-fold ( Yang et al., 2014 ). S. obliquus could also successfully grow on filtrate containing soluble product of wheat bran fermented by two fungal species ( EL-Sheekh et al., 2012 ). These studies support the ability of S. obliquus to uptake and assimilate several types of exogenous organic carbon. However, when considering the complexity and heterogeneity of lignocellulosic biomass, it may be necessary for S. obliquus to carry a suite of enzymes (e.g., glycosyl hydrolases and others) that, when working synergistically, hydrolyze the structural carbohydrates into free sugars for uptake and metabolism. Qualitative assessment of corn stem morphology via microscopy after incubation with S. obliquus indicated potential cellulolytic enzymatic production as there was noticeable degradation of the plant when compared to the plant incubated in just algal media. The pattern of degradation in these images was similar to observations in Nannochloropsis gaditana and corn stover, particularly regarding the missing and/or collapsed phloem ( Schambach et al., 2020 ). Multi-omic analysis of A. protothecoides revealed this alga has a suite of potential glycosyl hydrolases, including those from families 5 and 9, some of which were upregulated when grown in the presence of switchgrass or carboxymethylcellulose ( Vogler et al., 2018 ). Proteins from glycosyl hydrolase families 5 and 9 contain members with known cellulolytic activity, suggesting A. protothecoides could be directly degrading cellulose. In the current study, the consistent downregulation of potential endo-1,4-beta-D-glucanases in the presence of switchgrass, regardless of CO 2 concentration or time point, suggests S. obliquus may not be directly acting on the cellulose fraction of the plant. This finding could also indicate a lack of recycling or reorganizing of the cell wall of S. obliquus as it contains cellulose ( Voigt et al., 2014 ). This would allow the alga to maintain cell wall rigidity when in the presence of the switchgrass. Anecdotally, lysis of S. obliquus cells that were in culture with switchgrass required 2 to 3 additional cycles of mechanical homogenization before efficient lysis was seen under the microscope. Hemicellulose is another major component of plant cell walls, and can consists of xylan, xyloglucan, mannan, glucomannan, and beta-glucan ( Scheller and Ulvskov, 2010 ).Considering that xylans, polysaccharides of beta-1,4 linked xylopyranosyl residues, are the major hemicellulose structure in switchgrass, and that S. obliquus was reported to utilize xylose, we expected to see significant upregulation of xylanases and/or xylosidases. These enzymes are endo- and exoglycosidases that release xylooligosaccharides and xylose, respectively, from the xylan backbone ( Terrasan et al., 2016 ). We only observed one transcript encoding an endo-1,4-beta-xylanase significantly upregulated during mid-log phase in the presence of switchgrass in ambient CO 2 , whereas one xylan 1,4-beta xylosidase transcript was significantly downregulated in all other sample groups. However, significant upregulation of extracellular mannosidases, and beta-glucosidases under ambient CO 2 could suggest S. obliquus is acting upon the mannan fraction of switchgrass hemicellulose. Beta-1,4-mannosidases and beta-1,4-glucosidases are the major enzymes involved in mannan hydrolysis with complete hydrolysis requiring a synergistic effect of these enzymes ( Rodríguez-Gacio et al., 2012 ). While mannose content in most grasses is low, glucomannans are an important seed storage polysaccharide in grass species like switchgrass ( Scheller and Ulvskov, 2010 ; Rodríguez-Gacio et al., 2012 ). Note that beta-1,4-glucosidases also act on the exposed terminal glycosyl residues of cellulose. We hypothesized that when S. obliquus was grown in the presence of 1% CO 2 , there would not be a notable increase in growth rate or dry weight with switchgrass supplementation. The increased CO 2 concentration allows for the carboxylation reaction catalyzed by RuBisCO to be more energetically favorable, leading to increased carbon fixation and ultimately contributing to faster overall growth. This is exemplified by the 2-fold increase in dry weight when S. obliquus alone was grown in 1% CO 2 versus the ambient condition. Sforza et al., 2012 reported inhibition of mixotrophy in A. protothecoides and Nannochloropsis salina when grown under 5% CO 2 likely because CO 2 fixation was more energetically favorable under the high CO 2 concentration. While not statistically significant, we did see slight increases in specific growth rate and dry weight with switchgrass present under 1% CO 2 This could indicate the switchgrass derived carbon could be playing a role. Despite the lack of a statistically significant growth response to switchgrass under 1% CO 2 , we saw several transcriptomic responses with switchgrass present, that were also contrary to what was found when compared to ambient CO 2 , particularly regarding differential expression of potential glycosyl hydrolases. Several mannosidases, and a xylanase were significantly downregulated with switchgrass present under 1% CO 2 , but several alpha- and beta- galactosidases were significantly upregulated. Furthermore, many alpha- and beta-amylases, starch synthase and a branching enzyme, all of which are involved in starch metabolism, were significantly upregulated in the presence of switchgrass and 1% CO 2 . Stimulation of starch metabolism in microalgae grown under high CO 2 concentrations has been reported ( Tanadul et al., 2014 ), however the presence of switchgrass also appears to influence this response. These transcripts were downregulated in the presence of switchgrass in ambient CO 2 . These results could indicate that under higher CO 2 concentration, different carbon storage metabolic mechanisms are at play in the presence of switchgrass, compared to ambient air. We explored differential expression within mitochondrial oxidative phosphorylation, photosynthesis, and carbon fixation pathways, particularly during late-log phase growth in ambient CO 2 and mid-log in 1% CO 2, to determine if there were any discernable patterns in the overarching energy metabolism of S. obliquus when switchgrass was present. It has been reported under mixotrophic conditions, the metabolism of some algae, like N. gaditana and C. reinhardtii , is dominated by respiration with an associated decrease in photosynthetic activity ( Chapman et al., 2015 ; Bo et al., 2021 ). Differently, photosynthesis in C. sorokiniana is largely unaffected by the addition of acetate ( Cecchin et al., 2018 ). Upregulation of transcripts encoding inorganic pyrophosphatases and proton exporting ATP synthases in S. obliquus in the presence of switchgrass under both CO 2 concentrations, could be an indication of increased ATP production via respiration. In addition, significant upregulation of transcripts encoding key enzymes in the carbon fixation pathway, including RuBisCO, under both CO 2 concentrations could suggest reutilization of the CO 2 produced from respiration ( Smith et al., 2015 ). Interestingly, upregulation of PSI and PSII subunits in the presence of switchgrass was observed under ambient CO 2 , whereas, under 1% CO 2 these or related subunits were slightly downregulated. It could be that increases in expression of these subunits is a response to increased light attenuation because of the large insoluble switchgrass particles. However, when inorganic carbon is less limited under 1% CO 2 , this appears to be alleviated and could indicate that under ambient CO 2 , cells are perhaps more reliant on NADPH and ATP produced from the light reactions of photosynthesis for substrate utilization. Conversely, upregulation of these subunits could be an artifact of increased energy derived from the plant carbon being shuttled into chloroplast production. Cecchin et al., 2018 reported upregulation of Ferredoxin-NADP + reductase (FNR) when C. sorokiniana was grown with acetate, and we also observed this with S. obliquus in the presence of switchgrass under ambient CO 2 ( Cecchin et al., 2018 ). In linear electron flow, electrons are transferred from ferredoxin by FNR to produce NADPH in photosynthesis ( Chitnis, 2001 ). For the case of C. sorokiniana , increased reducing power (i.e., NADH) derived from acetate consumption could be transferred from mitochondria to the chloroplast causing an overreduction of plastoquinones and increasing demand of FNR and ultimately contributing to increased overall growth ( Cecchin et al., 2018 ). Similarly, upregulation of FNR could suggest increased electron transport in photosynthesis in S. obliquus in the presence of switchgrass but only under ambient CO 2 and more work would be needed to determine if it is in fact due to increased reducing power derived from switchgrass consumption."
} | 5,197 |
36242065 | PMC9563475 | pmc | 9,158 | {
"abstract": "Background Geothermal systems have contributed greatly to both our understanding of the functions of extreme life and the evolutionary history of life itself. Shallow-sea hydrothermal systems are ecological intermediates of deep-sea systems and terrestrial springs, harboring unique and complexed ecosystems, which are well-lit and present physicochemical gradients. The microbial communities of deep-sea and terrestrial geothermal systems have been well-studied at the population genome level, yet little is known about the communities inhabiting the shallow-sea hydrothermal systems and how they compare to those inhabiting other geothermal systems. Results Here, we used genome-resolved metagenomic and metaproteomic approaches to probe into the genetic potential and protein expression of microorganisms from the shallow-sea vent fluids off Kueishantao Island. The families Nautiliaceae and Campylobacteraceae within the Epsilonbacteraeota and the Thiomicrospiraceae within the Gammaproteobacteria were prevalent in vent fluids over a 3-year sampling period. We successfully reconstructed the in situ metabolic modules of the predominant populations within the Epsilonbacteraeota and Gammaproteobacteria by mapping the metaproteomic data back to metagenome-assembled genomes. Those active bacteria could use the reductive tricarboxylic acid cycle or Calvin-Benson-Bassham cycle for autotrophic carbon fixation, with the ability to use reduced sulfur species, hydrogen or formate as electron donors, and oxygen as a terminal electron acceptor via cytochrome bd oxidase or cytochrome bb3 oxidase. Comparative metagenomic and genomic analyses revealed dramatic differences between submarine and terrestrial geothermal systems, including microbial functional potentials for carbon fixation and energy conversion. Furthermore, shallow-sea hydrothermal systems shared many of the major microbial genera that were first isolated from deep-sea and terrestrial geothermal systems, while deep-sea and terrestrial geothermal systems shared few genera. Conclusions The metabolic machinery of the active populations within Epsilonbacteraeota and Gammaproteobacteria at shallow-sea vents can mirror those living at deep-sea vents. With respect to specific taxa and metabolic potentials, the microbial realm in the shallow-sea hydrothermal system presented ecological linkage to both deep-sea and terrestrial geothermal systems. \n Video Abstract Supplementary Information The online version contains supplementary material available at 10.1186/s40168-022-01351-7.",
"conclusion": "Conclusions Here, genome-centric metagenomic analysis of vent-associated microbiomes over a 3-year period allowed the recovery of 20 MAGs, substantially increasing the number of genomes sequenced from shallow-sea hydrothermal environments. The metabolic modules of the active and predominant populations within Epsilonbacteraeota and Gammaproteobacteria in shallow-sea vents could mirror those living in deep-sea vents, as suggested by the metagenomic and metaproteomic data. Although hydrothermal microbiomes exhibit different facets of functional traits including the adaption to regional environmental conditions, the microbial functional signatures in the shallow-sea hydrothermal system shed light on a linkage to deep-sea and terrestrial counterparts. Future investigations on the intrinsic functions of rare species from shallow-sea hydrothermal systems could reveal undiscovered metabolic capabilities and facilitate our understanding of microbial ecology and evolution in extreme environments.",
"discussion": "Discussion Endemicity of microbial populations in shallow-sea hydrothermal systems Hydrothermal systems present a variety of ecological niches enabling the colonization of diverse microorganisms, many of which are endemic to these systems. A total of 20 of the 57 culturable genera had a restricted distribution, appearing only in one category of geothermal system (Table S 10 ). Shallow-sea hydrothermal systems shared a total of 28 genera (15 bacterial and 13 archaeal) and 14 genera (5 bacterial and 9 archaeal) with deep-sea hydrothermal systems and terrestrial hot springs, respectively (Fig. 7 ). While only 5 genera (3 bacterial and 2 archaeal) were common to deep-sea hydrothermal systems and terrestrial hot springs, which were also found in shallow-sea hydrothermal systems (Fig. 7 ). Thermophilic cyanobacteria are endemic to terrestrial hot springs (Fig. 7 ) and have successfully colonized this extreme environment [ 75 ]. Culturable genera of Epsilonbacteraeota and Gammaproteobacterial isolates have been retrieved from marine hydrothermal systems; however, none of those genera have been observed in terrestrial hot springs (Fig. 7 ), suggesting a high degree of endemicity in marine hydrothermal vents. Members of the Nautiliaceae within Epsilonbacteraeota and Thiomicrospiraceae within Gammaproteobacteria have frequently been found to be the major active bacterial groups in the hydrothermal systems of Kueishantao Island [ 9 , 16 , 76 ]. Here we consistently recovered MAGs belonging to Nautiliaceae and Thiomicrospiraceae during our 3-year investigation period and successfully resolved their in situ protein expression at the population level for the first time. The genera of Nautiliaceae and Thiomicrospiraceae are also prevalent at deep-sea hydrothermal vents; however, their metabolic processes operating in deep-sea vents in situ at the individual population level are poorly described [ 77 , 78 ]. Cultured Nautiliaceae species can grow autotrophically via the rTCA cycle for carbon fixation, using S 0 reduction coupled with the oxidation of H 2 S, H 2 , or formate [ 79 ], and these processes were found to operate in situ in their relatives thriving at the Kueishantao vents (Fig. 5 A). Thiomicrorhabdus MAGs within Thiomicrospiraceae possessed the potential ability to generate energy via the oxidation of multiple reduced sulfur compounds and transfer electrons to oxygen or nitrate (Fig. 5 B). This result is in line with previous studies of deep-sea vent-associated Thiomicrorhabdus species which noted their metabolic versatility as aerobic chemoautotrophs [ 80 – 82 ]. Thus, the metabolic machinery of the active and predominant Nautiliaceae and Thiomicrospiraceae populations at shallow-sea vents can mirror those living at deep-sea vents. Linking microbial community functional potential to shallow-sea hydrothermal environmental settings Autotrophy The physical and chemical characteristics of geothermal systems play a significant role in constraining microbial functional metabolisms. Phototrophs could inhabit not only terrestrial hot springs but also shallow-sea hydrothermal systems where sunlight is present, including the Kueishantao system [ 9 , 76 ], which represents a striking difference with deep-sea vent systems [ 1 ]. Generally, the rTCA cycle is the major pathway for chemosynthesis in marine hydrothermal systems, including the Kueishantao vents (Figs. 2 and 6 ). However, the in situ chemosynthetic carbon fixation rates suggested only a minor contribution to the primary production of terrestrial hot spring ecosystem, accounting for approximately 3% of the local photosynthetic carbon fixation rates [ 83 ]. Many autotrophic hydrothermal microbes, similar to the gammaproteobacterial Thiomicrorhabdus , appear to have an incomplete oxidative TCA cycle, including but not limited to members of Thermodesulfobacteria and methanogens (Fig. 7 ). Generating energy chemosynthetically from reduced chemicals, instead of by the complete oxidation of exogenous organic carbon, might give an advantage to the living hydrothermal microorganisms to thrive in such extreme environments. The primary function of the incomplete TCA cycle might be responsible for producing metabolites for growth [ 84 ], reflecting adaptation to life in the vents. In contrast to a heterotrophic community, with its high expression of proteins involved in carbohydrate metabolism and organic matter transporter functions [ 60 , 70 , 72 ], the chemolithoautotrophs-dominant Kueishantao community could meet their growth requirements mainly via endogenous synthesis. Energy conversion Geothermal fluids often contain reduced sulfur species (mainly H 2 S and S 0 ), though the actual composition and concentration can vary dramatically depending on the geological setting [ 85 , 86 ]. The andesite-hosted Kueishantao hydrothermal system contained μM level concentrations of H 2 S and were rich in S 0 [ 87 ]. The metagenomic and metaproteomic results revealed that oxidation of reduced sulfur species, especially H 2 S, constituted the most abundant chemolithotrophic energy metabolism in all Kueishantao hydrothermal vents (Tables S 4 and S 6 ). Fluids from shallow-sea hydrothermal vents are often characterized by low H 2 concentrations, as compared to that of deep-sea hydrothermal vents, and trace H 2 was found in Kueishantao hydrothermal fluids [ 87 , 88 ]. The Kueishantao vent metagenomes contained a much lower proportion of putative hydrogenase sequences compared to those from H 2 -rich environments (Table S 3 ), such as serpentinizing deep-sea hydrothermal systems [ 89 ]. Nevertheless, hydrogenases were actively expressed in situ by microbes inhabiting this hydrothermal system (Table S 6 ), enabling them to utilize the available H 2 for energy generation. Geothermal autotrophs that possess the capability of aerobic respiration have a competitive advantage in microbial communities, due to the greater energy yields from the oxidation of reducing substrates with oxygen compared to nitrate or sulfate respiration [ 90 ]. It is noteworthy that the uncultured Nautiliaceae members recovered within this study could utilize O 2 as an electron acceptor (Fig. 5 A) and were persistently abundant in the O 2 -rich Kueishantao hydrothermal fluids (Fig. 2 A), while all cultured Nautiliaceae species could only grow under anaerobic conditions [ 79 ]. Given that a bacterial phenotype is the result of multiple factors, there would be distinctions in observations between laboratory or shipboard incubations and in situ processes. The culturable species of Nautiliaceae that possessed cydAB genes (Fig. 7 ), may present a tolerance for oxygen in submarine hydrothermal systems. Furthermore, the bd -type oxidase were the most abundant terminal oxidases in the metagenomes and metaproteomes of all Kueishantao vent colonizing microbial communities (Tables S 4 and S 6 ), suggesting their ability to grow in even low-oxygen habitats as well as maintain redox balance [ 91 ]. Under anaerobic or microaerobic conditions of geothermal habitats, several microorganisms that possess nitrate or nitrite reductase genes, mainly Epsilonbacteraeota and Gammaproteobacteria (Fig. 7 ), could utilize nitrate or nitrite as an electron acceptor. Epsilonbacteraeota could simultaneously express nitrate reductase and cytochrome oxidases, such as in the prevalent Nautiliaceae populations of the Kueishantao vents (Fig. 5 A), which may enable them to flourish across diverse redox gradients. High rates of sulfate reduction are frequently measured in terrestrial hot springs, enhanced by the presence of photosynthesis products [ 92 – 94 ], whereas the sulfate reduction rates at marine hydrothermal systems are much lower [ 95 , 96 ]. Although sulfate is plentiful in Kueishantao hydrothermal fluids, with concentrations of approximately 24.5-30.4 mM [ 87 ], only a small number of sulfate-reducing bacteria [ 16 ] and few gene sequences involved in sulfate reduction (this study; Table S 4 ) were present in the Kueishantao bacterial community. The functional potentials analyses results suggested the metabolic versatility of microbial communities as chemolithoautotrophs in the Kueishantao shallow-sea vents, generating energy via multiple redox reactions for adapting to low H 2 S, H 2 -poor, and O 2 -rich environments. Rare taxa of shallow-sea vents as future research hotspots The steep geochemical (oxic to anoxic) gradients at shallow-sea vents not only contains aerobic microorganisms but also harbors anaerobic ones that occupy a narrow range of ecological niches and may include rare populations within Thermotogae, Caldiserica, Thermococci, and Thermoprotei (Fig. 7 ). These taxa could serve as a reservoir of special functional potentials, like that in deep-sea vents [ 97 , 98 ]. It was recently shown that high CO 2 levels drive the TCA cycle backward (namely reversed oxidative tricarboxylic acid cycle) [ 99 ], allowing carbon fixation in bacteria inhabiting geothermal systems, such as the hot spring-associated Desulfurella acetivorans [ 100 ], deep-sea vent-associated Thermosulfidibacter takaii [ 101 ], and shallow-sea hydrothermal-associated Hippea martima [ 102 ]. Such a carbon fixation strategy was thought to operate in microorganisms on CO 2 -rich ancient Earth [ 102 ]. Here, we found that autotrophs from Kueishantao Island submarine vents harbored a similar capability for carbon fixation. For example, two bins (2017-1 and 2018-7) belonging to the thermophilic anaerobic bacterial genus Hippea (Fig. 7 ). Additionally, shallow-sea vents contain special and rare species, including thermophilic nitrogen fixing bacteria (such as Methanotorris ), a radiation-resistant bacterium (in Truepera ) [ 103 ], the only currently known hyperthermophilic archaeal host (in Ignicoccus ) [ 104 ] and a hyperthermophilic and neutrophilic archaeon (in Hyperthermu s) [ 105 ] (Fig. 7 ). These results indicated that the rare taxa of shallow-sea hydrothermal systems might curate a seed bank of functional genes involved in ancient metabolic pathways that have survived from when such microbes throve in the early anoxic Earth’s geothermal environments. Hence, shallow-sea hydrothermal rare taxa are worth further exploration to expand our knowledge of microbial metabolic functions in extreme environments and provide clues about the microbial life of early Earth."
} | 3,511 |
39203359 | PMC11356171 | pmc | 9,162 | {
"abstract": "Wheat is a vital global food crop, yet it faces challenges in saline–alkali soils where Fusarium crown rot significantly impacts growth. Variations in wheat growth across regions are often attributed to uneven terrain. To explore these disparities, we examined well-growing and poorly growing wheat samples and their rhizosphere soils. Measurements included wheat height, root length, fresh weight, and Fusarium crown rot severity. Well-growing wheat exhibited greater height, root length, and fresh weight, with a lower Fusarium crown rot disease index compared to poorly growing wheat. Analysis of rhizosphere soil revealed higher alkalinity; lower nutrient levels; and elevated Na, K, and Ca levels in poorly growing wheat compared to well-growing wheat. High-throughput sequencing identified a higher proportion of unique operational taxonomic units (OTUs) in poorly growing wheat, suggesting selection for distinct fungal species under stress. FUNGuild analysis indicated a higher prevalence of pathogenic microbial communities in poorly growing wheat rhizosphere soil. This study underscores how uneven terrains in saline–alkali soils affect pH, nutrient dynamics, mineral content, wheat health, and rhizosphere fungal community structure.",
"conclusion": "5. Conclusions In conclusion, this study highlights the significant influence of uneven terrain and saline–alkali conditions on wheat growth and microbial diversity. Compared to the rhizosphere soil of well-growing wheat, the soil of poorly growing wheat exhibited higher pH values and increased levels of Na, K, Ca, Cu, and Sr while showing lower levels of nitrogen, organic carbon, and P. Additionally, the proportion of unique OTUs was notably higher in the rhizosphere soil of poorly growing wheat compared to that of well-growing wheat. Furthermore, the rhizosphere soil of poorly growing wheat contained significantly higher proportions of the Pathotroph group. Our findings offer valuable insights into the complex interactions between soil properties and wheat growth, emphasizing the need for integrated soil and crop management strategies to enhance wheat production in saline–alkali fields.",
"introduction": "1. Introduction Wheat is one of the most important food crops worldwide and is vital for human sustenance and economic development [ 1 ]. However, wheat production faces numerous disease challenges, including Fusarium crown rot, Fusarium head blight, wheat stripe rust, powdery mildew, and wheat sharp eyespot [ 2 , 3 , 4 ]. In China, Fusarium crown rot has become one of the most severe soil-borne diseases affecting wheat [ 5 , 6 , 7 ]. This disease primarily affects the roots and basal stems of the wheat plant and can result in the entire plant’s death in severe cases. Heavily infected fields can suffer yield losses of up to 30% or more. In China, the predominant pathogen causing Fusarium crown rot is Fusarium pseudograminearum [ 8 , 9 , 10 ]. This pathogen also infects wheat heads and maize roots, leading to Fusarium head blight and maize seedling blight in natural settings [ 11 , 12 , 13 ]. Besides yield reduction and economic losses, the pathogen produces toxins such as nivalenol (NIV) and deoxynivalenol (DON), which can cause nausea and vomiting if ingested, posing significant health risks to humans and animals [ 14 , 15 ]. The incidence of Fusarium crown rot has been increasing due to the lack of resistant varieties, warming climate, straw returning to fields, and the emergence of fungicide-resistant strains [ 16 , 17 , 18 ]. China has extensive and widely distributed saline–alkali soils [ 19 ], mainly located in Northwest, Northeast, and North China and coastal regions. These saline–alkali conditions negatively impact wheat growth, yield, and quality [ 20 , 21 ]. High salt content, elevated pH levels, and poor water permeability in saline–alkali soils hinder water absorption by wheat roots, impact physiological activities, and impede the uptake of essential nutrients like K, Ca, and Mg, leading to nutrient imbalances and stunted growth [ 22 ]. Consequently, these conditions weaken wheat’s disease resistance, making it more susceptible to pathogens and increasing the incidence of diseases like Fusarium crown rot. Soil properties such as nitrogen content, organic carbon, and rhizosphere microorganisms significantly influence wheat’s growth, development, and health [ 23 , 24 ]. Nitrogen plays a vital role in photosynthesis and metabolism. Sufficient nitrogen supply enhances leaf growth and boosts the efficiency of photosynthesis in wheat [ 25 ]. Soil organic carbon provides continuous nutrients and helps improve soil structure, enhancing aggregation and buffering capacity [ 26 ]. This makes the soil more resilient to fluctuations in pH and salinity, thereby promoting healthy wheat growth. Rhizosphere microorganisms, which are primarily beneficial, aid in nutrient availability, though some pathogenic microorganisms can negatively affect wheat’s health [ 27 ]. Managing these soil properties is essential for ensuring robust and healthy wheat cultivation. In the Huanghuaihai region, wheat frequently faces drought conditions [ 28 ]. The common irrigation method used in this area is flood irrigation. Uneven land surfaces created during farming lead to variations in salinity and alkalinity when flood irrigation is applied to saline–alkali soils. These differences result in varying wheat growth and development across uneven regions of saline–alkali fields. Field surveys conducted in this study revealed significant disparities in wheat growth within the same saline–alkali fields. Statistical analysis showed notable variations in plant height, root length, and fresh weight of wheat in different areas. We propose a hypothesis that the rhizosphere soil properties and rhizosphere microorganisms may differ between wheat plants with good growth and those with poor growth. To validate our hypothesis, we collected rhizosphere soil samples from both well-growing and poorly growing wheat regions and analyzed their pH, total nitrogen content, organic carbon content, and mineral element content. Additionally, we conducted fungal amplicon sequencing of the rhizosphere soil. Our aim is to provide a comprehensive understanding of how soil characteristics and microbial diversity impact wheat growth and health. This research will offer valuable insights to support the development of more effective and sustainable agricultural practices.",
"discussion": "4. Discussion This study highlights the significant impact of soil properties on the development and health of wheat and the composition of rhizosphere soil. By examining the interplay between soil pH, nutrient content, and microbial diversity in areas of varied wheat growth, we can better understand the underlying factors influencing wheat health and productivity in saline–alkali environments. One of the key findings of this study is the higher alkalinity in the rhizosphere soil of poorly growing wheat compared to well-growing wheat. The pH levels measured in the poorly growing areas were significantly higher (more alkaline) than those in the well-growing regions. This increased alkalinity is often attributed to the low-lying topography of these areas, which can lead to the accumulation of salts and reduced drainage. Higher soil alkalinity likely plays a crucial role in impeding nutrient absorption and overall plant health [ 38 ]. Previous studies have shown that alkaline soils can restrict the availability of essential nutrients, such as Fe, Mn, and P, which are critical for plant development and metabolic functions [ 39 ]. These nutrient limitations may explain the poorer performance of wheat in these high-pH, saline–alkali environments. Under salt-induced conditions, plants undergo several adaptive responses, such as the regulation of phytohormones, melatonin, ion transport, mitochondrial respiration, and growth and development levels, as well as the activation of reactive oxygen species (ROS) cascades [ 40 , 41 , 42 , 43 ]. While these mechanisms are essential for plant adaptation to stress, they can simultaneously compromise the plant’s resistance to pathogens, making them more susceptible to diseases like Fusarium crown rot. Our study highlights a significant difference in mineral content, particularly the elevated levels of K, Na, Ca, Cu, and Sr in the rhizosphere soil of poorly growing wheat. This suggests that salinity and mineral imbalances in these areas may exacerbate the challenges faced by the plants. High concentrations of certain minerals, especially sodium, can cause osmotic stress and ion toxicity, adversely affecting plant growth and development [ 44 , 45 ]. These conditions likely create an environment that is less conducive to the healthy growth of wheat, leading to the observed variations in plant height, root length, and fresh weight. Interestingly, poorly growing wheat exhibited a higher proportion of unique OTUs in its rhizosphere soil. This suggests that the stressed environment in poorly growing areas might select for a more diverse range of fungal species, potentially including opportunistic pathogens that thrive in saline–alkali conditions. Some inorganic salts have been shown to significantly inhibit various fungal diseases, including wheat stripe rust, wheat powdery mildew, and grape powdery mildew [ 46 ]. In our study, the presence of pathogenic fungi such as Fusarium spp. and Dactylonectria torresensis [ 47 ] was found to decrease with higher levels of pH, Cu, Ca, K, and Sr. This inverse relationship suggests that while these conditions might be unfavorable for plants, they may also suppress the growth of certain pathogens. The FUNGuild analysis revealed that poorly growing wheat had significantly higher proportions of Pathotroph and Saprotroph–Symbiotroph groups compared to well-growing wheat. This points to a more pathogenic and decomposer-dominated microbial environment in the poorly growing wheat’s rhizosphere soil. Such conditions could further stress the plants and reduce their growth and yield potential. These findings underscore the critical importance of managing soil pH and nutrient levels to promote healthy wheat growth, especially in saline–alkali terrains. Enhancing soil organic carbon and nitrogen levels, along with careful management of mineral content, could mitigate the adverse effects of salinity and alkalinity. Practices such as soil amendments [ 48 , 49 ], crop rotation [ 50 ], and the use of salt-tolerant wheat varieties [ 51 , 52 ] might be effective strategies for improving wheat productivity in these challenging environments. Maintaining level terrain during cultivation is also crucial to minimize waterlogging and salt accumulation. Moreover, understanding the microbial dynamics in the rhizosphere can guide the development of biocontrol and biofertilization strategies to harness beneficial fungi and suppress pathogens [ 53 , 54 ]. For instance, promoting the growth of symbiotic fungi and reducing the prevalence of pathogenic fungi through targeted soil management practices could enhance wheat resilience and yield in saline–alkali soils."
} | 2,780 |
28243246 | PMC5303725 | pmc | 9,164 | {
"abstract": "Minimal tillage management of extensive crops like wheat can provide significant environmental services but can also lead to adverse interactions between soil borne microbes and the host. Little is known about the ability of the wheat cultivar to alter the microbial community from a long-term recruitment standpoint, and whether this recruitment is consistent across field sites. To address this, nine winter wheat cultivars were grown for two consecutive seasons on the same plots on two different farm sites and assessed for their ability to alter the rhizosphere bacterial communities in a minimal tillage system. Using deep amplicon sequencing of the V1–V3 region of the 16S rDNA, a total of 26,604 operational taxonomic units (OTUs) were found across these two sites. A core bacteriome consisting of 962 OTUs were found to exist in 95% of the wheat rhizosphere samples. Differences in the relative abundances for these wheat cultivars were observed. Of these differences, 24 of the OTUs were found to be significantly different by wheat cultivar and these differences occurred at both locations. Several of the cultivar-associated OTUs were found to correspond with strains that may provide beneficial services to the host plant. Network correlations demonstrated significant co-occurrences for different taxa and their respective OTUs, and in some cases, these interactions were determined by the wheat cultivar. Microbial abundances did not play a role in the number of correlations, and the majority of the co-occurrences were shown to be positively associated. Phylogenetic Investigation of Communities by Reconstruction of Unobserved States was used to determine potential functions associated with OTUs by association with rhizosphere members which have sequenced metagenomics data. Potentially beneficial pathways for nitrogen, sulfur, phosphorus, and malate metabolism, as well as antimicrobial compounds, were inferred from this analysis. Differences in these pathways and their associated functions were found to differ by wheat cultivar. In conclusion, our study suggests wheat cultivars are involved in shaping the rhizosphere by differentially altering the bacterial OTUs consistently across different sites, and these altered bacterial communities may provide beneficial services to the host.",
"introduction": "Introduction The roots of land plants are surrounded by complex communities of microorganisms ( Tringe et al., 2005 ; Peiffer et al., 2013 ; Edwards et al., 2015 ). There is a growing body of evidence that the root rhizosphere is a crucial zone for many host–microbe interactions ( Zachow et al., 2014 ; Bulgarelli et al., 2015 ; Tkacz et al., 2015 ; Müller et al., 2016 ; Saleem et al., 2016 ). Many of these interactions are mutualistic. For example, plants benefit from mobilization of minerals ( Vangronsveld et al., 2009 ; Sessitsch et al., 2013 ; Plociniczak et al., 2016 ) and other inorganic molecules, while the microbes benefit from root exudates which provide an energy source in the form of sugars and organic acids ( Bais et al., 2006 ; van Dam and Bouwmeester, 2016 ). Plant hosts further benefit from microbes with improved growth ( Bashan et al., 2014 ; Glick, 2014 ), drought and salt tolerance ( Castiglioni et al., 2008 ; Kasim et al., 2013 ; Han et al., 2014 ; Naveed et al., 2014 ; Sarma and Saikia, 2014 ; Pinedo et al., 2015 ), and protection against soilborne pathogens ( Weller et al., 2002 ; Cha et al., 2015 ; De Boer et al., 2015 ). There is further evidence to suggest that some of these microbial interactions are driven by root exudates, such as malic and hydroxamic acids, benzoxazinoids, and other phytochemicals ( Neal et al., 2012 ; Badri et al., 2013 ; Carvalhais et al., 2013 , 2015 ; Yin et al., 2013 ). The use of high-throughput sequencing and microbial-specific databases have allowed for greater characterization of the bacterial rhizosphere communities. It is possible to classify soil microbes down to the level of species or operational taxonomic unit (OTU) using microbial specific databases and efficient clustering algorithms. Next generation sequencing technologies have allowed for deeper sequencing of the microbial communities over previous methods, such as terminal restriction fragment length polymorphism (TRFLP). These newer approaches have provided better insights into the assemblage of the community in terms of alpha and beta diversity ( Peiffer et al., 2013 ; Edwards et al., 2015 ). The use of beta diversity analyses, such as ordination, have helped to describe the microbial community patterns over differing habitats such as field locations and by specific host genotypes ( Berg and Smalla, 2009 ; Peiffer et al., 2013 ; Zachow et al., 2014 ; Edwards et al., 2015 ). However, these ordinations only partially address some of the connections between the microbial taxa and their habitats. Newer methods like co-occurrence networks provide better insights into how these species co-occur, either positively or negatively, and the functional roles they have in the habitat ( Cardinale et al., 2015 ). Recently, co-occurrence networks were used to determine non-random associations between bacterial taxa and their plant hosts ( Cardinale et al., 2015 ; Edwards et al., 2015 ; Kuebbing et al., 2015 ; Tu et al., 2016 ). This method was further complemented with analyzing functional metagenomic data of sequenced bacterial genomes that were associated with species of interest using 16S rDNA markers ( Langille et al., 2013 ; Chen et al., 2016 ). The informative data these analyses provide could be used to characterize microbial species that may benefit host-plant through environmental services, such as mineral mobilization, nutrient assimilation or antimicrobial production ( Wissuwa et al., 2009 ; Langille et al., 2013 ; Chen et al., 2016 ; Raaijmakers and Mazzola, 2016 ). Microbial studies involving crop species such as maize, rice, and wheat have provided insight into how different bacterial communities respond to host, location, plant growth stage, field management strategies, and soil conditions ( Peiffer et al., 2013 ; Donn et al., 2015 ; Edwards et al., 2015 ; Corneo et al., 2016 ). Studies characterizing the bacterial communities of wheat roots have demonstrated host-dependent and cropping effects. For example, the age of a host plant, and crop rotations can impact the bacterial communities of wheat ( Yin et al., 2010 ; Donn et al., 2015 ). There is further evidence that tillage management, such as conventional tillage versus minimal tillage may alter the microbial community and their diversity ( Navarro-Noya et al., 2013 ), including the OTU level ( Yin et al., 2010 ). Wheat is grown on more acres than any other crop ( FAO, 2015 ) allowing for an excellent opportunity to manipulate soil biology and microbial communities through host-dependent mechanisms. We hypothesized that deeper sequencing and more cultivars would provide better insights into host-associated recruitment, and a “bacteria core” for the wheat rhizosphere, particularly in the high rainfall zone of the inland Pacific Northwest (PNW). We also explored the functional potential of the community by using 16S rDNA marker gene associations with species with extensive functional and genomic characterization ( Langille et al., 2013 ; Chen et al., 2016 ). Lastly, we searched what is known about functions of OTU we found to be responsive to specific wheat cultivars to reveal the potential of wheat breeding to enhance microbial communities for purposes like disease suppression.",
"discussion": "Discussion The reduction of environmental and financial costs associated with conventional agriculture requires novel management and breeding strategies that aim to shift current high input methods to more sustainable biological methods. Breeding for wheat cultivars that can recruit beneficial microbes could potentially reduce inputs and make production more sustainable. In this work, we characterized the rhizosphere bacteria associated with nine field-grown winter wheat cultivars by comparing the V1–V3 region of the 16S rDNA gene. This method of sampling and analysis allowed for testing host cultivar effects on the rhizosphere bacteria. OTU variation between wheat cultivars was characterized, and 24 out of the1305 most abundant OTU were found to vary in frequency in the rhizospheres of the different wheat cultivars. This result indicated that the host genotype played a minor but significant role in the bacterial diversification of the rhizosphere and that bacterial frequencies can be altered by selecting specific wheat cultivars. In previous reports, wheat cultivars Lewjain, Eltan, and Hill81 have been shown to differentially select members of the Pseudomonas spp. ( Mazzola and Gu, 2002 : Gu and Mazzola, 2003 ; Okubara et al., 2004 ). Interestingly, no significant differences for Pseudomonas spp. were observable for the same wheat cultivars in the present study. However, members of the Pseudomonas were present in the rhizosphere, accounting for 0.7% of the representative OTUs, and 11 OTUs in the core bacteriome ( Figure 1 ). Mazzola and Gu (2002) demonstrated Lewjain had a tendency to make soils more suppressive to soilborne pathogens like Rhizoctonia spp. and in soils conducive to apple replant disease ( Gu and Mazzola, 2003 ), as compared to other wheat cultivars like Eltan and Madsen. They provided evidence this suppression was mitigated through host-associated recruitment of Pseudomonas spp. into the rhizosphere. Biocontrol studies have often looked at using species like Pseudomonas fluorescence for their production of antimicrobial compound 2,4-diacetylphoroglucinol (2,4-DAPG) ( Landa et al., 2003 ). Interestingly, Landa et al. (2003) , observed species of genera Arthrobacter, Chryseobacterium , and Flavobacterium to be enriched in the presence of 2,4-DAPG producing species, such as P. fluorescence . In the present study, we identified OTU_10130 (genus, Arthrobacter ), OTU_1302 ( Chryseobacterium ), and OTU_256 ( Flavobacterium ) to be enriched in the rhizosphere of some wheat lines, especially line PI561725. Although, in significantly lower abundance than PI561725, OTU_1302 was observed to be in greater abundance for the wheat cultivar Lewjain than for other cultivars. However, we did not see the same higher relative abundance for OTUs 256 and 10130. It is likely that the different soil types and their microbial communities, or the methods employed in the previous studies, such as using a greenhouse or growth chamber, may have attributed to differences in these studies. Differences in the rhizosphere communities of various crop species have been observed ( Aira et al., 2010 ; Bouffaud et al., 2012 ; Peiffer et al., 2013 ; Edwards et al., 2015 ), including differences with two wheat cultivars using TRFLP on the V1–V3 region ( Donn et al., 2015 ). Interestingly, our results differed from a recent report by Corneo et al. (2016) on 24 wheat lines which found no microbial community differences at the vegetative stages of wheat using the TRFLP method on amplicons generated from the V3–V6 hypervariable regions. The lower resolution by the TRFLP method may have affected their ability to identify community differences between wheat lines. In this study, as well as the work by Donn et al. (2015) , rhizosphere samples were collected at the reproductive stages and examined after 2 years of growth in the same soil. In previous studies, no differences were found in the first planting cycle ( Donn et al., 2015 ), or in the vegetative stages ( Donn et al., 2015 ; Corneo et al., 2016 ) of wheat. Several lines of evidence suggest bacteria are recruited through root exudates, and the greatest release may occur during the reproductive stages ( Steinkellner et al., 2007 ; Rudrappa et al., 2008b ; Ferluga and Venturi, 2009 ; Zhang et al., 2009 ; Neal et al., 2012 ; Lakshmanan et al., 2013 ). Multiple growth cycles of different wheat genotypes to recruit responsive microbes may be necessary to see noticeable changes in their frequencies, depending on their initial frequencies in the soil. Changes of frequencies in bulk soil, not just soil closely associated with wheat roots, would presumably take longer but wheat genotypes with stronger effects on microbial communities may be able to affect these communities more rapidly. Plants have evolved mechanisms to tolerate the toxic effects of aluminum by excreting organic acids into rhizosphere soils ( Kochian et al., 2004 ). In wheat, the gene TaALMT1 , an Al-activated malate transporter has been found to confer such a tolerance ( Sasaki et al., 2004 ). A previous study involving an aluminum tolerant (Atlas 66) and intolerant line (Scout 66) found no significant differences after 60 days for the rhizosphere communities using DGGE methods ( Wang et al., 2013 ). We compared two sets of isolines and found very noticeable differences in the Century background (PI561725 and PI561727) but not in the Chisholm background (PI561722 and PI561726; Figure 3 ; Table 3 ). Differences in the genetic background of these pairs could have attributed to the observable changes through a differential response to aluminum or an induced root defense response ( Rudrappa et al., 2008a ; Millet et al., 2010 ). Lakshmanan et al. (2012) reported the bacterial MAMP flg22 or the phytotoxin coronatine can induce malic acid expression in Arabidopsis thaliana in the absence of a low pH or an aluminum rich environment. Additional experiments with the ALMT1 isolines are warranted to determine how the gene expression is regulated in response to environmental stimulus and if other factors that differ between the two genetic backgrounds affect malate secretion. Previous evidence has found that wheat and other plant species differ in the organic compounds they deposit into the soil rhizosphere ( Rengel and Römheld, 2000 ; Sasaki et al., 2004 ; Zuo et al., 2014 ). Multiple biochemical differences were found between the bacterial communities associated with roots of different wheat lines as indicated by KO functions and pathways ( Table 4 ). Zuo et al. (2014) found an increase in nitrogen-fixing and nitrifying bacteria, as well as microbial enzymes associated with nitrogen and carbon metabolism in response to different wheat lines (e.g., ‘22 Xiaoyan’). It was suggested that differences in root phytochemicals could have attributed to the observable differences. In A. thaliana , induction of the AtALMT1 gene, and subsequently the increase in root exudates of MA, had the effect of recruiting bacterial species Bacillus subtilis (strain FB17) into the rhizosphere ( Lakshmanan et al., 2013 ). In the present study, wheat cultivar differences in the bacterial sequence counts by KO terms could be the result of root exudates which select and increase bacteria associated with antibiotic production, sulfur, nitrogen, phosphorus, and malate responsive metabolisms ( Table 4 ). Some of the wheat cultivar-associated bacteria identified in this study may provide beneficial services to their host. Different strains of bacteria isolated from soil have been found to promote plant growth by producing plant hormones such as indole-3-acetic acid (auxin) and 1-aminocyclopropane-1-carboxylate, a precursor for ethylene ( Maimaiti et al., 2007 ; Madhaiyan et al., 2010 ; Marques et al., 2010 ; Soltani et al., 2010 ). It is theorized that the increase in plant growth creates a positive feedback which increases root exudates for bacterial metabolism. Two genera in this study Amycolatopsis and Sphingomonas were found to be wheat cultivar-associated and members of these genera have been previously shown to contribute to plant health by antibiotic production ( Wink et al., 2003 ; Chen et al., 2016 ) and disease suppressive effects ( Vogel et al., 2012 ), respectively. There is further evidence to suggest that strains of Chryseobacterium and Pedobacter produce antifungal compounds against soilborne Phytophthora and Rhizoctonia species ( Marques et al., 2010 ; Kim et al., 2012 ; Yin et al., 2013 ), although, the mechanism for this suppression is still unclear. In this study, the OTU_1302 corresponded to a C. soldanellicola isolate which was identified by Yin et al. (2013) and found to suppress R. solani in culture and in greenhouse assays. Several other suppressive bacteria were identified in that study but did not correspond to any of our wheat cultivar associated OTUs. These results indicated that disease suppressive soils, although not tested in our study, may be attributed to multiple species working in concert together ( Mendes et al., 2011 ; Yin et al., 2013 ; Shen et al., 2015 ; Chapelle et al., 2016 ). Sequencing depth played a major role in quantifying the bacterial community. Altogether, 26,604 OTUs with a ≥97% sequence similarity were found. These numbers are higher than reported in previous wheat rhizosphere studies using pyrosequencing and TRFLP ( Donn et al., 2015 ; Corneo et al., 2016 ) and are consistent with other rhizosphere sequencing studies ( Mendes et al., 2011 ; Peiffer et al., 2013 ; Edwards et al., 2015 ). Increasing sequencing depth would allow for rarer OTUs to be sequenced which may provide improved cataloging of unclassifiable or unculturable OTUs. When OTUs were filtered by a relative abundance of 0.01%, 1305 remained, suggesting a large portion of the community is composed of rarer OTUs. A recent study suggests that the roots of wild grasses such as oat ( Avena spp.) select for rarer bacterial communities, and despite deeper 16S rDNA sequencing, these rarer bacteria were not detectable in the bulk soil ( Nuccio et al., 2016 ). Our data is consistent with these findings and suggested a large portion of OTUs were not in high abundance but were detectable using deeper 16S rDNA sequencing. The most dominant taxa observed in our wheat rhizospheres were in agreement with previous wheat studies but differ from those of other species. The most abundant families in the present study were Sphingobacteriaceae (Bacteroidetes; 15.5% of total sequence reads) and Gemmatimonadaceae. Previous studies complement the current study demonstrating Sphingobacteriaceae as a dominant taxon in the wheat rhizosphere ( Yin et al., 2010 , 2013 ; Wang et al., 2013 ; Donn et al., 2015 ; Corneo et al., 2016 ). Alternatively, the most dominant taxa for Arabidopsis ( Bulgarelli et al., 2012 ; Lundberg et al., 2012 ) and Lettuce ( Cardinale et al., 2015 ) was Comamonadaceae (Proteobacteria), which accounted for only 0.01% of the total sequence reads for the present study. In studies comparing wild and domesticated maize, the family Burkholderiaceae has been found to be the most dominant ( Estrada et al., 2002 ; Szoboszlay et al., 2015 ). Other abundant taxa, such as Bradyrhizobiaceae and Sphingomonadaceae have also been found in the rhizospheres of wheat, Arabidopsis , lettuce, and sugarcane ( Bulgarelli et al., 2012 ; Lundberg et al., 2012 ; Yeoh et al., 2015 ) and are apparently less species dependent. However, it is difficult to determine how the representative OTUs in the present study compare with other studies due to the differences in sequenced regions, platforms, and the techniques used. We have shown wheat cultivars are involved in shaping the rhizosphere by differentially altering the bacterial community. Using deep 16S rDNA sequencing, we could characterize a larger portion of the rhizosphere community than has been previously reported. The differences in microbial communities observed on different wheat lines indicate that rhizosphere communities can be manipulated by wheat breeding. The specific community members found to be responsive to different wheat lines can be used as biomarkers for these community-altering traits. These biomarkers will assist in dissecting the genetic pathways used by these wheat hosts in recruitment, and provide the necessary tools for breeders to incorporate these favorable alleles into commercial production. A future challenge will be to determine which of these traits, and which recruited microbial community components, provide advantages to sustainable wheat production in various environments."
} | 5,077 |
34473943 | null | s2 | 9,166 | {
"abstract": "Due to limitations on high-resolution strain tracking, selection dynamics during gut microbiota colonization and transmission between hosts remain mostly mysterious. Here, we introduced hundreds of barcoded Escherichia coli strains into germ-free mice and quantified strain-level dynamics and metagenomic changes. Mutations in genes involved in motility and metabolite utilization are reproducibly selected within days. Even with rapid selection, coprophagy enforced similar barcode distributions across co-housed mice. Whole-genome sequencing of hundreds of isolates revealed linked alleles that demonstrate between-host transmission. A population-genetics model predicts substantial fitness advantages for certain mutants and that migration accounted for ∼10% of the resident microbiota each day. Treatment with ciprofloxacin suggests interplay between selection and transmission. While initial colonization was mostly uniform, in two mice a bottleneck reduced diversity and selected for ciprofloxacin resistance in the absence of drug. These findings highlight the interplay between environmental transmission and rapid, deterministic selection during evolution of the intestinal microbiota."
} | 298 |
34151570 | PMC7611660 | pmc | 9,168 | {
"abstract": "Synthetic hydrogels\nformed from poly(ethylene glycol) (PEG) are\nwidely used to study how cells interact with their extracellular matrix.\nThese in vivo -like 3D environments provide a basis\nfor tissue engineering and cell therapies but also for research into\nfundamental biological questions and disease modeling. The physical\nproperties of PEG hydrogels can be modulated to provide mechanical\ncues to encapsulated cells; however, the impact of changing hydrogel\nstiffness on the diffusivity of solutes to and from encapsulated cells\nhas received only limited attention. This is particularly true in\nselectively cross-linked “tetra-PEG” hydrogels, whose\ndesign limits network inhomogeneities. Here, we used a combination\nof theoretical calculations, predictive modeling, and experimental\nmeasurements of hydrogel swelling, rheological behavior, and diffusion\nkinetics to characterize tetra-PEG hydrogels’ permissiveness\nto the diffusion of molecules of biologically relevant size as we\nchanged polymer concentration, and thus hydrogel mechanical strength.\nOur models predict that hydrogel mesh size has little effect on the\ndiffusivity of model molecules and instead predicts that diffusion\nrates are more highly dependent on solute size. Indeed, our model\npredicts that changes in hydrogel mesh size only begin to have a non-negligible\nimpact on the concentration of a solute that diffuses out of hydrogels\nfor the smallest mesh sizes and largest diffusing solutes. Experimental\nmeasurements characterizing the diffusion of fluorescein isothiocyanate\n(FITC)-labeled dextran molecules of known size aligned well with modeling\npredictions and suggest that doubling the polymer concentration from\n2.5% (w/v) to 5% produces stiffer gels with faster gelling kinetics\nwithout affecting the diffusivity of solutes of biologically relevant\nsize but that 10% hydrogels can slow their diffusion. Our findings\nprovide confidence that the stiffness of tetra-PEG hydrogels can be\nmodulated over a physiological range without significantly impacting\nthe transport rates of solutes to and from encapsulated cells.",
"conclusion": "Conclusions The diffusivity of solutes in hydrogels is important for the viability\nand activity of encapsulated cells and will regulate the diffusion\nand/or local retention of secreted factors, which play important roles\nin regulating cell behavior. 22 , 47 , 48 Here, we show using theoretical estimates of hydrogel mesh size\nthat predictions for the diffusivity of solutes with known hydrodynamic\nradii reasonably match experimental diffusion behaviors. We also show\nthat altering polymer concentration in our tetra-PEG design allows\nus to produce hydrogels with different mechanical stiffnesses without\nsignificantly impacting diffusivity for hydrogels up to a polymer\nconcentration of 5%. Hydrogels are increasingly used to explore hypotheses\nregarding the role of mechanical stiffness in regulating cell behaviors.\nOur findings provide confidence that such questions can be addressed\nin tetra-PEG hydrogels without introducing the confounding effect\nof differing transport rates of solutes to and from encapsulated cells.\nMoreover, as the A 4 +B 4 hydrogel design allows\nfor stiffer hydrogels to be formed at low polymer concentrations compared\nto A 2 +B 4 designs, stiffer in vitro tissue models can be formed without compromising diffusivity. However,\nour findings also highlight the importance of solute size on diffusion\nrates, suggesting that it should be an important consideration in\nexperimental designs that aim to deliver factors to encapsulated cells\nor assay secreted proteins in culture supernatants. Combined with\nour previous work 11 showing that tetra-PEG\ndesigns can also allow for quick gelation at low polymer concentrations,\nproducing hydrogels that are sufficiently soft for encapsulation of\nhuman intestinal organoids, we have gone some way to demonstrate that\nthese hydrogels are suitable for a range of applications.",
"introduction": "Introduction Cells’ interactions\nwith their local environment are known\nto play central roles in regulating processes including proliferation,\nmigration, differentiation, and phenotypic maintenance. 1 − 3 By extension, these interactions are also involved in dysregulation\nof cell behavior in pathologies. Thus, understanding the impact of\nmechanical and biological cues cells receive from their surroundings\nis key in both disease modeling and the development of regenerative\ntherapies. 4 , 5 While the ability of whole organisms and\ntissue explants to provide physiologically relevant environments to\ncells are unrivalled, there is also a need for simpler reductionist\nmodels that allow for studies into how specific cues impact cellular\nbehaviors. Such models have the potential to identify underlying mechanisms\nthat govern complex tissue pathologies, can reveal fundamental insights\ninto cell-matrix interactions, and may inform methods to engineer\ntissues for regenerative applications. While in vitro cell cultures have revolutionized\nour understanding of mammalian biology, cells respond differently\nwhen within 3D structures akin to tissues compared to on 2D surfaces. 6 , 7 Indeed, among other factors, the transport of molecules to and from\ncells is markedly changed in 3D. Hydrated polymer networks called\nhydrogels can mimic many aspects of the 3D environment cells inhabit in vivo . Their compatibility hinges on their two-phase nature,\nwith a solid polymer scaffold mimicking the extracellular matrix (ECM)\nand the liquid phase available for transport of nutrients. 8 Moreover, the properties of the hydrogel network\ncan be tuned to mimic characteristics of the native tissue, including\ntheir stiffness, which is known to regulate a range of cellular behaviors,\nincluding fate specification. 1 − 3 Synthetic hydrogels formed\nfrom poly(ethylene glycol) (PEG) are\nsuitable for cell encapsulation due to PEG’s stability, hydrophilicity,\nand resistance to protein adsorption. 9 Furthermore,\nthe versatility with which PEG macromers can be cross-linked allows\ncontrol over theoretical mesh size by simply changing polymer concentration,\nmacromer arm size, and the number of arms. And, while native biochemical\ncues are missing in PEG hydrogels, the polymer can be modified to\ninclude ECM-mimicking anchorage sites. Furthermore, cross-linking\nthe network with matrix metalloproteinase (MMP)-sensitive peptides\nallows encapsulated cells to actively remodel and migrate through\nthem. 10 , 11 However, the introduction of bioactive motifs\noften leads to network inhomogeneities. 12 These irregularities are caused by missing cross-links, internal\nloops within individual polymer macromers, and dangling polymer ends. 13 , 14 Such inhomogeneities, although potentially useful as means to permit\ndiffusion, can lead to reduced stiffness. 15 , 16 Moreover, as inhomogeneities push gel structures further from the\nideal network, theoretical characterizations fall short, making predictions\nof hydrogel properties more complex and attributing them to biological\noutcomes more fraught. 17 Therefore, hydrogel\ndesigns that reduce inhomogeneities may provide a more effective and\ncontrolled platform for studying cellular behaviors in 3D. Many\ncovalently cross-linked hydrogel networks rely on Michael-type\nadditions between a cysteine residue at the end of a peptide and an\nalkene-containing end group on the PEG macromer arm (either 4-arm\nor 8-arm, B 4 /B 8 ). Peptide sequences susceptible\nto enzymatic degradation are then created with cysteine groups at\nboth termini (A 2 ), creating A 2 +B 4 /B 8 designs. In such designs, homobifunctional cross-linking\npeptides react with the polymer chain ends indiscriminately. In this\nscenario, primary loops in which one peptide reacts at both ends on\nthe same macromer are likely to form, particularly at low polymer\nconcentrations. Adhesive motifs, on the other hand, typically have\na single cysteine group, and thus are incorporated in a pendant fashion.\nIn the latter arrangement, as more pendant groups are introduced,\nthe number of arms available for cross-linking is reduced, increasing\ngel inhomogeneities. To circumvent these issues, it is possible\nto selectively functionalize\nend groups of both the polymer backbone and peptides, ensuring that\neach can only react in a desired manner. Indeed, the Shibayama group\nhas reported on highly homogeneous, high-strength “tetra-PEG”\nhydrogels that form upon mixing two polymer macromers with different\nreactive terminal groups (A 4 +B 4 ). 15 We hypothesized that it would also be possible\nto create efficiently cross-linked A 4 +B 4 hydrogels\nsuitable for supporting live cells. However, the implications of the\nA 4 +B 4 design on mass transport to and from encapsulated\ncells has not been investigated thoroughly. To create A 4 +B 4 /tetra-PEG hydrogels, we created\nheterobifunctional peptides and reacted an amine at the peptides’\nN-terminus with nitrophenyl carbonate (NPC) end-functionalized four-arm\nPEG (PEG-4NPC, A 4 ), creating PEG–peptide conjugates.\nWe then formed hydrogels by reacting a free thiol from a cysteine\nresidue located at the peptides’ C-terminus with vinyl sulfone\n(VS) end-functionalized four-arm PEG (PEG-4VS, B 4 ). We\nhave previously shown that when adhesive (RGD) and MMP-degradable\npeptide sequences are used to cross-link the PEG network, this design\nsupports the viability of encapsulated human induced pluripotent stem\ncell-derived intestinal organoids. 11 Importantly,\neven within these soft matrices (elastic modulus, ∼ 1 kPa),\ngelation was quick enough that organoids did not fall to the bottom\nof the hydrogel prior to gelation, suggesting that network formation\nwas more effective at polymer concentrations as low as 2.5% compared\nto similar A 2 +B 4 designs. 18 The tetra-PEG design allows physical and biological\nproperties\nof the hydrogel to be tuned independently, while maintaining network\nconnectivity. Indeed, as MMP-susceptible and adhesive peptides both\nparticipate in cross-linking, cellular response to mechanical stiffness\ncan be studied without altering adhesiveness or degradability. However,\nat higher polymer concentrations, the space between cross-links in\nthe polymer phase, known as the mesh size, is reduced. It therefore\nfollows that higher polymer concentrations may not only change cells’\nmechanical environment but also impact the mass transport of solutes.\nIndeed, others have shown that for some hydrogel systems, increasing\npolymer concentration impacts diffusivity. 19 , 20 For both in vitro models and regenerative applications,\nthe ability of nutrients to reach encapsulated cells over a reasonable\ntime scale is crucial. Diffusivity will also impact researchers’\nability to detect secreted molecules in the culture supernatant, which\nmay be of interest for monitoring cell behaviors. Moreover, time scales\nfor diffusion of biomolecules can impact cell–cell communication, 21 , 22 which may play a role in regulating autocrine versus paracrine signaling\neffects. Here, we combined predictive models with experimental\ncharacterization\nto study how altering polymer concentration in tetra-PEG hydrogels\nimpacts the network’s permissiveness to the diffusion of molecules.\nOur findings show that hydrogel stiffness can be modulated over a\nlarge range while only impacting diffusivity negligibly, as we only\nobserved significant changes in diffusion at high polymer concentrations\nthat are less suitable for encapsulating cells.",
"discussion": "Results and Discussion Tetra-PEG\nHydrogel Physical Properties Are Dependent on Polymer\nConcentration To build a model of solute diffusivity, we\nrequired baseline experimental parameters for the tetra-PEG hydrogel\nsystem. Therefore, we first measured the mass swelling ratio of hydrogels\nwith polymer concentrations of 2.5%, 5%, and 10% ( Figure 3 ). On the basis of these values,\nwe applied the Flory–Rehner model to calculate theoretical\nmesh sizes, which yielded values of 8.41, 7.41, and 6.43 nm for the\n2.5, 5, and 10% hydrogels, respectively. These findings were in line\nwith expected trends that hydrogels formed with higher polymer concentrations\nhave smaller mesh sizes. Figure 3 Mass swelling ratio ( Q m ) for 2.5%,\n5%, and 10% hydrogels calculated using the hydrogels’ wet weight\n(at swelling equilibrium) and dry weight ( n = 3 independent\nhydrogels, mean ± SD, one-way Anova with Tukey’s multiple\ncomparison test). Next, we characterized\nthe mechanical behavior of the tetra-PEG\nhydrogels. Mechanical studies using oscillatory rheology can provide\ninsight into hydrogel gelation kinetics and stiffness. 39 To determine the critical polymer concentration\nfor hydrogel formation, we tested hydrogels formed with varying polymer\nconcentrations and determined gelation from the point at which the\nstorage modulus ( G ′) was greater than the\nloss modulus ( G ″). These data show that tetra-PEG\nhydrogels form at polymer concentrations of 1.5% and higher (polymer\nconcentrations of 1% behave as viscous liquids; Figure 4 A). Time sweep measurements further revealed\nthat gelation occurs more quickly for higher polymer concentration\ngels. Ten-percent hydrogels formed in the short time frame between\nloading the sample and measuring the first data point. Alternatively,\n5% hydrogels reached plateau values of G ′\nand G ″ within 10 min, and 2.5% hydrogels reached\nplateau values in ∼20 min ( Figure 4 B). These findings are consistent with theoretical\npredictions that an increased concentration of reactive groups should\ndrive faster reaction kinetics. We also found that G ′ was significantly different for all three polymer concentrations\n( p < 0.0001 for 2.5% vs 5%, 2.5% vs 10%, 5% vs\n10%). However, the loss moduli did not differ significantly from one\nanother ( p > 0.9999 for 2.5% vs 5%, p = 0.2647 for 2.5% vs 10%, and p = 0.2659 for 5%\nvs 10%). The 10% hydrogel showed strain resistance up to 25%, whereas\nboth the 5% and 2.5% polymer concentrations showed strain resistance\nwithin the accessed range ( Figure 4 C). No frequency dependence in storage moduli was observed\nfor any of the three formulations ( Figure 4 D). Taken together, these data show that\ntetra-PEG hydrogels form at polymer concentrations ≥ 1.5% and\nthat their gelation kinetics and resulting equilibrium moduli follow\nexpected patterns based on polymer concentrations. Figure 4 Rheological measurements\nperformed on PEG hydrogels. (A) Mean plateau\nmoduli of hydrogels of varying polymer concentrations. Hydrogels form\n( G ′ > G ″) at polymer\nconcentrations ≥ 1.5%. Mean modulus was calculated from data\ncollected for 10 min after plateau values were reached. (B) Time sweep\nmeasurements of the gelation reaction. Higher polymer concentration\nhydrogels have a higher plateau value and form more quickly. (C) Strain\nsweep measurements, G ′ was significantly different\nbetween all three samples ( p < 0.0001 for 2.5%\nvs 5%, 2.5% vs 10% and 5% vs 10%); G ″ not\nsignificant ( p > 0.9999 for 2.5% vs 5%, p = 0.2647 for 2.5% vs 10%, and p = 0.2659\nfor 5%\nvs 10%), both by one-way ANOVA with Tukey’s correction for\nmultiple comparisons. (D) Frequency sweep measurements. The loss modulus\ncould not be determined for the full range of frequencies accessed.\nIn panels A–D, data are shown as means (dots) with the shaded\narea representing SD; n = 3 independent hydrogels\nfor each condition. Some errors are small and not visible. Mathematical Models Predict That Mesh Size Plays a Limited Role\nin Diffusivity for Small Solutes With the hydrogels’\nphysical properties well characterized, we next aimed to build a diffusion\nmodel treating the polymer chains as an obstruction to diffusing molecules.\nThe hydrodynamic radii of biologically relevant proteins are generally\nwithin the range of a few nanometers. Indeed, cytokines such as IFNγ\nand TNFα are reported to have hydrodynamic radii of 1.85 nm 40 and 3 nm, 41 respectively.\nBovine serum albumin is reported to be 3.56 nm 42 and MMP-9 4.5 nm. 43 Some secreted\nproteins, however, have hydrodynamic radii that are considerably larger.\nFor example, the ubiquitous iron-storing protein ferritin has a hydrodynamic\nradius of 7.17 nm, 44 and the ECM protein\nfibronectin is 8.7 nm. 45 Large proteoglycans\ncan be as much as an order of magnitude larger (∼80 nm 46 ), but these are generally accepted to not diffuse\nwithin hydrogels. Therefore, we considered solutes with hydrodynamic\nradii of 2.3, 4.5, and 6 nm, which correspond to the predicted values\nfor 10 kDa, 40 kDa, or 70 kDa FITC-labeled dextran. Using our\nmodel, we predicted the average concentration of 10 kDa, 40 kDa, or\n70 kDa solutes over time in the solution above the hydrogel normalized\nrelative to the steady state concentration prediction ( Figure 5 ). We found that in the 10\nkDa condition, diffusivity was negligibly impacted by changes in mesh\nsize with a maximum percentage difference between the 2.5% and 10%\nconditions of 4.6% ( Table 3 ). Changes in diffusivity in the 40 kDa condition were more\npronounced between different mesh sizes with the largest difference\nof 20.8%. For the 70 kDa condition, mesh size had the greatest impact,\nas in 10% hydrogels we found that diffusivity changed up to 46% between\nprofiles at each time point. In all conditions, models predicted that\nchanging polymer concentration from 2.5% to 5% only resulted in a\nmaximum change in diffusivity of 16%. Figure 5 Plots generated using mathematical models\nshowing the predicted\naverage concentration of (A) 10 kDa, (B) 40 kDa, and (C) 70 kDa solutes\nin the solution above the gel normalized to the steady state concentration\n(to calculate a percentage of endpoint concentration) for 2.5%, 5%,\nand 10% hydrogels. Table 3 Table Summarizing\nthe Maximum Percentage\nDifference between Diffusion Profiles As Predicted by Modeling Results\nfor All Solute Sizes and Hydrogel Compositions The time taken to reach a steady state concentration\nfor all hydrogel\ncompositions for the 10 kDa condition was ∼20 h and for 40\nkDa solutes was ∼50 h. However, for 70 kDa solutes, the smaller\nmesh size of 10% hydrogels impacted the time to a steady state, extending\nit beyond the ∼100 h found for the 5% and 2.5% conditions.\nFor the 10 kDa solute, the transition to a steady state was faster\ncompared to that of the 40 kDa solute, as half the steady state concentration\nwas reached in <2 h. For the 40 kDa solute, this took <5 h.\nWe then carried out parameter sweeps of solute size from 2 to 8 nm\nfor a fixed mesh size and found that the time to achieve steady state\nconcentration increased dramatically with increasing solute hydrodynamic\nradius ( Figure 6 ).\nIn short, our model predicts that in tetra-PEG/A 4 +B 4 hydrogels, mesh size does not have a large influence on the\ndiffusion of small solutes but can have a more dramatic influence\non larger solutes. Figure 6 Plots generated using mathematical models showing results\nof parameter\nsweeps with (A) showing an average concentration of 40 kDa solute\nin the solution normalized to the steady state concentration above\nthe hydrogel (to calculate a percentage of endpoint concentration)\nas the mesh network size is altered. (B) Average concentration of\nsolute in the solution above a 2.5% hydrogel for solutes with different\nhydrodynamic radii ( r s ). As the solute\nsize increases, the diffusivity is reduced. Experimental Measurements Confirm That Polymer Concentration\nOnly Minimally Impacts Solute Diffusion As our model had\npredicted that polymer concentration only substantially impacted diffusivity\nin the 10% condition for the larger solute, we next aimed to measure\ndiffusivity experimentally. We formed 2.5%, 5%, and 10% tetra-PEG\nhydrogels that contained 10 kDa, 40 kDa, or 70 kDa FITC-labeled dextran\nmolecules. Broadly, we observed that smaller molecules were not differentially\nhindered from diffusing over 100 h, with larger differences observed\nbetween polymer concentrations for the largest molecule. In the 10\nkDa condition, the diffusivity profiles for all polymer concentrations\nwere similar, with the time taken to plateau of ∼45 h ( Figure 7 ). Similarly, in\nthe 40 kDa condition, 2.5% and 5% hydrogels behaved similarly, with\na time taken to plateau of ∼100 h; however, the 10% profile\nappeared to be marginally slowed. For 70 kDa solutes, differences\nbetween polymer concentration were more apparent, again confirming\nthat changes in mesh size have a larger effect for larger molecules\n( Figure 7 ). Statistical\nanalyses comparing fluorescence values in the solution surrounding\nhydrogels after 2 h revealed significant differences between polymer\nconcentrations for all solute sizes (10 kDa: 2.5% vs 5% p = 0.0035, 5% vs 10% p = 0.0275, 2.5% vs 10% p = 0.0002; 40 kDa: 2.5% vs 5% p = 0.0009,\n5% vs 10% p < 0.0001, 2.5% vs 10% p < 0.0001; 70 kDa: 2.5% vs 5% p = 0.0072, 5%\nvs 10% p = 0.0360, 2.5% vs 10% p = 0.0005). However, by 24 h, no significant differences were detected\nbetween polymer concentration in the 10 kDa condition. In the 40 kDa\ncondition, we detected higher levels of fluorescence in the 5% and\n2.5% conditions compared to the 10% (5% vs 10% p =\n0.0056, 2.5% vs 10% p = 0.0008), but the 2.5% and\n5% conditions were no different. For the 70 kDa condition, we detected\nsignificant differences between polymer concentrations for all comparisons\n(2.5% vs 5% p = 0.0142, 5% vs 10% p = 0.0022, 2.5% vs 10% p = 0.0001). These observations\nsuggest that the diffusion of larger solutes is far more impacted\nby changing polymer concentration than that of smaller solutes. Figure 7 Plots showing\nabsolute fluorescence in the media surrounding 2.5%,\n5% and 10% hydrogels containing (A) 10 kDa, (B) 40 kDa, and (C) 70\nkDa solutes ( n = 3, mean ± SD). Some error bars\nare small and not visible. Overall, these data align with trends predicted by our models.\nThey also suggest that doubling the polymer concentration from 2.5\nto 5% produces hydrogels with faster gelling kinetics that are an\norder of magnitude stiffer without greatly affecting diffusivity in\nthe long term. Indeed, increased polymer concentration only appears\nto affect the diffusion of the largest FITC-dextran molecule at stiffnesses\nthat may not be suitable for cell encapsulation ( G ′> ∼ 10 kPa). These findings alleviate potential\nconcerns\nsurrounding PEG hydrogels that changes in stiffness may impact mass\ntransport. Moreover, our hydrogel system allows us to analyze these\nimpacts independently, thus taking advantage of the tunability of\nthe PEG system while minimizing possible confounding effects from\nchanges in mass transport. Our experimental and modeling findings\nboth identified that smaller\nsolutes diffuse faster than larger in our tetra-PEG network. Our findings\nalso show that the time required to achieve a steady state is non-negligible.\nThis is of importance for 3D cell cultures in which both delivery\nof factors to cells from outside the hydrogel (growth factors, e.g.)\nand detection of biomolecules produced by cells (cytokines, e.g.)\nin the surrounding media depend on diffusion. In particular, the latter\nshould only be sampled (or interpreted) at time scales that account\nfor these effects. Furthermore, our modeling and experimental data\nsuggest that differences between diffusion profiles tend to occur\nwithin the first hours. We suggest that differences in diffusion over\nthis time frame are likely to have a minimal impact on experimental\nsetups, as time scales are often longer. We fitted our model\nto experimental results for D eff , which\ndepends on both solute size and mesh size ( R 2 values all > 0.944; Figure 8 ). In some cases, differences between the\npolymer concentrations normalized versus experimental fluorescence\nvalues appeared more pronounced in models; however, this was likely\nattributable to our strategy of normalizing experimental data to a\npredicted steady state value from the data fitting. To determine which\nparameter plays the larger role in determining diffusivity, we analyzed\nthe sensitivity of the model to both parameters. Our results show\nthat network diffusivity, as initially predicted by our model, was\nsystematically overestimated. This is in agreement with others’\nfindings that obstruction theory overestimates diffusivity. 37 Such overestimations may extend from model assumptions,\nincluding that solutes are treated as hard spheres of fixed radius.\nIn reality, molecules like dextran have more nebulous structures,\nand their hydrodynamic radii are unlikely to remain fixed as they\ndiffuse through the network. 25 Furthermore,\nour model predictions rely on estimates of mesh size. Mesh sizes are\nnotoriously complex to obtain, and there remains debate concerning\nthe accuracy of different prediction methods. 23 A final limitation of our model is that potential interactions between\nthe polymer and solute molecules is not accounted for; however, others\nhave shown that this simplification is reasonable. 35 , 37 Nevertheless, despite discrepancies between our predictions and\nexperimental findings, we can have confidence in the trends predicted\nby our model as fitting only served to scale the value of the effective\ndiffusivity rather than change the predicted diffusion profile. Figure 8 Normalized\nfluorescence from experimentally acquired measurements\nof (A) 10 kDa, (B) 40 kDa, and (C) 70 kDa FITC-labeled dextran plotted\nwith fitted mathematical model predictions of solutes diffusing out\nof 2.5%, 5%, and 10% hydrogels. Experimental values are fitted to\nan exponential plateau function and normalized to the end point fluorescence\nto calculate a percentage of the total fluorescence for each time\npoint ( n = 3, mean ± SD). Some error bars are\nsmall and not visible. By performing parameter\nsweeps changing for mesh size and solute\nsize incrementally, we were able to observe the role of each parameter\nin determining overall diffusivity. Thus, our model provided additional\ninsight into mechanisms that drive network diffusivity. Indeed, the\nimpact of changing mesh size was negligible and only had an increasingly\nlarger effect as mesh size approached the hydrodynamic radius of the\nsolute. These results are in line with our experimental findings that\nincreasing polymer concentration to 10% had a greater impact on diffusion\ncompared to increasing from 2.5% to 5%. On the other hand, our models\npredicted that solute size played a far larger role in predicting\ndiffusion, with larger solutes taking longer to diffuse than smaller.\nFor example, our model predicted that solutes with a radius of 7 nm\nversus 8 nm diffusing out of a 2.5% hydrogel produced a maximum percentage\ndifference between profiles of 36%. Our theoretical estimates\npredict that 40 kDa and 70 kDa molecules\nshould remain trapped within hydrogels, as we predicted hydrogel mesh\nsize to be smaller than dextran molecules’ hydrodynamic diameter.\nHowever, though diffusion was increasingly hindered at higher polymer\nconcentrations for larger molecules, they were still able to escape\nthe hydrogels, in keeping with previous reports. 25 These findings suggest an underestimation of mesh size\nor an overestimation of solute effective radius. It is also possible\nthat encapsulating FITC-dextran within hydrogels impacts the mesh,\ndriving the inconsistencies. In short, while theoretical predictions\nare useful for estimating diffusivity, these discrepancies highlight\nthe importance of experimentally measuring diffusion. Parameters from\nthese experiments can then be used to improve models."
