pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
38946907 | PMC11211263 | pmc | 4,435 | {
"abstract": "Bryophytes, the second-largest group of plants, play a crucial role as early colonizers of land and are a prolific source of naturally occurring substances with significant economic potential. Microorganisms, particularly bacteria, cyanobacteria, fungi form intricate associations with plants, notably bryophytes, contributing to the ecological functioning of terrestrial ecosystems and sometimes it gives negative impact also. This review elucidates the pivotal role of endophytic bacteria in promoting plant growth, facilitating nutrient cycling, and enhancing environmental health. It comprehensively explores the diversity and ecological significance of fungal and bacterial endophytes across various ecosystems. Furthermore, it highlights the moss nitrogen dynamics observed in select moss species. Throughout the review, emphasis is placed on the symbiotic interdependence between bryophytes and microorganisms, offering foundational insights for future research endeavors. By shedding light on the intricate bryophyte-microorganism associations, this study advances our understanding of the complex interplay between plants, microbes, and their environment, paving the way for further research and applications in environmental and biotechnological realms.",
"conclusion": "6 Conclusions As a whole, the analysis of bryophyte-microorganism associations offers a sophisticated comprehension of the complex interactions that occurs between bryophytes and a range of microorganisms, such as cyanobacteria, fungus, and bacteria. This review clarifies the significant influence of these relationships on environmental optimization by carefully analyzing the literature and reveals exciting prospects. It is shown that these relationships are critical for the resilience and productivity of ecosystems in processes including organic matter decomposition, nitrogen fixation, and nutrient cycling. Further investigation into bryophyte-microorganism associations is imperative for advancing our comprehension of their intricate ecological roles and interactions. These studies have the potential to unveil profound insights into ecosystem dynamics, biodiversity, and the resilience of natural habitats. Understanding these complex relationships is vital for maintaining ecological equilibrium and addressing the pressing environmental challenges of our time. These applications highlight the transformative possibilities of leveraging these natural symbioses. To harness the full ecological benefits of bryophyte-microorganism associations for promoting sustainable environmental solutions, it is imperative to continue intensive research. Ongoing studies are essential not only to understand the complex dynamics of these interactions but also to develop innovative applications that address pressing environmental challenges. The expanding body of research indicating that fungi and other microbes that are symbiotic or otherwise associated with bryophytes affect their growth and development is not perfectly addressed in this study. It is important to remember that these unnoticed connections may have a significant impact on experimental research.",
"introduction": "1 Introduction Bryophytes, belonging to the Bryophyta division, boast an incredible diversity, with over 23,000 species distributed worldwide. Classified into three main groups - mosses (class Bryopsida), liverworts (class Hepaticopsida), and hornworts (class Anthocerotopsida), these plants thrive in a wide range of environments, making them crucial components of ecosystems (Bahuguna et al., 2013 ). They are considered the second most diverse group of plants after flowering plants, and are believed to be among the oldest terrestrial plants (Clarke et al., 2011 ). Being early colonizers of land, bryophytes faced numerous challenges, including pathogen attacks and insect predation, due to their exposure to adverse environmental conditions (Whitehead et al., 2018a ). Symbiotic relationships between bryophytes and microorganisms, including fungi, bacteria, and algae, are vital for the ecological functions and survival strategies of these plants. Fungi aid in nutrient absorption, especially in nutrient-poor soils, while nitrogen-fixing bacteria enable bryophytes to thrive in nitrogen-deficient environments. Algae, particularly cyanobacteria, contribute to photosynthesis, enhancing the production of organic compounds crucial for bryophyte growth. These symbiotic associations increase resilience to environmental stresses, enabling bryophytes to occupy diverse habitats worldwide, from rocky terrains to polar regions, and play essential roles in ecosystem functioning (Adams and Duggan, 2008 ; Rimington et al., 2018 ; Glime, 2019 ; Poveda, 2020 ). Microorganisms like algae (including cyanobacteria), bacteria, and some macro fungi are associated with bryophyte including all three classes i.e., liverwort, moss and hornwort. Bryophytes may potentially be parasitized by a wide variety of fungus, lichens, and microbes (During and Tooren, 1990 ). Since the majority of mosses are ectohydric, the gametophytes can absorb water and dissolved minerals onto their surfaces. The moss leaf surface is comparable to the rhizosphere in this manner. This could be among the factors responsible for the colonization of microorganisms (Opelt and Berg, 2004 ). Additionally, the related population of microorganisms varies according to the host's needs; for example, cyanobionts are in variable phases of forming symbiosis in different sections of the thallus, as shown by varying heterocyst frequency and enzymatic activity, this leads to metabolically diverse cyanobiont populations (Rai et al., 1989 ). Unlike flowering plants, liverworts are more often linked to ancient lineages of arbuscular mycorrhizal fungus, while being some of the closest extant relatives of the first land plants (Rimington et al., 2018 ). An essential component of the nitrogen economy of terrestrial arctic habitats is the biological fixation of atmospheric nitrogen by cyanobacteria associated with mosses (Solheim et al., 2004 ). The interactions that bryophytes have with a wide range of species are diverse and can range from obligatory symbioses to sporadic epiphytism. Over time, a qualitative understanding of the many kinds of interactions, the species involved, and certain structural traits of the connections have been described (During and Tooren, 1990 ). The greatest source of naturally occurring substances with potential economic value is microorganisms. In addition to being a source of polyunsaturated fatty acids, endophytes, such as fungi and bacteria that intracellularly colonize plant tissues, are known to be a rich source of new compounds, including anticancer drugs, antibiotics, antivirals, antioxidants, and immunomodulatory substances. (Brady and Clardy, 2000 ; Wrigley, 2000 ; Strobel and Daisy, 2003 ; Bérdy, 2005 ). In plants, endophytes are a widespread world. Over the course of long-term coevolution, they have developed a mutually beneficial connection with host plants. A complex microecosystem is made up of the diverse range of microbial species that make up the endophytic community. Numerous studies have demonstrated that endophytes directly produce bioactive compounds that protect their host plants from harmful microorganisms and herbivores, increasing the fitness of the host plants (Stelmasiewicz et al., 2023 ). It's possible that endophytic fungi are abundant sources of naturally occurring bioactive substances that have use in the pharmacological, medical, and agricultural sectors (Govindan and Venkatesan, 2022 ). The endophytic bacteria support the growth of host plants by fortifying their resilience to both biotic and abiotic stressors (Stelmasiewicz et al., 2023 ). Certain chemicals found in bryophytes have antibacterial and antibiotic qualities (Bahuguna et al., 2013 ). So, the use of bryophytes as a source of microorganisms to enhance crops and/or forest species is now understudied (Poveda, 2024 ). This review sheds light on the significant yet often overlooked associations between bryophytes and microorganisms. Exploring the bryophyte-microbe association would help in better understanding of the terrestrial ecosystems. Future research can prioritize this area by employing molecular techniques, investigating environmental influences, and integrating computational modeling. Such efforts deepen our understanding of ecological networks and symbiotic interactions, offering avenues for environmental sustainability and biotechnological innovation. The flow diagram ( Figure 1 ) of article analysis illustrates the systematic process of evaluating and synthesizing information from scholarly articles. It typically begins with the identification of relevant articles through database searches, followed by screening for eligibility based on predefined criteria. The next step involves assessing the quality and relevance of the selected articles, often through methods such as critical appraisal or risk of bias assessment. Subsequently, data extraction is performed to gather key information from each article, such as study design, participant characteristics, interventions, outcomes, and results. Figure 1 Flow diagram of article analysis [adapted from Chen and Nelson ( 2022 )]."
} | 2,321 |
39880896 | PMC11779866 | pmc | 4,437 | {
"abstract": "Collective behavior in biological systems emerges from local interactions among individuals, enabling groups to adapt to dynamic environments. Traditional modeling approaches, such as bottom-up and top-down models, have limitations in accurately representing these complex interactions. We propose a novel potential field mechanism that integrates local interactions and environmental influences to explain collective behavior. This study introduces dynamic potential fields, where individuals perceive and respond to local potential fields generated by environmental cues and other individuals. We develop a mathematical framework combining distributed learning and swarm control to simulate and analyze collective behavior under varying conditions. Our simulations span a variety of environmental conditions, including standard environments where organisms interact under typical conditions, high noise environments where interactions are disrupted by random fluctuations, high density environments with increased competition for space, high risk environments featuring areas of strong negative potential field, and multiple resource environments with varying degrees of resource availability. These simulations demonstrate the adaptability and resilience of biological groups to changing and challenging conditions. Results reveal how potential fields facilitate the emergence of stable and coordinated behaviors, providing insights into self-organization, cooperation, and competition in nature. This framework enhances our understanding of collective behavior and has implications for bio-robotics, distributed systems, and complex networks.",
"conclusion": "Conclusion This research presents a comprehensive exploration of dynamic potential field mechanisms as a model for understanding and simulating collective behavior in biological systems. Through extensive simulations, the study demonstrates the capacity of potential fields to capture the adaptive and self-organizing characteristics inherent in natural groups, reflecting behaviors observed in standard, noisy, dense, high-risk, and resource-rich environments. By integrating distributed learning and local interaction principles, the model offers a scalable framework that mirrors real-world biological phenomena, such as coordinated flocking, schooling, and risk-averse clustering. The findings align with empirical data, reinforcing the validity of the potential field approach as a tool for exploring the complexities of collective motion. Beyond biological insights, this research contributes valuable methodologies for the advancement of bio-robotics and distributed systems. The development of recursive algorithms and adaptive gain sequences highlights the potential for real-time behavioral adjustments in autonomous agents, facilitating enhanced synchronization and resource utilization. The adaptability and robustness demonstrated by the potential field model underscore its applicability in fields ranging from ecological management to autonomous robotics, offering a pathway for designing resilient, decentralized systems that emulate the efficiency of biological collectives. Finally, this study bridges the gap between theoretical modeling and practical application, laying the groundwork for future research that further explores the intersection of biological principles and technological innovation in collective behavior and swarm intelligence. Limitations and future work A critical limitation of the current model is its inability to account for non-linear or counterintuitive responses to multiple external stimuli. For example, when individuals encounter a combination of attractive stimuli, such as food odor and conspecific odor, the combined effect may not necessarily enhance attractiveness. Instead, it can lead to unexpected outcomes, such as reduced attractiveness or even repulsion. Studies by Günzel et al. 50 and Salazar et al. 51 provide compelling evidence of these phenomena. Günzel et al. demonstrated that cockroaches avoided vanillin-scented shelters when combined with social odors, even though both stimuli were individually attractive. This counterintuitive behavior was attributed to changes in neural activity within the antennal lobe, where social odor signals suppressed the response to vanillin. Similarly, Salazar et al. observed that while isolated cockroaches preferred vanillin-scented shelters, group contexts resulted in a collective preference for unscented shelters. This shift was driven by reduced attraction between individuals in the presence of vanillin, leading to aggregation in unscented shelters and an inversion of collective preference. These findings emphasize the limitations of our model, which assumes an additive response to stimuli in the potential field mechanism. Such an assumption simplifies the interplay between competing stimuli, potentially overlooking the sensory interference, inhibitory effects, or social modulation that shape real-world behaviors. While the current model performs well in environments where stimuli are distinct and non-overlapping, its predictive accuracy diminishes in scenarios involving overlapping, interacting, or competing stimuli. To address this limitation in future work, several strategies could be explored. First, interaction terms or weighting factors could be incorporated into the model to account for the inhibitory or competitive effects between stimuli. This would allow for a more nuanced representation of how individuals respond to mixed stimuli. Second, incorporating mechanisms inspired by neurobiological findings, such as sensory interference observed in Günzel et al. could enhance the model’s ability to simulate such behaviors. Finally, context-sensitive decision-making rules could be introduced to reflect the dynamic shifts in individual and collective behavior observed in studies like Salazar et al., where group size or social context influenced responses to environmental cues. These improvements will not only increase the applicability of models to more complex environments, but will also provide a more realistic representation of how competing stimuli affect individual and collective behavior in biological systems. Future research will expand the model’s applications to broader contexts and further refine its predictive accuracy through empirical testing. While theoretical extensions will explore applications in bio-inspired robotics, swarm intelligence, and distributed systems, an essential component of future work will be experimental collaborations with biologists. These collaborations will focus on testing the model’s predictions under controlled field conditions to ensure alignment with real-world behaviors. By partnering with biologists, we aim to conduct experiments that replicate the environmental variables and interaction dynamics found in natural systems, such as animal group formations under predation threat or resource-driven clustering behaviors. Controlled studies will allow for systematic testing of the model’s assumptions about interaction rules, adaptive clustering, and role differentiation, providing critical data to validate or adjust the model as necessary. These experiments will be designed to test key model predictions, such as the emergence of criticality in response to external stimuli and the adaptive flexibility of collective behavior under changing environmental conditions. For example, experiments may focus on measuring response times, cohesion, and resilience of animal groups in response to artificially induced perturbations that simulate high-risk or resource-limited environments. This empirical approach will not only validate the model’s core predictions but also enable adjustments that enhance its utility for specific ecological and technological applications. Through these experimental validations, future research will bridge theoretical modeling with empirical evidence, enabling a deeper understanding of collective behaviors and ensuring that our model remains robust and versatile across diverse real-world scenarios. This combined approach will enhance the model’s potential applications in fields requiring adaptive, bio-inspired solutions, including robotics, environmental monitoring, and networked systems.",
"introduction": "Introduction In nature, individual organisms form complex collective behaviors through local social interactions, enabling groups to adapt to changing environments 1 – 3 . Collective behavior is common at the molecular, cellular, and organismal levels, from flocks of birds and insect task allocation to the coordinated movement of cell groups during embryonic development 4 . The mechanisms of collective behavior in biology involve various complex interactions and feedback processes 5 . Individuals perceive their surroundings and the states of other individuals through visual, auditory, and chemical signals, making behavioral decisions accordingly 6 . These local interactions can lead to global collective behavior 7 . Different biological groups rely on different perception and communication methods, resulting in diverse and complex mechanisms of collective behavior 8 . Currently, there are two main approaches to studying collective behavior: bottom-up models and top-down models 9 . Bottom-up models describe individuals and their interactions, revealing how local interactions lead to the emergence of collective behavior 10 , 11 . However, these models are often overly simplified, making it difficult to accurately describe the actual mechanisms and interactions in real biological systems 12 . Typical bottom-up models include agent-based models (ABMs), such as the Boids model 13 , 14 . The Boids model employs simple local rules that govern individual agents’ behaviors based on their immediate surroundings, enabling the emergence of complex collective behaviors without central coordination. By defining basic rules for separation, alignment, and cohesion, the Boids model effectively simulates the dynamic formations observed in natural groups, such as bird flocks and fish schools, illustrating how local interactions lead to the global organization of the group. These models define simple local rules and successfully simulate the formation processes of fish schools, bird flocks, and other collective behaviors 15 . However, the complexity of interactions in real biological systems limits the assumptions and simplifications of these models 16 . On the other hand, top-down models directly describe group-level behavior. They effectively simulate and predict the functions of collective behavior but lack explanatory power regarding the underlying causes 17 . For example, hydrodynamic models simulate the overall movement patterns of bird flocks and fish schools through the assumption of a continuous medium, but they struggle to reveal the interaction mechanisms at the individual level. Additionally, the complexity and diversity of biological systems, along with the sensitivity of individual behavior to environmental changes, make it difficult for a single type of model to be universally applicable. To overcome these limitations, we introduce the concept of potential fields 18 – 20 . The formation of potential fields arises from local interactions between individuals and environmental influences. These interactions can be physical contact, visual signals, chemical signals (such as pheromones), or acoustic signals. For example, ants release pheromones to mark food sources, creating a chemical potential field that guides other ants to the resource. The pheromone trail forms a gradient, with higher concentrations near the food and lower concentrations along the path, attracting ants to follow it. This self-reinforcing loop of pheromone deposition and attraction enables the colony to efficiently locate and exploit food sources 21 . Fish perceive the position and movement of nearby individuals through vision, forming a visual potential field 22 . Organisms sense this potential field through their sensory systems and adjust their behavior based on the intensity and direction of the potential field. This sensing process is usually local, meaning that individuals can only perceive changes within a certain range around them, rather than the distribution of all potential fields. This local perception leads to spatial and temporal limitations in individual behavior adjustments, which in turn generates complex collective behaviors 23 . Therefore, the guiding role of potential fields in individual behavior is a core mechanism for the formation of collective behavior 24 . When individuals perceive a potential field, they move towards areas with higher potential field intensity or adjust their direction accordingly. For instance, when a food source creates an attractive potential field, individuals tend to move towards it, leading to spatial aggregation. This guiding effect is seen not only in individual attraction to resources (like food) but also in the tendency towards the group’s center. In fish schools and bird flocks, individuals are attracted not only to food sources but also to the overall movement trend of the group. This dual guiding effect enables individuals to maintain a certain distance from each other while forming a cohesive group structure. The guiding role of potential fields affects not only individual behavior but also generates self-organizing effects through interactions and feedback mechanisms 25 , 26 . This self-organizing effect refers to individuals spontaneously forming ordered group structures and behavior patterns through local interactions without central control 27 . For example, in ant foraging, individual ants move along pheromone gradients after sensing the potential field, reinforcing the field when returning to the nest, thus strengthening the potential field 28 . This positive feedback mechanism allows ant colonies to efficiently find and exploit food sources, forming stable foraging path networks. Potential fields are not static, they dynamically adjust with individual behavior and environmental changes 29 . In biological groups, individuals constantly perceive and respond to changes in the potential field, giving the field high flexibility and adaptability 30 . Environmental changes (such as shifts in food sources or predator appearance) lead to the redistribution of potential fields, prompting individuals to adjust their behavior accordingly, enabling the group to adapt flexibly to environmental changes 31 , 32 . For instance, bird flocks adjust their flight routes and speeds based on wind direction, terrain, and other environmental factors, maintaining group coordination through visual and acoustic signals. This dynamic adjustment allows bird flocks to maintain high collective coordination during long migrations, enhancing survival rates 33 . In addition, the guiding role of potential fields influences not only individual movement trajectories but also the overall characteristics of collective behavior patterns 34 . Depending on the nature of the potential field and the response rules of individuals, groups can form various behavior patterns, such as aggregation, dispersion, and synchronized movement 35 . For example, fish schools quickly form tight, ball-shaped groups when threatened by predators, increasing individual survival chances through collective action 36 . Bees form stable foraging path networks by following and reinforcing pheromone trails, efficiently utilizing food resources. Thus, introducing the potential field mechanism not only explains the formation process of biological collective behavior but also reveals the role of individual interactions and environmental factors in collective behavior. This mechanism provides a unified theoretical framework for understanding biological collective behavior and has been validated in various biological systems 37 . Potential fields can also explain more complex collective behaviors. In social insects like ants and bees, information transfer and task allocation among individuals through chemical signals and physical contact create a dynamic potential field, enabling the entire group to efficiently complete tasks like foraging and nest building 38 . This potential field mechanism enhances group coordination and allows flexible responses to environmental changes and resource distribution uncertainties 39 , 40 . While potential field mechanisms offer an alternative approach, they fundamentally represent a specific type of bottom-up model. As with traditional bottom-up approaches, potential fields rely on local interactions and sensory inputs that direct individual agents’ behaviors through environmental cues. Emphasizing local mechanisms, potential fields illustrate how individual interactions foster emergent group dynamics, aligning with bottom-up principles and providing an adaptive, flexible model for collective behavior. To enhance the adaptability and responsiveness of individual agents within a swarm, we introduce a combined approach of distributed learning and swarm control. Distributed learning refers to the process through which individuals, such as robots, animals, or autonomous systems, adjust their behavior based on local interactions with their environment and other individuals. This process is autonomous, meaning each individual independently modifies its behavior without the need for external control or recalibration. Rather than relying on fixed rules or centralized instructions, individuals learn from ongoing experiences and interactions, allowing them to adapt to their surroundings in real time. A central feature of distributed learning is the autonomy of each individual. Decisions are made based on local, real-time feedback, which can stem from the environment or from interactions with other individuals. This feedback may include changes in environmental conditions (e.g., temperature, obstacles, or resource availability) or social cues from other individuals (e.g., movement patterns, pheromone trails, or cooperation signals). Through such interactions, individuals continually refine their behavior, leading to progressively more effective outcomes. Another key characteristic of distributed learning is its dynamic adaptability, which allows the system to continuously adjust its behavior in response to changing conditions. In contrast to static models that rely on fixed rules, distributed learning enables individual individuals to modify their decision-making processes based on real-time feedback. This adaptability enhances the system’s responsiveness to unforeseen circumstances, such as fluctuations in environmental factors or shifts in inter-individual communication, ensuring its continued effectiveness and efficiency, even in the face of uncertainty or complexity. By allowing individuals to learn from their experiences and adjust their strategies, distributed learning fosters more flexible and resilient collective behavior, thereby improving the system’s ability to tackle dynamic challenges and optimize performance over time. Thus, distributed learning improves upon traditional bottom-up models, which typically depend on static, predefined rules to govern behavior in response to immediate stimuli. The inherent flexibility of distributed learning enhances the system’s adaptability, enabling it to respond more effectively to dynamic and evolving conditions. This aligns with the principles of swarm control, which focus on coordinating individual actions to achieve cohesive collective behavior. Together, these approaches create a system where agents respond more effectively to varying environmental conditions, making the swarm more flexible and resilient. This integration is essential because distributed learning empowers agents to learn optimal responses to specific environmental cues in real time, thereby refining their local interactions. This is particularly valuable in complex or unpredictable environments where static control parameters may be inadequate. By dynamically adjusting to new scenarios, the swarm collectively maintains robust and adaptive behavior, allowing for more efficient resource utilization and self-organization. The combined approach of distributed learning and swarm control offers distinct advantages, including enhanced control over collective movement and decision-making. The increased complexity of the proposed model enhances its predictive accuracy by incorporating distributed learning and swarm control mechanisms, which address the challenges commonly encountered in complex systems. Distributed learning enables individual agents to autonomously and dynamically adjust their interaction rules based on local cues and environmental changes, thereby reducing estimation bias and improving robustness in noisy or unpredictable environments. This decentralized approach reduces dependence on globally predefined parameters, which are often difficult to estimate in complex systems, and instead emphasizes adaptive, real-time updates that enhance predictive performance. Additionally, swarm control mechanisms ensure stability and cohesion within the collective, allowing the system to maintain consistent and accurate predictions, even under significant variability in external conditions or individual behaviors. By combining these features, the model effectively captures the intricate dynamics of both individual and group interactions with the environment, striking a balance between model complexity and predictive performance. The simulation results demonstrate reliable convergence, accurately representing collective behavior across a range of scenarios, including environments with high noise, high density, and multiple resource types. It also increases the predictive accuracy of models simulating biological collective behavior and provides greater flexibility for practical applications in bio-inspired systems. Bridging a gap in traditional bottom-up models, this approach supports local rule adjustments while ensuring swarm-wide coherence, advancing both theoretical understanding and practical applications of collective behavior in artificial systems. In summary, the concept of potential fields provides a cross-disciplinary framework for understanding various biological collective behaviors. This mechanism not only reveals the fundamental laws of individual interactions but also offers new methods for simulating and analyzing biological collective behavior. This research will enrich the theoretical understanding of biological collective behavior and provide new ideas and methods for future applications in bio-robotics, automation systems, and complex networks. Future research will further explore the application of potential field mechanisms in different biological systems and experimentally validate their predictive power and effectiveness. The rest of this paper is organized as follows. Section 2 introduces the methodologies employed in this study, including detailed descriptions of the individual dynamic behavior models and the potential field mechanism. Section 3 shows the emergence of collective behavior under various environmental conditions. Section 4 provides empirical comparisons with biological systems. Lastly, Sect. 5 summarizes the key findings of the study, emphasizing the significance of the dynamic potential field mechanism in modeling collective behavior.",
"discussion": "Discussion Adaptability of the potential field mechanism in complex systems This study presents a novel approach to modeling collective behavior in biological systems through the concept of dynamic potential fields. By integrating local interactions and environmental influences, our framework effectively simulates the self-organizing and adaptive nature of collective behaviors. The proposed model demonstrates robustness and efficiency across different environmental conditions, reflecting the resilience and adaptability of biological systems. Our findings not only contribute to theoretical insights but also align with empirical data on collective behaviors observed in biological systems. These empirical validations reinforce the potential field framework’s relevance in modeling real-world behaviors, such as risk-averse clustering in animal groups and resource navigation efficiency. In standard environments, our simulations reveal how individual organisms utilize local interactions to navigate towards optimal resources, illustrating the efficiency of collective movements observed in nature, such as birds flying in formation or fish schooling. In high-noise environments, the model shows that although noise initially disrupts coordination, biological systems can adapt and restore collective coherence through distributed learning and feedback mechanisms. This resilience to noise underscores the robustness of the potential field mechanism in maintaining group cohesion under challenging conditions. High-density environments present unique challenges, such as overcrowding and increased competition for resources. Our model demonstrates how individuals can rapidly adjust and synchronize their behaviors to achieve effective collective movement, even in these crowded settings. This finding highlights the ability of biological systems to balance individual needs with group efficiency, ensuring survival and resource optimization. In high-risk environments, characterized by strong negative potential fields, the model reveals a natural clustering behavior among individuals as they seek safety in numbers. This collective risk-averse strategy mirrors real-world scenarios where animals group together to enhance their defense against predators or adverse conditions. The ability to form tight clusters in response to threats demonstrates the adaptive strategies employed by biological systems to mitigate risks. In environments with multiple resources, the dynamic potential field mechanism shows that individuals do not move directly to the highest resource peak but instead exhibit stepwise movements towards intermediate potential fields. This gradual aggregation process reflects real-world behaviors where organisms use local cues to navigate complex environments efficiently. This behavior enhances group coordination and ensures optimal resource utilization. Adaptability of the potential field mechanism in ecological management Applications in predator-prey dynamics The application of potential field mechanisms in modeling predator-prey dynamics represents a significant advancement in ecological modeling and wildlife management. Predator-prey interactions constitute one of the most intricate and fundamental processes governing ecosystem balance and biodiversity. These interactions create a dynamic feedback loop, wherein prey species continually adjust their spatial movements and behavioral patterns to evade predation, while predators simultaneously refine their hunting strategies to optimize resource acquisition. This ongoing evolutionary arms race can be effectively simulated and analyzed through potential field models, providing insights into the spatial distribution and behavioral ecology of species in natural environments. Potential field models operate by representing prey aggregations as attractive forces that generate regions of increased potential, while predators are modeled as repulsive forces that generate avoidance gradients. This dual interaction mirrors the collective behaviors observed across various taxa, such as schooling in fish, flocking in birds, and herding in terrestrial mammals. These emergent patterns, driven by the interplay between attraction and repulsion, serve as evolutionary adaptations to minimize predation risk and maximize survival rates within populations. By simulating these dynamics, potential field models offer a robust framework for predicting how alterations in environmental variables—such as habitat fragmentation, climate change, or human intervention—affect ecosystem stability and predator-prey population dynamics. Marine ecosystem applications Marine ecosystems provide a compelling example of how potential field models enhance our understanding of predator-prey relationships. In pelagic environments, schooling fish exhibit collective movement behaviors that reduce individual predation risk through the “selfish herd” effect, wherein individuals seek the relative safety of the group’s center. Predatory species, such as sharks, tuna, or orcas, in turn, adapt their foraging strategies by targeting these aggregations, leading to dynamic predator-prey spatial distributions. Potential field models enable researchers to simulate these interactions by representing fish schools as cohesive attractors and modeling predatory species as mobile repulsive forces that respond to prey density. These simulations can predict spatial patterns of fish aggregation under varying environmental conditions, such as ocean temperature gradients or shifting resource availability. The insights derived from such models contribute to the design of marine protected areas (MPAs) and the establishment of sustainable fisheries management practices. By identifying critical zones where prey aggregations are vulnerable, policymakers can implement seasonal fishing restrictions or designate no-catch zones to ensure the resilience of prey populations, promoting long-term marine biodiversity conservation. A notable case study involves the modeling of tuna and sardine interactions in the Pacific Ocean, where dynamic potential field simulations have guided quota-setting policies and enhanced the sustainable harvesting of sardine populations while safeguarding predator species that rely on them. Such models also provide essential data for minimizing bycatch, an issue that frequently arises in pelagic fisheries, ensuring ecological balance and economic stability within the fishing industry. Terrestrial ecosystem applications Terrestrial ecosystems similarly benefit from the application of potential field models to understand and predict predator-prey dynamics. Large herbivores, such as wildebeest, elk, or deer, exhibit migratory behaviors that are shaped by both resource availability and predation risk. These herbivores tend to aggregate in areas with abundant vegetation, creating attractive potential fields, while predator species such as lions, wolves, or cougars generate repulsive fields that drive prey toward safer regions. Potential field simulations in terrestrial environments provide critical insights for wildlife conservationists and land managers seeking to design and implement effective wildlife corridors and protected areas. For instance, by simulating how prey species navigate landscapes in response to predation pressure, researchers can identify key migratory routes and habitats that serve as refuges. This information informs the strategic placement of protected zones, ensuring that prey species can safely traverse fragmented habitats, reducing mortality rates and promoting genetic diversity across populations. An illustrative example is the application of potential field modeling to the Serengeti wildebeest migration, where predators such as lions and cheetahs influence herd movement across the plains. By integrating potential field simulations with satellite tracking data, conservationists have successfully delineated migratory corridors that align with natural avoidance patterns, reducing the likelihood of predator-prey conflict at critical crossing points, such as riverbanks and open plains. This approach not only enhances prey survival but also maintains the ecological integrity of the predator population, ensuring a balanced and functioning ecosystem. Addressing human-wildlife conflict Beyond ecological modeling, potential field mechanisms offer practical solutions for mitigating human-wildlife conflict, a growing concern in regions experiencing urban expansion and agricultural development. In landscapes where wildlife encroaches on human settlements or farmland, potential field models simulate the spatial behavior of predators and prey in response to human-generated stimuli. For example, repulsive potential fields can be used to model predator aversion to human activity or barriers such as electric fencing, while attractive fields simulate the draw of agricultural crops for herbivores. By identifying areas where wildlife movement is likely to intersect with human land use, conservationists can implement targeted deterrent strategies or design buffer zones that minimize conflict. This proactive approach has been successfully applied in regions like Yellowstone National Park, where potential field models have guided the development of wildlife corridors that reduce elk-cattle competition, thereby lowering predator-livestock conflicts and fostering coexistence between ranchers and native carnivores such as wolves and grizzly bears. Adaptability of the potential field mechanism in bio-robotics Decentralized resource allocation and foraging models A primary application of potential field mechanisms in bio-robotics lies in decentralized resource allocation and foraging models. Autonomous robotic swarms utilize potential field algorithms to distribute themselves across environments, optimizing the collection, retrieval, and management of resources. This mimics biological foraging patterns, where animals locate and exploit resources by following environmental cues. In these models, attractive potential fields are used to represent valuable resources, while repulsive fields define obstacles, danger zones, or depleted areas. As robotic agents navigate the space, they are drawn toward regions of high resource density while simultaneously avoiding obstacles and areas of congestion. This adaptability ensures that resources are optimally harvested and distributed, balancing efficiency with sustainability. For example, autonomous agricultural robots deployed for precision farming use potential field mechanisms to locate areas requiring irrigation, fertilization, or pest control. By assigning attractive fields to areas with nutrient deficiencies or pest hotspots, robots can prioritize regions that require immediate attention, optimizing the use of agricultural inputs and minimizing waste. This dynamic adaptability reduces operational costs while enhancing crop yields, contributing to more sustainable farming practices. In warehouse logistics, automated guided vehicles (AGVs) utilize potential field algorithms to retrieve and distribute inventory. AGVs autonomously respond to attractive fields generated by high-priority tasks, optimizing their paths to reduce travel time and avoid collisions. This decentralized resource management approach allows large-scale logistics operations to dynamically adjust to changes in inventory demand and supply chain disruptions. Task allocation and dynamic path planning Beyond simple resource collection, potential field mechanisms facilitate task allocation and dynamic path planning in robotic swarms, ensuring that resources are allocated to agents based on real-time environmental feedback. This is particularly useful in search-and-rescue missions, where multiple robots must explore complex and hazardous environments to locate survivors or vital resources. In such scenarios, potential field algorithms guide robots toward high-probability areas (attractive fields) while repelling them from unsafe zones, such as debris or fire (repulsive fields). As new data emerges from sensor networks or drone surveillance, potential fields dynamically update to reflect the changing landscape, redirecting robots toward new targets or redistributing them across unexplored areas. This adaptability enhances the overall efficiency of search operations, ensuring maximum coverage and minimal redundancy. A notable example is the use of underwater drones in environmental monitoring. These drones autonomously explore ocean floors by responding to chemical or thermal gradients that signal pollution sources or ecological disturbances. The potential fields guide the drones toward areas requiring monitoring, optimizing data collection while avoiding regions of high risk, such as underwater cliffs or hazardous currents. Energy efficiency and collaborative resource management Potential field mechanisms also contribute to energy-efficient resource management by ensuring that robotic agents distribute workloads evenly across the swarm. In scenarios where multiple agents are required to collect resources or complete tasks, potential fields prevent clustering by spreading robots across available space. This reduces bottlenecks, minimizes idle time, and ensures that energy consumption is optimized across the system. For instance, in autonomous mining operations, robots equipped with potential field models collaboratively extract and transport materials from distributed mining sites. Robots are directed toward high-yield zones, with repulsive fields preventing over-concentration at any one site. This distributed approach improves the overall extraction rate while reducing wear and tear on individual robots, prolonging operational lifespans and enhancing sustainability. Beyond efficiency, the implementation of potential fields in resource management contributes to long-term sustainability. Distributing workloads evenly reduces the mechanical strain on individual robots, lowering the frequency of maintenance and extending their operational lifespans. This reduction in wear and tear not only cuts costs but also decreases material waste and resource consumption associated with frequent replacements or repairs. Over time, this approach fosters a more resilient and environmentally conscious system that aligns with the principles of sustainable development."
} | 9,453 |
37570143 | PMC10419362 | pmc | 4,438 | {
"abstract": "Triboelectric nanogenerators (TENG) have shown great potential in harvesting energy from water. For the TENG that harvests water energy, surface hydrophobicity is crucial for its performance. In this paper, we prepare a hydrophobic composite film of Polyvinylidene Fluoride/Polydimethylsiloxane/Polytetrafluoroethylene (PVDF/PDMS/PTFE) and an electrode of Polyaniline/Carbon nanotubes/Silver nanowires (PANI/CNTs/AgNWs) by electrospinning technology and a doping method, respectively, which are served as the friction layer and top electrode of TENG. The contact angle of the hydrophobic film and electrode both reach over 120°, which makes the separation process between water and the interface complete and promotes the output of TENG. The open-circuit voltage (V oc ) and short-circuit current (I sc ) can reach 150 V and 60 μA approximately. In addition, the composite electrode can be applied in the preparation of complex electrode shapes. Furthermore, the different reactions of TENG to different liquids indicate that it may contribute to liquid-type sensing systems. This work presents an efficient approach to fabricating hydrophobic films and electrodes, laying a foundation for the development of TENG for harvesting water energy.",
"conclusion": "4. Conclusions In conclusion, we successfully prepared a hydrophobic composite film of PVDF/PDMS/PTFE using electrospinning technology, and a hydrophobic composite electrode of PANI/CNT/AgNW using a doping method, respectively. Furthermore, the composite film and electrode were applied in a solid–liquid-based TENG device as the friction layer and the top electrode, and the influence of a material’s composition and hydrophobicity on its output performance were studied. For the TENG harvesting water energy, the hydrophobicity of the surface and electrode is the key to promoting the output performance of the TENG. In addition, complex-shape electrode shapes can be fabricated using the composite electrode. Finally, TENG can be applied in a liquid sensing system according to the different reactions of various liquid types. Our work demonstrates a new concept in fabricating hydrophobic materials and guides the application of TENGs in sensing systems.",
"introduction": "1. Introduction With the continuous development of the social economy, people’s demand for electric energy is increasing. As the main source of electricity generation, fossil fuels are non-renewable and diminishing. Therefore, there is an urgent need for sustainable energy alternatives. Water resources are abundant on Earth, encompassing a wide range of sources such as ocean energy and raindrop energy. These resources hold immense potential for harnessing renewable energy and providing sustainable solutions to meet our growing energy needs. Currently, one of the most common methods for harvesting water resources is through the construction of hydroelectric power stations. However, this approach does have certain limitations, such as geographical location, a high cost, and difficulty in collecting low-frequency water energy. The triboelectric nanogenerator (TENG) reported in 2012 shows great development potential in collecting water resources [ 1 , 2 , 3 ]. The TENG has demonstrated significant advantages in terms of its ability to harvest low-frequency energy, its ability to operate across unrestricted geographical locations and its low cost. These advantages make TENGs a promising complementary technology for electromagnetic power generation [ 4 , 5 , 6 , 7 , 8 ]. Furthermore, a TENG has a good performance in many aspects such as self-drive sensing, blue energy, and high-voltage power supply [ 9 , 10 , 11 , 12 ]. In terms of self-drive sensing, a TENG is capable of autonomously sensing and responding to its surroundings, allowing it to adapt to different environmental conditions. This feature enables a TENG to effectively harness energy from sources such as human motion, wind, and vibrations, ensuring optimal energy conversion and harvesting [ 13 , 14 , 15 , 16 , 17 ]. The research on water-driven TENGs focuses on two aspects: structural design and the study of surface charge transfer mechanisms [ 18 ]. Water-driven can be classified into multiple structures. The solid–liquid-based TENG has garnered significant attention due to its ability to directly interact with water, leading to a substantial reduction in friction loss. This unique characteristic has attracted considerable interest for harnessing energy from water-based sources more effectively [ 19 , 20 , 21 , 22 ]. Several structures of a solid–liquid-based TENG have been studied successively, including a single-electrode mode, double-electrode mode, etc. [ 23 , 24 , 25 ]. Xu et al. reported that a droplet-based electricity generator with a novel structure has attracted a lot attention due to its high-output performance, which has become the mainstream structure of a solid–liquid-based TENG [ 26 , 27 ]. Since the surfaces of solid–liquid-based TENG devices are in direct contact with water, the hydrophobicity of friction layers is highly required. Lots of research has reported that the hydrophobicity of the friction layer and electrodes plays a vital part in the performance of TENGs, which is contributed to by the thorough contact-separation process enhancing its charge transfer [ 28 , 29 , 30 ]. Wang et al. prepared a superhydrophobic FEP film based on a droplet electricity generator, and it was found that a TENG with film performs better [ 31 ]. When droplets and the TENG surface separate in time after contact, a layer of liquid film is formed on the surface, which will affect the output of the TENG. This fully demonstrates the influence of the hydrophobicity of the friction layer on the output performance of TENG [ 32 ]. At present, most methods for preparing hydrophobic materials use laser etching, template etching, electrochemical deposition, and chemical etching in chemistry [ 33 , 34 ]. However, most of these methods have the disadvantages of high cost and a complex preparation process, which is not good for large-scale use. Here, we prepared a hydrophobic film and investigated the effects of different components on its surface morphology, hydrophobicity, and triboelectricity. Furthermore, a hydrophobic electrode was prepared and we explored the influence of different conductive substances on the hydrophobicity and output performance applied to the TENG. The I sc and V oc of a TENG based on the hydrophobic film and an electrode can reach 60 μA and 150 V, respectively. This work may provide a guide for the preparation of hydrophobic films and hydrophobic electrodes, and increase the value of TENG in collecting water energy, such as raindrop energy and ocean energy.",
"discussion": "3. Results and Discussion The preparation process of the PVDF/PDMS/PTFE precursor fluid and electrostatic spinning device is shown in Figure 1 a,b. The specific preparation process is detailed in the experimental procedures section. Scanning electron microscope images of PVDF electrospun films showed that the fibers were evenly distributed, with a good morphology, and no granular material was observed, as exhibited in Figure 1 c. To further enhance the hydrophobicity of PVDF films, PDMS was selected as the hydrophobic enhancing material. As can be seen from Figure 1 d, the coarseness of PVDF/PDMS fibers is relatively uniform, and there is no adhesion phenomenon. The surface of a single fiber shows uneven fluctuations and a small size, and it was speculated that PDMS were distributed in PVDF fibers after THF volatilization and curing. PTFE particles were added into PVDF/PDMS solution to further enhance the triboelectric performance of the film. As shown in Figure 1 e, there are tiny particles on the surface of the fibers, and we assume that these tiny particles are PTFE, which could play an important role in increasing the hydrophobicity and electron-trapping capability. The insets of Figure 1 c,d show the contact angles of electrospun films with different components. Among them, Figure 1 c inset shows the hydrophobicity of pure PVDF electrospun films. Because PVDF itself has certain hydrophobicity, the film prepared by electrospinning also maintains its original hydrophobic property. Figure 1 d inset shows the contact angle of PVDF/PDMS electrospun film, which has the highest contact angle and the best hydrophobic performance. Combined with the SEM of PVDF/PDMS film, it is known that PDMS is evenly dispersed in the electrospinning solution by magnetic stirring before forming PVDF/PDMS films through electrospinning, which are dispersed in the interior and surface of the fibers in the form of many tiny aggregate structures in the process of electrospinning. After vacuum drying, stable micro/nanostructures can be formed and dispersed in the film, thus enhancing the roughness of the film and enhancing the hydrophobicity. The addition of PTFE has no evident influence on hydrophobicity because PTFE particles enlarged the size of the aggregate structure, which counteracts the hydrophilic property of PTFE. The addition of PTFE aims to enhance the triboelectric property owing to its strong electron-generating capacity. Therefore, three TENG devices based on three different films were prepared and the performances of the TENGs were compared, as shown in Figure 2 . The structure of the TENG is illustrated in Figure 2 a. Pt wire and indium tin oxide (ITO) were adopted as the top and the bottom electrode, respectively. Three electrospinning films were prepared for acting as the friction layer of the TENG. The tilt angle of the TENG was set at 30° and the water droplets were released from a height of 20 cm. In order to verify the triboelectric performance of the fiber skeleton PVDF, the I sc and V oc test was conducted on the TENG prepared from pure PVDF thin films, as shown in Figure 2 a,b. The V oc and I sc values are approximately 20 V and 5 μA, respectively. This is because the hydrophobic performance of pure PVDF is limited, which affects the output performance of TENG. To further verify the influence of hydrophobicity on the solid–liquid-based TENG, we prepared a TENG device based on PVDF/PDMS electrospinning film and tested its output. We can see from Figure 2 c,d that the V oc and I sc values can reach 60 V and 13 μA, respectively. Compared with pure PVDF, the output performance is remarkably improved, which further proves that increasing the hydrophobicity can improve the output performance of the TENG. Furthermore, to increase the output performance of the TENG, PTFE particles possessing a strong electron-accepting ability were added. Figure 2 f,g exhibits the influence of various PTFE contents on the TENG performance. The results show that the output performance of the PVDF/PDMS/PTFE-based TENG was relatively improved compared with that of the TENG without PTFE particles, due to the high electronegativity of PTFE. However, the output of the TENG decreased slightly with the increase in PTFE content, which resulted from the reduction in the contact angle. The power density was shown in Figure A1 . On the one hand, the increase in PTFE can increase the tribological properties of the film. At the same time, the slight decrease in hydrophobicity can increase the contact area of water droplets and film without affecting its contact separation at the same time. Therefore, the charge induced by water droplets and the film will increase. On the other hand, after reaching the maximum value, due to the decrease in hydrophobicity, the separation between water droplets and the friction layer is limited, and the output performance of the TENG will decrease even if the triboelectric performance increases gradually. According to previous research, for an instantaneous structure drop-based TENG, the key to generating a high current signal is the thorough contact-separation process of the waterdrop and top electrode, which places great demands on the hydrophobicity of the friction film. Furthermore, the electronegativity difference between water and film is another important factor influencing the output performance. In some cases, the two factors mentioned above may work against each other, and it is necessary to explore a balance between them. In the solid–liquid-based nanogenerator, the hydrophobic electrode is also an important component in addition to the hydrophobic surface. Water droplets transfer the accumulated charge instantaneously to the bottom electrode when they contact the top electrode. If the hydrophobicity of the top electrode is relatively low, water droplets will remain on the top electrode, affecting the charge transfer. In addition, solid-state electrodes, such as silver wire, are mostly adopted in the preparation of TENG. However, these electrodes are expensive and cannot be used in complex circuits. We prepared a hydrophobic composite electrode composed of AgNWs, CB, and PANI, and the preparation process is exhibited in Figure 3 a, with more details in the Experimental Procedures Section. Figure 3 c,e presents the morphology of the PANI/CNTs/AgNWs composite material at different magnifications. The finer fibers were CNTs, which were wound around the surface of PANI and dispersed throughout the material. As shown in Figure 3 d, the diameter of a single tube is approximately 100 nm. The presence of PANI makes CNTs interweave on their surface, which makes CNTs contact more closely, thus enhancing the conductive property of the material. In addition, PANI is tightly wrapped by CNTs, which can form micro- and nano-scale processes. Moreover, this structure is more stable and not easily damaged, thus enabling the material to achieve hydrophobic properties. The coarser fiber in the SEM image is AgNWs, and it has very strong conductivity. Dispersed in the whole system, AgNWs can enhance the conductivity of the whole material without affecting the hydrophobicity of the material. To more intuitively express the relationship between the three in the SEM image, three-dimensional graphics were established, as shown in Figure 3 f,g. In the figure, the yellow part is the AgNWs, which constitute the entire conductive network. The blue part is the CNTs, which are distributed in the conductive network formed by AgNWs. The purple part is the PANI particles, which are wrapped and attached by CNTs on the surface and dispersed in the entire composite material. The enlarged image in Figure 3 g shows the relationship between the three more clearly and directly. Furthermore, to verify the electrodes contained in PANI/CNTs/AgNWs composites, Fourier transform infrared (FTIR) analysis and Raman spectroscopy were employed to investigate the three composites. As presented in Figure 4 a, according to analysis, the absorption peak of 1117 cm −1 is for the C-H bending vibration of the benzene ring. The absorption peak of 1494 cm −1 and 1580 cm −1 are for the benzene ring skeleton. The absorption peak of 1291 cm −1 is for the PANI’s C-N stretching. The wave peaks of 3568 cm −1 and 3614 cm −1 are for the -N-H stretching vibration of the PANI, and the absorption peak of 1379 cm −1 is the -C-H bending vibration of the CNTs. The absorption peak of 2927 cm −1 and 2856 cm −1 are the -C-H stretching of the CNTs. According to Raman spectrum analysis ( Figure 4 b), it was proved that the existence of the -C-C- stretching vibration of the major functional groups of PANI and CNTs were not destroyed. Through the combination of the two spectra, it is sufficient to prove that there was no chemical reaction between the PANI and CNTs, and no new chemical bonds and new chemical substances were formed. Therefore, the hydrophobic and conductive properties formed by PANI/CNTs/AgNWS composites result from the properties of the material itself. Furthermore, we utilized the composite electrode to prepare the TENG acting as the top electrode. For comparison, two TENG devices with a Cu wire and Pt wire as the top electrodes were prepared, respectively. As shown in Figure 4 c,d, the results illustrate that the V oc and I sc of the PANI/CNT/AgNW-based TENG are approximately equal to that of the Pt-based TENG and slightly higher than that of the copper electrode. We can elucidate from the results that the PANI/CNT/AgNW hydrophobic electrode can not only replace expensive platinum electrodes but also shows great advantage in complex electrode preparation, promoting the development of a diversified structure and low-cost TENG. The PANI/CNT/AgNW composite electrode has a fluidity to a certain degree. Therefore, as a hydrophobic electrode, it can prepare complex shapes for large-scale preparation. Due to this property, we can prepare a TENG device with electrodes with complex shapes that fit the friction layer. To verify that it can be made into any shape, a two-dimensional code and grid pattern were fabricated; see details in Figure A2 . As Figure 5 b shows, the process of preparation can be summarized as follows. First, using CAD drawing software (2020.01.00.01), a two-dimensional code and grid patterns can be drawn, and be imported into the laser cutting machine for pattern cutting. A mask is then pasted on the required basement. After cutting, the two-dimensional code and grid pattern mask is removed to obtain the two-dimensional code and grid pattern. The PANI/CNT/AgNW composite material is coated on a two-dimensional code and grid pattern by an adjustable coater. The hydrophobic electrode with complicated shapes is obtained at heating at 200 °C for 10 min after drying at room temperature and removing the remaining mask. Another approach to do this is to cut out the required mask primarily and then cover it on the required substrate for electrode coating and fabrication. The thickness of the electrode can be adjusted according to the thickness of the mask, or any pattern can be drawn through CAD to prepare the hydrophobic electrode suitable for the application. The waterdrop-based TENG can be utilized to detect liquid types, using the electrical outputs for various kinds of water, including tap water, lake water, and seawater. As shown in Figure A3 , droplet TENG device was fabricated by hydrophobic electrode and hydrophobic membrane. As shown in Figure 5 c,d, tap water produces a high-output current and voltage, and seawater produces the smallest output. For the sea and lake water, high ion concentrations and other impurities lead to the formation of a shielding layer, decreasing the charge induction and transfer. The reaction of TENG to different liquids indicates that TENG has application potential in the field of self-driven sensing systems."
} | 4,666 |
36294652 | PMC9604847 | pmc | 4,443 | {
"abstract": "By dint of the development of agroecological practices and organic farming, stakeholders are becoming more and more aware of the importance of soil life and banning a growing number of pesticide molecules, promoting the use of plant bio-stimulants. To justify and promote the use of microbes in agroecological practices and sustainable agriculture, a number of functions or services often are invoked: (i) soil health, (ii) plant growth promotion, (iii) biocontrol, (iv) nutrient acquiring, (v) soil carbon storage, etc. In this paper, a review and a hierarchical classification of plant fungal partners according to their ecosystemic potential with regard to the available technologies aiming at field uses will be discussed with a particular focus on interactive microbial associations and functions such as Mycorrhiza Helper Bacteria (MHB) and nurse plants.",
"conclusion": "4. Conclusions According to several scientifical studies, reviews and calls, the agronomic potential of endophytic root fungi is attested. However, despite their major importance in soil health management, their large-scale use is limited due to various factors such as constraints related to the production of fungal inoculants or the variability of their effects on target plants. The management of the mycorrhizal soil infectivity should be addressed considering the concepts of soil microbial ecology, and the valorization of these biofertilizers should not follow the usual rules of standard fertilization. Studies must be developed to evaluate the different strategies to be implemented (holistic or reductionist approaches) to sustainably manage the biological and chemical fertility of agrosystem soils.",
"introduction": "1. Introduction Biofertilizers are a class of biostimulants for which there is a plethora of definitions: the European Biostimulant Industry Council (EBIC) proposes the following definition: “Plant biostimulants contain substance(s) and/or micro-organisms whose function when applied to plants or the rhizosphere is to stimulate natural processes to enhance/benefit nutrient uptake, nutrient efficiency, tolerance to abiotic stress, and crop quality” [ 1 ]. A recent report [ 2 ] recalled the abundance of terminology related to plant stimulation products and provided an exhaustive inventory, and we refer the reader to this document. In this review, we will consider microbial biofertilizers or, instead, the microbial component of biofertilizers. Until the early 1980s, agroecology was considered a desired goal for agricultural systems, aiming at solving the sustainability problem of agriculture. At that time, some transposable field practices were still quite limited, particularly in developed countries where large-scale agrosystems, in order to have a more profitable agriculture, used chemical inputs and an extremely high level of mechanization [ 3 ]. Since then, the advent of agro-ecology and organic farming, the awareness of the importance of soil health and the banning of a growing number of pesticide molecules have changed the plant bio-stimulants market. Remarkably, a number of scientifical research priority programs and calls, as well as a higher number of private companies and startups in the domains of (i) seed selection, (ii) phytopathology and (iii) chemical fertilizers, are now turning to the acquisition and characterization of microorganisms as alternatives to chemicals fertilizers or biocontrol agents against plant disease or pest attacks, aiming mainly at the improvement of crop production, the interface of partners in crop associations, the shared networking of soils for the rehabilitation of lands or the restoration of different ecosystems. This was made possible by the concomitant emergence of companies developing tools and markers in the (i) global chemistry, (ii) sequencing and (iii) microbiota analyzing domains. As microbial biofertilizers, both eucaryotes—such as (i) ecto (ECM) and (ii) endotrophic (AM) mycorrhizal fungi—and procaryotes—free or symbiotic (i) nitrogen-fixers, (ii) phosphate-solubilizing bacteria, (iii) plant-growth-promoting rhizobacteria (PGPR), etc.—are considered, and more recently, even viruses are used as bacteriophages for biocontrol purposes. Plants can also, sometimes, be used as supports to produce both cultivable and uncultivable microbes, such as Glomeromycetes in endotrophic mycorrhizal symbioses, normally produced on transformed root systems. More recently, the concept of nurse plants has been developed on the basis of the higher ability of particular plants to mobilize a diversified root microbiote, thus allowing a better soil reactivation and plant growth of their co-cultivated associates through a diversified range of bacterial or mycorrhizal functions. The use of these high potential holobionts was qualified as holistic by several authors [ 4 ] ( Figure 1 ). Challenges to the use of these beneficial microbes are wide and include a number of characters such as their (i) identification among a soil microbial complex or plant microbiota, (ii) the ease and difficulty of their cultivability and (iii) their survival as a conditioning inoculant. In the acquisition and use of microbial resources, several elements must be considered: - Access to resources: Among crop plant species, domestication and genetic selection often lasted for thousands of years, with more or less long distances of transportation from their native areas to the current zones of production [ 5 ]. Native areas should be regarded as a major source of information about the natural microbiota of the considered crop with a set of functions presumably essential to the plant holobiont life cycle naturally developed during evolution and early domestication. Of course, native areas of the considered crop plants must be known, the corresponding countries or zones must be accessible, the ancient ecosystems preserved, etc. [ 5 ]. The plant original diversities have to be explored across domestication steps, in several countries, with regard to the more or less progressive (e.g., intra- vs. intercontinental) dissemination. As early as possible in the development of microbial exploration projects, the terms and purposes of use must be established with the partner country or countries and adjusted as the databases evolve and their potential for use is assessed, both for plants and microbial strains (in accordance with ABS rules). - Cultivability or the survival of microbial strains: Having the microbial strain available as a pure culture is a typical objective of an agro-microbiologist; however, non-cultivability is not an indicator of non-viability as evidenced by Xu et al. (1982) on different bacterial taxa [ 6 ]. Non-cultivability is considered a general fate of AM fungi (although Glomeromycota are still partially explored in terms of cultivability), and ECM are also not always easy to cultivate or to maintain over time in pure cultures. Plate cultures of the ECM ascomycete Tuber spp. are possible [ 7 , 8 ], but the presence of bacteria of the genus Rhodopseudomonas sp. ( Figure 2 ) seems obligate [ 9 ]. Cultivability can thus constitute a real obstacle to agronomic use. Molecular methods of global analyses of microbial communities (metagenomic) can be used in parallel with microbial isolation trials to evaluate the relative rates of non-culturable microbial taxa and thus evaluate the representativity of the isolates. - Inoculation method: Depending on the plant and the cultivation methods (the need for a nursery stage, direct sowing, mechanized or not, cuttings, grafting, etc.), and also depending on the microbial strain and the form of inoculum, supply must be adapted. In this paper, we will summarize some of the approaches that can be used to stimulate or reconstitute soil microbial life with regard to the environmental situation in the frame of a collaborative program within a variety of countries. We will also propose in this paper to review and try to hierarchically classify strategies to restore soils ecosystemic potential with regard to the available technologies aiming at field uses."
} | 2,021 |
23125483 | null | s2 | 4,444 | {
"abstract": "Photosynthetic algae and cyanobacteria have been proposed for producing biofuels through a direct photoconversion process. To accelerate the efforts of discovering and screening microbes for biofuel production, sensitive and high throughput methods to measure photosynthetic activity need to be developed. Here we report the development of new ratiometric optical oxygen and pH dual sensors with three emission colors for measuring photosynthetic activities directly. The dual sensor system can measure oxygen (O(2)) generation and pH increase resulted from carbon dioxide (CO(2)) consumption simultaneously. The sensor was prepared by a copolymerization of three monomeric probes, an intra-reference probe (IRP) which does not respond to pH or O(2), a probe for pH sensing (pHS), and an O(2) probe for O(2) sensing (OS) with 2-hydroxyethyl methacrylate (HEMA) and acrylamide (AM). After polymerization, the three probes were chemically immobilized in an ion and O(2) permeable poly(2-hydroxyethyl methacrylate)-co-polyacrylamide (PHEMA-co-PAM) matrix. The resulted sensing films (membranes) exhibited three emission colors with well separated emission spectra, covering blue, green, and red emission windows, under 380 nm light excitation. Responses of the sensors to pH and dissolved O(2) were investigated in buffers and cyanobacterial cell cultures (Synechocystis sp. PCC 6803). In spite of the strong autofluorescence from cyanobacteria, the sensors were able to determine the pH values and dissolved O(2) concentrations accurately and reproducibly. The measured results using the optical sensors were well in accordance with measurements using electrodes with minimal experimental variations. The sensors were further applied for evaluation of photosynthetic activities of Synechocystis sp. PCC 6803 at the exponential and stationary phases. The results were consistent with biological observation that the photosynthetic activity in the exponential phase was higher than that in the stationary phase."
} | 501 |
33177547 | PMC7659003 | pmc | 4,446 | {
"abstract": "Over the past decades, one main issue that has emerged in ecological and environmental research is how losses in biodiversity influence ecosystem dynamics and functioning, and consequently human society. Although biodiversity is a common indicator of ecosystem functioning, it is difficult to measure biodiversity in microbial communities exposed to subtle or chronic environmental perturbations. Consequently, there is a need for alternative bioindicators to detect, measure, and monitor gradual changes in microbial communities against these slight, chronic, and continuous perturbations. In this study, microbial networks before and after subtle perturbations by adding S. acidaminiphila showed diverse topological niches and 4-node motifs in which microbes with co-occurrence patterns played the central roles in regulating and adjusting the intertwined relationships among microorganisms in response to the subtle environmental changes. This study demonstrates that microbial networks are a good bioindicator for chronic perturbation and should be applied in a variety of ecological investigations.",
"introduction": "Introduction The past decades have seen remarkable progress in understanding how human activities influence Earth’s ecosystems 1 . The loss of biological diversity, in terms of all life on earth, genetic variation among creatures, and entire ecosystems, impacts ecological processes, ecosystem services, human society, and economics. Most recently, the intertwined relationship between global environmental change and biodiversity dynamics has become a new impetus for ecological research 2 , 3 . Major strides have been made in rethinking how we conserve natural resources sustainably, redefining how we evaluate the quality of ecosystems, and refining biological indicators to accurately measure biodiversity 1 , 2 , 4 . A powerful bioindicator, e.g. plants, planktons, animals, and microbes, can show how quickly a natural surrounding is changing, or predict how it will change. For example, marine pollution can be detected by measuring changes in phytoplankton diversity 5 , 6 . The quality of aquatic and terrestrial habitats can be monitored by changes in rotifer 7 , leech 6 , and macrobenthos 8 populations, among others. Several studies recently highlighted the importance of microorganisms for determining low levels of contaminants and small biological changes, owing to their rapid growth 9 , 10 . Due to their rapid growth, easy to test and readily available, there is mounting evidence to integrate microbial biodiversity into studies of ecosystem processes or environmental changes 11 . Current negative trends in microbial biodiversity are mainly due to climbing anthropogenic pressures, e.g. resource consumption, invasive alien species, pollution, and habitat destruction, and rapid climate change 1 , 4 . These large perturbations directly affect microbial species, leading to compositional changes, and indirectly alter species’ behaviors and the strength of inter-species interactions, all of which result in a dynamic ecological network. However, most conventional indexes capturing taxonomic diversity based only on species abundance, richness, and evenness—such as the Shannon index 12 , Simpson index 13 , and Chao-1 index 14 —do not assess the effect of interrelatedness among species on the stability of the entire ecosystem, and therefore lack the sensitivity to respond to subtle and chronic environmental degradation. Microbes interact with their communities in a complicated way. Using correlation, co-occurrence or co-exclusion, to measure microbial relationships 15 is the simplest approach to potentially identifying pairs that are metabolically complementary 16 . Mutualistic microbes may benefit each other and correlate positively among samples. Competitive microbes may compete with one another, leading to a negative correlation trend 15 , 16 . Studies of co-occurrence networks in microbial communities have confirmed the connection between network structure and chronic (and subtle) environmental changes due to soil anthropization 17 , litter quality 18 , and air pollution 19 , 20 , in which microbial biodiversity, e.g. Shannon and inverse Simpson index, might therefore lead to the conclusion that the ecosystem remains unaltered if the perturbation is of subtle intensity. Over the past decades, researchers have become increasingly aware of subtle changes in environmental conditions during studies on ecological sustainability and the impacts of anthropogenic and natural processes 11 , 21 . A robust microbial ecosystem has four major drivers influencing microbial biodiversity, which can be informative or sensitive indicators of an ecosystem’s response to subtle perturbations: rare species effect, resistance/resilience effect, spatial effect, and microbial interactive effect 22 . Measuring microbial diversity at different spatial and temporal scales is another way of using microorganisms as indicators. Priority effects during microbial colonization have long-lasting consequences for the development of microbial communities and constitutes a major barrier to entry for microbes entering a community; this is also called colonization resistance 23 , 24 . The purpose of this study is to describe the application of microbial networks to detect a subtle and chronic environmental change. More specifically, this study was undertaken to understand how a subtle perturbation can illustrate the concept of colonization resistance using a novel network-based bioaugmentation approach in an anaerobic digestion system, and to suggest some practical implications and connections of co-occurrence patterns to future work. In this paper, we present a conceptual framework that links subtle perturbations; the unchangeableness of biodiversity; and the dynamics of microbial composition, co-occurrence, and networks. Their unique capabilities make the use of microbial networks to detect subtle perturbations much more useful than conventional biodiversity measurements.",
"discussion": "Discussion Microbial diversity is a common bioindicator of ecosystem functions 32 , but it is not sensitive enough to detect chronic and subtle environmental perturbations 5 , 9 . Detecting or monitoring gradual changes are important for preventing global changes, ecological disturbances, and human-induced pollution from worsening. Instead of biodiversity, non-random patterns in microbial species co-occurrence and associated metrics are being integrated to amplify the differences that subtle perturbation makes 22 , 33 , e.g. a temperature increase of 1 °C for 5 years 20 , soil contamination with mercury for several decades 17 , and annual litter decomposition 18 . Our analysis provides a framework for studying microbial communities, co-occurrence, and networks under subtle anthropogenic perturbation by adding S. acidaminiphila . In this study, a network based approach was first proposed to design and measure a microbial ecosystem after a subtle perturbation. Based on network topologies, predicted key species were added to a mesophilic anaerobic digestion microsystem to slightly interfere with the priority effect from the species that arrived first in the communities 24 , 34 . The addition of S. acidaminiphila yielded increased biogas and methane production; this represents the transitory disturbance to the microbial ecology, and the invariant microbial species represent a protective mechanism in this microsystem that prevents the colonization and overgrowth of new bacterial species. Therefore, this study takes a step toward designing a novel laboratory-scale that differentiates microbial topological niches in response to subtle perturbation pressure. This whole new way of artificially disturbing a microbial ecosystem at a small scale provides a chance for researchers to observe and investigate changes in microbial communities in response to gradual and subtle perturbation. In addition, the generalization of such laboratory-scale experimental findings to real environmental changes might help monitor and sound the alarm about subtle changes from chronic atmospheric pollution 20 , agricultural practices 18 , and metallic contaminants in the soil 17 at an early stage, which can then be countered with ecosystem management strategies. Although most taxa were maintained after S. acidaminiphila was added, changes in abundance level, differential abundance status, correlated abundance pattern ( \\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}$${\\text{r}}_{{{\\text{C}}_{{\\text{i}}} {\\text{B}}_{{\\text{i}}} }}$$\\end{document} r C i B i ), network motif, and microbial network were detected. For three abundance attributes, differential abundance status and correlated abundance did not associate with each other, but each was significantly associated with abundance level. The conventional way to decipher microbial communities with high abundance levels is not capable of adequately explaining intertwined microbial relationships. However, co-occurrence patterns and network topology results underscore the importance of recognizing regulatory interactive behaviors between microbes. In three co-occurrence clusters, seven correlated microbes ( \\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}$${\\text{r}}_{{{\\text{C}}_{{\\text{i}}} {\\text{B}}_{{\\text{i}}} }}$$\\end{document} r C i B i )— Enterococcaceae (ID5), Acidaminococcaceae (ID10), Enterobacteriaceae (ID16), Fibrobacteraceae (ID68), Clostridiales_Incertae_Sedis_XI (ID56), Syntrophorhabdus (ID86), and Methanomicrobiales_unclassified (ID19)—were linked with microbes containing the KEGG (Kyoto Encyclopedia of Genes and Genomes) reactions R10204 and R09339 (Supplementary Results 3) 35 . Syntrophorhabdus aromaticivorans (family Syntrophorhabdus (ID86)) was the first cultured anaerobe found to be responsible for degrading phenol into acetate and methane in a syntrophic relationship with hydrogenotrophic methanogens 36 . The methane generating potential of Syntrophorhabdus (ID86) was ignored in KEGG reactions; however, the co-occurrence patterns between Syntrophorhabdus (ID86) and microbes with R10204 and R09339— Pseudomonadaceae (ID22), Desulfuromonadaceae (ID64), and Dietziaceae (ID65)—addressed the underlying importance of generating methane, which was proven in a literature survey 36 . Furthermore, Syntrophorhabdus (ID86) increased in abundance with bioaugmentation but not consistently across all samples. Syntrophorhabdus (ID86) coupled with Methanomicrobiales_unclassified (ID19), Acidaminococcaceae (ID10), and Clostridiales_Incertae_Sedis_XI (ID56) were also shown to have critical roles in the network topology after S. acidaminiphila was added, whereas Methanomicrobiales_unclassified (ID19) might include uncultured hydrogenotrophic methanogens that are syntrophic with Syntrophorhabdus (ID86). Acetogenesis of Acidaminococcaceae (ID10) could definitely raise the efficacy of biogas and methane yields. Besides, some gram-negative short rods in Escherichia and Enterobacter 37 (family Enterobacteriaceae (ID16)) could enlarge the formation of methylmercury and indirectly improve methane production, even in the presence of oxygen. Concerning the daily feeding of swine manure to the mesophilic anaerobic digesters, it was inevitable that some oxygen entered the microsystem, thus allowing for some bacterial methanogenesis. These improvements mentioned above might be triggered by adding S. acidaminiphila , an aerobe that was first isolated from a lab-scale upflow anaerobic sludge blanket reactor treating petrochemical wastewater 38 . S. acidaminiphila has multiple carbon sources, including acetate, crotonate, fumarate, DL-Lactate, pyruvate, and succinate. Furthermore, it can degrade N-Acetylglucosamine 38 —the monomeric unit of the polymer chitin and a major component of the cell walls of most fungi and bacteria—and be resistant to various antimicrobial agents 39 . Consistent with the potential metabolic functions of the family of Xanthomonadaceae based on KEGG reactions, Xanthomonadaceae conveyed the ability to degrade cellulose (R11307) and chitin (R01206 and R02334). The degradation of N-Acetylglucosamine implies that S. acidaminiphila in anaerobic digesters increases fiber digestion and the ability to break down or reuse dead microbes. Bacteria-oriented methanogenesis and carbon degradation explain why biogas and methane yields increased in a low-abundance methanogenic population after S. acidaminiphila was added. In addition, the microbial networks suggest a transition from methanogens that consume a combination of acetate, hydrogen, and methanol ( Methanotrichaceae, Methanospirillaceae, and Methanobacteriac eae) to purely hydrogenotrophic types ( Methanomicrobiaceae and Methanomicrobiales_unclassified ). Although the concentration of organic acids could not be detected in this study, it is possible that the added S. acidaminiphila rapidly exhausted most of the acetate at anaerobic digesters and created a hydrogen-abundant environment for hydrogenotrophic methanogens. However, more rigorous and extensive laboratory experiments are necessary to support this proposition and make any definitive claims along these lines. Co-occurrence and model-based microbial networks are two popular approaches based on different rationales to decipher dynamic microbial ecosystems 15 , 40 – 43 . The biggest difference between the two prediction strategies is whether the interaction direction can be inferred. Commonly used methods that use co-occurrence networks to infer microbial interactions are based on non-directional association measurements, such as Pearson and Spearman correlation 17 , 19 , 44 , Bray–Curtis distance 45 , or covariance estimation 46 . Model-based networks can follow certain regression models, such as Bayesian statistics, Lotka-Volterra, and a variety of sparse regressions 40 , 47 – 49 . However, this is the first study to combine co-occurrence and model-based microbial networks (e.g. Lotka-Volterra), and it indicates the regulatory roles that co-occurrence microbes might play in a dynamic microbial network. Although the addition of S. acidaminiphila to anaerobic digesters did not directly interfere with microbial community diversities, it did change the microbial ecosystem by enhancing network motifs and motivating bacteria-mediated methanogenesis by feedback loop and cascade signaling motifs. Consistent with previous studies 50 , 51 , feedback loop regulation had a great impact on cell growth and microbial biofuel production, where the toxic effects of biofuels for cell growth could be mitigated by expressing efflux pumps to export biofuels from the microbes. Furthermore, the overall performance of biofuel production depended on a cascade process, including the efficient pretreatment of influent sludge, more short-chain fatty acids, and higher conductivities in the fermentative liquid 52 . In our study, the stronger microbial interactions highlighted long chain regulatory cascade motifs, M4-406, and parallelly the weaker interactions featured shorter chain cascade motifs, M4-404 and M4-4682, which may have boosted the microbial cascade process by switching to different interactive strengths after bioaugmentation. Our study demonstrates for the first time under subtly perturbed environments that the purpose of hierarchical regulatory motifs launched by co-occurrence members might form a functional module to respond to the dynamic surroundings instantly. Previous efforts to characterize ecological fitness and adaptation have primarily been conducted based on the response of microbial diversities to some disturbance. In the coming years, dynamic microbial ecological studies will increasingly be applied to detect subtle environmental perturbations. We present a systematic approach for handling time-series microbial communities to detect slight changes in microbial abundance between two populations with and without subtle perturbation. This method can be generalized to dynamic experiments in a wide variety of fields and provides a predictive direction and landscape for further research and experimental designs. In order to obtain more reliable and objective support, future research should search for evidence of these regulatory network motifs under a microbial ecological process and use them to decipher intertwined relations among microbes."
} | 4,222 |
39431241 | PMC11487525 | pmc | 4,447 | {
"abstract": "The concept of multifunctionality has enabled reservoir computers (RCs), a type of dynamical system that is typically realized as an artificial neural network, to reconstruct multiple attractors simultaneously using the same set of trained weights. However, there are many additional phenomena that arise when training a RC to reconstruct more than one attractor. Previous studies have found that in certain cases, if the RC fails to reconstruct a coexistence of attractors, then it exhibits a form of metastability, whereby, without any external input, the state of the RC switches between different modes of behavior that resemble the properties of the attractors it failed to reconstruct. In this paper, we explore the origins of these switching dynamics in a paradigmatic setting via the “seeing double” problem.",
"introduction": "1 Introduction Multifunctionality is the term used to describe a neural network that has the ability to perform multiple tasks without changing any of its connections. Multifunctionality is an essential property of certain biological neural networks and has been an active area of research in neuroscience since the mid-1980s, with seminal work published in Mpitsos and Cohan (1986) and Getting (1989), followed by further review papers by Dickinson (1995) and Marder and Calabrese (1996) , and more recently, reviewed in Briggman and Kristan, (2008) . These studies have identified that a multifunctional neural network in principle resembles a multistable dynamical system. In this sense, for each task that the network performs, there is an attractor associated with it. This attractor is in coexistence with several other attractors in the network’s state space, and each attractor is related to one of the tasks that the network performs. Therefore, in order to perform a given task, the multifunctional network requires a cue in the form of an initial condition in the basin of attraction of the attractor associated with the task. Taking all of the above into account, where this ability to harness multistability becomes immediately relevant is in the domain of machine learning (ML), as multifunctionality can be used to unlock additional computational capabilities of artificial neural networks (ANNs) that would otherwise have remained dormant. In Flynn et al. (2021b), multifunctionality was achieved in an artificial setting for the first time via the reservoir computing approach to ML. This involved training a “reservoir computer” (RC), which in this case was a dynamical system in the form of an ANN, to reconstruct a coexistence of chaotic attractors from different dynamical systems using the same set of trained weights. This RC was driven with input from these chaotic attractors, and the RC’s response dynamics to the different driving inputs were used to obtain a readout layer to replace the drive, after which the RC became a multistable system that reconstructed a coexistence of the chaotic attractors. In this example, to perform a particular task, i.e., to reconstruct a particular chaotic attractor, the multifunctional RC is like any other multistable dynamical system and only needs to be initialized with an initial condition in the basin of attraction of the corresponding attractor. There are many additional phenomena that can arise and also factors to consider when training an RC to reconstruct more than one attractor simultaneously. For instance, it was shown in Flynn et al. (2021b) that multifunctionality becomes increasingly difficult to achieve with the increase in the difference of the time scales of the attractors that the RC is trained to reconstruct. Furthermore, in Flynn et al. (2023), where the RC was trained to solve the “seeing double” problem that involves training the RC to construct a coexistence of attractors that describe clockwise and anticlockwise trajectories on two circular orbits, it was shown that by manually shifting the location of the training data describing these orbits, the closer the orbits are to one another, the more difficult it is for the RC to achieve multifunctionality. Remarkably, for a small range of training parameters, it was found that the RC achieves multifunctionality even when the orbits are overlapping in state space (in the sense that the training data used to drive the RC contain identical data points from the different orbits). In Flynn et al. (2023) and Flynn (2023) , it was shown that in certain cases, when the RC fails to achieve multifunctionality, it instead produces a variety of episodic switching patterns between different metastable states that resemble the dynamics it failed to reconstruct. Through further investigation of the seeing double problem, we have found a similar phenomenon to occur when the orbits are moved closer together. The purpose of this paper is to examine the origins of these switching dynamics in much greater detail. We explore the origins of the transition from multifunctionality to metastable switching dynamics in much greater detail. We find that for a small change in the spectral radius of the RC’s internal connectivity matrix, the RC first fails to reconstruct one of the orbits as the corresponding reconstructed attractor becomes unstable, and it is only after a relatively long transient that the RC approaches the other reconstructed orbit (which is the only stable attractor present in the system). After another small change in the spectral radius, the other reconstructed orbit also becomes unstable, and this results in RC switching between the dynamics of these two unstable states. On closer inspection, we find that when the second attractor becomes unstable, there is a new attractor created that facilitates these switching dynamics. This new attractor is created through this sequence of attractors becoming unstable because due to the RC’s design, and it is prohibited from becoming globally unstable. We show that these switching dynamics appear when the orbits are brought closer together, touch, and overlap. From computing the probability density of different residence times in each of the metastable states, we find a sawtooth-like pattern consisting of multiple branches of exponentially distributed points, where each branch describes a particular path taken by the RC on each of the metastable states.",
"discussion": "4 Discussion In this paper, we explore how switching dynamics emerge in a dynamical system in the form of an RC when trained to achieve multifunctionality by solving the seeing double problem. This problem involves training the open-loop RC in Equation (1) to reconstruct a coexistence of two circular orbits \n C A \n and \n C B \n . We find that as \n C A \n and \n C B \n are moved closer together, the state of the closed-loop RC ( Equation 8 ) begins to switch between what appears to be metastable states that resemble trajectories around regions of \n P \n associated with \n C A \n and \n C B \n . To be more specific, we find that these switching dynamics occur just before \n C A \n and \n C B \n touch (as shown in Figure 2 for \n x cen = 6.5 \n ), as they touch (as shown in Figure 3 for \n x cen = 5.0 \n ), and after they touch (as shown in Figure 5 for \n x cen = 2.0 \n ), whereby there is an overlap between \n C A \n and \n C B \n . However, as shown in Figure 4 , there is an intermediary regime whereby after \n C A \n and \n C B \n touch and begin to overlap (for \n x cen = 3.5 \n ), the RC recovers its ability to achieve multifunctionality and does not succumb to these switching dynamics. It is only after there is a sufficiently large amount of overlap between \n C A \n and \n C B \n (for \n x cen = 2.0 \n ) that the switching dynamics reappear. Our results also shed further light on the key role played by \n ρ \n in this RC design and its connection to the concept of memory in terms of how the larger the value of \n ρ \n , the greater the influence of previous states on the current state of the RC. What our results indicate is that if the orbits are close to touching each other, like for \n x cen = 6.5 \n , or touch each other at only one point when \n x cen = 5.0 \n , this requires the RC to place a greater weight on previous states (i.e., large \n ρ \n ) in order to achieve multifunctionality as the dynamics nearby these touching regions are quite similar. On the other hand, if the orbits overlap and touch each other in two locations that are sufficiently far but not too far apart, like for \n x cen = 3.5 \n , then the RC does not need to place such a large weight on previous states in order to achieve multifunctionality. However, once there is a larger amount of overlap between the orbits, like for \n x cen = 2.0 \n , then the RC needs to place greater weight on previous states in order to achieve multifunctionality once again. It is also worth noting that in panel (e) of Figures 2 – 5 , prior to \n C ^ A \n or \n C ^ B \n becoming unstable as \n ρ \n decreases, there is a noticeable difference in the obtained values for \n x m \n and the corresponding true values. This is most evident in panels (a)–(d) of Figure 4, where we see \n C ^ A \n and \n C ^ B \n stretched toward larger positive and negative values of \n x \n , respectively. As \n x cen \n is decreased further, this effect appears to becomes increasingly noticeable. A similar sequence of events was shown to occur in Figures 14, 15, and 21 in Flynn et al. (2023), where, for \n x cen = 0 \n , as \n ρ \n decreases, \n C ^ A \n and \n C ^ B \n are deformed in a similar way. This particular deformation may occur due to the design of \n W i n \n , as each neuron receives input from only one component of the driving input signal because each row contains only one nonzero element; therefore, as \n ρ \n decreases, this increases the influence of the input, and this may increase the likelihood that the resulting dynamics of the closed-loop RC are stretched along the \n y = x \n and \n y = − x \n diagonals. However, in order to provide a more rigorous answer, this requires conducting an extensive analysis across several random realizations of \n M \n and \n W i n \n and testing whether such a deformation effect persists when using different design principles to construct \n M \n and \n W i n \n . We believe that such an investigation is highly worthwhile to conduct and is better suited to appear in a paper where this is the main focus. From closer inspection of the transitions between these metastable states, which we refer to as \n C ~ A \n and \n C ~ B \n , we find that there is a common sequence of events that occurs in each case in order to produce the switchings between \n C ~ A \n and \n C ~ B \n . Starting from a set of training parameters where the closed-loop RC achieves multifunctionality, we track how the dynamics of \n C ^ A \n and \n C ^ B \n change with respect to changes in \n ρ \n , the spectral radius of the RC’s internal connectivity matrix. We find that by decreasing \n ρ \n from the point where \n C ^ A \n and \n C ^ B \n coexist and resemble \n C A \n and \n C B \n , there is a value of \n ρ \n , where, for instance, \n C ^ A \n collides with a nearby saddle and becomes unstable, but there still exists some transient dynamics that the state of the closed-loop follows when initialized from a point on the previously stable \n C ^ A \n . Then, by further decreasing \n ρ \n , we find that there is a value of \n ρ \n where \n C ^ B \n also becomes unstable by colliding with a nearby saddle. However, when \n C ^ B \n becomes unstable, there is a new attractor born that facilitates the switching dynamics between the metastable states, \n C ~ A \n and \n C ~ B \n , mentioned earlier. To be more specific, a trajectory on this new attractor consists of two regions of convergent flow where the trajectory inside these regions resembles a trajectory around \n C A \n and \n C B \n and a divergent flow whereby the state of the closed-loop RC switches from one region of convergent flow to the other. We also investigate the long-term behavior of some of these new attractors that are born during the sequence of events discussed above. We integrated the closed-loop RC forward in time until we obtained 10,000 transitions between \n C ~ A \n and \n C ~ B \n for the chaotic attractors illustrated in Figure 2C , denoted as case 1, and in Figure 3D , denoted as case 2. We construct an algorithm based on the concept of a non-ideal relay to determine the time of transition between \n C ~ A \n and \n C ~ B \n . In Figure 6, we provide an example of the transition times detected by this algorithm. Interestingly, by computing the probability density of residence times in \n C ~ A \n and \n C ~ B \n , we obtain several branches of exponentially distributed points, as shown in Figure 7 . From closer inspection, we find that each of these branches correspond to scenarios where the state of the closed-loop RC completes a given number of loops or partial loops around \n C A \n and \n C B \n . We remark that while these switching dynamics are found for a particular random realization of \n M \n and \n W i n \n (the internal and input connectivity matrices), the results presented in this paper are not solely dependent on these particular weights as we see similar behavior emerging from further experiments not shown here. Furthermore, there is a noticeable imbalance in the behavior of \n C ^ A \n and \n C ^ B \n despite the symmetry present in the training data. We believe that this is due to the particular random realization of the \n M \n and \n W i n \n matrices happening to favor the reconstruction of one orbit over the other at particular parameter settings. From further analysis (also not shown in this paper), we find some small differences in the values of \n ρ \n and the order of when \n C ^ A \n and \n C ^ B \n become unstable for different realizations of \n M \n and \n W i n \n . As a further point, while the switching dynamics are induced by moving \n C A \n and \n C B \n closer together, it is still possible for switching dynamics to emerge between a reconstructed attractor and untrained attractors (attractors that the closed-loop RC produces that was not present during the training), or between the attractor, an RC with symmetry trained to reconstruct, and its mirrored counterpart as shown in Figure 2 in Herteux and Räth (2020) . We suspect that when there is a competition between attractors, be it attractors that are manually moved closer together or attractors that compete with their mirrored counterpart or other untrained attractors, this sequence of attractors becoming unstable combined with the constraint that the RC is prohibited from exhibiting globally unstable dynamics (due to the choice of activation function) in turn creates a new attractor that is composed of different metastable states, which in turn produces these switching dynamics. Out of the many examples of routes to metastable dynamics discussed in Rossi et al. (2024) , there are a number of similarities between the results presented in this paper and phenomena such as chaotic itinerancy and heteroclinic cycles. In the case of chaotic itinerancy, which describes a switching process whereby the state of an autonomous dynamical system switches between several “attractor ruins” or “quasi-attractors” (these were previously coexisting attractors that retain much of their original features except trajectories on these quasi-attractors leak into each other), in our case, these quasi-attractors are described as the metastable states \n C ~ A \n and \n C ~ B \n . In terms of heteroclinic cycles, this typically occurs when the unstable manifold of one saddle intersects with a stable manifold of the other saddle, which, in our case, these saddles would be the chaotic transients associated with \n C A \n and \n C B \n . However, further work is required in order to determine which of these phenomena our results are most closely aligned with. Furthermore, a similar route to chaotic behavior has been observed in the past by Grebogi et al. (1985), whereby when two unstable orbits move toward each other by changing a parameter in the system, they coalesce at a bifurcation point and subsequently disappear; however, after the bifurcation, a chaotic transient is produced, which persists for parameter values far beyond the bifurcation point. In our case, we have one stable attractor and an unstable orbit/relatively long transient in the closed-loop RC that as \n ρ \n is varied, and there is a bifurcation where the stable attractor becomes unstable and a new attractor is born, which, depending on the circumstances, is either a chaotic attractor or limit cycle. Moreover, there is a valid reason why there is no transient produced after the second attractor becomes unstable. Due to the design of this closed-loop RC, it is prevented from ever becoming globally unstable, and since there is no other stable attractor present in the closed-loop RC when the second attractor becomes unstable, there is no option but for there to be a stable attractor born through these sequence of attractors becoming unstable. While the routes to metastable behaviour mentioned above are well-studied phenomena they only arise in certain circumstances and rather than relying on there being a parameter in a dynamical system that so happens to produce these switching dynamics, the major advantage of the multifunctional reservoir computing setup studied in this paper is that we are able to systematically induce these switching dynamics by adjusting the location of \n C A \n and \n C B \n . As a further remark, while the results presented in this paper are based on \n C A \n and \n C B \n rotating in opposite directions, this is not a necessary condition in order for switching dynamics to emerge in the RC. From additional experiments that are not reported on in the present paper, we find that when \n C A \n and \n C B \n rotate in the same direction then switching dynamics also emerge at particular values of \n ρ \n as \n C A \n and \n C B \n are moved closer together. In future work we intend to conduct a wider study that includes additional factors which may influence the emergence and behaviour of switching dynamics in a RC that are related to the training data, such as, in the context of the seeing double problem, differences in the frequency or relative size of \n C A \n and \n C B \n , and the relationship between the training data and RC training parameters. The benefit of conducting such a step-by-step sequence of increasingly sophisticated experiments is that is provides a reasonable point of reference when attempting to make sense of how switching dynamics in a RC can emerge in more exotic scenarios involving, for instance, multiple chaotic attractors, or working with experimental data where transitions occur between states and multistability is suspected to play a role. Given the rich variety of interesting dynamics that we see arise when training the RC to reconstruct a coexistence of two circular orbits we expect that in these more complicated scenarios there are even more interesting dynamics waiting to be explored. As a final comment, the work presented throughout this paper highlights the importance of studying the behavior of saddles and the bifurcations which take place as an RC, or any dynamical system-based machine learning approach, is trained to solve a given task. As strongly emphasized in Sussillo and Barak (2013) , in order to open the black-box of machine learning approaches, it is necessary that we improve our understanding of the interaction between stable and unstable dynamics and pay closer attention to the influence of saddles that are present in the system."
} | 4,929 |
36917283 | PMC10497456 | pmc | 4,448 | {
"abstract": "Globally, \nsubstantial research into endophytic microbes is being conducted to increase agricultural and environmental sustainability. Endophytic microbes such as bacteria, actinomycetes, and fungi inhabit ubiquitously within the tissues of all plant species without causing any harm or disease. Endophytes form symbiotic relationships with diverse plant species and can regulate numerous host functions, including resistance to abiotic and biotic stresses, growth and development, and stimulating immune systems. Moreover, plant endophytes play a dominant role in nutrient cycling, biodegradation, and bioremediation, and are widely used in many industries. Endophytes have a stronger predisposition for enhancing mineral and metal solubility by cells through the secretion of organic acids with low molecular weight and metal-specific ligands (such as siderophores) that alter soil pH and boost binding activity. Finally, endophytes synthesize various bioactive compounds with high competence that are promising candidates for new drugs, antibiotics, and medicines. Bioprospecting of endophytic novel secondary metabolites has given momentum to sustainable agriculture for combating environmental stresses. Biotechnological interventions with the aid of endophytes played a pivotal role in crop improvement to mitigate biotic and abiotic stress conditions like drought, salinity, xenobiotic compounds, and heavy metals. Identification of putative genes from endophytes conferring resistance and tolerance to crop diseases, apart from those involved in the accumulation and degradation of contaminants, could open new avenues in agricultural research and development. Furthermore, a detailed molecular and biochemical understanding of endophyte entry and colonization strategy in the host would better help in manipulating crop productivity under changing climatic conditions. Therefore, the present review highlights current research trends based on the SCOPUS database, potential biotechnological interventions of endophytic microorganisms in combating environmental stresses influencing crop productivity, future opportunities of endophytes in improving plant stress tolerance, and their contribution to sustainable remediation of hazardous environmental contaminants. Graphical Abstract",
"conclusion": "Conclusion and Future Perspectives The application of microbial endophytes in agriculture, as well as environmental sustainability, is a growing research field. During the past two and a half decades, many studies have revealed rising interest in endophytic microbes. Endophytic microbes are known to improve host plant performance under abiotic and biotic stress conditions by altering the plants’ response. Recent advances in biotechnology and bioinformatic tools such as CRISPR (Clustered Regularly Interspaced Palindromic Repeats)–Cas system, RNA interference (RNAi), metabolomics, and next-generation sequencing systems have made the possibility of studying endophytes at the molecular level [ 167 ]. The present concept of isolation, purification, and characterization of endophytes and the research connecting biology to chemistry is now being developed. This opens new interdisciplinary dimensions and actively allows bachelor and master research students to participate in this domain of research. Research must focus on microbial endophytes to come up with new ideas to improve crop productivity on a pilot scale. Endophytes play an important role in producing a wide variety of naturally occurring secondary metabolites (such as tyrosol, saadamycin, and munumbicins) showing the industrial application in pharmaceutics and thus human health. In this regard, researchers from all over the world are continuously exploring hidden endophytic microbes for novel potent bioactive compounds that can be used as potential therapeutics. Figure 7 shows the importance of the biological activities of endophytic metabolites. Endophytes are reported to be a warehouse of new metabolites that can be widely used as antimicrobial, anticancer, immunosuppressant, antiarthritic, and anti-insect drugs. Although several bioactive compounds produced by endophytes, such as camptothecin, vinblastine, hypericin, and podophyllotoxin, have already been commercialized, novel bioactive compounds seem promising in the case of most pathogenic microorganisms in overcoming the problem of antibiotic resistance. Fig. 7 Biological activities of importance to humans present in endophytes’ metabolites. Endophytes have been reported to have the ability to produce novel metabolites which can serve as anticancer agents, glucosidase inhibitors (antidiabetic), and immunosuppressive agents; some of these endophytes also show antioxidant, antituberculosis, anti-inflammatory, and antimalarial activity, and serve as inhibitors of viruses Taken together, new bioactive compounds emitted by endophytes, particularly endophytic actinomycetes, could make a significant contribution to the current and future challenges of agriculture, the environment, and medicine. To isolate and characterize new endophytes with specific features that could be useful for crop production, comprehensive bioprospecting research of endophytic microbes from various ecological niches (e.g., harsh habitats, the marine environment, etc.) is required. We anticipate a shift in practice in the future, with a greater emphasis on optimizing the interaction between plants and soil microorganisms and endophytes. However, molecular mechanisms that explain the interaction between plants and endophytes have yet to be discovered. They will open a new door to the isolation and characterization of new molecules for humans and provide a new way to improve crops and environmental sustainability.",
"introduction": "Introduction Plants interact with diverse microbial species thriving in the rhizosphere and phyllosphere, thereby resulting in altered vital biological activities together with defense strategies against various abiotic and biotic stresses [ 43 , 78 , 101 , 178 ]. Rhizosphere and phyllosphere plant growth–promoting bacteria (PGPB) and mycorrhizal fungi in the rhizosphere are capable to induce growth of the plants directly by increasing macronutrient and mineral uptake and concentrations of essential hormones and/or indirectly through minimizing the negative impacts of a myriad of pathogens [ 18 , 35 , 79 , 147 , 161 , 162 , 171 , 186 , 240 , 263 ] (Fig. 1 ). Fig. 1 Overview of the plant–microbe interactions at phyllospheric and rhizospheric zone: endophytic microbes and rhizospheric microbes are capable to induce growth of the plants directly by increasing macronutrient and mineral uptake or indirectly through plant protection against pathogens. Naturally synthesized bioactive compounds with antimicrobial activities can be exploited in various sectors, especially in the agricultural and medicinal sectors The microbial species surviving on plant surfaces are epiphytes, whereas endophytes are those that inhabit the plant tissues [ 149 , 203 , 253 ]. In 1866, De Barry introduced the term “endophyte” for those organisms, including bacteria, fungi, or their associations multiplying intracellularly or intercellularly into host plants at least once in a lifetime without producing any marked signs of disease. Recent studies have illustrated that the growth and development of host plants depend to a greater extent on such symbiotic microbial species [ 55 ]. For example, in the most widely studied endosymbiotic association of rhizobium and legume, the bacterial counterpart is reported to regulate and meet the host plant nitrogen requirement [ 200 , 201 ]. Endophytes facilitate the successful establishment of symbiotic association via the synthesis and secretion of plant growth–promoting compounds responsible for host adaptation under given environmental conditions. Several fungal, bacterial, and actinomycetes species are described to participate in the synthesis and secretion of biologically active compounds and secondary metabolites [ 7 , 14 , 46 , 56 , 64 , 144 , 189 , 198 , 230 , 273 ]. Biomolecules belonging to classes of alkaloids, phenols, peptides, etc. synthesized by bacterial endosymbionts show a promising future in agriculture and medicine [ 163 , 215 ]. For example, microbially synthesized bio-insecticide azadirachtin was found to be an effective inhibitor toward the desert locust ( Schistocerca gregaria ) [ 33 ]. Since its first discovery, azadirachtin has been found to be effective against more than 200 insect species and has become an active component of many commercial pesticides, including TreeAzin, AzaMax, BioNEEM, AzaGuard, and AzaSol [ 38 , 59 , 62 , 80 , 85 , 94 , 156 , 196 ]. Many experimental investigations have reported the differential impact of factors such as specific host tissue, climatic conditions, and soil characteristics on bioactive compounds synthesized by endophytic microbial species [ 205 ]. The clue about the important role of endophytic microorganisms in the governance of the composition of metabolic products of host plants has attracted plant biologists to decipher the complexities of endophytic associations to improve crop plants. Based on life strategies, endophytic bacteria were classified as facultative, obligate, opportunistic, and passenger endophytes [ 84 ] (Fig. 2 ). Currently, different biotic factors (e.g., insects and phytopathogens) and abiotic stress (e.g., extreme temperatures, salinity, drought, flood, low/excess nutrients, and organic/inorganic contamination) resulting from climate change have emerged as important limiting factors for agricultural and horticultural crop productivity worldwide [ 274 ]. Biotic stress has been estimated to reduce annual production of about 30% of crops [ 66 ]. In particular, combined effects of multiple abiotic stress factors such as drought and heat in a particular stage of growth of the plant are more detrimental than individual stress factors. Apart from abiotic stress factors, plants are constantly challenged with biological stresses through pathogenic bacteria, viruses, fungi, insects, and pests, causing considerable losses in food productivity worldwide [ 76 , 152 , 202 , 234 ]. Various approaches, such as the selection of tolerant varieties, molecular breeding, and genetic engineering are being used to improve crop varieties against different stressors. However, the majority of these methods are time consuming, costly, and not well accepted in some areas [ 12 ]. Therefore, to neutralize the negative consequences of various factors connected with abiotic and biotic stress, host plants have developed many biological mechanisms that can function simultaneously. In this context, the mutualistic association arising from interconnections between the host and the microbe is considered an effective and sustainable means of improving plant development and growth [ 54 , 132 , 173 , 195 ]. Fig. 2 Categorization of endophytic bacteria based on their lifestyle. Opportunistic endophytes: they are bacteria which occasionally enter plants for their own needs. Passenger endophytes: they are bacteria which enter the plant by chance. Obligate endophytes: they are bacteria which are strictly bound to life inside a plant. Facultative endophytes: they are bacteria which can live inside plants and in other habitats also Unlike other plant growth–promoting microorganisms, endophytes have a direct relation with plants. They possess rapid adaptability under given conditions of biotic and abiotic stress, thereby improving host plant growth and survivability [ 9 , 25 , 61 , 101 , 149 ]. Furthermore, endophytic microbes can be an integral part of the rhizospheric region with the potential to synthesize and secrete metabolic products and enzymes [ 27 , 188 ]. They facilitate in neutralizing harmful impacts of plant pathogens. They may also allow the host plant to multiply even in polluted soil by degradation of contaminants in a manner similar to those harbored by plant growth–promoting rhizobacteria (PGPR) [ 31 , 37 ]. The application of high-throughput current “omics”-based technology such as gene sequencing, metabolomics, and microarray could comprehend the complex associations existing between plants and their endophytes and can be a promising tool for sustainable environmental development [ 40 , 105 ]. Their high colonization efficacy and stability against abiotic stress make them a potential candidate for environmental management [ 12 , 47 , 116 , 128 ]. The novelty of the present review is the current understanding pertaining to the colonization strategy of endophytes into host plants and their promising role in the alleviation of multiple abiotic and biotic environmental constraints limiting crop productivity. Noteworthy, the review has included comprehensive bibliometric information using the “SCOPUS” research database to illustrate the current research trend in the area of endophyte and possible implications in environmental stress management. In addition, the extensive information dealing with the possible roles of endophytes in eco-friendly removal of contaminants of hazardous nature including heavy metals, and diverse organic pollutants along with the future opportunities of endophytic microbes in crop improvement under changing climatic conditions, not considered in previously published reviews, are extensively taken into account."
} | 3,356 |
38702785 | PMC11067214 | pmc | 4,449 | {
"abstract": "Background Living things from microbes to their hosts (plants, animals and humans) interact with each other, and their relationships may be described with complex network models. The present study focuses on the critical network structures, specifically the core/periphery nodes and backbones (paths of high-salience skeletons) in animal gastrointestinal microbiomes (AGMs) networks. The core/periphery network (CPN) mirrors nearly ubiquitous nestedness in ecological communities, particularly dividing the network as densely interconnected core-species and periphery-species that only sparsely linked to the core. Complementarily, the high-salience skeleton network (HSN) mirrors the pervasive asymmetrical species interactions (strictly microbial species correlations), particularly forming heterogenous pathways in AGM networks with both “backbones” and “rural roads” (regular or weak links). While the cores and backbones can act as critical functional structures, the periphery nodes and weak links may stabilize network functionalities through redundancy. Results Here, we build and analyze 36 pairs of CPN/HSN for the AGMs based on 4903 gastrointestinal-microbiome samples containing 473,359 microbial species collected from 318 animal species covering all vertebrate and four major invertebrate classes. The network analyses were performed at host species, order, class, phylum, kingdom scales and diet types with selected and comparative taxon pairs. Besides diet types, the influence of host phylogeny, measured with phylogenetic (evolutionary) timeline or “age”, were integrated into the analyses. For example, it was found that the evolutionary trends of three primary microbial phyla ( Bacteroidetes / Firmicutes/Proteobacteria ) and their pairwise abundance-ratios in animals do not mirror the patterns in modern humans phylogenetically, although they are consistent in terms of diet types. Conclusions Overall, the critical network structures of AGMs are qualitatively and structurally similar to those of the human gut microbiomes. Nevertheless, it appears that the critical composition (the three phyla of Bacteroidetes , Firmicutes, and Proteobacteria ) in human gut microbiomes has broken the evolutionary trend from animals to humans, possibly attributable to the Anthropocene epoch and reflecting the far-reaching influences of agriculture and industrial revolution on the human gut microbiomes. The influences may have led to the deviations between modern humans and our hunter-gather ancestors and animals. Supplementary Information The online version contains supplementary material available at 10.1186/s42523-024-00291-x.",
"introduction": "Introduction Organisms, from microbes through animals and plants to humans, aggregate and form communities, which can be abstracted as complex networks, usually with nodes for organisms and edges (links) for their interactions. In the case of animal gastrointestinal microbiomes (AGMs), network nodes can represent microbes on a host animal species, class, or phylum, and the edges can capture not only the interactions among microbial species, but also the influences of the phylogeny of animal hosts through multi-scale network modeling (from species, through class to phylum and kingdom). Such capabilities are particular important for modeling the AGMs because animal microbiomes and their hosts form so-called holobionts, i.e., animal species and their symbiotic microbes. The holobionts are more like superorganisms with hologenomes (i.e., the total genomes/metagenomes carried by the holobionts), and are subject to natural selection and genetic drifts, and the hologenomes can be passed over next generations with reasonable fidelity [ 5 , 59 , 60 , 65 ]. Besides host phylogeny, an equally important factors influencing the evolution or coevolution of holobionts is the animal diet types because microbes frequently determine what animals can eat and digest. In fact, the trophic relationships (who eats/digest who) form the backbones of food webs in ecosystems [ 11 , 55 , 56 ]. Specifically, the symbiotic microbes not only can regulate, modulate, and/or alter the various relationships in the food web networks, primarily the competition, predation and cooperation on ecological time scale (e.g., [ 13 , 34 , 66 ]), but also can shape the coevolution between animal microbes and their hosts within the holobionts [ 5 , 59 , 60 , 65 ]. In spite of the obvious multi-facet nature of the AGM-phylogeny—food-web relationships, virtually all existing studies in the field have been focused on the influences of phylogeny and diet types on the AGM biodiversity (e.g., [ 1 , 2 , 7 , 16 , 17 , 21 , 23 , 25 , 35 , 43 , 44 , 46 – 51 , 63 , 64 , 67 , 68 , 72 ]), virtually all of which investigated prokaryotes diversity but there are also studies on eukaryome (microeukaryotic organisms associated with animal hosts) diversity [ 6 ]. In a previous series of articles [ 41 , 42 , 44 ], we have also investigated the AGM diversity [ 44 ] and the underlying mechanisms [ 41 , 42 ], heterogeneity and their scaling patterns across the spectrums of host animal phylogeny (in terms of phylogenetic or evolutionary timeline) and diet types (herbivores, carnivores, omnivores and the other). However, few of the existing studies have involved comprehensive analyses of the interactions either from microbial, host, or holobiont perspectives. In the present article, our focus is on modeling the AGM with complex network approaches with the same datasets we previously analyzed, which covers 4903 AGM samples collected from 318 animal species representing all 6 vertebrate classes and 4 major invertebrate classes. There is no doubt that the biodiversity studies of AGM are important, and in the meantime, it is not the most informative in our opinion because it explicitly ignores the interactions among species and is simply an aggregation measure in the form of entropy of species abundance distribution. For example, measuring biodiversity with Shannon entropy or Simpson index is essentially similar to measuring income distribution in economics with arithmetic average or Gini index, either far from ideal. In fact, Simpson index for measuring biodiversity and Gini index can be derived from each other [ 37 , 38 ]. A major issue in using diversity or Gini indexes is that they treat individuals as discrete entities and ignore their relationships, instead, focusing on the number of entities, and they are moderately better than using simple statistical averages thanks to their using some kind of non-linear weighting schemes inherited from entropy functions. There have been extensive applications of network science in life sciences since its start, perhaps because, like social networks, interactions among organisms including their components such as cells and neurons, offer ideal testbed for developing and testing the methods of complex networks, primarily developed by mathematicians and physicists [ 40 , 61 ]. Consequently, there have been a wide range of network models for choosing to apply to the AGM studies, here we choose two of them: the core/periphery network (CPN) and high-salience skeleton network (HSN) [ 4 , 15 , 24 ]. The CPN distinguish network nodes as densely connected core nodes and loosely connected periphery nodes that are sparsely linked to core. It mirrors the virtually ubiquitous nested structures in ecological communities and highlights the heterogeneities of network nodes from node perspective. The HSN distinguishes network edges (links) as “backbones” (consisting of high-salience skeletons) and “rural roads” (consisting of regular or weak links). It mirrors the nearly universal asymmetricities in species (or other taxa) interactions in ecological communities [ 27 , 28 ] and highlights the heterogeneities of species interaction strengths from edge perspective. Integrated together, the CPN and HSN offer a powerful approach to detecting critical network structures from both node and link perspectives, covering the perspectives of two only elements of any network models. We further build a series of CPN/HSN models on the scales of host animal species, class, phylum, and diet types, which equivalently incorporate the animal phylogeny and diet types into the AGM network models, and enable us to analyze the effects of phylogeny and diet types on the interactions in AGMs in a network setting. The insights from such network analysis cannot be obtained from diversity or heterogeneity analyses with traditional community ecology approaches. Such insights are also of important practical significance. For example, recent conservation of wildlife advocates for the protection of the whole host-microbes, known as holobiont, rather than the animal per se in traditional conservation biology. This is because microbes obviously can influence host physiology, health, behavior, and psychology [ 8 ]. Indeed, animal microbiome should be a critical part of disease ecology of zoonoses [ 45 ]. In summary, the objective of this study is to gain insights, from multiple host taxonomic scales (species, order, class, phylum, and kingdom, or diet types from host animal perspective, or community/metacommunity and landscape scales from microbial perspective) on the interactions among animal gastrointestinal microbes by building their complex network models. The interactions include both their critical structures (core, periphery nodes and backbones), as well as their effects measured in the network properties. Although critical network structures are indeed critical for AGM functions, network theory stipulates that weak links or periphery nodes in complex networks are important too because it is the weak leaks that are indispensable for stabilizing the network (e.g., [ 14 ]), somewhat similar to the relationship between billionaires and middle classes in a national economy. Therefore, we do not ignore periphery or weak links in the AGM networks either. Finally, we also model the relationship between phylogenetic timeline (PT), also known as evolutionary timeline (ET) but different from familiar phylogenetic distance (PD), and network properties to gain quantitative insights on the effects of phylogeny on AGM structures.",
"discussion": "Conclusions and discussion From previous sections, we summarized the following findings: The AGM (animal gastrointestinal microbiome) networks follow typical core/periphery nested structures from the network node perspective and contain high-salience skeleton paths from the network link perspectives. That is, all node/links in the AGM network are not homogenous; instead, both nodes and links are heterogenous with differentiated importance. The CPS structures are likely to play major roles in maintaining the functionalities of the AGM networks, and non-critical structures (periphery nodes and low-salience links) are likely to play important roles in stabilizing the networks by offering the network redundancy. In addition, the core/backbone should be like housekeeping genes in genetic networks, being general and ubiquitous. Host phylogeny measured in phylogenetic timeline (PT) does not seem to have significant influences on the evolution of CPS network properties, and our interpretation for the lack of consistent evolutionary patterns in CPS parameters is that the complex networks capture the ecological interactions on ecological time scales, which may fail to emerge in the PT-network parameters models we could reconstruct. The CPS networks of different diet types seem to differ, but we could not determine their statistical significance. While we could not relate holistic network parameters such as core strengthen to host phylogeny, we successfully detected some interesting microbial phylum level relationships with host phylogeny and diet types, especially three primary phyla in animal and human gut microbiomes, as summarized below. The relationships between the abundances of three primary phyla (BFP or Bacteroidetes, Firmicutes and Proteobacteria ) and die types seem to be consistent with the findings in the human gut microbiomes [ 29 , 30 ]. The B/F ratio of herbivores (B/F = 0.416) is indeed the highest, followed by omnivores (B/F = 0.325) and carnivores (B/F = 0.266). The B/P exhibited the same consistent trend with B/F and that of the human gut microbiome. That is, herbivorous animals and vegetarians should have higher B/F (B/P) ratios than carnivores and meatarians. The ratio of FP is also consistent with the previous two ratios (B/F, B/P). That is, herbivorous animals and vegetarians should have higher BF/BP/FP ratios than carnivores and meatarians, while omnivores sit between them and behind herbivores. The evolution trends of the three key phyla (i.e., BFP) in the animal microbiomes, and especially the evolution of their ratios do not seem to be fully consistent with the patterns found in modern human populations, especially in meatarians or obese populations. Our analyses suggested that both B & F abundances appear to be negatively correlated with PT, and P abundance is positively correlated with PT, implying that more recent (modern) species should have higher B & F abundances, lower P abundance than ancient species, although the correlations between PT with B & F may not be statistically significant (the correlation with P is statistically marginally significant, P value = 0.079). Furthermore, the three ratios B/F, B/P, and FP are negatively correlated with PT, suggesting that more recent species should have higher ratios, although the relationship between B/F ratio and PT may be statistically insignificant. Therefore, if the evolutionary trend in animals is to continue in humans, then we should have higher B/P and F/P ratios, rather than lower ratios as exhibited by some obese populations [ 29 , 30 ]. Regarding pairwise correlations between B, F and P, the correlation between B and F is not statistically significant, that between B and P is statistically marginally significant ( P value = 0.097), but the negative correlation between F & P is indeed significant statistically ( P < 0.005). That is, the phyla of F and P are inhibitory with each other. Combined with the previous findings that the relationship between the abundance of phylum P and its PT is positively correlated, while the relationships of the other two phyla (B & F) and their PT were not statistically significant, we postulate that the evolution of phylum P among animals may have more far-reaching influences on the evolution of BFP ratios, than the phyla B & F per se may have on BFP ratios. In other words, more attention to P is deserved than to B & F, to deepen our understanding of the evolution of BFP and their ratios. In existing studies on the human microbiome, much attention has been on B/F or B/P, and little attention has been paid to F/P ratio, which is negatively correlated as this study has suggested. We further cross-verify and supplemented the CPS network analyses with the module detection technique based on MCODE algorithm that detect closely linked clusters. In general, the MCODE algorithm can detect multiple (usually at least 3–5 strong clusters or modules). In the case of microbial phylum-level AGM networks in this study, only one strong cluster was detected in virtually all networks we selected to construct. Furthermore, predominantly majority of the nodes in the strong clusters from MCODE were core nodes of the CPN networks. This confirms the robustness of the CPS analyses and high reliability of the CPS findings. Finally, we summarized, from existing literature, some biological information on the critical taxa (phyla, genera, species) in the critical CPS structures, such as the MAO (most abundant OTU), network hub, host-taxon specific unique or shared OTUs in critical network structures. Nevertheless, due to the current information scarcity on specific microbial OTUs in animal microbiomes, the information we summarized (Additional file 2 : Table S7 and S8) is mainly from literatures of human and environmental microbiomes, animal specific information is rather limited in our summary, which calls for more future studies on animal microbiomes. Cordero and Datta [ 12 ] argued that microbial species may co-aggregate for mutual benefits and may segregate to alleviate the effects of competitions. The balance between co-aggregation and segregation can establish distinct local microbial communities and regional metacommunities at larger scales through dispersal/migration. To fully understand the roles of species interactions may play in driving community functionalities, it is imperative to investigate the spatial distribution (organization or structure) with sufficient “resolution” or throughput to measure statistical correlations between taxa and possible alternative community states. We fully agree with Cordero and Datta [ 12 ] arguments and adopted a hierarchical design, from both animal host and microbial taxa perspectives besides diet types, to build and analyses the CPS networks. Vellend [ 69 ] proposed to synthesize the community ecology, similar to the synthesis of population genetics, based on the four ecological/evolutionary processes including local selection, local speciation/extinction, global dispersal (migration) and random drifts. In the context of animal or human microbiomes, Näpflin and Schmid-Hempel [ 52 ] identified two open questions: First, are there most protective microbiomes to hosts? Second, how much influences does the host exert on shaping the composition and structure of its microbiome? In other words, understanding the bidirectional interactions between animal hosts and their symbiotic microbiomes should be the key for animal microbiome research. Host phylogeny and diet types are arguably the top two most important host factors for animal microbiome research. The phylogenetic timeline (PT, also known as evolutionary timeline), which is different from commonly used PD (phylogenetic distance) and can be considered as an approximate “evolutionary age” of an animal taxon. The more recent species should have smaller PT value and be “young” in terms of the evolutionary “age”. For example, the human has a PT value of 0.6 according to http://timetree.org . The usage of PT information allowed us investigating not only the possible influences of host phylogeny on the CPS structures, but also the evolutionary trend of key AGM taxa. This enabled us to obtain important comparative insights on the BF ( Bacteroidetes / Firmicutes ), BP ( Bacteroidetes / Proteobacteria ) ratios in animal and humans, which has been a focus of studies on the relationship between human gut microbiomes and personalized nutrition [ 30 ]. Existing studies on the relationships between human gut microbiomes and host lifestyles have suggested that modern urban lifestyles (such as eating more high-saturated fat and lower-fiber diets) in the Anthropocene Epoch, especially after industrial revolution, may have led to the rise of Firmicutes and Proteobacteria and the decline of Bacteroidetes abundances in the human gut microbiome (e.g., [ 29 , 30 ]). Our analysis here is aimed to discover whether or not the trend in the human gut microbiome may have certain traces in the evolution of animal gut microbiomes. In perspective, our study suggests that the evolutionary trends of B/F and B/P ratios in animal microbiomes, from phylogeny perspective, do not mirror the observed patterns in modern human populations, especially in obese populations dieting on more high-saturated fat and lower-fiber foods [ 30 ]. This seems to suggest that the high BF/BP ratios in obese populations should be due to “artificial selection”, rather than natural selection. Our study also calls for more attention on the antagonistic relationships between Firmicutes and Proteobacteria or the F/P ratio, which is currently paid relatively little attention. From host diet type perspective, our finding from AGM networks seems to be consistent with the finding from humans—herbivores and vegetarians do exhibit highest B/F and B/P ratios, followed by carnivores and omnivores. Additional file 1 : Table S6A also listed the BFP ratios for a human population with B/F = 0.286, which is slightly smaller than carnivores (B/F = 0.325) but slightly higher than omnivores (B/F = 0.266) and seems to be a reasonable estimate for humans. The B/P (1.484) and F/P (5.179) for this human population are much larger than those for animals, which is puzzling. Another study on Ukrainian population suggested a B/F range between 0.63 and 1.42 depending on BMI index, age, gender, physic activity, and cigarette smoking [ 29 ]. Obviously, comparison with human data is difficult and should be treated with caution. Still from these comparative studies, we postulate that the evolution of the critical compositional phyla ( Bacteroidetes, Firmicutes and Proteobacteria ) from animals to humans may have broken the trend, which highlights the far-reaching influences of agriculture and industrial revolution on the human gut microbiomes. In other words, the balances among Bacteroidetes, Firmicutes and Proteobacteria in the human gut microbiomes in the Anthropocene epoch have been deviating from those of our hunter-gather ancestors and animals. Finally, we should note a limitation of this study. As introduced in the material and methods section, our study reanalyzed the AGM datasets from 108 published studies containing 6900 AGM samples covering 5 phyla and 19 classes of the animal kingdom. To deal with potential heterogeneities across the different studies, we implemented strict quality control, and only selected 4903 samples covering three primary animal phyla ( Nematoda , Arthropoda and Chordates ), 10 classes (including all six vertebrate classes and four major invertebrate classes), 59 orders, 142 families, 261 genera, and 318 animal species. To further minimize the influences of potential heterogeneities across the different studies, we recalculated the OTU tables with same bioinformatics pipelines and standard parameters from the original sequencing reads of the respective studies. We also designed and performed rigorous randomization tests whenever computationally feasible to differentiate treatment effects (the influences of phylogeny and diet types) from random noises. Despite these efforts, the results (findings) generated from our analyses may still be subject to possible influences from the heterogeneities of different studies we relied on. Indeed, dealing with the heterogeneity is a rather challenging problem, and we will be conducting additional heterogeneity investigation with these datasets from different perspectives."
} | 5,690 |
21059164 | null | s2 | 4,451 | {
"abstract": "Bacteria prefer to grow attached to themselves or an interface, and it is important for an array of applications to make biofilms disperse. Here we report simultaneously the discovery and protein engineering of BdcA (formerly YjgI) for biofilm dispersal using the universal signal 3,5-cyclic diguanylic acid (c-di-GMP). The bdcA deletion reduced biofilm dispersal, and production of BdcA increased biofilm dispersal to wild-type level. Since BdcA increases motility and extracellular DNA production while decreasing exopolysaccharide, cell length and aggregation, we reasoned that BdcA decreases the concentration of c-di-GMP, the intracellular messenger that controls cell motility through flagellar rotation and biofilm formation through synthesis of curli and cellulose. Consistently, c-di-GMP levels increase upon deleting bdcA, and purified BdcA binds c-di-GMP but does not act as a phosphodiesterase. Additionally, BdcR (formerly YjgJ) is a negative regulator of bdcA. To increase biofilm dispersal, we used protein engineering to evolve BdcA for greater c-di-GMP binding and found that the single amino acid change E50Q causes nearly complete removal of biofilms via dispersal without affecting initial biofilm formation."
} | 307 |
26435905 | PMC4586181 | pmc | 4,452 | {
"abstract": "Respiration is an important process in photosynthetic organisms, as it is in other organisms, for the supply of ATP and metabolites required for biosynthesis. Furthermore, individual enzymatic activity is subject to regulation by metabolic intermediates in glycolysis and the citric acid cycle. However, little is known about how glycolysis or catabolism are related to photosynthetic activity or accumulation of photosynthetic products. We previously developed a flat-plate culture apparatus assembled from materials commonly used for gel electrophoresis, which enables high-density culture of the unicellular red alga Cyanidioschyzon merolae . In this study, a stationary dense culture of C . merolae , when re-activated in this culture apparatus, exhibited an accumulation of photosynthetically produced starch. We demonstrated that respiratory activity increased during the culture period, while photosynthetic activity remained constant. Gene expression analysis revealed that the genes involved in cytosolic glycolysis and the citric acid cycle were selectively activated, compared to the genes for the oxidative pentose phosphate pathway and the Calvin–Benson cycle. Measurements of the respiratory rate after addition of various organic substances showed that C . merolae can utilize almost any exogenous organic compound as a respiratory substrate, although the effectiveness of each compound was dependent on the culture time in the flat-plate culture, suggesting that glycolysis was rate-limiting to respiration, and its activity depended on the level of photosynthetic products within the cells. We also demonstrated that organic substances increased the rate of cell growth under dim light and, interestingly, C . merolae could grow heterotrophically in the presence of glycerol. Obligate photoautotrophy should be considered an ecological, rather than physiological, characteristic of C . merolae . Electronic supplementary material The online version of this article (doi:10.1186/s40064-015-1365-0) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions Cyanidioschyzon merolae utilizes exogenous organic substances as substrates for respiration, the rate of which is mainly regulated by modulation of cytosolic glycolytic activity in response to the level of starch accumulation. This activity seems to metabolize the photosynthetic product GAP to maintain the rate of photosynthesis. Exogenous organic compounds indeed accelerate the growth of C . merolae under mixotrophic conditions. Furthermore, we demonstrated that C . merolae , which was thought to be an obligate photoautotroph, can heterotrophically grow in medium containing glycerol, but not other organic substances. The cellular state in which GAP is synthesized from exogenous glycerol may mimic the state of cells actively performing photosynthesis.",
"discussion": "Discussion Culture of C . merolae in the flat-plate culture apparatus We used the flat-plate culture system to measure respiratory rate in C . merolae cells with different levels of starch accumulation at high cell densities because effects of exogenous substrates on respiratory rate were not measurable if the cells were in the logarithmic phase (OD 750 <1). To obtain cells with a wide range of starch accumulation levels, the flat-plate culture was started using a shaken culture (OD 750 = 10). The cells at 0 h contained hardly any starch and did not undergo division, suggesting that the cell cycle was quiescent (i.e., in the G0 phase), and the cells reentered the G1 phase during the flat-plate culture with high-light irradiation. In the green alga Chlamydomonas reinhardtii , the protein compromised hydrolysis of triacylglycerols 7 (CHT7), which contains a DNA binding domain and is localized to the nucleus, is related to quiescence with respect to cellular nutritional status (Tsai et al. 2014 ). In synchronous culture of C . merolae , cell division occurred at 10–12 h from the beginning of the light phase (Fujiwara et al. 2009 ; Moriyama et al. 2010 ), and therefore cells in the flat-plate culture will divide if the culture time is prolonged. When physiological changes were measured in the flat-plate culture, the variance was small, indicating that individual cells were evenly irradiated by light. The flat-plate culture system consists of materials used commonly for gel electrophoresis, along with thin silicon tubes for aeration; thus, no special materials are required, and the culture system can be easily acquired by any laboratory. In addition, the system can be applied to other algae, as well as C . merolae . Utilization of exogenous organic substances by C . merolae As far as we examined, C . merolae could utilize almost any exogenous organic substance as a respiratory substrate (Fig. 4 ). In contrast, exogenous glucose and glycerol did not increase the photosynthetic rate, and excess substrate actually decreased this rate. The decrease in photosynthetic rate by addition of glucose and glycerol at high concentrations may have been caused by a rapid change in the osmotic pressure. Pyruvate and lactic acid slightly stimulated photosynthetic activity at low concentrations. Because photosynthetic rate is related to respiratory rate (Vedel et al. 1999 ), it is possible that in our study increased respiratory activity stimulated the photosynthetic activity. These results indicate that exogenous substrates were indeed imported into the cells. C . merolae encodes a single gene of sugar transporter (CMK066C). In A. thaliana , the monosaccharide/proton symporter AtSTPs, which have sequence similarity with CMK066C, transport various sugars, including glucose, fructose, galactose, mannose, xylose, and arabinose, into the cell (Truernit et al. 1999 ; Sherson et al. 2000 ). In C . merolae , monosaccharides were presumably taken up by the function of the sugar transporter, although the efficiency of uptake of exogenous sugars was low (i.e., the K s values for sugars were large, Table 1 ). Exogenous glycerol dramatically increased the respiratory rate, although the C . merolae genome encodes no glycerol permease, unlike G . sulphuraria (Barbier et al. 2005 ). The cell membrane is highly permeable to glycerol, and glycerol permease may not be required for glycerol uptake in C . merolae , which lacks a cell wall (Oesterhelt et al. 2008 ). Short-chain carboxylic acids with a carbon number less than three are not significantly ionized in 2 × Allen’s medium because the acid dissociation constant (p K a ) of these carboxylic acids is much greater than the pH value of the medium (pH 2.5). Therefore, the short-chain carboxylic acids should easily diffuse into C . merolae cells, and indeed, the K s values for these acids were much lower than those for the sugars (Table 1 ). Because of this high permeability, when short-chain carboxylic acids were added to the culture at a high concentration, it is plausible that the carboxylic acids rapidly entered the cells, and then, encountering a neutral pH (Zenvirth et al. 1985 ), ionized to release excess protons, resulting in cell death. Under mixotrophic conditions, the efficiency with which exogenous substrates affected the growth rate in C . merolae was different than in the algae G . sulphuraria (Oesterhelt et al. 2007 ) and Chlamydomonas acidophila (Tittel et al. 2005 ), which can grow heterotrophically. In these algae, the growth rate effectively increased upon addition of carbon sources, even when the algae were irradiated with sufficient light intensity for photosynthesis. These results suggest that C . merolae cells with high photosynthetic activity can undergo cell division. We also demonstrated heterotrophic growth in C . merolae in the presence of glycerol (Fig. 6 , Additional file 1 : S3). Exogenous glycerol was presumably converted into DHAP or GAP in the cytosol. Because GAP is a product of photosynthesis and is transferred into the cytosol from the plastid under lighted conditions, a large amount of GAP derived from exogenous glycerol might mimic the state of cells actively performing photosynthesis. Accordingly, genes necessary for progression of the cell cycle might be transcribed in cells supplied with exogenous glycerol in darkness. The failure to observe heterotrophic growth with exogenous glucose, l -lactic acid, and succinic acid might also be explained by the level of GAP. Exogenous glucose would hardly be catabolized into GAP if the activities of enzymes in the first half of glycolysis, such as glucokinase, PFK, or fructose 1,6-bisphosphate aldolase, are suppressed in darkness. For l -lactic acid and succinic acid, these substrates are catabolized in the citric acid cycle, and metabolism of these substrates is not related to that of GAP. For verification of this hypothesis, transcriptome and metabolome analyses are needed. Our results demonstrate that C . merolae can grow in darkness with 200 mM glycerol. Similar results have been reported for the haptophyte Prymnesium parvum (Rahat and Jahn 1965 ) and marine cryptomonad Chroomonas salina (Antia et al. 1973 ); these algae can grow heterotrophically with 250 mM exogenous glycerol. However, ecologically speaking, these algae are considered to be obligate photoautotrophs because very high concentrations of glycerol do not normally exist in their natural habitat (Droop 1974 ). Similarly, C . merolae may be considered an ecologically obligate photoautotroph. Regulation of respiration in C . merolae Figure 7 summarizes the results of this study. During flat-plate culture, the respiratory rate increased dramatically, while the photosynthetic rate was constant, irrespective of starch accumulation levels. The results of Fig. 5 suggest that glycolysis was rate-limiting to respiration. The respiratory process was subdivided into three pathways, namely, the pathway from G6P to GAP in glycolysis (G6P-GAP pathway), the pathway from GAP to pyruvate in glycolysis (GAP-PYR pathway), and the downstream pathway of the citric acid cycle. The GAP-PYR pathway was immediately activated in response to enrichment of the cellular nutritional status, which may reflect the necessity to rapidly catabolize GAP, which is a photosynthetic product and should be consumed as a respiratory substrate to maintain photosynthesis. The G6P-GAP pathway also increased, but the increase was milder than that in the GAP-PYR pathway and the increase continued throughout the 6 h of culture. The G6P-GAP pathway is used for both catabolism and anabolism of starch, which functions as storage for excess energy. The production of ATP and metabolites responsible for biosynthesis through the downstream GAP-PYR pathway might be prioritized over starch metabolism if the nutrient status of the cells is very low. The results of the gene expression analysis suggest that the regulation of the PFP transcript level is related to the regulation of the G6P-GAP pathway, and the regulation of the transcript levels of GAPC , PGAM , and PK are related to the regulation of the GAP-PYR pathway. In plants, the expression level of PFP (Groenewald and Botha 2007 ) and GAPC (Zaffagnini et al. 2013 ) is related to the regulation of the respiratory rate. In contrast to glycolytic activity, the activity of the downstream citric acid cycle was largely unaltered, although some citric acid cycle genes showed large changes in expression level during the flat-plate culture. Upon increase in the activity of these metabolic pathways, the respiratory rate increased dramatically during the flat-plate culture, indicating that the amount of respiratory substrate limits the respiratory rate and has an influence on the ratio of photosynthetic rate to respiratory rate. Fig. 7 Diagram showing the regulation of the activity of respiration and photosynthesis in C. merolae . Details of this diagram are explained in the last part in “ Discussion ” section We found that cell size and shape also changed during the flat-plate culture. Changes in the areas of the plastid and extra-plastid compartments reflect the photosynthetic and glycolytic (G6P-PYR) activities, respectively. Because starch is synthesized and accumulated in the cytosol in red algae, the increase in the area of the extra-plastid compartment appears to be due mainly to an increase in the accumulation of starch. Alternatively, the development of cytosolic glycolysis and/or mitochondrial TCA cycle/electron transfer system may contribute to the increase in the area of the extra-plastid space. Because C . merolae has no cell wall, the shape and size of the cell may be variable in response to cellular nutritional states."
} | 3,184 |
22363326 | PMC3282942 | pmc | 4,453 | {
"abstract": "Lifestyle adaptation of microbes due to changes in their ecological niches or acquisition of new environments is a major driving force for genetic changes in their respective genomes. Moving into more specialized niches often results in the acquisition of new gene sets via horizontal gene transfer to utilize previously unavailable metabolites, while genetic ballast is shed by gene loss and/or gene inactivation. In some cases, larger genome rearrangements can be observed, such as the incorporation of whole genetic islands, providing a range of new phenotypic capabilities. Until recently these changes could not be comprehensively followed and identified due to the lack of complete microbial genome sequences. The advent of high-throughput DNA sequencing has dramatically changed the scientific landscape and today microbial genomes have become increasingly abundant. Currently, more than 2,900 genomes are published and more than 11,000 genome projects are listed in the Genomes Online Database ‡ . Although this wealth of information provides many new opportunities to assess microbial functionality, it also creates a new array of challenges when a comparison between multiple microbial genomes is required. Here, functional genome distribution (FGD) is introduced, analyzing the diversity between microbes based on their predicted ORFeome. FGD is therefore a comparative genomics approach, emphasizing the assessments of gene complements. To further facilitate the comparison between two or more genomes, degrees of amino-acid similarities between ORFeomes can be visualized in the Artemis comparison tool, graphically depicting small and large scale genome rearrangements, insertion and deletion events, and levels of similarity between individual open reading frames. FGD provides a new tool for comparative microbial genomics and the interpretation of differences in the genetic makeup of bacteria.",
"conclusion": "Conclusion 16S rRNA and other gene subset analyses mainly focus on the determination of the line of descendants of a given gene or organism (Zhang et al., 2009 ) or on the identification of protein families (Enright and Ouzounis, 2000 ; Enright et al., 2002 ; Kelil et al., 2007 ). Such phylogenetic studies aim to reconstruct the relationship between organisms and are paramount to analyze the (changing) community structures of complex biological ecosystems. While this type of phylogenetic analysis is well accepted and widely used, it does not reflect the respective comprehensive genotypes. In contrast, FGD provides a different view of microbial similarities to each other. The example data set demonstrated the effects of lifestyle adaptation on genome content. FGD has shown the potential to provide new insights into the relationships between microbes from a comparative genomics perspective. The algorithm has already been used in a variety of analyses ranging from microbial (Goh et al., 2011 ) and archeal (Leahy et al., 2010 ) genomes to bacteriophage (Lu et al., 2010 ) which describe the impact of FGD analyses in more detail within their respective scopes. Functional genome distribution in combination with the graphical visualization in ACT using ORFeome distance files (pMSPs) and functionally annotated GenBank files offers a powerful tool for comparative genomics that allows comparisons of whole genomes within genome space, encompassing heritage based (vertical transmission), lateral gene transfer (HGT), and lifestyle-driven change (adaptation) in a single analysis. Therefore, rather than attempting to reconstruct the evolution of the core genome with its set of commonly shared genes, FGD allows a representation of whole genome similarity at the functional level. It will be possible to add further functionality to the algorithm, such as the ability to mask defined or dynamically created gene clusters within groups of organisms, thus identifying potentially important genetic elements independent from otherwise overpowering gene sets, such as central house-keeping or metabolism genes.",
"introduction": "Introduction Microbial genomes range in size from the smallest microbial genome known to date of Candidatus Tremblaya princeps with just under 139,000 nt (McCutcheon and von Dohlen, 2011 ) to large genomes with over 13,000,000 nt such as Sorangium cellulosum SO ce56 (Schneiker et al., 2007 ). With an average gene size of about 1,000 nt, microbial genomes harbor between 140 and 13,000 genes. Lifestyle adaptation is one of the major driving forces for microbial genome re-arrangement processes such as gene loss and gene acquisition, genome rearrangements, and the movement of whole genetic islands. Until recently, such processes could not be monitored comprehensively and observations were limited to either a few select genes (typing) or the analysis of large scale genome re-arrangement events. Typing methods such as multi-locus sequence typing (MLST; Chan et al., 2001 ), are based on the selection of a few widely distributed and conserved house-keeping genes. Today, MLST utilizes whole microbial genome predominantly to identify new select target regions for amplification, rather than analyzing the whole nucleotide sequence (Maiden, 2006 ). With the advent of high-throughput DNA sequencing techniques such as automated capillary sequencing, solid-state sequencing 1 , or pyrosequencing 2 or sequencing by synthesis 3 , access to high-quality draft, and complete bacterial genomes has become feasible and is a commonly used technique. Currently 2,943 complete genomes (including eukaryotic organisms) have been reported, with another 184 archeal and 5,490 bacterial genome projects in progress 4 . The availability of complete or nearly complete genomes triggered attempts to incorporate selected larger genetic subsets (Makarova et al., 2006 ; Makarova and Koonin, 2007 ) or complete genomes (Rohwer and Edwards, 2002 ; Henz et al., 2005 ) to infer evolutionary lineages, while less progress has been made in analysis of multiple whole microbial genomes from functional and comparative genomic perspectives. One of the most prominent examples of whole genome comparative analyses is based on Blast Score Ratios (Rasko et al., 2005 ). Here, we introduce a new analysis tool, functional genome distribution (FGD). FGD does not attempt to represent the evolutionary path a genome has taken, since different genes will have been acquired by different routes. Instead, FGD investigates the overall levels of similarity between microbial genomes based on amino-acid sequences of the predicted complete ORFeomes. This reflects the impact the evolutionary force has had on genome makeup in the past, resulting in the current level of niche adaptation (Thomson et al., 2003 ). Thus presence, absence, or modification of individual genes or genetic islands defines the phenotypic potential of a given organism at a given temporal snapshot. The comparison of these ORFeomes to each other ultimately defines the level of similarity of the genomes. This approach then also takes into account important genetic adaptations to specific ecological niches or even to human made environments such as industrial fermentation processes. Such common genotype adaptations might render organisms more similar by FGD analyses than their respective evolutionary heritage would indicate. The presented approach of a FGD is a BLAST-based ORF-position-independent algorithm, implemented in the compACTor software. In the context of FGD analyses the term “functional” is used in the sense of functionality based on sequence and sequence similarity and is not based on annotation classification [i.e., such as implied by COG (Tatusov et al., 2003 ) or KEGG (Kanehisa, 2002 ; Kanehisa et al., 2008 ) databases]. Research in functional genomics (defined as the investigation of gene function by gene inactivation, gene complementation, and in silico analyses) relies heavily on the identification of differences between two or more genomes, identifying differences in the presence or absence of individual genes. The compACTor software also creates all-vs.-all ORFeome distance data files which, in combination with the respective GenBank files of the query microbes and the Artemis comparison tool (ACT; Carver et al., 2005 ), facilitate visual qualitative comparative in silico analyses of complete and draft phase microbial genomes for the subsequent analysis of changes to gene synteny and operon structures. Furthermore, to identify genes shared and unique between selected clusters a mining tool, FGDfinder, is provided, facilitating rapid identification of relevant gene sets for further in vitro analyses.",
"discussion": "Results and Discussion Evaluation of functional distribution trees To assess the FGD concept 35 completed and four draft phase genomes from different phyla, including several archeal genomes, were selected (Table a2 in Appendix). The purpose of this diverse taxonomy is to investigate how FGD places individual genomes into clusters and how similar these genomes are to each other on a functional level. The majority of the genomes selected are members of the family Lactobacillaceae in the order Lactobacillales (12 genomes). To investigate the discriminatory power of the FGD algorithm, the range was expanded and five genomes of the family Bacillaceae (order Bacillales), three genomes of the family Streptococcaceae (order Lactobacillales) and two genomes of the family Leuconostocaceae (order Lactobacillales) were added. Eight more distantly related genomes of the order Clostridiales (class Clostridia) were chosen to broaden the taxonomic selection to different classes. All of these genomes are members of the phylum Firmicutes. Three representatives of the class Gammaproteobacteria, (phylum Proteobacteria) were included to investigate inter-phylum relationships. As a final outlier, six genomes of archeal Euryarchaeota were included in the analysis. Figure 2 represents the functional distribution of the selected taxa within an FDT. To test the influence of early, incomplete draft phase genomes on functional placement, the draft phase sequence of Anaerostipes caccae DSM14662 (Schwiertz et al., 2002 ) which consisted of ∼1.69 Mbp at the time of analysis was included. FGD penalized the missing genetic information and set A. caccae apart from the Clostridiales as a separate cluster. When the updated genome sequence of A. caccae encompassing ∼3.6 Mbp was included instead, A. caccae shifted its position and clustered with Ruminococcus obeum and Ruminococcus gnavus (branch length 1.99 du), while showing a deeper branching to Butyrivibrio proteoclasticus (branch length 3.13 du; data not shown). This clearly highlights the necessity of obtaining high-coverage genome sequence data for FGD analyses. However, with high-throughput genome sequencing techniques currently available, initial draft phase genomes usually encompass 85–95% of the genome, thus allowing an initial representative functional placement. Subsequent analysis of complete genomes included in the FDT revealed the deepest branching (20.6 du) for euryarcheal genomes. The selected methanogens form two distinct genome clusters (node at 5.3 du), separating Methanococcoides burtonii and Methanosarcina mazei (Deppenmeier et al., 2002 ) from the remaining four taxa. Initial comparisons of habitat, growth temperature, and GC content did not indicate a consistently shared denominator between the two groups. Further analyses will be necessary to determine the imminent functional similarities indicated by the FGD approach. Interestingly, Clostridiaceae, Bacillaceae, and a subcluster comprised of Rumincocci and B. proteoclasticus B316 formed a new functional node within the FDT (branch depth from node to Lactobacillales was 7.8 du), combining the taxonomic families into one genome cluster. Although the genomes of members of Clostridiaceae and Bacillaceae are still placed into distinct functional groups (internal cluster branch depth was 4 du) and no taxon shuffling was observed between both sub-clusters, it appears that lifestyle adaptation has led to similar genome content, potentially indicative of a high level of HGT between both families or from one family to the other. A survey of the nine ORFeomes of the Bacillus and Clostridium clusters revealed 154 ORFs that are highly conserved in both groups ( e -value threshold 1 e -100). As expected, most ORFs could be assigned to central house-keeping functions such as DNA synthesis and repair (21%), tRNA genes and related processes (12%), central metabolism (35%), transcription and translation (6%), cellular processes (9%), and molecule transport (8%). However, besides these central functions, a significant number of ORFs related to sporulation were present and highly conserved in both clusters (8%). Based on the algorithms used, it is reasonable to hypothesize that the presence of these highly conserved sporulation genes may be one of the key drivers for the observed clustering of bacilli and true clostridia. When compared to the Ruminococcus subcluster a similar conserved gene set was found with the notable absence of most sporulation genes ( e -value threshold 1 e -60). This is in agreement with the observed non-sporulating phenotype. It is noteworthy, that two conserved genes involved in sporulation (stage V sporulation protein D, spoVD, and Sporulation initiation inhibitor protein, soj) were identified in the Ruminococcus and Butyrivibrio genomes. This may indicate an ongoing genetic loss in response to adaptation to a new environment (rumen) where sporulation is no longer offering an advantage in fitness. The other identified shared genes are likely to be present in most of the other microbial genomes analyzed, and thus would contribute to higher-level genome clustering. In addition, the ORFeomes were analyzed for predicted genes which are conserved in one genome cluster but not in the other – and vice versa (threshold conserved: 1 e -100; threshold unique: 1 e -10). Overall, 84 ORFs were identified to be group specific. Seventy of these were found only in genomes assigned to the Bacillus cluster and 14 in genomes in the Clostridium group. Remarkably, genes involved in heme and cytochrome biogenesis ( hem E, hem H, hem Y, res B, and res C), cytochrome reduction ( gcr B, gcr C, cyp D), and cytochrome oxidation ( qox B, cyd A, cta B, cta D, cyd A) were identified, indicating a Bacillus -specific electron transport chain. It is thus tempting to speculate that, functionally, Bacilli are aerobic Clostridia, having acquired the capability of oxidative phosphorylation. Furthermore, a subset of the propionate metabolism pathways identified to be Bacillus -specific. This subset is involved in the conversion of propanoyl-CoA to succinate and succinyl-CoA (prpD, prpB, pccB, sucD, sucC) and might present an additional energy conversion option for Bacilli which is absent in Clostridia. Only a few ORFs were identified to be Clostridium cluster specific. Among those a cobyric acid synthase cobQ was identified to be Clostridium specific. CobQ is part of the porphyrin metabolic pathway, involved in converting cobyrinic acid into coenzyme vitamin B 12 . Notably, a central branching point in this pathway leads to the synthesis of hemes and cytochromes found in Bacilli (see above). Interestingly, CobQ is also absent in the Ruminococcus subcluster, providing further support for the proposed ongoing adaptation to the new rumen environment. In contrast, the Ruminococcus subcluster acquired a number of membrane and sugar utilization (e.g., beta-glucosidases bgl3A, bgl3B, and bgl3D and an l -fucose isomerase) which may aid in the adhesion to and degradation of plant fibers in the rumen (threshold conserved: 1 e -60; threshold unique: 1 e -10; Kelly et al., 2010 ). While the function of these genes has been well described in the past, they deliver the proof-of-concept that FGD analyses are able to identify gene sets involved in lifestyle adaptation processes. Because the initial similarity analysis does not rely on existing gene annotation, uncharacterized ORFs (e.g., genes annotated as “conserved hypothetical”) can be identified as potential targets to contributing to respective phenotypes. This is particular important for poorly annotated microbial genomes with a high level of conserved hypothetical ORFs. In summary, results obtained from the test dataset provide strong support for the usefulness of FGD analyses, by illustrating the ability of the method to draw together groups into common nodes based on shared core (shared by all genomes in a specific cluster) and lifestyle elements, yet distinguishing them into distinct sub-clusters based on relevant genotypic differences and lifestyle adaptation processes. Importantly, FGD subsequently allows identifying gene sets likely to be responsible for the observed clustering, providing meaningful new target selections for functional genomics analyses independent of other means of classification or prior annotation. Furthermore, distinct placements in genome clustering of Leuconostoc mesenteroides and Oenococcus oeni (Ze-Ze et al., 2000 ; Makarova et al., 2006 ; Leuconostocaceae) were observed in the FDT. Both genera can be found epiphytically on fruits, fruit mashes, and vegetables and are used in industrial and food fermentation processes. Interestingly, Functional Distribution placed L. casei and L. plantarum into a separate cluster, branching deeply within the Bacilli (Figure A1 in Appendix). Both genomes are significantly larger than the average Lactobacillus genome of 1.8–2.0 Mbp with 2.8 and 3.3 Mbp, respectively. The significantly increased genome size likely reflects a more generalized lifestyle, capable of thriving in a variety of habitats such as raw and fermented dairy products, plants, and the intestinal and reproductive tracts of animals and humans. In contrast, the smaller genomes of other lactobacilli often reflect the more specialized lifestyle to one habitat such as the human or animal gastrointestinal system or specific fermentation processes. These results may indicate that lifestyle adaptation can lead to a similar genetic makeup of taxa defined as being distinctively different by heritage based phylogeny. Limits of the FGD resolution In the previous example microbial genomes from a wider range of species were investigated. To determine if the algorithm can discriminate strains from the same species, 23 Chlamydia trachomatis genomes (host: human), three Chlamydia muridarum genomes (host: members of the family Muridae), and one Chlamydia pneumoniae genome (host: varied; see Table a3 in Appendix) were subjected to an FGD analysis. It is interesting to note that until 1999 C. muridarum , infecting only members of the family Muridae, was designated as C. trachomatis (Everett et al., 1999 ). C. pneumoniae , which can infect a wide variety of different hosts and causes atypical pneumonia, clusters away from both C. trachomatis and C. muridarum genomes, indicating a different – and possible more flexible – genome makeup ( Figure A2 in Appendix). In contrast to the other Chlamydia genomes, C. pneumoniae harbors an additional ∼200 kb of genetic information, and a more detailed analysis will be necessary to determine whether gene loss or gene acquisition is the major driving force. Similarly, the C. muridarum group is clearly forming its own cluster, albeit indicating a higher-level of similarity to C. trachomatis than to C. pneumoniae . On this high level, FGD can clearly resolve genomic differences and support observed host specificities (varied hosts – Muridae – human). Within the C. trachomatis cluster, three distinct sub-clusters could be identified with little reshuffling observed. Cluster 1 comprises serotypes E, F, G, and J (and C. trachomatis Ds2923), Cluster 2 harbors serotypes A, B, D, and L (and C. trachomatis E Sweden2), and Cluster 3 groups two serotype L and one serotype A (Figure A2 in Appendix). The observed positioning of the serovar D strain C. trachomatis Ds2923 into Cluster 1 supports the pairwise alignment of several chlamydial isolates (Jeffrey et al., 2010 ) which identified the least number of nucleotide substitutions between Ds/2926 and E/11023. In contrast, different major groups were identified in this nucleotide based analysis; two major clades (D/G/J and E/F; Jeffrey et al., 2010 ) are contrasted by three clusters (E/F/G/J, A/B/D/L, and L/A). Overall, the FGD analysis was able to resolve strains from the same species to a similar level and with similar results as other whole genome comparative approaches. However, one of the limitations seen in the analysis of very similar genomes from the same species was the difficulty in identifying cluster specific gene sets based on e -value using FGDfinder [only three hypothetical ORFs were found to be cluster specific to Cluster 3 (see Figure A2 in Appendix, threshold conserved: 1 e -20; threshold unique: 1 e -10)]. In its current version FGDfinder uses calculated e -values to determine respective conserved or cluster specific gene sets. A future version of the software will incorporate the FGD scoring algorithm, increasing the power of resolution when very similar ORFeome sets are compared. Topographical stability of functional distribution trees The overall stability of inferred trees and respective genome clusters was tested by a Jackknife analysis (James and McCulloch, 1990 ). Individual observations (genome entries) from the calculated distance matrix were removed sequentially and resulting UPGMA-based FDTs were approximated. This was done for each genome in the dataset, resulting in 39 subset trees. Respective topologies were assessed individually. To investigate the impact of whole genome clusters on tree robustness, a Jackknife analysis was performed defining observed genome clusters as individual observations. Again, FDTs were approximated and evaluated for each resulting subset (data not shown). In summary, tree topology remained stable and only a swapping of neighboring branches was observed while in no instance shuffling was found for individual genome entries between genome clusters in either Jackknife analysis. This strongly supports the overall stability and discriminatory power of the FGD analysis. Similarly, a Jackknife analysis was performed for the narrow strain FGD analysis (Figure A2 in Appendix). Individual entries, individual clusters, and complete serotypes were removed and the resulting tree topology investigated. Removal of individual entries and complete clusters did not change the topology of the FGD tree. Only minor reshuffling was observed within a respective cluster when complete serotypes were removed from the analyses (e.g., removal of serotype L caused a repositioning of C. trachomatis D UW3 CX into the serotype D subcluster within Cluster 2. As expected, the removal of individual entries reduces discriminatory power, resulting in grouping together previously more separated genome clusters, without any entry-reshuffling. Comparative analysis using the artemis comparison tool Comparison of the degree of genome similarity between two or more genomes relies on analyses of presence/absence of genes and their respective syntenies in an operon or genome-wide context. In combination with ORFeome based (ORF-to-ORF comparison on amino-acid level, a maximum of 20 similarity hits per ORF is permitted) MSPcrunch (Sonnhammer and Durbin, 1994 ) comparison format data files (pMSPs) created by the compACTor software and annotated GenBank files, the ACT (Carver et al., 2005 ) provides an excellent visualization platform for mobility through the entire genome. Figure 3 illustrates the differences and similarities found between closely related strains such as between members of the acidophilus-complex of the Lactobacilli or between subspecies as well as between less similar genomes of more distantly related microbes. For example, small scale [inversion of a specific gene locus between both L. delbrueckii ssp. bulgaricus genomes (Makarova et al., 2006 ; van de Guchte et al., 2006 )] and large scale (double inversion ∼700 and ∼150 kb, respectively) of the terminus of DNA replication between L. gasseri (Azcarate-Peril et al., 2008 ) and L. johnsonii (Pridmore et al., 2004 ) genome inversions, deletion, and insertion events [between L. acidophilus (Altermann et al., 2005 ) and L. delbrueckii ssp. bulgaricus ], localized gene synteny (between L. delbrueckii ssp. bulgaricus and L. brevis ), general presence/absence of individual genes and larger synteny trends (between L. casei and L. plantarum ) can be immediately identified. Figure 3 ORFeome based comparative ACT visualization of 11 Lactobacillus genomes . Based on the distribution observed in Figure 2 , 11 Lactobacillus genomes and their ORFeome similarities were visualized in ACT using pMSP-datafiles. Respective genome designations are indicated on the left hand side of each genome line. Genomes are shown in full and drawn to scale. Genomic nucleotide sequences are represented by gray lines indicating sense and anti-sense strands and position markers are shown in between. Predicted ORFs are shown on each strand in their respective orientation as arrowed boxes. Direct amino-acid similarity between individual ORFs of neighboring genomes are shown as red lines, inverted similarities are indicated by blue lines. Color shadings indicate the level of similarity, the more saturated a similarity line the more conserved are two ORF-pairs. A trust level value of 40 was employed as display threshold to visualize similarity hits below an e -value of 1 e -60. Comparison of the FGD algorithm to alternative methods Although similar algorithms exist to investigate phylogenetic relationships based on whole (or partial) genomes sequences (Snel et al., 1999 ; Wolf et al., 2001 ; Henz et al., 2005 ; Khiripet, 2005 ; Canchaya et al., 2006 ; Fuchsman and Rocap, 2006 ; Berger et al., 2007 ; Felis and Dellaglio, 2007 ; Blaiotta et al., 2008 ), their focus remains mostly to infer a heritage based phylogeny. Furthermore, often only subsets of ORFeomes are chosen for these analyses. These are then analyzed individually (with or without weighting) or as concatenated sequences. Thus an artificial restriction is introduced that may bias the analysis. An example of such a method has been published by Konstantinidis and Tiedje ( 2005 ). There, genome information to infer taxonomy of prokaryotes is used, calculating an average amino-acid identity of shared gene subsets. Only few publications investigate the functional relationship of microbial genomes, such as the development of the Blast Score Ratio which analyzes the complete ORFeome but is limited to three genomes at a time (Rasko et al., 2005 ). Other methods investigating the functional relationship and similarities between gene clusters have been used to address the problem of genomes with different sizes. One example of such a method, GRAST, explores the ongoing genome reductions and rearrangements by identifying clusters of functionally related genes (Toft and Fares, 2006 ). Subsets of orthologous gene pairs are identified to determine conserved genetic loci. Similar to Blast Score Ratio analyses, the number of input genomes is limited to two at a time. While the output in part shows similarity to FGD analyses (visual representation of a genome plot and the determination of common and non-common genes), the purpose of this method is distinctly different in specifically identifying genome plasticity trends. A combination of genome analysis and visualization tool, GeneComp, has been published earlier (Yang et al., 2003 ). GeneComp is able to use different BLAST flavors and then visualize the textual output with varying levels of alignment length stringencies. While this solution offers the advantage of providing a combined analysis and visualization package, a number of limitations exist when compared to FGD. Like BSR and GRAST, GeneComp is restricted to a maximum number of three genomes. Furthermore, the algorithm is sequence based, highlighting genome variations such as repeat regions, insertions, deletions, and rearrangements rather than specific similarities to predicted ORFs. Non-sequence based methods such as MLST use only a relatively small number of conserved genome loci with the primary aim to establish a highly discriminating (microbial) typing system (Chan et al., 2001 ; Maiden, 2006 ; Diancourt et al., 2007 ). Noteworthy, the ability of FGD analyses to identify cluster-conserved gene sets may provide a high-quality starting point for the selection of MLST targets."
} | 7,213 |
35630450 | PMC9142973 | pmc | 4,454 | {
"abstract": "Microbial electrolysis cells (MECs) are an emerging technology capable of harvesting part of the potential chemical energy in organic compounds while producing hydrogen. One of the main obstacles in MECs is the bacterial anode, which usually contains mixed cultures. Non-exoelectrogens can act as a physical barrier by settling on the anode surface and displacing the exoelectrogenic microorganisms. Those non-exoelectrogens can also compete with the exoelectrogenic microorganisms for nutrients and reduce hydrogen production. In addition, the bacterial anode needs to withstand the shear and friction forces existing in domestic wastewater plants. In this study, a bacterial anode was encapsulated by a microfiltration membrane. The novel encapsulation technology is based on a small bioreactor platform (SBP) recently developed for achieving successful bioaugmentation in wastewater treatment plants. The 3D capsule (2.5 cm in length, 0.8 cm in diameter) physically separates the exoelectrogenic biofilm on the carbon cloth anode material from the natural microorganisms in the wastewater, while enabling the diffusion of nutrients through the capsule membrane. MECs based on the SBP anode (MEC-SBPs) and the MECs based on a nonencapsulated anode (MEC control) were fed with Geobacter medium supplied with acetate for 32 days, and then with artificial wastewater for another 46 days. The electrochemical activity, chemical oxygen demand (COD), bacterial anode viability and relative distribution on the MEC-SBP anode were compared with the MEC control. When the MECs were fed with artificial wastewater, the MEC-SBP produced (at −0.6 V) 1.70 ± 0.22 A m −2 , twice that of the MEC control. The hydrogen evolution rates were 0.017 and 0.005 m 3 m −3 day −1 , respectively. The COD consumption rate for both was about the same at 650 ± 70 mg L −1 . We assume that developing the encapsulated bacterial anode using the SBP technology will help overcome the problem of contamination by non-exoelectrogenic bacteria, as well as the shear and friction forces in wastewater plants.",
"conclusion": "4. Conclusions In this study, a novel bacterial anode encapsulation technology based on a small bioreactor platform (SBP) was developed. To the best of our knowledge, and after a survey of the literature, this approach has not been reported before. The capsule membrane physically separates the microbial culture inside the capsule, including the carbon-cloth anode material from the natural microorganisms in the wastewater, while enabling nutrient diffusion. The MECs based on the SBP anode, as well as the MECs based on the nonencapsulated anode (MEC control), were fed with Geobacter medium supplied with acetate for 32 days, and then with artificial wastewater for another 46 days. LSV measurements of the MEC-SBP fed with artificial wastewater produced 1.70 ± 0.22 A m −2 (at 0.6 V), i.e., double the rate of the control. The HERs of the MEC-SBP, when supplied with acetate as the sole carbon or with artificial wastewater, were 0.027 and 0.017 m 3 m −3 day −1 , respectively. In comparison, the MEC control led to only 0.006 and 0.005 m 3 m −3 day −1 , respectively. The COD consumption rate for both MECs was about the same at 650 ± 70 mg L −1 . Biofilm viability on the control anode was twice as high as that observed on the SBP anode. The microbial diversity on the MEC-SBP and MEC-control anodes showed that the relative distribution of Geobacter was only 10%, which can explain the relatively low currents. To increase the currents in the MEC-SBP, SBP-anode technology must overcome several obstacles, such as expanding the carbon-cloth anode material surface area for biofilm attachment, a proper method for sterilization of the capsule, and increasing the contact of the external titanium wire with the carbon cloth. The SBP-anode approach may protect the exoelectrogenic bacterial anode from the invasion of non-exoelectrogenic bacteria that may reduce the electron transfer from the bacteria to the carbon-cloth anode. The non-exoelectrogenic bacteria can also compete with the exoelectrogenic bacterial anode for nutrients. In addition, the SBP anode is stable in withstanding the shear and friction forces found in domestic wastewater treatment plants. All these improvements may accelerate electron transfer and the HER.",
"introduction": "1. Introduction Large amounts of electrical energy are spent globally to operate domestic wastewater treatment plants (air blowers, hydraulics, and waste sludge stabilization and dehydration). Using appropriate technologies, the potential chemical energy contained in the organic compounds in domestic wastewater might help to improve the energy yield and cost efficiency of treatment plants [ 1 , 2 ]. Among the several chemicals that may be extracted from wastewater, hydrogen occupies a preeminent position because of its desirable potential as a fuel: it is a clean and neutral energy carrier, and can be converted directly into electrical energy very efficiently using fuel-cell technology [ 3 ]. Currently, most hydrogen production occurs via water splitting, and is considered a clean energy source. However, this technology is based on applying high voltage (1.5–2 V) and/or high temperatures (225–475 °C) [ 4 , 5 ]. Bioelectrochemical systems (BESs) in general, and microbial electrolysis cells (MECs) [ 6 , 7 ], as well as microbial fuel cells (MFCs) [ 8 ] in particular, represent an emerging technology capable of harvesting part of the potential chemical energy in organic compounds. The anodic reactions in MFCs and MECs are quite similar. Almost any source of organic matter, such as the carbohydrates and lipids usually found in wastewater, can serve as a suitable carbon source for bacterial anode activity [ 7 , 9 , 10 ]. In both MFCs and MECs, the bacterial anode activity leads to the generation of electrons and protons. The main differences between these technologies are in the cathode activity and the operation mode. In MFCs, the presence of an oxidative agent (oxygen) causes the electrical current to flow spontaneously [ 11 , 12 ], whereas MECs require a certain amount of electrical input (about 0.3–0.8 V) to drive the redox reactions [ 13 , 14 , 15 ]. Electricity and hydrogen are the outcome of MFCs and MECs, respectively. Therefore, they both have the ability to convert some of the negative-value waste streams into value-added products. Since a BES as a stand-alone treatment presents limited capabilities in removing carbon and nitrogen, it is likely that a BES operated in a wastewater treatment plant would require an additional treatment step. This leads to the operational concept that a BES application, which consists of bioelectrochemical treatment, can be operated in the anaerobic pond, prior to the anoxic pond, followed by an aerobic step [ 16 ]. Moreover, it is essential that MEC and MFC technology designs be suitable for long-term implementation in the continuously hostile environment found in domestic wastewater treatment plants. One of the main obstacles that has been identified is the bacterial anode of a BES, which usually contains mixed cultures in addition to the exoelectrogenic microorganisms [ 17 ]. Non-exoelectrogenic microorganisms can act as a physical barrier by settling on the anode surface and displacing the exoelectrogenic microorganisms. In addition, these non-exoelectrogenic microorganisms can induce negative interactions with the exoelectrogenic culture, such as competition for nutrients and carbon sources. These phenomena can reduce electron transfer from the exoelectrogenic bacteria to the anode material [ 18 ]. In addition, there are obstacles related to the operational needs of domestic wastewater plants that can reduce the feasibility of BES implementation. These can include shear forces, friction forces, and unstable inflow rates [ 19 ]. The anode material in BESs should be highly conductive and biocompatible. Carbon-based materials are commonly used as anode materials due to their chemical stability, conductivity, and cost-effectiveness [ 20 , 21 , 22 , 23 ]. However, the carbon materials’ hydrophobicity prevents stable bacterial attachment. A pretreatment with acid and high temperature was reported to overcome the hydrophobicity patterns of carbon materials. For example, high temperature in a muffle furnace at 450 °C for 30 min was used by Wang et al. (2009) [ 20 ]. Soaking in acid (H 2 SO 4 ) while heating was used by Scott et al. (2007) [ 24 ]. However, these pretreatments are performed in extreme conditions. The environmentally friendly method of employing cold nitrogen plasma showed enhancement of the hydrophilicity of a carbon-felt anode, which led to better biofilm formation [ 25 ]. Other ways to increase bacterial attachment include the immobilization of bacteria with organic polymers. Gandu et al. (2020) immobilized G. sulfurreducens on a carbon-cloth anode using alginate and chitosan. When the MEC, which was based on the immobilized bacterial anode, was fed with wastewater, the current density at a potential of 0.2 V was 11.52 A m –2 , 29% higher than the non-immobilized anodes, and the hydrogen evolution rate (HER) was 0.56, m 3 m −3 d −1 , 70% higher than the non-immobilized anode [ 26 ]. Another approach to immobilizing the bacterial anode was suggested by Rozenfeld et al. (2021). In their study, the bacterial anode was encapsulated in a dialysis bag. The anode material was made of carbon cloth combined with stainless steel, encapsulated in a dialysis bag including a suspension of G. sulfurreducens. The current densities obtained at 0.6 V vs. Ag/AgCl were 16.34 ± 0.42 A m −2 , whereas the MEC that employed a nonencapsulated anode led to only 12.19 ± 0.49 A m −2 [ 27 ]. To protect the exoelectrogenic bacterial anode from the shear and friction forces found in domestic wastewater treatment plants, a novel encapsulation technology in a small bioreactor platform (SBP) was adopted. The SBP technology was recently developed for achieving successful bioaugmentation in wastewater treatment plants. The SBP method is based on macro-encapsulation of bacterial culture in a confined environment using a microfiltration membrane as a protective barrier. A 3D capsule (2.5 cm in length, 0.8 cm in diameter) physically separates the bacterial culture, including the carbon cloth-anode material, from the natural microorganisms in the wastewater, while enabling the diffusion of nutrients through the capsule membrane [ 19 ]. The SBP capsule creates an appropriate growth environment by providing nutrients and physical protection and preventing competition with natural microorganisms in wastewater plants. This results in rapid biomass acclimation within the SBP capsules. Furthermore, the physical barrier prevents the washout of the selected microorganisms from the bioreactor in a continuous outflow [ 28 ]. Our study adopted SBP technology to operate as an anode for a novel MEC. The MEC based on the SBP anode was fed with Geobacter medium supplied with acetate for 32 days, and then with artificial wastewater for another 46 days. The electrochemical activity, COD removal, bacterial anode viability, and relative distribution of the MEC-SBP anode were compared with an MEC based on a nonencapsulated anode.",
"discussion": "3. Results and Discussion 3.1. LSV Measurements of MECs Which Were Fed with Acetate as the Sole Carbon Source A single-chamber MEC was constructed based on a carbon-cloth anode encapsulated in an SBP (SBP anode) and a platinum-coated carbon-cloth cathode; designated as MEC-SBP. An MEC with the same cathode and anode materials, but without encapsulation in SBP, served as a control, defined as the MEC control. The MEC-SBPs were inoculated with 0.6 mL (1.0 ± 0.05 OD) of G. sulfurreducens culture injected directly into the SBP capsule. Thus, the exoelectrogenic G. sulfurreducens were in a confined environment next to the carbon-cloth anode material. The MEC controls were inoculated by injection into the medium of the MEC facility (100 mL). The medium was replaced once a week, and twice a week the MECs were fed with acetate. The MECs were provided with acetate as the sole carbon source for 32 days and maintained under an external voltage of 0.3 V vs. Ag/AgCl. LSV measurements were performed once a week. The LSV analyses on the 14th and 27th days are shown in Figure 2 A,B, respectively. The results depicted in Figure 2 A,B show that the MEC-SBP, based on the encapsulated anode, led to higher currents compared to the MEC control. On the 14th and 27th days, under an applied voltage of 0.6 V vs. Ag/AgCl, the current densities of the MEC-SBP were 1.61 ± 0.11 and 1.64 ± 0.29 A m −2 , respectively. In comparison, the MEC control yielded currents of only 0.25 ± 0.02 and 0.48 ± 0.02 A m −2 , respectively. The higher observed currents of the SBP anode occurred despite the high onset potential (−0.2 V) compared with the control (−0.55 V); and are in line with the low resistance (ca. 0.4 Ω m 2 ), which was 6.25 times higher than the control (2.5 Ω m 2 ) ( Figure 2 B). The encapsulated anode exhibited electrochemical behavior different from the free-standing carbon electrode, as seen in their onset potentials and the above-mentioned currents. While the exact reason is not clear to us, we think it had to do with the entrapped bacterial cells, their released redox-active mediators, and the interaction of these two with the active biofilm on the carbon electrode inside the capsule. The enrichment of this micro-environment altered the equilibrium potential (and consequently the onset potential), but it also provided a more accessible and higher concentration of bio-electrochemical active species to support a high current. These conditions did not exist in the diluted supernatant surrounding the carbon-cloth anode in the MEC control, which was a free-standing electrode under the same physical and chemical conditions. 3.2. LSV Measurements of MECs Utilizing Artificial Wastewater as Carbon Source The MECs were fed with acetate as the carbon source for 32 days followed by artificial wastewater for another 46 days. In this second period, the artificial wastewater was replaced once a week and acetate was added twice a week. LSV measurements were performed at least once a week. LSV measurements on the 36th and 57th days of the MECs’ operation are shown in Figure 3 A,B. The results depicted in Figure 3 A,B also showed that the MEC-SBP with the encapsulated anode led to higher currents than the MEC control. On the 36th and 57th days, under an applied voltage of 0.6 V vs. Ag/AgCl, the current densities of the MEC-SBP were 1.72 ± 0.15 and 1.70 ± 0.22 A m −2 , respectively. In comparison, the MEC control yielded currents of only 0.92 ± 0.09 and 0.78 ± 0.01 A m −2 , respectively. A slight decrease in the onset potential (from −0.04 to −0.2 V) of the MEC-SBP was seen when the devices were fed with acetate on the 14th and 27th days, respectively. The onset continued to decrease (to about −0.35 V) when the MECs were fed with wastewater. This phenomenon could explain the increase in the currents on the 36th and 57th days, compared to the 14th and 27th days. Regarding the MEC control, the currents increased moderately from 0.3 A m −2 on the 14th day to 0.5 A m −2 on the 27th day. This may be attributed to the slow formation of the biofilm on the carbon-cloth anode, due to low bacterial inoculation, 0.6 mL (1.0 ± 0.05 OD) in a relatively high volume (100 mL) of the MEC control. In contrast, the MEC-SBP inoculation was introduced directly to the SBP capsule, which favored the biofilm formation because of the small void volume of the capsule (0.6 mL). To summarize, on the 14th and 27th days, when the MECs were supplied with acetate, the MEC-SBP led to currents higher by 6.44-fold and 3.42-fold, respectively, compared to the MEC control. On the 36th and 57th days, when the MECs were supplied with artificial wastewater, the MEC-SBP led to currents higher by 1.87-fold and 2.18-fold, respectively, compared to the MEC control. These results showed that the MEC-SBP demonstrated better electrochemical performance than the MEC control. In our previous studies, the bacterial anode was immobilized using natural polymers (alginate and chitosan) or encapsulated in a dialysis bag with different molecular-weight cutoffs. It was shown that when the bacteria on the carbon-cloth anode were immobilized using alginate and chitosan (1 mL alginate and chitosan was mixed with G. sulfurreducens (1 OD at 590 nm)), the current density at 0.6 V was 9.75 A m −2 , whereas when the optical density was only 0.1 OD, the MEC produced a current density of only 5.04 A m −2 [ 26 ]. In the other study, the anode was made of carbon cloth combined with stainless steel and encapsulated in a dialysis bag with different molecular-weight cutoffs of 2 kDa, 14 kDa, and 50 kDa. In these MECs, the encapsulated anodes were inoculated with a suspension of G. sulfurreducens (10 mL with an optical density of 1 at 590 nm). The current densities obtained at 0.6 V vs. Ag/AgCl were 13.79 ± 0.30, 14.94 ± 0.49, and 16.34 ± 0.42 A m −2 for the anode encapsulated in a dialysis bag with molecular-weight cutoffs of 2 kDa, 14 kDa, and 50 kDa, respectively. In the MEC that employed a nonencapsulated anode, the current density was only 12.19 ± 0.49 A m −2 [ 27 ]. In the study of Zikmund et al. (2018), the MECs (28 mL) were operated under 0.9 V and were based on graphite-fiber brush anodes (2.5 cm length, 1.5 cm diameter, encased volume of 4.4 cm 3 ) and carbon-felt anodes (7 cm 2 cross-sectional surface area). The anodes were placed close to the cathode to reduce the electrolyte resistance between electrodes. The MEC based on the brush anode led to a current density of I 90 = 4.2 ± 0.5 A m −2 , compared to the felt anode, which led to I 90 = 3.4 ± 0.1 A m −2 [ 34 ]. In summary, the inoculum concentration and favorable conditions for biofilm attachment are vital for the MEC electroactivity. As mentioned above, favorable conditions can be obtained by a high anode surface area [ 34 ], immobilization using alginate [ 26 ], inoculation in a dialysis bag [ 27 ], or inoculation into encapsulated SBP anodes, as described in our current study. These conditions also enabled earlier biofilm formation. 3.3. Reduction Currents and Hydrogen Production Reduction currents were measured at least once a week during the MECs’ operation, when the sole carbon source was acetate ( Figure 4 A), and when the MECs were fed with artificial wastewater ( Figure 4 B). The reduction current analyses were conducted when the MECs were in a configuration of a complete cell (two-electrode configuration). The cathodic current of hydrogen evolution on the Pt electrode in a two-electrode configuration is shown in Figure 4 A,B. At an applied maximum voltage difference of −0.8 V, the reduction currents to hydrogen gas in the MEC-SBP were higher than in the MEC control when cells were enriched with acetate, and to a lesser extent, also, when fed with artificial wastewater. The choice of −0.8 V represents a value at which the MEC had an advantage over conventional water-electrolysis cells (above 1.4 V). When the MECs were supplied with acetate, the reduction currents of the MEC-SBP were −0.64 ± 0.14 A m −2 (at −0.8 V), while the MEC control led to −0.12 ± 0.05 A m −2 . When the MECs were fed with artificial wastewater, the MEC-SBP produced reduction currents of −0.35 ± 0.07 A m −2 , 2.33-fold higher than the MEC control. The HERs (at −0.6 V) of the MECs, when supplied with acetate as the sole carbon or artificial wastewater, were calculated according to Equations (1) and (2) (given in the Materials and Methods section). In the MEC-SBP, the calculated HERs were 0.027 and 0.017 m 3 m −3 day −1 , respectively. In comparison, the MEC control led to only 0.006 and 0.005 m 3 m −3 day −1 , respectively. Hydrogen production currents under an applied potential of 0.6 V were recorded from steady-state polarization curves measured at various times during the MEC-SBP and MEC control operation ( Figure 5 ). The MECs were supplied with acetate as the carbon source for 32 days, followed by wastewater for another 46 days. The current changes seen in the graph are associated with the increase of the biofilm activity post-feeding steps. The current of the encapsulated anode in the MEC-SBP was higher (between 25–80%) than the bare carbon electrode in the MEC control. From these results, it is obvious that the encapsulation of bacterial anode in the MEC-SBP provides better long-term stability than the biofilm of the control anode. Lim et al. (2022), constructed an MEC with two plain carbon-felt structures as anode and cathode (size 4.8 × 4.8 × 0.2 cm, projected area 25 cm 2 ). The MEC working volume was 25 mL. The hydrogen evolution rate was 0.32 ± 0.01 m 3 m −3 day −1 (6–11 days) and 0.37 ± 0.02 m 3 m −3 day −1 (12–14 days) [ 13 ]. Yasri and Nakhla (2017) investigated an MEC employing granular activated carbon as a 3-dimensional (3D) anode. In different MECs, the anodes were doped with conductive calcium sulfide (CaS), iron sulfide (FeS), and magnetite (Fe 3 O 4 ), and were compared to granular activated carbon without doping. In all anodes, the granular activated carbon (12 g) had a surface area of 900 m 2 g −1 and a theoretical geometric surface area of the granular activated carbon per MEC anolyte chamber volume of 30.8 × 106 m 2 m −3 . The hydrogen production rate values were as follows, in decreasing order: 3-D CaS (0.54 ± 0.03 m 3 m −3 d −1 ) > 3-D FeS (0.46 ± 0.02 m 3 m −3 d −1 ) > 3-D Fe 3 O 4 (0.36 ± 0.02 m 3 m −3 d −1 ) > 3-D granular activated carbon (0.31 ± 0.01 m 3 m −3 d −1 ) [ 35 ]. Zikmund et al. (2018) compared the bio-electroactivity of MECs based on a flat-felt anode with an MEC brush anode in a two-chamber, cubic type facility. The MECs with the brush anodes had a higher HER of 0.38 ± 0.02 m 3 m −3 d −1 , while the flat-felt anodes had only 0.32 ± 0.02 m 3 m −3 d −1 . They suggested that the main reason for the flat-felt anodes’ lower performance was substrate-limited mass transfer [ 34 ]. Wang et al. (2021) examined MEC performance with alkaline thermally pretreated sludge. The pretreatment was done at 90 °C and at 180 °C, allowing the release of more organic matter. The hydrogen yield using pretreated sludge at 90 °C was 0.44 m 3 m −3 d −1 , while at 180 °C it was only 0.31 m 3 m −3 d −1 [ 36 ]. Gandu et al. (2020) examined HER performance of MECs based on immobilized anodes. The exoelectrogenic bacteria were immobilized using alginate and chitosan. This MEC led to a HER of 0.56 m 3 m −3 d −1 , while the MEC based on non-immobilized anodes led to 0.16 m 3 m −3 d −1 [ 26 ]. Rozenfeld et al. (2021), studied MECs based on encapsulated anodes (carbon cloth combined with stainless steel) with a dialysis bag. An MFC based on dialysis bags with molecular weight cut-offs of 50 kDa led to a HER of 0.160 ± 0.009 m 3 m −2 d −1 , while the bare anode led to only 0.122 ± 0.004 m 3 m −2 d −1 [ 27 ]. In conclusion, the HER of the MEC applying the SBP anode was higher than the MEC based on the control anode, which was not encapsulated. However, the HER of the MEC applying the SBP anode was 5 to10-fold less than the HER reported by other studies. It is important to note that the ratios of anode surface area to MEC working volume are important parameters for hydrogen evolution rates. We assume that expanding the carbon-cloth anode material surface area for biofilm attachment in the capsule and increasing the contact of the external titanium wire with the carbon cloth would improve the SBP-anode performance. 3.4. COD Removal The COD inlet in the MEC-SBP and MEC control was 7400 ± 478 mg L −1 . The COD was analyzed on the 59th, 63rd, and 69th days. COD removal on the 63rd day related to the COD on the 59th day, and the COD on the 69th day related to the COD on the 63rd day. All the samples were filtered, diluted according to the sample concentration, processed in the COD reactor, and analyzed for the absorbance intensity of the solution using spectrophotometry. As seen in Figure 6 , the COD consumption on the 63rd day vs. the 59th day was 33 ± 9.5% in the MEC control, and 25 ± 8.5% in the MEC-SBP. On the 69th day vs. the 63rd day, it was 67 ± 2.0% and 71 ± 1.2%, respectively. There was no significance ( p > 0.05) in the COD consumption in the MEC control vs. the MEC-SBP. The COD consumption rate for both was about the same, 650 ± 70 mg L −1 . Chaurasia and Mondal (2021) studied biohydrogen production using Ni, Ni-Co and Ni-Co-P electrodeposit cathodes in MFCs fed with sugar-industry wastewater. These MEC systems reportedly achieved ~47–50% COD removal. Initially, COD was 4850 ± 50 mg L −1 , and after 7 days of operation was reduced to ~2425 mg L −1 [ 37 ]. Xie et al. (2021), showed that 34.26% of the COD provided was converted to electrical current in an 80-day period when the MEC was fed with rendering wastewater [ 38 ]. Keruthiga et al. (2021), investigated an MEC based on a modified carbon-cloth anode pasted with char and fed with wastewater. The reduction of COD was found to be correlated with acid concentration; increasing the acid concentration from 0.5 to 1.5% increased the COD reduction. The optimum acid concentration of 1.5% hydrolysed the organics effectively, which increased COD reduction to 76.8% [ 39 ]. Yu et al. (2021), studied a cylindrical-chamber MEC with a bed volume of 28 mL and a graphite brush anode of various sizes (surface area was 0.22 m 2 ). COD removal efficiency in that case was 40.33% [ 40 ]. In conclusion, the COD removal in the MEC-SBP and the MEC control was similar to the COD removal reported in other studies. 3.5. Biofilm Viability on the Bacterial Anodes Biofilm viability was evaluated based on the reduction of tetrazolium salts by the bacterial hydrogenase. At the end of the experiment (day 78), the MEC-SBP capsule was cut, and the carbon cloth with the attached biofilm was removed. The bacterial anodes from the MEC control and MEC-SBP were gently washed in PBS (pH 6.8) to release the planktonic bacteria. The carbon-cloth anodes with their attached biofilms were transferred to MTT solution including tetrazolium salts. The bacterial hydrogenases reduced the yellowish crystals of the tetrazolium salts to purple, then the reduced tetrazolium salts were dissolved in a DMSO–ethanol solution. The absorbance intensity of the purple solution was examined using a spectrophotometer ( Figure 7 ). The results in Figure 7 show that the viability of the biofilm on the control anode was twice as high as that observed on the SBP anode, 0.71 ± 0.07 OD vs. 0.34 ± 0.04 OD, respectively. The lower bacterial viability on the MEC-SBP anode can be explained by the condensed carbon cloth at the bottom of the capsule. We assume that this pattern of folding inhibited bacterial attachment in the depth of the carbon cloth. However, the control carbon-cloth anode was bare and unfolded, with no limitation for bacterial attachment. In addition, the internal volume of the SBP was restricted to inoculation of only 0.6 mL (1 OD 590 nm). Our previous study of a semi-single-chamber MEC was based on an anode encapsulated in a dialysis bag inoculated with Geobacter sulfurreducens (10 mL of 0.35 ± 0.05 OD), and presented a significant difference in the bacterial viability (using MTT analysis) between the anode types. When the MECs were fed with wastewater, the encapsulated anode’s viability was 2.5-fold higher than the nonencapsulated anode [ 27 ]. This finding might determine the biofilm-forming potential of the MEC-SBP anode once technical challenges are overcome. 3.6. Microbial Diversity on the Carbon-Cloth Anode The MEC systems were inoculated with Geobacter sulfurreducens and fed with acetate for 32 days, followed by feeding with artificial wastewater for an additional 46 days. The artificial wastewater included Staphylococcus aureus , Escherichia coli , Enterobacter cloacae , and Pseudomonas putida to demonstrate wastewater flora. The microbial diversity on the MEC-SBP and MEC-control anodes was evaluated based on 16S rRNA at the end of the MECs’ operation (78 days). Operational taxonomic unit (OTU) readings were identified and phylogenetically classified. Five distinct phyla ( Proteobacteria , Firmicutes , Bacteroidetes , Actinobacteria , and Euryarchaeota ) were identified. The three most abundant phyla were Proteobacteria, Firmicutes , and Actinobacteria , with a relative distribution on the MEC-control anode of 32%, 32% and 14%, respectively. On the MEC-SBP anode, they were 40%, 15%, and 30%, respectively. Relative bacterial distribution with respect to the genus level of the anode biofilm is presented in Figure 8 . Unidentified species or sequences with relative abundances of <2% were grouped as “Others/NA”. Geobacter relates to the class Deltaproteobacteria ; its distribution on the MEC control and MEC-SBP was 8% and 9%, respectively. The low percentage of Geobacter can explain the low reduction currents obtained, as seen in Figure 4 . The currents obtained in the MEC control and MEC-SBP were −0.15 A m −2 and −0.35 A m −2 (at −0.8 V), respectively. However, Rhodococcus erythropolis was found in a relatively high distribution on the MEC-SBP (18%), while its distribution was negligible on the MEC control. We assume that the presence of R . erythropolis only on the MEC-SBP originated from self-contamination of the capsule. It is possible that Rhodococcus erythropolis also contributed to the currents found in the MEC-SBP. It was reported that Rhodococcus erythropolis was one of the dominant bacteria found on the biocathode of MFCs facilitating the mineralization of pentachlorophenol [ 41 ]. This bacterium was also identified in the microbial community of MFC biocathodes used for Cr(VI) reduction [ 42 ]. An additional species, Rhodococcus pyridinivorans, which was inoculated into MFCs, improved their power output. In this MFC, increasing the concentration of trehalose led to a 5.93-fold acceleration of the maximum power density, from 54.7 mW m −2 to 324.4 mW m −2 [ 43 ]. An interesting study by Taşkan and Taşkan showed that quorum quenching of the Rhodococcus sp. can control the biofilm thickness on the anode surface by inactivation of signal molecules among microorganisms, which reduces the production of extracellular polymeric substances. It was found that increases in Rhodococcus concentrations led to a reduction of the anode biofilm thickness and an abundance of dead bacteria. The best electrochemical activity (1924 mW m −2 ) was in an MFC with a biofilm thickness of 26 μm at 40 mg, using Rhodococcus immobilized in 10 mL sodium alginate [ 44 ]. In conclusion, we assume that the combination of Geobacter and Rhodococcus in the SBP anode led to a relatively higher current than in the control anode. Concerning bacteria included in the artificial wastewater, Staphylococcus was found in negligible percentages, and Escherichia and Enterobacter were less than 10%. In contrast, the abundance of Pseudomonas was 22% on the bacterial anode of the MEC control, and only 13% on the MEC-SBP. Pseudomonas species are abundant in the microbial community of MFCs, and they are known to secrete electron mediators such as pyocyanin and phenazine with a redox potential of −0.03 V (versus SHE) [ 45 , 46 , 47 ]. Pseudomonas alcaliphila can excrete phenazine-1-carboxylic acid, which transfers electrons under alkaline conditions in the MFC. Results indicated that phenazine-1-carboxylic acid was a key factor for extracellular electron transfer [ 48 ]. 3.7. SEM Analysis of the Bacterial Anodes At the end of the MECs’ operation (78 days), the anodes were dehydrated and prepared for visualization by SEM. The images are shown in Figure 9 . The SEM images of the MEC control ( Figure 9 (A1,A2); magnification 3 kx and 50 kx, respectively) showed bacterial aggregation, especially on the carbon fibers (A1); and bacterial cells with relatively low matrices (A2). However, the SEM images of the SBP anode showed a massive biofilm on and between the fibers ( Figure 9 (B1)), and a very dense biofilm ( Figure 9 (B2)). It is important to note that there was still space for substrate access despite the massive biofilm on the SBP anode. Moreover, the biofilm’s morphological structure was quite different in the two systems. MTT analysis showed higher biofilm viability (2-fold) on the control anode compared to the biofilm on the SBP anode. In contrast, SEM images showed less biofilm on the control anode, indicating different biofilm development over time. We assume that the biofilm on the control anode was looser, probably influencing electron flux to the carbon-cloth anode material. Ishii et al. revealed that the biofilm on the anode was nonhomogeneous at the beginning of MEC operation (11 days). There were many large aggregates, and the electrode was partially covered by the bacterial cells. After long-term operation (216 days), there was an increase in the coverage area of G. sulfurreducens cells, resulting in a dense biofilm on the anode. They indicated that the limiting current density changed proportionally to biomass densities on the anode [ 49 ]. Liu et al. showed bacterial anode images on different electrode materials, graphite rods, and carbon-fiber veils. On the carbon rod, a thick and dense biofilm was observed, but on the carbon-fiber veil, the biofilm colonized every carbon fiber with a thickness of more than 10 µm. There was more space between intersectional carbon-fiber biofilm on the carbon fibers. This porous structure may provide better substrate access, resulting in high current density [ 50 ]. Chang et al. modified a carbon-cloth anode surface by screen-printing reduced graphene oxide, and calcination using an atmospheric-pressure plasma jet. Both treatments significantly increased the hydrophilicity and surface area of the effective materials for bacterial adhesion [ 51 ]."
} | 8,537 |
36862240 | PMC9981844 | pmc | 4,455 | {
"abstract": "Background Enhancement of lipid accumulation is the major strategy to improve the commercial feasibility of microalgae as a source for biodiesel production. Pseudochlorella pringsheimii (Formally was named as Chlorella ellipsoidea ) green microalgae strain was chosen with respect to their ability as a potential source to produce high lipids content, could be used for the production of biofuel, which can be an alternative renewable energy source instead of fossil fuels. Results Initially, the Pseudochlorella pringsheimii microalgae was evaluated on the basis of tested at Lab scales 2 L by applicable different nutrient individual of N, P, Fe conditions in BBM medium concentrations for choosing the best concentrations induce lipid contents and productivity to cultivate in large scale in the 2000 L PBR. The suitable concentrations of nutrients with highest lipid contents were obtained under deficient of nitrogen (1.25 gL −1 , limited N) and phosphorus (0.1 mg L −1 , limited P) coupled with high iron concentration (10 mg L, rich Fe) and CO 2 (6%). Therefore, their collective of nutrients was applied to culture of microalgae cells at large scale in 2000 L photobioreactor (PBR model), which, this techniques was used to quantify high lipid contents (25% w/w) and high lipid productivity (74.07 mgL −1 day −1 ). The inducted lipid conversion to biodiesel via transestrification process was 91.54 ± 1.43%. The fatty acid methyl esters (FAMEs profile by means of GC/MS resulted in C16:0, C18:1, C18:2, C18:3 as a main constituents. With regard to physical–chemical property (such as density, kinematic viscosity, gravity, and certain number), the Pseudochlorella pringsheimii biodiesel have biofuel properties, in accordance with appropriate biodiesel properties, as ASTM and EU standards, that thereby referring to high quality biodiesel. Conclusions Pseudochlorella pringsheimii cultured in large scale in photobioreactor under stress condition have a high potential of lipids production with high quality of FAMEs that can be used as a promising biodiesel fuel. It has also a potential to be applied for commercialization based on the techno-economic and environmental impacts.",
"conclusion": "Conclusion The microalgae oils represent a promising feedstock for biodiesel. Microalgae can be cultured to economical production of lipid through the nutrient stresses. The lipid content and biomass productivity as well as lipid composition of microalgae can be made nutrient stress. Thus, the optimization of the nutrient stress and cultivation process was applied to increasing the lipid content with provable fatty acid compositions for biodiesel production from Pseudochlorella pringsheimii microalgae. Under lab scales (2 L medium), the microalgae cultured in N- and P -limitation as well as Fe richest medium (in individual set), resulted a relative increase those parameters. The optimized value of (limited of N + P combined with high Fe) was choosing to cultivate of microalgae in large scale in the 2000 L PBR. Their collective of nutrients greatly improves the quantify lipid contents (25% w/w) and lipid productivity (74.07 mgL −1 day −1 ). The inducted lipid was converted to biodiesel via transestrification process yielded of 91.54 ± 1.43%. The fatty acid profile analyzes by GC/MS resulted in C16:0, C18:1, C18:2, C18:3 as a main constituents. Physical–chemical property (such as density, kinematic viscosity, gravity and cetane number) of the Pseudochlorella pringsheimii biodiesel was found to comply with ASTM D6751 and EN 14,214 standards, that thereby referring to high quality biodiesel.",
"discussion": "Discussion Nutrient composition have an important role in metabolic pathway of microalgae constituent includes primary and second metabolites as sources for biofuel, fertilizer, cosmetics, nutraceuticals and pharmaceutical material [ 2 , 3 ]. Cultivation of macro or micro-algae in depletion and excessive sources of nutrient might affect the quality of biomass and quality and quantity of bio-molecule such as lipid, protein, pigments and polysaccharides. However, investigation on the impact of nitrogen, phosphorus and ferric, which are crucial for the growth of algae has been addressed [ 1 – 5 ]. On other hand, enhancing nutrient utilization efficiently for cultivated of microalgae at large scale cultivation. Hence, this study aims to highlight the concentration of N, P and Fe required for Pseudochlorella pringsheimii microalgae cultivation to increasing biomass productivity and lipid contents and to produce fatty acid quality for biodiesel production. In this study, Pseudochlorella pringsheimii was grow firstly at lab scales (2 L flasks) in order to optimizes the suitable nutrient condition, to get a high growth rate, high lipid content for biodiesel production. The carbon concentrations, and three elements availability (N, P, Fe) in the microalgae growth medium required for the metabolic pathway switch to lipid accumulation was determined.. The data microalgae revealed that the lipid content and lipid productivity of microalgae was significantly affects by treated with carbon source and concentration ( p < 0.05), the higher values was obtained in 8% and 16% gaseous CO 2 cultures (Table 1 ). The correlation coefficient (R 2 ) between the microalgae biomass and the carbon sources and level (Fig. 1 ) in nutrient medium throughout the cultivation period was found to be very high (R 2 ranged from 0.956 – 0.997). However, lipid content lipid productivity of Pseudochlorella pringsheimii cultures in sodium bicarbonate salt (NaHCO 2 ), the biomass yields (as dw) were found to be lower than in CO 2 cultures, that suggests the occurrence of NaHCO 2 stress conditions induced by high bicarbonate levels in the medium due to excess osmotic pressure [ 4 ]. The CO 2 gases could be more economically sustainable than that supplied by an exogenous organic carbon source [ 19 ]. The high lipid productivity and lipid contents are considered as the most important desirable characteristic of chosen the microalgae strains and nutrient for biodiesel production [ 20 ]. However, the high concentration of CO 2 gases had a significant effect on the growth of microalgae that the utilized CO 2 in culture will be converted to carbonic acid (H 2 CO 3 ), as result reducing the pH value of the culture [ 4 ]. Therefore, to obtain enhanced biomass and lipid production requires optimal CO 2 concentration. In this regard, the Dunaliella salina had high lipid content when cultivated at 6% to 10% CO 2 gases as compared with that cultivated in 1% CO 2 . However, 8% CO 2 was used as carbon source for cultivation of several microalgae species to provide the best result in terms of higher oil yield and lipid productivity [ 4 , 5 and 21 ]. Nitrogen is an essential macronutrient for microalgae growth and plays an important role in primary constituents like protein, lipid and nucleic acids, chlorophyll and in syntheses the energy transfer molecules viz adenosine triphosphate ATP [ 2 , 5 ]. As showed in Fig. 2 and Table 2 , Pseudochlorella pringsheimii grow in MMB medium with different concentrations of N (KNO 3 , free N (0 N free), ½ (0.125 g/L), and double (0.50 g/L) of optimum) had a significant differences in biomass and lipid content. AsN concentration increasing the biomasses and cells growth was increased with high correlation coefficient (R 2 ranged from 0.978 to 0. 934). In the absence of N free source (0 g/ L free N) little and slowly growth was observed and the cells appeared in bleached form. In contrast, in rich N medium (double N concentration 0.50 g/L, rich N), the maximum biomass of 1.86 g/L (dw) was recorded (Fig. 2 ). Thus, the Pseudochlorella pringsheimii cannot grow without a nitrogen source and its biomass as a function of algae growth is directly proportional to the concentration of N in the medium. Regard to lipid content, the higher (28.14%) lipid content was got when grown in limited N of 0.125 g/ L N than that in rich or free nitrogen culture (Table 2 ). However, the N level in medium demonstrated significant negative correlations with the lipid contents in algae cells ( p < 0.05). Thus, the Pseudochlorella pringsheimii can grow in N limited medium and accumulated lipid, with inverse relationship to the concentration of N in the medium. This result is in accordance with in same strain that the biomass was increased as function for increasing of N concentration [ 1 , 4 and 5 ]. In general, nitrogen concentration significantly influences microalgae growth rate and their lipid content and lipid productivities, that depletion of N in cultivation medium causes a decrease in growth rate, but lipid productivities was increased. Therefore, a good relationship between the nitrogen dose and lipid content in the microalgae biomass was reported, which with nitrogen deficiency, the metabolic pathway of carbon fixation changes from protein synthesis to lipid production [ 21 – 24 ]. It can be concluded that nitrogen concentration favors higher biomass productivity and depletion of nitrogen shifts the flux to lipid production [ 7 , 25 – 28 ]. Phosphorus is also essential element that plays an important role in microalgae growth, lipid production and fatty acid yield is required for metabolic processes such as energy transfer and signal transduction and photosynthesis (DNA, RNA and ATP) and nucleic acid synthesis [ 26 ]. As shown in Fig. 2 , P concentrations in BBM medium (0 P free), (0.175 g/L optimum and double (0.350 g/L) had a high impact on growth and lipid content of Pseudochlorella pringsheimii. The increases of biomass dry weight was increased with increasing of P concentration and time, with highest correlation coefficient ( R 2 = ranged from 0.954 to 0.982). The biomass as a function for cells growth was slightly increased and in the absence of P free medium (0 g/ L free P) very slowly culture growth was observed. In contrast, in rich P sources (double the P, 0.350 g/L, rich P), the maximum biomass (dry weight dw) concentration of 1.356 g/L was recorded. The cultures grow in rich P medium and standard P BBM BG11 medium had highest biomass concentrations that in free P culture. The microalgae cultured in P-rich medium produced less lipid contents of 4.38% compares with that (12.36%) in P standard medium and in Free P (15.65%) medium. Thus, statistically, there is high significant difference between the lipid contents of P-rich or P- stander cultures and in the P-deficient BBM medium ( p > 0.05). These results revealed that P is the most important nutrient for the cell proliferation and has significant impacts on cell growth of Pseudochlorella pringsheimii microalgae. The results were consistent with those previously reported in literature that the P deprived conditions led to increase the lipid content in Phaeodactylum tricornutum , Chaetoceros sp., Isochrysis galbana and Pavlova lutheri cultures [ 27 – 29 ]. In Chlorella sp . the higher lipid accumulation was observed with decreasing of P concentrations in nutrient media [ 30 ]. Fan et al. [ 31 ] who reported that the growth rate, biomass productivity and high accumulation lipids in Meyerella sp., Chlamydomonas reinhardtii , Botryococcus sp. and C. pyrenoidosa when cultured under nutrient stress condition. Abd El Baky and El Baroty [ 5 ] have also reported a decline in growth (dry weight) of microalgae under unfavorable conditions (P or N limitation) and is seems to be the most promising combination conditions for high lipid production. The Fe ion as the most important trace metals has a significant effect in the growth rate, lipids and carbohydrates content in numerous microalgae. Ferric ions are involved in fundamental enzymatic reactions of photosynthesis and in regulating the gene expression and in metabolism pathway in algae [ 32 ]. As shown in Fig. 2 , the biomass (dw) concentration in Pseudochlorella pringsheimii cells was increased gradually as Fe increased in nutrient medium. The biomass (dry weight dw) concentration were significantly inbreeded gradually by increases of Fe 3+ iron concentration, with high correlation coefficient ( R 2 = 0.967 to 0.978). The similar trend was observed that the total lipid content was significant increased with increasing Fe ion (Table 2 ).The high total lipid content was found to be 19.25% were obtained in 10.0 mg/L Fe culture, which to be 2-folds) than that in free Fe 3+ culture. Thus, in rich Fe medium, the microalgae biomass and its quality are closely related to microalgae growth rates, that a higher growth rate could produce a higher biomass and lipid content. This observation was corroborated was that reported by Abd El Baky et al. [ 5 ], who showed that Fe-rich conditions were suitable for lipid generation in S. obliquus . In microalgae S. obliquus , the total lipid content in cultures supplemented with high Fe 3+ concentration exhibited a higher biomass dry weight than that in media supplemented with lower iron concentration. In Botrycococcus spp , the lipid content and lipid productivity were increased when grow with high iron concentration in combination with limited nitrogen condition [ 24 ]. Chew et al. [ 33 ] have demonstrated that Monoraphidium sp . algae can simultaneous increase of both lipid content and growth rate by increasing of bio-available iron Fe 3+ concentration in the growth medium. It well know that that the increased concentration of iron in medium lead to an increased production of free radicals, which might have changed the metabolic pathway toward to lipids accumulation in the microalgae through increase the activity of acetyl-CoA carboxylase enzyme which, accelerating the biosynthesis of the fatty acids by carboxylation of acetyl-CoA to malonyl-CoA (precursors for lipid accumulation). Finally, high lipid content were obtained in either P starvation Nor in rich Fe ion cultures The relative levels of these effects on lipid content are similar to those on the growth as follows: N starvation > P starvation > Fe starvation. It interestingly, to known that the N and P starvation were the most effective strategy to increase the lipid content in microalgae cells. The individual experimental showed that relative higher lipid content and lipid productivity was obtained in Pseudochlorella pringsheimii cultured under: N starvation, P starvation and rich Fe ions condition. Consequently, the microalgae was cultivated in photobioreactor (2000 L) in medium contained a combined (N, P and Fe) either nitrogen (1.25 g/L) and phosphorus (0.35 g/L) deficient and high concentration of iron ion (10 mg/L) and aerated with carbon dioxide (8% CO 2 ) as a gases carbon source was selected as an effective strategy to production of biodiesel (Table 3 and Fig. 3 ). Under indusial scales condition, the values of biomasses dry weight (4.0e5 g/L), lipid content (24.98%), lipid productivity (1.01 g/L) and total Chl a (96.31 mg/L) was calculated at late log phase (15 days). The specific growth rate μ (d −1 ), biomasses productivity biomass productivity (g −1 L −1 d −1 ) was 0.274 and 0.270, respectively (Table 5 ). This result confirmed the fact that nitrogen and P starvation is the important nutrient for enhance of accumulation of lipid in microalgae. However, microalgae cultivated under optimal optimum conditions,, show high biomass growth but do not accumulate a large amount of reserve materials such as lipids which are useful for biofuel production [ 5 , 7 ]. The direct transesterification TE reaction condition was optimized as a way to improve the biodiesel yield of algae lipids [ 3 ]. As shown in Table 4 , The biomass of Pseudochlorella pringsheimii cultured at large scales in collective (N + Fe + P) medium subjected to a directly TE in one step to production of FAME biodiesel exhibited a high conversion percentage and biodiesel productivity was found to be about 97.64% and 0.067 mg −1 L −1 d −1 , respectively (Table 4 ). These values are in specified limit (96.5%) reported for vegetable seed oils [ 34 ]. In many report that the Feedstock lipids (vegetable oils and microalgae lipid) were achieved maximum biodiesel yield ranged 70% to 97%, depending on the extraction-transesterification conditions [ 2 – 4 , 35 ]. Pseudochlorella pringsheimii fatty acid FA) methyl esters profile was estimated by GC/MS analysis (Tables 5 , 6 ). The data revealed that the most common fatty acid profiles consist mainly of C16:0, C18:0, C18:1 and C18:2 fatty acids (FAs). The fatty acid constituents have a high influence on fuel property [ 4 , 5 ]. Since no individual fatty acid is responsible for any particular fuel property [ 36 – 38 ]. The composition percentage (%) of microalgae biodiesel were 33.98% saturated fatty acids (palmitic acid, stearic acid), 50.78% monounsaturated (C16 and C18 together) and 21.95% polyunsaturated fatty acids (C16 and C18 group together). A higher percentage of saturated fatty acids and monounsaturated could improve fuel property such as oxidative stability and cetane number of biodiesel of produced. Also, higher percentage of unsaturated and monounsaturated could be enhancing the biodiesel characteristic that is responsible for better cold flow properties and oxidative stability. Table 7 shows the physical–chemical property of biodiesel of Pseudochlorella pringsheimii includes : the iodine value (IV), acid value (AV), saponification value (SV) and peroxide value (PV) that was determined based on ACOS method [ 14 ]. The results revealed that the low iodine value (IV, 89.84 ± 3.33 g I 2 /100 g lipid) is a good indicating parameter of the degree of saturation in fuel which has a high influences on the fuel viscosity and cold filter plugging point properties. The lower of AV (0.41 ± 0.09 mg KOH −1 g) may be due to neutralized most of free fatty acids present in algae lipids during the trans-esterification process [ 3 – 5 ]. Also, the moderate saponification value (199.33 ± 1.66 mg KOH/g) indicates higher degree of lower molecular weight of the corresponding microalgae lipid and these results was similar that reported for olive (192) soybean (190)) and sesame (188 mg of KOH/g) crude oils that uses for production of commercial biodiesel. The CN value of Pseudochlorella pringsheimii was 53.46, which refluxed a good fuel property that the biodiesel quality is directly related to CN, an indicator of a fuel’s ignition quality in an engine and combustion quality [ 37 ]. In general, the value Pseudochlorella pringsheimii CN was within that values reported in Desmodesmus abundan s (54.89), D. abundans (58.36) and D. obtusus , (57.49) microalgae biodiesel, which could help ensure better cold start properties and minimize the formation of white smoke [ 38 ]. Also, similar results have been observed in other microalgae biodiesel of Nannochloropsis sp., S. pectinatus and S. obtusus . However, based on cetane hexadecane (C 16 H 34 ) as a long straight-chain hydrocarbon (it is assigned of CN of 100) the most of the vegetable oil biodiesel has CN higher than 51 while the CN of petroleum base diesel (is a mixture of C 8–12 chain hydrocarbon) usually ranges from 40 to 52 [ 39 ]. Pseudochlorella pringsheimii biodiesel is characterized by low PV (< 0.10 ± 0.01 (O 2 meq/kg 1 of lipid). is directly related to have a high oxidative stability of biodiesel against auto oxidation reaction. However, this value was expected to be low, due to no chance for generation of free radicals throughout the biodiesel preparation from algae lipids [ 2 , 5 ]. Moreover, the low values of auto oxidative (OS, 0.169) and degree unsaturation (0.97) of microalgae biodiesel indicted that it has high auto-oxidation stability, which not need to add of any synthetic anti-oxidants (tert-butyl hydroquinone (TBHQ) or 3-tertbutyl-4-hydroxyanisole). On other hand, oxidation stability of biodiesel depends greatly on fatty acid compositions and ratio degree of unsaturated/saturated FAs, biodiesel contained high saturated FA is more stable than unsaturated ones [ 2 ]. Thus, Chlorella biodiesel exhibit superior oxidative stability due to presence of relatively high levels of saturated (SFA, %) and lower polyunsaturated fatty acids (PUFA). The values of density (0.922 ± 0.080 at 15 °C, kgm 3 ) and viscosity (4.98 ± 0.24 at 40 °C, mm 2 /s) were comparable with that recorded in the vegetable (D 0.85 – 0.95; KV 1.9–6.0 mm 2 /s) oils and it was dependent on the quantity of saturated FAMEs, which could be responsible for the moderate value of viscosity and density. However, the lower values of those parameters were desirable for improvements of the low temperature properties of biodiesel [ 37 ]. The heating value (39.22), LCSF (31.42) and CFPP (82.25), are used to evaluate biodiesel quality in terms of ignition readiness; combustion performance and fuel-line plugging temperature were within an acceptable range of EU biodiesel standards. According to those values the Pseudochlorella pringsheimii biodiesel tends to have a better ignition quality (CN values), and exhibit better flow performance at low temperatures (CFPP). These values were found to be within the limit values of international biodiesel fuel standards ASTM D 6751 [ 40 ] and EN 14,214 [ 41 ]. In general, the data of physical and fuel properties (density, CN, CEPP, acidity, and oxidative stability and heating value) of biodiesel from microalgae are comparable to those of conventional diesel. Overall, these results concluded that the Pseudochlorella pringsheimii is a suitable feedstock for biodiesel production due to its high oil content and related to their levels of saturated (palmitic C16:0) and mono-unsaturated lipids (oleic acids C18:1), which is advantageous for higher biodiesel quality."
} | 5,458 |
36844038 | PMC9944057 | pmc | 4,456 | {
"abstract": "Lipid-bilayer nanodiscs and liposomes have been developed to stabilize membrane proteins in order to study their structures and functions. Nanodiscs are detergent-free, water-soluble, and size-controlled planar phospholipid-bilayer platforms. On the other hand, liposomes are curved phospholipid-bilayer spheres with an aqueous core used as drug delivery systems and model membrane platforms for studying cellular activities. A long-standing challenge is the generation of a homogenous and monodispersed lipid-bilayer system with a very wide range of dimensions and curvatures (elongation, bending, and twisting). A DNA-origami template provides a way to control the shapes, sizes, and arrangements of lipid bilayers via enforcing the assembly of lipid bilayers within the cavities created by DNA nanostructures. Here, we provide a concise overview and discuss how to design planar and curved lipid-bilayer membranes by using DNA-origami nanostructures as templates. Finally, we will discuss the potential applications of DNA-origami nanostructures in the structural and functional studies of large membrane proteins and their complexes.",
"introduction": "1 Introduction Over the past four decades, DNA nanotechnology has shown tremendous growth as an outstanding approach for engineering nanoscale molecules using synthetic DNA as constructing blocks ( Rothemund, 2006 ; Seeman, 2010 ; Seeman and Sleiman, 2017 ). As a building material, DNA molecules are considered one of the most defined, predictable, and programmable material due to their specific sequence programmability, synthesis accessibility, rigidity, self-assembly, biocompatibility, and thermodynamic stability ( Chen et al., 2015 ; Seeman and Sleiman, 2017 ; Bujold et al., 2018 ). This flexibility in the structural design by DNA allowed the exclusively de novo designing of precisely defined structures in nearly any shape and size and with additional capability to control self-assembly in both static and dynamic ways ( Jones et al., 2015 ). Several assembly methods have been developed to make DNA nanostructures, including DNA-origami ( Rothemund, 2006 ; Douglas et al., 2009 ), single-stranded DNA tiles ( Wei et al., 2012 ), supramolecular DNA assembly, and polyhedral mesh method ( Benson et al., 2015 ). Among all these DNA assembly methods, DNA-origami has been extensively used due to its robustness and versatility for the custom building of not only 2D and 3D static nanostructures ( Rothemund, 2006 ; Douglas et al., 2009 ) but also dynamic nanostructures, including nanodevices and robots ( Hong et al., 2017 ). Membrane proteins (MPs) are a category of proteins that are constituents of all biological membranes including the plasma membranes and membranes that envelope the intracellular organelles. Based on various predictions by multiple methods, up to 28% of the entire human protein-coding genes encode for MPs ( Lander et al., 2001 ; Fagerberg et al., 2010 ; Attwood et al., 2017 ). The structure of MPs is versatile to provide enormous functionality to the cells. MPs function as receptors, transporters, channels, enzymes, and redox facilitators. In addition, they are pivotal players in several biological processes such as ion transport, cellular signaling, and cell adhesion ( Cournia et al., 2015 ). Given their essential physiological roles, more than 60% of FDA-approved drugs target MPs ( Overington et al., 2006 ). Despite their current immense medical importance and relative abundance, only the structures of a small fraction of MPs have been determined. The main challenges for structural and functional studies of MPs are their insolubility, instability, and inactivity upon isolation from their native lipid-bilayer membranes. Indeed, MPs are neither soluble nor functional without their native lipid-bilayers or lipid-bilayer mimetics ( Yeagle, 2016 ). Various lipid-bilayer mimetics have been developed to stabilize MPs for studying their structures and functions. These membrane mimicking systems include detergent-based platforms (micelles) and detergent-free systems (liposomes and nanodiscs). Micelles can lower the stability of MPs and hence reduce or even abolish their biological functions ( Seddon et al., 2004 ). Additionally, detergents can interfere with the interactions between MPs and their soluble partners. Both liposomes and nanodiscs provide detergent-free phospholipid-bilayer platforms and thus save the stability and functionality of MPs. Liposomes are heterogeneous in size, and thus it is hard to control the number of copies of a given membrane protein for each particle. Nanodiscs are planar lipid-bilayer platforms with relatively homogeneous sizes and thus provide native-like environments for MP studies ( Bayburt and Sligar, 2010 ). In this review, we discuss how to harness DNA-origami nanostructures for constructing planar and curved lipid-bilayer membranes with well-defined shapes, sizes, and arrangements. In addition, we will discuss the applications of DNA-scaffolded lipid-bilayer membranes for studying the structure and function of large MPs."
} | 1,269 |
28296150 | PMC5658582 | pmc | 4,457 | {
"abstract": "Summary This study addresses the question of ecological interest for the determination of structure and diversity of microbial communities that degrade lignocellulosic biomasses to produce biofuels. Two microbial consortia with different history, native of wheat straw ( NWS ) and from a methanogenic digester ( MD ) fed with cow manure, were contrasted in terms of hydrogen performance, substrate disintegration and microbial diversity. NWS outperformed the hydrogen production rate of MD . Microscopic images revealed that NWS acted on the cuticle and epidermis, generating cellulose strands with high crystallinity, while MD degraded deeper layers, equally affecting all polysaccharides. The bacterial composition markedly differed according to the inocula origin. NWS almost solely comprised hydrogen producers of the phyla Firmicutes and Proteobacteria, with 38% members of Enterococcus . After hydrogen fermentation, NWS comprised 8% Syntrophococcus , an acetogen that cleaves aryl ethers of constituent groups on the aromatic components of lignin. Conversely, MD comprised thirteen phyla, primarily including Firmicutes with H 2 ‐producing members, and Bacteroidetes with non‐H 2 ‐producing members, which reduced the hydrogen performance. Overall, the results of this study provide clear evidence that the history of adaptation of NWS enhanced the hydrogen performance from untreated wheat straw. Further, native wheat straw communities have the potential to refine cellulose fibers and produce biofuels simultaneously.",
"introduction": "Introduction Currently, the bioenergy market is a necessity for society. First‐generation (1G) ethanol production has assisted in the transition towards a low‐carbon economy; however, feedstocks used to produce it compete directly for resources with food production. In contrast, it has been suggested that second‐generation (2G) biofuels could be sustainable, as the use of lignocellulosic residual biomasses derived from forestry, agriculture and agroindustry does not compete directly with food production. Technological processes for 2G biofuel production should fulfil the commitments of white biotechnology achieving more efficient biomass degradation, consuming less energy and resources, generating less waste and obtaining suitable profits. The current technologies for 2G biofuels do not entirely fulfill these conditions, as these processes discharge effluents with residual chemicals, typically use elevated processing temperatures and pressures (> 100°C and 1 atm) and are not commercially viable (Rabemanolontsoa and Saka, 2016 ). To identify white biotechnologies in the biofuel market, several authors have tested microbial communities or consortia from different sources to mediate biological transformations from lignocellulosic feedstocks into biofuels. For example, microbial communities in anaerobic digesters simultaneously perform various tasks to transform complex substrates, generating methane‐rich biogases (Werner et al ., 2011 ). Because microbial communities act in a single unit, the energy efficiency ratio of anaerobic digesters is higher than those bioprocesses with separate units dedicated to pretreatment, saccharification and fermentation (Börjesson and Mattiasson, 2007 ). Thus, the question remains whether a similar technology could be applied to produce another type of biofuel, such as bioalcohols or biohydrogen. Hydrogen is a versatile, carbon‐free fuel. Either burning it directly in internal combustion engines or providing electrons for fuel cells, hydrogen supplies a source of pollution‐free energy. Dark fermentation is by far the biological method for producing renewable hydrogen that has the best opportunities for scaling up (Sen et al ., 2016 ). New designs of 2G biorefineries now integrate H 2 ‐producing fermenters as part of a strategy to enhance the end‐use energy efficiency (Sanchez et al ., 2016 ). There are two routes to produce biohydrogen from lignocellulosic feedstocks: as mentioned earlier, configurations with separate units, or configurations using microbial consortia that integrate various operations (e.g. Valdez‐Vazquez et al ., 2015 ; Sanchez et al ., 2016 ). There have been several reports on the performance of microbial consortia for hydrogen production from lignocellulosic biomasses. Typical microbial communities tested for this purpose include sludge from anaerobic digesters, composts and ruminal fluids (Chu et al ., 2011 ; Pérez‐Rangel et al ., 2015 ; Reginatto and Antônio, 2015 ). In addition, recent studies have highlighted the suitability of using microbial communities to produce liquid biofuels (Ho et al ., 2011 ; Ronan et al ., 2013 ). However, the drawback of using microbial communities has been the low product yields and extended conversion times. There are two potential explanations for the poor performance of such microbial communities. Microbial ecologists suggest that historical factors are determinant in shaping the functioning of native microbial communities (Martiny et al ., 2006 ). For example, soil and sediment microbial communities perform better in their ‘native’ environment than when these communities are exposed to other new environmental conditions, suggesting the local adaptation of microbial communities to their original environments (Strickland et al ., 2009 ; Reed and Martiny, 2013 ). A recent study showed that the microbial community naturally present in lignocellulosic biomasses outperformed other communities, such as anaerobic sludge, ruminal fluids and the communities present in soil to produce hydrogen from untreated wheat straw (Pérez‐Rangel et al ., 2015 ). Members of the native microbial community were then isolated and used to integrate a synthetic microbial consortium for producing hydrogen from untreated wheat straw. The synthetic microbial consortium was unable to consume the entire fraction of the wheat straw xylan as the native microbial community did (Valdez‐Vazquez et al ., 2015 ). Because in natural lignocellulosic biomasses, lignin forms stable lignin–carbohydrate complexes (Kajikawa et al ., 2000 ), we assumed that still unknown members of the native wheat straw community are involved in the disintegration of the xylan–lignin complexes. The starting point of this study is the design of a cellulosic biorefinery with two sequential bioprocesses, both mediated by microbial consortia. In the first stage, a microbial consortium acts on the lignocellulosic substrate only long enough to consume non‐cellulosic polysaccharides. The partially fermented substrate is then intended for producing bioalcohols (Valdez‐Vazquez et al ., 2015 ). Keeping this in mind, in this study, we investigated two microbial communities of different origins, native of wheat straw (NWS) and from a methanogenic digester (MD), to characterize direct hydrogen production from wheat straw (WS), examine the microscopic arrangement to disintegrate lignocellulosic substrates and determine the composition of the communities using massive sequencing technologies. Overall, the three experimental approaches to characterize the microbial communities of interest enabled the elucidation of the relationship between the inocula origin (reflected as microbial diversity) and function, measured as the hydrogen production rate and substrate degradation of untreated wheat straw.",
"discussion": "Results and discussion Batch hydrogen production performance The hydrogen production performance of the NWS community was compared with that from a MD using untreated WS as the sole carbon source (Fig. 1 ). Both microbial communities positively responded to the increase in the substrate concentration; NWS had an increasing in 3.3 ml of H 2 with an increase of 1 g of total solids (TS) of substrate. In addition, NWS produced 1.15 times more hydrogen than the MD. In the range of substrate concentrations tested, neither of the two microbial communities exhibited substrate inhibition. Therefore, the substrate concentration at which hydrogen production reaches a maximum would be higher than 5% of TS for both populations. The profile of soluble products indicates that in NWS, already 70% of total soluble products comprised acetate (Fig. 1 B). In contrast, MD produced on average 25% of the total soluble products as propionate, a metabolite whose production route represents a hydrogen sink (Reichardt et al ., 2014 ). For both inocula, large quantities of acetate were recovered. The hydrogen/acetate ratio ranged from 0.5 to 1.5 for NWS and from 0.2 to 0.8 for MD. The hydrogen/acetate ratios suggest that acetogens were active in both microbial communities. Also, the metabolite profile in MD is a clear indicative of the existence of propionate producers competing for the substrate. A case study dealing with changes in hydrogen production with pH observed that the activity of propionate producers was stimulated at pH varying from 5.0 to 6.0. At this pH range, propionate was the major soluble metabolite positively related with the undermining of hydrogen production (Hwang et al ., 2004 ). Given the initial pH of 6.5 in the batch tests, the activity of propionate producers could be stimulated. Besides propionate, the metabolite profiles for NWS and MD showed differences in the production of butyrate. MD produced quantities of butyrate that could have resulted in more hydrogen than NWS. However, the opposite was observed: NWS produced more hydrogen than MD. There could be at least one explanation for this behaviour, if, for example, butyrate‐producing pathways occurred in MD without hydrogen formation. Catabolism of fructose, a sugar present in the water‐soluble fraction of WS, produces butyrate without hydrogen (Falony et al ., 2009 ; Tishler et al ., 2015 ). Also, acetate – the main metabolite in the batch tests – acts as precursors of butyrate (Pryde et al ., 2002 ). The results of microbial diversity revealed the identity of MD members whose metabolisms lead to the production of propionate and butyrate. Figure 1 Performance of two microbial communities, native of wheat straw ( NWS ) and of a methanogenic digester ( MD ) for hydrogen production from untreated wheat straw after 7 days of incubation. A. Hydrogen production and B. soluble products. Errors bars indicate standard deviation ( n = 4). NWS was highly competitive in short‐term fermentations, with a hydrogen yield 1.15 times higher than that achieved with MD (17 ml of H 2 per g of volatile solid (VS) for NWS versus 15 ml of H 2 /g‐VS for MD). After comparing the hydrogen performance under a mesophilic regime of different communities fermenting untreated wheat straw, the maximum yields reached by NWS and MD were at the same level than those previously reported ranging from 3 to 37 ml of H 2 /g‐VS (Chu et al ., 2011 ; Quéméneur et al ., 2012 ; Pérez‐Rangel et al ., 2015 ). However, in terms of hydrogen production rate (R m , determined from the modified Gompertz equation), NWS displayed a higher value of R \n m with 72 ml of H 2 /d than MD with a value of 55 ml of H 2 /d. NWS also outperformed the values of R \n m reported for the other microbial communities which varying from 8 ml of H 2 /d to 68 ml of H 2 /d (Chu et al ., 2011 ; Quéméneur et al ., 2012 ; Pérez‐Rangel et al ., 2015 ). In the present study, the increase in substrate concentration to nearly 5% TS resulted in an improvement in R \n m . Microscopic observations The total removal of volatile solids was similar for both microbial communities, showing on average 30% (± 3%) for NWS and 32% (± 9%) for MD. However, the extent degradation and functions of both hydrogen‐producing communities on the wheat straw differed. In Fig. 2 , scanning electron microscopy (SEM) micrographs show the epidermal zone and stomata of WS. Unfermented WS had a smooth surface, reflecting the waxy layer (cuticle) that covers the epidermal cells (Xu, 2009 ). The epidermis containing long cells and short cells with silica bodies and stomata were observed, consistent with Andrade et al . ( 2012 ). All anatomical structures observed were intact (Fig. 2.1 A,B). NWS superficially digested the WS, eliminating the cuticle and partially degrading the epidermis; thus, the tabular epidermal cells and silica bodies were exposed. Additionally, guard cells were partially degraded (Fig. 2.2 A,B). In contrast, MD degraded deeper layers, and the epidermis, where stomata are localized, was peeled off, exposing the next layer, the cortex, which contains collenchyma and parenchyma cells (Xu, 2009 ). In SEM images for MD, guard cells of deeply degraded stomata and holes in place of stomata were common in the analysed surfaces. Moreover, the anticlinal walls [walls separating the two adjacent cells in the same layer of the epidermal cells (Jääskeläinen et al ., 2013 )] disappeared in some regions of the epidermis, presumably causing the structure of long cells to collapse (Fig. 2.3 A,B). Figure 2 Scanning electron microscopy ( SEM ) photograph of wheat straw. A. Structure of the epidermis of wheat straw showing long cells, short cells with silica bodies and stomata. B. Structure of stomata from wheat straw. (1) Before hydrogen fermentation and after 7 days of hydrogen fermentation by (2) the native community of wheat straw ( NWS ) and (3) of a methanogenic digester ( MD ). Confocal laser scanning microscopy (CLSM) showed the site of polysaccharide degradation on these anatomical structures. Unfermented WS presented high green fluorescence with sinuous but well‐defined epidermal cells. The safranin staining of polysaccharides revealed irregular zones inside the epidermal cell anticlinal walls (arrowhead). These irregular zones play a relevant role in microbial degradation in NWS (Fig. 3.1 A,B). After hydrogen fermentation, NWS left sinuous polysaccharide strands primarily corresponding to cellulose fibrils (Pérez‐Rangel et al ., 2015 ). The activity of NWS was discontinuous, acting locally on irregular zones of the anticlinal walls and failing to penetrate deeper layers. Similar to the observation obtained from the SEM images, the stomatal guard cells were not extensively disturbed (Fig. 3.2 A,B). Figure 3 Confocal laser scanning microscopy ( CLSM ) images of polysaccharides in wheat straw stained with Safranine O (green fluorescence). A. Structure of the epidermis of wheat straw showing long and short cells. B. Structure of stomata and short cells with silica bodies. (1) Before hydrogen fermentation and after 7 days of hydrogen fermentation by (2) the native community of wheat straw ( NWS ) and (3) of a methanogenic digester ( MD ). Regarding MD, the pattern of waviness of the epidermal cells was fairly conserved, and the uniform decrease in polysaccharide fluorescence indicated that MD showed a wide range of action on the carbohydrate fraction (Fig. 3.3 A,B). The crystallinity index (CI) of cellulose after fermentation revealed that MD degraded equally amorphous and crystalline cellulose (CI of 50.1% for the unfermented WS versus 45.9% for the fermented WS). In contrast, NWS consumed amorphous cellulose leaving cellulose strands with high crystallinity (CI of 52.7% after fermentation, Fig. S1 in Supporting information). Bacterial diversity and composition Original microbes of the studied communities and those remaining after 7 days of hydrogen fermentation were characterized using massive DNA pyrosequencing analysis (Fig. 4 ). A total of 49508 reads were obtained. After quality filters, NWS presented 9706 and 8870 reads before and after fermentation, respectively, while MD showed 9987 and 16220 reads before and after fermentation respectively. The median read length was 455 bps for all samples. Good‐quality reads were classified as different operational taxonomic units (OTUs) at the 97% sequence similarity cut‐off. The rarefaction analysis of OTUs showed a sufficient sampling effort for both microbial communities (see Fig. S2 in Supporting information). For NWS, more than 99.9% OTUs could be taxonomically assigned to at phylum level (Fig. 4 ). For MD, the percentages of taxonomically assigned OTUs were 63.7% and 98.9% before and after fermentation respectively (Fig. 4 ). The total number of distinct bacterial OTUs observed for NWS and MD were 21 and 75 respectively. Figure 4 Proportion of major OTU s found in the H 2 ‐producing reactors from wheat straw. A. Native wheat straw community and B. community of a methanogenic digester. Members of Firmicutes and Proteobacteria evenly composed the original NWS (top panel in Fig. 4 ). The species richness observed for this microbial consortium was 19 with a Shannon index of 1.51. Notably, two species of Enterococcus dominated, composing 42% of the consortium. One member of the family Enterobacteriaceae represented 35% of the entire consortium. Clostridium , a typical hydrogen producer, represented only 7% of the original NWS including four different species (bottom panel in Fig. 4 ). Surprisingly, the population structure of NWS was nearly unaffected after hydrogen fermentation (species richness of 11 and Shannon index of 1.59). The two species of Enterococcus exhibited changes in abundance but together still accounted for almost 40% of total. The various species of Clostridium, Escherichia and Klebsiella also undergo minor changes. However, the most notable change was the increase in abundance of the genus Syntrophococcus from < 1% to 8%. Original MD comprised 13 phyla, where Firmicutes and Bacteroidetes dominated, with 23.7% and 19.0% respectively (top panel in Fig. 4 ). The species richness for the original MD was 74 with a Shannon index of 3.09. In the original MD, Clostridium and Bacteroides represented 21.5% of the total of bacterial species. After hydrogen fermentation, the diversity and composition of the MD showed visible alterations (bottom panel in Fig. 4 ). The species richness decreased to 63 with a Shannon index of 2.43. At the end of fermentation, Roseburia , Bacteroides and Escherichia accounted for 60.2% of the total. The bacterial composition greatly differed according to inocula origin. The most remarkable characteristic of the original NWS was the lack of aerobic members, which were expected based on previous reports of epiphytic bacteria, such as Methylobacterium , Sphingomonas and Pseudomonas (Vorholt, 2012 ). The only aerobic members observed in the original NWS that previously were reported as part of epiphytic communities were Pseudomonas and Pantoea (< 1%). Instead, the original NWS was enriched with facultative and strict anaerobes, likely reflecting the sun‐drying at which the wheat plants were subjected before harvest and the time in which the wheat straw was stored. Under indoor conditions (28°C and 55% relative humidity), all obligate aerobes died and only some genera, such as Enterococcus, survived on wheat straw. Previous reports have determined that Enterococcus faecalis survived well for extended periods under nutrient‐starvation conditions on solid substrates or water (Mackey and Hinton, 1990 ; Lebreton et al ., 2014 ). These fortuitous events selected for bacterial species adapting to the new environmental conditions inside the anaerobic bioreactors. After hydrogen fermentation, Enterococcus and members of the family Enterobacteriaceae remained without major changes in abundance. The genus Enterococcus comprises members typically found in human and animal gastrointestinal tracts, the guts of insects, such as termites, plants, soil and water, and fermented foods and dairy products (Lebreton et al ., 2014 ). Importantly, Enterococcus has been reported in a few hydrogen‐producing consortia (Liu et al ., 2009 ; Pendyala et al ., 2013 ). Recently, Valdez‐Vazquez et al . ( 2015 ) isolated and tested various strains of Enterococcus from the NWS. Such enterococcal strains efficiently convert soluble xylan. However, when cultivated under a natural polysaccharide matrix, these strains were incapable of completely degrading the xylan fraction consuming merely 30%. In natural lignocellulosic biomasses, xylan is linked to lignin via ether groups, forming xylan–lignin complexes (Kajikawa et al ., 2000 ; Lawoko et al ., 2006 ). The incapacity of Enterococcus to consume the entire xylan fraction could reflect the absence of some lignin‐releasing members from the NWS. After hydrogen fermentation, the abundance of obligate anaerobic, acetogen Syntrophococcus , increased. The genus Syntrophococcus, belonging to the family Lachnospiraceae, was originally isolated from rumen (Krumholz and Bryant, 1986 ). Syntrophococcus recognizes and cleaves the methyl groups within the polymeric structure of lignin as a one‐carbon source to release acetate and the corresponding hydroxyl derivatives. In anaerobic environments, this O ‐demethylating acetogen plays a relevant role in the mineralization of ligno‐aromatic compounds in conjunction with other anaerobes that metabolize the aromatic ring structure (Doré and Bryant, 1990 ; Frazer, 1994 ; Bernard‐Vailhé et al ., 1995 ). Because in native lignocellulosic substrates, xylan exists in the form of xylan–lignin complexes, Syntrophococcus twinned with fermentative bacteria could act as a catalyst for the degradation of the xylan fraction of native substrates, particularly in zones with highly lignified cells, such as the epidermal layer. In contrast, the cellulose fraction is not affected by the presence of these phenolic‐degrading acetogens. These observations support the findings of the present study, as Syntrophococcus, in conjunction with the remaining members of the NWS, primarily reported as H 2 producers ( Enterococcus , Enterobacter , Clostridium , Klebsiella , Escherichia and Citrobacter ), hydrolysed the non‐cellulosic fraction of the WS to produce hydrogen, leaving unconsumed crystalline cellulose. The original MD was distinguished based on the high level of diversity consistent with previous reports, where the phyla Firmicutes and Bacteroidetes were predominant (Klang et al ., 2015 ). After hydrogen fermentation, five bacterial genera predominated: Bacteroides, Roseburia, Escherichia, Enterococcus and Clostridium . The genera Escherichia, Enterococcus and Clostridium represented the hydrogen‐producing population in the MD. Regarding Roseburia , some isolates express xylanase and endoglucanase activities producing H 2 , CO 2 as well as formate, butyrate, succinate and lactate (Chassard et al ., 2007 , 2010 ). Similar to Syntrophococcus , Roseburia belongs to the family Lachnospiraceae, and these bacteria have been implicated in the disintegration of complex substrates. Roseburia along with other anaerobes belonging to the Clostridial clusters IV and XIVa are recognized as the main producers of butyrate by the microbial communities present in the human colon and rumen (Pryde et al ., 2002 ). The catabolism of carbohydrates by Roseburia produces hydrogen. However, as outlined above, it seems that butyrate‐producing species were active without hydrogen formation. Roseburia inulinivorans perform the oligofructose degradation producing butyrate and CO 2 , but not H 2 (Falony et al ., 2009 ). Roseburia intestinalis converts acetate into butyrate with oligofructose as the sole energy source without hydrogen formation (Pryde et al ., 2002 ). In this way, it appears that Roseburia representing almost 20% of the MD community could be responsible for butyrate formation, but with little or no contribution to the formation of hydrogen. Equivalent to one‐fifth of the abundance, the genus Bacteroides , belonging to the family Bacteroidaceae, was one of the most important bacteria in the MD. Some Bacteroides have been previously identified as relevant members of the fibrolytic microbial community in the human colon, degrading a wide range of polysaccharides, such as cellulose, xylan, starch and pectin generating acetate, propionate and succinate (Robert et al ., 2007 ). In methane‐producing populations, Bacteroides plays a dual role, degrading complex polysaccharides and contributing to volatile organic acids, which are subsequently converted into methane, poorly contributing to hydrogen formation (Chassard et al ., 2010 ). Bacteroides is primarily identified as a non‐H 2 ‐producing cellulose‐degrading species. Therefore, in MD, Bacteroides could be responsible for decreasing the hydrogen yield when substrates without hydrogen formation were consumed to produce propionate. As mentioned earlier, an initial pH of 6.5 along with protein from the medium stimulated the activity of Bacteroides during the first days of incubation decreasing the potential of hydrogen production (Hwang et al ., 2004 ; Walker et al ., 2005 ). At this point, it is important to stress that care is needed when comparing results of culture‐independent studies, such as the presented here, with the metabolic capabilities of isolated strains from the same genera or families. Some hypothesis from these comparisons can be derived and further investigated, like the role of specific species, genera or families in the overall functionality of bioreactors. To this end, it is in fact needed to continue strain isolation efforts to study directly strain metabolic capabilities, and also to pursue transcriptomic and proteomic efforts in bioreactors that will allow a better understanding of functionality. Biotechnological and ecological considerations The findings of the present study have biotechnological and ecological implications. From a technological point of view, the members that integrate the NWS community are interesting to characterize because they have developed specialized enzymatic machinery to hydrolyse the xylan fraction with minor alterations in the cellulose fraction (refer to the microscopic observations). Thus, the partially refined cellulose can then be intended for producing other biofuels under a biorefinery approach. Two members are of special interest: Enterococcus and Syntrophococcus, both of which were highly abundant in NWS communities. Enterococcus is an H 2 ‐producing, facultative anaerobe that survives during prolonged periods of starvation and produces bacteriocins (Leroy et al ., 2003 ). These characteristics make Enterococcus highly desirable for large robust facilities for 2G biofuel production. For H 2 ‐producing consortia enriched with Enterococcus , the anaerobic conditions for the regulation of growth could be relaxed, and the produced bacteriocins could limit the growth of undesired bacteria competing for the substrate. On the other hand, Syntrophococcus belongs to a family of acetogenic bacteria that catalyses O ‐demethylation of constituent groups on the aromatic components of lignin. Other O ‐demethylating bacteria also include Acetobacterium woodi , Clostridium pfennigii , Eubacterium callandri (Frazer, 1994 ). The specific cleavage of β ‐aryl ethers bonds in lignin, accounting for approximately 50% of all the linkages in lignin, has been implicated in recovering valuable aromatic groups of lignin while refining cellulose (Reiter et al ., 2013 ; Strassberger et al ., 2013 ). To some extent, Syntrophococcus in conjunction with selected members of the NWS consortium could serve as an efficient biological pretreatment to refine cellulose fibers and to produce bioenergy in the form of hydrogen. Instead of dedicated units for high energy‐demanding pretreatments, saccharification and fermentation, 2G biofuel production facilities could install anaerobic fermenters with NWS microbial communities that, as shown in the present study, can degrade and efficiently ferment WS. The added environmental benefits are the elimination of acid or alkali compounds, elimination of pretreatment units obviates the formation of fermentation inhibitors, specific cleavage of ethers in lignin would preserve aromatic compounds for further industrial applications, and the refinement of cellulose strands for liquid biofuel production (ethanol/butanol). The implementation of such dual bioprocesses for biological pretreatment and hydrogen production would generate higher energy efficiency ratios and product yields than bioprocesses with separate units. In any case, additional studies using native strains of lignocellulosic biomasses are necessary to explore the roles of these bacteria in improving the rates of substrate consumption and product formation with other lignocellulosic substrates. The results of the present study indicate that the inocula origin reflected differences in the microbial community composition, thereby affecting the hydrogen performance from untreated wheat straw. The primary population structure of NWS remained with minor alterations after hydrogen fermentations; only changes in abundances of some members were detected. In contrast, several members of MD become extinct after hydrogen fermentations. These results are a clear indication that NWS was already adapted to grow on WS. The time of storage for 6 months acted such as an adaptation process in which the microflora naturally present on the surface of WS could adapt to the lignocellulosic substrate before the hydrogen tests. In the course of these 6 months, major changes in the population structure of NWS could occur with the extinction of aerobic members and those with little resistance over long periods of storage. The long‐time storage resulted in a stabilized, adapted NWS consortium. These results resemble those microbial communities present in the stabilization period of fermentations or digestion processes. Microbial community dynamics of different bioprocesses involving lignocellulosic substrates indicate that microbial communities can be grouped into three categories based on pattern similarities: at the start‐up, during the growing period and at the stabilization period (Li et al ., 2015 ; Yan et al ., 2015 ). Also, the population structure remains stable over time if the operating conditions such as temperature and substrate are kept (Sun et al ., 2015 ). By comparison with previous of microbial communities acting on lignocellulosic substrates, NWS can be considered such as a microbial community coming from the stabilization period of an adaptation process to WS, whereas MD consisted of a microbial community experiencing an adaptation process to the new substrate. Thus, history of adaptation of each microbial community determined the short‐term shifts in population structure. Similar observations have previously been reported for other systems, such as soil and sediments, in terms of biogeochemical transformations and the activity of microbial communities (Strickland et al ., 2009 ; Reed and Martiny, 2013 ). Studies of microbial ecology stated that historical factors determine the functioning of microbial communities, as these microbes behave better under their original environments (Strickland et al ., 2009 ). In the present study, several unmeasured environmental conditions differed from the environment of which the microbial communities were derived; however, we assume that the substrate could be the main factor driving the behaviour of both consortia. The above‐mentioned result highlights the importance of the ecological context and perspective, while selecting for specific microbially mediated enzymatic processes, such as lignocellulosic fermentation. Thus, the ecological perspective of these results raises questions about what microbial composition is better to disintegrate and ferment different lignocellulosic biomasses into target biofuels, where such microbial communities prosper, and how the bioreactor conditions affect the functioning of such communities. The microbial composition of NWS almost entirely included hydrogen producers and lignin‐degrading members that presumably acted together to increase xylan fermentation into hydrogen and soluble end‐products. In contrast, MD comprised non‐H 2 ‐producing members that degraded substrates and therefore reduce the hydrogen yield. Based on these results, functionally robust H 2 ‐producing communities must integrate members of Firmicutes and Proteobacteria, avoiding the sampling of natural communities in which Bacteroidetes is present. In summary, this study presents the shifts in population structure during hydrogen production from a lignocellulosic substrate of two microbial communities with variations in life history traits. Differences between the NWS and MD communities were notable, before and after the hydrogen fermentations, where the former exhibited changes only in the proportion of their members, whereas the later had members that were extinct after fermentations. The history of the NWS community may have been decisive in shaping a better functioning than the MD community. Finally, new members in the NWS community were identified with potential for refining cellulose and producing hydrogen."