} | 6,868 |
36809583 | PMC10161047 | pmc | 9,169 | {
"abstract": "Abstract Integrating adaptative logic computation directly into soft microrobots is imperative for the next generation of intelligent soft microrobots as well as for the smart materials to move beyond stimulus‐response relationships and toward the intelligent behaviors seen in biological systems. Acquiring adaptivity is coveted for soft microrobots that can adapt to implement different works and respond to different environments either passively or actively through human intervention like biological systems. Here, a novel and simple strategy for constructing untethered soft microrobots based on stimuli‐responsive hydrogels that can switch logic gates according to the surrounding stimuli of environment is introduced. Different basic logic gates and combinational logic gates are integrated into a microrobot via a straightforward method. Importantly, two kinds of soft microrobots with adaptive logic gates are designed and fabricated, which can smartly switch logic operation between AND gate and OR gate under different surrounding environmental stimuli. Furthermore, a same magnetic microrobot with adaptive logic gate is used to capture and release the specified objects through the change of the surrounding environmental stimuli based on AND or OR logic gate. This work contributes an innovative strategy to integrate computation into small‐scale untethered soft robots with adaptive logic gates.",
"conclusion": "3 Conclusion In summary, this work introduced a strategy to construct a series of small‐scale untethered soft microrobots with adaptive logic gates based on stimuli‐responsive hydrogels. The various stimuli‐responsive hydrogels (i.e., temperature, pH, and salt‐responsive hydrogels) were used as modules to fabricate soft microrobots performing different logic gates (i.e., YES, AND, and OR gate). Besides, we successfully integrated different basic logic gates into the one soft microrobot as described (i.e., AND‐OR and OR‐OR gate). Most importantly, we developed two kinds of soft microrobots that can adapt logic operations between AND gate and OR gate according to the surrounding environmental stimuli intelligently and autonomously. Thus, combined with the simplicity of preparation and the capability of the adaptive logic gates, the soft microrobots own anticipated potential to perform advanced computation or analyses. The proposed strategy for fabricating the microrobots is able to build a soft adaptive computational system directly into the body of the microrobots. This would lead to a new generation of soft microrobots, paving the way for more sophisticated soft microrobots and intelligent compliant structures.",
"introduction": "1 Introduction The development of untethered microrobots is one of the holy grails in the field of robotics research. The microrobots have been intensively investigated in the last decade, they can autonomously perform specific and various tasks at a small scale and can be applied to representatively specific environments (e.g., the organism). Meanwhile, they are applied in various fields, such as ecological restoration, biotechnology, micro‐manipulation, targeted release, diagnosis, artificial muscles, and medical imaging. [ \n \n 1 \n , \n 2 \n , \n 3 \n , \n 4 \n , \n 5 \n , \n 6 \n , \n 7 \n , \n 8 \n \n ] Intelligent microrobots have reached a high‐level of complexity, which is interacted with the surroundings and can acquire information about the self and the environment and then output the analyzed data as expected. [ \n \n 9 \n , \n 10 \n \n ] As a promising strategy, the research of microrobots has currently focused on developing microrobots with advanced stimulus‐responsive abilities, which is a critical step toward micromachine intelligence. [ \n \n 11 \n , \n 12 \n , \n 13 \n \n ] Recently, many different kinds of soft microrobots based on responsive polymers have been developed with excellent performances such as programmable shape‐morphing, [ \n \n 14 \n , \n 15 \n \n ] drug delivery, [ \n \n 16 \n , \n 17 \n \n ] sensors, [ \n \n 18 \n , \n 19 \n \n ] and motions (e.g., jumping, swimming and crawling), [ \n \n 20 \n , \n 21 \n , \n 22 \n , \n 23 \n , \n 24 \n , \n 25 \n \n ] which provide tremendous potential to achieve intelligent adaptation in diverse fields. Integrating adaptative logic computation directly into soft microrobots is imperative for the next generation of intelligent soft microrobots as well as for the smart materials to move beyond stimulus‐response relationships and toward the intelligent behaviors seen in biological systems. [ \n \n 13 \n , \n 26 \n , \n 27 \n \n ] However, combining adaptive logic‐based computation with untethered soft microrobots is difficult because of the stiffness, flexibility, and size mismatch between rigid electrical components (e.g., electronic chips, control sensors, and batteries) and soft microrobots. [ \n \n 9 \n , \n 28 \n , \n 29 \n , \n 30 \n , \n 31 \n \n ] Therefore, it is an enormous challenge to integrate adaptive computing functions into soft microrobots. Combining chemical logic gates with microrobots is a promising strategy to enable microrobots with binary logic computations. Implementing basic logic gates such as YES gate, NOT gate, AND gate, OR gate, and NOR gate is generally supposed to be the precondition for achieving “artificial intelligence” in chemical systems. [ \n \n 32 \n , \n 33 \n \n ] Recently, various impressive chemical logic gates performed by chemical compounds or natural molecules have been reported in literature. [ \n \n 34 \n , \n 35 \n , \n 36 \n , \n 37 \n , \n 38 \n , \n 39 \n , \n 40 \n , \n 41 \n \n ] Remarkably, Wang and co‐workers further constructed a logic gate based on DNA that could be endowed with adaptiveness, meaning the same logic gate could exhibit two different logic gate functions when the surrounding temperature change. [ \n \n 41 \n \n ] Acquiring adaptivity is coveted for soft microrobots that can adapt to implement different works and respond to different environments either passively or actively through human intervention like biological systems. Meanwhile, adaptive logic gates enable the soft microrobots’ flexibility and reusability, which means that the same soft microrobot can implement different logic operations in response to external environmental stimuli without additional replacement. In this work, a strategy is contributed to fabricate the soft microrobots with logic‐based computation via a simple way and commonly used stimuli‐responsive hydrogels. Stimulus‐responsive hydrogels were widely used for constructing the logic system, which can controllably and reversibly swell and shrink according to external environmental stimuli such as current/voltage, temperature, light intensity, magnetic field, pH, and salt to a controllable and reversible shape transformation. [ \n \n 42 \n , \n 43 \n , \n 44 \n , \n 45 \n \n ] Thus, we chose three commonly used stimulus‐responsive hydrogels poly(N‐isopropylacrylamide) (PNIPAm), poly(acrylic acid) (PAAc), poly(2‐acrylamido‐2‐methylpropanesulfonic acid) (PAMPS) as building modules to construct a series of soft microrobots with logic gates, which are occupied by stimuli‐responsive hydrogels except for a hole in the middle. In this logic system, the states of stimuli‐responsive hydrogels (swell or shrink) as input signal, the microrobots can logically and autonomously analyze the input signal to open or close the hole in the middle as the output signal. In this way, the process of logical operation can be observed intuitively. Our work focuses on the proof‐of‐concept demonstration, the soft microrobots fabricated in this work are at the micron or millimeter scale. Our work provides a new approach to integrating adaptive logic gates into soft microrobots so that they can implement two different logic gate operations according to external environmental stimuli without replacement. Furthermore, employing stimuli‐responsive hydrogels as modules to construct soft microrobots with logic gates is a flexible approach. In practical applications, a variety of stimuli‐responsive hydrogels can be selected as modules in response to the different stimuli of the surroundings.",
"discussion": "2 Results and Discussion 2.1 Design and Construction of Soft Microrobots Processing YES, AND, and OR Gates As a presentation of our notion, a series of submillimeter/millimeter scale microrobots were fabricated including one or more pieces of soft stimuli‐responsive hydrogels with a hole in the middle for capturing objects, and a layer of coating around its sides. (Details on the materials and methods are shown in Supporting Information). In our system, when the hydrogel is in a shrunken state, it is defined as input of 0, when the hydrogel is in a swollen state, it is defined as input of 1. When the hole in the middle of the hydrogel was opened, we define the state as output of 0; when the hole in the middle was fully closed, we define the state as output of 1( Figure \n \n 1 a ). The YES gate was fabricated employing the temperature‐responsive hydrogel (PNIPAm‐1) or pH‐responsive hydrogel (PAAc‐1), respectively (Figure S1 , Supporting Information). When the temperature‐responsive hydrogel was placed in an aqueous solution at 40 °C, it was shrunk (input 0). Meanwhile, the hole in the middle of the hydrogel was opened (output 0). Conversely, the hydrogel was swollen at 20 °C (input 1), the hydrogel did not expand laterally outward due to the coating, instead it expanded inward into the hole; hence, the hole was fully closed (output 1). Similarly, the YES gate was performed employing a pH‐responsive hydrogel (Figure 1a ). The hydrogel was shrunk in an aqueous solution of pH = 2 (input 0), and the hole in the middle was opened (output 0). By contrast, the hydrogel was swollen in an aqueous solution of pH = 12 (input 1), the hydrogel expanded inward into the hole, and the hole was fully closed (output 1). The process was reversible. Figure 1 The construction of the microrobots with YES, AND, and OR logic gates employing PNIPAm and PAAc hydrogels. a) Schemes illustrating the microrobots with two YES gates. Inputs 0 and 1 were defined as the state of the hydrogel shrunk and swollen. Outputs 0 and 1 were defined as the state when the hole in the middle of hydrogel opened and closed; Schematic illustrations and images of the microrobots in the different states and the truth tables with b) AND gate and c) OR gate. Based on the response properties of PAAc and PNIPAm hydrogel, we constructed more complex logic gates. Figure S2 , Supporting Information, demonstrates the typical approaches for constructing the microrobots with AND and OR gates including pH‐responsive and temperature‐responsive hydrogels, each one of them responds to a different stimulus. For the AND gate, two kinds of responsive hydrogels take up half of the microrobot except for a hole in the middle. Half of the microrobot consisting of PAAc‐1 was actuated by pH change, and another half of the microrobot including PNIPAm‐1 was actuated by temperature change. When the temperature decreases or the pH increases, the half of microrobot was expanded to the center. However, it merely took up half of the hole without fully closing the hole. Hence, the hole could be fully closed only when both the hydrogels expanded (Figure 1b ). For the OR gate, the two stimuli‐responsive hydrogels (PAAc‐2 and PNIPAm‐2) occupy half of the microrobot except for a hole in the middle. Unlike the AND gate microrobot, whenever anyone (or both) hydrogel of OR gate microrobot was swollen under the influence of either one (or both) of the stimulus, the hole was closed to zero (Figure 1c ). 2.2 Design and Construction of Soft Microrobots Processing Combinational Logic Gates Most chemical logic gates can only exhibit basic logic functions, which limits their applications for performing more advanced operations. Connecting different basic chemical logic gates can construct multi‐switchable logical systems and be applied in various fields, such as molecular computation, [ \n \n 36 \n , \n 46 \n \n ] chemical detection, [ \n \n 47 \n , \n 48 \n \n ] and molecular capturing and releasing. [ \n \n 49 \n , \n 50 \n \n ] Reasonably advanced arithmetic operations can be performed by integrating basic logic gates. In our work, different classes of stimuli‐responsive hydrogels were employed to construct soft microrobots with combinational logic gates. Under different types of external stimuli, the outputs of the microrobot with AND and OR gate are the same: occupying space of the hole in the middle by expansion or contraction. Thus, we can integrate different basic logic gates to implement more advanced computations. First, we integrated AND gate and OR gate into the same microrobot, which was taken up by three kinds of responsive hydrogels except for a hole in the middle (Figure S3 , Supporting Information). The AND gate was implemented including a temperature‐responsive hydrogel (PNIPAm‐1) and a pH‐responsive hydrogel (PAAc‐1) as discussed in Figure 1b . The OR gate was fabricated employing a salt‐responsive hydrogel poly(2‐acrylamido‐2‐methylpropane sulfonic acid) (PAMPS). When both the hydrogels (PNIPAm‐1 and PAAc‐1) of the AND gate were swollen, the hole in the middle was fully closed regardless of whether the hydrogel (PAMPS) was swollen (in aqueous solution) or shrunk (in 0.5 m NaCl solution). However, when the hydrogel (PAMPS) was swollen, the hole was fully covered regardless of whether the hydrogels (PNIPAm‐1 and PAAc‐1) of the AND were swollen or not ( Figure \n \n 2 a ). Moreover, we also fabricate a soft microrobot that connected an OR gate to an OR gate. Figure 2 Schematic illustrations and images of the microrobots in the different states with connected logic gates and the truth tables. a) An AND gate connected with an OR gate. b) An OR gate connected with another OR gate. The first OR gate was implemented including hydrogels (PNIPAm‐2 and PAAc‐2) as discussed in Figure 1c . The second OR gate was also fabricated by salt‐responsive hydrogel (PAMPS). The preparation method was consistent with that of microrobot with AND‐OR gate. Whenever anyone (or both) hydrogel of the first OR gate was swollen, the hole in the middle was fully closed regardless of whether the hydrogel (PAMPS) was swollen or shrunk. Likewise, when the hydrogel (PAMPS) was swollen, the hole was also fully covered regardless whether the hydrogels (PNIPAm‐1 and PAAc‐1) of the OR gate were swollen or not (Figure 2b ). The reversibility of the microrobots that performed logic gates operation was tested by switching the different types of stimuli repeatedly for ten cycles. For the microrobot performing YES gate by pH‐responsive hydrogel, we changed the pH of the solution repeatedly from pH 2 to pH 12 and vice versa (Figure S4a , Supporting Information); For the microrobot performing YES gate by temperature‐responsive hydrogel, we changed the temperature repeatedly from 40 to 20 °C and vice versa (Figure S4b , Supporting Information). In both cases, the response times of the microrobots performing logic gates remained approximately constant throughout the ten cycles. These results show that the microrobots were able to operate logic gates without any decrease in performance after ten cycles of changing the stimuli. This conclusion is supported by previous studies in which the stimuli‐responsive hydrogels were known to be able to change their sizes reversibly for many cycles. [ \n \n 51 \n \n ] \n 2.3 Design and Construction of Soft Microrobots with Adaptive Logic Gates More importantly, many intelligent biological systems such as humans, animals, and even cells can choose appropriate logic operations to adapt to surrounding changes. For example, the lactose operon is only activated for the production of monosaccharides in the presence of lactose and interestingly, when both glucose and lactose exist, the lactose operon is inactivated. Therefore, the energy needed by cells was preferred to provide by glucose instead of the less efficient lactose. [ \n \n 40 \n , \n 52 \n \n ] Although the soft microrobots we fabricated can perform a series of logic operations (e.g., YES, OR, AND, and connected gates) upon the environmental stimuli, they are not able to choose logic gate intelligently under different situations since one microrobot can only perform one logic gate. Hence, we investigate the effect of amount of crosslinkers on the swell ratio of stimuli‐responsive hydrogels, and adjust the swell ratio of hydrogels to develop the microrobot that can smartly switch logic operation between AND gate and OR gate upon varied environmental stimuli ( Figure \n \n 3 a ). The microrobot was designed based on OR and AND gates. At first, the microrobot was placed in an aqueous solution of pH = 2 at 40 °C, it was in a fully shrunk state. As the surrounding environmental conditions change, when the temperature was between 27 and 29 °C, and the pH was between 4.5 and 5, only one type of responsive hydrogel of the microrobot was swollen, the hole in the middle cannot be fully closed (output 0). When two kinds of stimuli‐responsive hydrogel were swollen simultaneously, the hole in the middle was fully closed (output 1). The microrobot could identify this environmental information and perform the AND gate. However, when the temperature was lower than 25 °C and the pH was higher than 6, any one (or both) type of responsive hydrogel of microrobot was swollen, the hole in the middle of the microrobot was able to fully close (output 1). The microrobot could analyze the environmental stimuli and switch to the OR gate intelligently (Figure 3b ). Figure 3 The construction of soft microrobots with adaptive logic gate. a) The schematic diagram to switch logic operations between OR gate and AND gate and truth tables. b) Schematic illustrations and images of the soft microrobots switching AND gate to OR gate. The soft microrobot with adaptive logic gates demonstrated in Figure 3 was constructed by two independent stimulus‐response hydrogels (PNIPAm and PAAc), which contributes an advantage that autocephalous stimulus‐response modules were not affected by different stimulus signals. In order to show the universality of constructing the microrobots with adaptive logic gates, we introduce a more convenient strategy for developing the soft microrobot with adaptive logic gates employing the dual‐responsive hydrogel, which was prepared by pH‐responsive polymer (PAAc) and salt‐responsive polymer poly(2‐acrylamido‐2‐methylpropane sulfonic acid) (PAMPS) as shown in Figure \n \n 4 a . The construction method is the same as the YES gate (Figure S5 , Supporting Information). The microrobot also demonstrated the characterization of switching between AND and OR logic gates according to different NaCl concentrations and pH of the aqueous solution. In the same way, when the hydrogel is in a shrunken state, it is defined as an input of 0, when the hydrogel is in a swollen state, it is defined as an input of 1. In addition, the PAAc network and the PAMPS network of the hydrogel are labeled as Gel 1 and Gel 2, respectively. Figure 4 The construction of soft microrobots with adaptive logic gate employing the pH and salt dual‐responsive hydrogel. a) The schematic diagram to switch logic operation between AND and OR gate and truth tables. b) Schematic illustrations and images of the soft microrobots switching AND gate to OR gate. At first, the microrobot was exposed to a 0.2 m NaCl aqueous solution of pH = 2, it was in a fully shrunk state. As the surrounding environmental conditions change, when the concentration of NaCl solution was between 0.025 and 0.1 m , and the pH was between 4.5 and 6, the soft microrobot could identify this environmental information and perform the AND gate. However, when the concentration of NaCl solution was 0 m and the pH was higher than 8, the soft microrobot could analyze the environmental stimuli and switch to the OR gate intelligently (Figure 4b ). The process of the microrobot performing OR gate in an aqueous solution of pH = 8 is shown in Figure S6 , Supporting Information. 2.4 Magnetic Soft Microrobot with Adaptive Logic Gate Performing Capture, Delivery, and Release One important application of these microrobots is that they could be used for the targeted transport of therapeutic agents. As a proof of concept, we demonstrated the trapping, delivery, and release of a Teflon rod (diameter: 200 µm) using a pH and salt‐actuated microrobot controlled under tele‐operation. By adding iron oxide (Fe 3 O 4 ) nanoparticle to the dual‐responsive hydrogel as shown in Figure 4a , the soft microrobots are endowed with magnetic properties. Under the action of the magnetic field, the microrobot can reach the designated position. We simulated the changes in environmental conditions and chose the pH and salt concentration that the soft microrobot can perform the logic operation based on AND gate or OR gate as shown in Figure 4b . First, the microrobot was guided to the Teflon rod using a pulling motion at a 0.2 m NaCl aqueous solution of pH = 2. As the microrobot captured the Teflon rod and added the aqueous solution of pH = 6, the hole in the middle began to close. When the concentration of NaCl aqueous solution was reduced to 0.1 m , the hole in the middle fully closed and the Teflon rod was gripped tightly by the microrobot. Then, the trapped Teflon rod was then freely moved along the desired path using a swimming motion. Finally, with the addition of high‐concentration NaCl solution, the hole began to open, releasing the trapped Teflon rod. The process is based on an AND gate ( Figure \n \n 5 \n ). Similarly, the Teflon rod can be captured and released by the same microrobot based on an OR gate just by changing the pH of the solution ( Figure \n \n 6 \n ). Hence, the soft microrobot could analyze the environmental stimuli and choose the logic gate intelligently. Figure 5 a) Schematic illustrations and images of the process that a Teflon rod was transferred by the magnetic soft microrobot with adaptive logic gates based on an AND gate. b) The area of the AND gate and truth tables. Figure 6 a) Schematic illustrations and images of the process that a Teflon rod was transferred by the magnetic soft microrobot with adaptive logic gates based on an OR gate. b) The area of the OR gate and truth tables."
} | 5,576 |
39955649 | PMC11962713 | pmc | 9,170 | {
"abstract": "Abstract The unique structure of carbon nanotubes (CNTs) endows them with exceptional electrical and mechanical properties, along with a high surface area, making them highly beneficial for use as flexible, high‐performing thermoelectric materials. As a result, the application of CNTs in the thermoelectric field has become increasingly widespread. Considering the rapid advancements in this field, this review offers a timely overview of the most recent progress on CNT‐based thermoelectric materials and devices over the past five years. This review begins by introducing the fundamental concepts and thermoelectric mechanisms of CNT‐based thermoelectric materials. Then new strategies are explored to enhance their thermoelectric performance, focusing on doping and composites, while emphasizing the importance of CNT stability as a key research area. Additionally, the latest design concepts and expanded application scenarios for flexible and wearable CNTs‐based thermoelectric devices are summarized. Finally, the current challenges are addressed and future directions for the development of CNT‐based thermoelectric materials and devices are discussed.",
"introduction": "1 Introduction The development of human society is inextricably linked to energy, with the majority currently sourced from natural resources, particularly fossil fuels and coal. To mitigate the ongoing energy depletion crisis, extensive research over the past few decades has focused on harnessing green energy sources such as wind, solar, and hydropower for electricity generation. However, more than half of the energy generated in these processes is lost as heat, making the reuse of waste heat a critical area of research. Thermoelectric technology, which leverages the Seebeck effect to directly convert heat into electricity, has gained significant attention in this context. [ \n \n 1 \n , \n 2 \n , \n 3 \n , \n 4 \n , \n 5 \n \n ] Wearable and flexible thermoelectric generators (W‐ and F‐TEGs) have emerged as focal points of research, [ \n \n 6 \n \n ] particularly due to the availability of constant heat sources (such as body heat) and the potential for miniaturization and wearability of thermoelectric materials and devices. [ \n \n 7 \n \n ] These devices offer significant commercial potential as continuous power sources for medical sensors, human–machine interfaces, and connected devices. [ \n \n 8 \n \n ] The basic structure of a TEG, illustrated in Figure \n \n 1 a , consists of a closed‐loop circuit that connects a pair of p‐type (where holes are the majority carriers) and n‐type (where electrons are the majority carriers) materials. [ \n \n 1 \n \n ] A single TEG generates power by utilizing the temperature difference (Δ T ) across the thermoelectric materials, where carriers move from the hot side to the cold side, producing an output voltage ( V ) to power a load. A TEG can include multiple thermoelectric pairs, connected thermally in parallel and electrically in series, to multiply the Power output ( P ), as shown in Figure 1a . This enables TEGs to meet various power needs including energy storage in devices such as supercapacitors, regulation through voltage amplifiers, or direct power supply for low‐power electronics (such as sensors). [ \n \n 8 \n \n ] \n Figure 1 Review of thermoelectric performance of carbon nanotube (CNT)‐based materials. a) Schematic diagram of thermoelectric devices with multiple p–n pairs (up panel) and single p–n pair (down panel). Here, SWCNTs is abbreviated from single‐walled carbon nanotubes, MWCNTs represents multiwalled carbon nanotubes. Diagram of thermoelectric device with multiple p–n pairs. Reproduced with permission. [ \n \n 275 \n \n ] Copyright 2021, Wiley. Diagram of device with single p–n pair. Reproduced with permission. [ \n \n 118 \n \n ] Copyright 2021, Royal Society of Chemistry. Structure of SWCNTs and MWCNTs. Reproduced with permission. [ \n \n 169 \n \n ] Copyright 2022, De Gruyter. b) Summary of the power factor ( S \n 2 \n σ ) as a function of absolute Seebeck coefficient value (| S |) for CNT films reported within five years. [ \n \n 68 \n , \n 71 \n , \n 72 \n , \n 74 \n , \n 77 \n , \n 87 \n , \n 88 \n , \n 93 \n , \n 98 \n , \n 108 \n , \n 114 \n , \n 115 \n , \n 120 \n , \n 131 \n , \n 132 \n , \n 133 \n , \n 142 \n , \n 145 \n , \n 151 \n , \n 163 \n , \n 168 \n , \n 185 \n , \n 205 \n , \n 209 \n , \n 226 \n , \n 248 \n , \n 259 \n , \n 260 \n , \n 265 \n , \n 266 \n , \n 267 \n \n ] c) Summary of S \n 2 \n σ for CNT fibers reported within five years. [ \n \n 116 \n , \n 135 \n , \n 152 \n , \n 162 \n , \n 210 \n , \n 227 \n , \n 231 \n , \n 265 \n , \n 269 \n , \n 270 \n , \n 271 \n , \n 272 \n \n ] d) Summary of the output power density ( ω ) of reported flexible thermoelectric generators (F‐TEGs) based on CNT films and fibers within five years. [ \n \n 41 \n , \n 74 \n , \n 77 \n , \n 98 \n , \n 102 \n , \n 104 \n , \n 108 \n , \n 114 \n , \n 115 \n , \n 116 \n , \n 117 \n , \n 120 \n , \n 131 \n , \n 132 \n , \n 133 \n , \n 135 \n , \n 145 \n , \n 152 \n , \n 162 \n , \n 163 \n , \n 168 \n , \n 185 \n , \n 203 \n , \n 205 \n , \n 209 \n , \n 210 \n , \n 227 \n , \n 231 \n , \n 242 \n , \n 248 \n , \n 260 \n , \n 265 \n , \n 266 \n , \n 267 \n , \n 270 \n , \n 271 \n , \n 274 \n \n ] \n Thermoelectric materials are the key determinant of a TEGs’ performance. [ \n \n 8 \n \n ] Their efficiency is largely measured by the figure of merit, ZT , with higher ZT , corresponding to better P . ZT is defined by ZT = S \n 2 \n σT / κ , where S represents the Seebeck coefficient, σ is the electrical conductivity, S \n 2 \n σ is the power factor, T is the absolute temperature in Kelvin, and κ is the thermal conductivity. [ \n \n 9 \n \n ] Typically, a higher S \n 2 \n σ and lower κ result in a larger ZT . The total κ consists of both electronic thermal conductivity ( κ \n e ) and lattice thermal conductivity ( κ \n l ). However, optimizing ZT is challenging due to the strong coupling between parameters, such as σ , S , and κ \n e . [ \n \n 10 \n \n ] In general, the S can be expressed as S = 8 π 2 k B 2 3 e h 2 m ∗ T ( π 3 n c ) 2 3 , where k \n B , e , h , m *, and n \n c represent the Boltzmann constant, charge, Planck's constant, effective carrier mass, and carrier concentration, respectively. [ \n \n 10 \n \n ] \n σ is defined as σ = n \n c \n eµ , where µ is the carrier mobility, and κ \n e = LσT , with L being the Lorenz constant. [ \n \n 10 \n \n ] Thus, practical thermoelectric materials with high ZT require careful tuning of n \n c to achieve large S \n 2 \n σ , along with meticulous structural design to enhance the scattering of phonons of various wavelengths for low κ \n l . [ \n \n 10 \n \n ] However, these strategies may also scatter carriers, which can reduce both µ and σ . As a result, designing and optimizing thermoelectric materials remains a significant challenge. Inorganic thermoelectric materials have been studied since the 1950s due to their relatively high ZT values. [ \n \n 11 \n , \n 12 \n \n ] In the mid‐to‐high temperature range, materials such as GeTe, [ \n \n 13 \n \n ] SnSe, [ \n \n 14 \n \n ] PbTe, [ \n \n 15 \n \n ] and Cu 2 Se, [ \n \n 16 \n \n ] have demonstrated ZT values as high as 2–3 between 600 and 1000 K. [ \n \n 8 \n \n ] However, these materials typically show lower ZT values (<1) at near room temperature (below 350 K). Since W‐TEGs operate at near room temperature, developing thermoelectric materials with high ZT at this range has become a major research focus in recent years. [ \n \n 17 \n \n ] Bismuth telluride‐based thermoelectric materials are commonly used for near‐room‐temperature applications, with ZT values exceeding 1.2 for n‐type materials and 1.5 for p‐type materials. [ \n \n 18 \n , \n 19 \n \n ] However, these materials tend to be expensive, and their rigidity makes them difficult to apply in highly flexible TEGs unless produced as thin films or fibers, [ \n \n 20 \n , \n 21 \n , \n 22 \n \n ] which can compromise their intrinsic high ZT . To overcome these limitations, recent research has focused on alternative near‐room‐temperature thermoelectric materials that offer higher ductility, such as n‐type Ag 2 Q [ \n \n 23 \n \n ] and p‐type AgCuQ (Q = S, Se, Te), [ \n \n 24 \n \n ] as well as Mg 3 (Bi, Sb) 2 . [ \n \n 25 \n , \n 26 \n \n ] These materials typically have greater plasticity and/or lower cost than bismuth telluride, while maintaining comparable ZT values. However, inorganic thermoelectric materials share common drawbacks, [ \n \n 27 \n \n ] including high toxicity, high processing costs, and inherent rigidity, which limit their practicality in F‐TEGs. [ \n \n 28 \n \n ] As a result, organic thermoelectric materials, such as conducting polymers (e.g., poly(3,4‐ethylenedioxythiophene):poly(styrene sulfonate), PEDOT:PSS), [ \n \n 29 \n \n ] have recently attracted attention for their high flexibility and low κ . Nevertheless, their relatively low S and high cost still pose challenges for widespread practical applications. In TEGs, the impact of κ in thermoelectric materials becomes less critical if the temperature differential between the hot and cold sides is well maintained. This is especially true for many TEGs, particularly F‐TEGs, which are designed to power low‐grade wearable electronic devices. For such applications, the focus shifts to identifying thermoelectric materials with high σ , low cost, and good flexibility. Carbon‐based materials are considered promising candidates due to their high σ , excellent processability, low cost, light weight, and environmentally friendly nature. [ \n \n 30 \n , \n 31 \n , \n 32 \n \n ] They are also promising candidates for hybridizing with many other thermoelectric materials to improve the overall thermoelectric performance. [ \n \n 33 \n , \n 34 \n , \n 35 \n , \n 36 \n \n ] Among carbon materials, carbon nanotubes (CNTs) have attracted significant attention since their discovery due to their unique 1D electron transport properties. [ \n \n 35 \n , \n 36 \n , \n 37 \n , \n 38 \n , \n 39 \n , \n 40 \n , \n 41 \n , \n 42 \n , \n 43 \n , \n 44 \n , \n 45 \n , \n 46 \n , \n 47 \n , \n 48 \n , \n 49 \n , \n 50 \n , \n 51 \n , \n 52 \n , \n 53 \n , \n 54 \n , \n 55 \n , \n 56 \n , \n 57 \n , \n 58 \n , \n 59 \n , \n 60 \n , \n 61 \n , \n 62 \n , \n 63 \n , \n 64 \n , \n 65 \n , \n 66 \n , \n 67 \n , \n 68 \n , \n 69 \n , \n 70 \n , \n 71 \n , \n 72 \n , \n 73 \n , \n 74 \n , \n 75 \n , \n 76 \n , \n 77 \n , \n 78 \n , \n 79 \n , \n 80 \n , \n 81 \n , \n 82 \n , \n 83 \n , \n 84 \n , \n 85 \n , \n 86 \n , \n 87 \n , \n 88 \n , \n 89 \n , \n 90 \n , \n 91 \n , \n 92 \n , \n 93 \n , \n 94 \n , \n 95 \n , \n 96 \n , \n 97 \n , \n 98 \n , \n 99 \n , \n 100 \n , \n 101 \n , \n 102 \n , \n 103 \n , \n 104 \n , \n 105 \n , \n 106 \n , \n 107 \n , \n 108 \n , \n 109 \n , \n 110 \n , \n 111 \n , \n 112 \n , \n 113 \n , \n 114 \n , \n 115 \n , \n 116 \n , \n 117 \n , \n 118 \n , \n 119 \n , \n 120 \n , \n 121 \n , \n 122 \n , \n 123 \n , \n 124 \n , \n 125 \n , \n 126 \n , \n 127 \n , \n 128 \n , \n 129 \n , \n 130 \n , \n 131 \n , \n 132 \n , \n 133 \n , \n 134 \n , \n 135 \n , \n 136 \n , \n 137 \n , \n 138 \n , \n 139 \n , \n 140 \n , \n 141 \n , \n 142 \n , \n 143 \n , \n 144 \n , \n 145 \n , \n 146 \n , \n 147 \n , \n 148 \n , \n 149 \n , \n 150 \n , \n 151 \n , \n 152 \n , \n 153 \n , \n 154 \n , \n 155 \n , \n 156 \n , \n 157 \n , \n 158 \n , \n 159 \n , \n 160 \n , \n 161 \n , \n 162 \n , \n 163 \n , \n 164 \n , \n 165 \n , \n 166 \n , \n 167 \n , \n 168 \n , \n 169 \n , \n 170 \n , \n 171 \n , \n 172 \n , \n 173 \n , \n 174 \n , \n 175 \n , \n 176 \n , \n 177 \n , \n 178 \n , \n 179 \n , \n 180 \n , \n 181 \n , \n 182 \n , \n 183 \n , \n 184 \n , \n 185 \n , \n 186 \n , \n 187 \n , \n 188 \n , \n 189 \n , \n 190 \n , \n 191 \n , \n 192 \n , \n 193 \n , \n 194 \n , \n 195 \n , \n 196 \n , \n 197 \n , \n 198 \n , \n 199 \n , \n 200 \n , \n 201 \n , \n 202 \n , \n 203 \n , \n 204 \n , \n 205 \n , \n 206 \n , \n 207 \n , \n 208 \n , \n 209 \n , \n 210 \n , \n 211 \n , \n 212 \n , \n 213 \n , \n 214 \n , \n 215 \n , \n 216 \n , \n 217 \n , \n 218 \n , \n 219 \n , \n 220 \n , \n 221 \n , \n 222 \n , \n 223 \n , \n 224 \n , \n 225 \n , \n 226 \n , \n 227 \n , \n 228 \n , \n 229 \n , \n 230 \n , \n 231 \n , \n 232 \n , \n 233 \n , \n 234 \n , \n 235 \n , \n 236 \n , \n 237 \n , \n 238 \n , \n 239 \n , \n 240 \n , \n 241 \n , \n 242 \n , \n 243 \n , \n 244 \n , \n 245 \n , \n 246 \n , \n 247 \n , \n 248 \n , \n 249 \n , \n 250 \n , \n 251 \n , \n 252 \n , \n 253 \n , \n 254 \n , \n 255 \n , \n 256 \n , \n 257 \n , \n 258 \n , \n 259 \n , \n 260 \n \n ] CNTs are one of the allotropes of carbon, formed by rolling graphene sheets with covalent sp 2 ‐hybridized C─C bonds into a honeycomb‐structured, 1D nanotube. [ \n \n 38 \n \n ] Due to the unique quasi‐1D structure of CNTs, phonon transport is highly facilitated, resulting in CNTs possessing an exceptionally high κ of 2000–3000 W m −1 K −1 . [ \n \n 261 \n \n ] This unique structure also provides a high aspect ratio, promoting charge transport while limiting the Seebeck effect, leading to high σ and low S of <30 µV K −1 . [ \n \n 169 \n , \n 204 \n , \n 262 \n , \n 263 \n , \n 264 \n \n ] Over the decades, various strategies have been reported to enhance the thermoelectric performance of CNT films and fibers by reducing κ while increasing S and σ . [ \n \n 169 \n , \n 204 \n , \n 263 \n , \n 264 \n \n ] Figure 1b summarizes the relationship between S \n 2 \n σ and absolute Seebeck coefficient values (| S |) for CNT‐based films within the last five years. [ \n \n 68 \n , \n 71 \n , \n 72 \n , \n 74 \n , \n 77 \n , \n 87 \n , \n 88 \n , \n 93 \n , \n 98 \n , \n 108 \n , \n 114 \n , \n 115 \n , \n 120 \n , \n 131 \n , \n 132 \n , \n 133 \n , \n 142 \n , \n 145 \n , \n 151 \n , \n 163 \n , \n 168 \n , \n 185 \n , \n 205 \n , \n 209 \n , \n 226 \n , \n 248 \n , \n 259 \n , \n 260 \n , \n 265 \n , \n 266 \n , \n 267 \n \n ] Despite a series of effective treatments, pure CNT‐based films still exhibit relatively low S \n 2 \n σ (<3 µW cm −1 K −2 ) due to their low S . To further enhance their thermoelectric performance, treated CNTs are often combined with other functional materials, including inorganic nanomaterials, organic materials, and organic/inorganic hybrids. [ \n \n 118 \n \n ] This approach potentially improves S and decouples S from σ by forming heterointerfaces that invoke an energy filtering effect, [ \n \n 268 \n \n ] thereby filtering out low‐energy carriers. This process can lead to significant improvements in S \n 2 \n σ . For instance, a high S \n 2 \n σ value of 10.79 µW cm −1 K −2 was achieved by forming PANI/SWCNT films, where single‐walled CNTs (SWCNTs) are composited with polyaniline (PANI). [ \n \n 248 \n \n ] \n Table \n \n 1 \n further summarizes the detailed thermoelectric performance of CNT‐based films over the last five years. Regarding CNT‐based thermoelectric fibers, Figure 1c provides an overview of the S \n 2 \n σ of CNT fibers over the past five years. [ \n \n 116 \n , \n 135 \n , \n 152 \n , \n 162 \n , \n 210 \n , \n 227 \n , \n 231 \n , \n 265 \n , \n 269 \n , \n 270 \n , \n 271 \n , \n 272 \n \n ] Due to their 1D characteristics, fibers are well‐suited for wearable or wearable thermoelectric textile‐type W‐TEGs, [ \n \n 273 \n \n ] offering superior wearability compared to film materials. The optimization strategies are almost the same, for example, an S \n 2 \n σ value of 34.25 µW cm −1 K −2 was achieved with Sb 2 Te 3 /CNT composite fibers. [ \n \n 265 \n \n ] \n Table \n \n 2 \n provides a detailed summary of the thermoelectric performance of CNT‐based fibers over the past five years. Additionally, we review the performance of F‐TEGs based on CNT films and fibers. Figure 1d presents the reported power output density ( ω ) of CNT film‐ and fiber‐based F‐TEGs over the last five years. [ \n \n 41 \n , \n 74 \n , \n 77 \n , \n 98 \n , \n 102 \n , \n 104 \n , \n 108 \n , \n 114 \n , \n 115 \n , \n 116 \n , \n 117 \n , \n 120 \n , \n 131 \n , \n 132 \n , \n 133 \n , \n 135 \n , \n 145 \n , \n 152 \n , \n 162 \n , \n 163 \n , \n 168 \n , \n 185 \n , \n 203 \n , \n 205 \n , \n 209 \n , \n 210 \n , \n 227 \n , \n 231 \n , \n 242 \n , \n 248 \n , \n 260 \n , \n 265 \n , \n 266 \n , \n 267 \n , \n 270 \n , \n 271 \n , \n 274 \n \n ] These results indicate that flexible thermoelectric materials and TEGs based on CNTs hold significant practical value. Table 1 Summary of thermoelectric properties of carbon nanotube (CNT)‐based films reported within five years. Here, the units for κ , σ , S , and S \n 2 \n σ , are W m −1 K −1 , S cm −1 , µV K −1 , and µW cm −1 K −2 , respectively. Abbreviations: SWCNT, single‐walled carbon nanotube; FcMA, (dimethylamino) methyl group; MOF, metal–organic framework; CNT, carbon nanotube; SFX‐2, C 33 H 21 O; Spiro‐MeOTAD, (C 65 H 40 O 8 N 4 ) \n n \n ; MXene, Ti 3 C 2 T \n x \n –H 2 O; [HMIM][BF 4 ], 1‐hexyl‐3‐methylimidazolium tetrafluoroborate; CNTs, carbon nanotubes; PBDTT‐FTTE, (C 49 H 57 FO 2 S 6 ) \n n \n ; MWCNT, multiwalled carbon nanotube; P(BDTC), macrocyclic crown ether; nCB, nanocarbon black; TPP, riphenylphosphine; PYB, pyridineborane; DMSO, dimethyl sulfoxide; PEDOT:PSS, poly(3,4‐ethylene dioxythiophene): polystyrene sulfonate; PANI, emeraldine base polyaniline; PEDOT‐Tos, poly(3,4‐ethylenedioxytiophene)‐tosylate; a‐SWCNT, acidified‐single‐walled carbon nanotube; PSSH, polystyrene sulfonate; APA, aniline tetramer‐ b ‐polyethylene glycol‐ b ‐aniline tetramer; a‐TM, acid‐treatment of tourmaline; E7, a eutectic nematic mixture; CAS, sulfurochloridic acid; PEDOT, poly(3,4‐ethylenedioxytiophene); PE, polyethylene. Material Type \n σ \n \n S \n \n S 2 σ \n \n κ \n \n ZT \n Refs. SWCNTs/nCB p 1710 4644.2 151 – – [ 274 ] CAS‐MWCNT p 10 000 23 46.6 45.9 0.0304 [ 186 ] PANI/SWCNT – DMSO p 3980 55 10.79 – – [ 248 ] PANI/SWCNT – DMSO n 3980 −50 10.31 60 0.005 [ 248 ] [HMIM][BF 4 ]/CNTs p 1154 81 7.62 3.84 0.060 [ 131 ] CNT/Se/PEDOT:PSS p 1825.02 66.38 5.842 5.64 0.034 [ 267 ] SWCNT/FcMA n 2674.86 −46.07 5.6754 52.98 0.0032 [ 77 ] PEDOT:PSS/SWCNT p 4717.8 32.6 5.0131 0.062 0.012 [ 185 ] PEDOT:PSS/SWCNTs p 2333.7 46.2 5.00 164.4 0.0009 [ 266 ] SWCNT/E7 p 1665.5 50.48 4.2859 – – [ 87 ] PEDOT:PSS/SWCNTs n 1718.2 −49 4.1140 0.4 0.3 [ 145 ] a‐TM/SWCNT p 2837.4 34.2 3.35 – – [ 170 ] SiO 2 @MoS 2 /SWCNT p 1646.4 39.2 2.532 61.1 0.00123 [ 41 ] Spiro‐MeOTAD/SWCNT p 873.3 51.7 2.392 – – [ 168 ] PEDOT:PSS sheet/SWCNTs p 3085.4 26.95 2.24 – – [ 88 ] SWCNTs/PYB n 766.7 −48.5 2.238 40.1 0.00154 [ 117 ] SFX‐2/SWCNT p 762.2 53.4 2.186 – – [ 98 ] SWCNT/MXene p 1293.76 39.64 2.0329 9.764 0.0062 [ 132 ] SWCNTs/DMSO n 3490 −23.7 1.95 19.88 – [ 163 ] PEDOT:PSS/CNT p 1602.6 33.4 1.827 – – [ 72 ] PEDOT‐Tos/a‐SWCNT p 4731.6 21.1 1.685 – – [ 74 ] PE/PEDOT/SWCNT p 459.82 47.58 1.5881 0.0879 – [ 151 ] SnSe nanobelt/SWCNT p 586.37 49.3 1.45 14 0.003 [ 68 ] P(BDTC)/SWCNT p 864.8 39.9 1.377 – – [ 133 ] PEDOT‐Tos/Te/SWCNTs p 578.4 47.75 1.3191 – – [ 108 ] CNTs/PEDOT p 910.1 37.9 1.311 – – [ 226 ] PEDOT:PSS/SWCNT n 1410 −29.54 1.23 – – [ 205 ] PSSH/SWCNT p 3749 17.7 1.177 – – [ 259 ] APA/SWCNT n – −49.0 1.063 – – [ 93 ] PEDOT:PSS/SWCNT p 1514 25.3 0.96 – – [ 205 ] MOF/CNT n 347.4 −56.4 0.844 0.18 0.071 [ 260 ] PEDOT:PSS/SWCNT p 1562 21.9 0.739 0.3 0.07 [ 142 ] PEDOT:PSS/SWCNT p 1130 18.9 0.486 – – [ 120 ] PBDTT‐FTTE /MWCNT p 210.8 47.7 0.4821 – – [ 114 ] PEDOT:PSS/SWCNT p 510.6 20.18 0.2068 – – [ 71 ] CNT/PEDOT:PSS n 1938 −9.8 0.164 – – [ 209 ] TPP/CNTs n 48.1 −53.9 0.136 0.123 0.0091 [ 115 ] MWCNT p 84 19.8 0.0004 0.58 0.0000142 [ 64 ] John Wiley & Sons, Ltd. Table 2 Summary of thermoelectric properties of CNT‐based fibers reported within five years. Here, the units for κ , σ , S , and S \n 2 \n σ , are W m −1 K −1 , S cm −1 , µV K −1 , and µW cm −1 K −2 , respectively. Abbreviations: CNTFs, CNT‐based fibers; DWCNT, double‐walled carbon nanotube; PU, polyurethane; N‐DMBI, 4‐(1, 3‐dimethyl‐2, 3‐dihydro‐1H‐benzimidazole‐2‐yl) phenyl) dimethylamine; NBP, nickel‐backboned polymer; PEI, polyethylenimine; Au NPs, gold nanoparticles. Material Type \n σ \n \n S \n \n S \n 2 \n σ \n \n κ \n \n ZT \n Refs. Sb 2 Te 3 /CNT p 4620 85 34.25 – – [ 265 ] CNT/Bi 2 Te 3 \n n 4675 −80 27.30 – – [ 265 ] CNT p 6440.3 64.5 26.19 17 0.05 [ 270 ] N‐DMBI/CNT n 1182 −113 15.34 27.3 – [ 227 ] PEI/Au NPs/CNTFs n 2100 −80 14.00 – – [ 231 ] CNTs/PANI p 3651 59.5 12.94 – – [ 271 ] NBP/CNT p 1236.85 76.26 71.948 – – [ 272 ] PEDOT:PSS/CNT p 2110 36.4 2.80 – – [ 162 ] CNT‐modified p 747 57 2.42 63.0 0.00115 [ 269 ] DWCNT‐PU p 285.1 72.5 1.497 – – [ 135 ] CNT/PEDOT:PSS p 26.3 44 0.0536 – – [ 152 ] CNTs/PEDOT:PSS p 14.62 57.27 0.0479 – – [ 210 ] CNTFs p 17.4 51.5 0.0459 4.17 0.0000016 [ 116 ] John Wiley & Sons, Ltd. To date, significant progress has been made in CNT‐based thermoelectric materials and devices, with various practical films and fibers emerging as promising candidates for F‐TEGs, particularly in wearable technologies. Given the rapid developments in this field, periodic reviews of developments in CNT‐based thermoelectric materials and devices are crucial for guiding future research. This review, drawing on the authors’ years of research and achievements in the field, aims to provide an overview of the latest progress in CNT‐based thermoelectric technology. It focuses on exploring underlying mechanisms, optimizing thermoelectric performance, integrating advanced devices, and identifying novel applications. Finally, we address the challenges and bottlenecks that remain in CNT‐based thermoelectric materials and devices and offer fresh perspectives for future research directions."
} | 5,308 |
35492658 | PMC9048857 | pmc | 9,171 | {
"abstract": "Biomimetic synthetic functional materials are valuable for a large number of practical applications with improved or tunable performance. In this paper, we present a series of mussel-inspired biomimetic catechol-containing copolymers synthesized from dopamine methacrylamide (DMA) and 2-(2-ethoxyethoxy)ethyl acrylate (EEA) and abbreviated as poly(PDMA-PEEA). The successfully synthesized adhesive polymers allow adhering polytetrafluoroethylene (PTFE) and were used for coating PTFE particles in organic solvent and re-dispersion in an aqueous medium. Adhesive polymer coated PTFE particles were efficiently used as a nanoreactor for generating silver (Ag) metal nanoparticles (NPs).",
"introduction": "Introduction Mussels derived adhesive proteins which contain catecholic amino acids, 3,4-dihydroxy- l -phenylalanine (DOPA), are useful materials for industrial, biomedical and pharmaceutical applications owing to their characteristic binding applicability, hydrophilicity, permeability, and stability towards a wide variety of surfaces both in air and water. 1–3 Catechol moieties in DOPA has a well-known ability to form various types of chemical interactions, e.g. , cross-linking by oxidation, and hence, shows excellent adhesion properties for a wide variety of materials, such as inorganic metals, metal oxides, alloys, polymers, glass, wood and ceramics. 4,5 Strong research efforts have been made in the last decade to develop new biomimetic catechol based adhesive materials for better adhesion of components and surface coatings or modifiers motivated by the growing needs and demands of the automotive and aerospace industries. 6,7 Polymers containing catechol groups have been reportedly used for surface modification, 8–10 surface coatings, 11–13 stabilization of nanoparticles, 14–16 and adhesives. 17,18 PTFE, which is a synthetic hydrophobic fluoropolymer of tetrafluoroethylene, has several successful engineering practical applications such as non-stick coatings for cookware and other appliances, as friction-reducing lubricants, graft materials in biomedical field, chemical and electronic fields, 19–21 due to its attractive properties of low surface energy, high thermal and chemical stability, low dielectric constant, low water adsorption and high potential biological inertness. However, high hydrophobicity and poor wettability of PTFE limits its performance in real applications. Various reinforcement and modification of PTFE such as coating with functional polymer, 25,26 chemical etching, 27 ion-beam, 28 ozone treatment, 29 plasma modifications 30 have been tried to intrinsically improve the material performance. Many of the above-mentioned approaches require complicated processes and specially made instruments. To date, there has been a large number of polymer materials which could disperse extremely hydrophobic and poorly adhesive PTFE 22–24 in organic solvents, however, PTFE dispersion in an aqueous medium without using fluorinated surfactants or additional fluorinated reagents remains a challenging scope. Inspired from the reducing properties of catechol moieties, the composite materials prepared by coating PTFE particles in nm range coated with amphiphilic polymers containing catechol moieties can be exploited as nanosized reactors that can automatically reduce metal ions. In the present study, a series of adhesive copolymers, poly(PDMA–PEEA) based on dopamine methacrylamide (DMA) and 2-(2-ethoxyethoxy)ethyl acrylate (EEA) was synthesized and modified by varying copolymerization ratios. The synthesized polymers, abbreviated as poly(PDMA–PEEA) show strong adhesion properties on PTFE substrates; and the surfaces of hydrophobic PTFE particles were successfully modified with those copolymers in tetrahydrofuran solution. The surface-modified PTFE particles can be dispersed in many kinds of solvents, especially in aqueous media. Furthermore, the remaining catechol groups on the surface of modified PTFE particles show reduction property of silver ions and formed silver nanoparticle shells on the surface of them ( Fig. 1 ). Results of silver ions reduction allow tailoring the design of such composite materials as nanoreactors to replace, totally or partially expensive catalytic materials to ensure electrochemical operation. We envisage that this new class of composite materials with simple and easy preparation will facilitate tuning of properties for a host of other polymer-nanoparticles as nanoreactors without any additional reducing agents. Fig. 1 (a) Reaction scheme for the synthesis of co-polymer, poly(PDMA–PEEA). (b) Sketched representation of coating of the PTFE nanoparticles with the adhesive polymer in THF solvent (surface modification), the addition of Ag + ions in the aqueous medium containing polymer-coated PTFE particles, and plausible illustration of reduced Ag nanoparticles anchored on to hydrophilic catechol surface.",
"discussion": "Results and discussion A series of adhesive polymer, poly(PDMA–PEEA) was synthesized using DMA and EEA as monomers using thermal initiated free-radical polymerization as shown in Fig. 1a . The chemical structure of all the polymers were determined using 1 H-NMR (Fig. S2 † ). The location of proton peaks in all adhesive polymers were similar, which indicated that they all had the similar chemical shifts in the main structure. Peaks were attributed at around 8.64, 6.40–6.63 ppm, 4.10 ppm, 3.37–3.57 ppm, 2.27–2.51 ppm, 1.09–1.58 ppm, 0.86–0.88 ppm as per the chemical structure ( Fig. 1a ). The characteristics peaks of DOPA in the range of 6.40–6.63 ppm shows that DOPA groups were successfully grafted onto adhesive copolymers. Some minimal peaks above 8.64 ppm shows oxidized DOPA forms (quinone or semi-quinone). The copolymerization was further investigated by FT-IR measurements. As shown in Fig. S3, † the decay of C \n \n\n<svg xmlns=\"http://www.w3.org/2000/svg\" version=\"1.0\" width=\"13.200000pt\" height=\"16.000000pt\" viewBox=\"0 0 13.200000 16.000000\" preserveAspectRatio=\"xMidYMid meet\"><metadata>\nCreated by potrace 1.16, written by Peter Selinger 2001-2019\n</metadata><g transform=\"translate(1.000000,15.000000) scale(0.017500,-0.017500)\" fill=\"currentColor\" stroke=\"none\"><path d=\"M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z\"/></g></svg>\n\n C stretching vibrations of methacrylate or acrylate group at 1638 cm −1 after polymerization, and the appearance of broad absorbance between 3600 and 3100 cm −1 which is ascribed to N–H/O–H stretching vibrations in catechol moiety in copolymer. The weight-average molecular weight of ( M w ), the number-average molecular weight ( M n ) and the molecular weight distribution (PDI) were determined by GPC measurements and summarized in Table 1 . Poly(PDMA–PEEA) had high M w and M n values in the range of 1.41 × 10 5 to 1.97 × 10 5 and 6.83 × 10 3 to 4.61 × 10 5 , respectively, and the PDI ranged from 2.07 to 3.27. Corresponding unimodal plots from the GPC results are shown in Fig. S4. † Molecular weight of adhesive copolymers, poly(PDMA–PEEA) with different DMA and EEA ratios Sample \n M \n n [kg mol −1 ] \n M \n w [kg mol −1 ] \n M \n w / M n 1 : 2.5 6.82 14.1 2.1 1 : 5 13.6 33.6 2.5 1 : 7.2 79.2 19.7 2.5 1 : 10 18.7 61.2 3.3 The solubility of adhesive polymers were tested by dissolving them in three groups of solvents: 1 nonpolar solvent such as: toluene, hexane, diethyl ether, dichloromethane, 2 polar protic solvents such as distilled water, methanol, ethanol, isopropanol, and 3 polar aprotic solvent such as tetrahydrofuran (THF), dimethyl formamide, dimethyl acetamide (DMAc), N -methyl-2-pyrrolidone, dimethyl sulfoxide, and dioxane at room temperature. All the adhesive polymers were soluble in polar aprotic solvent and insoluble in non-polar solvent. As the content of EEA in the copolymer increased, the adhesive polymer showed fair solubility in aqueous medium due to increase in hydrophilicity. The thermal degradation of all the adhesive polymers, poly(PDMA–PEEA) with different monomers ratio was investigated by thermogravimetric analysis (TGA) in a nitrogen atmosphere at a heating rate of 10 °C min −1 , and the 5% and 10% weight-loss temperatures, T 5 and T 10 , were determined ( Fig. 2a ). The slight weight loss of about 0.6% below 110 °C is due to the moisture that had been absorbed from the atmosphere during the handling of samples for testing. As revealed by the graphs, the main degradation of the adhesive copolymer from about 200–480 °C corresponds to the decomposition of DMA and EEA. Residual weight loss of 5% and 10% of the network polymers are in the ranges of 282.64–336.72 °C and 319.35–356.58 °C, respectively. Addition of higher amounts of EEA enhances the thermal stability of the co-polymer matrix possibly which is clearly visible in the thermograms. The overall weight loss is consistent with the copolymer composition taking into account the experimental errors related to the measurement and the sample preparation. Fig. 2 Thermal properties of adhesive copolymer, poly(PDMA–PEEA). (a) TGA and (b) DSC plots respectively. The differential scanning calorimetry (DSC) experiments were conducted to study the phase changes that were indicated by the glass transition temperature ( T g ) in the adhesive copolymers, poly(PDMA–PEEA). In a typical measurement, the polymer samples were heated at 10 °C min −1 from room temperature up to 100 °C as 1st run, then again cooled down to −100 °C as 2nd run and finally heated up to 150 °C as 3rd run. T g values were calculated from the 3rd run as the cross-point of the heat capacity change observed during the transition from glassy state to rubbery state in the DSC trace. The DSC results are summarized in Table 2 , and the corresponding profiles are shown in Fig. 2b . As evident in the DSC curves of all the polymer specimens, the T g values ranged from −7.2 °C to −47.3 °C. The combined results of both DSC and TGA profiles show that no unusual phase changes or weight losses occur within the temperature range of 30 to 200 °C, which makes the adhesive copolymers thermally stable and useful over several synthetic adhesives which readily decomposes over lower range of temperatures. Thermal properties of poly(PDMA–PEEA) with different EEA contents as a measurements of force–displacement values Sample \n T \n 5 [°C] \n T \n 10 [°C] \n T \n g [°C] 1 : 2.5 282 319 −7.2 1 : 5 327 348 −47.3 1 : 7.2 337 357 −39.6 1 : 10 328 356 −43.2 The adhesive strength of poly(PDMA–PEEA) polymers were determined using force–displacement curves obtained after lubricating adhesive polymer on to rectangular glass and PTFE sheet or PTFE and PTFE substrates pulled apart at an elongation speed of 1.0 mm min −1 at room temperature. The results ( Fig. 3a and b ) shows that on poly(PDMA–PEEA) coated PTFE – PTFE sheets or PTFE-glass substrates, the force increases initially quite steeply, and in that range, high force is required to overcome the friction and adhesion at the interface, as well as the internal stiffness of the adhesive polymer. Therefore, in the initial range (up to 5 mm), the PTFE sheet moves very slightly with respect to PTFE sheet or glass substrates. The maximum stress (MPa) decreases gradually with the increase of EEA content because of the increased viscosity of adhesive copolymer. The maximum stress (MPa) and displacement values are summarized in Table 3 . Fig. 3 Mechanical properties of adhesive copolymer, poly(PDMA–PEEA). Force–displacement curves (a) using glass-co-polymer-PTFE and (b) PTFE-co-polymer-PTFE substrates respectively. Adhesive strength of poly(PDMA–PEEA) with different EEA contents coated on glass substrate and PTFE sheet (in the table, the values in parenthesis are for polymer coated on PTFE and PTFE substrates) Sample Max. stress [MPa] Displacement [mm] 1 : 2.5 5.1 × 10 −2 (3.9 × 10 −2 ) 20.0 (22.4) 1 : 5 3.5 × 10 −2 (2.6 × 10 −2 ) 22.3 (24.0) 1 : 7.2 3.2 × 10 −2 (2.5 × 10 −2 ) 22.3 (26.3) 1 : 10 2.3 × 10 −2 (2.4 × 10 −2 ) 25.2 (41.4) The differences in surface properties were also demonstrated by the changes in water contact angles measurements. The water contact angle of smooth PTFE sheets was 114°, while it was 70° after coating with adhesive copolymer, poly(PDMA–PEEA). The results of time dependence contact angle measurements shows the contact angle decreased from 70° to 30°. This suggests that the surface of PTFE became hydrophilic after coating with poly(PDMA–PEEA) which has catechol moieties. The effects of different weight ratios of monomers in poly(PDMA–PEEA) on surface properties of PTFE and time-dependence change in contact angles on glass substrates coated with adhesive polymers were also studied. The values are summarized in Table S1. † The measurement shows increase of hydrophilicity (decrease of contact angles) of adhesive polymer owing to the swelling of polymers. Scanning electron microscopy (SEM) and scanning transmission electron microscopy (STEM) images of the PTFE powdered samples and coated PTFE particles are shown in Fig. 4 . In order to disperse PTFE particles in aqueous medium, the adhesive polymer was coated onto PTFE particles in an organic solvent. The uniform coating was ensured in SEM and STEM images where the original spherical morphologies of PTFE with diameter ca. 100–150 nm was clearly distinguishable from the crumpled surface of coated PTFE with diameter ca. 150–200 nm. Fig. 4c shows the photograph of poly(PDMA–PEEA) coated PTFE particles and pristine PTFE particles in aqueous medium. Fig. 4 SEM images of (a) PTFE and (b) PTFE@poly(PDMA–PEEA); STEM images of (c) PTFE and (d) PTFE@poly(PDMA–PEEA); (e) shows the photograph of poly(PDMA–PEEA) coated PTFE particles and pristine PTFE particles in aqueous solution. Catechol moieties in adhesive polymer also acts as a reducing agent 33–35 and can stabilize metal nanoparticles surfaces because of their strong adhesion properties. In this context, the reduction of silver nanoparticles was carried out using polymer-coated PTFE particles in the procedure followed. In brief, 2 mL of as-synthesized PTFE@poly(PDMA–PEEA) particles in distilled water (5 mg mL −1 ) was mixed with 2 mL of aqueous AgNO 3 solution. The color of PTFE@poly(PDMA–PEEA) particles changed to orange within few minutes. The solution was kept at 25 °C for overnight in order to react catechol moieties from poly(PDMA–PEEA) completely. The mixture was subjected to centrifugation and washed with distilled water 3 times in order to remove excess of Ag ions. The particles were redispersed in water for further analysis. Fig. 5 shows the photograph of orange Ag nanoparticles adsorbed on the surface of PTFE@poly(PDMA–PEEA) suspended in water, and an absorption spectra of solution of PTFE@poly(PDMA–PEEA)@Ag and poly(PDMA–PEEA). As shown in Fig. 5b , the absorption peak at around 300 nm in both the spectra is attributed to absorption of catechol moieties in poly(PDMA–PEEA). Additionally, there is a strong absorption peak at around 400 nm in the spectrum of PTFE@poly(PDMA–PEEA)@Ag, which was attributed to the surface plasmonic resonance (SPR) due to reduction of Ag ions to Ag NPs. Fig. 5c shows TEM image of Ag NPs on the surface of poly(PDMA–PEEA) coated PTFE particles. Furthermore, the spectroscopic evidence (Fig. S5 † ) shows the uniform coating of PTFE particles by poly(PDMA–PEEA). This composite PTFE@poly(PDMA–PEEA) particles act as a nanoreactors for carrying out reduction of inorganic Ag ions without use of any additional reducing agents or surfactants as stabilizers as excess of catechol moieties in adhesive polymer that were not oxidized in reducing Ag ions adhered to the surface of Ag NPs and stabilized them. Fig. 5 (a) shows UV-Vis spectra of THF solutions of poly(PDMA–PEEA) and PTFE@poly(PDMA–PEEA)@Ag nanoparticles, whereas, (b) photograph and (c) TEM images of PTFE@poly(PDMA–PEEA)@Ag nanoparticles dispersed in water. In conclusion, inspired by biological mussel-adhesion phenomenon, we present the simple, mild and facile chemical route for the synthesis of biomimetic adhesive polymer poly(PDMA–PEEA) from DMA and EEA as starting monomers in presence of thermal initiator AIBN. As dopamine-based polymers have adhesive and mild reductant abilities for metal salts due to the presence of abundant catechol and amine groups, therefore, neither special surface modification procedures of templates nor additional toxic or reducing agents are needed in this procedure. The synthesized adhesive polymers were directly used for coating hydrophobic polytetrafluoroethylene (PTFE) particles as template spheres in an organic solvent, and then dispersing coated PTFE particles in an aqueous medium. Silver precursor ions Ag + are added and absorbed onto the surfaces of poly(PDMA–PEEA)@PTFE composite spheres by the active catechol and amine groups of poly(PDMA–PEEA) coating. Meanwhile, the absorbed Ag + ions are in situ reduced to Ag NPs and are home positioned. In other words, the adhesive polymer-coated PTFE composite particles were applied as a nanoreactor for generating silver metal nanoparticles (Ag NPs) without additional reducing agents as the catechol moieties in the copolymer acted as reductants and stabilizers for dispersing inorganic Ag NPs. It is possible to envisage that the composite nanostructure based on adhesive coating will contribute in simpler and facile models for surface modification of hydrophobic surfaces, and for inorganic NPs reductions eliminating the use of complex systems for catalytic reductions or miscellaneous applications."