} | 8,264 |
36364309 | PMC9655766 | pmc | 4,458 | {
"abstract": "Microbial fuel cells (MFCs) are an environmentally friendly technology and a source of renewable energy. It is used to generate electrical energy from organic waste using bacteria, which is an effective technology in wastewater treatment. The anode and the cathode electrodes and proton exchange membranes (PEM) are important components affecting the performance and operation of MFC. Conventional materials used in the manufacture of electrodes and membranes are insufficient to improve the efficiency of MFC. The use of nanomaterials in the manufacture of the anode had a prominent effect in improving the performance in terms of increasing the surface area, increasing the transfer of electrons from the anode to the cathode, biocompatibility, and biofilm formation and improving the oxidation reactions of organic waste using bacteria. The use of nanomaterials in the manufacture of the cathode also showed the improvement of cathode reactions or oxygen reduction reactions (ORR). The PEM has a prominent role in separating the anode and the cathode in the MFC, transferring protons from the anode chamber to the cathode chamber while preventing the transfer of oxygen. Nanomaterials have been used in the manufacture of membrane components, which led to improving the chemical and physical properties of the membranes and increasing the transfer rates of protons, thus improving the performance and efficiency of MFC in generating electrical energy and improving wastewater treatment.",
"conclusion": "5. Conclusions and Future Directions This review deals with the applicability of nanomaterials in improving the performance and efficiency of MFC in the manufacture of anode and cathode electrodes and proton exchange membranes. CNTs, CNF, graphene, GAC, CNS, and metal oxides such as iron oxide, MnO 2 , gold, silver, cobalt, tin oxide, and spinel oxide are among the most famous nanomaterials that have been used to improve the efficiency of the anode and cathode electrodes by increasing the surface area of the electrodes, increasing the rates of electron transfer, and increasing biocompatibility with microorganisms, biofilm formation, oxidation reactions, and ORR. The addition of nanomaterials to the components of proton exchange membranes, such as Fe 2 O 3 , Fe 2 O 4 , SiO 2 , MnO 2 , GO, TiO 2 , CNF, and ACNF, led to an improvement in the chemical and physical properties and thermal stability of the membranes, thus increasing the transfer rates of protons and preventing the transfer of oxygen from the cathode chamber to the anode chamber compared to the performance of commercial membranes. The use of nanomaterials had a substantial effect in improving the performance and efficiency of MFC in the production of electrical energy and wastewater treatment. From the above, it becomes clear the importance of using nanomaterials to improve the efficiency of electrodes and membranes used in the operation of MFCs. Therefore, we, the authors of this review, recommend the use of nanomaterials to improve the operating efficiency of microbial fuel cells in the future.",
"introduction": "1. Introduction Microbial fuel cell (MFC) is an innovative and eco-friendly technology. It is classified as one of the ways to produce renewable energy using organic waste, as it is considered environmentally friendly. In MFCs, a chemical is converted energy to electrical energy using microbes [ 1 ]. The MFC consists of either one or two chambers separated by a membrane. In the case of two chambers of MFC, the anode chamber contains the positive electrode, the cathode chamber contains the negative electrode, and the proton exchange membrane (PEM) is used to separate the two chambers and allows the proton transfer from the anode to the cathode. Organic or chemical waste is placed in the anode chamber. Bacteria play an important and effective role in the analysis of organic compounds present in the anode chamber and produce electrons and protons. Electrons transfer from the anode to the cathode through an external circuit, while the protons transfer through the PEM to the cathode [ 2 ]. There are many factors that affect the performance and efficiency of MFC, such as the organic material used, bacteria (biocatalyst), internal resistance, properties of the electrode material, the membrane, and the concentration of ions [ 3 , 4 , 5 ]. The electrode materials (anode and cathode) are important in the performance and efficiency of MFC in terms of the type of electrode material, electrical conductivity, biocompatibility, surface area, and non-toxicity to bacteria, which affects the efficiency of the oxidation and reduction processes of organic matter, electron transfer, and electric power generation. In addition, the PEM is an important component in the efficient operation of the MFC in terms of the type of membrane material and the size of the membrane pores, which affects the process of transferring protons and reducing the transfer of oxygen to the cathode [ 5 ]. In MFC systems, anode performance is a critical component that can influence the level of power output that can be reached. In general, the surface qualities of the anode material used have a significant impact on the anode response. As a result, the development of new materials and designs for anode electrodes is crucial to increase the efficiency of this technology for practical use. One goal of employing nanomaterials for anode manufacturing is to improve electron transfer mechanisms between microorganisms that function as biocatalysts in the anode chamber and the material that forms the anode itself to improve power output. Nanostructured materials can be used to modify the surface of an anode electrode composed of different sorts of materials or as the base anode material [ 6 ]. At the cathode, the final electron acceptor (e.g., oxygen) combines with the protons to generate the final product (e.g., H 2 O). The cathode materials directly dominate the kinetics of the oxygen reduction reactions (ORR) in the cathode chamber. ORR is primarily determined by the surface area, electrical conductivity, chemical stability, and activity of the cathode materials, among other factors. As a result, the cathode materials used and their alterations are critical for the optimization and improvement of MFC performance [ 7 ]. The anode and the cathode reactions of the MFC must be separated using the membrane that is fixed between the anode and the cathode chambers. The membrane acts as a physical conductor or separator, which facilitates the transfer of protons from the positive electrode to the negative electrode [ 8 ]. The membrane used in MFC must be characterized by low internal resistance, high proton conductivity, high energy recovery, and good chemical and physical stability, which helps prevent the transfer of oxygen from the cathode chamber to the anode chamber [ 9 ]. The current trend is to improve the performance of membranes and their physical and thermal properties by incorporating nanomaterials into membrane structures, which helps in achieving the desired properties and improving the performance of MFC [ 10 ]. Figure 1 exhibits schematic for the microbial fuel cell (MFC) [ 11 ]. The present review focuses on previous studies of nanotechnology applications and their impact on the performance and efficiency of MFC in energy production and wastewater treatment. There are three factors or basic parts in the installation and operation of the MFC. The first factor is the anode or the positive electrode, and the focus is on the use of nanomaterials that improve the performance and efficiency of the anode, such as increasing the surface area, increasing electron transfer, and the generation of electrical energy. The second factor is the cathode or the negative electrode, and the focus is on the use of nanomaterials that improve cathode reactions or ORR. The third factor is the PEM proton membrane and the use of materials and nanoparticles in the manufacture and installation of proton exchange membranes and their effect on proton transfer and preventing the transfer of oxygen from the cathode to the anode. The effect of anode, cathode, and membrane on improving the performance and efficiency of MFC in energy production and wastewater treatment is the focus."
} | 2,067 |
20423483 | PMC2873267 | pmc | 4,459 | {
"abstract": "Background Microbial consortia are a major form of life; however their stability conditions are poorly understood and are often explained in terms of species-specific defence mechanisms (secretion of extracellular matrix, antimicrobial compounds, siderophores, etc.). Here we propose a hypothesis that the primarily local nature of intercellular signalling can be a general mechanism underlying the stability of many forms of microbial communities. Presentation of the hypothesis We propose that a large microbial community can be pictured as a theatre of spontaneously emerging, partially overlapping, locally recruited microcommunities whose members interact primarily among themselves, via secreted (signalling) molecules or cell-cell contacts. We hypothesize that stability in an open environment relies on a predominantly local steady state of intercellular communication which ensures that i) deleterious mutants or strains can be excluded by a localized collapse, while ii) microcommunities harbouring useful traits can persist and/or spread even in the absence of specific protection mechanisms. Testing the hypothesis Some elements of this model can be tested experimentally by analyzing the behaviour of synthetic consortia composed of strains having well-defined communication systems and devoid of specific defence mechanisms. Supporting evidence can be obtained by in silico simulations. Implications of the hypothesis The hypothesis provides a framework for a systematic comparison of bacterial community behavior in open and closed environments. The model predicts that local signalling may enable multispecies communities to colonize open, structured environments. On the other hand, a confined niche or a host may be more likely to be colonized by a bacterial mono-species community, and local communication here provides a control against spontaneously arising cheaters, provided that survival depends on cooperation. Reviewers This article was reviewed by G. Jékely, L. Aravind and E. Szathmáry (nominated by F. Eisenhaber)"
} | 511 |
18218112 | PMC2259305 | pmc | 4,460 | {
"abstract": "Background Despite the prevalence of horizontal gene transfer (HGT) in bacteria, to this date there were few studies on HGT in the context of gene expression, operons and protein-protein interactions. Using the recently available data set on the E. coli protein-protein interaction network, we sought to explore the impact of HGT on genome structure and protein networks. Results We classified the E. coli genes into three categories based on their evolutionary conservation: a set of 2158 Core genes that are shared by all E. coli strains, a set of 1044 Non-core genes that are strain-specific, and a set of 1053 genes that were putatively acquired by horizontal transfer. We observed a clear correlation between gene expressivity (measured by Codon Adaptation Index), evolutionary rates, and node connectivity between these categories of genes. Specifically, we found the Core genes are the most highly expressed and the most slowly evolving, while the HGT genes are expressed at the lowest level and evolve at the highest rate. Core genes are the most likely and HGT genes are the least likely to be member of the operons. In addition, we found the Core genes on average are more highly connected than Non-core and HGT genes in the protein interaction network, however the HGT genes displayed a significantly higher mean node degree than the Core and Non-core genes in the defence COG functional category. Interestingly, HGT genes are more likely to be connected to Core genes than expected by chance, which suggest a model of differential attachment in the expansion of cellular networks. Conclusion Results from our analysis shed light on the mode and mechanism of the integration of horizontally transferred genes into operons and protein interaction networks.",
"conclusion": "Conclusion To our knowledge, our analysis represent the first time that the HGT events are investigated in the context of protein-protein interaction and cellular networks. This is important because horizontal gene transfer in known to be prevalent in bacterial genome evolution in shaping the genome content, and they had an impact on the stability and evolution of the protein interactions and network. From our analyses, the distinguishing characteristics which sets the HGT gene category apart from the Non-core and Core gene categories are (i) higher evolutionary substitution rates ( Ka/Ks ), (ii) protein interaction network statistical properties such as protein degree connectivity, average clustering coefficients and betweeness centrality , (iii) preferential attachment with regards to the number of interactions formed by HGT genes, which indicate that HGT proteins preferentially neither self-associate nor do HGT proteins associate with Non-core proteins within the E. coli protein interaction network. Results from our study revealed a clear relationship between gene expressivity, evolutionary rate and protein connectivity for the three evolutionary classes of genes (Figure 8 ). The conserved Core set of genes generally display higher gene expressivity and protein connectivity than strain-specific Non-core and HGT genes. However, both gene expressivity and protein connectivity are inversely related to evolutionary rates, with the most highly conserved genes evolving the slowest. In contrast, horizontally transferred genes evolve at considerably higher evolutionary rates, and have lower gene expressivity and protein connectivity. In addition, proteins encoded by horizontally transferred genes attach preferentially to Core proteins within the E. coli protein interaction network. Consistent with this finding is the general idea that Core genes are the oldest resident genes and form the backbone of the protein interaction network to which new proteins are attached. These results may also suggest that a proportion of the lowest connectivity proteins in bacterial protein interaction networks are those genes which are more likely to have recently been transferred and incorporated into the E. coli genome. Figure 8 Summary of the relationship between protein connectivity, gene expressivity (CAI) and evolutionary rates ( dN/dS ) in E. coli . This is reminiscent of the so-called \"Complexity Hypothesis\", which was proposed to explain why the successful horizontal transfer of a gene is less probable if the connectivity of the protein it encodes is large [ 52 ], and its later modification called the 'Extended Complexity hypothesis' [ 53 ] which aims to explain why adaptive evolution is the least likely for proteins with high complexity. Although the Complexity Hypothesis and its modified version aim to describe which types of genes are more or less likely to be subjected to horizontal gene transfer, it fails to provide a mode and mechanism for subsequent integration of the horizontally transferred gene into it new recipient genome. The results from our analysis support these hypotheses with genomics and evolutionary data. Considering the prevalence of HGT in bacteria, the relative contribution of HGT as an additional mechanism to gene duplication may become more important on network evolution. Thus, with the availability of proteomics data for more bacteria, we will most likely gain more insight on the impact of HGT on the evolution of networks.",
"discussion": "Results and Discussion HGT genes evolve faster and have lower expression levels To investigate the selective pressure acting on organizational units, we classified E. coli genes according to their evolutionary conservation into three categories, namely, (i). Core Set : a conserved core set of genes that exist in all E. coli strains. (ii). Non-core Set : genes that are found in at least one strain but not in all strains, and (iii). HGT Set : genes that are derived from putative recent horizontal gene transfer events after the divergence of E. coli and Salmonella . By delineating genes according to their evolutionary conservation, we can more clearly identify the evolutionary forces to which the various evolutionary classes of genes are subjected. Direct measurements of E. coli gene expression were obtained from microarray gene expression experiments (see Methods ). In addition, we have also used the codon adaptation index (CAI) as a proxy for gene expression data, which we referred as \"gene expressivity\" [ 43 ]. Figure 1 shows that the Core genes have higher CAI gene expressivity levels (Figure 1A ) as well as log2 expression values (Figure 1B ) than Non-core and HGT genes (t-test and Wilcox rank test, p-value < 0.001). This can be explained by the different evolutionary histories of these three groups of genes. The Core set of genes, being the oldest resident genes in the genome have thus had sufficient time to adapt and optimise their codon usage patterns, explaining the higher levels of gene expressivity; whereas the recent horizontally transferred genes may need an adaptation period during which their base composition and codon usage patterns may need to be optimised to their new resident genome. Figure 1 Box plot of (A) gene expressivity (CAI) values and (B) log2 gene expression values between Core , Non-core and HGT genes. Core genes display higher expressivity than Non-core and HGT genes (P-value < 0.001). Figure 2 shows that amongst the three categories of gene sets, the Core set of genes evolve at the lowest substitution rates ( dN/dS ) and HGT genes evolve at the fastest rates, using E. coli K12 as reference for comparison (Wilcoxon-Mann-Whitney test, p-value < 0.001). The high evolutionary rates observed for HGT genes may be explained by either one of the following two hypotheses: (i) result of reduced negative selection pressure, which enable the invading genes to be purged from the genome, or (ii) result of increased positive selection whereby HGT genes contribute to the phenotypic character of E. coli strains [ 14 ]. Accordingly, it is thought that the strain-specific Non-core genes and HGT genes may contribute to the pathogenic character separating the enterohemorrhagic and uropathogenic from the benign E. coli K12 strain, therefore these genes are under positive selection pressure. Figure 2 Distribution of evolutionary rates ( dN/dS ) for various E. coli strains overlaid on a phylogenetic tree using E. coli K12 as reference for genome comparisons. Core genes evolve slower than Non-core and HGT genes (P-value < 0.001). Genes in Operons and Networks Display Higher Gene Expressivity There is increasing evidence to suggest that the chromosomal gene order in organisms is not always random [ 44 ]. It is known that proteins of linked genes evolve at comparable rates, and that natural selection may promote the conservation of linkage of co-expressed genes [ 45 ]. Accordingly, genes in the same operon occur in close physical proximity and are often known to be co-transcribed as units. In addition, genes encoding subunits of protein complexes also need to be expressed at similar times. To investigate the relative contributions of the various evolutionary gene categories on organizational structures, we surveyed both operons and the protein interaction network for their content of Core , Non-core and HGT genes. The Core set form a predominant portion of operons with 47% (2129 out of 4506 genes catalogued in RegulonDB version 5.7) of the operons consisting of Core genes, whereas 21% (948 out of 4506) of Non-core genes and 23% (1020 out of 4506) HGT genes, respectively, accounted for the remaining gene constituents of operons (Figure 3A ). Similarly, proteins encoded by Core genes account for a 67.5% (852 out of 1262) of the protein interaction network as reported by Butland et al [ 46 ] whereas Non-core genes and HGT genes account for 14.1% (178 out of 1262) and 18.4% (232 out of 1262) respectively (Figure 3B ). Figure 3 Number of E. coli genes in the genome organizations: (A) operons, (B) protein interaction network (PIN). Genes are classified into three evolutionary categories Core , Non - core and HGT genes. Core genes predominantly occur in both operons and protein interaction networks (P-value < 0.001). The tendency of operons to be enriched in Core genes can be explained by a need to simplify regulation, since genes residing in operons known to be under control of the same promoter (Chi-squared test, p-value < 0.001). This may facilitate horizontal gene transfer by enabling genes to be inherited as a physical and functional cohesive group rather than separate individual genes. In regard to the protein interaction network, it is thought that the Core genes form the ancestral backbones of the protein interaction network to which new functionalities are added via protein nodes and thus strengthens a model by which pathways expand [ 47 ]. To understand the impact of higher order organization of genes (i.e. operons) and proteins (i.e. interaction complexes) on properties such as expression or evolution, we investigated the gene expressivity characteristics and evolutionary substitution rates of the various categories of gene sets. We found that Core genes in organizational clusters (both operons and protein interaction network or PIN) have higher gene expressivity (CAI) values (Figure 4A and 4B ) and as well as log2 expression values (Figure 4C and 4D ) relative to Non-core and HGT genes (t-test and Wilcox-test for both operons and PIN, p-value < 0.001). For the PIN, this trend was robust against removal of ribosomal proteins. Figure 4 Gene expressivity (CAI) values and log2 gene expression values between Core , Non-core and HGT genes in different genome organizations. (A) Box plot of CAI values between Core , Non-core and HGT genes in operons; (B) Box plot of gene CAI values between Core , Non-core and HGT genes in protein interaction network (PIN); (C) Box plot of log2 gene expression values between Core , Non-core and HGT genes in operons; (D) Box plot of log2 gene expression values between Core , Non-core and HGT genes in protein interaction network (PIN). The overall trend from surveying operons and the protein interaction network indicates that Core genes tend to be found more often in organizational units such as operons and networks. The evolutionary composition may be the reason that highly clustered proteins in the protein interaction network display apparently high gene expressivity and low substitution rates. Distribution of COG Functional Categories between Core, Non-core and HGT genes within the Operons and Protein Interaction Network We have analyzed and compared the distribution of the Cluster of Orthologous (COG) functional categories of the Core , Non-core and HGT genes within the E. coli K12 genome, protein interaction network and operons (Figures 5 , 6 and 7 ). The various gene categories differ significantly in their COG distribution in the genome, the protein interaction network and operons (Scheirer-Ray-Hare test, p-value < 0.001, see Additional File 5 ). Figure 5 Distribution among the COG categories for all the E. coli genes. Counts were estimated for each evolutionary gene category, and expressed as percentages per total number of genes per COG category. The Core , Non-core and HGT gene sets differ in their distribution of COG functional categories (P-value < 0.001). Figure 6 Distribution among the COG categories for those E. coli genes that are members of operons. Counts were estimated for each evolutionary gene category, and expressed as percentages per total number of genes per COG category. The Core , Non-core and HGT gene sets contained within operons differ in their distribution of COG functional categories (P-value < 0.001). Figure 7 Distribution among the COG categories for those E. coli genes that are members of proten-protein interaction network. Counts were estimated for each evolutionary gene category, and expressed as percentages per total number of genes per COG category. The Core , Non-core and HGT gene sets contained within the protein interaction network differ in their distribution of COG functional categories (P-value < 0.001). In the overall gene comparison of the E. coli K12 Core , Non-core and HGT gene sets, the Core genes constituted the major evolutionary gene set present in all COG categories (Figure 5 ). The Non-core gene set in comparison to the HGT gene set was markedly abundant in the two COG categories: O ( Posttranslational modification, protein turnover, chaperones ) and T ( Signal transduction mechanisms ). The HGT gene set was more abundant than the Non-core gene set in the COG categories C ( Energy production and conversion ), F ( Nucleotide transport and metabolism ), G ( Carbohydrate transport and metabolism ), I ( Lipid metabolism ), K ( Transcription ) and V ( Defense mechanisms ). For the operons, the Core genes occur predominately in all COG functional categories, whereas the Non-core genes are over-represented in COG categories S ( Function unknown ) and U ( Intracellular trafficking, Secretion, and vesicular transport ) and the HGT genes are over-represented in comparison to the Non-core genes in COG functional categories C , E ( Amino acid transport and metabolism ), G , H ( Coenzyme metabolism ), R ( General function prediction only ) and V ( Defence mechanisms ) (Figure 6 ). For the protein interaction network, the HGT genes are over-represented in COG functional categories most notably C , G , H , and V (Figure 7 ). A most notable example in this regard is the COG category V in which the HGT gene set within the E. coli protein interaction network has a significantly higher mean node degree than the Core and Non-core genes sets. The overall statistical difference in distribution of COG functional categories between the Core , Non-core and HGT gene sets therefore seems to argue against the notion of a Core -versus- Non-core or Core -versus-acquired gene category consisting of Non-core and HGT genes, but rather strengthens the notion of a distinct separate category for Non-core genes. Network topology of the E. coli genes To investigate the mode and mechanism of integration of horizontally transferred genes into the E. coli protein-protein interaction network, we systemically investigated the network characteristics of proteins encoded by the various evolutionary categories of genes (Table 1 ). We found that proteins corresponding to the Core gene set represent the most highly connected protein nodes, which have an average connectivity of 11.0 interactors (Chi-squared test, p-value < 0.05). In contrast, Non-core proteins and proteins encoded by HGT genes have on average lower connectivities of 4.0 and 3.0 interactors respectively. This is consistent with our hypothesis that Core genes being the most highly conserved genes have resided in the genome for much longer, and thus had more opportunities to evolve interactions. The result of the network analysis is consistent with this theory. Table 1 Protein interaction network characteristics of E. coli Core, Non-core and HGT genes Characteristic Core Non-core HGT Total nodes (1276) 852 (66.8%) 178 (14.0%) 232 (18.2%) Ave. Node degree 10.9 4.1 3.4 Ave. clustering coefficient 0.100 0.072 0.039 Ave. betweenness centrality 2.59e-3 6.3e-4 5.5e-4 We also analyzed two additional network properties: betweenness centrality and clustering coefficient (Table 1 ). Betweenness centrality characterizes how essential a node is in maintaining communication between each pair of nodes in a network [ 48 ]. Depending on its position within the network, removal of a node can have very different effects on the connectivity, topology and flux of the network. Some nodes can be removed without any harmful effect, while others separate a connected network into disconnected sub-graphs. Betweeness centrality is a measure devised to describe the fraction of shortest paths going through a given node, with high values indicating that a node can reach many other nodes. Removal of nodes with high centrality will make it difficult to reach from one node to another, thus lengthen the path between nodes. The clustering coefficient describes the local transitivity in a network, with two nodes having a common neighbour in a network being more likely to be neighbours [ 49 ]. Table 1 shows that the HGT genes have lower betweenness centrality than the Core and Non-core genes, which suggests that they are less important in cellular communications. Interestingly the Non-core genes have higher betweeness centrality than the Core genes, the implication of which need to be further explored. On the other hand, Core genes have the highest clustering coefficients , with any two Core genes having a common neighbour being more likely to be neighbours of each other. The results in Table 1 indicate the HGT genes are the least important in maintaining the overall connectivity of the protein interaction network, in other words they are more likely to be peripheral nodes . Our analysis of the distribution of COG functional categories of the Core , Non-core and HGT nodes within the E. coli protein interaction network reveal that the Core genes are the most abundant and cover all the major COG functional categories in comparison to the Non-core and HGT gene sets (Figure 7 ). Although, the Non-core and HGT genes show similar COG distribution profiles within the protein interaction network, differences exist in COG categories C , G, H and V in which the HGT genes are relatively more abundant than Non-core genes. A most notable result in this regard is the COG defense category ( V ) in which the HGT gene set within the E. coli protein interaction network has a significantly higher mean node degree than the Core and Non-core genes set. Preferential Attachment of HGT proteins to Core proteins We further investigated the evolutionary profiles of the interaction partners in the network. Table 2 shows that about 74% of all the interactions are between a pair of Core genes, 11.2 % of the interactions are between a Core gene and a Non-core gene. In other words, in total about 85% of the interactions involve at least one Core gene. Among the interactions involving HGT genes, a large percentage (89%) was between a HGT genes and a Core gene, while interactions between Non-core and HGT genes only account for 1%. This is surprising since the ratio between Core genes and Non-core genes is only ~5:1, much smaller than the 9:1 ratio (89%: 10%) that we observed in the network. This discrepancy in ratio implies that an HGT gene have a higher propensity to establish interaction with a Core gene than with a Non-core gene. Indeed, the proportions of HGT-Core interactions are higher than expected by chance (Chi-squared test, p-value < 0.001). Table 2 Classification of interactions based on the evolutionary profiles of interaction partners. Category of interacting partners Number of Interactions Core to Core 3981 (74.0%) Non-core to Non-core 35 (0.6%) HGT to HGT 24 (0.4%) Core to Non-core 606 (11.2%) Core to HGT 687 (12.8%) Non-core to HGT 55 (1.0%) Total Interactions 5388 (100%) Such a model of preferential attachment has previously been proposed to explain the growth of protein interaction networks in S. cerevisiae [ 20 , 50 , 51 ] and was also suggested recently for E. coli [ 52 ]; however it has remained mostly unproven since it was difficult to trace back the evolution history of protein networks. Along this line, the HGT genes in E. coli offer a unique opportunity to test this theory since these genes are indeed \"new genes\" that were only added to the network after the HGT event ~100 millions ago [ 5 ]. Our observation provided direct evidence and support for this model, which has not been reported previously. Data Availability Additional file 6 contains the data used and produced in this study."
} | 5,547 |
31542835 | PMC6754823 | pmc | 4,461 | {
"abstract": "Neolamarckia cadamba is an important fast growing tree species used for pulping and wood material in industry for it’s desirable wood properties. As one of the most important content in wood, lignin provides structural integrity, strength, and hydrophobicity to the thickened cell walls and is the major factor contributing to biomass recalcitrance. It does not reduce the palatability of forage grass for animals, but it hinders the isolation of cellulose fibers and the efficient enzymatic depolymerization of cellulose and hemicellulose into fermentable sugars in biorefining processes by limiting the access by hydrolytic enzymes to their polysaccharide substrates. This work focused on analyzing the functions of NcCSE (Caffeoyl Shikimate Esterase, GenBank accession number: MG739672) and NcH CT (Hydroxycinnamoyl Transferase,GenBank accession number: MG739673) in the lignin biosynthetic process in order to improve the potential for utilization of leaves and wood from N. cadamba . The mutant phenotype of cse - 2 was dramatically complemented to WT in the stable transgenic lines cse - 35S::NcCSE , but overexpression of NcHCT in the cse - 2 mutant did not have the same result as cse - 35S::NcCSE , providing only partial complementation.",
"introduction": "Introduction Neolamarckia cadamba (Rubiaceae , Anthocephalus ) is a tropical evergreen tree species with a natively distribution in South and Southeast Asia. It is famous for its fast growing and perfect trunk, which can reach a height of 45 m with a DBH (diameter at breast height) of 100–160 cm (Krisnawati et al. 2011 ). It is excellent for use in building, furniture fabrication and papermaking. Moreover, it is characterized by an abundance of alkaloids in the leaves and bark, such as cadambine and dihydroconchonine, which can be used as treatments for several diseases like fever, anemia, diabetes, tumors, and have been studied for more than 100 years (Dwevedi et al. 2015 ). In our previous research, we found that the leaves of N. cadamba contained an abundance of protein and fat, and the compositional parameter index as a forage is even better than that of the common forage species Alfalfa ( Medicago sativa Linn) (Additional file 1 : Table S1), so it is also good as a woody forage plant. The plant cell wall, which is the most abundant renewable energy in nature, the main component of plant biomass and essential mechanical support for the plant, is a complex matrix of several different polymers enclosing the plasma membrane, maintaining cell shape and protecting the cell (Hückelhoven 2007 ). The major components making up the framework of the cell wall are cellulose organized into microfibrils, hemicelluloses and lignin filling the space between the microfibrils (Salmén 2015 ; Adel et al. 2016 ). Of these compounds, lignin is the major factor limiting the digestibility of forage dry matter, the conversion of lignocellulosic biomass to fermentable sugars, and the effectiveness of pulping in papermaking (Powell et al. 2017 ; Zeng et al. 2014 ; Constant et al. 2016 ). Lignin is a complex polymer made from different kinds of monomeric subunits linked via ether bonds or carbon–carbon bonds, and the most abundant of these units are guaiacyl (G), syringyl (S), and p -hydroxyphenyl (H) groups (Shuai et al. 2016 ). Lignin is synthesized via shikimate through the general phenylpropanoid pathway and a monolignol-specific pathway, which include 10 well-known enzymes (PAL, phenylalanine ammonia-lyase; C4H, cinnamate 4-hydroxylase; 4CL, 4-coumarate: CoA ligase; HCT, p-hydroxycinnamoyl-CoA: quinate shikimate p-hydroxycinnamoyltransferase; C3H, p -coumarate 3-hydroxylase; CCoAOMT, caffeoyl-CoA O-methyltransferase; CCR, cinnamoyl-CoA reductase; F5H, ferulate 5-hydroxylase; COMT, caffeic acid O-methyltransferase; CAD, cinnamyl alcohol dehydrogenase) (Boerjan et al. 2003 ; Vanholme et al. 2010 ; Ouyang et al. 2016 ), and a more recently identified enzyme, CSE (Caffeoyl shikimate esterase), which also participates in this process (Vanholme et al. 2013 ). Many studies have identified the role of the first 10 enzymes in lignin biosynthesis process, and the model of lignin biosynthesis based on these 10 enzymes has been accepted for more than 10 years. However, for the new member, more work needs to be done in order to verify its position in lignin biosynthesis. It is believed that CSE and HCT may have a similar function in this biosynthetic process, as both of them can hydrolyze caffeoyl shikimate, a well-known intermediate in lignin biosynthesis, to produce caffeic acid or caffeoyl-CoA (Vanholme et al. 2013 ). In Saleme’s study, down-regulation of CSE in hybrid poplar result in a reduced lignin deposition and relatively higher cellulose content (Saleme et al. 2017 ). But, the relation of CSE and HCT is not mentioned. In this study, we are trying to determine the functions and relationship of NcCSE and NcHCT in the process of lignin biosynthesis and there potential influence in biomass accumulation, cell wall deposition, lignin per percentage and composition by heterologous express the NcCSE and NcHCT in Arabidopsis cse - 2 mutant and wild type plant. We find that NcCSE can fully complement ate the cse-2 while NcHCT can only providing partial complementation.",
"discussion": "Discussion CSE is a key enzyme in lignin biosynthesis The lignin biosynthetic pathway has been well known for years, but it was not found that CSE, together with 4CL, bypassed the second HCT reaction, or that its preferred substrate was caffeoyl shikimate, until 2013 (Vanholme et al. 2013 ). The cse mutation was shown to reduce the expression levels of lignin biosynthesis genes upstream of it, and so block the lignin biosynthetic pathway (Vargas et al. 2016 ). Similarly, in this study, the lignin biosynthesis genes upstream of AtCSE , including C3H , C4H , HCT , were down regulated, but the gene CCoAOMT1 , which lies downstream of CSE in the pathway, was not, the inflorescence stems were smaller and lighter, the CWR and lignin content were reduced in both the leaves and the stem, and the S/G ratio increased dramatically in the cse mutant compared with WT. However, the increases in expression levels of C3H , C4H , HCT and CCoAOMT1 (Fig. 1 ), and the restoration of lignin deposition (Additional file 1 : Table S8) in cse - 35S::NcCSE lines were also accompanied by restored plant growth (Fig. 3 a, d). The mutant phenotype of cse was thus complemented in stable transgenic lines expressing NcCSE. All of these findings indicate that the N. cadamba gene NcCSE has a similar function to AtCSE in A. thaliana . CSE affects lignin monomer content In this study, the cse mutant was found to have a much higher content of H monomer and lower content of G and S monomer in the stem compared with WT, resulting in the S/G ratio increasing from 0.44 (WT) to 0.84 ( cse ). This is likely to be because CSE plays an important role in the part of the phenylpropanoid pathway leading to the G and S units after the branching off of H unit biosynthesis (Vanholme et al. 2013 ). The mutant phenotype of cse was greatly complemented but not completely restored to WT in the stable transgenic lines cse - 35S::NcCSE , though there was still a significant difference in the amount of each lignin unit between WT and cse - 35S::NcCSE . The lignin unit content of cse was less fully restored in the stable transgenic lines cse - 35S::NcHCT , which had more H units, less of the G and S units and a higher S/G ratio compared with WT (Table 2 , Fig. 5 ). Furthermore, lignin biosynthesis genes including C3H , C4H , HCT , CCoAOMT1 were up regulated in the stable transgenic lines cse - 35S::NcCSE but were not affected in the transgenic lines cse - 35S::NcHCT compared with the cse mutant (Fig. 1 ), and alterations in the activity of these related enzymes is also likely to have affected the composition of lignin units (Franke et al. 2002 ; Besseau et al. 2007 ; Li et al. 2010 ). These findings again suggest that NcCSE has a similar function to AtCSE and also that it is more effective than NcHCT in restoring the distribution of lignin units content in A. thaliana , and that NcCSE may be more active than NcHCT in the section of the N. cadamba phenylpropanoid pathway that leads to to G and S units after the branching off of of H unit biosynthesis. NcCSE may be more active than NcHCT In the study, the mutant phenotypes of cse - 2 were complemented, either completely or partially, in the stable transgenic lines cse - 35S::NcCSE ; these characters included height and dry weight of the main inflorescence stem, dry weight of the total aboveground part of the plant, and CWR and lignin content of stems and rosette leaves. However, the phenotypes of the stable transgenic lines cse - 35S::NcHCT were not restored to WT, in comparison to which they showed significant differences (Additional file 1 : Tables S7, S8). Additionally, the xylem of cse - 35S::NcCSE and WT was red, but the color of cse - 35S::NcHCT xylem lay in between blue and red, in the safranin and fast green staining experiment (Fig. 4 ). When stained with phloroglucinol, the xylem of the cse mutant was light red but the the xylem of WT was dark red. The color of cse - 35S:NcHCT and cse - 35S:NcCSE lines was intermediate, but cse - 35S:NcCSE lines were much darker in color compared with cse - 35S:NcHCT lines (Fig. 5 ). All of these observations indicate that the lignin content was higher in the cse - 35S::NcCSE lines compared with the cse - 35S::NcHCT lines. When lignin content was measured, the results showed that it was reduced by 23.3% in the stem of the cse mutant compared with WT and this severe deficiency was completely complemented by NcCSE overexpression, but cse - 35S::NcHCT lines did not show full complementation, with the exception of cse - 35S::NcHCT1 (Additional file 1 : Table S8). CSE and HCT have the same substrate, caffeoyl shikimate (Vanholme et al. 2013 ). Both NcCSE and NcHCT were expressed at the highest level in the stem segment of N. cadamba with the greatest degree of lignification; however, the expression level of NcCSE was much higher than that of NcHCT in the same segment (Table 1 ) (Ouyang et al. 2016 ). These results suggest that NcCSE may be more active than NcHCT in lignin biosynthesis in N. cadamba . Potential application of NcCSE down-regulation Lignin does not reduce the palatability of forage grass for animals (Cornelissen et al. 2014 ), but it hinders the isolation of cellulose fibers and the efficient enzymatic depolymerization of cellulose and hemicellulose into fermentable sugars by limiting the access of the hydrolytic enzymes to their polysaccharide substrate in the biorefining industry (Vanholme et al. 2013 ). In a previous study, silencing CSE in poplar did not drastically affect plant growth or development, but it reduced lignin deposition and flux into G and S units, increased the cellulose content and improved the saccharification efficiency for stems (Saleme et al. 2017 ). We found that the function of NcCSE was similar to that of AtCSE in lignin biosynthesis. A highly efficient in vitro regeneration system has been successfully established for N. cadamba (Huang et al. 2014 ), and in recent years the CRISPR–Cas9 system for genome editing has been established and applied widely to create a loss-of-function mutants affecting of specific genes (Mali et al. 2013 ; Hsu et al. 2014 ). These factors suggest that the lignin content in both stems and leaves of N. cadamba could be reduced by creating the loss-of-function mutant nccse with the CRISPR–Cas9 system, resulting in improvements in the nutrient absorption of the leaves for animals and in the total saccharification yield from the stems."
} | 2,974 |
38283868 | PMC10821166 | pmc | 4,465 | {
"abstract": "Microalgal-indigenous bacterial wastewater treatment (MBWT) emerges as a promising approach for the concurrent removal of nitrogen (N) and phosphorus (P). Despite its potential, the prevalent use of MBWT in batch systems limits its broader application. Furthermore, the success of MBWT critically depends on the stable self-adaptation and synergistic interactions between microalgae and indigenous bacteria, yet the underlying biological mechanisms are not fully understood. Here we explore the viability and microbial dynamics of a continuous flow microalgae-indigenous bacteria advanced wastewater treatment system (CFMBAWTS) in processing actual secondary effluent, with a focus on varying hydraulic retention times (HRTs). The research highlights a stable, mutually beneficial relationship between indigenous bacteria and microalgae. Microalgae and indigenous bacteria can create an optimal environment for each other by providing essential cofactors (like iron, vitamins, and indole-3-acetic acid), oxygen, dissolved organic matter, and tryptophan. This collaboration leads to effective microbial growth, enhanced N and P removal, and energy generation. The study also uncovers crucial metabolic pathways, functional genes, and patterns of microbial succession. Significantly, the effluent NH 4 + -N and P levels complied with the Chinese national Class-II, Class-V, Class-IA, and Class-IB wastewater discharge standards when the HRT was reduced from 15 to 6 h. Optimal results, including the highest rates of CO 2 fixation (1.23 g L −1 ), total energy yield (32.35 kJ L −1 ), and the maximal lipid (33.91%) and carbohydrate (41.91%) content, were observed at an HRT of 15 h. Overall, this study not only confirms the feasibility of CFMBAWTS but also lays a crucial foundation for enhancing our understanding of this technology and propelling its practical application in wastewater treatment plants.",
"conclusion": "4 Conclusions This study represents the inaugural comprehensive investigation into the performance and mechanisms of CFMBAWTS. Our findings reveal that, with HRTs set at 15, 12, 8, and 6 h, the CFMBAWTS exhibited advanced treatment capabilities for sanitary effluent over a stable operational period of 120 days. As the HRT increased, we observed a corresponding improvement in N and P removal efficiency, elevated microalgae biomass concentration, heightened CO 2 fixation, and excellent energy production. Moreover, the effluent NH 4 + -N and P concertation complied with the strict wastewater discharge standards. Notably, microalgal biomass productivity was decreased with rising HRTs due to low nutrient loading. The primary constituents of microalgal cells in the CFMBAWTS, irrespective of HRTs, included glucose, C18:0, C18:1, C18:2, and C18:3. Crucially, self-adaption and synergistic interaction were major mechanisms for microbial growth, N and P removal, CO 2 fixation, and carbohydrate and lipid accumulation. Overall, this work could provide new insights into the application of CFMBAWTS for future SE treatment and serve as a strong support for the development of MBS.",
"introduction": "1 Introduction Currently, nitrogen (N) and phosphorus (P) in secondary effluent (SE) usually fail to reach the stringent wastewater discharge standards due to low carbon (C) source content and imbalanced C:N:P ratio in the secondary wastewater treatment process of the wastewater treatment plants (WWTPs) ( Table S1 ) [ 1 , 2 ]. Therefore, it is necessary to polish the overloaded N (especially nitrate nitrogen (NO 3 − -N) and ammonia nitrogen (NH 4 + -N)) and P in SE using an effective advanced treatment [ 3 ]. However, the common physio-chemical and membrane advanced treatment technologies, such as advanced oxidation, activated carbon adsorption, and membrane process, face enormous challenges due to high operational costs and considerable N and P resource waste [ [4] , [5] , [6] , [7] ]. As reported, total phosphorus (TP) and NH 4 + -N recovery efficiencies are 35.7% and 35.8% in China [ 8 ]. Moreover, massive carbon emission from conventional biological advanced N and P removal processes is also a major issue for WWTPs, with documented carbon emission intensities exceeding 0.99, 1.17, 0.68, and 0.72 kg CO 2 eq kWh −1 in South Africa, China, Germany, and the United States, respectively [ 9 ]. Based on these, an economical, ecologically friendly, and efficient advanced treatment technology should be developed to overcome these bottlenecks. Microalgal-bacterial wastewater treatment systems (MBWTS) are a proven sustainable alternative for achieving high CO 2 fixation and value-added component accumulation while removing P and N in wastewater without providing additional carbon sources [ 10 , 11 ]. Nevertheless, some demerits must be overcome for large-scale application of MBWTS. Artificial wastewater has been widely used in MBWTS, hindering its practical application. Mixing microalgae with activated sludge is usually an approach to establishing MBWTS, which will enhance biomass harvesting, prolong the culture time, increase the operating cost, and limit further development [ 12 ]. Researchers have recently tried exploiting a novel MBWTS (microalgae-based real wastewater treatment system) to promote practical application and sustainability. Microalgae have been shown to cooperate with bacteria in real wastewater (RW) to generate microalgal-bacterial flocs, accumulate carbohydrates and lipids, and improve the effluent quality of RW [ 13 ]. Despite this promising breakthrough, the self-adaptation and synergistic interactions between microalgae and indigenous bacteria remain unclear. Undeniably, self-adaptation between microalgae and indigenous bacteria in MBWTS is a prerequisite for sustaining microbial activity, function, and interaction, as well as enhancing the production of energy-rich compounds (carbohydrate or lipid) and the removal of N and P. Microalgae and indigenous bacteria can provide stable self-adaption environments for each other by inhibiting undesirable factors, uptaking N and P, and supplying cofactors and nutrients. The synergistic interaction of these self-adaptation factors can promote microbial growth, nutrient exchange, complete nutrient removal, and energy production. However, the detailed mechanisms, corresponding metabolic pathways, and related functional genes remain unclear. Furthermore, research on microbial succession, the mechanisms of N and P removal between microalgae and indigenous bacteria, CO 2 fixation, and biomass harvesting is still limited, hampering robust support for practical applications. Besides, microalgal species and batch systems are always major limiting factors for the large-scale expansion of MBWTS. Not all microalgal species can coexist with bacteria and improve biomass harvest in complex wastewater environments [ 14 ]. The uncertainty surrounding the optimal microalgal species selection persists due to practical considerations. Batch systems, such as fixed volumes, complex conditions, and long-term operation, pose obstacles to their feasibility on a large scale [ 15 ]. In contrast, continuous flow systems (CFS) offer a promising solution by addressing these drawbacks, making them an ideal choice for transitioning MBWTS from lab-scale to practical implementation. Despite this, the feasibility of a continuous flow microalgal-indigenous bacterial advanced wastewater treatment system (CFMBAWTS) for treating real SE is unclear. For CFMBAWTS, hydraulic retention times (HRTs) affect nutrient loading and microbial activity, diversity, and interaction, thereby affecting the overall performance of the wastewater treatment process—an aspect that warrants further exploration [ 16 ]. To fill this research gap, this study constructed a CFMBAWTS and determined appropriate microalgal species for treating SE. The study delved into the performance of the CFMBAWTS under different HRTs. Specifically, the variations in N and P, biomass, chlorophyll content, CO 2 fixation, carbohydrates, and lipids were monitored under long-term continuous operation (120 days). HRTs were selected to meet different N and P discharge standards, including Class-II, Class-V, Class-IA, and Class-IB. Finally, the stable self-adaptation mechanisms, synergistic interaction, N and P removal processes, and microbial composition were determined to thoroughly understand the stability and feasibility of CFMBAWTS. This work provides a valuable reference for the applicability of achieving low energy consumption and high-efficiency microalgae-based advanced wastewater treatment, which contributes to promoting the development of microalgae wastewater treatment units under strict wastewater discharge standards.",
"discussion": "3 Results and discussion 3.1 Selection of the preferable microalgal strain for CFMBAWTS As shown in Table S2 and Figs. S1–2 , the bacteria in SE did not inhibit microalgal growth in the CFMBAWTS. The biomass concentration of F2 (1.64 g L −1 ) was higher than that in MA1 (1.57 g L −1 , p < 0.05) and JSC4 (1.53 g L −1 , p < 0.05), and no significant differences were observed ( p > 0.05) between MA1 and JSC4. These results suggest that F2 has a higher tolerance to real SE than MA1 and JSC4. Consistent with previous reports, Scenedesmus sp. had higher wastewater tertiary treatment capabilities than Chlorella sp., exhibiting higher biomass levels (1.58–1.79 g L −1 ) and a wider tolerance range for N:P ratio [ 21 ]. CO 2 fixation also followed the trend F2 > MA1 > JSC4, with statistically significant differences noted between each pair of strains (F2 vs. MA1, p < 0.05; F2 vs. JSC4, p < 0.05), except for MA1 vs. JSC4 ( p > 0.05). This implied that F2 excelled in mitigating CO 2 emissions. This finding aligns with the work of Arbib et al. (2014), who reported a lower CO 2 consumption rate for Chlorella sp. compared to Scenedesmus sp [ 21 ]. Additionally, F2 had superior sedimentation, followed by JSC4 and MA1 ( Table S2 ). This observation could be attributed to the distinct morphology of each strain. Specifically, MA1 with lower dry weight and smaller cell size may prefer suspension in the solution. Conversely, F2 with higher dry weight, larger cells, and unique morphological characteristics, such as flagellum, aggregates of four cells, and large contact surface, tend to form microbial flocs conducive to gravity settling. The influent's average NH 4 + -N, NO 3 − -N, and TP concentrations were 23.22, 4.29, and 2.75 mg L −1 , respectively ( Figs. S1a–c ). At an HRT of 15 h, TP concentrations in the effluents from all reactors were zero, whereas the NH 4 + -N and NO 3 − -N concentrations of F2-CFMBAWTS, JSC4-CFMBAWTS, and MA1 - CFMBAWTS were below 0.30 and 0.5, 4.5 and 0.7, and 3 and 0.7 mg L −1 , respectively. Notably, only the F2-CFMBAWTS upgraded the NH 4 + -N and P concentration in effluent from Class-II of GB18918-2002 to Class-II of GB3838-2002 discharge standard, demonstrating better potential for treating SE. These findings underscore the self-adaption ability of strain F2 in CFMBAWTS due to its superior N and P removal abilities and ease of biomass harvesting. Consequently, F2 was chosen for further experiments. 3.2 Effect of HRTs on CFMBAWTS performance To explore the performance of CFMBAWTS, microalgal biomass, chlorophyll content, CO 2 fixation, and N and P concentrations in the effluents were investigated under different HRTs ( Fig. 1 ). The microalgal biomass and chlorophyll content were consistent with changes in HRTs. High microalgal biomass (1.65 g L −1 ) and chlorophyll content (14.54 mg L −1 ) were obtained at 15 h of HRT, while the lowest microalgal biomass (1.38 g L −1 ) and chlorophyll content (8.45 mg L −1 ) were obtained when HRT was 6 h. These results indicate that higher influent flow rates can increase nutrient load and enhance microalgal biomass productivity, consistent with previously reported literature [ 22 ]. Similarly, the CO 2 fixation rate (0.12 g L −1 h −1 ) was significantly increased under a higher nutrient load. The highest CO 2 fixation (1.23 g L −1 ) was observed at an HRT of 15 h, indicating that CO 2 fixation ability was closely related to microalgal growth. Fig. 1 Variations of NH 4 + -N, NO 3 − -N, and TP concentrations and characteristics of strain F2 in CFMBAWTS: a , Influent and effluent of NH 4 + -N; b , Influent and effluent of NO 3 − -N; c , Influent and effluent of TP; d , Biomass concentration, chlorophyll content, and carbon fixation ability of strain F2. Fig. 1 HRT influenced the removal of NH 4 + -N, NO 3 − -N, and TP during long-term operation. TP could be fully removed when the HRT was higher than 6 h, while NH 4 + -N and NO 3 − -N concentrations were 0.07–0.33 and 0.22–0.46, 0.56–1.14 and 0.45–0.77, 2.73–4.12 and 0.70–1.56, and 4.70–6.40 and 1.26–2.64 mg L −1 when HRTs were 15, 12, 8, and 6 h, respectively. This indicates that microalgae can assimilate more NH 4 + -N, NO 3 − -N, and P and achieve satisfactory water quality for the effluent discharge at longer HRTs. Notably, the NH 4 + -N and P concentrations in effluent complied with Class-II and Class-V (GB3838-2002), as well as Class-IA and Class-IB (GB18918-2002) wastewater discharge standards when HRTs were 15, 12, 8, and 6 h, respectively ( Table S1 ). 3.3 Effect of HRTs on energy production potential in CFMBAWTS Biodiesel and bioethanol are alternative energy sources [ 23 ]. Microalgae with rich lipid or carbohydrate contents can be superior precursors for producing biodiesel or bioethanol [ 24 ]. The lipid and carbohydrate content and composition, along with the energy yield, were investigated to understand the energy production potential of CFMBAWTS ( Fig. 2 ). The highest total energy yield (32.35 kJ L −1 ) was achieved when HRT increased from 6 to 15 h, composed of lipids (20.36 kJ L −1 ) and carbohydrates (11.99 kJ L −1 ). The highest lipid (33.91%) and carbohydrate (41.91%) contents were obtained when HRT was 15 h. In addition, similar fatty acid and carbohydrate compositions were observed under different HRTs. The glucose content was highest, followed by arabinose and xylose, regardless of HRT changes, indicating that the microalgal biomass could provide a great source material for generating bioethanol ( Fig. 2 a) [ 25 ]. Moreover, microalgae can synthesize considerable amounts of C16–C18 fatty acids suitable for making biodiesel [ 26 ]. The percentage of oleic acid (C18:1) is the key factor influencing biodiesel quality as it can largely improve oxidation stability and cold flow properties [ 27 ]. Fig. 2 b showed that C18:1 was the dominant fatty acid in strain F2-CFMBAWTS, implying that this strain could effectively convert C, N, and P sources from SE to high-quality bioethanol or biodiesel. This characteristic enhances the environmental friendliness of CFMBAWTS and contributes to its economic feasibility. Fig. 2 Energy production of F2-CFMBAWTS: a , The compositions of carbohydrates; b , The compositions of fatty acids; c , The contents of lipids and carbohydrates, and the energy yield. Fig. 2 3.4 Effect of HRTs on microbial succession in CFMBAWTS To reveal the biological self-adaption and synergistic interaction mechanism of CFMBAWTS, the changes of microbial communities under different HRTs were investigated ( Fig. 3 ). As seen in Fig. 3 d, Chlorophyta dominated the composition at various HRTs, significantly surpassing its presence in SE. This observation suggests that the N and P levels in SE are favorable nutrients for growing microalgae, and microalgae can adapt to survive in SE due to their excellent self-adaptation abilities [ 28 ]. Meanwhile, bacteria supply vitamins, siderophores, growth hormones, and other compounds that promote microalgal growth [ 29 ]. Furthermore, the percentage of Chlorophyta increased with increasing HRTs, demonstrating that microalgae could assimilate more N, P, and nutrients from wastewater to achieve higher cell growth, ultimately resulting in lower effluent N and P concentrations. Fig. 3 a – c , The microbial succession pattern of bacteria at the phylum ( a ), class ( b ), and genus ( c ) levels. d , The microbial community compositions of eukaryotes at the phylum level. Fig. 3 In Fig. 3 a, Proteobacteria, Bacteroidetes, and Firmicutes were the most abundant bacterial phyla in all groups. Compared with the Raw, the relative abundance of Proteobacteria increased, but the relative abundances of Bacteroidetes and Firmicutes decreased under different HRTs, demonstrating that these three main phyla could establish stable long-term synergistic interaction and self-adaption relationship with microalgae. Proteobacteria promote the formation of microalgal-bacterial flocs, enhancing biomass harvesting and avoiding microorganisms being washed out of the CFMBAWTS [ 30 ]. Bacteroidetes and Firmicutes participated in the nitrifying and denitrifying processes, indicating that they could compete with microalgae for N sources, while microalgae produce dissolved oxygen to inhibit their growth [ 31 ]. The relative abundance of Cyanobacteria showed a similar tendency under different HRTs and was higher than that in the Raw group, demonstrating that the photosynthesis of Cyanobacteria may also occur with microalgae in CFMBAWTS. At the class level, the percentages of Alphaproteobacteria and Gammaproteobacteria were varied in all groups ( Fig. 3 b). The total percentages of Gammaproteobacteria and Alphaproteobacteria in the CFMBAWTS at different HRTs were higher than those in the Raw group. This is likely because Gammaproteobacteria and Alphaproteobacteria could supply vitamins and acyl homoserine lactone signaling molecules to create a favorable self-adaption environment for microalgal growth [ 32 ]. In terms of genera level, microalgae changed the domain genera in the Raw group ( Fig. 3 c). In the CFMBAWTS, the relative abundances of Sphingopyxis and Caulobacter increased with increasing HRT. Sphingopyxis and Caulobacter secrete extracellular polymeric substances (EPSs) that could enhance the formation of microalgal-bacterial flocs to strengthen the cross-talk between microalgae and bacteria, achieving lower effluent N and P concentrations and higher microalgal growth at a longer HRT [ 33 ]. The relative abundance of Hyphomicrobium decreased while the relative abundance of Brevundimonas increased when HRT increased from 6 to 15 h, suggesting that microalgae harmed Hyphomicrobium by releasing DO [ 34 ]. Brevundimonas has been shown to boost microalgae growth by a threefold increase [ 35 ]. Interestingly, irregular changes in Thauera and Acinetobacter were observed under different HRTs. The relative abundance of Thauera was lower at HRTs of 12 and 15 h. Meanwhile, the relative abundance of Thauera was higher under HRTs of 8 and 6 h. Thauera is an aerobic denitrifying bacterium that could improve the removal of NO 3 − -N and competes with microalgae for NO 3 − -N assimilation in an aerobic environment [ 36 ]. Acinetobacter was enriched at an HRT of 6 h, which could promote microalgal growth by secreting IAA [ 36 ]. The relative abundance of Nitrincola (nitrifying bacterium) was lower in the CFMBAWTS than in the Raw group, insinuating that microalgae may have a stronger ability to compete for NH 4 + -N assimilation than Nitrincola . Together, stable self-adaption and synergistic interactions between microalgae and bacteria were obtained, which is conducive to the long-term stability of CFMBAWTS. 3.5 Effect of HRTs on functional genes analysis in CFMBAWTS To better understand self-adaption and synergistic interaction mechanisms in CFMBAWTS, a series of functional genes was analyzed. In general, the removal pathways of NH 4 + -N were mainly nitrification and assimilation [ 37 ]. Microalgae could create favorable self-adaption conditions (oxygen-rich) for nitrifying bacterial growth, converting NH 4 + -N into NO 2 − -N and NO 3 − -N. NH 4 + -N could be removed by microalgal assimilation. As seen in Fig. 5 a, with the increase of HRTs, the abundances of functional genes related to nitrification decreased, demonstrating that microalgae could compete with nitrifying bacteria for NH 4 + -N removal and may inhibit them, leading to improve microalgal self-adaption ability and enhance NH 4 + -N assimilation. As reported, the intervention of microalgae significantly increased NH 4 + -N removal through the assimilation of microalgae, with rates largely higher than the bacteria-alone group [ 31 ]. NO 3 − -N removal pathways mainly involved assimilatory NO 3 − -N reduction (converted to NH 4 + -N and N 2 ), dissimilatory NO 3 − -N reduction, and denitrification [ 38 ]. As seen in Fig. 5 a, the functional genes related to assimilatory NO 3 − -N reduction were found in the lowest abundance compared to dissimilatory NO 3 − -N reduction and denitrification, meaning the assimilatory NO 3 − -N reduction was not a main pathway to remove NO 3 − -N in this system. The initial step in dissimilatory NO 3 − -N reduction and denitrification involves the conversion of NO 3 − -N to NO 2 − -N. The abundance of dissimilation-related genes was higher than those related to denitrification, suggesting that the dissimilatory NO 3 − -N reduction was the main route to remove NO 3 − -N. Notably, the abundance of dissimilation-related genes in NO 3 − -N reduction remained steady under different HRTs, indicating that microalgae could provide a stable self-adaption condition for dissimilatory NO 3 − -N reduction. The denitrification-related genes decreased when HRTs increased from 6 to 15 h, indicating that microalgae have a greater effect on influencing denitrification in CFMBAWT than dissimilatory NO 3 − -N reduction. This trend may be attributed to the lower oxygen sensitivity of dissimilatory bacteria, which are largely non-strict anaerobic. For the whole NO 3 − -N removal pathway, the abundance of functional genes and the effluent NO 3 − -N concentrations were decreased when HRT increased from 6 to 15 h, implying that microalgae could make up for the deficiency of bacterial NO 3 − -N removal in the presence of DO. P accumulation and assimilation are usually presented as the two main biological pathways in microalgal-bacterial systems (MBS) [ 38 ]. As depicted in Fig. 5 b, the total abundance of P-related functional genes decreased with rising HRT, implying that P removal occurred mainly through the assimilation of microalgae in CFMBAWTS. This aligns with previous reports highlighting higher P removal efficiency in MBS compared to microalgae or bacteria alone [ 39 , 40 ]. Together, microalgae affected the N and P metabolism of bacteria, effectively compensated for the deficiency of functional bacteria, and enhanced N and P removal by the stable self-adaption and synergistic interactions between microalgae and bacteria. More importantly, this symbiosis improves microalgae biomass accumulation and considerable carbohydrate and lipid production. However, N and P assimilation are not the only mechanisms for biological self-adaption, synergistic interaction, and carbohydrate/lipid accumulation in microalgae-bacteria systems. As reported, bacteria can supply cofactors (such as vitamins, siderophores, and phytohormones) to promote microalgal growth and carbohydrate and lipid accumulation [ 41 , 42 ]. The metabolic pathways related to the cofactors were investigated in CFMBAWTS ( Fig. 4 ). Cobalamin (vitamin B12), thiamine (vitamin B1), and biotin (vitamin B7) are the main vitamins benefitting microalgal growth [ 43 ]. However, most microalgae lack the synthetic pathways to produce vitamins. Therefore, these vitamins needed for microalgae are usually taken from exogenous environmental sources [ 44 ], and the vitamin-producing bacteria could exist in MBS to supply the above vitamins to microalgae [ 45 ]. This study detected thirty-five, ten, and thirteen genes involved with bacterial vitamin (vitamin B12, B1, and B7, respectively) metabolisms ( Fig. 6 a,d,c). As a critical step for vitamin B12 synthesis [ 46 ], the cobNST genes were detected, and the abundance increased with rising HRT. The abundance of functional genes associated with vitamin biosynthesis and transport in bacteria was the highest at an HRT of 15 h, whereas the lowest abundance was discovered when HRT decreased from 15 to 6 h, implying that the exchange of vitamins between bacteria and microalgae had a tight relationship with microalgal growth. More importantly, these vitamins can be directly utilized by microalgal metabolisms; vitamins B7, B1, and B12 would individually participate with the microalgal metabolisms of fatty acid synthesis, primary carbohydrate metabolism, and lipid and carbohydrate metabolism [ 47 ] to improve the accumulation of carbohydrate and lipid in microalgae. In addition, riboflavin (vitamin B2) is the essential component of flavin adenine dinucleotide and flavin mononucleotide and could play a vital role in microalgal growth [ 48 ]. In this study, six genes ( ribA , ribD2 , ribD , ribB , ribH , and ribE ) related to the metabolic pathway of vitamin B2 were observed ( Fig. 6 b). The abundance of functional genes was consistent with cell growth trends, implying vitamin B2 may have a beneficial impact on the growth of microalgae. Fig. 4 The key metabolic pathways between microalgae and bacteria. Fig. 4 Fig. 5 The metabolic pathways and functional genes related to N ( a ) and P removal ( b ). Total: The total abundance of the functional genes in the metabolic pathways. Fig. 5 Fig. 6 The biosynthesis and transport of vitamin B12 ( a ), vitamin B2 ( b ), vitamin B7 ( c ), and vitamin B1 ( d ). Fig. 6 Additionally, iron is an indispensable element for microalgal metabolism, affecting carbon fixation and cell growth rates. However, the low solubility of iron (in the form of ferric ion, Fe 3+ ) in water may prevent iron uptake by microalgae. Interestingly, siderophores (specific iron chelators) produced by bacteria can be considered vital mediators between bacteria and microalgae and could exhibit high iron binding affinity to increase its solubility [ 49 ]. The detailed process is described in Fig. 4 . In brief, siderophores bind to iron, converting it to generate Fe 3+ , and this form is subsequently reduced to ferrous ion (Fe 2+ ) before microalgae uptake through ferrireductases/adjacent Fe 2+ transporters. Meanwhile, in return, microalgae would provide dissolved organic matter (DOM) to promote the growth of bacteria through “carbon-for-iron mutualism” [ 50 ]. As depicted in Fig. 7 b, three siderophores (vibriobactin, bacillibactin, and enterobactin) were discovered, known to significantly support iron uptake and growth of microalgae [ 50 , 51 ]. Notably, ten genes ( entF , entE , entB , entA , entD , vibe , vibB , DhbF , DhbE , and DhbB ) related to siderophore synthetase were increased when HRT increased from 6 to 15 h. This indicates that siderophores can enhance microalgae metabolism and accelerate carbohydrate and lipid accumulation [ 52 ]. Fig. 7 The biosynthesis and transport of indole-3-acetic acid ( a ) and siderophores ( b ). Fig. 7 Another key mechanism involved in our study revolved around the synthesis and utilization of indole-3-acetic acid (IAA) ( Fig. 4 ). Specifically, bacteria can use endogenous tryptophan (Trp) as a precursor of IAA biosynthesis as well as secret IAA to promote microalgal growth, Calvin cycle, and fatty acid synthesis. Meanwhile, microalgae secreted Trp to induce the production of more IAA by bacteria, forming a favorable cross-talk between microalgae and bacteria [ 53 ]. In CFMBAWTS, bacterial IAA was synthesized from Trp by four pathways: indole-3-pyruvate (IPA), tryptamine (TPM), indole-3-acetonitrile (IAN), and indole-3-acetamide (IAM) ( Fig. 7 a). Not all genes involved in these pathways have been discovered, such as IpdC and ALDH (in IPA pathway), DDC, MAO , and ALDH (in TPM pathway), nthA (in IAN pathway), and iAAM (in IAM pathway). The percentage of these genes increased when HRT increased from 6 h to 15 h, with the abundance of genes in the IPA and IAN pathways notably high. This observation suggests that IPA and IAN are vital pathways for secreting IAA in CFMBAWTS regardless of HRTs. A stable self-adaption process and synergistic interaction always occurred simultaneously, completing continuous pollutant removal and energy production. 3.6 Perspective and improvement directions for practical application While the lab-scale CFMBWT has shown notable advancements in effectively removing N and P from real SE, certain critical hurdles in the practical application of CFMBWT still warrant thorough investigation. Future research should make a more significant effort in the following areas. (1) How to ensure the stability and high quality of the system's effluent during the natural light-dark cycle? The impact of the natural light-dark cycle on the stability and effectiveness of microalgae-based wastewater treatment technology has posed a significant challenge, primarily due to the limitations in microalgae's responses to darkness. Fortunately, researchers found that the surplus oxygen produced by microalgae in the daytime could maintain an aerobic atmosphere at nighttime to promote microalgal and bacterial metabolism, thereby achieving simultaneous organics, N, and P removal in MBS [ 54 ]. The average removal efficiencies of organics, N, and P in the light-dark cycle were 93.1%, 62.5%, and 80.8%, respectively, when the HRT was 4 h during the nighttime [ 55 ]. More importantly, the COD removal efficiency in the daytime was slightly lower than in the nighttime due to the high supersaturation of dissolved oxygen (DO) in the daytime [ 56 ]. In summary, strengthening the photosynthesis of microalgae in the daytime and reducing its dependence on light are the keys to ensuring the stability and high quality of the system's effluent during the light-dark cycle. Accordingly, future advancements in CFMBWT should focus on two aspects. Firstly, improving light availability for microalgae during the daytime. Enlarging illumination surface areas and increasing light transmittance are the important factors affecting microbial metabolism [ 57 ]. In engineering applications, omnidirectional transparent plexiglass is recommended for CFMBWT. Moreover, the orientation of photobioreactors is crucial, with those oriented north-south receiving more light throughout the year, especially at higher latitudes [ 58 ]. Planar waveguides have demonstrated the ability to enhance light availability for microalgal cells in photobioreactors [ 59 ]. Given their cost-effectiveness, long lifespan, and superior luminance uniformity, planar waveguides offer a promising industrial application prospect [ 60 ]. In conclusion, introducing a planar waveguide into CFMBWT can effectively dilute and homogeneously distribute light within microalgae suspension, presenting a potential solution for industrial-scale applications. Secondly, achieving the appropriate microalgal biomass concentration with a shorter HRT. High microalgal biomass will inhibit its photosynthetic efficiency, while low microalgae biomass is not conducive to removing N and P in wastewater. Moreover, the high influent flow rate can increase the nutrient load and promote microalgae biomass productivity. Therefore, exploring the appropriate microalgae biomass concentration and short HRT is necessary to avoid photoinhibition and improve effluent quality. Collectively, further improvement and validation of CFMBWT on a larger scale are necessary for its practical application. (2) How to collect and deal with the residual microalgae and bacteria from the system? Compared to conventional microalgae biomass harvesting techniques, gravity sedimentation is considered an economical solid separation method, accounting for less than 5% of the total cost. It has been widely used for clarifying the treated wastewater [ 61 ]. Our study reveals that within the CFMBWT system, microalgal-bacterial flocs can be settled by gravity without additional chemicals. In fact, inside microalgal-bacterial flocs, the interaction of microorganisms provides naturally occurring processes inducing its spontaneous flocculation [ 62 , 63 ]. However, selecting microalgae strains with favorable characteristics (such as cell size and dry weight) is crucial for efficient harvesting. Microalgae with small sizes and low dry weights are prone to suspension in solutions, impeding their collection. Conversely, microalgae with large sizes and high dry weights hinder efficient mixing with the solution. According to our experimental results and extensive experience, we recommend microalgae with a dry weight of about 400–800 mg L −1 and a cell size of about 10–20 μm for optimal wastewater treatment. Notably, more excess biomass can be produced during large-scale wastewater treatment, and the harvested biomass can serve as a valuable feedstock for generating biodiesel, bioethanol, biogas, animal feed, fertilizer, protein, pigment, and other value-added products [ 54 ]. The effective utilization of biomass resources hinges on factors such as the amount of available biomass, local demand, and economic factors."
} | 8,315 |
38276707 | PMC10820155 | pmc | 4,466 | {
"abstract": "The present work reports on an empirical mathematical expression for predicting the digital porosity (DP) of electrospun nanofiber veils, employing emulsions of poly(vinyl alcohol) (PVOH) and olive and orange oils. The electrospun nanofibers were analyzed by scanning electron microscopy (SEM), observing orientation and digital porosity (DP) in the electrospun veils. To determine the DP of the veils, the SEM micrographs were transformed into a binary system, and then the threshold was established, and the nanofiber solid surfaces were emphasized. The relationship between the experimental results and those obtained with the empirical mathematical expression displayed a correlation coefficient (R 2 ) of 0.97 by employing threshold II. The mathematical expression took into account experimental variables such as the nanofiber humidity and emulsion conductivity prior to electrospinning, in addition to the corresponding operation conditions. The results produced with the proposed expression showed that the prediction of the DP of the electrospun veils was feasible with the considered thresholds.",
"conclusion": "4. Conclusions The image analysis method is a feasible option for establishing the surface DP of SEM micrographs of electrospun nanofiber veils. However, the method accuracy can be affected if a suitable selection of the threshold is not done; such an effect was observed when thresholds I and III were employed with R 2 = 0.71 and 0.64, respectively, which indicated that the DP calculated with the equation can produce a certain error margin. Furthermore, the mathematical expression resulted highly reliable with threshold II, when surface and intermediate layers in the SEM micrographs of the electrospun veils were considered, yielding R 2 = 0.97. For this reason, it is necessary that the right threshold be identified to reproduce as close as possible the pore areas during the electrospinning process of emulsions. With the proposed mathematical model, it was found that one of the important parameters to be considered is the moisture of the electrospun veils, for a dehydration process occurs during the electrospinning as a consequence of the volatilization of solvents, and another relevant parameter was the time.",
"introduction": "1. Introduction Among the different techniques that have been employed to obtain fibrous membranes from synthetic or natural polymers for the development of materials that can be applied in the food, pharmaceutical and biomedical industries, among others, electrospinning has played a major role [ 1 , 2 , 3 ]. Membranes for their industrial use, synthesized from micro- and nano- electrospun fibers, have offered advantages due to the fact that their surface area per unit of volume and porosity can vary. Furthermore, these features are a function of the properties of the solution or emulsion to be electrospun: concentration, density, viscosity, conductivity and surface tension [ 4 ]. Additionally, electrospinning conditions such as flow rate, voltage, temperature and injector-collector distance have to be taken into account [ 5 ]. As for the characterization of porous materials, conventional techniques like mercury intrusion porosimetry (MIP), X-ray diffraction, centrifugal porosimetry and nitrogen sorption porosimetry (NSP), among others, are based on physical methods that try to represent the totality of the sample [ 6 , 7 , 8 ]. With the use of the MIP and NSP techniques, it is very likely that the membrane be destroyed at high pressures, for the pores are not rigid enough, given the characteristics of the electrospun veils [ 9 ]. Digital image analysis (DIA) has been gaining importance in the study of porous materials. Different authors have discussed the importance of employing image analysis techniques by means of SEM micrographs to estimate the porosity of cellulose-based foams and aerogels [ 10 ] and in ceramic and organic materials, among others [ 11 ]. For these reasons, DIA is considered as an alternative method for estimating porosity by seizing micro- and macropores present in materials. Wu et al. [ 12 ], for characterizing quantitatively the morphology of membranes through SEM DIA, defined parameters such as geometrical distribution of pore size, surface porosity and fractal dimension of membrane pores; the latter reflects the irregularity degree of the membrane pores. Likewise, the obtained results contributed to understanding the membrane morphology and pore formation mechanisms. Crawford et al. [ 13 ] constructed a Java plugin codified for the open code program ImageJ intended for the automated analysis of alluvial mineral and tracer images to record morphological parameters like area, perimeter and Fourier analysis parameters. The results of studying gold morphological changes during alluvial transport showed that the data produced by this model defined a quantitative relationship between the distance and transport form. Grove et al. [ 14 ] determined the total optical porosity of thin sections impregnated with blue resin by employing a macro jPOR file for ImageJ. The results were compared with the point counting method; however, jPOR provided results that were comparable to those obtained by point counting, which requires more time. Haeri et al. [ 15 ] employed the Java-based-open-code software ImageJ 1.51j8 developed by the National Institutes of Health to calculate the mean size and distance between particles and concluded that this software is particularly useful in the analysis of synthetic and natural porous constructions known as scaffolds that are usually used in the field of regenerative medicine and tissue engineering. The architecture of these scaffolds, including the size and pore density, affected significantly their interaction with biological cells and scaffold mechanical properties. Pal et al. [ 16 ], by means of representative rock fragments, obtained images of rock cores by SEM. These images were processed and analyzed with the software ImageJ to produce 2D and 3D porosity. The porosity data by the DIA technique were compared with the porosity given by using a helium gas porosimeter, finding that the 2D porosity is between 14.543–45.328% for carbonates and 3.895–35.561% for sandstone, whereas 3D porosity is between 7.8–9892% for carbonates and 3.52–9.75% for sandstone. The estimated values of 2D porosity fell within the expected interval, whereas the 3D porosity values were underestimated in comparison with the employed technique; for this reason, it was concluded that this technique is useful for establishing 2D porosity. Daraei et al. [ 17 ] analyzed the microstructure of blood clots, because the fiber diameter and clot porosity can be altered by medicaments or cardiovascular disease. They used the ImageJ complement called DiamterJ and analyzed SEM images of fibrin meshes, reporting the diameter measurements and porosity and comparing them with manual measurements and concluded that the algorithms resulted suitable for establishing the diameter through image analysis; the measurements were adjusted to the clot biophysical characteristics and manually determined values. Tan et al. [ 18 ] estimated the porosity of natural rocks by means of SEM data and a pore and crack analysis system. These authors concluded that the digital analysis was capable of identifying accurately the pore size and porosity values, which agreed with experimental data. The relationship between the bidimensional porosity estimated from digital analysis and the tridimensional porosity obtained from laboratory experiments was established. In many applications of nanofiber membranes as filters and scaffolds for tissue engineering, it is important to know the porosity of several layers, because the materials not only have high specific surface area, but also provide an inductive structure for tissue engineering. Understanding the dynamic effects of the tridimensional matrix structure and pore size in the veils is the first step that requires the optimal design of materials for tissue or membrane engineering. In the literature, a reduced number of mathematical models capable of describing a specific phenomenon or of predicting the characteristics of electrospun nanofibers is reported. This fact is due to the fact that the electrospinning process is complex and involves electrostatic processes of momentum and mass transfer. In this sense, Fridrikh et al. [ 19 ] presented equations for predicting the formation of nanofibers as a function of material properties such as conductivity (k), electric permittivity (ε), dynamic viscosity (μ), surface tension (γ) and density (ρ), and also of operative characteristics like flow rate (Q), applied electric filed (𝐸∞) and electric current (I). Conversely, Stepanyan et al., in 2014 [ 20 ] and 2016 [ 21 ], based their studies on the nanofiber elongation kinetics and determined the nanofiber radio in a single injector. Maurya et al. [ 22 ] proposed a predictive model for the fiber diameter considering the application of artificial neural networks, due to the existence of nonlinear relationships between the process variables, and poly(vinyl alcohol) and ferrous compounds. However, mathematical models for predicting the porosity of electrospun membranes are not reported in the literature. As a result, it can be concluded that a micrometric morphological study of materials can enable researchers to approach and deepen the knowledge of transport phenomena and diffusion mechanisms associated with the mass transfer of porous materials. Parameters such as diameter and pore size can be determined quantitatively through the analysis of images, which allows the characterization of the microstructure and complexity of materials. With the help of image analysis, the present study proposes the possibility of measuring the DP of various veil surface layers of experimentally electrospun nanofibers as a function of the main parameters of the electrospinning process, such as nanofiber moisture and electrospinning time, which are considered as critical experimental parameters in the mathematical expression employed to determine the DP of veils. The electrospun nanofiber veils were synthesized by employing PVOH emulsions due to their amphiphilic, emulsifying and encapsulating properties [ 23 , 24 ]. Furthermore, since PVOH possesses a polar structure, because of the presence of the OH - group, it gives to it surfactant properties that are useful in the formation of emulsions with hydrophobic compounds [ 25 , 26 ].",
"discussion": "3. Results and Discussion 3.1. Physical Properties of the Nanofibers The physical properties of the PVOH-OO and PVOH-OEO emulsions were established prior the electrospinning process, and the results are shown in Table 1 . It can be observed that the aqueous emulsions presented high electrical conductivity values, which are related to PVOH. These values fell within the interval ranging from 0.43 to 0.55 mS cm −1 , and it was found that they depended on the vegetal oil concentration, where the higher oil concentration in the sample, the lower the conductivity. For OO, this phenomenon is associated with the alkyl chain in its structure and its hydrophobic character. As for OEO, the electrical conductivity values were due to benzene functional groups, double bonds and chemical composition of molecules such as d-limonene, α-himachalene, trans-verbenol, linalool, eugenol, acetyl isoeugenol and methyl chavicol, which are, in general, molecules with low hydrophilic capacity [ 29 , 30 ]. The use of the PVOH solution at 10% w / w allowed the formation of a more defined veil with thicker and homogeneous morphology with nanofiber diameters between 290–307 nm. Unlike the PVOH emulsion at 8% w / w , the nanofibers presented diameters between 173 and 179 nm. The reported diameters are close to those obtained by Rošic et al. [ 31 ] for nanofibers electrospun with PVA at 8 and 10% and whose diameters were found between 110 and 360 nm, respectively. The SEM images of the electrospun nanofibers shown in Figure 2 indicate that the PVOH solutions at 8 and 10% w / w , Figure 2 a,b, formed smooth nanofibers with homogeneous diameters, because their polymeric structure facilitated the stretching of the nanofiber during the electrospinning process [ 32 , 33 ], whereas the incorporation of vegetal oil during the formation of the nanofibers, Figure 2 c–g, produced heterogeneous diameters as a result of the encapsulation of the vegetal oil in the nanofiber body. The mean diameters of the PVOH-OO and PVOH-OEO nanofibers oscillated between 208–492 nm and 266–300 nm, severally. With the increasing oil concentration, the viscosity of the emulsions grew and with it, the diameter of the electrospun nanofibers [ 34 ]. As for the moisture content, in Table 1 , it is observed that it diminished as the PVOH concentration increased from 8 to 10% w / w ; similar results occurred by increasing the oil concentration in the PVOH-OO and PVOH-OEO emulsions [ 35 ]. The diameters of the nanofibers displayed a wide distribution due to the electrospinning process and the different variables that affect it. These results are in good agreement with those reported by Kalantary et al. [ 36 ] who stated that the major contribution to the nanofiber diameters came from the polymer concentration in the solution to be electrospun. Xu et al. [ 37 ] reported that the viscosity increase was reflected in higher nanofiber diameters; on the other hand, Khajavi et al. [ 38 ] asserted that the conductivity diminution increased the nanofiber diameters. As for the orientation of the nanofibers in the studied systems, the analysis of the SEM images in Figure 2 by means of the software Orientation J 1.51j8 showed that the nanofibers are oriented randomly at different angles throughout the sample. Furthermore, it is observed that the nanofibers of the analyzed systems possess anisotropy, for the micrographs display multimodal histograms. This result was predictable, because the random orientation occurs when a spinning collector with low rotation rate of 150 rpm is used. Similar results were reported by Nitti et al. [ 39 ] who concluded that the angular velocity of the spinning drum affects significantly the orientation of the nanofibers and their anisotropic mechanical properties [ 40 , 41 ]. 3.2. Determination of the Digital Porosity (DP) The results corresponding to the DP calculation of the electrospun nanofibers from the SEM micrographs in Figure 1 are shown in Table 2 . The DP of the surface layers employing threshold I presented an average value of 40.12 ± 3.30, which is higher than the DP of intermediate layers obtained with threshold II with an average value of 27.44 ± 1.40, and finally, the DP of the intermediate and internal layers calculated with threshold III, which is lower than the previous ones with an average value of 19.66 ± 1.13, which was due to the fact that more nanofibers overlap and for this reason, the empty areas have lower magnitude. The physical characterization of the electrospun PVOH veils shows that the concentration and viscosity of the solution are important variables for controlling the nanofiber porosity. Notwithstanding, other factors can also affect the morphology of the electrospun material like the solution feeding rate, the injector-collector distance, voltage, etc. [ 42 ]. 3.3. Proposed Model for Predicting the DP With the analysis of the SEM micrographs and the image analysis technique, the development of a new mathematical expression was proposed to estimate the DP with experimental data obtained by image analysis. The preceded DP prediction model ( D P P ) is shown in Equation (7): (7) D P P = − ln V × k × d I ( ln ( W f W i ) e θ n \nwhere V is the voltage (V), k is the conductivity (S/m), W f and W i are the final and initial moisture contents of the veil (g H 2 O)/(g b. s.), severally, θ is the electrospinning time (h), I is the electric current (A), d is the injector-collector distance (m) and n is a fractal exponent. The mathematical expression was established according to the following criteria: In the literature, it has been reported that the morphology and properties of the nanofibers, including the diameter, porosity, alignment and mechanical behavior, depend on the polymer solution properties (concentration, viscosity, surface tension and dielectric properties); on processing parameters (voltage, volumetric flow rate, injector-collector distance and intensity of the applied electric field); and environmental conditions (temperature, atmospheric pressure and moisture) [ 43 , 44 ]. Additionally, the electrospinning time was considered, because it is an important variable during the electrospinning process. Essaldi et al. [ 45 ] reported that the mean size of the space between nanofibers was smaller for longer electrospinning times. Finally, the final moisture content of the veil was taken into account, because during the electrospinning process, solution dehydration occurs as a consequence of solvent volatilization [ 46 ]. It is observed that Equation (7) is a function of the operation and solution parameters and moisture of the electrospun veils. Table 3 shows the values used in Equation (3) to predict the DP mathematically. Figure 3 , Figure 4 and Figure 5 show the validity of the mathematical expression proposed to predict the DP. As can be observed in Figure 3 (threshold I), through the image analysis of the electrospun veils, there is a relationship between the calculated pore size and the DP calculated with the mathematical equation, obtaining R 2 = 0.71, which indicates that the equation can predict the pore sizes in the surface layer of the electrospun veil; the RSS and MSE values featured in Table 4 represent the highest values with respect to those obtained with thresholds II and III, which reveals a certain error margin in the estimation of the DP. Conversely, Figure 4 shows that the mathematical expression improved the prediction (R 2 = 0.97) with threshold II when the surface and intermediate layers of the electrospun veil SEM micrograph were employed, presenting the lowest RSS and MSE values, as observed in Table 4 . Finally, in Figure 5 , when threshold III was employed, once again the prediction diminished, due to the fact that the nanofibers overlapped, and the DP was reduced by the color contrasts. It is worth emphasizing that, for each threshold, the fractal exponent ( n ) of the mathematical expression is different. The statistical parameters (R 2 , R, RSS and MSE) for establishing the correlation between the experimental data and those predicted by the proposed equation are displayed in Table 4 . The statistical parameters for threshold II indicate that the equation correctly fits the calculated values, giving a prediction error of approximately 3%, whereas the mathematical expression for thresholds I and III had a prediction percentage error of 32 ± 3%. In a similar analysis, Powell et al. [ 47 ] proposed a correlation between the nanofiber porosity and composition of the solution of gelatin electrospun nanofibers, obtaining R 2 = 0.70. Likewise, by means of response surface studies and RNA for porosity prediction of electrospun nanofibers, R 2 = 0.94 and 0.89 were calculated, severally [ 43 ]. Furthermore, the physical characterization of the electrospun PVOH veils shows that the concentration and viscosity of the solution are fundamental variables for controlling the nanofiber porosity. Other factors, such as solution feeding rate, injector-collector distance and voltage, can affect the morphology of the electrospun veil [ 48 ]. Additionally, the voltage and conductivity affect the porosity of the electrospun nanofibers, for they are variables that are involved in the equation; however, the final moisture of the nanofibers is the most important factor because of the fact that, during electrospinning, a solution dehydration process occurs as a consequence of the volatilization of the solvents. The electrospinning time has an inversely proportional relationship with the DP. This is because the DP diminishes as the electrospinning time increases, which is a consequence of higher overlapping of the veil nanofibers."
} | 5,069 |
38826158 | PMC11139967 | pmc | 4,468 | {
"abstract": "Abstract Understanding what regulates ecosystem functional responses to disturbance is essential in this era of global change. However, many pioneering and still influential disturbance‐related theorie proposed by ecosystem ecologists were developed prior to rapid global change, and before tools and metrics were available to test them. In light of new knowledge and conceptual advances across biological disciplines, we present four disturbance ecology concepts that are particularly relevant to ecosystem ecologists new to the field: (a) the directionality of ecosystem functional response to disturbance; (b) functional thresholds; (c) disturbance–succession interactions; and (d) diversity‐functional stability relationships. We discuss how knowledge, theory, and terminology developed by several biological disciplines, when integrated, can enhance how ecosystem ecologists analyze and interpret functional responses to disturbance. For example, when interpreting thresholds and disturbance–succession interactions, ecosystem ecologists should consider concurrent biotic regime change, non‐linearity, and multiple response pathways, typically the theoretical and analytical domain of population and community ecologists. Similarly, the interpretation of ecosystem functional responses to disturbance requires analytical approaches that recognize disturbance can promote, inhibit, or fundamentally change ecosystem functions. We suggest that truly integrative approaches and knowledge are essential to advancing ecosystem functional responses to disturbance.",
"conclusion": "6 CONCLUSIONS Disturbances are changing in frequency, intensity, and cause worldwide (e.g., in forests: Weed et al., 2013 , Seidl et al., 2017 ; grasslands: Joyce et al., 2016 ; Chen et al., 2023 ; drylands: Maestre et al., 2022 ; coral: Vercelloni et al., 2020 ; Chen et al., 2023 ). In addition to advancing fundamental knowledge in disturbance ecology (Wohlgemuth et al., 2022 ), updated and more integrative theories relevant to ecosystem functioning are needed to guide disturbance management, and better anticipate and simulate ecosystems' responses to disturbance in this era of rapid global change. The effect of disturbance on ecosystem processes will be a primary determinant of the future functioning and service provisioning of ecosystems in the face of these changing disturbance regimes (Seidl et al., 2016 ). Understanding the varied impacts of disturbance on ecosystem functions will be an essential component of both recognizing and mitigating the effects of climate and global change factors on the health of ecosystems (Thom et al., 2017 ). For example, monitoring of ecosystem functions can provide an “early warning system” of potential ecosystem transitions or state changes (Contosta et al., 2023 ; Keen et al., 2022 ). The frameworks discussed here highlight the value of integrative theory when considering applications and illustrate a potential roadmap for incorporating multiple response types and trajectories into long‐term ecosystem monitoring practice. In addition to monitoring, ecosystem functional response to disturbance can be used as both a predictor and outcome assessment tool for evaluating the impact of management focused on promoting ecosystem adaptation to climate change and related stressors (Seidl & Turner, 2022 ). For example, climate‐adaptive management in forested ecosystems is generally conducted using silvicultural plans that focus on forest structure and species composition and diversity (Janowiak et al., 2014 ; Nagel et al., 2017 ), but often with the goal of promoting stability in functions such as carbon or water cycling (Halofsky et al., 2018 ; Ontl et al., 2020 ). Understanding how disturbance structural outcomes and changes in species identities, traits, and diversity are linked with the response of ecosystem functions is therefore essential to understanding both near‐term responses of forests to climate‐adaptive management and also the longer‐term response of future ecosystems to projected changes (Aquilué et al., 2020 ; Clark et al., 2022 ; Messier et al., 2019 ). While disturbance occurs at all scales of biological organization, disciplinary science has sometimes resulted in disparate rather than integrative theories, terminology, and concepts. Comprehensively updating disturbance theories relevant to ecosystem ecologists requires outside‐of‐the‐disciplinary‐box thinking, and such thinking necessitates reading, discussion, and research that spans disciplines. While not exhaustive, Table 1 provides a sampling of literature from biological disciplines outside of ecosystem ecology that is relevant to the four theoretical areas discussed in this commentary. We invite your contributions to this list via https://osf.io/a5zvp/ . TABLE 1 Disturbance theoretical frameworks originating outside of ecosystem ecology with applicability to ecosystem functioning. Theory Origin What it said: How it applies to ecosystem ecology: References Biogeochemical dynamics Biogeo‐chemistry The partitioning ratio of soil and plant nutrient stocks will undergo a predictable trajectory after disturbance. Offers a framework to assess ecosystem biogeochemical response to disturbance using nutrient partitioning ratios. Kranabetter et al., ( 2016 ) Multidimensional stability Population and community ecology There are multiple, quantifiable dimensions of community and population response to disturbance. Provides a conceptual and mathematical framework for interpreting and comparing ecosystem functional responses to disturbance. Hillebrand et al., ( 2018 ); Mathes et al., ( 2021 ) Intermediate disturbance hypothesis Community ecology Moderate intensity disturbances may increase species diversity by augmenting or diversifying habitat and resource availability. Species diversity, habitat breadth, and resource availability affect ecosystem functional responses to disturbance, suggesting moderate intensity disturbance could increase mass and energy fluxes. Huston, ( 2014 ) Disturbance legacies Population and community ecology Traits and adaptations, as well as the residual abiotic and biotic materials that persist through disturbance determine ecological responses. Disturbance legacies may be critical determinants of ecosystem functional responses to disturbance. Johnstone et al., ( 2016 ) Tipping points, thresholds, and alternate stable states Population and community ecology High intensity or frequency disturbance may force a permanent (i.e., stable) shift in population or community structure. Ecosystems may exhibit similar non‐linear threshold responses to disturbance, changing long‐term functioning. Scheffer & Carpenter, ( 2003 ) Diversity and resilience Community ecology Diverse communities respond to disturbance with greater functional stability. Diversity, broadly defined, may increase the stability of ecosystem functioning. Oliver et al., ( 2015 ) Landscape dynamics Landscape ecology Spatially and temporally asynchronous disturbance responses, when balanced, may have a stabilizing influence over landscape level structure and function. Patchy disturbance within an ecosystem may not be functionally destabilizing when uniform in time and space. Turner, ( 2010 ) Functional buffering Cellular biology The functional redundancy of cellular components rescues whole‐cell function. Functional buffering mechanisms exist across levels of biological organization, from cellular to ecosystems Lin et al., ( 2022 ) Abrupt Changes in Ecosystems System theory Interactions among multiple drivers often produce abrupt change in ecosystems. Suggests research priorities to advance understanding of abrupt changes in ecosystems in the face of climate change. Turner et al., ( 2020 ) Net effects and indeterminate directionality of successional processes Community ecology Ecological restoration and succession more generally, is informed by synthetic and updated vegetation dynamic theories that consider net effects and indeterminate successional pathways. Just as the net effects of multiple interacting community processes influence overall vegetation dynamics, ecosystem processes such as net primary production and net ecosystem production are determine by aggregate and sometimes opposing flux responses. Pickett et al., ( 2009 ) \n Note : We invite additional recommendations and comments from the community here: https://osf.io/a5zvp/ .",
"introduction": "1 INTRODUCTION Disturbances affect every scale and level of biological organization. However, disturbance studies are generally guided by discipline‐specific theories, terminology, and literature, limiting coherence across fields of ecology. In ecosystem ecology, prominent historical and enduring theoretical frameworks emphasize disturbance effects on systems‐level biomass, and energy pools and fluxes over time and space (Bormann & Likens, 1979 ; Odum, 1969 ; Whittaker et al., 1974 ). While the influence of these theories continues, their inception did not account for interactions with rapidly changing climate or climate extremes, permanent (i.e., state‐) changes in biogeochemical cycles, species introductions, or novel disturbances (Corman et al., 2019 ; Sala et al., 2000 ). Yet, the multitemporal and spatially integrative nature of ecosystem ecology requires long‐term consideration (Gaiser et al., 2020 ) of uncertain future conditions (Stern, 2008 ), dynamic resource ratios and stoichiometries (Jentsch & White, 2019 ), and community‐to‐landscape structural reorganization (Carpenter et al., 2001 ; Pickett et al., 2011 ; Scheffer et al., 2001 ; Scheffer & Carpenter, 2003 ). Moreover, many historical conceptual models still embraced by ecosystem ecologists were not testable when proposed because of technological constraints and more limited quantitative metrics and methods. For example, Odum's ( 1969 ) seminal work, “The Strategy of Ecosystem Development,” which is cited more now than it was a half century ago, long‐preceded meteorological “flux” tower networks (Baldocchi, 2008 ; Novick et al., 2018 ) measuring ecosystem processes such as net ecosystem CO 2 exchange and ecosystem respiration, nomenclature that was standardized in the 21st Century (Chapin et al., 2006 ). Now, following decades of observations and theoretical advances (Gaiser et al., 2020 ; Jentsch & White, 2019 ; Kranabetter et al., 2016 ; Lin et al., 2022 ), we consider how contemporary disturbance theory and knowledge can inform core themes addressed by ecosystem ecologists. Here, the term functioning encompasses system‐wide processes, such as net primary production, ecosystem respiration, evapotranspiration, and energy balance. Rather than an exhaustive review, we present a broadly accessible outline for the novice in advance of a more comprehensive dive into a rich but technical literature spanning multiple biological disciplines and decades. We conclude by inviting readers to contribute their own commentary and suggested readings, acknowledging that interdisciplinary perspectives, theories, and observations are necessary to enrich and unify disturbance ecology paradigms."
} | 2,782 |
34227665 | PMC8632791 | pmc | 4,472 | {
"abstract": "ABSTRACT Most swimming bacteria are capable of following gradients of nutrients, signaling molecules and other environmental factors that affect bacterial physiology. This tactic behavior became one of the most-studied model systems for signal transduction and quantitative biology, and underlying molecular mechanisms are well characterized in Escherichia coli and several other model bacteria. In this review, we focus primarily on less understood aspect of bacterial chemotaxis, namely its physiological relevance for individual bacterial cells and for bacterial populations. As evident from multiple recent studies, even for the same bacterial species flagellar motility and chemotaxis might serve multiple roles, depending on the physiological and environmental conditions. Among these, finding sources of nutrients and more generally locating niches that are optimal for growth appear to be one of the major functions of bacterial chemotaxis, which could explain many chemoeffector preferences as well as flagellar gene regulation. Chemotaxis might also generally enhance efficiency of environmental colonization by motile bacteria, which involves intricate interplay between individual and collective behaviors and trade-offs between growth and motility. Finally, motility and chemotaxis play multiple roles in collective behaviors of bacteria including swarming, biofilm formation and autoaggregation, as well as in their interactions with animal and plant hosts.",
"conclusion": "CONCLUDING REMARKS Although motility is among the most studied bacterial behaviors under defined laboratory conditions, its multifaceted importance for the physiology of individual bacteria and microbial communities only recently became appreciated. In this review, we provided an overview of multiple functions of motility, with a primary focus on chemotaxis. As evident from studies of the E. coli model, even for the same species chemotaxis might make multiple contributions to physiology, including nutrient acquisition, expansion of the population range, biofilm formation and host colonization. Importantly, these different functions of motility and chemotaxis are not mutually exclusive but context-dependent. Even a single E. coli chemoreceptor Tsr can mediate chemotaxis to the preferentially consumed amino acid serine, to bacterial signaling molecules AI-2 and indole, and to animal hormones (Hegde et al . 2011 ; Lopes and Sourjik 2018 ; Mesibov and Adler 1972 ; Orr et al . 2020 ; Yang et al . 2020 ). Consequently, knockouts of individual receptors frequently show pleiotropic defects, from reduced growth fitness under conditions where chemotaxis is important to reduced biofilm formation and virulence. Moreover, the deletion of general chemotaxis genes not only impairs chemotaxis but also changes the swimming pattern of bacteria, either making them smooth swimming or tumbly, which can affect surface attachment, collective behaviors or spreading even in absence of specific chemotactic responses. These intertwined effects complicate the mechanistic understanding of the observed impacts of chemotaxis and motility in such complex environments as the rhizosphere or the GI tract, where grains, surfaces or the mucus affect swimming patterns (de Anna et al . 2020 ; Figueroa-Morales et al . 2019 ; Frangipane et al . 2019 ; Galajda et al . 2007 ; Makarchuk et al . 2019 ; Sipos et al . 2015 ; Spagnolie et al . 2015 ), and need to be kept in mind while interpreting such data. Another outstanding challenge in understanding the physiological and environmental importance of bacterial chemotaxis lies in the characterization of ligand specificity for the many chemotaxis receptors present in different bacterial species. Whereas signaling domains of receptors are highly conserved and can be easily identified bioinformatically, their ligand binding domains and corresponding sensing mechanisms are highly diverse (Ortega, Zhulin and Krell 2017 ). Although several different approaches to systematically identify ligands for various chemoreceptors have been recently established (Bi et al . 2016 ; Boyeldieu et al . 2021 ; Lehning et al . 2017 ; Luu et al . 2019 ; Matilla, Martin-Mora and Krell 2020 ), only a tiny fraction of chemoreceptor ligands are currently known and even fewer have an established mode of binding (Ortega, Zhulin and Krell 2017 ). With the increasing number of characterized ligand-receptor interactions and better understanding of ligand binding by the major structural classes of ligand-binding domains of receptors, computational prediction of ligand specificity should ultimately become possible.",
"introduction": "INTRODUCTION Swimming bacteria are able to monitor changes in environmental conditions as they move and to adapt their swimming pattern accordingly, in order to swim towards their preferred environment. Such biased movement in chemical gradients, called chemotaxis, is one of the longest and most thoroughly studied bacterial behavioral responses. Understanding of the molecular mechanisms controlling the chemotactic behavior has become highly refined over the years, especially in the model organism Escherichia coli (Wadhams and Armitage 2004 ; Colin and Sourjik 2017 ; Bi and Sourjik 2018 ). As a consequence, the typical behavior of a bacterial cell in a simple gradient and the underlying biochemistry and biophysics are well understood, and they could be mathematically modelled down to minute quantitative details (Tu 2013 ; Micali and Endres 2016 ; Colin and Sourjik 2017 ; Waite, Frankel and Emonet 2018 ; Wong-Ng, Celani and Vergassola 2018 ). Flagellated bacteria typically swim in a series of more or less straight runs interrupted by short reorientations (Fig. 1A ). In peritrichous bacteria like E. coli , runs occur when all flagella rotate unidirectionally (counterclockwise in the case of E. coli ) and form a bundle that propels the cell body forward (Berg 1975 ; Macnab 1977 ). Tumbles, which result from transient reversal of the rotary direction of flagellar motors, cause the flagellar bundle to fall apart and lead to reorientation of the cell body. The strategies of reorientation in polarly flagellated bacteria are more complex and diverse, with several distinct cell motility states, which might have been evolutionary selected to match the respective bacterial habitat (Altindal, Xie and Wu 2011 ; Constantino et al . 2018 ; Grognot and Taute 2021 ; Stocker 2011 ; Taktikos, Stark and Zaburdaev 2013 ; Xie et al . 2011 ). Regardless of the specific mechanism of reorientation, such run-reorientation behavior results over long times in an active diffusion that enables bacteria to efficiently spread in their environment. Figure 1. Chemotactic behavior and signaling pathway. (A) , Two prominent types of bacterial flagellar motility patterns, run-tumble and run-reverse-flick swimming. Both types of swimming lead to effective diffusion in homogeneous environments and get biased by the chemotaxis pathway to climb up physicochemical gradients. (B) , Schematic representation of the chemotaxis pathway of E. coli , featuring clustered chemosensory complexes formed by receptors bound to histidine kinase CheA and adaptor protein CheW. Chemoreceptors detect chemical ligands, either directly via their ligand binding domains or indirectly upon interactions with periplasmic binding proteins (PBPs), and modulate activity of CheA. The signal is transmitted to flagellar motor by phosphorylation of the diffusible response regulator CheY, which modulates the direction of motor rotation. The signal is terminated by the phosphatase CheZ. Receptor methylation enzymes, the methyltransferase CheR and the methylesterase CheB carry out adaptation to steady stimulation and provide short-term memory for temporal concentration comparisons. The chemotaxis system modulates the duration of the runs according to perceived changes in environmental conditions, making them longer or shorter, if conditions get better or worse, respectively, to bias the average cell motion towards favorable environments (Berg and Brown 1972 ; Larsen et al . 1974 ; Macnab and Koshland 1972 ). The signaling pathway controlling this behavior is highly conserved among bacteria and even archaea (Fig. 1B ). Nevertheless, several distinct classes of chemotaxis pathways could be distinguished based on their detailed molecular composition and evolutionary relatedness, some of which control behaviors other than flagellar motility (Gumerov, Andrianova and Zhulin 2021 ; Wuichet and Zhulin 2010 ). In contrast to many other bacteria, the genome of E. coli encodes only a single motility and chemotaxis system, which moreover functions with a nearly minimal set of chemotaxis proteins. Such comparative simplicity turned E. coli into a preferred model for studying signal transduction in bacterial chemotaxis (Bi and Sourjik 2018 ; Parkinson, Hazelbauer and Falke 2015 ). In general, bacterial chemotaxis pathways consist of two modules – one for rapid signal transduction and another for slower adaptation (Shimizu, Tu and Berg 2010 ). The signal transduction module is composed of transmembrane chemoreceptors that change conformation upon ligand binding or other environmental perturbations and together with the adaptor protein CheW modulate the activity of a histidine kinase CheA (Parkinson, Hazelbauer and Falke 2015 ). Together with CheA and CheW, chemoreceptors form stable supramolecular sensory complexes that primarily cluster at cell poles in E. coli and other bacteria (Yang and Briegel 2020 ). The kinase CheA phosphorylates the diffusible response regulator CheY, which, when phosphorylated, binds to the flagellar motor to induce its clockwise rotation and thus cell tumbling. This signaling core is highly conserved among all chemotaxis pathways (Wuichet and Zhulin 2010 ). Many bacterial systems, including that of E. coli , also possess a specific phosphatase CheZ that rapidly dephosphorylates CheY, thereby ensuring that the phosphorylation level of CheY reflects the kinase activity with only short delay. In other chemotaxis pathways, CheY dephosphorylation is carried out by alternative phosphatases, CheC or CheX (Silversmith 2010 ). Some of the chemotaxis pathways, including the closely related pathway in Salmonella enterica , include an additional component of the sensory complexes CheV, which has the CheW-like scaffolding domain and the CheY-like regulatory domain (Alexander et al . 2010 ). The signal transduction module of the chemotaxis pathway belongs to a larger family of two-component systems (TCSs) that enable environmental sensing in prokaryotes and are also present in fungi and plants (Stock, Robinson and Goudreau 2000 ). One important difference between the canonical TCSs and the chemotaxis pathways is that in the former the sensory, kinase and phosphatase activities are typically executed by a single sensory kinase protein, whereas in chemotaxis these functions are carried out by different proteins within one stable complex (Gumerov, Andrianova and Zhulin 2021 ; Sourjik and Armitage 2010 ). Such segregation of sensory and signaling activities likely facilitates evolutionary adaptation of the chemotaxis pathway to new environmental niches with different chemoeffector requirements, where specific chemoreceptors could be rapidly acquired or lost without affecting the function of the signaling core. Indeed, both specificities and the number of chemoreceptors apparently correlate with the respective lifestyles of bacterial species (Ortega, Zhulin and Krell 2017 ). The chemotaxis pathway further includes an adaptation module that is composed of two enzymes CheR and CheB, which respectively methylate and demethylate specific residues on the receptor and thus counterbalance the effect of ligand binding on receptor conformation (Goy, Springer and Adler 1977 ; Kehry and Dahlquist 1982 ; Terwilliger, Wang and Koshland 1986 ). The (de)methylation rates depend primarily on the current activity of the receptor-kinase complex and they are slow compared to the other reactions within the pathway (Block, Segall and Berg 1983 ; Sourjik 2004 ; Sourjik and Berg 2002 ). Consequently, CheR and CheB provide a delayed integral negative feedback, which allows the cell to respond to temporal changes in experienced conditions over a wide range of backgrounds (Barkai and Leibler 1997 ; Berg and Purcell 1977 ; Kalinin et al . 2009 ; Lazova et al . 2011 ; Mesibov, Ordal and Adler 1973 ; Segall, Block and Berg 1986 ; Yi et al . 2000 ). This methylation-dependent adaptation module is unique in comparison to other TCSs, and present in the vast majority of bacterial chemotaxis pathways. In addition to CheR and CheB, adaptation in other chemotaxis pathways such as that of Bacillus subtilis involves CheV and a receptor deamidase CheD that are absent in E. coli , but the interplay between these different levels of adaptation remains poorly understood (Walukiewicz et al . 2014 ). Clustering of chemoreceptors and associated chemotaxis proteins appears to be a universal feature of all studied prokaryotic chemotaxis systems (Sourjik 2004 ; Yang and Briegel 2020 ). Clustering allows receptor-kinase complexes to respond cooperatively and thus highly sensitively to changes in environmental conditions (Sourjik 2004 ; Tu 2013 ). Since receptors with different ligand specificities are mixed within clusters, clustering further facilitates signal integration (Parkinson, Ames and Studdert 2005 ). Finally, many bacteria express multiple chemotaxis systems, and hence spatial segregation provided by clustering might help to separate proteins belonging to different systems and thus prevent their undesired interference (Sourjik and Armitage 2010 ). In contrast to this highly detailed molecular understanding of signal processing and motility control in E. coli and several other bacteria, the physiological importance of chemotaxis and flagellar motility are not well established even for model bacterial systems. In this review, we thus aim to summarize the current state of knowledge about different physiological aspects of chemotactic behavior. These range from importance of chemotaxis for enhanced nutrient acquisition by individual bacteria and population range expansion to the role that chemotaxis plays in bacteria-bacteria and bacteria-host interactions. We further illustrate how better understanding of bacterial chemotactic behavior in its physiological context(s) might help to rationalize many of its observed properties, from ligand specificity of chemoreceptors to growth-dependent regulation of motility gene expression."