} | 4,399 |
33478390 | PMC7818742 | pmc | 9,172 | {
"abstract": "Background Forest trees have important economic and ecological value. As a model tree, poplar has played a significant role in elucidating the molecular mechanisms underlying tree biology. However, a lack of mutant libraries and time-consuming stable genetic transformation processes severely limit progress into the functional characterization of poplar genes. A convenient and fast transient transformation method is therefore needed to enhance progress on functional genomics in poplar. Methods A total of 11 poplar clones were screened for amenability to syringe infiltration. Syringe infiltration was performed on the lower side of the leaves of young soil-grown plants. Transient expression was evaluated by visualizing the reporters β-glucuronidase (GUS) and green fluorescent protein (GFP). The experimental parameters of the syringe agroinfiltration were optimized based on the expression levels of the reporter luciferase (LUC). Stably transformed plants were regenerated from transiently transformed leaf explants through callus-induced organogenesis. The functions of Populus genes in secondary cell wall-thickening were characterized by visualizing lignin deposition therein after staining with basic fuchsin. Results We greatly improved the transient transformation efficiency of syringe Agrobacterium infiltration in poplar through screening for a suitable poplar clone from a variety of clones and optimizing the syringe infiltration procedure. The selected poplar clone, Populus davidiana × P. bolleana , is amenable to Agrobacterium syringe infiltration, as indicated by the easy diffusion of the bacterial suspension inside the leaf tissues. Using this technique, we localized a variety of poplar proteins in specific intracellular organelles and illustrated the protein–protein and protein–DNA interactions. The transiently transformed leaves could be used to generate stably transformed plants with high efficiency through callus induction and differentiation processes. Furthermore, transdifferentiation of the protoxylem-like vessel element and ectopic secondary wall thickening were induced in the agroinfiltrated leaves via the transient overexpression of genes associated with secondary wall formation. Conclusions The application of P. davidiana × P. bolleana in Agrobacterium syringe infiltration provides a foundation for the rapid and high-throughput functional characterization of Populus genes in intact poplar plants, including those involved in wood formation, and provides an effective alternative to Populus stable genetic transformation. Supplementary Information The online version contains supplementary material available at 10.1186/s12870-021-02833-w.",
"conclusion": "Conclusions By widely exploring suitable Populus clones for syringe infiltration and optimizing the experimental parameters, we developed a syringe agroinfiltration assay in poplar. The highest transient expression in this study was obtained at 5 dpi when P. davidiana × bolleana leaves LPI 4 from plants PI 11–12 were infiltrated with A. tumefaciens strain GV3101 cells suspended in infiltration media containing 1.6 mM AS. The infiltrated leaves in one single plant were sufficient for both RNA and protein analysis. This approach will be useful for the rapid and high-throughput characterization of Populus genes, such as analyses of the subcellular localization of gene products and the interaction between proteins and proteins or DNA, the production of stable transformants, and the elucidation of gene biological function and molecular mechanisms, e.g., in the developmental process of protoxylem tracheary elements and the biosynthesis of SCW. Since the transient transformation is conducted in intact plants, this system allows gene function to be elucidated in diverse genetically modified backgrounds, especially in overexpression transgenic lines, RNAi-based gene silencing lines, artificial microRNA-based gene silencing lines, and genome editing lines, either via transient overexpression or silencing of the target genes through syringe agroinfiltration. This makes it possible to manipulate multiple genes in perennial trees, in which crossing between mutant (or transgenic) lines normally takes years.",
"discussion": "Discussion Agroinfiltration has been widely used for high-throughput gene functional studies in many species due to its simplicity, speed, and efficiency [ 16 , 31 – 35 , 60 ]. Although Agrobacterium vacuum infiltration had been established in hybrid aspen P. tremula × tremuloides [ 29 ], the complex operation and typically weak expression have limited its usefulness in functional genomics research in poplar. In this study, we enhanced the Agrobacterium syringe infiltration method using the aspen hybrid clone, P. davidiana × bolleana . Throughout the optimization of the key experimental conditions, this clone exhibited a high level of transient expression and was as easy to work with as the more widely used N. benthamiana . The high transformation efficiency enabled subcellular localization of the Populus proteins, allowing protein–protein interactions and transcriptional regulation analysis to be fulfilled in a homologous plant system. Furthermore, this method provided an effective alternative to stable genetic transformation as well as a new approach for characterizing the genes involved in secondary wall formation in poplar. The aspen hybrid clone P. davidiana × bolleana , also called Shanxin Yang in Chinese, is widely grown in the northern part of China. It was selected as the preferential clone for Agrobacterium syringe infiltration since it was the most easily infiltrated and showed relatively high transient expression efficiency. In this clone, the bacterial suspension spread easily through the leaf lamina, which was previously reported to be a key factor for high-level transient expression in agroinfiltration because it maximizes the physical access of the agrobacteria to leaf cells [ 12 , 17 , 61 ]. This speculation is supported by data in Arabidopsis [ 16 ], grapevine [ 35 ], and potato [ 33 ], in which the highest transformation efficiency is consistent with the amenability of the agrobacterial suspension to diffuse inside the leaf tissue, as demonstrated in the grapevine cultivar ‘Aleatico’ and potato cultivar ‘Katahdin’. Through investigating the inner structure of the leaves of all the tested poplar clones, we found that the spreadability of the agrobacterial suspension was associated with the volume of the intercellular air spaces within the leaves, which facilitated agrobacterial spread inside the leaf, and sometimes over the vein networks, since the air there was easily replaced with the agrobacterial suspension with gentle pressure on the lower side of the leaf. Good spreadability typically resulted in a high level of transient expression, as demonstrated in clones P. davidiana × bolleana and P. alba var. pyramidalis (Fig. 1 and Fig. S 1 ). However, there was an exception in that the clone P. trichocarpa showed the largest intercellular air space and good spreadability, but a very low level of transient expression (Fig. 1 and Fig. S 1 ). We proposed that, compared with the other clones, the vast and continuous air spaces made the leaves of P. trichocarpa more likely to be damaged severely during agroinfiltration. With the vast intercellular air space, the weight of the large amount of bacterial suspension within the leaves caused separation of the lower epidermis from the rest of the leaf tissue, as described in the Results section. As a result of the poor physiological state of the leaves, transformation of the leaf cells often failed despite the wide spread of the agrobacterial suspension inside the leaves (Fig. 1 and Fig. S 1 ). In addition to the volume of the intercellular air space within the leaves, we found that the transient transformation efficiency of agroinfiltration was also affected by the arrangement of the mesophyll cells and leaf vein networks (Fig. 1 ). In the clone P. davidiana × bolleana , the loosely arranged mesophyll cells afforded them a better chance to make contact with the agrobacterial cells and then be transformed. On the contrary, the relatively smaller and compartmented intercellular space, and the compacted mesophyll cells, restricted the spread of the infiltrated suspension and transformation of the leaf cells in clones P. alba × glandulosa ‘84 K’, P. tomentosa ‘741’, and P. euramericana ‘74/76’. Additionally, the restriction of the agrobacterial suspension by leaf vein networks has also been demonstrated in lettuce and tomato [ 16 ]. These data demonstrate that the amenability of a plant to syringe agroinfiltration is associated with the interior structure of the leaves. Interestingly, many of the factors reported to be important for vacuum agroinfiltration in hybrid aspen P. tremula × tremuloides , such as bacterial density, growth stages, and infiltration medium [ 29 ], did not have obvious effects in aspen hybrid P. davidiana × bolleana under our conditions (Fig. S 3 ). In this study, we found that the physiological condition of the plants played essential roles in efficient syringe agroinfiltration in P. davidiana × bolleana . Specifically, we learned that the young aspen hybrid plants, which underwent 3 weeks of growth on MS medium in a growth chamber and then 2 weeks of growth in soil in a climate chamber, reaching an approximate plant age of PI 12, exhibited the highest levels of expression efficiency (Fig. 2 a). Among leaves of different ages in plant PI 12, LPI 4 was the easiest to infiltrate and showed the highest expression level (Fig. 2 b). There are two possible explanations for this result. First, the good performance of leaf LPI 4 from plant PI 12 (Figs. 2 a–b) was attributed to its specialized physiological state. This leaf was normally initiated and had grown to less than 1 cm in length on MS medium under the growth chamber conditions and developed more rapidly in the soil under climate chamber conditions, and had fully expanded by the time of infiltration. Further, its vigorous cells that recently experienced rapid cell expansion facilitated a high level of transient transformation, as previously suggested [ 16 , 33 ]. Second, this leaf was found to have less pubescence compared to the leaves that developed later in the climate chamber, which further facilitated the syringe infiltration. The variation in pubescence might be the result of the differences in water availability between the growth chamber and the climate chamber, with leaf LPI 4 and the younger leaves undergoing organogenesis in the former and latter, respectively. The effect of water availability on pubescence development was also reported for the desert shrub Encelia farinose [ 62 ]. In this study, we developed an alternative procedure for effective Populus genetic transformation using agroinfiltrated leaves as explants. The integration of the transferred genes by agroinfiltration was also reported in tobacco [ 31 ] and grapevine [ 35 ]. This procedure increased the transformation frequency of aspen hybrid P. davidiana × bolleana by up to 41–67% (Table 1 ), which is much higher than that obtained from the routine genetic transformation procedure where leaf disk explants were co-cultivated with Agrobacterium liquid culture and shoots were regenerated via direct organogenesis (16.4% transformation frequency) [ 63 ]. The higher transformation frequency in this study was attributed mainly to the effectiveness of early selection for transformants during the callus-induction stage, which has been reported to be beneficial for successful transformation [ 36 , 64 ]. During the indirect organogenesis process, calli formed on the explant’s cut surface, grew slowly along the medium surface, and made close contact with the selective medium during the callus-induction stage, which allowed the transformed cells to multiply under the selective pressure and then minimized the number of non-transgenic escapes (Fig. S 5 ). On the contrary, in the direct organogenesis process, callus-like tissues were normally initiated on the cut of the midrib and the secondary veins in a leaf, probably from cambium cells inside these major veins, which the agrobacterial cells were not able to reach through syringe agroinfiltration due to the tightly aligned vascular bundle sheath cells around the veins. These callus-like tissues were visible on the upper side of the explant’s cut surface after 10 days of culture and grew rapidly upward from the surface of the leaf lamina, which prevented them from directly contacting the selective medium (Fig. S 5 ). As a result of the ineffective selection, all the regenerated plants via direct organogenesis were confirmed to be non-transgenic in this study. The importance of effective selection for transformants in the early stage of the culture process may be further illustrated by the high frequency of positive calli producing at least one transgenic plant (67–75%) (Table 1 ). Additionally, the generation of stably transformed plants further verified that Agrobacterium syringe infiltration was able to target the heterologous genes in mesophyll cells beyond the epidermal cell, as shown in Fig. 3 b, since the epidermal cells were resistant to dedifferentiation and had no potential to form callus and further develop into plants [ 35 ]. Although the generation of stably transformed lines normally takes 2–4 months, much longer than that for transient expression, the method set out for generating stably transformed poplar lines in this protocol provides a convenient approach to study genes in cell types other than the leaf epidermis. In this case, after the transient expression analysis of the fluorescent fusion protein was performed in the leaf epidermis, the sterilized infiltrated leaves could be directly used for callus induction and then shoot regeneration, circumventing the routine steps of explant inoculation with Agrobacterium and co-cultivation of Agrobacterium -mediated poplar transformation. Furthermore, we showed that the agrobacterial syringe infiltration method could be used for in vivo activation of the specialized processes of SCW biosynthesis in the epidermal cells of poplar leaves by overexpressing master activators of secondary wall formation, PdbVNS07/WND6A, PdbVNS09/WND2A, and PdbMYB020 (Fig. 6 ), in which the activation activity of these key regulators was enhanced through fusion with the activation domain of the herpes virus VP16 protein, as reported in Arabidopsis [ 65 ]. The overexpression of PdbNVS07/WND6A induced transdifferentiation of the epidermal cells into protoxylem-like vessel elements (Figs. 6 a–b), and PdbVNS09/WND2A and PdbMYB020 resulted in ectopic secondary wall deposition in the epidermal cells (Figs. 6 c–f). Since vascular tissue is deeply embedded in the plant, it is difficult to analyze the process of vessel element development in detail. For this reason, the in vitro induction system of xylem vessel elements from Zinnia suspension cells [ 66 ], Arabidopsis suspension cells [ 55 ], and Arabidopsis excised cotyledons [ 67 ] was established with effort and has provided fundamental information on xylem vessel element development. Thus, the success in inducing secondary wall formation in poplar leaves provides a powerful tool for dissecting the molecular mechanisms regulating vascular development in poplar. For example, the comprehensive gene expression profile analysis in those SCW-producing epidermal cells will contribute to elucidating the specialized regulatory mechanism of SCW formation and vessel element differentiation of woody plants in a high-throughput manner in the near future."
} | 3,952 |
35694518 | PMC9178751 | pmc | 9,173 | {
"abstract": "Lignin is a natural\naromatic compound in plants. Several lignin\nstructural models have been proposed in the past years, but all the\nmodels cannot be converted to benzene carboxylic acids (BCAs) for\nall aromatic rings connected to oxygen. This inspired us to explore\nthe structures of lignin. Based on the yields of BCAs, the results\nof 13 C NMR and ethanolysis residues, and gas chromatography–mass\nspectrometry and electrospray ionization mass spectrometry of ethanolysis\nof lignin, we have constructed a structural model of lignin with a\nformula C 6407 H 6736 O 2590 N 147 S 3 . The model not only satisfies the results of analyses,\nbut also explains the generation of BCAs from lignin oxidation and\nthe ethanolysis products. Importantly, double-ring and triple-ring\naromatic clusters are found in lignin, and some of them are connected\nby alkyl bridges, which results in conventional low conversions of\nlignin. Our findings in the structures of lignin may significantly\ninfluence the structures and applications of lignin.",
"conclusion": "3 Conclusions In summary, we constructed a new model\nmolecular formula of enzymatic\nlignin with a molecular formula of C 6407 H 6736 O 2590 N 147 S 3 and a model molecular\nweight of 127,214 Da. The structural model of lignin includes single-ring,\ndouble-ring, and triple-ring clusters. In addition, only a small part\nof aromatic clusters where at least one benzene ring is not directly\nconnected with oxygen can be oxidized to BCAs. Most single-ring clusters\nare connected with weak C–O bonds, while most multiring aromatic\nclusters are connected with C–C bonds. The model could reflect\nnot only the yield of BCAs, but also the products of ethanolysis and\nthe 13 C NMR result of enzymatic lignin.",
"introduction": "1 Introduction Biomass\nis highly attractive for a sustainable source of chemicals,\nmaterials, and fuels. 1 As the most abundant\nform of biomass, lignocellulose has a production of around 170 billion\nmetric tons per year. 2 Lignocellulose is\ninedible for humans, and it is mainly found in agricultural and forestry\nwaste. The efficient use of this cheap and abundant carbon-neutral\nresource can greatly alleviate the energy crisis and environmental\nproblems. Therefore, lignocellulose is a highly promising alternative\nto fossil energy sources. As we know, cellulose, hemicellulose,\nand lignin are the three\nmain components in lignocellulose. Among them, hemicellulose and cellulose\nare composed of five-carbon and six-carbon sugars, 3 but the structure of lignin is very complicated and there\nis no exact structural model. 4 Lignin acts\nas a binder in lignocellulose and holds cellulose and hemicellulose\ntogether. 5 More specifically, lignin is\na three-dimensional reticulated macromolecular structure made up of\nrandomly crosslinked oxygenated aromatic units. By crosslinking with\ncellulose and hemicellulose, lignin provides strength, rigidity, and\nflexibility with lignocellulose as well as aiding in water transport\nand protecting against attack by marauding insects and microorganisms.\nNowadays, in human’s daily life, the majority of lignin is\nused for direct combustion for heat and power because of its complicated\nstructure. Only by clarifying the structure of lignin can we put it\nto better use. Traditionally, as shown in Scheme 1 , three phenylpropane structures are considered\nas basic structural units in lignin. 4 , 6 − 8 The most prominent feature is that the structural units of lignin\nare all single-ring aromatic clusters and directly connected with\noxygen functional groups, such as −OH and −OCH 3 . Therefore, lignin is attracting much attention because of its potential\nas a renewable aromatic, 9 , 10 especially monophenols. 11 However, we produced almost all types of benzene\ncarboxylic acids (BCAs) from enzymatic lignin by alkali-oxygen oxidation\nin this work. The structures of 12 BCAs are shown in Scheme S1 . To the best of our knowledge, the production of\nBCAs from lignin has not been reported. On the other hand, BCAs cannot\nbe produced from the existing lignin structures (as shown in Scheme 1 ) for every aromatic\nring directly connected with oxygen functional groups. What are the\nreal structural units in lignin leading to BCAs’ formation? Scheme 1 Structures of the Primary Building Blocks of Lignin The most accredited theories about the lignin structure\ngo back\nto the early 1960s. Freudenberg performed the process of lignin biosynthesis\nin plants by coniferyl alcohol 12 and then\nplotted the first structural model of lignin through detecting the\nintermediates during lignin synthesis. 13 Then Nimz proposed another structural model of lignin by detecting\nthe products of thioacetic acid depolymerization of lignin. 14 The above two most famous structural models\nput a foundation for later research studies about lignin structures\nand many transformation routes of lignin. After that, several structural\nmodels of lignin were further proposed. 15 − 17 Recently, several review\narticles 18 , 19 have discussed the structures of lignin,\nbut the features of lignin structures are similar to those previously\nproposed. In summary, these structural models in the literature are\nvery similar in terms of structure units and chemical bonds, mainly\nincluding β-O-4 ether bonds and β-5, β-1, β-β,\n5-5 strong carbon bonds; and the structural units are all single-ring\naromatic clusters directly connected with −OH or −OCH 3 . There are two main methods to analyze the lignin structure:\ndirect\nspectrum characterization and inversion of depolymerization products.\nAs for the spectrum method, solid 13 C NMR is an effective\ntool for characterizing the structures of organic macromolecules.\nThe signal areas are proportional to the amount of carbon contained\nin each functional group. 20 Peak fitting\nof solid 13 C NMR spectra is then used to study the structures\nin detail. Nowadays, solid 13 C NMR plays an important role\nin the structure characterization of oil shale, 21 coal, 22 and lignin. 23 − 29 Hence, the method was also applied in this work to study the structure\nof lignin. As for the depolymerization method, to better retrieve\nlignin’s\nstructure, it is expected to find a suitable way to produce the real\nmolecular structure originating from lignin, that is to say, neither\nover depolymerization nor repolymerization happened. At present, many\nchemicals are produced from lignin through hydrolysis, 30 hydrogenolysis, 31 and pyrolysis. 32 Here, we obtained BCAs\nthrough the oxidation of lignin. However, we only make sure that in\nlignin there exist aromatic rings that are not connected directly\nwith oxygen functional groups. The oxidation method is too harsh to\ndepolymerize lignin to relatively small acids, so we cannot obtain\nthe macromolecular structure in lignin. Supercritical ethanolysis\nis an effective way to depolymerize organic\nsubstances by breaking weak bonds (such as C–O bonds) without\nfurther reactions or called the second reactions. It has been widely\nused in the depolymerization of lignin. 33 , 34 As we know,\nC–O bonds are the most common linkages in lignin. Hence, we\ncan obtain the macromolecules from lignin through supercritical ethanolysis.\nAfter the depolymerization of lignin, the detection of products is\nalso a pivotal factor in deducing the structure of lignin. Gas chromatography–mass\nspectrometry (GC–MS) is usually used in most cases; however,\nsome strongly polar and/or less volatile products cannot be detected\nwith GC–MS. 35 Thus, we applied electrospray\nionization mass spectrometry (ESI-MS) to detect ethanolysis products\nof lignin, which is adapted for the investigation of both the primary\nand supramolecular structures of biopolymers, 36 and it is also suitable to detect solutions containing nonvolatile\nand thermally labile compounds with high molecular weights. 37 As mentioned above, all the proposed structural\nmodels and structural\nunits of lignin before cannot be oxidized to BCAs. In this work, we\nproposed a new lignin structure model based on the distribution of\nBCAs combined with the results of ultimate analysis, Fourier-transform\ninfrared spectroscopy, 13 C NMR, and ethanolysis of lignin.\nThe result showed that there were not only single-ring aromatic clusters\ndirectly connected to oxygen atoms, but also multiring aromatic clusters,\nincluding double-ring aromatic clusters and triple-ring aromatic clusters\nin lignin.",
"discussion": "2 Results and Discussion The alkali-oxygen\noxidation of the lignin was performed, and the\neffects of temperature, initial oxygen pressure, reaction time, and\nthe alkali/enzymatic lignin on reaction were investigated. The results\nare shown in Figures S1–S4 in the\nSupporting Information (SI). It can be found that 11 kinds of BCAs\nhave been obtained. The distribution of BCAs at the maximum yield\nis shown in Figure S5 and Table 1 . The total mass yield of BCAs\nwas 8.14% obtained under the optimum conditions. It can be seen that\nBCAs containing three or more carboxyl groups are main products, especially\nbenzene pentacarboxylic acid. This result is undoubtedly determined\nby the structure of the lignin. Table 1 Yields of BCAs from\nEnzymatic Lignin\nvia Oxidation and Their Distribution in the Structural Model products mass yield/wt % molar yield/mmol·g –1 distribution\nin products distribution in the model benzoic acid 0.182 1.49 × 10 –2 1.99 2 phthalic\nacid 0.686 4.13 × 10 –2 4.49 5 isophthalic\nacid 0.148 0.89 × 10 –2 1.19 1 trimellitic\nacid 1.14 5.43 × 10 –2 7.24 7 hemimellitic\nacid 0.488 2.32 × 10 –2 3.09 3 trimesic\nacid 0.390 1.86 × 10 –2 2.48 2 prehnitic\nacid 0.191 0.75 × 10 –2 1 1 pyromellitic\nacid 0.963 3.79 × 10 –2 5.05 5 mellophanic\nacid 0.967 3.81 × 10 –2 5.08 5 benzene\npentacarboxylic acid 2.05 6.88 ×\n10 –2 9.17 9 mellitic acid 0.937 2.74 ×\n10 –2 3.65 4 We constructed the new structural\nunits of the lignin based on\nBCA distribution, and the data processing of BCAs is shown in Section 1.2 of the SI. The amounts of BCA distribution\nin the model are also shown in Table 1 . From Table 1 , we\ncan obtain the total molecular weight (10,604) of BCAs by adding the\nnumber in column 5 of Table 1 multiplied by BCA’s molecular weight. Because the\nmaximum yield of BCAs is 8.14%, so the molecular weight of the lignin\nstructure model can be roughly calculated as 10,604/0.0814 = 130,270\nDa (a hypothetical value proposed to construct the lignin model to\nensure that the lowest-BCA, isophthalic acid, has one aromatic structural\nunit). Then, based on the ultimate analysis of the lignin (shown in Table 2 ), the atom numbers\nof C, H, O, N, and S were calculated as 6404, 7035, 2763, 147, and\n3, respectively, in the structural model. Table 2 Proximate\nand Ultimate Analyses of\nEnzymatic Lignin a proximate\nanalysis (wt %) ultimate\nanalysis (in daf. Basis, wt %) M ad A ad V daf C H O b N S 8.31 1.84 61.29 58.99 5.41 33.94 1.58 0.08 a ad: air-dry basis; d: dry basis;\ndaf: dry-and-ash-free basis. M : moisture; A : ash; V : volatile matter content. b By difference. CP/MAS 13 C NMR was utilized\nto analyze the lignin to\nobtain structural parameters of various carbons. The 13 C NMR spectrum and peak fitting spectra of enzymatic lignin are shown\nin Figure S6 , and every position of carbon\nis shown in Table S1 . Based on the\npeak fitting spectra of 13 C NMR, carbon\ntypes and molar content have been determined and are shown in Table 3 . The specified carbon\nnumbers in the structural model are shown in column 6 of Table 3 . Table 3 Different Types of Carbon between\nOrigin Lignin and Ethanolysis Residues Based on the data in Table 3 , several structure parameters of the lignin\nwere calculated\nas follows. (1) Ratio of aromatic carbon ( f a ): f a = f ar H + f ar B + f ar C + f ar O1 + f ar O2 = 55.3% (2) Molar fraction\nof aromatic bridgehead\ncarbon ( X b ): X b = f ar B / f a = 0.160 (3) Ratio of aliphatic carbon ( f al ): f al = f al 1 + f al 2 + f al 3 + f al 4 + f al O1 + f al O2 + f al O3 = 35.3% (4) Substituted degree of\naromatic ring\n(δ): δ = ( f ar C + f ar O1 + f ar O2 )/ f a = 0.56 As a natural aromatic\nsubstance, the ratio of aromatic carbon is\nas high as 55.3%. The substituted degree of aromatic ring (δ)\nis 0.56, which indicates that more than half aromatic carbons are\nsubstituted. The molar fraction of aromatic bridgehead carbon ( X b ) is 0.160, which means that there are double-\nor more ring aromatic clusters in the lignin. The molar fraction of\naromatic bridgehead carbon X b is always\nan index for multiring aromatic clusters, and the X b of single-ring, double-ring, and triple-ring aromatic\nclusters is 0, 0.2, and 0.286, respectively. To identify a more\nexact number of aromatic rings, we applied supercritical\nethanolysis to break weak bonds (β-O-4 and α-O-4) and\nobtained liquid products in ethanol. The effects of temperature and\nreaction time on the yield of ethanolysis products were investigated.\nThe results are shown in Figures S7 and S8 . The results of the two figures indicate that the carbon yield of\nethanolysis products reaches 54.65% under relatively mild conditions\n(290 °C, 120 min). The ethanolysis liquid and residue were characterized\nseparately, and it is found that the aromatic bridge carbon of residues\nover 300 °C is more than that of raw lignin, which may indicate\nrepolymerization. Thus, the ethanolysis liquid and residue obtained\nat 290 °C were investigated. The total ion chromatogram spectrum\nobtained using GC–MS is shown in Figure S9 , and the compounds identified by the NIST11 database are\nshown in Table S2 . It should be noted that\nthe substances with relative contents higher than 0.5% are considered. However, only single-ring aromatic clusters were detected by GC–MS;\nall these structural units conform neither to X b from 13 C NMR nor to produce 11 BCAs via oxidation.\nAs mentioned above, because of the limitation of GC–MS, some\nlarge-molecule compounds may not be detected, which may result from\nthe residue in the column of GC. Hence, we applied ESI-MS to detect\nif there were multiring aromatic clusters existing in lignin ethanolysis\nproducts. The ESI-MS spectrum is shown in Figure 1 . Figure 1 Negative-mode ESI-MS spectrum of ethanolysis\nproducts. As shown in Figure 1 , the m / z peaks of ethanolysis\nproducts focus on 100 to 400, which indicates that the relative molecular\nmass focuses on 100 to 400. The relative molecular mass is larger\nthan that detected by GC–MS. We could find that single-ring\naromatic compounds detected by GC–MS were also detected by\nESI-MS as shown in Table S2 . In addition,\nseveral bigger compound formulas were detected based on the error\nand isotope analysis, and we could speculate the structures of different\naromatic ring clusters based on these formulas. The molecular formulas\nand intensity are shown in Table S3 , which\nincludes double-ring and triple-ring aromatics. Then we depicted the\nstructures of these multiring aromatic clusters as shown in Scheme S2 . There may be other multiring aromatic\ncluster isomers based on the formulas, and we just provided ones of\nthe possible structures, which are double- and triple-ring aromatic\nclusters. The results demonstrate that there exist double-ring and\ntriple-ring aromatic clusters in the lignin. The ethanolysis\nresidue was also characterized through 13 C NMR ( Figure S10 ) and Fourier-transform\ninfrared (FTIR) spectroscopy ( Figure 2 ). The 13 C NMR data were processed using\nthe same way as that of the enzymatic lignin. The total carbon atoms\nof the structural model are 6404, so the amount of carbon atoms in\nthe ethanolysis residue in the structural model is 2904 for the 54.65%\ncarbon yield of liquid products. The peak fitting of spectrum and\nthe amount of different carbon are shown in Figure S10 and Table 3 . We also obtained the structure parameters of the residue as follows. (5) Ratio of aromatic\ncarbon ( f a ): f a = f ar H + f ar B + f ar C + f ar O1 + f ar O2 = 66.9% (6) Molar fraction of aromatic\nbridgehead\ncarbon ( X b ): X b = f ar B / f a = 0.252 (7) Ratio of aliphatic carbon ( f al ): f al = f al 1 + f al 2 + f al 3 + f al 4 + f al O1 + f al O2 + f al O3 = 30.46% (8) Substituted degree of\naromatic ring\n(δ): δ = ( f ar C + f ar O1 + f ar O2 )/ f a = 0.55 Figure 2 FTIR spectra of enzymatic\nlignin and ethanolysis residue. The substituted degree of aromatic ring (δ) is almost the\nsame as that of the lignin, which means more substitution groups in\nthe ethanolysis residue, considering that multiring aromatic clusters\nhave fewer aromatic carbons that can be substituted. X b is 0.252, which indicates that double-ring aromatic\nclusters and triple-ring aromatic clusters exist in the ethanolysis\nresidue at least. During ethanolysis depolymerization, owing to the\nweak bonds (C–O) being broken, some single- and multi-ring\naromatic clusters were extracted into the liquid solution (as shown\nin Table S2 and Scheme S2 ), which indicates\nthat some multiring aromatic clusters are also connected with C–O\nbonds. From the NMR result of ethanolysis residue and liquid products\nof lignin, we concluded that most multiring aromatic clusters were\nconnected with strong C–C bonds and could not be broken during\nthe ethanolysis process. The numbers of carbon atoms of origin lignin\nand the ethanolysis residue are compared in Table 3 . From Table 3 , the\nnumber of oxygen-attached aliphatic carbon decreases notably, which\nindicates that the most connection manners (β-O-4) in lignin\nare broken to form phenols and then dissolve in the ethanol solution.\nIn addition, most of the aromatic bridgehead carbon is left in the\nresidue, and a small portion is extracted to liquid products. The\nmechanism of lignin ethanolysis is shown in Scheme 2 based on lignin alcoholysis results reported\nin the literature. 38 , 39 Scheme 2 Possible Mechanisms\nduring Ethanolysis The FTIR spectra of\nthe lignin and the ethanolysis residue are\nshown in Figure 2 .\nIn the comparison of these two FTIR spectra, the intensity of the\nassociative hydroxyl peak becomes narrow at 3500–3300 cm –1 , which can be attributed to some aromatics containing\nphenolic groups extracted to supercritical ethanol; the intensity\nof the methyl group peak increases slightly at 2937 cm –1 for esterification; the carbonyl groups at 1701 cm –1 also decrease after ethanolysis; the C–O–C asymmetric\nstretching vibration of the ether group decreased at 1127 cm –1 after ethanolysis, which indicates the break of ether bonds (β-O-4)\nduring the ethanolysis process. The amounts of total carbon,\naromatic carbon, and bridgehead carbon\nare shown in Table 4 . Based on the result of ESI-MS, we detected large molecules and\nclassified them into single-, double-, and triple-aromatic clusters.\nBased on elemental analysis and 13 C NMR spectra, there\nare 565, 490, and 75 bridgehead carbon atoms of the structural model\nof the lignin, ethanolysis residue, and ethanolysis liquid product,\nrespectively. The X b (0.252) of the ethanolysis\nresidue indicates that the average number of aromatic rings is between\ntwo and three. This suggested that almost single rings were extracted\nand the ethanolysis residue contained only double and triple-aromatic\nclusters. Thus, it can be concluded that the ratio of double-ring\naromatic clusters: triple-ring aromatic clusters in the ethanolysis\nresidue of lignin is 1:1.09 based on the X b of the ethanolysis residue. Hence, it can be calculated that the\nnumber of the double-ring aromatic clusters is 77, and that of the\ntriple-ring aromatic clusters is 84. According to 75 atoms of bridgehead\ncarbon in the liquid product, the numbers of double- and triple-ring\naromatic clusters could be identified as 12 and 13 based on the 1:1.09\nratio of double-ring aromatic clusters to triple-ring aromatic clusters.\nMeanwhile, we could calculate the number of single-ring aromatic clusters\nin the liquid product as 216 ((1600 – 12 × 10 –\n13 × 14)/6 = 216). Table 4 Distribution of Carbon\nin the Raw\nMaterial, Liquid Products, and Residue carbon type enzymatic lignin ethanolysis liquid product ethanolysis residue total carbon number 6404 3500 2904 aromatic carbon number 3543 1600 1943 bridgehead carbon number 565 75 490 Based\non the calculation of liquid products and ethanolysis residues,\nthere are 216 single-ring aromatic clusters, 89 double-ring aromatic\nclusters, and 97 triple-ring aromatic clusters in the structural model\nof the lignin. Thus, the number of all aromatic carbons is adjusted\nas 3544 (216 × 6 + 89 × 10 + 97 × 14 = 3544). Based on the yield distribution of BCAs, 44 units in enzymatic\nlignin can be converted into BCAs during the oxidation process, which\nindicates that some multiring aromatic clusters cannot be converted\ninto BCAs for their every aromatic ring connected with oxygen. Based\non the analysis of the ethanolysis liquid and residue, the multiring\naromatic clusters are connected by strong C–C mostly, which\ncauses most of them to be left in the ethanolysis residue, while C–O\naccounts for a large percentage in the connections of single-ring\naromatic clusters. In addition to C, H, and O, there are N (1.58%)\nand S (0.08%) elements\nthat exist in the lignin sample based on ultimate analysis ( Table 2 ). X-ray photoelectron\nspectroscopy (XPS) has been applied to characterize the surface composition\nof coals 22 , 40 and biomass. 41 , 42 Here, the\nS content is too little to be detected by XPS, and the lignin model\nonly contains three S atoms. We applied XPS to characterize the valence\nand form of N in the lignin sample. As displayed in Figure 3 and Table 5 , the XPS N 1s spectrum of the lignin sample\nis fitted with two peaks at 399.50 and 402.04 eV, corresponding to\namino and chemisorbed nitrogen oxides, respectively. Based on the\narea of two peaks and the result of ultimate analysis, the numbers\nof amino and chemisorbed nitrogen oxides are found to be 97 and 50.\nThe amino form of nitrogen may result from amino acids and proteins. 42 In addition, as the source of lignin we used,\ncorn cob contains 17 amino acids besides cellulose, hemicellulose,\nand lignin. 43 The trace S may also result\nfrom the methionine that exist in the corn cob. Figure 3 XPS spectrum (N 1s) of\nenzymatic lignin and their fitting curves. Table 5 Distribution of Carbon in the Raw\nMaterial, Liquid Product, and Residue elemental\npeak functionality binding energy/eV mole\ncontent/% N 1s amino 399.50 66.0 chemisorbed nitrogen oxides 402.04 34.0 Based\non the above information, we constructed a structural model\nof the enzymatic lignin (as shown in Figure 4 ) with a formula C 6407 H 6736 O 2590 N 147 S 3 . In the structural model,\nthere are 44 aromatic units in red, which can be converted to BCAs\nvia alkali-oxygen oxidation. In addition, the detected aromatic units\nby GC/MS and ESI/MS were depicted, while they are connected with C–O\nbonds. Many multiring aromatic clusters are connected with C–C\nbonds. In Table S4 , the results of the\nstructural model are compared with those from experiments of lignin,\nand it can be found that the average absolute relative deviation is\n1.4%. The formation of BCAs and the ethanolysis result can be reflected\nin the constructed model of lignin. Figure 4 Proposed structural model of organic matter\nof enzymatic lignin.\nAromatics clusters in red are convertible to BCAs via oxidation. The structural model presented in this work has\nthe following uses.\nFirst, multiring aromatic clusters and their connections with C–C\nbonds can explain the low conversions of lignin to aromatic compounds.\nSecond, the model provides information that more valuable aromatic\nchemicals like naphthene- and anthanthrene-based compounds can be\nobtained potentially from lignin. In the future experiments, it is\nnecessary to consider to cleavage C–C bonds between aromatic\nclusters during the depolymerization of lignin and then yield aromatic\nchemicals. Third, it is necessary to consider the formation mechanism\nand functions of lignin in plants according to the new model because\nthere are multiring aromatic clusters in lignin. Fourth, the findings\nof multiring aromatic clusters could be helpful for computational\nanalyses, such as new model construction and parameter optimization."
} | 6,063 |
36751405 | null | s2 | 9,174 | {
"abstract": "A large number of neural network models of associative memory have been proposed in the literature. These include the classical Hopfield networks (HNs), sparse distributed memories (SDMs), and more recently the modern continuous Hopfield networks (MCHNs), which possess close links with self-attention in machine learning. In this paper, we propose a general framework for understanding the operation of such memory networks as a sequence of three operations: "
} | 115 |
21750494 | PMC3196164 | pmc | 9,178 | {
"abstract": "Polysaccharides that make up plant lignocellulosic biomass can be broken down to produce a range of sugars that subsequently can be used in establishing a biorefinery. These raw materials would constitute a new industrial platform, which is both sustainable and carbon neutral, to replace the current dependency on fossil fuel. The recalcitrance to deconstruction observed in lignocellulosic materials is produced by several intrinsic properties of plant cell walls. Crystalline cellulose is embedded in matrix polysaccharides such as xylans and arabinoxylans, and the whole structure is encased by the phenolic polymer lignin, that is also difficult to digest 1 . In order to improve the digestibility of plant materials we need to discover the main bottlenecks for the saccharification of cell walls and also screen mutant and breeding populations to evaluate the variability in saccharification 2 . These tasks require a high throughput approach and here we present an analytical platform that can perform saccharification analysis in a 96-well plate format. This platform has been developed to allow the screening of lignocellulose digestibility of large populations from varied plant species. We have scaled down the reaction volumes for gentle pretreatment, partial enzymatic hydrolysis and sugar determination, to allow large numbers to be assessed rapidly in an automated system. This automated platform works with milligram amounts of biomass, performing ball milling under controlled conditions to reduce the plant materials to a standardised particle size in a reproducible manner. Once the samples are ground, the automated formatting robot dispenses specified and recorded amounts of material into the corresponding wells of 96 deep well plate (Figure 1). Normally, we dispense the same material into 4 wells to have 4 replicates for analysis. Once the plates are filled with the plant material in the desired layout, they are manually moved to a liquid handling station (Figure 2). In this station the samples are subjected to a mild pretreatment with either dilute acid or alkaline and incubated at temperatures of up to 90°C. The pretreatment solution is subsequently removed and the samples are rinsed with buffer to return them to a suitable pH for hydrolysis. The samples are then incubated with an enzyme mixture for a variable length of time at 50°C. An aliquot is taken from the hydrolyzate and the reducing sugars are automatically determined by the MBTH colorimetric method.",
"discussion": "Discussion Variations of the standard saccharification protocol can be used in the same platform for determining the activity of cellulolytic enzymes (i.e. by variation of the enzyme concentrations used in a plate as well as using paper as substrate); comparison of the efficiency of several enzyme mixtures on a specific material; time course for saccharification; etc. A standard saccharification protocol to compare the saccharification potential in different plant materials involves an eight hour hydrolysis (Figures 3 and 4). Most of the plant materials analysed requires four replicates. Under these conditions, the platform can process 80 samples/day. This analysis is being used to screen large populations of barley, maize, and brachypodium in order to establish variability in saccharification potential and the genes involved in its determination 4 ."
} | 840 |
24682152 | PMC4007538 | pmc | 9,179 | {
"abstract": "Naegleria gruberi is a free-living heterotrophic aerobic amoeba well known for its ability to transform from an amoeba to a flagellate form. The genome of N. gruberi has been recently published, and in silico predictions demonstrated that Naegleria has the capacity for both aerobic respiration and anaerobic biochemistry to produce molecular hydrogen in its mitochondria. This finding was considered to have fundamental implications on the evolution of mitochondrial metabolism and of the last eukaryotic common ancestor. However, no actual experimental data have been shown to support this hypothesis. For this reason, we have decided to investigate the anaerobic metabolism of the mitochondrion of N. gruberi . Using in vivo biochemical assays, we have demonstrated that N. gruberi has indeed a functional [FeFe]-hydrogenase, an enzyme that is attributed to anaerobic organisms. Surprisingly, in contrast to the published predictions, we have demonstrated that hydrogenase is localized exclusively in the cytosol, while no hydrogenase activity was associated with mitochondria of the organism. In addition, cytosolic localization displayed for HydE, a marker component of hydrogenase maturases. Naegleria gruberi , an obligate aerobic organism and one of the earliest eukaryotes, is producing hydrogen, a function that raises questions on the purpose of this pathway for the lifestyle of the organism and potentially on the evolution of eukaryotes.",
"conclusion": "Concluding Remarks on the Anaerobic Metabolism of Naegleria The recent completion of the genome project of N. gruberi has indicated that this amoeboflagellate is an organism with unique biology of its genome, cell biology, and biochemistry, leading to the capacity to alternate between aerobic and anaerobic metabolism ( Fritz-Laylin et al. 2011 ). The genome encodes a [FeFe]-hydrogenase along with three enzymes required for its maturation; all four proteins were predicted to harbor N-terminal mitochondrial targeting signals suggestive of potential localization of this pathway in the organelles of N. gruberi ( Fritz-Laylin et al. 2010 ; Fritz-Laylin et al. 2011 ; Opperdoes et al. 2011 ). As a result, having the capacity of aerobic respiration along with the anaerobic metabolism, the mitochondria of N. gruberi were considered to resemble the evolutionary intermediate proposed to have arisen within the ancestor of all extant eukaryotes ( Mentel and Martin 2008 ; Fritz-Laylin et al. 2011 ). Even though this was a concrete hypothesis, there were still questions on the functionality of this pathway within mitochondria. For example, the pathway requires the presence of an enzyme producing reducing equivalents and a protein, which accepts these electrons and transfers them to the [FeFe]-hydrogenase that subsequently produces molecular hydrogen. In anaerobic organisms, the proteins responsible for such function are the pyruvate:ferredoxin oxidoreductase (PFO) and the [2Fe-2S] ferredoxin. Even though two genes encoding [2Fe-2S] ferredoxin proteins have been discovered in the N. gruberi genome, a PFO gene has not been identified ( Fritz-Laylin et al. 2010 ; Fritz-Laylin et al. 2011 ), raising questions on the functionality of the anaerobic metabolism within mitochondria. Moreover, the data presented on this study clearly demonstrate that functional [FeFe]-hydrogenase is not present in mitochondria; instead, the active enzyme is localized in the cytosol of the organism. Because a marker component of the maturases HydE is localized in the cytosol as well, we can confidently predict that maturation of hydrogenase takes place in the cytosol of N. gruberi . This is the first time that the [FeFe]-hydrogenase along with its maturases have been found in the cytosol of an organism, a result that raises more questions on the evolution and/or acquisition of this pathway within eukaryotes. Along with Naegleria, other organisms have a cytosolic version of [FeFe]-hydrogenase, including M . balamuthi ( Nyvltova et al. 2013 ), E . histolytica , and G . intestinalis. Even in T . vaginalis , some of the [FeFe]-hydrogenases might be present outside of hydrogenosomes ( Sutak et al. 2012 ); however, hydrogenase maturases were found exclusively within organelles, thus far. The recent phylogenetic analysis of [FeFe]-hydrogenases proteins suggested multiple origins of [FeFe]-hydrogenases in eukaryotic cell evolution with lateral gene transfer (LGT), placing a huge part in it ( Hug et al. 2010 ). If so, an interesting question is: What was the original cellular localization of hydrogenase upon its acquisition by LGT? The most plausible scenario seems to be that hydrogenase operated first in the cytosol and later, upon acquisition of mitochondrial targeting signals, it was imported together with hydrogenase maturases to the mitochondrion. If so, the situation in Naegleria may represent a stage that was conserved just after LGT of hydrogenase gene. Alternatively, acquisition of [FeFe]-hydrogenase could arise with the endosymbiotic ancestor of mitochondrion ( Martin and Muller 1998 ), later in evolution, the ancient gene was replaced by LGT, and in the case of N. gruberi and other extant protists with cytosolic hydrogenases, the organellar hydrogenases gradually lost functional MTSs and localized to the cytosol. Indeed, [FeFe]-hydrogenase and HydE of N. gruberi possess unusual N-terminal extensions that reminds MTSs; however, the presence of these extensions did not facilitate protein delivery into the tested mitochondria. The situation will become clearer as we assemble more genomic data from diverse eukaryotes (especially those living in anaerobic or microaerophilic niches) within a range of lineages to provide a lucid picture on the evolution of this anaerobic pathway within eukaryotes. In addition, further investigations on the localization of proteins involved in anaerobic pathways of N. gruberi should come about, because we have clearly demonstrated that the current predictions are not consistent with experimental data. Last but not least, we should continue expanding our knowledge on the cell biology of N. gruberi , because it could subsequently provide us with new insights on how to combat its close relative N. fowleri , a deadly human pathogen without treatment.",
"introduction": "Introduction Naegleria gruberi is a noteworthy microbial eukaryote for evolutionary, biochemical, and biomedical reasons. Naegleria gruberi is a nonpathogenic relative to Naegleria fowleri , the “brain-eating amoeba,” and the causative agent of the primary amoebic meningoencephalitis (PAM), a disease currently with no efficient treatment. Both organisms have the ability to transform from an amoeba to a biflagellate form or a cyst depending on its habitat, alternating its biochemical functions in each stage ( Cable and John 1986 ; Fritz-Laylin et al. 2011 ). Evolutionarily, N. gruberi is considered to be one of the earliest eukaryotes and consequently close to the last eukaryotic common ancestor ( Koonin 2010 ). Recent analysis of its genome has backed up this hypothesis with the discovery of a metabolically flexible mitochondrion that possesses both classical aerobic pathways including branched respiratory chain and oxidative phosphorylation, and enzymes that are known to mediate a substrate-level phosphorylation in the hydrogenosome, an anaerobic form of mitochondrion ( Embley et al. 2003 ; Embley 2006 ). Most importantly, in silico predictions strongly suggested that Naegleria ’s mitochondrion possesses an [FeFe]-hydrogenase, a marker of energy metabolism in anaerobic or microaerophilic organisms that produced molecular hydrogen ( Vignais and Billoud 2007 ; Fritz-Laylin et al. 2010 ). This discovery provided new support for the hydrogen hypothesis of the origin of mitochondria, in which it has been postulated that the endosymbiotic ancestor of mitochondrion possessed both anaerobic and aerobic pathways ( Martin and Muller 1998 ). During adaptation of eukaryotes to aerobic or anaerobic niches, these pathways were selectively lost, resulting in the formation of mitochondrion or hydrogenosome. However, a mitochondrion with both types of metabolisms operating in any contemporary eukaryote has not been found thus far. For this reason, we have decided to investigate the localization and functional characterization of the [FeFe]-hydrogenase of N. gruberi to provide experimental data in addition to previous in silico predictions. The [FeFe]-hydrogenase is an enzyme that acts as a sink to remove reducing equivalents from oxidative decarboxylation of pyruvate or malate. Electrons generated during these reactions are accepted by low-redox potential electron carriers (usually ferredoxins) and transferred to the hydrogenase that synthesizes molecular hydrogen. In eukaryotes, these enzymes are found in the hydrogenosomes of several anaerobic protists (for further reading, see Embley and Martin [2006] , Hug et al. [2010] , and Muller et al. [2012] ) including chytridiomycetes, anaerobic ciliates, trichomonads, Blastocystis , as well as in the cytosol of others such as Giardia and Entamoeba . In addition, an [FeFe]-hydrogenase-dependent hydrogen production has been found in the chloroplasts of green algae ( Kamp et al. 2008 ). The [FeFe]-hydrogenase is generally associated with three maturase proteins: the radical S -adenosylmethionine enzymes HydE and HydG, and the small GTPase HydF ( Putz et al. 2006 ). These maturases are required for the assembly of H cluster at the catalytic site of the enzyme that is essential for hydrogen synthesis. The H cluster includes two subclusters; a diiron subcluster with several nonprotein ligands and a [4Fe-4S] subcluster ( Peters and Broderick 2012 ). Interestingly, the maturases have been identified in protists with hydrogenosomes ( Trichomonas vaginalis , Mastigamoeba balamuthi ; Putz et al. 2006 ; Nyvltova et al. 2013 ), but they are absent in organisms with exclusively cytosolic [FeFe]-hydrogenase ( Giardia intestinalis and Entamoeba histolytica ; Lloyd et al. 2002 ; Nixon et al. 2003 ). Unsurprisingly, the activities of cytosolic hydrogenases in these organisms are very low ( Giardia ) or could be detected only upon overexpression of the hydrogenase gene in transformed cells ( Entamoeba ). Thus, hydrogenases in Giardia and Entamoeba might not be involved in the production of molecular hydrogen as have been proposed ( Meyer 2007 ; Nicolet and Fontecilla-Camps 2012 ). In the current article, we have combined immunolocalization techniques along with cell biology and biochemistry to clarify the cellular localization of [FeFe]-hydrogenase in the aerobic excavate N. gruberi . We demonstrated that N. gruberi is able to generate molecular hydrogen when grown under aerobic conditions. Unexpectedly, [FeFe]-hydrogenase as well as HydE were detected exclusively in the cytosol of the organism.",
"discussion": "Results and Discussion Characteristics of the N. gruberi [FeFe]-Hydrogenase Sequence In the recent genomic and transcriptomic study of N. gruberi ( Fritz-Laylin et al. 2010 ), the authors have identified the full sequence of [FeFe]-hydrogenase gene, along with three predicted genes associated with the maturases of this enzyme. Bioinformatic analyses of the hydrogenase sequence have demonstrated that the encoded protein could have an N-terminal mitochondrial-targeting sequence (MTS; Fritz-Laylin et al. 2010 ) ( supplementary figs. S1 and S2 , Supplementary Material online). Interestingly, the Naegleria hydrogenase has an N-terminal region that is considerably longer than typical eukaryotic [FeFe]-hydrogenases ( supplementary fig. S1 , Supplementary Material online). This suspiciously long N-terminal expansion raised the question whether the predicted length of the sequence was of the correct size. To tackle this question, we used the rapid amplification of cDNA ends on N. gruberi cDNA. Using internal reverse primers to the coding region of the gene, we amplified the complete 5′-end, and after sequencing multiple clones, we demonstrated a 100% identity to the predicted coding region of the gene ( supplementary fig. S2 , Supplementary Material online). Similar approach has been used for verification of N-terminal regions of the hydrogenase maturases, with the same results ( supplementary fig. S3 , Supplementary Material online). By examining the protein sequence, the N. gruberi [FeFe]-hydrogenase possesses conserved cysteine residues in the H-domain of the enzyme that are required for coordination of the H-cluster ( supplementary fig. S1 , Supplementary Material online). In addition, the N-terminal part of the protein possesses typical cysteine motives for coordination of two accessory [4Fe-4 S] clusters and an [2Fe2S] cluster that are involved in electron transfer from the electron donor to the active site ( Vignais and Billoud 2007 ; Tsaousis et al. 2012 ). All these predictions suggest that the N. gruberi hydrogenase is a functional enzyme and could be localized in the mitochondria of the cell as previously predicted ( Fritz-Laylin et al. 2010 ). Immunolocalization of N. gruberi [FeFe]-Hydrogenase Using an Ab raised against T. vaginalis [FeFe]-hydrogenase ( Bui and Johnson 1996 ), we demonstrated its high specificity for N. gruberi [FeFe]-hydrogenase that was recognized as a protein with an expected size of 83.8 kDa in Naegleria ’s cell lysate ( fig. 1 B ) as well as in a lysate of Escherichia coli cells heterologously expressing a fragment of N. gruberi ’s hydrogenase ( fig. 1 C ). Surprisingly, immunofluorescent microscopy of N. gruberi cells using α-hydrogenase Ab showed a cytosolic distribution of this enzyme, and the hydrogenase labeling do not colocalize with the MitoTracker ( fig. 1 A ). The same cytosolic localization we also observed for hydrogenase maturase HydE ( fig. 1 A and supplementary fig. S4 , Supplementary Material online). As controls, we used Abs against SdhB, the mitochondrial protein involved in TCA (tricarboxylic acid) cycle and against Nbp35, the cytosolic Fe-S cluster assembly factor ( Netz et al. 2012 ). Immunofluorescent visualization of SdhB revealed its colocalization with the signal for MitoTracker in organelles corresponding to mitochondria ( fig. 1 A ). Signal of the α-Nbp35 antibody had shown the same distribution as the α-hydrogenase and α-HydE Abs: it did not colocalize with MitoTracker and consequently showed a pattern corresponding to the cytoplasm of Naegleria ( fig. 1 A ).\n F ig . 1.— Cellular localization of hydrogenase and HydE in N. gruberi cells . ( A ) Immunofluorescent microscopy. [FeFe]-Hydrogenase, HydE, and Nbp35 (cytosolic marker) were detected in the cytosol of the cell. The signals for these proteins do not colocalize with the MitoTracker red. The Ab against SdhB (mitochondrial marker) labeled discrete structures corresponding to the N. gruberi mitochondria and colocalized with the MitoTracker red. 4′,6-Diamidino-2-phenylindole (DAPI) staining shows the presence of nucleus and mitochondrial DNA. DIC, differential interference contrast. Scale bar: 10 μm. ( B ) Localization of [FeFe]-hydrogenase, HydE, and SdhB in N. gruberi cellular fractions. Lys, whole cell lysates; Mito, mitochondrial fraction, Cyto, cytosolic fraction. SdhB, HydE and [FeFe]-hydrogenase were visualized using M. balamuthi α-SdhB, α-HydE and T.vaginalis α-hydrogenase, antibodies, respectively. ( C ) Specificity of α-[FeFe]-hydrogenase and α-HydE Abs were tested using partial recombinant N. gruberi [FeFe]-hydrogenase (55 kDa), and complete N. gruberi HydE (53 kDa) heterologously produced in E. coli and subsequently were probed using corresponding heterologous Ab. The localization of [FeFe]-hydrogenase and HydE observed by immunofluorecent microscopy was verified by cell fractionation experiments. Mitochondrial and cytosolic fractions were isolated from cell homogenate using differential centrifugation, followed by a immunoblot analysis. Using the α-hydrogenase Ab and α-HydE Ab, we have detected the signal for hydrogenase and HydE, respectively exclusively in the cytosolic fraction ( fig. 1 B ). In contrast, SdhB was detected in the mitochondrial fraction. Furthermore, we have decided to employ an alternative approach to evaluate the localization of the [FeFe]-hydrogenase and examine the function of predicted MTS. We expressed the 5′-end region of N. gruberi ’s [FeFe]-hydrogenase (1,251 bp) and HydE (690 bp) in S. cerevisiae and investigated whether these proteins have the ability to deliver into mitochondria. As a positive control, we expressed the 5′-end region of SdhB (690 bp). Immunofluorescent microscopy revealed that the GFP-tagged N. gruberi SdhB localizes in the mitochondria (colocalization with MitoTracker; fig. 2 A ), whereas the N. gruberi [FeFe]-hydrogenase-GFP and HydE-GFP were found to be present exclusively in the cytosol of the cell ( fig. 2 A ), which was consistent with the previous results. In addition, we examined the localization of hydrogenase-GFP, HydE-GFP, and NgSdhB-GFP in yeast subcellular fractions. The immunoblot analysis showed the signal for hydrogenase-GFP and HydE-GFP only in cytosolic fractions ( fig. 2 B ). As expected, the signal for SdhB-GFP was found in isolated mitochondria. The NgSdhB topology within mitochondria was further tested by protease protection assay. The NgSdhB-GFP signal was not affected when the mitochondria were treated with trypsin; however, the signal disappeared after trypsin treatment when the organellar membranes were disintegrated with a detergent, which is consistent with the presence of NgSdhB-GFP in the mitochondrial matrix ( fig. 2 B ). As a control for mitochondrial integrity, α-Tom20 was used as the outer membrane marker (digested with trypsin without the addition of detergent), and α-Tim17 was used as an inner membrane marker (not digested with trypsin without the addition of detergent; fig. 2 B ). These data demonstrated that even though N. gruberi’s [FeFe]-hydrogenase and HydE proteins possess N-terminal extensions, these putative signals are not recognized by yeast mitochondria, and both proteins remain in yeast cytosol, which is consistent with their cytosolic localization in N. gruberi . In contrast, NgSdhB is recognized by yeast mitochondrial translocases and is delivered into the mitochondrial matrix.\n F ig . 2.— N-Terminal presequence-dependent targeting of N. gruberi hydrogenase, HydE, and SdhB into S. cerevisiae mitochondria. ( A ) The N-terminal region of the N. gruberi proteins (SdhB, Hydrogenase and HydE) were expressed with GFP tags in S. cerevisiae (green). Mitochondria were labeled with Mitotracker (red). DIC, differential interference contrast. ( B ) Western blot analysis of cellular fractions. Cyt, cytosol; Mit, mitochondrial fraction. Protease protection assays was performed for mitochondrial fraction: Mit+tryp, mitochondria were treated with trypsin; Mit+tryp+TX-100; mitochondria were treated with trypsin together with Triton X-100. Hydrogenase, NgSdhB, and HydE were detected using anti GFP tag antibody, and specific antibody was used to detect Tom20 (outer-membrane marker), and Tim17 (inner-membrane marker). Enzymatic Activity of [FeFe]-Hydrogenase Both in vivo and in vitro experiments have revealed that the [FeFe]-hydrogenase localizes in the cytosol of N. gruberi . However, the possibility that there is a minor pool of the hydrogenase in the N. gruberi’s mitochondria cannot be excluded. Therefore, to obtain independent line of evidence, we next tested the enzymatic activity of [FeFe]-hydrogenase in N. gruberi cellular fractions. Consistently, activity of [FeFe]-hydrogenase was found exclusively in the cytosolic fraction, and no activity was detected in the mitochondrial fraction of N. gruberi ( table 1 ). For control reasons, the abundance and purity of mitochondrial and cytosolic fractions was tested using marker enzymes. GDH was used as a mitochondrial marker enzyme, which has been predicted to be present in the mitochondria of N. gruberi ( Fritz-Laylin et al. 2010 , 2011 ). Indeed, the GDH activity was exclusively associated with mitochondria ( table 1 ). As a cytosolic marker, we used a glycolytic enzyme hexokinase that displayed about 9.36 nmol/min/mg activity in the cytosol and only 0.46 nmol/min/mg of activity was associated with mitochondria. The minor mitochondrial hexokinase activity is most likely due to its association with the outer mitochondrial membrane ( table 1 ) ( Majewski et al. 2004 ). These biochemical data not only support the cytosolic localization of N. gruberi ’s [FeFe]-hydrogenase but also demonstrated that the protein is enzymatically active. Noteworthy, the activity of hydrogenase, which is an oxygen-sensitive enzyme, was measured in N. gruberi that was grown under standard aerobic conditions in axenic culture. Our attempts to maintain Naegleria under anaerobic atmosphere have been unsuccessful.\n Table 1 Biochemical Localization and Characterization of Naegleria gruberi [FeFe]-Hydrogenase Whole Cells (nmol/min/mg) Mitochondrion (nmol/min/mg) Cytosol (nmol/min/mg) Localization Hydrogenase 3.839 ± 0.233 0.0000 0.893 ± 0.241 Cytosol GDH 5.132 ± 0.149 5.469 ± 0.152 0.000 Mitochondrion Hexokinase 9.367 ± 1.714 0.465 ± 0.188 15.552 ± 3.24 Cytosol Hydrogen Production Even though we have demonstrated the functional activity of N. gruberi [FeFe]-hydrogenase, we have also questioned whether the cells can actually produce molecular hydrogen under standard cultivation conditions. For this reason, N. gruberi cells were incubated in the M7 growth medium under atmospheric oxygen, and after 4 h, the hydrogen concentrations were determined in the gas phase. As a result, we found that N. gruberi cells are able to produce ∼6 nmol of hydrogen/min/10 7 cells ( table 2 ). This level is about three times higher than hydrogen production observed in microaerophilic organisms G. intestinalis (2 nmol/min/10 7 cells; Lloyd et al. 2002 ) and about five times lower than in T. vaginalis ( Sutak et al. 2012 ).\n Table 2 Hydrogen Production in Microbial Eukaryotes Organism Hydrogen Production (nmoles/min per 10 7 cells) Naegleria gruberi 5.814 ± 0.57 This study Trichomonas vaginalis 29.098 ± 1.549 This study Ellis et al. (1992) Giardia intestinalis 2 Lloyd et al. (2002) Concluding Remarks on the Anaerobic Metabolism of Naegleria The recent completion of the genome project of N. gruberi has indicated that this amoeboflagellate is an organism with unique biology of its genome, cell biology, and biochemistry, leading to the capacity to alternate between aerobic and anaerobic metabolism ( Fritz-Laylin et al. 2011 ). The genome encodes a [FeFe]-hydrogenase along with three enzymes required for its maturation; all four proteins were predicted to harbor N-terminal mitochondrial targeting signals suggestive of potential localization of this pathway in the organelles of N. gruberi ( Fritz-Laylin et al. 2010 ; Fritz-Laylin et al. 2011 ; Opperdoes et al. 2011 ). As a result, having the capacity of aerobic respiration along with the anaerobic metabolism, the mitochondria of N. gruberi were considered to resemble the evolutionary intermediate proposed to have arisen within the ancestor of all extant eukaryotes ( Mentel and Martin 2008 ; Fritz-Laylin et al. 2011 ). Even though this was a concrete hypothesis, there were still questions on the functionality of this pathway within mitochondria. For example, the pathway requires the presence of an enzyme producing reducing equivalents and a protein, which accepts these electrons and transfers them to the [FeFe]-hydrogenase that subsequently produces molecular hydrogen. In anaerobic organisms, the proteins responsible for such function are the pyruvate:ferredoxin oxidoreductase (PFO) and the [2Fe-2S] ferredoxin. Even though two genes encoding [2Fe-2S] ferredoxin proteins have been discovered in the N. gruberi genome, a PFO gene has not been identified ( Fritz-Laylin et al. 2010 ; Fritz-Laylin et al. 2011 ), raising questions on the functionality of the anaerobic metabolism within mitochondria. Moreover, the data presented on this study clearly demonstrate that functional [FeFe]-hydrogenase is not present in mitochondria; instead, the active enzyme is localized in the cytosol of the organism. Because a marker component of the maturases HydE is localized in the cytosol as well, we can confidently predict that maturation of hydrogenase takes place in the cytosol of N. gruberi . This is the first time that the [FeFe]-hydrogenase along with its maturases have been found in the cytosol of an organism, a result that raises more questions on the evolution and/or acquisition of this pathway within eukaryotes. Along with Naegleria, other organisms have a cytosolic version of [FeFe]-hydrogenase, including M . balamuthi ( Nyvltova et al. 2013 ), E . histolytica , and G . intestinalis. Even in T . vaginalis , some of the [FeFe]-hydrogenases might be present outside of hydrogenosomes ( Sutak et al. 2012 ); however, hydrogenase maturases were found exclusively within organelles, thus far. The recent phylogenetic analysis of [FeFe]-hydrogenases proteins suggested multiple origins of [FeFe]-hydrogenases in eukaryotic cell evolution with lateral gene transfer (LGT), placing a huge part in it ( Hug et al. 2010 ). If so, an interesting question is: What was the original cellular localization of hydrogenase upon its acquisition by LGT? The most plausible scenario seems to be that hydrogenase operated first in the cytosol and later, upon acquisition of mitochondrial targeting signals, it was imported together with hydrogenase maturases to the mitochondrion. If so, the situation in Naegleria may represent a stage that was conserved just after LGT of hydrogenase gene. Alternatively, acquisition of [FeFe]-hydrogenase could arise with the endosymbiotic ancestor of mitochondrion ( Martin and Muller 1998 ), later in evolution, the ancient gene was replaced by LGT, and in the case of N. gruberi and other extant protists with cytosolic hydrogenases, the organellar hydrogenases gradually lost functional MTSs and localized to the cytosol. Indeed, [FeFe]-hydrogenase and HydE of N. gruberi possess unusual N-terminal extensions that reminds MTSs; however, the presence of these extensions did not facilitate protein delivery into the tested mitochondria. The situation will become clearer as we assemble more genomic data from diverse eukaryotes (especially those living in anaerobic or microaerophilic niches) within a range of lineages to provide a lucid picture on the evolution of this anaerobic pathway within eukaryotes. In addition, further investigations on the localization of proteins involved in anaerobic pathways of N. gruberi should come about, because we have clearly demonstrated that the current predictions are not consistent with experimental data. Last but not least, we should continue expanding our knowledge on the cell biology of N. gruberi , because it could subsequently provide us with new insights on how to combat its close relative N. fowleri , a deadly human pathogen without treatment."
} | 6,822 |
31099139 | PMC6851840 | pmc | 9,183 | {
"abstract": "Abstract Habitat spatial structure has a profound influence on bacterial life, yet there currently are no low‐cost equipment‐free laboratory techniques to reproduce the intricate structure of natural bacterial habitats. Here, we demonstrate the use of paper scaffolds to create landscapes spatially structured at the scales relevant to bacterial ecology. In paper scaffolds, planktonic bacteria migrate through liquid‐filled pores, while the paper’s cellulose fibres serve as anchor points for sessile colonies (biofilms). Using this novel approach, we explore bacterial colonisation dynamics in different landscape topographies and characterise the community composition of Escherichia coli strains undergoing centimetre‐scale range expansions in habitats structured at the micrometre scale. The bacteria‐in‐paper platform enables quantitative assessment of bacterial community dynamics in complex environments using everyday materials.",
"introduction": "Introduction The intricate spatial structure of habitats has a decisive influence on the populations they support. Community dynamics are shaped by myriad factors, from macroscopic (i.e. many body lengths) to mesoscopic (i.e. several body lengths) and microscopic ( c . 1 body length) scales, dispersal, resource abundance, physical structure and interactions between individuals sculpt the assembly and composition of communities (Levin 1992 ; Leibold et al . 2004 ). Furthermore, processes occurring at different scales are intricately linked as single genes in individual cells can shape patterns and processes at the ecosystem level and vice versa (Whitham et al . 2006 ; Wymore et al . 2011 ). For bacteria, this means that the smallest ecological scale at which they interact with their environment is on the order of micrometres, while environmental gradients extend over millimetres and beyond (Cordero & Datta 2016 ). Physical and chemical heterogeneities through space and time, and interactions between bacteria give rise to the diverse and architecturally complex bacterial communities we find in nature (Dethlefsen et al . 2017; Stocker 2012 ; Hol et al . 2013 ; Vos et al . 2013 ; Aleklett et al . 2017 ). While the composition and structure of bacterial communities inhabiting various habitats, including the human body (Costello et al . 2009 ) and soil (Raynaud & Nunan 2014 ), have been characterised in detail, we currently lack generally available tools to systematically study how the micro‐ and mesoscopic structure of habitats drives the assembly and dynamics of such communities (Widder et al . 2016 ). Many important bacterial habitats, including biological tissues and soil matrices, consist of a microscale network of connected pores and cavities through which individuals migrate, while their abundant surfaces facilitate the growth of biofilms. Traditional laboratory tools to culture bacteria, however, are not well suited to mimic such landscapes, and furthermore typically only support planktonic or surface‐associated growth (not both simultaneously), and thus suppress the coexistence of these distinct lifestyles. In the past decade, various microfabrication‐based approaches to culture and study bacteria have emerged, enabling the study of bacterial ecology at the micrometre to millimetre scale by engineering synthetic landscapes (Rusconi et al . 2014 ). While such approaches have resulted in exciting insights regarding, for example, spatial competition between bacteria, the evolution of antibiotic resistance, microbial community assembly and biofilm growth (Balagaddé et al . 2005 ; Keymer et al . 2008 , 2006 ; Connell et al . 2013 ; Drescher et al . 2013 ; Park et al . 2003 ; Wessel et al . 2013 ; Coyte et al . 2016 ; Hol et al . 2013 ; Hol & Dekker 2014 ; Kim et al . 2008 ; Nagy et al . 2018 ), microfabrication‐based approaches to study bacterial ecology have not been adopted widely. This is largely due to the fact that the laboratory infrastructure necessary to create microfabricated landscapes is expensive, specialised and not readily available in microbiology laboratories. To overcome this barrier, we here demonstrate the use of paper scaffolds as a versatile and easy‐to‐use platform for studying bacterial communities in environments that are spatially structured at the relevant microscopic scales. Paper is a widely available material consisting of cellulose fibres. Interestingly, the characteristic length scales of paper (Derda et al . 2011 ) and many bacterial habitats (e.g. the soil matrix (Carson et al . 2010 )) are very similar, having pores from a few to several tens of micrometres. Furthermore, paper can be easily cut, either by hand or using a laser cutter, into any two‐dimensional geometry at the milli‐ to centimetre scale, while layers of patterned paper can be stacked to make three‐dimensional geometries (Derda et al . 2009 ; Mosadegh et al . 2014 , 2015 ; Truong et al . 2015 ). Paper furthermore can be creased and folded to create yet other geometries. The spatial scales at which paper is structured (microns) and can be manipulated (milli‐ to centimetre) correspond very well to the range of scales that are intrinsic to bacterial ecology, suggesting that paper may provide an excellent substrate for mimicking the complex structure of natural bacterial habitats. We here demonstrate this novel use of paper through two case studies: a range expansion in a dendritic landscape, and the colonisation of an archipelago of uninhabited islands from an inhabited main land. Dendritic (or branching) networks, such as rivers or cave systems, comprise a ubiquitous class of landscapes and their connectivity has been suggested to influence community dynamics and biodiversity across many taxa (Campbell Grant et al . 2007 ). Theoretical studies indicate that a dendritic landscape connectivity strongly impacts community dynamics, generally leading to decreased local species richness and an increase in between‐community diversity (i.e. local communities being less similar) when compared to a two‐dimensional lattice with uniform connectivity (Carrara et al . 2012 ; Muneepeerakul et al . 2007 ). Dendritic ecosystems are also found at microscopic scales in, for example, lungs, capillary networks or soil, yet experimental studies investigating the ecological dynamics of bacteria in dendritic landscapes are scarce. Furthermore, as theoretical investigations of dendritic landscapes are often inspired by observations of macroscopic systems (e.g. river networks (Fernandes et al . 2004 ; Muneepeerakul et al . 2008 )), it is interesting to experimentally test if similar mechanisms operate at microscopic scales. Experiments of protist communities undergoing experimentally imposed dendritic versus two‐dimensional lattice dispersal regimes (Carrara et al . 2012 ), or colonising dendritic versus linear microcosms suggest that observations made in macroscopic systems (e.g. branching increases β diversity) translate to microbial systems, although the observed differences may be transient, eventually resulting in similar community compositions in dendritic and linear ecosystems (Seymour & Altermatt 2014 ; Seymour et al . 2015 ). In case study 1, we explore this question in the bacterial context by using the bacteria‐in‐paper approach to contrast a two‐species range expansion unfolding in a dendritic landscape to a non‐branching linear landscape. In case study 2, we revisit island biogeography from a bacterial perspective. Biogeography theory posits that the community composition on islands depends on factors including the distance to the mainland and the size of the island, and for many taxa, a positive power‐law scaling between habitat size and the number of species is clearly established. Observations from the field (Horner‐Devine et al . 2004 ; Bell et al . 2005 ; Barreto et al . 2014 ) suggest that these generalities extend to the bacterial world, yet this notion has been a topic of debate (Fenchel & Finlay 2005 ), and has rarely been tested in the laboratory context (Fenchel & Finlay 2005 ), The bacteria‐in‐paper approach allows the experimental construction of bacterial archipelagoes and we demonstrate through a proof‐of‐concept study that paper scaffolds can be used to investigate the dynamics of bacteria that colonise initially uninhabited islands from an inhabited mainland. As the approaches presented here do not require any specialised equipment nor specific training, we anticipate that it will enable bacterial ecologists to transition their model systems from unstructured test tubes and petri dishes, to the intricate microscopic world inside a sheet of paper.",
"discussion": "Results and Discussion To facilitate bacterial growth and motility in the paper matrix, we saturated paper with bacterial growth medium (LB broth). Confocal imaging of fluorescently labelled Escherichia coli demonstrated that bacteria can swim in the liquid medium that fills pores in the cellulose mesh, allowing bacterial growth, migration and colonisation through paper scaffolds several centimetres in length. Figure 1 shows sessile colonies (biofilms) formed by E. coli (Fig. 1 b and c) and Bacillus subtilis (Fig. 1 d) after a 15 h incubation period at 37 o C. Cellulose fibres act as anchor points for surface‐associated growth, giving rise to dense colonies that form in the pores (see Supplementary Movie 1 for a confocal Z‐stack showing bacterial aggregates that formed 0–33 μm into the paper). Before inoculation of bacteria at one central point of the paper scaffold, the entire scaffold was saturated with growth medium forming an initially homogeneous nutrient landscape. After 15 h of growth cellular aggregates had formed scattered throughout the paper scaffold – even at the extremities, centimetres away from the original inoculation point. Figure 1 Bacteria‐in‐paper. (a) A photograph showing a paper scaffold cut to a predesigned shape with a laser cutter. A pencil is shown for scale, the scale bar is 5 mm. (b) Confocal scan of bacteria‐in‐paper showing GFP ‐expressing E. coli (green), RFP ‐expressing E. coli (red), and paper (blue). (c) Zoom in of the area indicated with dashed lines in (c). (d) Confocal scan of GFP ‐expressing B. subtilis (green) and paper (blue). Scale bars in (b–d) are 20 μm. Confocal imaging penetrates up to c . 100 micrometres into the paper, enabling the high‐resolution visualisation of communities of fluorescently labelled bacteria inhabiting the paper. The paper in use here weighs 87 g/m 2 and is 180 micrometres thick, images taken at multiple focal distances (Z‐stacks) from both sides can thus be used to visualise the entire community. However, cellulose fibres may obscure a fraction of cells when imaging beyond several tens of micrometres into the paper. To enable quantitative assessment of the bacterial communities inhabiting the paper matrix independent of the penetration depth of imaging, we took advantage of the fact that bacterial DNA can easily be extracted from paper to assess the community composition by, for example, quantitative PCR (qPCR) or sequencing‐based methods (e.g. (Cira et al . 2018 )). As we demonstrate below, qPCR provides an economical and convenient means to spatially resolve community composition, albeit at a lower resolution compared to confocal microscopy. Case study 1: Dendritic network connectivity Having established that bacteria are motile and grow in paper containing growth medium, we used this approach to investigate the colonisation dynamics of E. coil in a range expansion in two distinct types of landscapes. To probe the effect of a dendritic landscape topology on bacterial range expansions, we cut paper scaffolds (26 × 15 mm) that consist of a central inoculation zone providing access to both a dendritic and a non‐dendritic landscape on opposite sides (top and bottom in Fig. 2 a respectively). As both landscapes are colonised from the same inoculation zone, and thus by the same initial community, the effect of branching on the range expansions can be assessed by comparing the community composition at the extremities of both landscapes. Paper scaffolds were saturated with rich growth medium (LB) and inoculated in the centre with a 1:1 mixture of neutrally labelled E. coli , isogenic except for a green fluorescent protein (GFP) versus red fluorescent protein (RFP) insertion in the Lac operon (Keymer et al . 2008 ; Hol et al . 2013 ; Van Vliet et al . 2014 ). A challenge to confining bacteria to liquid‐saturated paper is the liquid film that forms when wet paper comes in contact with a surface (e.g. a glass coverslip). In order to prevent bacteria from growing in or migrating through such a liquid film, we suspended the wet paper on thin wires ( c . 5 mm pitch) in a chamber with saturated humidity. This ensures that no liquid interfaces are formed and all bacterial migration happens through the paper matrix. Figure 2 Range expansions in branching and non‐branching landscapes. (a) Cartoon of a paper scaffold consisting of a branching landscape and a non‐branching landscape connected to the same inoculation zone (indicated by a dashed circle), arrows indicate the direction of migration and population expansion upon inoculation. (b) Confocal scans of GFP ‐ and RFP ‐labelled E. coli at the landscape’s extremities labelled L, R, 3 and 8 in panel (a). Scale bars are 20 μm. (c) Fraction of GFP ‐labelled E. coli relative to the GFP fraction at the inoculation zone measured at the branch extremities by qPCR on gDNA extracted from the most distal 2 mm of each branch (i.e. all eight branches for the branching landscape, and the left‐ and rightmost corners of the non‐branching landscape). Data are plotted for three replicate experiments ( n = 3), the central line indicates the median, the bottom and top edges indicate the 25th and 75th percentile respectively. Confocal imaging of the branches demonstrated that bacteria successfully colonised the full length of both landscapes during a 15 h incubation period and revealed mixed (both colours) cellular assemblages at the branch extremities indicating coexistence of the two strains (Fig. 2 ). To determine the community composition at the ends of the range expansion, we extracted genomic DNA from 2 mm 2 paper fragments cut from the branch extremities. By utilising qPCR, we assessed the population fraction of GFP‐ versus RFP‐labelled E. coli in the branches using primer pairs amplifying a fragment of the respective genes encoding for the fluorescent proteins. Quantitative PCR showed that the average (global) community composition at the branch extremities did not differ from the community composition at the far edge of the linear system (rank sum test, P < 0.01) nor did either average deviate from the composition of the population in the inoculation zone. Although the averages were similar, the variation in community composition between branches was larger when compared to the variation among patches of equal size in the non‐branching system (one‐sided F ‐test, P = 0.03). The observed increase in interbranch variation in the dendritic network is in agreement with theoretical predictions, yet the effect is rather modest despite the fact that the range expansion covers centimetre distances, that is, c . 10 3 body lengths, indicating that local populations are more similar than would be expected from theory (Muneepeerakul et al . 2007 ; Carrara et al . 2012 ; Paz‐Vinas & Blanchet 2015 ). A possible reason for this is provided by recent theoretical work showing that the distribution of the distance of dispersal events can have a profound influence on the spatial composition of populations, suggesting that a broad distribution of dispersal events (a co‐occurring short‐ and long‐distance dispersal) may increase the local diversity of populations (Paulose et al . 2019 ) decreasing the dissimilarity between local populations. The suggestion that a broad distribution of dispersal distances increases population mixing and thus positively impacts local diversity is of relevance to the experiments performed here, as bacteria switch between two lifestyles each having a distinct dispersal mode (biofilm growth characterised by short‐range dispersal only, and a planktonic form capable of much longer distance dispersal). Interestingly, the relatively balanced population composition at the branch extremities, and the low variation between branches we observe, contrasts findings from a different experimental system commonly used to study bacterial range expansions, namely bacterial colonies growing on solid agar (Hallatschek et al . 2007 ; Hol et al . 2015 ). E. coli are non‐motile on solid agar, and a range expansion of two neutral strains growing on solid agar starting from a mixed point inoculation is governed by a stochastic coarsening process in which a small number of pioneers quickly dominates the expanding front, diminishing local diversity (Hallatschek et al . 2007 ; Hol et al . 2015 ). The coarsening is driven by priority effects in which competition for space and resources at the colony’s perimeter favours the expansion of those already present. In contrast to range expansions on solid agar, the current paper‐based system supports local coexistence of the two strains throughout the range expansion. Two strain coexistence is even observed at micrometre scales within an individual branch (Fig. 2 ), suggesting that coarsening along the range expansion is completely absent in paper scaffolds. As alluded to above, dispersal dynamics can strongly impact community assembly (Paulose et al . 2019 ); it is therefore possible that the stark differences in colonisation dynamics in paper scaffolds compared to solid agar originate from the different modes of dispersal and growth that the two systems support: dispersal by growth and division only on solid agar, versus swimming motility and co‐occurrence of sessile and planktonic lifestyles in paper scaffolds. It is interesting to note that the distinct lifestyles that the paper scaffolds support are an important ingredient of bacterial community assembly in natural habitats (Kolter & Greenberg 2006 ). Taken together, these results suggest that when growth and division are the only modes of dispersal, this leads to a coarsening of the community composition along the range expansion, while habitats that support the full range of dispersal modes promote community mixing to a much larger extent resulting in a high degree of diversity even at local scales. By virtue of the presence cellulose fibres (anchors for sessile bacteria) and liquid‐filled pores, paper substrates may provide a higher degree of environmental heterogeneity when compared to a solid agar surface. As spatial and environmental heterogeneity have been linked to the persistence of transient ecological dynamics (Hastings & Higgins 1994 ; Doebeli & Ruxton 1998 ; Hastings et al . 2018 ), increased environmental heterogeneity (especially in concert with long‐distance dispersal) may be another factor contributing to the long‐term local persistence of two‐strain coexistence in paper scaffolds (which in essence may be transient) while transient dynamics are rapidly quenched on solid agar leading to local extinction of one of the two strains. Case study 2: Bacterial island biogeography We took advantage of the versatile nature of growing bacteria‐in‐paper to explore colonisation in a second, quite different ecological scenario, an archipelago of islands. We constructed a landscape consisting of a ‘mainland’ (used to inoculate the system) and several uninhabited islands situated 5–15 millimetres from the mainland. The (non‐inoculated) islands were impregnated with 1 microgram of glucose and dried. The mainland was inoculated with a 1:1 mix of GFP‐ and RFP‐labelled E. coli , the archipelago was subsequently sandwiched between glass slides and the remaining space (the ‘sea’) filled with minimal medium (lacking a carbon source). The bacteria thus initially faced a low‐nutrient environment dotted with nutrient‐rich islands. Upon wetting the landscape, the solid glucose slowly dissolved and diffused out of the non‐inoculated islands, creating a dynamic and heterogeneous resource landscape. Figure 3 shows that after 15 h of incubation, E. coli from the mainland had successfully colonised the non‐inoculated paper scaffolds and established colonies in the islands. The mainland was coloured yellow due to a uniform mix of green and red cells. Interestingly, a very different distribution of single‐coloured colonies can be seen scattered across the three islands. The single‐colour colonies likely derive from individual colonisers, which gave rise to distinct founder populations. Community structure at the islands may thus be driven by priority effects and differs from the mainland, exhibiting a much lower local diversity (i.e. the characteristic length scale of clonal single‐colour patches is much larger) due to relatively rare colonisation events. The presence of homogeneous (i.e. single colour) regions spanning tens to several hundred micrometres suggests that the number of colonising species on small islands would be severely constrained indicating a positive scaling between island size and species richness, as predicted by classical island biogeography theory (MacArthur & Wilson 2001 ). As this proof‐of‐concept study shows, the ease with which different archipelagoes can be constructed positions the bacteria‐in‐paper approach well to examine the mechanisms giving rise to, for example, species–area and distance–decay relationships in bacterial communities (Green & Bohannan 2006 ). Figure 3 Colonisation of an archipelago. E. coli inoculated on the mainland colonise an initially uninhabited archipelago of three paper islands. Paper scaffolds are surrounded by liquid minimal medium, and (non‐inoculated) islands were pretreated with glucose which starts to diffuse out of the scaffolds upon wetting. Diffusing glucose promotes bacterial migration by creating a temporary glucose gradient increasing towards the non‐inoculated islands. Confocal scans correspond to the areas indicated with 1, 2 and 3 in the cartoon. Scale bars are 200 μm. Dendritic networks and archipelagos are canonical landscapes in ecology. Using no more than paper and scissors, such diverse ecological scenarios can now be explored in habitats structured at the microscopic scales relevant to bacterial ecology. Given the ease with which paper can be cut in milli‐ to centimetre shapes, the platform presented here can be used to address a wide range of questions on how multi‐scale landscape geometry and topology affect bacterial community dynamics. In addition to spatial structure, resource heterogeneity can be incorporated by seeding nutrients locally in the paper giving rise to a rich repertoire of ecosystems that can be modelled in paper. The bacteria‐in‐paper platform provides excellent opportunities to investigate the mechanisms underlying the assembly of experimental (model) communities and characterising how (scaling) relations used to describe macroscopic organisms, relate to microorganisms. These aspects of microbial biogeography have been studied in field samples, yet the level of control and manipulation experimental model systems afford will provide new avenues for the quantitative testing of hypotheses posed by theoretical and modelling studies (Jessup et al . 2004 ). The approach presented here is for instance well suited to study the influence landscape topology has on the spatial scaling of microbial diversity. In addition to assessing the abundance and location of individuals/species (the ‘who and where’ question), high‐resolution fluorescent imaging can provide a view on the physiology of a cell and the expression of relevant genes. Such measurements may provide insight into how the environment influences the physiological state of a cell, and how its state, in turn, influences its neighbours (e.g. by secreting toxins). As the state of a cell and its interactions with neighbours may have profound influences on (competitive) ecological processes, the opportunity to measure such parameters in controlled, yet realistic landscapes, presents a powerful tool for experimental bacterial ecology. Combining the bacteria‐in‐paper platform with next‐generation sequencing will provide a means to scrutinise the genetics of local populations and explore evolutionary dynamics in spatially structured ecosystems. The evolution of cooperation in structured communities, the production of costly common goods and the emergence of resistance in spatial gradients of antibiotics are examples where the combination of bacteria‐in‐paper and deep sequencing could lead to novel observations. By providing a versatile, easy‐to‐use and virtually zero‐cost alternative to microfabrication‐based approaches to experimental microbial ecology, this work fits in a broader push towards democratising science by eliminating the need for expensive and specialised equipment by providing inexpensive alternatives that rely on generic materials and tools."
} | 6,315 |
19703284 | PMC2754497 | pmc | 9,184 | {
"abstract": "Background Acidithiobacillus ferrooxidans gains energy from the oxidation of ferrous iron and various reduced inorganic sulfur compounds at very acidic pH. Although an initial model for the electron pathways involved in iron oxidation has been developed, much less is known about the sulfur oxidation in this microorganism. In addition, what has been reported for both iron and sulfur oxidation has been derived from different A. ferrooxidans strains, some of which have not been phylogenetically characterized and some have been shown to be mixed cultures. It is necessary to provide models of iron and sulfur oxidation pathways within one strain of A. ferrooxidans in order to comprehend the full metabolic potential of the pangenome of the genus. Results Bioinformatic-based metabolic reconstruction supported by microarray transcript profiling and quantitative RT-PCR analysis predicts the involvement of a number of novel genes involved in iron and sulfur oxidation in A. ferrooxidans ATCC23270. These include for iron oxidation: cup (copper oxidase-like), ctaABT (heme biogenesis and insertion), nuoI and nuoK (NADH complex subunits), sdrA1 (a NADH complex accessory protein) and atpB and atpE (ATP synthetase F0 subunits). The following new genes are predicted to be involved in reduced inorganic sulfur compounds oxidation: a gene cluster ( rhd, tusA, dsrE, hdrC, hdrB, hdrA, orf2, hdrC, hdrB ) encoding three sulfurtransferases and a heterodisulfide reductase complex, sat potentially encoding an ATP sulfurylase and sdrA2 (an accessory NADH complex subunit). Two different regulatory components are predicted to be involved in the regulation of alternate electron transfer pathways: 1) a gene cluster ( ctaRUS ) that contains a predicted iron responsive regulator of the Rrf2 family that is hypothesized to regulate cytochrome aa 3 oxidase biogenesis and 2) a two component sensor-regulator of the RegB-RegA family that may respond to the redox state of the quinone pool. Conclusion Bioinformatic analysis coupled with gene transcript profiling extends our understanding of the iron and reduced inorganic sulfur compounds oxidation pathways in A. ferrooxidans and suggests mechanisms for their regulation. The models provide unified and coherent descriptions of these processes within the type strain, eliminating previous ambiguity caused by models built from analyses of multiple and divergent strains of this microorganism.",
"conclusion": "Conclusion • Bioinformatic analysis coupled with gene transcript profiling extends our understanding of the iron and reduced inorganic sulfur compounds oxidation pathways in A. ferrooxidans . • Novel genes predicted to be involved in iron oxidation (Figure 1 ) include those potentially encoding: i) heme biosynthesis and insertion into the terminal electron acceptor cytochrome aa 3 oxidase used in the downhill flow of electrons from Fe(II) to reduce oxygen to water, ii) subunits of the F o ATP synthetase that may be an adaptation to extremely low pHs encountered during iron oxidation, iii) a copper-containing protein that may be involved in electron transfer or in copper insertion, iv) two subunits and one predicted accessory protein of the NADH complex that may promote the flow of electrons from the quinone pool to the NADH complex during uphill electron flow and v) two potential regulators of alternate electron pathways predicted to respond to iron and the status of the quinone pool respectively. • Novel genes predicted to be involved in RISCs oxidation (Figure 2 ) include those potentially encoding: i) three sulfurtransferases, ii) a heterodisulfide reductase complex, iii) an ATP sulfurylase and iv) a NADH complex accessory protein that may promote uphill electron flow from the quinone pool to the NADH complex during RISCs oxidation. • The models (Figures 1 and 2 ) provide unified and coherent descriptions of iron and RISCs oxidation and suggests mechanisms for their regulation within the type strain, eliminating previous confusion caused by models built from analyses of multiple and divergent strains of this microorganism. • The identification of differentially expressed genes of unknown function predicted to be involved in iron and RISCs oxidation direct the experimental biologist to future research.",
"discussion": "Results and Discussion General Features of the Transcriptional Profiles RNA, isolated from mid-log A. ferrooxidans grown in either sulfur (S 0 ) or ferrous iron (Fe(II)) medium, was used to probe gene expression using microarrays displaying unique oligonucleotides representing about 3000 predicted genes of the A. ferrooxidans type strain genome. Using statistical criteria described previously [ 14 ], a 1.5 log ratio of median cut-off (corresponding to genes induced more than 2.8 fold) was selected as indicating differential gene expression in the two growth conditions (Table 1 [ 7 , 24 , 25 ], Table 2 [ 24 - 29 ] and Additional file 1 [ 7 , 24 - 26 , 29 ]). The expression patterns observed with the microarrays were validated for some relevant genes by real-time quantitative PCR (Table 3 [ 7 , 24 - 26 , 28 , 29 ]). One hundred and ninety four genes presented a differential expression profile, of which 110 were upregulated (up to 38 fold) while 84 were downregulated (up to 10 fold) in iron compared to sulfur medium. Genes exhibiting differential expression were grouped by hierarchical clustering and were found to be mostly associated with unknown functions, energy metabolism, cell envelope and central intermediary processes (Additional file 1 [ 7 , 24 - 26 , 29 ]). Table 1 Microarray expression data for iron induced genes ID NC011761 Gene Function log 2 ratio median One sample t-Test (p-value) Proteomic data: strain/references pet I operon AFE_3111 petC1 ubiquinol-cytochrome c reductase, cytochrome c 1 subunit 3,7 0,00* AFE_3110 petB1 ubiquinol-cytochrome c reductase, cytochrome b subunit 4,4 0,00* AFE_3109 petA1 ubiquinol-cytochrome c reductase, iron-sulfur subunit 3,7 0,00* CCM 4252/[ 24 ] AFE_3108 sdrA1 oxidoreductase, short-chain dehydrogenase/reductase family 3,9 0,00* AFE_3107 cycA1 cytochrome c 4 3,7 0,00* rus operon AFE_3153 cyc2 cytochrome c 2,5 0,00* ATCC 33020/[ 7 ] AFE_3152 cyc1 cytochrome c 552 2,8 0,00* ATCC 33020/[ 7 ] CCM 4253/[ 24 ] ATCC 19859/[ 25 ] AFE_3151 cup conserved hypothetical protein 3,1 0,00* ATCC 33020/[ 7 ] AFE_3150 coxB cytochrome c oxidase, aa 3-type, subunit II 2,9 0,00* ATCC 33020/[ 7 ] AFE_3149 coxA cytochrome c oxidase, aa 3-type, subunit I 2,5 0,00* ATCC 33020/[ 7 ] AFE_3148 coxC cytochrome c oxidase, aa 3-type, subunit III 1,7 0,02* ATCC 33020/[ 7 ] AFE_3146 rus rusticyanin 1,6 0,00* ATCC 33020/[ 7 ] CCM 4253/[ 24 ] ATCC 19859/[ 25 ] AFE_3146 rus rusticyanin 2,1 0,00* ATCC 33020/[ 7 ] CCM 4253/[ 24 ] ATCC 19859/[ 25 ] AFE_3146 rus rusticyanin 1,9 0,00* ATCC 33020/[ 7 ] CCM 4253/[ 24 ] ATCC 19859/[ 25 ] Cytochrome c oxidase complex biogenesis operon AFE_3144 ctaA heme A synthase 2,4 0,00* AFE_3143 ctaB heme O synthase, protoheme IX farnesyltransferase 0,4 0,18 AFE_3142 ctaT major facilitator family transporter 0,8 0,00* AFE_3141 ctaR iron responsive regulator of the Rrf2 family ND ND AFE_3139 ctaU hypothetical protein 1,8 0,00* AFE_3138 ctaS oxidoreductase, 2OG-Fe(II) oxygenase family 1,4 0,00* Sensor/regulator two-component signal transduction system AFE_3137 regA DNA-binding response regulator 0,7 0,02 AFE_3136 regB sensor histidine kinase 2,0 0,00* NADH complex operon AFE_2630 nuoA NADH-quinone oxidoreductase, A subunit -0,1 0,80 AFE_2629 nuoB NADH-quinone oxidoreductase, B subunit 0,3 0,13 AFE_2628 nuoC NADH-quinone oxidoreductase, C subunit 0,0 0,69 AFE_2627 nuoD NADH-quinone oxidoreductase, D subunit -0,5 0,64 AFE_2626 nuoE NADH-quinone oxidoreductase, E subunit -1,1 0,05 AFE_2625 nuoF NADH-quinone oxidoreductase, F subunit 0,4 0,01 AFE_2624 nuoG NADH-quinone oxidoreductase, G subunit -0,2 0,24 AFE_2623 nuoH NADH-quinone oxidoreductase, H subunit -0,7 0,00 AFE_2622 nuoI NADH-quinone oxidoreductase, I subunit 1,4 0,00* AFE_2621 nuoJ NADH-quinone oxidoreductase, J subunit -0,1 0,21 AFE_2620 nuoK NADH-quinone oxidoreductase, K subunit 0,9 0,03* AFE_2619 nuoL NADH-quinone oxidoreductase, L subunit -0,9 0,00 AFE_2618 nuoM NADH-quinone oxidoreductase, M subunit -0,8 0,00 AFE_2617 nuoN NADH-quinone oxidoreductase, N subunit -0,4 0,19 ATP synthetase complex operon AFE_3209 atpB ATP synthase F0, A subunit 1,8 0,00* AFE_3208 atpE ATP synthase F0, C subunit 1,4 0,00* AFE_3207 atpF ATP synthase F0, B subunit 0,4 0,01 AFE_3206 atpH ATP synthase F1, delta subunit 0,4 0,02 AFE_3205 atpA ATP synthase F1, alpha subunit -0,2 0,99 AFE_3204 atpG ATP synthase F1, gamma subunit -0,5 0,23 AFE_3203 atpD ATP synthase F1, beta subunit 0,0 0,75 AFE_3202 atpC ATP synthase F1, epsilon subunit -0,6 0,01 Gene expression values (log 2 ratio of median) for all genes/operons alluded in the revised model of Fe(II) oxidation in A. ferrooxidans ATCC 23270. Genes with a log 2 ratio of median larger than |1.5| (corresponding to genes induced more than 2.8 fold) are considered differentially expressed (indicated with *) and genes p-value <0,005 are considered significant. Gene ID is that of Genbank genome annotation NC 011761 . The reference and the strain in which the level of the gene product has been shown to be higher in Fe(II) than in S 0 conditions are indicated in the last column. ND: not determined. Table 2 Microarray expression data for sulfur induced genes ID NC011761 Gene Function log 2 ratio median One sample t-Test (p-value) Proteomic data: strain/references pet II operon AFE_2732 hip High potential iron-sulfur protein -1,8 0,00* AFE_2731 petC2 ubiquinol-cytochrome c reductase, cytochrome c 1 subunit -0,3 0,09 AFE_2730 petB2 ubiquinol-cytochrome c reductase, cytochrome b subunit -1,7 0,00* AFE_2729 petA2 ubiquinol-cytochrome c reductase, iron-sulfur subunit -1,9 0,00* AFE_2728 sdrA2 oxidoreductase, short-chain dehydrogenase/reductase family -1,1 0,00* AFE_2727 cycA2 cytochrome c 4 -0,5 0,17 Heterodisulfide reductase complex operon AFE_2586 hdrB heterodisulfide reductase subunit B, homolog -1,5 0,00* AFE_2558 rhd rhodanese-like domain protein 0,1 0,38 AFE_2557 tusA conserved hypothetical protein -2,4 0,00* AFE_2556 dsrE conserved hypothetical protein -2,1 0,00* AFE_2555 hdrC iron-sulfur cluster-binding protein -2,1 0,00* AFE_2554 hdrB heterodisulfide reductase subunit B, homolog -2,0 0,00* AFE_2553 hdrA pyridine nucleotide-disulfide oxidoreductase -2,6 0,00* AFE_2552 orf2 conserved hypothetical protein ND ND AFE_2551 hdrC iron-sulfur cluster-binding protein -2,4 0,00* AFE_2550 hdrB succinate dehydrogenase/fumarate reductase, C subunit -1,9 0,00* Sulfide-quinone reductase AFE_1792 sqr sulfide-quinone reductase, putative -1,6 0,00* CCM 4253/[ 24 ] NASF-1/[ 26 ] Cytochrome bd ubiquinol oxidase AFE_0955 cydA cytochrome d ubiquinol oxidase, subunit I -2,0 0,00* AFE_0954 cydB cytochrome d ubiquinol oxidase, subunit II -2,6 0,00* Cytochrome bo 3 ubiquinol oxidase AFE_0634 cyoD cytochrome o ubiquinol oxidase, subunit IV -2,3 0,00* AFE_0633 cyoC cytochrome o ubiquinol oxidase, subunit III -3,0 0,00* AFE_0632 cyoB cytochrome o ubiquinol oxidase, subunit I -2,7 0,00* AFE_0631 cyoA cytochrome o ubiquinol oxidase, subunit II -3,2 0,00* Sulfate adenylyltransferase AFE_0539 sat sulfate adenylyltransferase, putative/adenylylsulfate kinase 0,7 0,00* Thiosulfate-quinone oxidoreductase complex operon AFE_0046 conserved hypothetical protein -1,7 0,00* AFE_0045 sulfur/pyrite/thiosulfate/sulfide-induced protein -1,1 0,00* ATCC 19859/[ 25 ] ATCC 19859/[ 28 ] AFE_0044 doxDA Thiosulfate-quinone oxidoreductase, DoxD-like family protein -2,3 0,00* AFE_0043 periplasmic solute-binding protein, putative -2,2 0,00* CCM 4253/[ 24 ] ATCC 23270/[ 25 ] ATCC 23270/[ 29 ] AFE_0042 Tat pathway signal sequence domain protein -1,5 0,00* AFE_0041 C4-dicarboxylate transporter/malic acid transport protein -1,5 0,00* Tetrathionate hydrolase AFE_0029 tetH Tetrathionate hydrolase -0,6 0,22* ATCC 23270/[ 27 ] Gene expression values (log 2 ratio of median) for all genes/operons alluded in the revised model of sulfur oxidation in A. ferrooxidans ATCC 23270. Genes with a log 2 ratio of median larger than |1.5| (corresponding to genes induced more than 2.8 fold) are considered differentially expressed (indicated with *) and genes p-value <0,005 are considered significant. Gene ID is that of Genbank genome annotation NC 011761 . The reference and the strain in which the level of the gene product has been shown to be higher in S 0 than in Fe(II) conditions are indicated in the last column. Table 3 Q-PCR expression data for relevant validated genes ID NC011761 Gene (locus) Function log 2 (Fe/S) Induced Proteomic data: strain/references rus operon AFE_3146 rus rusticyanin 3,8 Fe ATCC 33020/[ 7 ] CCM 4253/[ 24 ] ATCC 19859/[ 25 ] AFE_3151 cup conserved hypothetical protein 4,9 Fe ATCC 33020/[ 7 ] Cytochrome c oxidase complex biogenesis operon AFE_3141 ctaR iron responsive regulator of the Rrf2 family 3,9 Fe Sensor/regulator two-component signal transduction system AFE_3137 regA DNA-binding response regulator 4,3 Fe petI operon AFE_3108 sdrA1 oxidoreductase, short-chain dehydrogenase/reductase family 3,7 Fe AFE_3109 petA1 ubiquinol-cytochrome c reductase, iron-sulfur subunit 6,4 Fe CCM 4252/[ 24 ] AFE_3110 petB1 ubiquinol-cytochrome c reductase, cytochrome b subunit 5,7 Fe AFE_3111 petC1 ubiquinol-cytochrome c reductase, cytochrome c 1 subunit 5,5 Fe Others AFE_2599 - 1,2 Fe AFE_3116 - 1,2 Fe AFE_3119 - 3,4 Fe AFE_3124 cysD sulfate adenylyltransferase, small subunit 4,8 Fe CCM 4253/[ 24 ] Thiosulfate-quinone oxidoreductase complex operon AFE_0043 - periplasmic solute-binding protein, putative -2,5 S CCM 4253/[ 24 ] ATCC 19859/[ 25 ] ATCC 23270/[ 29 ] AFE_0045 - sulfur/pyrite/thiosulfate/sulfide-induced protein -1,2 S ATCC 19859/[ 25 ] ATCC 19859/[ 28 ] Cytochrome bd ubiquinol oxidase AFE_0955 cydA cytochrome d ubiquinol oxidase, subunit I -2,0 S Cytochrome bo 3 ubiquinol oxidase AFE_0632 cyoB cytochrome o ubiquinol oxidase, subunit I -3,2 S Heterodisulfide reductase complex operon AFE_2553 hdrA pyridine nucleotide-disulfide oxidoreductase -1,3 S AFE_2555 hdrC iron-sulfur cluster-binding protein -1,6 S AFE_2586 hdrB heterodisulfide reductase subunit B, homolog -0,8 S Sulfide-quinone reductase AFE_1792 sqr Sulfide-quinone reductase -0,1 ≈ CCM 4253/[ 24 ] NASF-1/[ 26 ] Others AFE_0049 - periplasmic solute-binding protein, putative 0,3 ≈ ATCC 23270/[ 29 ] AFE_1663 glcF glycolate oxidase, iron-sulfur subunit -1,7 S AFE_1677 cbbOIa von Willebrand factor type A domain protein -1,9 S AFE_2971 cysN2 sulfate adenylyltransferase, large subunit -1,3 S AFE_0282 fur ferric uptake regulator 0,6 ≈ AFE_2324 pgm phosphoglucomutase 0,6 ≈ AFE_0445 galU UTP-glucose-1-phosphate uridylyltransferase 0,2 ≈ AFE_1342 epsS UDP-glucose 4-epimerase 0,0 ≈ AFE_2840 malQ glycosyl hydrolase -0,2 ≈ AFE_2836 glbB 1,4-alpha-glucan branching enzyme -0,6 ≈ AFE_3054 cbbOIb von Willebrand factor type A domain protein -0,1 ≈ AFE_2157 cbbOII von Willebrand factor type A domain protein -0,6 ≈ AFE_0539 cysN3 sulfate adenylyltransferase, large subunit 0,1 ≈ AFE_2602 - hypothetical -0,1 ≈ ATCC 19859/[ 25 ] Real time PCR gene expression values (log 2 ratio of median) for relevant genes alluded in the revised iron or sulfur oxidation models for A. ferrooxidans ATCC 23270. Genes with a log 2 ratio of median larger than |1.5| (corresponding to genes induced more than 2.8 fold) are considered differentially expressed (indicated with *). Gene ID is that of Genbank genome annotation NC 011761 . The reference and the strain in which the level of the gene product has been shown to be higher in one of the conditions tested (Fe(II) or S 0 ) are indicated in the last column. Extending the current model of Fe(II) oxidation The rus operon was induced in Fe(II)-grown cells (Table 1 [ 7 , 24 , 25 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]) supporting the current model of the involvement of Cyc2, rusticyanin, Cyc1 and cytochrome oxidase in the oxidation of Fe(II) and the downhill electron transfer chain terminating in the reduction of oxygen to water (Figure 1 ). Embedded in the rus gene operon, is a hypothetical gene of unknown function ( ORF1 , AFE_3151). Its genetic linkage and congruent transcriptional activity suggest that it is involved in Fe(II) oxidation (Table 1 [ 7 , 24 , 25 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]). In agreement with these data, highest expression of ORF1 was detected in iron-compared to sulfur-grown ATCC 33020 cells [ 7 ]. ORF1 exhibits weak similarity to the putative type-3 multicopper oxidase from Halorubrum lacusprofundi (30% similarity, e-value: 8e-05) and to the predicted outer membrane type 3 multicopper oxidase protein Pan1 from Halobacterium sp. NRC-1 (32% similarity; e-value: 3e-04). It also exhibits weak similarity to rusticyanin including conservation of three out of its four critical copper binding residues. In addition, EPR analysis suggests that ORF1 contains copper [ 30 ]. We propose the name Cup ( cup redoxin-like) for ORF1. A possible function for Cup is to deliver copper either to aa 3 cytochrome oxidase or to rusticyanin. The well documented proteins Sco and Cox1 that are involved in copper delivery to copper-containing proteins in other organisms [ 31 ] have been not detected in A. ferrooxidans [ 8 ] and Cup may have assumed their role. Given its similarity to rusticyanin, an alternate hypothesis is that Cup is involved in electron transfer perhaps between Cyc2 and Cyc1 bypassing rusticyanin. Two arguments in favor of this hypothesis are: 1) Cup has been shown experimentally to be physically associated with Cyc1 and not with rusticyanin [ 15 ], and 2) Cup is bound to the outer membrane likely facing the periplasm [Amouric, Yarzabal and Bonnefoy, unpublished results]. Cup could provide an alternative route for electron flow during iron oxidation and an additional point for its regulation. Immediately downstream of the rus operon is a cluster of six genes, upregulated in iron (Table 1 [ 7 , 24 , 25 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]), that are predicted to be involved in cytochrome aa 3 oxidase biogenesis ( ctaABT ) and iron-responsive regulation of cytochrome aa 3 oxidase biogenesis ( ctaRUS ): ctaA (AFE_3144) encoding an integral membrane protein with 7 out of 8 His located in transmembrane regions similar to heme A synthase CtaA ([ 32 ] and references therein), ctaB (AFE_3143) encoding a heme O synthase and ctaT (AFE_3142) encoding an integral membrane protein belonging to the major facilitator family transporter that could be involved in the exportation of heme A to cytochrome oxidase [ 33 ] (Figure 1 ); ctaR (AFE_3141) predicted to encode an iron responsive regulator [ 34 ] of the Rrf2 family that in T. denitrificans is clustered with cbb 3 cytochrome oxidase biogenesis genes (data not shown), ctaU (AFE_3139) encoding a hypothetical protein of unknown function and ctaS (AFE_3138) encoding a predicted Fe(II)-dependent oxygenase superfamily member of unknown function. Immediately downstream but transcribed in the other direction, are two genes regBA (AFE_3136-3137), also upregulated in iron (Table 1 [ 7 , 24 , 25 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]), that are predicted to encode a sensor/regulator two-component signal transduction system of the RegB/RegA family with similarity to RegBA of Rhodobacter capsulatus . RegA directly controls synthesis of cytochrome cbb 3 and ubiquinol oxidases that function as terminal electron acceptors in a branched respiratory chain [ 35 ]. Given the sequence similarity of the predicted A. ferrooxidans RegA with that of R. capsulatus , the conservation of the quinone binding site in the membrane spanning domain and of the redox active cysteine in the cytoplasmic transmitter domain of RegB, we propose that it is also involved in redox sensing. Because of the regBA juxtaposition to genes predicted to encode cytochrome aa 3 oxidase biogenesis, this cytochrome oxidase is a likely candidate for the target of RegBA regulation in A. ferrooxidans . However, in R. capsulatus , RegBA also regulates other genes involved in respiratory electron components such as cytochromes c 2 , c(y) and the cytochrome bc 1 complex, so that the actual RegBA target(s) in A. ferrooxidans requires experimental evaluation. RegBA of R. capsulatus have been shown to respond to the status of the aerobic respiratory chain, most likely the ubiquinone pool in the membrane [ 36 ] and, if this also proves to be the case in A. ferrooxidans , RegBA could help in making regulatory changes to balance electron equivalents between uphill and downhill electron flow, perhaps by adjusting the proportion of cytochrome Cyc1 and cytochrome oxidase aa 3 (downhill electron flow) to the cytochrome CycA1 and the cytochrome bc 1 complex (uphill electron flow). Alternately, RegBA could play a role in switching between iron and sulfur oxidation or between aerobic and anaerobic oxidation. It is clear that the discovery of the predicted iron and redox responsive regulators CtaRUS and RegAB will now allow the experimental biologist to focus on important regulatory switches that are most likely to be involved in cellular decisions related to energy metabolism. The petI operon was also induced in Fe(II) medium (Table 1 [ 7 , 24 , 25 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]) supporting the current model for the role of the bc 1 complex in the uphill flow of electrons during iron oxidation (Figure 1 ). Embedded within the petI operon is sdrA1 (AFE_3108) whose function remains unknown. SdrA1 has the characteristic NAD(P) binding site at its N-terminus ( TG AGE G I G ) of Ndu9 which is a subunit of the NADH complex involved in this complex assembly and stability in a variety of eukaryotes [ 37 ]. However, SdrA1 exhibits the conserved catalytic residues (N125, S147, Y166, K170) involved in electron input into the NADH complex or electron transfer within the complex, suggesting that this protein could function as an oxidoreductase [ 38 , 39 ] (Additional file 2 ). SdrA1 has been predicted to be situated in the cytoplasm in A. ferrooxidans with a possible hydrophobic region embedded in the inner membrane [ 12 ] and we hypothesize that it transfers electrons from the quinone pool to the NADH complex (Figure 1 ). Whereas most of the predicted genes encoding the subunits of the NADH complex (AFE_2630-2617) are equally expressed in Fe(II) and S 0 growth conditions, nuoI (AFE_2622) and nuoK (AFE_2620) are upregulated in Fe(II) medium (Table 1 [ 7 , 24 , 25 ]). The nuoI encodes a ferredoxin located in the cytoplasmic arm of the NADH complex and is involved in the intramolecular electron transfer between FMN and quinone, whereas nuoK encodes a membrane subunit thought to be involved in quinone reduction [ 40 ] and in proton translocation [ 41 ]. Given the predicted locations of NuoI and NuoK in the hinge region of the NADH complex [ 42 ] and the upregulation of nuoI , nuoK and sdrA1 in Fe(II) growth conditions, we suggest that they interact to facilitate uphill electron flow from quinone to the NADH complex (Figure 1 ). Whereas most of the genes predicted to encode the ATP synthetase complex were similarly expressed in Fe(II) and in S 0 grown cells, membrane embedded F0 subunits A and C, encoded by atpB (AFE_3209) and atpE (AFE_3208) respectively, were upregulated in Fe(II) growth conditions (Table 1 [ 7 , 24 , 25 ]). These subunits are involved in proton translocation across the membrane [ 43 ] and their upregulation could allow more protons to pass through the ATP synthetase complex during Fe(II) oxidation, provided an increase in ATP synthesis. However, the resulting increase of intracellular protons requires a concomitant increase in intracellular electrons for their neutralization, so as not to compromise the internal pH of the cell. These electrons could come from the downhill pathway during Fe(II) oxidation as shown in Figure 1 . The organization and regulation of the components of Fe(II) oxidation in A. ferrooxidans appear to be unique to this organism, although Thiobacillus prosperus strain V6 possesses a transcriptional unit upregulated in iron conditions with some similarity to the rus operon [ 44 ]. However, the T. prosperus operon lacks the cyc1 and the rus genes encoding cytochrome c 4 and rusticyanin, respectively. The latter is located downstream from the cluster, is monocistonic and is expressed in iron and sulfur growth conditions [ 44 ], suggesting different regulatory mechanisms compared to A. ferrooxidans . In Leptospirillum group II, the electron transfer chain involved in Fe(II) oxidation contains two cytochromes c and a cbb 3 cytochrome oxidase [ 45 - 47 ] but no blue copper protein such as rusticyanin. In the archaea Ferroplasma acidarmanus a blue copper protein (sulfocyanin) has been suggested to transfer the electrons from Fe(II) to a cbb 3 -type terminal oxidase [ 48 ]. While the genes encoding four blue copper proteins (two sulfocyanin-like and two rusticyanin-like) have been identified in Metallosphaera sedula [ 49 ], none of them respond to the presence of iron in the medium [ 50 ]. However, foxA (cytochrome oxidase subunit I) and soxNL (cytochrome b and [2Fe-2S] Rieske) - cbsAB (cytochromes b ) clusters have been predicted to be important for Fe(II) oxidation in M. sedula . Genes encoding cytochrome c oxidase subunits I and II ( foxAB ) and CbsA-like cytochrome b ( foxC ) have been also proposed to be involved in iron oxidation in Sulfolobus metallicus [ 51 ]. It appears therefore that different pathways for ferrous iron oxidation have evolved in prokaryotes. Extending the current model of the oxidation of reduced inorganic sulfur compounds (RISCs) Predicted genes, proteins or enzymatic activities previously identified in the oxidation of RISCs in different species of A. ferrooxidans include: sqr (AFE_1792) encoding sulfide quinone reductase from NASF-1 strain [ 26 , 52 ], doxDA (AFE_0044) encoding thiosulfate quinone oxidoreductase from ATCC23270 [ 53 ] and from the CCM4253 strain [ 54 ], tetH (AFE_0029) encoding tetrathionate hydrolase from ATCC23270 [ 27 ], cydAB (AFE_0955-0954) encoding a bd oxidase and cyoABCD (AFE_0631-0634) encoding a bo 3 oxidase from ATCC19859 [ 11 ] (See Figure 2 ). All these genes are upregulated in sulfur-containing media relatively to Fe(II) (Table 2 [ 24 - 29 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]). This is also the case for the petII operon (AFE_2727-2732) encoding a second bc 1 complex, a cytochrome c 4 , SdrA2 and a high potential iron-sulfur protein, Hip, as already reported for ATCC19859 [ 11 ], ATCC33020 [ 13 ] and ATCC23270 [ 14 ] strains (Table 2 [ 24 - 29 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]). This second bc 1 complex has been proposed to function directly transferring electrons from sulfur to oxygen (Figure 2 ), and possibly in the aerobic and anaerobic oxidation of sulfur and formate described by Pronk et al . [ 55 ]. In that case, the bc 1 complex receives the electrons from the quinol pool and transfers them to the membrane-bound cytochrome c 4 CycA2 and/or to the periplasmic high potential iron-sulfur protein Hip that subsequently gives the electrons to the terminal oxidase [ 2 , 13 , 14 ] (Figure 2 ). SdrA2, like SdrA1 (see above), may promote electron flow from the quinone pool to the NADH complex (Figure 2 ). These findings support earlier models of the branched electron transfer flow during S 0 oxidation [ 2 , 11 , 13 , 14 ] (Figure 2 ). However, this model is far from being complete and several outstanding questions remain unanswered, including the identification of the enzymes that oxidize sulfur and sulfite. Figure 2 Model of sulfur oxidation in A. ferrooxidans ATCC 23270 . Reduced inorganic sulfur compound (RISC) oxidation pathways are predicted to involve various enzymes, enzyme complexes and a number of electron carriers located in different cellular compartments: in the outer membrane facing the periplasm (tetrathionate reductase, TetH), in the periplasm (high potential iron-sulfur protein, HiPIP), attached to the cytoplasmic membrane on the periplasmic side (cytochrome c , CycA2), in the cytoplasmic membrane (sulfide quinone reductase (SQR), thiosulfate quinone reductase (TQR), bc 1 complex, NADH complex I, bd and bo 3 terminal oxidases) and in the cytoplasm (heterodisulfide reductase (HDR), and ATP sulfurylase (SAT)). Insoluble sulfur is first converted to sulfane sulfate (GSSH) which is then transferred to the heterodisulfide reductase (HDR) through a cascade of sulfur transferases (DsrE, TusA and Rhd). Electrons coming from sulfide (H 2 S), thiosulfate (S 2 O 3 2- ) or sulfane sulfate (GSSH) are transferred via the quinol pool (QH 2 ) either (1) directly to terminal oxidases bd or bo 3 , or indirectly throught a bc 1 complex and a cytochrome c (CycA2) or a high potential iron-sulfur protein (HiPIP) probably to the aa 3 oxidase where O 2 reduction takes place or (2) to NADH complex I to generate reducing power. Sulfur Seven genes, potentially encoding a heterodisulfide reductase complex HdrABC (AFE_2586 and AFE_2555-2550) were highly upregulated in cells grown in sulfur medium (Table 2 [ 24 - 29 ] and Table 3 [ 7 , 24 - 26 , 28 , 29 ]). This complex catalyzes the reversible reduction of the disulfide bond X-S-S-X coupled with energy conservation in methanogenic archaea ([ 56 ] and references therein) and sulfate reducing archaea and bacteria [ 57 ]. This complex in A. ferrooxidans is predicted to have three different cytoplasmic HdrB subunits (AFE_2586, 2554 and 2550) with the cysteine-rich domain which binds the unusual type [4Fe-4S] cluster [ 58 ], involved in disulfide reduction [ 56 , 59 ]. In addition, genes potentially encoding the ferredoxin-like and the flavoprotein subunits, HdrC (AFE_2555 and 2551) and HdrA (AFE_2553) are present in the same locus. The flavoprotein HdrA subunit exhibits a possible N-terminal membrane spanning region, a conserved FAD binding site (GXGXXGX 16–19 (D/E)), and the conserved four cysteine cluster (CXGXRDX 6–8 CSX 2 CC) that binds a Fe-S center, typical of the cytoplasmic iron-sulfur flavoprotein type enzyme. Three genes encoding for sulfur metabolism accessory proteins, placed immediately upstream of the heterodisulfide reductase complex, are similarly upregulated (Table 2 [ 24 - 29 ]). These encode a cytoplasmic rhodanase-related sulfurtransferase (AFE_2558, COG0607) [ 60 ], a cytoplasmic SirA-like disulfide bond formation regulator (AFE_2557, pfam01206, COG0425, IPR001455) and an inner membrane located peroxiredoxin of the DrsE superfamily (AFE_2556, COG2210 and 2044). The rhodanase AFE_2558 is 48% similar to the Sud protein from Wolinella succinogenes which binds and transfers polysulfide sulfur to the polysulfide reductase located in the cytoplasmic membrane [ 61 , 62 ]. In turn, the SirA-like protein AFE_2557 is 54% similar to the TusA protein from Vibrio fischeri and other bacteria which belong to a complex sulfur-relay system that facilitates specific sulfur flow/trafficking from various pathways [ 63 - 65 ]. Finally, DsrE family proteins such as that predicted to be encoded by AFE_2556 are small proteins recently shown to be involved in sulfur transfer reactions during sulfur oxidation [ 66 ]. Significant sequence similarity and conserved gene organization of the A. ferrooxidans heterodisulfide reductase complex and accessory proteins is restricted only to Aquifex aeolicus and known acidophilic sulfur oxidizing microorganisms Hydrogenobaculum sp. Y04AAS1, Hydrogenivirga sp. 128-5-R1-1, Metallosphaera sedula [ 50 ], Sulfolobus acidocaldarius , S. tokodaii and S. solfataricus (Figure 3 ), associating the whole gene cluster with sulfur oxidation. In methanogenic and sulfate reducing archaea, HdrA receives the electrons from the hydrogenase and transfers them through HdrC to the heterodisulfide reductase catalytic site located in HdrB. Accompanying the reduction of heterodisulfide, protons are extruded across the membrane creating a proton motive force. We hypothesize that in A. ferrooxidans and the sulfur oxidizers referred above, the Hdr complex, driven by the naturally existing proton gradient, could be working in reverse and oxidizing disulfide intermediaries (from sulfur oxidation) to sulfite and delivering the collected electrons to the membrane quinol pool (Figure 2 ). In addition, the three accessory sulfurtransferases are likely involved in the transfer of the proposed sulfane sulfur [ 67 , 68 ] to the heterodisulfide reductase (Figure 2 ). Figure 3 Comparison of the hdr cluster between A. ferrooxidans ATCC 23270 and other sulfur oxidizers . Heterodisulfide reductase complex (HdrC 1 B 1 AOrf2HdrC 2 B 2 ), accessory proteins (Rhd, TusA, DsrE) and ATP sulfurylase (Sat) in AF: A. ferrooxidans ATCC 23270 (NC_011206), AA: Aquifex aeolicus (NC_000918) and known acidophilic sulfur oxidizing microorganisms HB: Hydrogenobaculum sp. Y04AAS1 (NC_011126), HV: Hydrogenivirga sp. 128-5-R1-1 (NZ_ABHJ00000000), MS: Metallosphaera sedula (NC_009440), SA: Sulfolobus acidocaldarius (NC_007181), ST: S. tokodaii (NC_003106) and SS: S. solfataricus (NC_002754). Percentage of amino-acid similarity is indicated. Blue triangles represent inversion in the gene order. The predicted A. ferrooxidans heterodisulfide reductase complex has additional features that are consistent with its proposed role in RISCs oxidation. Sulfur oxidation in a variety of different A. ferrooxidans strains has been shown to require glutathione [ 67 , 68 ] and requires a neutral pH optimum [ 67 ], suggesting a cytoplasmic activity in agreement with the predicted localization of the heterodisulfide reductase catalytic site (Figure 2 ). Second, non-heme iron and labile sulfur have been shown to be present in a sulfur oxidizing enzyme preparation of A. ferrooxidans [ 67 ] and iron-sulfur clusters are predicted to be present in the A. ferrooxidans HdrB and C subunits. Third, sulfur oxidation has been shown to be inhibited by HQNO in A. ferrooxidans [ 69 ], in agreement with the proposal that the quinone pool is the physiological electron acceptor (Figure 2 ). Fourth, the actual substrate of the sulfur oxidizing enzyme in A. ferrooxidans is thought not to be elemental sulfur, which has poor water solubility and cannot enter the cell, but rather sulfane sulfur of GS n H species ( n >1) and most likely GSSH [ 67 , 68 ]. Sulfane sulfate would thus provide the necessary disulfide bond X-S-S-X to serve as a substrate for the predicted catalytic activity of the A. ferrooxidans heterodisulfide reductase (Figure 2 ). Conserved within the heterodisufide reductase gene cluster, between hdrA and hdrC2 , is a putative gene of unknown function (AFE_2552) whose product is predicted to reside in the cytoplasm (Figure 2 ). Given its conserved gene context, we propose that it is also involved in RISCs oxidation and it is now pinpointed for experimental investigation. Sulfite Another step in the sulfur oxidation model that awaits genetic characterization is sulfite oxidation. Being metastable and considered short-lived in mine waste environments, one possibility is that sulfite rapidly oxidizes non-enzymically to sulfate, thiosulfate or glutathione S-sulfonate in the presence of Fe(III) [ 70 , 71 ] or sulfur [ 72 ]. However, the involvement of a protein catalyzing this reaction is more likely since a sulfite oxidase activity was purified from three different strains of A. ferrooxidans , namely TM [ 73 ], ATCC13661 [ 74 ] and AP19-3 [ 75 , 76 ] Genes coding for known periplasmic enzymes involved in the direct oxidation of sulfite during sulfur dissimilatory metabolism ( sorAB or soxCD [ 77 ]) have not been detected in the A. ferrooxidans genome. In our model (Figure 2 ), sulfite is hypothesized to be produced in the cytoplasm by heterodisulfide reductase. Therefore, subsequent oxidation of sulfite is likely to occur in this cellular location and therefore would not be expected to proceed via the classical periplasmic Sor or Sox. One possibility for the cytoplasmic activity is that sulfite is converted to adenosine-5'-phosphosulfate (APS) via the well characterized APS reductase complex encoded by aprBA [ 78 - 80 ]. However, the genome contains no candidates with significant similarity to aprBA , although it does have a predicted sat (AFE_0539) which, in other microorganisms, encodes an ATP sulfurylase responsible for the second step in this pathway. The A. ferrooxidans Sat shares 44% identity and 60% similarity with both domains of the bifunctional SAT/APS kinase from Aquifex aeolicus that catalyzes the production of ATP and sulfate from APS and pyrophosphate [ 81 , 82 ]. If Sat is indeed catalyzing APS to sulfate (Figure 2 ), an enzyme catalyzing the oxidation of sulfite to APS is required. This missing function could be accomplished by the conserved hypothetical gene embedded in the hdr locus of sulfur oxidizers (Figure 3 ). The concordance of gene occurrence and organization between A. aeolicus , Hydrogenobaculum sp. Y04AAS1 the Sulfolobales and A. ferrooxidans including 1) the hdr locus with a gene of unknown function 2) sat and 3) a lack of aprBA , strongly suggests that these microorganisms have a novel sulfur oxidation pathway. Our data agree with the model proposed for S 0 oxidation in M. sedula , including a heterodisulfide reductase ( hdr ), a tetrathionate hydrolase ( tetH ), a terminal oxidase complex based on both quinol oxidase ( soxCL ) and aa 3 -type cytochrome oxidase ( soxAB ) components [ 50 ]. In S. metallicus , a gene ( sor ), encoding sulfur oxygenase reductase, which is absent in A. ferrooxidans , is the dominant transcript in sulfur-grown cells and is therefore proposed to be involved in sulfur oxidation [ 51 ]. In conclusion, the RISCs oxidation pathways of acidophiles are not only different from those of neutrophilic sulfur oxidizers [ 17 , 18 ] but also appear to be different among the acidophilic sulfur-oxidizers including between members of the Acidithiobacillus genus [ 83 ]. In that sense, the conservation of the hdr locus in different acidophilic sulfur oxidizers, archaea and eubacteria, is noteworthy and merits further investigation. Additional Discussion Genomic and transcriptomic (microarrays and real-time quantitative PCR) studies of iron and sulfur energetic metabolism of A. ferrooxidans , not only confirm previous data, but elaborate on the complexity of these pathways. Both iron and RISCs respiratory chains are branched and redundant [ 2 , 11 , 13 , 14 ] (Figures 1 and 2 ). This provides A. ferrooxidans with a flexible respiratory system that may allow it to adapt efficiently to environmental changes by modulating gene expression according to the growth conditions (substrate, oxygen concentration, growth phase, etc.). Another way to adapt efficiently to a change in the growth conditions is by modifying the association of complexes as suggested recently for iron respiration in A. ferrooxidans [ 15 ]. In the case of the cytochrome c oxidase complex of Dictyostelium discoideum , the oxygen concentration induces a switch between two interchangeable subunit isoforms of the cytochrome c oxidase [ 84 - 86 ]. This switch has been shown to be due to transcriptional regulation and also to different stabilities of the two subunits toward oxygen [ 86 ]. Both iron and RISCs oxidation pathways involve outer membrane, periplasmic and inner membrane components (Figures 1 and 2 ). According to our hypothesis, several super-complexes spanning the outer and/or the inner membranes are expected to conduct either the electrons to the oxygen, or the sulfane-sulfur to the catalytic side of the herodisulfide reductase, from pyrite (FeS 2 ), which is a natural substrate of A. ferrooxidans . While such a super-complex has been isolated recently for iron oxidation [ 15 ], no biochemical data are available until now to substantiate the existence of a sulfane-sulfur transfer supramolecular structure. While a sulfurtransferase complex encoded by the hdr locus is likely to be involved in the transfer of the proposed sulfane sulfur to the heterodisulfide reductase from the inner membrane to the cytoplasm (Figure 2 ), expression data suggests no obvious upregulated outer membrane protein as proposed by Rohwerder and Sand [ 68 ] to allow it to cross the cell wall. We propose also the existence of a cytoplasmic super-complex catalyzing both the oxidation of sulfane-sulfur to sulfite and of sulfite to APS, preventing the accumulation of sulfite in the cytoplasm (Figure 2 ). While the existence of such a complex has not been demonstrated in A. ferrooxidans , a thiosulfate-oxidizing system oxidizing hydrogen sulfide, thiosulfate, sulfur and sulfite directly to sulfate without the presence of free intermediates has been evidenced in Paracoccus versutus and Paracoccus pantotrophus ([ 17 , 87 ] and references therein). Moreover, the methanogenic and sulfate reducing archaea heterodisulfide reductase forms a tight complex with the hydrogenase, which catalyzes its reduction with H 2 ([ 56 , 57 ] and references therein). Such supramolecular structures will allow (1) stabilization of the different components (2) electron, or sulfane sulfur, channeling leading to more efficient transfer, and (3) diffusion avoidance preventing toxic compound leakage.\n\nAdditional Discussion Genomic and transcriptomic (microarrays and real-time quantitative PCR) studies of iron and sulfur energetic metabolism of A. ferrooxidans , not only confirm previous data, but elaborate on the complexity of these pathways. Both iron and RISCs respiratory chains are branched and redundant [ 2 , 11 , 13 , 14 ] (Figures 1 and 2 ). This provides A. ferrooxidans with a flexible respiratory system that may allow it to adapt efficiently to environmental changes by modulating gene expression according to the growth conditions (substrate, oxygen concentration, growth phase, etc.). Another way to adapt efficiently to a change in the growth conditions is by modifying the association of complexes as suggested recently for iron respiration in A. ferrooxidans [ 15 ]. In the case of the cytochrome c oxidase complex of Dictyostelium discoideum , the oxygen concentration induces a switch between two interchangeable subunit isoforms of the cytochrome c oxidase [ 84 - 86 ]. This switch has been shown to be due to transcriptional regulation and also to different stabilities of the two subunits toward oxygen [ 86 ]. Both iron and RISCs oxidation pathways involve outer membrane, periplasmic and inner membrane components (Figures 1 and 2 ). According to our hypothesis, several super-complexes spanning the outer and/or the inner membranes are expected to conduct either the electrons to the oxygen, or the sulfane-sulfur to the catalytic side of the herodisulfide reductase, from pyrite (FeS 2 ), which is a natural substrate of A. ferrooxidans . While such a super-complex has been isolated recently for iron oxidation [ 15 ], no biochemical data are available until now to substantiate the existence of a sulfane-sulfur transfer supramolecular structure. While a sulfurtransferase complex encoded by the hdr locus is likely to be involved in the transfer of the proposed sulfane sulfur to the heterodisulfide reductase from the inner membrane to the cytoplasm (Figure 2 ), expression data suggests no obvious upregulated outer membrane protein as proposed by Rohwerder and Sand [ 68 ] to allow it to cross the cell wall. We propose also the existence of a cytoplasmic super-complex catalyzing both the oxidation of sulfane-sulfur to sulfite and of sulfite to APS, preventing the accumulation of sulfite in the cytoplasm (Figure 2 ). While the existence of such a complex has not been demonstrated in A. ferrooxidans , a thiosulfate-oxidizing system oxidizing hydrogen sulfide, thiosulfate, sulfur and sulfite directly to sulfate without the presence of free intermediates has been evidenced in Paracoccus versutus and Paracoccus pantotrophus ([ 17 , 87 ] and references therein). Moreover, the methanogenic and sulfate reducing archaea heterodisulfide reductase forms a tight complex with the hydrogenase, which catalyzes its reduction with H 2 ([ 56 , 57 ] and references therein). Such supramolecular structures will allow (1) stabilization of the different components (2) electron, or sulfane sulfur, channeling leading to more efficient transfer, and (3) diffusion avoidance preventing toxic compound leakage."