} | 3,680 |
33184377 | PMC7665216 | pmc | 4,474 | {
"abstract": "Petroleum refinery wastewater (PRW) that contains recalcitrant components as the major portion of constituents is difficult to treat by conventional biological processes. Microbial fuel cells (MFCs) which also produce renewable energy were found to be promising for the treatment of PRW. However, due to the high total dissolved solids and low organic matter content, the efficiency of the process is limited. Labaneh whey (LW) wastewater, having higher biodegradability and high organic matter was evaluated as co-substrate along with PRW in standard dual chambered MFC to achieve improved power generation and treatment efficiency. Among several concentrations of LW as co-substrate in the range of 5–30% (v/v) with PRW, 85:15 (PRW:LW) showed to have the highest power generation (power density (PD), 832 mW/m 2 ), which is two times higher than the control with PRW as sole substrate (PD, 420 mW/m 2 ). On the contrary, a maximum substrate degradation rate of 0.420 kg COD/m 3 -day (ξCOD, 63.10%), was registered with 80:20 feed. Higher LW ratios in PRW lead to the production of VFA which in turn gradually decreased the anolyte pH to below 4.5 (70:30 feed). This resulted in a drop in the performance of MFC with respect to power generation (274 mW/m 2 , 70:30 feed) and substrate degradation (ξCOD, 17.84%).",
"introduction": "Introduction The constituents of wastewater generated from the petroleum industry are complex and having slow to decompose carbon (i.e. recalcitrant). The major contaminants of petroleum refinery industries are volatile phenols, sulphides, benzene, ammonia, dissolved solids, suspended solids, cyanides and nitrogen compounds 1 , 2 . All the hydrocarbons present in the petroleum refinery wastewater (PRW) are referred to as total petroleum hydrocarbons (TPHs) which include both aliphatic and aromatic hydrocarbons 3 , 4 . Treatment of such recalcitrant contaminants is challenging and demands high energy. Approximately 3.5 to 5 m 3 of wastewater is generated from one tonne of crude oil processed 5 , 6 . Biological processes such as membrane bioreactor (MBR) 7 , 8 , upflow anaerobic sludge blanket 9 , 10 and biological aerated filter reactor 2 , 11 has been used to treat oily wastewaters. However, these processes require long operational periods and energy input. Several studies were also performed integrating MBRs for improved efficiency 12 . On the other hand, facultative stabilization ponds were also studied for biological degradation of carbon and phenol in petroleum based wastewaters 13 . Microbial fuel cells (MFCs) are proven as sustainable options for the treatment of such recalcitrant wastewaters, which also produces bioelectricity simultaneously 14 – 16 . MFCs are also suitable processes for treatment of various types of wastewater with low biodegradability 17 – 19 . Various MFC studies reported treatment efficiency of PRW in the range of 30–60% 1 , 20 . Few studies reported more than 80% degradation efficiency for hydrocarbon components and COD of PRW 21 – 24 . However, due to the poor biodegradability of the PRW, low rate of removal was identified in MFCs. This anticipated more research in this area to achieve efficient and sustainable processes for bioelectricity generation from the petroleum based wastewaters. Several strategies were studied to improve the MFC performance in treating petroleum based wastewaters. Reactor configuration, use of highly conductive electrodes, cell immobilization strategies, development of efficient anodic biofilm etc. were studied to improve MFC performance in treating petroleum wastewaters 1 , 25 – 28 . Co-substrate addition is one of the strategies that used in the wastewater treatment by combining a wastewater to another wastewater by complementing the scarce component. This strategy was well studied in anaerobic digestion (AD) and acidogenic fermentation for methane and hydrogen production 29 , 30 . It was also identified that co-digestion is an interesting option for improving yields of AD. In most cases, the use of a co-substrate improves the biogas yields by establishing positive synergisms in the digestion medium and the supply of missing nutrients by the co-substrates. In addition to process advantages, economic advantages of co-substrate addition are quite significant 29 . The addition of co-substrate to the wastewater can increases the biodegradable fraction that helps to increase the total efficiency and economics of the process 29 , 31 . In MFC studies, it was also suggested that considering two hydrocarbons of different homologues achieved improved degradation efficiency 32 , 33 . Addition of electron acceptors further improves the degradation of hydrocarbons. Additions of electron acceptors such as nitrate, sulfate, iron and carbon dioxide under anaerobic conditions, link various microbial processes including nitrification, sulfate reduction, iron reduction and methanogenesis 34 – 36 . Dissimilar efficiency of hydrocarbons degradation due to addition of electron acceptor is documented through the following three aspects, (i) degradation activation, (ii) preferential degradation with different hydrocarbon structures and carbon chains, and (iii) degradation rate 36 – 39 . The improved efficiency was due to diverse metabolic processes involved in the degradation of petroleum hydrocarbons. Here, the effect of co-substrate interactions on microbial uptake is not inhibitory but rather promoted simultaneous degradation of both substrates. Since PRW is found to exhibit poor biodegradability, selecting complimentary source of wastewater is rational. Labaneh whey (LW) wastewater that is produced in large quantities is found to have higher organic matter content and is readily biodegradable 8 , 40 . LW was also found to act as suitable substrate for bioelectrogenesis under different operating conditions 41 . With this background, the present study was aimed to use LW as co-substrate for PRW to improve the substrate degradation of recalcitrant PRW. Additionally, LW was added as co-substrate that is having higher biodegradability in several ratios and operated in dual chambered MFC system. The system was clearly evaluated for biodegradability and concomitant conversion of oxidized organic matter to bioelectricity. The results were also compared with PRW as sole source of carbon for bioelectricity generation to evaluate the range of improvement due to the addition of LW as co-substrate.",
"discussion": "Results and discussion Co-substrate influence on bioelectricity generation Biological oxidation of wastewater is mainly depending on the nature of the substrate. The same is applicable for bioelectrochemical oxidation in MFCs. The substrates (PRW and LW) chosen in the present study are having contrast biodegradable nature. LW that has good biodegradability was added as co-substrate to low biodegradable PRW as substrate and the function of MFC was evaluated. This exhibited positive influence on bioelectricity generation and simultaneous improved treatment efficiency (Fig. 1 ). The initial three operating cycles operated with 100% PRW is considered as control, which exhibited closed circuit voltage of 410 mV (at 100 Ω) and current density (CD) of 1024 mA/m 2 (power density (PD), 420 mW/m 2 ) (Table 1 ). Bioelectrogenesis takes place in the control operation is due to the sole function of bioelectrochemical degradation of organics present in the PRW (Fig. 1 ). Further, MFC was operated with 5% LW and 95% PRW and the performance was compared with the control. This was resulted in improvement in bioelectrogenesis to 484 mV (CD, 1210 mA/m 2 ; PD, 587 mW/m 2 ). Here, bioelectrogenesis is due to the degradation of both PRW and LW that generated more number of electrons from the degradation. Hence, improved bioelectrogenesis in 95:05 (PRW:LW) ratio compared to the control operation (100% PRW). Along with boosting of organic matters present in the LW for bioelectricity generation, higher total dissolved solids (TDS) values of PRW might have mutually helped for the efficient electron transfer mechanism in anode chamber 42 , 43 . It was understood from other studies that the dissolved ions and bacterial activity help to deliver electrons effectively from substrate degradation 44 – 46 . This condition helps to enhance current generation in MFCs. High concentrations of dissolved ions present in PRW also contribute as charge carrier and reduce the solution resistance, which also provide suitable conditions for controlled utilization of organic matter and bioelectrogenesis with high power densities. In the present condition, LW is simple substrate that generates higher electrons from oxidation and PRW with high TDS assists for effective electron transfer. This way PRW and LW are complementing each other for improved efficiency of MFC for sustainable energy generation. Figure 1 Bioelectrogenic behaviour observed during MFC operation for co-substrate influence. ( a ) Current density (mA/m 2 ) during 6 different combinations of (PRW:LW) and control operations studied for co-substrate influence on bioelectrogenesis (refer to Table 1 for exp. conditions for C1 to C21), ( b ) potentials and power density registered during the 3 cycles of each experimental variation. Table 1 Consolidated results from the bioelectrochemical treatment of petroleum refinery wastewater (PRW) and Labaneh whey (LW) as co-substrate. Experiment No PRW (%) LW (%) Inlet COD (mg/L) HRT (Days) Outlet COD (mg/L) COD degradation rate (kg COD/m 3 -day) COD removal efficiency (ξCOD, %) Outlet pH a Voltage (mV) PD (mW/m 2 ) SPY (W/Kg COD R ) Cell design point (Ω) C1-C3 100 00 2150 5 1579 0.114 26.54 7.31 410 420 2.95 200 C4-C6 95 05 3010 5 1873 0.227 37.77 7.18 484 587 2.06 200 C7-C9 90 10 3742 6 2071 0.278 44.65 6.82 530 702 1.68 100 C10-C12 85 15 4475 6 2153 0.387 51.90 6.63 577 832 1.43 100 C13-C15 80 20 5328 8 1966 0.420 63.10 6.34 497 618 0.74 100 C16-C18 75 25 6226 6 3321 0.415 46.66 5.60 438 479 0.66 100 C19-C21 70 30 7235 4 5944 0.323 17.84 4.20 330 274 0.86 300 All the values presented here are average of 3 cycles. HRT, hydraulic retention time; VPD, volumetric power density; SPY, specific power yield; PD—power density. a Inlet pH (7.0) was maintained constant in all the experiments. Further, increasing the LW concentration in PRW was evaluated in different ratios (PRW: LW). An improvement in power generation was recorded up to 85:15 ratio (577 mV, 1441 mA/m 2 ). Further increase in LW fractions, a gradual drop in the power generation was recorded. However, the bioelectrogenesis was found to be higher than PRW as the sole carbon source (80:20–497 mV, CD, 1242 mA/m 2 ; 75:25–438 mV, CD, 1094 mA/m 2 ) (Fig. 1 ). In the next substrate loading condition of 70:30, power generation was found to drop significantly than the control operation (70:30–330 mV, CD, 825 mA/m 2 ), indicating that higher concentration of LW yields less energy from MFC operation. In few studies, depending on the type of wastewater used in the anode chamber, it was observed that higher concentrations of readily degradable organic matter may results in lower power generation 47 , 48 . A study on treatment of liquid fraction of municipal solid waste through bioelectrochemical process evidenced that the highest energy yields could be attained at the lowest input COD concentrations 47 . Similar study with vegetable market waste also evidenced that the high concentration of COD showed relatively lower power generation than the low COD concentrations of same the waste 30 , 48 . In the present study, LW that used as co-substrate might also showed similar effect at high concentration along with PRW. Due to this a drop in bioelectrogenesis was identified with 70:30 substrate condition. After completing the 70:30 substrate condition, the MFC was shifted to 85:15 condition to recheck if the system is resuming to previous efficiencies. It took a continuous operation of 5 cycles with 85:15 feed condition to exhibit the comparable bioelectricity generation efficiency (578 mV; CD, 835 mA/m 2 ). Similar study was also done by other research group with produced water having petroleum hydrocarbons. A preliminary study by Shrestha et al., was performed using produced water (PW) Bakken shale, USA as major substrate along with municipal sewage in a dual chamber MFC configuration for 53 days 32 . PW as the sole carbon source reported to generate 3 ± 1 mW/m 2 . Further, addition of sewage as co-substrate was resulted in several folds improvement in power generation (77 ± 4 mW/m 2 ). The nutrients present in municipal sewage likely helped for the improved performance of MFC. Addition of co-substrate was also showed to enhance the anaerobic biodegradation of polycyclic aromatic hydrocarbons (PAHs) which are one of the important components of petroleum hydrocarbons 36 , 49 , 50 . The improved power generation was attributed to co-substrate (sewage) addition, which also improved substrate degradation efficiency 32 . Co-substrate influence on substrate degradation Substrate degradation is the source of electron generation required for bioelectricity production in MFCs. In the control experiment, PRW was found to be solely contributing for the electrons and resulted in power density of 420 mW/m 2 with substrate degradation rate of 0.114 kg COD/m 3 -day (ξCOD, 26.54%, 5 days) (Fig. 2 ). As the ratio of wastewater was varied according to experimental design, LW was found to have higher COD than PRW and the resultant wastewater feed exhibited considerable improvement in COD concentration (Table 1 ). This variation also requires extending the operation time (HRT), which was fixed based on the bioelectrogenesis of that particular variation (described in later section titled, pH). To normalize the substrate degradation with time of operation and volume of the reactor, substrate degradation rate (kg COD/m 3 -day) was used as an important parameter for the evaluation. Further, MFC fed with 95% PRW; along with 5% LW showed about 100% improvement in substrate degradation rate (0.227 kg COD/m 3 -day) and registered ξCOD of 37.77% in 5 days of operation. As the ratio of LW is increasing in the feed, HRT was found to increase from 5 to 6 days with 90:10 and 85:15 conditions. In the case of 80:20 condition, maximum HRT of 8 days was maintained. As steep drop in potential was observed with 75:25 and 70:30 conditions, the HRT was limited to 6 days and 4 days, respectively. Among all the variations studied, the maximum substrate degradation rate of 0.420 kg COD/m 3 -day was registered with 80:20 operation (ξCOD, 63.10%), followed by 75:25 condition (SDR, 0.415 kg COD/m 3 -day; ξCOD, 46.66%), 85:15 condition (SDR, 0.387 kg COD/m 3 -day; ξCOD, 51.90%), 90:10 condition (SDR, 0.278 kg COD/m 3 -day; ξCOD, 44.65%) and 70:30 condition (SDR, 0.323 kg COD/m 3 -day; ξCOD, 17.84%). In waste/wastewater treatment, co-substrate or co-digestion is considered as an interesting choice to achieve higher substrate degradation efficiencies. Co-substrate in anaerobic digestion is regarded as positive synergy establishing option for improved biogas production 51 . Higher fraction of organic matter is available in the waste produced from agricultural processes and associated activities are found to be viable for co-digestion to generate energy, which additionally delivers economic and environmental benefits 51 , 52 . Experimental studies by Zhang and Lo 36 , revealed that anaerobic biodegradation of petroleum hydrocarbons in marine sediments was improved by addition of acetate and methanol as co-substrates 36 . In the present study, LW generated from dairy industry is certainly providing additional nutrients to the system and resulted in improved bioelectrochemical degradation of organics in the petroleum refinery wastewater. In previous studies with phenol as substrate and glucose as co-substrate, by Luo et al. 31 , two distinct peaks were identified for voltage generation in each cycle of operation. During the first peak, 20% phenol degradation was recorded, whereas during the second peak, phenol degradation reached 90%. Both glucose and phenol were found to degrade simultaneously during the first cycle of operation. However, glucose removal was higher during the first peak and phenol degradation was higher during the second peak 31 . In another study by Shen et al., phenol co-metabolism was found to be efficient with acetate as co-substrate compared to other four substrates studied 53 . A dual chambered MFC using industrial acid mine drainage was treated effectively with municipal wastewater as co-substrate 54 . Similar distinct observation is infeasible during the present study, due to the complex nature of wastewaters (both PRW and LW) that were used as substrate and co-substrate. Figure 2 The trend of COD removal efficiency and substrate degradation rate with respect to different (PRW:LW) ratios for bioelectricity generation. pH During the degradation of wastewater having complex molecules, simple molecules that are metabolites will be produced as a result of oxidation. The nature of these products influence the pH of the treated/outlet wastewater. pH is one of the important factors affecting MFC operation. pHs in the range of 6 to 8.5 are considered as more favorable condition for bioelectricity generation 55 – 57 . pH conditions below 5.0 and above 8.5 showed to have adverse effect on the overall performance of MFC 48 , 58 , 59 . In the case of MFC operation, the metabolites present in the wastewater change the pH of the electrolyte. In the case of PRW as the sole carbon source, the pH of the electrolyte was found to shift from neutral pH (inlet) to slightly alkaline pH (7.31) by the end of the cycle of operation (Fig. 3 ). Similarly, when 5% of LW was used as feed, the pH was slightly moved to alkaline pH and recorded as 7.18. The dairy-based wastewater including LW contains high amount of lactose sugar and mild organic acids 60 – 62 . During anodic oxidation process, the lactose sugar generates volatile fatty acids (VFAs) such as lactic, acetic, butyric and propionic acid. Due to these VFAs, the effluent generated from the 5% LW showed relatively less shift to alkaline conditions. This phenomenon was more evident when the MFC operation was conducted at higher concentrations of LW. With 90:10 ratio of PRW and LW, the effluent pH was found to exhibit shift in anolyte pH to mild acidic conditions (pH 6.82) by end of the cycle operation. Figure 3 The shift of outlet pH (end of the cycle operation) from the neutral condition at the beginning of cycle (inlet pH 7.0) at different (PRW:LW) ratios for bioelectricity generation. Further, increase in LW concentration along with PRW in the feed was resulted in pH shift towards more acidic conditions [85:15, pH-6.63 (6 days of HRT)]; [80:20, pH-6.34 (8 days HRT)]; [75:25, pH-5.60 (7 days of HRT)]. In the case of 70:30 condition, the outlet pH was found to be 4.20 (4 days of HRT). Higher drop in the pH is due to higher concentration of LW available in the anolyte, this resulted in higher production of VFAs from the oxidation. Figure 3 clearly demonstrated the gradual drop in the pH with gradual increase in the LW concentration. Acidic pH that prevailed in the anolyte also influenced the bioelectrogenesis process. At 75:25 and 70:30 conditions, the pH drop towards more acidic conditions that were associated with lower substrate degradation rate and lower power generation. Acidic pH condition decreases the performance of the anodic biofilm that is acting for the degradation of pollutants 4 , 63 . Since the acidic conditions (< pH 4.5) are unfavorable for the activity of electroactive biofilms and substrate degradation, it also resulted in drop of current density. The operating time, at which more than 20% drop in current density was registered, a new operating cycle was started with new feed. This has led to stop the batch operation intermittently. Compared to 8 days of HRT with 80:20 feed conditions, the batch operation was ended by 6 and 4 days of operation for 75:25 and 70:30 conditions respectively. Bioelectrochemical evaluation Polarization behaviour of MFC during the 6 differnet concentrations of co-substrate (LW) along with PRW was evaluated and compared with the control operation. It was analyzed by recording the voltage and discharge current at a range of external resistances (50 Ω to 30 kΩ) 64 . To achieve stable performance and to avoid stress in the MFC operation, polarization behavior was recorded in the final operating cycle of each experimental variation (Fig. 4 a). Electron discharge in MFCs is inversely proportional to the external resistance used in the closed circuit. At higher external resistance, electron discharge is neglegible, due to which, lower current density and higher voltage will be recorded. Similarly, at lower resistance in the circuit, higher current density and low volatages were identified 65 , 66 . In the present study, 100% PRW case showed maximum current density (CD) and maximum volumetric power density (VPD Max ) of 1225 mA/m 2 and 4.97 W/m 3 (at 200 Ω resistance), respectively (Fig. 4 b, Table 2 ). Cell design point (CDP) is determined as the resistance point at which maximum volumetric power density (VPD Max ) is registered. In the case of MFC operation with 100% PRW, it can be noticed as 200 Ω. When 5% LW was added to PRW, higher performance was registered and VPD Max improved to 6.30 W/m 3 (200 Ω). This indicates the positive role of LW in improving stable electron discharge function of MFC. Figure 4 ( a ) Polarization behaviour at different co-substrate conditions evaluated at different (PRW:LW) ratios in MFC. ( b ) Maximum volumetric power density and cell design point recorded. Table 2 Volumetric power density (VPD Max ) and cell design point (CDP) observed from polarization behavior recorded from the six different co-substrate combinations studied. Experiment No PRW (%) LW (%) VPD Max (W/m 3 ) Cell design point (Ω) C1–C3 100 00 4.97 200 C4–C6 95 05 6.30 200 C7–C9 90 10 6.98 100 C10–C12 85 15 7.20 100 C13–C15 80 20 5.95 100 C16–C18 75 25 4.05 100 C19–C21 70 30 1.86 300 Similar to the power generation efficinecy, VPD Max was also improved with increase in LW concentration up to 85:15 feed condition. Maximum VPD Max of 7.20 W/m 3 was registered with 85:15 feed condition followed by 90:10 feed condition (6.98 W/m 3 ), 95:5 feed condition (6.30 Table 2 ). Among all variations evaluated, the minimum VPD Max was registered with 70:30 (1.86 W/m 3 ) and 75:25 variations (4.05 W/m 3 ). When CDP was compared among the different LW additions evaluated, it was shifted to lower resistances with increase in LW concentration. In the case of 100% PRW and 95:5 condition, CDP was observed at 200 Ω, which later dropped to 100 Ω (90:10, 85:15, 80:20 and 75:25). Improvement in power density along with shifting of CDP to lower resistances demonstrated the improved degradability of the substrate in the anode. On the contrary, 70:30 feed condition showed lowest VPD Max at CDP of 300 Ω. This also correlating well with the substrate degradation and power generation. Specific and volumetric power production As the major objective of MFCs is applied towards developing a unit operation for sustainable wastewater treatment along with power generation, it is required to evaluate the system efficiency with respect to practical parameters for large-scale applications. Specific power yield (SPY, W/kg COD R ) was calculated by normalizing the power generated to the amount of COD degradation at different concentrations of co-substrate added to PRW (Fig. 5 ). A maximum SPY of 2.95 W/kg COD R was registered with PRW alone as the substrate that is higher than the SPY produced from all the co-substrate addition experiments. Minimum SPY was registered with 75:25 feed condition (0.66 W/kg COD R ). Among the control and co-substrate addition conditions studied, the improvement in power generation was not directly correlated with the amount of COD degradation, which is the major factor for showing higher SPY with PRW alone. In MFCs, this was identified as one of the limitations. However, more studies are needed to optimize the effective ratio of co-substrate in relation to electrode surface area and volume of the anode chamber of MFC. The volumetric power density is derived as the maximum power generated per unit anode volume. A maximum volumetric power density of 9.51 W/m 3 was registered with 85:15 feed condition, which is two times higher than the PRW as sole substrate (Fig. 5 ). Similar to power generation, 70:30 feed conditions registered minimum volumetric power density of 3.13 W/m 3 . The results obtained were in good agreement with power generation and substrate degradation observed from the experimental study. Figure 5 ( a ) Cycle wise performance of MFC with respect to specific power yield and volumetric power density, ( b ) consolidated representation of specific power yield and volumetric power density during operation at 6 different substrate combinations and control operations. Normalized energy recovery 67 (NER, kWh/kg COD removed) or energy yield that evaluated with respect to total power generated in individual cycle of operation in relation to total COD degraded/removed in the respective cycle was provided different insights (Fig. 6 ). The highest NER was registered with 100 PRW as substrate (C1, 1.24 kWh/kg COD removed). In the case of first co-substrate addition (95:5 condition), 0.89 kWh/kg COD removed was registered. Further increase in LW concentration in PRW resulted in drop of NER. This phenomenon was found contrary to the power generation identified across all co-substrate variations studied. The minimum NER of 0.24 kWh/kg COD removed was registered with 70:30 condition (Cycle 20). Figure 6 Energy yield evaluated in relation to power produced and COD removed during all cycles of operation. In the present study, it was observed that bioelectrochemical degradation efficiency of petroleum refinery wastewater (PRW), a highly recalcitrant wastewater was improved by the addition of LW wastewater, an organic rich wastewater as co-substrate, which eventually also effectively improved bioelectricity generation. The optimum concentration of LW as co-substrate with PRW was evaluated under optimal MFC operating conditions at several combination ratios of the two wastewaters. The maximum power generation (current density, 1441 mA/m 2 ; power density, 832 mW/m 2 ) were achieved with 85:15 combination of PRW and LW as feed. On the contrary, 80:20 ratio resulted in higher substrate degradation rate (0.420 kg COD/m 3 -day) with ξCOD of 63.1%. The function of bioelectricity generation and substrate degradation were mainly limited by the electrolyte pH. Higher LW concentration resulted in highly acidic pH that hampered both power generation and substrate degradation. The maximum volumetric power yield with co-substrate addition was 9.51 W/m 3 , which is two times higher than PRW as sole substrate. This study paves the way for utilizing such combinations of different types of wastewaters with varying composition to increase the biodegradability of one due to the stimulatory effect of the other. Even though such combinations have been very effectively used in traditional bioremediation studies, this study shows that it is also effective in a bioelectroremediation using MFC. Other similar or diverse combinations of wastewaters should be explored to make this a generic practice in this field."
} | 6,891 |
37039578 | PMC10141414 | pmc | 4,475 | {
"abstract": "Methods for fabricating\nsuper-liquid-repellent surfaces\nhave typically\nrelied on perfluoroalkyl substances. However, growing concerns about\nthe environmental and health effects of perfluorinated compounds have\ncaused increased interest in fluorine-free alternatives. Polydimethylsiloxane\n(PDMS) is most promising. In contrast to fluorinated surfaces, PDMS-coated\nsurfaces showed only superhydrophobicity. This raises the question\nwhether the poor liquid repellency is caused by PDMS interacting with\nthe probe liquid or whether it results from inappropriate surface\nmorphology. Here, we demonstrate that a well-designed two-tier structure\nconsisting of silicon dioxide nanoparticles combined with surface-tethered\nPDMS chains allows super-liquid-repellency toward a range of low surface\ntension liquids. Drops of water–ethanol solutions with surface\ntensions as low as 31.0 mN m –1 easily roll and bounce\noff optimized surface structures. Friction force measurements demonstrate\nexcellent surface homogeneity and easy mobility of drops. Our work\nshows that fluorine-free super-liquid-repellent surfaces can be achieved\nusing scalable fabrication methods and environmentally friendly surface\nfunctionalization."
} | 302 |
26896131 | PMC4959497 | pmc | 4,476 | {
"abstract": "ABSTRACT A chemolithoautotrophic strain of the family Beggiatoaceae , Beggiatoa sp. strain 35Flor, was found to oxidize molecular hydrogen when grown in a medium with diffusional gradients of oxygen, sulfide, and hydrogen. Microsensor profiles and rate measurements suggested that the strain oxidized hydrogen aerobically when oxygen was available, while hydrogen consumption under anoxic conditions was presumably driven by sulfur respiration. Beggiatoa sp. 35Flor reached significantly higher biomass in hydrogen-supplemented oxygen-sulfide gradient media, but hydrogen did not support growth of the strain in the absence of reduced sulfur compounds. Nevertheless, hydrogen oxidation can provide Beggiatoa sp. 35Flor with energy for maintenance and assimilatory purposes and may support the disposal of internally stored sulfur to prevent physical damage resulting from excessive sulfur accumulation. Our knowledge about the exposure of natural populations of Beggiatoaceae to hydrogen is very limited, but significant amounts of hydrogen could be provided by nitrogen fixation, fermentation, and geochemical processes in several of their typical habitats such as photosynthetic microbial mats and submarine sites of hydrothermal fluid flow. IMPORTANCE Reduced sulfur compounds are certainly the main electron donors for chemolithoautotrophic Beggiatoaceae , but the traditional focus on this topic has left other possible inorganic electron donors largely unexplored. In this paper, we provide evidence that hydrogen oxidation has the potential to strengthen the ecophysiological plasticity of Beggiatoaceae in several ways. Moreover, we show that hydrogen oxidation by members of this family can significantly influence biogeochemical gradients and therefore should be considered in environmental studies.",
"introduction": "INTRODUCTION Members of the family Beggiatoaceae are colorless sulfur bacteria known to oxidize reduced sulfur compounds and organic substances for chemolithoautotrophic, chemoorganoheterotrophic, and mixotrophic growth ( 1 ). The use of various organic substances, such as mono- and dicarboxylic acids, sugars, amino acids, and alcohols, has been studied repeatedly in different strains of the family ( 2 – 6 ), but inorganic electron donors other than reduced sulfur compounds were never reported to support growth. The only indication of the oxidation of a nonsulfuric, inorganic electron donor was the stimulation of sulfur reduction by molecular hydrogen in a microaerophilic Beggiatoa strain under short-term anoxic conditions ( 7 ). Hydrogen oxidation or hydrogen-supported growth has been reported for many other well-known sulfur oxidizers such as members of the families Chromatiaceae ( 8 ), Acidithiobacillaceae ( 9 , 10 ), Aquificaceae ( 11 – 13 ), and Sulfolobaceae ( 14 ), the genus Sulfurimonas ( 15 , 16 ), the SUP05 clade ( 17 ), and endosymbionts of mussels ( 18 ). This suggests that hydrogen oxidation may be a widespread metabolic trait among sulfur oxidizers and as such may also be realized in the family Beggiatoaceae . Substantial amounts of molecular hydrogen are produced and consumed in many microbial habitats, so H 2 is considered to be an important electron transfer agent in oxic and anoxic environments ( 19 ). Nevertheless, there is little information about the environmental exposure of Beggiatoaceae populations to hydrogen and the potential importance of hydrogen oxidation for members of the family in situ . Despite high conversion rates, in situ studies on hydrogen cycling and availability are difficult due to the generally very low ambient concentrations ( 20 ). Steep biogeochemical gradients, which are typical for habitats of Beggiatoaceae , pose an additional problem because these necessitate a sampling resolution on the micrometer scale for meaningful conclusions. Microsensors are typically used for this purpose, and a microsensor for hydrogen has been available for more than two decades ( 21 ). However, the hydrogen microsensor has the critical disadvantage of being sensitive to hydrogen sulfide ( 22 ). This cross-reactivity disqualifies the sensor from many in situ applications, in particular, from measurements in habitats of sulfur bacteria, where the concentrations of sulfide are usually considerably higher than those of hydrogen. In the present study, we investigated the consumption of molecular hydrogen in cultures of a chemolithoautotrophic Beggiatoa strain using microsensors. Culture-based experiments allowed us to adjust the concentrations of hydrogen and sulfide to levels at which reliable measurements with the hydrogen microsensor are possible. We discuss here how hydrogen oxidation can contribute to the ecophysiological plasticity of the strain and point out environmental settings in which members of the family Beggiatoaceae may be able to use hydrogen as an electron donor and energy source.",
"discussion": "DISCUSSION We showed that a chemolithoautotrophic strain of the family Beggiatoaceae , Beggiatoa sp. 35Flor, consumed molecular hydrogen at the oxygen-sulfide interface. Microsensor profiles and rate measurements suggested that the strain oxidized hydrogen aerobically. With 5 to 17 nmol H 2 per μg protein and hour or 417 to 523 nmol H 2 per cm 3 mat volume and hour (7 to 21 days; see Fig. S5 in the supplemental material), the average hydrogen oxidation rates were substantial and in fact exceeded the sulfide oxidation rates at all times ( Fig. 4 ). Hydrogen is a valuable electron donor for Beggiatoa sp. 35Flor, as illustrated by the significantly higher protein content in hydrogen-supplemented cultures ( Fig. 5 ). Similarly to other members of the family Beggiatoaceae ( 24 , 35 ), Beggiatoa sp. 35Flor is capable of nitrogen fixation (Henze, unpublished). Because this process releases hydrogen as a byproduct ( 28 ), many diazotrophs couple the expression of nitrogenase to the expression of uptake hydrogenases on a transcriptional level ( 36 – 39 ). Hydrogen oxidation occurring under conditions of repression of nitrogen fixation (see Fig. S1 in the supplemental material) showed, however, that Beggiatoa sp. 35Flor does not merely recycle internally produced hydrogen but is able to use externally supplied hydrogen as a genuine electron donor. Beggiatoa sp. 35Flor grew in a defined coculture with Pseudovibrio sp. FO-BEG1, but several lines of evidence suggest that the Pseudovibrio strain did not contribute to the consumption of hydrogen. We did not observe hydrogen oxidation in gradient cultures that contained only Pseudovibrio sp. FO-BEG1 (see Fig. S2 in the supplemental material), and hydrogen oxidation was never observed in liquid cultures of the strain, irrespective of the incubation conditions tested (V. Bondarev, unpublished data). In addition, hydrogenase genes could not be identified in the essentially closed genome of Pseudovibrio sp. FO-BEG1 (Bondarev, unpublished). Hydrogen oxidation clearly influenced the mat position, oxygen consumption, and growth of Beggiatoa sp. 35Flor. This is of particular importance for environmental studies, because it illustrates that the measurement of oxygen and sulfide gradients alone does not necessarily suffice to gain a comprehensive picture of Beggiatoaceae metabolism. In contrast, the use of alternative electron donors such as hydrogen or electron acceptors such as nitrate ( 40 – 44 ) can significantly influence biogeochemical gradients as well as the position of Beggiatoaceae populations with respect to these. Hydrogen versus sulfur as an electron donor at the oxygen-sulfide interface. In order to assess the influence of hydrogen oxidation on the electron turnover in Beggiatoa sp. 35Flor, electron budgets were calculated on the basis of the measured consumption rates of oxygen, sulfide, and hydrogen as well as the estimated CO 2 fixation rates ( Fig. 7 A). In hydrogen-unsupplemented cultures, the average contribution of sulfide oxidation to the total electron supply decreased from 36% to 20% within the first 3 weeks of incubation. The absolute rates of sulfide oxidation were similar in hydrogen-supplemented cultures, but the relative contribution to the total electron supply was lower because hydrogen-supplemented cultures showed an overall higher electron demand. Within the first 3 weeks, the average contribution of sulfide oxidation decreased from 30% to 11% of the total electron supply in hydrogen-supplemented cultures and dropped to only 6% after 4 weeks. Concurrently, the average contribution of hydrogen oxidation to the total electron supply increased from 36% after 1 week to 102% after 4 weeks. Hydrogen oxidation was already the main electron-supplying reaction after 2 weeks, fulfilling on average 59% of the total electron demand. Other electron donors and CO 2 fixation as an electron sink were insignificant after 4 weeks such that hydrogen oxidation explained the total oxygen demand according to the Knallgas reaction (2H 2 + O 2 → 2H 2 O). FIG 7 Influence of hydrogen oxidation on electron turnover and pH in Beggiatoa sp. 35Flor cultures. (A) Electron budgets in hydrogen-supplemented (+) and -unsupplemented (–) cultures over 4 weeks. The total electron demand (shown in red) was calculated based on the measured oxygen consumption rate and the estimated rate of CO 2 fixation into biomass (<CH 2 O>). Weekly averages of CO 2 fixation rates were estimated on the basis of the increase in Beggiatoa protein levels and a cell carbon-to-protein ratio of 1.13 (wt/wt; determined for closely related strain Beggiatoa sp. MS-81-6 under similar growth conditions; [ 31 ]). According to this estimation, CO 2 fixation accounted for ≤6.4% of the total electron demand at all times. Hatched areas indicate the contribution of sulfide oxidation (2 electrons per H 2 S → S 0 ) to the electron supply; dotted areas represent the contribution of hydrogen oxidation (2 electrons per H 2 ). The electron demand, which cannot be fulfilled by the reactions described above, is most likely met by the oxidation of sulfur to sulfate (6 electrons per S 0 ). (B) Average pH profiles (± standard deviation; n = 6) measured in hydrogen-supplemented (black) and -unsupplemented (white) cultures after 7 days of incubation. Mats in hydrogen-supplemented cultures were situated 5.2 to 6.2 mm below the air-agar interface; mats in hydrogen-unsupplemented cultures were located at a depth of 6.6 to 7.5 mm. The total electron demand in hydrogen-supplemented and -unsupplemented cultures during the first 3 weeks was higher than what could be supplied by the oxidation of sulfide and hydrogen alone ( Fig. 7A ). This excess demand was most likely fulfilled by the oxidation of elemental sulfur to sulfuric acid as sulfur inclusions disappeared over time ( Fig. 3 ), and pH profiles showed a pronounced acidification in the region of the mat ( Fig. 7B ). Sulfur oxidation in oxygen-sulfide gradient cultures of Beggiatoa sp. 35Flor was recently also demonstrated by the production of large amounts of sulfate ( 45 ). Notably, the excess electron demand was lower in hydrogen-supplemented cultures throughout the incubation ( Fig. 7A ). Together with a less pronounced acidification of the medium ( Fig. 7B ), this suggests that less sulfur was oxidized to sulfuric acid in the presence of hydrogen. In addition to the production of sulfuric acid, higher CO 2 fixation rates ( Fig. 5 ) contribute to higher pH values in hydrogen-supplemented cultures. However, it is unlikely that the observed pH difference resulted mainly from differences in CO 2 fixation rates, because the estimated contribution of CO 2 fixation to the total electron demand was low in general (≤6.4%; Fig. 7 ). Overall, the influence of hydrogen oxidation on the sulfur metabolism of Beggiatoa sp. 35Flor points to a very efficient and purposeful use of the different electron donors in an environment, in which sulfide toxicity, competition for resources, and fluctuating supplies with oxidants and reductants are the major challenges. Sulfide and hydrogen, which cannot be stored, are oxidized immediately when available, while sulfur may be kept in reserve when the current energy requirements can be met by using other electron donors. Aerobic hydrogen oxidation occurs in the presence of reduced sulfur compounds. The presented results clearly show that Beggiatoa sp. 35Flor used energy from aerobic hydrogen oxidation for growth when reduced sulfur compounds were available. In contrast, growth on hydrogen in the absence of reduced sulfur compounds could not be shown. The apparent inability of hydrogen to support growth as an exclusive electron donor was unexpected, given that electrons from hydrogen ( 46 , 47 ) enter the electron transport chain either on the same level as or upstream of electrons from reduced sulfur compounds ( 48 – 50 ) and thus should be able to support at least the same metabolic processes. So far, the reason for absent or discontinued growth of Beggiatoa sp. 35Flor on hydrogen and oxygen alone is unclear. Possible explanations are the potential inability to assimilate sulfate, the accumulation of waste products in older cultures, or the missing abiotic oxygen removal by sulfide and the resulting lack of a microoxic niche in fresh sulfide-free gradient media. Beggiatoa sp. 35Flor oxidizes hydrogen also under anoxic conditions, presumably through sulfur respiration. Beggiatoa sp. 35Flor filaments did not oxidize hydrogen only aerobically in mats at the oxygen-sulfide interface ( Fig. 3 ). In cultures with a high sulfide flux, hydrogen was also oxidized in the fully anoxic section of the medium by a subpopulation of filaments that had migrated downward from the oxygen-sulfide interface ( Fig. 6 ). Several members of the family Beggiatoaceae are known to store nitrate in large amounts and use it as an alternative electron acceptor under anoxic conditions (e.g., 40 , 43 , 44 , 51 ). However, nitrate can be excluded as an electron acceptor in the present study. Gradient media for precultures and experiments were prepared without fixed nitrogen compounds, and Beggiatoa sp. 35Flor filaments from such precultures were previously shown to be free of NO X compounds ( 25 ). Hence, neither external nor internal nitrate was available for hydrogen oxidation under anoxic conditions. In addition to the use of nitrate, several members of the family Beggiatoaceae are known to use stored sulfur as an electron acceptor under short-term anoxic conditions. Previous studies showed that sulfur respiration in Beggiatoaceae can be supported by organic electron donors such as acetate ( 52 ) and internally stored polyhydroxyalkanoates ( 7 , 25 ) but also by molecular hydrogen ( 7 ). Sulfur respiration was recently shown in the strain Beggiatoa sp. 35Flor by Schwedt et al. ( 25 ) under incubation conditions very similar to the ones used here. Schwedt and colleagues showed that Beggiatoa sp. 35Flor filaments that had migrated into the anoxic section of a gradient medium under high sulfide fluxes reduced stored sulfur with stored polyhydroxyalkanoates. We assume that the same population of filaments as was studied by Schwedt et al. ( 25 ) used sulfur also as an electron acceptor for hydrogen oxidation under anoxic conditions in our experiments. Total sulfide profiles recorded in hydrogen-supplemented and -unsupplemented cultures did, however, not show significantly higher sulfide concentrations in the region of the anoxic subpopulation when hydrogen was present (see Fig. S3 in the supplemental material). This may have been due to the fact that the increase in sulfide production through hydrogen oxidation was too low compared to the background sulfide flux and the variability among replicate cultures. The average expected sulfide production rate, which is equal to the average measured hydrogen consumption rate (H 2 + S 0 → H 2 S; 1.16 × 10 −3 nmol cm −2 s −1 ), was only 11% of the background sulfide flux (10.50 × 10 −3 nmol cm −2 s −1 ) in hydrogen-supplemented cultures. The standard deviation of total sulfide concentrations was 7% to 18% ( n = 3; hydrogen-supplemented cultures) and 9% to 23% ( n = 3; hydrogen-unsupplemented cultures) of the average values in the region of the anoxic subpopulation located at a depth of 8 to 12 mm. For this reason, a significant increase in sulfide production through hydrogen oxidation may not have been detectable. Sulfur respiration in Beggiatoaceae has been suggested to serve two purposes: the generation of metabolic energy under short-term anoxic conditions ( 7 , 52 ) and the disposal of excess internal sulfur to avoid cell rupture ( 25 ). The use of hydrogen as an electron donor would enable an uncoupling of sulfur respiration from the oxidation of organic carbon compounds, thus leading to a higher flexibility in energy generation and sulfur disposal under anoxic conditions. Thus, hydrogen oxidation has the potential of increasing the ecophysiological plasticity of Beggiatoa sp. 35Flor and possibly of other members of the family Beggiatoaceae in two ways, both of which are tightly coupled to the sulfur metabolism. In the presence of a low sulfide flux and electron acceptors with a more positive redox potential such as oxygen, hydrogen can partially replace sulfur as an electron donor and thereby increase the amount of sulfur available for storage. In contrast, hydrogen may support sulfur respiration and disposal under conditions of high sulfide flux and anoxia in order to provide metabolic energy and prevent physical damage from excessive sulfur accumulation. Environmental relevance of hydrogen oxidation for members of the family Beggiatoaceae . A variety of biotic and abiotic environmental processes are associated with the production of molecular hydrogen ( 19 ). Nevertheless, significant amounts of hydrogen are probably available to Beggiatoaceae in only certain environments. Members of this family are very often found in organic-rich sediments, in which hydrogen is produced by fermentation. However, the preferred habitat of Beggiatoaceae , the oxygen-sulfide interface, is usually well and permanently separated from the zone of hydrogen production in these sediments. Even though large quantities of hydrogen are produced by fermentative processes in deeper, anoxic layers, H 2 is rapidly and efficiently reoxidized by the local community of hydrogenotrophic prokaryotes ( 20 ). Beggiatoaceae , which populate the oxygen-sulfide interface, are therefore unlikely to experience high concentrations or fluxes of hydrogen in such systems. In contrast, nitrate- or sulfur-respiring members of the family, which are residing in or traveling through fermenting sediment layers, could exploit hydrogen as an electron donor. The hypersaline cyanobacterial mats of the Guerrero Negro evaporation lagoons (Baja California Sur, Mexico) are a prominent example of an environment in which large amounts of hydrogen are frequently available to members of the Beggiatoaceae . The biogeochemical conditions in these mats follow a strong diel cycle ( 53 – 56 ), which involves the presence of exceptionally high hydrogen concentrations at the mat surface during nighttime ( 57 ). Reacting to the changing biogeochemical conditions, filamentous Beggiatoaceae migrate to the anoxic and sulfidic surface of the Guerrero Negro mats at night ( 58 , 59 ) and thus are regularly exposed to high hydrogen concentrations. Extensive cyanobacterial mats resembling those of the Guerrero Negro lagoons were present on earth for most of life's history, once dominating the biosphere ( 53 , 60 , 61 ). Substantial genetic exchange between cyanobacteria and Beggiatoaceae ( 62 , 63 ) strikingly evidences a historically frequent co-occurrence of these taxa. This suggests that hydrogen transfer from nitrogen-fixing and fermenting cyanobacteria to members of the family Beggiatoaceae could indeed be an ancient and once-widespread process. In addition, chemosynthetic ecosystems in the deep sea are sites at which hydrogen, specifically, H 2 of geothermal origin, could potentially serve as a source of metabolic energy for Beggiatoaceae . Members of the family are regularly encountered in the deep sea at sites of hydrothermal fluid flow (e.g., 64 – 69 ), and hydrogen is extruded at several of such places ( 18 , 70 , 71 ). In fact, H 2 of geothermal origin was suggested to be a key energy source in deep-seawater masses ( 17 ) and has been shown to fuel CO 2 fixation in sulfide-oxidizing endosymbionts of deep sea mussels ( 18 ). Yet seep-dwelling populations of Beggiatoaceae have apparently never been tested for exposure to or even consumption of H 2 . Similarly to submarine sites of hydrothermal fluid flow, members of the Beggiatoaceae thrive in terrestrial sulfidic springs ( 1 , 5 , 72 , 73 ), sites at which hydrogen is frequently emitted ( 74 ). However, further studies are necessary to evaluate the importance of molecular hydrogen for members of the family Beggiatoaceae on a broader scale. These studies will need to investigate the availability of H 2 to environmental populations as well as the ability of different strains to oxidize this electron donor."
} | 5,327 |
37115261 | PMC10497666 | pmc | 4,478 | {
"abstract": "Arbuscular mycorrhizal fungi (AMF) in the roots and soil surrounding their hosts are typically independently investigated and little is known of the relationships between the communities of the two compartments. We simultaneously collected root and surrounding soil samples from Cryptomeria japonica (Cj) and Chamaecyparis obtusa (Co) at three environmentally different sites. Based on molecular and morphological analyses, we characterized their associated AMF communities. Cj was more densely colonized than Co and that root colonization intensity was significantly correlated with soil AMF diversity. The communities comprised 15 AMF genera dominated by Glomus and Paraglomus and 1443 operational taxonomic units (OTUs) of which 1067 and 1170 were in roots and soil, respectively. AMF communities were significantly different among sites, and the root AMF communities were significantly different from those of soil at each site. The root and soil AMF communities responded differently to soil pH. At the genus level, Glomus and Acaulospora were abundant in roots while Paraglomus and Redeckera were abundant in soil. Our findings suggest that AMF colonizing roots are protected from environmental stresses in soil. However, the root-soil-abundant taxa have adapted to both environments and represent a model AMF symbiont. This evidence of strategic exploitation of the rhizosphere by AMF supports prior hypotheses and provides insights into community ecology. Supplementary Information The online version contains supplementary material available at 10.1007/s00248-023-02223-9.",
"conclusion": "Conclusion In this study, we validated the hypothesis of strategic exploration of the rhizosphere by AMF and described the associations in the AMF community of roots and the surrounding soil. Root and soil AMF communities responded differently to environmental factors, suggesting that soil AMF taxa directly reflect the physical condition of the soil, whereas root AMF taxa are selected and protected by the host. This strategic root versus soil association pattern in the AMF community may sustain the mutual benefits to host and symbionts. Also, host plants may collaborate and share an AMF community via proximal networks, but this disappears upon geographical separation.",
"introduction": "Introduction Arbuscular mycorrhizal fungi (AMF) are ubiquitous symbiotic microorganisms that live in both the soil and in roots of their hosts upon which they bestow diverse benefits [ 1 , 2 ]. AMF are a monophyletic group of fungi in the Glomeromycota or Glomeromycotina [ 3 , 4 ]. These fungi have wide host ranges and are obligate plant symbionts [ 5 ], which hampers investigation of their community ecology. The development of high-throughput sequencing tools has made studies of plant-microbe interactions possible without the need for culture [ 2 , 6 ]. AMF communities and species richness may be similar or dissimilar between the roots and surrounding soil [ 7 ]. Different AMF communities in roots and surrounding soil may be a result of differences in, for example, strategic intraradical versus extraradical biomass allocation, sampling season, site conditions, host species, and biological material (spore or hyphae) [ 2 , 8 ]. Paired root-soil paired samples of host plants collected from natural ecosystems and characterization of the associated AMF communities would provide insights into ecological patterns [ 2 ]. Such an approach may also shed light on fungal colonization strategies. Among the few studies that compared AMF community composition between roots and surrounding soil, only those by Faghihinia et al. [ 9 ], Ji et al. [ 7 ], and Djotan et al. [ 10 ] were based on Illumina’s next-generation amplicon sequencing (NGS). Also, except for the woody host plants Camellia japonica [ 11 ], Juglans mandshurica [ 7 ], and Cryptomeria japonica (Japanese cedar) [ 10 ], most studies focused on annual or perennial herbs. Such studies were carried out at local scales and only one provided evidence that the intraradical AMF community originated from the roots of host plant species ( Cryptomeria japonica ) [ 10 ]. Many AMF exhibit host specificity and some host plants select AMF from an AMF pool in soil [ 12 ]. AMF are obligate symbionts, and intra- and extraradical AMF communities are typically distinct. However, plants preferentially supply photosynthate to AMF taxa that deliver the most phosphorus [ 13 ]. The structure and composition of the root-soil AMF communities that maintain the mutually beneficial associations between hosts and symbionts remain to be characterized. In this study, we performed plant barcoding and NGS-based metabarcoding of fungal DNA from two related, co-planted, and important forest tree species in Japan. We hypothesized that any differences between the root and soil AMF communities of host plants are related to AMF taxon-based colonization strategies [ 14 ]. To test this hypothesis, we collected paired root and soil samples at three different sites with different environmental conditions, molecularly confirmed root identity, and morphologically analyzed root colonization. Next, we used NGS to characterize and analyze the composition and structure of the AMF communities in and between the roots and surrounding soil. \n Cryptomeria japonica (Sugi or Japanese cedar, Cj) and Chamaecyparis obtusa (Hinoki or Japanese cypress, Co), which belong to Cupressaceae, were used as host tree species. They are both planted throughout Japan and their planted area is about 7 million hectares, constituting 69% of the total artificial forests in the country [ 15 ]. They occur naturally in warm to cool temperate regions of Honshu, Kyushu, and Shikoku Islands [ 16 ]. Morphotypes of arbuscular mycorrhiza (AM) have been reported in Cj and Co [ 17 ] and the AMF colonization rate of Cj root has been assessed [ 18 ]. However, no study has assessed Co root colonization. Furthermore, few studies such as those by Zou et al. [ 19 ], Matsuda et al. [ 20 ], and Djotan et al. [ 10 ] have investigated the AMF communities associated with Cj. To our knowledge, no study has compared the AMF community of the roots and surrounding soil of Cj and Co.",
"discussion": "Discussion Cupressaceaous conifers, which have AMF, have been poorly investigated for their mycorrhizal partners. Before this study, no quantitative assessment of AMF colonization of Co roots had been conducted, unlike Cj. Because the formation of arbuscules, hypha, and vesicles differs among AMF species [ 29 ], and these components play different roles in symbiosis [ 30 ], information on how each morphological type colonizes the roots of tree species is crucial to understanding the ecophysiology of AMF colonization. Our results indicated that whether planted separately or together, Cj and Co are differently colonized by AMF. Soil conditions, mainly pH, play a crucial role in AMF symbiosis [ 31 ]. The soil pH and EC were significantly different between Cj and Co in this study. These results could explain the differences between Cj and Co in terms of AMF root colonization. The root AMF species richness had a significant correlation with AC and that of the soil with AC and HC, suggesting that the AMF inoculum in soil determines AMF root colonization (Table S 12 ). The composition and structure of the intraradical AMF communities of Cj and Co differed significantly from those in the surrounding soil (Fig. 2 ). These results are consistent with most previous findings [ 7 , 9 , 32 , 33 ]. By contrast, the non-significant difference reported by Djotan et al. [ 10 ] between the root and surrounding soil AMF communities associated with Cj may be a result of the small sample size and/or sampling season. The AMF communities in roots and corresponding surrounding soil can be affected by methodological differences [ 8 ]. Here, root and surrounding soil samples were collected simultaneously under the same trees in different physical environments (Table S 1 ). Also, obtaining DNA from root and soil samples overcomes the imperfect proxy problem raised by Stevens et al. [ 34 ]. Thus, the difference between the root and soil AMF communities could be attributed to a strategic root-soil exploration and biomass allocation in AMF [ 14 ], as well as the selection of AMF inocula in soil by their hosts [ 12 ]. AMF colonizing roots appear to be protected from environmental stresses present in soil. This assumption is supported by the Mantel test results which showed that soil pH and geographical separation have stronger effects on soil than the root AMF community (Fig. 2 , Table S 9 ). Selection and protection by the host explain the more homogenous AMF community in the root than soil across sites (Fig. 2 ), and why AMF communities reflect local environmental conditions and spatial distance between sites [ 35 ]. Our result is consistent with the report of Stevens et al. [ 34 ] that root and soil AMF communities respond differently to environmental factors. In addition, the variation in soil AMF community does not necessarily induce variation in the related root AMF community (Table S 9 ). Therefore, the host plants act as biotic (selection and physiological influence) and abiotic (physical protection against direct effects of environmental factors) filters and alter the AMF community composition between the soil and the root. Fig. 2 Multidimensional scaling plots of the intra- and extraradical communities of arbuscular mycorrhizal fungi (AMF) associated with Cryptomeria japonica (Cj) and Chamaecyparis obtusa (Co) collected from three sites in Japan. a and b Sample groupings by compartment (root and soil) at the OTU and genus levels, respectively. Glomus and Acaulospora were significantly associated with roots whereas soil was significantly associated with Paraglomus and Redeckera (Table 5 ). c and d The effects of soil pH and geographical separation on the root and soil AMF communities, respectively. Geographical separation significantly affected the soil, but not the root AMF community; pH correlated significantly with both communities but had a stronger effect on the soil than the root community (Table S 9 ) We detected three classes of AMF OTUs or genera using CLAM, the root explorers (more abundant in roots than in soil), the soil explorers (more abundant in soil than in roots), and the explorers of both, thereby validating the hypothesis of strategic taxon-based colonization in the AMF community (Online Resource 4 , Table S 8 ). The root versus soil fungal exploration patterns, which suggest a topological connection between root and soil, may sustain the mutual benefits to the host and symbionts. Glomus and Acaulospora were significantly associated with the roots, while Paraglomus and Redeckera were significantly associated with the soil (Fig. 2 , Table 5 ). These results are consistent with a report that different AMF taxa are differently distributed in the root and soil during their life history [ 36 ]. Glomeraceae and Glomus first infest and colonize roots, where they rapidly become the most abundant AMF symbionts [ 7 , 14 ], whereas Paraglomeraceae and Paraglomus are reportedly more abundant in soil [ 7 , 37 ]. In this study, there were more AMF OTUs in the surrounding soil than in the roots (Online Resource 1 ). However, other studies reported different AMF OTU richness values and community similarities between roots and surrounding soil [ 7 ]. This discrepancy can be explained by the use of different hosts, sites, seasons, AMF quantification proxies, and overall approaches [ 2 , 4 , 8 ], which varied among prior studies but were controlled in this work. In previous studies of the AMF communities of Cj and Co [ 19 , 20 , 38 ], root and soil OTU richness were not both evaluated, thus precluding comparison of intra- and extraradical AMF communities. In Venn diagrams, the number of AMF OTUs exclusive to the roots of Cj or Co decreased when data from all sites were considered (Online Resource 2 ). This indicates spatial OTU turnover in the intraradical AMF community of Cj and Co and supports the spatiotemporal hypothesis of AMF community dynamics [ 39 , 40 ]. The lower Shannon index values in UTTF than UTCBF and UTCF (Table 2 ) support the unification of island biogeography and niche theories [ 41 ]. The AMF community was significantly different between sites and hosts (Table S 4 ). The significant differences in AMF communities among sites could be explained by differences in site-related factors and variables (Table 1 and Table S 1 ). Similar variations were reported for secondary forests and Co plantations in Japan [ 38 ]. They found that the plant community composition affected the AMF community composition, which also varied between sites. We also detected differences in the understory plant communities among sites, which supports their conclusion. In contrast, Matsuda et al. [ 20 ] found no variation among sites in the AMF communities in Cj roots. The size of the amplicon used by Matsuda et al. [ 20 ] to characterize the AMF community was smaller than in this study, which probably failed to capture the variation in molecular diversity of the AMF community associated with Cj between their study sites. The host effect was significant only in UTTF, where Cj and Co plantations were adjacent and physically separated (Table 1 , and Tables S 1 and S 6 ). These results suggest that Cj and Co may be involved in a mycorrhizal network in which they share AMF symbionts when in proximity (Tables S 6 and S 9 ). These findings support host-related variation in AMF communities [ 12 ] and the greater effect of space than host identity [ 40 ] on AMF communities. Among the 15 AMF genera detected in this study using the GenBank and Maarj AM databases (Table 4 ) and phylogenetic analysis (Online Resource 3 ), Glomus and Paraglomus were the most abundant in the AMF community (Table 3 ). Glomus or Glomeraceae was most abundant in the majority of previous investigations of AMF communities associated with Cj or Co [ 10 , 20 , 38 ] or in many other host plants in different regions globally [ 35 , 42 ]. In this study, we further provided a compared composition and structure of AMF community between root and surrounding soil. MZ479751 (VTX00444), MZ479752 (VTX00080), MZ479753 (VTX00166), and MZ479754 (VTX00219) were the four most dominant OTUs recorded in the current study (Table 3 ). According to the global distribution of virtual AMF taxa based on Maarj AM database, all but MZ479751 were globally distributed, suggesting that the corresponding AMF species are cosmopolitan. In contrast, MZ479751 which was the most dominant OTUs recorded in our study was previously recorded in Estonia only. Therefore, the corresponding AMF species may have a restrain geographical distribution or data on the species are not submitted to publicly available databases. In this study, several dominant OTUs corresponded to the same virtual taxa defined in the Maarj AM database (Table 4 ). Miyake et al. [ 38 ] used the same OTU clustering threshold (97%) and reported similar results. Compared to previous studies of Cj and Co AMF communities, our work yielded larger numbers of AMF OTUs and dominant AMF OTUs, possibly because of the sampling design. In addition, Japan has ecosystems with large numbers of AMF taxa. For example, Öpik et al. [ 43 ] indicated in a review that Saito et al. [ 44 ] recorded the second-greatest AMF taxon richness (24 AMF taxa) from two temperate grassland sites in Japan. We recorded 15 taxa from three sites in planted Cj and Co forests. So, contrary to the conclusion of Miyake et al. [ 38 ], AMF communities in Japan are not composed of small numbers of taxa."
} | 3,938 |
32222320 | null | s2 | 4,481 | {
"abstract": "Isotopically nonstationary metabolic flux analysis (INST-MFA) provides a versatile platform to quantitatively assess in vivo metabolic activities of autotrophic systems. By applying INST-MFA to recombinant aldehyde-producing cyanobacteria, we identified metabolic alterations that correlated with increased strain performance in order to guide rational metabolic engineering. We identified four reactions adjacent to the pyruvate node that varied significantly with increasing aldehyde production: pyruvate kinase (PK) and acetolactate synthase (ALS) fluxes were directly correlated with product formation, while pyruvate dehydrogenase (PDH) and phosphoenolpyruvate carboxylase (PPC) fluxes were inversely correlated. Overexpression of enzymes for PK or ALS did not result in further improvements to the previous best-performing strain, while downregulation of PDH expression (through antisense RNA expression) or PPC flux (through expression of the reverse reaction, phosphoenolpyruvate carboxykinase) provided significant improvements. These results illustrate the potential of INST-MFA to enable a systematic approach for iterative identification and removal of pathway bottlenecks in autotrophic host cells."
} | 302 |
37635954 | PMC10457423 | pmc | 4,482 | {
"abstract": "Microbial electrosynthesis (MES) is a promising carbon utilization technology, but the low-value products (i.e., acetate or methane) and the high electric power demand hinder its industrial adoption. In this study, electrically efficient MES cells with a low ohmic resistance of 15.7 mΩ m 2 were operated galvanostatically in fed-batch mode, alternating periods of high CO 2 and H 2 availability. This promoted acetic acid and ethanol production, ultimately triggering selective (78% on a carbon basis) butyric acid production via chain elongation. An average production rate of 14.5 g m −2 d −1 was obtained at an applied current of 1.0 or 1.5 mA cm −2 , being Megasphaera sp. the key chain elongating player. Inoculating a second cell with the catholyte containing the enriched community resulted in butyric acid production at the same rate as the previous cell, but the lag phase was reduced by 82%. Furthermore, interrupting the CO 2 feeding and setting a constant pH 2 of 1.7–1.8 atm in the cathode compartment triggered solventogenic butanol production at a pH below 4.8. The efficient cell design resulted in average cell voltages of 2.6–2.8 V and a remarkably low electric energy requirement of 34.6 kWh el kg −1 of butyric acid produced, despite coulombic efficiencies being restricted to 45% due to the cross-over of O 2 and H 2 through the membrane. In conclusion, this study revealed the optimal operating conditions to achieve energy-efficient butyric acid production from CO 2 and suggested a strategy to further upgrade it to valuable butanol.",
"conclusion": "4 Conclusion This study shows that fed-batch operation with alternating periods of high pCO 2 and pH 2 is suitable for promoting butyric acid production from CO 2 in MES cells. A microbial population dominated by Megasphaera sp. achieved 78% selectivity, the highest reported so far for butyric acid, at an applied current density of 1.5 mA cm −2 . The enriched cathodic community can act as inoculum to obtain butyric acid production in successive cells with reduced start-up time. An efficient cell design with low ohmic resistance helped to reduce the electric power requirement to 34.6 kWh kg −1 butyric acid. The produced butyric acid can be further converted to valuable butanol by ceasing the CO 2 supply and maintaining high pH 2 (>1.7 atm) and low pH (<4.8).",
"introduction": "1 Introduction The development of efficient carbon capture and utilization (CCU) technologies with low power demand is compulsory to achieve the Sustainable Development Goal (SDG) 13 on climate action, while also avoiding negative impacts on other SDGs [ 1 ]. Among these technologies, microbial electrosynthesis (MES) has emerged as a promising approach for the electricity-driven CO 2 reduction to biofuels and platform chemicals by chemolithoautotrophic organisms [ 2 , 3 ]. To date, acetate and methane have been selectively produced from CO 2 and electricity in MES cells using a mixed culture of microorganisms as inoculum. Remarkable coulombic efficiencies approaching 100% have been reported for both products [ 4 , 5 ]. However, their low market value (<€0.7 kg −1 ) compared to their production costs challenges the industrial adoption. Thus, recent studies focus on producing more valuable compounds, such as ethanol and middle-chain fatty acids. Ethanol can be produced from acetic acid by solventogenic microorganisms mainly belonging to the genus Clostridium [ 6 , 7 ]. Ethanol production in MES cells has often been reported upon acetic acid accumulation. Romans-Casas et al. [ 8 ] achieved the highest ethanol production rate of nearly 11 g m −2 d −1 ) after acetic acid accumulated to 6 g L −1 in the catholyte by operating an H-type cell under high hydrogen partial pressure (pH 2 > 3 atm) and low pH (<4.5). The produced ethanol can be extracted, purified and commercialized, or further converted into higher-value products, such as butyric acid, via chain elongation pathways [ 9 ]. A thorough metagenomic survey confirmed the genetic potential of MES microbiomes to produce elongated (C 4 and C 6 ) carboxylic acids and alcohols from CO 2 [ 10 ]. Chain elongation is carried out by anaerobic microorganisms, such as some Clostridium , Megasphaera , and Caproiciproducens strains that convert a C n to a C n +2 carboxylic acid using ethanol or lactate as electron donors. This occurs via reverse β -oxidation or fatty acid biosynthesis pathways [ 9 ]. Thus, in the first chain elongation cycle, acetic acid is elongated to butyric acid, a chemical with a market value of circa €1.8 kg −1 that finds application in the pharmaceutical, farming, perfume, and chemical sectors [ 11 ]. Butyric acid production from CO 2 in MES cells was first reported by Ganigué et al. [ 12 ] and later in several studies [ [13] , [14] , [15] , [16] ]. However, butyric acid was always produced in mixtures with other carboxylic acids without exceeding 50% selectivity, making the follow-up downstream step difficult. Various techniques can separate carboxylic acids, including liquid-liquid extraction, membrane separation, adsorption, and electrodialysis [ 17 ]. However, regardless of the technique applied, the selective extraction of butyric acid from a mixture of carboxylic acids remains a challenge. Therefore, improving the selectivity of the production stage is crucial to achieving cost-effectiveness. Butanol production was occasionally reported along with butyric acid in MES cells [ 10 , 12 , 18 , 19 ]. Butanol has great industrial importance as a drop-in biofuel compatible with the current gasoline infrastructures [ 20 ] and also finds application in the pharmaceutical and chemical industries as a precursor to acrylate and methacrylate [ 21 ]. Selective extraction and purification of butanol from mixtures is simpler than doing the same for butyric acid, as it can be accomplished through distillation or by innovative, more energy-efficient methods like pervaporation [ 22 ]. Rovira-Alsina et al. [ 23 ] developed a thermodynamic model to predict the best conditions to trigger the production of elongated carboxylic acids and alcohols and reported that butanol production should be favoured at low pH and high pH 2 . Although thermodynamics is not the only factor affecting production in biological systems, setting up operational conditions enhancing the thermodynamics of a specific reaction is an effective starting point to optimize production rates and specificity experimentally. Besides the conversion efficiency, selectivity, and market value of the final product, the techno-economic feasibility of MES widely depends on the electric efficiency of the cells. Most of the studies on MES have been performed in inefficient H-type cells, characterised by high ohmic resistance and requiring as high as 7 V to deliver current densities of 3–4 mA cm −2 [ 8 ]. In addition to high operational costs, technologies with a high power demand have been demonstrated to have a collateral effect on population health, water scarcity, and mineral/metal resources on a large scale [ 1 ]. Therefore, compact, scalable electrochemical cells with low ohmic resistance are imperative to reduce overpotentials, minimizing the electric power input for product synthesis. Efficient, scalable, flat-plate cells with a minimum distance between the cathode and anode electrodes (i.e., low- and zero-gap configurations) have been designed to perform electrochemical CO 2 reduction [ 24 ] and recently proposed for bioelectrochemical systems [ 25 ]. Previous studies revealed the potential of MES for the conversion of CO 2 to valuable butyric acid and butanol [ 10 , 18 ]. However, increasing product selectivity to facilitate downstream processing while minimizing the electric energy expenses is imperative to bring this technology towards commercialization. This study aimed to develop a strategy to achieve high-selectivity butyric acid production from CO 2 in electrically efficient low-gap (with a 2-mm distance between the electrodes) MES cells and unravel the combination of operation conditions (pH 2 , pCO 2 , and pH) that trigger its further conversion to butanol.",
"discussion": "3 Results and discussion 3.1 Bioelectrochemical CO 2 conversion to butyric acid After inoculating EC-1, microbial growth was detected from the first days of operation, as suggested by the linear OD increase from 0.027 to 0.161 by day 21 ( Fig. 2 ). Despite this, no organic products were synthesised from CO 2 on days 0–12 at an applied current of 0.3 mA cm −2 . Acetic acid production was detected immediately after increasing the applied current density to 0.6 mA cm −2 ( Fig. 2 ), with a linear production rate of 10.0 g m −2 d −1 (0.13 g L −1 d −1 ) on days 12–19 ( Table 1 ), before slowing down on days 19–21. Acetic acid production resumed at a constant rate of 5.5 g m −2 d −1 after increasing the applied current to 1.0 mA cm −2 until day 54. The highest concentration obtained at the cathode was 2.6 g L −1 ( Fig. S2 ). Concomitantly, the pH naturally decreased from 5.8 (day 23) to 4.5 (day 54) and eventually led to the inhibition of acetic acid production ( Fig. 2 ). Fig. 2 Partial pressures ( a ), pH and OD ( b ), and carboxylic acid ( c ) and alcohol ( d ) production profiles (sum of cathode and anode concentrations) in EC-1 (left) and EC-2 (right) at the different current densities applied. Fig. 2 Table 1 Operation conditions and average performance parameters of the EC-1 and EC-2 cells during the periods of linear ( R 2 > 0.95) production. Table 1 Cell Period (d) Product Operational conditions Performance parameters Current applied (mA cm −2 ) Cell voltage (V) Cathode potential (V vs. SHE a ) pH range pH 2 (atm) OD b (AU) Production rate (g m −2 d −1 ) Production rate (mg L −1 d −1 ) Power requirement (kWh kg −1 ) CE c (%) Selectivity (%) EC-1 12–19 Acetic acid 0.6 2.40 −0.55 5.4–6.1 0.91 0.12 10.0 125.4 34.4 25.5 95.8 21–54 Acetic acid 1.0 2.57 −0.51 4.5–5.8 1.14 0.11 5.5 68.8 111.1 7.2 59.2 65–79 Butyric acid 1.0 2.58 −0.50 4.9–5.6 0.85 0.28 14.5 181.5 34.6 45.3 74.2 86–105 Butyric acid 1.5 2.75 −0.52 4.7–5.2 1.36 0.44 10.0 124.4 94.3 17.7 60.7 77–82 Caproic acid 1.5 2.75 −0.51 4.8–5.1 1.08 0.45 2.0 25.2 434.8 4.6 23.3 EC-2 7–11 Acetic acid 1.5 2.67 −0.42 5.3–6.8 1.34 0.22 27.3 340.9 34.5 27.6 82.6 14–28 Butyric acid 1.5 2.66 −0.47 4.7–5.3 1.38 0.38 14.2 177.4 65.4 24.6 77.6 18–28 Caproic acid 1.5 2.67 −0.46 4.7–4.9 1.50 0.41 0.8 10.1 1111.1 1.7 4.6 46–51 Butanol 1.5 2.68 −0.40 4.6–4.7 1.80 0.17 2.2 27.0 416.7 5.6 27.9 67–72 Butanol 1.5 2.70 −0.43 4.7–4.9 1.74 0.27 1.4 17.8 625.0 3.8 18.3 a Standard hydrogen electrode. b Optical density in absorbance units. c Coulombic efficiency. Acetic acid accumulation and the pH decrease triggered the onset of solventogenesis [ 28 ]. Ethanol was produced concomitantly with acetic acid from day 30 onwards, although its concentration remained low (<20 mM C). On day 63, acetic acid and ethanol concentrations decreased, and longer C-chains compounds (mainly butyric acid) were formed. This shift was due to the onset of chain-elongating pathways [ 29 ]. After increasing the current to 1.5 mA cm −2 on day 75, ethanol was only detected in low amounts (<2 mM C), suggesting its rapid consumption by chain-elongating microorganisms. Butyric acid concentrations increased, although with some fluctuations, likely due to the increased availability of H 2 in the reactor. The observed fluctuations in butyric acid concentration were presumably due to competitive reactions such as solventogenic butanol production and further elongation to caproic acid ( Fig. 2 ). Additionally, a portion of butyric acid was likely oxidized due to O 2 diffusion from the anodic to the cathodic chamber, an ongoing challenge in MES [ 30 ]. Despite these factors, a linear ( R 2 > 0.95) butyric acid production with a rate of 14.5 g m −2 d −1 (0.17 g L −1 d −1 ) was observed on days 65–79 ( Table 1 ). This rate was independent of the applied current density of 1.0 mA cm −2 (on days 65–75) or 1.5 mA cm −2 (on days 75–79). These results suggest that butyric acid synthesis was limited by the ethanol production rate, despite the increased availability of reducing equivalents at higher currents. Butyric acid was produced until day 117, reaching a total of 234 mM C. The highest butyric acid concentration in the catholyte was 3.2 g L −1 ( Fig. S2 ). Caproic acid production started after day 70 when butyric acid concentration at the cathode was near 1 g L −1 . It was produced at a rate of 2.0 g m −2 d −1 on days 77–82, up to a maximum of 24 mM C on day 91, before declining until day 126. No net carboxylic acid or alcohol production was achieved after increasing the applied current density to 3.0 mA cm −2 on day 126, despite the increased H 2 production and the pH rise from 4.5 to 5.3. This can be attributed to a progressive biofilm detachment caused by the increased H 2 production [ 31 ], as suggested by the increase in OD ( Fig. 2 ). Furthermore, doubling the applied current likely intensified O 2 diffusion from the anode, causing an increased product consumption by aerobic organisms. During cell operation, the gas pressure at the cathode followed a cyclic trend, where pCO 2 decreased from the initial set value of 1.3–1.5 atm (feast conditions) towards negative pressures (famine conditions) due to consumption by bacteria for their metabolism and product synthesis. Concomitantly, pH 2 increased over time proportionally to the current applied ( Fig. 2 ). The average pH 2 at the end of each batch cycle was 0.73 ± 0.20, 0.87 ± 0.27, 1.02 ± 0.34, 1.33 ± 0.26, and 2.52 ± 0.38 atm at an applied current of 0.3, 0.6, 1.0, 1.5, and 3.0 mA cm −2 , respectively. The alternate periods with high and low H 2 availability (and the opposite CO 2 trend) favoured the solventogenic and acetogenic pathways, respectively, ultimately leading to butyric acid production via chain elongation with over 70% selectivity ( Table 1 ). To further assess the replicability of the process, a second cell (EC-2) was inoculated with 8 mL (10%) of catholyte from EC-1 and operated under similar conditions, excluding the current density, which was set to 1.5 mA cm −2 from the beginning. Acetic acid production began on day 7, and its concentration rapidly increased to 62 mM C (day 12) and then slowed down, reaching 97 mM C by day 46. Ethanol concentration remained below 8 mM C, as it was consumed for butyric acid production by chain elongation. Butyric acid production was observed from day 12 ( Fig. 2 ). The lag phase was 82% lower than in EC-1, confirming that enriched cultures of chain-elongating microorganisms can start up MES cells in a relatively short time. Butyric acid was produced linearly on days 14–28, with an average rate of 14.2 g m −2 d −1 , similar to the rate obtained in EC-1 ( Table 1 ). During this period, it was produced with a remarkable 78% selectivity (on a carbon basis), well above those achieved in previous studies [ 16 , 18 , 32 ]. From day 28 onwards, the production rates declined, likely due to the increased butyric acid concentrations in the catholyte (around 2.3 g L −1 ) ( Fig. S2 ) in combination with low pH (<4.8). Nevertheless, butyric acid was still produced until day 44, reaching 196 mM C. Butanol and caproic acid were detected from day 14 onwards and reached 8.3 and 6.4 mM C, respectively. The concentration of caproic acid was considerably lower than those typically achieved from acetic acid and ethanol in fermentation bioreactors [ 9 ]. It appears that the further elongation of butyric acid to caproic acid was impeded by the low ethanol:acetic acid ratio in the catholyte resulting in the slow ethanol production kinetics in MES cells [ 32 ], despite the relatively high pH 2 achieved in this study. 3.2 Butyric acid upgrade to butanol In EC-1, butanol production fluctuated on days 88–125, with a concentration above 2 g L −1 of butyric acid and pH below 5. The fluctuation was due to the intermittent high pH 2 and pCO 2 conditions. The butanol production peaks were obtained in concomitance with high pH 2 (1.5–1.7 atm) and low pCO 2 (<0.05 atm) ( Fig. 2 ), whereas it was consumed in periods with high carbon availability. A highly positive Pearson correlation (T-score: 6.742, p -value <0.05) between butanol selectivity and pH 2 , as well as a positive Pearson correlation (T-score: 3.903, p -value < 0.05) between butanol production rate and pH 2 , were observed in this phase ( Fig. S3 ). To further investigate this phenomenon, the CO 2 supply was interrupted for one week in EC-2 on days 46–56 (ButOH test 1, Fig. 3 ) when the butyric acid concentration at the cathode was around 3 g L −1 ( Fig. S2 ), pH was 4.6, and a stable pH 2 of 1.78 ± 0.04 atm was maintained in the cell through a pressure release valve. Butanol was linearly produced for five days with an average production rate and selectivity of 2.2 g m −2 d −1 and 28%, respectively, before stopping due to a pH increase to 4.9. Similarly, Ganigué et al. [ 12 ] and Srikanth et al. [ 18 ] reported butanol production rates of 2–3 g m −2 d −1 at pH values of 4.5–4.6, though with substantially lower selectivity than in this study. This suggests that maintaining a pH below the pK A of butyric acid (i.e., 4.82) is imperative to trigger butanol production. At pH 4.6, nearly 70% of the butyric was in the undissociated form, which is the most toxic form for the microorganisms, triggering the defensive solventogenic mechanism [ 33 ]. Therefore, high butyric acid accumulation, low pH, lack of inorganic carbon, and high pH 2 (at least 1.5–1.7 atm) are simultaneously needed to trigger butanol production from butyric acid via solventogenesis, as previously reported for ethanol production from acetic acid [ 7 ]. Fig. 3 H 2 accumulation tests (ButOH test 1 and ButOH test 2) performed in EC-2. The white area represents the period where the CO 2 supply was stopped and pH 2 accumulated to 1.78 ± 0.04 atm. The grey area represents the recovering period where CO 2 was supplied again. Fig. 3 As expected, acetogenesis ceased during the test due to the lack of CO 2 , but acetic acid was progressively converted to butyric acid, suggesting that chain elongating pathways were still active in the absence of CO 2 . This is an expected result, as acetogenesis is the only CO 2 -requiring metabolic reaction in the system. Ethanol concentration also increased during CO 2 depletion, confirming that solventogenic pathways are favoured under those conditions when the corresponding carboxylic acid is present. From day 50 onwards, the increasing pH resulted in a depletion of alcohol concentrations, which became more evident after resuming the CO 2 feeding on day 56, suggesting the reversibility of the bioreaction [ 7 ]. Repeating the test on days 67–79 (ButOH test 2, Fig. 3 ) confirmed the same trends, although the production rate was 35% lower due to the slightly higher initial pH (4.73 vs. 4.61). 3.3 Microbial community analysis The suspended community of EC-1, originally dominated by species belonging to the genus Clostridium , radically changed within the first 12 days of operation at an applied current of 0.3 mA cm −2 ( Fig. 4 ). During this period, the relative abundance of Clostridium sp. dropped. Accordingly, neither carboxylic acids nor alcohols were produced at such low current density ( Fig. 2 ). Desulfovibrio sp. , an electrotrophic microorganism capable of catalysing H 2 production at the cathode, was detected on day 12 with a relative abundance of 7%, suggesting that H 2 -mediated acetogenesis was taking place [ 34 ]. A meta-analysis of MES microbiomes has shown that Desulfovibrio is commonly found as the syntrophic partner of acetogenic microorganisms in biocathodes [ 35 ]. Increasing the current from 0.3 to 0.6 mA cm −2 resulted in an increased relative abundance of acetogenic microorganisms identified as Sporomusa sp. (from 1% to 10%) and Acetobacterium sp. (from 2% to 5%), with the resulting onset of acetic acid production ( Fig. 2 ). Sporomusa sp. has a lower H 2 threshold than most acetogenic microorganisms [ 36 ], allowing these organisms to perform acetogenesis through the Wood-Ljungdahl pathway at low applied potential or, in this case, at low applied current. Fig. 4 Genus-level community composition of EC-1 and EC-2 at different time points. ASVs found in a relative abundance <1% were grouped and graphed as Others. S and A stand for suspended and attached community, respectively. Fig. 4 Further increasing the current to 1 mA cm −2 resulted in the revival of Clostridium sp. (most likely Clostridium ljungdahlii ) and Eubacterium sp., which were the dominant acetogens in the inoculum, at the expense of Sporomusa (<1%), probably due to their faster kinetics for H 2 consumption [ 36 ]. Since the cathode potential remained fairly constant at the increasing current density applied ( Fig. S4 ), this result suggested that the availability of reducing equivalents was the main driving force in selecting the acetogenic communities in the E-cells. In EC-1, after 75 days of operation, Megasphaera sp. was detected at a relative abundance of 32%, which increased to 54% on day 111. Megasphaera has been shown to perform chain elongation from various sugars and lactic acid [ 9 , 37 ], and it was suggested to use ethanol as the electron donor in MES cells to form butyric acid [ 16 ]. Megasphaera was the most abundant genus in EC-2, with relative abundances between 30% and 44%, and contributed by 24% to the attached community by the end of the experiment ( Fig. 4 ). Such a high relative abundance indicates its key role in butyric acid production in the two cells. Butanol production tests did not substantially affect the composition of the microbial community. Therefore, it can be hypothesized that the shift from acidogenesis to solventogenesis was due to metabolic shifts rather than species substitution. Current densities of 1.0 mA cm −2 and above had the collateral effect of increasing O 2 production at the anode, likely resulting in its faster diffusion of O 2 through the membrane towards the cathode. This was confirmed by the increased relative abundance of facultative or strictly aerobic microorganisms acting as oxygen scavengers [ 16 , 38 ]. In particular, Acetobacter sp. reached 22% of sequence reads in EC-1 after increasing the current density to 3.0 mA cm −2 . Interestingly, in EC-2, the relative abundance of Acetobacter sp. among the attached community was remarkably lower than in the bulk community ( Fig. 4 ). This suggests that the scavengers were effectively shielding the cathodic community from oxygen intrusion, protecting the anaerobic members of the community at the expense of the CEs ( Table 1 ). Overall, the key species found in the attached and suspended communities were similar, although the relative abundances were different, as previously reported [ 39 ]. 3.4 Perspectives and challenges Fed-batch operation with alternating periods of high pCO 2 and pH 2 was a suitable strategy to target high-selectivity butyric acid from CO 2 in MES cells. The E-cell was designed to maximise electric efficiency with only a 2-mm distance between the cathode and the anode and a high A cat /V cat of 250 m 2 m −3 for microbial adhesion. This resulted in an ohmic resistance of 15.7 mΩ m 2 , estimated by EIS analysis ( Fig. S4 ). To our knowledge, a lower ohmic resistance (2.4 mΩ m 2 ) has been reported only in a methanogenic zero-gap MES flow cell equipped with membrane-electrode assembly and solid anodic electrolyte [ 40 ]. This allowed us to achieve an electric power requirement of 34.6 kWh kg −1 butyric acid, 2-fold lower than the 64.3 kWh kg −1 reported in the literature [ 13 ]. As confirmed by CV analysis ( Fig. S4 ), the H 2 onset potential was around −0.5 V, higher than in previous studies performed under galvanostatic conditions [ 14 , 41 ], suggesting a low overpotential for the cathodic hydrogen evolution reaction (HER). Interestingly, increasing applied current density had only a marginal effect on the cathode potential ( Fig. S4 ), suggesting that H 2 production can be sustained at relatively high current densities without significantly increasing the overpotential. Conversely, the anode potential increased significantly with the current applied (from 1.52 V at −0.3 mA cm −2 to 2.96 V at −3.0 mA cm −2 ). Hence, the anodic reaction was the main contributor to the overall cell voltage, which can be reduced by mitigating the pH split between anode and cathode and considering less demanding anodic reactions, such as oxidation of organics [ 42 ]. Although the process may not be competitive with the current electricity prices, it should be considered that the cost of renewable energy (mostly solar and wind power) is expected to drop to somewhere between €10 MWh −1 and €50 MWh −1 by 2050 [ 43 ]. Assuming an electricity price of €30 MWh −1 , a production cost of around €1 kg −1 can be achieved, which would then require additional expenses for extraction and purification, typically accounting for 30–40% of the production cost [ 44 ]. The power consumption can be theoretically reduced to as low as 7.5 kWh kg −1 when assuming the minimum cell voltage of 1.23 V for water splitting and 100% CE. However, a more realistic voltage of 1.8–2.0 V, as achieved in state-of-art electrolysers [ 45 ], would produce an electric power requirement of 10–15 kWh kg −1 , potentially making the process competitive with chemical synthesis. Further conversion of the produced butyric acid to butanol, a promising green alternative to traditional fuels [ 46 ], will further improve the economics of the process. However, production rates, selectivity, and titer must be improved. This can be accomplished by implementing a controlled system that maintains the required pH (<4.8) and pH 2 (>1.7 atm) following butyric acid accumulation. Employing compact and modular cells with minimum distance between the electrodes, connected in a multi-stack configuration, represents the most efficient approach to scaling up electrochemical systems, such as MES [ 47 ]. However, despite the considerable improvements in electric efficiency, minimizing the distance between electrodes can increase the exposure of the cathodic community to O 2 . Cation exchange membranes cannot completely circumvent O 2 diffusion to the cathode, hampering the activity of strictly anaerobic members of the microbial community and/or causing product consumption by microaerophilic microorganisms [ 30 ]. Additionally, if not immediately utilized by the microbial community, the H 2 produced at the cathode can permeate towards the anode and escape into the atmosphere. This issue is particularly relevant when the cathodic chamber is maintained under pressure. In this study, it was experimentally measured that 4.6 μmol h −1 of H 2 (2.5% of the total produced at 10 mA cm −2 applied current) were lost through the membrane at a pH 2 of only 1.12 atm. Considering that the cells reached pH 2 values up to 3 atm, this percentage was inevitably higher, particularly during operation at 1.5 and 3 mA cm −2 applied current. The same phenomenon was observed for CO 2 , which was shown to permeate through the membrane at a rate of 22 mmol h −1 when the pCO 2 was around 1.5 atm. This issue could potentially be solved by employing low-gas permeability membranes. However, to the best of the authors’ knowledge, such membranes are yet to be developed and implemented. It is essential to address this problem in the near future and find a solution, as higher current densities are required to scale up MES cells."
} | 6,941 |
30034576 | null | s2 | 4,483 | {
"abstract": "Biological production of hydrocarbons is an attractive strategy to produce drop-in replacement transportation fuels. Several methods for converting microbially-produced fatty acids into reduced compounds compatible with petrodiesel have been reported. For these processes to become economically viable, microorganisms must be engineered to approach the theoretical yield of fatty acid products from renewable feedstocks such as glucose. Strains with increased titers can be obtained through both rational and random approaches. While powerful, random approaches require a genetic selection or facile screen that is amenable to high throughput platforms. Here, we present the use of a high throughput screen for fatty acids based on the hydrophobic dye Nile red. The method was applied to screening a transposon library of a free fatty acid overproducing strain of "
} | 216 |
27190206 | PMC4898794 | pmc | 4,484 | {
"abstract": "The spatial distribution of microbes on our planet is famously formulated in the Baas Becking hypothesis as “everything is everywhere but the environment selects.” While this hypothesis does not strictly rule out patterns caused by geographical effects on ecology and historical founder effects, it does propose that the remarkable dispersal potential of microbes leads to distributions generally shaped by environmental factors rather than geographical distance. By constructing sequence similarity networks from uncultured environmental samples, we show that microbial gene pool distributions are not influenced nearly as much by geography as ecology, thus extending the Bass Becking hypothesis from whole organisms to microbial genes. We find that gene pools are shaped by their broad ecological niche (such as sea water, fresh water, host, and airborne). We find that freshwater habitats act as a gene exchange bridge between otherwise disconnected habitats. Finally, certain antibiotic resistance genes deviate from the general trend of habitat specificity by exhibiting a high degree of cross-habitat mobility. The strong cross-habitat mobility of antibiotic resistance genes is a cause for concern and provides a paradigmatic example of the rate by which genes colonize new habitats when new selective forces emerge.",
"conclusion": "Conclusions By adopting a similarity network approach on a comprehensive set of environmental sequences, we revealed the absence of an overall distance effect in the level of sequence sharing among microbial samples; even distant microbial communities may share more homologous sequences than geographically closer DNA pools. Metagenome gene composition is therefore strongly affected by ecology. Interestingly, inland water samples occupy a “bridge-like” position in the overall metagenome network ( fig. 3 a ). Hence, despite maintaining their own (specific) gene pool as assessed by clustering analyses, these samples connect microbial communities that otherwise would remain disconnected (e.g., host and seawater samples). This is in agreement with previous findings on the horizontal flow of plasmid genes ( Fondi and Fani 2010 ) and speculations on the role of aquatic environments in the spreading of AR-related determinants ( Baquero et al. 2008 ). These trends were confirmed when the SSN was converted into a putative HGT network by maintaining only those connections linking very similar sequences (identity ≥ 98%) in distantly related microorganisms (i.e., belonging to different genera). Ecology strongly influences the network of HGT in microbes even when samples not strictly related to human are considered, as has also been preliminarily observed in terrestrial and aquatic environments ( Hooper et al. 2008 ). Moreover, HGT events mainly involve molecules of the same kind (i.e., either plasmids or chromosomes) with promiscuous gene exchange being less frequent. Our work shows the possible use of SSN for studying patterns in microbial ecology and also lays foundations for integrating such networks with other environmental parameters (e.g., temperature, pH, pressure, and physical barriers) on the structure of the gene sharing and HGT networks. Finally, our findings provide support for the Baas Becking hypothesis (formulated in 1934), suggesting that it also applies to genes, besides microbes for which it was originally formulated. Overlapping microbial gene pools are likely to be found in widely geographically disparate environments, and tighter associations are observed among gene pools from similar habitats. This holds true regardless of microbial evolutionary lineages (i.e., their common evolutionary history) since we have shown that the same patterns of common gene pools still remain when only genes likely shared by means of HGT events are maintained in the network. This suggests that it is not so important which organism transcribes and translates a gene and it matters more where that organism is located, demonstrating that at least some genes act as public goods ( McInerney et al. 2011 ). Accordingly, they are available for all organisms to integrate into their genomes although the kind of ecological niches occupied and the type of informative molecules harboring them might impose some constraints on the overall possibility of gene pools to undergo HGT. Finally, besides drafting an overall scheme of pathways for the global distribution of gene pools, results presented here provide important biological insights into the spreading of antibiotic-resistance-related genes across multiple hosts and habitats.",
"introduction": "Introduction The spatial distribution of microorganisms on the planet is often expressed according to Baas Becking’s famous tenet “everything is everywhere but the environment selects” ( Baas Becking 1934 ). “Everything is everywhere” alludes to the remarkable dispersal potential of microorganisms, whereas “the environment selects” implies that only specifically adapted organisms will thrive and proliferate in a particular environment ( Fuhrman 2009 ). The Baas Becking hypothesis does not rule out the possibility of strong geographic patterns but rather suggests that geography per se does not drive the distribution of species—geographic patterns could simply reflect an association between geography and ecology. Empirical testing of the Baas Becking hypothesis has focused mainly on specific microorganisms and/or specific environments ( Reno et al. 2009 ; Sul et al. 2013 ). Because most members of microbial communities resist cultivation, understanding of molecular and ecological details of microbial biogeography remains vague ( Staley and Konopka 1985 ; Martiny et al. 2006 ; Raes et al. 2011 ; Hanson et al. 2012 ). However, the recent increase in the number of metagenomes in public repositories offers an opportunity to explore the global distribution of coding sequences, universally shared phylogenetic marker genes, and horizontally transferred genes, including genes of clinical importance such as antibiotic resistance genes ( Fondi and Fani 2010 ). Furthermore, many studies have highlighted the importance of network theory and approaches based on sequence similarity networks (SSNs) in studying large-scale evolutionary relationships, including the influence of habitat and ecology in the distribution of gene pools, evolution of organisms, and horizontal gene transfer (HGT, Lima-Mendez et al. 2008 ; Halary et al. 2010 ; Dagan 2011 ; Tamminen et al. 2012 ; Alvarez-Ponce et al. 2013 ; Forster et al. 2015 ). However, in most cases, only completely sequenced genomes (including plasmids and phages) were used for these analyses, thus limiting the scope of the studies to mainly cultivable microorganisms or specific phyla (i.e., ciliates). Indeed, often the initial habitat assignment stems from where the organism was first isolated, which may not be its only, or even its preferred, habitat ( Hooper et al. 2009 ). Here, we empirically test the Baas Becking hypothesis by applying it to genes as well as organisms. By studying 339 metagenomes (pooled into roughly 100 sampling points) using an SSN approach ( Fondi and Fani 2010 ; Halary et al. 2010 ), we offer a culture-independent view of microbial gene pool commonalities and differences and investigate whether the distributions of genes are limited to particular ecological niches or whether they display a cosmopolitan or geographically defined distribution. Geographical influence on overall patterns of gene distribution is measured as the correlation between the physical distance and the degree of shared homologous sequences between the metagenomes. A positive or negative correlation indicates a distance-effect on global macroscale patterns of gene distributions, whereas absence of such correlation suggests independence between geographical distance and proportion of shared sequences. While gene dispersal may depend on the distribution patterns of microbial species, genes can also rapidly move between phylogenetically distant cells by means of HGT. To test whether the putative horizontally transferred genes follow the distribution of their hosts or form their own distribution, we converted the reconstructed SSN into an HGT network and investigated its main topological features. By applying a network-oriented analysis pipeline on culture-independent environmental data, we here demonstrate the cosmopolitan distribution of genes and the influence of ecology on their distribution and, in parallel, we show that the same patterns hold for “mobile” genes. Our findings have important implications in several areas of biology, from environmental microbiology to antibiotic resistance, to microbial evolution and to the structure of present day common gene pools.",
"discussion": "Results and Discussion General Features We built an SSN using metagenome sequences from 97 sampling sites (representing 339 metagenomic projects, see supplementary material S1, Supplementary Material online) where nodes represent sampling points and links reflect the number of shared homologous sequences (see Materials and Methods for network construction details). We used different sequence identity thresholds in building these SSNs (i.e., 70%, 80%, 90%, 95%, and 99%). Results presented here refer to the 90% network, although the results are valid for all identity thresholds (see supplementary material S1, Supplementary Material online). In figure 1 , the extent of sequence sharing among the different samples is presented as a network, together with the geographical location of each sampling site. To test whether physical distance and the number of homologous DNA fragments shared by the different metagenomes correlate, we calculated Pearson-product-moment correlation coefficients for samples from different (Pearson Correlation Coefficient [PCC] = −0.038 and P value = 6 × 10 − \n 3 ) and same habitats (from PCC = −0.2 in soil samples to 0.04 in fresh water samples, P values < 6 × 10 − \n 3 ; supplementary fig . S2, Supplementary Material online). Therefore, physical distance at the spatial resolution provided by the available metagenomes does not explain the distribution of the links in the metagenome-derived SSN, suggesting a relatively marginal role of physical distance in the shaping of the biological relationships. Exemplars of this situation are reported in figure 1 b and c for host- and sea water-derived samples. Metagenomes of the subnetwork of figure 1 b (samples no. 77, 25, 88 and 89, see supplementary material S2, Supplementary Material online), although connected to almost all the other metagenomes in the network, share many more sequences among themselves. The sequences embedded in these metagenomes were obtained from microbiomes of geographically distant Arthropods: Dendroctonus ponderosae (samples 88 and 89), D. frontalis (sample 25), Xyleborus affinis (sample 77), and Sirex noctilio (sample 54). We observed a similar trend in geographically disparate specimens of sea squirt Ciona intestinalis ( Dishaw et al. 2014 ), consistent with the selection of a core community by that particular ecosystem. We observed the same feature for metagenomes displayed in figure 1 c (samples no. 2, 97, 10, 39, 14, 28, 27, 2, and 8, see supplementary material S2, Supplementary Material online), all from seawater samples and all sharing heavy connections despite most being separated by large geographical distances. Accordingly, we speculate that the similarity of the ecological niches in which samples were collected explains the high level of gene sharing among these two sets of metagenomes. Figure 1 c also shows that, within samples sharing the same source niche, some nodes that are close in the network (e.g., 10, 7, and 97) display fewer connections among them in respect, for example, to those shared with nodes 28 and 2 (being far away in the map). This, in turn, might suggest the limit of using physical distances as a proxy for estimating the “real” distance among gene pools. Indeed, other barriers and forces (besides geographical distance) might account for the actual dispersal. This is the case, for example, of sea currents that may contribute to creating quite different environments in two close points in the network of metagenomic samples. Similarly, mountains might create a separation among physically close terrestrial DNA pools. On the other hand, these features are quite hard to be confidently modeled on a large, global scale as the one used in this work.\n F ig . 1.— ( A ) Overall SSN among the 97 sampling points together with their geographical positions. Each node represents a metagenome project and the links represent the presence of homologous sequences between them. Node and link sizes are proportional to the number of sequences embedded in the sample and the (normalized) number of shared sequences, respectively. In ( B ) and ( C ) specific study cases are reported (see text for details) for host-(red nodes) and sea-water (blue)-derived samples. The connections among samples from the same ecological niche and those among samples from different ecological niches are shown in ( D ) and ( E ), respectively. A preliminary visual inspection of the network revealed that samples from same ecological niches ( fig. 1 D ) are more tightly connected than samples from different niches ( fig. 1 E ). Thus, to explicitly test the ecological niche versus geographical distribution hypotheses, we evaluated the correlation between the grouping of the different metagenomes (i.e., the habitat composition of the major clusters in the network of fig. 1 ) and their source habitat. We first clustered the metagenomes according to the Markov Cluster (MCL) algorithm (see Materials and Methods) and then evaluated whether metagenomes belonging to the same ecological niche tended to (significantly) cluster together using recall (R), precision (P), and accuracy (A) measures. This analysis ( fig. 2 ) revealed relatively high values of both R and P across all the different networks (average R = 0.588 and average P = 0.71). A similar trend was observed also when measuring clustering accuracy (A) ( fig. 2 ). Such high values of P, R, and A were never obtained during 1,000 random permutations (label shuffling, see Materials and Methods) of the original networks, giving a P value estimate < 10 − \n 3 . The same results were observed for networks obtained with lower sequence identity thresholds ( supplementary fig. S6 , Supplementary Material online) and when evolutionary distances were considered for a set of 10,000 randomly sampled coding sequences in the data set ( supplementary fig. S7 , Supplementary Material online).\n F ig . 2.— Recall, precision, and accuracy values for real and random network at 90% sequence identity threshold. \n F ig . 3.— ( A ) Force-directed layout representation of the metagenome network (at 90% sequence identity threshold). Each metagenome is colored according to its source habitat as indicated in the legend and major coherent clusters are highlighted. ( B ) The putative HGT network derived from network shown in ( A ) (see text for details on HGT network construction). Additionally, assortativity was used to evaluate the tendency (if any) of nodes of the same type (i.e., sequences from the same source habitats) to cluster together in the network. Briefly, assortativity coefficient measures the preference for a network's nodes to attach to others that share a particular attribute (source environment in our case) and can be comprised between −1 (disassortative network) and 1 (assortative network). Assortativity for the network in figure 3 was found to be 0.157, thus confirming a general pattern of preferential connections between nodes of a particular ecological niche. Importantly, higher assortativity values were never encountered when (1,000) randomization of the original network were performed (edge rearrangement, see Network Analysis and Visualization), allowing to infer a rough estimation of a P value lower than 10 − \n 3 . From this we conclude that the source habitat of the different sequence samples is a key factor in determining their clustering within the different SSNs. A force-directed layout of the network ( fig. 3 a ) reveals a clear separation between sea samples (in dark blue) and samples coming from other sources such as host (red), soil (yellow), waste waters (black), and air filters (light blue). Interestingly, inland-water samples (blue) appear to lay half way between these two major clusters. As listed in table 1 , metagenomes from inland water samples possess the highest betweenness values in the SSN in comparison to all the other sample sources, expressing that these nodes have a central position in the network and that, in turn, they serve as connectors among otherwise separated regions of the network (Mann–Whitney U test, P values in table 1 ). These results were confirmed by randomizations (edge replacement, see Network Analysis and Visualization) of the original graph ( table 1 ) according to which inland water metagenomes, and (to a lower extent) sea water metagenomes, have betweenness centrality values higher than is expected by chance. Inland water metagenomes are also less prone to form clusters within the network, since they show, on average, the lowest clustering coefficient (Mann–Whitney U test, table 1 ). Inland water metagenomes possess also the highest closeness centrality values in the SSN (Mann–Whitney U test, table 1 ). This suggests that, in water, bacteria from different origins (human, animal, and environmental) may be able to mix, co-exist, and travel to an extent that is higher than in other ecological niches. This could give rise to exchange and shuffling of genes, genetic platforms, and genetic vectors ( Baquero et al. 2008 ). This result confirms and extends previous findings on the horizontal flow of the plasmid encoded resistome ( Fondi and Fani 2010 ).\n Table 1 Centrality Measures in Relation to Sample Environmental Origin in Observed and Random Networks Network Metric Soil Sea Host Inland water Real Random Real Random Real Random Real Random Betweenness 4.6 16.8(6.02) 52.46 42.20(4.05) 50.52 61.14(5.24) 102.67 63.76(6.9) P = 2*10 −3 P = 2*10 −2 P = 2*10 −2 Closeness 0.42 0.44(0.008) 0.46 0.48(0.07) 0.49 0.50(0.009) 0.50 0.48(0.01) P = 1*10 −3 P = 9*10 −3 P = 4*10 −3 Clustering c. 0.68 0.18(0.03) 0.6 0.24(0.03) 0.56 0.24(0.02) 0.39 0.20(0.03) P = 1*10 −3 P = 3*10 −1 P = 2*10 −3 N ote .—Values in parentheses after randomized values indicate standard deviation. Values after real values for soil, host, and sea metagenomes indicate P values for comparisons to inland water samples (Mann–Whitney U test). As shown in figure 3 a , nine metagenomes remained disconnected from the overall network. These metagenomes included five seawater samples, two soil samples, one host, and one inland water samples. Not surprisingly, these metagenomes embed fewer sequences than others present in the data set. Indeed, although it has been shown that the metagenome size has a negligible effect on the overall connectivity within the network (see supplementary material S1, Supplementary Material online, and Materials and Methods), some exceptions may still exist. These metagenomes are connected to the others at lower identity thresholds (data not shown). HGT Networks The extent of sequence sharing among the metagenomes can be partially explained by the overlapping taxonomical space of the different samples; indeed, similar habitats may tend to be colonized by the same major taxonomical groups. This latter observation is supported by the results obtained repeating the same analysis pipeline for marker genes retrieved in the studied metagenomic samples ( supplementary fig S4 , Supplementary Material online) and likely with a reduced susceptibility to HGT. Nevertheless, the assembled data set permits us the opportunity to assess the relationships (if any) between physical proximity, ecological niche, and HGT. To account for this task, a second set of networks was constructed, accounting for putative HGT events among the analyzed sequence data sets. We identified putative HGTs as blocks of nearly identical DNA (≥500 nucleotides and ≥98% sequence identity) in otherwise distantly related contigs (i.e., contigs from different genera inferred by a composition-based, semisupervised, taxonomic binning algorithm). Since the method adopted for taxonomic binning of metagenome sequences is mainly suited to microbial sequences ( Nalbantoglu et al. 2011 ), only prokaryote to prokaryote putative gene exchanges will be considered in the following sections. Importantly, trends in sequence sharing described below were observed also when a composition-oriented method (based on the evaluation of differences in tetranucleotide frequency distribution between two contigs, see Contig taxonomic annotation and source molecule identification) was used for the identification of (putative) HGT. The network of HGT among metagenomes is reported in figure 3 b , displaying a topology very similar to the network of gene sharing ( fig. 3 a ) although, as might be expected, possessing fewer links. The HGT network also proves that sequence sharing between metagenomes is not just due to overlapping taxonomical space. To further investigate the HGT network, we built a second type of network in which each node represents a single contig, whereas links account for (putative) HGT events. This network contains 34,555 nodes (contigs) and 34,398 edges (putative HGT events, supplementary material S3, Supplementary Material online) and can be divided into 8,017 connected components (CC), the great majority embedding only few contigs (≤10). We identified 46 larger CCs, embedding 50 or more contigs. Functional annotation was missing for 38% of the genes involved in putative HGT events. Among those that were successfully annotated using Pfam database, the two most represented functional categories were ABC transporters and transposase DDE domain. Considering the biological role of genes embedded into these categories (resistance to xenobiotics and horizontal transfer of genes) this finding highlights the dangerous implications of the horizontal flow of genes in the spreading of microbial resistance (and resistance to xenobionts in general) in natural environments ( Baquero et al. 2008 ; Fondi and Fani 2010 ). Two examples of this are provided below. To investigate the influence of ecology shaping the HGT network, we estimated whether each CC was either homogeneous or heterogeneous in terms of the habitat of the embedded contigs. Results shown in figure 4 a revealed that almost 90% of the CCs (6,814 CCs) contain contigs belonging to the same environment. Heterogeneous clusters are less frequent, although interesting exceptions do exist (see below). The observed distribution of homogeneous clusters was compared against the (averaged) distribution of the same measure from 1,000 networks, obtained through random label reshuffling (see Computational strategy for clusters identification and testing). The distinctness of the two distributions is shown in figure 4 a and was assessed by a Mann–Whitney U test ( P value < 2.2e-16). A high number of interconnections inside each of the examined habitats (e.g., host–host and sea water–sea water) were observed for most of the samples ( fig. 5 ; see below), in agreement with overall samples clustering reported in figure 4 a and with previous findings concerning the possible presence of barriers or trends to HGT ( Popa and Dagan 2011 ). According to this whole body of data, ecology seems to exert a broad influence on recent gene exchange in environmental samples. This is in agreement with the theory according to which ecological similarity shapes networks of gene exchange by selecting for the transfer and proliferation of adaptive traits or by increasing physical interactions between community members ( Aravind et al. 1998 ; Caro-Quintero et al. 2011 ; Smillie et al. 2011 ). For example, strong geographical differentiation apparently caused by recent gene transfer among co-occurring bacteria was observed for Vibrio representatives ( Boucher et al. 2011 ).\n F ig . 4.— Composition of network clusters in terms of habitat and molecule categories. One hundred percentage values on the X axis indicate clusters with contigs belonging to the same category; conversely, lower values indicate more heterogeneous clusters (i.e., contigs belonging to different habitat or to different molecules). The cluster composition is shown for ( A ) habitat coherence and ( B ) molecule coherence (i.e., plasmid–plasmid and chromosome–chromosome). \n F ig . 5.— Adjacency matrix showing the relationships among the different habitat types in the putative HGT events network. For each habitat, the proportion of connections of that habitat with all the other habitats has been computed. The proportion of connections connecting habitat A with habitat B ( PC A , B ) is given by this formula: PC A , B = Weight ( Edge A , B ) ∑ i Weight ( Edge A , i ) Since the denominator represents the amount of sequences in one of the two analyzed samples, this measure is specific to each of the analyzed environments and is not symmetric ( PC A , B ≠ PC B , A ). Color gradient within the matrix refers to the proportion of connections of contigs from a given habitat with all the others from other habitats, with lighter tones representing less abundant interconnections among the corresponding habitats. An adjacency matrix was built to explore more thoroughly the interconnections that link sequences from different habitats and common patterns of gene exchange among samples retrieved from different ecosystems ( fig. 5 ). Two major clusters can be identified on the basis of the dendrogram topology (Clusters 1 and 2 in fig. 5 ). Contigs embedded in each of these clusters have similar connections toward the other environments present in the HGT network. This suggests the presence of a common pool of genes in ecosystems embedded in these clusters. Cluster 1, for example, embeds Host, Sludge waste, and Air ecosystems. This particular clustering is supported by Smillie et al. (2011) and studies showing that fecal coliforms and other animal pathogens are indeed present in sludge waste samples ( Jones 1980 ; De Luca et al. 1998 ; Shanahan et al. 2010 ) and that opportunistic pathogens commonly isolated from human-inhabited environments have been identified in airborne environments ( Tringe et al. 2008 ). Also, the fact that activated sludge microbiomes are characterized by high microbial density and high levels of various HGT associated traits (e.g., AR-related genes and plasmids/integrons/transposons) ( Schluter et al. 2007 ; Zhang et al. 2011 ) indirectly supports the observed clustering of sludge waste samples together with microbes from other (diverse) ecological niches (e.g., clinical environment). Similarly, Cluster 2 contains ecosystems that embed overlapping microbial communities (i.e., biotransformation, bioremediation, and soil environments) and thus showing similar patterns of interconnections against microbes from other ecosystems. Exceptions to ecologically homogeneous clusters can be highlighted within our data set. Two paradigmatic examples of cross-habitat putative HGT were chosen in the overall putative HGT network and are shown in figure 6 . In detail, figure 6a reports putative HGTs among contigs embedding tetracycline resistance determinants ( tet 34) in samples isolated from host and inland waters. Tetracycline resistance is often associated with conjugative transposons or other transferable elements (e.g., pheromone-inducible plasmids) ( Clewell et al. 1995 ; Dunny et al. 1995 ) and plasmid-mediated HGT events involving such determinants have been previously identified ( Fondi and Fani 2010 ; Bosi et al. 2011 ). Similarly ( fig. 6 b ), contigs embedding chloramphenicol resistance determinants belong to samples of very different origin (soil and host). This latter finding shows possible pathways for cross-habitat chloramphenicol-resistance propagation in the environment and is in line with previous observations on swine feedlot wastewater as a possible source of chloramphenicol-resistance genes ( Li et al. 2013 ) and the overall capability of this class of genes to undergo HGT ( Sermonti et al. 1978 ; Takamatsu et al. 2003 ). Taken together, these two cases show that interhabitat barriers and taxonomic distance can be overcome by certain genes since phylogenetically unrelated bacteria, and those inhabiting distinct environments were found to share common antibiotic resistance determinants, probably as a result of (one or multiple) HGT event( s) (Halary et al. 2010 ; Smillie et al. 2011 ).\n F ig . 6.— Examples of putative cross-habitat HGT events among contigs (nodes) embedding ( A ) tetracycline resistance determinants and retrieved from inland waters (blue nodes) and host (red nodes) and ( B ) chloramphenicol resistance in host (red) and soil (yellow) derived samples. The network-based approach adopted here allows testing the role of plasmids and chromosomes in the overall gene exchange pattern within environmental samples. Indeed, the importance of plasmids and chromosomes in shaping the microbial HGT network has been assessed in recent works ( Halary et al. 2010 ; Smillie et al. 2011 ). Halary et al. (2010) showed that gene sharing mostly occurs among molecules of the same type (molecule coherence), meaning that plasmid-plasmid and chromosome–chromosome gene sharing is more frequent than cross-molecule sharing. Accordingly, we investigated whether contigs embedded in the same CC belonged to the same or different molecules (i.e., plasmids or chromosomes). Contig sequences were assigned to their source molecule adopting a composition-based strategy as implemented in cBar ( Zhou and Xu 2010 ) and the source molecule composition of each cluster was evaluated. Results reported in figure 4b show an overall coherence within the CCs identified in the network. In particular, 5,199 CCs (∼65% of all the CCs) are highly homogeneous: more than 90% of the embedded contigs belong to the same type of DNA molecule. Conversely, heterogeneous clusters (those in which contigs are almost evenly distributed among the two types of molecules) represent 24.3% of the total number of clusters. Again, the observed distribution of homogeneous clusters was compared against the same (averaged) distribution obtained from 1,000 networks, obtained through label reshuffling (red line in fig. 4 b ). The distinctness of the two distributions was assessed by a Mann–Whitney U test ( P value < 2.2e-16). This finding indicates that DNA pools are mainly transferred between molecules of the same type. Notably, general trends (i.e., molecule and habitat coherence) among the various clusters were not affected by the method used for estimating the number of HGT events as adopting a composition-based (i.e., tetranucleotide frequencies, see Materials and Methods) approach led to the same overall results (data not shown)."