} | 11,262 |
19369953 | null | s2 | 9,185 | {
"abstract": "Many bacterial systems rely on dynamic genetic circuits to control crucial biological processes. A major goal of systems biology is to understand these behaviours in terms of individual genes and their interactions. However, traditional techniques based on population averages 'wash out' crucial dynamics that are either unsynchronized between cells or are driven by fluctuations, or 'noise', in cellular components. Recently, the combination of time-lapse microscopy, quantitative image analysis and fluorescent protein reporters has enabled direct observation of multiple cellular components over time in individual cells. In conjunction with mathematical modelling, these techniques are now providing powerful insights into genetic circuit behaviour in diverse microbial systems."
} | 195 |
27557104 | null | s2 | 9,186 | {
"abstract": "Large multiscale neuronal network simulations are of increasing value as more big data are gathered about brain wiring and organization under the auspices of a current major research initiative, such as Brain Research through Advancing Innovative Neurotechnologies. The development of these models requires new simulation technologies. We describe here the current use of the NEURON simulator with message passing interface (MPI) for simulation in the domain of moderately large networks on commonly available high-performance computers (HPCs). We discuss the basic layout of such simulations, including the methods of simulation setup, the run-time spike-passing paradigm, and postsimulation data storage and data management approaches. Using the Neuroscience Gateway, a portal for computational neuroscience that provides access to large HPCs, we benchmark simulations of neuronal networks of different sizes (500-100,000 cells), and using different numbers of nodes (1-256). We compare three types of networks, composed of either Izhikevich integrate-and-fire neurons (I&F), single-compartment Hodgkin-Huxley (HH) cells, or a hybrid network with half of each. Results show simulation run time increased approximately linearly with network size and decreased almost linearly with the number of nodes. Networks with I&F neurons were faster than HH networks, although differences were small since all tested cells were point neurons with a single compartment."
} | 364 |
35994371 | PMC9447851 | pmc | 9,187 | {
"abstract": "Abstract Bacteria and lytic viruses (phages) engage in highly dynamic coevolutionary interactions over time, yet we have little idea of how transient selection by phages might shape the future evolutionary trajectories of their host populations. To explore this question, we generated genetically diverse phage-resistant mutants of the bacterium Pseudomonas syringae . We subjected the panel of mutants to prolonged experimental evolution in the absence of phages. Some populations re-evolved phage sensitivity, whereas others acquired compensatory mutations that reduced the costs of resistance without altering resistance levels. To ask whether these outcomes were driven by the initial genetic mechanisms of resistance, we next evolved independent replicates of each individual mutant in the absence of phages. We found a strong signature of historical contingency: some mutations were highly reversible across replicate populations, whereas others were highly entrenched. Through whole-genome sequencing of bacteria over time, we also found that populations with the same resistance gene acquired more parallel sets of mutations than populations with different resistance genes, suggesting that compensatory adaptation is also contingent on how resistance initially evolved. Our study identifies an evolutionary ratchet in bacteria–phage coevolution and may explain previous observations that resistance persists over time in some bacterial populations but is lost in others. We add to a growing body of work describing the key role of phages in the ecological and evolutionary dynamics of their host communities. Beyond this specific trait, our study provides a new insight into the genetic architecture of historical contingency, a crucial component of interpreting and predicting evolution.",
"introduction": "Introduction Pathogens are ubiquitous and exert strong selection on their hosts to evade infection ( Stenseth and Maynard Smith 1984 ; Gómez et al. 2010 ). These selection pressures are constantly in flux, and defense-related traits are often detrimental when pathogens are not present at high levels ( Clay and Kover 1996 ; Sheldon and Verhulst 1996 ). The loss of costly resistance under relaxed selection has been the focus of a plethora of theoretical and empirical studies, in large part because it helps to explain the observed coexistence of resistant and sensitive host types in many natural populations ( Waterbury and Valois 1993 ; Stahl et al. 1999 ; Rodriguez-Brito et al. 2010 ; Lourenço et al. 2020 ). Whether host populations readily regress to susceptibility after escape from pathogen pressure or retain a signature of their coevolutionary history will depend on several factors, including the environmental conditions and the strength of selection. For example, resistance may persist if it is not costly to maintain; or if compensatory mutations reduce the fitness costs without reversing the trait itself, as is often observed in drug-resistant bacteria ( Faria et al. 2015 ; Durão et al. 2018 ). In other cases, even when reversion to sensitivity would be favorable, it may not be possible if the host population has since acquired other mutations that would be deleterious on the wild-type genetic background ( Shah et al. 2015 ). In a more general sense, any mutation that affects an organism’s phenotype and is selected for (even temporarily) can alter the selection acting on subsequent mutations and can therefore shape the evolutionary trajectory of the population. For example, garter snakes that prey on tetrodotoxin-bearing newts have evolved high levels of toxin resistance, but only within lineages that already carried a prior substitution—an ancient modification to a sodium channel that took place long before the newts arose ( McGlothlin et al. 2016 ). In a laboratory evolution experiment with Escherichia coli , different substitutions in a DNA topoisomerase enzyme were shown to have different consequences for the subsequent accumulation of other beneficial mutations. In fact, this second-order selection for evolvability was more important than the initial effects of the substitutions themselves in determining which lineage prevailed in the long term ( Woods et al. 2011 ). Historical contingencies such as these can make it particularly challenging to interpret patterns of genetic divergence across populations or predict the lasting consequences of short-term coevolutionary interactions. To explore how previous selection by phages can alter future bacterial evolution, we tracked phage resistance over time in experimental evolution populations of the bacterium Pseudomonas syringae . Bacteria and phages are a tractable model system frequently used for studying coevolution in the laboratory ( Brockhurst et al. 2007 ; Dennehy 2012 ). Phages initiate infection by recognizing and binding to proteins on the surfaces of bacterial cells. Bacteria can evade phages by altering or deleting phage receptors, but in doing so, often compromise other fitness-related traits such as nutrient uptake, adhesion, and virulence ( Dy et al. 2014 ; Mangalea and Duerkop 2020 ). In these cases, as resistance spreads in a population and phage densities decrease, resistance is predicted to be lost over time as a result of relaxed selection ( Koskella 2018 ). Laboratory fluctuation assays show that the rates of spontaneous genetic reversion from resistance to susceptibility can be high ( Chaudhry et al. 2018 ), suggesting that phage sensitivity can (re)emerge within resistant populations. Despite these predicted dynamics, studies of natural bacterial communities sometimes find that bacteria remain resistant to phages that they coexisted with many months or even years in the past ( Koskella and Parr 2015 ; LeGault et al. 2021 ; Dewald-Wang et al. 2022 ). In cases where laboratory studies propagated resistant bacteria for many generations in the absence of phages, whether phage sensitivity re-emerged has remained inconsistent and difficult to explain ( Meyer et al. 2010 ; Avrani and Lindell 2015 ). For example, over 45,000 generations of relaxed selection did not reduce the observed resistance of E. coli to T6 phage ( Meyer et al. 2010 ). In contrast, some experimental populations of Prochlorococcus became less resistant in the absence of phage, but these changes were difficult to explain in terms of compensatory adaptation, as they often occurred independently of fitness gains ( Avrani and Lindell 2015 ). On the basis of these observations, we hypothesized that bacteria can access many different genetic pathways to phage resistance, each with different implications for the subsequent evolutionary potential of the bacteria, including whether they compensate for fitness costs by re-evolving phage sensitivity. To test this idea, we isolated and sequenced P. syringae colonies that had evolved resistance across a panel of lytic phages and measured the fitness costs of each resistance mutation. We then used each resistant strain to seed a different population that was experimentally evolved in the absence of phages. Through a series of laboratory evolution experiments, we demonstrate that phage resistance can either be reversible or entrenched depending on the initial genetic path to resistance.",
"discussion": "Discussion Phages are ubiquitous in microbial communities and are expected to play a central role in bacterial evolution, with critical implications for bacteria–bacteria and bacteria–host interactions ( Thurber 2009 ; Koskella and Brockhurst 2014 ). Predation by phages can maintain population and community diversity in their bacterial hosts ( Weinbauer and Rassoulzadegan 2004 ), regulate the dissemination of antibiotic resistance genes ( Burmeister, Fortier, et al. 2020 ; LeGault et al. 2021 ), select for hypermutator strains ( Pal et al. 2007 ), and alter competitive outcomes among bacterial species ( Bohannan and Lenski 2000 ). Here, we show that even transient exposure to phages can have lasting consequences for the evolutionary trajectories of bacterial populations. Through a series of laboratory evolution experiments, we demonstrate that phage resistance in the bacterium P. syringae can be reversible or entrenched depending on the original genetic path to resistance. Phage resistance in our study primarily occurred through mutations in lipopolysaccharide biosynthesis genes, which have been previously implicated in phage resistance in this bacterial species and others ( Picken and Beacham 1977 ; Evans et al. 2010 ; Meaden et al. 2015 ; Kulikov et al. 2019 ; Holtappels et al. 2020 ). One population did not appear to have any fixed genetic differences from the ancestor despite its resistant phenotype. Resistance in this case may be conferred by mutations that are not adequately resolved by short-read sequencing, such as copy number variation or sequence region inversions, or through unstable genetic changes such as phase variations ( Kircher and Kelso 2010 ; Bull et al. 2014 ). Of note, this population remained phenotypically resistant throughout the evolution experiment, possibly because it did not experience detectable fitness costs relative to its phage-sensitive ancestor. Even though the resistance mutations in our study were concentrated within a handful of genes, they occurred at many unique positions and altered the amino acid sequences in a variety of ways, including missense and nonsense substitutions and frameshift mutations of vastly different sizes. Such convergence in resistance mechanisms at the gene level, and diversity at the sequence level, is consistent with the expectation that phage infection relies on specific receptor structures and that recognition can easily be disrupted by modifying or deleting these structures in one of many ways ( Dy et al. 2014 ). However, resistance was generally costly in the absence of phages, as is often observed in this system and others ( Koskella et al. 2012 ; Meaden et al. 2015 ; Scanlan et al. 2015 ; Burmeister, Fortier, et al. 2020 ; Burmeister, Sullivan, et al. 2020 ). Costs of resistance may stem from alterations to lipopolysaccharide molecules that destabilize the bacterial membrane or reduce surface adhesion ( Mangalea and Duerkop 2020 ). The commonly observed costs of phage resistance suggest that resistance might be selected against in the absence of phage pressure, yet previous studies of this question have produced inconsistent results ( Meyer et al. 2010 ; Avrani and Lindell 2015 ; Wielgoss et al. 2016 ). When we propagated the bacterial populations for an extended period in the absence of phages, we observed that phage sensitivity re-appeared and swept to high frequencies in several populations but that the majority of populations remained resistant. In the case of one population, this reversion to sensitivity was due to a mutation that restored the reading frame of the original sequence, but in other cases, populations re-evolved phage sensitivity without reversing the original mutation. The diversity of phage-resistant mutants in this study allowed us to ask whether the re-evolution of phage sensitivity was contingent on the mechanisms and/or costs of phage resistance. Costs of trait maintenance are often expected to predict trait loss under relaxed selection ( Lahti et al. 2009 ; Meyer et al. 2010 ), yet we did not observe a relationship between the magnitude of fitness costs and the re-evolution of phage sensitivity in our study. This observation, along with the fact that populations that did not re-evolve sensitivity nevertheless improved their fitness to match their phage-sensitive counterparts, suggests the existence of compensatory mutations that reduce the costs of resistance. Compensatory mutations could eventually restore fitness levels to a point where phage sensitivity is no longer advantageous ( Teotónio and Rose 2007 ; McCandlish et al. 2016 ; Pennings et al. 2022 ), or may even be disadvantageous or lethal on the ancestral background, further discouraging reversion of the original trait ( Rojas Echenique et al. 2019 ). Therefore, it appears that trade-offs between phage resistance and other aspects of fitness can be strong, yet bacteria are able to access two evolutionary pathways (reversion and compensation) in response. We hypothesized that the probabilities of these two pathways could be contingent on the genetic mechanism underlying phage resistance. Some traits might have a greater supply of reversion mutations than others; for example, there may be more mutations that restore the ancestral expression levels of a gene than those that reconstitute the exact 3D structure of a receptor ( Wielgoss et al. 2016 ), or more mutations that reverse duplications than point mutations ( Chaudhry et al. 2018 ). Similarly, resistance mechanisms may impact bacterial fitness in different ways, thus requiring different sets of compensatory mutations to restore their costs ( Rojas Echenique et al. 2019 ). We did not find an overall correlation between the initial resistance gene and whether phage sensitivity re-evolved but with two important caveats. First, when resistance was acquired through a large deletion in a receptor biosynthesis gene, it was never reversed, whether through a complementary insertion or other mutations. This suggests that the mutations that would reverse such a large change are so vanishingly rare that other compensatory mutations are much more likely to appear first. Our second observation was that the genetic replicates in the mutant panel (i.e., the populations that independently acquired resistance via the same mutation) all followed the same phenotypic trajectories. Specifically, they all remained resistant throughout experimental evolution in the absence of phage. This suggested that historical contingencies—if they existed here—might be generated not from the identity of the resistance gene but from the exact genetic sequence. Even within the same resistance gene, some mutations might be more reversible than others. As the original experiment was not explicitly focused on this possibility, we designed a follow-up experiment that “replayed” experimental evolution ten times per founding genotype (inspired by Stephen Jay Gould’s famous thought experiment about replaying the tape of life, [ Gould 1990 ]). Strikingly, we found that the evolutionary outcomes of the populations in our replay experiment closely mirrored those of their founders. Populations whose founders had re-evolved phage sensitivity also tended to re-evolve phage sensitivity at high rates, whereas populations whose founders had remained resistant tended to remain resistant as well. Variation in the reversibility of different resistance mutations may occur if different sets of compensatory mutations are required to restore their costs. To explore this possibility, we compared the similarity of acquired mutations among pairs of the experimentally evolving populations. We found that populations with the same initial resistance gene evolved more in parallel with one another than populations with different resistance genes, suggesting that compensatory adaptation in phage-resistant bacteria also depends on evolutionary history. Several other studies have observed greater genomic parallelism among experimental evolution populations with similar starting genotypes for other traits, including in antibiotic resistance evolution ( Card et al. 2021 ) and in compensatory adaptation after gene deletion (Rojas Echenique et al. 2019). And outside of the laboratory, convergence in sequence evolution appears to be more common among populations with a recent common ancestor than a distant one ( Conte et al. 2012 ; Goldstein et al. 2015 ). Notably, the effect we observed was strong when the phage selection event was recent but was no longer detectable after populations had spent an extended period of time in the same environment. Thus, recent historical differences appear to be more important than distant historical differences in shaping subsequent evolution ( Travisano et al. 1995 ; Yen and Papin 2017 ; Santos-Lopez et al. 2021 ). Our study provides evidence for an evolutionary ratchet in bacteria–phage coevolution, where compensatory adaptation enables the persistence of certain resistance mutations even after the original selection pressure has ceased to operate. Why did history constrain evolution in this study but not in some others ( Weinreich et al. 2006 ; Feldman et al. 2012 ; Toledo et al. 2016 )? One important clue may lie in the underlying genetic structure of phage resistance. The phage-sensitive ancestor in this study could mutate to phage resistance through any one of at least 17 individual mutations, including nonsense mutations and large deletions. Phage adsorption relies on highly specific molecular interactions ( Sharma et al. 2008 ), suggesting that there are many ways to change lipopolysaccharide structure to avoid phage recognition, and that at least some of these mutations are not lethal to the cell. In other cases, traits can evolve in parallel across lineages despite their historical differences. For example, several distantly related groups of animals have identical substitutions conferring tetrodotoxin resistance ( Toledo et al. 2016 ). In this case, mutagenic screens have identified additional possible mutations that confer toxin resistance, but these additional mutations are so detrimental to sodium channel performance—an essential function—that they are not observed in nature ( Feldman et al. 2012 ). The role of history in evolution is therefore likely related to the target size of mutations that confer novel functions yet are not overly disruptive to the original trait. Our findings add to a growing body of work that selection by phages plays a key role in the ecological and evolutionary dynamics of bacterial communities. Further, many of the bacterial populations in our study acquired mutations in genes known to interact with eukaryotic immune systems. Phage resistance was directly mediated in many cases by changes to lipopolysaccharide molecules, which are recognized by both animal and plant immune systems ( Triantafilou and Triantafilou 2005 ; Newman et al. 2007 ), and many populations also acquired mutations in Type III effector proteins that underlie bacterial virulence. It will thus be important to characterize whether and how phages are indirectly responsible for shaping coevolution in between bacteria and eukaryotes as well ( Wahida et al. 2021 )."
} | 4,635 |
26991294 | PMC4874681 | pmc | 9,188 | {
"abstract": "Little is known regarding how the increased diversity of nitrogen-fixing bacteria contributes to the productivity and diversity of plants in complex communities. However, some authors have shown that the presence of a diverse group of nodulating bacteria is required for different plant species to coexist. A better understanding of the plant symbiotic organism diversity role in natural ecosystems can be extremely useful to define recovery strategies of environments that were degraded by human activities. This study used ARDRA, BOX-PCR fingerprinting and sequencing of the 16S rDNA gene to assess the diversity of root nodule nitrogen-fixing bacteria in former bauxite mining areas that were replanted in 1981, 1985, 1993, 1998, 2004 and 2006 and in a native forest. Among the 12 isolates for which the 16S rDNA gene was partially sequenced, eight, three and one isolate(s) presented similarity with sequences of the genera Bradyrhizobium , Rhizobium and Mesorhizobium , respectively. The richness, Shannon and evenness indices were the highest in the area that was replanted the earliest (1981) and the lowest in the area that was replanted most recently (2006).",
"introduction": "Introduction Nitrogen-fixing bacteria are an extremely important group of microorganisms for various ecosystems because they promote the entry of nitrogen into the soil. The capacity to fix atmospheric nitrogen is widely distributed among microorganisms with different levels of phylogenetic relationships, including representatives of Archaea and Eubacteria. However, the capacity to fix atmospheric nitrogen and induce nodule formation in leguminous plants is restricted to members of the proteobacteria phylum. 1 , 2 , 3 , 4 Legume nodulating nitrogen-fixing bacteria, which are commonly known as rhizobia, are abundant in the soil of many ecosystems 5 and have a high diversity and variability regarding symbiotic efficiency. 6 , 7 , 8 The importance of symbiotic biological nitrogen fixation (BNF) in agricultural systems is well-documented in plant species such as soybeans, common beans and peanuts. 6 , 9 , 10 However, the role of this group of microorganisms in natural ecosystems is poorly understood. 11 , 12 Little is known about the contribution of the increased diversity of nitrogen-fixing bacteria to the productivity and diversity of plants living in natural communities. Melloni et al. 13 reported that a greater diversity of bacteria in the soil results in greater resilience of the system and that a higher diversity of legume nodulating bacteria can favor symbiosis with various leguminous plant species and maximize the biological fixation of nitrogen in degraded areas. Previous research by van der Heijden et al. 11 demonstrated that symbiotic nitrogen-fixing bacteria promote evenness, productivity and nitrogen capture in systems that are rich in leguminous species, which suggests that the presence of nodulating bacteria is necessary for different species of leguminous and non-leguminous plants to coexist. Although mining activities generally alter a proportionally smaller area than other human activities, such as farming and planting pastures for livestock, the level of environmental degradation is very high because of the intense disturbance of the soil. This makes it necessary to take measures to restore these degraded areas at the end of the mining operations. In Brazil, to promote the rapid revegetation of highly degraded mined areas, the planting of leguminous species inoculated with nitrogen-fixing bacteria and arbuscular mycorrhizal fungi has been successfully employed. 14 The planting of leguminous species with selected isolates of these microorganisms enables the initial colonization of substrates that have been subjected to high chemical, physical and biological degradation. 15 The colonization with legumes leads to the deposition of litter and increases the concentrations of nutrients in the soil surface, enabling the replanted sites to enter the initial stages of plant succession. 15 Within this context and in this study, we assessed the diversity of these bacteria in areas that were revegetated after bauxite mining to better understand their role in degraded ecosystems under the recovery process. The areas studied were revegetated between 1981 and 2006 on soil consisting of overburden or tailings and used mixes of native species and inoculated leguminous species.",
"discussion": "Discussion Several studies have demonstrated that species belonging to the Mimosa genus have a high specificity for rhizobia isolates belonging to the beta proteobacteria subdivision, notably, isolates of the Burkholderia genus. 28 , 29 In addition, some species of Mimosoideae must associate with arbuscular mycorrhizal fungi for nodulation to occur. 30 The absence or low density of beta-rhizobia in the sampled areas and the absence of arbuscular mycorrhizal fungi in the Leonard jars used in this study can be possible explanations for the lack of nodulation in M. acutistipula . However, the majority of siratro plants showed nodulation, which demonstrates that the density of rhizobia present in the inoculum was sufficient to induce nodulation. This species, because of its wide host range, has been previously used to assess the diversity of isolates from nodules. 31 , 32 ARDRA was initially proposed as a useful tool for the rapid identification of the taxonomic position of rhizobia isolates because it is based on the restriction of a gene that codes for ribosomal RNA, which permits separation at the genus level and at the species level in some cases. It is mainly employed for this purpose in situations where the 16S gene cannot be sequenced. 8 , 20 , 33 , 34 In the present study, the majority of the isolates presented sequences close to those of the Bradyrhizobium genus (groups 10–22, Fig. 1 ). A greater abundance of Bradyrhizobium isolates have been observed by other authors under a variety of conditions, 5 , 32 , 35 which suggests that there is high variability within this genus in the areas studied. In addition to the isolates proximately related to the Bradyrhizobium genus, we also observed isolates proximately related to the Rhizobium and Mesorhizobium genera. Previous results showed the capacity of isolates belonging to these genera to withstand adverse conditions of pH, temperature and the presence of toxic elements. 36 , 37 , 38 The presence of these genera in the areas studied confirms their capacity to survive and establish symbiosis fixing-nitrogen under adverse environmental conditions. The BOX-PCR analysis allows the simultaneous evaluation of distinct genomic regions to identify intraspecific variability. This feature was demonstrated in this study because the isolates were distributed into 25 clusters by the ARDRA and 40 clusters by the BOX-PCR analysis. Using ARDRA, even based on data from three restriction enzymes, 56 isolates positioned in different genetic clusters showed 100% similarity, whereas this number was only 12 for the BOX-PCR technique. This is because the 16S rDNA gene is relatively small and presents highly conserved regions. Additionally, this gene is poorly discriminated within the genus Bradyrhizobium . 39 These findings show that the BOX-PCR technique was more discriminating between isolates, demonstrating the applicability of this tool to separate isolates that are taxonomically proximate 40 , 41 , 42 , 43 and its utility in studies aiming to compare the level of diversity between different sites. Among the areas that were revegetated on overburden, the OTU richness, Shannon and evenness indices were higher in the area that was revegetated the earliest (1981). The area revegetated more recently (2006) showed a low number of isolates, which impaired the diversity estimate because it was not possible to perceive stabilization of the Shannon index ( Fig. 4 ) when presented with the lowest OTU richness, Shannon and evenness indices value. The other areas presented intermediate values ( Fig. 3 ). The forest area showed OTU richness, Shannon and evenness indices lower than all of the overburden areas, and the two tailings waste areas that were revegetated using Acacia species presented similar indices to the overburden area revegetated in 1981 and higher indices than in the overburden area that was revegetated in the same year (1993) using a mixture of plant species. A lower diversity of nodule isolates in forest areas compared cultivated areas or fallows has been previously reported in several studies. 13 , 30 , 32 Jesus et al. 31 evaluated the effect of the type of land use on the diversity of nodule isolates captured by siratro and observed that the area cultivated with cassava presented a higher diversity than the forest areas and those cultivated with peach palm, which did not differ between one another. Lima et al. 32 also investigated the community of nodule isolates from siratro in areas with different uses in the Amazon region and observed that the richness index of the nodule isolates in the primary forest area was 12, which was the lowest value among the areas analyzed. The values in the other areas were 46 for the cultivated area, 48 in the agroforestry area, 24 for the area identified as old secondary forest, 28 in new secondary forest and 29 in the pasture. These results show that the revegetation strategies used in these areas enabled the establishment of a plant community that was able to sustain an increasing diversity of root nodule isolates. According to Melloni et al., 13 studies of the diversity of key groups of microorganisms in reclaimed areas are important because they supply an indication of the effects of different rehabilitation methods on the diversity of these microorganisms. The tailings pond areas were replanted with Acacia mangium in 1993. Since then, the changes caused by the coverages with leguminous species have led to the deposition of plant material on the soil, which provided conditions for the establishment of a diverse community of root nodule isolates. A higher diversity of legume-nodulating and nitrogen-fixing bacteria was also found in areas that were replanted with the leguminous species Mimosa scabrella and Cajanus cajan after mining activities. 13 The capacity of leguminous species to colonize degraded environments has been reported by other authors 14 , 15 and has been linked to the ability of these species form symbiotic associations with nitrogen-fixing bacteria in the soil, as well as with arbuscular mycorrhizal fungi. This three-way interaction favors the nutrition of plants by enhancing their uptake of nutrients. As a result of the increase of nitrogen in the biomass and the absorption of important nutrients, such as phosphorous, leguminous plants favor plant succession and the diversity of microorganisms."
} | 2,719 |
29955139 | PMC6155033 | pmc | 9,189 | {
"abstract": "The fate of carbon sequestered in permafrost is a key concern for future global warming as this large carbon stock is rapidly becoming a net methane source due to widespread thaw. Methane release from permafrost is moderated by methanotrophs, which oxidise 20–60% of this methane before emission to the atmosphere. Despite the importance of methanotrophs to carbon cycling, these microorganisms are under-characterised and have not been studied across a natural permafrost thaw gradient. Here, we examine methanotroph communities from the active layer of a permafrost thaw gradient in Stordalen Mire (Abisko, Sweden) spanning three years, analysing 188 metagenomes and 24 metatranscriptomes paired with in situ biogeochemical data. Methanotroph community composition and activity varied significantly as thaw progressed from intact permafrost palsa, to partially thawed bog and fully thawed fen. Thirteen methanotroph population genomes were recovered, including two novel genomes belonging to the uncultivated upland soil cluster alpha (USCα) group and a novel potentially methanotrophic Hyphomicrobiaceae . Combined analysis of porewater δ 13 C-CH 4 isotopes and methanotroph abundances showed methane oxidation was greatest below the oxic–anoxic interface in the bog. These results detail the direct effect of thaw on autochthonous methanotroph communities, and their consequent changes in population structure, activity and methane moderation potential.",
"conclusion": "Conclusion The methanotrophs found within the Stordalen Mire thaw gradient include canonical and novel populations that encode diverse metabolisms. The low CH 4 environment of the palsa is populated by the likely facultative and high affinity atmospheric methane-oxidisers USCα. The Methylocystaceae (MC1 and MC2) in the bog can cope with fluctuating CH 4 conditions due to the pMMO isozymes (pMMO and pMMO2) and capability for acetate uptake. In the high CH 4 fen, the obligate Methylococcaceae (MB1 and MB2) are dominant and active, while the metabolically diverse HYP1 are in low abundance and appear relatively inactive. It is evident that an evolutionarily complex, diverse and shifting methanotroph community is at the forefront of climate change as permafrost thaws. The low abundance of USC1 and HYP1 precluded their enrichment and visualisation, however it is hoped that the metabolic inferences determined from these genomes will guide future efforts to target and eventually isolate these elusive microorganisms, and experimentally confirm the extent of their ecological impact. Data availability Data used in this manuscript are submitted under NCBI BioProject accession number PRJNA386568.",
"discussion": "Results and discussion Identification and distribution of canonical and novel methanotrophs across the thaw gradient To investigate methanotroph diversity and the succession of the community as thaw progresses from palsa, to bog and fen, 188 raw metagenomes were searched for pMMO ( pmoA gene) and sMMO ( mmoX gene) reads using GraftM [ 42 ]. GraftM classified the reads into taxonomic groups following placement in a curated pmoA or mmoX gene tree (Fig. 1 ). The bog had the highest abundance of methanotrophs and was dominated by the well-characterised Methylocystaceae . Several members of the Methylocystaceae are suited to soil and bog environments exhibiting low pH and nitrogen, oligotrophy, or variable methane levels where the ability to oxidise atmospheric methane using a high affinity isozyme of the pMMO (pMMO2) is advantageous [ 52 , 53 ]. In the minerotrophic and pH neutral fen, members of the Methylococcaceae comprised the largest fraction of the low abundance but taxonomically diverse methanotroph community. Methylocystaceae and Methylococcaceae methanotrophs occasionally possess pMMO, with the canonical operon structure of pmoCAB , in addition to a divergent monooxygenase of undetermined function, pXMO, encoded by a reordered pxmABC operon [ 54 ]. The pxmA gene was detected in the palsa, fen and bog, indicating the presence of multiple methanotrophic lineages encoding the divergent pXMO. Beijerinckiaceae possessing mmoX appeared in low abundance in both fen and bog (Fig. 1 ). Fig. 1 Methanotroph diversity across the Stordalen Mire thaw gradient. a Heatmap of the relative abundance of methanotrophs as a proportion of the total metagenome based on the particulate methane monooxygenase (pMMO; left) and the soluble methane monooxygenase (sMMO; right). Abundances are indicated by the coloured scale (from white, to blue, to red). The trees used by GraftM to classify the reads are shown for pmoA ( b ) and mmoX ( c ). The colour of the clades indicates the environment where these clades are most often found (green = bog, blue = fen). Asterisks indicate significantly different abundances based on one-way ANOVA tests and Bonferroni corrected pairwise t -tests ( p value < 0.05; see Supplementary Table 6 ) Of the uncharacterised methanotroph clades, USCα had greatest abundance in the bog and was the predominant methanotroph in the palsa. Minimal methane is produced in the oxic palsa environment [ 35 ], so methane is likely taken up from the atmosphere. Reads from two additional novel groups, ‘novel pmo’ ( pmoA ) and ‘novel smo’ ( mmoX ), were found only in the fen (Fig. 1 ). The ‘novel pmo’ reads clustered alongside a collection of sequences recently recovered from a Sacramento-San Joaquin Delta wetland metagenome [ 43 ]. The ‘novel smo’ group has not previously been recognised. Interestingly, methanotrophs were often present below the water table in the deep (bog: >5 cm below water table, fen: >10 cm below peat surface) and extra deep (>30 cm below peat surface) metagenomes (Fig. 1 ), suggesting adaptations to cope with low oxygen conditions. The palsa remains mainly oxic to the depths sampled, whereas the bog includes an oxic surface and then likely becomes primarily anoxic below the water table (maximum depth 20 cm) [ 55 ]. Although the fen water table is above the peat surface, sedge roots are known to transport oxygen, which likely supports methanotrophs near live roots [ 56 , 57 ]. Recovery of methanotroph population genomes To recover the genomes of canonical and novel methanotrophs, assemblies of the 214 Stordalen Mire metagenomes (188 active layer, and 26 palsa core series) were binned into population genomes using differential coverage binning [ 37 ]. This led to the recovery of 1529 medium to high-quality population genomes (>70% completeness, and <10% contamination [ 37 ]). Of these genomes, 12 were found to encode pMMO and/or sMMO. An additional genome was recovered from a co-assembly of the palsa samples ( n = 78). Phylogenetic analysis of the genomes using a genome tree created from the concatenated alignment of 120 single copy marker genes classified all 13 as proteobacterial. Three of these genomes belonged to the Methylococcaceae (Methylobacter-like MB1-2; Methyloglobulus-like MG1), seven to the Methylocystaceae (Methylocystis-like MC), two to the USCα (USC1 and USC2) and surprisingly one to the Hyphomicrobiaceae (HYP1) (Fig. 2 ). Six of the Methylocystaceae genomes were recovered in the mid or deep core layers of different bog samples. These genomes had ANI of >99% indicating that they represent the same Methylocystis-like population and will be referred to as MC1 accordingly (Supplementary Fig. 1 ). One Methylocystis-like genome (MC2) was recovered from a palsa sample, and represented a distinct population at 94% ANI to MC1. Fig. 2 Genome tree of the methanotroph population genomes recovered from Stordalen Mire. Solid circles represent bootstrap support of over 70%. Heatmap bars indicate average relative abundance per environment (P = palsa (brown), B = bog (green), F = fen (blue)) as a percentage of the total metagenome library (Supplementary Information). Raw values and standard deviation are presented in Supplementary Table 8 . The highest abundances are observed for MC1 (1.85% in a deep bog sample) and USC1 (1.6% in a mid-depth bog sample) USC1 and USC2 are the first genomic representatives of the USCα lineage [ 25 ], as former representation of this group was restricted to partial pmoA sequences and a collection of short (<43 kb) genome fragments [ 22 , 26 ]. The previous hypothesis that these microorganisms fall within the Beijerinckiaceae was confirmed by placement of USC1 and USC2 in a genome tree [ 26 ] (Fig. 2 ). Average amino acid identity (AAI) of USC1 and USC2 was 72% to their closest isolated taxonomic neighbours, the Methylocapsa spp. [ 26 ] (Supplementary Table 1 ), making it likely these populations belong to a novel species or potentially novel genus within the Beijerinckiaceae [ 58 ]. The USCα PmoCAB protein sequences derived from the genomes (USC1 and USC2), and additional sequences from unbinned contigs, clustered with both partial and full length translated USCα pmoA , pmoB and pmoC gene sequences recovered from the active layer of mineral cryosols in the Canadian high Arctic [ 22 ] and an acidic forest soil [ 26 ] (Fig. 3 ; Supplementary Fig. 2 ). Fig. 3 Phylogenetic tree of PmoA proteins recovered from Stordalen Mire. This tree is constructed from isolate-derived protein sequences (colour strip = black), Stordalen sequences (colour strip: palsa = brown; bog = green; fen = blue) and translated environmental sequences compiled by Knief [ 18 ] (colour strip = grey), with additional sequences from He et al. [ 43 ], Ricke et al. [ 26 ] and Lau et al. [ 22 ] (colour strip = grey). The asterisks indicate sequences that are within population genomes An additional potential methanotroph genome (HYP1) was recovered from a fen sample, and was distinct from all canonical methanotroph families. Phylogenetic analysis revealed Rhodomicrobium spp. within the Hyphomicrobiaceae as the closest isolated taxonomic neighbours with an AAI of 67.5% (to both R. vannielii and R. udaipurense ), indicating likely taxonomic distinction at the genus level (Fig. 2 ; Supplementary Table 1 ) [ 58 ]. Interestingly, HYP1 encodes novel methanotrophy operons for both pMMO and sMMO. Contamination from poor binning was discounted as these genes were also found in an additional medium-quality Hyphomicrobiaceae genome (HYP2; 60.53% completeness, 5.54% contamination) recovered from a deep fen sample with 80% AAI to HYP1 (Supplementary Figs. 3 and 4 ). Protein trees showed that HYP1 PmoA and PmoB sequences, and several additional PmoAB sequences from partial genomes and unbinned contigs, clustered closely with the wetland ‘novel pmo’ clade that appears basal to the Methylocystaceae and Beijerinckiaceae (Fig. 3 ; Supplementary Fig. 2a ). Full-length PmoC sequences of this ‘novel pmo’ group have not been recovered until now, and were found to be similarly basal to the Methylocystaceae and Beijerinckiaceae PmoC proteins (Supplementary Fig. 2b ). The MmoX from HYP1 and additional unbinned MmoX sequences clustered within the ‘novel smo’ group that formed a monophyletic clade with the Beijerinckiaceae (Supplementary Figs. 5 and 6 ). Recovery of methanotrophy marker genes in HYP1 suggests that the Hyphomicrobiaceae should be added to the recognised alphaproteobacterial methanotrophic families which were previously limited to the Beijerinckiaceae and Methylocystaceae . To investigate other possible habitats of HYP1, publicly available SRA (Sequence Read Archive) datasets (24,636 sequence files) and the NCBI nr database were searched for pmoA sequences falling within the HYP1 clade. Only one sequence from NCBI nr was identified as belonging to HYP1, and was from a Finnish peat soil (NCBI accession: CAC84777). However, HYP1 was detected (at >1 read hits) in 34 SRA metagenomes from a variety of habitats, including Sacramento-San Joaquin Delta wetland, New York City MTA subway, peat, bog and sand metagenomes (Supplementary Table 2 ), suggesting that HYP1 is a rare but widely distributed group. Metabolic characterisation of Stordalen Mire methanotrophs Metabolic reconstruction of the Stordalen Mire methanotroph genomes revealed diverse functional potential. Methane oxidation pathways, carbon assimilation, dissimilation and other potentially habitat-specific metabolisms were examined in the gammaproteobacterial Methylobacter-like (MB) populations MB1, MB2 and Methyloglobulus-like (MG) population MG1 recovered from the fen samples. MB1 and MB2 encode pMMO ( pmoCAB ), however MB2 also has sMMO ( mmoXYZBCD ). MG1 has only pxmA and B that appear consecutively at the end of a contig, indicating that pxmC and the pmoCAB operon are likely in the unrecovered portion (21%) of the genome. MG1 also encodes a hydroxylamine reductase (HAO) allowing for the conversion of hydroxylamine to nitrite, which is likely a toxicity response measure for this population [ 59 ]. Methane monooxygenases can co-oxidise ammonia to hydroxylamine, an intermediate toxic to cells unless further processed by an enzyme such as HAO [ 60 ]. MB1, MB2 and MG1 all possess the ribulose monophosphate (RuMP) pathway for carbon assimilation, the tetrahydromethanopterin (H 4 MPT) carbon dissimilation pathway for the oxidation of formaldehyde to formate and nitrogen fixation genes ( nifHDK ). MB1 also has dissimilatory nitrate reduction, which has been identified in the gammaproteobacterial methanotrophs Methylomicrobium , Methylomonas and Methylobacter tundripaludum [ 61 ]. This ability appears to be widespread within the Methylococcaceae , and has been linked to survival in low oxygen conditions where nitrate or nitrite can be used instead of oxygen as terminal electron acceptors in order to direct all available oxygen to methane oxidation [ 61 ]. Metabolic reconstruction of MB1-2 and MG and the observed distribution of Methylococcaceae in the metagenomes suggest that these microorganisms can function in the microaerobic environment of the deeper permafrost thaw layers (Fig. 1 ). The alphaproteobacterial Stordalen methanotrophs had larger genomes than their gammaproteobacterial counterparts, and had an expanded metabolic diversity indicative of a facultative, rather than obligate, methanotrophic lifestyle (Supplementary Table 4 ). Metabolic analyses were conducted on the most complete representatives of the alphaproteobacterial lineages, USC1, HYP1 and MC1 (Fig. 4 ). Similar to USC1 and HYP1, MC1 encodes pMMO genes ( pmoCAB ) but further possesses a pMMO2 ( pmoCAB2 ) variant that is associated with high affinity methane oxidation [ 52 ]. MC1 also encodes the pXMO ( pxmABC ) homologue, as does USC1. This discovery adds Beijerinckiaceae to the list of pXMO-encoding methanotroph families, and explains the high relative abundance of pxmA genes observed in the palsa environment (Fig. 1 ). Fig. 4 Metabolic reconstruction of the alphaproteobacterial population genomes MC1, HYP1 and USC1. Colours indicate the genome or combination of genomes (Venn diagram) in which the cycle or enzymes are found. Abbreviations: H 4 F tetrahydrofolate pathway, H 4 MPT tetrahydromethanopterin pathway, TCA tricarboxylic acid cycle, EMC ethylmalonyl-CoA pathway, EMP Embden–Meyerhof–Parnas pathway (glycolysis), CBB Calvin–Benson–Bassham cycle, PHB polyhydroxybutyrate pathway, LPS lipopolysaccharide, CODH carbon monoxide dehydrogenase, NiFe nickel iron hydrogenase, nitrogenase (NifHDK), pMMO particulate methane monooxygenase, pMMO2 particulate methane monooxygenase isozyme II, sMMO soluble methane monooxygenase, pXMO homologue of particulate methane monooxgyenase, CH 3 OH methanol, I complex I NADH dehydrogenase, II complex II succinate dehydrogenase, III complex III cytochrome bc1, IV cytochrome c oxidase, IV cbb3 complex IV cytochrome cbb3 oxidase, cyd complex IV cytochrome bd oxidase, FDH formate dehydrogenase, EHR energy converting hydrogenase related (part of formate hydrogenlyase complex), SO 4 2− sulphate, MoO 4 2− molybdate, Zn zinc, PO 4 3− phosphate, CHOH formaldehyde, sulphate adenylyltransferase (SatAB), APS adenosine 5'-phosphosulphate, SO 3 2− sulphite, adenylylsulphate reductase (AprAB), dissimilatory sulphite reductase (DsrAB), H 2 S hydrogen sulphide, dissimilatory sulphite reductase electron transport complex (DsrK representing the DsrKMJOP complex), SoxYZ/SoxAX/SoxB/SoxCD = sulphur oxidising proteins. See Supplementary Fig. 7 for detailed gene presence/absence and Supplementary Table 3 for the list of additional abbreviations Genes for the calcium-dependent methanol dehydrogenase (MDH) mxaF -encoded large subunit of the mxaFI-MDH complex, which catalyses the second step of methane oxidation, could not be identified in any of the Stordalen Mire methanotroph genomes. Instead, at least one copy of xoxF , a homologue of mxaF [ 62 ], was identified in each genome except MG1. This gene differentiates into five clades ( xoxF 1-5)[ 62 , 63 ] and encodes the lanthanide-dependent xoxF–MDH complex known to oxidise methanol in several methanotrophs [ 62 , 64 ]. The H 4 MPT carbon dissimilation and tetrahydrofolate (H 4 F) carbon assimilation pathway, which facilitates entry into the serine cycle [ 63 ], are present in USC1 and MC1. Alternatively, HYP1 appears to have a thiol-dependent (glutathione (GSH)-linked) pathway for formaldehyde oxidation to formate using glutathione synthase ( gfa ), S-(hydroxymethyl)glutathione dehydrogenase ( frmA ) and S-formylglutathione hydrolase ( frmB ) genes [ 63 , 65 ]. This pathway is present as a detoxification and energy generation mechanism in the methylotroph Paracoccus denitrificans [ 63 , 66 ]. Similar to P. denitrificans and several other alphaproteobacterial methylotrophs [ 62 , 67 ], in HYP1 these genes are located directly adjacent to a PQQ-dependent alcohol dehydrogenase identified as the xoxF 5 type (Supplementary Fig. 8 ; Supplementary Fig. 9 ) [ 62 ]. This suggests that formaldehyde is produced by the xoxF-MDH of HYP1 and then oxidised to formate using the thiol-dependent pathway. The ethylmalonyl-CoA (EMC) pathway is used to regenerate glyoxylate from acetyl-CoA for use in the serine cycle [ 68 ], and was present in MC1 and HYP1 but not USC1 (Fig. 4 ). USC1, similar to the Beijerinckiaceae methanotrophs [ 20 ], possesses isocitrate lyase ( icl ), which forms part of the glyoxylate shunt with malate synthase and allows the formation of glyoxylate from isocitrate [ 68 ]. The glyoxylate cycle enables the assimilation of acetate, and acetyl-CoA can be produced from acetate using acetate kinase ( ack ) and phosphotransacetylase ( pta ) or acetyl-CoA synthetase ( acs ), which are present in USC1. These genes, in conjunction with demonstrated acetate uptake in isotope studies [ 69 ], implicate USCα populations as likely facultative methanotrophs. Additionally, MC1 and HYP1 possess genes involved in transforming acetate to acetyl-CoA ( ack , pta , acs ), suggesting these populations are also facultative. A facultative methanotrophic lifestyle likely allows USCα and MC1 to survive on acetate, which can diffuse freely across the cell membrane in the acidic conditions of the bog, under methane limited conditions [ 70 ]. Although USC1, HYP1 and MC1 contain a near complete EMP pathway, the lack of glucose transporters suggests these populations do not use glucose as a carbon and energy source. Transporters are missing in known methanotrophs and growth of methanotrophs on glucose has not been recorded [ 20 ], indicating that this pathway is anabolic in the Stordalen Mire genomes. Poly-β-hydroxybutyrate (PHB) is a storage compound commonly used as a carbon and energy source by methanotrophs under nutrient limiting conditions [ 71 ]. Genes for the PHB storage pathway ( phaABCZ , bdh , acsA ) are present in all three genomes. HYP1 may have additional and unusual means of energy generation. A complete dissimilatory sulphate reduction pathway was identified, including the canonical markers of dissimilatory sulphite reductase subunits A and B ( dsrAB ) (Fig. 4 ). These reductases have high AAI to oxidative dsr genes of the purple non-sulphur photolithotrophic bacteria Rhodomicrobium vannielii (83% WP_013418283) and R. udaipurense (86% KAI93440) [ 72 , 73 ]. Like Rhodomicrobium spp., HYP1 possesses an incomplete sulphur oxidation system (SOX; soxAXYZD ), however sulphur oxidation could still be possible as Rhodomicrobium spp. can use sulphide and thiosulphate [ 74 ]. The potential for sulphur oxidation in a methanotroph was recently described in a Methylococcaceae genome from a deep-sea hydrothermal plume [ 75 ]. Biogeochemical measurements from the site show that the fen, particularly the fen surface, has significantly higher sulphate concentrations than the bog (Supplementary Table 5 ; fen surface average = 10.44 µM, average fen = 6.25 µM, bog = 1.83 µM; p value < 10 −4 ), potentially indicating increased sulphur oxidation activity. In the fen, it may be possible for HYP1 to oxidise sulphate that has been reduced during anaerobiosis in the deeper layers [ 76 ]. In addition to sulphur, Rhodomicrobium spp. can use hydrogen as an electron donor [ 74 ]. HYP1, USC1 and MC1 encode numerous hydrogenases, indicating potential hydrogen use (Fig. 4 ). Hydrogen uptake and production has been recorded in several verrucomicrobial, alpha- and gammaproteobacterial methanotrophs, and has been linked to nitrogen fixation, which produces the required hydrogen as a by-product [ 77 , 78 ]. Both HYP1 and USC1 also encode coxLMS for carbon monoxide (CO) oxidation. The ability of these microorganisms to perform CO oxidation could allow CO to be used as both an energy and carbon source [ 79 ]. Stable-isotope probing or successful culturing would be required to confirm this metabolism. However, atmospheric CO levels, or CO produced by the photochemical degradation of matter, are sufficient for carboxydotrophs in many soil environments [ 79 ]. Several hydrogenotrophic [ 80 ], methanotrophic [ 81 ] and carbon monoxide oxidising [ 79 ] bacteria are known to use these respective pathways to generate energy for carbon fixation using the Calvin–Benson–Bassham (CBB) cycle. Ribulose bisphosphate carboxylase (RuBisCO) is the key enzyme and genetic marker for the CBB cycle. HYP1 encodes RuBisCO form IV, a RuBisCO-like protein of unknown function [ 82 ], and form II ( cbbM ). Form II is known to operate under high CO 2 and low O 2 concentrations [ 82 , 83 ], which would suit fixation in the deeper fen layers (Supplementary Fig. 10 ). Of the other genomes, MC1 possesses only form IV, whereas USC1, similar to the characterised methanotrophic Beijerinckiaceae [ 84 ], encodes the catalytic form I (both cbbL and S ; Supplementary Fig. 11 ). HYP1, MC1 and USC1 are also predicted to be capable of nitrogen fixation, as phylogenetic analysis of the NifH proteins suggests they belong to the functional Type 1D clade (Supplementary Fig. 12 ). Given the wide distribution of nitrogen fixation genes in the Stordalen Mire methanotrophs, it appears that this metabolism may be a selective advantage for methanotrophs in the nitrogen limited thaw environment. Metatranscriptomic analysis of in situ activity Gene presence in metagenomes reveals the metabolic potential of the community but does not indicate which genes are active at the time of sampling. In order to investigate the activity of methanotrophs in the system, expression of key genes for methane oxidation ( pmoA and mmoX ) were examined in 24 metatranscriptomes spanning different sites and depths. Despite high abundance in the bog metagenomes, low relative transcript expression was observed for the Methylocystaceae (Fig. 5 ). Low activity of Methylocystaceae has been recorded in other environments [ 85 ], and is likely a consequence of harsher environmental conditions in the low pH and ombrotrophic bog that would favour stress-tolerance over productivity [ 23 , 86 ]. In the fen, the diversity of methanotrophs was not reflected in the metatranscriptomes, which were dominated by Methylococcaceae even in the deeper layers (Fig. 5 ). This unexpected prevalence of Methylococcaceae activity at depth has been observed in Arctic soils, which further supports hypotheses that these microorganisms are active under microaerophilic conditions [ 61 , 87 ]. Methylococcaceae pmoA transcripts in the fen also comprised a greater proportion of the total metatranscriptome reads than in the bog or palsa, indicating that methanotrophs make up more of the active community in the fully thawed site. This suggests that Methylococcaceae are oxidising a substantial amount of CH 4 and/or responding to increased CH 4 availability in the fen, which has the highest CH 4 flux of all three sites [ 35 ]. Fig. 5 Methanotroph abundance and activity in the 24 samples with paired metagenomes and metatranscriptomes from Stordalen Mire. For spatial orientation, distance from the water table and peat surface is shown in a . The metagenome abundances are indicated in b and the transcript expression in c . Methanotroph pmoA and mmoX read abundances are presented as a percentage of total reads normalised by HMM length for both metagenomes and metatranscriptomes Consistent with previous findings from isolates [ 68 ] and similar environments [ 23 , 43 , 85 ], very few pxmA or mmoX transcripts were detected in the Stordalen Mire metatranscriptomes. The ‘novel pmo’ and ‘novel smo’ of the HYP1 group had minimal or no transcript expression (12 ‘novel pmo’ reads detected in one surface fen sample, and 25 ‘novel smo’ reads in one mid fen sample). USCα had very low pmoA expression in the palsa core mid depth sample only (112 reads detected), which fits with low expression observed in samples of an Arctic cryosol environment [ 22 ]. Transcript expression of the second step of methanotrophy, methanol oxidation to formaldehyde, was high for xoxF , with virtually no representation in the metatranscriptome reads of mxaF (Supplementary Fig. 13 ). This strongly suggests that the xoxF form of MDH is the most important complex for methanol oxidation at Stordalen Mire, which supports recent findings that this form is likely dominant in lanthanide non-limiting environments [ 64 ]. Further, research into methylotroph and methanotroph co-cultures has revealed that high mxaF expression over xoxF can be linked to cross-feeding and syntrophy between these microorganisms [ 88 ]. The dominance of xoxF–MDH expression at Stordalen suggests that lanthanide is non-limiting in this environment, and potentially that methanol is not being secreted for use by syntrophic partners. Mapping metatranscriptomes to the population genomes enabled transcription analysis of entire metabolic pathways for specific lineages, and revealed the functionality of methanotrophs beyond the expression of marker genes for primary methane oxidation. The highest relative transcript expression was 0.27% for the MC1 population genome in a bog mid-depth sample, and 0.24% for MC2 in a palsa deep (calculated as proportion of total transcript reads in the metatranscriptome; Supplementary Table 7 ). MC1 and MC2 had high transcript expression for a variety of genes in the bog, and some expression in the fen and one palsa sample, confirming active carbon assimilation (Supplementary Fig. 14 ). Activity through transcript expression was also observed for MB1 and MB2 in the fen, which had high transcript coverage of pmoCAB genes and downstream methane processing genes (Supplementary Fig. 14 ). xoxF was consistently expressed in MC1-2 and MB1-2, indicating use of the xoxF–MDH for methanol oxidation in these populations. Surprisingly, MC1-2 and MB1-2 appeared to be actively fixing nitrogen through high expression of nifH , reaffirming the importance of this function in environments with low nitrogen availability [ 23 ]. The novel metabolic inferences could not be confirmed for MG1, HYP1 and USC1 due to limited transcript expression across all metatranscriptome samples. It is likely that these populations were inactive, or below detection, at the time of sampling. Relationship between methanotrophs and biogeochemistry Previous work at the site revealed that average CH 4 flux increases as thaw progresses, from minimal emission in the palsa, to ~1.46 mg CH 4 m 2 /h and ~8.75 mg CH 4 m 2 /h in the bog and fen, respectively [ 35 ]. Here, methanotroph metagenome abundances, determined using the pmoA and mmoX genes, were analysed alongside depth resolved biogeochemical data from the sites to evaluate relationships between community and methane oxidation. At the fen and bog sites the concentration and δ 13 C signature of dissolved CH 4 was analysed in porewater samples collected in parallel with the peat samples. The δ 13 C signature of CH 4 is influenced by the combined effect of the methanogenic pathway (acetoclastic versus hydrogenotrophic) that produced the CH 4 , and the activity of the methanotrophs [ 35 , 89 ]). The highest abundance of methanotrophs occurred in the region of the peat profiles where there was maximum dissolved CH 4 , but still periodic oxygen availability either due to varying water tables in the bog [ 57 ] or root transport in the fen [ 56 ] (bog = 15–20 cm, fen = 10–15 cm; Fig. 6 ). Fig. 6 Depth profiles of palsa, bog and fen methanotroph community abundances (as pmoA and mmoX gene reads normalised by total library and HMM length) and relationship to water table, dissolved CH 4 concentration and δ 13 C signature (no porewater present at the palsa site). Metagenome and porewater chemistry data from 2011 and 2012 was averaged across all dates by depth category; error bars represent ± 1 standard error. This shows a link between methanotroph abundances and dissolved CH 4 concentration across the thaw gradient. At the bog site, δ 13 C-CH 4 patterns track the depth distribution of the methanotroph community, with the heaviest (most oxidised) CH 4 occurring just above the maximum water table depth where methanogen populations are highest and the lightest (least oxidised) CH 4 occurring in permanently inundated peat where methanotroph abundances are low Since hydrogenotrophic methanogens were dominant in the bog samples [ 35 ], most of the variation in δ 13 C-CH 4 at this site was likely due to variation in methane oxidation (not variation in production pathway), with less negative δ 13 C-CH 4 values indicating the preferential use of lighter 12 CH 4 by methanotrophs, and consequently greater CH 4 oxidation. While methanotroph transcript expression in the bog revealed no depth associated trends (Fig. 5 ), comparison of methanotroph abundances and δ 13 C-CH 4 suggest that methanotrophy increased with depth across the region of peat that is periodically above the water table and decreased in the permanently inundated peat (Fig. 6 ). The heaviest (most oxidised) CH 4 and the highest methanotroph abundances occurred in peat that is inundated >90% of the time (15–20 cm) (Fig. 6 ). This was surprising, as CH 4 oxidation was expected to be greatest nearer the theoretical optimal oxygen and CH 4 conditions of the oxic–anoxic boundary at the average water table depth (6 cm) [ 85 ]. Instead, it appears that CH 4 concentration, which increases with depth, is the key driver of methanotroph community patterns and that a highly variable water table in the bog allows infrequent oxic events that provide sufficient oxygen to support specialised methanotroph populations. In the fen, methanotroph abundances correlated with distance from the water table (Supplementary Fig. 15 ), however low variability in δ 13 C-CH 4 indicated no clear isotopic evidence for activity of methanotrophs even though the highest methanotroph abundances were found between 10 and 15 cm, and Methylococcaceae were active (Figs. 5 and 6 ). This disconnection of abundances, activity and δ 13 C-CH 4 may be related to the capacity of plant roots to provide substrates for methanogenesis within the peat, a conduit for oxygen into the peat and CH 4 out of the peat, which could consequently bypass oxidation and disrupt clear methane oxidation gradients [ 57 ]."
} | 8,069 |
26489859 | PMC4620461 | pmc | 9,190 | {
"abstract": "ABSTRACT A better understanding of how bacteria resist stresses encountered during the progression of plant-microbe symbioses will advance our ability to stimulate plant growth. Here, we show that the symbiotic system comprising the nitrogen-fixing bacterium Bradyrhizobium diazoefficiens and the legume Aeschynomene afraspera requires hopanoid production for optimal fitness . While methylated (2Me) hopanoids contribute to growth under plant-cell-like microaerobic and acidic conditions in the free-living state, they are dispensable during symbiosis. In contrast, synthesis of extended (C 35 ) hopanoids is required for growth microaerobically and under various stress conditions (high temperature, low pH, high osmolarity, bile salts, oxidative stress, and antimicrobial peptides) in the free-living state and also during symbiosis. These defects might be due to a less rigid membrane resulting from the absence of free or lipidA-bound C 35 hopanoids or the accumulation of the C 30 hopanoid diploptene. Our results also show that C 35 hopanoids are necessary for symbiosis only with the host Aeschynomene afraspera but not with soybean. This difference is likely related to the presence of cysteine-rich antimicrobial peptides in Aeschynomene nodules that induce drastic modification in bacterial morphology and physiology. The study of hopanoid mutants in plant symbionts thus provides an opportunity to gain insight into host-microbe interactions during later stages of symbiotic progression, as well as the microenvironmental conditions for which hopanoids provide a fitness advantage.",
"introduction": "INTRODUCTION A variety of plants, including leguminous ( 1 ), actinorhizal ( 2 ), and land-dwelling ( 3 ) plants, rely on bacterial symbiotic partners for assimilation of nitrogen, an essential macronutrient. These symbioses require bacterial invasion of plant tissues and adaptation of the bacterial symbiont to the plant host environment, processes in which several microbial membrane lipids play key roles. For instance, phosphatidylcholine is critical for efficient nitrogen fixation in several legume-rhizobial partnerships such as soybean- Bradyrhizobium diazoefficiens (formerly named Bradyrhizobium japonicum [ 4 ]) and alfalfa- Sinorhizobium meliloti ( 5 ). Similarly, an intact outer membrane (OM) lipopolysaccharide (LPS) is essential for all stages of rhizobial symbiosis, including root hair infection, symbiotic tissue (nodule) establishment, and survival within the plant cell ( 6 ). Whether and to what extent other microbial membrane lipids regulate the establishment and maintenance of plant-microbe symbioses are unclear. Here, we consider whether hopanoids ( 7 ), steroid-like pentacyclic triterpenoid lipids, support plant-microbe interactions. Hopanoids are the progenitors of hopanes, molecular fossils that exhibit intriguing yet poorly understood abundance patterns in the rock record ( 8 ). In part, our interest in hopanoids derives from a desire to interpret ancient biomarkers and the conviction that this requires a nuanced understanding of the biological functions of diverse hopanoids in modern niches. The capacity for hopanoid biosynthesis is statistically enriched in the (meta-)genomes of bacteria associated with plants ( 9 ), and hopanoids have been found in high abundance in membranes of well-studied plant symbionts of the Bradyrhizobium (40% of total lipid extract [TLE]) and Frankia (87%) genera ( 10 , 11 ). Hopanoids promote membrane rigidity ( 12 ) and confer protection against numerous stresses, including acidic or alkaline pH, high temperature, high osmolarity, oxidative stress, detergents, and antibiotics ( 13 – 16 ). They have varied structures, formed via methylation (2Me or 3Me), unsaturation, and/or attachment to a ribose-derived side chain (C 35 ) ( Fig. 1A ; also see Fig. S1 in the supplemental material) ( 7 ). Whether there are functional distinctions under specific environmental conditions for diverse hopanoid types is unclear, yet some evidence suggests that they have nonoverlapping roles. For example, in Rhodopseudomonas palustris ( 17 ) and Burkholderia cenocepacia ( 14 ), C 35 hopanoids are critical for OM stability and for resistance to low pH, detergent (sodium dodecyl sulfate [SDS]), and polymyxin B, respectively. Though the absence of 2Me-hopanoids did not manifest a stress phenotype in previous tests of R. palustris , their biosynthesis is transcriptionally induced under stress ( 13 ). This suggests that 2Me-hopanoids might contribute to stress resistance under conditions yet to be identified in this and other organisms; consistent with this notion, 3Me-hopanoids contribute to late-stationary-phase survival in Methylococcus capsulatus ( 18 ). In vitro , 2Me-hopanoids rigidify membranes of varied compositions ( 12 ). However, until now, no study has explored whether different hopanoids impact fitness in a natural ecological context. FIG 1 (A) Structures of hopanoid and tetrahymanol. B. diazoefficiens makes C 30 hopanoids, such as diploptene (C-22=C-30) and diplopterol (OH at C-22); C 35 hopanoids, such as bacteriohopanetetrol (BHT; R 2 =OH) and aminobacteriohopanetriol (aminotriol; R 2 =NH 2 ); and tetrahymanol. All these compounds can be methylated at C-2 (2Me, R 1 =CH 3 ). (B) Hopanoid biosynthetic gene cluster of B. diazoefficiens . In this study, we focused on the genes colored in gray: the shc (squalene hopene cyclase) product catalyzes squalene cyclization to hopene, the first reaction in the hopanoid biosynthetic pathway; the hpnH product catalyzes addition of adenosine to hopene, the first reaction in the synthesis of C 35 hopanoids; and the hpnP product catalyzes C-2 methylation. (C) GC-MS and LC-MS (inset) total ion chromatograms of total lipid extracts from aerobically grown B. diazoefficiens strains. For GC-MS, main hopanoid peaks are numbered and the methylated counterparts elute 0.2 to 0.5 min earlier. I, pregnane acetate (standard); II, (2Me) hop-17(21)-ene; III, (2Me) hop- x -ene; IV, (2Me) hop-22(29)-ene (diploptene); V, (2Me) hop-21-ene; VI, (2Me) hopan-22-ol (diplopterol); VII, (2Me and 20Me) tetrahymanol; and VIII, BHP-508. LC-MS: a, aminotriol; b, BHT; c, 2Me-aminotriol; d, adenosylhopane; e, 2Me-BHT. Lipid analysis for each strain was performed in triplicate. For chemical structures of hopanoids, refer to Fig. S1A in the supplemental material. Recently, it was shown that elimination of hopanoid biosynthesis in photosynthetic Bradyrhizobium strain BTAi1 impairs its symbiosis with the legume Aeschynomene evenia ( 15 ). Because the absence of all hopanoids likely had a broad and drastic impact on cellular physiology and hence symbiosis, in this study we tested specific effects of 2Me- and C 35 hopanoids ( 7 ) using B. diazoefficiens USDA110, the best-studied Bradyrhizobium strain. In addition to C 30 and C 35 hopanoids ( 10 ), B. diazoefficiens makes tetrahymanol, a triterpenoid with a gammacerane skeleton ( 19 ) ( Fig. 1A ; also see Fig. S1 in the supplemental material). Intriguingly, while most hopanoids are thought to occur free within membranes, the C 35 hopanoid, (2-Me) 34-carboxyl-bacteriohopane-32,33-diol, was found to be covalently attached to LPS lipidA, a well-established player in a broad range of host-microbe interactions, to form a compound called hopanoid-lipid A (HoLA) ( 15 , 20 ) (see Fig. S1B ). B. diazoefficiens exhibits two different lifestyles, free living in soil or symbiotic within legume root nodule cells ( 1 , 21 ). In addition to its native soybean host, B. diazoefficiens can engage in nitrogen-fixing symbioses with the stems and roots of the tropical legume Aeschynomene afraspera ( 22 ). In both of the these hosts, development of the symbiosis progresses through a series of defined stages: (i) colonization and invasion of host root tissue; (ii) internalization of bacteria by plant cells to form an organelle-like structure called the symbiosome, comprising endosymbiotic bacterial cells termed “bacteroids” that are surrounded by a plant-derived “peribacteroid” membrane (see Fig. S2 in the supplemental material); and (iii) initiation of nitrogen fixation by bacteroids, during which there is a high rate of nutrient exchange across the symbiosome membranes between plant-supplied carbon sources and fixed atmospheric nitrogen produced by bacterial nitrogenase ( 21 , 23 ). Supporting such high levels of bacterial nitrogenase activity requires both extensive host control of bacteroid physiology and establishment of a specific host microenvironment, defined by low oxygen, low pH, hyperosmosis, and oxidative stress ( 24 ). There is some evidence suggesting that the extent of these environmental stresses varies between plant hosts. For example, A. afraspera was recently shown to produce nodule-specific, cysteine-rich antimicrobial peptides (NCR peptides) that induce differentiation of the bacteroid into an enlarged, elongated, and polyploid state, whereas in the soybean host, NCR peptides are absent and bacteroid morphology and ploidy are similar to those of the free-living state ( 25 ). Thus, it is likely that the microbial adaptations required for survival within root nodules are host specific, and we hypothesized that specific hopanoid mutants may exhibit variable phenotypes in diverse plant hosts. Here, we focus on the phenotypic consequences of the inability of B. diazoefficiens to produce two specific hopanoid classes, 2Me- and C 35 hopanoids. We compare its hopanoid-dependent stress phenotypes in the free-living state to those of other hopanoid producers. We further explore the fitness effects of 2Me- or C 35 hopanoid production within a natural ecological context: the symbiotic microenvironment of soybean and A. afraspera cells. These studies begin to define the role of 2Me- and C 35 hopanoids during the progression of plant-microbe symbioses and provide insight into microbial membrane factors that facilitate adaptations to particular microenvironments.",
"discussion": "DISCUSSION Bacteria that provide plants with fixed nitrogen represent an attractive agronomical alternative to nitrogen fertilizers ( 36 ). Recently, we found a statistically significant correlation between the presence of hopanoid biosynthetic genes and organisms, metabolisms, and environments known to support plant-microbe interactions ( 9 ). This raised the hypothesis that hopanoids could support bacterial fitness in the context of symbiosis for certain organisms. Building on our recent observation that hopanoids promote symbiosis between the legume A. evenia and the Bradyrhizobium BTAi1 strain ( 15 ), we explored the generality of this finding by studying the phylogenetically different and better-known Bradyrhizobium species B. diazoefficiens , focusing on the roles of two dominant hopanoid classes. In the free-living state of B. diazoefficiens , 2Me-hopanoids contribute to growth under microaerobic and acidic conditions and C 35 hopanoids are required for microaerobic growth and tolerance to diverse stresses found in the symbiotic microenvironment. Consistent with these phenotypes, C 35 hopanoids are critical for symbiosis between B. diazoefficiens and A. afraspera , and yet they are dispensable for symbiosis with soybean. This intriguing finding suggests that the microenvironment encountered by plant symbionts varies between hosts. Bradyrhizobium strain BTAi1 and B. diazoefficiens differ physiologically in important ways. Shc proteinss from these two strains fall in distinct phylogenetic clades ( 15 ). Unlike Bradyrhizobium strain BTAi1, B. diazoefficiens is unable to photosynthesize ( 37 , 38 ). Moreover, B. diazoefficiens infects plants via a Nod factor-dependent pathway, whereas Bradyrhizobium strain BTAi1 uses alternate symbiotic strategies ( 39 ). Our inability to delete shc in B. diazoefficiens suggests that hopanoids are essential in this species, in contrast to Bradyrhizobium strain BTAi1, where shc mutants are viable. This fundamental difference between the species likely reflects differences in the niches that they inhabit as a consequence of their metabolic differences and what is required for survival therein. What insights can we gain about 2Me- and C 35 hopanoid functions in natural contexts based on ex planta experiments? This is the first study to identify conditions under which synthesis of 2Me-hopanoids is important for any cell type. These include hypoxic and acidic conditions that B. diazoefficiens possibly perceives as stress, thus upregulating 2Me-hopanoid production as has been seen in R. palustris ( 13 ) to enhance membrane rigidity and stability ( 12 ). However, it is puzzling that 2Me-hopanoids are dispensable under similar conditions within the plant cell, and a priority for future work will be to explain this paradox. Prior work in B. cenocepacia has shown that C 35 hopanoids promote stress tolerance and antibiotic resistance ( 14 ). Similarly, we found that synthesis of C 35 hopanoids is important for growth under oxic and hypoxic conditions and for tolerance to diverse stressors. Consistent with our whole-cell membrane rigidity measurements, these phenotypes could be due to higher membrane fluidity resulting from the absence of C 35 hopanoids, the inability to make HoLA, and/or the accumulation of the C 30 hopanoid diploptene. We hope to tease apart these possibilities going forward. Intriguingly, the abundance of 2Me-hopanes in ancient sediments peaks during oceanic anoxic events ( 8 ), and today, C 35 hopanoids appear to be enriched in hypoxic regions of the oceans ( 40 ). Why do we observe context-dependent hopanoid phenotypes in planta ? The phenotypic difference of the requirement for C 35 hopanoids in A. afraspera but not in soybean likely stems from differences between the plants’ intracellular environments. In both plants, the bacterium is exposed to a variety of stresses, including oxidative, osmotic, and acidic stresses, within the microaerobic niche of the infected plant cell ( 24 ). Although such an environment is less than ideal for the Δ hpnH mutant, it is able to colonize as evidenced by its successful symbiosis with the soybean plant. However, unlike soybean, in A. afraspera , B. diazoefficiens undergoes terminal differentiation due to the action of NCR peptides ( 25 ). Because the Δ hpnH mutant is highly sensitive to NCR peptides, exposure within the host might reduce the viable mutant population by causing cell death or increasing susceptibility to this and other plant defense mechanisms. Consistent with this hypothesis, the Δ hpnH mutant only partially infects the A. afraspera nodule tissue. In A. afraspera , synthesis of C 35 hopanoids is critical for several aspects of the symbiosis, including evasion of plant defense reactions, efficient utilization of plant photosynthates, and nitrogen fixation. Two reasons why the plant host mounts an immune response against the Δ hpnH mutant may be that the altered mutant surface layer, as seen in TEM images, is unable to suppress this response ( 41 ) and/or the host induces nodule senescence prematurely on detecting an underproductive symbiont ( 42 ). Consistent with this, nitrogenase activity is reduced in the Δ hpnH mutant relative to the WT, a likely consequence of poor cell viability. Similarly, the buildup of plant carbon as starch in Δ hpnH mutant nodules might indicate slow metabolism and/or perturbation of membrane transport processes that facilitate bacteroid carbon acquisition. The global agricultural economy is largely based on nitrogen fertilizers, with the United States alone consuming 13,000 tons per annum ( 43 ). However, the usage of nitrogen fertilizers comes with a price, as their production requires burning of fossil fuels and their runoff from soils leads to surface and groundwater contamination ( 44 ). Rhizobia, which naturally fix nitrogen in association with common crops, are an environmentally and economically feasible alternative to nitrogen fertilizers. Our results reveal that hopanoids affect the ability of B. diazoefficiens to cope with environmental stresses as well as its nitrogen fixation efficiency in a plant host-dependent manner. This observation is relevant to interpreting ancient patterns of hopane deposition, which correlate with paleoenvironmental conditions where nitrogen fixation may have provided a selective advantage ( 8 ). In addition, understanding the roles of hopanoids in bacterial stress resistance and how they facilitate nitrogen fixation may enable the engineering of agronomically useful strains with enhanced tolerance to rising temperature and salinity."
} | 4,186 |
29018238 | PMC5635011 | pmc | 9,192 | {
"abstract": "The emissivity of common materials remains constant with temperature variations, and cannot drastically change. However, it is possible to design its entire behaviour as a function of temperature, and to significantly alter the thermal emissivity of a surface through the combination of different patterns and materials. We show that smart patterned surfaces consisting of smaller structures (motifs) may be designed to respond uniquely through combinatorial strategies by transforming themselves. The smart surfaces can passively manipulate thermal radiation—without the use of electronics—because their modus operandi has already been programmed into their intrinsic characteristics; the environment provides the energy required for their activation. Each motif emits thermal radiation in a certain manner, as it changes its geometry; however, the spatial distribution of these motifs causes them to interact with each other. Therefore, their combination and interaction determine the global behaviour of the surfaces, thus enabling their a priori design. The emissivity behaviour is not random; it is determined by two fundamental parameters, namely the combination of orientations in which the motifs open (n-fold rotational symmetry) and the combination of materials (colours) on the motifs; these generate functions which fully determine the dependency of the emissivity on the temperature.",
"conclusion": "Conclusions By identifying and handling these fundamental properties—orientation and colour sequences—we designed the thermal emissivity function of a patterned surface. The classification of the generated curves and the similarities owing to the existence of invariant properties limit the number of combinations that need to be considered. We drastically altered the emissivity value (approximately Δε ≈ 0.47) within the temperature span of ΔΤ ≈ 80 °C, Fig. 6 and ΔΤ ≈ 37 °C, Fig. S4 ; thus, we developed integrated, low-weight, cost-effective, and programmable thermal-management materials/surfaces. Using these materials and through the combinatorial design strategy, we may design any bounded function within certain interval. Of course, it is not possible to generate emissivity values larger than unit and smaller than zero (0 < ε < 1), (practically 0.03 ≤ ε ≤ 0.96). Theoretically, we can approximately design any strictly monotonic or non-monotonic function with one turning and one inflection point (i.e. theoretically we could predict all the characteristics of the material to approximately design a trigonometric function, ε(T) = sin(T) but within certain intervals, namely [0 < T ≤ π/2] or [0 < T < 2π/3]; however, we cannot return to the initial emissivity value). Τhis is logical because the emissivity must receive a global minimum or maximum value, as well as a steady value ε > ε min , beyond a temperature level (T). In addition, a steady value for T > T max is achievable; practically, however, the maximum temperature is determined by the physical properties of the materials. More complex functions may be designed with limitations (particular intervals). Hence, in our future investigation, we will focus on the design of more complex functions, implementing more complex motifs (with more degrees of freedom) with or without nonlinear material properties. These smart patterned surfaces present the following advantages compared with the aforementioned materials/ΜΕΜS/devices. A) Their most important advantage is that we may use these fundamental properties to design the entire emissivity vs temperature curve; currently, there is no study or technological achievement in which the aforementioned advantages have been introduced. B) These smart patterned surfaces function passively; this means that they use energy from their environment, as opposed to MEMs, which use high-voltage power supplies in order to change the effective thermal emissivity. C) They have the ability to change their emissivity to a great extent; the change may either be positive or negative (Δε < 0 or Δε > 0), as opposed to certain materials for which their emissivity decreases as the temperature increases. D) The material can be developed to passively react over a very broad range of thermal requirements (i.e. −270 °C to +350 °C). E) They are low-weight compared with various other devices that are considerably efficient; however, these devices have a limited range of application and are very heavy. Radiative thermal management is crucial for every application subsystem on Earth and in Space where in the latter, thermal radiation is the only heat transfer mechanism. The present work can significantly contribute to the future thermal design of various energy systems— such as buildings and satellites/spacecraft for Space exploration—and sensors for the directional identification of a heat source or for handling different wavelengths. The potential applications are numerous. More specifically, in Space applications, all approaches mentioned in the introduction suffer from major drawbacks: the devices are extremely complex and heavyweight; MEMs are extremely complex, high-cost and low-performing; certain materials require power supplies and are low-performing or incapable of activation at various temperature ranges. These smart materials may change their emissivity from 0.03 to 0.9, as well as their absorptivity for small temperature differentials (ΔΤ < 20 °C) at any temperature level. Meanwhile, we can predetermine the behaviour of the effective thermo-optical properties as a function of temperature. In addition, practically, we can design these materials to resist in UV radiation or to absorb particular wavelengths. Using these low-cost materials, we can passively control the temperature of the systems and sub-systems of a satellite through the regulation of the absorptivity/emissivity ratio, and we may reduce the weight and the complexity of the overall system. Furthermore, this work may lead to the development of “4D materials” and thermal adaptable materials.",
"introduction": "Introduction Temperature control is one of the most common processes in Nature and man-made systems. By observing the manner in which plants and animals control their temperature 1 – 5 and their geometrical characteristics 6 – 8 , we may deduce that Nature is a specialist in purely mechanistic thermal management strategies. Nature addresses thermal management issues—prevention of overheating and damages—through an efficient holistic and mechanistic design approach. Man-made and natural strategies combine the geometrical characteristics, the materials, and the patterns of the outer surface of a body to handle the thermal energy exchange in the nano-, micro- or macro-scale. By utilising common paints 9 and by forming arrays in micro or in macroscale 10 – 14 , certain emissivity values can be achieved that would allow the temperature regulation of a body and emittance direction. On the other hand, to achieve variable thermal emissivity properties, different approaches have been adopted according to the target application, namely the development of: i) advanced materials 15 – 17 , ii) active metamaterials 18 – 20 through the formation of patterns in microscale, iii) active micro-electromechanical systems which incorporate materials of different thermo-optical properties 21 , 22 , and iv) the development of systems, such as mechanical louvers 23 and morphing radiators, which conceal/reveal materials of different emissivity values using shape memory alloys 24 , 25 . Furthermore, in architecture, large morph-able facades have been proposed for the shading of buildings 26 – 29 . The present authors have preliminary studied the feasibility to achieve variable thermal emissivity behaviour of a single self-shape structure 30 . The self-shape structures interact with their environment passively, as opposed to MEMs, which use high-voltage power supplies in order to change the effective thermal emissivity. While the extremely heavy mechanical systems and complex MEMs use hinges, actuators in order to develop simple movements, in Nature extremely complex movements can be realized through the materials’ self-shaping & self-folding capabilities in response to a stimulus 30 . Thermonastic movements, meaning ‘folding caused by a temperature stimulus’, are realized in order to transform the shape of leaves/petals under a temperature stimulus. The view factor (geometrical characteristics of the radiative object), and the material that is exposed to the environment are regulated in Nature for prevention from overheating and damages 3 , 4 . More specifically, the drooping of rhododendron leaves protects them from damage due to high irradiance and cold temperatures 3 , while poplar’s leaves present dual thermal reflectivity values on both of their leaves’ surfaces for damage prevention 4 . Based on the aforementioned mechanistic strategies, and taking them one step further, we theoretically and experimentally studied smart patterned surfaces (in a broader context “patterned surfaces” may be referred to as “metasurfaces”) and identified their fundamental properties, which govern the global behaviour of the overall thermal energy emission. We prove that smart patterned surfaces consisting of smaller structures (motifs) may be designed to respond uniquely through combinatorial design strategies. The combination and interaction of the motifs determine the global behaviour of the surfaces, thus enabling their a priori design of the effective thermo-optical properties as a function of temperature. Currently, there are no studies in radiative heat transfer in which a property (i.e. emissivity) can be fully designed as a function of another parameter (temperature). In the present research work, we studied theoretically and we proved experimentally that we can design the entire emissivity function of a patterned surface through the combination of the orientation and the colour sequences of its motifs. Through this mechanistic approach and the combination of at least two different materials with different thermo-optical properties (colours) and motifs with two different orientations, entire families of different emissivity curves can be generated. The emissivity value could be significantly altered (ε max ≥ 20ε min ) passively within a small temperature change (ΔΤ ≈ 20 ο C), and the generated functions can be classified according to their behaviour. In contrast, MEMs can be change their emissivity dynamically (ε max ≥ 5ε min ) 21 , 22 . The change may either be positive or negative (Δε < 0 or Δε > 0). Moreover, these surfaces are low-weight and they interact passively with their environment. Despite the fact that this study is not directly correlated with other types of metamaterials (mechanical metamaterials), it is important to be mentioned that in the field of mechanical metamaterials 31 – 33 and thermal metamaterials 34 , interesting studies have revealed the importance of the manner in which a value of a property can be designed using combinatorial strategies 32 . Furthermore, through the combination and the interaction of oriented unit cells 31 of a mechanical metamaterial, it is possible to design the local or global mechanical response. In our study, the entire function of the effective emissivity of a surface ε eff (T) can be designed by controlling its global maximum or minimum, its linearity, convexity, inflection point, and other characteristics. Ultimately, the handling of the effective thermo-optical properties of a surface through a material that interacts with light and temperature will lead to the development of advanced materials and structures for optimized thermal applications in satellites and other energy systems."
} | 2,930 |
27242504 | PMC4869607 | pmc | 9,193 | {
"abstract": "In situ concept-based computing is based on the notion that conceptual representations in the human brain are “ in situ .” In this way, they are grounded in perception and action. Examples are neuronal assemblies, whose connection structures develop over time and are distributed over different brain areas. In situ concepts representations cannot be copied or duplicated because that will disrupt their connection structure, and thus the meaning of these concepts. Higher-level cognitive processes, as found in language and reasoning, can be performed with in situ concepts by embedding them in specialized neurally inspired “blackboards.” The interactions between the in situ concepts and the blackboards form the basis for in situ concept computing architectures. In these architectures, memory (concepts) and processing are interwoven, in contrast with the separation between memory and processing found in Von Neumann architectures. Because the further development of Von Neumann computing (more, faster, yet power limited) is questionable, in situ concept computing might be an alternative for concept-based computing. In situ concept computing will be illustrated with a recently developed BABI reasoning task. Neurorobotics can play an important role in the development of in situ concept computing because of the development of in situ concept representations derived in scenarios as needed for reasoning tasks. Neurorobotics would also benefit from power limited and in situ concept computing.",
"conclusion": "Conclusion Brain inspired forms of computing could play an important role in combining the need for new forms of computing with more sophisticated reasoning capabilities for AI. In situ concept computing is an example of brain inspired computing, because it is based on the notion that concept representations are in situ , as found with concept representations in the brain. Conceptual forms of processing can be achieved by embedding in situ concepts in specialized neurally inspired blackboards. As illustrated with the BABI tasks, in situ concepts directly influence processing (without the need to search for them in memory), which reduces the amount of processing and power requirements needed. Neurorobotics could play an important role in developing in situ concepts and the scenarios underlying basic and common sense forms of reasoning. Neurorobotics could also benefit from power limited and concept base in situ concept computing.",
"introduction": "Introduction Important progress has been made in neurorobotics on topics such as processing sensory information [e.g., Yan et al. ( 2013 ) and Chou et al. ( 2015 )], motor control [e.g., Burms et al. ( 2015 ) and Grinke et al. ( 2015 )], and implementation with neuromorphic hardware (Stewart et al., 2016 ). In this way, neurorobotics can use brain research to develop models that process information in a neurally inspired way. Furthermore, the possibility of parallel implementation and neuromorphic hardware may be crucial for further development of robotics, because these forms of hardware can reduce the power of computing while maintaining the ability to process complex information. This allows robots to move around freely without the need for continuous energy take up. Parallel and neuromorphic forms of hardware [e.g., Benjamin et al. ( 2014 ) and Chicca et al. ( 2014 )] are also important given the problems with the further development of standard computer hardware. The past development of Von Neumann computing (more and faster processing, yet power limited) will likely not continue over the next decades (SIA and SRC, 2015 ). Therefore, new forms of hardware and new computing architectures are needed [e.g., see Nano.gov (2015)]. Williams and DeBenedictis ( 2015 ) argue for the development of dedicated “accelerators,” consisting of specific forms of computing that can interact with standard computing to enhance performance on certain tasks. Examples are GPUs for graphical processing. Other examples could be neuromorphic hardware for sensory processing and motor control. However, robots would also need to develop a form of understanding of the environment they operate in [e.g., Law et al. ( 2014 )]. At some level, they need to acquire concepts of the world around them and use these concepts in basic common sense like reasoning capabilities. However, Davies and Marcus ( 2015 ) argued recently that conceptual knowledge and basic forms of common sense reasoning is still lacking in artificial intelligence (AI), and hence also in robotics. In their view, this is even true for a system like IBMs Watson, even though that defeated humans in the game of Jeopardy, which does seem to be concept based."
} | 1,176 |
26527732 | PMC4702838 | pmc | 9,194 | {
"abstract": "The MetaCyc database ( MetaCyc.org ) is a freely accessible comprehensive database describing metabolic pathways and enzymes from all domains of life. The majority of MetaCyc pathways are small-molecule metabolic pathways that have been experimentally determined. MetaCyc contains more than 2400 pathways derived from >46 000 publications, and is the largest curated collection of metabolic pathways. BioCyc ( BioCyc.org ) is a collection of 5700 organism-specific Pathway/Genome Databases (PGDBs), each containing the full genome and predicted metabolic network of one organism, including metabolites, enzymes, reactions, metabolic pathways, predicted operons, transport systems, and pathway-hole fillers. The BioCyc website offers a variety of tools for querying and analyzing PGDBs, including Omics Viewers and tools for comparative analysis. This article provides an update of new developments in MetaCyc and BioCyc during the last two years, including addition of Gibbs free energy values for compounds and reactions; redesign of the primary gene/protein page; addition of a tool for creating diagrams containing multiple linked pathways; several new search capabilities, including searching for genes based on sequence patterns, searching for databases based on an organism's phenotypes, and a cross-organism search; and a metabolite identifier translation service.",
"introduction": "INTRODUCTION MetaCyc ( MetaCyc.org ) is a highly curated reference database of small-molecule metabolism from all domains of life. It contains data about enzymes and metabolic pathways that have been experimentally validated and reported in the scientific literature ( 1 ). MetaCyc is a uniquely valuable resource due to its exclusively experimentally determined data, intensive curation, and tight integration of data and references. It is commonly used as a reference in various fields including biochemistry, enzymology, genome and metagenome analysis and metabolic engineering. In addition to its role as a general reference on metabolism, MetaCyc can be used by the PathoLogic component of the Pathway Tools software ( 2 ) as a reference database to computationally predict the metabolic network of any organism that has a sequenced and annotated genome ( 3 ). During this automated process, the predicted metabolic network is captured in the form of a Pathway/Genome Database (PGDB). In addition to the automated creation of PGDBs, Pathway Tools enables scientists to improve and update these computationally generated PGDBs by manual curation. SRI has used MetaCyc to create more than 5700 PGDBs (as of September 2015), which are available through the BioCyc ( BioCyc.org ) website. In addition, many groups outside SRI have generated many thousands of additional PGDBs. Interested scientists may adopt any of the SRI PGDBs through the BioCyc website for further curation ( biocyc.org/intro.shtml#adoption )."
} | 720 |
28840937 | null | s2 | 9,195 | {
"abstract": "The archaeon Pyrococcus furiosus is emerging as a metabolic engineering platform for production of fuels and chemicals, such that more must be known about this organism's characteristics in bioprocessing contexts. Its ability to grow at temperatures from 70 to greater than 100°C and thereby avoid contamination, offers the opportunity for long duration, continuous bioprocesses as an alternative to batch systems. Toward that end, we analyzed the transcriptome of P. furiosus to reveal its metabolic state during different growth modes that are relevant to bioprocessing. As cells progressed from exponential to stationary phase in batch cultures, genes involved in biosynthetic pathways important to replacing diminishing supplies of key nutrients and genes responsible for the onset of stress responses were up-regulated. In contrast, during continuous culture, the progression to higher dilution rates down-regulated many biosynthetic processes as nutrient supplies were increased. Most interesting was the contrast between batch exponential phase and continuous culture at comparable growth rates (∼0.4 hr"
} | 277 |
37372055 | PMC10295459 | pmc | 9,196 | {
"abstract": "Simple Summary Hydrothermal vents are regions such as hot springs found on the seafloor in the mid-ocean and near tectonic plates. They contain fluids with highly enriched carbon dioxide, which is the central element of life on Earth. Many organisms live in this environment and can survive in extreme conditions (extremophiles), such as up to 400 °C or higher, low pH, and high pressure. All organisms need the carbonic anhydrase (CA) enzyme to handle the acid-base imbalance through the hydration of carbon dioxide and the production of bicarbonate necessary for pH homeostasis and many cellular functions. The CAs have been categorized into eight families. In this study, we focused on α-, β-, and γ-CAs from the thermophilic microbiome of marine hydrothermal vents. Microorganisms in this environment need CA to capture CO 2 , which is an important contribution to marine hydrothermal vent ecosystem functioning. Previously, we showed the transfer of β-CA gene sequences from prokaryotes to protozoans, insects, and nematodes via horizontal gene transfer (HGT). HGT is not only the transfer and movement of genetic information between organisms but is also a powerful tool in natural biodiversity. If the CA coding gene is transferred horizontally between microorganisms in hydrothermal vents, it is hypothesized that CA is essential for survival in these environments and one of the key players in the carbon cycle in the ocean. Abstract Carbonic anhydrases (CAs) are metalloenzymes that can help organisms survive in hydrothermal vents by hydrating carbon dioxide (CO 2 ). In this study, we focus on alpha (α), beta (β), and gamma (γ) CAs, which are present in the thermophilic microbiome of marine hydrothermal vents. The coding genes of these enzymes can be transferred between hydrothermal-vent organisms via horizontal gene transfer (HGT), which is an important tool in natural biodiversity. We performed big data mining and bioinformatics studies on α-, β-, and γ-CA coding genes from the thermophilic microbiome of marine hydrothermal vents. The results showed a reasonable association between thermostable α-, β-, and γ-CAs in the microbial population of the hydrothermal vents. This relationship could be due to HGT. We found evidence of HGT of α- and β-CAs between Cycloclasticus sp., a symbiont of Bathymodiolus heckerae, and an endosymbiont of Riftia pachyptila via Integrons. Conversely, HGT of β-CA genes from the endosymbiont Tevnia jerichonana to the endosymbiont Riftia pachyptila was detected. In addition, Hydrogenovibrio crunogenus SP-41 contains a β-CA gene on genomic islands (GIs). This gene can be transferred by HGT to Hydrogenovibrio sp. MA2-6, a methanotrophic endosymbiont of Bathymodiolus azoricus , and a methanotrophic endosymbiont of Bathymodiolus puteoserpentis. The endosymbiont of R. pachyptila has a γ-CA gene in the genome. If α- and β-CA coding genes have been derived from other microorganisms, such as endosymbionts of T. jerichonana and Cycloclasticus sp. as the endosymbiont of B. heckerae , through HGT, the theory of the necessity of thermostable CA enzymes for survival in the extreme ecosystem of hydrothermal vents is suggested and helps the conservation of microbiome natural diversity in hydrothermal vents. These harsh ecosystems, with their integral players, such as HGT and endosymbionts, significantly impact the enrichment of life on Earth and the carbon cycle in the ocean.",
"conclusion": "5. Conclusions According to the results of this big data mining and bioinformatics study, α-, β-, and γ-CAs from the thermophilic microbiome of marine hydrothermal vents have a reasonable evolutionary relationship. The α- , β- , and γ-CA genes can be transferred to other microorganism habitats in hydrothermal vents via HGT and cause natural biodiversity in this extreme ecosystem. Given the presence of an integron with an integrase coding gene in the Cycloclasticus sp. symbiont of Bathymodiolus heckerae, it is highly possible that the α- CA coding gene is transferred between Cycloclasticus sp. as the symbiont of B. heckerae and endosymbiont of Riftia pachyptila . This evolutionary phenomenon can also be applied to β-CA -coding genes. According to the β- CA gene on the endosymbiont of T. jerichonana and the endosymbiont of R. pachyptila and the evolutionary relationship between them, the HGT of the β- CA gene from the endosymbiont of T. jerichonana to the endosymbiont of R. pachyptila and conversely is highly possible. In addition, the endosymbiont of R. pachyptila has a γ- CA gene on the chromosome; if α- and β- CA coding genes are derived from other microorganisms, such as the endosymbiont of T. jerichonana and Cycloclasticus sp. as the symbiont of B. heckerae, the theory of the necessity of the CA enzyme for survival in this extreme ecosystem and its effect on preserved natural biodiversity is proposed. Despite the presence of the α- CA gene in R. pachyptila and the α- , β -, and γ- CA genes in its endosymbiont, this theory is suggested for this giant marine worm. Therefore, the prokaryotic endosymbionts of mussels and giant marine worms have evolutionary relationships through HGT. With more focus on the HGT phenomenon, endosymbionts are integral parts of natural biodiversity and ecosystem functioning of marine hydrothermal vents.",
"introduction": "1. Introduction Deep-sea hydrothermal vents are one of the best environments for evolutionary studies. Hydrothermal vents are regions such as hot springs found on the seafloor. These are located in the mid-ocean and near tectonic plates initially discovered in 1977 at a depth of 2.5 km around a hot spring on the Galápagos volcanic rift (spreading ridge) off the coast of Ecuador [ 1 , 2 ]. Based on their characteristics, deep-sea hydrothermal vents are called either black smokers or white smokers [ 3 ]. Black smokers’ fluid temperature goes up to 400 °C or above and has a low pH, but white smokers have an alkaline pH, and their temperature is approximately 40–75 °C [ 3 ]. Hydrothermal vents contain fluids with highly enriched carbon dioxide (CO 2 ), which are discharged into the deep sea by these vents [ 4 ]. CO 2 is a very stable form of carbon, the central element of life on Earth, and consists of a carbon atom covalently double-bonded to two oxygen atoms. Carbonic acid (H 2 CO 3 ) is derived from the reaction of CO 2 and water molecules, so the product is an unstable compound that spontaneously splits into bicarbonate (HCO 3 − ) and protons (H + ). Many organisms live in this environment, especially bacterial and archaeal species that can survive in extreme conditions such as high temperatures and pressure. The organisms adapted to this habit are called extremophiles. All organisms need carbonic anhydrases (CAs) to handle the large amount of CO 2 and, consequently, the related acid-base imbalance [ 5 , 6 , 7 ]. CA is the metalloenzyme that catalyzes the reversible hydration of CO 2 to HCO 3 and H + as follows: CO 2 + H 2 O ↔ HCO 3 − + H + \n CAs are encoded by eight evolutionarily divergent gene families, including alpha (α), beta (β), gamma (γ), delta (δ), zeta (ζ), eta (η), theta (θ), and iota (ι) CA. α-CA has been reported in vertebrates, prokaryotes, fungi, algae, protozoa, and plants [ 7 ]. β-CA is expressed in prokaryotes, plants, fungi, protozoa, arthropods, and nematodes [ 8 , 9 , 10 , 11 , 12 , 13 ]. γ-CA is present in many plants, fungi, and prokaryotes. δ-CA and ζ-CA are present in marine diatoms [ 7 , 12 ]. η-CA was identified in the causative agent of malaria, Plasmodium spp., and θ-CA was identified in marine diatoms [ 7 , 14 , 15 ]. Iota(ι)-CA was recently reported to be expressed in diatoms and bacteria [ 16 ]. In this study, we focused on α-, β-, and γ-CAs from the thermophilic microbiome of marine hydrothermal vents. These metalloenzymes have an active site containing a Zn(II) metal ion cofactor [ 17 ], while Co(II) and Fe(II) can be included in α- and γ-CA, respectively [ 7 ]. The structures of α-CAs are frequently monomers and rarely dimers [ 18 ]; β-CAs are dimers, tetramers, or octamers [ 19 ]; and γ-CAs are trimers [ 20 ]. A previous study showed that β-CA gene sequences could be transferred from prokaryotes to protozoans, insects, and nematodes via HGT [ 21 ]. Additionally, the involvement of bacterial β-CA gene sequences in the gastrointestinal tract and their horizontal transfer to their host during evolution has been demonstrated [ 22 ]. HGT, also called lateral gene transfer (LGT), is the transfer and movement of genetic information between organisms and thus is differentiated from the vertical transmission of genes from parent to the next generations [ 23 ]. HGT plays a crucial role in natural biodiversity as a general mechanism [ 24 , 25 ], and it often causes dramatic changes in the ecological and pathogenic properties of bacterial species, thereby promoting microbial diversification and speciation [ 25 ]. HGT may occur via mobile genetic elements (MGEs) such as integrons, genomic islands (GIs), integrative conjugative elements (ICEs), transposable elements (TEs), plasmids, and phages [ 26 , 27 , 28 , 29 , 30 ]. MGEs are parts of DNA that encode enzymes and other proteins that interpose the transfer of DNA in HGT within genomes (intracellular mobility) or between bacterial cells (intercellular mobility) [ 26 ]. Intercellular transfer of DNA takes three forms in prokaryotes: transformation, conjugation, and transduction [ 31 ]. Integrons are MGEs that allow the capture and expression of exogenous genes. Integrons have three essential core features: intI, attI, and Pc [ 32 , 33 ]. IntI is the gene encoding an enzyme for catalyzing recombination between incoming gene cassettes called integron integrase (IntI) [ 32 ]. AttI is an integron-associated recombination site [ 33 ], and Pc is an integron-associated promoter that is expressed once a gene cassette is recombined [ 34 ]. The length of GIs is more than 10 kb, a part of a chromosome, recognized as discrete DNA segments, and can be different from closely related strains, and transposase is a primary tool for HGT through GIs [ 32 , 33 , 34 ]. Another family of MGEs is the integrative conjugative element (ICE), called the conjugative transposon. ICEs have two features: first, they are integrated into a host genome, and second, they encode a type IV secretion system (functional conjugation system) [ 28 , 35 ]. TEs are DNA sequences that can move from one location to another in the genome [ 30 ]. TEs fall into two classes: retrotransposons (Class I) or RNA transposons [ 36 ] and true transposons (Class II) or DNA transposons that consist of a transposase gene with two terminal inverted repeats (TIRs) on either side [ 30 ]. Additionally, insertion sequences (IS) are small MGEs that carry more than one or two transposase genes [ 37 ]. The CA genes may be transferred between organisms living in hydrothermal vents and their endosymbionts via HGT. Endosymbiotic bacteria are located in the trophosome of the host, which contains animal cells, so-called bacteriocytes [ 38 ]. For instance, one of the important living organisms living in deep-sea hydrothermal vents is the giant tubeworm Riftia pachyptila , which lives with its symbiont bacteria. Nitrate, oxygen, hydrogen sulfide, and inorganic carbon are taken up from the environment and it feeds its symbiotic bacteria with these substances in an organ known as the trophosome [ 39 ]. In addition, about one-third of the added carbon from atmospheric CO 2 uptake into the ocean increases dissolved CO 2 in seawater [ 40 ]. The accompanying acidification may reduce the seawater saturation of calcite, thus affecting marine calcifications. CA helps the concentration of inorganic carbon in the fluid from which calcium carbonate is sedimented and directly affects the calcification in some calcifiers, such as gastropods, oysters, and giant clams as well as coral calcification. The calcification can be reduced by 40%, which has been affected by high atmospheric CO 2 levels. Even a modest impact on producing carbonate shells and skeletons may have important consequences on the global carbon cycle [ 41 ]. Microorganisms in this environment need CA to capture CO 2 , which is an important contribution to marine hydrothermal vent ecosystem functioning [ 42 ]. It has been suggested that α-CA evolution may contribute to the vulnerability to environmental changes of bivalves and their diversity [ 43 ] since HGT would create a large variability acted on by natural selection [ 39 ]. If the coding gene of this enzyme is transferred horizontally between hydrothermal vent microorganisms, it is hypothesized that CA is essential for survival and for preserving natural biodiversity in this extreme ecosystem. For this purpose, we investigated the evolutionary relationship and the possibility of HGT in the hydrothermal vent ecosystem. We conducted a large data mining and bioinformatics study focusing on the HGT of α-, β- and γ-CA genes in the microbial population of deep-sea hydrothermal vents.",
"discussion": "4. Discussion The evolutionary process in hydrothermal vent ecosystems and the role of viruses in the biodiversity in this harsh environment have been studied previously. A study performed by Cheng et al. [ 145 ] revealed that bacteriophages are the most predominant viruses across the global hydrothermal vents, while single-stranded DNA viruses, including Microviridae and small eukaryotic viruses, have been located in the next steps. The metagenomics analysis showed that this virome plays a crucial role in the evolution and biodiversity of the microbiome of hydrothermal vents, especially Gammaproteobacteria and Campylobacterota [ 145 ]. Although the bacteriophages have no role in the HGT of CA genes in the hydrothermal ecosystems, our previous studies showed the HGT of β-CA genes from prokaryotic endosymbionts to their protozoan, insects, and nematodes hosts. In addition, the genomic islands have been shown to have a potential role in the HGT of β-CA genes from ancestral prokaryotes to protists. Since then, no further study has been performed on the HGT of CA genes. Since hydrothermal vent ecosystems have been reported as potent environments for HGT and biodiversity, these harsh deep-sea fissures were studied. According to the heatmap and phylogenetic analysis ( Figure 2 ) of α-CAs, Bpm-ACA and Ca-ACA showed no significant relationship with the other α-CAs. Hc(XLC-2)ACA, Hc(SP41)-ACA, and Sca-ACA clustered together with branch bootstrap values of 1.00, showing significant relationships. Additionally, G(HR-1)-ACA and G(EPR-M)-ACA had a branch bootstrap value of 1.00, indicating a robust evolutionary relationship similar to that between Sr-ACA and S(NBC37-1)-ACA, whose bootstrap value was also 1.00. Similar to a previous study, the branch for Pm-ACA and Ph-ACA was observed to have a high bootstrap value of 0.99. A high branch bootstrap value of 0.86 was observed for CsBh-ACA and Erp-ACA. It is necessary to mention that all the α-CAs above belong to the Proteobacteria phylum except for Pm-ACA and Ph-ACA, which belong to the Aquificae phylum. According to the heatmap and phylogenetic analysis ( Figure 3 ) of β-CAs, in clade A, Opr-BCA and It-BCA have poor relationships with other clade members, showing a branch bootstrap value of 0.37. All members of clade B have the same root, but Mfe-BCA, Mfo-BCA, and Mb-BCA have poor relationships with other clade members. In clade C, a significant relationship between Sr-BCA, S(NBC37)-BCA, and Si-BCA showed a branch bootstrap value of 1.00, in which pairwise sequence identities of more than 88.6% were revealed. Although these three cases with a branch bootstrap value of 0.92 have a significant relationship with Ns-BCA, they have a poor relationship with other members of clade C. In clade D, a relationship between Erp-BCA, Et-BCA, and CsBh-BCA was observed with a 1.00 branch bootstrap value, in which a pairwise sequence identity of more than 83.5% was observed for all three. In clade E, the cluster containing Hts-BCA, Ts-BCA, Hc(XCL-2) –BCA, and Btt-BCA was observed with a branch bootstrap value of 1.00. Clade F with a 0.3 branch bootstrap value did not show a good relationship with other clades, while Di-BCA and Ta-BCA have the same root as Tp-BCA, a member of archaea. According to the heatmap and phylogenetic analysis of γ-CAs ( Figure 4 ), clades A and B, with branch bootstrap values of 0.02 and 0.001, respectively, have a very poor relationship with other clades, including Erpi-GCA, Erp-GCA, and ET-GCA with different branch bootstrap values of more than 0.98 and a pairwise sequence identity value of more than 97.78, which have a significant relationship together. In addition, a meaningful relationship was observed for Hts-GCA, Ts(S5)-GCA, and Sca-GCA with a branch bootstrap value of 0.99. In clade C, archaea and bacteria have the same root, and according to the heatmap, all archaea have high pairwise sequence identity values. In clade D, Gbah-GCA and Ggac-GCA have a good relationship with a branch bootstrap value of 0.99 and a pairwise sequence identity value of 69.18. According to the heatmap of γ-CA ( Figure 4 B) in clade D, Gs-GCA with a branch bootstrap value of 0.99 and a pairwise sequence identity value of 94.29 had a significant relationship with Gk-GCA. CALIN elements ( Table 1 ) might have arisen from a missing integrase in a previously complete integron. The α- and β- CA genes from CALIN may be cut by the integron-integrase and reinserted in the integron at an attI site. Since the stable circular form of CALINs can survive in the environment, these genetic elements can be taken up by transformable bacteria through a transformation mechanism [ 146 ]. On the other hand, integrons often capture cassettes from CALIN elements [ 129 ], so the α- and β- CA genes can be derived from different microorganisms or transferred to other hosts. According to the phylogenetic trees of α- and β-CA ( Figure 2 A and Figure 3 A), Erp-ACA has the highest relationship with CsBh-ACA, with a bootstrap value of 0.40 and a pairwise sequence identity value of 57.61, which is a weak relationship. In addition, it has a relatively weak relationship with G(EPR-M)-ACA and G(HR-1)-ACA twins with a bootstrap value of 0.44 and pairwise sequence identity values of 55.19 and 57.39, respectively. The β-CAs in clades A and B, with 0.02 and 0.001 branch bootstrap values, respectively, did not have a good relationship with other clades. At the same time, Erp-BCA is the highest related compound to Et-BCA and CsBh-BCA, with a 1.00 branch bootstrap value and pairwise sequence identity of 99 and 100, respectively. Moreover, the β- CA gene ( Et-BCA ) from the endosymbiont of T. jerichonana is related the highest to Erp-BCA and CsBh-BCA, with a branch bootstrap value of 1.00 and pairwise sequence identity of 99 and 84, respectively, which indicates the possibility of horizontal gene transfer of β- CA coding genes in these microorganisms. It should be noted that inorganic carbon from CO 2 is first obtained from the environment via diffusion through the plume, a branchial organ [ 147 ]. Next, CO 2 is transformed to HCO 3 − and transported to trophosome cells, particularly bicarbonate, at the surrounding branchial plume interface. Then, HCO 3 − is transformed to CO 2 on the body fluids and bacterial cells [ 148 ] and adhered via the bacterial symbiont enzyme RuBisCO form II. In the arginine biosynthesis and pyrimidine pathways, carbamylphosphate synthetase uses inorganic HCO 3 − to start the biosynthesis process. Since the metabolic relationship between R. pachyptila and its endosymbiont is vital for the survival of each organism, this issue can explain the cause and importance of HGT of CA in these organisms. Furthermore, R. pachyptila contains an α- CA gene [ 149 ] with UniProt ID: Q8MPH8, which is not similar to Erp-ACA. Additionally, T. jerichonana has no reported CA family. Identification of the β- CA gene beside three transposase genes on one of the GIs of H. crunogenus SP-41 could lead to the theory that this gene may be transferred with plasmids and phages or occur through transposon accumulation in recombination sites. Experimental studies have suggested the release of about 1.5 billion symbionts from dead tubeworm clumps into the environment [ 47 ], which provides the opportunity for the spread and HGT of CA genes in the environment and preparing the biodiversity condition. In addition to the β-CA phylogenetic tree, the heatmap showed that the Hc(SP41)-BCA in clade B is closely related to Hs(MA2-6)–BCA with a branch bootstrap value of 0.99 and a pairwise sequence identity value of 82.9. In addition, MeBa-BCA and MeBp-BCA showed a close relationship with Hc(SP41)-BCA with branch bootstrap values of 1.0 and pairwise sequence identity values of 76.56 for both cases. The HGT of hydrogenase-coding genes between H. crunogenus SP-41 and H. crunogenus XCL-2 was studied previously [ 150 ]; however, in this study, H. crunogenus SP-41 ( Hc(SP41)-BCA ) had no HGT relationship with H. crunogenus XCL-2. R. pachyptila has cytosolic α-CA in the trophosome. Although these organisms need secretory CA for their physiological needs and use Erp-ACA, this theory must be experimentally studied. The significance of this study revealed that there is an evolutionary relationship between Hc(XLC-2)ACA, Hc(SP41)-ACA, and Sca-ACA; G(HR-1)-ACA and G(EPR-M)-ACA; Sr-ACA and S(NBC37-1)-ACA; Pm-ACA and Ph-ACA; and CsBh-ACA and Erp-ACA in α-CAs. In addition, there is an evolutionary relationship between Sr-BCA, S(NBC37)-BCA, and Si-BCA; Erp-BCA, Et-BCA, and CsBh-BCA; and Hts-BCA, Ts-BCA, Hc(XCL-2) –BCA, and Btt-BCA in β-CAs. Additionally, there is an evolutionary relationship between Erpi-GCA, Erp-GCA, and ET-GCA; Hts-GCA, Ts(S5)-GCA, and Sca-GCA; Gbah-GCA and Ggac-GCA; and Gs-GCA and Gk-GCA in γ-CAs. Elevated CO 2 pressure in seawater can affect marine organisms by disrupting acid-base physiology and decreasing mineralization rates (affecting calcium carbonate saturation and calcification). Ocean uptake of anthropogenic CO 2 and associated changes in seawater chemistry adversely affect biodiversity, other ecosystem processes, and the global carbon cycle [ 151 ]. The HGT and distribution of CA genes in the hydrothermal vent area may also help the survival and diversity of the organisms in this environment."
} | 5,628 |
35621971 | PMC9147911 | pmc | 9,197 | {
"abstract": "Microalgae host varied microbial consortium harboring cross-kingdom interactions with fundamental ecological significance in aquatic ecosystems. Revealing the complex biofunctions of the cultivable bacteria of phycosphere microbiota is one vital basis for deeply understanding the mechanisms governing these dynamic associations. In this study, a new light-yellow pigmented bacterial strain LZ-28 was isolated from the highly-toxic and harmful algal bloom-forming dinoflagellate Alexandrium catenella LZT09. Collective phenotypic and genotypic profiles were obtained to confidently identify this strain as a new Mameliella \n alba member. Comparative genomic analysis showed that strain LZ-28 shared highly similar functional features with other four marine algae-derived M. alba strains in spite of their distinctive isolation sources. Based on the bioactivity assaying, the mutual growth-promoting effects between bacterial strain LZ-28 and algal strain LZT09 were observed. After the culture conditions were optimized, strain LZ-28 demonstrated an extraordinary production ability for its bioflocculanting exopolysaccharides (EPS). Moreover, the portions of two monosaccharides glucose and fucose of the EPS were found to positively contribute to the bioflocculanting capacity. Therefore, the present study sheds light on the similar genomic features among the selected M. alba strains, and it also reveals the potential pharmaceutical, environmental and biotechnological implications of active EPS produced by this new Mameliella alba strain LZ-28 recovered from toxic bloom-forming marine dinoflagellate.",
"introduction": "1. Introduction The term “phycosphere” was initially coined in 1972 to describe the region immediately surrounding an individual algal cell that is enriched in organic matters, which are exuded by the cell into its surroundings [ 1 ]. It is known as the aquatic analogue of the rhizosphere in the soil ecosystem [ 2 ]. The aquatic phycosphere niche is regarded as the boundary of phytoplankton holobionts and the ecological interface for algae–bacteria interactions (ABI) [ 3 ]. The relationships that usually span mutualism, commensalism, antagonism, parasitism, and competition, usually involve cross-kingdom exchanges of nutrients, secondary metabolites, infochemicals, and gene-transfer agents (GTA) [ 4 ]. It has subsequently become apparent that these microscale interactions are far more sophisticated than previously thought. It usually requires the close spatial proximity of two sides involved in the interactions, which are usually facilitated by bacterial colonization in the phycosphere niches [ 5 ]. Within this microscopic interface, the exopolysaccharides or extracellular polysaccharides (EPS) contributed by both partners are the main components of the matrix, which embeds the proliferating cells and promotes colony formation [ 5 ]. It is also the essential chemical intermedia connecting the microscale interactions [ 1 , 2 , 6 ]. Presently, it is becoming increasingly clear that these complex inter-species or intra-species associations usually exert further ecosystem-scale influences on fundamental processes including primary production, phycotoxin biosynthesis, biogeochemical cycles and the microbial loop in the oceans [ 7 ]. Currently, the rapidly increasing observations show that phytoplankton-associated bacterial communities are often limited to a small handful of taxa groups, including specific members of the Roseobacter group ( Rhodobacteraceae ), Flavobacteraceae , and Alteromonadaceae [ 8 , 9 , 10 ]. These observations also provide emerging evidence for the species-specific associations between the phytoplanktons and their closely-associated bacterial consortium [ 2 , 3 , 4 , 5 , 6 ]. Rhodobacteraceae usually accounts for an overwhelming proportion of the bacterioplankton communities in marine environments [ 8 , 9 ]. Within this family, the genus Mameliella has become one key and representative member in the Roseobacter group, although it was initially established one decade ago with Mameliella alba JLT354-W T as the type species [ 11 ]. Moreover, Mameliella members widely contributed to the marine environments in aromatic compound degradation, biogeochemical cycles of carbon and sulfur, dimethylsulfoniopropionate demethylation, and the production of diverse secondary metabolites [ 12 ]. In recent decades, four type species, Mameliella phaeodactyli , Mameliella atlantica , Ponticoccus lacteus , and Alkalimicrobium pacificum , were reclassified later as heterotypic synonyms of M. alba based on their phylogenomic characterizations [ 13 ]. Nowadays, the genus Mameliella contains only two type species with correctly published names, including M. alba and M. sediminis ( https://lpsn.dsmz.de/genus/mameliella , accessed on 10 May 2022). Currently, over 1500 species of free-living dinoflagellates have been described, and the harmful algal blooms (HABs)-causing group occupies over 300 species [ 14 ]. Among them, over one quarter are toxic species that produce diverse types of phycotoxins, such as paralytic shellfish-poisoning (PSP) toxins [ 15 , 16 ]. Within the toxic dinoflagellate species, the globally distributed genus Alexandrium , in which the members were widely found in sub-polar, temperate and tropical coastal water environments, is a particularly well-known bloom-forming group due to its large and widespread threat to seafood production and human health [ 17 , 18 ]. The importance of the phycosphere has been postulated for four decades, yet only recent new technological progress in high-throughput pyrosequencing made it possible to start teasing apart the complex nature of the microbial composition within this unique microbial habitat [ 3 , 4 , 5 , 6 , 19 , 20 , 21 , 22 , 23 , 24 ]. The culture-independent isolation of the cultivable strains is still the first critical step for exploring the dynamics of algae–bacteria interactions [ 3 , 4 , 5 , 6 ]. Previously, we conducted the Phycosphere Microbiome Project (PMP) to convey the compositions of varied microbial consortiums that were closely-associated with diverse HAB-forming dinoflagellates [ 25 , 26 , 27 , 28 , 29 , 30 ]. During our investigation, the genus Mameliella was found to be one predominant group and was widely distributed in the bacterial communities associated with various marine dinoflagellates [ 10 , 19 , 20 , 21 , 22 , 23 ]. Consequently, a new red-pigmented bacterial strain, LZ-28, was isolated from the highly toxic Alexandrium catenella LZT09. Strain LZ-28 produced active bioflocculanting exopolysaccharides, which were also discovered in other cultivable bacterial strains isolated from marine dinoflagellates [ 27 , 28 , 29 , 30 , 31 ]. Additionally, it demonstrated the obvious growth-promoting ability of the algal strain LZT09, as previously verified by other marine bacteria strains derived from the same algal host [ 21 , 30 ]. Thus, the present work aims to characterize the taxonomic status of this new bacterial strain using combined taxonomic and phylogenomic approaches. Moreover, comparative genomic analysis is then emphatically performed for the five marine algae-derived strains among the ten selected M. alba strains to reveal their shared functional nature. Based on bioactivity assaying after culture condition optimization, the contribution of the individual monosaccharide portion of the EPS produced by strain LZ-28 to the bioflocculanting bioactivity is also characterized.",
"discussion": "2. Results and Discussion 2.1. Characterization of the Composition of the Bacterial Community of Algal Strain LZT09 To reveal the microbial composition of the bacterial community associated with LZT09, the high-throughput sequencing of the V3–V4 variable region of bacterial 16S rRNA was performed. Based on the obtained result, the phylum Pseudomonadota represented the overwhelming dominant bacterial fraction (53.9%) of the total bacterial community of LZT09, followed by Bacteroidota (26.7%) and Cyanobacteria (15.2%). Other lineages, including Actinomycetota, Bacillota, and Spirochaetota, were also present, but each constituted less than 1.0% of the total ( Figure 1 ). Similarly, previous studies showed that two phyla, Pseudomonadota and Bacteroidota, were the most major bacterial groups affiliated with the investigated bacterial community associated with marine dinoflagellates [ 20 , 21 ]. Moreover, members of Rhodobacteraceae were reported to be frequently present during various stages of HABs [ 32 , 33 ], and also regarded as a major taxa associated with some diatom cultures [ 34 , 35 ]. In this study, Rhodobacteraceae was found to occupy only 8.7% of the total bacterial community of LZT09, which held the fifth dominant group following the family Cryomorphaceae (22.9%), one unidentified Cyanobacteria (16.8%), Saccharospirillaceae (14.7%), and the Hyphomonadaceae (9.7%). Cryomorphaceae is a member of the order Flavobacteriales within the phylum Bacteroidota, and has been found to distribute in a wide range of marine and terrestrial systems from tropical to polar regions [ 36 ]. Moreover, molecular phylogenetic studies show that the phylotypes related to the family Cryomorphaceae belong to the abundant coastal phytoplankton bloom-responding flavobacterial group [ 37 ]. In addition, at the genus level, five dominant bacterial groups, including one unidentified Cryomorphaceae (22.9%), one unidentified Cyanobacteria (16.8%), Saccharospirillum (14.6%), Maricaulis (5.6%), and Mameliella (5.1%), were identified, and totally accounted for 65.0% of the total bacterial community of LZT09 ( Figure 1 ). It is worth noting that only one genus, Microcystis , was found in the Cyanobacteria group, which indicated that some cryptic cyanobacterial lineages were potentially hidden by the dinoflagellate host [ 37 ]. 2.2. Phenotypic and Biochemical Characteristics of Bacterial Strain LZ-28 Previously, nine cultivable bacterial strains, including strain LZ-28, were isolated from the cultivable PM of LZT09 [ 10 ]. However, no strain was found to belong to the dominant group Cryomorphaceae. Previous studies have shown that the most cultured Cryomorphaceae species were mainly isolated from the low-temperature ecosystems, although the family Cryomorphaceae was widespread in a wide range of non-extreme ecosystems. These findings indicate the resistance of the Cryomorphaceae members towards be cultivation [ 36 ]. Thus, the constant optimization of the isolation procedure for the cultivable bacteria strains isolated from marine dinoflagellates still needs to be performed, especially by means of the emergence and prevalence of modern multi-omics data [ 38 , 39 ]. Based on the phenotypic characterization, the cellular colonies of strain LZ-28 grown on marine agar (MA) were circular, smooth, and convex with light-yellow colors. Cells of strain LZ-28 were Gram-stain-negative, rod shaped, and non-motile with the sizes of 0.7–1.0 μm for the width and 2.0–2.9 μm for the length ( Figure 2 ). The physiological characterization revealed that strain LZ-28 grew at the pH ranging from 5.0 to 10.0 with the optimum growth at pH 7.0. The growth temperature ranged from 15 °C to 40 °C with the optimum value of 25–28 °C, at the presence of 1–10% ( w/v ) NaCl with the optimum value of 2.5%. No bacterial growth was observed under anaerobic conditions when grown on MA, even after three week-long incubations. The hydrolysis of starch, L-tyrosine, and gelatin were observed. Nitrate was reduced to nitrite, but the reduction of nitrite to nitrogen was not observed. Chemotaxonomic analysis showed that the major cellular fatty acids of strain LZ-28 consisted of C 18:0 , C 18:1 ω7c 11-methyl, and C 19:0 cyclo. Detailed comparisons of the morphological, biochemical, and physiological characteristics of strain LZ-28 and the other five M. alba strains are summarized in Table 1 . The differential characteristics between strain LZ-28 and other M. alba members can be easily found, even though they share very high 16S rRNA gene-sequence similarities, for example, the distinguished colony color, the ability for citrate utilization, the presence/absence of a minor fatty acid of C 17:1 ω8c, as well as the apparently varied polar lipid profiles among the six compared M. alba strains. 2.3. Phylogenetic Analysis Based on 16S rRNA Gene Sequences In the phylogenetic tree constructed by the maximum-likelihood (ML) algorithm using the 16S rRNA gene sequences, strain LZ-28 formed a monophyletic branch together with the type and other non-type strains of M. alba , as well as the selected representative Rhodobacteraceae members ( Figure 3 ). Previously, M. phaeodactyli (type-strain KD53) [ 12 ], M. atlantica (type-strain L6M1-5) [ 40 ], Ponticoccus lacteus (type-strain JL351) [ 41 ], and Alkalimicrobium pacificum (type-strain F15) [ 42 ] were later reclassified as heterotypic synonyms for M. alba based on the phylogenomic characterizations [ 13 ]. In this study, the phylogenetic analysis showed that strain LZ-28 shared high 16S rRNA gene similarity values of 99.70, 99.77, 99.92, 99.60, and 99.40% with five other M. alba strains, JLT354-W T ; KD53; L6M1-5; JL351; and F15, respectively, although they were obviously isolated from different sources ( Table 1 ). These values were all exceeding the threshold values (98.65%) generally accepted for species delineation [ 43 ]. Thus, it collectively indicated that strain LZ-28 was affiliated with the genus Mameliella , and was probably a new member of M. alba . 2.4. Genomic Features of the Selected M. alba Strains Due to the high-similarity values of the 16S rRNA gene sequence between strain LZ-28 and its phylogenetic neighbors in the family Rhodobacteraceae , the extra phylogenomic characterization was performed to ensure its taxonomic status. The genome sequence of strain LZ-28 was obtained by our previous study [ 10 ], and the other nine M. alba strains with available genomes were obtained from the NCBI database ( https://www.ncbi.nlm.nih.gov , accessed on 2 April 2022). The isolation sources and the general genomic characteristics of the ten selected M. alba strains are summarized in Table 2 . Among the ten selected M. alba strains, only strain KU6B has a complete genome containing a 5.386 Mbp circular chromosome and three circular plasmids of 256, 112, and 76 Kbp, respectively [ 44 ]. The genomic size of strain LZ-28 was 5.66 Mb with a G+C content of 64.94%. It contained 5497 protein-coding DNA sequences (CDSs) and 61 RNA genes, including 8 rRNA, 50 tRNA, and 3 other RNA genes, respectively, as well as 86 pseudogenes. The genome of strain LZ-28 was the fifth largest, with a genome range of 5.26–5.90 Mb. However, the genomic DNA G+C content of strain LZ-28 was the lowest (64.94 mol%) among the ten selected M. alba strains. 2.5. Phylogenomic Characterization of Bacterial Strain LZ-28 To further infer the phylogenetic relationship among the ten selected M. alba strains with available genome sequences, the phylogenomic tree using an up-to-date bacterial core gene set (UBCG) consisting of 92 bacterial core genes was also performed. The constructed phylogenetic tree is shown in Figure 4 . It clearly shows that, among the five marine algae-derived M. alba strains, strain LZ-28 clusters together with strain KD53, which is isolated from marine diatom Phaeodactylum tricornutum [ 12 ]. The other three strains, PBVC088, UMTAT08, and Ep20, were also clustered together in the UBCG tree. Two clusters were also observed in the phylogenomic tree. One cluster was composed of strains JLT354-W T , JL-351, and KU6B, which were all isolated from the surface seawater, and the second cluster included two strains, L6M1-5 and F15, which were both isolated from the deep-sea sediment. It indicated that the genome-based phylogeny plays a more definitive role in the construction of a natural and objective taxonomy [ 43 ]. Additionally, the comparison of the three key phylogenomic parameters, including average nucleotide identity (ANI, Figure S1A ), average amino acid identity (AAI, Figure S1B ), and digital DNA–DNA hybridization (dDDH, Figure S1C ) values of strain LZ-28 and the type strain of M. alba JLT354-W T were 98.0%, 98.4%, and 84.3%, respectively. All the values were higher than the thresholds (95–96% for ANI, 97% for AAI, and 70% for dDDH) generally accepted for new species delineations [ 43 ]. Accordingly, it clearly confirmed that strain LZ-28 was a new member of M. alba , based on the obtained taxogenomic evidences. 2.6. Comparison of the Core- and Pan-Genomic Profiles among the Selected M. alba Strains The genome sequences of the other nine selected M. alba strains were retrieved from the NCBI database in addition to strain LZ-28. The numbers of accessory and unique genes of the ten M. alba strains were created and shown in Figure 5 A. It shows that 4472 core genes are shared by the ten selected M. alba strains, which account for 80.8% to 90.3% of the genome repertoire of each strain. Additionally, only 0.1% to 5.1% of the unique genes were distributed in individual strains, depending on the varied genomic sizes. Two functional accumulation curves were constructed according to the core- and pan-genome analyses, respectively. The pan genome of the ten M. alba strains was fitted into a power law function with an exponent γ = 0.15 [Fp( x ) = 5160.29 × x 0.15 ], and did not appear to reach saturation even with the increasing genome number. Thus, it indicated that the pan genome was in an open state. This kind of open pan genome often exists in bacterial species residing in multiple ecological environments, and has multiple ways of exchanging genetic material through horizontal gene transfer (HGT) [ 45 ]. The core genome was fitted into an exponential regression [Fc( x ) = 4965.15 + e −0.01 x ]. Based on the two constructed functional curves, the gene number of the pan genome increased when the species number of the M. alba strain accumulated. Meanwhile, the tendency of the core genome, on the contrary, implied that more strains tended to result in additional numbers of unique genes. The phylogenetic trees of the core and pan genomes of the ten selected M. alba were constructed and shown in Figure 6 . It indicates the highly conservative evolution among the M. alba members investigated in this study. The obvious difference of the number of unique genes among the ten selected M. alba strains may be explained by their adaptions to the growth conditions and possible horizontal gene-transfer events [ 45 ]. Moreover, strain LZ-28 showed a close phylogenetic relationship with the marine algae-derived strains, either with strains KD53 and UMTAT08 in the core-genome phylogenetic tree ( Figure 6 A), or with strains KD53 and Ep20 in the pan-genome phylogenetic tree ( Figure 6 B). Thus, it indicates the genome-based phylogenetic analysis provides robust evidence for the evolutionary history of the individual M. alba strain [ 46 ]. 2.7. Comparison of the Functional Classes of Predicted Genes among the Selected M. alba Strains The functions of gene families in the genomes of the ten selected M. alba strains were evaluated by performing the COG and KEGG categories analyses. The COG distribution profile showed that most core ( Figure S2A ) and unique ( Figure S2B ) genes were both mainly related to five functional groups, including carbohydrate transport and metabolism [G], transcription [K], replication, recombination and repair [L], general function prediction only [R], and unknown function [S]. Three strains, including strain KU6B isolated from surface seawater, strain KD53 from marine diatom Phaeodactylum tricornutum [ 12 ], and strain UMTAT08 from marine dinoflagellate Alexandrium tamiyavanichii [ 47 ], harbored the largest core gene number among the ten selected M. alba strains during COG analysis ( Figure S2A ). Moreover, most unique genes were found in strains KU6B and UMTAT08 ( Figure S2B ). With respect to the KEGG assignment, based on the characterization of the core and unique genomes, the genes related to general function, amino acid metabolism, and carbohydrate metabolism, accounted for the major types of KEGG categories. In addition, two strains, UMTAT08 and F15, harbored the largest unique gene number among the ten selected M. alba strains during COG analysis ( Figure S2D ). Additionally, no unique gene sorted in COG and KEGG categories was observed in strain JL351 isolated from the surface seawater [ 41 ]. For the core-gene groups of the COG category, the five marine algae-derived M. alba strains demonstrated a more scattered distribution pattern ( Figure S2A ). However, for the unique gene groups of COG, the principal components analysis (PCA) showed two obviously separated groups among the five marine algae-derived M. alba strains. One group consisted of strains Ep20 and KD53, and the second group included strains LZ-28 and PBVC088. For strain UMTAT08, isolated from A. tamiyavanichii [ 47 ], it was far away from the two clustered groups ( Figure S2B ). In addition, based on the two PCA analyses of the core- ( Figure S2C ) and unique-KEGG ( Figure S2D ) profiles, it was highly noteworthy that strain LZ-28 both showed a functional relationship similar to the other three marine algae-derived M. alba strains, except for strain UMTAT08. Thus, it indicated that the four selected marine algae-derived M. alba strains shared similar functional destinations in spite of their distinctive isolation sources. 2.8. Growth-Promoting Effects of Bacterial Strain LZ-28 and Algal Strain LZT09 Based on the microalgae growth-promoting (MGP) assay [ 21 ], strain LZ-28 demonstrated obvious growth-promoting activity when co-cultured with algal host LZT09 ( Figure 7 A). Interestingly, the mutual promoting effect of the algal culture extract from LZT09 on the bacterial growth of strain LZ-28 was also observed ( Figure 7 B). It indicated that the potential association between bacterial strain LZ-28 and algal strain LZT09 in spite of the detailed nature of action mechanism was still unclear. Furthermore, deeply unearthing the multi-omics information for both sides in co-culture circumstances is believed to offer substantial insights into the detailed mechanism governing these dynamic interactions [ 48 , 49 , 50 , 51 , 52 ]. 2.9. Optimization of Bacterial Growth and EPS Accumulation Bacterial exopolysaccharides (EPSs) were revealed to serve as one essential chemical intermedia within the microscopic phycosphere niches and mediate host–microbe interactions [ 2 , 3 , 4 , 5 , 53 ]. During our previous investigations, several novel bacterial species, which produce bioactive EPSs, were also recovered from the freshwater [ 26 ] and marine phycosphere [ 27 , 28 , 29 , 30 , 31 ], as well as the gut microbiota of Antarctic emperor penguin [ 48 , 54 , 55 ]. In this study, the preliminary experiment showed that temperature and carbon sources were the main two factors influencing the bacterial growth and EPS accumulation. Therefore, ten carbon sources, including cellbiose, fructose, galactose, glucose, glycerol, lactose, maltose, mannose, sucrose, and trehalose, and the pH range of 5.0–9.0 were further used for the optimization of the culture conditions. Based on the obtained result, bacterial incubation at 37 °C promoted bacterial growth and shortened the incubation period of strain LZ-28, although the general trends of bacterial growth that were cultured at 28 °C and 37 °C demonstrated a similar pattern. For the pH tests, strain LZ-28 grew better in the pH values ranging from 6.0 to 9.0. Additionally, strain LZ-28 was found to achieve the fastest growth rate when the cellobiose was used as a carbon source, and cultured at 28°C or at 37 °C and at pH 7.0 ( Figure 8 ). Based on the EPS accumulation measurements, when cultured at 28 °C, the higher EPS yield of strain LZ-28 was achieved both at pH 5.0 and pH 6.0. However, when the culture temperature was changed to 37 °C, the EPS yields were observed to gradually enhance with the increasing pH values, and reached the maximum at pH 9.0. Under the optimized conditions, the highest EPS yield of 47.7 μg/mL was obtained when sucrose (10 g/L) was used as a carbon source and cultured at 37 °C and at pH 9.0 ( Figure 8 ). 2.10. Bioflocculanting-Activity Evaluation and Correlation Analysis Based on the bacterial bioflocculanting assaying, EPS extracted from strain LZ-28 demonstrated obvious bioflocculanting activity, which showed a concentration-dependent manner ( Figure 9 A). The highest efficiency of the bioflocculanting rate of 95.7 ± 8.5% was achieved when 0.60 g·L −1 of EPS was applied. It exhibited a higher bioflocculanting capacity compared to the other bacterial strains, which were also recovered from the marine dinoflagellates previously reported [ 27 , 28 , 29 , 30 , 56 , 57 , 58 , 59 ]. To further infer whether the type and abundance of the monosaccharides of the polymer EPSs were related to the bioflocculanting activity, the characterizations of the monosaccharide compositions of the EPSs and their correlations with the bioflocculanting activity were then performed. Base on the obtained result, the relative portions of two monosaccharides, glucose and fucose, in the crude polysaccharides demonstrated significantly positive highly strong correlations (both p < 0.05) with the bioflocculanting capacities with the correlation coefficients of 0.9 ( p = 0.03734) and 0.8 ( p = 0.03101), respectively, whereas the other four monosaccharides were found to be negatively contributed without statistical significance ( Figure 9 B). Despite these findings, the detailed chemical structures of the EPSs produced by strain LZ-28 remain to be further elucidated. Moreover, more experimental data are still needed to dig out reliable clues to explore the structure–activity relationship of the bacterial EPSs as promising and natural microbial bioflocculants. In addition, the genomic mining also revealed the presence of several typical biosynthesis genes (wzx, exo, and muc) for bacterial EPS biosynthesis in strain LZ-28. Thus, the present study indicated that strain LZ-28 could serve as a new, fresh bacterial candidate with natural potential for the production of versatile EPS bioflocculants derived from marine microalgae with potential pharmaceutical, environmental, and biotechnological implications [ 27 , 28 , 29 , 30 , 55 , 56 , 57 , 58 , 59 , 60 , 61 ]."
} | 6,652 |
31320420 | PMC6639622 | pmc | 9,200 | {
"abstract": "Little information on poly( l -lactic acid) (PLLA) treatment-associated microbiota in thermophilic anaerobic digestion reactors is available. Here, we provide 16S rRNA gene sequence data on microbiota in a thermophilic anaerobic digestion reactor converting PLLA to methane for 336 days. Data comprising 99,566 total high-quality reads were tabulated at the taxonomic class level."