} | 7,812 |
26568784 | PMC4644332 | pmc | 4,488 | {
"abstract": "Strain K22 T is the type species of the recently- described genus Pyrinomonas , in subdivision 4 of the phylum Acidobacteria (Int J Syst Evol Micr. 2014; 64(1):220–7). It was isolated from geothermally-heated soil from Mt. Ngauruhoe, New Zealand, using low-nutrient medium. P. methylaliphatogenes K22 T has a chemoheterotrophic metabolism; it can hydrolyze a limited range of simple carbohydrates and polypeptides. Its cell membrane is dominated by iso- branching fatty acids, and up to 40 % of its lipid content is membrane-spanning and ether lipids. It is obligately aerobic, thermophilic, moderately acidophilic, and non-spore-forming. The 3,788,560 bp genome of P. methylaliphatogenes K22 T has a G + C content of 59.36 % and contains 3,189 protein-encoding and 55 non-coding RNA genes. Genomic analysis was consistent with nutritional requirements; in particular, the identified transporter classes reflect the oligotrophic nature of this strain. Electronic supplementary material The online version of this article (doi:10.1186/s40793-015-0099-5) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions Acidobacteria is one of the most widely-distributed bacterial phyla, particularly in soils [ 30 – 32 ]. Despite the wide distribution, the number of cultivated and sequenced representatives within most subdivisions within Acidobacteria remains low [ 33 ]. The sequencing and annotation of the P. methylaliphatogenes K22 T genome presented here links the phenotypic traits of P. methylaliphatogenes K22 T [ 4 ] with its genetic characteristics, and represents a step that will assist future studies describing the ecological and metabolic capabilities of this widespread phylum.",
"introduction": "Introduction Phylotypes from the phylum Acidobacteria 1 are commonly detected across a range of ecosystems, including marine and freshwater bodies, sediments, geothermal systems, and soils. Despite the apparent ubiquitous distribution acidobacterial phyotypes, particularly in soil environments, only 17 acidobacterial genera (represented by formal description and publication of respective type strains, in accordance with the International Code of Nomenclature of Prokaryotes [ 1 ]) have been validly published [ 2 , 3 ]. Here we present a description of the complete genome sequence and annotation of Pyrinomonas methylaliphatogenes strain K22 T (= DSM 25857 = ICMP 18710 ), the type species of the genus Pyrinomonas within subdivision 4 of Acidobacteria . Pyrinomonas methylaliphatogenes K22 T was isolated from a fumarole on the outer crater rim of the stratovolcano Mt. Ngauruhoe [ 4 ]. It exhibits a Gram-negative cell wall, is non-spore-forming, and is catalase- and oxidase-positive (Table 1 ). It is a thermophilic and moderately acidophilic obligately aerobic chemoorganotroph. Of particular note is its unusual lipid composition that is dominated by odd-numbered saturated iso- branching fatty acids ( iso- C 15:0 , iso- C 17:0 , iso- C 19:0 and iso- C 21:0 that total >88.5 % of the total fatty acid extract) [ 4 ]. In addition, >40 % of the total membrane lipid content is made up by iso- branching glycerol ether analogues of the cellular fatty acids and membrane-spanning iso- diabolic acids [ 5 ]. Membrane-spanning and ether lipids occur ubiquitously in Archaea , but in recent studies have also been commonly detected in cultivated representatives in subdivision groups 1, 3 and 4 of Acidobacteria [ 5 , 6 ]. Table 1 Classification and general features of P. methylaliphatogenes K22 T \n MIGS ID Property Term Evidence code a \n Current classification Domain Bacteria \n TAS [ 35 ] Phylum Acidobacteria \n TAS [ 36 ] Class ‘Insertae sedis 99’ Order ‘Insertae sedis 100’ Family ‘Insertae sedis 101’ Genus Pyrinomonas \n TAS [ 4 ] Species Pyrinomonas methylaliphatogenes \n TAS [ 4 ] Type strain K22 T (=DSM 25857 T =ICMP 18710 T ). TAS [ 4 ] Gram stain negative TAS [ 4 ] Cell shape rod TAS [ 4 ] Motility non-motile TAS [ 4 ] Sporulation non-sporulating TAS [ 4 ] Temperature range thermophilic (50–69 °C) TAS [ 4 ] Optimum temperature 65 °C TAS [ 4 ] pH range moderately acidophilic (4.1–7.8) Optimum pH 6.5 Carbon source peptides, proteins, carbohydrates TAS [ 4 ] Terminal electron receptor oxygen TAS [ 4 ] Energy metabolism chemoorganotroph TAS [ 4 ] MIGS-6 Habitat geothermal soil TAS [ 37 ] MIGS-6.3 Salinity non-halophile (no growth > 1 % (w/v) NaCl) TAS [ 4 ] MIGS-22 Oxygen requirement obligate aerobe TAS [ 4 ] MIGS-15 Biotic relationship free-living TAS [ 4 ] MIGS-14 Pathogenicity not reported NAS MIGS-4 Geographic location Mt Ngauruhoe, New Zealand TAS [ 37 ] MIGS-5 Sample collection 2006 NAS MIGS-4.1 MIGS-4.2 Latitude – Longitude 39° 9’25.31”S - 175° 38’6.74”E IDA MIGS-4.3 Depth not reported IDA MIGS-4.4 Altitude 2,270 m IDA \n a Evidence codes - IDA inferred from direct assay, TAS traceable author statement (i.e., a direct report exists in the literature), NAS non-traceable author statement (i.e., not directly observed for the living, isolated sample, but based on a generally accepted property for the species, or anecdotal evidence). These evidence codes are from the Gene Ontology project [ 38 ] Subdivision 4 of the Acidobacteria has five validly-named species: P. methylaliphatogenes K22 T ,[ 4 ] Chloracidobacterium thermophilum [ 7 , 8 ], Blastocatella fastidiosa [ 9 ], Aridibacter famidurans , and Aridibacter kavangonensis [ 3 ]. The latter three species are phylogenetically distant from P. methylaliphatogenes K22 T , are mesophilic and have differing pH ranges and substrate utilization profiles from that of P. methylaliphatogenes K22 T . Chloracidobacterium thermophilum is a moderately thermophilic facultatively anoxygenic photoheterotroph isolated from a hotspring microbial mat at Yellowstone National Park [ 7 , 8 ]. An additional strain, Ellin6075 was isolated from an Australian pasture soil, and is a mesophilic heterotroph that derives its energy from complex carbohydrate sources, but has little information available regarding its phenotypic traits [ 10 ]. Common features shared by subdivision 4 strains include an aerobic and heterotrophic phenotype [ 3 , 4 ], and membrane lipid iso -diabolic acids [ 5 ]."
} | 1,566 |
32708813 | PMC7407388 | pmc | 4,489 | {
"abstract": "Microalgae have a wide industrial potential because of their high metabolic diversity and plasticity. Selection of optimal cultivation methods is important to optimize multi-purpose microalgal biotechnologies. In this research, Chlorella sorokiniana AM-02 that was isolated from a freshwater lake was cultured under various high photosynthetic photon flux density (PPFD) conditions and CO 2 gas levels in standard Bold’s basal medium (BBM). Furthermore, a wide range of nitrate levels (180–1440 mg L −1 ) was tested on the growth of C. sorokiniana . Microalgae growth, pigment concentration, medium pH, exit gas composition, as well as nitrate, phosphate, and sulfate levels were measured during an experimental period. The preferred high PPFD and optimal CO 2 levels were found to be 1000–1400 μmol photons m −2 s −1 and 0.5–2.0% ( v / v ), respectively. The addition of nitrate ions (up to 1440 mg L −1 ) to the standard growth medium increased final optical density (OD 750 ), cell count, pigment concentration, and total biomass yield but decreased the initial growth rate at high nitrate levels. Our findings can serve as the basis for a robust photoautotrophic cultivation system to maximize the productivity of large-scale microalgal cultures.",
"conclusion": "4. Conclusions A study was conducted comparing the effect of various PPFD and CO 2 concentration values on the growth of the new isolated strain C. sorokiniana strain AM-02. Thus, the photoautotrophic cultures of C. sorokiniana in modified BBM (720 mg L −1 of NO 3 − ) showed better growth performance. It was found that the preferred high PPFD and CO 2 concentration values are 1000–1400 μmol photons m −2 s −1 and 0.5–2.0%, respectively. In addition, the high growth of C. sorokiniana AM-02 makes this strain particularly suitable for the rapid production of biomass, attractive for bioremediation of aquatic environments contaminated with high levels of nitrate, phosphate, and sulfate. Moreover, our results show that adding an additional nitrogen source can significantly increase final biomass and pigment yield compared to standard BBM. Finally, our findings prove that the new isolated strain C. sorokiniana AM-02 is a fast-growing microalgae strain that can also be used in industry.",
"introduction": "1. Introduction Microalgae-based biotechnologies have received much attention worldwide, and they are presented as complex processes for purification of air and wastewater as well as for the production of substantial amounts of lipids, polysaccharides, proteins, commercially important pigments (such as lutein and astaxanthin), and other useful products. Pharmaceuticals for various purposes, nutraceuticals, fertilizers, biodiesel, bioethanol, and biogas are obtained from different microalgae species [ 1 , 2 , 3 , 4 ]. Currently, individual microalgae that are capable of adsorbing or transforming certain pollutants have become excellent candidate species for simultaneously capturing carbon dioxide and removing nutrients, hydrocarbons, pesticides, and cyanide compounds [ 1 , 5 ] as well as for remediation of aquatic ecosystems polluted with heavy metals [ 6 ]. The practical application of microalgae metabolic potential can be limited by certain factors, such as photoinhibition and a slow response to different light irradiance, which lead to a low yield of biomass or target products. To increase the cell growth rate, the efficiency of biomass production, and the yield of target products, which means the effectiveness of various microalgae-based technologies, light intensity, photoperiod, temperature, pH, adequacy of macro- and micronutrients, and cultivation regimen (autotrophy, heterotrophy, and mixotrophy) must be thoroughly studied [ 4 , 7 , 8 , 9 ]. Multifarious microalgae species, including algae of the genus Chlorella ( Chlorophyta ; Trebouxiophyceae ; Chlorellales ; Chlorellaceae ), have been studied in various experimental projects to assess their potential for nutrient bioextraction, bioremediation, and bioenergy processes [ 7 , 10 ]. Species in the genus Chlorella are thought to have high potential for industrial use because of their fast and stable growth characteristics [ 11 , 12 , 13 ]. Chlorella sorokiniana is considered to be one of the promising Chlorella species that is used for wastewater treatments as well as for producing valuable products, including lipids, polysaccharides, proteins, and pigments [ 10 , 14 , 15 , 16 ]. Chlorella sorokiniana UTEX 1230 was originally isolated from a stream in Texas in 1951 by Dr. Constantine Sorokin [ 17 ], and more detailed information about this species of Chlorella was presented in subsequent research studies [ 18 , 19 , 20 ]. C. sorokiniana can grow effectively in a wide temperature range (25–40 °C) and under different light intensities [ 11 , 18 , 19 , 20 ]. Representatives of freshwater microalgae have the potential for large-scale cultivation. Researchers have developed various methods of cultivation depending on further practical application goals for the object that is being studied. In photoautotrophic regimens, microalgae cultures assimilate inorganic carbon, and light intensity can significantly affect CO 2 assimilation and biomass production [ 3 , 10 , 21 ]. Among cultivation trends for various industrially attractive microalgae, attention to heterotrophic (substrate dependent) or mixotrophic cultivation (auto- and heterotrophic) is given, which explains the choice with a higher growth rate and lower operating costs of green microalgae [ 7 , 11 , 15 ]. Although microalgae cultivation under heterotrophic and mixotrophic conditions has some advantages, this process is at risk of contamination with heterotrophic microbes because organic compounds are used as carbon and energy sources [ 22 ], and this leads to the release of carbon dioxide as one of the major greenhouse gases. The use of gaseous CO 2 as a carbon source by photoautotrophic microalgae contributes to sequestration of carbon dioxide, which is one way to reduce the risks that are associated with global warming [ 23 , 24 ]. Furthermore, when synthesis of the target products is the cell’s physiological response to high light stress, it is important to develop and apply effective strategies for photoautotrophic cultivation [ 10 ]. Therefore, additional studies are needed on the characteristics of algae metabolism at different light intensities and CO 2 concentrations for each microalgae culture. Moreover, the isolation of new strains with high rates of nitrogen, phosphorus, and sulfur removal is important for the development of new wastewater treatment technologies. Despite recent success in this research area, it is important to develop novel methods and optimize previous methods of growing and harvesting of microalgae so that many attractive species for biotechnology can be grown in a short time and with less corresponding energy use. C. sorokiniana as one of the attractive species for wastewater treatment was selected for the current study. Thus, this study evaluated the optimal high photosynthetic photon flux density (PPFD) conditions and CO 2 levels on the growth rate and the ability to utilize important nutrients, such as nitrate, phosphate, and sulfate, by indigenous C. sorokiniana AM-02 that tolerates high irradiance. In addition, based on the optimal growth parameters for C. sorokiniana , the influence of additional nitrogen and sulfur sources on the biomass productivity was investigated. The results presented here offer a new strategy for improving productivity and decreasing resource costs.",
"discussion": "3. Results and Discussion 3.1. Effects of PPFD and CO 2 Concentration Values on Growth of Algae In this research study, various PPFD and CO 2 concentration values as well as nutrient requirements for maximum growth and biomass accumulation of C. sorokiniana AM-02 were determined. Phylogeny of the strain AM-02 based on rbcL gene is presented in Figure 1 . Based on these data, this strain was assigned to C. sorokiniana . To control the algal growth, OD 750 was measured and number of cells was counted (results are summarized in Figure 2 ). Three high PPFD conditions were chosen (1000, 1200, and 1400 μmol m −2 s −1 ) for the experiments. We chose high PPFD conditions, because we plan to continue testing the growth of this strain in digested agricultural waste materials (such as pig manure and chicken manure), which are dark in color. In experiments, when cultures were bubbled with air supplemented with final 0.5%, 1.0%, and 2.0% CO 2 (for all tested PPFD conditions), a 1.7–2.4-fold increase in growth in standard BBM was observed compared to that detected in treatments with non-CO 2 augmented cultures. The mean OD 750 values of C. sorokiniana grown at 1000, 1200, and 1400 μmol photons m −2 s −1 and sparged with air after 112 h of cultivation were 2.52, 2.46, and 2.44, respectively, whereas in experiments with the addition of extra CO 2 , the growth values were above 4.3 ( Figure 2 a–c). The cultures in experiments supplied with different concentrations of carbon dioxide (under different PPFD conditions) increased in cell number until stationary growth was achieved at 1.22–1.46 × 10 8 cells mL −1 (after 88 h), while in treatments supplied with atmospheric carbon dioxide, the cultures increased in cell number to 5.7–6.4 × 10 7 cells mL −1 only (after 112 h) ( Figure 2 d–f). Overall, OD 750 correlated with the calculated cell numbers ( Figure 2 ). The concentration of total pigments (chlorophyll a, b, and total carotenoids) also increased to 14–17 mg mL −1 with higher accumulation in treatments supplemented with 2% carbon dioxide but decreased thereafter ( Figure 2 g–i). Table 2 presents the tested regimens of cultivation and the average yield of dry biomass of microalgae after photoautotrophic cultivation. As can be seen from the cell density, cell number, average biomass yield of dry microalgae, as well as pigment content, the growth had the similar trends under all lighting conditions with an additional CO 2 supply; however, the growth of C. sorokiniana was slightly higher in experiments in which PPFD and CO 2 concentration values were 1200 μmol m −2 s −1 and 2.0%, respectively. In this regard, we used cultures grown at 1200 μmol photons m −2 s −1 and sparged with 2.0% carbon dioxide for further selection of optimal nutrient concentrations in modified BBM. At 1000 μmol photons m −2 s −1 and 0.5% CO 2 , the lower growth rate was additionally observed (in treatments with a changing level of CO 2 ). However, at 1400 μmol photons m −2 s −1 , the growth was almost identical regardless of the level of CO 2 addition (0.5%, 1.0%, or 2%). It is believed that, from an industrial point of view, transportation, storage, and delivery of gaseous carbon dioxide are expensive for large-scale algae production [ 12 ]. Therefore, the cultivation of C. sorokiniana AM-02 on an industrial scale can also be achieved with lower levels of CO 2 supplementation. In several previous studies, other PPFD conditions (100–500 μmol photons m −2 s −1 ) were tested for culturing other C. sorokiniana strains in flasks and bottles, and no significant differences were found between different conditions for photoautotrophic cultures [ 6 , 11 , 13 , 31 ]. Due to the shading of the cells one above the other, the design of photobioreactors is also important to ensure adequate light distribution, as reported by Leon-Vaz et al. [ 32 ]. An increase in light intensity is not a solution to overcome shading in many photoautotrophic cultures with a high cell density, since many microalgae cannot grow efficiently at extremely high light intensities. However, Cuaresma et al. [ 33 ] reported a high growth rate of C. sorokiniana CCAP 211/8K under continuous illumination of 2100 μmol photons m −2 s −1 using red light emitting diodes. In our research, we show that our strain has optimal cultivation conditions in standard BBM at a high surface irradiance between 1000–1400 μmol photons m −2 s −1 with Gro-Lux tubes in a 3.6 L photobioreactor with a working volume of 2.4 L. Figure 3 demonstrates the changes in pH and the percentage of released CO 2 under selected conditions. Thus, the pH increased from an initial ~6.6 to 10.4–10.9 under all lighting conditions in the non-CO 2 augmented cultures (autotrophic cultivation leads to hydroxyl formation, which increases the pH), whereas the pH initially dropped to values 6.1–6.4 but ultimately did not increase above 7.8, 7.3, and 7.0 under all lighting conditions with the addition of CO 2 up to 0.5%, 1.0%, and 2.0%, respectively ( Figure 3 a–c). During the light period, pH increased, while cultivation in the dark period led to a slight decrease in pH (in CO 2 augmented cultures). In the non-CO 2 augmented cultures, pH decreased substantially during the dark period. In the case of CO 2 release from the photobioreactor, a decrease and an increase in the CO 2 level were observed during the light period and the dark period under all lighting conditions, respectively ( Figure 3 d–f). The increase in pH found in the culture medium can be explained by the growth of C. sorokiniana and the uptake of carbon from dissolved HCO 3 − , leaving OH – ions. Similarly, higher pH levels in the non-CO 2 augmented cultures can be attributed to lower dissolved CO 2 and less dissociated H + ions [ 7 , 34 ]. In addition, sodium nitrate nutrition leads to an increase in pH of the medium [ 35 ]. Figure 4 shows the nutrient levels in the BBM with nitrate levels of about 180 mg L −1 at the beginning of cultivation. After 64 h and 88 h, nitrogen was completely utilized in standard BBM in the CO 2 augmented and the non-CO 2 augmented cultures, respectively, and, therefore, decreases in the growth of the microalgae cultures and the pigment concentration were noted. It should also be reported that nitrate, orthophosphate, and sulfate were consumed faster by culture of C. sorokiniana AM-02 when air containing 2.0% CO 2 was continuously bubbled into the reactor. Ion chromatography analysis demonstrated that the added amounts of phosphate (around 160 mg L −1 ) and sulfate (around 35 mg L −1 ) slightly decreased but were in excess during the entire experimental period, while the nitrate concentration sharply decreased, and NO 3 – completely was utilized after 64–88 h of cultivation ( Figure 4 ). Thus, the concentration of NO 3 – was initially insufficient for the active growth and development of the culture in photoautotrophic treatments; therefore, the growth values were low, and the following experiments were performed to identify the effects of increasing NO 3 – content in the culture medium. These data confirm several reports in the literature on the rapid growth and rate of removal of nutrients but under other, different conditions [ 11 ]. In addition, some previous studies have demonstrated that various other C. sorokiniana strains can grow in wastewater with the effective removal of nitrates and phosphates [ 34 , 36 ], providing high potential for their application in developing wastewater treatment technologies. 3.2. Effects of Nutrient Concentration on Algal Growth In standard BBM (a medium that has been used to grow a variety of green algae cultures), NaNO 3 is used as a source of nitrogen. The growth of strain AM-02 in treatments containing 180 mg L −1 NO 3 – (standard concentration) under the selected PPFD and CO 2 concentration values (1200 μmol m −2 s −1 and 2% CO 2 ) was low, as measured by low cells number (1.47 × 10 8 cells mL −1 ), pigment concentration (16.8 mg L −1 ), and dry biomass weight (1.10 g L −1 ) ( Figure 5 ; Table 2 ). No NO 3 – was detected in standard BBM sparged with air containing 2.0% CO 2 on 66 h and onwards. Figure 5 shows that all key parameters were improved by adding an additional level of NO 3 – . In the 360 mg L −1 NO 3 – -treatments (double NO 3 – concentration), number of cells, pigment concentration, and dry biomass weight increased significantly ( p < 0.001; up to 1.9 × 10 8 cells mL −1 , 42.8 mg L −1 , and 1.76 g L −1 , respectively). However, even at a double concentration of introduced NO 3 – , nitrogen began to be limited after 66 h, and the cell concentration did not exceed 1.9 × 10 8 cells mL −1 . Under both conditions, PO 4 3– and SO 4 2– levels also decreased but remained abundant for culture growth ( Figure 5 ). A further increase in the concentration of the nitrogen source to 720 mg L −1 of NO 3 – led to a significant increase in the number of cells and was reflected in the yield of final dry biomass ( p < 0.001; Figure 5 a; Table 2 ). After 160 h, the culture increased in cell concentration until reaching stationary growth at 4.13 × 10 8 cells mL −1 . After 112 h, the concentration of total pigments increased to 82.1 mg mL −1 and remained almost stable during cultivation. As shown in Figure 5 b, the pigment concentrations in C. sorokiniana cultivated in the presence of high NO 3 – levels were generally higher than those observed in cells cultured at lower levels of nitrate ions. A clear positive correlation was observed between chlorophyll levels and the color of the microalgae culture. High concentrations of NO 3 – promoted the synthesis of pigments. Under these conditions, nitrate and sulfate were completely utilized by 88 h and 160 h, respectively ( Figure 5 d,f). Phosphate levels also decreased from an initial 160 to 66 mg L −1 by 160 h and remained at around these levels until the end of growth ( Figure 5 e). In 6×NO 3 – and 8×NO 3 – -treatments, the final biomass yield and pigment concentration were significantly higher than previously reported ( p < 0.001); however, the best growth rate was observed under 4×NO 3 – conditions. Since SO 4 2– was completely utilized under 4×NO 3 – and 6×NO 3 – conditions, we increased its level to 95 mg L −1 in 8×NO 3 – treatments. Under all conditions, nitrate, phosphate, and sulfate were efficiently utilized by culture of C. sorokiniana ( Figure 5 ). These results clearly demonstrate that C. sorokiniana AM-02 can grow at high nitrate concentrations and can be further tested for wastewater treatment. However, remarkably high NO 3 – concentrations added to the medium reduced the specific growth, which can be explained by increased osmotic stress and inhibition by the intermediate [ 37 ]. The pH increased to 7.1, 7.3, 7.6, 7.8, and 8.0 in 1× NO 3 – , 2× NO 3 – , 4× NO 3 – , 6× NO 3 – , and 8× NO 3 – -treatments, respectively ( Figure 5 c). In the case of CO 2 emission from the photobioreactor, a decrease and an increase in the CO 2 level were observed during the light period and the dark period, respectively, with an increase in the CO 2 level by the entire period in experiments with a high nitrogen content (from an initial 2.0% to 2.6%, data not shown). Therefore, based on our results, nitrogen and sulfur are the main nutrients that become insufficient in standard BBM, while the phosphorous level is sufficient to support the growth of microalgae. Our findings also demonstrate that high level of inorganic anions removal is possible (100% for NO 3 – and SO 4 2– and 91% for PO 4 3– (146 mg L −1 )), which indicates the potential use of the strain for wastewater treatment. Nutrient availability affects the growth of photoautotrophic organisms. One of the important components affecting the growth of microalgae and physiological activity is nitrogen [ 14 , 38 , 39 ]. For numerous applications of promising photosynthetic microorganisms, an increase in biomass productivity is considered to be an important step. Various potential nitrogen sources and their concentrations, such as urea, nitrate, and ammonium ions, have been evaluated by several other researchers for the cultivation of other C. sorokiniana strains. Lizzul et al. [ 34 ] observed that C. sorokiniana UTEX1230 prefers ammonium ions rather than nitrate ions as a source of nitrogen when cultured in wastewater and BBM medium. Ramanna et al. [ 36 ] reported that urea provided the highest biomass yield during cultivation of C. sorokiniana strain NIES: 2173 in wastewater. Kim et al. [ 7 ] found that growth rate and nitrogen and phosphorus removal rates by C. sorokiniana UTEX 1670 were higher under heterotrophic conditions than under autotrophic conditions. The wastewater nitrogen and phosphate were also effectively utilized by several other C. sorokiniana strains [ 40 ]. Another species of the same genus, C. vulgaris , also effectively removed nitrogen and phosphorus from domestic wastewater [ 41 ]. In this work, C. sorokiniana AM-02 was isolated from a freshwater lake and therefore can be adapted to different environmental changes and thus potentially provides a competitive advantage when grown in open water."
} | 5,221 |
21863015 | PMC3265368 | pmc | 4,490 | {
"abstract": "Over the last decade electrical batteries have emerged as a critical bottleneck for portable electronics development. High-power mechanical energy harvesting can potentially provide a valuable alternative to the use of batteries, but, until now, a suitable mechanical-to-electrical energy conversion technology did not exist. Here we describe a novel mechanical-to-electrical energy conversion method based on the reverse electrowetting phenomenon. Electrical energy generation is achieved through the interaction of arrays of moving microscopic liquid droplets with novel nanometer-thick multilayer dielectric films. Advantages of this process include the production of high power densities, up to 10 3 W m −2 ; the ability to directly utilize a very broad range of mechanical forces and displacements; and the ability to directly output a broad range of currents and voltages, from several volts to tens of volts. These advantages make this method uniquely suited for high-power energy harvesting from a wide variety of environmental mechanical energy sources.",
"discussion": "Discussion The results described above establish the fundamentals of energy production using REWOD process and clearly demonstrate that the energy-generation process can be readily scaled upwards to achieve high power output in excess of 1 W in a relatively small package. Indeed, scaling up the REWOD-based energy-generation process entails increasing the number of droplets working in parallel to generate electrical current. In this work, we have experimentally demonstrated that this scaling can be readily achieved over two orders of magnitude, from a single droplet to 150 droplets. In terms of the liquid-substrate area, the scaling was demonstrated over almost three orders of magnitude, from 0.28 mm 2 (single droplet in the sliding plates set-up) to 1.28 cm 2 (22 droplets in a channel set-up, each droplet having 5.8 mm 2 liquid-substrate overlap area). As the total capacitance of the system C linearly scales with the area, it was demonstrated to scale over three orders of magnitude as well, from 14 pF (single droplet in the sliding plates set-up, 5 nF cm −2 dielectric stack) to 20 nF (22 droplets in a channel set-up, 16 nF cm −2 dielectric stack). Devices with even larger number of droplets can be readily fabricated by exploiting a natural synergy between the REWOD process and droplet-based microfuidics 10 . Indeed, parallel actuation of a large number of micro-droplets required for scaling-up of the generated power is routinely performed in channel-based droplet macrofluidic devices, where thousands of droplets can be synchronously moved in microchannels with a great degree of control over their position and velocity 10 . Combination of REWOD and droplet microfluids offers important advantages such as easy scaling, very flexible force–displacement relationship, and extremely simple device design with no moving solid parts. Microfluidic power generators based on the REWOD process can take advantage of many previously inaccessible environmental mechanical energy sources. Two specific examples will be illustrated: energy harvesting from human locomotion and high-power harvesting of mechanical vibration energy. Energy harvesting from human locomotion using footwear-embedded harvesters is a long-recognized concept 1 11 12 13 . Data available in the literature indicate that up to 10 W per foot can be generated without adversely affecting a person's gait 13 . For comparison, relatively high-power mobile electronic devices, such as cell phones and mobile computers, typically consume power on the order of 1 and 15 W respectively. The following simple estimate illustrates the power that can be produced by a footwear-embedded microfluidic harvester using the REWOD process. Let us consider 2-m-long train of 1,000 conductive droplets, each 1 mm long separated by 1 mm spacers and positioned inside 1-mm-diameter circular cross-section channel with the total length of 4 m, Figure 4a . The total area covered by such channel is about 40 cm 2 or less than ¼ of the area of a typical human footprint. The total volume of the liquid contained in the channel would be about 4 ml, which makes it readily compatible with footwear. Assuming that the heel area is about 20 cm 2 , we estimate that the total midsole compression required to achieve 4 ml volume displacement is around 2 mm. Such a displacement is well below the level that might affect the person's gait. Let us consider the case of the film stack with the capacitance of 16 nF cm −2 . The average generated power calculated using equation (2) is shown in Figure 3b . The average power per foot can exceed 2 W for bias voltages in excess of 35 V and 10 W for bias voltages in excess of 75 V. The bias voltage can be substantially reduced by increasing the capacitance of the dielectric film stack. However, it is important to mention, that even at its current level the bias voltage does not present a substantial practical issue. A wide range of commercially available DC–DC boost converter components can be used to convert the 3.7 V output of standard Li-ion batteries to the required bias voltage. Thus, this example clearly supports the use of footwear designed for high-power-energy harvesting based on reverse electrowetting. The other common source for mechanical energy harvesting is vibration energy. It has been demonstrated that the energy of mechanical vibrations present in floors, stairs, vehicles and equipment housings can be used for electrical power generation 1 3 14 . Currently, the majority of experimental vibration harvesters have output power in the range from 10 −6 to 10 −2 W (refs 1 , 3 , 14 ). The REWOD process can enable the use of novel harvester architectures with greatly increased power output. One example of the REWOD-based vibration harvester device consists of an array of conductive droplets squeezed between two dielectric-coated electrodes, as shown in Figure 4b . The electrodes are separated by a millimeter-thick elastic spacer so that the resulting structure can be used as a mounting 'pad' for the load device. Mechanical vibration of the load device causes periodic change in the solid–liquid contact area and, thus, electrical current generation. For the film stack with a capacitance of 10 2 nF cm −2 , the resulting power density can be scaled up to 10 −1 W cm −2 at 50 Hz vibration frequency, thus enabling the fabrication of practical vibration harvesters with power output of several watts. The above examples illustrate new possibilities in portable high-power energy harvesting that can be opened by utilizing the REWOD process. High-power energy harvesting can potentially provide a valuable alternative to the use of batteries. Even though energy harvesting is unlikely to completely replace batteries in the majority of mobile applications, it can have a very important role in reducing cost, pollution, and other problems associated with battery use. We believe that the REWOD-based mechanical to electrical energy conversion process, which we have developed, can go a long way in achieving this goal."
} | 1,783 |
33980683 | PMC8125057 | pmc | 4,491 | {
"abstract": "Kara B. De Leόn works in the field of microbial ecology, environmental biofilms, and microbial genetics."
} | 26 |
21461344 | null | s2 | 4,493 | {
"abstract": "Polymer nanofibers exhibit properties that make them a favorable material for the development of tissue engineering scaffolds, filtration devices, sensors, and high strength lightweight materials. Electrospinning is a versatile method commonly used to manufacture polymer nanofibers. Collection of electrospun nanofibers across two parallel plates is a technique useful for creating nanofiber structures because it allows for the collection of linearly oriented individual nanofiber arrays and these arrays can be easily transferred to other substrates or structures. It is of importance to have some understanding of the capabilities of this collection method, such as the maximum length of fibers that can be collected across two parallel plates. The effect of different electrospinning parameters on maximum fiber length, average fiber diameter, diameter uniformity, and fiber quality was explored. It was shown that relatively long continuous polycaprolactone (PCL) nanofibers with average diameters from approximately 350 nm to 1 µm could be collected across parallel plates at lengths up to 35-50 cm. Experimental results lead to the hypothesis that even longer continuous nanofibers over 50 cm could be collected if the size of the parallel plates were increased. Extending the maximum fiber length that can be collected across parallel plates could expand the applications of electrospinning. Polymer solution concentration, plate size, and applied voltage were all shown to have varying effects on maximum fiber length, fiber diameter, and fiber uniformity."
} | 391 |
26568700 | PMC4623359 | pmc | 4,496 | {
"abstract": "Methanogens, a key contributor in global carbon cycling, methane emission, and alternative energy production, generate methane gas via anaerobic digestion of organic matter. The methane emission potential depends upon methanogenic diversity and activity. Since they are anaerobes and difficult to isolate and culture, their diversity present in the landfill sites of Delhi and marshlands of Southern Assam, India, was analyzed using molecular techniques like 16S rDNA sequencing, DGGE, and qPCR. The sequencing results indicated the presence of methanogens belonging to the seventh order and also the order Methanomicrobiales in the Ghazipur and Bhalsawa landfill sites of Delhi. Sequences, related to the phyla Crenarchaeota (thermophilic) and Thaumarchaeota (mesophilic), were detected from marshland sites of Southern Assam, India. Jaccard analysis of DGGE gel using Gel2K showed three main clusters depending on the number and similarity of band patterns. The copy number analysis of hydrogenotrophic methanogens using qPCR indicates higher abundance in landfill sites of Delhi as compared to the marshlands of Southern Assam. The knowledge about “methanogenic archaea composition” and “abundance” in the contrasting ecosystems like “landfill” and “marshland” may reorient our understanding of the Archaea inhabitants. This study could shed light on the relationship between methane-dynamics and the global warming process.",
"conclusion": "4. Conclusions In the sequencing of the molecular marker for archaeal diversity, 16S rDNA identified the orders, named as Methanobacteriales and Methanosarcinales in both landfill sites and the phylum Crenarchaeota (thermophilic) in marshland. Quantitative PCR indicated a higher abundance of methanogens in landfill compared to that of marshland sites. The knowledge about the composition and abundance of methanogenic Archaea in a landfill may provide information on the decomposition mechanism of municipal solid waste and the subsequent generation of methane. This information can be exploited for controlling methane emission from landfill by mitigation process. The increasing knowledge about the genomic content of microbes belonging to the phylum Thaumarchaeota (mesophilic) will enrich our understanding of their adaptative behavior in the transposition from thermophily to mesophily. This indicates whether they follow a similar or different evolutionary pattern with respect to the phylum, Euryarchaeota.",
"introduction": "1. Introduction Methane is an important greenhouse gas because it is 25 times more powerful than CO 2 in global warming potential (i.e., the ability of the gas to trap heat in the atmosphere) and thus plays a crucial role in climate change and carbon cycling [ 1 , 2 ]. Methane emission has contributed approximately 20% to global climate change from preindustrial times [ 1 , 3 ]. About 500–600 Tg of methane is emitted annually to the atmosphere of which 74% is biogenic, produced by methanogenic Archaea [ 4 ]. The methanogenic Archaea (methanogens) usually occurs in highly reduced, anoxic environments such as landfills, wetlands, rice fields, rumen, and marine sediments where they serve as a terminal electron sink [ 5 , 6 ]. Methanogens are strict anaerobes and the presence of oxygen leads to the formation of reactive oxygen species (ROS), which damage their cell membranes, DNA, and proteins [ 7 , 8 ]. Methanogens are phylogenetically divided into 5 families within the phylum Euryarchaeota and are comprised of 31 known genera [ 9 , 10 ]. Methanogens can utilize a wide range of compounds for methane production, but, in most natural systems, there are two major pathways for methanogenesis, reduction of CO 2 (hydrogenotrophic methanogenesis) and cleavage of acetates (acetoclastic methanogenesis). A third pathway for methane generation is called methylotrophic methanogenesis that occurs in marine sediments and salt lakes where methane is produced from methylated compounds such as trimethylamine [ 11 , 12 ]. Landfill sites are the third largest source of methane. It constitutes about 30 and 24% of the anthropogenic methane production in Europe and US, respectively [ 4 , 13 ]. In comparison to the western countries, the composition of municipal solid waste (MSW) in developing countries like India is higher (40–60%) in organic waste. This has more potential to emit higher GHGs (Green House Gases) per ton of MSW compared to the developed world [ 14 ]. Moreover, landfills in India are neither well planned nor engineered and are often found in low-lying open areas, where municipal waste is haphazardly and indiscriminately disposed. These sites have neither landfill lining to avoid percolation of leachate to groundwater table nor leachate collection facility. The city generates about 6000 tonnes of solid waste per day and the expected quantity of solid waste generation in Delhi would be about 12,750 tonnes per day by 2015 [ 15 ]. Due to scarcity of land in big cities, municipal authorities are using the same landfill for nearly 10–20 years. Thus, the possibility of anaerobic emission of GHGs further increases [ 16 ]. Microbial decomposition, climatic conditions, MSW wastes characteristics, and landfilling operations are among the many factors that contribute to the generation of methane [ 2 , 17 ]. The migration of gas and leachate away from the landfill boundaries and their release into the surrounding environment present serious environmental threats, including potential health hazards, fires and explosions, damage to vegetation, unpleasant odors, landfill settlement, ground water pollution, air pollution, and global warming [ 18 – 20 ]. Wetlands (marshland) are the largest source of natural methane emissions contributing about 10–231 Tg methane per year accounting for 20–39% of annual global CH 4 emission [ 4 , 21 ]. Methanogens in the moist, anoxic (oxygen-free) wetland soil produce CH 4 as they decompose dead plant material. The methane emission from wetland was increased by 7% from 2003 to 2007 [ 2 , 19 ]. Methane production in wetlands is affected by the acetate supply through acetate fermentation or the CO 2 reduction potential [ 22 , 23 ]. The exponential increase in the rate of CH 4 production with temperature is due to the availability of more substrates and is not associated with changes in the composition of methanogens [ 24 ]. Methanogens belonging to the groups Methanomicrobiales and Methanosarcinales performing acetoclastic and methylotrophic pathway were found to be dominant in landfill sites [ 25 – 27 ]. In acidic conditions, due to the presence of acid tolerant hydrogenotrophic methanogens, H 2 /CO 2 is efficiently converted to methane compared to acetate, and methanogenic activity decreases with decrease in pH regardless of the substrates [ 28 ]. The prokaryotic diversity in our planet dictates our planet's ecosystems by acting as key functional drivers [ 29 ]. The understanding of the functional potential of the most individual microbial flora residing within the ecosystem is extremely limited because of our inability to isolate and culture them in laboratory conditions [ 30 ]. Since the methanogens are anaerobes and are difficult to culture, they are identified by culture independent molecular techniques like PCR amplification, denaturing gradient gel electrophoresis (DGGE), and quantitative real-time PCR, using molecular markers such as 16S rDNA genetic locus [ 31 – 34 ]. Hence, the present study was aimed at detecting the methanogenic Archaea inhabitants (richness) (by DGGE), identification by DNA sequencing, and quantification by qPCR in both the landfill sites of Delhi and marshland sites of Southern Assam, India.",
"discussion": "3. Results and Discussions 3.1. Identification of Methanogenic Archaea in Landfill and Marshland Sequences of MET1 LAND and MET2 LAND obtained from the Bhalswa landfill site are clustered with Methanoculleus thermophilus methanogens belonging to the order Methanomicrobiales which are hydrogenotrophic in nature. Third sequence of MET3 LAND from the Bhalswa landfill site clustered with the Candidatus Methanomethylophilus alvus Mx1201 , which is H 2 -dependent methylotrophic methanogens. In Figure 1 , it is shown that these three sequences from the landfill sites of Delhi are clustered with Euryarchaeota cluster (Cluster I). Sequence METG1 LAND obtained from the Ghazipur landfill site, Delhi, clustered with Methanomassiliicoccus luminyensis (Cluster II). Sequences obtained from marshland sites of Southern Assam were clustered (Cluster III) separately with Crenarchaeota (Cluster IIIa) and Thaumarchaeota (Cluster IIIb). There are five more sequences from the landfill sites of Delhi. They are related to two different species of methanotrophs (methane oxidizing bacteria) (see Table 4 ), Methylobacillus arboreus (marked as grey triangle) and Methylobacillus flagellatus (marked as grey circle), and are clustered separately, as shown in Figure 1 . Phylogenetic analysis of 16S rDNA clones indicates the presence of methanogens belonging to the phylum Euryarchaeota, order Methanomicrobiales, Methanobacteriales-1, and seventh order of methanogens in the landfill sites [ 42 – 44 ]. Both Candidatus Methanomethylophilus alvus Mx1201 and Methanomassiliicoccus luminyensis represent a monophyletic lineage that is not phylogenetically associated with any of the previously known orders of methanogens or the anaerobic methanotrophic ANME1 lineage [ 43 , 45 ]. They belong to the Mx order clusters with two lineages: the planktonic Marine Group II (MG-II) and the sediment dwelling Marine Benthic Group D (MBG-D) [ 45 – 47 ]. The other five sequences from Ghazipur landfill sites revealed presence of methanotrophs belonging to class Betaproteobacteria , family Methylophilaceae. 16S rDNA clones obtained from marshland sites of Southern Assam revealed a cluster of Archaea that are distantly related to two different phyla, Crenarchaeota and Thaumarchaeota. Microorganisms belonging to the phylum Thaumarchaeota (recently proposed) are thermophilic and mesophilic in nature and are found to be present in a wide variety of ecosystems, including marine and fresh waters, soils, and also hot environment [ 44 , 48 – 52 ]. 3.2. Culture Independent Molecular Analysis of Methanogenic Diversity Microbes dominated in the history of living organisms and they are a fundamental part of the biosphere. The study of microbial diversity has been, therefore, essential for understanding the evolution of life. Traditionally, cultivation based methods have contributed to our knowledge about their whereabouts and diversity of microbes in naturally occurring communities. However, only a small fraction of the prokaryotes has been cultivated in vitro by standard methods. Therefore, this knowledge may not reveal the actual composition and/or diversity associated with an ecosystem [ 31 , 33 ]. In the present study, we used culture independent molecular techniques like 16S rDNA PCR, cloning-sequencing, DGGE, and qPCR for estimation of the richness and diversity of the methanogenic Archaea in the landfill site of Delhi and marshland areas of Southern Assam. These techniques are widely used for molecular community analysis of microbes present in various types of habitats [ 32 , 42 , 53 – 56 ]. A combination of DNA sequencing, DGGE, and quantitative PCR (qPCR) can provide valuable information about microbial consortia associated with a specific ecosystem. Denaturing gradient gel electrophoresis (DGGE) is used to determine the genetic diversity of microbial communities. The procedure is based on electrophoresis of PCR-amplified 16S rDNA fragments in polyacrylamide gels containing a linearly increasing gradient of denaturants. In DGGE, DNA fragments of the same length but with different base-pair composition can be separated. Separation is based on the electrophoretic mobility of partially melted DNA molecules in a polyacrylamide gel and resulting into a band pattern [ 57 – 60 ]. DGGE can reveal 1-2% of the actual diversity present in the samples [ 61 ]. 3.3. Estimation of Methanogenic Richness by Quantitative Real-Time PCR DNA extracted from the three sampling points, that is, two landfill sites Okhla and Bhalswa of Delhi and Silcoorie Lake (Silchar) of Southern Assam, was screened for the quantification of methanogens. The copy number of all methanogens (pure culture) was higher in the two landfill sites than that of marshland in Southern Assam ( Table 5 ). Methanogenic pathway associated with the methanogens order and its reactions involved in the process are included in Table 6 . The copy numbers of Methanomicrobium mobile belonging to the order Methanomicrobiales and Methanobacterium bryantii belonging to the order Methanobacteriales-1 were found to be higher in both landfill sites in comparison to the Silcoorie Lake (Silchar) of Southern Assam. Copy number of Methanobrevibacter arboriphilus (order Methanobacteriales-1) and Methanosarcina mazei (acetoclastic) (order Methanosarcinales) was found to be higher in the Bhalswa landfill site than Okhla landfill site and Silcoorie Lake (Silchar) marshland site. The value of R \n Sq and slope dR for standard curve was 0.948 and −2.641, and the efficiency of the reaction was 139.1%. The R \n Sq and slope dR values for “absolute” quantification of Methanobrevibacter arboriphilus are 0.903 and −2.128. R \n Sq ( dR ) and slope dR values for this quantification of Methanobacterium bryantii are 0.877 and −1.384. R \n Sq ( dR ) and slope dR values of Methanomicrobium mobile are 0.956 and −2.563, respectively. The values of R \n Sq ( dR ) and slope dR for Methanosarcina mazei were 0.394 and −2.051, respectively. Methanogens pertaining to both acetotrophic and hydrogenotrophic decomposition pathways were detected in MSW landfills, which have been reported earlier [ 25 , 26 , 42 ]. Acetate serves as a precursor for more than 70% of CH 4 (methane) formation in the most anaerobic digestion process [ 62 ]. Therefore, acetoclastic methanogens, which utilize acetate as substrate, play a key role in stabilizing the pollution load of wastewater by methanogenesis. In the present study, quantitative PCR indicates the higher methanogenic richness in both landfill sites of Delhi compared to marshland of Silcoorie Lake, Silchar. 3.4. Diversity of Methanogenic Archaea by Denaturing Gradient Gel Electrophoresis Abundance and diversity of methanogenic Archaea were studied in three landfill and four marshland sites situated at different location in Delhi and Southern Assam, India. 16S rDNA amplicons were cloned and then analyzed on the DGGE gel for estimation of the archaeal richness in respective samples as shown in Figure 2 . Band patterns of 16S rDNA amplicons obtained from the landfill sites (OK, BH, and GZ) of Delhi and marshland samples (SON, SIL, KRM, and BDR) of Southern Assam were compared for methanogens richness and diversity analysis using Gel2K software. Analysis of DGGE image revealed the presence of total 38 bands. There are some unique bands in each lane, which indicates the variation of methanogens community residing in those particular samples. Cluster analysis of bands using Jaccard analysis indicated the presence of three main clusters consisting of localities that differ in number of similarity versus DGGE bands ( Figure 3 ). In the first cluster, Badarpur beetle-nut pond and Silcoorie Lake (Silchar) of Southern Assam clustered together showing similar band pattern. In the second cluster, interestingly, despite being two different ecosystems, Ghazipur landfill sites of Delhi clustered with wetland of Sonbill, Southern Assam, India. In the third cluster, the two landfill sites of Delhi (Okhla and Bhalswa) clustered together showing similar band pattern. In terms of richness, number of bands from the respective samples from Bhalswa landfill and Sonbill wetland have maximum of 11 bands, followed by Ghazipur landfill site and Silcoorie Lake (Silchar) having 10 bands each. Okhla landfill and Badarpur beetle-nut pond showed 9 bands each in the cluster. In the Karimganj rice paddy field sample, only four bands were observed showing the least diversity. Microbial diversity within contaminated ecosystems like landfill should be less diverse than those in natural systems like a wetland because the diversity may be influenced by the complexity of toxic chemical mixtures, heavy metals present, and duration of time the populations have been exposed. In the present study, after analyzing DGGE gel banding pattern and the number of bands, we found that the methanogenic diversity present in both landfills (anthropogenic system) and marshland (natural) is quite similar, except for the samples obtained from the Karimganj rice paddy field where only four bands appeared. The number of total bands observed in this study was in accordance with the number of DGGE bands reported previously [ 42 , 51 , 56 , 57 ]. It strongly indicates that the methanogenic Archaea diversity in both landfill and marshland is influenced by sampling location rather than type."
} | 4,280 |
27399705 | PMC4970092 | pmc | 4,497 | {
"abstract": "In this paper, we study energy harvesting from the mouse click motions of a robot finger and a human index finger using a piezoelectric material. The feasibility of energy harvesting from mouse click motions is experimentally and theoretically assessed. The fingers wear a glove with a pocket for including the piezoelectric material. We model the energy harvesting system through the inverse kinematic framework of parallel joints in a finger and the electromechanical coupling equations of the piezoelectric material. The model is validated through energy harvesting experiments in the robot and human fingers with the systematically varying load resistance. We find that energy harvesting is maximized at the matched load resistance to the impedance of the piezoelectric material, and the harvested energy level is tens of nJ.",
"conclusion": "5. Conclusions In this paper, we have analyzed energy harvesting from mouse click motions using a piezoelectric material. We have developed a mathematical model for the electromechanical behavior of the system to predict the energy harvested from the finger motion during the mouse clicks. The finger motion has been described as a three degree of freedom mechanism consisting of three links in a plane. The electromechanical coupling of the piezoelectric material has been considered a relationship between its stored charge and rotation angle. To validate the modeling framework, we have conducted experiments about harvested energy at the varying load resistances. As the test subject, we have used a robot finger and a human index finger. The energy harvester has been attached to the both fingers by using a glove with a pocket. Theoretical predictions of energy harvesting have been found to be in good agreement with experimental results for the both human and robot fingers, corroborating the validity of the proposed modeling approach. Our results indicate that energy harvesting is optimized when the load resistance matches the impedance of the piezoelectric material for the fundamental harmonic, and the maximum harvested energy is in the range of 1– 10 nJ . We anticipate that the energy level can be improved by using multiple layers or alternative smart materials with high efficiency. We expect that the experimental results and the modeling framework presented in this study can find application in the design of self-powered mouse or wearable devices through energy harvesting from human motions.",
"introduction": "1. Introduction Recent advancements and developments in the area of wearable devices have stimulated the demand for energy harvesting system [ 1 , 2 , 3 ]. Various energy sources, such as heat [ 4 ] and motions [ 5 , 6 , 7 ], in human body can be converted into useful electric energy by using energy transducers. In this context, energy harvesting to support the power of wearable devices using the human body energy sources can offer several benefits: permanent lifetime and weight reduction through the needlessness of batteries [ 8 ]. Piezoelectric materials are a good solution as energy transducer [ 9 , 10 ]. Especially, Lead zirconate titanate (PZT) [ 11 , 12 , 13 ], Polyvinylidene fluoride (PVDF) [ 14 , 15 , 16 , 17 ], Macro-fiber composite (MFC) [ 18 , 19 , 20 , 21 ], Aluminum nitride (AlN) [ 22 ], and Zinc oxide (ZnO) [ 23 ] have been used as typical energy transducers. They convert the kinetic energy of surrounding environment into electric energy. For example, energy harvesting from PVDF when subjected to various wind speeds and water droplets has been evaluated in [ 14 ]. The feasibility of energy generation of PVDF cantilever with a magnetic mass has been experimentally studied in [ 15 ]. PVDF is the most flexible piezoelectric material among them although it has the lowest electromechanical coupling [ 24 , 25 ]. The flexibility can offer important advantage in the development of wearable devices by reducing the inconvenience of wearing. Here, we theoretically and experimentally study energy harvesting using the flexible energy transducer, PVDF, attached to robot and human index fingers during mouse click motions. The ability of harvesting energy from mouse click motions may be applicable to the development of self-powered computer mouse [ 26 ], wearable mouse glove [ 27 ], or hand motion recognition device [ 28 ]. Various researches about energy harvesting using human motions have been reported. For instance, energy harvesting from foot strike during human walking by using shoes including PZT and PVDF has been introduced in [ 5 ]. In [ 29 ], human walking motion has been also used for energy harvesting by using a backpack instrumented with piezoelectric shoulder straps. In [ 30 ], a device using plucked piezoelectric bimorphs for energy harvesting from knee motions during human walking has been reported. Energy harvesting on human limb motions has been demonstrated in [ 6 ]. In [ 7 ], energy harvesting using the jaw movements of human through the piezoelectric chin strap has been demonstrated. Energy harvesting from two different piezoelectric transducers attached to the human body for five human activities has been studied in [ 31 ]. Interestingly, a glove to harvest electric energy from the kinetic energy of the fingers has been reported in [ 32 ]. Therein, four couples of piezoelectric transducers have been integrated into the glove in correspondence with the fingers joints, but the finger motions for energy harvesting have not been specified. Moreover, ZnO based energy harvester using stretched and released states of index finger has been demonstrated in [ 33 ]. In this work, we focus on the mouse click motions of fingers for energy harvesting. Specifically, we model the mouse click motions and the energy transducing of the piezoelectric material. The proposed model is validated through experiments using robot and human index fingers. The energy transducer, PVDF, is attached to fingers by wearing a glove with a pocket for piezoelectric material. In the robot finger, we can reduce the effect by the variation of the human motion, and quantitatively analyze the energy harvesting from the mouse click motions. From a practical point of view, this work addresses the untapped research question of energy harvesting from mouse click motions. From a methodological point of view, the main contributions of this effort are: (i) developing an electromechanical model to study energy harvesting of flexible piezoelectric materials bent by mouse click motions; (ii) performing a thorough experimental campaign to validate the proposed modeling framework in various conditions, whereby one and double click motions, human and robot fingers, and shunting load resistance; and (iii) conducting a systematic analysis of its energy harvesting capacity. This paper is organized as follows. In Section 2 , we introduce the proposed modeling framework, including the inverse kinematic framework of parallel joints in a finger and the electromechanical coupling for the energy transducer. In Section 3 , we describe the experimental scheme developed to study energy harvesting from the mouse click motions. In Section 4 , we report and discuss the experimental results toward the validation of the model and the analysis of harvested energy from the mouse click motions of a robot finger and a human index finger. Conclusions are summarized in Section 5 ."
} | 1,839 |
25450192 | null | s2 | 4,498 | {
"abstract": "Advances in reading, writing and editing genetic materials have greatly expanded our ability to reprogram biological systems at the resolution of a single nucleotide and on the scale of a whole genome. Such capacity has greatly accelerated the cycles of design, build and test to engineer microbes for efficient synthesis of fuels, chemicals and drugs. In this review, we summarize the emerging technologies that have been applied, or are potentially useful for genome-scale engineering in microbial systems. We will focus on the development of high-throughput methodologies, which may accelerate the prototyping of microbial cell factories."
} | 160 |
21481583 | null | s2 | 4,499 | {
"abstract": "Evolution results from molecular-level changes in an organism, thereby producing novel phenotypes and, eventually novel species. However, changes in a single gene can lead to significant changes in biomolecular networks through the gain and loss of many molecular interactions. Thus, significant insights into microbial evolution have been gained through the analysis and comparison of reconstructed metabolic networks. However, challenges remain from reconstruction incompleteness and the inability to experiment with evolution on the timescale necessary for new species to arise. Despite these challenges, experimental laboratory evolution of microbes has provided some insights into the cellular objectives underlying evolution, under the constraints of nutrient availability and the use of mechanisms that protect cells from extreme conditions."
} | 212 |
24478193 | null | s2 | 4,500 | {
"abstract": "Many bacteria regulate gene expression through a cell-cell signaling process called quorum sensing (QS). In proteobacteria, QS is largely mediated by signaling molecules known as N-acylated L-homoserine lactones (AHLs) and their associated intracellular LuxR-type receptors. The design of non-native small molecules capable of inhibiting LuxR-type receptors (and thereby QS) in proteobacteria is an active area of research, and numerous lead compounds are AHL derivatives that mimic native AHL molecules. Much of this previous work has focused on the pathogen Pseudomonas aeruginosa, which controls an arsenal of virulence factors and biofilm formation through QS. The MexAB-OprM efflux pump has been shown to play a role in the secretion of the major AHL signal in P. aeruginosa, N-(3-oxododecanoyl) L-homoserine lactone. In the current study, we show that a variety of non-native AHLs and related derivatives capable of inhibiting LuxR-type receptors in P. aeruginosa display significantly higher potency in a P. aeruginosa Δ(mexAB-oprM) mutant, suggesting that MexAB-OprM also recognizes these compounds as substrates. We also demonstrate that the potency of 5,6-dimethyl-2-aminobenzimidazole, recently shown to be a QS and biofilm inhibitor in P. aeruginosa, is not affected by the presence/absence of the MexAB-OprM pump. These results have implications for the use of non-native AHLs and related derivatives as QS modulators in P. aeruginosa and other bacteria, and provide a potential design strategy for the development of new QS modulators that are resistant to active efflux."
} | 396 |
34577990 | PMC8468103 | pmc | 4,501 | {
"abstract": "Renewable polymers with self-healing ability, excellent elongation, hydrophobicity, and selective oil absorption attributes are of interest for an extensive range of applications, such as e-skin, soft robots, wearable devices, and cleaning up oil spills. Herein, two fully renewable eco-friendly polyamide (PA)-based self-healing elastomers (namely, PA36,IA, and PA36,36) were prepared by a facile and green one-pot melt polycondensation of itaconic acid (IA), Pripol TM 1009, and Priamine TM 1075 monomers. The molecular structures of these PAs were analyzed by FITR, 1 H NMR, and 13 C NMR. The distinct structure of these PAs shows superior strain values (above 2300%) and high ambient temperature autonomous self-healing ability. Interestingly, the synthesized renewable PA36,36 showed zero water absorption values and hydrophobic properties with a contact angle of θ = 91° compared to the synthesized PA36,IA and other previously reported PAs. These excellent attributes are due to the low concentration of amide groups, the highly entangled main chains, the intermolecular diffusion, the manifold dangling chains, and the numerous reversible physical bonds within the renewable PAs. Furthermore, the hydrophobic properties may aid in the selective oil absorption of the PA36,36-based foam, for which PA36,36 foam is produced by the green supercritical carbon dioxide (scCO 2 ) batch foaming process. The PA36,36 foam with a microporous cellular structure showed better absorption capacity and high stability in repeated use. Due to these advantages, these bio-based PAs have potential for the production of eco-friendly self-healing materials, superabsorbent foams, and other polymeric materials.",
"conclusion": "4. Conclusions In summary, highly stretchable, autonomous self-healing, hydrophobic, and wholly renewable PA-based thermoplastic elastomers were successfully prepared by one-pot green melt polycondensation under catalyst and solvent-free conditions. The M w values of the elastomers were above 30,000 g mol −1 , and 1 H NMR, 13 C NMR, and FTIR verified their molecular structures. PA36,IA was found to have T m of 134.6 °C, whereas PA36,36 showed an amorphous nature, and both elastomers were found to have T d5% above 420 °C. The elastomers showed low T g values of 24 °C for PA36,IA and 11.7 °C for PA36,36, which aids the autonomous self-healing capability at ambient temperatures (healing efficiency of ~97% in 48 h). The stress–strain (S-S) test revealed that the elastomers had excellent strain values (PA36,IA: 2330 ± 112% and PA36,36: 2698 ± 108%). These elastomers consisted of an abundance of van der Waals forces from long entangled alkylene chains, manifold dangling chains, and intermolecular diffusion, which led to better self-healing behavior. Interestingly, the PA36,36 elastomer is hydrophobic, as confirmed by its water contact angle (θ = 91°) and the water absorption study. This is a unique attribute compared to traditional water-absorbing PAs. The high M w and special hydrophobic attributes of the PA36,36 elastomer permit the fabrication of PA36,36 foam. SEM images of the PA36,36 foam indicate a microporous cellular structure, which assists in better absorption capacity and greater stability in the repeated use for selective oil absorption. This novel generation of fully renewable elastomers can be rated as a class of renewable eco-friendly self-healing material, and shows significant potential to replace fossil-based alternative products.",
"introduction": "1. Introduction Recently, active interest in bio-based polymeric materials has increased, making the innovation of bio-based plastics one of the most fascinating fields in materials science [ 1 ]. Bio-based polymers are prepared from monomers derived from biomass feedstock [ 2 , 3 ]. Compared to fossil fuel-based commodity polymers, the use of biomass polymers results in the reduction in carbon dioxide (CO 2 ) emissions into the atmosphere and oil consumption [ 4 ]. Therefore, the growth in these kinds of sustainable polymers not only alleviates the over-reliance on petroleum feedstock, but also offers new products with a notably improved ecological footprint and performance compared to their fossil-based counterparts [ 5 ]. Service life and implementation are two vital factors affecting the practical applications of bio-based polymers. Currently, the self-healing effect is considered one of the most effective methods to extend the life of bio-based polymers. In recent decades, remarkable advances have been made in the development of self-healing bio-based polymer elastomers [ 6 , 7 ]. Elastomers are one of the most studied self-healing polymer materials due to their versatile applications, such as electronic devices, coatings, sealings, clothing, medical devices, tires, and daily commodities [ 8 , 9 , 10 ]. The name “elastomer” is derived from rubber, which is typically defined as a highly stretchable material that can retract forcibly and quickly to maintain its original dimensions while releasing considerable force. Elastomers are viscoelastic polymers (i.e., characterized by both elasticity and viscosity), having weak intermolecular forces, commonly high failure strain, and low Young’s modulus compared to other materials. Rubber is an elastic material derived from natural and petroleum gas (synthetic) or acquired from the exudations of some tropical plants (natural rubbers). Elastomers and rubbers are commonly used terms to refer to any materials with rubber-like attributes [ 11 ]. A milestone study in self-healable elastomer systems was documented in 2008 [ 12 ]. Subsequently, healable elastomer materials have attracted considerable attention because they can fulfil the needs associated with sustainable growth [ 13 ]. The excellent characteristics of self-healing elastomer materials create prospects for their use in high-end applications, such as wearable devices, electronic skin/smart flexible electronics, and soft robots [ 14 , 15 , 16 , 17 , 18 ]. Specifically, the properties of autonomous self-healing elastomers are preferable because they may self-heal without any exterior intervention after experiencing damage. Self-healing capability elastomers are activated by physical bonds such as van der Waals forces [ 19 ], ionic interaction [ 20 ], hydrogen bonds [ 21 ], host–guest interaction [ 22 ], metal–ligand interaction [ 23 ], the combination of various bonds [ 24 ], covalent bonds such as the disulfide bond [ 25 ], the Diels Alder reaction [ 26 ], and boronic oxide/ester [ 27 ]. However, the integration of most autonomous self-healing elastomeric materials includes multiple steps using noxious compounds, which are undesirable for green polymer production, create challenges for scaling up production, and are not cost effective [ 28 ]. Furthermore, few fully bio-based autonomous self-healing elastomeric materials have been reported to date; published studies involve bio-based instances that are not 100% bio-based or are partially bio-based [ 29 , 30 , 31 ]. The most general method employed in the polymer processing industry is the melting process, which is eco-friendly. Solution polymerization based on acid chlorides/dimethyl esters results in greater toxic condensates and the raw materials utilized in this kind of process require additional synthetic steps and work-up. Additionally, toxic solvents are required to remove the homogeneous catalyst after manufacturing [ 32 ]. Therefore, solvent-free melting polycondensation is one of the most extensively utilized synthetic methods for the production of commercial engineering plastics such as polyamides (PAs), polycarbonates (PCs), and polyesters (PEs) [ 33 , 34 ]. In this melting polymerization, monomers are melted and subjected to a polycondensation reaction to yield a highly viscous resin, which can be directly processed into the desired materials. In 1946, Whinfield first developed a melting polymerization path for polyester production by the transesterification polycondensation of diols with diester monomers [ 35 ]. This method is still used for the production of millions of tons of polyester each year. Moreover, the preparation of healable elastomers via solvent-free melting polymerization remains a vitally important process. Therefore, interest in self-healing elastomer research aimed at green, eco-friendly, and energy-efficient manufacturing has increased due to the scarcity of petroleum-based resources and significant concerns related to the effect of greenhouse emissions [ 33 ]. In addition, significant attention has also focused on using hydrophobic self-healable polymer foams to produce oil absorbents due to their facile, rapid, and effectual absorption and oil/water separation capability [ 35 , 36 ]. This is motivated by the large-scale oil spills that have occurred in line with the development of modern industry, which cause acute harm to local ecosystems. For instance, more than 600 crude oil barrels spilled into a river in Santander, Colombia, in March 2018, leading to a catastrophe for the local ecological system [ 37 ]. To assist with this issue, many innovative polymeric porous materials with superhydrophobic properties have been developed, including microspheres [ 38 ], aerogels [ 39 ], membranes [ 40 ], and monoliths [ 41 ], which can separate oil from water. Nonetheless, most research studies have the following issues. First, the manufacturing process of these products is highly complex and costly, which prevents their large-scale use. Second, most of the previously mentioned products are composites, thus creating issues in the recycling of the products after their use. Ultimately, modern industrial manufacture faces the issue of scarcity of fossil-based non-renewable resources [ 37 ]. Hence, it is considered to be highly attractive to develop a pristine fully bio-based polymeric elastomer with multifunctional characteristics, such as autonomous self-healing abilities and oil/water separation properties, which can be easily manufactured and are derived from green resources. In this study, we report on the development of two classes of fully biomass-derived polyamide (PA)-based self-healing thermoplastic elastomers via a green melt-polycondensation approach without the use of solvents and additives, using bio-based monomers (itaconic acid (IA) derived from the fermentation of glucose, and Pripol TM 1009 and Priamine TM 1075 derived from natural fatty acids). Both Pripol TM 1009 and Priamine TM 1075 contain four methylene chains with a total of 36 carbon atoms. Due to the numerous methylene dangling chains and long methylene main chains present in Pripol TM 1009 and Priamine TM 1075 monomers, it is thought that robust physical and reversible cross-links can be formed between the methylene chains via van der Waals forces. This produces highly stretchable and autonomous self-healing polyamide elastomers. Interestingly, PA36,36 exhibits zero water uptake properties due to the significant content of non-polar parts. Therefore, the foams were prepared via a green supercritical carbon dioxide (scCO 2 ) process using this hydrophobic PA36,36 elastomer. As a proof of concept, the use of hydrophobic PA36,36 foams to effectively eliminate oil from water was proven.",
"discussion": "3. Results and Discussion 3.1. Design, Synthesis, and Structure Characterization of the Bio-Based Thermoplastic PA Elastomers The prime aim of the present work was to synthesize and characterize two types of fully bio-based PA elastomer derived from renewable monomers, and to apply them to self-healing and oil absorption applications. The bio-based PAs, namely PA36,IA and PA36,36, were synthesized by the polycondensation of Priamine TM 1075 with IA or Pripol TM 1009 at the temperature of 220 °C ( Scheme 1 and Scheme 2 ), respectively. Melt-polycondensation was used to diminish the ecological effect of the synthesis of the preferred PAs. The bio-based PAs showed a slightly yellow transparent appearance ( Figure 1 a,b) and high elongation. The transformation of monomers into bio-based PAs was confirmed by FTIR, 1 H NMR, and 13 C NMR spectroscopy. Figure 1 c–e shows the FTIR spectra of bio-based PA36,IA, and its monomers. The IR band in the regions of 1437, 1622, and 1695 cm −1 are associated with the =CH 2 scissoring bending vibration, and C=O and C=C stretching vibrations, respectively, and they are allocated to the chemical structure of the IA monomer ( Figure 1 c). In the FTIR spectrum of Priamine TM 1075 ( Figure 1 d), the bands at 3377 and 3300 cm −1 related to the stretching vibration of the primary amino group. The band in the region at 3300 cm −1 is related to the secondary amino group of Priamine TM 1075 [ 43 ]. However, in the FTIR spectrum of PA36,IA ( Figure 1 e), primary and secondary amino group peaks (3377 and 3300 cm −1 ) were not observed and a new broad single absorption peak appeared in the region at 3293 cm −1 . This peak is assigned to the amide group (-NH-) in PA36,IA, which indicates that a successful polycondensation reaction took place between IA and Priamine TM 1075. The single absorption band at 3293 cm −1 also represents the strongest H-bond between the PA amide groups. The apparent difference between the bands of the carbonyl groups in IA ( Figure 1 c) and PA36,IA ( Figure 1 e) can be witnessed in the FTIR spectra. The characteristic carbonyl band of IA at 1695 cm −1 has almost completely disappeared, and the carbonyl bands of amide-I (1645 cm −1 ) and amide-II (1547 cm −1 ) were raised in the PA36,IA FTIR spectrum ( Figure 1 e), respectively [ 42 ]. These outcomes indicate the conversion of the monomers into bio-based PA36,IA. For PA36,36 ( Figure 1 g), the existence of the amide bond was detected in several wavenumber regions: at 3289 cm −1 (N–H stretching in amide-I); 1641 cm −1 (C=O stretching in amide-II); 1548 cm −1 (in-plane N–H bending coupled with C–O, and C–N stretching in amide-III); and 1261 cm −1 (N–H bending in amide-IV), indicated the reaction between Priamine TM 1075 and Pripol TM 1009 monomers. The disappearance of the band at 1710 cm −1 (the C=O stretching of carboxylic acid) ( Figure 1 f) confirmed that the Pripol TM 1009 had been completely amidized [ 44 ]. The transparency of the PAs was supported by the outcomes from the UV-Vis spectrophotometer ( Figure S2, Supplementary Material ), which evidence the high optical transparency. The 1 H NMR spectra of IA, Priamine TM 1075, and PA36,IA are presented in Figure 2 . The chemical shifts at 3.32, 5.77, and 6.26 ppm are allocated to the hydrogen atoms of the IA ( Figure 2 a). The chemical shift of proton in the CH 2 adjacent to NH 2 was witnessed at 2.67 ppm for the Priamine TM 1075. The peaks in the region at 1.00–1.52 ppm are assigned to the –CH 2 protons in the middle of the Priamine TM 1075 chains. The peak at 0.89 ppm is related to the –CH 3 protons at the end of the Priamine TM 1075 dangling chains ( Figure 2 b). In the 1 H NMR spectra of PA36,IA ( Figure 2 c), the chemical shift at 0.89 ppm, and at 1.00–1.52 ppm, are attributed to the aliphatic –CH 2 protons of the PA36,IA segments, whereas the chemical shifts at 2.21 ppm represent the –CH 2 proton adjacent to the pyrrolidone ring. The chemical shifts in the region at 2.91, 3.29, and 3.77 ppm refer to the protons of the pyrrolidone ring, i.e., the creation of pyrrolidone ring [ 42 ]. These results further support the successful synthesis of the PA36,IA elastomer. In the 1 H NMR spectrum of Priamine TM 1075 and Pripol TM 1009 ( Figure 3 a,b), the chemical shift of Pripol TM 1009 is similar to that of Priamine TM 1075, with the exception that the chemical shift of proton in the –CH 2 adjacent to –NH 2 was witnessed at 2.67 ppm for Priamine TM 1075, and the proton of the CH 2 adjacent to –COOH was witnessed at 2.36 ppm for Pripol TM 1009. For PA36,36 ( Figure 3 c), the –CH 2 protons adjacent to the –CO and –NH, which signify the amide linkage, were detected at 2.17 and 3.26 ppm, respectively [ 45 ]. The chemical structures of PA36,IA, and PA36,36 were further supported by the results of 13 C NMR ( Figure S3, Supplementary Material ), which show the chemical shift for the PAs, providing evidence of the –CO groups and the alkyl dangling chains. The resulting PAs were then analyzed by GPC ( Figure 4 a) and the relevant information is presented in Table 1 . The outcomes display that the synthesis led to PA36,IA with M n , M w , and Ð values of 19,262 g mol −1 , 30,889 g mol −1 , and 1.60, respectively. For PA36,36, the M n , M w , and Ð values were 21,045 g mol −1 , 32,938 g mol −1 , and 1.57, respectively. The greater reactivity of Pripol TM 1009 may be a factor in the M w gain. 3.2. Thermal Properties TGA analysis was performed to study the thermal decomposition behavior of bio-based PA elastomers. Figure 4 b shows the TGA traces of Priamine TM 1075, Pripol TM 1009, IA, and the bio-based PAs (PA36,IA, and PA36,36) derived using solvent and catalyst-free melt-polycondensation. Greater thermal stability of the bio-based PAs in comparison with starting monomers indicates the formation of PA elastomers ( Table 2 ). Both PAs show traces of similar degradation with a single-stage decay profile, denoting similar thermal decomposition behavior. The initial degradation temperature (T d5% ) of the bio-based PAs begins at 429.3 °C for PA36,IA, and 427 °C for PA36,36, after which the decomposition accelerates until there is no remarkable residue. The DTG traces of PA36,IA, and PA36,36 samples ( Figure 4 c) show that the maximal degradation temperature (T dmax ) occurs at 471 and 470.7 °C. In addition, T dmax is comparable to those of traditional polyamide 11 (PA11), for which T dmax is 470 °C [ 46 ]. Thermal decay of linear bio-based PAs starts either in the –CH 2 group adjacent to –NH or the carbonyl methylene group [ 47 ]. Inorganic gases such as water vapor (H 2 O), carbon dioxide (CO 2 ), hydrogen cyanide (HCN), and ammonia (NH 3 ) are the major products of thermal decay in PAs [ 48 ]. The high T d5% and T dmax values of bio-based PA36,IA, and PA36,36 prove their superior resistance to thermal decay. DSC analysis was employed to study the crystallization and thermal transition behaviors of bio-based PAs. The DSC thermograms of the bio-based PAs are shown in Figure 4 d. Due to the existence of multiple dangling chains, the PA36,36 is amorphous in nature, with no T m and T c peaks noticed upon heating and cooling by DSC. For the PA36,IA polymer, the T m peak at 134.6 °C (∆H = 4.3 J g −1 ) was witnessed and the T c peak was not observed because the T c of PA36,IA is mainly affected by the generation of the pyrrolidone ring and the concentration of –NH groups, and also affected by the existence of dangling side chains. First, we observed that it is difficult to determine the T g of all PA elastomers from DSC thermograms. This may be because the sensitivity of the DSC thermogram device is not adequate to detect the T g characteristics of the amorphous domains. Thus, DMA analysis was used instead to evaluate T g . Figure 5 shows the DMA traces of the PAs as functions of temperature in the range from −75 to 80 °C. There was a distinct variation in the storage modulus (E’) and loss factor (tan δ) of the PA samples ( Table 2 ). DMA analysis shows a higher storage modulus value of PA36,IA, and a slightly lower value for PA36,36 ( Figure 5 a,b). Furthermore, the small loss factor (dumbing value) of PA36,IA compared to PA36,36 signifies that PA36,IA is a stiffest material and PA36,36 is a more flexible material. The PA elastomers show T g , determined as the maximal peak value of tan δ, of 24 °C for PA36,IA and 11.7 °C for PA36,36. The high stiffness of PA36,IA prepared here may be ascribed not only to robust interchain interaction through -NH bonds but also to the PA36,IA network comprising a rigid pyrrolidone ring; however, steric hindrance against interchain H-bonding may be considered by the nonplanar pyrrolidone ring [ 49 ]. The T g of PA is associated with the monomers’ chain length and the amide group concentration in the polymer chains. Therefore, longer monomer chains result in greater mobility of PA36,36, meaning it is much softer [ 50 ], so the synthesized bio-based PAs possess much lower T g values than the traditional rigid PAs, i.e., PA11 (T g = 68 °C), PA6,6 (T g = 80 °C), and PA6 (T g = 78 °C) [ 51 ]. Due to the existence of a rigid pyrrolidone ring in PA36,IA, it exhibits a higher storage modulus value than PA36,36. This may be due to the presence of a hard pyrrolidone ring on the PA36,IA backbone. The low T g value ensures the mobility of the PA elastomers at room temperature, aiding autonomous self-healing [ 52 , 53 ]. 3.3. Mechanical Properties The descriptive stress–strain (S-S) graphs of PA36,IA and PA36,36 are presented in Figure 6 , and the data are summarized in Table 3 . The results show an S-S curve shape that is compatible with the thermoplastic elastomers (TPEs), comprising PA-based TPEs [ 54 ]. At plastic deformation, when the stress reaches above a certain value, the value of stress linearly increases with the strain, causing plastic deformation. Figure 6 a also shows that the stress of each PA continually increases with the strain (strain hardening). The presence of long alkyl dangling chains and a pyrrolidone ring in the PA backbone enhanced the physical cross-linking points, van der Waals forces, and hydrogen bonds, thus effectively improving the strain hardening and leading to the non-linear S-S graphs of PA elastomers. This was reflected in the S-S graphs by the appearance of the evident plastic deformation and yield point. Both the bio-based PAs show notable strain values ( Figure 6 a,b). However, there was a significant difference in the stress and strain of the PA36,IA and PA36,36 samples. The PA36,IA sample showed significantly lower strain (2330 ± 112%) and higher strength (1.29 ± 0.04 MPa) values in comparison with PA36,36 (strain: 2698 ± 108%, strength: 1.22 ± 0.02 MPa). The rigidity of the pyrrolidone ring in the backbone of PA36,IA may contribute to the higher stress and lower strain values compared to PA36,36. Nonetheless, PA36,IA also shows outstanding stretchability. The entangled and long methylene chains with “hidden lengths” present inside the dynamic physical cross-links of PA36,36 confirm the remarkably higher strain values than those of PA36,IA [ 55 ]; when stretched, the long methylene chains detach, and the physical bonds then break, releasing to further extend the hidden chains. By comparison, PA36,IA also exhibits outstanding elongation, which may be due to the existence of long alkyl entangled chains and the formation of the pyrrolidone ring in the polymer backbone structure, which leads to decreased amide group concentration and greater elongation. Additionally, the low M w of PA36,IA is another reason for the high elongation, which facilitates molecular motion and flexibility. The bio-based PA36,IA and PA36,36 possess the greatest strain values among aliphatic PAs and stretchability that is comparable with that of existing PA-based TPEs. The strain value was ~600% for fatty dimer acid bio-PAs [ 44 ], 245 ± 14.4% for PA10,12, 186 ± 12.3% for PA6,10, 197 ± 11.7% for PA10,10 [ 56 ], and 350–2200% for fossil-based thermoplastic poly(ether- b -amide) elastomers with diverse PA ratios. The strain was also greater than that of other dimer acid-based self-healing rubber-like elastomers (strain ~620%) [ 30 ]. 3.4. Self-Healing Properties of the Bio-Based PA Elastomers Another interesting property of PA36,IA and PA36,36 is their ability to self-heal. Traditional PA-based TPEs are not self-healing and require alteration to have such characteristics. Both PA36,IA and PA36,36 showed autonomic self-healing properties under ambient temperature, without any alterations or external stimuli. Samples of PA36,IA and PA36,36 were cut into two parts and then reconnected for 1 min before being left for different times at room temperature (~20 °C) ( Figure 7 a). The photographs in Figure 7 a displayed excellent autonomic self-healing capability at ambient temperature. Additionally, the self-healing behavior of bio-based PA elastomers was intuitively monitored by POM ( Figure 7 b–e and Figure S4a–d ) and SEM ( Figure 7 f–i and Figure S4e–h ). As shown in Figure 7 b–i, the PA36,IA was first cut into two parts with a 0.4 mm razor blade, after which the two parts were reattached without stress to record the self-healing phenomena using POM and SEM. Results in Figure 7 b–e and Figure S4a–d show that the fracture surfaces were sufficiently healed such that the gaps disappeared, thus providing robust proof of the effectiveness of the overall autonomic self-healing process. A similar autonomic self-healing trend was witnessed in PA36,36, for which the POM and SEM images are presented in Figure S4 (Supplementary Material) . Next, the self-healing attributes of the bio-based PA elastomers were further elaborated by S-S traces. As demonstrated in Figure 8 a,b, the mechanical attributes of the PA elastomers were recovered with increasing self-healing time. After 1 h, the PA36,IA and PA36,36 samples were stretched to 1029 ± 217% and 1196 ± 195%, respectively, whereas, after 48 h, they were stretched to 2265 ± 156% and 2638 ± 147%, which is ~97% of the average strain value of the original PA specimens, before being cut into two parts ( Table 3 ). The strain value of self-healed PA36,IA and PA36,36 samples showed a self-healing ability of 44% and 44%, respectively, after 1 h, and 97% and 98% after 48 h. This autonomic self-healing ability is remarkable for fully bio-based polymer materials. The self-healing attributes of PAs were compared. With an identical healing time (48 h), the healing ability of the two PAs was almost identical. Interestingly, PA36,IA, which contains a hard pyrrolidone fraction, also exhibited a healing ability of 97%. The probable cause is a decrease in M w of PA36,IA compared to PA36,36, which enhances the intermolecular segment migration. Compared with a number of previous studies on bio-based self-healing polymers [ 57 , 58 ], PA36,IA and PA36,36 displayed advantages in terms of more remarkable healing ability, shorter healing time, and no need for external stimuli. As illustrated in Figure 8 c, the fatty acid monomer of PA36,IA and PA36,36 consists of dangling side chains with several alkyl groups. After polycondensation, the key long methylene and dangling chains in the PA form numerous van der Waals bonds amid the methylene groups similar to a supramolecular network, which was found to be robust, reversible, and dynamic [ 59 ]. After physical damage to the PA specimens, the van der Waals bonds break, and the attached chains become free dangling chains leading to an entropic rise [ 60 ]. When the two broken PA elastomer specimen parts are brought back together, the main and dangling chains spread from one part to the other, and again form robust van der Waals forces/interactions between the elastomer chains; thus the PA elastomer self-heal until the entropy of the system returns to equilibrium [ 61 ]. 3.5. Viscoelastic Properties The viscoelastic features play a key role in the foaming process. Rheological analyses were carried out at 150 °C to evaluate the viscoelastic attributes of bio-based PAs. Both of the bio-based PAs are thermostable up to 420 °C, which means the PAs do not thermally decompose during viscoelastic experiments. Complex viscosity (η*), storage modulus (G’), loss modulus (G”), and loss factor (tan δ) of PA36,IA and PA36,36 were derived as a function of angular frequency (ω). Both the elastomers acted in a similar way and the η* values decreased, whereas the G” and G’ values increased with increasing ω. The rheological properties of molten polymers strongly depend on M w [ 62 ]. A shear-thinning behavior of bio-based PA elastomers is witnessed in Figure S5a (Supplementary Material) . The η* values of PA36,IA declined compared to those of PA36,36, resulting from the reduction in M w . The greater η* values can improve the melting strength and better inhibit the foam cell rapture throughout the cell growing stage during the scCO 2 foaming process. Figure S5b showed the associations of G’ and ω. It can be seen that the G’ curve of PA36,36 is higher than that of PA36,IA due to its higher M w . A similar trend was observed for G” values ( Figure S5c ). The definition of tan δ is the contribution ratio (G”/G’) at a specific value of ω. For the PA36,IA elastomer, the increasing trend in tan δ is observed in Figure S5d . It is shown that the melting elasticity deteriorates and the melting viscosity loss progressively improves. Polymers with higher melting strength are anticipated to improve the foamability. 3.6. Water Uptake Study Figure 9 a displays the percentage of water uptake of PA36,IA and PA36,36 after being submerged in water for 24 h compared with other PAs reported in previous literature [ 51 ]. PA36,IA displays a water absorption value of 0.2%, and PA36,36 displays water absorption of zero percent after 24 h of water submersion, representing hydrophobicity. However, compared to other traditional PAs, our integrated PA36,IA also exhibits a lower water absorption value, indicating that PA36,IA is significantly less sensitive to water. A lower ratio of –NH- linkages in the PA chains contributes to reduced water absorption. Based on the comparatively low –NH– ratio of PA36,36, this polymer has relatively zero water absorption properties and hydrophobicity. The origin of the monomer is another cause of the hydrophobicity of PA36,IA and PA36,36. The Priamine TM 1075 and Pripol TM 1009 monomers are derived from long-chain fatty acids, which are generally hydrophobic. Priamine TM 1075 and Pripol TM 1009 both comprise non-polar hydrocarbon chains and polar chain ends, which control water uptake. The water contact angle (θ) measurement was utilized to affirm the hydrophobicity of PAs ( Figure 9 b,c). Polymers with θ > 90° are defined as hydrophobic materials. The θ values of PA36,IA and PA36,36 were 84° and 91°. PA36,36 has a higher θ value compared to other traditional PAs such as PA6, PA66, and PA6,10 [ 63 ]. 3.7. Absorption Performance In this work, PA36,36 only exhibited higher M w and hydrophobic attributes compared to PA36,IA. Therefore, as a proof of concept, PA36,36 was selected as the substrate for the production of foam for the oil absorption study. For this test, PA36,36 foams were prepared by the scCO 2 foaming process. The prepared foam shows a microporous cellular structure ( Figure 9 d). To ensure the use of PA 36,36 foam in the selected oil absorption, the wettability test was first performed. After immersion in water, no water was seen to be absorbed by the PA36,36 foam ( Figure 10 a–c, Movie S1 in Supplementary Material ). Accordingly, the PA36,36 foam has excellent potential for oil absorption applications. To test the oil absorption behavior of PA36,36 foam, dyed oils (hexane and motor oils dyed using Oil Red O dye and Oil Blue A dye) were used. Figure 10 d-f shows a piece of PA36,36 foam entirely absorbed the dyed oils from the surface of the glass. When in contact with the dyed oils, the PA36,36 foams absorbed the oils quickly and entirely. Moreover, PA36,36 foam is expected to absorb oils easily from oil/water mixtures ( Figure 10 g–i). Hence, PA36,36 foam was also used to selectively absorb dyed hexane on top of the water. This absorption process is displayed in Movie S2 (Supplementary Material) , in which PA36,36 foams can be seen to absorb the dyed hexane rapidly from the water. In addition, no water was discovered in the PA36,36 foam, and no dripping of absorbed oils was found during the absorption process, indicating the high selectivity of the oils. The absorption characteristics of PA36,36 foam were also estimated by measuring the weight of oil, i.e., the ratio of the mass of absorbed oil to the PA36,36 foam. The absorption capacity of hexane and motor oils of PA36,36 foam was found to be 2.1 and 1.8 g/g, respectively, depending on the viscosity and density of the oils ( Figure 10 j). Moreover, it is vital to examine the consistency of the absorption capacity in repeated usage. The PA36,36 foam was deployed to absorb hexane and motor oil for five cycles, and the results are presented in Figure 10 k. It is shown that the absorption capacity of PA36,36 foam was the same when the motor oil and hexane were absorbed. Due to its eco-friendliness, the bio-based PA36,36 foam holds significant potential for utilization as an oil absorber in practical use."
} | 8,184 |
39836381 | PMC11833322 | pmc | 4,502 | {
"abstract": "Abstract Land-use changes threaten ecosystems and are a major driver of species loss. Plants may adapt or migrate to resist global change, but this can lag behind rapid anthropogenic changes to the environment. Our data show that natural modulations of the microbiome of grassland plants in response to experimental land-use change in a common garden directly affect plant phenotype and performance, thus increasing plant tolerance. In contrast, direct effects of fertilizer application and mowing on plant phenotypes were less strong. Land-use intensity-specific microbiomes caused clearly distinguishable plant phenotypes also in a laboratory experiment using gnotobiotic strawberry plants in absence of environmental variation. Therefore, natural modulations of the plant microbiome may be key to species persistence and ecosystem stability. We argue that a prerequisite for this microbiome-mediated tolerance is the availability of diverse local sources of microorganisms facilitating rapid modulations in response to change. Thus, conservation efforts must protect microbial diversity, which can help mitigate the effects of global change and facilitate environmental and human health.",
"introduction": "Introduction Land-use changes such as mowing or fertilizer application in grasslands are a major driver of species loss and responsible for alterations in community structure, with ensuing consequences for ecosystem functioning and human well-being [ 1 , 2 ]. Plant species are increasingly challenged to cope with these environmental alterations. Migration to more suitable habitats or adaptations to novel conditions are frequently discussed mechanisms to avoid extinction [ 3 , 4 ], which may, however, lag behind rapid anthropogenic alterations of environmental conditions [ 5 ]. Natural modulations of plant-associated microbial communities in response to environmental stresses can represent an alternative and possibly faster way for plants to cope with alterations in land-use and other components of global change [ 6–9 ]. In fact, ecological and evolutionary microbiome responses can easily keep pace with environmental changes [ 10 ] and land-use history, intensity, and agricultural practices have been shown to affect the diversity and composition of bacterial and fungal communities in the soil or associated with plants [ 11–15 ]. These rapid responses of plant-associated microbiomes may then, in turn, have positive effects on plant phenotype and productivity, increasing plant species’ tolerance to changes [ 16 , 17 ]. As a consequence, plants may invariantly perform well despite being challenged by stressors [ 18–20 ]. Accordingly, inoculation experiments in the context of climate change showed that a preadapted microbiome can increase survival rates of seedlings and promote plant growth under environmental stress [ 9 , 20 , 21 ]. Positive microbial effects on plant species could even promote species richness and productivity in plant communities [ 12 , 14 , 22 ]. Therefore, spontaneous natural changes in plant-associated microbiomes may be equivalent to recent advances in microbiome engineering, which is used to promote plant fitness and support plant tolerance to rapid changes of the environment [ 23 ]. For grasslands in particular, land degradation and changes in land-use intensity such as mowing or fertilizer application were identified as the main causes of the decline in species diversity, the disruption of interactions between species, and the loss of ecological functions [ 2 , 24–26 ]. Different components of land-use (e.g. mowing or fertilizer application) have been shown to affect plant and microbial diversity and composition differently [ 13 , 27–29 ]. Although some plant species and other taxa benefit from an increase in land-use intensity, many plant species decline under intensive land-use and are eventually displaced from their original habitat [ 2 ]. Plants in natural and agricultural systems may benefit from microbes, which can help plants individuals to resist abiotic stresses or protect them against diseases and herbivores [ 20 , 21 , 30 , 31 ]. Both individual strains of microbes as well as diverse consortia can affect plant traits, plant interactions, and ecosystem processes and functions [ 22 , 32–34 ], but inoculated consortia usually outperform individual strains [ 35 ]. Based on the previous knowledge about the vulnerability of grassland species [ 27 , 36 ] and the responses of bacteria and fungi to land-use changes [ 13 , 37 ], we tested whether natural rapid modulations of the plant microbiome have the potential to mediate land-use effects on plant phenotype and performance, and thus might represent an alternative to the “adapt or migrate” strategies. In a common garden experiment, grass sods from different regions in Germany were subjected to different mowing and fertilizer regimes [ 29 ] to test the effects of land-use treatments on morphological and physiological features of plant communities. Additionally, the sods were complemented with Fragaria vesca plants as phytometers to also test effects of land-use treatments on F. vesca phenotype and performance, as well as the diversity and composition of the bacterial and fungal communities associated with F. vesca . Such experiments may reveal strong microbial contributions to land-use effects on plant phenotype and performance [ 9 , 21 , 22 ], but clear experimental evidence for direct effects of the plant microbiome modulated by global change on plant responses requires additional, more controlled laboratory experiments. Therefore, we inoculated F. vesca plants grown from surface sterilized seeds in germ-free containers with the whole microbiome extracted from field-grown phytometers, testing microbial effects on the plant phenotype under gnotobiotic conditions in the absence of environmental variation. With these common garden and laboratory experiments, we tested a set of hypotheses that specifically address the prerequisites underlying the rationale of microbiome-assisted plant tolerance to land-use changes: (i) microbial diversity and composition as well as single plant phenotype and plant community features rapidly respond to major components of land-use, mowing and fertilization; (ii) effects of the microbiome on plant phenotype and performance are as strong or even stronger than effects of mowing and fertilization; (iii) if rapid alterations in the plant’s microbiome increase the tolerance of plants to land-use changes, plant performance should be relatively stable across land-use treatments; (iv) microbial effects are also measurable in the absence of other biotic or abiotic influences (as tested in the lab experiments). By verifying these hypotheses, we show that rapid natural modulations of the plant microbiome may increase plant tolerance to land-use changes and therefore may also buffer grasslands against stressors of global change.",
"discussion": "Discussion Our results largely verified our initial hypotheses on the prerequisites of microbiome-assisted plant tolerance against land-use changes. The microbiome of the F. vesca phytometers rapidly altered composition and diversity in response to mowing and fertilization. These alterations in the plant-associated microbiome directly affected the phytometers’ phenotype – an effect that was stronger than direct land-use effects on morphological and physiological traits of F. vesca . Note that the direct effect of plant communities (which may be an indirect effect of land-use treatment) is stronger than the effect of microbial diversity. As the performance of the plants was solely related to the plant phenotype and invariant across land-use treatments, we concluded that the plant microbiome may mediate land-use effects on plant phenotype and performance and thus may strongly contribute to plants’ tolerance to environmental changes. This conclusion was clearly supported by our laboratory experiments that demonstrated the microbiome’s potential to alter plant phenotypes in the absence of other biotic or abiotic influences associated with land-use changes. While our experiments clearly demonstrate that land-use treatment-specific microbiomes affect the phenotype of F. vesca plants, future studies may benefit from tracking shifts in the microbiome and associated changes in phenotype in response to land-use changes – as opposed to inoculating sterilized plants. In contrast to previous studies that transplanted microbiomes from sites with different levels of environmental stress to increase plant tolerance [ 9 , 18–21 ], our data indicate the stress-mitigating potential of rapid modulations of the plant microbiome occurring on site. These natural modulations of the plant microbiome can be key in plant tolerance to global change components and thus a potential pathway in addition to “adapt or migrate” strategies [ 3 ] for species persistence and ecosystem stability. Land-use intensification poses a major threat to many ecosystems, leading to species loss and alterations in community structure, which can disrupt ecosystem functioning [ 1 , 2 ]. Relocation as a strategy to avoid environmental changes might often fail because of time constraints, continued habitat shrinking, or a lacking spatial connectivity within a species’ range of dispersal [ 73–76 ]; likewise, the process of evolutionary adaptation to rapid alterations also takes several generations and might be too slow in most cases [ 5 ]. Our data experimentally support recent findings that land-use not only affects plants and plant communities, but also the diversity and composition of plant-associated bacterial and fungal communities [ 11 , 13 , 15 ]. Our data additionally demonstrate the pace and small scale of such alterations: Within three months of exposure to different land-use treatments in the small area of the common garden, surrounded by the same environment, microbial communities clearly diverged to treatment-specific compositions. Thus, our data support the idea that alterations in the microbiome can easily keep pace with changes in environmental conditions [ 10 ]. These rapid alterations are the prerequisite for microbiome-assisted plant tolerance to changing environmental conditions, which has been proposed in transplantation studies [ 8 , 9 ]. Our lab experiment provides important evidence—lacking in many other studies—of the contribution of the microbiome to the plant phenotype, which translates, according to our data, to plant performance. Therefore, small-scale natural variation in the plant microbiome in response to environmental conditions may be an untapped resource for microbiome engineering [ 23 , 77 ] that aids ecosystem restoration [ 78 ]. Microorganisms play essential roles in ecosystem tolerance and resilience against global change components [ 79–81 ]. Our data indicate that rapid alterations of the plant microbiome may be key in plant species’ persistence despite changing environments and thus in ecosystem stability. However, a prerequisite for this microbiome-mediated strategy is locally available diverse sources of microorganisms that provide the source for rapid modulations in response to change. The need for a diverse selection of microbes is further supported by the finding that the strongest effects on plants are usually mediated by microbial consortia, not individual strains [ 35 ], and that microbial diversity can enhance plant species richness and productivity [ 14 , 22 ]. Therefore, we conclude that conservation efforts will benefit from considering and protecting microbial diversity in addition to efforts to maintain plant and animal communities. Given the multiple beneficial roles of microbes in natural and anthropogenically altered environments, such “microbiome stewardship” [ 82 ] can help mitigate the effects of global change and facilitate environmental and human health."
} | 2,975 |
34940148 | PMC8704280 | pmc | 4,503 | {
"abstract": "Simple Summary Comparisons of plant and insect pollinator networks along elevational gradients can help predict future impacts of changing climate on pollinator distribution on local scales. We compare the pollination network structure along the altitudinal gradient of the San Francisco Peaks in Arizona. We evaluate shifts in network connectance, nestedness, modularity, and overall generalization with increased elevation. We conclude that plant–pollinator networks become more nested and generalized with elevation and identify the insect pollinator species most critical for network stability at the higher elevation pollination community. The variation in plant–pollinator network structure at different elevation zones of the San Francisco Peaks helps unveil which local communities currently support the most stable systems in the face of climate change. Abstract The structural patterns comprising bimodal pollination networks can help characterize plant–pollinator systems and the interactions that influence species distribution and diversity over time and space. We compare network organization of three plant–pollinator communities along the altitudinal gradient of the San Francisco Peaks in northern Arizona. We found that pollination networks become more nested, as well as exhibit lower overall network specialization, with increasing elevation. Greater weight of generalist pollinators at higher elevations of the San Francisco Peaks may result in plant–pollinator communities less vulnerable to future species loss due to changing climate or shifts in species distribution. We uncover the critical, more generalized pollinator species likely responsible for higher nestedness and stability at the higher elevation environment. The generalist species most important for network stability may be of the greatest interest for conservation efforts; preservation of the most important links in plant–pollinator networks may help secure the more specialized pollinators and maintain species redundancy in the face of ecological change, such as changing climate.",
"introduction": "1. Introduction Insects, especially bees, are critical for pollination services worldwide [ 1 , 2 , 3 ]. Many fly species replace bees as the dominant pollinators at high elevation communities due to their ability to maintain functionality in cooler conditions [ 3 , 4 , 5 ]. Hoverflies (Diptera: Syrphidae) are especially important dipteran pollinators who feed almost exclusively on the nectar and pollen of many wild plants and important food crops [ 6 ]. However, increased global warming, climate variability, and land use change is leading to higher temperatures, habitat loss, and resource loss for insect pollinators, already affecting species distribution and diversity across time and space [ 7 , 8 , 9 , 10 , 11 ]. With temperatures predicted to warm ~3 °C over the next 80 years [ 12 ], there may be increased instances of phenological asynchrony, or a disruption in the overlap of pollinator foraging time and host plants’ flowering period [ 13 , 14 ]. Most plant species in temperate environments rely on air temperature as a trigger for flowering, and even a six-day premature blooming period has been shown to disrupt pollinator–plant associations and negatively impact the fitness of solitary bee species [ 15 ]. Additionally, a rise in ambient temperature may cause insect pollinators to shift upward in elevation or latitude to follow the climate with which they are adapted, potentially leaving pollinators with a new suite of host plants in their new range [ 13 , 14 , 16 , 17 , 18 , 19 ]. Insect pollinator species range from being dietary specialists to extreme generalists; obligate specialists depend on one or a few plant species while extreme generalists may use numerous floral resources available in their community [ 20 , 21 , 22 ]. Previous research has shown that pollinator specialization can vary along productivity gradients [ 22 , 23 ], and while some bee, fly, and butterfly species may be able to act as opportunistic generalists in the face of reduced host partner availability, others may be more sensitive to resource loss. The structural variability of plant–pollinator network interactions along environmental gradients can help predict how future shifts in temperature, extreme climatic events, or changes in species composition may impact pollinator community robustness. Elevation gradients offer unique opportunities to study pollination network properties across different habitats and temperatures within a small geographic area, and these gradients represent suitable proxies for climate change by essentially replacing time with space [ 3 , 4 , 16 , 24 , 25 ]. In this study, we compare the structure of pollination networks at three elevation zones of the San Francisco Peaks in northern Arizona to determine which plant–pollinator communities in this local montane environment are less vulnerable to changing climate [ 4 , 26 , 27 , 28 , 29 ]. Namely, we analyzed network connectance, nestedness, modularity, specialization, and pollinator robustness. Network connectance reveals the proportion of observed plant–pollinator interactions out of those that have the potential to occur [ 26 , 30 ]. Networks consisting of more generalized species with a wider diet breadth are typically more connected due to an increased number of observed linkages with plants [ 23 , 31 , 32 ]. Nestedness describes the degree to which specialist species interact with partners who also associate with the highly generalized species [ 13 , 16 , 26 , 31 , 33 ]. More nested networks can offer built in redundancy and buffer the plant–pollinator system against species loss [ 34 ]. Network modularity characterizes the extent to which clusters of interacting species form tightly linked compartments that share very few interactions with other distinct compartments [ 35 , 36 ], and network specialization describes the level of overall species selectiveness within the plant–pollinator system [ 23 , 37 ]. Finally, robustness describes a network’s susceptibility to community collapse based on species extinction. Specifically, we measure overall pollinator robustness to plant species loss [ 38 ]. Previous studies have shown increases in connectance, nestedness, and species generalization, as well as decreased modularity, with increasing altitude [ 4 , 23 , 31 ], often driven by reduced partner availability because of decreased species richness. Increased elevation in temperate zones is typically associated with colder temperatures, reduced land area, shorter growing seasons, or unpredictable reproductive success of flowering plants, which can impose harsher conditions for flora and fauna [ 16 , 36 , 39 ]. Conversely, warmer conditions at lower altitudes may allow for longer flowering times and potentially increased reproductive success, positively affecting mass flowering of plant species [ 16 ]. Since different pollinator taxa groups have varying responses to weather and climate, some high elevation insect communities may experience increased environmental filtering and/or decreased species richness [ 16 , 23 , 24 , 28 , 36 , 39 ]. Larger-bodied, cold-adapted pollinators such as bumble bees and non-syrphid flies may dominate higher elevations of temperate regions whereas smaller-bodied, solitary bees drop out of the species pool. [ 3 , 40 , 41 ]. We have three predictions: (1) Plant–pollinator networks will become more connected and nested with increased elevation zone on the San Francisco Peaks; with fewer available host plant species, insect pollinators may be required to act as opportunistic generalists and share more plant resources [ 28 ]. (2) Network specialization and modularity will decrease with elevation zone. Modularity tends to increase with increasing link specificity [ 23 ], and therefore, lower overall selectiveness of plant partners at higher elevations may also lead to networks with fewer distinct modules. (3) The more generalized pollinators at the high elevation network will be the most critical species for plant–pollinator community stability and robustness. Dietary generalist pollinators often drive the more nested pollination networks [ 42 ], and with a wider flexibility in interaction partners, they may offer greater species redundancy in the event of host plant loss [ 20 , 28 , 43 , 44 , 45 ]. Identifying which pollinator species and interactions on the San Francisco Peaks currently support the pollination communities most robust to species loss may reveal which pollinator taxa contribute most to facilitating network success and structure [ 46 ]. Regardless of current population status, any event that could lead to the loss of these critical generalist species may lead to a cascading decline in diversity and/or community collapse [ 13 , 26 , 43 , 47 ].",
"discussion": "4. Discussion Our results confirm variations in plant–pollinator network structure along the elevational gradient of the San Francisco Peaks. Total species richness of both plants and pollinators loosely follows the unimodal, hump-shaped pattern often observed along elevational gradients [ 39 , 71 ], with the highest species richness present at the mixed conifer life zone. This may be attributed to an overlap in species ranges at the mixed conifer life zone; for example, 74% of lower-elevation pollinator species and 66% of higher-elevation pollinator species are also present at this life zone ( Supplemental Table S2 ). Similarly, 65% of the spruce fir life zone host plants and 63% of the ponderosa pine life zone host plants are also host species at the mixed conifer life zone ( Supplemental Table S3 ). The subsequent decrease in species richness from the mixed conifer life zone to the higher zone spruce fir is consistent with patterns observed in other studies of plant and pollinator richness along elevational gradients [ 16 , 23 , 28 ]. A combination of fluctuating climatic factors at higher elevations that determine the productivity of a plant–pollinator community, including precipitation, temperature, and solar radiation, may drive the taxonomic groups and/or number of individuals able to exist at certain geographic zones [ 39 , 72 ]. One recent study [ 16 ] found significant decreases of plant and insect species at higher elevations of Jonaskop Mountain, South Africa, and similarly, research in the Andes of Mendoza, Argentina [ 31 ], showed that the number of pollinator species decreased by over 33% from a lower to higher elevation zone. Further, decreasing temperatures at the higher elevations of the Mexical scrublands supported declines in bee species richness, with mean annual temperature as the best predictor of declines [ 1 ]. Cooler temperatures can affect the physiological and foraging capabilities of insects, which can lead to overall environmental filtering on bee or other insect communities along altitudinal gradients [ 1 , 16 , 72 , 73 ]. Average pollinator species richness also followed a significant hump-shaped pattern, but average plant species richness was not significantly different between the mixed conifer and spruce fir life zones. While both average and total pollinator abundance followed a hump-shaped pattern, average and total flower abundance was equal across life zones. This suggests that (a) there is a greater proportion of flowers per species at higher elevations or (b) the extremely high number of one or two dominant plant species at the spruce fir life zone, including Oxytropis lambertii and Eremogone fendleri , may be inflating the number of flowers counted at the higher elevation life zone. When pollinator species communities vary locally across the range of plant species, such as along environmental gradients, there may be intraspecific variation within the plants that reflect adaptive shifts to different pollinator presence [ 74 , 75 ]. Previous research has shown that plant communities may vary along an elevational gradient; because of limiting climatic factors at higher altitudes, such as decreased temperature and decreasing atmospheric pressure [ 72 , 75 ], there may be intraspecific morphological changes within plant species, such as decreased overall size, longer growing periods, shorter maximum height of individuals, and larger seed mass [ 16 , 75 , 76 ]. Local selection of beneficial traits for plant individuals at different climatic zones may assist with survival in harsher habitats [ 75 ]. Additionally, some plants produce larger flowers at higher elevations, where larger-bodied insects, such as bumblebees and flies in the family Bombyliidae , often dominate high altitudes [ 16 ]; an increase in flower size may be more attractive to these specific pollinators and increase chance of pollination [ 74 ]. Besides morphological variation, plants may become more specialized at lower elevations, which could be occurring within our study area. For example, Oxytropis lambertii (purple locoweed) was highly abundant at all elevations on the San Francisco Peaks. At the high elevation spruce fir, O. lambertii interacted with 33% of pollinator species and 14% of individuals. While it had a similar importance at the mixed conifer life zone, O. lambertii only interacted with 2% of individuals and 5% of species at the ponderosa pine life zone. This suggests that this plant becomes more generalized with increased elevation. Some network features showed variation as predicted along the gradient. For example, there was a significant decrease in network specialization (H 2 ) with elevation. This supports overall lower plant selectiveness at the higher elevation life zone of the San Francisco Peaks. It is possible that the more generalized pollinator species at the spruce fir life zone have greater weight within the network and help to secure the poorly linked (more selective) species across the community [ 7 , 13 , 27 , 77 ]. Further, nestedness was significantly lower in the lowest life zone. These patterns may be driven by a decrease in total species richness of flowering plant partners at the higher elevation life zone, spruce fir; with fewer host-plant options, insect pollinators may choose to interact with the more stable plant species in a community. These findings are consistent with those of recent studies of plant–pollinator network structure along elevational gradients; for example, research in the Canary Islands [ 23 ] found a decrease in network specialization at higher elevations of the El Teide stratovolcano, likely explained by reduced partner availability leading to a wider niche breadth. Similarly, research in Germany [ 28 ] found that higher elevation plant–pollinator communities of the northern limestone Alps, especially when confronted with an experimental treatment of delayed snowmelt, showed greater generalization than lower elevations. Contrary to our expectations, however, connectance did not show significant differences with life zone. This could be a product of small sample sizes when running analyses at a site level. For example, at the spruce fir life zone, when data are kept separate across sites, some pollinator species appeared to be singletons who only interact with one plant, including Bombus huntii . However, when data are pooled, many of these pollinators are actually more generalized as they interact with multiple plant species across the life zone. Thus, the contribution of what appear to be, but are not actually, singleton interactions at individual spruce fir sites may be depressing connectance values of this higher elevation life zone. This underscores the importance of adequate sample sizes to accurately identify the plant–pollinator interactions comprising a pollination system. Additionally, although we expected modularity to be higher at the lower elevation life zone ponderosa pine due to greater specialization, there were no significant differences in modularity across life zone. It is possible that high modularity did not emerge in any of our pollination systems due to small network sizes when analyzed at a site level. It has been shown that modularity may be less pronounced or absent in smaller networks (<50 species), whereas it is common in larger networks (>150 species) [ 26 , 78 ]. At individual sites, even the most species rich network only had a total of 56 plant and pollinator species sampled throughout our study period, which may make it difficult for patterns of modularity to emerge in our small pollination networks. Other studies of network structure along elevational gradients have also identified patterns that differ from what we expected. Research in Chile [ 36 ] found that modularity was actually conserved along an elevational gradient in the Andes, and that nestedness decreased, even though there was a simultaneous increase in connectance. Similarly, while studies on Mt. Olympus, Greece [ 4 ] did show increased nestedness in bumblebee-plant networks at high elevation communities, there was no correlation between network specialization and elevation. This highlights how network indices are not always predictably correlated and that results may vary depending on the pollinator taxa groups included in the network analyses. Typically, networks with higher nestedness and generalization show greater robustness when it comes to ecological disturbances such as changing climate or decreasing available habitat [ 35 , 43 ]. Although the robustness coefficient (R) was not significantly different along the altitudinal gradient, of interest at the spruce fir life zone are the highly interactive generalist species identified, including Plebejus melissa and Oarisma garita , as they may facilitate the construction of nested subsets and higher overall generalization. It is important to acknowledge the relationship between species Gc values and abundance at all life zones; recent studies suggest that abundance may influence species persistence in a habitat and that this bias is often ignored when identifying generalist species [ 32 ]. Specifically, the more abundant species are likely to be sampled more frequently [ 32 ], potentially skewing the results of which pollinators have the widest diet breadth. At all three life zones in our study, species abundance showed significant positive correlations with calculated Gc values, such that the species considered “core generalists” are typically more abundant in the community. This may have played a role in the calculation of generalist pollinator species used in our analyses ( Supplemental Figure S2 ). The simulations of pollinator species extinction at the spruce fir life zone showed that a loss of the three most generalized pollinators resulted in lower network nestedness. Although these pollinators may currently be abundant and stable in these pollination communities, increasing temperatures and greater instances of extreme climatic events could put their population health at risk. For example, increased drought frequency and intensity is predicted across regions worldwide, which can severely lower the amount of available floral resources for even the more stable, generalized pollinators [ 79 ]. This has already occurred in places that affect the area represented in this study; in June 2021, a 6-day period of record-breaking heat in the southwestern United States gave Arizona some of its highest temperatures in the state’s history [ 80 ]. Further, pollinators in montane environments may be especially susceptible to changing climate; competition between pollinators may become more prevalent with increased elevation if the range expansion of lower elevation pollinators encroaches on the space typically reserved for the higher-elevation pollinator groups, such as bumblebees, non-syrphid flies, or cold-adapted butterflies [ 4 , 81 ]. These higher-elevation insect pollinators often have narrow niches restricted to upper altitudes with cooler temperatures and low seasonal temperature variation [ 81 , 82 , 83 ]. Unfortunately, areas supporting these conditions are expected to shrink disproportionately under future climate scenarios [ 84 ], potentially causing certain pollinator taxa to reduce their ranges to remain within optimal habitat [ 81 , 85 ]. This may impact the stability of important generalist pollinators at higher elevation environments, ultimately leading to an increased risk of community collapse. Conservation management of pollinator species inhabiting the local mountain gradient of the San Francisco Peaks should focus on preserving and understanding the limitations of the most strongly interactive species identified in this study [ 23 , 46 ]. To accurately pinpoint factors that may limit these species with future climate change, their host-plant relationships and overall distribution should be studied on grander spatial scales. The response of various pollinator groups to changing climate will depend on life history traits such as sociality, body size, or nesting requirements [ 16 ], but factors that affect these species may be difficult to identify if focusing only on local studies [ 46 ]. The future development of more intensive plant–pollinator monitoring programs and pollinator inventory studies along the San Francisco Peaks will help create a more robust data set of plant–pollinator interactions in this area. Identifying the most important pollinator taxa and plant–pollinator associations could help outline steps for protecting and ensuring population health of the pollinators most responsible for network stability in this unique montane environment."