} | 95 |
38404882 | PMC10884847 | pmc | 9,201 | {
"abstract": "Surface modification of electrically neutral hydrophilic polymers is one of the most promising methods for preventing biofouling and biological contamination by proteins and bacteria. Surface modification of inorganic materials such as silica-based glass can render them more durable and thus help in achieving the sustainable development goals. This study reports a novel method for the simple and effective surface modification of glass surfaces with amphiphilic block copolymers possessing the silane coupling segment composed of 3-(methacryloyloxy)propyltris (trimethylsilyloxy) silane and 3-methacryloxypropyltrimethoxysilane. The ability of hydrophilic segments composed of either 2-methacryloyloxyethyl phosphorylcholine (MPC) or poly(ethylene glycol) methyl ether methacrylate (mOEGMA) to prevent bacterial adhesion was investigated. The target block copolymers were prepared by reversible addition-fragmentation chain transfer polymerization and the monomer units of the hydrophilic segments were controlled to be either 120 or 160. The polymers were modified on the substrate by dip-coating. Contact angle measurements indicated that the block copolymer with the PMPC hydrophilic segment formed a hydrophilic surface without pre-hydration, while those with the PmOEGMA hydrophilic segment-coated surface became hydrophilic upon immersion in water. The block copolymer-coated surfaces decreased S. aureus adhesion, and a significant reduction was observed with the MPC-type block copolymer. The following surface design guidelines were thus concluded: (1) the block copolymer is superior to the random copolymer and (2) increasing the hydrophilic segment length further decreases bacterial adhesion.",
"conclusion": "4 Conclusions This study reports the fabrication of bacterial adhesion-suppression coatings from amphiphilic block copolymers. The hydrophilic free chain (PMPC or PmOEGMA) acts as an antibacterial property, and hydrophobic polymer with a silane coupling unit (P(MPTMSi- r -MPTSSi)) works for bonding the substrate to become stable. The monomer units of the hydrophilic segments were controlled to be either 120 or 160 by RAFT polymerization. Characterization of the obtained polymer-coated surfaces indicated that the block copolymer with the hydrophilic PMPC segment formed a hydrophilic surface without prehydration, whereas those with the PmOEGMA hydrophilic segment-coated surface became hydrophilic upon immersion in water. The block-copolymer-coated surfaces successfully decreased S. aureus adhesion, whereas the MPC-type block copolymer resulted in a significant decrease in adhesion. By comparing the polymer structures, the following guidelines to suppress bacterial adhesion were indicated: (1) the block copolymer is superior to the random copolymer, and (2) increasing the hydrophilic segment length further decreases bacterial adhesion. Our amphiphilic block copolymer coating will be useful for silica-based glass in social infrastructure applications for long-term use, which can contribute to solving current global environmental issues for SDGs.",
"introduction": "1 Introduction In light of the current global environmental issues, products with long lifetimes have been gaining significant attention for achieving sustainable development goals (SDGs) [ 1 ]. Most of the currently employed products are in contact with water or are exposed to biological agents (bacteria, viruses, and proteins) as well as organic (nutrients) and inorganic matter (mineral ions), because water dissolves various substances. Notably, bacterial adhesion initiates the formation of biofilms which can cause serious environmental problems [ [2] , [3] , [4] ]. Consequently, inhibiting the biofouling of materials to extend their lifetime while maintaining their original function can assist in achieving the SDGs. One promising method involves the modification of the surface of the material with anti-biofouling polymers so it can be employed in medical devices such as artificial hearts and joints [ [5] , [6] , [7] , [8] , [9] ]. Silica-based glass is an inorganic solid material which has been widely employed in electronic and optical devices in addition to construction materials owing to its lifetime which is much longer than that of organic materials. The glass surface is hydrophilic and negatively charged because the material mostly consists of oxides, and water droplets pass beneath both water-soluble and oil-based contaminants and wash them away. Consequently, hydrophilicty can induce anti-fouling properties [ 10 ]; however, suppressing protein and bacteria biofouling on glass surfaces remains challenging. Several studies have investigated the use of antibiofouling materials on glass surfaces to modify electrically neutral hydrophilic polymer chains [ [11] , [12] , [13] , [14] ]. For example, coatings of polymeric materials comprising poly (polyethylene glycol) (PEG) have been intensively investigated [ 15 , 16 ]. Yoshikawa et al. demonstrated that a poly[poly(ethylene glycol)methyl ether methacrylate] (PPEGMA) brush can prevent biofouling and have blood compatibility [ 17 ]. Zwitterionic polymers, including poly(sulfobetaine methacrylate), poly(carboxybetaine methacrylate), and poly(2-methacryloyloxyethyl phosphorylcholine) (PMPC) have excellent hydrophilicity and anti-biofouling properties [ [18] , [19] , [20] , [21] ]. Surface chemistry enables the modification of hydrophilic polymers on material surfaces [ 22 ]. Controlled radical polymerizations such as surface-initiated atom transfer radical polymerization (SI-ATRP) enables the formation of densely packed polymer brush surfaces with the grafting density over 0.1 chains nm −2 [ 23 ]. Moreover, the protein adsorption in the zwitterionic PMPC brush surfaces decreased significantly with increasing graft density and/or chain length [ 24 ]. In these ways, dense polymer brush surfaces are favorable for obtaining anti-biofouling properties [ 12 , [24] , [25] , [26] ]; however, this approach is expensive and cannot be performed on a large scale. Hence, the “grafting to” method using an anchoring moiety has been gaining attention as a promising method to coat hydrophilic polymers. Our group has previously reported the synthesis of a block copolymer composed of a PMPC segment and hydrophobic silane coupling segment of poly(3-methacryloxypropyl trimethoxysilane- random -3-(methacryloyloxy)propyl-tris(tri(methylsilyloxy))silane) (P(MPTMSi- r -MPTSSi)) [ 27 ]. The P(MPTMSi- r -MPTSSi) segment enabled a stable bonding with the glass surface via silane coupling and the polymer (PMPC- b -P(MPTMSi- r -MPTSSi)) was successfully coated onto glass substrates by simple dip coating. Moreover, the obtained surface exhibited hydrophilic and protein-repellent properties which can suppress bacterial adhesion. Nevertheless, the polymer design strategy requires further investigation. This study reports the synthesis of a polymer coating for the suppression of bacterial adhesion. The polymer was composed of a hydrophilic polymer segment and a hydrophobic silane coupling P (MPTMSi- r -MPTSSi) segment ( Fig. 1 ). The hydrophilic segment was designed to be either PMPC or poly(oligo(ethylene glycol) methyl ether methacrylate) (PmOEGMA), and the segment length was controlled by reversible addition–fragmentation chain transfer (RAFT) polymerization. The surface wettability and antibiofouling properties were investigated by comparing them with those of random copolymers. From these investigations, surface design guidelines for suppressing bacterial adhesion were deduced. Fig. 1 Chemical structure of the amphiphilic block copolymers with the silane coupling segment. The hydrophilic segment was either PMPC or PmOEGMA. The silane coupling segment was P(MPTMSi- r -PMPTSSi). Fig. 1",
"discussion": "3 Results and discussion The target amphiphilic block copolymers were composed of a hydrophilic segment (either PMPC or PmOEGMA) and a hydrophobic-silane coupling segment (P(MPTMSi- r -MPTSSi)). The block copolymers were synthesized by the two-step RAFT polymerization [ 27 ]. The hydrophilic segment (PMPC or PmOEGMA) was firstly polymerized, followed by the second RAFT polymerization of MPTMSi and MPTSSi (the detail shown in Supporting Information). To compare the copolymer structures, ternary random copolymers composed of hydrophilic monomers (MPC or mOEGMA), MPTMSi, and MPTSSi were synthesized via RAFT polymerization. Table 1 summarizes the monomer composition, weight-averaged molecular weight ( M w ), and polydispersity index ( M w / M n , M n: number-averaged molecular weight) of each polymer. The samples were annotated as b (or r) X–Y, where X and Y indicate the monomer unit number of the hydrophilic segment and hydrophilic segment composition, respectively, while “r” and “b” indicate “random copolymer” and “block copolymer,” respectively. The compositions and molecular weight distributions of the obtained copolymers were successfully controlled. In this manner, block and random copolymer structures were designed, and the monomer units in the hydrophilic segment of the block copolymer were controlled to be either 120 or 160. These parameters were investigated to clarify the effects of the polymer coatings on the surface properties and bacterial adhesion suppression. Table 1 Characterization of the amphiphilic block copolymers. Table 1 Code a MPC or mOEGMA/MPTMSi/MPTSSi M w c M w / M n c In feed In copolymer b r120-MPC 120/40/40 112/36/34 9.5 × 10 4 1.6 b120-MPC 120/40/40 114/36/31 8.4 × 10 4 1.5 b160-MPC 160/20/20 158/18/16 8.3 × 10 4 1.5 r120-PEG 120/40/40 147/27/27 2.5 × 10 4 1.3 b120-PEG 120/40/40 119/39/40 1.5 × 10 4 1.3 b160-PEG 160/20/20 155/24/24 2.1 × 10 4 1.2 a The samples are referred to as b (or r)X–Y, where X and Y indicate the monomer unit number of hydrophilic segment, and the hydrophilic segment composition, respectively. “r” and “b” indicate “random copolymer” and “block copolymer”, respectively. b bDetermined by 1 H NMR. c Determined using GPC. The obtained copolymers were coated onto the surfaces of Si/SiO 2 and glass substrates. The thickness of the coating obtained on Si/SiO 2 was quantitatively determined using spectroscopic ellipsometry ( Fig. 2 ). Regardless of the polymer structure, the thickness of the MPC-type copolymer was approximately 5 nm ( Fig. 2 (a)). The block copolymers with the PmOEGMA hydrophilic segment were thicker than those with the PMPC hydrophilic segment ( Fig. 2 (b)). The zwitterionic group can interact with charged surfaces, such as SiO 2 or glass surfaces, via dipole-dipole or ion-dipole interactions [ 28 , 29 ]. Unlike the PEG-type copolymers, the MPC-type copolymers could interact with substrates, which resulted in a several-nanometer-thick coating, regardless of the polymer structure of the MPC copolymers. Thus, nanoscale polymer coatings were successfully prepared. Fig. 2 Thicknesses of the obtained block copolymer-modified surfaces, (a) MPC-type and (b) PEG-type. The data are presented as the mean ± SD ( n = 3). Fig. 2 The hydrophilic properties of a surface play a pivotal role in the potential application of the antifouling coating. Fig. 3 (a) shows the water contact angles in air on the polymer-coated surfaces. For the MPC-type copolymers, the contact angle on the r120-MPC-coated surface was approximately 90° in air, whereas those on the b120-MPC- and b160-MPC-coated surfaces decreased to 30° and 20°, respectively. The r120-MPC-coated surface was hydrophobic because the hydrophobic moiety of MPTSSi was expressed on the surface in air to minimize the surface energy. The block copolymer-coated surfaces were hydrophilic because a sufficiently long PMPC segment covered the surface. For the PEG-type copolymers, the contact angles of the polymer-coated surfaces were approximately 100° regardless of the polymer structure. The PmOEGMA segment is thought to be expressed on the surface; however, the hydrophobic moiety in the ethylene glycol unit affects the hydrophobicity of the surface. The air contact angles of the polymer-coated surfaces in water ( Fig. 3 (b)) revealed that all the polymer-coated surfaces were hydrophilic in water. Importantly, PEG-type block copolymer surfaces exhibited hydrophilic properties in water. Although the surface coated with the amphiphilic copolymers was hydrophilic in air, the hydrophilic moieties in the copolymer were present on the surface in water. Overall, the hydrophilicity in water increased in the order of r120, b120, and b160. Fig. 3 (a) Static contact angles of water in air, (b) static contact angles of air in water, and (c) dynamic contact angles of water of the block copolymer-modified surfaces. For each sample, the left and right bars indicate the advancing and receding contact angles, respectively. The data are presented as the mean ± SD ( n = 3). Fig. 3 To further evaluate the structural changes in air and water, the dynamic contact angles of the surfaces were evaluated ( Fig. 3 (c)). For the MPC-type copolymers, the advancing contact angle decreased in the order of r120-MPC, b120-MPC, and b160-MPC and so did the hysteresis which is the difference between the advancing and receding contact angles. This result indicated that the MPC-type block copolymer-coated surfaces were hydrophilic in air. The PEG-type copolymer-coated surfaces exhibited larger hysteresis compared with those of the MPC-type thus further confirming the structural change in the PEG-type block copolymer-coated surface between air and water. The changes in thickness between air and water were also investigated for the PEG-type copolymers ( Fig. S1 ). The increase in the water thickness indicated that the polymer chains were swollen and extended in water, and consequently the b120-PEG and b160-PEG coated surfaces exhibited large hysteresis and became hydrophilic in water. Subsequently, the adsorption of fibrinogen as a plasma protein was investigated. The copolymer coating successfully decreased fibrinogen adsorption compared to the bare glass ( Fig. 4 ), as expected from the hydrophilic surfaces obtained by coating the amphiphilic copolymers. Among the MPC-type copolymers, fibrinogen adsorption decreased in the following order: r120-MPC b120-MPCand b160-MPC. Among the PEG-type copolymers, fibrinogen adsorption was significantly reduced by coating with b160-PEG. These results indicated that block copolymers with long hydrophilic segments can effectively decrease fibrinogen adsorption. Since the block copolymer coating decreased plasma protein adsorption, we expected it to decrease bacterial adhesion to the substrates. Fig. 4 Amount of the fibrinogen adsorbed on the block copolymer-modified surfaces (* p < 0.05, ** p < 0.01.). The data are presented as the mean ± SD ( n = 3). Fig. 4 Fig. 5 (a) shows the microscopic images and coverage of S. aureus after the contact of the polymer-coated substrates with S. aureus in nutrient-rich TSB. The r120-MPC coating decreased S. aureus adhesion to some extent compared to bare glass. Moreover, MPC-type block copolymer (b120-MPC and b160-MPC) coatings dramatically suppressed S. aureus adhesion ( Fig. 5 (b)). The block copolymer also more effectively decreased the S. aureus adhesion than r120-PEG. Although the PEG-type block copolymer-coated surfaces were hydrophobic in air, they exhibited hydrophilic properties owing to the extension of the polymer chain in water, which enabled bacterial adhesion suppression. Bacterial adhesion further decreased upon increasing the hydrophilic segment length from b120-PEG to b160-PEG. This result is consistent with the decrease in fibrinogen adsorption ( Fig. 4 ). From these investigations, the following surface design guidelines for bacterial adhesion suppression were obtained: (1) the block copolymer is superior to the random copolymer and (2) increasing the hydrophilic segment length further decreases bacterial adhesion. Fig. 5 (a) The microscope images of the adhered S. aureus on the block copolymer-modified surfaces incubated in TSB. Scale: 200 μm. (b) Bacterial coverage on the block copolymer-modified surfaces (* p < 0.05, ** p < 0.01.). The data are presented as the mean ± SD ( n = 3). Fig. 5 We further investigated the anti-adhesive properties of bacteria on the polymer-coated surfaces. Fig. 6 shows the SEM images of adhered S. aureus , which was incubated in TSB at 37 °C for 24 h. A large number of S. aureus adhered and formed multiple layers on the bare glass substrate; however, the bacterial adhesion dramatically decreased on the polymer-coated substrates even after incubation for 24 h. On the random copolymer-coated surface, S. aureus partially adhered and formed a multilayer (indicated by the red box in Fig. 6 ), while the adhesion of S. aureus was a monolayer on the block copolymer-coated surfaces. Thus, the block copolymer coatings successfully exhibited superior antibacterial properties. Fig. 6 SEM images of the adhered S. aureus on the MPC-type block copolymer-modified surfaces. The incubation was proceeded at 37 °C for 24 h. Scale: 5 μm. Fig. 6"
} | 4,270 |
29050374 | PMC7059795 | pmc | 9,202 | {
"abstract": "Abstract The root microbes play pivotal roles in plant productivity, nutrient uptakes, and disease resistance. The root microbial community structure has been extensively investigated by 16S/18S/ITS amplicons and metagenomic sequencing in crops and model plants. However, the functional associations between root microbes and host plant growth are poorly understood. This work investigates the root bacterial community of foxtail millet ( Setaria italica ) and its potential effects on host plant productivity. We determined the bacterial composition of 2882 samples from foxtail millet rhizoplane, rhizosphere and corresponding bulk soils from 2 well-separated geographic locations by 16S rRNA gene amplicon sequencing. We identified 16 109 operational taxonomic units (OTUs), and defined 187 OTUs as shared rhizoplane core OTUs. The β-diversity analysis revealed that microhabitat was the major factor shaping foxtail millet root bacterial community, followed by geographic locations. Large-scale association analysis identified the potential beneficial bacteria correlated with plant high productivity. Besides, the functional prediction revealed specific pathways enriched in foxtail millet rhizoplane bacterial community. We systematically described the root bacterial community structure of foxtail millet and found its core rhizoplane bacterial members. Our results demonstrated that host plants enrich specific bacteria and functions in the rhizoplane. The potentially beneficial bacteria may serve as a valuable knowledge foundation for bio-fertilizer development in agriculture.",
"introduction": "Introduction The root surface sets the environment for complex interactions among soil, the host plant, and microbes [ 1 ]. The root microbiota (rhizosphere, rhizoplane, and endophytic bacteria) mainly derive from surrounding soil and are influenced by geographical locations, nutrient status, and host genotype [ 1 – 9 ]. These microbes play pivotal roles in plant productivity, nutrient uptakes, and disease resistance. Observations from multiple research teams imply that plants selectively “cultivate” specific and potentially beneficial microbes through root exudates and deposits, which act as a carbon source and nutrients for microbial growth, as well as altering soil pH structure [ 10 , 11 ]. However, how the root microbiota influence plant growth and yield remains largely unknown. Comprehensive association studies between the root microbiota and crop traits are needed to identify beneficial or harmful microbes. Foxtail millet ( Setaria italica ) is an important crop in arid and semiarid regions due to its water use efficiency and drought tolerance [ 12 ]. It is crucial to understand the genetic and environmental factors of foxtail millet; 2540 foxtail millet cultivars with the phenotypic traits and genomes were deposited in the China National Gene Bank-Shenzhen [ 13 – 15 ].These collections facilitate a deep understanding of how the genotypes and root microbiota affect the foxtail millet growth, development, and yield. In this study, we sequenced the rhizosphere and rhizoplane bacterial microbiota in 1219 foxtail millet cultivars, which were grown in 2 far-separated fields in China, Yangling and Zhangjiakou. We evaluated the effects of geographic location and microhabitat on root bacterial communities and predicted root bacterial functions according to their taxonomy. We performed systematic association analysis between rhizoplane bacteria and foxtail millet productivity traits and identified specific bacterial taxa correlated with host plant productivity. Our work provides a basis for potential agricultural improvement of foxtail millet by root microbiota modification. Data description The root microbiota of foxtail millet remains largely unknown. In this study, we collected rhizosphere, rhizoplane, and corresponding unplanted bulk soil samples from the foxtail millet cultivars in 2 well-separated locations in China. We recorded 12 foxtail millet traits related to growth and productivity (Tables S1 and S2). We sequenced the V4-V5 region of the 16S rRNA gene for 2882 samples, which yielded 98 750 591 high-quality reads for subsequent analysis, an average of 34 264 sequences per sample. After discarding low-abundance and non-bacterial operational taxonomic units (OTUs), we obtained 16 109 OTUs, 2998 OTUs per sample (Table S3), with 97% similarity of the entire V4-V5 region; 81.3% of the OTUs could be assigned to 34 bacterial phyla, mainly including Acidobacteria, Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria. In total, 624 genera belonging to 254 families were recorded from these samples. Rhizoplane and rhizosphere contained a large number of total OTUs, and only 30 and 26 OTUs per sample showed relative abundances higher than 0.5% on average. Bulk soil samples contained the lowest number of total OTUs because of the small sample size. A summary of the data is in Table 1 . Taxa at the phylum level of rhizosphere and rhizoplane microbiota are shown in Fig. 1 and Table S4. The sequencing data have been deposited in the European Nucleotide Archive (accession PRJEB16061). Figure 1: Dominant bacterial phyla detected in foxtail millet root compartments and bulk soils. BS: bulk soil; RP: rhizoplane; RS: rhizosphere; YL: Yangling; ZJK: Zhangjiakou. Table 1: Summary of samples and OTUs assigned to different taxonomic levels % of OTUs assigned to Sample group Sample number Average reads/sample Average OTUs/sample No. of OTUs Genus Family Phylum BS.YL 8 23 473 2980 8132 59.7 68.8 92.2 RS.YL 1219 33 373 3333 15 676 62.2 78.8 92.8 RP.YL 1219 36 396 2479 15 110 54.1 88.4 97.7 BS.ZJK 8 54 756 4414 9426 56.7 64.3 88.6 RS.ZJK 214 26 218 3555 14 696 63.7 73.8 92.8 RP.ZJK 214 34 879 3433 14 558 65.6 80.1 95.3 Total 2882 34 264 2998 16 109 YL: Yangling; ZJK: Zhangjiakou; BS: bulk soil; RS: rhizosphere; RP: rhizoplane.",
"discussion": "Discussion We defined the taxonomic structure of the foxtail millet root microbiota, mainly comprising Acidobacteria, Actinobacteria, Bacteroidetes, Firmicutes, and Proteobacteria. Bacterial alpha diversity decreased from foxtail millet rhizosphere to rhizoplane microbiota. These findings were consistent with reports in other plant species such as Arabidopsis thaliana [ 6 , 8 ], maize [ 2 ], rice [ 3 ], barley [ 4 ], grapevine [ 36 ], and soybean [ 37 ], indicating that the foxtail millet root bacteria follow the general rule of microbiota establishment. In this study, both CAP and principal coordinates analysis (PCoA) demonstrated that microhabitat was the major factor affecting the bacterial microbiome, rather than geographic locations. Our result was consistent with the findings reported by Lundberg et al. [ 6 ], in which the microhabitat was the most important factor driving root bacterial community composition. However, our result was different from the field study of rice, which suggested that the geographic location was a larger source of variation than soil structure and might be a major determining factor shaping the composition of the root microbiome in a setting where the distance between planted locations was up to ∼125 km [ 3 ]. Although the distance between Yangling and Zhangjiakou was about 1000 km and the soil types of the 2 fields were obviously different (the soil in the Yangling cropping field was loessal soil, and in Zhangjiakou it was cinnamon soil; this information could be found in a database called Sciences Database of the Chinese Academy of Science [ 38 ]), the largest source of the variation of microbial structure was attributed to the rhizospheric compartmentalization. Our result indicated that the foxtail millet was the major factor that drove the assembly of the root-associated microbiome. We analyzed the association between grain weight of foxtail millet and the corresponding rhizoplane bacterial microbiota using methods similar to those applied in human-associated studies [ 39 , 40 ]. Through this process, we found that grain weight of foxtail millet could be predicted with explicable variance of 31% (in test data, 29.3%) using abundance of only 75 bacterial OTUs as information. Although 31% was not a high measure of association, it may be considered a reasonable and significant result as it is widely accepted that apart from environmental factors like biological factors and abiotic factors, crop productivity is mainly influenced by plant genotype. Since crop productivity is associated with multiple genes, it is more difficult to identify those key genes/single nucleotide polymorphisms by using genome-wide association study (GWAS), compared to some traits that are associated with a single gene. So far there are no publications showing how the genes determine the foxtail millet yield. The importance of the root-associated microbial community, as the major representative of biological factors, for plant growth and development has been widely recognized [ 41 , 42 ]. Of the positive markers, 3 main positive genera markers comprised of Bacillus, Falsibacillus , and Paenibacillus , in which many bacteria were reported with a characteristic that functions as a biocontrol against soil-borne pathogens or with the capability of secreting auxin to promote plant growth [ 43 , 44 ]. Interestingly, we found that positive marker OTUs showed a strong and complex correlation network, while negative marker OTUs showed a more loose and simple network (Fig. 5 ). In addition to yield traits, most of these marker OTUs show a positive correlation with growth traits such as main stem width, and top second leaf width (Fig. S3). Our results indicate that cooperative microbial interactions may play critical roles in microbial assembly of the plant microbiome and may benefit plant growth and development. We found that a large number metabolic pathways related to nutrient uptake, environmental responses, and density control were enriched in the microbiota of the root surface, while many of metabolic pathways related to carbon fixation and amino acid synthesis were enriched in the microbiota of the rhizosphere (Fig. 6 ). The roots of plants excrete 10–44% photosynthetically fixed carbon, which may serve as an energy source, signaling molecules or antimicrobials for soil microorganisms [ 45 ], fitting the hypothesis that plants actively shape their root microbiome by root exudates and other rhizodeposits stimulating and/or inhibiting various microbes. Taken together, our work has systematically characterized the root bacterial microbiota of foxtail millet and identified potential beneficial root bacterial and genomic pathways, which will provide a basis for the application of beneficial root bacteria for agricultural improvement. So, isolation and functional verification of root microbes is necessary for future work, especially for marker species. On the other hand, we will perform metagenomic sequencing of root samples, which will provide more precise qualitative and quantitative functional information of the root microbes than predicted data. It will help further our understanding of this microecosystem in the root zone. Potential implications In this work, we characterized bacterial OTUs’ composition in the root zone using more than 1000 foxtail millet cultivars from 2 well-separated geographic locations. The large sample size allowed us to capture the bacterial diversity in rhizosphere and rhizoplane compartments and assess the existence of a core rhizoplane microbiome for foxtail millet. The data collection serves as an important basis for studying the association of microbial organisms in the foxtail millet root zone with host phenotypes. Large-scale association analysis identified the potential beneficial bacteria correlated with plant high productivity using methods similar to those applied in human-associated studies [ 39 , 40 ]. The methods for deciphering the association between root microbiome and plant phenotype are applicable to other crops. A better understanding of the various interactions between microbes and the host plant may serve as a valuable knowledge foundation for bio-fertilizer development in sustainable agriculture. Isolation and functional identification of the potentially beneficial bacteria are necessary for future work. Metagenomics combined with quantitative functional genomic approaches like transcriptomics, proteomics, and metabolomics will provide deep insights into microbial interactions and help us fully understand the microecosystem in the root zone."
} | 3,126 |
34028885 | null | s2 | 9,203 | {
"abstract": "A wide portfolio of advanced programmable materials and structures has been developed for biological applications in the last two decades. Particularly, due to their unique properties, semiconducting materials have been utilized in areas of biocomputing, implantable electronics, and healthcare. As a new concept of such programmable material design, biointerfaces based on inorganic semiconducting materials as substrates introduce unconventional paths for bioinformatics and biosensing. In particular, understanding how the properties of a substrate can alter microbial biofilm behavior enables researchers to better characterize and thus create programmable biointerfaces with necessary characteristics on demand. Herein, the current status of advanced microorganism-inorganic biointerfaces is summarized along with types of responses that can be observed in such hybrid systems. This work identifies promising inorganic material types along with target microorganisms that will be critical for future research on programmable biointerfacial structures."
} | 264 |
38579990 | PMC11067327 | pmc | 9,204 | {
"abstract": "Cyanobacteria harvest light by using architecturally complex, soluble, light-harvesting complexes known as phycobilisomes (PBSs). PBS diversity includes specialized subunit paralogs that are tuned to specific regions of the light spectrum; some cyanobacterial lineages can even absorb far-red light. In a recent issue of the Journal of Biological Chemistry , Gisriel et al. reported the cryo-electron microscopic structure of a far-red PBS core, showing how bilin binding in the α-subunits of allophycocyanin paralogs can modify the bilin-binding site to red shift the absorbance spectrum. This work helps explain how cyanobacteria can grow in environments where most of the visible light has been filtered out."
} | 178 |
33151998 | PMC7644084 | pmc | 9,209 | {
"abstract": "The implementation and monitoring of management strategies is integral to protect coastal marshes from increased inundation and submergence under sea-level rise. Sediment addition is one such strategy in which sediment is added to marshes to raise relative elevations, decrease tidal inundation, and enhance ecosystem processes. This study looked at the plant and invertebrate community responses over 12 months following a sediment addition project on a salt marsh located in an urbanized estuary in southern California, USA. This salt marsh is experiencing local subsidence, is sediment-limited from landscape modifications, has resident protected species, and is at-risk of submergence from sea-level rise. Abiotic measurements, invertebrate cores, and plant parameters were analyzed before and after sediment application in a before-after-control-impact (BACI) design. Immediately following the sediment application, plant cover and invertebrate abundance decreased significantly, with smothering of existing vegetation communities without regrowth, presumably creating resulting harsh abiotic conditions. At six months after the sediment application treatment, Salicornia bigelovii minimally colonized the sediment application area, and Spartina foliosa spread vegetatively from the edges of the marsh; however, at 12 months following sediment application overall plant recovery was still minimal. Community composition of infaunal invertebrates shifted from a dominance of marsh-associated groups like oligochaetes and polychaetes to more terrestrial and more mobile dispersers like insect larvae. In contrast to other studies, such as those with high organic deposition, that showed vegetation and invertebrate community recovery within one year of sediment application, our results indicated a much slower recovery following a sediment addition of 32 cm which resulted in a supratidal elevation with an average of 1.62 m (NAVD88) at our sampling locations. Our results indicate that the site did not recover after one year and that recovery may take longer which illustrates the importance of long-term monitoring to fully understand restoration trajectories and inform adaptive management. Testing and monitoring sea-level rise adaptation strategies like sediment addition for salt marshes is important to prevent the loss of important coastal ecosystems.",
"conclusion": "Conclusions Sediment addition projects are a unique approach to restoring and building marsh elevations. The impacts from these projects will vary depending on initial marsh bio-geomorphic properties, the depth of application, and the composition of sediment used. Considering our measured negative effects of this deeper sediment addition on vegetation and invertebrate communities in the first twelve months, continued monitoring and the identification of adaptive management strategies is important. Deeper sediment application projects on the Pacific Coast, USA and other areas may be necessary especially for those marshes located in urban estuaries where SLR adaptation strategies are limited. Urbanized estuaries with marshes around the world are at risk of submergence from SLR due to sediment starvation, human development, and the interaction of these complex processes [ 27 , 72 ]. Without management actions, many of these marshes may be converted to mudflat and subtidal habitats as SLR rates outpace accretion (e.g. [ 27 ]). This study provided a unique, large-scale opportunity to explore the short-term impacts of sediment addition on important ecosystem features and processes. We measured a relatively slow trajectory of site recovery when compared with other regional systems and thinner application methods. Understanding that the recovery of the marsh is not occurring within 12 months is crucial for planning for effective restoration and monitoring programs in the future. Also, studying the effects of sediment application on a range of ecological settings will allow for a more complete understanding of its utilization as a management strategy for wetlands threatened by SLR. Evaluation of management strategies such as sediment application at various time scales with thorough pre- and post-monitoring is important to help design resilience strategies in the region and for subsiding wetlands in urban regions moving forward.",
"introduction": "Introduction Coastal wetlands, including salt marshes, are among the most productive ecosystems in the world and are recognized for their ecological functions including habitat provision, water filtration, flood abatement, and fishery and biodiversity support [ 1 , 2 ]. Despite their importance, over half of global wetlands have already been lost as a result of human activities [ 2 , 3 ], and remaining coastal wetlands are threatened by climate change, specifically sea-level rise (SLR) [ 4 ]. Under moderate SLR scenarios, up to 59% of the remaining global coastal wetlands could be lost by 2100 [ 5 ]. Under conditions of sufficient sediment supply and accommodation space, feedbacks between biological and physical processes control accretion, allowing the salt marshes to maintain their elevation relative to tidal inundation and increasing SLR rates [ 6 – 13 ]. However, today, many coastal salt marshes are unable to build their elevations enough to outpace accelerated rates of SLR. This is especially key for those marshes that are located in modified landscapes that are experiencing subsidence and/or alterations of sediment dynamics, such as decoupling rivers from estuarine systems and river channelization [ 14 , 15 ]. Additionally, many intensely urbanized estuaries are surrounded by developed land that limits the ability of salt marshes to migrate landward [ 16 ]. For these marshes, management strategies have been developed to minimize their loss due to SLR. One such strategy to artificially increase salt marsh elevations is sediment addition, a process that mechanically adds a layer of sediment to increase surface elevations, reducing overall inundation and often increasing primary productivity [ 17 ]. Impacts and results of sediment addition have been studied in other regions of the world as well as marshes on the East and Gulf Coasts of the United States [ 18 – 25 ]. Many of these sediment application studies have been conducted in microtidal deltaic salt marsh systems where there is some available natural sediment supply. Fewer studies have been conducted in urbanized, sediment-limited salt marsh ecosystems. To date, only a few limited studies have analyzed a thin-layer sediment addition in salt marshes on the Pacific Coast of the United States [ 26 , 27 ], making this research timely. Our study highlights the plant and invertebrate community change under a deeper sediment application project with an elevation goal designed to offset high SLR risk and local subsidence. For coastlines experiencing higher relative SLR rates, there is a sense of urgency to develop effective adaptation strategies to prevent salt marsh loss. Coastal salt marshes have been projected to become submerged over the coming century often due to limited sediment supply or landscape modifications that prohibit upland migration [ 13 , 15 , 28 – 30 ]. These at-risk ecosystems provide an opportunity to implement and test SLR adaptation strategies to prevent loss of these ecosystems. Thus, additional studies are needed to assess the effects of large-scale sediment addition projects that deposit thicker sediments to offset elevation deficits in locations that are experiencing higher SLR rates. Monitoring and studies of sediment application projects often focus on elevation change and vegetation but do not monitor other key ecosystem components including the invertebrate communities. Yet, the abundance, diversity, and community structure of invertebrate communities (polychaetes, oligochaetes, amphipods etc.) can serve as indicators for marsh health [ 31 ]. Earlier studies on subsiding coastal marshes in Louisiana [ 20 , 32 ] found that moderate levels of sediment addition (12–14 cm) could allow for the macro-invertebrate community to recover to pre-sediment application conditions; however, too much sediment (~17 cm) could impair recovery. Despite the recognized importance of invertebrates, few studies have experimentally studied the short-term responses of belowground invertebrates, including micro-invertebrates, to plant disturbance and thin-layer sediment addition. Given that other studies had seen recovery of plants within a one to two-year time frame (as summarized in Raposa et al. 2020 [ 33 ]), the objectives of this study were to monitor the shorter-term effects (within 12 months) of large scale deeper sediment addition project on (1) the vegetation community (percent cover, community composition), (2) the benthic invertebrate community (abundance, species richness, diversity and community composition) and (3) associated abiotic parameters (elevation, porewater salinity, temperature and light intensity). We hypothesized that all plant and invertebrate parameters would decrease following sediment application. However, we predicted that the plant community would recover but on a longer time scale as seen in non-sediment application restoration studies and that this would facilitate recovery of invertebrate communities.",
"discussion": "Discussion In the short-term (12 months), the addition of a deep sediment layer had dramatic negative effect on both vegetation and invertebrate communities in this marsh ecosystem. The lack of vegetation at 12 months’ post-sediment application as well as the significant decline in invertebrate abundance and alteration of the invertebrate community composition is likely due to a combination of interrelated factors and processes: the thickness of the sediment applied, the resulting supratidal elevation, harsh abiotic parameters, dispersal inhibition, and the characteristics of the dredge material. Given the relationship between plant cover and sediment-dwelling organisms in tidal marshes, we hypothesize that the slow recovery of invertebrates was related to the lack of plant cover (e.g. [ 51 ]). Thickness of sediment applied Sediment addition experiments in marshes have been used to raise elevations and restore plant communities. Most of the ecosystems studied for this work were in the deltaic microtidal marshes of Louisiana, USA [ 17 – 19 , 52 – 55 ] and barrier marshes in North Carolina, USA [ 21 ], with a goal of sediment addition for restoration due to large-scale disturbances, such as marsh dieback, drought or regional subsidence. These studies used sediment depths ranging from 2–10 cm and found that, within a short timeframe (averaging one year), plant cover, plant productivity, aboveground biomass, and soil mineral matter all quickly increased following sediment application. Additional studies that monitored change over seven years indicated similar improvements in deltaic marshes post-application [ 20 ], and fourteen years in salt marsh islands of New York, USA [ 11 ], indicating that time scale of recovery is variable even within similar marsh types. Overall, thinner sediment applications may damage plants but are unlikely to smother existing plant communities while thicker sediment applications seem to smother and significantly impact plant communities, resulting in slower vegetation recovery due to a need for plants to recolonize. Specifically, in a submerging coastal marsh in the Mississippi River Delta region of coastal Louisiana, USA, Ford et al. [ 19 ] found that a thin-layer sediment addition (2.3 cm) knocked down plants initially but the plant community recovered after 12 months. Thicker applications (e.g. 10–38 cm of sediment) studied by Cahoon and Cowan [ 54 ] smothered and killed most vegetation with recovery via recolonization just starting at 14 months after the sediment application. Thus, if restoration designs call for deeper sediment addition, slower plant community recovery is expected without active restoration strategies like replanting (as discussed later). Previous studies have shown the impacts of adding thin-layer sediment slurry on marsh plant communities, but more research is needed to investigate thicker sediment applications, such as those used in this study (32 cm). Non-deltaic, rapidly submerging marshes likely require deeper depths of applied sediment to prevent SLR submergence. The recommended application depth will vary with management concerns, predicted SLR, and existing marsh elevations. It has generally been postulated that sediment additions that are less than 15 cm can be recolonized by re-sprouting of existing marsh vegetation, while additions that are greater than 15 cm are recolonized by new plants colonizing the area [ 56 ]. In our study with a sediment application depth of 32 cm, there is no evidence that the same plants that existed pre-sediment application grew through the dredge sediment [ 19 , 21 ]. This depth of sediment application combined with resultant harsh abiotic properties presumably prevented the re-emergence of existing vegetation and slowed recovery. Supratidal elevation Similarly, prior studies conducted in deltaic microtidal salt marshes rarely raised the marsh platform above the tidal range. One such study, Stagg and Mendelssohn (2011), found that salt marshes subsidized with sediment to elevations 2–11 cm above MSL in deltaic marshes of coastal Louisiana, USA experienced less flooding stress, but sediment addition greater than 11 cm above MSL led to decreased platform stability [ 52 ]. In our study, sediment application resulted in a reduced inundation depth and less frequent inundation regime whereas the control site experienced a greater percentage of time under water in the same year [ 39 ]. Pre-sediment application, the experimental and control sites were inundated at most tide heights greater than approximately 0.46 m and 0.57 m above MSL, respectively, which was calculated to be 5.2 and 7.9% of the time [ 39 ]. Post-sediment application, the control site was inundated by tides greater than 0.36 m, which was 9.2% of the time while the experimental site was inundated only by tides higher than 0.48 m, which was calculated to be 2.3% of the time [ 39 ]. The sediment application raised the marsh elevation from intertidal to supratidal, which likely contributed to the slow establishment of the plants and invertebrate communities when compared to other short term studies [ 39 ]. Supratidal marshes occur above the intertidal zone and don’t receive daily tidal inundation but are flooded only during spring high tides and other high water events which can directly influences abiotic factors and the type of plants and invertebrates that develop [ 50 ]. While increasing marsh platform elevations was the ultimate purpose for this project and may be necessary to prevent submergence due to SLR, the short-term ecological implications of such a substantial elevation change had several negative outcomes for vegetation and invertebrate recovery. Colonization This study was undertaken knowing that post-sediment application elevations may place the marsh surface at supratidal elevations. Supratidal elevations can limit colonization of vegetation and invertebrates. For plants the literature demonstrates that recovery is governed by optimal inundation and hydroperiod, high organic matter content concurrent with appropriate elevation, and rhizome survivability following burial (e.g. [ 50 , 57 , 58 ]). Spartina foliosa is found at the SBNWR at elevations up to 0.71 m above MSL [ 39 ]. Post-sediment application elevations range from 0.86 below MSL to 1.27 m above MSL [ 39 ], and it is therefore possible that certain areas on the sediment application site may not be suitable for the growth of Spartina foliosa and are above the optimal elevation range from some studies (e.g. [ 9 ]) surface vegetation. Although there is no evidence that within one year of sediment application Spartina foliosa or other high marsh plants colonizing the sediment application site, at 6 MAT newly colonized Salicornia bigelovii was just beginning to colonize throughout the site. This annual succulent species is an opportunistic colonizer that has a broad ecological tolerance and was not characteristic of the plant community before sediment application took place [ 59 ]. Long-lived perennials that are more reflective of the pre-sediment application community ( Salicornia pacifica , Spartina foliosa , and Batis maritima) may take longer to return to the site and persist because seed delivery may be limited by the decreased flooding regime. The rate at which invertebrates colonize an area is also influenced by the hydroperiod [ 57 ]. Restored marshes at higher elevations were found to have significantly fewer invertebrates than those at lower elevations when the elevation differed by 20 cm [ 60 ]. The invertebrates that were present are predominantly the larvae and pupae of mobile, algal-feeding insects, which are frequently early colonizers in marshes sites (e.g. [ 43 – 45 , 50 , 51 ]). The most abundant invertebrates (oligochaetes and polychaetes) found in the control site were almost entirely absent from the sediment application site. Several species of polychaetes exhibit a planktonic larval stage and may have been unable to recruit due to the decreased tidal inundation at the experiment site [ 57 ]. Delivery of both oligochaete and polychaete juveniles via sea grass rafts and algal mats has also been described as a mechanism for larval dispersal [ 50 ]. Rafted plant material was not encountered on the sediment application site, suggesting that elevation may also be limiting delivery of invertebrates by this mechanism. Once tidal inundation periods become longer and more frequent on the sediment application site due to further subsidence and SLR, opportunistic species like members of Capitellidae will likely begin to increase in abundance due to increased larval recruitment [ 61 ]. The initial decline in invertebrate abundance following sediment application was likely due not only to the smothering of existing communities and harsh abiotic environment conditions [ 19 ], but also the slow recovery of plants as trophic support and refuge to the site. Decaying plant matter serves as refugia for oligochaetes, and their root mats loosen the soil making it easier for oligochaetes to burrow, thus the absence of plants diminishes suitable habitat [ 60 ]. Over time, as vegetation develops in the marsh the invertebrate community generally begins to reflect communities more comparable to natural marshes. In restored Spartina foliosa marshes, it can take eight to ten years for the development of infaunal communities that resemble those of natural marshes [ 45 , 62 ]. Changes in abiotic factors Further limiting the colonization and influencing the community structure of plants and invertebrates is the high porewater salinity and temperature throughout the sediment application site. While we did not measure evaporation directly, the infrequent inundation combined with increased light intensity and temperature due to the absence of plants presumably resulted in increased evaporation ( Fig 3 and S3 Table ). Prior studies in northeastern USA [ 63 ], southern California, USA [ 51 , 64 ], Georgia in the southern USA [ 65 ], and in a coastal lagoon in central Argentina [ 66 ] have demonstrated that in the absence of shading, higher soil temperature, increased porewater salinities, and lower sediment pore water content can result and are most likely due to the increased sun exposure and subsequent evaporation. Lack of plant cover has been shown to impact infauna communities by altering the light and evaporation regime, decreasing species richness, increasing the proportion of insects, and decreasing the proportion of oligochaetes and polychaetes [ 51 ]. In addition to sediment application depth, the composition of the dredge material in this study may also have contributed to the lack of short-term recovery. Substrate conditions, both physical (e.g. grain size) and biochemical (e.g. redox potential), directly affect the growth and colonization of plants and invertebrates in marshes [ 57 , 67 , 68 ] and as reviewed in [ 62 ]. The dredged material used in this study was coarse-grained and low in organic matter. Initial increases in bulk density, indicating higher percentage sand, have been observed in previous sediment application studies (e.g. [ 19 ]). In this study, however, at 12 MAT, soil bulk density and percent organic matter had returned to or exceeded levels measured prior to sediment application despite percent silt and clay content remaining low ( Fig 2 ). The discrepancies in sediment parameters between the control and post-sediment application in our study was greater than other studies and likely contributed to slower recovery of other parameters that we observed. Coarse-grained sediments drain water quickly creating aerobic conditions that lead to high rates of decomposition and low organic matter content [ 57 ]. Since organic matter is the main source of soil nutrients used for plant growth and serves as a major food source for invertebrates, coarse-grained sediment with low organic matter does not provide conditions conducive for salt marsh plants and invertebrates. Additionally, the faster drainage of coarse sediments leads to higher evaporation rates which contributed to the increase in salinity in this study ( Fig 3 and S3 Table ). These properties of sandy, salty sediment on the sediment application site are likely contributing to the delayed establishment of vegetation and consequently invertebrates in this marsh. A previous study of coarse-grained sandy sediment application to a predominantly sandy-mud back barrier North Carolina, USA marsh found that after two years, new plants were able to establish and grow to reflect pre-sediment application parameters after silty sediments were incorporated into the augmented sediment through natural marsh processes [ 21 ]. Over time, through bioturbation and sedimentation processes, natural sediments with higher silt and clay contents are expected to accumulate throughout the site, leading to increased nutrient levels and organic matter content which provide more favorable conditions for plants to continue to grow throughout the site. Future work and lessons learned The overarching goal of this project was to build elevation capital to combat SLR submergence by using a thicker sediment addition method, and we will continue monitoring for additional years to monitor recovery. Given the relationships between sediment application depth and the delayed plant and invertebrate recovery (e.g. [ 17 , 52 ]), a thinner application of sediment may have expedited recovery; however, then the resultant elevation gain may not have been sufficient to combat SLR submergence for this site. Adjusting monitoring expectations to reflect a longer time frame for recovery may be important when assessing trade-offs in these high risk urbanized estuaries with limited sediment supply. Longer-term monitoring programs to fully understand recovery trajectories are important to invest in and implement for deep sediment application projects. Adaptive management strategies to accelerate plant and invertebrate recovery time may be an important approach if the community does not recover at the pace expected or needed. There is evidence suggesting that the succession of a restored marsh relies on the overall amount of plant cover, so developing planting strategies for the site may help speed up recovery [ 42 , 57 ]. This has also been shown to be effective for long-term recovery following other thin-layer sediment addition projects [ 69 ]. Implementing a planting strategy using polyculture plots in salt marshes has been shown to increase plant cover by 80–100% in one year and increase species richness and canopy complexity [ 70 ]. Soil amendments such as nitrogen fertilization have been suggested for enhancing plant growth; however, it was found that despite immediately stimulating plant growth, nitrogen is not retained in the system and enhanced growth is not sustained once fertilization ends [ 71 ]. Rather, amending the substrate by importing finer marsh soils that incorporates rhizomes, roots, and microorganisms to the site is a preferred method to enhance plant establishment and growth [ 57 , 58 ]. One or more of these active restoration strategies can be considered as an important adaptive management restoration tool to achieve management goals and speed functional recovery following sediment application and will be considered for this project."
} | 6,172 |
27091306 | PMC4835754 | pmc | 9,212 | {
"abstract": "This work presents a novel coating technique to manufacture ceramic superhydrophobic coatings rapidly and economically. A rare earth oxide (REO) was selected as the coating material due to its hydrophobic nature, chemical inertness, high temperature stability, and good mechanical properties, and deposited on stainless steel substrates by solution precursor plasma spray (SPPS). The effects of various spraying conditions including standoff distance, torch power, number of torch passes, types of solvent and plasma velocity were investigated. The as-sprayed coating demonstrated a hierarchically structured surface topography, which closely resembles superhydrophobic surfaces found in nature. The water contact angle on the SPPS superhydrophobic coating was up to 65% higher than on smooth REO surfaces.",
"conclusion": "Conclusions In summary, we present a promising technique to fabricate superhydrophobic coatings using precursor solutions as feedstock in a plasma spray deposition process. It offers a fast, simple and low cost method to produce large area hydrophobic surfaces on a variety of substrates. The superhydrophobicity of the SPPS coatings results from the combination of the hydrophobic material and a hierarchically structured coating topography, which is similar to superhydrophobic surfaces found in nature.",
"discussion": "Results and Discussion Effect of standoff distance From the results for conditions 1, 3, and 5 the effect of standoff distance (SD) can be investigated. Figure 2b–d show the cross sectional microstructures of coatings deposited with experimental conditions 1, 3, and 5 respectively and Fig. 2a shows the substrate temperature history for these three conditions. The coatings are porous and rough for all three conditions. Particles observed in the coatings have irregular shapes which is an indication of incomplete melting. This suggests that the coatings were formed mainly by the sintering of incompletely melted particles and aggregates of fully/partially decomposed precipitates from the droplets. Nano-particles were also observed in the coatings. Individual particles of this size would not have sufficient inertia to penetrate the gas boundary layer at the surface of the substrate. The thermophoresis force may have allowed these particles to pass through the boundary layer, or the nano-particles may have formed on the substrate from the condensation of vaporized material. When the standoff distance was increased while all the other parameters were held constant, an increase in the coating porosity, and decreases in the coating thickness and substrate temperature were observed. At long standoff distances, the plasma plume is cooled by the ambient air, resulting in cooler feedstock particles arriving at the substrate. The gas and particle velocities are also reduced at longer standoff distances. The combination of lower momentum and lower feedstock temperature at long standoff distances decreased the adhesion of the feedstock particles when they arrive at the substrate, which resulted in a reduction in the coating thickness and deposition efficiency. This agreed with the SEM images of deposits collected after a single torch pass, that show less material was deposited under condition 5 compared to conditions 1 and 3 (see Supplementary Fig. S2 ). The lower feedstock temperature at longer standoff distances also explains the increase in coating porosity, due to fewer molten droplets arriving at the substrate and less sintering of the fine particles after deposition. Water contact angles were higher on the coatings deposited at the short standoff distance, which correlates with a higher surface roughness ( Supplementary Table S1 and Fig. S5 ). Effect of torch power When the torch power was increased by increasing the arc current, a reduction in the coating porosity was observed (see Supplementary Table S1 ). Conditions 5 and 6 showed the most pronounced effect of torch power on coating porosity ( Fig. 2d,e ). The coating microstructure produced with a higher torch power ( Fig. 2e ) still contained many incompletely melted particles, but denser regions formed by molten splats were seen, resulting in a reduction in porosity of over 10% ( Fig. 2d ). In contrast with the reduction in coating porosity, the change in torch power has negligible effect on the coating thickness and water contact angle. Effect of number of torch passes As the coating thickness increased during deposition, the roughness of the surface gradually increased, as shown in Fig. 2d . When the coating thickness exceeded approximately 20 microns, a feathery structure began to appear, as seen for both conditions 1 and 3 ( Fig. 2b,c ). This structure results from a shadowing effect, whereby small particles approaching the surface along a path which is not perpendicular to the surface deposit preferentially at high points on surface. Once the feathery structure begins to grow, the shadowing effect becomes more and more dominant. Therefore, a further increase in the number of torch passes during deposition leads to longer feathery structures, as shown in Fig. 2f,g . However, the exaggerated growth of the feathery structure was not beneficial to the hydrophobicity of the coating. The water contact angle was reduced to 140° for condition 7 ( Fig. 2f ). The water droplet penetrated into the spacing between the feathery structure columns, and the contact angle hysteresis increased to over 30°, an indication that the water droplets entered the Wenzel regime 10 . Effect of solvents It has been reported that higher density coatings can be obtained by using a solvent consisting of a mixture of alcohol and water 8 . Adding ethanol to the solution can enhance heat transfer between the plasma and the feedstock solution 11 . Introducing alcohol into the solution also decreases the surface tension of the solution, improving the secondary breakup of the atomized solution in the plasma plume, resulting in smaller size droplets 8 . Smaller droplets require less total heat to evaporate the solvent and fully melt the solute, which should increase the density of the coating. This effect can be seen by comparing the porosity of the coatings deposited under conditions 2 and 7, and conditions 4 and 8. Thermogravimetric analysis and differential scanning calorimetry (TGA-DSC, NETZSCH STA F3, see Methods) were performed to examine the precipitate powders that resulted from drying both solutions (see Supplementary Fig. S4 ). From these results, the effect of ethanol on the nature of the initial precipitate is negligible; therefore the difference in coating structure must be attributed to the change in the break-up behaviour of the droplets and the difference in the enthalpy required vaporizing the two solvents. Effect of plasma velocity The effect of plasma velocity was investigated in conditions 9 and 10. To decrease the plasma velocity compared to conditions 7 and 8, the total plasma gas flow rate was reduced from 250 slpm to 200 slpm and the nozzle size was increased from 10 mm to 13 mm. Figure 2g shows an SEM image of the coating deposited under condition 9. Among all the spraying conditions investigated, conditions 9 and 10 have the highest coating porosities (see Supplementary Table S1 ). Note that when the total plasma flow rate was reduced, the total torch power also decreased. However, conditions 9 and 10 have similar torch power, slightly higher enthalpy, and the same standoff distances as conditions 1 and 3 respectively, but higher porosity. The high coating porosity resulting from the lower gas flow rate agrees with previous findings of solution precursor flame-sprayed coatings 12 . When the plasma gas flow rate was decreased, the feedstock solution experienced less secondary breakup, resulting in larger droplets during the spraying process, which required more heat to melt. Also, the particles would have lower momentum before impacting the substrate. These factors together resulted in a lower packing density of the particles and higher porosity in the coating. The coating thickness for conditions 7 and 8 was similar to conditions 9 and 10. The water contact angle in conditions 9 and 10 is approximately 160°. Even though the coating was porous, the spacing between the feathery structures was reduced, and the water droplet remained in the Cassie regime during the contact angle measurement 13 (see Supplementary Fig. S5 ). The optimized spraying condition Based on the previous results, deposition of a dense superhydrophobic coating requires: a short standoff distance; a high arc current; a low number of torch passes; the addition of ethanol to the solvent; and a high plasma velocity. Conditions 11 and 12 include these requirements, and furthermore a lower feedstock flow rate was used to reduce the cooling of the plasma ( Table 1 ). Much denser coatings were deposited under these conditions ( Fig. 2h , and Supplementary Table S1 ). From the SEM images of the deposit formed from a single torch pass (see Supplementary Fig. S3 ), good coverage of the substrate was observed at the center of the plasma plume, and the deposit consisted mainly of pancake shaped splats, which indicate complete melting of the feedstock material. As Fig. 2h shows, the molten droplets can be seen to have flowed into a crater in the substrate surface formed by roughening of the substrate surface prior to deposition, which improves the mechanical interlocking between the coating and the substrate 14 . In between the dense regions, incompletely melted particles were observed. From the examination of the single torch pass, these particles resulted from feedstock that travelled at the perimeter of the plasma plume. A uniform distribution of micro-scale irregular clusters ranging from 5 microns to 30 microns in size was observed on the surface of the coating ( Fig. 3b and Supplementary Fig. S5 ). The clusters are agglomerates of individual particles less than 100 nm in diameter. This topography is consistent with the cross sectional images of the coating. The hierarchical structured top surface of the coating with a multi-scale roughness is very similar to the surface of superhydrophobic leaves in nature, such as the quaking aspen leaf ( Fig. 3a ). The as-sprayed coating surface was initially hydrophilic, but after vacuum treatment at 1 Pa for 48 hours the coating surface became superhydrophobic as shown in Fig. 3c . This transition is believed to be dependent on the surface oxygen-to-metal ratio 15 . The combination of the surface structure and the intrinsic hydrophobicity of the material gives the coating an excellent water repellent property ( Fig. 3d ). A 0.1 m 2 surface could be uniformly coated with an average coating thickness of 15 μm in less than 20 minutes in air using our laboratory set-up. This is over two orders of magnitude faster than other recently developed techniques for producing superhydrophobic surfaces such as laser ablation 16 17 and physical vapor deposition 18 19 ."
} | 2,751 |
25228891 | PMC4151088 | pmc | 9,213 | {
"abstract": "The problem of deriving the processes of perception and cognition or the modes of behavior from states of the brain appears to be unsolvable in view of the huge numbers of elements involved. However, neural activities are not random, nor independent, but constrained to form spatio-temporal patterns, and thanks to these restrictions, which in turn are due to connections among neurons, the problem can at least be approached. The situation is similar to what happens in large physical ensembles, where global behaviors are derived by microscopic properties. Despite the obvious differences between neural and physical systems a statistical mechanics approach is almost inescapable, since dynamics of the brain as a whole are clearly determined by the outputs of single neurons. In this paper it will be shown how, starting from very simple systems, connectivity engenders levels of increasing complexity in the functions of the brain depending on specific constraints. Correspondingly levels of explanations must take into account the fundamental role of constraints and assign at each level proper model structures and variables, that, on one hand, emerge from outputs of the lower levels, and yet are specific, in that they ignore irrelevant details.",
"introduction": "1. Introduction Any attempt to derive the processes of perception and cognition or the modes of behavior from sets of neural activities is confronted with the problem of mapping an incredibly large set of possible brain states to a very large number of observables. Simply put, the numbers are staggering: although estimates vary, there are purportedly about N = 10 11 neurons in the human brain (Sporns, 2012 ) and even with the very drastic simplification that a neuron is a binary device, possible states are 2 N = 2 10 11 . This enormous set of states must be mapped into the possible observables and even in this case numbers are huge: for instance even with a conservative estimate the number of possible postures is 10 30 (Stephens et al., 2011 ). The sheer orders of magnitude involved seem to prevent the possibility of finding any correspondence among elements of the two sets, i.e., the matching of states to observable processes. Fortunately there are factors that somewhat simplify the problem: for instance a given behavior can result from many different brain states, as redundancy is a well known evolutionary feature to make living systems more robust. Furthermore brains are made up of very complex networks (connections are of the order of 10 15 ), thus neural states are not independent variables and they tend to form spatio-temporal patterns, rather that disordered sequences of activity. Indeed, fMRI measures have shown that spatial maps of activity are formed even in resting state situations, without any external stimulus (Raichle, 2010 ). In addition, as suggested in Ganguli and Sompolinsky ( 2012 ), states of the dynamical systems describing the activity of cortical areas (e.g., motor cortex, or sensory cortex) are limited by the dimensionality of the inputs (e.g., motor task to be performed, or sensory inputs), which is often much lower than the dimensionality of the cortical dynamical system. These simplifying factors notwithstanding, the brain is so complex that to explain cognitive and behavioral functions philosophers and scientists have often resorted to conceptual metaphors (Daugman, 1993 ); modern examples are the computer and information metaphor (see Werner, 2011 ) for a critical review. An earlier version of the computation metaphor, based on the seminal work of McCulloch and Pitts ( 1943 ), on the equivalence between networks of formal neurons and Turing machines, was centered on the notion that neural activity implements logical calculus via formal rules for the transformation of for the manipulation of symbols (Daugman, 1993 ), an idea which has provided much impulse to the development of artificial neural networks and their applications (Haykin, 1994 ; Werner, 2011 ). The computation metaphor later has given rise to the so called “computational theory” of the brain whose aim is to explain and to simulate the mechanisms by which the brain performs a variety of tasks such as, for instance, edge detection or stereo vision (Marr, 1982 ). This version of the computation metaphor has became so popular that the term “computational” is nowadays used to characterize almost any model including task analysis (Daugman, 1993 ). Complementary to this approach is the information metaphor, that views the brain as an information processing device and focuses on the input–output relations among neurons in the framework of information theory. The central issues in this framework are those of coding and decoding of the neural stimulus, namely which feature of a neural spike train (rate, correlations, etc.) carries the information (in Shannon's sense) and, next, how this information is decoded by the brain, revealing the nature of the external (physical) stimulus (Jacobs et al., 2009 ; Werner, 2011 ). The latter problem is known to be an inference problem (Knill and Pouget, 2004 ), to solve which Bayesian techniques have proven to be very successful. This has lead to the “Bayesian coding hypothesis”: the brain represents sensory information in the form of probabilities and derives posterior probabilities of the configurations of the external world (Knill and Pouget, 2004 ; Doya, 2007 ; Friston, 2012 ). Computation and information metaphors are useful to elucidate important aspects of brain function, but, as pointed out in Werner ( 2011 ), they fail to provide the fundamental link between the dynamics of neural activity and computational and information processing properties of the brain. Thus, a different approach has emerged which maintains that real comprehension of cognitive and behavioral functions can only follow from the analysis and explanation of the collective dynamics of neural systems (Werner, 2011 ; Parker and Srivastava, 2013 ). This is also the point of view taken in the present work: specific models related to this approach will be reviewed in more detail later. Neuronal activity takes place at different scales and a rough classification can distinguish between microscopic (neurons and synapses), mesoscopic (networks and local interactions between neurons), and macroscopic levels (areas of the brain) (Deco et al., 2008 ). All these levels have their own specificity determined by different types of activity patterns. Then understanding the dynamics of the nervous system requires insights into processes occurring at different scales and that must be matched by appropriate levels of description or representation, characterized by specific variables and model structures. Different neural models can be represented as elements of a two dimensional space (Cessac and Samuelides, 2007 ). The first axis of this space describes the type of neuron and its proximity to biology, starting from the Hodgkin–Huxley equations followed by excitable systems with continuous state and finally binary neurons of the McCulloch–Pitts type. The other axis takes in account the collective aspect of neural networks in a hierarchy of ordering: one neuron, few neurons, one population of weakly coupled neurons and finally one population with arbitrary coupling. Large neural populations present an obvious similarity with physical systems composed of very large number of elements (atoms or molecules) subjected to mutual interactions. In physics the answer to challenges posed by such systems is to resort to mechanical statistical methods, which do not try to solve models at the microscopical level of individual elements, but, instead, use laws of probability to derive a set of collective variables, whose properties can then be studied at the macroscopic level. The success of this approach requires, and indeed depends on, finding the right variables, which can lead to meaningful macroscopic representations, while disregarding irrelevant ones. This, in turn, involves simplifying the system under consideration, from a detailed description to a more abstract representation in which some properties of the elements forming the system are disregarded. It must be kept in mind, however, there are crucial differences when considering physical vs. neurobiological systems. First, neural systems of the brain are part of living organisms. The problems concerning the transitions from inert to living states of matter and the characterization of life (Smith and Szathmary, 1997 ; Longo and Montévil, 2012 ) are outside the scope of this paper. It is enough to say that, at a fundamental level, activity of neural systems is constrained by the amount of metabolic energy available and by the need to limit entropy production (Schrödinger, 1956 ; Longo and Montévil, 2012 ). More relevant for our work is the fact that animal brains have been shaped by evolutionary pressures and, therefore, neural systems are subjected to many cost-benefits trade-offs, the most basic involving the balance between the speed of respose against the accuracy of identification of a stimulus (Geary, 2005 ). These constraints affect the topology of the connections: empirical evidence suggests that brain anatomical connectivity is locally clustered with a few long-range connections between any pair of regions, and this can be explained by the need to minimize wiring costs while maintaining the possibility of long range interactions among different areas (Bassett and Bullmore, 2006 ). Neurons interact with the rest of the organism and among themselves in ways, in general, more complex than interactions among elements of physical systems. Furthermore, neurons are computational units, able to perform non trivial computations (Koch, 2004 ). Differently from physics where the elements of a system can be considered all equal (“all electron are the same” as Fermi put it), neural systems are characterized by heterogeneity, e.g., excitatory vs. inhibitory neurons or electrical vs. chemical coupling. Neural systems are endowed with specific architectures, gauged to specific sensory, motor, and cognitive tasks. Networks can learn by changing the strength of their mutual connections. In physical systems the global behavior can be represented by simple scalars, for instance critical exponents and correlation lengths in non-equilibrium phase transitions, whereas models of large networks in the brain must explain the complex spatio-temporal patterns that make up physiological or behavioral responses. Therefore the question arises of what constitutes the relevant definition of system activity for a given level of explanation. These differences notwithstanding, a statistical mechanics approach is almost inescapable, since dynamics of the brain as a whole are obviously determined by patterns of neural activities occurring at a lower level, and, indeed statistical mechanics tries to derive the laws at the macroscopic level from interactions among microscopic components. A classical example are, in the theory of artificial neural networks, the so called Hopfield networks of binary units, (see Hopfield, 1982 ; Amit, 1992 ) and, for more recent results, (Advani et al., 2013 ). Statistical mechanical techniques are not restricted to the Hopfield model (Coolen and Del Prete, 2003 ): they have been applied also to biological neural systems both to explain experimental data (Masoller et al., 2009 ; Montani et al., 2009 ; Deco et al., 2012 ) and to provide general models of the brain (Ingber, 1981 ; Freeman and Vitiello, 2006 ; Parker and Srivastava, 2013 ). It will be argued here that the problem of modeling and representing neural systems of increasing size and complexity is akin to the problem of statistical mechanics and that the way out of the problem of intractability is the same: to assign at each scale proper variables, namely variables that emerge from outputs of the lower level, while ignoring details which are irrelevant for the higher level. In particular, the main claims of this paper are: Systems at each level obey to the laws holding for the lower levels, but they are subjected to new constraints that in turn generate new features, like novel patterns of activity, requiring adequate levels of representations. Constraints derive from the neural connections whose complexity increases with the dimension of neural circuits, whose topology then plays a central role in determining neural dynamics. This approach is inspired by the ideas of Jacob ( 1977 ) on the structure of natural systems:\n “ Nature functions by integration…. Each system at a given level uses as ingredients some systems of the simpler level but some only. The hierarchy in the complexity of objects is thus accompanied by a series of restrictions of limitations. At each level new properties may appear that impose new constraint on the system… Those (constraints) that operate at a given levels are still valid at a more complex level .”",
"discussion": "3. Discussion It is often said that the human brain is the most complex structure in the known universe, even though how such complexity can be computed is still a open question. In Tononi et al. ( 1994 ), complexity is derived by measures of mutual information, but other definitions could be considered, based on the entropy of the states of neural populations (Shiner et al., 1999 ). In any case the complex nature of the brain reveals itself in the structure of its connections and patterns of activities. These two aspects are inextricably linked: the structure of interactions among elements of a neural population generates patterns of activity of increasing complexity. If the single neuron can just perform a scale transformation of the inputs, pairs of mutually connected neuron can give rise to a variety activity patterns, characterized by the presence of attractors and sustained oscillations. These patterns result from the constraints that weights impose on activities. Also, we have shown that in large neural systems the processes of integration and segregation of connections give rise to a greater variety of activities of neurons and neuron groups. As mentioned earlier, in models of biological networks events are usually supposed to occur at three canonical scales, namely: microscopic, mesoscopic, and macroscopic, to which correspond different levels of explanation. Inside each scale some finer subdivision can be considered. For instance, motifs, small repetitive networks occurring in large neural populations and supposed to be building blocks of larger networks (Sporns and Kötter, 2004 ; Battaglia et al., 2012 ) can be thought of as an intermediate level between microscopic and mesoscopic scales. Also networks devoted to specific behavioral or cognitive tasks can provide a link between mesoscopic and macroscopic levels. An interesting suggestion has been presented in West and Deering ( 1994 ): in many physical systems “ exists a critical dimension above which fluctuations have only a quantitative effect, but below which the fluctuation can be amplified to modify the qualitative behavior of the phenomenon .” In the context of neurobiology, this observation could be translated to mean that domains in the cortex in which variations of activity are amplified into sharp transitions implicitly determine a proper scale for the explanation, for instance, of the sensory or cognitive responses to an input. The focus of the present work is on the connectivity among neurons in large neural populations and considers a simple neuron model with complex connections, so it can be thought of as situated close to one end of the conceptual space proposed in Cessac and Samuelides ( 2007 ); moving across this space one can find models with different emphasis on the neuron/connectivity relationships. At the opposite end of the spectrum with respect to the approach presented here is the analysis of the computational properties of the single neuron, which appears to be able to perform also complex computations (Rieke, 1999 ; Dayan and Abbott, 2001 ; Koch, 2004 ). Each specific model can be backed (or disproved) by specific types of data, from recording of electrical activity for single neurons or small networks to activity maps, for instance obtained with fMRI techniques, for large populations. Conflict of interest statement The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest."