} | 5,392 |
30016366 | PMC6049921 | pmc | 4,505 | {
"abstract": "Here we describe development of a microfluidic viscometer based on arrays of magnetically actuated micro-posts. Quantitative viscosities over a range of three orders of magnitude were determined for samples of less than 20 μL . This represents the first demonstration of quantitative viscometry using driven flexible micropost arrays. Critical to the success of our system is a comprehensive analytical model that includes the mechanical and magnetic properties of the actuating posts, the optical readout, and fluid-structure interactions. We found that alterations of the actuator beat shape as parameterized by the dimensionless “sperm number” must be taken into account to determine the fluid properties from the measured actuator dynamics. Beyond our particular system, the model described here can provide dynamics predictions for a broad class of flexible microactuator designs. We also show how the model can guide the design of new arrays that expand the accessible range of measurements.",
"conclusion": "Conclusion We have demonstrated the first microfluidic viscometer using driven flexible micro-actuators. Our device does not require externally driven flow and is capable of successful measurements of viscosity on sample volumes of 20 μL for fluids in the range of 0.005 to 5 Pa*s. We found that alterations of the actuator beat shape as parameterized by the dimensionless “sperm number” must be taken into account to back out fluid properties from the measured actuator dynamics. Critical to the success of our system is a comprehensive analytical model that includes post viscoelasticity, magnetics of the post, optical readout, and fluid structure interactions. The model is applicable beyond our specific system, and can provide dynamics predictions of any flexible microactuator system.",
"introduction": "Introduction Rheometry, the measurement of the mechanical properties of fluids, is critically important in areas ranging from basic biology and medicine [ 1 , 2 ] to industrial contexts such as polymers [ 3 ], oil production [ 4 ], and food processing [ 5 ]. Conventional rheometers use relatively large volumes of material (hundreds of microliters), making studies of precious and novel materials difficult or impractical [ 6 ]. In the past decade there has been increasing interest in microfluidics-based rheometers to enable medical diagnostics and basic biological studies on small volume or precious samples [ 6 ]. The current push towards lab-on-a-chip technologies that can perform a battery of chemical and rheological tests on a single chip has increased interest in integrating micro-fluidic rheometers into these systems [ 7 – 10 ]. Here we describe a rheometer based on magnetically actuated surface attached post (ASAP) arrays ( Fig 1 ). The post structure consists of a PDMS core that provides mechanical flexibility with a nickel shell on the apical half of the post that drives post motion when exposed to an external magnetic field. We show that quantitative viscosities over a range of three orders of magnitude are determined for samples of less than 20 uL without the need for fluid flow during the measurement. In addition to measuring viscosity, we have previously used these arrays to measure the elastic properties of blood clots [ 11 ], and to pump and mix fluids [ 12 , 13 ]. 10.1371/journal.pone.0200345.g001 Fig 1 ASAP based Rheometer. (A) Computer generated 3D rendering of the experimental setup. The micro-fluidic channel is placed in between the C-shaped electromagnet core, and is imaged with the microscope objective. The intensity of the transmitted light from the light source is monitored using either a camera or photo diode (not shown). (B) Front view of the microfluidic chamber in between the magnetic poles. The face of the electromagnet is 10 mm X 10 mm and the gap is 16 mm. The magnet is tilted at a 10 o angle from the horizontal. (C) A diagram showing the microfluidic chamber. The chamber has a height, H , of 250 um. The PDMS base and posts have a total height of 125 um so that the gap height, G , between the top of the posts and the ceiling of the channel is 125 um. The channel is 1 mm in width and 16 mm long. (D) SEM of an array of micro-posts. The posts are 2 um in diameter and 23 um tall. Nearest neighbor distance between posts varies between a few um and 15 um. The success of our ASAP viscometer depends critically on the aid of a comprehensive analytical model we’ve developed that includes the device post material viscoelastic and magnetic properties, the system’s optical readout, and the fluid-structure interactions between the device posts and a viscous analyte. Through use of this model, we find that the beat shape of our actuating posts alters with the relative dominance of viscous drag forces over elastic restoring forces that occurs with an increase of the viscosity and/or beat frequency. For the purposes of this discussion, we use the terminology “beat shape” as distinct from beat amplitude; a change in beat shape denotes a qualitative difference in the beat profile. Taking into account the altered beat shape is essential for predicting actuator behavior across a useful range of viscosity and drive frequency. The model enables successful viscometry, but also provides guidance for future array design that will expand our device’s utility beyond current limitations imposed by our post aspect ratio and the bandwidth of our electromagnet system. The model is more generally applicable to any driven post system where fluid structure interactions affect beat shape, including actuated slender bodies in biological systems. In the last 10 years, there have been numerous papers on biomimetic cilia arrays and their potential applications for micro-fluidics applications. Like their biological counterparts, artificial cilia arrays have been demonstrated as effective mixers and pumps [ 12 , 14 – 16 ], and computational studies have predicted them to be capable of manipulating particle settling [ 17 , 18 ]. These artificial cilia are typically on the scale of a few microns in diameter and tens of microns tall, and therefore operate in the low Reynolds number regime for typical operational frequencies (< 100 Hz). Arrays have been fabricated using a variety of techniques including using self-assembled micro beads [ 19 ], magnetic polymer composites [ 12 ], electrostatic and magnetic flaps [ 20 , 21 ], and core-shell structures [ 22 ]. Several numerical and modeling studies have focused on the flow around post arrays [ 15 , 16 , 23 – 28 ]. Additionally, there is a considerable body of computational work that focuses on the dynamics of biological cilia arrays [ 29 – 31 ]. These computational studies typically either model the fluid dynamics using a prescribed motion of the posts, or are complex numerical simulations that cannot be used to enable experimental measurements of fluid viscosity. The dynamics of single filament fluid structure interactions have been studied analytically [ 32 – 35 ], and models built on these analytical treatments have been used to describe artificial swimmers and fluid shear sensors [ 36 – 38 ]. In general, the dynamics of these systems depend on a dimensionless parameter known as the “sperm number” [ 33 ]. The sperm number, or Sp , represents the relative influence of viscous drag forces between the surrounding fluid and the actuator with respect to the elastic restoring forces within the actuator, and determines the shape of driven post motions:\n S p = L ( 4 π ω η E I ) 1 4 (1) \nwhere L is the length of the filament, ω is the angular frequency of the motion, η is the viscosity of the fluid, E is the elastic modulus of the filament, and I is the second moment of inertia. For Sp << 1, the elastic forces dominate, and the deflection of the post is well described by quasi-static Euler beam mechanics. At large sperm numbers, Sp >> 1, the viscous forces dominate and the elastic waves formed in the filament become shorter than the filament length. The Sp can be thought of as the number of wavelengths that fit within the length of the elastic filament [ 33 ]. Changes in beat shape affect fluid flow around the actuator and consequently affect post-post interactions within actuator array settings. Computational studies of the fluid flow generated by actuating post arrays show that the nature of the flow depends strongly on Sp [ 25 ]. In other studies, simulations show that particle interactions with the post arrays also depend on Sp [ 17 , 39 , 40 ]. Multiple lab on a chip rheometer designs have been developed that operate on principles ranging from pressure drop across channels to the onset of turbulence [ 7 , 41 , 42 ]. Microposts employed as fluid motion sensors have been driven by steady and time varying fluid flows, and in high frequency applications near resonance driven by magnetic force at the pillar tip [ 37 , 43 , 44 ]. Artificial cilia arrays however, have not been reported as fluid rheometers. The work presented here represents the first demonstration of a micro-viscometer employing actuated flexible microposts. As previously noted, in prior work, we used our Ni composite arrays as a sensor for the relative elastic properties of human blood clots but did not measure the calibrated absolute viscous or elastic properties of the clot [ 11 ]. Here we expand on our analytical model of actuating surface attached posts (ASAP) arrays developed in our previous work [ 11 ], to include fluid-structure interactions ( Fig 2 ). From here on, we will refer to our previous analytical model as the “ASAP-1 model”, and the new analytical model including fluid-structure interactions described in this paper as the “ASAP-2 model”. In the experimental system, the bandwidth of the photodiode measurement of the transmitted light through the post array is 10 KHz. This optical measurement is synchronized with the magnetics (driven at 1–30 Hz), enabling precise determination of relative phase of the post response and magnetic driving force ( Fig 3 ). Using the phase data and the ASAP-2 model, we determine the viscosity of the fluid. We show that at low Sp ( Sp < 1 ), we are able to successfully measure viscosity for fluids ranging from 0.005 to 5 Pa*s. The fluid-structure interaction model that is included in ASAP-2 is necessary to achieve success near Sp ~1. For Sp > 1 , the model fails to predict post array behavior. This suggests a fundamental change in the fluid dynamics at Sp = 1 that may be due to post-post interactions or post-boundary interactions that are not accounted for in the analytical model. This is consistent with computational studies of post array driven fluid flow at Sp ~ 1 [ 25 ]. 10.1371/journal.pone.0200345.g002 Fig 2 Analytical and computational models. (A) Cartoon of post with analytical model parameters defined. The lower segment (blue) of the post is made of PDMS, and the top (gray) is the Ni shell. The black center line represents the deflection of the beam from the relaxed upright position (parameterized by w(s)). Note that since the Ni segment does not bend, w(s) for the Ni shell for all time points is entirely determined by the position and angle of the PDMS segment at its top end. The applied magnetic field applies a moment to the Ni segment (M mag ). (B) The analytical model focuses on the PDMS portion of the ASAP and incorporates the magnetic and drag forces experienced by the Ni segment as shear force and moment boundary conditions at the top end of the PDMS segment (S drag , M mag and M drag ). There is an applied moment from the magnetic field, and there are shear and moment drag terms generated by the motion of the Ni portion that are entirely determined by the time dependent position and angle of the top end of the PDMS portion. (C) Rendering from COMSOL showing the mesh and geometry of the finite element model. The model has no slip boundary conditions on the top and bottom, blue; while the left and right sides can have either no-slip or slip boundary conditions, red. The entire post, both the elastic lower portion and the stiffer upper shell, are explicitly modeled. The surfaces perpendicular to the plane of the post motion, orange, have zero pressure boundary condition. Box Dimension: 50 um x 50 um x 100 um. 10.1371/journal.pone.0200345.g003 Fig 3 ASAP deflection and phase. (A) Example images of the post array with the field off (top) and on (bottom). Images of this type are used to calibrate the post deflection. Scale bar = 50 um. (B) Raw ASAP results: the magnetic field and the transmitted intensity are measured simultaneously using a DAQ board. The intensity data comes either from integrated whole-frame intensity of video data, or from the photodiode signal (the latter case in this figure). See Materials and Methods . (C) The measured post angle and applied torque are calculated using the tilt and magnetic model explained in detail in [ 11 ]. From these results we get the post amplitude, Δθ , torque amplitude, Δ τ, and the phase, Δφ , between the drive signal and the post response.",
"discussion": "Discussion Viscometry analysis The success of the ASAP-2 model for Sp < 1 enables the use of ASAP as a viscometer. We note that this constraint ( Sp < 1) still provides experimental accessibility to very wide range of fluid viscosities. As Eq 1 indicates, drive frequency (or ASAP geometry) can be adjusted to accommodate for large viscosities to keep Sp below 1. Fig 5B depicts the useable range of our ASAP system for reliable viscosity measurement, in viscosity vs. frequency space. The two overlapping diagonal regions (dark grey-blue and light grey) represent Sp ranges for particular ASAP-geometries. The grey-blue region represents our current ASAP geometry, while the light-grey represents an alternative geometry that provides access to a complementary region of the phase space. In the right margin of the plot, the viscosity ranges of relevant biofluids are depicted and fall within the usable range of our current ASAP or that of the alternative design. In both cases, the lower boundary of the shaded regions corresponds to Sp = 0 . 3 and upper boundary corresponds to Sp = 1 . 0 . The lower boundary ( Sp = 0 . 3 ), is due to the confounding effect of the viscoelastic properties of the PDMS which dominate and swamp the signal in low viscosity/low frequency range. At high viscosity/high frequency we enter the large sperm number, Sp > 1 , regime where the ASAP-2 model begins to show its limitations. The red circles represent viscosity measurements of a nominal 2 Pa*s sample of Karo taken at different ASAP beating frequencies derived using the ASAP-2 model. Below Sp = 1, the measured values are accurate and consistent. For higher frequency measurements above Sp = 1, the measured viscosity deviates from the true value. This deviation at Sp = 1 is interesting because previous computational studies of driven post arrays indicate that Sp = 1 is a critical number. Computational studies have shown that for actuator arrays, the direction of directed flow is predicted to change at Sp ~ 1 [ 25 ]. It is intriguing that the deviation from experimental results occurs at this number and suggests that the deviation could be caused by a change in the flow dynamics in the chamber that invalidates the single post assumptions implicit in the ASAP-2 model. It is also possible that the PDMS calibration includes effects of post-post interactions. We calibrated for PDMS properties using the post response in water, which may include the influence of post-post interactions. These interactions may explain why we measured PDMS properties on the high end of the range that is found in the literature [ 51 , 52 ]. As the posts transition through Sp = 1 , this calibration may become invalid as the flow in the channel fundamentally changes. The limit of Sp = 1 for our results limits the experimental range of the system, as shown in Fig 5B . It is possible to design the post arrays in order to bring Sp into the valid range for any viscosity. The simplest and most direct way to adjust the post properties is by adjusting the aspect ratio of the PDMS section. Sp has two geometric parameters, L PDMS and the moment area of inertia, I (See Eq 1 ). For a cylinder I = π r 4 4 , so Sp scales linearly with the aspect ratio of the PDMS region, L P D M S r . Sp also depends on the frequency. This gives us a simple way to adjust the design of the ASAP post arrays to enable measurement of higher or lower viscosity samples while keeping Sp < 1 ( Fig 5B ). In the case of design of ASAP, we are limited by geometrical constraints set by ground and lateral collapse [ 53 , 54 ]. The low driving frequency limit is constrained by phase accuracy as it scales with the number of cycles measured. Our high frequency limit is determined by electromagnetic bandwidth of our magnetics system (200 Hz). In an array of posts, the fluid flow generated by neighboring posts and its impact on ASAP motion should be considered. Within this work, we show that it is not necessary to take this effect into account to for successful viscometry. However, the FEM model indicates that the posts generate significant fluid flow for their neighbors ( S4 Fig ). In posts that are separated by 30 μm , the fluid velocity drops by less than 20% between the posts. This suggests that post-post interactions in our dense arrays (average spacing 5–6 μm ) are extremely important. Future work will focus on understanding how these post-post interactions effect the analytical model. Other applications The post arrays have a number of unique advantages. They are easily integrable into micro-fluidic devices, and do not require an externally established flow, thereby reducing the amount of fluid necessary for experiments. Additionally the same ASAP can be used as a mixer or micro-fluidic pump while simultaneously measuring fluid viscosity. For mixtures that change viscosity in time upon mixing, the post arrays could be used as a both a mixer and a measure of the level of mixing. Alternatively, the ASAP elements could monitor the viscosity of the system while acting as a pump in a microfluidic system. We’ve now demonstrated the ASAP system as an effective elastometer [ 11 ] and a viscometer, and believe that future improvements to the model and system will allow us to measure the moduli of visco-elastic materials. ASAP arrays could have potential applications in shear thinning biofluids such as mucus. Given our experimental setup we can approximate the maximum shear that we can produce. If we model the tip of the post as a 2 um sphere we can use the standard equation for the maximum shear rate for a translating sphere [ 55 ],\n | γ ˙ | = 3 v s 2 r s (14) \nwhere r s is the radius of the sphere, or post in our case, and v s is the velocity. The max velocity at the tip of the post is:\n v t i p = θ max L p o s t ω (15) \nwhere ω is the drive frequency of the posts, θ max is the max post deflection, and L post is the total length of the post. Substituting Eq 15 into Eq 14 , we get the following equation for the tip shear.\n | γ ˙ | = 3 θ m a x L p o s t ω 2 r s (16) \nAssuming a reasonable post amplitude of 10° at 30 Hz, we obtain a shear rate on the order 10 3 s -1 , which is large enough to induce shear thinning behavior in biological samples such as mucus and blood, which both have significant shear thinning at shear rates above 10 1 s -1 [ 56 , 57 ]. For mucus, the ASAP arrays are particularly interesting as they are capable of measuring rheological properties in a way that directly mimics biological cilia in the airway. Rheological measurements of heterogeneous samples such as mucus can vary with the scale of the measurement and the actuation geometry. The current post arrays are not stiff enough to effectively measure the full range of viscosities found in mucus while remaining in the low sperm number regime. This could be adjusted by changing the aspect ratios of the posts. Posts with twice the diameter and 75% of the length of our current ASAP arrays could potentially measure mucus rheology properties while remaining under the Sp = 1 limit."
} | 5,055 |
28765600 | PMC5539299 | pmc | 4,506 | {
"abstract": "An industrial waste, 1,2,3-trichloropropane (TCP), is toxic and extremely recalcitrant to biodegradation. To date, no natural TCP degraders able to mineralize TCP aerobically have been isolated. In this work, we engineered a biosafety Pseudomonas putida strain KT2440 for aerobic mineralization of TCP by implantation of a synthetic biodegradation pathway into the chromosome and further improved TCP mineralization using combinatorial engineering strategies. Initially, a synthetic pathway composed of haloalkane dehalogenase, haloalcohol dehalogenase and epoxide hydrolase was functionally assembled for the conversion of TCP into glycerol in P. putida KT2440. Then, the growth lag-phase of using glycerol as a growth precursor was eliminated by deleting the glpR gene, significantly enhancing the flux of carbon through the pathway. Subsequently, we improved the oxygen sequestering capacity of this strain through the heterologous expression of Vitreoscilla hemoglobin, which makes this strain able to mineralize TCP under oxygen-limited conditions. Lastly, we further improved intracellular energy charge (ATP/ADP ratio) and reducing power (NADPH/NADP + ratio) by deleting flagella-related genes in the genome of P. putida KT2440. The resulting strain (named KTU-TGVF) could efficiently utilize TCP as the sole source of carbon for growth. Degradation studies in a bioreactor highlight the value of this engineered strain for TCP bioremediation.",
"conclusion": "Conclusions In this work, an efficient TCP-mineralizing strain was created using combinatorial engineering strategies. In brief, we first integrated a synthetic pathway for aerobic TCP mineralization into the chromosome of P. putida KT2440 by functional assembly of various TCP-degrading enzymes from different bacteria. We then enhanced the utilization of TCP as a carbon source by eliminating the growth lag-phase of using GLY as a growth precursor. We additionally engineered the strain with the ability to mineralize TCP under oxygen-limited conditions through the heterologous expression of VHb. Lastly, we further improved the utilization of TCP by increasing intracellular ATP and NADPH levels resulting from deletion of the flagellar operon. We envision that these combinatorial engineering approaches could provide insights into devising engineering strategies to improve the degradation of anthropogenic chemicals by P. putida .",
"introduction": "Introduction Synthetic biology has become a powerful tool to construct complete heterologous metabolic pathways in a host cell by functional assembly of various enzymes from different organisms 1 . Currently, most studies have focused on implantation of diverse synthetic pathways into microbial cells to synthesize high value-added products 2 . Although synthetic biology tools are not widely applied for engineering the organics catabolic pathways in microorganisms so far, they have enormous potential for rational tuning of pathways for efficient degradation of environmental pollutants. A toxic persistent pollutant, 1,2,3-trichloropropane (TCP), has been recognized as an emerging contaminant in groundwater. The global yield of TCP can reach about 50,000 tons annually, and TCP is widely used as a solvent and as building block for the synthesis of other chemicals 3 . TCP is frequently detected in groundwater due to improper waste disposal, its moderate water solubility, and its recalcitrance to biodegradation 4 . TCP contamination of groundwater poses a serious threat to drinking water sources and human health. Therefore, there is an urgent need to develop efficient technologies for the remediation of TCP-contaminated sites. Disposal of TCP by physical or chemical methods is expensive. Bioremediation, which is a simple, safe and cost-effective technique to fight pollution, utilizes and manipulates the biodegradation abilities of living organisms to transform toxic organic pollutants into harmless products 5 . To date, there are no reports on the aerobic degradation of TCP by natural microorganisms. Metabolic engineering has opened up new avenues for the evolution of efficient degradation pathways 6 , which allows the construction of recombinant microorganisms with the capability to degrade TCP under aerobic conditions. Haloalkane dehalogenases (DhaA) catalyze the dehalogenation of TCP to the ( R ) and ( S ) enantiomers of 2,3-dichloropropane-1-ol (DCP) 7 . To improve the catalytic efficiency and enantioselectivity, two DhaA mutants 8 , 9 , DhaA31 and DhaA90R, were generated by directed evolution. Furthermore, haloalcohol dehalogenases (HheC) and epoxide hydrolases (EchA) are responsible for the conversion of DCP to glycerol (GLY) 10 , 11 . Recently, a synthetic biodegradation pathway capable of aerobic biotransformation of TCP into harmless GLY was assembled by the heterologous expression of DhaA, HheC and EchA in Escherichia coli \n 12 , 13 . However, expression of the synthetic pathway is dependent on IPTG induction, which is not suitable for practical application in the large-scale degradation of TCP. Moreover, the plasmid-based expression systems tend to lose the introduced heterologous genes in the absence of selective pressure and the diffusion of antibiotic resistance markers on plasmids poses potential risks to the environment. More importantly, lab-born E. coli strains are not regarded as good candidates for in situ bioremediation due to their strict nutrient demand, poor adaptability and weak competitiveness. In another study, a dhaA31 gene encoding DhaA31 was integrated into the chromosome of the DCP-degrading bacterium Pseudomonas putida MC4 using a transposon delivery system 4 . However, the engineered strain needs to be intensively researched with regard to its biosafety prior to its application for in situ bioremediation due to legislative barriers on the field release of genetically engineered microorganisms. The GRAS (generally recognized as safe) strain P. putida KT2440 is considered as a potential agent for environmental bioremediation of industrial waste. The genome of P. putida KT2440 has been sequenced and multiple tools for genome editing have been devised and implemented 14 – 18 , which have laid a foundation for the metabolic engineering of P. putida KT2440. Recently, P. putida KT2440 has been highlighted as a robust metabolic chassis for catabolic pathway assembly 19 . In this work, we engineered P. putida KT2440 for the efficient degradation of TCP using combinatorial engineering strategies by carrying out the following four tasks: (1) constructed a synthetic pathway for the conversion of TCP into GLY (Fig. 1 ), in particular, prochiral TCP was converted predominantly into ( R )-DCP (ee 90%) by the enantioselective DhaA90R and the accumulation of ( S )-DCP in the pathway caused by the poor activity of HheC toward ( S )-DCP was avoided; (2) deleted the glpR gene to eliminate the growth lag-phase of using GLY as a carbon source; (3) enhanced aerobic metabolism by promoting oxygen delivery through the heterologous expression of Vitreoscilla hemoglobin (VHb); (4) increased intracellular energy charge (ATP/ADP ratio) and reducing power (NADPH/NADP + ratio) by deleting the flagellar operon (a ~ 70 kb DNA segment of the genome). Through the engineering of TCP biodegradation pathway, the enhancement of oxygen transport, and the modulation of energy status and reducing power availability, the resulting strain could mineralize TCP aerobically and utilize TCP as the sole source of carbon for growth. Figure 1 Construction of a synthetic pathway for aerobic mineralization of TCP in P. putida KT2440. Enzyme sources: DhaA from Rhodococcus rhodochrous NCIMB 13064, HheC from Agrobacterium radiobacter AD1, and EchA from Agrobacterium radiobacter AD1.",
"discussion": "Results and Discussion Construction of a synthetic TCP mineralization pathway in P. putida KT2440 Currently, synthetic biology approaches can be applied for implantation of the synthetic biodegradation pathways into an ideal host strain for bioremediation 20 . In this work, a synthetic pathway for aerobic mineralization of TCP comprising DhaA from Rhodococcus rhodochrous NCIMB 13064, HheC from Agrobacterium radiobacter AD1, and EchA from Agrobacterium radiobacter AD1 was assembled in P. putida KT2440 (Fig. 1 ). All these genes ( dhaA90R , hheC and echA ) were integrated into the chromosome of P. putida KT2440 using a previously developed genome editing method with upp as a counter-selectable marker 16 . To improve the efficiency of the synthetic biodegradation pathway, the vgb gene was inserted into the P. putida chromosome, and both the glpR gene and the flagellar operon were knocked out. The resulting mutant strain was designated as P. putida KTU-TGVF. The successful construction of strain KTU-TGVF was verified by PCR and DNA sequencing. The correct amplicons were obtained by PCR with chromosomal DNA of strain KTU-TGVF as the template (Fig. S1 ). The nucleotide sequences of the inserted DNA fragments on chromosome were in accordance with those of four synthetic gene cassettes (Fig. S2 ). The poor activity of DhaA toward TCP and cellular toxicity of TCP and intermediates represent primary bottlenecks of the TCP mineralization pathway, limiting the flux of carbon through the pathway 3 , 12 . To overcome the bottlenecks of aerobic TCP mineralization, previous studies focused on engineering of DhaA, the first enzyme of the TCP mineralization pathway. Two DhaA mutants, DhaA31 and DhaA90R, were obtained using computer-assisted directed evolution. Mutant DhaA31 8 showed 26-fold higher catalytic efficiency toward TCP than the wild-type enzyme (DhaAwt: k \n cat = 0.08 s −1 , K \n m = 2.2 mM, k \n cat / K \n m = 36 M −1 s −1 ; DhaA31: k \n cat = 1.26 s −1 , K \n m = 1.2 mM, k \n cat / K \n m = 1050 M −1 s −1 ), while the ( S )-DCP was accumulated in the pathway because of equimolar production of ( R )-DCP and ( S )-DCP by the non-enantioselective DhaA31 and the poor activity of HheC toward ( S )-DCP. Mutant DhaA90R 9 possessed similar catalytic efficiency toward TCP as the wild-type enzyme (DhaA90R: k \n cat = 0.16 s −1 , K \n m = 6.5 mM, k \n cat / K \n m = 25 M −1 s −1 ), while DhaA90R converted prochiral TCP predominantly into ( R )-DCP (ee 90%), which is the preferred substrate for the enantioselective HheC. In a previous study, the synthetic pathway with DhaA90R reconstructed in E. coli BL21 (DE3) failed to provide sufficient carbon flux to support bacterial growth in minimal medium with TCP 12 . In another study, a novel TCP biodegradation pathway including a chromosome-borne dhaA31 gene and an intrinsic nonselective DCP degradation pathway was reconstructed in a natural DCP-degrading bacterium P. putida MC4. Because of cellular toxicity of TCP and intermediates, the engineered P. putida MC4 showed a prolonged growth lag-phase when cultured in minimal medium with TCP 4 . To construct a practically applicable TCP-degrading bacterium, in this study, a robust environmental bacterium P. putida KT2440, which is resistant to toxic organic solvents, was chosen as a host strain for reconstructing the TCP mineralization pathway. To avoid the accumulation of ( S )-DCP in the pathway, in this study, the enantioselective DhaA90R was selected for the dehalogenation of TCP to DCP. As discussed later in detail, the insufficient carbon flux through the pathway caused by the less active DhaA90R was further improved by eliminating the growth lag-phase of using GLY as a carbon source, enhancing aerobic metabolism, and increasing intracellular energy charge and reducing power. Heterologous expression of the synthetic TCP mineralization pathway in P. putida KT2440 To achieve the optimal expression levels, the optimized gene expression regulatory elements 21 , including a strong constitutive promoter, two Shine-Dalgarno sequences, and a terminator, were designed for controlling the transcription and translation of exogenous genes (Fig. S3 ). In order to verify whether these exogenous genes ( dhaA90R , hheC , echA and vgb ) are transcribed to mRNA in P. putida KTU-TGVF, RT-PCR assays were performed. As expected, these target products were obtained by PCR with cDNA or genomic DNA as template. In contrast, no PCR products were obtained with mRNA or ddH 2 O as template (Fig. S4 ). These results indicated that transcription of these exogenous genes had occurred in P. putida KTU-TGVF. Production of DhaA, HheC and EchA in KTU-TGVF cells was demonstrated by Western blot analysis. Immunoreactive bands, which matched well with the theoretical molecular weights of DhaA (34 kDa), HheC (29 kDa) and EchA (35 kDa), were detected in whole-cell lysates using antibodies against DhaA, HheC and EchA (Fig. S5 ). Function validation for the synthetic TCP mineralization pathway The functionality of DhaA, HheC and EchA in KTU-TGVF cells was validated by GC-MS analyses of resting-cell transformation products. Resting cells of KTU-T1 containing dhaA90R produced a new chromatographic peak with a retention time (RT) of 11.7 min when incubated with TCP, and this peak was identified as DCP by comparing its MS pattern with that of DCP standard (Fig. 2A ). Transformation of DCP by resting cells of KTU-T2 containing hheC produced a new chromatographic peak with a RT of 5.1 min, which had the same RT and MS pattern as ECH standard (Fig. 2B ). After the transformation of ECH by resting cells of KTU-T3 containing echA , a new chromatographic peak appeared at a RT of 12.3 min, the RT and MS pattern of this peak were the same as those of CPD standard (Fig. 2C ). Therefore, KTU-T1, KTU-T2 and KTU-T3 cells expressed functional DhaA, HheC and EchA, respectively. Figure 2 ( A ) GC-MS analysis of transformation of TCP by resting cells of P. putida KTU-T1. Two chromatographic peaks representing TCP and DCP had a RT of 11.3 min and 11.7 min, respectively. ( B ) GC-MS analysis of transformation of DCP by resting cells of P. putida KTU-T2. Two chromatographic peaks corresponding to ECH and DCP appeared at a RT of 5.1 min and 11.7 min, respectively. ( C ) GC-MS analysis of transformation of ECH by resting cells of P. putida KTU-T3. Two chromatographic peaks representing ECH and CPD had a RT of 5.1 min and 12.3 min, respectively. \n TCP degradation experiment was performed with KTU-TGVF in M9 minimal medium supplemented with 0.5 mM TCP. As shown in Fig. 3A , 0.5 mM TCP was almost completely converted to GLY and the vast majority of GLY were further metabolized within 32 h, as revealed by quantification of DCP, ECH, CPD and GLY throughout the experiment. Accompanying with TCP degradation, cell growth was observed (Fig. 3B ), which indicated that this strain could utilize TCP as the sole source of carbon for growth. In contrast, the concentration of TCP did not change and no growth was observed when inoculated with KTU. These results indicated that TCP could be converted to GLY by the heterologous pathway assembled in P. putida KT2440 (Fig. 1 ). Furthermore, GLY could be utilized as the carbon source for cell growth via the intrinsic pathway of KT2440. Taken together, we concluded that the pathway for aerobic mineralization of TCP was functionally assembled in P. putida KT2440. Figure 3 ( A ) Aerobic mineralization of TCP by P. putida KTU-TGVF. P. putida KTU-TGVF was incubated in M9 minimal medium supplemented with 0.5 mM TCP in a shaking incubator at 200 rpm and 30 °C. The initial inoculum density was OD 600 = 0.05. TCP, DCP, ECH and CPD were quantified by GC-MS. GLY was quantified by colorimetric analysis. ( B ) Growth curve of P. putida KTU-TGVF. The OD 600 was measured to estimate cell growth. Bars represent the mean values ± standard deviation of triplicate measurements from three independent experiments. \n Elimination of the growth lag-phase of using GLY as a carbon source GlpR functions as a negative regulator for controlling the utilization of GLY as a growth precursor in P. putida KT2440 and the regulatory mechanisms have been elucidated 22 . Deletion of glpR may eliminate the growth lag-phase of using GLY as a carbon source. In this work, the strain KTU-TGVF (Δ glpR ) could grow aerobically without the lag-phase in minimal medium with TCP as the sole carbon source. In contrast, no growth was observed with the strain KTU-T123 ( glpR \n + ) when using TCP as the sole carbon source. These results suggest that the efficient utilization of GLY may support bacterial growth, alleviate cellular toxicity of TCP and intermediate metabolites and improve the flux of carbon through the TCP mineralization pathway. VHb enhances TCP mineralization under oxygen-limited conditions CO-difference spectrum assays were used to detect VHb activity. The cell extracts from the strain KTU-TGVF showed a characteristic absorption peak at 420 nm when fed with CO, while the peak was not observed with the cell extracts from the strain KTU fed with CO (Fig. S6 ). These results indicated that the strain KTU-TGVF expressed functional VHb. VHb delivers the O 2 to the respiratory chain, enhancing ATP generation and NADH consumption at low O 2 concentrations. This process could improve the efficiency of bacterial aerobic respiration, increase the carbon flux through the tricarboxylic acid cycle and improve the aerobic growth of bacteria. Heterologous expression of VHb in a variety of hosts has been shown to improve cell growth, protein synthesis, metabolite productivity, and bioremediation under oxygen-limited conditions 23 , 24 . In a previous study, integrating vgb into the chromosomes of P. aeruginosa and Burkholderia sp. strain DNT could improve growth and degradation of 2,4-dinitrotoluene or benzoic acid under hypoxic conditions 25 . In another study, P. putida KT2440 was engineered for the anoxic biodegradation of 1,3-dichloroprop-1-ene by introducing various genes retrieved from facultative anaerobe and aerotolerant bacteria 26 . Sufficient oxygen is the key factor for complete oxidative mineralization of TCP. The success of aerobic mineralization of TCP in oxygen-restricted environments will largely depend on the oxygen sequestering capability of TCP-degrading bacteria. In this work, significant growth was observed for the strain KTU-TGVF expressing VHb under oxygen-limited conditions, while no growth was observed for the strain KTU-TGF without VHb expression under oxygen-limited conditions (Fig. 4 ). These results suggest that VHb may improve the ability of P. putida KT2440 to compete for limited oxygen in hypoxic environments, thus making this bacterium more competitive in actual environments such as a packed-bed bioreactor. This study not only underscores the value of P. putida KT2440 as a versatile biocatalyst for biotransformation under oxygen-limited conditions but also highlights the value of metabolic engineering for expanding the catalytic repertoire of P. putida KT2440. Figure 4 Growth curves of P. putida KTU-TGF and KTU-TGVF under oxygen-limited conditions. P. putida strains were incubated in 25 ml glass vials with a screw cap mininert valve containing 20 ml of M9 minimal medium supplemented with 0.2 mM TCP in a shaking incubator at 80 rpm and 30 °C. Cell growth was estimated by measuring the OD 600 of the culture broth. Bars represent the mean values ± standard deviation of triplicate measurements from three independent experiments. \n Increased intracellular ATP and NADPH levels improve aerobic mineralization of TCP Both the production and the motion of flagella are energy-consuming processes for the cell. In E. coli , flagellar production consumes about 2% of the biosynthetic energy expenditure of the cell, while flagellar motion demands about 0.1% of the total energy cost 27 . In this work, a non-flagellated strain KTU-TGVF was generated from the original strain KTU-TGV by deleting the flagellar operon in the genome. Accordingly, the flagellar operon knockout mutant strain showed the complete absence of the flagellum structure when observed using TEM after negative staining. In contrast, the flagella were observed in the wild-type strain KT2440 (Fig. S7 ). These results indicated that the flagellar synthesis was completely blocked in the mutant strain. It has been reported that the non-flagellated strain KT2440 showed increased intracellular ATP and NADPH levels 28 . In our study, the ATP/ADP ratio in the strain KTU-TGVF was 1.3-fold higher than that in the strain KTU-TGV (Fig. 5A ). Since the ATP levels in bacterial cells are tightly regulated, a difference of 30% could significantly influence cellular functions. These results indicated that eliminating the consumption required for flagella synthesis and rotation greatly changed the energy status of the cells. Figure 5 Determination of the ATP/ADP ( A ) and NADPH/NADP + ( B ) molar ratios in P. putida . The intracellular levels of ADP, ATP, NADP + and NADPH were determined as described in Materials and methods. Bars represent the mean values ± standard deviation of triplicate measurements from three independent experiments. \n Except for the energy consumption, a considerable amount of reducing power, mainly in the form of NADPH, is indispensable for basic anabolic processes, such as the biosynthesis of building blocks for biomass 29 . In this work, the strain KTU-TGVF had a 1.2-fold higher NADPH/NADP + ratio than did the strain KTU-TGV (Fig. 5B ). Furthermore, the growth kinetics of the strain KTU-TGVF and KTU-TGV were compared. The strain KTU-TGVF showed a higher growth rate in M9 minimal medium containing 0.5 mM TCP than did the strain KTU-TGV. The KTU-TGVF and KTU-TGV cultures reached a maximum OD 600 of 0.104 and 0.088 at 32 h, respectively (Fig. S8 ). Obviously, the strain KTU-TGVF could utilize TCP more efficiently as the sole carbon source for growth, which may be a reflection of increased resistance to cellular toxicity of TCP and intermediate metabolites. Increased resistance to toxic substrates may be due in part to enhanced anabolic capability resulting from the increase of ATP and NADPH levels within the cell. Taken together, the efficient mineralization of TCP by the strain KTU-TGVF may be attributed to the efficient utilization of GLY as a growth precursor and the increased intracellular ATP and NADPH levels. Treatment of TCP-contaminated water in a lab-scale bioreactor The flagella-deficient strain KTU-TGVF showed a distinct change in biofilm formation. The quantities of biofilm formed by the strain KTU-TGVF were 4.3- and 9.5-fold higher at 24 and 96 h, respectively, than those formed by the strain KTU-TGV, as judged by the CV staining assay (Fig. S9 ), when these strains were grown in M9 minimal medium containing 0.4% (w/v) glucose. The observation is in agreement with a previous report 28 , in which the non-flagellated P. putida KT2440 cells form more biofilm than the wild-type KT2440 cells. Recently, the biofilm formation capability of P. putida KT2440 was improved for boosting biodegradation of haloalkanes by the modulation of the intracellular c-di-GMP level 30 . Biofilm offers protection against a hostile environment and helps bacteria persist within the environment. Therefore, the strain KTU-TGVF has potential for use in a packed-bed bioreactor treating TCP-contaminated water because its high biofilm formation ability is favorable for the immobilization of cells on the support material in wastewater treatment. To explore the feasibility of using strain KTU-TGVF for TCP bioremediation, we designed a packed-bed bioreactor and tested the removal of TCP from wastewater streams in a consecutive process (Fig. S10 ). KTU-TGVF cells were grown in M9 minimal medium containing 0.5 mM TCP plus 0.4% glucose to OD 600 = 0.4 and were subsequently inoculated into the reactor for immobilization on ceramic rings. Previous studies have shown that ceramic rings as the support material are suitable for the immobilization of microbial cells 4 , 31 . After inoculation, the reactor was operated under fed-batch conditions for 4 days to promote attachment of the cells to the support material. Within a period of 4 to 18 days, TCP (0.05 to 0.2 mM) was supplied continuously at a rate of 0.1 ml/min. During the operating period, the maintenance of a high hydraulic retention time (HRT, 133 h) may be favorable for cell growth and biofilm formation. HRT is considered as an important operating parameter directly influencing the bioreactor performance 4 , 31 . After 18 days, the influent concentration of TCP was maintained at 0.2 mM, and the reactor was continuously operated for another 30 days. The reactor performance was evaluated by continuous monitoring of the effluent concentrations of TCP, DCP and chloride (Fig. 6 ). Under these operating conditions, the removal efficiencies of TCP were 95 to 97% and quantitative release of chloride (TCP/chloride, molar ratio 1:3) was observed. We detected minute quantities of DCP in the reactor effluent. When inoculated with the strain KTU, the TCP effluent concentration remained unchanged relative to the influent concentration. Figure 6 Degradation of TCP by P. putida KTU-TGVF in a lab-scale bioreactor. Reactor performance was assessed by continuously monitoring the effluent concentrations of TCP, DCP and chloride during a 30-day operating period. Symbols: ◾, TCP influent; ⚫, TCP effluent; ▵, DCP effluent; ◊, chloride effluent. \n Cells from the reactor effluent were spread on LB agar plates and colonies formed were checked for their identity. All 20 colonies tested were identified as P. putida KT2440 by the 16 S rRNA gene sequencing. Furthermore, both the four gene insertions ( dhaA90R \n + , hheC \n + , echA \n + , vgb \n + ) and the two gene deletions (∆ glpR , ∆flagellar operon) were detected by PCR from all 20 colonies tested, and the detection results of one colony are shown in Fig. S11 . These results demonstrated that the inoculated KTU-TGVF cells thrived and accounted for the observed TCP degradation during a 30-day operating period in the bioreactor. In the future, the efficient TCP-mineralizing strain KTU-TGVF coupled with a field-scale reactor has enormous potential to be applied for the treatment of industrial wastewater containing TCP."
} | 6,584 |
38389536 | PMC10883772 | pmc | 4,507 | {
"abstract": "Microbial communities have been used as important biological tools for a variety of purposes associated with agriculture, the food industry and human health. Artificial engineering of microbial communities is an emerging field of research motivated by finding stable and efficient microbial systems. However, the successful design of microbial communities with desirable functions not only requires profound understanding of microbial activities, but also needs efficient approaches to piece together the known microbial traits to give rise to more complex systems. This study demonstrates the bottom-up integration of environmentally isolated phototrophic microalgae and chemotrophic bacteria as artificial consortia to bio-degrade selected volatile organic compounds (VOCs). A high throughput screening method based on 96-well plate format was developed for discovering consortia with bioremediation potential. Screened exemplar consortia were verified for VOCs degradation performance, among these, certain robust consortia were estimated to have achieved efficiencies of 95.72% and 92.70% and near 100% removal (7 days) of benzene, toluene, and phenol, respectively, with initial concentrations of 100 mg/L. VOCs degradation by consortia was mainly attributed to certain bacteria including Rhodococcus erythropolis , and Cupriavidus metallidurans , and directly contributed to the growth of microalgae Coelastrella terrestris ( R = 0.82, p < 0.001). This work revealed the potential of converting VOCs waste into algal biomass by algae-bacteria consortia constructed through a bottom-up approach. The screening method enables rapid shortlisting of consortia combinatorial scenarios without prior knowledge about the individual strains or the need for interpreting complex microbial interactions.",
"conclusion": "4 Conclusion This research has successfully demonstrated the bioremediation potential of environmentally isolated individual strains, particularly in the degradation of benzene, toluene, and phenol, and in the efficient conversion of VOCs into algal biomass through well-combined algae-bacteria consortia. The screening method, which utilises total algal chlorophyll content as a proxy, has proven to be highly efficient and rapid in identifying suitable algae-bacteria combinations for VOC degradation. The assessing parameters associated with this method, especially algae growth contribution index ( CI ) and functional stability ( FS ), showed the effectiveness of shortlisting high-performance consortia. Notably, Rhodococcus erythropolis and Cupriavidus metallidurans emerged as key bacteria in VOC catabolism within these consortia. The screening method provides a benchmark for engineering more stable and effective microbial systems for bioremediation. Moreover, it aids in narrowing the targets for subsequent metabolic or proteomic studies, offering deeper insights into microbial interactions. This study, however, faced limitations including low-resolution microbial identification and a narrow focus on specific pollutants and strains. Research gaps remain in understanding the real-world application and scalability of industrial VOC waste treatment. Future research should explore a broader range of pollutants and microbes, employ advanced analytical techniques, and focus on practical applications.",
"introduction": "1 Introduction The majority of biological treatment technologies for pollutant remediation utilise bacteria due to their diversity and metabolic versatility in utilising a variety of substrates ( Antizar-Ladislao, 2010 ; Numberger et al., 2019 ). Many well-characterised bacterial genera such as Pseudomonas , Rhodococcus and Bacillus , have been used to degrade a variety of toxic compounds including formaldehyde ( Roca et al., 2008 ), benzene, phenol, and toluene ( Reardon et al., 2000 ; Abuhamed et al., 2004 ) from wastewater, gaseous waste streams and polluted soils ( Antizar-Ladislao, 2010 ; Lu et al., 2012 ; Numberger et al., 2019 ). Whilst bacteria dominate pollutant bioremediation, the inclusion of microalgae in the process presents attractive benefits. Microalgae are versatile organisms that can capture CO 2 from biodegradation processes, providing oxygen in return to support bacterial enzymatic activites. Microalgal biomass also serves as a rich source of lipids for biofuel production, carbohydrates for ethanol and biogas ( Chandrasekhar et al., 2023 ), and fibrous polymers for biodegradable plastics ( Zanchetta et al., 2021 ). They can also provide high-value products like pigments ( Stachowiak and Szulc, 2021 ) and vitamins ( Arora and Philippidis, 2023 ). Given these benefits, microalgae are viewed as solar-driven cell factories for sustainable biomanufacturing. Algae-bacteria communities are some of the most common and fundamental forms of mixed microbial systems in nature, in which, nutrient exchange is the most basic interaction associated with their co-existence. In aquatic environments, for example, algae are responsible for providing a large amount of dissolved organic carbon (DOC) that can work as the major carbon source for the survival of heterotrophic bacteria ( Weinberger et al., 1994 ; Takemura et al., 2014 ). An early study by Humenik and Hanna (1971) showed that combined green alga Chlorella and bacteria in activated sludge resulted in sufficient algal-originated O 2 supply for bacteria and increased algal protein production within a wastewater treatment (WWT) system. This early finding became the starting point of many studies motivated by seeking other superior properties over bacteria-only systems, such as self-supporting systems and enhanced nutrient removal ( Min et al., 2011 ; Zhang et al., 2012 ; Inoue and Uchida, 2013 ). These studies followed a top-down approach for consortia design, i.e., using naturally occurring algae–bacteria consortia as the inoculant, which usually involves highly complex microbial communities. Integrating individual strains to form multi-strain systems provides a bottom-up route of algae-bacteria consortia engineering. Usually, prior understanding and characterisation of the biological phenotype of the strains is vital. An early study by Mouget et al. (1995) suggested that two Pseudomonas strains isolated from laboratory algal cultures, Pseudomonas diminuta and Pseudomonas vesicularis , could stimulate the growth of co-cultured green microalgae Scenedesmus bicellularis and Chlorella sp., simply by providing CO 2 and consuming oxygen to maintain a preferable condition for photosynthesis. As the knowledge base surrounding the algae-associated microbiome (or phycosphere) becomes more available, intentionally selected bacterial partners for specific algae species have emerged as a viable option for the bottom-up construction of algal-bacterial consortia. In another study, Dong and Zhao (2004) obtained 12.92 mg/L of astaxanthin production in a mixed culture of Haematococcus pluvialis and Phaffia rhodozyma , which was 3.5- and 11-fold higher than those in axenic cultures of the two strains, respectively. Wirth et al. (2015) mixed microalgae, Chlamydomonas sp. and Scenedesmus sp., with bacteria Rhizobium sp., and increased algal biomass yields by up to 24% as the result of O 2 -CO 2 exchange. Due to the same mechanism, Papone et al. (2012) obtained enhanced lipid production in Chlorella sp. KKUS2 by choosing fungi Toluraspore YU5/2 , and Toluraspore Y30 for co-culturing. Apart from wild-type strains, engineered bacterial strains have even greater potential for applications in algae-bacteria co-culture systems. For example, Therien et al. (2014) designed an artificial cross-feeding system consisting of glycogen synthase knockout mutant Synechococcus sp. 7002 that provides acetate to support lipid-producing Chlamydomonas reinhardtii . In addition to function enhancement and productivity promotion ( Xie et al., 2013 ; Amin et al., 2015 ; Segev et al., 2016 ), the adverse interaction between certain algae and bacteria can also be utilised to address certain environmental challenges. For example, harmful algal bloom control can be achieved by bacteria that release algicidal molecules including 1-acetyl-β-carboline ( Kim et al., 2015 ), Orfamide A ( Aiyar et al., 2017 ) and urocanic acid ( Kim et al., 2018 ; Zhuang et al., 2018 ). Despite these achievement, challenges remain for both top-down and bottom-up approaches to form algae-bacteria consortia. Top-down approaches struggle with the specificity of target interactions and often fail to elucidate community interactions or underlying mechanisms, hindered by the sheer number of possible interactions. Efforts to bridge this gap using omics and evolutionary tools have been made ( Okurowska et al., 2021 ; Qu et al., 2021 ; Si et al., 2022 ), but more efficient methods integrating top-down data with functional insights are needed for designing specific algae-bacteria consortia. In bottom-up approaches, predictability is the main hurdle. Microbiome biology is influenced by various ecological principles like competition ( Griffin et al., 2004 ; Hibbing et al., 2009 ), ecosystem succession ( Fierer et al., 2010 ; Jiménez et al., 2017 ), and mass cycling ( Gralka et al., 2020 ), which shape microbial community assembly. While computational tools have advanced our understanding and prediction capabilities ( Coyte et al., 2015 ; Zomorrodi and Segrè, 2016 ; Marsland et al., 2019 ; Daly et al., 2022 ), experimental validation remains crucial. The complexity of multi-species microbial systems, with diverse interacting components and pathways, leads to unpredictable system-scale behaviours ( Gilbert and Henry, 2015 ). Olson et al. (2012) highlight the difficulty in designing or optimising complex microbial systems without extensive data. Consequently, most synthetic algae-bacteria communities studied are limited to two-member systems ( Mujtaba et al., 2017 ; Contreras-Angulo et al., 2019 ), with multi-strain studies being rare ( Le Chevanton et al., 2013 ). Here we aimed to design a screening method to enable the identification of high-performance algal-bacterial consortia that are capable of biodegradation. We targeted some of the most recognised health-threatening volatile organic compounds (VOCs) related to biomass combustion ( Kim and Shim, 2010 ; Wielgosiski, 2012 ) and the isolation of co-existing strains in a relevant polluted environment. This work is motivated by limited knowledge of how to construct stable synthetic algae-bacteria consortia and the absence of efficient methodologies accessible in most research laboratories. We followed a bottom-up route to integrate environmentally isolated microalgae and chemotrophic bacteria to form VOCs-degrading consortia. By implementing a microplate-scale screening method and measuring chlorophyll abundance in this closed system, a high throughput data generation approach was developed which facilitated rapid shortlisting of algae-bacteria combinations. In addition to algae growth data, our experimental design takes advantage of the consortia combination structure to give rise to two scoring indexes which uncovered the algae growth-promoting potential and functional stability of consortia without the need for interpreting complex interaction networks. The screening method exhibits fair predictive accuracy for consortia performance against technical error and variation of other potential affecting factors during up-scaling, evidenced by the quantification of algae growth and VOCs degradation in flask-grown exemplar consortia. In addition, an end-point algae-bacteria community structure analysis was conducted to provide insight into the association between bacterial strains and the observed behaviour in exemplar consortia.",
"discussion": "3 Results and discussion 3.1 Microbial enrichments and identification Environmental isolation, together with one Pseudomonas putida KT2440 obtained from lab collection, resulted in a total of 26 bacteria ( Supplementary Figure S1 and Supplementary Table S1 ) with VOC-resistant features and 6 exhibited potential VOC-degrading phenotypes. Two soil algae isolates obtained from autotrophic cultures, were characterised for their growth characteristics ( Supplementary Figures S5 , S6 ). One strain which survived up to 200 mg/L of each VOC was selected as the candidate algae strain ( Supplementary Figure S7 ). The 16S rDNA sequencing results revealed the dominance of gram-negative (GN) Pseudomonas and gram-positive (GP) Rhodococcus in these 26 bacteria isolates ( Supplementary Figure S1 ), both known for their capability to degrade aromatic compounds and use them as growth substrates ( Kauppi et al., 1998 ; Cébron et al., 2008 ). Additionally, two isolates of Delftia , a GN bacterium capable of degrading low molecular weight phenolic compounds and aniline ( Sheludchenko et al., 2005 ; Juárez-Jiménez et al., 2010 ), were also classified as degraders. These genera of bacteria possess multiple enzyme systems for breaking down aromatic compounds ( Kauppi et al., 1998 ; Shigematsu et al., 2003 ; Patrauchan et al., 2008 ) with Rhodococcus jostii RHA1 predicted to have over 200 oxygenases and 30 pathways for aromatic compound catabolism ( McLeod et al., 2006 ). Although bacterial isolates like Plantibacter , Achromobacter , and Ochrobactrum anthropic did not exhibit direct signs of VOC degradation, they demonstrated VOC resistance in this experiment, along with other beneficial characteristics such as plant growth promotion identified in previous research ( Chakraborty et al., 2009 ; Meng et al., 2014 ; Mayer et al., 2019 ; Jiménez-Vázquez et al., 2020 ), underscore their potential significance in an algae bacteria consortium. The employment of these strains may result in the broader unknown role beyond VOC degradation. However, considering the theoretical possibility of 2 26 (or 268,435,455) combinations, it was impractical to explore all of these in-vitro . To manage this complexity, a final collection of 6 VOC-degrading bacteria (degraders) and 7 non-VOC-degrading bacteria (non-degrader) and Coelastrella terrestris ( Table 2 ) were chosen as the “building blocks” for the algae-bacteria consortia which resulted in 6 × 2 7 = 768 different combinations. Growth data of a total of 1920 samples were measured, including biological duplicates for each consortium (768 × = 1,536), 90 controls, 102 blanks and 32 × 6 consortia containing only algae and degraders. Details of microplate layouts are presented in the Supplementary Table S5 . Table 2 Strains involved in constructing algae-bacteria consortia. Microbe type Strain label Identification Sequence accession number(s) Bacteria: VOC degraders A \n Pseudomonas fluorescens \n PP106129 B \n Rhodococcus erythropolis \n PP106147 C Pseudomonas sp. PP106139 D Delftia sp. PP106142 E \n Rhodococcus sp 1 . \n PP106148 F \n Rhodococcus sp 2 . \n PP106152 Bacteria: VOC non-degraders 1 \n Pseudomonas syringae \n PP106131 2 \n Agromyces atrinae \n PP106151 3 \n Cupriavidus metallidurans \n PP106143 4 \n Ochrobactrum anthropi \n PP106150 5 \n Plantibacter flavus \n PP106144 6 Plantibacter sp. PP106153 7 \n Rhodococcus sp 3 . \n PP106149 Microalga Ag \n Coelastrella terrestris \n PP106155 3.2 Monitoring algae growth as a proxy for consortia performance in the screen The screening experiment was performed as a closed system without air exchange, therefore the axenic algal control samples showed no growth due to their autotrophic nature and the lack of access to atmospheric CO 2 as a carbon source. Bacterial degradation of VOCs, therefore, determines the availability of CO 2 as a carbon source, which directly limits the maximum potential growth of algae. The VOCs per litre of medium contributed only a theoretically 16.12 kJ of chemical energy under aerobic conditions ( NIST Chemistry WebBook, n.d. ) and 27.16 mmol of CO 2 equivalent, the fixation of which requires at least 127.11 kJ of photonic energy under the 10% of photosynthetic efficiency (PE), the theoretical upper limit of microalgae ( Roux, 2016 ; Masojídek et al., 2021 ). For this reason, actual algae growth mainly reflects the amount of photonic energy fixed into consortia, which suggests the consortium’s capability for energy capture, primary production and suitability for scale-up. This mimicked closed ecosystem is hypothetically illustrated in the conceptual diagram ( Figure 2 ). Figure 2 An ecosystem model of the algae-bacteria consortium in the high throughput screen. An algae-bacteria consortium and its containing environments is an ecosystem where the majority of carbon sources for the growth of the entire microbial community are from VOCs degradation by bacteria with algal photosynthesis as the main energy input and algae biomass as the final VOC-originated carbon sinks. 3.3 Strains configuration for screening By categorising characterised bacteria into VOC degraders and non-degraders, it is possible to significantly reduce the screening load by minimising the members of all possible combinations while analysing as many non-repeated consortia as possible. This helps to balance high throughput and analysis rate. Although VOC degradation may emerge among non-degrading bacteria because of microbial interactions, the ‘algae + degrader + non-degrader’ structure is expected to increase the possibility of discovering high-performance VOC removal consortia. 3.4 Comparison of screening results Distributions of the three consortia assessment parameters across the 768 consortia are visualised in Figure 3 . The GSs of 768 consortia exhibited a bimodal distribution, characterised by 679 positive values and 87 negative values ( Figure 3 A4 ). Since GS was based on a direct comparison of chlorophyll accumulation in tested consortia versus axenic algae control, its distribution showed that most consortia had more algae growth than the control. Consortia with high algae growths were found to be associated with two Rhodococcus strains (Degraders B and F), and contrastingly, two Pseudomonas strains, (Degraders A and C), were found in some of the consortia with the lowest growth. Non-degrader 3 ( C. metallidurans ) had a stronger association with high- GS consortia by presenting in 62.8% (241/384) of those with GS value higher than 5.59 (median), while other non-degraders were present in 50% or less. GS -negative consortia are distinctly characterised by the absence of degraders B ( R. erythropolis ), E ( Rhodococcus sp 1 ) and F ( Rhodococcus sp 2 ) and non-degrader 3 . Figure 3 Screening data of 768 different consortia. Growth score data ( GS ) were visualised in the bar chart (A1) , polar plot (A2) , average values under 6 different degraders (A3) and demonstration of GS value distribution via histogram (A4) . The same visualisations are also adopted for parameters CI (B1–B4) and FS (C1–C4) . The distribution of CI ( Figure 3 B4 ) spanned both negative and positive values, with a slight trend of being centred around zero. This suggests the existence of both algae growth promotion and inhibition across the screened consortia, a greater number (587) of which, however, had neutral effects on algae growth. Unlike GS , the comparative nature of CI allows it to uncover the behaviour of a consortium against the potential influence of different co-cultured bacterial strains while also improving the measurement accuracy due to an increased number of observations, which led to very distinct patterns between GS and CI of the same consortia. Interestingly, positive CI values were found evenly in consortia containing degraders A, B, C and D, while high CI values (≥ 1.97, median) seem to be subject to the presence of non-degrader 3. In contrast, CI values were significantly lower in consortia containing degraders E and F among which half (138/256) were found negative. This could be explained by their high algae growth promotion effects being dominant and leaving little room for further effects beyond their original contribution. Interestingly, consortia based on another Rhodococcus (degrader B) were found to have much higher CI values despite already having high GS values. This suggests the potential existence of special positive interactions associated with this degrader being synergistically improved. FS uncovered extra information about the stability of a consortium by assessing the difference between its CI and GS . Theoretically, stable consortia are relatively independent systems with unchanged behaviour when co-cultured with different bacteria and their GS and CI values tend to be close to each other, thus having FS values close to 1. In contrast, FS values deviating from 1 indicate consortia of weaker resilience which are more subject to the influence of additional bacteria and end up in either algae growth promotion ( FS > 1) or growth inhibition ( FS < 1) effects. Most of the screened consortia (517) had low stability with FS values lower than 1 featured by a peak ( Figure 3 C4 ) lying between the range of 0 and 0.6. The trailing pattern of FS ( Figure 3 C4 ) towards higher values greater than 1 indicates that fewer consortia (249) had synergistic algae growth promotion effects and were closely associated with degraders A, C and D ( Figures 3 C1 , C2 ). FS value in consortia based on the three Rhodococcus degraders (B, E and F) were homogeneously below 1, indicating that they were not significantly affected by any synergistic effects regardless of the varying combinations of non-degraders. A few stable consortia were found to be strongly related to degrader B, and non-degrader 3 ( C. metallidurans ) which had average FS values close to 1. 3.5 Combination significance and single strain effect test Significance test results ( Figure 4 ) revealed how the three parameters vary as the bacteria combinations change. After false positive rate (FPR) correction, GS was only found most affected ( p < 0.05) by 63 consortia ( Figure 4C ) which are associated with the presence/absence of degrader B, F, and non-degrader 3 ( Figure 4C , insert). CI values vary significantly under the presence and absence of 273 consortia wherein degraders A, C and D, and non-degraders 3 and 7 appear to be the primary bacteria strains that led to CI variation. A larger number (475) of consortia were found to significantly affect FS values. These consortia were featured by the absence of Degrader B. Non-degraders, 3 and 7. Figure 4 Combination significance test result. (A,B) Effects on the three parameters. The bar charts presented the normalised value of the three parameters with (light colour, marked as logic 1) and without (dark colour, marked as logic 0) specific strain. (C–E) Significance test on all possible combinations for GS , CI , and FS where histograms show the number of consortia hitting the p -values < 0.05 threshold (HB-corrected). Number of replicates? The effect of individual strains ( Figures 4A , B ) indicates that two Pseudomonas (degrader A, C) had a very similar effect on all three assessment indexes. Specifically, consortia involving A and C tend to have lower GS but higher CI and FS compared to consortia that do not contain the two strains. This trend opposes what was observed in consortia containing Rhodococcus degrader (degrader B, E and F) which are featured by higher GS and significantly lower CI and FS . Delftia (degrader D) had a similar effect on CI and FS to the two Pseudomonas degraders but higher GS . Among non-degraders, C. metallidurans had the strongest effects on the consortia, evidenced by significantly higher GS and CI and reduced FS values associated with its presence. Also, non-degrader 7 resulted in higher CI and FS values. The rest of the non-degraders strains had a weak influence on GS and CI albeit no obvious effect on the FS . Based on the observation of the three parameters, degraders appear to be the fundamental functional unit in the consortia although their effects vary with the strains. Rhodococcus has stronger growth-promoting effects on C. terristeris evidenced by their association with high GS values, which could be a result of the efficient single-ring aromatics metabolic systems in this genus ( McLeod et al., 2006 ; Patrauchan et al., 2008 ; Táncsics et al., 2008 ). However, due to the same reason, Rhodococcus could easily dominate the VOC degrading processes thus outcompeting and shedding the effect of other non-degraders and thus leading to enhanced functional stability ( Coyte et al., 2015 ). Contrary to this, two Pseudomonas (degrader A, C) and Delftia (degrader D) were considered to have lower performance in supporting algae photosynthesis, evidenced by their frequent presence in low- GS consortia despite many species in these genera having been recognised as strong aromatic hydrocarbon degraders ( Kauppi et al., 1998 ; Shigematsu et al., 2003 ). Conversely, while these strains may be less efficient in VOC degrading, their reduced competitiveness left niches for the activities of non-degraders which in turn, may have improved the overall consortia performance, as reflected by high FS . Among the 7 non-degraders, Cupriavidus metallidurans and Rhodococcus Sp 3 seem to be the only two strains affecting the consortia behaviour and thus were very likely to directly participate in the VOCs degradation process due to their frequent presence in consortia of high GS and CI value and metabolic capability reported in other studies ( Cébron et al., 2008 ; Alviz-Gazitua et al., 2019 , 2022 ). Other non-degraders including Pseudomonas syringae , Agromyces atrinae , Ochrobactrum anthropic and two Plantibacter strains were seen as ‘weak’ bacteria members in the consortia with little effect on any of the parameters regardless of their presence or absence, nor the variation of their combinations. 3.6 VOCs degradation Exemplar consortia aligned with screening results and their VOC degradation performance are listed in Table 3 . Table 3 Exemplar consortia and their average daily VOCs degradation within 7 days. Consortia Screening parameters VOCs degradation rate (mg/L) OD 685 Growth pattern \n GS \n \n CI \n \n FS \n B T P THF Total B14 7.19 2.72 0.62 30.85 32.79 96.54 *3.20 163.39 0.33 \n \n E14 9.86 −0.45 0.19 34.49 37.76 84.28 *25.09 181.61 0.13 B23 6.86 3.25 0.72 85.49 85.81 91.38 *10.95 273.62 0.91 E23 12.80 −1.42 0.07 95.72 92.70 93.76 *26.74 308.91 0.53 B27 3.03 1.25 0.92 20.98 23.02 100 *11.36 157.99 0.32 E27 10.97 −2.16 −0.03 63.41 65.62 100 *9.17 243.14 0.27 B123 9.19 6.30 0.84 92.14 87.48 100 *11.97 291.77 0.86 E123 8.49 1.85 0.45 87.89 87.80 93.92 *12.11 281.71 0.56 B125 10.15 4.00 0.56 25.03 43.86 94.56 *33.38 196.83 0.45 E125 10.59 −0.13 0.20 38.46 54.83 97.68 *19.81 210.78 0.28 B123456 5.23 4.35 0.69 65.26 69.67 91.22 *25.55 251.71 0.84 E123456 5.13 4.07 0.67 82.52 84.86 91.86 *21.09 280.34 0.68 B124567 5.41 1.02 0.30 21.69 37.06 96.53 *5.48 160.76 0.36 E124567 6.31 0.27 0.19 26.90 40.43 98.45 *9.06 174.84 0.37 CT 0 N/A N/A *32.35 *34.09 *5.60 *6.87 *78.92 0.01 BK N/A N/A N/A *23.13 *27.23 *3.06 *3.28 *56.70 0.00 Consortia name indicates the bacterial strains of the consortia following the labels listed in Table 2 ; All 14 consortia contain microalgae C. terristeris . CT: axenic algae control; BK: media-only blank; B: benzene, T: toluene, P: phenol, THF: tetrahydrofuran; Data marked with * are estimated results based on experimental fact, for example, a ‘negative’ VOC concentration indicate zero degradation. After considering sampling loss and aqueous-gas distributions, GC-FID analysis ( Figure 5 ) confirmed that the VOCs degradation by consortia was primarily biological since no natural degradation was observed in the blank groups despite their high volatility and even light-sensitivities (phenol). Rapid degradation of benzene and toluene occurred mainly between days 2 and 4, with daily degradation rates of 20.61 to 39.86 mg·L −1 ·D −1 for benzene and 18.58 to 35.45 mg·L −1 ·D −1 for toluene. Consortia B23, E23, B123, E123, B123456, and E123456 showed high benzene removal (63.41–95.71%) and toluene removal (65.62–92.70%) over 7 days. In contrast, other consortia degraded no more than 34.48% of benzene and 40.43% of toluene. Phenol appears to be the most readily degradable VOC due to its relatively central position on aromatic compounds catabolic pathways of associated bacteria ( Selvakumaran et al., 2011 ; Jiang et al., 2015 ; Alviz-Gazitua et al., 2022 ). Although phenol degradation was slower in consortia E14, E27 and E125, it was nearly 100% removed within 7 days and served indirectly as the basic carbon source for algae growth in all 14 exemplar consortia. THF showed no obvious sign of degradation by any of the consortia. Figure 5 Degradation of VOCs in exemplar consortia. (A) Benzene, (B) Toluene, (C) Phenol, (D) THF. Consortia with high VOC removal, including B23, E23, B123, E123, B123456, and E123456, also had high algae biomass content, as indicated by OD 685 measurements. A strong correlation ( R = 0.82, p < 0.001) was found between total VOC degradation and OD 685 . The performance of consortia in VOC degradation and OD 685 was influenced by the presence of non-degraders, especially C. metallidurans , while no significant difference in VOC degradation was observed between consortia with different degraders ( p = 0.71). Surprisingly, no significant correlation was found between total algae biomass and GS ( R = −0.18, p = 0.54) of these flask-grown exemplar consortia. One possible cause of this deviation is the different sensing mechanisms from which data were generated since the fluorescence signal is measured more sensitively with less interference on the optical pathway caused by suspended bacteria cells ( Lakowicz, 2006 ). Also, larger flask-scale culture volume may influence the growth of algae due to differences in factors such as growth space, nutrient availability and luminance patterns. Conversely, CI demonstrated a higher level of efficacy when predicting the algae growth in flask scale exemplar consortia, evidenced by its strong positive correlation with OD 685 ( R = 0.73, p < 0.001), which could be attributed to its comparative nature, as well as its extended number of observations. A moderate positive correlation lies between OD 685 and FS ( R = 0.58, p = 0.003) which, however, reflects the stability of a consortium against the potential influence of additional bacteria as an important reference for the consortia robustness. 3.7 Microbial community structure analyses The microbial community structure was profiled in terms of total algae biomass measured as OD685 over a 7-day sum ( Figure 6A ), total bacterial cell densities ( Figure 6B ), and the relative abundance of each bacterial strain on day 7 ( Figure 6C ) within the exemplar consortia. Figure 6 Microbial community structure analyses. Total algae biomass (A) calculated as the 7-day sum of OD 685 values, estimated total bacterial cell density (B) and relative abundance (C) . Both degraders were still detected on day 7, however, their relative abundance varied depending on co-cultured non-degraders. Although both belong to the same genus, the presence of degrader E tends to result in larger overall bacteria populations in the consortia than degrader B (B124567 as an exception). However, little difference was found in the VOCs degradation performance between flask-grown consortia containing the two Rhodococcus degraders, suggesting that B is a more competitive strain and may inhibit the growth of other co-cultured bacteria. Among non-degraders, C. metallidurans exhibited the strongest dominance evidenced by its high relative abundance and absence of other co-cultured non-degraders in the associated consortia. The presence of C. metallidurans also saw faster degradation of benzene and toluene, suggesting that its aromatic compounds catabolic traits reported in other studies were highly likely to have taken effect in the consortia. ‘Weaker’ non-degraders such as Agromyces sp. and Ochrobactrum sp. were only found contributing to small proportions of the overall community populations in consortia where P. syringes and C. metallidurans were absent (except in consortium E14). Neither of the two Plantibacter strains (non-degrader 5 and 6) were detected in their associated consortia on Day 7. Notably, degraders B and E, both from the Rhodococcus genus, resulted in distinct community structures. Consortia with degrader B had smaller bacterial populations compared to those with degrader E, suggesting degrader B’s competitive nature, possibly inhibiting co-cultured bacteria growth while favouring algae. This competitive behaviour of degrader B, identified as R. erythropolis , might be due to its production of antibiotics effective against various bacteria, including other Rhodococcus species ( Kitagawa and Tamura, 2008 ) a trait shared by several Rhodococcus species ( Kurosawa et al., 2008 ; Ward et al., 2018 ). These interactions suggest complexities beyond mere competition for VOCs as a bacterial growth carbon source. However, the limited resolution of 16S rDNA sequencing and the challenges of qPCR analysis in distinguishing between bacteria 7, B, and E in various consortia make it difficult to precisely identify the species involved. Thus, a metabolic profile analysis is recommended for more detailed insights. Interestingly, consortia with high total VOC degradation had very small bacteria populations and reduced bacteria species diversity, regardless of the number of bacteria strains initially inoculated. This phenomenon was particularly observed with both degrader B and non-degrader 3 ( C. metallidurans ), two competitive strains which support the growth of algae by controlling bacteria populations to small size, which, in turn, enlarged the algae population by making more VOC-originated carbon source photosynthetically available, which reinforces the ‘ecosystem’ model defined in session 3.2. Such consortia were found to have relatively high CI values and FS values closer to 1, which further evidence the predictive potential of the two parameters for algae growth and stability. In summary, consortia exhibiting high VOC degradation performance, particularly those involving C. metallidurans and R. erythropolis , show great promise in converting VOCs into biomass of Coelastrella terrestris . This algae species is beneficial due to its lipid-rich composition ( Hodač et al., 2016 ) thermal resistance ( Hu et al., 2013 ), and carotenoid production capabilities ( Rauytanapanit et al., 2019 ). These characteristics align well with large-scale algae cultivation efforts, particularly those selecting strains for high lipid content for biodiesel production. These consortia would also be a microbial candidate for industrial-scale carbon capture in facilities like power plants, and not only serve as a biological emission control process ( Iglina et al., 2022 ) but also offer the added benefit of VOC remediation."
} | 8,812 |
37903266 | PMC7615278 | pmc | 4,510 | {
"abstract": "Significance Bacillus subtilis , a common soil bacterium, forms water-repellent multicellular biofilm communities. A key component of this hydrophobicity is the presence of a self-assembled surface layer of protein known as BslA. Previous work has shown that BslA forms a highly ordered elastic layer at an interface, but the molecular interactions that drive self-assembly and film formation were not understood. In this work, we identify the key protein–protein interactions that facilitate film assembly using both experiment and simulation. The insights gained from this work will inform engineering and design principles to control the biophysical properties of protein surface layers which has the potential to impact technologies reliant on interfacial stability such as emulsions, foams, and surface coatings.",
"discussion": "Discussion During biofilm formation, microbes produce and secrete a diverse mixture of molecules that create a protective and structured matrix for the community. The matrix molecules also contribute to robustness and the emergent properties associated with biofilm formation ( 3 ). B. subtilis is a well-studied model for biofilm formation with the main components of the matrix established in the literature ( 2 ). One such component, the secreted protein BslA, has been studied in detail ( 9 – 11 , 13 , 14 , 24 ). The ability of BslA to form protein films and stabilize emulsions is interesting from a translational perspective, and moreover, this trait is necessary for its role in B. subtilis biofilm formation, structure, and protection ( 9 , 12 , 25 ). Despite the interest in BslA film properties, little was known about how the films assemble and the molecular interactions underpinning these elastic assemblies. Herein, we used a combination of structural, biophysical, and microbiological approaches to determine specific protein interactions and their impact on BslA function. The crystal structures of BslA, and its paralogue YweA, revealed two unique protein–protein interfaces that allow BslA to form two-fold symmetric dimers with the same hydrophobic cap orientation. Based on these dimerization interfaces, we built a hypothetical molecular model of the BslA interfacial lattice. Our model is supported by multiple independent lines of experimental evidence: First, the model quantitatively fits into the average lattice unit cell determined from previously published TEM images of BslA films ( 11 ); MD simulations confirm the stability of these BslA dimers at interfaces, including the one modeled from the YweA crystal structure; mutations at the key interfaces weaken the BslA elastic films in vitro, and, for one interface, also alter the morphology of the biofilm; finally, YweA mutations made to resemble BslA at one key interface contribute to rescuing the biofilm morphology in Δ bslA strains. Confirmation of the atomistic details of our molecular model would require additional structural characterization using, e.g., cryo-EM, however, collectively our data show that the identified molecular interactions form the basis of BslA self-assembly both in vitro and within the actual biofilm. We note that the lattice model suggests an arrangement of BslA monomers with four interaction interfaces (each monomer binds four neighbors) ( Fig. 4 ), but from the available crystal structures, we could only characterize two of them in detail. Future work could be done to explore the other two interfaces and their impact on film strength, stability, and dynamics. The reason why these two interfaces do not appear in the available crystal structures could be due to their relative weakness compared to the other interfaces, although it could simply be due to unfavorable crystal packing or the crystallization conditions which may weaken or strengthen ionic/nonionic interactions. The identification of dimer1 interface from the BslA crystal structure, and its higher free energy of binding compared to BslA dimer2 from the MD simulations, supports that this interface could be the strongest of the four BslA protein–protein interactions. The MD simulations of the orientation of monomers and dimers at the air/buffer interface revealed a good correlation. The dimers were more constrained compared to monomers, which makes sense as there is a greater hydrophobic surface created by the two exposed cap regions. The orientation of dimer1 and dimer2 were overlapping, supporting that these two dimers could coexist within the same lattice structure. Importantly, most of the monomers inhabited orientations that aligned with the dimer populations, suggesting that monomers likely associate with the hydrophobic phase and then laterally assemble into a lattice ( Fig. 6 ). Finally, this shows that hydrophobic-to-hydrophilic cap mutations that weaken protein adsorption ( 11 ) can additionally interfere with film formation by altering the orientation of individual BslA monomers at the interface so that it is not consistent with the orientation of the monomers within the 2D lattice ( 12 ). Fig. 6. Model of BslA film formation. BslA exists in solution in a soluble “cap in” configuration [for simplicity, BslA is only shown in its monomeric form but in solution can also exist as dimers and tetramers mediated via disulfide bonds through their CxC motifs, unrelated to the interfacial lateral interactions ( 13 )]. BslA absorbs onto an interface and undergoes a limited structural rearrangement into the “cap out” form ( 11 ), which exposes the cap hydrophobic residues and reorientates the protein ( 12 ), facilitating lateral self-assembly. The final cartoon illustrates the D1 and D2 lateral interactions that our experimental in vitro and in vivo data support as the molecular basis for interactions between monomers that hold the film together. Our results support that the dimer1 and dimer2 interfaces are necessary for elastic film formation in vitro ( Fig. 4 E ) but the in vivo effects were less evident ( Fig. 4 F ) and amino acid substitutions in the dimer1 interface did not impact colony biofilm architecture. In the MD simulations, the binding is weakened, but both BslA D1– and BslA D2– dimers are still stable for about 10 ns before dissociation occurs. These data suggest that the film could in principle still form if stabilized by the remaining unaffected lateral interactions. Based on this evidence, the D1- and D2- mutants could form weak but partially ordered 2d films. This is consistent with previous work wherein BslA mutations that weaken the elastic film could still form regular 2D films according to TEM images, although the size of the regular 2D domains was significantly decreased compared to WT BslA ( 11 ). The low in vivo impact of the dimer mutations may be due to the biofilm BslA films being stabilized by other factors such as interactions with other matrix components, making the effects of the dimer interface mutations muted when in the context of the whole biofilm. For instance, it has been postulated that the matrix protein TasA may interact with BslA ( 26 ). Whether these interactions are direct or specific has not been explored. To this end, our model of the BslA lattice may aid in future work as interaction in a mature biofilm matrix could occur between the assembled BslA film and its partners. The limited in vivo effects led us to seek further validation of our findings using YweA. It was previously shown that YweA could form films in vitro but that the overexpression of the gene could not complement the absence of BslA in vivo. It had been hypothesized that this was due to the differences in film stability between the paralogues as well as the differences in the C-termini. Herein, we show that YweA can rescue the Δ bslA strain when engineered to have the C-terminal region of BslA complemented by enhanced lateral interactions. These results further support that the film strength is important for BslA function in the biofilm. The increase in film strength seen with the YweA D1+ variant is congruent with the conserved packing structure in the lattice between paralogues. The question remains of what the role is of secreted YweA? One study linked both bslA and yweA to sporulation repression in planktonic stationary conditions ( 27 ). This was attributed to the Spo0A pathway where YweA was shown to directly inhibit KinA autophosphorylation in vitro. Since KinA autophosphorylation occurs in the cytoplasm, it is possible that secreted YweA may play an additional, alternative role. In our current study, Δ bslA has the opposite effect with a huge decrease in sporulation rather than derepression. These differences may be down to growth conditions or strain variation between PY79 ( 27 ) and NCIB 3610 which was used in this study. To our knowledge, there is no literature on the role of secreted YweA besides the slight change to colony morphology seen between Δ bslA and the Δ bslA Δ yweA strain, which indicates that YweA has a small additive effect to BslA in the biofilm matrix ( 14 ). The difference in colony biofilm architecture seen between the YweA D1+ vs. YweA CxC and YweA D1+ CxC shows that the C-terminus of BslA has a unique role in BslA function outside of monolayer formation. This tail is essential for colony hydrophobicity but not film strength which is determined by the lateral interactions identified in this study. Previous investigation showed that covalently dimerized BslA (BslA C-C BslA) is associated at the interface using only one cap region of the two ( 13 ). Based on this, lateral interactions between BslA C-C BslA dimers could lead to a BslA bilayer in which one monomer of each pair is associated with the hydrophobic phase. The covalently bound monomers that remain in the hydrophilic phase could laterally interact since they are brought into proximity by their dimer partner. Whether this occurs, and leads to colony hydrophobicity, would need further investigation and is outside the limits of this study."
} | 2,495 |
37196337 | PMC10278179 | pmc | 4,513 | {
"abstract": "The inner physicochemical\nheterogeneity of bacterial cells generates\nthree-dimensional (3D)-dependent variations of resources for effective\nexpression of given chromosomally located genes. This fact has been\nexploited for adjusting the most favorable parameters for implanting\na complex device for optogenetic control of biofilm formation in the\nsoil bacterium Pseudomonas putida .\nTo this end, a DNA segment encoding a superactive variant of the Caulobacter crescendus diguanylate cyclase PleD expressed\nunder the control of the cyanobacterial light-responsive CcaSR system\nwas placed in a mini-Tn5 transposon vector and randomly inserted through\nthe chromosome of wild-type and biofilm-deficient variants of P. putida lacking the wsp gene cluster.\nThis operation delivered a collection of clones covering a whole range\nof biofilm-building capacities and dynamic ranges in response to green\nlight. Since the phenotypic output of the device depends on a large\nnumber of parameters (multiple promoters, RNA stability, translational\nefficacy, metabolic precursors, protein folding, etc.), we argue that\nrandom chromosomal insertions enable sampling the intracellular milieu\nfor an optimal set of resources that deliver a preset phenotypic specification.\nResults thus support the notion that the context dependency can be\nexploited as a tool for multiobjective optimization, rather than a\nfoe to be suppressed in Synthetic Biology constructs.",
"discussion": "Discussion That different locations of the bacterial chromosome\ngive rise\nto very different levels of expression of heterologous genes has been\nobserved for a long time and often attributed to differential supercoiling 48 − 50 or proximity to the origin of replication. 17 − 19 However, the\n3D heterogeneity of protein distribution 51 along the genome (including RNAP 52 ) and\nthe patchy intracellular allocation of ribosomes, 53 metabolites, and pathways 54 , 55 produce a\nmuch more variable molecular environment for every chromosome site\nthan anticipated before. 56 The synthetic\noptogenetic device OPT·FILM on which this work is based was first\noptimized in vitro to exhibit a present input (green\nlight) to output (biofilm formation) transfer function in P. putida when expressed from a low-copy number plasmid\n( Figure 3 ). Yet, the\ndata above shows that the same device dramatically diversifies its\nphenotypic outcome when placed in different chromosomal locations.\nWe attribute such variations to the molecular heterogeneity of the\nbacterial cytoplasm, which results in a landscape of dissimilar resources\nfor the gene expression flow (ribosomes, RNAP, metabolic availability)\nat given locations of the 3D space. Such physicochemical unevenness\nof the cell inside concurs with proximal setting-dependent effects\n(e.g., local supercoiling, readthrough transcription) at the sites\nof chromosomal insertion, which can modify vicinal intermolecular\ninteractions. Finally, insertions of the mini-Tn5 [OPT·FILM]\ntransposon can hit native functions that either decrease or enhance\nthe observable phenotype. Because of such hypervariability, random\ninsertions of functional DNA segments in the host genomes may afford\nexploration of a much denser solution space for multiobjective optimization\nof genetic devices than that achieved with combinatorial libraries\nof regulatory parts. One important aspect of the experiments\nconducted is the likely\ncell-to-cell variation in intracellular 3D granularity, given that\nclones were interrogated at a whole population level. While the phenotype\nof different insertions was maintained through three biological replicates\ngenerated on different days (suggesting that the main repeatability\nfrom the insertion at a locus comes from its physical location in\nthe genome rather than the 3D milieu of proteins, factors, and metabolites\naround it), our results did not directly prove a\nheterogeneous availability of resources. However, some data can be\nbetter comprehended under this light. For instance, some insertions\nclose to ribosomal genes did not produce regulated biofilm formation\n( Supporting Information Tables S2 and S3 ). This suggests that chromosomal locations might be rich in some\ncomponents of the gene expression flow, such as RNAP and ribosomes\nbut not in the metabolic building blocks required for biofilm formation.\nThis result challenges the conventional wisdom of targeting heterologous\nexpression to such chromosomal sites, which are high in transcriptional\nand translational resources. Full display of a complex engineered\nphenotype, such as biofilm formation in response to light, is not\nsolely about producing the proteins involved but also requires tuning\ntheir biochemical activities to the physiological background, as well\nas providing metabolic precursors for synthesizing, for example, chromophores\nfor the correct function of the light-responsive system and extracellular\npolymeric substances for surface attachment. Under this light,\nthe wealth and variety of molecular contexts\navailable in each 3D chromosomal spot can be leveraged to become a\nphenomenal asset for adjusting the boundaries and parameters of an\nengineered function in a fashion that possibly resembles a natural\nmechanism to the same outcome. As a matter of fact, the physical location\nof genes in the bacterial chromosome seems not to be casual 57 − 59 but instead likely to reflect one more molecular stratagem for cracking\notherwise intractable adaptation challenges. In sum, the data above\nadds to the growing evidence indicates that genetic context effects\ncan override cis regulatory elements, 60 pinpointing such context variability as a principal\nmechanism to evolutionarily optimize input–output functions in vivo ."
} | 1,431 |
31372223 | null | s2 | 4,514 | {
"abstract": "Catechol-bearing polymers form hydrogel networks through cooperative oxidative crosslinking and coordination chemistry. Here we describe the kinetics of cation-dependent electrochemical-mediated gelation of precursor solutions composed of catechol functionalized four-arm poly(ethylene glycol) combined with select metal cations. The gelation kinetics, mechanical properties, crosslink composition, and self-healing capacity is a strong function of the valency and redox potential of metal ions in the precursor solution. Catechol-bearing hydrogels exhibit highly compliant mechanical properties with storage moduli ranging from "
} | 157 |
35145889 | PMC8816665 | pmc | 4,516 | {
"abstract": "Highlights • All the Bradyrhizobium species were effective; but, B. shewense was the most effective. • Inoculation of Erythrina brucei with all inoculants increased shoot length and dry weight. • E. brucei inoculated with B. shewense +A. soli increased the shoot N by 260%. • E. brucei inoculated with B. shewense + Glomus sp.1 increased shoot P by 1200%.",
"conclusion": "5 Conclusion In this experiment root nodule bacteria A. soli (AU4) exhibited multiple phyto-beneficial properties and the three Bradyrhizobium species were effective nitrogen fixers, but B. shewense (AU27) was the best performer. Multiple inoculation of E. brucei with microbial inoculants comprised of B. shewense (AU27) + Glomus sp.1 (AMF1) + Acaulospora sp.1 (AMF2) + A. soli (AU4) highly increased plant growth with reference to the shoot length and dry weight. Similarly, total shoot nitrogen content was highly improved as result of inoculation with B. shewense (AU27) and A. soli (AU4) and the plant biomass was enriched with fixed nitrogen as result of dual and/ormultiple inoculations. Likewise, shoot phosphorus content was highly improved due to dual inoculations comprised of B. shewense (AU27) and AMF. In general, shoot P content was improved as a result of dual and/or multiple inoculations. Therefore, inoculation of the host plant with phyto-beneficial microbial inputs can enhance its growth and improve the shoot nitrogen and phosphorous and the enriched plant biomass can be applied as low cost agricultural inputs by small holder farmers in order to achieve sustainable and eco-friendly agriculture.",
"introduction": "1 Introduction Nitrogen (N), phosphorous (P) and Potassium (K) are the three most important nutrients that determine soil fertility and limit plant growth. However, it is established that plant growth and health is not only determined by the availability of these nutrients, but also by the presence of consortium of microorganisms in the vicinity of the root surface known as the rhizosphere. About 2-5% of the rhizosphere competitive microbes exert phyto-beneficial effects. The use of plant growth-promoting microbes (PGPM) is a potentially advantageous technique for improving crop productivity, food quality and security in more sustainable and eco-friendly agricultural systems [ 2 , 14 , 24 ]. These microorganisms are engaged in symbiotic relationships with a multitude of above- and belowground plant parts that constitute phyto-beneficial microbes, including rhizobia, mycorrhizal fungi, and endophytes [34] . Rhizosphere associated bacteria referred to as rhizobacteria and the Mycorrhizha majorly contribute to plant growth promoting functions. Somers et al. [38] have classified these rhizosphere associated microorganisms based on their roles as (i) biofertilizers (increasing the availability of nutrients to plant), (ii) phytostimulators (plant growth promotion, generally through phytohormones production), (iii) rhizoremediators (degrading organic pollutants) and (iv) biopesticides (controlling diseases, mainly by the production of antibiotics, antifungal metabolites, synthesis of fungal cell wall and its component degrading enzymes). Plant growth promoting rhizobacteria have the potential to produce different types of metabolites that help the host plants to improve minerals such as phosphorus and iron, promote growth and protect them from phytopathogens, and enhance their tolerance to abiotic stresses [ 3 , 32 ]. The symbiotic association between leguminous plants and rhizobium that leads to biological nitrogen fixation (BNF) [44] , is very important in nitrogen nutrition in any ecosystem. Arbuscular mycorrhizal fungi (AMF) are present in nearly all soils, forming a symbiotic association with roots of approximately more than 80% of terrestrial plant species [36] . Arbuscular mycorrhizal fungi mediates up to 70% of total P uptake by plants and other immobile nutrients by producing extensive hyphae that grow out from roots and effectively increase exploratory areas of roots and help in harnessing fixed/immobile P [10] . The symbiotic association involving rhizobium, AMF and the legume plant is referred to as tripartite symbiosis, and is a mutually beneficial interactions and plays a pivotal role in natural ecosystems by influencing plant productivity, nutrition, and community structure [3] . Dual inoculation of leguminous plants with rhizobium and AMF is being recommended to improve legume plant growth, nodulation and to increase shoot N and P content [ 3 , 45 ]. However, the effectiveness of the rhizobium-AMF-plant interactions varies with host plant species, Rhizobium strains, arbuscular mycorrhizal fungal species and soil conditions [ 30 , 39 ]. Erythrina brucei is a woody legume tree characterized by very important agro forestry attributes such as rapid establishment, tolerance to light, possession of spreading canopy, high rate of litter production, rapid litter decomposition and very soft woody nature [ 16 , 28 ]. It is extensively planted in the farmlands in southern Ethiopia in the agro forestry systems. Its biomass has been used as green manure for crop growth enhancement. The local farmers in southern and southwestern Ethiopia commonly plant E. brucei inside their farmlands, while cultivating crops such as barley, wheat and maize. The farmers prune the branches and leaves of E. Brucei and mulch it under soil before sowing the grains. They also collect the branches and leaves of E. brucei from forests, shades, home gardens and land boundary fences and transport to farmlands as low-cost agricultural inputs. The legume is well integrated in Sidama agro-forestry system [12] in southern Ethiopia. The plant is reported to improve soil fertility as it fixes atmospheric nitrogen in association with root nodulating bacteria [ 1 , 43 ]. The symbiotic association of this host plant with arbuscular mycorrhizal fungi has also been reported [ 12 , 25 ]. Megersa and Assefa [25] have also shown that the dual inoculation of root nodulating bacteria and AMF (Gigaspora and/or Glomus) improved its biomass production threefold compared to the control plants under greenhouse conditions. Despite the agro-forestry importance of the tree species E. brucei, there is still a dearth of information regarding enrichment of E. brucei biomass with N and P through single inoculation and consortial inoculations with selected and effective endosymbionts to improve its growth, nodulation and N and P status in the biomass. This work was a part of a long term plan that targets on understanding the rhizosphere microbes of E .brucei for use in its growth promotion. The rhizosphere associated indigeneous (Rhizobia, AMF and rhizobacteria) can be used as microbial inputs for the enhancement of the symbiotic association among the host legume, Rhizobium and AMF which could lead to improved nitrogen fixation, and utilization of phosphorus and other immobile plant nutrients. The use of effective and compatible microbial inoculants could be sustainable and ecofriendly biotechnological inputs to improve the traditional agro-forestry practices in the southern and southwestern Ethiopia. Therefore, this study was conducted to investigate the effects of combined inoculation of B. shewense (AU27) and Glomus sp.1 (AMF1) and/or Acaulospora sp.1 (AMF2) and/or Acinetobacter soli (AU4) on growth, nodulation and shoot nitrogen and phosphorus contents of E. brucei under greenhouse condition.",
"discussion": "4 Discussion Rhizosphere associated plant growth promoting bacteria and AMF are reported to enhance growth, development and improve biomass nitrogen and phosphorus contents of woody legume E. brucei. This host woody legume is used in agro-forestry as a low cost agricultural input for improving soil fertility by smallholder farmers in southern Ethiopia. It is being planted in the home gardens, farmlands, land boundaries and forests. Smallholder farmers use the biomass of E. brucei in farmland for mulching, as green manure and cover plant material and as a substitute to chemical fertilizers to improve crop productivity. Several symbiotic and non-symbiotic bacteria have been previously reported from the root nodules of E. brucei (Amsalu et al., 2012, 2013; [5] a). These authors also reported various phyto-beneficial properties like IAA production, inorganic phosphate solubilization, synthesis of volatile secondary metabolites and hydrolytic enzymes exhibited by E. brucei root nodule bacteria. Similarly, Dobo et al. [12] and Berza et al. [5] have reported other phyto-beneficial microorganisms, arbuscular mycorrhizal fungi (AMF) from the rhizosphere of the same host plant implying that the rhizosphere and root nodules of this particular plant are rich in phyto-beneficial microbes which can be inoculated to E. brucei to enhance its growth and improve its biomass nitrogen and phosphorous and to be used as low cost agricultural input by smallholder farmers in Ethiopia. As revealed by the shoot dry matter accumulation, B. shewense (AU27), and B. cytisi (HU3) are highly effective since they accumulated shoot dry matter more than 80% of the N-fertilized plants [27] ). However, B. cajani (HO2) enabled the host plant to accumulate 73% of the shoot dry matter of the positive control plants, and is rated as effective. As indicated in the results, a significant differences in plant height and shoot biomass between the inoculated treatments and the control under greenhouse conditions. This can be explained by an increase in the supply of nitrogen by symbiotic association with the inoculated nitrogen fixing bacterial strains. This increment in the supply of nitrogen was reflected in better plant growth. Similar results were reported by [ 18 ] who have showed that rhizobial inoculation improves nitrogen fixation, photosynthetic capacity and total biomass of cowpea plants. Other researchers have also emphasized that symbiotic nitrogen fixation is directly related to shoot dry matter produced as a result of rhizobium inoculation [ 27 , 37 ]. It is interesting to note that inoculation with B. shewense (AU27) accumulated shoot dry matter more than the N-fertilized control plants (129%) ( Table 1 ) indicating that the inoculant not only fixed atmospheric nitrogen through the symbiotic interaction with the host plant, but also enhanced the host plant growth with a possible production of different types of phyto-beneficial traits by the Rhizobium species. The accumulation of 3.16±0.1 to 5.45±0.1 g shoot dry weight per plant by the Bradyrhizobia species in this study was much higher than those reported by Megersa and Assefa [25] , which was 2.49 g shoot dry weight per plant. Apart from the inherent effectiveness of the inoculants, the big difference could be attributed to the long growing period (90 days) of the plant growth in this experiment compared to the 60 days by the other study. Although all the Bradyrhizobium species were effective and highly effective, B. shewense (AU27) and B. cytisi (HU3) could be used as potential candidates for field trial to enhance the growth, development and biomass nitrogen content of the host plant. In the sand culture experiment, we recorded nodule number between 6±1.5 and 30±1 per plant. However, between 25±1 and 50±1.1 nodules per plant were recorded during isolation of root nodule bacteria using plant infection method from E. brucei rhizosphere soil samples. On the other hand, in the multiple inoculation experiment, we recorded between 7±0.1 and 18±0.1 nodules per plant. These small numbers of root nodules could be attributed to the absence of competing ineffective native E. brucei nodulating bacteria as revealed by the lack of nodules in negative controls. Megersa and Assefa [25] have reported the presence of root nodules between 125 and 143 per plant in greenhouse experiment that involved E. brucei single, dual and consortia inoculations in soil. The number of root nodules produced by a legume plant may be determined by the variety of the host plant and the type of the Bradyrhizobium strain [30] . Hence, the number of nodules produced could depend on the growth stage of the host plant , the demand of fixed nitrogen by the host plant and its rhizosphere micro flora and symbiotic nitrogen fixing efficiency of nodulating rhizobial species. Interestingly, the inoculated plants in this study exhibited fast growth as revealed by shoot length improvement records across the growth periods. The highest shoot length, 39% and 23% increment compared to the un-inoculated control was recorded by treatments that involved the consortia of all inoculants (T7) and consortia of B. shewense (AU27) + A. soli (AU4) + AMF1 (T5), respectively during the first 30 DAP. The same treatments increased plant height by 19.5% and 5.7%, respectively compared to the dual inoculations comprised of B. shewense (AU27) + AMF during the same growth period. Similarly, inoculation treatments T7 and T5 increased shoot length by 42% and 27.6%, respectively compared to the un-inoculated control plants during the first 60 DAP. Likewise, consortia inoculation treatments, T7 & T5 increased plant height by 20% and 8%, respectively compared to dual inoculation treatments that comprised of B. shewense (AU27) and AMF during 60 days’ growth period. The role of AMF inoculation is well expressed among dual inoculation treatments (T2, T3 & T4). AMF + phyto-beneficial bacteria dual inoculations exhibited better plant performance with regard to shoot length compared to phyto-beneficial bacteria +phyto-beneficial bacteria co-inoculation (Fig.1) across the growth periods. Consortia inoculation with B. shewense (AU27) + A. soli (AU4) + Glomus sp.1 (AMF1) (T5) increased shoot length by 130% and the other treatment that comprised of all inoculants (T7) exhibited the highest increase in shoot length (140%) compared to un-inoculated control at 90 DAP. These treatments (T7&T5) increased shoot length by 13% and 9%, respectively compared to dual inoculation treatments that comprised of B. shewense (AU27) and AMF during the same growth period. Dual inoculation of E. brucei with B. shewense (AU27) + Glomus sp.1(AMF1) (T2) and B. shewense (AU27) + Acaulospora sp.1 (AMF2) (T3) increased plant shoot length by 113% and 111%, respectively compared to un-inoculated control 90 DAP. However, dual inoculation with B. shewense (AU27) + A. soli (AU4) (T4) increased the shoot length by 72% compared to un-inoculated control 90 DAP. This indicates that AMF species ( Glomus sp.1 or Acaulospora sp.1) co-inoculations enhanced plant shoot length by 39% compared to A. soli (AU4)co-inoculation. Inoculated plants showed better development (shoot length and shoot dry weight) compared to un-inoculated plants. These results can be explained by the improved mineral nutrition availability to the inoculated plants,which was reflected in improved vertical growth and biomass. The variations between AMF and/or A. soli (AU4) inoculations could be attributed to the differences in the rhizosphere function between AMF and phyto-beneficial bacteria. Arbuscular mycorrhizal fungi are well known for their mobilization of available P, macro and/or micronutrients and water beyond the root depletion zone and translocation to the associated host plants. AMF also play vital role in the rhizosphere by solubilizing organic phosphate by producing phosphatase enzyme [ 3 , 31 ], while phyto-beneficial bacteria have crucial role in inorganic phosphate solubilization. Inoculations with all inoculants highly enhanced plant growth with reference to plant height which might be attributed to the synergistic interactions among B. shewense (AU27), AMF species and the A. soli (AU4). Vafadar et al. [41] have reported that legumes benefit very much from dual symbiosis (rhizobia and AMF) to improve their growth, biomass and nutrient assimilation. In this experiment, the increased shoot length could be attributed to AMF species (Glomus sp.1 and Acaulospora sp.1)and A. soli (AU4) (which is producer of multiple plant growth promoting traits such IAA, phosphate solubilizing, production of volatile secondary metabolites and synthesis of hydrolytic enzymes ( Table 1 ) which improved plant growth parameters and nutrient uptake. In general, inoculation of this woody legume tree with microbial consortia enhanced growth which is characterized by shoot length increment and production of high amount of litter compared to un-inoculated control plants. The consortia inoculations which comprised of all inoculants (T7) and triple inoculation treatment which contained all inoculants except AMF2 (T5) increased the shoot dry weight by 268% and 230%, respectively compared to the un-inoculated control plants. Similarly, co-inoculations involved AU27 + AMF1 (T2) and AU27 + AMF2 (T3) increased shoot dry weight by 143% and 127%, respectively compared to the un-inoculated control. In addition, co-inoculation of AU27 + AU4 (T4) increased shoot dry weight by 66% compared to the un-inoculated control. Co-inoculations that comprised of AMF (T2 & T3) increased plant biomass accumulation at least 61% compared to inoculations that involved AU27 + AU4 (T4). Moreover, consortia inoculation treatments, T7, T5 & T6 increased shoot dry weight by 56%, 40% and 20%, respectively compared to the dual inoculation treatments that involved B. shewense and AMF. Previously, Megersa and Assefa [25] have reported shoot dry weight increment between17% and 45.3% using the same plant as result of single, dual and consortia inoculations compared to un-inoculated control in the greenhouse experiment. Phyto-beneficial interactions are observed when a microbial consortium is inoculated to enhance plant growth which could be expressed as an additive or synergistic interaction, in part, are due to the fact that multiple microbial species can perform a variety of tasks in an ecosystem like in the rhizosphere [34] . Therefore, the phyto-beneficial mechanisms of plant growth stimulation like enhanced nutrient availability, phytohormone modulation, biocontrol, biotic and abiotic stress tolerance are exerted by different microbial players within the rhizosphere, such as phyto-beneficial bacteria and arbuscular mycorrhizal fungi [ 17 , 35 ]. The disparities in plant biomass accumulation as indicated by shoot dry weight in the present study and Megersa and Assefa [25] could be attributed to several different factors. The soils used in this greenhouse experiment were characterized by a higher organic matter content (5.48%) compared to soil used by the other authors, (with 1.53%) organic content. Higher soil organic matter content has a multifaceted strategy to improve soil quality such as increment in bioavailability of soil nutrients like P [11] . The soil organic matter also provides substrates and energy and therefore, increases bioavailability of macro and micro nutrients, allowing the maintenance of soil quality and ecosystem functionality [ 15 , 22 ]. These could enhance symbiotic nitrogen fixation and photosynthetic rates that could be expressed in terms of biomass accumulation. In addition, relatively higher AMF root colonization in this study might have contributed to the enhanced nitrogen fixation compared to Megersa and Assefa [25] . As demonstrated by results, AMF inoculated treatments performed higher compared to A. soli (AU4) inoculated treatments. This difference could be attributed to the fact that AMF infection often results in increased allocation of C to the root system, implying increased root biomass, respiration and mycelial biomass. Thus, the root and mycelia mass could explore larger soil volume, beyond the depletion zone for nutrients, resulting in a higher nutrient uptake rates [7] . The mean root length colonization as the result of inoculation with microbial inputs varied between 27% and 55%. Similarly, Megersa and Assefa [25] have recorded between 25% and 28% AMF root length colonization in inoculation studies carried out in greenhouse using the same host plant. The higher AMF root length colonization in the present study could be attributed partly to the soil physicochemical properties such as available P and soil organic carbon. The soils used in this experiment were characterized by an available P concentration of 27.05 mg kg −1 which falls in medium soil available P range [ 19 , 23 ]. However, Megersa and Assefa [25] have used soils with 37mg kg −1 of available P which falls in higher soil available P range [ 19 , 23 ]. It is well established that higher soil P content inhibits AMF root length colonization. In addition, the higher soil organic carbon content (3.18 %) in this experiment might have also enhanced AMF root infection compared to soil organic carbon content (0.88 %) in the other authors. The root nodule bacterial inocula in the presentexperiment might have assisted the germination of AMF spores, thus leading to higher infection percentage [44] . In this context, the plant growth promoting bacteria might have acted as mycorrhiza helper bacteria (MHB). The MHB could promote mycelial growth, improve host recognition and change root system architecture and improve receptivity of roots [40] .The root nodule bacteria, A.soli (AU4) applied in the present study was IAA producer ( Table 1 ) . Hence, this strain might have enhanced E. brucei root growth and development, which in turn stimulated the AMF root length colonization [40] . The E. brucei root growth and development as indicated by root length was also varied among different inoculation treatments. The consortia inoculations in treatments (T5) and (T7) increased the root length by 261.9% and 250.8% respectively compared to un-inoculated control plants. The increased root length in the treatments (T5 and T7) which received consortial inoculation could be associated to the phytohormone production (IAA) by A. soli (AU4) that probably enhanced root extension, growth and development. The stimulation of root hairs growth and lateral roots elongation by IAA might provide more active sites and access for dual symbiotic association with rhizobia and AMF to improve root architecture, length and dry weight [13] . The inoculation of different microbial inputs enhanced E. brucei biomass, nitrogen and phosphorus contents as exhibited remarkable improvements compared to un-inoculated control. The dual and triple inoculations with B. shewense (AU27) + A. soli (AU4) (T4) and B. shewense (AU27) + Acaulospora sp.1 (AMF2) + A. soli (AU4) (T6) increased the shoot total nitrogen content by 260%, and 214.4%, respectively compared to un-inoculated control. The higher shoot total nitrogen content in the dual inoculated treatments might be due to a higher nitrogenase activity as a result of better P nutrition corroborating to the enhanced nitrogen fixation reported by Bai et al. [3] . The higher shoot total nitrogen and phosphorus contents could allow for increased shoot growth and photosynthetic rates and lead to increased total N and P levels that resulted from the improved biological nitrogen fixation [41] . Likewise, shoot P content significantly (p<0.05) differed among inoculated treatments. The inoculation treatments consisted of B. shewense (AU27) + Glomus sp.1 (AMF1) (T2) and B. shewense (AU27) + Acaulospora sp.1 (AMF2) (T3) increased the shoot P content by 1200% and 1141.6%, respectively compared to un-inoculated control. Though, A. soli (AU4) was previously confirmed as good inorganic phosphate solubilizer ( Table 1 ), inoculations of B. shewense (AU27) with Glomus sp.1 (AMF1) or Acaulospora sp.1 (AMF2) enhanced better shoot P accumulation compared to inoculation with B. shewense (AU27) and A. soli (AU4) ( Table 4 ). The results clearly indicated that inoculation of AMF contributed to higher shoot P accumulation compared to bacterial inoculations. This difference could be attributed to increased available P exploration beyond the depletion zone and translocation to the plant roots by AMF compared to bacterial species which play role in inorganic P solubilization [ 3 , 31 ]."
} | 6,038 |
38443414 | PMC10914721 | pmc | 4,517 | {
"abstract": "Coral reef ecosystems supported by environmentally sensitive reef-building corals face serious threats from human activities. Our understanding of these reef threats is hampered by the lack of sufficiently sensitive coral environmental impact assessment systems. In this study, we established a platform for metabolomic analysis at the single-coral-polyp level using state-of-the-art mass spectrometry (probe electrospray ionization/tandem mass spectrometry; PESI/MS/MS) capable of fine-scale analysis. We analyzed the impact of the organic UV filter, benzophenone (BP), which has a negative impact on corals. We also analyzed ammonium and nitrate samples, which affect the environmental sensitivity of coral-zooxanthella (Symbiodiniaceae) holobionts, to provide new insights into coral biology with a focus on metabolites. The method established in this study breaks new ground by combining PESI/MS/MS with a technique for coral polyps that can control the presence or absence of zooxanthellae in corals, enabling functions of zooxanthellae to be assessed on a polyp-by-polyp basis for the first time. This system will clarify biological mechanisms of corals and will become an important model system for environmental impact assessment using marine organisms.",
"introduction": "Introduction Currently, there is great concern about adverse effects on ecosystems due to discharge of anthropogenic chemicals into the environment and disruption of nutrient cycles. In terms of the planetary boundary framework demarcating a global safe operating space for humanity, six of the nine boundaries have already been transgressed, including not only climate change and ocean acidification, but also discharge of anthropogenic chemicals and disruption of nutrient cycles 1 , and there is a growing awareness of the necessity to regulate chemicals in the ocean, including microplastics. Against this backdrop, there is a growing need for assessment systems to determine how chemicals adversely affect marine organisms. Tropical coral reef ecosystems provide a range of important ecosystem services to human society, including coastal protection, fisheries, and tourism 2 , 3 , but a global decline of reef-building corals is occurring because of coral sensitivity to environmental change at global and regional scales 4 – 7 . Corals may also be adversely affected by anthropogenic chemicals 8 – 10 . Considering that corals are keystone organisms of marine ecosystem, establishment of a reliable environmental impact assessment system using corals is urgently needed. In recent years, sunscreen products and their ingredients have also been reported to adversely affect corals 10 – 18 . The organic UV filter, benzophenone (BP or oxybenzone), harms corals in laboratory studies 11 , 12 , 18 , and bans have been issued in some areas 19 – 21 . However, previous studies assessing effects of sunscreens on corals continue to be controversial due to lack of analytical verification of exposure concentrations, poor controls, and lack of environmental relevance 12 . Moreover, establishment of standard toxicological assessment protocols is still wanting. Previous studies assessing chemical effects on corals have used mainly mature coral fragments, but rearing experiments using coral fragments are difficult without a flowing seawater system, and it is also difficult to ensure sufficient space for rearing an adequate number of replicates. Coral primary polyps obtained during simultaneous spawning are smaller than coral fragments (Fig. 1 ), so they can be reared in petri dishes. Necessary numbers of replicates can easily be achieved and chemical exposures can be controlled 8 , 22 . Rearing of healthy coral primary polyps is an essential step in coral larval recruitment, and assessment of impacts during the primary polyp stage is extremely important from the perspective of maintaining coral populations. This life history stage is considered more sensitive to environmental changes than mature coral colonies (reviewed in 23 ), making them excellent for sensitive toxicological assessments. Another advantage is that coral primary polyps produced from Acropora species can be artificially infected with zooxanthellae (Symbiodiniaceae) to carry out response assessments in the presence or absence of zooxanthellae, as larvae of Acropora species are zooxanthella-free 24 . As corals maintain a symbiotic life history with zooxanthellae, environmental responses also need to be assessed in coral holobionts. The state of this coral-zooxanthellae symbiosis varies in sensitivity to high water temperature stress depending on surrounding nutrient concentrations (ammonium and nitrate as nitrogen sources) 25 . However, it is unclear how metabolite levels that respond to changes in seawater nutrient concentrations are altered by zooxanthellae. Figure 1 Schematic of the experimental flow using coral polyps and probe electrospray ionization/tandem mass spectrometry (PESI/MS/MS). Holobionts can be created by artificially adding zooxanthellate. Metabolomic data can be acquired within 2.4 min by fixing the sample plate under a fine needle that acquires metabolite data. Biological parameters such as survivorship, growth rate (calcification rate) and tentacle movement are used to assess coral health in coral rearing experiments 8 , 22 , 24 , 26 , but there are limitations to assessing health based on morphological and behavioral observations. Metabolomic analysis, which is commonly used in medicine and other fields, allows numerous metabolites to be quantified simultaneously, making it possible to extract metabolites of interest according to the target life phenomenon. In addition, metabolomic analysis can also be interpreted as phenotypic analysis and is therefore excellent for detecting potential effects that are not discoverable by other methods, making it suitable for environmental impact and stress assessment. Conventional metabolomic analysis requires a relatively large amount of sample, whereas probe electrospray ionization/tandem mass spectrometry (PESI/MS/MS), a newly developed technique in medicine, enables local analysis with an ultra-fine metal needle (tip diameter: 700 nm) without sample pre-treatment. It has been applied to fine-scale analysis of mouse liver 27 , mouse brain 28 , 29 and agricultural crops 30 . Considering the above background, if metabolomic analysis can be performed on a single coral primary polyp by PESI/MS/MS, it would provide a powerful new platform for simple and rapid assessment of exposure to various chemical substances. We have therefore developed a new method for metabolomic analysis of single coral primary polyps of a reef-building coral, Acropora sp., to assess the impact of chemical exposure.",
"discussion": "Results and discussion We used coral primary polyps produced from gametes obtained during coral spawning, and artificially added zooxanthellae to create primary polyps with and without zooxanthellae (Figs. 1 and S1 ), and conducted exposure experiments using BP, which has adverse effects on corals. Metabolite information was obtained for seven zooxanthella-free polyps in the control group, four BP-exposed polyps (Table S1 ), and for four holobionts in the control and six experimental groups (Table S2 ). For some samples, polyps were not punctured successfully with the needle and data could not be obtained. As for experiments involving nitrate or ammonium exposure, metabolite information was obtained for five polyps in the control, five in the nitrate-exposed and five in the ammonium-exposed zooxanthella-free polyps (Table S3 ), and for three polyps in the control, four in the nitrate-exposed and five in the ammonium-exposed polyps with zooxanthella-present polyps (Table S4 ). The variation in the number of data acquired may be due to the somewhat larger size of the holes in the plate (Fig. S1 ) compared to the polyps, which may have biased the position, and therefore requires smaller hole size in the future. Using multivariate metabolite data, PCA readily segregated zooxanthella-free polyps with and without BP exposure (Fig. 2 A), with nine metabolites showing significant differences between treatments (Fig. S2 ). On the other hand, zooxanthellate polyps exposed and unexposed to BP were not segregated (Fig. 2 B) and no metabolites showed significant differences between treatments. In a previous study, toxicity of BP to cnidarians differed in the presence and absence of zooxanthellae 11 , and this was replicated in the present experiments. A previous study 11 showed that symbiotic algae mitigate effects on host cnidarians by sequestering the phototoxic oxybenzone metabolite. In other words, symbiotic algae may prevent adverse effects of reactive oxygen species on corals caused by oxybenzone metabolites. In our metabolomic analysis, we found a reduction of some amino acids in zooxanthella-free-polyps under BP treatment (Fig. S2 ), and a previous study also found a reduction of amino acids such as glutamate under oxidative stress caused by high-water-temperature exposure 31 , suggesting suppression of metabolism due to oxidative stress. In addition, metabolomic analyses of coral species that inhabit relatively high-stress areas also show a trend toward low levels of arginine and other amino acids 32 . Reduced glutathione, an important intracellular antioxidant, is oxidized to eliminate reactive oxygen species 33 , 34 , and the decrease of reduced glutathione in BP treatment is a result of the reaction with ROS produced by phototoxicity of oxybenzone metabolites. Figure 2 Principal component analysis (PCA) of the BP treatment. Dotted and solid circles in score plots indicate 95% and 99% confidence intervals for all plots, respectively. Colored circles indicate 95% confidence intervals for each group; ( A ) primary polyps without zooxanthellae; ( B ) primary polyps with zooxanthellae. PC1, first principal component; PC2, second principal component; BP, BP treatment, C, control and S, zooxanthellae present. PCA using multivariate data on metabolites obtained in nitrate and ammonium experiments confirmed group segregation in zooxanthella-free polyps with and without ammonium exposure (Fig. 3 A). Among zooxanthella-free polyps, significant differences were observed in 13 metabolites between control and ammonium-exposure groups. Ten of these corresponded to amino acids and increased in the treatment group. In contrast, 3 of them corresponded to fatty acids and decreased in the treatment group (Fig. S3 ). On the other hand, in polyps with zooxanthellae, no group segregation was observed upon nitrate or ammonium exposure (Fig. 3 B), and the only metabolite that showed significant differences in the treatment group was asparagine (Fig. S4 ). Previous studies have reported that enrichment of nitrate and ammonium is associated with effects on the energetic and redox status of corals 35 . In corals, nitrate enrichment induces oxidative stress and ammonium enrichment promotes amino acid synthesis and protein turnover 35 . Indeed, our experimental results showed a significant increase in amino acids such as lysine and valine upon ammonium exposure (Fig. S3 ). The decrease in fatty acids may have been due to their consumption for energy, as a result of accelerated amino acid synthesis caused by ammonium. Indeed, it has been reported that all three fatty acids that were reduced in the present study are abundant in corals and may be used for cell structure and energy storage 36 , 37 . Nitrate exposure did not change metabolite profiles associated with oxidative stress, but these differences from previous studies may reflect the different coral species and coral fragments used in those studies compared to this study 38 . Therefore, we recommend that PESI/MS/MS be applied to the mature corals and comparisons with the results obtained in the present study should be made. Recent studies also highlighted limitations of analyzing metabolites in holobionts 39 , emphasizing the need for analysis in independent component partners. With regard to symbiotic systems between cnidarians and zooxanthellae, experimental systems are well established in the anemone Aiptasia 40 , 41 , which is an important future target for PESI/MS/MS. Figure 3 Principal component analysis (PCA) of nitrate and ammonium treatments. Dotted and solid circles in score plots indicate 95% and 99% confidence intervals for all plots, respectively. Colored circles indicate 95% confidence intervals for each group; ( A ) primary polyps without zooxanthellae; ( B ) primary polyps with zooxanthellae. PC1, first principal component; PC2, second principal component; apo, sym, without and with zooxanthellae, respectively; NaNO 3 , nitrate treatment; NH 4 Cl, ammonium treatment; ctrl, control. The platform established in this study, which combines a coral primary polyp experimental system with PESI/MS/MS, succeeded in comprehensively capturing metabolite changes in single coral polyps. Furthermore, by comparing responses in the presence and absence of zooxanthellae, it was possible to verify at the metabolite level whether effects of substance exposure occur in the coral itself or in the holobiont. One disadvantage of using coral primary polyps is that gametes can only be collected when a limited number of coral taxa spawn simultaneously, but Acropora sp. 1 used in this study, spawns approximately two months later than most Acropora species 42 , so we were able to observe coral mass spawning twice, allowing us to carry out multiple experiments. Although sample sizes were somewhat limited in this study, the method is scalable, as it is possible to assay other metabolites as well. The platform has the potential to be used not only for chemical exposure, but also to investigate other biological phenomena, such as responses of coral holobionts to high seawater temperatures and acidified seawater, tissue regeneration, and calcification of corals. It can potentially become an important model system to assess effects of environmental changes on marine organisms. This assessment system could be employed not only for the risk assessment of chemicals and other substances but also for evaluation of positive impacts, including enhanced growth and promotion of metabolic processes."
} | 3,576 |
34946025 | PMC8709402 | pmc | 4,519 | {
"abstract": "The genus Gemmobacter grows phototrophically, aerobically, or anaerobically, and utilizes methylated amine. Here, we present two high-quality complete genomes of the strains con4 and con5 T isolated from a culture of Anabaena . The strains possess sMMO (soluble methane monooxygenase)-oxidizing alkanes to carbon dioxide. Functional genes for methane-oxidation ( prmAC , mimBD , adh , gfa , fdh ) were identified. The genome of strain con5 T contains nirB , nirK , nirQ , norB , norC , and norG genes involved in dissimilatory nitrate reduction. The presence of nitrite reductase gene ( nirK ) and the nitric-oxide reductase gene ( norB ) indicates that it could potentially use nitrite as an electron acceptor in anoxic environments. Taxonomic investigations were also performed on two strains through polyphasic methods, proposing two isolates as a novel species of the genus Gemmobacter . The findings obtained through the whole genome analyses provide genome-based evidence of complete oxidation of methane to carbon dioxide. This study provides a genetic blueprint of Gemmobacter fulva con5 T and its biochemical characteristics, which help us to understand the evolutionary biology of the genus Gemmobacter .",
"conclusion": "4. Conclusions In this study, the strains con5 T and con4, representing methane oxidizing species from an Anabaena culture belonging to the genus Gemmobacter , were investigated using genomic and polyphasic methods. The findings obtained through the whole genome analyses provide genome-based evidence of complete oxidation of methane to carbon dioxide. This study provides a genetic blueprint of Gemmobacter fulva con5 T and its biochemical characteristics, which help us to understand the evolutionary biology of the genus Gemmobacter . Based on the phylogenetic position and the genotypic, chemotaxonomic, and physiological differences, we propose that strains con5 T and con4, Gemmobacter fulva sp. nov., should be assigned as a novel species of the genus Gemmobacter in the family Rhodobacteraceae ( Table 3 ).",
"introduction": "1. Introduction Gemmobacter is of interest because of its metabolic pathways and its habitats. Members of the Gemmobacter species are able to grow phototrophically, aerobically, and anaerobically [ 1 , 2 , 3 , 4 , 5 , 6 ]. The first species, Gemmobacter changlensis , isolated from a snow sample collected in the Indian Himalayas, was proposed as psychrotolerant and phototrophic bacteria [ 1 ]. Another phototrophically growing species, Gemmobacter aestuarii , recovered from estuarine surface water, was described as containing a complete gene cluster for photosynthesis [ 3 ]. The members of the genus Gemmobacter have been discovered in a wide range of natural environments, such as freshwater, sulphuric cave waters, estuaries, forest ponds, artificial fountains, tidal flats, activated sludge, a white stork nestling, coastal planktonic seaweed, and conserved forages, indicating that members of this genus are widely distributed in natural and artificial environments [ 2 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 , 14 , 15 ]. Comparative genomics analyses revealed that members of the genus Gemmobacter , including Gemmobacter aquatilis , Gemmobacter lutimaris , Gemmobacter sp. HYN0069, Gemmobacter caeni , and Gemmobacter sp. LW-1, utilize methylated amine. These species have related genes encoding the enzymes trimethylamine (TMA) dehydrogenase, TMA monooxygenase, and TMA demethylase in their genome, indicating metabolic potential of using the TMA oxidation pathway to convert trimethylamine to dimethylamine [ 3 , 8 ]. In a recent study, we isolated two alphaproteobacterial strains, con4 and con5 T , from an Anabaena culture that contained all genes related to the oxidization of methane to carbon dioxide. The species in the genus Gemmobacter were characterized as Gram-negative, oxidase and catalase-positive, non-spore-forming, and rod-shaped containing ubiquinone-10 as the major respiratory quinone, with G + C content in the range of 61.4–69.1 mol% [ 1 , 3 , 4 , 5 ]. Here, we report a comparative analysis of the genomes of two strains, con4 and con5 T , together with a taxonomic proposal based on their phylogenetic, genomic, physiological, and chemotaxonomic characteristics.",
"discussion": "3. Results and Discussion 3.1. Physiological Tests The two strains con5 T and con4 were Gram-negative, non-motile, aerobic, and rod-shaped ( Supplementary Figure S1 ). The colonies appeared yellow, convex, circular, and smooth, with entire edges, after being grown for two days at 30 °C on R2A agar. Cell growth was observed at temperatures ranging from 4 to 37 °C and at pH 5–10 (weak at pH 5). Oxidase and catalase activities were present. The cells were found to assimilate N -acetyl-glucosamine, d -glucose, d -mannitol, d -mannose, malate, and maltose but not adipate, l -arabinose, caprate, citrate, gluconate, and phenyl acetate (API 20NE). The cells were found to be positive for the following enzyme activities (API ZYM test strip): esterase (C4), esterase lipase (C8), α -glucosidase, leucine arylamidase, and naphthol-AS-BI-phosphohydrolase. However, the cells were found to be negative for N -acetyl- β -glucosaminidase, acid phosphatase, alkaline phosphatase, α -chymotrypsin, cystine arylamidase, α -fucosidase, α -galactosidase, β -galactosidase, β -glucosidase, β -glucuronidase, lipase (C14), α -mannosidase, trypsin, and valine arylamidase ( Table 1 ). The cells were found to be susceptible to amikacin (30 µg mL −1 ), ampicillin/sulbactam (1:1; µg mL −1 ), chloramphenicol (30 µg mL −1 ), erythromycin (30 µg mL −1 ), gentamicin (30 µg mL −1 ), kanamycin (30 µg mL −1 ), nalidixic acid (30 µg mL −1 ), rifampicin (30 µg mL −1 ), spectinomycin (25 µg mL −1 ), streptomycin (25 µg mL −1 ), teicoplanin (30 µg mL −1 ), tetracycline (30 µg mL −1 ), and vancomycin (30 µg mL −1 ) but resistant to lincomycin (15 µg mL −1 ). The major fatty acids were summed feature 8 (comprising C 18:1 ω 7 c and/or C 18:1 ω 6 c ) for both strains (data not shown). The major fatty acids in strains con5 T and con4 were consistent with the major fatty acid components in species from the genus Gemmobacter . Notably, con5 T and con4 differed from the four closest relatives in the proportions of some minor fatty acids. The major predominant respiratory ubiquinone was Q-10. The polar lipids consisted of phosphatidylethanolamine (PE), phosphatidylglycerol (PG), phosphatidylcholine (PC), two unidentified glycolipids (GL1, GL2), three phospholipids (PL1, PL2, and PL3), and three unidentified aminophospholipids (APL1, APL2, and APL3) for the type strain con5 T ; and phosphatidylethanolamine (PE), phosphatidylglycerol (PG), phosphatidylcholine (PC), two unidentified glycolipids (GL1, GL2), three unidentified phospholipids (PL1, PL2, and PL3), and two unidentified aminophospholipids (APL1, APL2) for the strain con4 ( Supplementary Figure S2 ). This profile is similar to that of closely related species G. changlensis and G . aquaticus , with major components of PE, PG, and PC, and strains con5 T and con4 contain two unidentified glycolipids that differentiate these two novel strains from close members in the genus Gemmobacter . 3.2. Phylogenetic and Genomic Analysis: The Taxonomic Status The 16S rRNA sequences of strains con5 T and con4 share 100% identity between them and 94.9–97.9% identity with those of the closest species within the genus Gemmobacter ( Table 2 ). The 16S rRNA gene sequences of strains con5 T and con4 were compared with the 16S rRNA gene sequences of representative species within the genus Gemmobacter and related genera in the EzTaxon-e server. The two strains share over 97.0% similarity with G. aquaticus A1-9 T (97.9%), G. caeruleus N8 T (97.7%), G. lutimaris YJ-T1-11 T (97.4%), and G. tilapiae KCTC 23310 T (97.3%) and less than 97% with the remaining species within the genus Gemmobacter . Strains con5 T and con4 also share high similarities with other species than members of Gemmobacter : 96.7% with Cypionkella collinsensis 4-T-34 T , 96.7% with Cypionkella psychrotolerans PAMC 27389 T , and 96.4% with Cypionkella aquatica DC2N1-10 T . However, it was clear from the topology of the phylogenetic tree ( Figure 1 ) that strains con5 T and con4 clustered clearly with the species of Gemmobacter . In addition, the phylogenomic tree reconstructed on the TYGS provided clearer evidence for the taxonomic position of the two strains within the genus Gemmobacter ( Supplementary Figure S3 ). The genomic DNA G + C content of the two strains was 64.1 mol%, which is in the range reported previously for the genus Gemmobacter (61.4–69.4 mol%) [ 3 , 4 ]. The ANI and dDDH values of strains con5 T and con4 with other available type strains of Gemmobacter were 73.31–80.61 % and 19.03–23.15 % ( Table 2 , Supplementary Figure S4 ), respectively, which were much lower than the species boundaries of ANI or dDDH of 95–96% and 70%, respectively, and fall in the intergeneric range [ 41 , 42 , 43 ]. 3.3. Genome Properties The genome of strain con5 T was 4.7 Mb and contained a circular chromosome of 3.4 Mb and six plasmids sized 34.0–425.5 kb (CP076361–CP076367) ( Supplementary Tables S1 and S2 ). Of 4534 genes, 4472 were protein-coding genes and 62 were RNA genes (nine rRNA genes, 52 tRNA genes, and one tmRNA gene). For strain con4 (JAHHWR000000000), of 4417 genes, 4317 were protein-coding genes and 53 were RNA genes (four rRNA genes, 46 tRNA genes, one tmRNA gene, and two ncRNA genes) ( Supplementary Figure S5 ). The DNA G + C content of both strains was 64.1 mol% ( Supplementary Tables S1 and S2 ). 3.4. Genome Analyses for Denitrification and Methane Oxidation It is generally understood that methanotrophic bacteria are mostly active at the oxic-anoxic transition zone in stratified lakes, using oxygen to oxidize methane. The methanotrophs produce methane monooxygenase to utilize methane as a carbon source [ 44 , 45 ]. The methanotrophs express two kinds of methane monooxygenase, soluble methane monooxygenase (sMMO) and particulate methane monooxygenase (pMMO), and these enzymes can also oxidize various alkanes [ 46 , 47 , 48 ]. Strain con5 T possesses sMMO ( prmAC, mimBD ) but does not have a pMMO gene in its genome sequence. The presence of one PQQ-dependent alcohol dehydrogenase (ADH) (PQQ-MDH: adh , adh1, and adhB ), two NAD + -dependent ADH ( adh , adh1 ), and another alcohol dehydrogenase ( adhB ) in strain con5 T makes it a good candidate for the conversion of methane to aldehyde ( Figure 2 ). Genes involved in the glutathione (GSH)-dependent pathway to metabolize formaldehyde are identified. Formaldehyde (HCHO)-activating enzyme ( gfa ), a GSH-dependent formaldehyde dehydrogenase ( fdh ), and S -formyl-GSH hydrolase ( fgh ) convert formaldehyde to formate, and then the formate is oxidized by formate dehydrogenase ( fdh ) to the final product, carbon dioxide (CO 2 ) ( Figure 2 and Figure 3 ). Those genes were also found in the strain con4 ( Figure 4 ). Some methanotrophs comprise genes encoding enzymes for the nitrate reduction pathway, which was confirmed to be related to methane oxidation under anoxic conditions [ 49 ], and these methanotrophic species encode a complete nitrate reduction pathway to use nitrate as a terminal electron acceptor when oxygen is depleted [ 50 ]. The genome of strain con5 T contains nirB , nirK , nirQ , norB , norC , and norG genes involved in dissimilatory nitrate reduction, but no dissimilatory nitrate reductase ( narG ) or nitrous oxide reductase ( nosZ ) genes are detected, which is incomplete for the denitrification pathway ( Figure 2 and Figure 3 ). This suggests that strain con5 T cannot perform nitrate reduction, itself, unless it utilizes unknown or incorrectly classified reduction pathways up to date. The genome of strain con5 T contains the nitrite reductase gene, nirK , and the nitric-oxide reductase gene norB , and it could potentially use nitrite as an electron acceptor in anoxic environments."
} | 3,028 |
32953424 | null | s2 | 4,521 | {
"abstract": "The development of power generators that can function in harsh snowy environments and in contact with snow can be beneficial but challenging to accomplish. Herein, we introduce the first snow-based triboelectric nanogenerator (snow-TENG) that can be used as an energy harvester and a multifunctional sensor based on the principle of snow-triboelectrification. In this work, we used a 3D printing technique for the precise design and deposition of the electrode and triboelectric layer, leading to flexible, stretchable and metal-free triboelectric generators. Based on the single electrode mode, the device can generate an instantaneous output power density as high as 0.2 mW/m"
} | 169 |
22428027 | PMC3302856 | pmc | 4,522 | {
"abstract": "Background Coral bleaching events vary in severity, however, to date, the hierarchy of susceptibility to bleaching among coral taxa has been consistent over a broad geographic range and among bleaching episodes. Here we examine the extent of spatial and temporal variation in thermal tolerance among scleractinian coral taxa and between locations during the 2010 thermally induced, large-scale bleaching event in South East Asia. Methodology/Principal Findings Surveys to estimate the bleaching and mortality indices of coral genera were carried out at three locations with contrasting thermal and bleaching histories. Despite the magnitude of thermal stress being similar among locations in 2010, there was a remarkable contrast in the patterns of bleaching susceptibility. Comparisons of bleaching susceptibility within coral taxa and among locations revealed no significant differences between locations with similar thermal histories, but significant differences between locations with contrasting thermal histories (Friedman = 34.97; p<0.001). Bleaching was much less severe at locations that bleached during 1998, that had greater historical temperature variability and lower rates of warming. Remarkably, Acropora and Pocillopora , taxa that are typically highly susceptible, although among the most susceptible in Pulau Weh (Sumatra, Indonesia) where respectively, 94% and 87% of colonies died, were among the least susceptible in Singapore, where only 5% and 12% of colonies died. Conclusions/Significance The pattern of susceptibility among coral genera documented here is unprecedented. A parsimonious explanation for these results is that coral populations that bleached during the last major warming event in 1998 have adapted and/or acclimatised to thermal stress. These data also lend support to the hypothesis that corals in regions subject to more variable temperature regimes are more resistant to thermal stress than those in less variable environments.",
"introduction": "Introduction Coral reefs are critically important for the ecosystem goods and services they provide to maritime tropical and subtropical nations [1] . However, major coral bleaching events – caused by a breakdown in the relationship between scleractinian corals and their algal symbionts – have led to widespread coral mortality on reefs in recent decades [2] . Global warming poses a particularly significant threat to the future of coral reef ecosystems because large-scale coral bleaching episodes are strongly correlated with elevated sea temperatures [3] , [4] . Indeed, among Earth's ecosystems, coral reefs are one of the most severely threatened by global warming [5] . Coral bleaching severity varies in space and time as a consequence of the magnitude of thermal stress [6] , levels of irradiance [7] , [8] , symbiont types [9] , the species composition of the coral assemblage [10] , [11] , [12] and thermal history of the site [13] , [14] . Species composition is one of the strongest drivers of this variation due to a predictable hierarchy of susceptibility among coral taxa [10] , [11] , [12] . Fast growing branching taxa, such as Acropora and Pocillopora , are normally highly susceptible to thermal stress; they bleach rapidly and experience high rates of whole colony mortality [15] . In contrast, massive taxa such as Porites and some faviids are more resistant to bleaching, they take longer to bleach, and although they may stay bleached for longer, few entire colonies die [15] . This consistency has led to the prediction that hardier, slow-growing massive species will replace less hardy, fast-growing branching species on reefs in the future [10] , [16] . The thermal history of a site may also play an important role in determining bleaching severity. For example, on reefs with naturally higher temperature fluctuations, corals are frequently exposed to stressful temperatures for short periods, and this may lead to greater tolerance during episodes of more prolonged thermal stress [14] , [17] . Scleractinian corals are the major framework builders of reefs and provide most of the structural complexity in reef ecosystems. Therefore, the capacity of coral species to adapt and acclimatise to increasing episodes of thermal stress will greatly influence rates of reef degradation [5] . Several studies cite repeated bleaching episodes in the same coral assemblages, the increasing scale and frequency of coral bleaching and the low overall evolutionary potential of scleractinians as evidence that corals have exhausted their capacity to adapt to rising sea temperatures [18] , [19] . In contrast, other studies show considerable spatial and temporal variation in bleaching susceptibility within scleractinian taxa, suggesting an underappreciated capacity for corals to adapt and/or acclimatise to thermal stress [20] . If the hypothesis that corals still have the capacity to adapt to elevated sea temperatures is correct, we would expect to find increases in thermal tolerance on reefs that have previously experienced major bleaching with the most susceptible species exhibiting the greatest increases in thermal tolerance [21] . Furthermore, we would expect reefs in more thermally variable environments to bleach less severely during episodes of elevated sea temperatures [14] . Here we examine the bleaching and mortality responses of corals at sites with contrasting thermal histories during a large-scale bleaching event in 2010. Our data provide evidence in support of both hypotheses as we documented an unprecedented reversal in the susceptibility of coral genera, but only at sites where bleaching occurred in 1998. Furthermore we show that corals generally bleached less severely at locations where temperature variability has been greater and warming rates lower over the last 60 years.",
"discussion": "Discussion Fast growing branching coral taxa, such as Acropora and Pocillopora , are normally highly susceptible to thermal stress and to date there has been a predictable hierarchy of bleaching susceptibility that was consistent over a wide geographic range and among bleaching events [10] , [11] , [12] . The hierarchy of susceptibility in Pulau Weh in 2010 was typical of previous bleaching episodes, for example the 1998 event on the GBR [11] , but was in marked contrast with the patterns of susceptibility observed in Singapore and Tioman Island (see Figure 1 ). Comparisons of the BMI of taxa among locations revealed significantly different patterns between Pulau Weh and the South China Sea locations and confirmed a reversal in the normal patterns of susceptibility between Singapore and Pulau Weh. This is the first time such a reversal has been reported during a major warming-induced bleaching event. The remotely sensed temperature data corroborate reports indicating that corals in Singapore and Tioman Island bleached in 1998 but those in Pulau Weh did not. Extensive bleaching was documented in Singapore, in several nearby Indonesian sites [23] and Tioman Island [24] during 1998; whereas there are no reports of bleaching from Pulau Weh prior to 2010 despite numerous reports from elsewhere in the Indonesian archipelago ( www.reefbase.org ). Local dive operators have not witnessed mass bleaching in the area in the last 30 years and in nearby Andaman Sea sites, severe bleaching was not observed during 1997–1998 [8] . A parsimonious explanation for the contrasting bleaching responses among locations, therefore, is that removal of susceptible individuals from populations that bleached during 1998 in Singapore and Tioman Island, followed by reproduction and successful recruitment of the remaining, more thermally tolerant individuals, has led to adaptation through natural selection within an ecological time frame [13] , [25] . Recurring bleaching episodes of increasing magnitude and frequency within coral assemblages have been cited as evidence that corals have exhausted their capacity to adapt and it is often stated that the generation times of corals are too long to allow rapid adaptation to a changing climate [18] , [19] . In contrast, a growing body of evidence indicates that the capacity for adaptation and acclimatisation in corals has been underestimated [13] , [21] , [26] . Even for highly susceptible coral species, variation in specific characteristics of the symbiotic zooxanthellae [27] and the coral host [28] lead to different bleaching responses among colonies. Selective mortality among individuals within populations suggests there is sufficient genetic variability upon which natural selection can act [29] . Several studies have documented increasing thermal tolerance and declining rates of bleaching induced mortality over successive bleaching episodes [21] , [30] . Similarly, thermal history and previous exposure to thermal stress have been shown to determine bleaching responses to contemporary thermal stress [13] . The most compelling evidence of an adaptive response at our study locations is that the taxa that showed the greatest contrast in response ( Acropora and Pocillopora ), have life history traits most likely to lead to rapid adaptation. For example, these taxa become sexually mature within 2 to 3 years [31] , [32] and typically experience high rates of whole colony mortality following thermal stress [15] . Most taxa bleached much less severely and far fewer corals died in Singapore and at Tioman Island than in Pulau Weh in 2010. The 2010 episode in Pulau Weh was greater in magnitude compared to previous major bleaching episodes, such as that on the GBR in 1998 [11] , and surveys from other sites in the Andaman Sea indicate that the 2010 event is the most severe for this region on record [8] , [33] . The differences in overall bleaching severity among the three study locations in 2010 are not readily explained by differences in the magnitude of the thermal anomaly but may have been influenced by long-term differences in thermal histories at each location. In environments with naturally higher temperature fluctuations, the coral holobiont is frequently exposed to stressful temperatures for short durations, and this may lead to greater tolerance during episodes of prolonged thermal stress [14] , [17] , [34] . Consequently, acclimatisation of corals driven by greater thermal variability and facilitated by slower warming rates may also have led to overall differences in the severity of bleaching responses among locations. If our findings apply more generally then locations that are more resistant to bleaching can be identified from their thermal histories. Such knowledge can be used to inform protected area planning by aiding in the identification of sites with lower relative vulnerability to global warming [35] , [36] . It is often stated that corals have exhausted their capacity to adapt to thermal stress [18] , [19] . Here we provide evidence in support of the alternative hypothesis, i.e., taxa that, to date, have been consistently the most thermally susceptible possess an underappreciated capacity for adaptation to thermal stress [26] . Identification of genes that respond to thermal stress and are under selection, followed by studies to quantify changes in gene expression and gene frequency among coral populations are required to assess the likelihood that adaptation has driven the response seen in these populations [37] . Our study also highlights the critical importance of comparing rates of bleaching induced mortality within coral populations and spanning repeated bleaching episodes; indeed, such data are essential if we hope to assess the capacity of coral populations to adapt to rising temperatures. We cannot rule out the possibility that differences in irradiance [8] , turbidity and thermal stress among locations also contributed to the spatial variation in the severity of bleaching in 2010. For example lower bleaching severity in Singapore may in part be explained by lower thermal stress and higher turbidity relative to the other sites – however, the differences in environmental stress do not explain the reversal in the hierarchy of susceptibility among taxa. An adaptive response in certain taxa at a few locations does not mean that the global threat to reefs from climate change has lessened. There are likely to be limits to thermal adaptation and acclimatisation, and these may incur costs in life history traits such as growth, fecundity and competitive ability [20] . In addition, reefs continue to be threatened by numerous other factors including overfishing, pollution, disease, acidification, and severe storms [16] . The results of the present study do indicate however that the effects of bleaching will not be as uniform as anticipated [20] and fast-growing branching taxa such as Acropora and Pocillopora are likely to persist in some locations despite increases in the frequency of thermal stress events."
} | 3,237 |
39985278 | PMC11934084 | pmc | 4,523 | {
"abstract": "ABSTRACT Climate‐driven range extensions of animals into higher latitudes are often facilitated by phenotypic plasticity. Modifications to habitat preference, behaviour and diet can increase the persistence of range‐extending species in novel high‐latitude ecosystems. These strategies may be influenced by changes in their gut and stomach microbial communities that are critical to host fitness and potentially adaptive plasticity. Yet, it remains unknown if the gut and stomach microbiome of range‐extending species is plastic in their novel ranges to help facilitate these modifications. Here, we categorised stomach microbiome communities of a prevalent range‐extending coral reef fish along a 2000‐km latitudinal gradient in a global warming hotspot, extending from their tropical core range to their temperate cold range edge. At their cold range edge, the coral reef fish's stomach microbiome showed a 59% decrease in bacterial diversity and a 164% increase in the relative abundance of opportunistic bacteria ( Vibrio ) compared to their core range. Microbiome diversity was unaffected by fish body size, water temperature, physiology (cellular defence and damage) and habitat type (turf, barren, oyster, kelp and coral) across their range. The observed shifts in microbiome composition suggest dysbiosis and low plasticity of tropical range‐extending fishes to novel environmental conditions (e.g., temperate prey and lower seawater temperature) at their novel range edges, which may increase their susceptibility to disease in temperate ecosystems. We conclude that fishes extending their ranges to higher latitudes under ocean warming can experience a simplification (i.e., reduced diversity) of their stomach microbiome, which could restrict their current rate of range extensions or establishment in temperate ecosystems.",
"conclusion": "5 Conclusions We reveal that the stomach microbiome of a prevalent range‐extending coral reef fish shows decreased diversity and increased abundance of pathogenic bacterial species, which indicates dysbiosis and low plasticity of their microbiome at their novel temperate cold‐range edge. Dysbiosis and low plasticity of the microbiome might be a present‐day mediator of the rate of colonisation and persistence of coral reef fishes in the early stages of range extensions into temperate ecosystems, irrespective of the immediate drivers of gastrointestinal microbiome changes.",
"introduction": "1 Introduction Anthropogenic warming has facilitated the global redistribution of marine and terrestrial species (Parmesan and Yohe 2003 ; Chen et al. 2011 ; Pecl et al. 2017 ). Climate‐driven species redistributions have already altered species interactions and entire ecosystem functioning (Pecl et al. 2017 ). Marine species are shifting their distributions poleward at a faster rate than terrestrial species (Chen et al. 2011 ; Burrows et al. 2011 ; Poloczanska et al. 2013 ). Poleward range shifts can act as a mechanism to escape thermally stressful conditions at lower latitudes or allow species to inhabit previously inaccessible higher latitudes (Poloczanska et al. 2016 ). Animals range shifting into high‐latitude environments often modify their diet (Kingsbury et al. 2019 ; Monaco et al. 2020 ), habitat preference (Hayes et al. 2024 ), behaviour (Coni, Booth, and Ferreira et al. 2021 ), physiology (Mitchell et al. 2023a ) and/or morphology (Smith et al. 2016 ) to enhance their performance in novel ecosystems. However, a fundamentally overlooked response to their range‐shift success is changes in host‐specific microbiomes—the bacterial communities harbouring the internal and external surfaces of organisms (Wilkins et al. 2019 ). Microbial communities shape host physiology (Gould et al. 2018 ), immunity (Gerardo et al. 2020 ), behaviour (Ezenwa et al. 2012 ) and metabolism (Dvergedal et al. 2020 ) and can respond to environmental change faster than their host. Under rapid environmental change, shifts in microbial communities can promote acclimatisation and genetic adaptation (Alberdi et al. 2016 ; Webster and Reusch 2017 ; Peterson et al. 2023 ). Adaptive responses are apparent when microbial communities show a high degree of plasticity in response to environmental change, which can benefit host resilience or adaptation (Alberdi et al. 2016 ). Species fitness in changing environments can be mediated by host‐associated microbial communities (Pinnow et al. 2023 ), whereby beneficial microbial communities can enhance thermal tolerance (Jarmillo and Castañeda 2021 ) and modulate host pathogenic immunity (Fleischer et al. 2022 ). Hence, shifts in host microbial communities could indirectly mediate host resilience or vulnerability to changing environments. Yet, whether mutualistic relationships between host fitness and host‐associated microbial communities are plastic in animals exposed to novel or changing climatic conditions remains largely unknown. Microbiome dysbiosis arises when there is an imbalance or shift in the host's natural microbial composition (Petersen and Round 2014 ). When dysbiosis occurs, the mutually beneficial interaction between the host and their microbiome community is disrupted, leading to a reduction in microbiome diversity and an increase in pathogenic bacteria (Petersen and Round 2014 ). Across both marine and terrestrial taxa, previous research has shown that increased temperature can alter the host's microbiome composition (Bestion et al. 2017 ; Watson et al. 2019 ; Scanes et al. 2021 ; Moore et al. 2024 ), which can promote dysbiosis (Greenspan et al. 2020 ; Suzzi et al. 2023 ), decreased fitness (Steiner et al. 2022 ; Risely et al. 2023 ) and increased susceptibility to disease (Brown et al. 2012 ) in animals. Thus, future ocean warming may alter microbiome compositions across a wide range of animal taxa, which may disadvantage species fitness under future climate change. Climate‐driven shifts in microbiome communities may facilitate adaptive plasticity, providing benefits to the host adjusting to novel environmental challenges, shaping their overall adaptation to the environment (Kolodny and Schulenburg 2020 ). While the response between host fitness and host‐associated microbiome communities is well understood in mammals (Suzuki 2017 ) and other vertebrates (Ley et al. 2008 ), including fishes in aquaculture settings (Infante‐Villamil et al. 2020 ), there is still limited research on how fishes that are extending their ranges to higher latitudes under ocean warming experience shifts in their stomach microbiome composition. Most microbiome studies on fish use controlled experimental designs to understand the relationship between fish and their microbiome under climate change. However, these studies may not capture the ecological complexities of natural ecosystems, whereby species are challenged by novel species interactions (Smith et al. 2018 ; Mitchell et al. 2022 ), resource competition (Nagelkerken and Munday 2016 ; Coni, Booth, and Nagelkerken 2021 ) and habitat degradation/loss (Stuart‐Smith et al. 2021 ; Coni, Nagelkerken, and Ferreira et al. 2021 ) under climate change. This emphasises the need for a more comprehensive understanding of how host–microbiome interactions respond to climate change in natural complex ecological settings. Coral reef fishes contribute to one of the most diverse assemblages of vertebrates globally and are considered increasingly vulnerable to environmental change (Comte and Olden 2017 ). Climate change has already increased the poleward dispersal of coral reef fishes into subtropical and temperate ecosystems (Booth et al. 2011 ; Vergés et al. 2014 ), which has disrupted temperate ecosystem functionality (Nakamura et al. 2013 ; Vergés et al. 2016 ) and generated novel species interactions between local and range‐extending species (Smith et al. 2018 ). There has been substantial focus on what facilitates the poleward movement of coral reef fishes (e.g., increased ocean temperatures, strengthening of boundary currents and species traits; Booth et al. 2007 , 2018 ; García Molinos et al. 2022 ) and the outcome of their range shifts to temperate ecosystems (e.g., novel species interactions and resource competition, Smith et al. 2018 ; Coni, Nagelkerken, and Ferreira et al. 2021 ). Despite this, how fish microbiome diversity and functioning may facilitate or limit their range shifts into novel environments remains relatively unknown (but see Jones et al. 2018 ). The microbiome of fishes can shape their physiology and ecology (Clements et al. 2014 ) and correlate strongly with diet and phylogeny (Sullam et al. 2012 ). Additionally, fish microbiome can modulate host immune responses to pathogenic and environmental stressors (Butt and Volkoff 2019 ). Thus, it is of great ecological importance to understand how the microbiome of coral reef fish assemblages responds to environmental change. In Australia, over ~150 coral reef fish species have been observed range shifting into nearshore marine temperate ecosystems of the southeast Australian coastline during summer months (Booth et al. 2011 ; Feary et al. 2014 ). Many of these coral reef fishes interact with local temperate fishes (Smith et al. 2018 ; Coni, Booth, and Nagelkerken 2021 ), which can both increase (Smith et al. 2018 ; Paijmans et al. 2020 ) and decrease (Coni, Booth, and Nagelkerken 2021 ) the behavioural performance of coral reef fishes in novel temperate ecosystems. Yet, many coral reef fishes still fail to permanently establish in southeast Australian temperate ecosystems, because winter temperatures fall below their thermal critical minima and prevent overwintering success (Figueira et al. 2009 ; Booth et al. 2011 ). Additionally, these temperate ecosystems introduce novel prey, predators and competitors (Beck et al. 2016 ; Coni, Booth, and Nagelkerken 2021 ). In response, coral reef fishes can show behavioural (Coni, Booth, and Ferreira et al. 2021 ), dietary (Kingsbury et al. 2019 ; Monaco et al. 2020 ), habitat (Hayes et al. 2024 ) or physiological (Kingsbury et al. 2020 ; Hayes et al. 2024 ) plasticity to enhance their establishment in temperate ecosystems or to reduce competition with the local temperate fish species. Such responses to enhance coral reef fish persistence in their novel temperate ecosystems may be strongly shaped by their microbiome structure and plasticity, although this remains unknown. Here we investigate the stomach microbiome of a range‐extending coral reef fish collected in situ in a global warming hotspot along a 2000‐km latitudinal gradient encapsulating their tropical core range and their temperate novel leading range edge in eastern Australia. We chose the most prevalent and successful range‐extending coral reef fish species, the sergeant major damselfish (Abudefduf vaigiensis; Hayes et al. 2024 ). Understanding the degree of plasticity in the stomach microbiome of tropical range‐extending fishes remains unknown but is a critical component in predicting the rate and success of their range extension and population dynamics in temperate ecosystems under future climate change.",
"discussion": "4 Discussion We here show that the microbiome of a prevalent range‐extending coral reef fish is simplified at its novel temperate cold‐range edge. Decreased microbiome diversity and a shift in microbiome community structure associated with an increased prevalence of opportunistic bacteria ( Vibrio ) at the cold temperate region compared to its historical range together indicate simplification and dysbiosis, respectively, of the fish stomach microbiome in their novel ranges. Although high variability in microbial community structure can also indicate dysbiosis (Zaneveld et al. 2017 ), we found no consistent trends of variability across sampling regions. Dysbiosis of the microbiome and increased prevalence of Vibrio species can negatively affect the health and fitness of range‐extending coral reef fishes by increasing immune suppression (Moore et al. 2024 ), disease occurrence (Belden and Harris 2007 ) and mortality (Greenspan et al. 2020 ; Risely et al. 2023 ). Additionally, this can compromise their behavioural (Florkowski and Yorzinski 2023 ) and physiological (Gould et al. 2018 ) responses, both of which can underpin successful range extensions into temperate ecosystems. Coral reef fishes extending their ranges into temperate ecosystems experience increased susceptibility to cold stress (Figueira et al. 2009 ) and novel interactions with temperate competitors, prey or predators (Beck et al. 2016 ); therefore, simplification and dysbiosis of their microbiome could exacerbate vulnerability to novel stressors at their leading range edge. This suggests that microbiome simplification may mediate the colonisation and persistence of range‐extending coral reef fish in novel temperate ecosystems. Microbiome plasticity of range‐extending species may enhance their adaptive potential and persistence in novel ecosystems. At the tropical and subtropical regions, the microbiome of the range‐extending fish species showed a heterogeneous community structure, with no distinct genera dominating their stomach microbiome. However, at the novel cold temperate region, the community structure was simplified (i.e., less diverse) with two genera ( Vibrio and Pseudarthrobacter ) contributing to ~58% of the relative microbial abundance. Although the function of Pseudarthrobacter in fishes remains unknown, it is well understood that some Vibrio species are pathogenic and can cause body malformation, slow growth and increased disease prevalence and mortality in fishes (Ina‐Salwany et al. 2018 ). This change in microbiome community structure and diversity suggests the range‐extending coral reef fish exhibits low microbial plasticity at their cold‐range edge. Whilst this species shows high dietary and behavioural plasticity at their cold‐range edge (Kingsbury et al. 2019 ; Coni, Booth, and Ferreira et al. 2021 ), the observed low microbiome plasticity could reduce their ability to respond to novel challenges (competition, predation, prey and cold stress) in temperate ecosystems. Therefore, the inability to maintain the integrity of their stomach microbiome may mitigate their adaptive potential and persistence at their cold‐range edge. Stomach microbiome diversity of the range‐extending coral reef fish was unaffected by water temperature, habitat types, cellular defence, cellular stress and body size towards their cold‐range edge compared to their tropical native range. This suggests that the observed reduced microbiome diversity at their cold‐range edge occurs independently and is not influenced by novel temperate water temperatures, habitat types or host physiological performance. Host habitat (Kim et al. 2021 ; Clever et al. 2022 ), physiology (Clements et al. 2014 ), diet (Miyake et al. 2014 ) and behaviour (Trevelline and Kohl 2022 ) have previously been identified as major determinants of microbial diversity. Our focal species experiences increased oxidative stress (the combination of decreased cellular defence and increased cellular damage) at their cold‐range edge compared to their tropical native range (Hayes et al. 2024 ), as well as reduced feeding and activity levels (Kingsbury et al. 2020 ; Coni et al. 2022 ). The increased cellular damage may diverge energy away from other important fitness‐related traits such as reproduction and growth (Birnie‐Gauvin et al. 2017 ; Zhang et al. 2023 ). This species also consumes a wide variety of prey groups across the regions and shows a high degree of dietary generalism in novel temperate ecosystems (Kingsbury et al. 2019 ; Monaco et al. 2020 ). At the cold temperate region compared to the subtropical region, consumption increased for zooplankton (from ~38% to ~66%) and crustaceans (from ~4% to ~16%), but that of macroalgae decreased (from ~50% to ~7%; see figure S7 in Kingsbury et al. 2019 ). The observed simplification of the microbiome community structure of the fish could be influenced by an indirect response to other environmental changes along the gradient, such as altered prey communities. Factors influencing microbiome shifts, such as water temperatures, food sources and habitat types, are expected to change and simplify under future climate change scenarios (Nagekerken et al. 2020 ; Agostini et al. 2020 ; Coni, Nagelkerken, and Ferreira et al. 2021 ). Direct and indirect changes in abiotic and biotic variables are both driven by changes in climate, such as rapid ocean warming at our warm and cold temperate study sites. Therefore, irrespective of the underlying mechanisms influencing the microbiome structure, the observed shifts in the microbiome of a common range‐extending coral reef fish may mediate their persistence in novel temperate ecosystems. Future ocean warming will likely relax the thermal stress of coral reef fishes residing in novel temperate ecosystems and increase their likelihood of successful persistence, which may reduce microbial disturbances. Climate‐driven warming and strengthening of the East Australian Current (Wu et al. 2012 ) are projected to expand the prevalence of tropical microbes into temperate waters (Messer et al. 2020 ), potentially mediating beneficial microbial taxonomic shifts that could relieve current dysbiosis in their microbial structure. However, ocean warming drives higher abundances of Vibrio species (Baker‐Austin et al. 2013 ) because their abundance is positively correlated with increasing water temperature (Williams et al. 2022 ). Despite this, tropicalisation of microbial communities in temperate ecosystems could introduce beneficial microbes capable of suppressing pathogenic Vibrio (Messer et al. 2020 ), although this remains unknown. Additionally, ocean warming can benefit range‐extending coral reef fishes in temperate ecosystems through increased physiological function (Mitchell et al. 2023a ), growth (Djurichkovic et al. 2019 ; Mitchell et al. 2023b ) and foraging performance (Coni, Booth, and Nagelkerken 2021 ), overall enhancing successful establishment in their future ecosystems. Therefore, when future water temperatures track the thermal optima of range‐extending coral reef fishes, negative alterations to microbial communities may be alleviated and benefit the establishment of coral reef fishes in temperate ecosystems."
} | 4,613 |
35646842 | PMC9136054 | pmc | 4,524 | {
"abstract": "Microalgae have drawn much attention for their potential applications as a sustainable source for developing bioactive compounds, functional foods, feeds, and biofuels. Diatoms, as one major group of microalgae with high yields and strong adaptability to the environment, have shown advantages in developing photosynthetic cell factories to produce value-added compounds, including heterologous bioactive products. However, the commercialization of diatoms has encountered several obstacles that limit the potential mass production, such as the limitation of algal productivity and low photosynthetic efficiency. In recent years, systems and synthetic biology have dramatically improved the efficiency of diatom cell factories. In this review, we discussed first the genome sequencing and genome-scale metabolic models (GEMs) of diatoms. Then, approaches to optimizing photosynthetic efficiency are introduced with a focus on the enhancement of biomass productivity in diatoms. We also reviewed genome engineering technologies, including CRISPR (clustered regularly interspaced short palindromic repeats) gene-editing to produce bioactive compounds in diatoms. Finally, we summarized the recent progress on the diatom cell factory for producing heterologous compounds through genome engineering to introduce foreign genes into host diatoms. This review also pinpointed the bottlenecks in algal engineering development and provided critical insights into the future direction of algal production.",
"conclusion": "5 Concluding Remarks and Future Perspectives This review summarizes the current systems and synthetic biology progress for diatom cell factories toward biotechnology applications ( Figure 4 ). The systems biology approaches, including the GEMs, have facilitated metabolic prediction and rapid development of strain engineering and metabolic engineering in diatoms. Unlike GEMs of well-studied model organisms like E. coli , the establishment of diatom GEM is still in its early stage. The lack of high-quality and high-quantity genome annotations impedes the development of diatoms’ system biology ( Vavitsas et al., 2021 ). PE and biomass productivity enhancements are essential foundations and boosters for cost-effective algal production. There are also considerable gaps between diatoms and some model green microalgae in the photosynthetic field, such as the modification on CCMs and Rubisco. Although more studies have to be done on diatoms, it is clear that the long-awaited breakthrough in PE will be realized under the recent progress in synthetic biology, therefore boosting biomass productivity to a brand-new level in diatoms. FIGURE 4 Conceptual scheme for diatom cell factory. (A) systems biology: understanding the whole picture of metabolism in diatoms; (B) synthetic biology: genetic manipulation and genome engineering; (C) photosynthetic efficiency: fundamental principle of energy conversion and biomass production; (D) value-added products: output from diatom cell factory including homologous and heterologous compounds. Although the knowledge of the biosynthetic steps of bioactive compounds and genetic engineering help improve the yield and productivity of value-added compounds, diatom synthetic biology is still in its infancy. The essential enzymes and genes of interest, together with their regulation, are not fully understood. The diversity of the genome also poses a significant challenge to developing efficient, convenient, and stable transgenic tools for practical application since there has been no report of using the popular CRISPR technology for heterologous expression in diatoms yet. However, with the continuous development of gene-editing technology, new tools such as TALENs and CRISPR/Cas9 play a critical role in producing value-added compounds in diatoms. The combination of synthetic and systems biology will also allow us to understand better the compound conversion pathways and energy flows in diatom cell factories. These new technologies will facilitate the discovery of new compounds and the improvement of their yield, making diatoms one of the best candidates for cell factories. In the foreseeable future, we can anticipate that more compounds for bioenergy, food, feed, and pharmaceutical industries will be developed through algal cell factories, which could play an essential role in the green economy. In short, diatom cell factories may provide solutions to global challenges like rising CO 2 and energy crises. The diversity of diatom species offers us tremendous possibilities for digging potential value-added products. More discoveries and technological breakthroughs are needed to drive diatom research from the laboratory to commercialization. Moreover, this progress requires support from government policymakers, investors, educators, and other stakeholders. We believe a sustainable diatom cell factory is promising and achievable under continuous development of systems and synthetic biology to combine PE enhancement and compound yield improvement for biosustainability and carbon neutrality.",
"introduction": "1 Introduction Diatoms are natural cell factories that synthesize various value-added compounds, such as polyunsaturated fatty acids, pigments, terpenes, and sterols ( Yang et al., 2020 ). Recent years have witnessed tremendous progress in genome sequencing and editing of diatoms. The genome sequencing of two model diatoms, Phaeodactylum tricornutum and Thalassiosira pseudonana , has offered the genetic basis for the biotechnology development of diatoms. The reconstruction of genome-scale metabolic models (GEMs) in diatoms has promoted metabolic engineering strategies to improve algal productivity. Genome engineering tools such as CRISPR (clustered regularly interspaced short palindromic repeats)/Cas (CRISPR associated system), transcription activator-like effector nucleases (TALENs), zinc-finger nucleases (ZFNs), microRNA (miRNA), and small interfering RNA (siRNA) are in rapid development to harness the genomic potential. For instance, the genome of diatom P. tricornutum could be efficiently edited by CRISPR/Cas9 system ( Ng et al., 2017 ) and the TALEN-mediated system ( Serif et al., 2017 ). Thus, diatoms have broad prospects in the production of value-added compounds with systems and synthetic biology tools enabling the characterization of many biosynthetic pathways and related genes. Nevertheless, the bottleneck of commercializing commodities like biomaterials (such as polyhydroxyalkanoates) and biofuels by microalgal cell factories lies in the biomass productivity in high-density culture. Optimizing photosynthetic efficiency (PE) may achieve high overall productivity as microalgae are considered efficient solar energy converters with metabolic flexibility. For compound production in the high-density culture of diatoms, it is feasible to utilize a two-phase culture, which possesses high biomass productivity under optimal conditions in the first stage and then accumulates value-added compounds such as EPA and fucoxanthin under stress conditions in the second stage, showing the potential for industrial applications ( Yang and Wei, 2020 ; Parkes et al., 2021 ). As a group of oil-producing algae, diatoms can be used as a source to produce biofuels and bioactive compounds such as fatty acids and carotenoids. The omega-3 polyunsaturated fatty acids such as docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA) from diatoms have critical biological functions for fetal growth and development based on their roles in maintaining the healthy function of the brain and retina ( Marella and Tiwari, 2020 ). In addition, diatoms can produce non-native value-added compounds through heterologous expression systems, such as plant triterpenoids, monoclonal antibodies, and bioplastics ( Butler et al., 2020 ; Lu et al., 2021 ). The traditional heterologous expression platforms like Escherichia coli and Saccharomyces cerevisiae have already been widely used. However, these classical expression systems require external organic carbon sources, limiting their carbon neutrality contributions ( Hempel and Maier, 2012 ). Compared to other cell factories, such as bacteria, yeast, plants, and mammals ( Table 1 ), diatoms have shown promise to become an ideal natural expression system due to their advantages of high efficiency, low cost, and the ability to synthesize complex value-added compounds. Green microalgae, such as the model species Chlamydomonas reinhardtii , have also been reported for the heterologous synthesis of value-added compounds. For comparison, diatoms could achieve stable genetic modifications with an available genetic toolbox and have a number of features; for example, P. tricornutum does have a free intracellular pool of GPP, while C. reinhardtii doesn’t accumulate GPP naturally ( Fabris et al., 2020 ). Further, the rapid progress of diatom sequencing also provides new insights into diatom systems and synthetic biology. TABLE 1 Advantages and disadvantages of different types of cell factories. Platform Advantages Disadvantages Example References bacteria low cost; high growth rate; easy to perform without post-translational modifications; require external nutrition recombinant human insulin (Humulin®) \n Walsh, (2018) \n yeasts have post-translational modifications require external nutrition humanized Ab (Herceptin®) \n Liu et al., (2018) \n plants no external nutrition needed; low cost occupy the land for crops; restricted by-laws; low growth rates and yields recombinant taliglucerase alfa \n Grabowski et al., (2014) \n mammals mature commercial applications; similar to humans; have post-translational modifications high cost; low efficiency; susceptible to pathogens recombinant c1- esterase inhibitor \n van Veen et al., (2012) \n diatoms rapid biomass accumulation; have post-translational modifications low efficiency of autotrophy at high density; transgene-silencing; fewer Synthetic biology tools monoclonal antibodies \n Hempel et al., (2017) \n In this review, we specifically focus on 1) diatom systems biology, including genome sequencing and GEMs, 2) recent progress in PE optimization, and 3) native and non-native bioactive compounds and their biosynthetic pathways in diatoms manipulated by synthetic biology. Finally, we highlight challenges and discuss possible directions for developing algae-based value-added products."
} | 2,599 |
24118108 | null | s2 | 4,525 | {
"abstract": "Intercellular chemical signaling in bacteria, commonly referred to as quorum sensing (QS), relies on the production and detection of compounds known as pheromones to elicit coordinated responses among members of a community. Pheromones produced by Gram-positive bacteria are comprised of small peptides. Based on both peptide structure and sensory system architectures, Gram-positive bacterial signaling pathways may be classified into one of four groups with a defining hallmark: cyclical peptides of the Agr type, peptides that contain Gly-Gly processing motifs, sensory systems of the RNPP family, or the recently characterized Rgg-like regulatory family. The recent discovery that Rgg family members respond to peptide pheromones increases substantially the number of species in which QS is likely a key regulatory component. These pathways control a variety of fundamental behaviors including conjugation, natural competence for transformation, biofilm development, and virulence factor regulation. Overlapping QS pathways found in multiple species and pathways that utilize conserved peptide pheromones provide opportunities for interspecies communication. Here we review pheromone signaling identified in the genera Enterococcus and Streptococcus, providing examples of all four types of pathways."
} | 326 |
25648201 | PMC4316194 | pmc | 4,526 | {
"abstract": "Exploiting biomass as an alternative to petrochemicals for the production of commodity plastics is vitally important if we are to become a more sustainable society. Here, we report a synthetic route for the production of terephthalic acid (TPA), the monomer of the widely used thermoplastic polymer poly(ethylene terephthalate) (PET), from the biomass-derived starting material furfural. Biobased furfural was oxidised and dehydrated to give maleic anhydride, which was further reacted with biobased furan to give its Diels-Alder (DA) adduct. The dehydration of the DA adduct gave phthalic anhydride, which was converted via phthalic acid and dipotassium phthalate to TPA. The biobased carbon content of the TPA was measured by accelerator mass spectroscopy and the TPA was found to be made of 100% biobased carbon.",
"discussion": "Results and Discussion Oxidation of furfural to fumaric acid and maleic acid The oxidisation of furfural with NaClO 4 as an oxidant, and V 2 O 5 as a catalyst, gave a mixture of fumaric acid and maleic acid in 58% yield, lower than the 72% yield reported in the literature 30 . This lower yield could be caused by vigorous oxidation, which is difficult to control. The oxidation of furfural to maleic acid and fumaric acid has been performed by other means with improved yields and better selectivity for the products, either maleic acid or fumaric acid, elsewhere in the literature 31 32 . However, the oxidation with NaClO 4 is, for our purposes, a more practical laboratory process, and, therefore, we adopted this traditional oxidation method in this study. Dehydration of fumaric acid and maleic acid to maleic anhydride The dehydration of the mixture of maleic acid and fumaric acid to maleic anhydride was performed using P 2 O 5 as a dehydration agent. As the ratio of maleic acid and fumaric acid, determined by 1 H NMR, was 1:7, the yield of maleic anhydride could potentially be less than 30%. However, once maleic acid was dehydrated to give maleic anhydride, phosphoric acid, produced by the reaction of P 2 O 5 and water, could isomerise fumaric acid to maleic acid 33 . Consequently, the mixture of maleic acid and fumaric acid was quantitatively converted to maleic anhydride. Diels-Alder (DA) reaction of anhydrous maleic acid and furan to the exo -DA adduct The synthesis of furan from furfural could have been demonstrated in this study. However, commercially available furan is a biobased chemical and has been verified as such in previous work 13 . The biobased carbon contents of the furan and furfural used in this study are shown below. Therefore, although we have not performed the process ourselves, for the purposes of this work, commercially available furan is assumed to be biobased. The DA reaction of maleic anhydride and furan readily gave the DA adduct. At the beginning of the reaction, the regioselectivity of DA cyclisation is for the endo -adduct due to its kinetic stability. However, the DA reaction is reversible and, after some time, the product is converted to the more thermally stable exo -DA adduct 34 35 36 . Consequently, the reaction was carried out for 12 h at room temperature, yielding the DA adduct in almost quantitative yield. Melting point analysis showed the m.p. of the product to be 127–129°C, corresponding to the exo adduct. Dehydration of the exo -DA adduct to phthalic anhydride The oxabicyclo moiety in the exo -DA adduct is readily dehydrated with an acid. We attempted to dehydrate the exo -DA adduct using sulfonic acid, phosphoric acid, and P 2 O 5 , but the yield and purity of the phthalic anhydride produced were not high enough to isolate it. However, a more effective protocol for the dehydration of the exo -DA adduct has been recently developed in which it is treated with a mixture of trifluoromethane sulfonic acid and acetic anhydride 37 . We employed this new method and obtained phthalic anhydride in 84% yield. Hydrolysis of phthalic anhydride to dipotassium phthalate Phthalic acid was readily hydrolysed with aqueous potassium hydroxide to give dipotassium phthalate quantitatively. Transfer reaction and acidification of dipotassium phthalate to TPA Half a century ago, a transfer reaction known as the Henkel method, which converts dipotassium phthalate to dipotassium terephthalate at high temperature (above 400°C) with CdI 2 as a catalyst, was the most common industrial process for the production of TPA 38 39 . The development of an alternative method involving the oxidation of p -xylene to TPA led to the Henkel method losing its competitive advantage and falling out of favour. Consequently, it is rarely used industrially. However, in this study, we adopted the Henkel method to convert biobased phthalate to biobased TPA, as we found it to be a practical method for obtaining biobased TPA from furfural. The reaction was carried out with CdI 2 at 420°C, and the resulting mixture was acidified to give biobased TPA. At 44%, the yield of TPA obtained in this study is not sufficient. However, this figure is obtained at the milligram scale, but the process was optimised industrially, so, therefore, it is reasonable to expect that the yield would increase for industrial production. Additionally, the Henkel method is proven as an industrial process, and, therefore, the commercial viability of this synthetic route from furfural to TPA is already established. Biobased carbon content The biobased carbon contents of the reagents and products are summarised in Table 1 . The synthesis of fully biobased TPA is verified by the fact that the values of furfural, furan, and TPA were almost 100%. Thus, we can conclude that both the starting materials and product are fully biobased chemicals. On the other hand, the values for furfural and furan reported previously were 100.8 and 105.0% 11 13 and slightly higher than those measured in this study. These values were obtained in 2010, while those in this study were obtained in 2014. Since the lot numbers of furfural and furan used in this study are different from those used in the previous study, the actual value of the 14 C/ 12 C ratio could be slightly different. This difference could be explained by the difference in the definition of biobased carbon content between ISO 16620-2 and ASTM D6866 and the manufacturing year of furfural and furan. In principle, the percentage of modern carbon (pMC) calculated from the 14 C/ 12 C concentration ratios, is the biobased carbon content. However, the pMC for biomass produced by fixation of CO 2 in the atmosphere by photosynthesis was 108–110% in 2002 27 28 29 . The pMC is possibly slightly higher than 100% because of the continuing but diminishing effects of nuclear testing in the atmosphere in the 1950s, during which large amounts of 14 C were emitted into the atmosphere. Because the 14 C in all the samples is referenced to a “prebomb” standard, i.e., modern carbon-based oxalic acid radiocarbon [Standard reference material (SRM) 4990c, National Institute of Standards, USA], all pMC values must be multiplied by a cofactor to correct for the bomb carbon and to obtain the true biobased carbon content of the sample. In our previous study, the biobased carbon contents were calculated using a strong cofactor of 0.93, that being the old value based on ASTM D6866 (2008) because the furfural and furan used were purchased before 2009. Nevertheless, the reason why the value of furan was above 105% is that it was produced before 2008. In this study, as the reagents used were purchased in 2013, the biobased carbon contents were calculated using the new value for the weak cofactor of 0.95 based on ISO 16620-2. These indicate that the nuclear testing effect on the old reagents was strong and the biobased carbon content was above 100%, even though the strong cofactor 0.93 was used, while the effect on the new reagents was weaker, giving a biobased carbon content of almost 100%. Therefore, the biobased carbon contents of furan and furfural in this study were slightly different from the values measured in the previous study. The precise method for the measurement of biobased carbon content is detailed in ISO 16620-2 and ASTM D6866 and is an industrially indispensable verification procedure 29 40 . It is important, not just to prevent mistakes by researchers, but also to detect whether supposedly biobased materials have undergone some contamination from, or carbon exchange with, petrochemical sources such as other reaction reagents or non-biobased solvents. For example, in the case of the Henkel method, the transfer reaction could involve the incorporation of a carbonyl carbon from carbon dioxide, produced as a by-product of a petrochemical process 38 41 . In addition, biobased carbon content measurement is also an invaluable method for identifying materials mistakenly or falsely supplied as biobased. Therefore, we propose that the measurement of biobased carbon content should be necessary when biobased chemicals are used, especially when the products can be synthesised from commercially available petroleum-derived starting materials or involve the use of non-biobased reagents or solvents. In summary We successfully synthesised biobased TPA from furfural and furan using viable and proven organic synthetic procedures. Furthermore, the biobased carbon content of the TPA that we synthesised confirmed that it is a truly biobased product. Using furfural as a single resource is a novel and interesting concept, since furfural can be produced from inedible cellulosic biomass. The aim of this study was to propose a viable synthetic route from furfural alone to TPA, and we have succeeded in this. It is our hope that more research, conducted by both ourselves and, perhaps, other groups, will optimise this process so that it may be industrialised. We finally propose that the measurement of biobased carbon content is indispensable as a verification method in the research area of biobased synthesis."
} | 2,491 |
35727987 | PMC9245616 | pmc | 4,527 | {
"abstract": "Significance Advances in DNA sequencing have provided an unprecedented view of the complex microbial communities that populate global ecosystems. We present a metatranscriptomic analysis of samples from the boreal forest—the largest terrestrial carbon store—capturing the seasonally resolved transcriptomes of Norway spruce roots and more than 350 root-associated fungal species. Our findings link the functional response of host-trees to increased nutrient availability, with profound perturbations in the fungal community. Notably, we observed an exchange in prevalence and host-coordination of specialist ectomycorrhizal species critical for enzymatic cycling of recalcitrant carbon, to metabolically versatile species with resilient melanized cell walls. This research unites kingdom-spanning taxonomic and functional details of the boreal root microbiome, contributing a missing perspective toward modeling global carbon cycling.",
"discussion": "Discussion Nutrient availability is a critical factor limiting plant productivity and C storage in boreal forest soils. An evolutionary solution to nutrient limitation was the establishment of mutualistic associations, linking plant fitness with root-associated fungal activity. Like most economies, this exchange is sensitive to market forces that can alter the equilibrium of supply and demand ( 52 , 53 ). Increased nutrient availability, due to anthropogenic N deposition or higher rates of mineralization linked to global warming, is predicted to destabilize this trading status ( 21 ), although the severity and lasting consequences of this altered paradigm in boreal symbiosis remains unknown. At the study site used here, 25 y of NE conditions led to a fourfold increase in aboveground tree biomass compared to NL grown trees ( 54 ), which coincided with decreases in the flux of photosynthates directed belowground ( 55 ), and decreases in autotrophic and heterotrophic soil respiration ( 56 ) ( SI Appendix , Fig. S1 H ). We deployed a high-resolution metatranscriptomic analysis to consolidate these environmental observations with the underlying mechanisms impacted by this symbiotic restructuring. We observed functional evidence of host trees redefining the economic status of their relationship with the fungal community via reduction of sugar transporters that are integral to mycorrhizal establishment ( 57 , 58 ) ( SI Appendix , Fig. S6 B ), coupled with an enhanced defense response involving effector-triggered immunity, downstream signal propagation, and stress-associated transcription factors ( Fig. 3 , modules 2 and 12). The consequence of these alterations on the fungal community was profound, resulting in a reduced relative abundance of sequencing reads of fungal origin ( Fig. 1 A ), with a notable decrease in the proportion of reads assigned to ECM Basidiomycota in contrast to an increased proportion of melanotic ECM Ascomycota species ( Fig. 1 C ). The sensitivity and resolution of this approach revealed that these changes in community structure were accompanied by, and likely the result of, prominent shifts in broad-scale functional coordination between specific ECM fungi and the host tree ( Fig. 4 ). Perhaps most strikingly, we identified dramatic alterations in the coordination between predicted effectors of the three most prevalent fungal species and over 5,000 Norway spruce transcripts. This finding provides a mechanistic link between the broad structural and functional changes observed in the fungal community, with alterations in the complex molecular dialogue that coordinates the host with its prospective fungal partners. The response of a single tree to a confluence of effectors from competing fungal species is necessarily an emergent property ( 59 ), as individual microbes employ a repertoire of effectors (the “effectome”) that can function pleiotropically ( 60 ), redundantly ( 61 ), or in concert with each other ( 62 ) to affect generic or species-specific processes ( 63 ). Global analyses of environmental samples, such as the one described here, can capture these tree-level emergent properties in action, providing crucial insight into their complex in vivo activities and sufficient resolution to identify individual species within the ECM community driving these emergent responses. For Co. glaucopus and P. olivaceum , we observed the transcript abundance of a number of predicted effectors decrease drastically when the resource trading conditions were less favorable in NE conditions, particularly during the period of active fertilization (weeks 23 to 34). In contrast, although the number of Ce. geophilum effectors expressed in NE conditions was the same in NL conditions, the magnitude of their expression was greater in NE conditions ( Fig. 5 , Ce. geophilum heatmap). Taken together, this suggests the decoupling of P. olivaceum and Co. glaucopus from their host becomes most pronounced following fertilizer application, while the absence of such a response in Ce. geophilum effectors indicates this species is particularly well adapted to N-rich environments. To date, what little is known of fungal effector function is the result of detailed experimental characterization of individual effectors under controlled conditions. These approaches have provided insights into their diverse modes of action ( 64 ), their spatial and temporal kinetics ( 65 ), and the host processes commonly targeted ( 66 ). Many of the candidate effectors identified in this study possessed qualities linked to interference with protein posttranslational modification (Tyr phosphatase dual domain, protein kinase domain, and Tyr kinase active site), protein stability/folding (DnaJ, calreticulin/calnexin, and peptidyl-prolyl cis-trans isomerase FKBP2/11), or protein degradation (proteasome subunit-β5 and UBQ-conjugating E2). Such mechanisms have been widely described during phytopathogenic interactions ( 67 ), for example, during infection the Phytophthora sojae effector Avr1d competitively binds with a soybean E3 ubiquitin ligase, preventing its self-ubiquitination and degradation, leading to increased plant susceptibility to infection ( 68 ). Similarly, in Pseudomonas syringae , the effector HopBF1 phosphorylates plant HSP90 to inhibit the chaperone’s ATPase activity, which blocks its initiation of the plant hypersensitive response ( 69 ). We also identified five candidate effectors putatively involved in host transcriptional regulation [e.g., leucine zipper domain, Zn (2)-C6 fungal-type DNA-binding domain, and subunits of the Mediator complex and transcription initiation factor IID]. Compromised immunity arising from effector-mediated interference of host transcription has been described in several plant–microbe interactions, including mutualistic symbioses. For example, the interaction of the symbiosis effectors SP7 of Rhizophagus irregularis with the pathogenesis-related transcription factor ERF19 of Medicago truncatula , and MiSSP7 and MiSSP7.6 of Laccaria bicolor with the jasmonic acid transcriptional corepressor PtJAZ6 and the transcription factors PtTrihelix1 of Populus trichocarpa , respectively ( 16 , 70 , 71 ). Notably, we identified a number of effectors with potential targeting to mitochondria, two of which (one from Co. glaucopus and one from P. olivaceum ) possess a domain common to translocase complex proteins of the inner membrane (TIM), which has been linked to mitochondrial-mediated plant immunity in A. thaliana ( 72 , 73 ). Although mitochondria play a central role in the plant defense response, it is only recently that the effector proteins MoCDIP4 and Avr-Pita from Magnaporthe oryzae were shown to modulate rice susceptibility by targeting a mitochondrial DnaJ protein ( 74 ) and a cytochrome c oxidase ( 75 ), respectively, revealing a novel pathogen infection strategy targeting host mitochondria. Future efforts to experimentally validate candidate fungal effectors (such as those identified here) ( Fig. 5 and SI Appendix , Table S1 ) would be bolstered by the integration of these functional predictions with the foreknowledge of their expression dynamics, not only in regard to treatment and colonization-checkpoints, but in relation to the greater fungal effectome (including self and fungal competitors) and their coexpression with host functions. The reduction of total fungal reads in NE soils (consistent across growing seasons and methodologies) ( Fig. 1 A and SI Appendix , Fig. S2 B ) concurs with previous observations that increasing nutrient gradients or active fertilization in boreal soils decreases ECM mycelium and sporocarp production ( 76 , 77 ). Extramatrical mycelia represent a substantial proportion of total belowground C ( 78 ). However, beyond absolute fungal biomass, mycelial taxonomic composition has also been shown to influence the accumulation of C through the production of degradative enzymes ( 79 ) and the varying recalcitrance of fungal necromass ( 25 , 80 , 81 ). The Basidiomycota Cortinarius underwent the most dramatic reduction in assigned fungal reads in response to NE conditions ( Fig. 1 and SI Appendix , Fig. S2 ), coincident with a decoupling of broad-scale fungal and tree functions related to growth and solute transport ( Fig. 4 ), as well as highly specific predicted fungal effectors and tree functions associated with supporting mycorrhizal partners (e.g., sugar efflux transporters and symbiont-response processes) ( Fig. 5 ). This genus comprises well characterized ECM fungi with energetically demanding lifestyles that employ medium-distance fringe exploration strategies and secrete oxidative enzymes capable of degrading lignin ( 79 , 82 – 85 ). Due to the elaborate foraging strategies of Cortinarius spp., they have some of the highest C demands of all ECM ( 86 ) and are thus highly sensitive to reduced C flow from host roots ( 83 , 87 – 89 ). The capacity of members of the genus Cortinarius to degrade recalcitrant soil organic matter is unique among ECM fungi and they exert a disproportionate impact on soil C turnover ( 90 ). A recent study demonstrated the presence of Cortinarius species-complex members in the organic topsoil of boreal forests resulted in a 33% decrease in local C storage ( 90 ). This finding disputes the prevailing consensus that the functional redundancy inherent to complex microbial communities will offset the loss of an individual species. Thus, the collapse in symbiotic coordination and diminished presence of Cortinarius spp. in the fungal community observed in our study suggests that increased atmospheric N deposition and enhanced mineralization rates linked to warming soils will reduce the abundance and activity of these keystone degraders of lignified plant matter and humus in boreal soils. Contrasting this reduction in Basidiomycete abundance, NE resulted in increases in the Ascomycetes Ce. geophilum ( Fig. 1 ), accompanied by an intensification in transcriptomic coordination with the host tree, at both a broad-scale with fungal processes linked to metabolism, cell wall remodeling, and nutrient transport ( Fig. 4 and SI Appendix , Fig. S8 ), and a fine-scale involving a suite of predicted fungal effectors with putative functions, ranging from transcriptional regulation to protein ubiquitylation ( Fig. 5 and SI Appendix , Table S1 ). Ce. geophilum is a globally distributed and highly versatile short-range exploratory ECM fungus with distinctive black hyphae resulting from heavily melanized cell walls ( 91 , 92 ). This heavy melanization makes these fungi resilient to adverse environmental conditions, such as drought ( 93 , 94 ), and can act as virulence factors during root colonization ( 95 , 96 ). Notably, melanized necromass is also highly recalcitrant to decomposition ( 97 , 98 ), with the mycelia of Ce. geophilum demonstrated to persist in soils up to 10 times longer than that of other ECM species ( 92 , 99 ). This suggests that while NE resulted in reduced overall fungal biomass ( 31 ), it triggered an absolute increase in recalcitrant fungal necromass that is more likely to contribute to long-term C storage. This store of organic C would necessarily benefit microorganisms with the capacity to access it. Functional analysis of Ce. geophilum transcripts highly expressed in both nutrient treatments revealed a significant enrichment of chitinases and functions associated with polycyclic aromatic hydrocarbon degradation ( SI Appendix , Fig. S4 ) ( Ce. geophilum ), a cohort of functions that have previously been implicated in the degradation of environmental pollutants and, notably, melanins ( 100 ). These enzymes play essential roles in accessing stores of melanized fungal necromass and the self-digestion processes (autolysis) utilized for remodeling of mycelial architecture ( 80 ). The expression of these functions in both NL and NE conditions suggests that the capability to oxidatively degrade melanin could be a crucial function for Ce. geophilum , regardless of soil nutrient status, with the corollary that access to these growing reserves of recalcitrant C in NE soil has the potential to further reinforce this species’ position in the rhizospheric community. However, strategic experimentation will be needed to discriminate the overlapping roles these degradative enzymes play in mycelial autolytic remodeling and nutrient scavenging. As boreal forests harbor globally significant amounts of C, understanding the biogeochemistry of boreal soils is essential for informing strategic responses to future climate scenarios. One area where such understanding is needed was revealed by a recent metaanalysis of over 100 elevated CO 2 experiments ( 12 ) that has challenged the consensus of existing ecosystem modeling. Previous models had predicted increases in terrestrial C sequestration under elevated CO 2 conditions, while this metaanalysis suggests instead that the increased biomass of ECM-associated plants will be supported by augmented nutrient uptake and result in reduced soil organic C accumulation. Recent field experiments have shown that trees exposed to elevated CO 2 increased belowground C allocation, resulting in enhanced production of fine roots ( 101 ) and higher rates of soil CO 2 efflux ( 102 ). This alteration in belowground C allocation represents a revealing counterpoint to the impact of soil nutrient enrichment examined in the present study. CO 2 -driven increases in belowground C-availability could cause reciprocal changes in the dominance and activity of Basidiomycetes (such as Cortinarius and Piloderma species) at the expense of Ascomycetes (such as Ce. geophilum and Meliniomyces species). Potentially, this CO 2 -driven shift in community composition could result in a gradual increase in oxidative decomposition of soil organic matter ( 103 ) and a decrease in recalcitrant fungal necromass, triggering a reduction in the capacity of boreal forest soils to sequester C but an increase in tree biomass. This conflict between observed and modeled C cycling highlights the uncertainty of climate projections that underestimate the importance and scale of rhizospheric processes and emphasizes the need to monitor and integrate the diverse biogeochemical processes contributed by microbial communities and their host plants. Metatranscriptomic approaches, such as the one described here, can aid in functionally resolving these processes and will add to our conceptual framework of global C sequestration and ecosystem stability in future climate conditions. Notably, this approach advances studies of complex microbial communities from being purely taxonomic, using DNA-based methods, to identifying causative process-driven links and alterations in the transcriptional coordination of mycorrhizal associations. Additionally, minor user adjustments to this opensource and reproducible methodology will enable high-resolution monitoring of a range of host–microbe systems, from pathologies arising from dysbiosis in the human body, to applications in a diversity of threatened ecosystems."
} | 4,032 |
36469764 | PMC9897461 | pmc | 4,529 | {
"abstract": "Significance Phototrophic organisms provide most of the metabolic energy powering life on this planet. There exist two fundamentally different systems for harvesting light energy: (bacterio)chlorophyll-based photosynthetic complexes and proton-pumping rhodopsins. Different phototrophic groups use one system or the other. Here, using transcriptomics, infrared variable fluorescence, and flash photolysis measurements, we show that the bacterium Sphingomonas glacialis AAP5 isolated from an alpine lake is able to use xanthorhodopsin and bacteriochlorophyll-based photosystems simultaneously. Our results suggest that the possession of two light-harvesting systems may also be beneficial in other environments where organisms are exposed to extreme changes of light and low temperature.",
"discussion": "Discussion Here, we have demonstrated that the bacterium S. glacialis AAP5 is capable of dual phototrophy. It possesses all the necessary genes for BChl- and XR-based light energy conversion and, under appropriate physiological conditions, uses them both and even at the same time to generate a proton gradient across the membrane for the production of ATP. Thereby, the need to generate energy through aerobic respiration is reduced and saves available organic carbon for growth, which is scarce in oligotrophic aquatic ecosystems such as alpine lakes ( 31 ). This effect has been documented for both BChl-containing ( 32 – 34 ) and rhodopsin-containing photoheterotrophs ( 35 – 37 ). Yet, why does S. glacialis AAP5 possess two different phototrophic systems that basically fulfill the same function? Zeng et al. proposed that differences in the absorption spectra of rhodopsins and BChl a –based photosystems might explain the simultaneous genomic presence of both ( 20 ). The penetration of blue and green light differs between snow and ice and might favor one system over the other ( 20 ). We think that this hypothesis is not correct. While BChl a molecules absorb in the UV and IR parts of the spectrum, the photosystems contain the additional carotenoid spirilloxanthin, which absorbs blue–green light and transfers the energy to BChl a ( 21 ). XR only utilizes the visible part of the spectrum as it contains a green-absorbing rhodopsin and blue-absorbing nostoxanthin antenna ( Fig. 4 B ). Thus, the absorption spectra of XR and BChl a –based photosystems largely overlap in the visible spectrum ( Fig. 4 B and C ), and small changes of the spectral irradiance should not have any dramatic effect. BChl-based systems are large pigment–protein complexes, which require a complex machinery for its synthesis, assembly, and regulation ( 1 ). On the other hand, due to the large number of pigments, they can effectively operate even under low-light conditions. In contrast, XR has only two chromophores: retinal and nostoxanthin antenna. The utilization of simple xanthophyll nostoxanthin is unusual as the original XR from Salinibacter ruber contains a more complex glycosylated ketocarotenoid salinixanthin ( 30 ). While XR is much more straightforward to assemble, metabolic energy balance calculations suggest that rhodopsins are efficient only at higher irradiance ( 13 ). Another remarkable difference between XR and BChl-based photosynthesis is their contrasting regulation ( Fig. 6 A ). In a previous study, we documented that PGC expression was inhibited by the presence of light, glucose, or galactose ( 21 ). By contrast, XR is not inhibited by the presence of sugars, and the XR promoter is activated only at temperatures below 16°C, with the gene being fully transcribed only in the presence of light. The studied lake is an oligotrophic lake with temperatures usually lower than the determined threshold for XR activity for the whole year ( 9 ). Thus, it is likely that S. glacialis AAP5 cells contain and utilize both XR and BChl-based photosystems in its natural habitat. Fig. 6. Expression regulation of dual phototrophy in S. glacialis AAP5. ( A ) Proposed scheme of anoxygenic photosynthesis (AP) and XR regulation by nutrients, light, and temperature. ( B ) Hypothetical model of an alpine lake with the main environmental factors (temperature, irradiance, and snow cover) affecting the expression of dual phototrophy during the seasons. The thin black lines indicate ice cover. Cyt, cytochrome; QH 2 , ubiquinol. Physical conditions in alpine lakes periodically change from moderate temperatures and high light intensities in summer to cold temperatures and coverage by ice and snow in winter that alters the light spectrum and greatly reduces its penetration ( 38 ). The possibility to use two phototrophic systems with different photochemical properties and contrasting regulation may represent an adaptation strategy allowing for perpetual growth in this challenging, dynamically changing habitat. The counterbalanced regulation of BChl- and XR-based phototrophy may in particular help to tune the ratio of both systems to light intensity, day length, and temperature ( Fig. 6 B ). During short winter days, light reaching the lake water column is strongly attenuated by snow and ice, which is favorable for the PGC expression and synthesis of the BChl a –containing photosystems. In late spring to early summer, ice starts to melt, and the water is still cold especially at the surface, where it is in contact with ice ( 39 , 40 ). Photosystems probably cannot use the full potential of increasing light intensities as the cyclic electron transport is slower at these temperatures. Thus, XR expression becomes more advantageous. During long summer days, both systems work equally well. However, due to its simple biosynthesis, XR is the more cost-effective system ( 13 ). Reaching autumn, warmer water temperatures and shorter days with lower light influx might favor PGC expression again. This is consistent with the peak of BChl-containing bacteria in the GKS found in the middle of September ( 9 ). In addition to the natural seasonal cycle, light intensity may rapidly fluctuate during the day due to cloud cover changes ( 41 ), which may impact the transcriptional regulation of both photosystems differently. The constitutive high activity of the XR promoter at low temperatures and the light-dependent full transcription of the gene seem to be very costly regulatory strategies but could be beneficial due to a rapid reactivation in response to light. In contrast, reactivation of PGC expression in the dark takes hours ( 42 – 44 ), making its repression insensitive to fluctuations in light intensity. Are dual phototrophic organism present also in other environments? A survey of 215,874 bacterial genomes identified both PGC and rhodopsin genes in 55, mostly alphaproteobacterial, genomes. Almost half of them were of alpine or glacial origin ( 20 ), but they were also found in other environments. Recently, a BChl a –producing mesophilic Rhodobacter strain M37P was isolated from Yellowstone springs. It also contains an XR gene, but its expression has not yet been documented ( 45 ). BChl a –based phototrophy is inherited mostly vertically, ( 46 ) and distant horizontal transfers are very rare ( 47 ). In contrast, rhodopsin genes have been found in Bacteria, Archaea, Eukaryota, and even Viruses ( SI Appendix, Fig. S9 ) and are relatively easy to transfer horizontally ( 48 ). Thus, a BChl a –containing bacterium may eventually receive rhodopsin genes horizontally. This process may have occurred repeatedly during the evolution. However, whether these species retain and express the obtained rhodopsin gene will depend on the new genes providing a competitive advantage in a particular environment. Thus, dual phototrophy may also be beneficial in other environments with highly dynamic physicochemical conditions with extremes favoring one system over the other."
} | 1,957 |
38225698 | PMC10933641 | pmc | 4,530 | {
"abstract": "Abstract Memristors are regarded as promising candidates for breaking the problems including high off‐chip memory access delays and the hash rate cost of frequent data moving induced by algorithms for data‐intensive applications of existing computational systems. Recently, organic–inorganic halide perovskites (OIHPs) have been recognized as exceptionally favorable materials for memristors due to ease of preparation, excellent electrical conductivity, and structural flexibility. However, research on OIHP‐based memristors focuses on modulating resistive switching (RS) performance through electric fields, resulting in difficulties in moving away from complex external circuits and wire connections. Here, a multilayer memristor has been constructed with eutectic gallium and indium (EGaIn)/ MAPbI 3 /poly(3,4‐ethylenedioxythiophene): poly(4‐styrenesulphonate) (PEDOT: PSS)/indium tin oxide (ITO) structure, which exhibits reproducible and reliable bipolar RS with low SET/RESET voltages, stable endurance, ultrahigh average ON/OFF ratio, and excellent retention. Importantly, based on ion migration activated by sound‐driven piezoelectric effects, the device exhibits a stable acoustic response with an average ON/OFF ratio greater than 10 3 , thus realizing non‐contact, multi‐signal, and far‐field control in RS modulation. This study provides a single‐structure multifunctional memristor as an integrated architecture for sensing, data storage, and computing.",
"conclusion": "3 Conclusion In summary, we reported single‐structure multifunctional memristors with EGaIn/MAPbI 3 /PEDOT:PSS/ITO sandwich‐like structure. Under ambient conditions, the memristors successfully presented robust RS behaviors, including low SET/RESET voltages (+0.56 V/−0.87 V), stable endurance (5 × 10 3 cycles), ultrahigh electrical and acoustic ON/OFF ratios (10 4 and 10 3 ), and long data retention time (5 × 10 4 s). Moreover, the RS model, originating from the formation/diffusion of CFs derived from V I s migration, was proposed based on direct observations and compositional analysis of the nanoscale conductive regions. The presence of the polar molecule MA + at the center of the cage brought the directional disorder and polarization, accounting for the piezoelectricity of MAPbI 3 (d 33 = 3.44 pm V −1 ). Consequently, piezo‐acoustic RS effects can tune the evolution dynamics of V I s by inhibiting their formation or accelerating their annihilation, eventually switching the memristor to HRS. The underlying principles related to the frequency and SPL dependences of the piezo‐acoustic‐RS effects were investigated in depth. This work provides an attractive idea for the design and fabrication of an integrated architecture for sensing, data storage, and computing within a single memristor.",
"introduction": "1 Introduction The continuous evolution of a smarter world encompassing neuromorphic computing, [ \n \n 1 \n \n ] cloud robotics, [ \n \n 2 \n \n ] and the Internet of Things (IoT), [ \n \n 3 \n \n ] demands comprehensive, and inevitable innovations of the electronic memories that are currently used on a large scale. However, traditional memory devices, e.g., static random‐access memory (SRAM), dynamic random‐access memory (DRAM), and flash, are approaching their physical limits when scaled down due to enhanced quantum tunnel‐induced charge leakage at the 10‐nanometer scale. Consequently, it is difficult to shrink their sizes to increase storage density storage in a bit‐cost scalable manner. Moreover, modifying the MOS (Metal‐Oxide‐Semiconductor) technology as well as optimizing memory operation to overcome these challenges and achieve further upgrades are also quite difficult. [ \n \n 4 \n \n ] As a solution to challenges posed by new non‐volatile memory (NVM) storage devices, Resistive Random‐access Memory (RRAM) has garnered attention for meeting the demands of efficiently handling massive data. Due to its rapid switching speed, ultra‐large‐scale‐integration (ULSI) densities, simple structure, and low power consumption, memory technology experts believe that RRAM based on the memristors is the optimal choice in the mass memory market. This selection aims to streamline production costs by focusing on a select few technologies. [ \n \n 5 \n \n ] \n Memristors are two‐terminal devices with an electrode/active material/electrode capacitor‐like structure. They have received enormous attention for data storage applications due to their advantages in simplistic structure, high integration density, fast switching speed, low power consumption, and high density. [ \n \n 6 \n , \n 7 \n \n ] The SET (switching the device from high resistance state to low resistance state, LRS) and RESET (switching the device from low resistance state to high resistance state, HRS) processes correspond to the information storage operation of write and erase, respectively, and can be obtained by the external stimulation with varying voltage amplitude, bias polarities, and other stimuli. The formation and rupture of conductive filaments (CFs) is a common form of resistance change. [ \n \n 8 \n , \n 9 \n \n ] To date, a variety of materials have been employed to construct the active material including polymers, [ \n \n 10 \n \n ] sulfide, [ \n \n 11 \n \n ] graphene oxide, [ \n \n 12 \n \n ] and chalcogenide. [ \n \n 13 \n \n ] Devices based on these materials have made full progress in the field of RRAM and artificial synapses. [ \n \n 14 \n , \n 15 \n \n ] However, most of them have complex structures and cumbersome preparation processes, causing problems such as increased manufacturing costs, decreased production efficiency, and decreased production yields. Recent research indicates that the development of halide perovskites (HPs) memristors holds the potential to improve current semiconductor process integration and lead to a reduction in operating power consumption. As the core unit to obtain high‐performance memristors, the HPs are commonly the compound with the stoichiometric formula ABX 3 (A:, e.g., formamidinium (FA + ), methylammonium (MA + ), Cs + , B:, e.g., Pb 2+ , Sn 2+ ; X:, e.g., Cl − , Br − , I − ), which assembles in a lattice with the coordination numbers 12 for A, 6 for B, and 8 for X. [ \n \n 16 \n \n ] Due to unique crystal structure of organic–inorganic HPs (OIHPs), they have drawn tremendous attention for excellent optical absorption, low exciton binding energy, long electron‐hole diffusion, high defect tolerance, and structural and compositional flexibilities. Given these merits, OIHPs have been widely employed in photodetectors, light‐emitting diodes, solar cells (SCs), and so on. [ \n \n 17 \n , \n 18 \n \n ] Although OIHPs SCs displayed superior optical and electrical properties to traditional semiconducting materials, the development is somewhat hindered by the hysteresis during I–V scans. It is speculated that this phenomenon should be closely related to ferroelectricity and ion migration, providing the basis for the realization of memristors. [ \n \n 19 \n \n ] As a prominent member of OIHPs, MAPbI 3 has been observed to exhibit a piezoelectric response at room temperature in theoretical calculations and experiments. The asymmetry of MA + would result in a pseudo‐cubic structure rather than an inversely symmetric one. [ \n \n 20 \n , \n 21 \n \n ] Therefore, the absence of an inversion center makes self‐polarization particularly pronounced. Consequently, with the 4 mm point group and I4/mcm space group for the MAPbI 3 at room temperature, the measured effective piezoelectric coefficient d 33 can exceed 5 pm V −1 , significantly increasing to 25 pm V −1 under illumination due to larger light‐induced dipole moments of MA + . [ \n \n 22 \n , \n 23 \n \n ] Combining an excellent piezoelectric effect with high tolerance to substrates, OIHPs emerge as a focal point for potential applications in acoustic sensors and transducers. Furthermore, they offer new insights into piezoelectric modulation of ion migration in MAPbI 3 film. Moreover, owing to its ion mobility phenomenon, OIHPs exhibit resistive switching (RS), making them highly attractive for emerging RRAM. [ \n \n 24 \n \n ] Building on this foundation, the modulation of ion migration through other physical fields to bring about changes in the resistance state facilitates achieving non‐contact, far‐field RS modulation. Novel memristors based on piezo‐acoustic RS effects can be obtained based on accurate control and optimization. As we all know, the OIHPs memristors solve the problems of existing computational systems based on Von Neumann architecture such as high off‐chip memory access delays and inefficient algorithms for data‐intensive applications. [ \n \n 25 \n \n ] The OIHPs memristor can avoid the processes of module sampling and transmission to the processing unit after acquiring external information. Previously, various electric field‐driven OIHPs memristors were gradually applied in the fields of multilevel storage, neural computing, hardware security, and so on. [ \n \n 26 \n \n ] Electric‐field‐driven OIHPs memristors have difficulties in moving away from complex external circuits and wire connections. By utilizing the excellent photonic response, optogenetics‐inspired tunable synaptic functions, photo‐induced logic gate devices, and artificial retina systems have been achieved by the perovskite films. [ \n \n 27 \n \n ] The potential drawbacks of optical control in OIHPs memristors include complex equipment requirements, such as lasers, optical lenses, and photodetectors, sensitivity to ambient light and temperature, relatively high energy consumption, and a need for precise alignment. These factors may limit the applicability of optical control and increase the complexity of the system. Moreover, magnetic fields also provide an opportunity to control perovskites‐based memristors in a remote way and environmentally robust devices capable of operating at high temperatures have been prepared. The magnetic field may be influenced by surrounding elements such as metals and other magnetic materials in the environment, introducing potential instability in certain practical applications. Compared to other field‐modulated memristors, acoustic memristors can operate in complex environments with significantly simplified circuits, showcasing superior adaptability. By adjusting acoustic parameters, memristor control can be achieved without the need for high‐energy devices. This will also greatly improve the efficiency of the system and complete the construction of an integrated architecture of sensing, storage, and computing. In this work, we systematically studied the piezo‐acoustic RS effect based on the MAPbI 3 memristors under ambient conditions. The single‐structure multifunctional memristors with eutectic gallium and indium (EGaIn)/MAPbI 3 /poly(3,4‐ethylenedioxythiophene): poly(4‐styrenesulphonate) (PEDOT:PSS)/indium tin oxide (ITO) sandwich‐like structure have been successfully fabricated, where the MAPbI 3 RS layer was synthesized by a low‐temperature all‐solution process. The devices exhibited reproducible bipolar RS behavior with low SET/RESET voltage (+0.56 V/−0.87 V), stable endurance (5 × 10 3 cycles), much higher electrical or acoustic ON/OFF ratio (10 4 /10 3 ), and excellent data retention property (5 × 10 4 s), revealing prominent RS characteristics along with stability in air. Based on our principle design and experimental verifications, the devices possessed reproducible acoustic‐HRS which is speculated to be due to the piezo‐acoustic‐RS effect of MAPbI 3 through the sound waves. A physical model was proposed to inspire the understanding and applications of the piezo‐acoustic‐RS behavior. Based on the multi‐signal, and far‐field control in RS modulation, the MAPbI 3 ‐based memristors could be a promising candidate for the future construction of an integrated architecture of sensing, storage, and computing.",
"discussion": "2 Results and Discussion For details about the schematic drawing of the memristors, the atomic force microscopy () images, the scanning electron microscopy (SEM) images, the X‐ray diffraction (XRD) patterns, the Raman spectra, the optical absorption spectra, and the photoluminescence (PL) spectra of the samples, see Figures S1–S7 (Supporting Information). The selection of EGaIn material as the top electrode offered the advantage of requiring no additional annealing processing, thereby avoiding significant mechanical or thermal damage to the RS layer. The adaptive deformation capability of the EGaIn electrode enabled the establishment of a stable and tight electrode‐perovskite interface. This adaptability not only helped maintain the integrity of the RS layer but also enhances the reliability under different operating conditions. Due to its low‐temperature solution processability, excellent flexibility, high stability, and remarkable film‐forming ability, the PEDOT:PSS emerged as an ideal choice for crafting uniform and crystalline MAPbI 3 RS layers. Then, the prepared MAPbI 3 ‐based memristors were conducted systematical electrical characterizations. Generally, the FORMING process was necessary to initiate the memory cells to achieve consistent RS behaviors ( Figure \n 1 a ). A high voltage was employed to induce the generation of vacancy defects and to control the distribution of vacancies for subsequent processes. A positive cyclic sweep (0 V→+6 V→0 V) was applied to complete the FORMING operation and the switching from initial resistance state (IRS, 50.21 MΩ) to LRS (84.60 Ω) occurred at a voltage of ≈+3.32 V (forming voltage), named V forming . It has been proved that the RS behavior of a MAPbI 3 ‐based memristor is purely associated with the formation and rupture of the CFs rather than the changes in interface valence. [ \n \n 28 \n \n ] Afterward, the I–V characteristics were collected by applying DC voltage sweeps with a forward SET stop voltage of +3 V and a reverse RESET stop voltage of −3 V. As shown in Figure 1b , the I–V curves, illustrated in semilogarithmic scale, exhibited excellent bipolar RS behaviors under a compliance current (I CC ) of 10 mA. A positive voltage sweeping from 0 to +3 V applied to the top electrode resulted in a sharp current increase and the device was switched from the HRS to the LRS, referred to as the “writing” or SET process. With the voltage scan from +3 to 0 V, the resistance value tended to be constant, proving the non‐volatile feature. For voltage sweep from 0 to −3 V, the resistance slowly increased to the HRS, termed as “erasing” or RESET process. Subsequently, no significant fluctuation had been observed in the switching voltages and resistance for five cycles, suggesting a high reproducibility. RS power consumption of SET process (36.9 mW) and RESET process (1.9 mW) was obtained from I–V curves. Additionally, to confirm the uniformity of the memristors, the I–V measurements were carried out at five randomly selected areas on the sample. Despite that the RESET voltages and ON/OFF ratios were slightly different, all the devices maintained stable RS operations (Figure S8 , Supporting Information). To investigate the switching speed, transient measurements were performed, as shown in Figure S9 (Supporting Information). The tests confirmed that the device has a switching speed of ≈340 ns, much faster than flash memory devices (order of µs). The memristors are competitive with the conventional flash memory and NVMs in switching speed. [ \n \n 29 \n \n ] \n Figure 1 RS properties of the memristors. a) Typical FORMING process. The inset shows the re‐plotted FORMING process of the logarithmic scale of Y. b) Series I–V behaviors of the device. c) statistics of the forming, set, and reset voltage distributions depicted as a box‐whisker plot for 50 consecutive cycles. d) Fitted logarithmic I–V behavior of positive voltage sweep. e) Endurance test for 5 × 10 3 cycles. f) Retention test for 5 × 10 4 s. The switching voltages had been collected to investigate the statistics characterizations of the RS performance. The memristors can maintain the switching voltages for more than 50 consecutive cycles (Figure 1c ). The average values of the V forming , V set , and V reset were +3.32, −0.56, −and 0.87 V, respectively, superior to most memory devices. [ \n \n 30 \n \n ] Low switching voltage, originating from the low activation energy (E a ) of iodine vacancies (V I s) in the CFs‐dominated memristors, was a primary parameter for the potential applications in low energy‐consuming devices. The mechanism was systematically discussed in the following section. Basically, the data was stored using two/more very distinct resistance levels assumed by the reversible soft breakdowns. The completion of RS behavior in MAPbI 3 ‐based memristors was determined by states of CFs which can be modified by ion migration. Previous research revealed that the MAPbI 3 layer contained defects, such as vacancies, interstitials, cation substitutions, and anti‐site substitutions, that work as charge‐trapping centers. [ \n \n 31 \n , \n 32 \n \n ] Among them, V I s were the most active ones in MAPbI 3 due to their relatively low level of E a (≈0.08 eV). Usually, the ion migration rate ( k ) in solids is estimated from the Arrhenius Equation 1 . [ \n \n 33 \n \n ] \n \n (1) \n k = A e − E a / K B T \n where A is the pre‐exponential factor, E a \n is the activation energy, K B \n is the Boltzmann constant, and T is the absolute temperature, respectively. Due to the high E a of V Pb s (0.80 eV) and V MA s (0.46 eV), their migration rates were 1.2 × 10 0 and 6.5 × 10 5 s −1 , much lower than the migration rate of V I s (1.7 × 10 12 s −1 ). [ \n \n 34 \n \n ] Therefore, the lowest E a of V I s corresponded to the highest migrating rate. Under external stimulations, migrations of V I s were so readily that affected the transportation of electrons quite significantly in the close‐packed structure, which generated conductive paths by positively‐charged V I s. The FORMING and SET processes were often accompanied by migrations of V I s under an electric field. During the forward sweep, the iodide ions hopped toward the top electrode while V I s moved to the bottom electrode and formed the CFs gradually. Once the conductive paths of electrons formed, electrons transported along with the CFs and resulted in a significant resistance decrease and converted the device to LRS. Since the SET process occurs on pre‐formed CFs, it was possible to obtain a lower V set (≈0.56 V) than V forming (≈3.32 V). [ \n \n 35 \n \n ] On the contrary, I − s can also migrate back to the perovskite interior under the reverse voltage sweep during the RESET process, which could recombine with V I s and accelerate the annihilation process. As a result, the CFs were gradually dissolved by the ion migration and the current gradually decreased along with the disappearance of the conductive paths. According to the previous reports, [ \n \n 36 \n \n ] self‐limited behaviors might be responsible for the fluctuation of resistance. That is to say, the operation variability of the device was generally caused by random growth and dissolution of CFs. However, this phenomenon during the RESET process mentioned above may be caused by a partial rupture of the multiple CFs. [ \n \n 37 \n \n ] \n The I–V characteristics of the devices under positive voltage sweeps were drawn in the logarithmic form to interpret the switching mechanism, as depicted in Figure 1d . For the HRS, the ohmic and space charge limited current (SCLC) conduction mechanisms dominated before the switching. At a low voltage bias (0−0.46 V), the slope of 1.03 was indicative of a linear ohmic conduction region ( I ∝ V ), where the number of free carriers generated by thermal fluctuation exceeds the injected carriers. In the high bias region (0.46−0.73 V), all the traps were filled, and the injected carriers dominated over thermally generated ones, corresponding to the SCLC conduction. This classical trap‐controlled SCLC relationship ( I ∝ V \n 2 ) can be described by Equation 2 . [ \n \n 38 \n \n ] \n \n (2) \n J = 9 8 n ε μ V 2 d 3 \n where J , n , ε, μ, V , and d are the current density, free carrier concentration, dielectric constant, electronic mobility, biased voltage, and thickness, respectively. Furthermore, inside the perovskite layer, SCLC was a current mechanism closely related to the traps consisting of vacancies, especially V I s of the lowest E a \n . [ \n \n 39 \n \n ] Subsequently, the current increased dramatically by switching HRS to LRS at 0.73 V. In contrast, ohmic behavior consistently dominated in LRS that consistent with the CFs formation as RS mechanism. [ \n \n 40 \n \n ] As shown in Figure S10 (Supporting Information), the data in the RESET process exhibited a similar mechanism to the SET process. To verify the reproducibility of the MAPbI 3 ‐based memristors, the cycling endurance measurement was carried out under a read voltage (−0.1 V, 50 ms). As shown in Figure 1e , the endurance property was obtained by applying the continuous SET/RESET stop voltage pulse of +3.0 V/−3.0 V, 50 ms. We have included statistical measures such as mean, standard deviation (SD), and coefficient of variation (CV) of endurance data in Table S1 (Supporting Information). Additionally, we have incorporated a cumulative probability plot of endurance data. As shown in Figure S11 (Supporting Information), uniform switching with low variability in the narrow distribution and well‐separated HRS and LRS were obtained in the endurance. Under the ambient conditions, the memristors can operate for 5 × 10 3 cycles with fairly stable resistance states during the switching processes. For the beginning 1.3 × 10 3 cycles, the HRS had a window between ≈100 and ≈10 MΩ achieving a staggering average ON/OFF ratio of ≈10 4 . Nonetheless, in subsequent cycles, the HRS window dropped to between ≈20 and ≈500 KΩ, which resulted in a decrease in the ON/OFF ratio. The drop of the HRS window (cycling‐induced degradation) might be triggered by the excessive accumulation of V I s. [ \n \n 41 \n \n ] The concentration of local vacancies had a significant impact on diminishing HRS compared to the intrinsic state as they always promote the reduction of E \n a and the formation of partial rupture of the multiple CFs (the augmentation in residual CFs). [ \n \n 42 \n \n ] Figure 1f showed the retention distributions (readout voltage: −0.1 V, 50 ms) of the devices with a time of ≈5 × 10 4 s. Note that the retention level was kept at the same level without obvious deviation and the devices maintained a decent ON/OFF ratio of over 1.5 × 10 4 . In addition to the previous evidence of the RS mechanism, direct observation of V I s CFs during the switching operation was also conducted by cross‐sectional SEM. Figure \n 2 a showed the SEM image of an as‐fabricated device where the MAPbI 3 layer with a thickness of ≈170 nm, corresponding to the region a in I–t curve (Figure 2g ). CFs assembled by V I s were formed in response to the forward sweep (0 V→+6 V→0 V), and unimpeded electron channels led the memristors to enter LRS state in Figure 2b . This was in accord with the features of CFs observed at region b in Figure 2g . The shape of the CFs resembled that of lightning, and it was also noticed that the diameter of the filament close to the bottom electrode side (∼44.5 nm) was wider than that close to the top electrode side (∼26.7 nm). The morphology of the CFs is also consistent with other research. [ \n \n 43 \n \n ] When the negative bias had been applied to the top electrode, the recombination of V I s with I − s was promoted (region c in Figure 2h ). Despite the apparent rupture of the CFs, the residual vacancies still formed partial CFs (Figure 2c ). According to Figure 2h , the current dropped gradually and we assumed that the CFs broke from the top to bottom. Eventually, there would be residual CFs retained. [ \n \n 44 \n \n ] Moreover, due to the partially formed CFs, the resistance value of HRS was generally not as high as that of IRS, which also corresponded to our experimental results. It can be demonstrated that the devices achieved RS based on the formation and breaking of CFs. Figure 2 SEM and EDS analysis of the formation and rupture of CFs in MAPbI 3 perovskite layer. a) The pristine state, corresponding to IRS. scale bar, 100 nm. b) The state after FORMING (SET) process, corresponding to LRS. scale bar, 100 nm. c) The state after the RESET process, corresponding to HRS. scale bar, 100 nm. d) The cross‐sectional SEM image containing the CFs region and non‐CFs region. Point 1: CFs. Point 2: non‐CFs. e) Schematic illustration of the CFs morphology in Point 1. The inset table shows the weight and atomic percentage of the I and Pb elements in Point 1. f) Schematic illustration of the non‐CFs morphology in Point 2. The inset table shows the weight and atomic percentage of the I and Pb elements in Point 2. g) Corresponding I–t curve during the forward voltage sweep that led to the image in a and b. (h) Corresponding I–t curve during the reverse voltage sweep that led to the image in c. i) Line profile of the EDS intensity of I, C, and Pb elements along the yellow line in the inset from top to bottom. Inset: CFs region of the corresponding line profile of the EDS. Understanding the RS behavior is critical to device optimization, and in particular, whether I − s migration is involved. Energy dispersive X‐ray spectroscopy (EDS) measurement provided the necessary elemental composition information. Considering the conclusion that the Pb element was difficult to migrate while the I element was easy to migrate, we selected point 1 and point 2 for EDS (Figure 2d ). [ \n \n 45 \n \n ] A schematic illustration of the CFs region (Figure 2e ) and non‐CFs region (Figure 2f ) was provided to clarify the RS dynamic processes. The specific EDS collection positions were displayed in the dotted red box. Interestingly, EDS results of Points 1 and 2 showed a clear distinction of the weight and atomic percentage of the I and Pb elements. As illustrated in the inset tables of Figure 2e,f , the weight and atomic percentages of I − s in Point 1 are 56.6% and 68.0%, respectively, compared with 69.1% and 78.5% in Point 2. These differences revealed that the CFs region formed by the excess V I s existed in the film. [ \n \n 46 \n \n ] Subsequently, we conducted an EDS line‐scan across the CFs region. It was also observed that the EDS line‐scan results revealed a decrease in I − s content with increasing distance from the electrode. From the 1st measurement point to the 11th, the weight percentage of the I− − s decreased by 30.9% unexpectedly. Additionally, the I − s at the 10th measurement point site was 34.8%, compared with 35.0% at the 11th measurement point. This difference can be attributed to the fact that there may be fewer defects in the preparation process near the 10th site, which will reduce the ion mobility near this point and lead to the reduction of V I s concentration. [ \n \n 47 \n \n ] Here we reported the direct observation of V I s CFs via SEM imaging combined with compositional EDS analysis of the nanoscale CFs, which provided critical insight into the complex RS dynamic mechanisms. According to previous reports, the displacement of positive and negative electric charge will lead to an electric polarization inside the MAPbI 3 under pressure and produce the electric charge by the deformation. [ \n \n 48 \n , \n 49 \n \n ] In the pursuit of a better understanding of piezoelectric properties, effective longitudinal piezoelectric coefficients (d 33 ) were extracted from the V ac ‐dependent piezo‐responses measurements. As shown in Figure \n 3 a , the d 33 of the MAPbI 3 films was ≈3.44 pm V −1 . The spontaneous polarization of MAPbI 3 may originate from the orientational polarization of A + dipole, ionic polarization induced by displacements of the positive charge center of MA + relative to the negative charge center of the PbI 3 cages, and ionic polarization induced by the off‐center displacement of Pb 2+ in the PbI 6 octahedron. [ \n \n 50 \n \n ] Fan et al. reported that Pb 2+ showed only 0.01 Å off‐center motion in the PbI 6 octahedron, which was quite weak for spontaneous polarization. [ \n \n 51 \n \n ] The presence of the polar molecule MA + at the center of the cage, creating directional disorders, was a more important factor that accounts for a major portion of the polarization. Structurally, MA + (C 3v point groups) would result in the highly symmetrical pseudo‐cubic lattice, causing MAPbI 3 to lose its inverse symmetry. This asymmetry of MA + meant that an inversion center was absent, leading to particularly pronounced self‐polarization. [ \n \n 52 \n \n ] Subsequently, when the MAPbI 3 was strained by an external force, the internal asymmetric center was further shifted, leading to the destruction of inverse symmetry. Finally, the positive and negative charges would be equally concentrated on the crystal surface, resulting in the macroscopic generation of a built‐in electric field that influenced ion migration and accumulation. [ \n \n 53 \n \n ] Hence, utilizing non‐contact acoustic signals to modulate ion migration became an idea to achieve RS behavior. Figure 3 Measurement of piezoelectric coefficient and piezo‐acoustic‐RS. a) The effective piezoelectric coefficient measurement results. b) Reduction of device current stimulated by sound waves of the EGaIn/MAPbI 3 /PEDOT:PSS/ITO memristor. Inset shows the structure used to measure piezo‐acoustic‐RS. The fabricated MAPbI 3 ‐based memristors were attached to the front of a sound speaker employed to play the sound waves at various frequencies. As expected, the acoustic waves can regulate the RS behaviors due to the piezoelectric effects. The inset of Figure 3b presented the schematic illustration of the configuration used to measure piezo‐acoustic‐RS under ambient conditions. In Figure 3b , when a sine acoustic wave with a sound pressure level (SPL) of 90 dB was generated at 0.954 s, the current of memristors suddenly increased. It should be noted that this phenomenon was substantially different from other methods proposed to improve MAPbI 3 memristor performance, such as reducing V forming and V set through external stimulation assistance, which still depended on the bias voltage. [ \n \n 54 \n \n ] In contrast, the piezo‐acoustic RS produced an irreversible and permanent HRS independently, which did not need to be combined with voltage at all. To understand the RS effect, we investigated the frequency responses of the memristors at 15, 75, 150, 750, and 1500 Hz. As shown in Figure \n 4 a,b , the piezo‐acoustic‐RS effect was indistinct, and the RS window was imperceptible with an electric‐SET voltage of +3 V and stimulation of 15 and 75 Hz sine sound waves. Therefore, the small average ON/OFF ratio (≈10 1 ) and continuous piezo‐acoustic‐RS failure increased the bit error rate of the storage device. At 150 Hz (Figure 4c ), there were still many piezo‐acoustic‐RS failures during the cycles although the average ON/OFF ratio increased to ≈10 2 . In contrast, the phenomenon was well demonstrated under the stimulations of 750 and 1500 Hz signals (Figure 4d,e ). The average ON/OFF ratio reached an impressive value of ≈10 3 under 750 Hz, significantly reducing piezo‐acoustic‐RS failures. To further explore this phenomenon, the statistical distribution of acoustic‐HRSs was investigated in a box‐whisker plot, as shown in Figure 4f . The average value of the HRS can reach an astonishing 179.7 KΩ when stimulated by 750 Hz sound waves, providing a satisfactory window to obtain superior error tolerance. [ \n \n 55 \n \n ] \n Figure 4 Frequency dependence of the acoustic‐HRSs. a–e) Frequency dependence of the acoustic‐HRSs in successive electric‐SET and acoustic‐RESET cycles. (a) 15 Hz. (b) 75 Hz. (c) 150 Hz. (d) 750 Hz. (e) 1500 Hz. f) Frequency‐dependent statistical distribution of acoustic‐HRSs. (g‐k) Evolution of the device resistance stimulated by different frequency sound waves. g) 15 Hz. h) 75 Hz. i) 150 Hz. j) 750 Hz. k) 1500 Hz. l) Retention test result at different frequencies during 5 × 10 4 s. Longitudinal sound waves transmitted through the ambient air and caused local compression in the memristors. [ \n \n 56 \n \n ] When stimulated by acoustic waves, the memristors were subjected to compressive forces that brought deformations in the sample. In reaction, a piezoelectric potential was created between the two electrodes. [ \n \n 57 \n \n ] The polarization charges generated by piezoelectric potential contribute to the migration of I − s to balance this potential. Low‐frequency sound waves made it difficult to promote the recombinations of V I s and I − s and brought about small changes in resistance and even RS failure. Under the external sound pressure, each different film should correspond to different sound frequencies suitable for regulating ion migration. Stronger piezo‐acoustic‐RS will occur when the driving frequency of external forced vibration is close to this frequency. Therefore, the film would vibrate violently, and recombination of excess V I s with I − s would be maximum, promoting the annihilation process of the V I s and enhancing piezo‐acoustic‐RS behavior. Continuous RS behaviors were more pronounced at 750 Hz than that at 1500 Hz, which provided a higher volume of energy density for realizing the rupture of V I s CFs. [ \n \n 58 \n \n ] The closer the sound frequency is to 750 Hz, the higher the success rate and degree of CFs fractures are. Suitable sound frequencies for regulating ion migration will result in a stable RS effect. [ \n \n 59 \n \n ] The I–V characteristics of the devices under negative voltage (0 V→−3 V) without the sound, and with the sound at 15, 75150, 750, and 1500 Hz of 90 dB were drawn in the logarithmic form to show the trap‐filled limit voltage ( V TFL \n ), as depicted in Figure S12 (Supporting Information). The current increased from the linear ohmic region through a trap‐filled limit, then eventually to the quadratic Child's region along with the increasing bias voltage. Moreover, V TFL \n was calculated the trap density ( n trap \n ) by the Equation 3 : [ \n \n 60 \n \n ] \n \n (3) \n n t r a p = 2 ε ε 0 q L 2 V T F L \n where n trap \n , q , L , ε, ε 0 , and V TFL \n denoted the trap density, elemental charge, MAPbI 3 layer thickness, dielectric constant, and vacuum permittivity of perovskite film, trap‐filled limit voltage, respectively. Therefore, n trap \n is related positively to V TFL \n . The V TFL \n of MAPbI 3 ‐memristors without the sound and with the sound at 15, 75150, 750, and 1500 Hz of 90 dB were 1.581, 1.499, 1.297, 1.216, 0.932, and 1.054 V, respectively. The MAPbI 3 ‐memristor with sound at 750 Hz has a smaller V TFL \n than the others, suggesting that the device has fewer defect states, such as the V I s. [ \n \n 61 \n \n ] This also provided evidence for a reduction in the number of V I s caused by sound. To further investigate the piezo‐acoustic RS behaviors of the memristors and their impact on the switching performances, we studied the piezo‐acoustic‐RS speed at different acoustic frequencies. As illustrated in Figure 4g–k , the sound waves applied to the memristors were 90 dB at 15, 75, 150, 750, and 1500 Hz, respectively. Moreover, with the sampling frequency set at 25 Hz, the piezo‐acoustic‐RS speed was highly dependent on the frequency. At 15, 75, and 150 Hz, the RS speed was ≈0.28, ≈0.12, and ≈0.08 s, respectively (Figure 4g–i ). [ \n \n 62 \n \n ] After that, the device reached stable HRSs (≈5 × 10 2 , ≈6 × 10 3 , and ≈2 × 10 4 Ω, respectively). As depicted in Figure 4j,k , the memristor achieved piezo‐acoustic‐RS in less than 0.04 s at 750 and 1500 Hz, heading to achieve stable HRSs. The relationship between the piezo‐acoustic‐RS speed and the sound frequency showed from the side that it was indeed the sound waves that caused the RS behavior. To provide further evidence, piezo‐acoustic‐RS speed under sound waves between 25 and 65 Hz in steps of 10 Hz was studied in Figure S13 (Supporting Information). The piezo‐acoustic‐RS speed at 25, 35, 45, 55, and 65 Hz was 0.28, 0.28, 0.20, 0.20, and 0.12 s, respectively, clearly demonstrating the high‐frequency dependence. Ion mobility can be suppressed by the strain, and higher frequency acoustic signals may provide a higher probability for vacancy annihilation. [ \n \n 63 \n \n ] Therefore, the higher sound frequency in a certain range, the easier it was to promote the CFs rupture. High‐frequency sound waves would squeeze the MAPbI 3 more vigorously, producing a more effective piezoelectric effect. In this case, the ion migrations were suppressed, and the recombinations of ions and vacancies were accelerated, resulting in a shorter CFs rupture time. Additionally, the E a of V I s enhanced by a significant reduction of V I s under the compressive strain area, meaningful for the rupture of CFs and the transition to HRS. [ \n \n 64 \n \n ] It is obvious to find that the switching speed of non‐volatile piezo‐acoustic‐RS and acoustic‐HRS tend to be affected by frequency variations. [ \n \n 65 \n \n ] \n The retention performances were obtained under the ambient condition to further evaluate the nonvolatile properties of piezo‐acoustic‐RS. Not surprisingly, the acoustic‐HRSs at different frequencies were found to be stable without any detectable degradation over 5 × 10 4 s with a constant readout voltage of −0.1 V, 50 ms, as schematically illustrated in Figure 4l . The acoustic‐HRSs induced by different frequencies are stable and show a narrower distribution (Figure S14 , Supporting Information). The well‐maintained acoustic‐HRSs at different frequencies indicated that the regulation of piezo‐acoustic‐RS to V I s CFs was stable, laying a foundation for long‐term data storage. This phenomenon provided potential evidence for the RS behavior dominated by ion migrations under the external stimulus. In contrast to electrical retention, acoustic retention results induced a float in the resistance due to non‐uniformity and complexity CF caused by the vibrational motion of sound waves. In order to investigate SPL dependence of the acoustic‐HRSs, a commercial loudspeaker and an adjustable amplitude were employed as acoustic sources. As depicted in Figure \n 5 a–e , the dual‐dimensional hybrid RS cycle results were obtained by the electric‐SET and acoustic‐RESET to show the SPL dependence. The acoustic‐HRSs at 750 Hz and different SPL (45, 60, 75, 90, and 105 dB, respectively) were investigated under ambient conditions. Under the condition of 45 and 60 dB, the average ON/OFF ratio was ≈1.9 and 4.5 (Figure 5a,b ). Ion migrations would not be restricted under low SPL conditions and the CFs formed by electric‐RS did not break. [ \n \n 66 \n , \n 67 \n \n ] In this case, HRS simply increased from ≈100 to ≈300 Ω under the influence of sound waves of 45 and 60 dB. As the SPL rose from 60 to 75 dB, the average ON/OFF ratio increased to ≈50 (Figure 5c ). It was clear that indestructible CFs under previous SPL conditions ruptured under higher SPL of 45 and 60 dB. [ \n \n 68 \n \n ] However, there were still some piezo‐acoustic‐RS failures affecting the operation stability of memristors. Subsequently, similar trends could be observed where the SPL further increased to 90 and 105 dB (Figure 5d,e ). The acoustic‐HRSs became uniform during electric‐SET and acoustic‐RESET cycles. The average ON/OFF ratios of 90 and 105 dB at 750 Hz reached ≈10 3 , which is 20 times higher than the result at 75 dB. Moreover, RS exhibited small differences between 90 and 105 dB because they were close to the IRS, where localized vacancy concentration reached an unprecedented minimum, even comparable to the initial concentration. The acoustic‐HRSs hardly increased when the SPL reached 105 dB, which is close to the IRS of the material determined by the intrinsic vacancy concentration. The residual amount of multiple or incomplete rupture CFs would almost disappear under the sound waves with high SPL. [ \n \n 69 \n \n ] Figure 5f showed the retention results (read voltage: −0.1 V, 50 ms) at different SPL during 5 × 10 4 s. The corresponding statistical distribution was shown in Figure S15 (Supporting Information). The reproducible RS with a large ON/OFF ratio was far superior to other pure electric‐driven HPs memristors such as MA 3 Sb 2 Br 9 and Cs 3 Bi 2 I 9 , indicating ultralow energy consumption. [ \n \n 70 \n , \n 71 \n \n ] Generally, this phenomenon proved that the higher SPL stimulation could produce more energetic sound waves for regulating the RS behavior. [ \n \n 72 \n \n ] Based on the above analysis, we speculated that piezo‐acoustic mechanical stress affected the migration of ions. Higher frequencies and SPL lead to a more thorough annihilation of V I s, resulting in rare residual CFs and ultimately higher HRS. In Table S2 (Supporting Information), we have aggregated the performance parameters of the MAPbI 3 ‐based memristors alongside those of other memristors. Notably, our device demonstrated exceptional performance, surpassing or at least comparable to the performance of other memristors in the comparison. Figure 5 SPL‐dependence of the acoustic‐HRSs and retention property. a–e) SPL‐dependence of the acoustic‐HRSs in successive electric‐SET and acoustic‐RESET cycles. (a) 45 dB. (b) 60 dB. (c) 75 dB. (d) 90 dB. (e) 105 dB. f) Retention test results at different SPL during 5 × 10 4 s. Consequently, we proposed the RS mechanism for the formation and rupture of the CFs based on hybrid electric and piezo‐acoustic RS effects, as illustrated in Figure \n 6 \n . Initially, a small quantity of V I s was randomly dispersed in the MAPbI 3 films, and the device was in LRS due to the intrinsic high resistance characteristic (Figure 6a ). After applying a positive bias to the top electrode in the FORMING process (The subsequent sweep was the SET process), V I s with positive charge migrated toward the bottom electrode and accumulated until CFs were constructed (Figure 6b ). Subsequently, sound signal stimulation promoted the recombination of excess V I s with I − s and accelerated the V I s annihilation (Figure 6c ). Therefore, the V I s concentration significantly decreased. Eventually, the CFs ruptured, and the device switched to HRS again (Figure 6d ). Figure 6 Schematic of the RS mechanism in the EGaIn/MAPbI 3 /PEDOT:PSS/ITO memristor. a) Initial state showing randomly distributed V I s. b) CFs construction after FORMING or SET process. c) Recombination of excess V I s with I − s and annihilation of V I s stimulated by sound waves. d) Stable acoustic‐HRSs."