} | 4,141 |
33531388 | PMC7858071 | pmc | 9,214 | {
"abstract": "Recent advances in our understanding of c-di-GMP signaling have provided key insights into the regulation of biofilms. Despite an improved understanding of how biofilms initially form, the processes that facilitate the long-term maintenance of these multicellular communities remain opaque.",
"introduction": "INTRODUCTION Pseudomonas aeruginosa is a Gram-negative opportunistic pathogen that is found both environmentally and within clinical settings. Able to transition from planktonic lifestyles to a biofilm mode of growth, P. aeruginosa biofilms develop via a number of discrete steps generally defined as initial attachment, irreversible attachment, microcolony formation, maturation and dispersal ( 1 ). Flagella mediate the initial polar attachment of the cell to the surface, whereas pili facilitate irreversible attachment and commitment to surface growth ( 2 , 3 ). Once on the surface, increased production of extracellular polysaccharide (EPS) facilitates increased surface adhesion and intercellular cohesion and provides both protection and structural integrity required by mature biofilms ( 3 – 6 ). An important element in the transition of P. aeruginosa from a planktonic to the biofilm mode of growth is bis-(3′,5′)-cyclic dimeric GMP (c-di-GMP), a second messenger that coordinates the regulatory control of virulence and behaviors needed for surface growth such as motility and the production of EPS ( 7 – 10 ). The concentration of intracellular c-di-GMP is controlled enzymatically by c-di-GMP-synthesizing diguanylate cyclases (DGCs) which contain GGDEF domains, and c-di-GMP degrading phosphodiesterases (PDEs) that harbor EAL or HD-GYP domains ( 8 , 11 ). The genome of P. aeruginosa PA14 encodes ∼40 different DGCs and PDEs involved in the regulation of c-di-GMP ( 12 ). Despite the rigorous investigations of the contributions of these different enzymes to biofilm formation and dispersion ( 13 , 14 ), our understanding of how they are coordinated and integrated temporally to affect bacterial surface behavior remains incomplete. Select DGCs and PDEs have been shown to be important during different phases of biofilm formation. For example, the DGC SadC and the PDE BifA both contribute to irreversible attachment and early biofilm formation ( 15 , 16 ). During subsequent biofilm growth, the DGC WspR mediates biofilm maturation by increasing the production of EPS ( 17 ). Regulation of c-di-GMP also facilitates the return to the planktonic lifestyle, as evidenced by the PDEs NbdA and DipA, which are required for dispersion in response to changes in levels of nutrients and nitric oxide, respectively ( 18 , 19 ). While the available literature provides substantial insights into how biofilms form and disperse, our understanding of biofilm maintenance—the process by which existing biofilms regulate themselves to persist on a surface—remains rudimentary. Indeed, it is not even clear if maintenance of the biofilm is an active process. To date, information regarding biofilm maintenance is largely informed by proteomic analysis of biofilms at specific stages of development ( 20 ). Such analyses are primarily conducted using biofilms grown under steady-state conditions, leaving open the questions of whether and how the regulation of established biofilms respond to starvation. Here, we provide evidence that the PDEs RmcA and MorA are needed for the maintenance of P. aeruginosa biofilms: the loss of either of these PDEs resulted in robust biofilm formation under nutrient-sufficient conditions, but increased cell death and compromised biofilms during starvation.",
"discussion": "DISCUSSION Exploiting previous findings from our lab in which a CRISPR-activated strain exhibited a defect in biofilm maintenance ( 21 ), we discovered that two PDEs, RmcA and MorA, were essential for maintaining late-stage biofilms. The Δ rmcA and Δ morA mutants exhibit phenotypes consistent with the inability to degrade c-di-GMP, specifically, elevated c-di-GMP, increased Pel production, and the ability to initiate a robust biofilm. However, the Δ rmcA and Δ morA mutants fail to maintain the biofilm in long-term static assays or when established biofilms are deprived of a carbon source in a microfluidic chamber. In addition, we have shown that the inability to maintain biofilms in these mutant backgrounds is driven by widespread cell death during nutrient limitation. Consistent with the hypothesis that cell death in these mutants is due to an aberrant nutrient limitation response, we showed that the Δ relA Δ spoT mutant, which lacks the ability to induce a stringent response, demonstrates a biofilm maintenance defect during nutrient limitation similar to that observed for the Δ rmcA and Δ morA mutants. Taken together, these data suggest a model ( Fig. 7 ) whereby the production of the energetically expensive Pel polysaccharide, required for the initial steps of biofilm formation, is downregulated by RmcA and MorA during biofilm maintenance when nutrient limitation conditions predominate. As such, while the loss of either PDE results in increased EPS levels and enhanced biofilm growth, a boon to these microorganisms in resource-rich environments typical of early biofilm formation, it leaves the cells unable to adapt to later nutrient-limited conditions in the context of a mature biofilm. FIG 7 Model for MorA- and RmcA-mediated biofilm maintenance in P. aeruginosa . Typical biofilm development (top panel) involves surface attachment, after which increased c-di-GMP and EPS synthesis mediate microcolony development and increased biofilm biomass. During the maturation and maintenance phase of biofilm development, regulatory changes reflect growth in a nutrient-limited environment and result in a decrease in the production of energetically expensive products like Pel, reduced c-di-GMP (cdG) and induction of the stringent response. The loss of RmcA or MorA (bottom panel) results in enhanced c-di-GMP and Pel production and increased biomass in nutrient-rich environments. The Δ rmcA and Δ morA mutant biofilms are unable to appropriately respond to nutrient limitation, resulting in cell death and loss of biofilm biomass. The mechanisms by which RmcA and MorA are regulated in nutrient-limited conditions remain unknown; however, recent findings in Pseudomonas putida provide a potential signaling framework. Work by Carlos Díaz-Salazar et al. found that RelA and SpoT-dependent synthesis of (p)ppGpp mediates dispersal during nutrient-limited conditions ( 37 ). This group also found that (p)ppGpp increased transcription of the PDE bifA and that a Δ bifA mutant was defective in starvation-induced biofilm dispersal ( 37 , 38 ). It is possible that in P. aeruginosa RmcA and MorA operate analogously to that of BifA in P. putida by acting as effectors of stringent response signaling under nutrient-limited conditions. Unlike BifA in P. putida , RmcA and MorA do not coordinate dispersal in P. aeruginosa but rather participate in effectively maintaining the biofilm in the face of nutrient limitation. It is also possible that the specific impact of RmcA and MorA on late-stage biofilms is due to their expression specifically in mature biofilms. Reported transcriptional analyses suggests that rmcA transcript levels are elevated in 48 h biofilms versus planktonic culture ( 39 ), consistent with this hypothesis. There is no such evidence for enhanced morA transcription in late-stage biofilms, but it is also important to keep in mind that much of c-di-GMP mediated control is via posttranscriptional regulation, as we and many others have shown ( 9 , 13 , 40 ). The high levels of c-di-GMP in the Δ rmcA and Δ morA mutants may have adverse impacts on the cells in nutritionally limited, mature biofilms. While we do not yet completely understand how the regulation of Pel may contribute to biofilm maintenance, there is strong evidence for the physical interaction of RmcA and MorA with PelD and/or other elements of the Pel biosynthetic machinery ( Fig. 6 ), suggesting that RmcA and MorA may have a direct role in regulating Pel synthesis. While artificially increasing pel expression alone to high levels does not result in a biofilm maintenance defect, pel gene overexpression combined with enhanced c-di-GMP production (via the wspF mutation) resulted in reduced biomass and increased cell death for these biofilms grown in microfluidics chamber when nutrient-limited. These data are consistent with our findings that overexpression of c-di-GMP alone cannot induce a maintenance defect (see Fig. S3 ) and suggests that it is unregulated production of Pel which makes P. aeruginosa biofilms vulnerable in nutrient-limited conditions. Together, with the data presented here indicating that RmcA and MorA physically interact with elements of the Pel machinery, our findings suggest a model whereby RmcA and MorA are components of a Pel regulatory complex required for biofilms to successfully navigate starvation conditions. Finally, we believe that the importance of RmcA- and MorA-mediated regulation of biofilm maintenance extends beyond the in vitro studies we show here, as P. aeruginosa strains carrying mutations in rmcA and morA genes were the only PDEs to show a defect in a mouse model of catheter infection from among the 15 strains lacking genes encoding proteins with EAL or HD-GYP domains ( 41 ). Further studies are needed to elucidate whether the potential role of Pel in biofilm maintenance is related to the stringent response. It is possible that the inability to appropriately regulate c-di-GMP and EPS production during nutrient limitation impacts (p)ppGpp levels, eventually resulting in extensive cell death and biofilm degradation. Alternatively, the ability to induce a stringent response may be part of a coordinated downregulation of metabolic activity required for the long-term maintenance of a mature biofilm, particularly when carbon/energy sources are limiting. In addition, Pseudomonads appear to have developed catabolic pathways for the utilization of arginine and lactate for “maintenance energy” in mature biofilms ( 42 , 43 , 58 ), as well as a pathway to downregulate flagellar motility, another early-stage biofilm factor, in mature biofilms ( 1 ). Taken together, these data indicate that pseudomonads, and likely other microbes, have active, well-regulated mechanisms necessary to maintain a mature biofilm in the face of changing environmental conditions. Finally, the apparent role for c-di-GMP-metabolizing enzymes RmcA and MorA later in the biofilm lifestyle suggests the interesting possibility that the plethora of these enzymes in pseudomonads stems from their roles in regulating discrete aspects of the biofilm lifestyle—from formation to maturation to maintenance to dispersal. The finding that the loss of different PDEs would result in different phenotypes may be expected given the varied impacts that these enzymes have on the regulation and timing of c-di-GMP signaling and biofilm formation ( 9 , 13 ). Here, while we assessed all PDE and dual-domain mutants in P. aeruginosa in our initial screen, we only observed consistent and significant defects in biofilm maintenance for the Δ rmcA and Δ morA mutants. Previous work identified a number of DGCs and PDEs apparently required for early biofilm formation, including SadC, RoeA, BifA, and SiaD ( 15 , 16 , 44 ). In contrast, the PDE DipA has been shown to mediate biofilm dispersion in response to elevated nutrient concentrations and this protein localizes to the cell pole during division, resulting in the asymmetric distribution of c-di-GMP ( 18 , 45 ). Thus, our data are consistent with the hypothesis of stage-specific roles for DGCs/PDEs in the biofilm life cycle. The network which controls c-di-GMP levels in P. aeruginosa is complex. Identified first in P. putida , MorA was found to repress motility in swim assays ( 35 ). This same work found that the enhanced motility of the morA mutant was not observed in P. aeruginosa ( 35 ), but previous work from the Hogan and O’Toole labs showed that the Δ morA mutant exhibited a significant decrease in flagellum-dependent swimming and swarming motility ( 23 ). The basis of this difference in phenotypes may be due to the fact that different species of Pseudomonas were used in the two studies. Nevertheless, given the role of motility in early biofilm formation, the observation that MorA contributes to swimming and swarming motility indicates that this PDE also likely contributes to the initiation of biofilm communities. Insightful work from the Dietrich lab provided evidence that RmcA is activated by phenazine’s availability to mediate a decrease in c-di-GMP levels during oxidative stress conditions ( 24 ). In this model, RmcA can act as a redox sensor and may behave as a switch to translate this signal into decreased levels of c-di-GMP and EPS. This model for the role of RmcA in the context of the colony biofilm used by Okegbe et al. ( 24 ) is largely in agreement with the experimental evidence we have provided, which suggests that RmcA is important for biofilm maintenance. Specifically, it is likely that in mature biofilms with elevated biomass, nutrient limitation coincides with oxygen depletion and the production and utilization of phenazines as electron shuttles. Indeed, the direct regulatory signal sensed by RmcA appears to be a change in redox state that is likely secondary to the loss of a catabolizable carbon source ( 24 ). Thus, in this environment, we hypothesize that RmcA downregulates the production of energetically expensive Pel EPS and that failure to do so could result in the observed cell death and biofilm maintenance defect. Together, these data suggest that further examination of these enzymes will generate a more nuanced view of the model presented in Fig. 7 , wherein specific c-di-GMP metabolizing enzymes work at one or more stages of the biofilm life cycle, with the potential to perform several overlapping functions across these various stages (i.e., biofilm initiation and biofilm maintenance)."
} | 3,532 |
37540744 | PMC10403213 | pmc | 9,216 | {
"abstract": "How dynamic bacterial calcium is regulated, with kinetics faster than typical mechanisms of cellular adaptation, is unknown. We discover bacterial calcium fluctuations are temporal-fractals resulting from a property known as self-organized criticality (SOC). SOC processes are poised at a phase transition separating ordered and chaotic dynamical regimes and are observed in many natural and anthropogenic systems. SOC in bacterial calcium emerges due to calcium channel coupling mediated via membrane voltage. Environmental or genetic perturbations modify calcium dynamics and the critical exponent suggesting a continuum of critical attractors. Moving along this continuum alters the collective information capacity of bacterial populations. We find that the stochastic transition from motile to sessile lifestyle is partially mediated by SOC-governed calcium fluctuations through the regulation of c-di-GMP. In summary, bacteria co-opt the physics of phase transitions to maintain dynamic calcium equilibrium, and this enables cell-autonomous population diversification during surface colonization by leveraging the stochasticity inherent at a boundary between phases.",
"introduction": "INTRODUCTION Compared to eukaryotic systems, little is understood regarding how bacteria manage ion flux ( 1 – 4 ). Because of their size, small ionic changes are sufficient to drastically alter bacterial electrophysiology ( 3 ). In the case of free Ca 2+ , bacteria must balance an external concentration of up to 1 mM against an internal concentration as low as 300 nM ( 5 ). As was noted by Benarroch and Asally ( 3 ), the small size of a bacterium is consequential for its electrophysiology. For a femtoliter cell with a membrane capacitance of 1 μF/cm 2 , a surface area of 6 μm 2 , and an internal Ca 2+ concentration of 300 nM ( 5 ), the addition of a mere 180 free Ca 2+ ions to the cytoplasm would double the molar concentration and increase the membrane potential by 1 mV ( V m = q / C where q = 180*2*1.6 × 10 −19 Coulombs/proton and C = 1 μF/cm 2 × 6 × 10 −8 cm 2 ). The kinetics of Ca 2+ fluctuations are much faster (~ms) ( 6 ) than typical mechanisms of biological adaptation such as transcription and posttranslational modifications. Yet, despite these kinetic challenges, bacteria use dynamic Ca 2+ in numerous processes including motility, energy production, and virulence ( 2 , 3 , 7 ). How is this dynamic equilibrium maintained without enormous energy loss to continuously coordinate ion flux? Here, we find evidence that Ca 2+ flux is governed by the physics of phase transitions, specifically a transition between dynamical regimes in a process termed self-organized criticality (SOC) ( 8 ). SOC has been studied in a wide range of seemingly disparate phenomena ranging from flocking birds to ferromagnets to the architecture of the brain ( 9 – 16 ). Rather than separating states of matter, SOC systems are characterized by the separation of dynamical regimes, colloquially termed “order at the edge of chaos” ( 17 ). The classic example of SOC is sand falling and forming into a pile; as sand continues to fall, the pile alternates between avalanches and growth ( 8 ). While the avalanches’ timing is stochastic, the statistical relationship between its size and duration is predictable. All SOC systems have universal properties permitting their collective behavior to be described with statistical models while neglecting the intricacies of the system’s subcomponents. For example, the size and duration of avalanches in a SOC system have probability distributions well described by a power law ( 9 , 17 ). Power-law scaling leads to scale invariance whereby the statistical behaviors of these systems do not depend on the size of the spatial or temporal domain (i.e., changing the size of the sand pile does not alter the relationship between avalanche size and duration) ( 9 , 17 ). Here, we investigate whether SOC governs Ca 2+ flux within a single-celled bacterium. This inquiry was prompted by the observation of anomalous structures in the power spectra of a Ca 2+ sensor expressed in Escherichia coli when characterizing the type of noise in bacterial Ca 2+ regulation.",
"discussion": "DISCUSSION The dynamic nature of bacterial electrophysiology and its role in biofilm growth and maintenance has become increasingly clear over the last decade ( 3 , 4 , 50 ). B. subtilis communicates metabolic needs electrically through a biofilm ( 51 , 52 ), and these communications are governed by percolation theory ( 53 , 54 ), demonstrating that single-cell organisms can co-opt physical principles which increase their collective fitness. Here, we find another elegant example of the complexity of a bacterium using the physics of SOC to regulate Ca 2+ flux. By selecting for ion channels with the biophysical characteristics necessary to establish SOC, single-cell organisms have co-opted the physics of phase transitions for biological sensing. Our observation of SOC governing Ca 2+ dynamics is not merely a phenomenological parallel to criticality but, in a deep statistical sense, a true phase transition ( 55 ) defined by the coupling strength between channels. Mechanisms that have evolved to homeostatically tune the strength of the Ca 2+ channel coupling remain to be elucidated. However, the evolutionary advantages of SOC are readily apparent. The dynamics span several orders of magnitude while simultaneously maintaining long-term stability in the face of diverse, rapidly changing environments ( 25 , 26 ). Our work suggests an additional evolutionary advantage of SOC: emergent diversification of individual actors even in the absence of communication. Leveraging the stochasticity inherent at the boundary between phases, bacteria can coordinate a variable commitment to sessile growth versus the maintenance of a motile lifestyle in unpredictable environments. Prior work has shown power-law–like residence times of single-cell attachment dynamics in Pseudomonas aeruginosa ( 56 ). Our work points to the potential role of scale-free, mechanically stimulated fluctuations in Ca 2+ in a bacterium’s decision to invest in a sessile lifestyle. Additional investigation is necessary to determine how unique this phenomenon is to Ca 2+ as compared with other ions such as K + or Na + , as well as how fluctuations influence other Ca 2+ -regulated elements of cell physiology beyond c-di-GMP ( 2 , 3 ). In addition, we found that SOC is maintained across environmental conditions; however, the critical exponent can vary. Thus, the critical exponent is not unique but exists on a continuum depending on the internal and external environments. We discovered moving along the continuum changed the collective information capacity of the population (i.e., its entropy) as increasing the critical exponent leads to decreasing the entropy in the avalanche size up to a threshold. This relationship suggests an overarching lens through which to re-examine how critical exponents and information capacity are modified in diverse SOC systems. Limitations of study The primary limitation of this study is the inability to directly manipulate cytosolic Ca 2+ in E. coli using reagents such as bis-aminophenoxy-ethane-tetraacetic acid-acetoxy-methylester (BAPTA-AM) due to limited uptake across the double membrane. Because the flux in cytosolic Ca 2+ originates from the periplasm in Gram-negative rods like E. coli , the cells are robust to external changes in Ca 2+ . In addition, the paucity of information on the identity of Ca 2+ effectors in E. coli highlights the need for further discovery of the key players in regulating bacterial electrophysiology. Without the ability to directly manipulate cytosolic Ca 2+ , much of our study depended on altering the membrane voltage and SOC via the protonophore CCCP. Last, previous attempts to measure rapid Ca 2+ dynamics at single-cell resolution in Gram-negative species using fluorescent dyes such as Fura2-AM have proven infeasible due to no uptake. Therefore, our studies are limited by reliance on GCaMP6f. Another limitation was the necessity of defining an avalanche construct to use existing SOC analytics. While we observed the hallmarks of SOC for two different avalanche definitions, both approaches were pursued to leverage existing SOC frameworks ( 31 ). This limitation points to the need for extensions of classic SOC analytics to continuous signals. Also, we found the SOC analytics, in particular, the power law fits of the avalanche size ( S ) and duration ( T ), to be sensitive to slight changes in the data preprocessing. As a result, the crackling noise relationship proved to be the least reliable measure of the critical exponent. Last, there are several areas for future improvement in our approach to model bacterial SOC. The communication between channels in our model cannot be derived from biophysical measurements but rather is patterned on the discrete, Ising-like models developed to model SOC in neural systems. As is the case in our study, the binarization of a neuron and nonthermodynamic equilibrium of these models substantially reduces their biological “realism”; nevertheless, these models have proven useful in predicting responses of neural systems to perturbations ( 29 ). Similarly, our model predicted the relationship of the entropy as a function of the critical exponent ( Fig. 5, D and E ) as well as the salient features observed in the GCaMP6f traces including the scale-free power spectra, branching function, and ACF ( Table 2 ) despite this limitation. In particular, the prediction of the relationship between the critical exponent and entropy of the avalanche sizes depended on connecting the Ising-like model developed for neuroscience with the Hodkin-Huxley model of ion flux thereby revealing a continuum of critical attractors."
} | 2,466 |
29755329 | PMC5934941 | pmc | 9,217 | {
"abstract": "Honey bees transfer different informational components of the discovered feeding source to their nestmates during the waggle dance. To decode the multicomponent information of this complex behavior, dance followers have to attend to the most relevant signal elements while filtering out less relevant ones. To achieve that, dance followers should present improved abilities to acquire information compared with those bees not engaged in this behavior. Through proboscis extension response assays, sensory and cognitive abilities were tested in follower and non-follower bees. Individuals were captured within the hive, immediately after following waggle runs or a bit further from the dancer. Both behavioral categories present low and similar spontaneous odor responses (SORs). However, followers exhibit differences in responsiveness to sucrose and odor discrimination: followers showed increased gustatory responsiveness and, after olfactory differential conditioning, better memory retention than non-followers. Thus, the abilities of the dance followers related to appetitive behavior would allow them to improve the acquisition of the dance surrounding information.",
"introduction": "Introduction The waggle dance is a stereotyped behavior performed by Apis mellifera foragers which consists in an eight-shape figure on the vertical comb inside the hive (von Frisch, 1967 ). This complex behavioral display is considered a multicomponent signal (Grüter and Farina, 2009 ) which not only attracts nestmates to the dance surrounding but also informs the presence of a profitable food source (von Frisch, 1967 ; Seeley, 1989 ). Honey bees can acquire information about the location of the feeding site by following these maneuvers from behind or laterally (Michelsen, 2003 ; Díaz et al., 2007 ). The dance followers can also perceive and learn the odors of the collected food during interactions with the dancer (von Frisch, 1967 ; Farina et al., 2005 ; Díaz et al., 2007 ). In this way, both naïve and experienced foragers can acquire information from the waggle dance (Biesmeijer and de Vries, 2001 ; Biesmeijer and Seeley, 2005 ). The honey bee dance takes place in particular comb areas named “dance floor” (Tautz and Lindauer, 1997 ), located at approximately 4–20 cm from the hive entrance, where dancers and dance followers come into contact (von Frisch, 1967 ; Seeley, 1995 ). Dance maneuvers increase the activity of bees in the dancer’s vicinity (von Frisch, 1923 ; Božič and Valentinčič, 1991 ; Thom et al., 2007 ). Thus, in this informational context, the levels of motivation and attention of the bees located in the dance surrounding might be enhanced by the presence of the excited dancers. As a result, follower bees are motivated to start foraging. The shift from in-hive tasks to foraging involves changes in the responsiveness to external stimuli (Robinson, 1987 ; Robinson and Page, 1989 ; Seeley, 1989 ). Furthermore, bees performing different tasks within the colony also present different response thresholds (Ramírez et al., 2010 ). In this sense, Katz and Naug ( 2016 ) have shown that dancer and follower bees present different sensitiveness according to the individual and colony nutritional states. Creating a mismatch between these nutritional states and using a conditioning assay, they evaluated the proboscis extension response of fed and starved bees. Followers showed to be less sensitive to changes in the colony nutritional condition than dancers. However, incoming foragers can adjust their response to the nutritional status of the colony (Lindauer, 1954 ; Seeley, 1989 ; Farina, 2000 ; De Marco, 2006 ). Therefore, the presence of individuals responding differentially according to nutritional states would allow a better adjustment of the foraging activity of the entire colony. However, until now, it is unknown how different are the chemosensory and olfactory learning abilities of those bees located in the hive areas where the information related to the incoming resources is transmitted. Bearing this in mind, our aim is to study the sensory and cognitive capacities of bees involved in dance following and unemployed bees located next to the dance floor that did not follow dances at the moment they were captured. For this, we carried out behavioral assays testing sucrose responsiveness, spontaneous response to odors and ability to discriminate odors.",
"discussion": "Discussion Through PER assays, we evaluated olfactory and gustatory responsiveness besides the ability to discriminate odors through classical conditioning in honey bee workers with different probabilities to be recruited as foragers. A brief time after capturing both dance follower and non-follower bees showed similar SOR levels but significant differences in the sucrose responsiveness and odor memory retention. Specifically, dance followers presented higher gustatory responsiveness (higher GRS levels) and better memory retention after olfactory conditioning than non-followers. We did not take into account the age of the experimental bees in this study. Previous reports suggest that age seems not to be relevant within this social context. For instance, individualized bees that followed dances showed a wide age range (i.e., from 9 to 32 days old; Balbuena et al., 2012 ). Likewise, a recent study evaluated the sucrose responsiveness of hive bees captured from the dance floor or delivery area within an age interval of 2–15 days old (Mengoni Goñalons et al., 2016 ). No age-dependent relationship was found for the sucrose responsiveness measured at the per setup in that study. In this sense, the gustative responsiveness of honey bees may be affected by the informational context where individuals were caught from (i.e., dance floor) more than by their age. Indeed, Martinez and Farina ( 2008 ) showed that hive bees captured after receiving food from a donor bee presented different sucrose responsiveness according to the food quality received during the oral contact. In the present study, bees captured in the dance context (following dances but did not interact orally with a dancer) showed higher gustatory responsiveness compared with those bees captured at a distance of 10–20 cm from the dancer. High GRS values correlate positively with improved performances during olfactory conditioning as it was previously reported (Scheiner et al., 2004 ; Mengoni Goñalons and Farina, 2015 ). Consistent with this evidence, we found that dance followers showed improved levels of memory retention after an olfactory PER conditioning. However, when sucrose responsiveness correlates with memory retention, also correlates with learning performance (Scheiner et al., 2001 , 2004 ; Mengoni Goñalons and Farina, 2015 ). Here, we only found a correlation between memory retention and gustatory responsiveness. Those previous studies used absolute conditioning procedures to test it. In this study, differential conditioning has been used. This protocol evaluates the ability to associate an odor to a reward, but also the capacity to distinguish it from another which is not linked to an unconditioned appetitive stimulus. Here, our unrewarded odor, CS−, is the same odor used that we used as the hive odor and it would represent a nonappetitive context for the experimental subjects. Thus, the differences in response to rewarded and unrewarded odors would be bigger for those bees that are more motivated to acquire appetitive information. Although it was not significant, this tendency can be observed at the end of the learning performance for dance followers and clearly visualized while the trained odors were tested 15 min later. It is worth mentioning that we only evaluated memory retention at a medium-term scale. It is expected that highly motivated individuals not only learn faster but also recall longer, an issue which was not cover in the present study. Changes in the motivation and attention levels of the dance followers would turn out to be more sensitive to any other environmental stimuli, a fact that might facilitate the decoding of spatial information transmitted as well as the acquisition of incidental cues such as floral odors carried by the waggle dancer. The role of the early odor-rewarded experiences acquired in the beehive as a stimulus that facilitates the decoding of waggle dance information at elder ages has been suggested (Balbuena et al., 2012 ). In that study, honey bees preferred to follow dancers scented with an early exposed and rewarded odorant, and even they were recruited to the feeding site scented with the early experienced odors more successful. As the presence of reward affects physiological states in honey bees within a short-term period (Hammer, 1997 ), the most vigorous dances, which indicate the presence of a highly profitable food source, might represent appetitive stimuli that facilitate a prompt acquisition of information within the dance context. This study shows a correlation between sensory and cognitive performances and behavioral category based on the dance context. Nevertheless, it does not show if there is a causal relationship. To do that, the life history of individual bees should be considered to determine whether bees with a low gustatory responsiveness tend to follow dances or even whether the sucrose responsiveness changes after dance following. Our results are a first approach to understand abilities of the dance surrounding bees, but this issue requires further analysis."
} | 2,373 |
31551977 | PMC6737463 | pmc | 9,219 | {
"abstract": "Legume plants have colonized almost all terrestrial biotopes. Their ecological success is partly due to the selective advantage provided by their symbiotic association with nitrogen-fixing bacteria called rhizobia, which allow legumes to thrive on marginal lands and nitrogen depleted soils where non-symbiotic plants cannot grow. Additionally, their symbiotic capacities result in a high protein content in their aerial parts and seeds. This interesting nutritional value has led to the domestication and agricultural exploitation of several legumes grown for seeds and/or fodder for human and domestic animal consumption. Several cultivated legume species are thus grown far beyond their natural geographic range. Other legume species have become invasives, spreading into new habitats. The cultivation and establishment of legume species outside of their original range requires either that they are introduced or cultivated along with their original symbiotic partner or that they find an efficient symbiotic partner in their introduced habitat. The peanut, Arachis hypogaea , a native of South America, is now cultivated throughout the world. This species forms root nodules with Bradyrhizobium , but it is unclear whether these came with the seeds from their native range or were acquired locally. Here we propose to investigate the phylogeography of Bradyrhizobium spp. associated with a number of different wild and cultivated legume species from a range of geographical areas, including numerous strains isolated from peanut roots across the areas of peanut cultivation. This will allow us to address the question of whether introduced/cultivated peanuts associate with bacteria from their original geographic range, i.e., were introduced together with their original bacterial symbionts, or whether they acquired their current associations de novo from the bacterial community within the area of introduction. We will base the phylogenetic analysis on sequence data from both housekeeping and core genes and a symbiotic gene ( nif ). Differences between the phylogenetic signal of symbiotic and non-symbiotic genes could result from horizontal transfer of symbiosis capacity. Thus this study will also allow us to elucidate the processes by which this symbiotic association has evolved within this group of Bradyrhizobium spp.",
"introduction": "Introduction The symbiosis between legume plants and nitrogen-fixing bacteria called rhizobia is a major ecological process in the nitrogen biogeochemical cycle. This symbiosis allows legume plants to colonize N-limited environments, to accumulate large amounts of protein in their seeds and aerial parts and to enrich the soil with an input of fixed nitrogen at the end of their life cycle. Legume plants, with these very useful characteristics, were among the first domesticated plants at the dawn of agriculture, constituting, together with cereals (such as barley, emmer, and einkorn wheats) and flax (the first fiber crop), the so-called Neolithic founder crops. Lentil, chickpea, pea, and bitter vetch were domesticated and have been cultivated for food and feed and selected for high protein content for about 9,500–10,000 years ( Zohary and Hopf, 1973 ). Throughout human history several legumes including pulses, forage crops and trees have been spread all over the world, often intentionally introduced to areas far from their native range. In these new environments, legumes need symbiotic partners to establish a functional symbiosis. If the habitual partners are not introduced along with the host plants nodulating bacteria must be acquired locally from the soil bacterial community in the area of introduction. The availability of appropriate partners will depend on the biogeography of soil microbes. Are all taxa, potentially, everywhere, but their presence and abundance locally determined by environmental conditions, as was suggested by Baas-Becking in 1934 ( de Wit and Bouvier, 2006 ) with the famous “Everything is everywhere, but the environment selects”? Indeed, a broad survey of the bacterial communities from over 2000 soil samples across France revealed that presence and abundance of most phyla and genera was best explained by environmental drivers such as land use and soil properties ( Karimi et al., 2018 ). However, for four phyla, including beta-proteobacteria that include some rhizobia, spatial parameters explained most of the variation, implying limited dispersal. Thus, a biogeographical signal can be expected in the distribution and abundance of some soil bacteria such that introduced host plants may not find their habitual bacteria at a site of introduction. Furthermore, at the level below genus, several observations belie the absence of biogeography. First, for pathogenic organisms, quarantine often works, so clearly these pathogenic strains are not everywhere simply waiting for an appropriate host. Secondly for mutualists, several symbiotic legume crops required inoculation, at least initially, to grow well when they were first planted in new areas (see below). Indeed, the introduction success of a symbiotic legume in a new habitat depends on its capacity to find a local bacterial nodulator, and thus symbiosis may represent a barrier to the establishment of symbiotic legumes in a new region ( Parker, 2001 ; Simonsen et al., 2017 ). To our knowledge there is no general overview of how commonly introduced symbiotic legume species acquire local bacteria at the site of introduction versus how often their symbiotic partner is also introduced. In this paper, we compile information available in the literature about the bacteria that nodulate some major legume crops and trees. Comparing bacterial symbionts from the native and introduced ranges of these host plants allows inference as to whether these plants were introduced with their original symbionts or whether they acquired local bacteria from the community of bacteria nodulating indigenous legume plants. Finally, we focus on peanut ( Arachis hypogaea ), the second pulse and oilseed legume crop after soybean in terms of global production. Peanut is native to the Americas, although the main production areas today are Asia and Africa, making it an interesting case to study the history of legume introduction and nodulation outside the native range. Do introduced legumes bring their partners with them or associate with new partners? When rhizobial-associated plants are transported to new environments, they either are introduced together with their nodulating bacteria or they find and recruit mutualistic bacteria from the local rhizobial community. Identifying which process has occurred involves comparing strains of bacteria that nodulate the same plant species in its native and introduced range with bacteria that nodulate other legume species from both ranges. Indeed, if plants are introduced with their habitual nodulating bacteria, the bacterial strains from the native and introduced ranges should cluster together on a tree of genetic similarity, as depicted in Figure 1A . On the other hand, for plants that acquire new rhizobial partners in the introduced area, their bacterial strains from the introduced area should cluster together with bacterial strains from other species of host plant in that same introduced area, with clusters presenting geographical areas and a clear biogeographical pattern ( Figure 1B ). Several researchers have addressed this question directly, and several studies on the genetic diversity of rhizobial bacteria provide data from which these patterns can be deduced. Here we discuss what we could find for a number of introduced, invasive or cultivated legume species that grow in several different regions of the world. FIGURE 1 Phylogenetic trace of co-introduction (A) versus acquisition of a new symbiotic partner (B) in a new geographical zone. (A) Depicts scenario 1: If the bacteria found associated with a host species in a new geographic area arrived via co-introduction with the host plant these bacteria should be genetically most similar to the bacteria from the same host plant in its original geographical range. In contrast in (B) , scenario 2, if hosts acquire new symbionts from the soil bacterial community in their new geographic range, we expect the strains isolated from the introduced host species to most closely resemble bacterial strains associated with one or more host species within the introduced range. We assume some geographic pattern to relationships of the bacteria, with species from the same geographic range showing more similarity. Some nomadic legumes bring their symbionts with them. Several cultivated species are or were systematically inoculated when planted in new areas. Lupins ( Lupinus spp.) and Serradella ( Ornithopus spp.) were introduced as forage crops in Australia and South Africa but their successful cultivation required inoculation with co-introduced rhizobia for several decades. Now these plants can grow in both areas without inoculation, but the bacteria that nodulate them are closely related to strains from southern Europe, part of the native range of these forage species ( Stȩpkowski et al., 2005 ), suggesting that the current population of nodulating bacteria stem from the co-introduced inoculum. Similarly, in New Zealand, introduced and invasive Acacia , Cytisus and Ulex species associate with Bradyrhizobium microbial partners that are very different from the Sinorhizobium that nodulate native species in New Zealand ( Weir et al., 2004 ), while Dipogon lignosus , an invasive from South Africa, is nodulated by Burkholderia that closely resemble South African strains. Australian Acacia longifolia and A. melanoxylon , invasive in coastal sand dune habitats of Portugal, are nodulated by their habitual Australian rhizobia that must have been co-introduced with them. Furthermore, these introduced bacteria also can nodulate native Portuguese Cytisus and Ulex but these native plants derive far less benefit from the foreign bacteria than they do their habitual native bacteria ( Rodríguez-Echeverría et al., 2012 ). These patterns indicate that several introduced species arrived together with their bacterial symbionts, with which they continued to associate. Indeed, the fact that successful introduction and cropping of some legume species required the concomitant introduction and inoculation of their rhizobial symbiotes suggests that recruiting local bacteria is not always easy. In addition, novel associations between host plant and bacteria may be less efficient than old, coevolved associations ( Rodríguez-Echeverría et al., 2012 ), so even if new partners were available they may provide less advantage than co-introduced, habitual mutualists. Not surprisingly then, there are fewer examples of plants that, in their introduced range, are nodulated only by novel bacteria that do not resemble the bacteria from their native range. Mimosa pigra , a native of the neotropics, presents a mixed picture. This plant is invasive in many areas including Taiwan and Australia. In both invasive areas and in the native range it is nodulated by bacteria of the genus Burkholderia . Whereas the bacteria that nodulate this plant in Taiwan closely resemble one of the dominant strains that nodulates this plant in its native range ( Chen et al., 2005 ), the strains in Australia represent several divergent lineages that are unrelated to the native strains. Thus it appears that M. pigra was cointroduced with a nodulating strain to Taiwan ( Chen et al., 2005 ) but was not accompanied by its habitual Burkholderia on colonizing Australia, where it acquired novel symbiotic Burkholderia ( Parker et al., 2007 ). Several other species associate both with strains that appears to have been co-introduced, being very similar to strains found in the native range of the plant, and have adopted additional new symbiotic partners at the area of introduction. Within introduced populations or areas of cultivation this can be observed by an increase in the genetic diversity of symbiotic bacteria between the native and introduced range. Acacia pycnantha , an Australian invader of South Africa and Robinia pseudoacacia , a North American native that has been introduced throughout the world and is deemed invasive in some areas, both show a higher diversity of nodulating bacteria in the introduced than the native range. For both plant species bacteria characteristic of the native range were recovered also in the invaded range, together with additional bacterial taxa that represented additional, new partners from the native bacterial communities in the areas of introduction ( Chen et al., 2000 ; Ulrich and Zaspel, 2000 ; Wei et al., 2009 ; Tang et al., 2012 ; Ndlovu et al., 2013 ). In addition, horizontal exchange of the genes involved in the symbiotic process can also occur during the introduction of legumes to new habitats. Thus the two possibilities discussed above, of introduced, cultivated or invasive legume species either arriving accompanied by their rhizobial mutualists or acquiring novel ones from the microbial communities in their new geographic range do not cover all the observations of the rhizobia associated with legumes far from their native ranges. At the large evolutionary scale of bacterial genera, symbiotic and non-symbiotic genes in rhizobial bacteria appear to have distinct evolutionary histories. Symbiotic genes show G-C content and codon usage patterns distinct from the rest of the genome in Sinorhizobium meliloti ( Galibert et al., 2001 ), and homologous symbiotic genes occur in very distinct bacterial lineages, implying horizontal transfer, though from an as yet unknown source. The homologous system of dialog with legume hosts using Nod factors and of nitrogen fixation within nodules has been acquired by only ten distinct genera of alpha- and two distinct genera of beta-proteobacteria. Thus acquisition of nodulation induction, though repeated, appears rare at the level of bacterial genera ( Remigi et al., 2016 ). Within genera, however, transfer of these genes appears to occur frequently. Symbiotic genes often show far less genetic diversity than do core genes, suggesting very recent horizontal transfer ( Laranjo et al., 2008 ; Kumar et al., 2015 ), and congruent topologies of trees based on several different Nod genes ( nod , nol , and noe ) imply that these genes generally evolved together and are transferred among bacteria as a single unit ( Moulin et al., 2004 ), although exceptions are known ( Steenkamp et al., 2008 ). For example, two recently described species of Bradyrhizobium nodulating Jicama , Pachyrhizus erosus , in Central America closely resemble Bradyrhizobium strains isolated from soybean and Vicia for the nifH gene but are highly divergent from these strains for the nodD gene ( Ramírez-Bahena et al., 2009 ), implying different evolutionary histories for these two types of symbiotic genes in these species. Several events of apparently recent recombination have been observed between introduced and native microbes, such that the rhizobial bacteria are chimeric constructs, combining genetic features of co-introduced strains and strains derived from the bacterial communities at the site of introduction. Such recombination can be inferred from incongruence among the topologies of the gene genealogies of 16S and housekeeping genes on the one hand and symbiotic genes on the other for the same strains of bacteria. In Figure 2 we map out the possible types of recombination events and how they can be recognized from incongruences in the phylogenetic trees constructed from core or housekeeping genes versus symbiotic genes. Figure 2A shows the tree topologies consistent with the adoption of a new symbiont from the microbial community at the site of introduction that, however, bears the symbiotic machinery from the original bacterial partner. This is a pattern that makes ecological sense. If there is local adaptation in these soil microbes ( Kraemer and Boynton, 2017 ) conditions in the new habitat may well be less favorable for bacteria from the original distribution than for local bacteria. If a local bacterium, adapted to the local conditions in the site of introduction, were to acquire the symbiotic genes involved in the dialog with the host plant and nitrogen fixation in the specific environment of its nodule, this new chimera would be adapted both to local abiotic conditions at the site and to the biotic interaction with this host species. Figure 2B shows another possible scenario of recombination. FIGURE 2 Phylogenetic trace of co-introduction or acquisition of a new symbiotic partner together with recombination between bacterial strains to generate chimeric symbiotic partners with core genetic markers and symbiotic genes that show different phylogenetic patterns. (A) Depicts what we subsequently call scenario 3, with a new symbiotic partner acquired in the new geographic range that resembles some local bacteria for core genes but that has recombined with the original symbiont to acquire the symbiotic genes specific to the original interaction, and thus resembles the original symbiotic partner at its symbiotic genes. (B) Depicts the opposite scenario, hereafter referred to as scenario 4 with co-introduced symbionts that resemble the bacteria from the native range of the host plant for core genes but has recombined to acquire new symbiotic genes from bacterial species present in the introduced range. As in Figure 1 we assume some geographic pattern to relationships of the bacteria, with species from the same geographic range showing more similarity. There are several observations of introduced or invasive legumes nodulated by bacteria that are chimeras, bearing housekeeping genes of bacteria from the site of introduction but symbiotic genes of the habitual symbionts from the plants’ native ranges. Thus bacteria in the non-native ranges acquire the tools for interacting with introduced host plant from bacteria that must have been co-introduced. Indeed, genetic exchange requires that both types of bacteria, i.e., those from the native range of the host plant as well as those from the bacterial communities at the site of introduction co-occur at some point, hence presupposing co-introduction. For example, the invasive European Cytisus scoparius in the United States shows all three phenomena of co-introduction, acquisition of new symbiotic partners and chimeric bacterial symbionts. Some Bradyrhizobium species isolated from C. scoparius in the United States resemble those from European C. scoparius , some resemble strains that nodulate the native American Lupinus lepidus , indicating the acquisition of a native symbiont in the introduced range, and some are chimeric, with housekeeping genes characteristic of North American Bradyrhizobium but symbiotic genes identical to a single strain isolated from a C. scoparius growing in Spain ( Horn et al., 2014 ). R. pseudoacacia in China is also nodulated by bacteria that resemble members of the Chinese rhizobial community at a number of genetic markers. These strains, however, bear nodC genes characteristic of strains that nodulate R. pseudoacacia in its native range ( Wei et al., 2009 ), implying horizontal transfer of symbiotic genes from the habitual symbiont into the novel one adopted in the new distribution range. Other recombination patterns can also be observed, even in the absence of the habitual symbiotic partner, generating chimeric bacteria that combine features of two different bacteria both foreign to the introduced species. Acacia mangium , an Australian native has been planted throughout the tropics, where it can be nodulated by many local bacteria. In Brazil, at uninoculated sites this tree was nodulated by Bradyrhizobium that resemble B. elkanii strains from soybean, possibly from inoculum for their core genome. Their nodA gene sequences, however, resemble most closely bacteria that nodulate other Acacia species in Africa as well as other tropical legumes ( Perrineau et al., 2011 ). Thus chimeric nodulating bacteria may be combinations of only novel foreign partners for introduced legume species. Three of the most widely cultivated legumes are soybean, Glycine max , an Asian native, common bean, Phaseolus vulgaris , from Meso- and South America and peanut, A. hypogaea , originally from South America. Studies of the nodulating bacteria from soybean show both co-introduction and acquisition of novel bacterial strains from native plants in the introduced range. Soybean is usually inoculated at sowing, so finding apparently co-introduced rhizobia is unsurprising. For example, Shiro et al. (2013) , studying nodulating strains from soybean throughout the United States, found that all of their isolated strains clustered together with reference strains from inocula, suggesting little implication of rhizobial strains from the North American soil bacterial community. Similarly, many effective nitrogen fixing strains isolated from the promiscuous soybean cultivar developed for cultivation in Africa to be compatible with African rhizobial bacteria were identified as B. japonicum or B. elkanii ( Chibeba et al., 2017 ). On the other hand, some strains nodulating soybean in South Africa were distinct from known reference strains, suggesting that new strains had been adopted from the native soil community ( Naamala et al., 2016 ). Other studies of soybean-nodulating bacteria from Brazil, Paraguay, and Canada recovered some strains nodulating soybean that resembled neither inoculum nor reference strains and that most likely had been acquired from native American P. vulgaris in South America ( Chen et al., 2000 ; Hungria et al., 2001 ) or Amphicarpaea bracteata in Canada ( Tang et al., 2012 ). Common bean ( P. vulgaris ), a native of Central and South America, is now grown throughout the world as one of the most important legume crops. This promiscuous legume host can interact with a wide variety of bacterial species belonging mostly to the genus Rhizobium ( Amarger, 2001 ). Within its centers of origin in Meso- and South America, beans are nodulated mainly by Rhizobium etli ( Martínez-Romero, 2003 ). This bacterial species can also be found nodulating bean in introduced areas as widespread as Nepal ( Adhikari et al., 2013 ), Ethiopia ( Aserse et al., 2012 ), China ( Cao et al., 2014 ), Tunisia ( Mhamdi et al., 2002 ), Spain ( Rodriguez-Navarro et al., 2000 ), Iran ( Rouhrazi et al., 2016 ) and Jordan ( Tamimi and Young, 2004 ). Throughout this plant’s range several other species of Rhizobia can be recovered from nodules. These include several closely related Rhizobium species, R. tropici ( Martinez-Romero et al., 1991 ; Anyango et al., 1995 ; Grange and Hungria, 2004 ), R. leguminosarum ( Adhikari et al., 2013 ; Ribeiro et al., 2013 ), but also bacteria belonging to the genera Bradyrhizobium ( Han et al., 2005 ) and Sinorhizobium ( Mnasri et al., 2007 ; Zurdo-Piñeiro et al., 2009 ). The overriding dominance of R. etli on the common bean throughout the world argues for a co-introduction scenario, unsurprising since bean seeds are known to carry R. etli ( Pérez-Ramírez et al., 1998 ). On the other hand, the large spectrum of bacterial species that nodulate Phaseolus in its introduced ranges as well as their genetic distinctiveness indicates that this host species also regularly adopts new bacterial strains ( Mwenda et al., 2018 ), sometimes from other crop species ( Flores-Félix et al., 2018 ), in its area of introduction. In addition, bacterial strains nodulating Phaseolus can show incongruent phylogenetic relationships for housekeeping versus symbiotic genes, suggesting horizontal transfer of symbiotic genes ( Aserse et al., 2012 ). For example, some rhizobial strains isolated from nodules of the common bean on Hispaniola Island resemble no other known bean nodulators for several housekeeping genes though they bear the typical nodC alleles of R. phaseoli ( Díaz-Alcántara et al., 2014 ), suggesting adoption of new symbiotic partners that acquired the symbiotic genes adapted to interaction with this host. The peanut ( A. hypogaea L.) is the second most widely cultivated grain legume in the world, after soybean. This South America native, is nodulated by a large range of Bradyrhizobium and Rhizobium species in its native range ( Taurian et al., 2006 ; Muñoz et al., 2011 ). Peanut is now an important oil and seed crop in several regions of Africa and Asia, where it is nodulated by several newly described species of bacteria ( Wang et al., 2013 ; Grönemeyer et al., 2015 , 2016 ; Li et al., 2015 ). One of these new species, B. guangdongense , clusters with Bradyrhizobium isolated from Vigna unguiculata from Brazil ( Li et al., 2015 ), while the rest and additional strains from Ghana, resemble most closely B. yuanmingense originally isolated in China from Lespedeza , an Asian native, for housekeeping and 16S sequences ( Wang et al., 2013 ; Grönemeyer et al., 2015 , 2016 ; Li et al., 2015 ; Osei et al., 2018 ). This could suggest the acquisition of Asian bacteria after being introduced there, though how such strains arrived in Africa remains puzzling, as the strains isolated in Argentina, within the native range, also resemble these same species. Thus the source of bacteria that nodulate peanut in its introduced ranges in Asia and Africa remains unsolved and may point to a widespread cosmopolitan group of bacteria capable of effectively nodulating this promiscuous species. Isolates from Ghana appear to bear both a core genome and the symbiotic genes similar to those of B. yuanmingense from the host species Lespedeza ( Osei et al., 2018 ) but B. arachidis from China has divergent and distinct nifH sequences ( Wang et al., 2013 ), and both B. guangxiense and B. guangdongense have nodA genes identical to those reported from peanut isolates in Argentina ( Muñoz et al., 2011 ; Li et al., 2015 ) suggesting recombination between housekeeping and symbiosis genes that generate chimeric strains as depicted in Figure 2A . On the other hand, strains isolated from peanut in South Africa and Botswana resemble most closely an inoculum strain widely deployed in Africa and originally isolated from the African native Macrotyloma africanum at a number of housekeeping genes. The various nod-genes appear to be of diverse African and Asian origin, suggesting that recombination has occurred, but that the nod-genes do not correspond to those from the native bacteria that had evolved in association with peanut ( Steenkamp et al., 2008 ). Thus peanut nodulating bacteria also provide examples of unusual events that bring together novel, chimeric combinations of genes that had not previously evolved in the context of this host species, as depicted in Figure 2B . To further explore the interesting case of peanut nodulators, we constructed phylogenetic trees of 16S and housekeeping genes and compared these with trees constructed with the nodC and nifH gene sequences for those strains of peanut nodulators for which all sequence data were available. This allows us to explore the various scenarios found for these bacteria.",
"discussion": "Discussion What is the nature of the bacterial symbionts of the legumes that human agricultural practices have moved around the world? In the literature we found most examples of plant species that appeared to have been co-introduced with their nodulating bacteria, i.e., for which the strains isolated in the introduced range were most similar to strains from the original native range. This is unsurprising for legume species that are inoculated either systematically or when they are first planted in a new area ( Stȩpkowski et al., 2005 ) and soybean is systematically inoculated in large parts of its cultivation area ( Chang et al., 2015 ). The importance of soil bacteria as plant growth promotors and in nitrogen fixation in the root nodules of many legumes has been known since the late 19th century and inoculation was practiced even before people understood why ( Brown, 1918 ). This long history of moving soil together with plants could well explain why many introduced legumes can be nodulated by their habitual bacteria from their native range. Similarly, unintentional introduction or invasion that was initiated from soil contaminated with seed of a non-native species would have been likely to also include their nodulating bacteria. A co-introduction scenario is thus compatible with the many observations of introduced plants associating with bacteria that strongly resemble their native symbionts, even in the introduced range. On the other hand, the close similarity between bacteria that nodulate an introduced species in its native and introduced range could, alternatively, indicate that bacteria very similar to their habitual nodulators were already present at the site of introduction. Indeed, distinguishing between a scenario of co-introduction and the “Everything is everywhere but…” hypothesis of Baas-Becking ( de Wit and Bouvier, 2006 ) is virtually impossible. It was a far greater challenge to find examples of plants that were NOT co-introduced with their symbionts, or for which the symbiont in the introduced range resembled more closely the symbionts of other native species in that range. The single case we found, of M. pigra in Australia ( Parker et al., 2007 ), does not even completely meet our criteria, because we have no phylogenetic information for the Burkholderia that associate with other species in the introduced range. Nonetheless. it is clear that the lineages of this bacteria that associate with M. pigra in Australia are distinct from those in the neotropics, the native range of these plants. Co-introduction does not preclude additional associations with members of the new bacterial community. We found numerous examples of introduced plant species that associate both with their habitual symbionts but that have clearly adopted additional bacterial strains that are distinct from those found in their native range and that resemble bacteria from the soil community of the introduced range ( Chen et al., 2000 ; Ulrich and Zaspel, 2000 ; Hungria et al., 2001 ; Wei et al., 2009 ; Tang et al., 2012 ; Ndlovu et al., 2013 ; Chibeba et al., 2017 ). The more surprising result, perhaps, was the amount of recombination and lateral gene transfer evident in the bacterial symbionts of these plants. Such lateral gene transfers and exchanges of symbiotic genes is already discussed in the literature ( Steenkamp et al., 2008 ; Ramírez-Bahena et al., 2009 ; Wei et al., 2009 ; Muñoz et al., 2011 ; Perrineau et al., 2011 ; Aserse et al., 2012 ; Díaz-Alcántara et al., 2014 ; Horn et al., 2014 ). Our ability to even ask these questions about the frequency and nature of chimeric bacterial strains has been made possible by technological breakthroughs that allow rapid sequencing at low cost that provide sequence data on multiple genetic markers per individual. Indeed, few publications on bacterial diversity from before the current millenium presented data from multiple markers, which made it impossible to even address these questions. Based on the strains of Bradyrhizobium isolated from peanut for which we had adequate sequence data for our analyses, we found poor congruence between the tree based on 16S and housekeeping genes and both the nodC and nifH trees. Though such incongruence is not always found ( Jaiswal et al., 2017 ) it is a general finding for these and other strains ( Menna and Hungria, 2011 ; Zinga et al., 2017 ; Tong et al., 2018 ) implying widespread recombination with horizontal exchange of the symbiotic mechanism, as has been noted on both large and small scales in these bacteria ( Galibert et al., 2001 ; Kumar et al., 2015 ; Remigi et al., 2016 ). Indeed, from our analysis of these peanut-nodulating strains mostly from China, we found patterns that were consistent with co-introduction of bacteria from the native range, with adoption of new strains and novel associations in the introduced range and with recombination that generated new, chimeric symbionts. Surprisingly, we found no cases of recombination with adopted novel symbionts acquiring the symbiotic mechanisms from the habitual nodulating strains, which we had expected to be the usual case for chimeric bacteria. Indeed one would expect chimeric strains that combine the genes responsible for dialog with the host plant that are conserved from the original association, but with genes responsible for adaptation to the abiotic environment to be acquired from the local bacterial community at the site of introduction. Such a pattern has been found for bacteria associated with several plant species including peanut, with bacteria resembling species from the local rhizobial community in the introduced range for housekeeping genes but resembling the original nodulating bacteria from the plants’ native range for their symbiotic genes ( Wei et al., 2009 ; Díaz-Alcántara et al., 2014 ; Horn et al., 2014 ; Li et al., 2015 ; Wang et al., 2016 ). Here, on the other hand, nodC and nifH genes similar to those from the native range in South America were not found in other genetic backgrounds of bacteria adopted from the local communities. However, there were cases of peanut-nodulating bacteria that clustered together with bacterial strains from their native South America for housekeeping genes but, for their symbiotic genes, were most similar to strains that have symbiotic genes like those of bacteria nodulating soybean. For example, B. japonicum was isolated from soybean and B. arachidis , a bacteria associated with peanut first described by Wang et al. (2013) , has nodC and nifH genes that cluster with strains from soybean. Thus we identified bacterial symbionts of peanut in China that combined the housekeeping genes of an introduced bacterium foreign to Chinese soils with symbiotic genes new for peanut. One group of peanut-nodulating bacteria isolated in China formed a clade, on the species tree, with B. subterraneum isolated from peanut and B. vignae , isolated from V. unguiculata , both from Namibia, marked as “African association” on Figures 3 , 4 . For the nodC and nifH genes the group of peanut isolates formed an isolated clade on its own. The close relationship with bacterial isolates from Namibia poses an interesting problem. Either these strains colonized China via southern Africa, or vice versa. Peanut cultivation in China may be very old. Though peanut originated in the new world, and as such should have been unknown outside of the Americas before the Columbian Exchange, it is unambiguously mentioned in Chinese horticultural literature in the early sixteenth century, which seems too soon to have followed a European route of introduction to China. This suggests that peanuts may have arrived in China in pre-Columbian times, possibly via a western route out of South America and Chinese mariners’ voyages of discovery of the islands of the South China Sea ( Ho, 1955 ). Thus it is not impossible that peanut was first cultivated in China and subsequently moved to Africa, possibly bringing with it the bacteria it had adopted in China. This analysis and interpretation of the available genomes of peanut-associated rhizobia, though a valid first step, has failed in clearly elucidating the origin of the bacterial symbionts of peanut in China, let alone in other introduced areas. Furthermore, though we were able to identify events of host shift and recombination, our approach was mainly qualitative, i.e., particular events could be observed or not. Understand the evolutionary forces and processes that shape this complex interaction between host and symbiont would necessitate a more quantitative approach to determine how often particular phenomena occur. The latter requires a population genetics approach, which should become increasingly accessible through technological advances. We are left with a large number of intriguing questions about peanut-nodulators, which of course can be posed for any species of legume that has spread beyond its native range: What is the origin of the symbiotic bacteria, and for which genes, the core genome or the symbiotic genes? Can bacterial phylogeography elucidate patterns of agricultural exchange? How do abiotic conditions influence host range for bacteria and symbiont choice for the plants? How broad is the host range of bacterial strains that can nodulate peanut. How interchangeable are symbiotic genes? How large is the range of core genomes in which symbiotic genes can function? Does the compatibility between core and symbiotic genes influence symbiotic efficiency? Is the unexpected extent of the recombination and the acquisition of nitrogen-fixation genes from divers sources an exception? Is the case of peanuts unusual or are many symbiotic bacteria comprised of chimeric populations of such diverse origins for symbiotic as well as core genes? Addressing these questions would require far more extensive sampling of bacteria that nodulate peanut as well as a large range of native species that could harbor the sources of peanut-nodulating bacteria in the introduced ranges. Additional sampling of bacterial symbionts of a range of host plants as well as of peanuts from more introduced areas would improve the scope of the study and allow us to explore the intriguing patterns we uncovered of rampant recombination in all directions in the bacteria nodulating this promiscuous host species in its introduced range. Peanut rhizobia, at least in China, appears to present a convoluted history of host shift and recombination to an unusual degree. Only more systematic study of peanut, but also other bacterial symbionts of widespread crop and introduced species confronted with bacterial diversity with which they have not evolved, can answer whether this is a rare or a habitual occurrence. Though the complete sampling of adequate representatives of all potential sources of bacteria is probably out of reach, we hope however, to stimulate a more systematic approach to this question."
} | 9,604 |
36711600 | PMC9882018 | pmc | 9,221 | {
"abstract": "Spatial structure within microbial communities can provide nearly limitless opportunities for social interactions and are an important driver for evolution. As metabolites are often molecular signals, metabolite diffusion within microbial communities can affect the composition and dynamics of the community in a manner that can be challenging to deconstruct. We used encapsulation of a synthetic microbial community within microdroplets to investigate the effects of spatial structure and metabolite diffusion on population dynamics and to examine the effects of cheating by one member of the community. The synthetic community was comprised of three strains: a ‘Producer’ that makes the diffusible quorum sensing molecule ( N -(3-Oxododecanoyl)-L-homoserine lactone, C12-oxo-HSL) or AHL; a ‘Receiver’ that is killed by AHL and a Non-Producer or ‘cheater’ that benefits from the extinction of the Receivers, but without the costs associated with the AHL synthesis. We demonstrate that despite rapid diffusion of AHL between microdroplets, the spatial structure imposed by the microdroplets allow a more efficient but transient enrichment of more rare and slower growing ‘Producer’ subpopulations. Eventually, the Non-Producer population drove the Producers to extinction. By including fluorescence-activated microdroplet sorting and providing sustained competition by the Receiver strain, we demonstrate a strategy for indirect enrichment of a rare and unlabeled Producer. The ability to screen and enrich metabolite Producers from a much larger population under conditions of rapid diffusion provides an important framework for the development of applications in synthetic ecology and biotechnology.",
"conclusion": "Conclusion Understanding how natural and synthetic microbial communities arise and how to program them for specific tasks would accelerate innovations in biotechnology, medicine, and biomanufacturing. Spatial structure is an important aspect of natural multi-strain communities and contributes strongly to their robustness and diverse phenotypes 35 - 41 . Earlier studies have shown that when secreted signals or cell types are highly constrained (low or no diffusion) within microdroplets, slow-growing but desirable high-yielding cells producing valuable biomolecules can be enriched efficiently 23 , 34 . However, it is likely that many small molecules of interest, especially signaling molecules or antimicrobials could readily diffuse between microdroplets and alter population dynamics. To engineer both natural and synthetic microbial communities will require an understanding of the molecular principles of social interactions as well as strategies to shape communities to desired characteristics. We used a three-member microbial synthetic community in which the Producers secrete AHL as signaling molecules that can diffuse between the microdroplets to investigate the role of signaling within a defined synthetic community and develop experimental and mathematical modeling approaches to control emergent Producer population dynamics. Using emulsions of microdroplets as individually compartmentalized microenvironments, we were able to quantitatively assess and model the interplay of population dynamics when diffusible signals are present. Of particular importance was the development of strategies that could allow for the enrichment of valuable Producer subpopulations despite moderate to fast diffusion. The production of valuable biomolecules or controlling phenotypes of synthetic communities for translation applications like bioremediation will typically incur fitness costs to the Producer that invite the rise of cheating Non-Producers. We showed that during serial propagation in a spatially structured environment, the window of opportunity for the selection of the AHL Producer population is small as the Non-Producers can rapidly outcompete Producers. This is caused by rapid diffusion of AHLs, which, in turn, leads to reduced Receiver competition in all the microdroplets. While AHL production is metabolically costly to the Producers, the benefits of reduced Receiver competition are shared by both the Producers and the Non-Producers alike. In the absence of spatial structure, as in the case of a well-mixed batch culture, Non-Producers rapidly outcompete both Producers and Receivers. Even with fast diffusion, the spatial structure imposed by microdroplets did slow the rise of Non-Producers. Unlike in batch culture, the Non-Producers within microdroplets are limited by the physical confinement of the microdroplet and cannot access the entire culture volume of the emulsion and therefore expand more slowly. The mathematical model developed using this microbial synthetic system and our experimental findings showed that Producer success in the community when AHL diffuses among the microdroplets is transient. This may seem counterintuitive as it might have been expected that the community might reach a stable steady-state. Our data clearly show that in the case of rapid diffusion, cheaters (Non-Producers) will eventually force the Producers and Receivers to extinction. Importantly, the simulation also suggested that the window of opportunity for Producer enrichment and isolation can be affected by different environmental factors such as initial population size, AHL secretion rate, sensitivity of the population to the signal, composition of the droplets, the duration of the interaction, and dilution factor. Our results pointed toward several parameters that could be adjusted to increase the probability of extending the window of opportunity for the AHL Producer to succeed. We tested predictions of the simulation and showed that we could extend the success of the Producers by reducing AHL production, and by using Fluorescence Activated Droplets Sorting (FADS) sorting. Using FADS to sort for microdroplets with lower green fluorescence (Receivers), we observed a 50-fold improvement for the enrichment of Producers. We deliberately chose to sort the microdroplets based on Receiver fluorescence (indirect) rather than the Producers (direct, the cell type of interest) with a specific application in mind. Translational applications of synthetic communities to produce valuable molecules will likely make use of the abundant diversity of both natural and engineered microbial life. Indirectly enriching for Producers within the context of fast diffusion could be used to enrich for wild strains of bacteria without the need for genetic manipulation such as introducing a fluorescent reporter. By avoiding genetic manipulation of the Producer or strain of interest, we can dramatically increase our ability to evaluate wild strains and communities rapidly and at scale. Further, if we define a Producer as a strain, strain variant, or community that produces a valuable activity or molecule(s), then using an indirect approach to enrich Producers allows an investigator to engineer a reporter strain or molecule that reports on the desired activity. In experiments where the population of interest is labeled, it can be sorted directly 33 , 42 - 44 , and therefore population ratio increases dramatically. By evaluating an indirect method, we develop opportunities to screen for bacteria of interest based on their activity towards a directed task. Given sufficient accuracy of the sorter, a better enrichment of Producers can be achieved through increasing the sorting threshold and/or iterating the growth cycles with sorting. Since the indirect sorting is done based on the survived Receiver’s fluorescence in droplets, the upper limit of the sorter for enrichment is bounded by the number of Producers that’s high enough to kill all Receivers in the system. Therefore, when enriching for Producers with iteration, one can increase the Receiver cells spiked-in to further extend the enrichment window for Producers. With accurate sorting and adaptive Receiver spike-in, the platform is theoretically capable of enriching for 1 Producer from 1 million Non-Producer cells. In summary, we have tested and validated that enrichment of rare but potentially valuable subpopulations within a much large population of competitors and cheaters can be achieved even when the desired rare subpopulation is not directly labeled. As microfluidic technologies for microdroplet production, fluorescence sorting, and microdroplet capture improve, new opportunities for selection and enrichment of rare strains and synthetic microbial communities with desirable phenotypes will also improve.",
"introduction": "Introduction Locey and Lennon estimated microbial biodiversity on Earth to be as high as 1 trillion species 1 . In nature, microbes live in multispecies communities where they compete or cooperate in the utilization of resources within a localized and often structured environment 2 , 3 . Microbial communities comprise the predominant social structures of life and are critical players in our global ecosystems, yet their potential as contributors to a sustainable future remains largely untapped. Social microbial interactions are dynamic and vary in response to a variety of environmental cues 2 , 4 . Factors such as spatial patterning, resource availability, and metabolite diffusion are major drivers of microbial community structure across ecological contexts 2 , 5 , 6 . Understanding the role of metabolites both as resources, and often as signals, within microbial communities is challenging especially when diffusion plays a significant role in shaping the population dynamics of the community. A quantitative understanding of how diffusion within structured systems affects population dynamics can significantly advance the use of synthetic ecologies for ecological, evolutionary, or biomedical applications 7 - 10 . Recent studies of natural microbial interactions have provided key insights into the role of community structure, metabolites, and metabolic networks 11 , 12 . Such studies can be difficult because of the complexity, inaccessibility, and difficulty of genetic manipulation of highly complex natural communities such as those found in the gut, soil, or aquatic microbiomes 13 , 14 . Therefore, to provide more tractable experimental approaches for quantitative modeling of community interactions, researchers have begun to use synthetic biology to produce synthetic ecologies using model organisms such as Escherichia coli \n 5 , 10 , 15 - 18 . Using synthetic biology tools, studies are developed to resemble key features of natural ecosystems such as competition or predator-prey interactions in terms of logic and dynamics 5 , 19 . Such model systems can be used to develop predictive models and principles to explore the ecological, functional, and key structural features of a community 16 , 20 . To assess how diffusible molecular signals, such as AHL, affect population dynamics, we constructed a synthetic community of three competing E. coli strains: 1) a ‘Producer’ that synthesizes and secretes AHL; 2) a ‘Non-Producer’ that does not synthesize or respond to AHL, and 3) a ‘Receiver’ that dies in the presence of AHL. Here, the Producer incurs the fitness cost associated with producing the AHL 21 . In a well-mixed environment such as test tubes, the molecules secreted by the Producer are shared by all the members of the population through diffusion, creating an opportunity for cheating by the Non-Producers who do not pay the cost of production, but benefit from the reduced Receiver population 21 , 22 . However, in a structured environment like microdroplets with reduced permeability of AHL signals, cheating by Non-Producers encapsulated in other nearby microdroplets is minimized. Thus, Producers can preferentially benefit by killing the Receiver despite the cost. Bachmann and colleagues demonstrated a successful use of spatial segregation to identify more efficient, but slower growing variants of Lactococcus lactis by serial propagation of the cultures in microdroplets 23 . To better understand the population dynamics of a community subject to social interactions affected by diffusible metabolites, we encapsulated synthetic communities comprised of the engineered Producers, Non-Producers, and Receivers under various initial conditions. We also set out to identify conditions where a less fit Producer strain that is not labeled could be recovered from a much larger population (which we model as Non-Producers). Indirect Producer enrichment within the context of fast diffusion can be useful for studies in which investigators seek to enrich for wild/unlabeled strains as Producers of valuable biomolecules like antimicrobials. Our results demonstrate that in the presence of AHL, we observe an increase in the Producer population at the expense of Receivers that die allowing access to more nutrients by the Producers. However, as the number of Producers increases, so does the secreted AHL. The diffusible AHL favors the rise of Non-Producers, which benefit from the metabolites at no cost to their fitness. We developed a mathematical model to examine how different levels of effective diffusion across the microdroplets affects the rate of Producer enrichment. Guided by the modeling results and using precisely controlled bottom-up experiments of simple communities, we also evaluated the effect of initial population ratios, concentrations of metabolite produced, and duration of incubation on population dynamics. Finally, we propose a microfluidic platform incorporating sorting to enrich a rare Producer population secreting diffusible metabolites within a larger mixed community. Insights from this research provide design principles for community studies with unknown or varying degrees of metabolite diffusion in structured environments and guide in screening for rare subpopulations of microbes in both natural and engineered communities."
} | 3,461 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.