} | 10,684 |
36352542 | PMC9646523 | pmc | 4,531 | {
"abstract": "Abstract Background Coral reefs house about 25% of marine biodiversity and are critical for the livelihood of many communities by providing food, tourism revenue, and protection from wave surge. These magnificent ecosystems are under existential threat from anthropogenic climate change. Whereas extensive ecological and physiological studies have addressed coral response to environmental stress, high-quality reference genome data are lacking for many of these species. The latter issue hinders efforts to understand the genetic basis of stress resistance and to design informed coral conservation strategies. Results We report genome assemblies from 4 key Hawaiian coral species, Montipora capitata, Pocillopora acuta, Pocillopora meandrina , and Porites compressa . These species, or members of these genera, are distributed worldwide and therefore of broad scientific and ecological importance. For M. capitata , an initial assembly was generated from short-read Illumina and long-read PacBio data, which was then scaffolded into 14 putative chromosomes using Omni-C sequencing. For P. acuta, P. meandrina , and P. compressa , high-quality assemblies were generated using short-read Illumina and long-read PacBio data. The P. acuta assembly is from a triploid individual, making it the first reference genome of a nondiploid coral animal. Conclusions These assemblies are significant improvements over available data and provide invaluable resources for supporting multiomics studies into coral biology, not just in Hawaiʻi but also in other regions, where related species exist. The P. acuta assembly provides a platform for studying polyploidy in corals and its role in genome evolution and stress adaptation in these organisms."
} | 436 |
22399947 | PMC3280839 | pmc | 4,532 | {
"abstract": "Wireless Sensor Networks consisting of nodes with limited power are deployed to gather useful information from the field. In WSNs it is critical to collect the information in an energy efficient manner. Ant Colony Optimization, a swarm intelligence based optimization technique, is widely used in network routing. A novel routing approach using an Ant Colony Optimization algorithm is proposed for Wireless Sensor Networks consisting of stable nodes. Illustrative examples, detailed descriptions and comparative performance test results of the proposed approach are included. The approach is also implemented to a small sized hardware component as a router chip. Simulation results show that proposed algorithm provides promising solutions allowing node designers to efficiently operate routing tasks.",
"conclusion": "5. Conclusions In this paper we have presented a new protocol for WSN routing operations. The protocol is achieved by using an ACO algorithm to optimize routing paths, providing an effective multi-path data transmission method to achieve reliable communications in the case of node faults. We aimed to maintain network life time at a maximum, while data transmission is achieved efficiently, so an adaptive approach is developed according to this goal. The proposed approach is compared to a well-known ant based algorithm named EEABR using an event-based simulator. The results show that our approach offers significant reductions of energy consumption which is used as a performance metric for different sized WSNs. We also implemented our approach on a small sized hardware component requiring minimum connections suitable for tiny node designs and we developed an easy method for handling the routing tasks by using the proposed router chip. We also tested the ACO algorithm running on the router chip and obtained its performance results, including response times of the chip. Response time of the header request for the routing operation would be satisfactory for many WSNs where transmission speed is not essential. The proposed ACO approach for WSN routings and its hardware implementation seem to be a promising solution for node designers. As future work, it is planned to improve our routing approach to be effective in proper WSN settings, including nodes having high mobility. The improved approach will also be studied in network types that include multiple sink nodes.",
"introduction": "1. Introduction Due to advances in low-power wireless communications, low-power analog and digital electronics, the development of low-cost and low-power sensor nodes that are small in size has received increasing attention. Sensor nodes have the ability to sense the environment nearby, perform simple computations and communicate in a small region. Although their capacities are limited, combining these small sensors in large numbers provides a new technological platform, called Wireless Sensor Networks (WSNs). WSNs provide reliable operations in various application areas including environmental monitoring, health monitoring, vehicle tracking system, military surveillance and earthquake observation [ 1 - 2 ]. Although WSNs are used in many applications, they have several restrictions including limited energy supply and limited computation and communication abilities. These limitations should be considered when designing protocols for WSNs. Because of these considerations specific to WSNs, many routing schemes using end-to-end devices and MANETs [ 3 ] are inappropriate for WSNs. In sensor networks, minimization of energy consumption is considered a major performance criterion to provide maximum network lifetime. When considering energy conservation, routing protocols should also be designed to achieve fault tolerance in communications. In addition, since channel bandwidth is limited, protocols should have capability of performing local collaboration to reduce bandwidth requirements [ 4 ]. The basic method to transfer information from a sensor node to the base is called flooding. In this method, information is disseminated by all the nodes as well as the base node. The broadcasting operation to all over the network consumes too much node resources such as energy and bandwidth. Heinzelman et al. proposed SPIN family protocols that disseminate all the information in the network assuming that all nodes are potential base nodes [ 5 ]. However SPIN's data advertisement operation does not guarantee data delivery. In this respect multi-path routing protocols promise advantages. The use of multiple paths to transfer data to the base enhances the reliability of WSNs. Directed diffusion [ 6 ] is a candidate method for multi-path routing. However, directed diffusion may not be suitable for those monitoring applications which require periodic data transfers. The optimization of network parameters for a WSN routing process to provide maximum network lifetime might be considered as a combinatorial optimization problem. Many researchers have recently studied the collective behavior of biological species such as ants as an analogy providing a natural model for combinatorial optimization problems [ 7 - 10 ]. Ant colony optimization (ACO) algorithms simulating the behavior of ant colony have been successfully applied in many optimization problems such as the asymmetric traveling salesman [ 11 ], vehicle routing [ 12 ] and WSN routing [ 8 , 13 , 14 ]. Singh et al. [ 15 ] proposed an ant based algorithm for WSN routings. However, this algorithm does not consider the main specifics of WSN structures, including energy related issues. Zhang et al. [ 16 ] proposed ant based algorithms for WSNs; their study includes three routing algorithms named SC, FF, and FP. The algorithms are successful with initial pheromone settings to have a good system start-up, but the SC and FF algorithms are not quite effective in latency, while providing better energy efficiency. Besides, the FP algorithm, while providing high success rates of data delivery, consumes much higher energy than the FC and FF algorithms. The Energy Efficient Ant Based Routing Algorithm for WSNs (EEABR) [ 17 ], based on a ACO metaheuristic, is another proposed ant based algorithm to maximize the lifetime of WSNs. The algorithm uses a good strategy considering energy levels of the nodes and the lengths of the routed paths. In this paper, we have compared the performance results of our ACO approach to the results of the EEABR algorithm. Various differently sized networks are considered, and our approach gives better results than EEABR algorithm in terms of energy consumption. The main goal of our study was to maintain network life time at a maximum, while discovering the shortest paths from the source nodes to the base node using a swarm intelligence based optimization technique called ACO. A multi-path data transfer is also accomplished to provide reliable network operations, while considering the energy levels of the nodes. We also implement our approach on a hardware component to allow designers to easily handle routing operations in WSNs. The preliminary report on this work may be seen in [ 18 ]. The rest of the paper is organized as follows. In Section 2, the proposed routing scheme using ACO is explained by an illustrated example scenario. In Section 3, performance results obtained from the simulations are given. In Section 4, implementation of the approach is presented with hardware simulation results. Finally, in Section 5 we conclude our study and give our future work plan."
} | 1,872 |
28651051 | null | s2 | 4,533 | {
"abstract": "Robust self-assembly across length scales is a ubiquitous feature of biological systems but remains challenging for synthetic structures. Taking a cue from biology-where disparate molecules work together to produce large, functional assemblies-we demonstrate how to engineer microscale structures with nanoscale features: Our self-assembly approach begins by using DNA polymerase to controllably create double-stranded DNA (dsDNA) sections on a single-stranded template. The single-stranded DNA (ssDNA) sections are then folded into a mechanically flexible skeleton by the origami method. This process simultaneously shapes the structure at the nanoscale and directs the large-scale geometry. The DNA skeleton guides the assembly of RecA protein filaments, which provides rigidity at the micrometer scale. We use our modular design strategy to assemble tetrahedral, rectangular, and linear shapes of defined dimensions. This method enables the robust construction of complex assemblies, greatly extending the range of DNA-based self-assembly methods."
} | 262 |
23300594 | PMC3530589 | pmc | 4,535 | {
"abstract": "Fumaric acid (FA) is a promising biomass-derived building-block chemical. Bio-based FA production from renewable feedstock is a promising and sustainable alternative to petroleum-based chemical synthesis. Here we report on FA production by direct fermentation using metabolically engineered Saccharomyces cerevisiae with the aid of in silico analysis of a genome-scale metabolic model. First, FUM1 was selected as the target gene on the basis of extensive literature mining. Flux balance analysis (FBA) revealed that FUM1 deletion can lead to FA production and slightly lower growth of S. cerevisiae . The engineered S. cerevisiae strain obtained by deleting FUM1 can produce FA up to a concentration of 610±31 mg L –1 without any apparent change in growth in fed-batch culture. FT-IR and 1 H and 13 C NMR spectra confirmed that FA was synthesized by the engineered S. cerevisiae strain. FBA identified pyruvate carboxylase as one of the factors limiting higher FA production. When the RoPYC gene was introduced, S. cerevisiae produced 1134±48 mg L –1 FA. Furthermore, the final engineered S. cerevisiae strain was able to produce 1675±52 mg L –1 FA in batch culture when the SFC1 gene encoding a succinate–fumarate transporter was introduced. These results demonstrate that the model shows great predictive capability for metabolic engineering. Moreover, FA production in S. cerevisiae can be efficiently developed with the aid of in silico metabolic engineering.",
"introduction": "Introduction Fumaric acid (FA) is widely used in food, pharmaceutical and chemical industries, and is attracting increasing attention because it can be converted into therapeutic drugs and is a starting material for polymerization and esterification. FA is mainly produced petrochemically from maleic anhydride at present. Increasing petroleum prices, concerns about climate change and advances in the field of metabolic engineering have fueled renewed interest in the production of organic acids by microbial fermentation [1] . Although high FA yields have been obtained from fungi such as Rhizopus oryzae [2] and Rhizopus arrhizus [3] , the process might be limited at the industrial scale because these fungi are difficult to grow and their morphology can strongly affect production characteristics. The yeast Saccharomyces cerevisiae was regarded as a suitable microorganism for biotechnological production of carboxylic acids [4] , and significant progress has been made in exploring metabolic engineering for the production of carboxylic acids such as lactic [5] , malic [6] , [7] , and succinic acids [8] , [9] by S. cerevisiae. At least two metabolic strategies can be used for FA production by S. cerevisiae . In the first, FA can be produced via a reductive tricarboxylic acid (TCA) cycle, which provides a maximum theoretical yield of 2 moles of FA per mole of glucose. Moreover, this process involves biological CO 2 fixation instead of release, which is of great interest because of increasing concerns about climate change. In our previous study, an exogenous fumarate biosynthetic pathway involving reductive reactions of the TCA cycle was successfully introduced in S. cerevisiae via a series of simple genetic modifications and pyruvate carboxylase was identified as one of the factors limiting fumarate production [10] . However, the energy balance for FA synthesis via a reductive TCA cycle is barely even and does not provide any ATP for maintenance and active transport processes, and the redox balance is uneven. In the second strategy, FA can be produced via an oxidative TCA cycle and the engineered strain is stable in the fermentation process. It was reported that cells of a fumarase-deficient mutant accumulated extracellular FA when fermenting glucose [11] . Similarly, a concentration of 3.62 g L –1 at a yield of 0.11 moles of succinic acid per mole of glucose was achieved for oxidative production of succinic acid in yeast by deletion of the SDH1 , SDH2 , IDH1 and IDP1 genes [8] . Recent advances in genomics and other -omics technologies combined with computational analysis have opened new avenues for strain improvement [12] – [15] . Metabolic engineering combined with systems biology has been successfully applied to the development of strains capable of enhanced production of chemicals and materials by redistributing and optimizing metabolic fluxes [16] . Identification of genes for manipulation is an essential step in metabolic engineering for strain improvement for enhanced production of target bioproducts. In the present study, the target gene for FA production in S. cerevisiae was identified via literature mining. Then iND750, a validated genome-scale metabolic model (GSMM) of S. cerevisiae [17] , was used for in silico simulation of the metabolic response to deletion of the target gene by flux balance analysis (FBA) [18] and robustness analysis ( Figure 1 ) [19] . Rational metabolic engineering [20] was then applied to develop a S. cerevisiae strain capable of efficient FA production. In addition, to further improve FA production, the model combined with literature surveys was used as a tool to indentify the controlling steps, and experimental validation was performed. 10.1371/journal.pone.0052086.g001 Figure 1 The major metabolic pathway leading to the formation of fumaric acid and in silico carbon flux distribution in the central metabolism of S. cerevisiae during fumaric acid production on glucose. a/b, “a” represent the flux of parent strain, “b” represent the flux of the mutant strain FMME-002Δ FUM1 . Fluxes are shown relative to a glucose uptake rate of 100.",
"discussion": "Discussion None of the natural fumarate-producing microorganisms seem to be suitable for large-scale commercial production although high FA yields have been obtained [2] , [3] . S. cerevisiae is an excellent platform for biologically based chemicals such as organic acids. The aim of the present study was to construct a genetically engineered S. cerevisiae strain that can produce FA. First, the target gene FUM1 was identified in an extensive literature search and then FBA was used to predict the effect of FUM1 disruption using the S. cerevisiae i ND750 GSMM. The simulated results revealed that FUM1 deletion could lead to FA accumulation with only a slight influence on cell growth (∼1.95% lower). Then gene deletion was carried out and engineered S. cerevisiae cells produced FA at concentrations up to 610±31 mg L –1 . Meanwhile, cell growth and glucose consumption were slightly lower compared to the parent strain, in accordance with the simulated result. Simulated results also showed that pyruvate carboxylase could be one of the factors limiting higher FA production, and an improved FA yield was obtained when the RoPYC gene was introduced. Furthermore, a significant improvement in FA production was achieved when the SFC1 gene was introduced. The final FA concentration obtained was 1675±52 mg L –1 . Thus, the engineered strain provides a potential new route for FA production. However, the concentration and yield are low in comparison with R. oryzae \n [2] , so further work is required before this approach is economically feasible. The number of GSMMs available is increasing sharply [24] . Because a GSMM represents nearly all the metabolic activities of an organism, it can be of great help in understanding metabolism on a global level [25] . Thus, GSMMs are widely used in metabolic engineering [26] , [27] and can be used to predict and evaluate genetic manipulations in advance (dry experiments) when combined with certain algorithms [18] , [28] . This can greatly improve the efficiency and directionality of metabolic engineering in various phases by predicting gene targets to be manipulated throughout the whole cellular network. In S. cerevisiae , metabolic engineering strategies aided by GSMM have led to improved production of various metabolites such as bioethanol, purine, proline/pyrimidines and vanillin [29] . In addition to direct improvements in production capacity, GSMMs can also be used to predict cellular properties or phenotypic traits such as growth and glucose consumption. In previous studies, growth behavior, and ethanol, succinate, citrate and fumarate concentrations were determined in various media (rich and minimal media) under aerobic and anaerobic conditions [11] , [30] , but the effect of FUM1 deletion or fumarase deficiency on fermentation profiles (growth and glucose uptake rate) has not been studied in detail. In the present study, the phenotypic trait of slightly lower growth caused by FUM1 deletion in S. cerevisiae was successfully predicted by FBA analysis; this trait is important for metabolic engineering because unwanted side effects can be induced. Metabolic models are also useful in identifying targets for further strain improvement. We identified pyruvate carboxylase as a factor restricting higher FA yield. A higher FA yield was obtained by increasing the flux through pyruvate carboxylase. The increased flux induced by overexpression of pyruvate carboxylase is linked to increased transport of cytosolic oxaloacetate into mitochondria and supply to oxidative reactions [31] . When pyruvate carboxylase and the transporter encoded by SFC1 were coexpressed, a higher growth rate and FA yield were obtained, which suggests that insufficient FA export is another controlling step that would lead to higher FA production in steady-state metabolism. The model showed some restrictions; however, the physiological characteristics observed for engineered organisms can be used to update the model. In the present study, there was very good accordance between in silico predictions and experimental results. The discrepancy between experimental and predictive yields was primarily caused by lack of model knowledge for yeast metabolism, regulatory mechanisms and feedback inhibition, which requires specific further experimental investigation. In conclusion, the metabolic pathways in S. cerevisiae were rationally engineered for FA production with the aid of in silico simulations. The strategy described here can be useful for improved production of organic acids and other metabolites by direct microbial fermentation from renewable resources."
} | 2,584 |
33791573 | PMC7750977 | pmc | 4,536 | {
"abstract": "Synopsis Bioeroding organisms play an important part in shaping structural complexity and carbonate budgets on coral reefs. Species interactions between various bioeroders are an important area of study, as these interactions can affect net rates of bioerosion within a community and mediate how bioeroders respond to environmental change. Here we test the hypothesis that the biomass of endolithic bioeroding microalgae is positively associated with the presence of a macroboring bivalve. We compared the biomass and chlorophyll concentrations of microendolithic biofilms in branches of the coral Isopora palifera (Lamarck, 1816) that were or were not inhabited by a macroboring bivalve. Those branches with a macroborer present hosted ∼80% higher microbial biomass compared to adjacent branches from the same coral with no macroborer. Increased concentrations of chlorophyll b indicated that this was partly due to a greater abundance of green microalgae. This newly described association has important implications for the coral host as both the bivalve and the microalgae have been hypothesized as symbiotic.",
"introduction": "Introduction The process of bioerosion (the decay, degradation, or dissolution of calcium carbonate by living organisms) modulates multiple aspects of reef ecological function. Bioeroders modify the structural complexity of reefs at scales from microns to meters ( Glynn and Manzello 2015 ; Davidson et al. 2018 ; Roff et al. 2019 ), which influences coral reef herbivory ( Vergés et al. 2011 ), coral and fish larval settlement ( Coker et al. 2012 ; Kegler et al. 2017 ), and a reef’s economic value ( Graham and Nash 2013 ; see also Schönberg et al. 2017 ). Bioeroders can be characterized into guilds depending upon their habitat (epilithic or endolithic), size (micro and macro), and mechanism of bioerosion ( Schönberg et al. 2017 ), with both mechanical and chemical mechanisms being employed by various guilds. Some bioeroding taxa such as excavating parrotfish, scraping urchins, and boring sponges physically break down carbonate substrates and produce coarse carbonate sediments on a reef. Sediment turnover is important in maintaining sediment porosity and permeability, which affects the productivity of coral reef sands ( Miyajima et al. 2001 ; Santos et al. 2010 ). Other bioeroders dissolve, rather than degrade, carbonate substrates and so affect reef carbonate budgets by modulating the availability of dissolved inorganic carbon ( Perry et al. 2014 ). Bioerosion, therefore, affects both reef growth and productivity. Importantly, there is evidence that ecological interactions between bioeroders are capable of enhancing or suppressing overall rates of erosion on a coral reef. As such these are key considerations when seeking to understand how bioerosion is shaping a reef environment. For example, grazing urchins may target endolithic microalgae (i.e., living within the rock) as a food source and scrape the substrate as they feed ( Chazottes et al. 1995 ). Despite this, predation by urchins ultimately increases bioerosion by microendoliths due to the increased light field in the substrate which extends the compensation depth for algal colonization ( Chazottes et al. 1995 ; Tribollet and Golubic 2005 ). Parrotfish and urchin grazing have also been found to control recruitment and succession in endolithic macroborer communities in Kenya, resulting in reduced overall rates of macrobioerosion on the reef ( Carreiro-Silva and McClanahan 2012 ). These inter-guild interactions between microborers, macroborers, and grazers can take the form of feedback loops in which one guild might enhance or suppress bioerosion by a different guild ( Carreiro-Silva and McClanahan 2012 ; Schönberg et al. 2017 ). Environmental change can disrupt the balance between these species and alter pathways in these loops, leading to shifts in ecosystem bioerosion ( Perry et al. 2014 ; Schönberg et al. 2017 ). Therefore, we can benefit from a greater understanding of the number and structure of bioeroder interactions within an ecological web. Recently, Rice et al. (2020) identified an interaction between two bioeroding organisms, excavating parrotfish and endolithic lithophagine mussels inhabiting the skeleton of live massive Porites spp. corals. Similar to the results presented by Rotjan and Lewis (2005) , the authors identified a positive relationship between the density of macroborers in a live coral and the frequency of parrotfish bite scars on the same colony. Both studies suggested that targeted feeding by parrotfish on macroborers drove this association, and Rice et al. (2020) went further to hypothesize that this interaction might be partly mediated by endolithic microalgae living alongside the bivalves inside the coral skeleton. Excretion by the bivalve, which underlies its putative beneficial role to corals ( Mokady et al. 1998 ), could effectively fertilize the surrounding skeleton and so increase the abundance of endolithic microalgae. The blue mussel, Mytilus edulis , has been shown to enrich sediment porewater through the biodeposition of ammonium and phosphates, boosting the growth of co-occurring seagrass ( Reusch et al. 1994 ). By enriching the endolithic habitat of a coral colony with waste products, bivalves are potentially increasing the abundance of endolithic algae which increases the nutritional value of that patch of coral colony for a grazing parrotfish. This is especially true for “microphagous” parrotfish, a termed used to describe their preferential feeding habits on areas abundant in microalgae ( Bruggemann et al. 1994 ; Clements et al. 2016 ). Here we investigate the potential for this undescribed association occurring inside the skeleton of a living coral, with the potential to influence coral health and skeletal integrity. We tested the hypothesis of Rice et al. (2020) that the presence of macroborers is associated with an increased biomass of microalgae in the endolithic habitat of Isopora palifera (Lamarck, 1816) coral colonies.",
"discussion": "Discussion Lithophagine bivalve macroborers, identified by the figure-eight shape of their boreholes ( Kleemann 1980 ), were present in all but one of the surveyed I. palifera colonies on the Heron Island reef flat ( Fig. 1E ). The median density of boreholes per cubic meter of coral was slightly lower than those recorded on previous surveys of date mussel density in massive Porites spp. ( Rotjan and Lewis 2005 ; Rice et al. 2020 ). However, our method for approximating coral volume might be expected to overestimate the true volume of substrate available for macroboring. In the I. palifera branches inhabited by a macroboring bivalve, the microendolithic biomass was almost double that of adjacent, uninhabited branches from the same colony ( Fig. 2A ). Additionally, the concentrations of chlorophylls a and b were ∼4- and 2-fold greater in the presence of a macroborer ( Fig. 2B ). Chlorophyll b is the primary accessory pigment found in microendolithic green algae, such as Ostreobium spp., which commonly dominate the microboring communities of coral skeletons ( Marcelino and Verbruggen 2016 ; del Campo et al. 2017 ; Ricci et al. 2019 ; Pernice et al. 2020 ). Our data, therefore, suggest that the increased microbial biomass was in part due to a higher abundance of green microalgae in the skeleton. The data also indicate the presence of chlorophyll d -containing cyanobacteria within our I. palifera samples. The only recorded genus of alga known to use chlorophyll d is Acaryochloris ( Larkum and Kühl 2005 ), which has been identified previously from endolithic habitats under crustose coralline algae at this location ( Behrendt et al. 2011 ). We have found evidence of a positive association between the presence of a lithophagine bivalve macroborer and the biomass of endolithic microalgae, within an I. palifera coral host. Both Rotjan and Lewis (2005) and Rice et al. (2020) found that parrotfish bite frequency on a coral colony was correlated with the density of resident macroboring bivalves. All authors suggested that these relationships could reflect targeted feeding of parrotfish on nutrient-rich macroborers, as Rotjan and Lewis (2005) found no difference in the nutritional quality of overlying coral tissue. Rice et al. (2020) went on to hypothesize that it may also be partly mediated by microendolithic algae which live alongside the macroborer in the coral skeleton. Our results lend credence to this hypothesis. Higher microalgal biomass in the skeleton concomitant with macroborer infestation would increase the nutritional value of a particular patch of coral. This is especially true if the parrotfish species show preferential feeding on microalgae ( Bruggemann et al. 1994 ; Clements et al. 2016 ) and/or is omnivorous ( Bellwood and Choat 1990 ). Similarly, the bivalve–microendolith association could help explain the results presented by Simon-Blecher et al. (1996) . They recorded higher chlorophyll fluorescence in healthy coral tissue adjacent to a lithophagine borehole in the coral Goniastrea sp. compared to tissue without a borehole next to it. They suggest that this might reflect nitrogen-enrichment of coral tissue through bivalve excretion ( Simon-Blecher et al. 1996 ). However, Rotjan and Lewis (2005) found no difference in coral tissue nitrogen content whether next to or away from a borehole. Instead the stronger fluorescence signal may have been due to the microendolithic “halo,” as described here ( Fig. 1 ), beneath the coral tissue. Fine et al. (2005) have previously shown that endolithic algae beneath the tissue of the coral Oculina patagonica can influence chlorophyll fluorescence measured in the coral tissue. There are several possible mechanisms driving the association between macroboring bivalves and algal microendoliths inside the skeleton of I. palifera . First and foremost is the potential for fertilization of the endolithic microhabitat by excretion of nitrogenous waste from the macroboring bivalve. This dynamic underlies the hypothetical mutualism between corals and boring bivalves ( Mokady et al. 1998 ) and occurs in seagrass beds where infaunal blue mussels, M. edulis , enhance the growth of Zostera marina by fecal biodeposition ( Reusch et al. 1994 ). Additionally, artificial enriching the endolithic habitat with nitrogenous inorganic matter increases colonization and bioerosion by microendoliths in the shells of giant clams Strombus gigas ( Carreiro-Silva et al. 2005 , 2009 , 2012 ). It is also possible that CO 2 produced by bivalve respiration diffuses into the skeleton and enhances daytime microbial photosynthesis. This could be coupled with oxygen produced by photosynthesis diffusing into the macroborer burrow; endolithic algae have been previously shown to cause skeletal porewater to become supersaturated with respect to oxygen ( Kühl et al. 2008 ). Finally, there is evidence that photoassimilates produced by endolithic algae are a source of sugars for the coral host ( Schlichter et al. 1995 ; Fine and Loya 2002 ; Sangsawang et al. 2017 ) and this may also be the case for lithophagine bivalves. Taken together, this is suggestive of metabolic exchange between endolithic bivalves and microalgae in the form of series of positive feedback loops ( Fig. 3 ). The limited diffusion of metabolic waste products through the skeleton may also explain the shape of the green “halos” around macroborer boreholes. This has important implications for the spatial extent of this association within a coral colony. Fig. 3 Conceptual diagram illustrating the hypothetical mechanisms driving the relationship described here. ( 1 ) Biodeposition of nitrogenous compounds through bivalve excretion enhances microalgal growth. ( 2 ) Exchange of the products of photosynthesis and respiration for mutual benefit. ( 3 ) Each bioeroder reduces the energetic cost of boring for its skeletal co-habitant. ( 4 ) The open borehole increases irradiance for and promotes settlement of microalgal endoliths. Bioerosion by microendolithic algae can promote colonization by macroborers such as polychaetes and sponges, by weakening substrates and thereby reducing the energetic cost of macroborer colonization ( Che et al. 1996 ; Schönberg et al. 2017 ). Equally the secretion of an acidic mucus by the macroborer, which is the primary form of chemical erosion in boring bivalves ( Kleemann 1996 ), weakens skeletal matrices ( Scott and Risk 1988 ), and so may reduce the energetic cost of boring for microendoliths. Additionally, the presence of an external opening on the coral surface may increase the endolithic light field and promote more settlement by microendoliths that colonize new substrates from the water column as opposed to from neighboring substrates ( Massé et al. 2018 ) ( Fig. 3 ). Therefore both members of this association have the capacity to promote colonization and bioerosion by each other. In the freshwater bivalve Lignopholas fluminalis , cooperation with co-occurring microorganisms was found to promote the bioerosion of silicate siltstone ( Daval et al. 2020 ). The association described in this study may therefore be maintained through metabolic exchange and/or by the combined weakening of the coral skeleton ( Fig. 3 ). It is beyond the scope of this study to state how the relationship is first established. It is possible that the initial trigger is macroborer settlement which enhances growth in the already present microendolith community, which then serves to reduce the energetic cost of burrowing by the bivalve. In this vein, pre-existing microendolithic biomass (i.e., before infestation by a macroborer) might be an important factor affecting settlement success (i.e., recruitment) in bivalves. The bivalve–microendolith association described here is comparable to the relationship between macroboring polychaetes and microendoliths, wherein each guild promotes bioerosion by the other ( Che et al. 1996 ; Schönberg et al. 2017 ). These inter-guild relationships affect overall rates of bioerosion on a reef through the “bioerosion loop” ( Carreiro-Silva and McClanahan 2012 ; Schönberg et al. 2017 ). While it is not clear how this relationship between boring bivalves and microendoliths affects net bioerosion, it does have some interesting implications for the health of the coral host. Both lithophagine bivalves and endolithic microalgae have been previously independently proposed as symbiotic to their host coral ( Mokady et al. 1998 ; del Campo et al. 2017 ). Describing and understanding these multi-species interactions are a promising area for discovery and continues to be an important step in understanding the role of bioeroders on coral reefs."
} | 3,699 |
26184838 | PMC4505329 | pmc | 4,540 | {
"abstract": "To elucidate how geothermal irregularities affect the sustainability of high-temperature microbiomes we studied the synecological dynamics of a geothermal microbial mat community (GMMC) vis-à-vis fluctuations in its environment. Spatiotemporally-discrete editions of a photosynthetic GMMC colonizing the travertine mound of a circum-neutral hot spring cluster served as the model-system. In 2010 a strong geyser atop the mound discharged mineral-rich hot water, which nourished a GMMC continuum from the proximal channels (PC) upto the slope environment (SE) along the mound’s western face. In 2011 that geyser extinguished and consequently the erstwhile mats disappeared. Nevertheless, two relatively-weaker vents erupted in the southern slope and their mineral-poor outflow supported a small GMMC patch in the SE. Comparative metagenomics showed that this mat was a relic of the 2010 community, conserved via population dispersal from erstwhile PC as well as SE niches. Subsequently in 2012, as hydrothermal activity augmented in the southern slope, ecological niches widened and the physiologically-heterogeneous components of the 2011 “seed-community” split into PC and SE meta-communities, thereby reclaiming either end of the thermal gradient. Resilience of incumbent populations, and the community’s receptiveness towards immigrants, were the key qualities that ensured the GMMC’s sustenance amidst habitat degradation and dispersal to discrete environments.",
"conclusion": "Conclusions GMMCs have been explored extensively from diverse high temperature habitats using multiple methodologies 26 27 28 29 30 . However, this study is distinctive in using large volume of metagenomic sequence in combination with an equally substantive amount of 16S rDNA sequence to unearth microbial diversity associated with a relatively uncommon type of circum-neutral hot spring system. Though globally infrequent, such hot springs, which are typically poor in TDS, silicate and chloride, but often rich in sodium/calcium and various compounds of sulfur, are common in the geothermal areas of North-Western Himalayas and Tibet 8 31 32 33 . But very few explorations have hitherto been conducted to characterize their microbiota 34 35 . However, the main importance of the present study was its elucidation of the survival dynamics of a GMMC in the face of drastic environmental changes. In conclusion, that survival story can be summed up as follows. In 2010 a graded continuum of intricate ecological conditions existed along large swathes of the western slope of the PT sinter accretion. This resulted in wide habitat heterogeneity within the ecocline comprised of several gradients of opportunity (mainly in the form of various nutrients dissolved in the hot water outflow) as well as resistance (mainly in the form of thermal stress). While some of the opportunity gradients lay parallel to the gradient of environmental resistance, others were aligned antiparallel to the latter. Since almost none of the GMMC constituents were hyperthermophilic, the temperature gradient, running from the proximal channels to the slope environment, acted as a potential gradient of environmental resistance for nearly all the members of the community. In this scenario, ecological growth of the GMMC in time and space was mainly driven by the dispersion of populations, which in turn depended on the innate ability (resilience) of the species to cope with the environmental resistance. Eventually, on the multivariate ecological continuum (or the ecocline) the diverse biotypes (taxonomic and/or metabolic types) potentially consolidated their positions at their respective optimum niches, while their populations tapered off gradually on either side of the consolidation zones. Then between November 2010 and 2011 the large geyser atop the PT got extinguished, thereby causing the hot water flow on western slope of the mound to dry up completely. Even as substratum desiccation turned the existing habitat of the GMMC hostile, during the same period a slender prospect of survival (habitable space) opened up in the form of two new vents in the southern slope of the PT. Amidst this spatial drift, as well as shrinkage, of habitable space (defined by the shorter length and breadth of the new OFC) dispersal emerged as the main driver of ecological sustenance of the community. While some incumbent populations of the sprawling 2010 GMMC compromised their niche specialization (e.g. Deinococcus-Thermus ) others gave up their dispersion (e.g. Proteobacteria ) to launch spatiotemporal dispersal and build a minimally-dispersed seed-like community (viz. the 2011_GM_SE) on the southern-slope OFC. The apparent aim of the whole process was to conserve maximum-possible phylogenomic diversity from both ends of the 2010 GMMC continuum. However, in course of this en masse dispersal several incumbent populations of the 2010 GMMC got eliminated or acutely marginalized, even as many of them managed to maintain significant representation in the GM_SE construct of 2011. Besides embracing internally-displaced populations from the 2010 GMMC, 2011_GM_SE showed exceptional receptivity in countenancing external immigrants. This openness potentially enhanced the ecological fitness (survival value) of the community by inducting new metabolic pathways and/or variants of existing ones. Presumably, 2011_GM_SE must have positioned itself at such an optimum point on the new ecocline that allowed its diverse constituents to make best use of all available environmental opportunities and overcome existing resistances. Growth (dispersive scope) of 2011_GM_SE, however, was constrained by additional environmental resistances such as limited habitable space and scarce nutrients due to diminutive volume, flow rate, and chemical-content of the hydrothermal discharge. Presumably in order to sustain the equilibrium of its existence this recuperating GMMC limited its dispersion to a restricted zone of the already short ecological continuum, and therefrom started developing resilience adequate enough to encounter the existing resistances and prepare for future growth. Finally in 2012, geothermal activity in the southern-slope of the PT gained impetus (in the form of new vents, and increased flow rates and mineral contents of discharges), thereby widening the available ecological niches. The eco-physiologically miscellaneous components of 2011_GM_SE took this opportunity to split into discrete GM_PC and GM_SE metacommunities. As many expatriated 2010_GM_PC populations vacated the SE refuge, community growth space in the SE niche increased further. Thus, many GM_SE natives (e.g. many Proteobacteria ) who were reduced to minuscule minorities in 2011_GM_SE regained preeminence in 2012_GM_SE. However, the second round of population dispersal was more complex than the 2010–2011 transition since many more new species joined the GMMC, especially through the revived GM_PC window. In this way the GMMC seeds were again successfully sown at either end of the thermo-nutritional gradient. Whether this niche-partitioned GMMC can regain dispersive growth and regenerate an expansive continuum depend on the vitality of the spring system.",
"discussion": "Results and Discussion Site description and sample identity The Dhauliganga valley in the Garhwal Himalayas (in the state of Uttarakhand, India) encompasses several circum-neutral hydrothermal vents, which discharge up to 95 °C hot waters having low chlorine and silica content 8 . The hot spring cluster of our interest, christened as the “Pilgrim Terrace” (PT) after the holy shrines of the nearby areas, is located at Tapovan (30.492542 N / 79.631002 E), 15 km south-east of the Joshimath town on the Malari-Joshimath Road. This site was explored in the November months of 2010, 2011 and 2012, and data gathered thereof constituted the basis of this report. Topographically, PT is a massive mound of hydrothermally deposited travertine interspersed with iron, nickel, aluminium, silver and sulfur. Spatiotemporally unstable vents seated within this mound discharge hot waters with fluctuating vigor and mineral content. In 2010, vigorous hydrothermal discharge and fumarolic activity occurred from a large geyser situated on the flat top of the PT ( Fig. 1A1 ). Thermal water (95 °C, pH 7.5) ejecting out of this vent (at a flow rate of ~20 l s −1 ) was rich in thiosulfate, sulfate and ferrous iron ( Table 1 , Fig. 2A ). Active deposition of vertically-laminated red and creamy white sediments ( Figure S1A ) could be observed around the geyser mouth. Downslope, the laminated pattern gave way to fragile bulbous reddish yellow spherulites that entrapped fumarolic and/or microbially-produced gases ( Figure S1B ). Hyaline green microbial mats (GMs) grew all along the proximal as well as distal outflow channels (OFC) on the west-facing slope of the PT (area demarcated by green arrows in Fig. 1A1 ). GMs were sampled from the proximal channels (PC) as 2010_GM_PC, and also from the OFC in the slope environment (SE) as 2010_GM_SE ( Fig. 1A2 ). During our survey in 2011 we found that there was zero discharge from the PT-top geyser as its crater was completely clogged by sinter deposition. Consequently, there was no visible sign of the erstwhile GMMC along the west-facing slope of the PT ( Fig. 1B1 ). Large chunks of the west-slope travertine had already been quarried, exposing hard old sinters from the deeper layers. The active sedimentary fabrics observed in the 2010 PCs were found vividly preserved in the texture of the old sinter layers exposed by fresh vertical cuts ( Figure S1C ). Elsewhere, on the eroded slopes of the remnant sinter, algae and mosses had started colonizing. Notwithstanding this scenario on the west-face, two relatively-weaker mid-terrace hot water vents were found to have formed anew on the south-facing slope of the PT where there was neither any hydrothermal feature nor any microbial mat in the previous year. Both had flow rates of ~4 l s −1 , temperatures ~85 °C and pH 7.0. Their combined mineral-poor outflow ( Table 1 , Fig. 2A ) drained down the southern slope and supported a small patch of deep green mat (sampled as 2011_GM_SE) on the thin layer of freshly-precipitated travertine ( Fig. 1B2 ). Noticeably, this was the only visible counterpart of the sprawling 2010 GMMC in the entire PT microbiome of 2011. Until 2012, there was no revival of the PT-top geyser ( Fig. 1C1 ) and the western slope had started hosting angiospermous herbs ( Figure S1D ). However, hydrothermal manifestations in the southern slope had gained momentum as the two vents of 2011 origin had become more vigorous (flow rates ~6 l s 1 , 85–90 °C, pH 7.0–7.25), and just beneath the dead cavern of the 2010 geyser a small mud pool (~85 °C; pH 8.0) and a very strong but small vent (flow rate ~8 l s −1 , ~95 °C, pH 7.5) formed anew. Discharge from this new vent joined the combined OFC of the 2011 vents below; and together the three created an outflow more vigorous and mineral-rich than last year. A small patch of deep green mat (sampled as 2012_GM_PC) grew on the PC close to the vents of 2011-origin ( Fig. 1C2 ), while a couple of discrete GM stretches (sample 2012_GM_SE) also grew along the OFC in the SE. Rationale of the study and work-plan For all practical purposes, our central objective was to know whether the small GM_SE patch of 2011 was a remnant of the expansive 2010 GMMC, and then whether the two niche-partitioned mats (GM_PC and GM_SE) of 2012 were derivatives of that 2011 GM_SE propagule. If these conjectures were found to be incorrect then it was imperative to confirm whether the 2011 and/or 2012 mats were distinct pioneer communities trying to colonize the PT afresh. It is further noteworthy that even if there was a spatiotemporal dispersal of the original west-slope community to the southern slope, its structure and function(s) were most likely to undergo alterations owing to the differences in the physicochemical conditions of the two sites. As such, the two types of changes needed to be identified and their potential correlations investigated. Now, over the three assessment years five mat samples were retrieved from two distinct ecological niches, viz. the PC and SE OFCs. Incidentally, the SE niche was consistently occupied by mats throughout the study period but the PC niche was vacant in-between. Accordingly, the three GM_SE editions were chosen as the central models of this investigation and studied by both (i) shotgun metagenome analysis and (ii) amplified 16 S rDNA V3 sequence analysis. The two GM_PCs, on the other hand, were investigated mainly for comparative purposes, and V3 sequence-based taxonomic diversity analysis was sufficient for the same. Bacterial predominance in the GM_SE >1 Gb metagenomic sequence was generated for each GM_SE sample and used to draw synecological inferences ( Table S1 ). As such, percentage distribution of metagenomic reads over various groups of a particular taxonomic category was considered as a direct measure of the relative abundance of those groups. Initially, reads were classified up to the domain level ( Fig. 3A ) by searching against the non-redundant ( nr ) protein database using the Organism Abundance tool of MG-RAST. Similarly, 16S rRNA gene (rDNA) reads within the datasets were classified by searching against the RDP database using the same tool ( Fig. 3B ). Read distribution patterns obtained from either analysis was broadly in agreement with each other. Unassigned and unclassified sequences constituted ~15% of every metagenome, while reads assigned to eukaryotes made up only 0.2–0.8% and those attributable to viruses were <0.05%. Archaeal contribution in the GM_SE metagenomes of 2010, 2011 and 2012 was 0.01%, 1% and 0.5% respectively. Whatever little archaeal reads were there in the 2010 dataset were all attributable to Euryarchaeota . In contrast, out of the 1% archaeal share in the 2011_GM_SE metagenome, 0.37% came from Thaumarchaeota , 0.32% from Euryarchaeota , and 0.29% from Crenarchaeota . In 2012_GM_SE Euryarchaeota was the predominant phylum accounting for 0.23% of the total 0.5% archaeal share in the community. Out of the remaining 0.27%, Thaumarchaeota alone contributed 0.2%, while Crenarchaeota put in 0.07%. In both the 2011 and 2012 GM_SEs the thaumarchaeon Nitrosopumilus was the dominant genus accounting for 18–20% of all archaeal reads. Two other thaumarchaea Cenarchaeum and Nitrososphaera , alongside the Euryarchaeota member Methanosarcina , were the other dominant constituents of the 2011 and 2012 archaeal populations, collectively accounting for another 18% and 20% of all archaeal reads in the respective datasets. The exceptional hike in the proportion of Crenarchaeota in 2011_GM_SE was mainly due to increased representation of Pyrobaculum , which contributed ~50% of the all Crenarchaeota reads in this sample. 98–99% of the taxonomically classifiable reads of all three GM_SE editions were from Bacteria . Hence it was deemed appropriate to focus only on bacterial community dynamics for the rest of the study. Bacterial community dynamics in the GM_SE A broad consensus was witnessed between the pictures of phylum-level read abundance emerging from comparison against the RDP ( Fig. 3C ) and the nr protein ( Fig. 3D ; Table S2 ) databases. As illustrated by these figures, the three GM_SE editions varied considerably in the relative abundance of the various bacterial phyla. The 2011 edition showed maximum heterogeneity, while 2010_GM_SE was the least diversified. Proteobacteria dominated all the three GM_SEs, but the extent of their predominance was exceptionally high in 2010_GM_SE where 95% of all reads were ascribed to this phylum. The only other phylum having >1% representation in 2010 was Actinobacteria . Again, within Proteobacteria , Alphaproteobacteria was always the most abundant class except in 2010_GM_SE where Betaproteobacteria predominated ( Fig. 3E ). There was a sharp decline in the abundance of Proteobacteria in 2011_GM_SE where ~41% of total classifiable reads were affiliated to the phylum. This was accompanied by remarkable upturns in the representation of Chloroflexi , Bacteroidetes , Cyanobacteria , Firmicutes , Nitrospirae , Verrucomicrobia , Chlorobi , Deinococcus-Thermus , Acidobacteria and Planctomycetes . Several marginal phyla, which did not have >1% representation in any of the three GM_SE metagenomes, also increased their abundance by ≥10 times ( Fig. 2B ; Table S2 ). Interestingly again, Proteobacteria rebounded in 2012, complemented by concomitant fall in the abundance of many of those phyla which had surged in 2011. Notably, Actinobacteria was the only phylum whose oscillation was in sync with this trend of Proteobacteria . On the other hand, most of the phyla that had increased their representation in 2011_GM_SE could not hold on to their gains in 2012. Chloroflexi and Deinococcus-Thermus were the biggest casualties of proteobacterial resurgence in 2012, suffering maximum losses and retreating almost back to their 2010 levels. Cyanobacteria , Chlorobi , Nitrospirae , Gemmatimonadetes , Planctomycetes , Spirochaetes and Acidobacteria also experienced considerable downturn between 2011 and 2012, but were still above their 2010 levels. Firmicutes suffered minimum cutback from their 2011 level, whereas only Bacteroidetes and Verrucomicrobia managed to further increase their representations in 2012. Different mathematical indices were calculated ( Table S3 A through S3 C) to quantify the phylum-level ecological diversity (which involves ‘what types’ as well as ‘how many’) of the GM_SEs. Corroborating the relative abundance trends of Fig. 3 , Simpson dominance index D was exceptionally high for 2010_GM_SE (0.95), moderate for 2012_GM_SE (0.44), and lowest for 2011_GM_SE (0.22) ( Fig. 4A ). Expectedly, fluctuations of Shannon diversity index H ( Fig. 4B ) and Shannon equitability index E H ( Fig. 4C ) were inverse to the trend observed for D . Concurrence between genus- and phylum-level trends It was remarkable to observe that >50% of all taxonomically classifiable reads of any given GM_SE metagenome were assigned to the 50 most-abundant putative genera identified in that metagenome ( Fig. 5 ). In 2010_GM_SE, top-50 genera accounted for ~80% of all classifiable reads, thereby justifying its high dominance index D . In 2011 and 2012 GM_SE respectively top-50 genera accounted for 56% and 58% of all classifiable reads. Interestingly again, year-on-year constancies and fluctuations exhibited by the top-50 genera mirrored the trends observed for the relative abundance of GM_SE phyla in Fig. 3D . As a mark of stability, 13 names (written in green font in Fig. 5 ) were common to all the three top-50 generic lists (referred to hereafter as the ‘leaderboard’). The leaderboard of 2010_GM_SE ( Fig. 5A ) encompassed representatives from only two phyla, while that of 2011_GM_SE ( Fig. 5B ) was taxonomically most inclusive (encompassing genera from eight phyla), followed by that of 2012_GM_SE that included genera from five different phyla ( Fig. 5C ). These data also corroborated the synecological indices of the three GM_SE editions. In 2010_GM_SE, all but only one genus in the top-50 were from Proteobacteria . Apart from the 13 genera common to all three leaderboards and the single genus Azoarcus that was common to the 2010 and 2011 GM_SEs, the rest of the 36 members of the 2010_GM_SE leaderboard dropped from the corresponding list of 2011. Interestingly, 10 out of these 36 (names in blue font in Fig. 5 ) reappeared in the 2012_GM_SE leaderboard. Moreover, none of those 36 leaderboard members of 2010_GM_SE which dropped from the list in 2011 (including those 26 which failed to make a comeback in 2012) were actually dispelled from the community. The fact that they continued to occupy significant positions in the 2011 and 2012 GM_SEs was evident from their high read counts in these two metagenomes (see the numbers given in parenthesis in Fig. 5A ). 2011_GM_SE had only 23 Proteobacteria among its top-50 genera. Concomitant to this proteobacterial retreat there were 36 new entries in the 2011 leaderboard, out of which, 21 remained predominant in 2012_GM_SE (names in red font in Fig. 5 ). Again, among the 36 new leaders of 2011 only eight were completely absent in the 2010_GM_SE metagenome (see names in Fig. 5B whose first number in parenthesis is zero). Again, all these eight newcomers maintained significant presence in 2012_GM_SE, with four of them even managing to retain their place in that leaderboard. Proteobacterial resurgence was again conspicuous in 2012_GM_SE as 35 genera affiliated to the phylum hit the leaderboard. Concomitantly, all genera from Deinococcus-Thermus (three) , Chloroflexi (two), Planctomycetes (one) and Firmicutes (one), which had made it to the top-50 in 2011_GM_SE, were relegated from the chart. Number of genera from Cyanobacteria and Bacteroidetes in the top-50 list also decreased in 2012_GM_SE. Additionally, there were six new leaderboard entrants in 2012_GM_SE, all of which except Opitutus were already there in the community since 2010 but gained preeminence only in 2012. These numbers highlighted the resilience of the dominant components of the GM_SE, even as the community remained consistently receptive towards newcomers, which in their turn got stably incorporated into the guild. The last inference was evidenced by the fact that despite several transpositions in and out of the top-50 list over the three assessment years, not a single instance was apparent where an erstwhile leaderboard member went totally undetected in a subsequent GM_SE edition (follow the numbers in parentheses of Fig. 5 ). Correlating populational and environmental fluctuations It is presumable that in between 2010 and 2011 disappearance of the PT-top geyser and desiccation of the substratum ushered wide-ranging ecological limitations for the GMMC and eventually rendered irreparable damage to its habitat. In this scenario, even if a substantive fraction of the 2010 community managed to render dispersal to the southern slope, this calamity must have already impacted the community’s structure and function(s) significantly. In course of time, the distinct physicochemical conditions of the newfound habitat ought to have impacted the community further. However, it must be acknowledged that we lacked data on the environmental constraints, or for that matter, community alterations within the degenerating mats, around the time points when the west slope environment was actually drying up. As such, direct impact assessment of habitat degradation was not possible and environment-community correlations could only be based on comparative geochemistry and metagenomics of the spatiotemporally discrete hot water outflows and mat samples respectively. On top of that, comparative physicochemical information (see Table 1 or Fig. 2A ) used in the correlation studies were averages of data collected at three discrete time points of only one particular day. Hence there was no data in hand to prove that the apparent year-on-year geochemical variations between the niches did not also occur on a more regular, say monthly or seasonal, basis. The above caveat notwithstanding, it is noteworthy that nine out of the ten tested physicochemical parameters of the SE outflow showed identical patterns of oscillation over the three assessment years. Sulfate concentration was the only exception to the U-shaped curves observed for the nine other factors. Correspondingly, all the GM_SE phyla except Proteobacteria (particularly Gammaprotoebacteria ) and Actinobacteria exhibited bell-shaped trends of population fluctuation over the three assessment years (see Fig. 2B & 3E ). Temperature, pH, total dissolved solid (TDS), and concentrations of thiosulfate, sulfide, Fe 2+ and Fe 3+ iron, acetate and formate, all were at their relative highs in 2010. Next year all these values dipped sharply, only to increase again in 2012. While in 2012 most of these parameters were still below their 2010 levels, ferric iron recovered to its 2010 high while acetate went past that level. As such, the physicochemical milieu of 2011_GM_SE was quite different from that of 2010_GM_SE, while environmental parameters for 2012_GM_SE tended partly back towards the 2010 values. Here it may be recalled that a wide divergence between 2010_GM_SE and 2011_GM_SE, and an intermediary status of 2012_GM_SE, was already evident from community structure comparisons. These prima facie impressions were quantitatively established when the environmental conditions experienced by the three GM_SE editions were statistically correlated with their community compositions. Bray-Curtis similarities and Euclidean distances between GM_SE sample pairs ( Table S4 ) evidenced the wide divergence between 2010 and 2011 GM_SEs and the intermediary status of 2012_GM_SE, while Canonical Correspondence (CCA) ( Fig. 6 ) and Principal Coordinate (PCoA) ( Figure S2 ) analyses with phylum/class-level data graphically illustrated the same. CCA further revealed that the trifurcated relationship of the three GM_SEs stemmed mainly from the concentration of Betaproteobacteria in 2010_GM_SE, and Deinococcus - Thermus and Chloroflexi in 2011_GM_SE, while there was no such skewed population in 2012_GM_SE. Besides such large-scale population growths, a few instances of modest but significant population expansion were also apparent, e.g. (i) Actinobacteria and Zetaproteobacteria in 2010_GM_SE, (ii) Acidobacteria , Deltaproteobacteria , Cyanobacteria and Planctomycetes in 2011_GM_SE, and (iii) Verrucomicrobia and Bacteroidetes in 2012_GM_SE. Separation of the three GM_SE editions in CCA was further explained by the fact that some environmental factors were exceptionally high or low for certain particular samples. Such examples included (i) higher sulfide and Fe 2+ in the 2010 outflow, (ii) lower pH and temperature alongside higher sulfate in 2011, and (iii) higher acetate in 2012. As observable in Fig. 6 , several cases of population fluctuation in the GM_SE were correlated to the oscillations of one or more environmental factors. In the following section we discuss only those cases which were further corroborated by strong Pearson correlation values, i.e. r numerically >0.8 alongside P < 0.05. Fluctuations in the abundance of Deltaproteobacteria , Acidobacteria and Gammaproteobacteria vis-à-vis sulfur species concentrations and pH were some of the most conspicuous correlations observed in these analyses. Decrease in the pH of the SE outflow in 2011 coincided with ~50% increase in sulfate concentration. These were accompanied by matching decreases in the flow rate and, sulfide and thiosulfate content of the fluid ( Table 1 ). On the contrary, the 2012 pH upturn towards neutrality concurred with ~22% slump in sulfate content alongside relative hikes in flow rate, sulfide and thiosulfate content. However, none of these reversals were high enough to take the concerned parameters back to their 2010 levels. Within this gamut of fluctuations, increase in sulfate concentration of the 2011 SE outflow could have particularly ushered the five and a half fold increase in the relative abundance of Deltaproteobacteria in 2011_GM_SE ( Fig. 3E ), which in its turn was mainly attributable to rise in read counts of sulfate-reducing Desulfuromonadales and Desulfovibrionales members such as Geobacter , Pelobacter , Desulfovibrio , etc. The fact that deltaproteobacterial abundance in 2012_GM_SE, despite being halved from its 2011 level, still remained more than twice the level of 2010_GM_SE could be attributed to the moderately high sulfate concentration of 2012. It was further noteworthy that upturns in the abundance of sulfate-reducing bacteria coincided with the steep population growth of Cyanobacteria . This observation was in agreement with earlier reports of syntrophism between the two groups within microbial mats, where cyanobacterial excretory products (such as glycolate) are used by sulfate-reducing bacteria as electron sources 9 . The Acidobacteria population oscillated ( Fig. 2B ) in sync with Deltaproteobacteria ( Fig. 3E ), even as the former’s fluctuations were sharper. While both the groups showed strong positive and negative correlations with sulfate concentration and pH respectively, the small dip in the pH of the OFC could have driven the influx and/or proliferation of Acidobacteria in 2011_GM_SE. Acidobacteria can slow down metabolism and live long under low-nutrient conditions and substratum desiccation 10 . This together with their diverse mechanisms of respiration may have helped them overcome similar adversities between 2010 and 2011. The gammaproteobacterial population oscillated in a pattern opposite to that of Deltaproteobacteria and Acidobacteria , thereby showing strong positive and negative correlations with thiosulfate and sulfate respectively. These data were consistent with the fact that decrease in the overall population of Gammaproteobacteria in 2011_GM_SE involved sharp decrease in lithotrophic sulfur-oxidizing bacteria (SOB) like Thioalkalivibrio , Acidithiobacillus , Allochromatium and Halothiobacillus . Actinobacteria was the only phylum other than Proteobacteria that showed U-shaped oscillation over the years. Its abundance correlated positively with Fe 2+ , which was remarkable since iron oxidation is reportedly not widespread in Actinobacteria , and that too mostly confined to Acidimicrobidae 11 , a subclass absent from the Actinobacteridae -dominated PT GM_SE. Other significant canonical correspondences and linear correlations observed in the GM_SE community dynamics included (i) Cyanobacteria versus pH, TDS and formate ( r = −0.998 for all three); (ii) Firmicutes versus sulfide and ferrous ( r = −0.999 and r = −0.998 respectively); (iii) Nitrospirae versus pH, TDS and formate ( r = −1.000 for all); and (iv) Planctomycetes versus pH, TDS and formate ( r = −1.000 for all); (v) Deinococcus-Thermus versus temperature and ferric iron ( r = −0.997 for both); and (vi) Chlorobi versus sulfide ( r = −1.000). Negative correlations between Deinococcus-Thermus and temperature, and Chlorobi and sulfide, appeared strange. However, subsequent 16S rDNA sequence-based diversity estimation of the GM_SEs versus the GM_PCs ( Fig. 7 ) showed that the growth of Deinococcus-Thermus in 2011_GM_SE could be due to their dispersal from the 2010_GM_PC. Again relative abundance of Chlorobi or green sulfur bacteria (GSB) was lowest in 2010_GM_SE ( Fig. 2B ) when sulfide concentration in the OFC was at its highest ( Fig. 2A ). Unusually again, the GSB population increased to its highest level in 2011_GM_SE when sulfide was at its lowest. This could be explained by fact that Chromatiaceae and Ectothiorhodospirilaceae , or the purple sulfur bacteria (PSB), accounted for >10 times, 1.25 times and 1.4 times more reads than Chlorobi in the metagenomes of 2010, 2011 and 2012 GM_SE respectively. Now, despite their common requirement for sulfide, the PSB are nutritionally more versatile than the GSB by virtue of their ability to utilize organic compounds and also grow chemotrophically on thiosulfate in the dark by reducing oxygen 12 . So in 2010_GM_SE, the PSB might had outnumbered the fastidious GSB by making full use of thiosulfate (that was also abundant that year) and then by virtue of their numbers usurped the available sulfide also. However, with thiosulfate going down next year, the PSB potentially lost this advantage, thereby allowing the GSB to avail whatever sulfide was available in the 2011 discharge and grow from ~0.1% of all bacteria in 2010_GM_SE to ~0.5% in 2011_GM_SE. Constancy and variability of metabolic types in the GM_SE metagenomes Detection of photosynthetic genes in all GM_SE metagenomes concurred with their hyaline to deep green appearance. As such, the detectable repertoire of genes involved in electron transport and photophosphorylation, and chlorosome and phycobilisome light-harvesting complexes was higher in 2011 and 2012 GM_SEs than in 2010_GM_SE. Notably again, photosynthetic genes detected in 2010_GM_SE were mostly proteobacterial in origin, while those of the next two editions were mainly from Cyanobacteria . Invariable presence of multiple homologs of all the sox (sulfur oxidation) structural genes of chemo- as well as photo-lithotrophic bacteria in all the three GM_SE metagenomes indicated that reduced sulfur species, which were by and large abundant in this spring system, could be a key source of energy and/or electron for the community. Molecular hydrogen could also be an important source of energy and/or electron for the autotrophic as well as mixotrophic members of the GM_SE. This was suggested not only by the detection of energy-generating NAD + -reducing [NiFe] hydrogenases in all three GM_SE metagenomes but also by the occurrence of the hydrogen-oxidizing “Knallgas” bacterium Ralstonia in all the three lists of top-50 genera ( Fig. 5 ). Other such organisms as Rhodococcus , Azotobacter , Chloroflexus etc., which too can utilize H 2 as a source of energy and/or electron via the oxyhydrogen “Knallgas” reaction 2H 2 + O 2 → 2H 2 O 13 14 were also significantly represented in all these metagenomes. Consistent presence of facultatively hydrogen-oxidizing chemolithoautotrophic diazotrophs like Bradyrhizobium , and phototrophs like Rhodobacter and Rhodopseudomonas 14 , in significant proportions further underscored the potential importance of hydrogen as an energy source of the community. Besides sulfur- or hydrogen-based lithotrophy, anaerobic carbon monoxide (CO) oxidation could also be another important source of energy and electron for the GM_SE. This was evidenced by the constant predominance of the anoxygenic phototroph Rhodospirillum [which can grow anaerobically in the dark on CO as the sole source of energy 14 ] and the presence of several homologs of CO dehydrogenase (CODH) and CO-insensitive [NiFe] hydrogenases in all GM_SE metagenomes. Notably, other carboxydotrophs like Streptomyces 15 16 , Carboxydothermus and Oligotropha 14 also accounted for significant fractions of metagenomic reads in all GM_SE editions. Hydrothermal supply of iron (particularly Fe 2+ ) could be another key determinant of the GM_SE community structure. While an apparent parallelism existed between Fe 2+ concentration in OFCs ( Table 1 , Fig. 2A ) and the iron richness of the GM_SE substrata ( Figure S3 ), fluctuations of these two factors correlated well with the oscillation of iron-oxidizing bacteria (IOB) within the community (see lower panel of Figure S3 ). In 2010 the hydrothermal discharge and the deposited sinters were remarkably rich in Fe 2+ and Fe 3+ respectively ( Table 1 ). Correspondingly, there were at least eight lithotrophic or organotrophic IOB (viz. Pseudomonas , Rhodobacter , Thiobacillus , Leptothrix , Sideroxydans , Rhodopseudomonas , Gallionella and Marinobacter ) 17 in the list of top-50 genera of 2010_GM_SE. Out of these eight genera three were also present in the corresponding list of 2011_GM_SE, while four made the cut in 2012_GM_SE ( Fig. 5 ). IOB, as such, accounted for >10% of all bacterial metagenomic reads of 2010_GM_SE, while their read share was ~3% in the next two GM_SE editions. Drop in IOB abundance in the GM_SE clearly coincided with the sharp decline in Fe 2+ availability in 2011, but, due to some unknown reason, their population did not recover appreciably in 2012 despite considerable increase in Fe 2+ abundance in the discharge as well as the substratum. Significant synchrony was also observed between the oscillation of the Shewanella population ( Fig. 5 ) and that of iron, thiosulfate or TDS in the OFC ( Fig. 2A ), or for that matter, iron in the substratum ( Figure S3 ). Under oxygen-stressed conditions Shewanella respire by reducing diverse oxidized metals and other substrates like fumarate, nitrate, trimethylamine N -oxide, dimethyl sulfoxide, sulfite, thiosulfate and elemental sulfur 18 . As such, this facultatively aerobic alphaproteobacterium, or its close phylogenomic relatives, could complement the biogeochemical roles of IOB and SOB and help complete the iron and sulfur cycles in the PT slope environment. They could also play central role in the biomineralization process in the slope microfacies by reducing dissolved metal ions from the hydrothermal fluid to their insoluble sulfide or oxide forms. So far as carbon sources were concerned, all three GM_SE metagenomes encompassed large number of CO 2 fixation-related genes such as those involved in carboxysome, CO 2 uptake, Calvin-Benson cycle, and photorespiration/oxidative C2 cycle. Potentials for autotrophic acetogenesis via the acetyl-CoA “Wood-Ljungdahl” pathway 19 were noticeable in the GM_SE throughout the assessment period. However, standard acetogens like Clostridium , Moorella , Eubacterium and Thermoanaerobacter were present in significant proportions only in the 2011 and 2012 editions of the GM_SE, but not in 2010_GM_SE. As such, these four genera together accounted for 0.01%, 0.34% and 0.27% of all taxonomically classifiable reads of the 2010, 2011 and 2012 GM_SE metagenomes respectively. Nonetheless, genes for some of the key enzymes of the bacterial acetyl-CoA pathway, viz. (i) formate dehydrogenase, (ii) methenyltetrahydrofolate cyclohydrolase and methylenetetrahydrofolate dehydrogenase complex, and (iii) methylenetetrahydrofolate reductase were present in all GM_SE metagenomes, including the 2010 edition. This observation illustrated that in 2010 cryptic potentials for acetogenesis could have well been dispersed up to the SE community, even as the main hub of acetogens lied somewhere up in the reductive gradient closer to the venting point. On the other hand, although both 2011 and 2012 GM_SE had equally increased abundance of acetogenic genera, genes for the three sub-units of key acetogenesis enzyme acetyl CoA synthase 19 , viz. (a) acetyl-CoA synthase corrinoid iron-sulfur protein; (b) CO dehydrogenase/acetyl-CoA synthase, CO dehydrogenase subunit; and (c) CO dehydrogenase/acetyl-CoA synthase, acetyl-CoA synthase subunit, appeared only in the 2012_GM_SE metagenome, plausibly due to sheer stochasticity. Anyway, these observations certainly implied that the altered SE conditions of 2011, which appreciably lingered until 2012, promoted the growth of acetogenic bacteria. Now this population growth could have been driven by elements that were already there in subdued abundance in 2010_GM_SE and/or those which migrated from more reduced and hotter niches of 2010. Remarkably again, the apparent rise of acetogens in 2011 as well as their significant retention through 2012 coincided with identical population dynamics of methanogens and sulfate-reducing bacteria, which like the acetogens also employ the Acetyl-CoA pathway either in the direction of acetate/biomass synthesis or that of acetate degradation 19 . In this connection it is noteworthy that the summation of potential methanogens, acetogens and sulfidogens (sulfate-reducers) did not account for more than 5% of total metagenomic reads in any GM_SE sample, but their steep rise between 2010 and 2011 (followed by strong retention through 2012) was relatively significant. For example, the representation of methanogens went up from <0.01% of all classifiable reads in 2010_GM_SE to 0.21% in 2011_GM_SE and 0.14% in 2012_GM_SE, corresponding to which 6, 10 and 16 different methanogenesis-associated genes were identified in the respective metagenomes ( Table S5 ). Similarly, proportion of potential sulfidogens (which included probable members from Deltaproteobacteria , Firmicutes and Nitrospirae ) increased from 1.25% of all reads in 2010_GM_SE to almost 4.0% in 2011 and 3.3% in 2012. As envisaged for the acetogens, the observed upturn of methanogenic and sulfidogenic populations could have also been driven by elements already present in 2010_GM_SE (albeit, as minuscule minorities) and/or by immigrants from more reduced and hotter niches of 2010 (as apparent from subsequent analyses). However, simultaneous rise of these three biotypes in the same SE niche was bioenergetically intriguing since all the three metabolisms are hydrogen-requiring processes and the populations in question ought to compete with each other for hydrogen. Sulfate-reducing bacteria are known to utilize H 2 at concentrations lower than that required by methanogens and their ability to outcompete the methanogens plausibly arise from the more-positive reduction potential of SO 4 2− than that of CO 2 20 . From this point of view, the spike in methanogenic archaea in 2011_GM_SE was all the more intriguing because it happened when sulfate and sulfide contents of the SE outflow were at their three-year highest and lowest levels respectively. In other oxygen-stressed environments with sufficient quantities of SO 4 , H 2 S is reportedly the predominant reduced product, and the major fate of biodegradable organic carbon is oxidation to CO 2 (formed from the CH 3 group of acetate) 20 . Conversely, CH 4 replaces H 2 S as the main reduced product typically when SO 4 is limiting (as organic carbon is disproportionated to CO 2 and CH 4 via H 2 - or formate-using aceticlastic reactions, which form CH 4 from the CH 3 of acetate) 20 . As such, the unusual expansion of the GM_SE methanogenic population amidst copious sulfate availability could have been rendered feasible by metabolisms based on methylated substrates like methylamines, which are monopolistic substrates rapidly fermentable (to CH 4 , CO 2 and NH 3 ) by methanogens. So far as the supply of trimethylamine is concerned, it can form from betaine glycine or other related osmoprotectants 20 , which in their turn are likely to be produced profusely by geothermal bacteria to balance the osmolarity of their cytoplasm with that of the thermal fluid 21 22 . Predominance of Methanosarcinaceae among the GM_SE methanogens in tandem with the steady detection of methylamine fermentation genes (monomethylamine permease, and trimethylamine:corrinoid and methanol:corrinoid methyltransferases) insinuated the prevalence of methylotrophic methanogenesis in the community. Nevertheless, since the observed oscillation trend of acetate in the thermal discharge was opposite to that of acetogenic/methanogenic populations in the GM_SE it was quite likely that aceticlastic reactions were also rendered significantly. All these circumstantial evidences collectively hinted that H 2 was a limiting factor for the community, which again could plausibly be compensated by the formate (an equally efficient reductant for methanogens) available in the system ( Table 1 ). Anyway, even if the methanogens were out of the race for H 2 , the number of potential H 2 -utilizing processes and the count of corresponding bacteria were still significantly high in the GM_SE. As such, the community ought to have ensured sufficient and steady supply of H 2 , either from the native H 2 -generating bacteria and/or by inorganic reaction between Fe 2+ and water. So far as biotic H 2 contribution is concerned, phototrophic Proteobacteria like Rhodobacter and Rhodospirillum , perpetually predominant in the GM_SE, could transform carbon substrates like lactate, acetate, butyrate, malate, etc. (using dinitrogen, glutamate or aspartate as nitrogen source) to CO 2 and H 2 , which in their turn could also be reused for photoautotrophic growth 23 . Fermentation of organic matter could be another major energy-yielding as well as H 2 -forming process in the community. Constant occurrence of several genes for [NiFe] hydrogenases and formate hydrogenlyase indicated that mixed-acid fermentations could be in vogue in the GM_SE. Cyanobacteria like Cyanothece and Microcoleus and anoxygenic phototrophs like Rhodospirillum , Rhodobacter and Rhodopseudomonas (which ferment endogenous reserves in the dark to produce H 2 as one of the fermentation products) could significantly contribute to H 2 production via this route. Chemotrophic CO oxidation by anaerobic anoxygenic phototrophs like Rhodospirillum could be another viable source of molecular hydrogen in the GM_SE. Detection of CODH and CO-insensitive [NiFe] hydrogenase (which together catalyze the net reaction CO + H 2 O → CO 2 + H 2 ) in all three metagenomes spoke volubly in favor of this process. H 2 could also be produced substantively as a byproduct of nitrogenase-mediated N 2 fixation without involving a hydrogenase 14 . This was insinuated not only by the abundant dinitrogen-fixing rhizobia but also the various nitrogen-fixation genes detected in all three GM_SE metagenomes. Last but not the least; metagenomic data also suggested that alkaline phosphatase-mediated phosphite oxidation (H 3 PO 3 + H 2 O → H 3 PO 4 + H 2 ) could also be a viable source of H 2 in the GM_SE. Besides a large number of heterotrophic bacteria capable of growing on complex sugars and amino acids, several methylotrophic and methanotrophic genera 24 predominated the GM_SE consistently. Although the concentration of methylated compounds in the PT environment was not measured their importance as carbon source for the community was apparent from the occurrence of Methylibium , Methylobacillus , Methylobacterium , Methylotenera , Methylovorus and Methylococcus in the top-50 genera list of all three metagenomes (see Fig. 5 and Table S6 ). In addition, all the GM_SE metagenomes encompassed genes for at least one of the three subunits of methane monooxygenase, the key enzyme oxidizing the C-H bond of methane and other alkanes 25 . Taxonomic diversity flux corroborated the fluctuations in ecological diversity To get a precise picture of the taxonomic diversity (which only involves ‘what types’ and not ‘how many’) of the discrete GM_SE editions we estimated their species richness by analyzing amplified 16S rDNA fragments. V3 regions of all potential bacterial 16S rDNA present in a GM_SE metagenome was PCR-amplified using Bacteria –specific primers, but no amplicon could be obtained with Archaea –specific primers plausibly due to their extremely low abundance. The bacterial V3 amplicon pools were sequenced by Ion PGM up to such depths which ensured that plateaus of rarefaction curves were reached. Table S7 documents all data pertaining to the clustering of OTUs from the PGM reads. The GM_SEs of 2010, 2011 and 2012 encompassed 1478, 1248 and 1220 OTUs respectively ( Fig. 7 ). The species richness of 2010_GM_SE, though numerically highest among the three GM_SEs, was confined to only a few higher-level groups. In contrast, 2011_GM_SE was taxonomically most dispersed, followed by 2012_GM_SE. As such, the classifiable OTU diversity of 2010, 2011 and 2012 GM_SE was distributed over 9, 16 and 11 phyla respectively ( Fig. 7 , Table S8 ). Likewise, 12, 22 and 14 classes ( Table S9 ), and 19, 52 and 27 genera ( Table S10 ) were identifiable in 2010, 2011 and 2012 GM_SE in that order. These numbers were suggestive of an oscillation in the GM_SE taxonomic diversity in line with the trend observed for its ecological diversity. They further highlighted that higher OTU count did not necessarily mean greater taxonomic diversity if the spread of the OTUs over various higher-level taxa was included in the concept of diversity. However, since in the existing literature there is no mathematical scale to quantify the taxonomic spread of OTU sets we first had to formulate such an index and then use it to analyze potential swings in the GM_SE alpha diversity. It needs to be clarified at this juncture that this new Relative Taxonomic Diversity Index ( T r ) only gives a comparative measure of the alpha diversities of closely related communities at a chosen hierarchic level (‘phylum’ in the present case) and is not an absolute measure of taxonomic diversity of a community. As such, it can only quantify the taxonomic diversity of a community relative to other spatially- or temporally-linked communities. So far as calculating T r was concerned, first, diversity pertaining to the i th phylum within a given GM_SE edition (p i ) was estimated as a proportion of the cumulative diversity of that phylum encompassed by all three GM_SEs. Equation 1 gives the general expression for this ratio. Notably, the denominator (N i ) of the term p i represented the taxonomic scope of the i th phylum in the whole GM_SE system, and was determined as follows: (i) First, the three multifasta files that respectively contained the consensus sequences of all the OTUs of the three GM_SE editions were merged into a single multifasta. (ii) Pair-wise alignment of all the sequences of this merged file was performed, followed by their tree-based clustering at 97% sequence similarity level. The new OTU superset thus formed gave the full scope of species diversity in the whole system. (iii) Finally, the consensus sequences in this OTU superset were classified via the RDP Classifier, and those OTUs which were found to be affiliated to the i th phylum gave the taxonomic scope (N i ) of that phylum in the whole GM_SE system. After p i value was calculated for any phylum identified in a given OTU set it was multiplied by the square roots of their corresponding n i value to account for the actual species count for that phylum. The resulting products were then summed across phyla, and the summation gave the final T r value for the OTU set or GM_SE edition in question ( equation 2 ). All raw calculations involved in the current analysis are given in Table 2 . As such, T r value for 2011_GM_SE happened to be the highest, i.e. 54.3 out of a maximum possible 101.85, which was achievable only if a community edition by itself encompassed the whole diversity scope of the community system. 2012_GM_SE had the lowest T r value (28.1), while 2010_GM_SE (39.51) enjoyed intermediate status. Thus, the year on year fluctuations in the T r value of the GM_SE ( Fig. 4D ) almost mirrored the oscillations witnessed for the ecological diversity indices H and E H . The only distinction was that the T r of 2012_GM_SE was less than that of 2010_GM_SE. Although the T r values suggested an eventual loss of taxonomic diversity in the GM_SE, subsequent inclusion of the two GM_PC samples in a comprehensive OTU analysis of the entire PT GMMC revealed that a considerable portion of taxonomic diversity concentrated in 2011_GM_SE actually migrated to the PC niches in 2012 rather than being lost from the system. Constancy, resilience and flux of bacterial groups in the entire GMMC The above analyses comprehensively showed that a substantive fraction of GM_SE constituents were spatiotemporally constant while others were transitory in nature. However, the key questions that remained unresolved were the potential source(s) of so many newcomers into the GM_SE and the eventual fate of the incumbent populations that were apparently marginalized in course of community restructuring. Answers to these questions required a thorough idea about all plausible paths of dispersion and dispersal of populations as well as influx and efflux of bacteria within the entire GMMC continuum, and not just the GM_SE alone. This could be achieved only via a holistic comparison of the taxonomic diversity of the all GM_SEs and GM_PCs encountered in the PT microbiome over the three assessment years. Thus, V3 regions of all bacterial 16S rDNAs present in the two GM_PC samples were PCR-amplified and sequenced by Ion PGM up to depths ensuring plateaus in the rarefaction curves. The resultant OTU sets were then compared with the existing GM_SE OTU sets to generate comprehensive pictures of phylum ( Fig. 7 ) and genus ( Fig. 8 ) level diversities of the three GMMCs as a whole. Figure 8 additionally retraced the putative paths via which genus-level entities could have transmitted from one PT niche to another. The quotient of all demographic interrelations between pertinent mat sample pairs was determined by identifying their shared and unique OTUs ( Fig. 9A ). This was done by first merging the consensus sequences of all the OTUs of the two samples in question and then doing tree-based clustering at 97% similarity level. The doubletons of the new consensus sequence superset represented the OTUs common to the two samples. Similarly, final quotient of year on year addition, deduction and transmission of OTUs at the level of the whole GMMC was also determined ( Fig. 9B ). First, following the above principles a consensus sequence superset was created for 2010 or 2012 by taking all OTUs from the GM_PC and GM_SE samples of that year. The resultant superset, representing the total diversity of the 2010 or 2012 GMMC, was then merged with the consensus sequence set of 2011_GM_SE, and the clustering process repeated as above to get the common and unique OTUs between the GMMCs of 2010 and 2011, or 2011 and 2012 respectively. Figure 8 revealed that a number of genera were dispersed throughout the GMMC continuum (see light green-highlighted names which are common to all PC or SE mat samples in question), while others were restricted to either PC or SE microenvironments (names highlighted in crimson, orange or light blue). While Fig. 8 identified only a few genera as common to all five mat samples, Fig. 9A indicated that the total count of such consistent broad niche-width OTUs could be quite high. Holistic comparison of the five OTU sets also offered a plausible reason for the sharp hike in the relative abundance of Deinococcus-Thermus , Chloroflexi , Verrucomicrobia , Acidobacteria or Firmicutes in the GM_SE edition of 2011 as against its 2010 counterpart. Since the 2010 PC niche was significantly rich in members of these phyla (see Fig. 7 ) it was natural to conjecture that between 2010 and 2011 spatiotemporal dispersal had occurred from the PC to the SE, thereby boosting the species richness and/or abundance of such bacteria in 2011_GM_SE. Fig 9A reinforced this notion by revealing that the total count of such niche-transcending OTUs (represented by the overlap between 2010_GM_PC and 2011_GM_SE) could be as high as 397, while the total number of native GM_SE OTUs conserved over 2010 to 2011 was 464. Concurrent to this, Fig. 8 identified seven such genera in 2011_GM_SE (names highlighted in deep blue) which were not present in the 2010_GM_SE but apparently migrated to the former from 2010_GM_PC. Subsequently, it was further interesting to detect all these seven genera in 2012_GM_PC while only two of them were archived in 2012_GM_SE. This gave the inkling that a sizeable fraction of the diversity accumulated in 2011_GM_SE from the previous year’s GM_PC actually migrated back to its original PC niche once that window of opportunity revived in 2012. The last notion was buttressed by the observation that in Fig. 9A , 562 out of the total 1248 OTUs of 2011_GM_SE overlapped with 2012_GM_PC, whereas only 438 OTUs of 2011_GM_SE overlapped with 2012_GM_SE. Corroboratively, the final quotient of diversity conservation in the PC niche over 2010–2012 was also as high as 515 ( Fig. 9A ). Bacterial community structure of the mat-adjacent sediments was further studied to check whether the hypothesis of long-term survival and dispersal of the GMMC constituents was indeed true and the 2011 or 2012 mats were not constructed by similar founder bacteria from anywhere other than the GMMC. 2011_GM_SE-adjacent sediments were sampled thrice by scooping out 2.5 cm 3 of freshly-precipitated mineral deposits in each round from three discrete points around, and at least 10 cm beyond, the mat structure. The three sub-samples were mixed thoroughly and labeled as the 2011_sediment sample. In 2012, sediments around the GM_SE and GM_PC growths were sampled separately as above; subsequently the six sub-samples were mixed thoroughly and treated as the 2012_sediment sample. Total community DNA was extracted from these two composite sediment samples and V3 regions of all bacterial 16S rDNAs present therein PCR-amplified and sequenced by Ion PGM up to depths that yielded plateaus in rarefaction curves. Table S11 documents the numerical details pertaining to the clustering of OTUs from the PGM reads of the 2011 and 2012 sediment samples. Taxonomic composition of the two sediment samples was quite similar to each other and at the same time clearly distinct from any of the five green mat samples. Bulk of their OTUs belonged to the phyla Actinobacteria and Firmicutes ( Fig. 10 ), which, notably, had contributed very little to the alpha diversity of the GMMC samples (see Fig. 7 ). Most importantly, the 2011_sediment sample was found to include only one (viz. Limnobacter ) out of the 20 genera hypothesized to have been handed over to 2011_GM_SE from 2010 GM_PC and/or GM_SE (compare the sedimentary genera listed in Table S12 with data shown in Fig. 8 ). Similarly, the 2012_sediment community included only four (viz. Limnobacter , Meiothermus , Brevundimonas and Chloroflexus ) out of the 28 genera hypothesized to have been carried over to 2012 GM_PC and/or GM_SE from 2011_GM_SE (compare information given in Table S12 and Fig. 8 ). These observations, along with the fact that there was no microbial mat in the southern slope in 2010, reinforced the hypothesis that long-term survival and dispersal of 2010 GMMC constituents, and not the founder effect of similar bacteria from other mat-adjacent niches, created the framework of the 2011 and 2012 mats. Figure 9A showed 679 OTUs to be common between 2010 and 2012 GM_SEs, thereby insinuating extensive species conservation within the GM_SE itself. Intriguingly, this number was much higher than the number of OTUs common between 2010 and 2011 GM_SE (464), or 2011 and 2012 GM_SE (438). These numbers were indicative of a scenario where several species native to the SE niche were pushed below detectable levels in 2011_GM_SE as this minimally-dispersed community was overcrowded by the influx of immigrants from discrete ecological niches, potentially extending beyond the GMMC framework. Then in 2012, when a large number of species apparently vacated the GM_SE space and migrated back to the revived PC niche, some of the GM_SE components marginalized in 2011 restored their populations above detectable limits. Similar results were also observed for the two GM_PCs, where the number of 2010 OTUs recovered from the 2012 edition was higher than the number of 2010_GM_PC OTUs which could have been transmitted via 2011_GM_SE. As such, in Fig. 9A , 515 OTUs were common to the 2010 and 2012 GM_PCs, while only 397 were common to 2010_GM_PC and 2011_GM_SE. These observations were further mirrored in Fig. 8 where some of the 2010_GM_PC and 2010_GM_SE genera that had gone undetected in 2011_GM_SE resurfaced in the 2012 GM_PC and/or GM_SE samples (see the hypothetical migration paths in dotted lines in Fig. 8 ). This suggested that within 2011_GM_SE, members coming from both 2010_GM_SE and 2010_GM_PC were equally starved for space and pushed below detectable levels. The crux of the above analyses showed that a considerable portion of the 2011_GM_SE diversity was actually derived from 2010_GM_PC. At the same time, the apparent decline in the taxonomic diversity of the SE green mats in 2012 (or for that matter decline in the relative abundance of Chloroflexi , Deinococcus-Thermus and Acidobacteria in 2012_GM_SE) could be attributable to the partitioning of a substantial portion of 2011_GM_SE diversity into two discrete meta-communities. Since 2012_GM_PC was significantly richer than 2011_GM_SE in terms of the species diversity of Chloroflexi , Deinococcus-Thermus or Acidobacteria (see Fig. 7 ) it was natural to presume that between 2011 and 2012 bulk of these bacteria relocated to their preferred PC niches. In a nutshell, potential species loss notwithstanding, the overall diversity of the PT GMMC, over the three years, actually augmented from the 2010 level. The final quotient of the entire diversity dynamics of the PT GMMC is summarized below. In 2010, the GMMC harbored a sum total of 2078 OTUs, while in 2012 the number went up to 2636 ( Fig. 9B ). These numbers, in conjunction with the data presented in Fig. 7 , indicated a net increase in the alpha diversity of the GMMC, both in terms of OTU count and taxonomic spread of OTUs. Notably, the number of OTUs (618) common between 2010 and 2012 GMMC exceeded the overlaps between the GMMCs of 2010 and 2011 (409 OTUs) as well as 2011 and 2012 (451 OTUs). On the other hand, when the entire OTU set of the 2010 GMMC was compared with that of 2011_GM_SE, 1669 OTUs were unique to the former and only 839 were exclusive to the latter. The 839 OTUs detected exclusively in the 2011_GM_SE were most likely to represent foreign recruits that were not present in the PT GMMC earlier. While some idea about the generic identity of these OTUs can be made from the 32 names highlighted in yellow in Fig. 8 , it is not impossible that some of these 839 were also present in the GMMC in 2010 but were overwhelmed by the preponderant Proteobacteria . Conversely, there were several pointers demonstrating that a substantive fraction of the 1669 OTUs found to be unique to the 2010 GMMC (versus 2011_GM_SE) were not totally lost from the 2011 community but marginalized below detection limits. Remarkably again, in Fig. 9B , the number of OTUs common to 2011_GM_SE and the whole GMMC of 2012 (451) was higher than the overlap between 2010 GMMC and 2011_GM_SE (409 OTUs). This implied that many new recruits of 2011 were retained in the GMMC through 2012, a fact that concurred with the genus-level comparisons (note that in Fig. 8 , 11 out of the 32 unique genera of 2011_GM_SE were also detected in 2012 GM_PC and/or GM_SE). Comparison between 2011_GM_SE and the 2012 GMMC further showed that 797 OTUs may have been lost in transition from 2011 to 2012, while 2185 OTUs apparently joined the GMMC during that period (idea about the generic identity of these OTUs can be made from the names highlighted by bright green, white and purple in Fig. 8 ). However, the net influx between 2010 - 2012 could have actually been lower than 2185, since this number also included those 2010 OTUs which were marginalized in 2011 (note that in Fig. 9B only 2018 OTUs were unique to the 2012 GMMC in comparison to 2010 GMMC). Most importantly, net diversity addition in the GMMC over three assessment years was surely higher than the net loss represented by the 1460 OTUs unique to the 2010 GMMC in comparison to the 2012 GMMC."
} | 15,773 |
27230116 | PMC4882505 | pmc | 4,541 | {
"abstract": "3-Hydroxypropionic acid (3-HP) is an important platform chemical proposed by the United States Department of Energy. 3-HP can be converted to a series of bulk chemicals. Biological production of 3-HP has made great progress in recent years. However, low yield of 3-HP restricts its commercialization. In this study, systematic optimization was conducted towards high-yield production of 3-HP in Klebsiella pneumoniae . We first investigated appropriate promoters for the key enzyme (aldehyde dehydrogenase, ALDH) in 3-HP biosynthesis, and found that IPTG-inducible tac promoter enabled overexpression of an endogenous ALDH (PuuC) in K. pneumoniae . We optimized the metabolic flux and found that blocking the synthesis of lactic acid and acetic acid significantly increased the production of 3-HP. Additionally, fermentation conditions were optimized and scaled-up cultivation were investigated. The highest 3-HP titer was observed at 83.8 g/L with a high conversion ratio of 54% on substrate glycerol. Furthermore, a flux distribution model of glycerol metabolism in K. pneumoniae was proposed based on in silico analysis. To our knowledge, this is the highest 3-HP production in K. pneumoniae . This work has significantly advanced biological production of 3-HP from renewable carbon sources.",
"conclusion": "Conclusions In this work, we engineered recombinant K. pneumoniae strains expressing native ALDH (PuuC) under constitutive lac promoter or inducible tac promoter. Tac promoter outperformed lac promoter for the production of 3-HP in K. pneumoniae , which was evidenced by high expression of PuuC, the increased transcription of RNAP and related genes, and particularly the high ALDH activity towards 3-HPA. When a fed-batch culture was carried out under microaerobic conditions at pH 7.0 in a 5 L bioreactor, this tac -driving recombinant K. pneumoniae strain produced 73.4 g/L 3-HP with a cumulative yield of 52% on glycerol carbon and 1.53 g/l/h productivity in 48 h. Blocking the lactic acid and acetic acid synthesis genes slightly repressed cell growth, but elevated 3-HP titer to 83.8 g/L in 72 h. Given the efficient expression system and the obvious advantages of K. pneumoniae as a host, including super ability to utilize glycerol, natural ability to generate cofactor B 12 , and low consumption of IPTG, we believe that this work has substantially advanced biological production of 3-HP.",
"discussion": "Discussion Here we developed an efficient system for biological production of 3-HP in K. pneumoniae . The success of this system attributed to the concert of promoters, key enzymes and fermentation processes. It should be noted that the IPTG-inducible tac promoter was shown to enable high expression of PuuC and thus 3-HP overproduction in K. pneumoniae . In a 5 L bioreactor, the recombinant K. pneumoniae strain overexpressing PuuC under tac promoter produced 73.4 g/L 3-HP in 48 h with 52% glycerol conversion ratio and 1.53 g/L/h productivity on glycerol. To reduce the formation of byproducts, the lactic acid and acetic acid synthetic genes were eliminated. Although the resultant mutant strain compromised growth ( Fig. 5 ), it produced 83.8 g/L 3-HP in 72 h. To our knowledge, this is the highest 3-HP titer reported thus far. Below are the reasons behind this high production. The first reason lies in the biochemical attributes of K. pneumoniae. In K. pneumoniae , glycerol conversion to 3-HP involves only two sequential reactions. Presumably, when glycerol is the sole carbon source, the 3-HP biosynthesis is actually the core metabolism because glycerol can be easily converted to 3-HP when a high-activity ALDH is available. In contrast, the 3-HP biosynthesis from glucose undergoes at least four reactions 10 . It is conceivable that manipulating multiple enzymes usually leads to metabolic flux imbalance and thus increases burden on the host. In addition to the difference of pathway steps, K. pneumoniae manifests striking attributes such as remarkable capacity to metabolize glycerol, active cell proliferation, and native ability to synthesize vitamin B 12 . These advantages empower K. pneumoniae to be a promising host for the production of 3-HP. For instance, since vitamin B 12 is the cofactor of glycerol dehydratase, K. pneumoniae is obviously superior to E. coli which needs the addition of vitamin B 12 to the fermentation medium. More importantly, the aggressive cell growth, powerful capability to metabolize glycerol, and high-activity of ALDH in K. pneumoniae , jointly provide a powerful driving force that pushes glycerol towards 3-HP biosynthesis. Although Lactobacillus reuteri also naturally produces 3-HPA 22 , its growth is significantly slower than K. pneumoniae , thereby increasing the production cost. In this study, K. pneumoniae grew actively due to vigorous consumption of glycerol. A total of 1553 mM glycerol was consumed with 82% glycerol conversion to valuable chemicals including 3-HP, 1,3-PD and 2,3-BD ( Table 2 , Fig. 4 ). The second reason behind high 3-HP production is the effective expression system developed in this study. Previously, the lack of appropriate promoter restricts ALDH expression and accordingly hinders the conversion of 3-HPA to 3-HP. The importance of this study is the finding that tac promoter is functional in K. pneumoniae . This is evidenced by PuuC overexpression and enhanced PuuC activity (26.31 U/mg) compared with the recombinant strain recruiting lac promoter, whereby the PuuC activity was only 7.66 U/mg. Due to multiple copies of native ALDHs in K. pneumoniae genome, wild type K. pneumoniae also exhibited limited ALDH activity (3.82 U/mg) ( Fig. 3B ). To clarify the interplay between tac promoter and transcription machinery, quantitative analysis of RNAP was performed. We found that the transcription levels of RNAP and the five related genes were increased ( Fig. 3C,D ). RNAP is known to be consisted of sigma subunit and core enzyme which includes α, β, β′ and ω subunits forming a complex and encoded by rpoA , rpoB , rpoC and rpoZ , respectively 23 . It is well known that the rpoC -encoded β′ subunit mediates RNAP assembly. In this present study, we found that the rpoC gene in tac -driving strain showed notable transcription ( Fig. 3C ), indicating its likely involvement in RNAP assembly. Besides, sigma factors rpoS and rpoE were also strongly transcribed, implying that they may bind to tac promoter and initiate PuuC expression. In fact, tac promoter is hybridized by trp and lac UV5 promoters, and the affinity of tac with rpoS has been reported 16 . In view of this, we concluded that sigma factors might contribute to tac activity. Apart from RNAP, the five RNAP-related genes were also significantly transcribed. This can be explained by the fact that nusA , greA and greB govern transcription elongation. Another interesting finding in this study was that dksA showed similar transcription level to rpoS ( Fig. 3D ), which is consistent with the previous report that dksA induces rpoS at translational level 24 . The third reason behind the high production of 3-HP may be the fermentation conditions, including IPTG concentration, pH value and medium composition. To maximize PuuC expression, IPTG concentration was optimized ( Fig. S1 ). Compared with the IPTG concentrations used for triggering gene expression in other bacteria, only 0.02 mM IPTG (4.7 mg/L) was used in this study, clearly indicating low production cost. Indeed, 0.5 mM, 1 mM or 2 mM IPTG was usually used for tac- driving gene expression in E. coli 16 and up to 5 mM IPTG were used for tac- driving gene expression in Zymomonas mobilis and Pseudomonas putida 17 25 . Due to the 3-HP accumulation in fermentation broth, pH value decreased and cell growth nearly ceased. To alleviate the stress of 3-HP, pH value was maintained at 7.0 by automatic addition of NaOH, which benefited cell growth and facilitated 3-HP formation. To further promote cell growth, fermentation medium was optimized, and the optimum medium contains a little more nitrogen sources compared with those previously reported ( Fig. S2 ) 26 27 . Owning to above measures, both biomass and 3-HP titer were enhanced. Given that the PuuC activity (26.31 U/mg) was still relatively low compared with those previously reported 7 , 3-HP titer can be enhanced in the future work by taking a suite of measures, including employment of a high-activity ALDH, alleviation of feedback inhibition, and further optimization of fermentation conditions. Collectively, all above results point to the fact that the biosynthesis of 3-HP depends on multiple factors, including cell growth, substrate provision, enzymatic activity, cofactor availability, redox balance, and cell tolerance to substrate and metabolites. For example, 3-HP production was shown to be closely coupled with cell proliferation especially at exponential phase 19 , and glycerol dissimilation is mediated by the dha regulon which steers parallel glycerol oxidation and reduction pathways. Interestingly, there exists a tradeoff between these two pathways. Intensifying the reduction pathway will simultaneously alter the flux toward oxidation pathway. Of the enzymes that execute glycerol dissimilation, DhaT, GDH, and DhaD, are three vital enzymes 10 . Their activities ascribe to multiple factors, including promoter strength, redox cofactors, substrate provision, and metabolite inhibition. Due to the buildup of metabolites including 3-HP, lactic acid and acetic acid, together with the increasing ionic strength, the cell tolerance to organic acids has emerged as a determinant for 3-HP production. Provided that the yield of metabolites could be considered as a quantitative trait controlled by multiple factors, it is conceivable that systematic optimization strategy is efficient for improving 3-HP production. Given the high 3-HP titer, high activity of tac promoter, low production cost, as well as novel insights into glycerol metabolism, we believe that this study is a big stride towards industrial production of 3-HP and polyesters consisting of 3-HP monomers 28 . More broadly, the efficient expression system developed here could be extended to the overproduction of 1,3-PD and 2,3-BD, both are top-valued bulk chemicals native to K. pneumoniae ."
} | 2,593 |
29346651 | PMC5888944 | pmc | 4,544 | {
"abstract": "Abstract Extensive microbial gene flows affect how we understand virology, microbiology, medical sciences, genetic modification, and evolutionary biology. Phylogenies only provide a narrow view of these gene flows: plasmids and viruses, lacking core genes, cannot be attached to cellular life on phylogenetic trees. Yet viruses and plasmids have a major impact on cellular evolution, affecting both the gene content and the dynamics of microbial communities. Using bipartite graphs that connect up to 149,000 clusters of homologous genes with 8,217 related and unrelated genomes, we can in particular show patterns of gene sharing that do not map neatly with the organismal phylogeny. Homologous genes are recycled by lateral gene transfer, and multiple copies of homologous genes are carried by otherwise completely unrelated (and possibly nested) genomes, that is, viruses, plasmids and prokaryotes. When a homologous gene is present on at least one plasmid or virus and at least one chromosome, a process of “gene externalization,” affected by a postprocessed selected functional bias, takes place, especially in Bacteria. Bipartite graphs give us a view of vertical and horizontal gene flow beyond classic taxonomy on a single very large, analytically tractable, graph that goes beyond the cellular Web of Life.",
"conclusion": "Conclusion Our work complements the findings of other recent studies using networks. Cong et al. (2017) , for example, analyzed genetic exchange communities, by decomposing a k-mer based LGT network using notions of graph theory. They inferred functionally biased LGT between cells from various prokaryotic data sets, with up to 144 chromosomes. Our strategy was similar to this inspiring work, but aimed at a different goal. In contrast with ( Cong et al. 2017 ), we represented a diversity of sharings (vertically transmitted genes, horizontally acquired genes, or simply shared genes when the mode of inheritance could not be determined), between a diversity of hosts (between cells, between cells and MGE, and between MGE). In that regard, we did not attempt to reconstruct genetic exchange communities of prokaryotes, united by LGT, but rather to offer a broader picture of CHG distribution across microbial genomes. This was achieved by two means: first, we used a broader data set than Cong et al. (2017) , since a total of 8,214 genomes (7,832 MGEs and 382 chromosomes) were included in the network. Second, we used a different type of network, that is, a genome-CHG bipartite graph. In our analyses, we further focused on genes shared between extrachromosomal elements and chromosomes (externalized genes), because these types of sharings would not be necessarily detected by the identification of LGT between cells, whereas our work shows that the sharing of genes between chromosomal and extrachromosomal elements constitutes an important process of microbial communities, which is also functionally biased. Namely, homologous genes which are present, after selection, on both chromosomal and extrachromosomal elements have biased functional profiles, in particular a high turn-over of genes associated with the L category, because genes from this COG category, whereas overrepresented in the pools of conserved externalized genes are less abundant in the pools of divergent externalized genes than genes with, for example, the “M” and “T” functions. Thus, genes that are recently externalized are not enriched in metabolism, regulation, and transport, contrasting with the genes that Cong et al. reported as overrepresented in LGT within genetic exchange communities. Therefore, our work underscores that two distinct introgressive processes affect the evolution of microbial communities, that is, gene externalization and LGT. Our findings also strongly echoe with the remarkable study by Popa et al. (2017) . These authors used a bipartite CHG-genome network (in their case, directed between donors and hosts), with 3,982 prokaryotic genomes and phages genomes (including prophages of prokaryotic chromosomes, individualized as nodes in their network) to investigate lateral gene transfer by transduction. These authors report that transduction is mostly restricted between closely related donors and recipients, consistent with experimental observations and with our observation that viral genomes are generally peripheral on the genome network, that is, associated with specific cellular host lineages ( Halary et al. 2010 ), which is also confirmed in our present work with an expanded data set. Interestingly, Popa et al. accurately decomposed transduction events in two phases: the uptake of a gene from a donor chromosome into a phage, and the acquisition of a gene from a prophage into a recipient chromosome. According to our own terminology, each of these phases corresponds to an externalization event (one from a cell to a virus, and another from a virus to a cell, respectively). For that reason, the edges in their network correspond to polarized externalization events between prokaryotes and phages. Interestingly, Popa et al. (2017 ) stress that 9% of the transduction events they detected are autologs, that is, duplication of genes mediated by mobile DNA vectors. This autology is consistent with our claim that externalization introduces genetic redundancy (and possibly resilience) within microbial communities. Of note, our work further underscores that the process of externalization (i.e., the sharing of genes on different types of supports, extrachromomal, and chromosomal elements) is very general. We report that externalized genes amount to even higher proportions of the chromosomes (i.e., >60% in Starkeya novella , which are not due to inserted prophages) than the proportion of transferred genes detected by Popa et al. (who inferred 15,298 edges between 2,573 bacteria and 4,650 phages). This is logical, because only events of gene externalization 1) beginning in a chromosome and ending in a chromosome, 2) mediated by a phage, were described in their directed bipartite networks of chromosomes and phages ( Popa et al. 2017 ). By contrast, in our analysis, the first constraint is alleviated, and plasmids, rather than phages, appear as the main carriers of externalized genes (85–99% of externalized genes match on plasmids at ≥30% ID, see supplementary fig. S9 -3, Supplementary Material online). To summarize, we introduced an analysis of gene sharing between chromosomal and extrachromosomal elements from the microbial world based on a bipartite graph. This strategy allowed us to independently recover major trends of microbial evolution (i.e., the taxonomical and typological biases in the pattern of gene sharings, the existence of genetic worlds, in particular in networks built at low stringency thresholds, the promiscuity of transposases), to propose some novel HGT candidates, and to bring forward the general process of gene externalization. We thus show that the web of life is disconnected, with major prokaryotic kinds, largely, but not absolutely isolated from one another in terms of gene sharing, surrounded by spider-webs of mobile genetic elements, with which chromosomes share externalized genes. Moreover, transposases have recently run across this web of life. Consequently, bipartite graphs appear as a powerful way to study the processes of microbial life beyond classic taxonomy and customary genomic analyses, and we propose that beyond coding CHG, the use of bipartite graphs could be further generalized to small RNA families-genome networks and gene families-metagenome networks, and applied to even larger data sets to keep up with the impressive accumulation of molecular sequences. Finally, gene families-genome-metagenome tripartite graphs constitute an exciting horizon for expanded multilevel analyses.",
"introduction": "Introduction A major problem for biology is to understand short and long taxonomical range sharing of genes, be they acquired by vertical descent or introgression ( Beiko et al. 2005 ; Kunin et al. 2005 ; Puigbo et al. 2010 ; Andam and Gogarten 2011 ; Kloesges et al. 2011 ; Popa et al. 2011 ; Smillie et al. 2011 ; Cong et al. 2017 ). Another problem is to acknowledge and evaluate the role of viruses (and other mobile genetic elements) in the evolution of cells ( Simmonds et al. 2017 ). Considered from a genetic perspective, mobile elements impact cellular evolution first by moving genes into cells and this often results in a kind of paralogy within microbial communities, when a gene copy is carried on the genome of a mobile element as well as on a chromosome. This process can create new phenotypes ( Busby et al. 2013 ), opportunities for the coming of additional genes in genomes ( Roux et al. 2015 ), and contribute to the resilience of microbial communities by dispersing physiologically important genes on multiple vectors ( Sullivan et al. 2010 ; Biller et al. 2015 ). Moreover, the number of prophages inserted in prokaryotic genomes (and their biological impact on their hosts) is likely underestimated ( Roux et al. 2015 ). However, we have a poor framework to study dynamics of gene flow between mobile elements and cells. Here, we exploited a novel approach to study gene sharing in the microbial world, which expands over the numerous approaches currently being developed. These latter display some limits when it comes to study together gene sharing between mobile genetic elements and cells. Tracking the multitude of transfer paths (be they vertical or horizontal) is difficult but important, especially since sets of genes can be transferred together for functional reasons ( Jain et al. 1999 ). On the one hand, phylogenetic methods such as the one using highways of gene sharings ( Kunin et al. 2005 ; MacLeod et al. 2005 ; Beiko et al. 2006 ; Dagan and Martin 2006 ) can be used to investigate these genes movements, yet phylogenetic approaches are limited to related entities, that is, related genomes diverged from a last common ancestral genome, which belong to a monophyletic group. The simultaneous study of viruses, plasmids, plasmids and viruses, viruses and cells, plasmids and cells together remains therefore difficult. On the other hand, binary matrices ( Nelson-Sathi et al. 2015 ), similarity networks, such as genome networks ( Fondi and Fani 2010 ; Halary et al. 2010 ; Tamminen et al. 2012 ) and bipartite gene–genome networks analyzed with heuristic community detection method ( Iranzo, Koonin, et al. 2016 ; Iranzo, Krupovic, et al. 2016 ), have provided exciting results. Such networks have been used to test hypotheses about the phylogenetic or environmental drivers of genomic diversity ( Kloesges et al. 2011 ; Cheng et al. 2014 ; Forster et al. 2015 ) and the selective advantages of introgressed genes ( Bapteste 2014 ). These approaches can in principle include all key evolutionary players in a common framework (although so far the vast majority of genome networks, binary matrices, and bipartite graphs have been limited to either chromosomes; Kloesges et al. 2011 ) or genomes of mobile elements ( Lima-Mendez et al. 2008 ; Desnues et al. 2012 ; Yutin et al. 2013 ). These binary matrices are mathematically equivalent to bipartite graphs; however, it is more natural to speak in terms of graph when using tools from graph theory. Bipartite graphs consist of nodes of two fundamentally different kinds. These nodes are connected by edges such that two nodes on either end of an edge are never of the same kind. Bipartite graphs have been successfully used to explore gene–disease relationships ( Hwang et al. 2008 ), the evolution of malaria parasites ( Larremore et al. 2013 ), recipe ingredient data sets ( Ahn et al. 2011 ), and social media networks ( Murata 2009 ). In the context of gene flow, cluster of homologous genes (CHG)–genome networks ( fig. 1 , panel A) can reveal, like binary matrices, which exact groups of homologous genes are shared exclusively by certain groups of genomes (a pattern called “twins” in graph theory, formally defined as sets of CHG nodes that have identical connectivity to genome nodes). Network analyses can also reveal “articulation points,” that is, CHG nodes that are connected to parts of the graph that otherwise share no CHG nodes.\n Fig . 1. Construction of the bipartite graphs and identification of the twins and articulation points. ( A ) Construction of the genome-CHG bipartite graphs. Top nodes represent genomes of cells and mobile genetic elements. Bottom nodes represent CHG: we display the corresponding connected component of the sequence similarity network (see Materials and Methods), edge color (from yellow to red) indicates increasing % ID. ( B ) Bottom twins and articulation points: bottom nodes forming a twin class and their incident edges are drawn in the same shade of gray. Nodes 7 and 8 have the same neighbors (nodes 1 and 2) thus form twin class 1. Twins 2 and 4 are trivial since they contain only one node. Node 9 is an articulation point since its removal disconnects the graph. Specifically, bipartite graphs can be used to track sets of genes that have been laterally transferred together, genes that unite genomes that otherwise have no homologs in common at a given threshold, and also to uncover biases in transfer and/or retention of genes between mobile elements and cells. Importantly, bipartite graphs can be analyzed in two complementary ways ( Barber 2007 ; Alzahrani and Horadam 2016 ; Iranzo, Krupovic, et al. 2016 ; Jaffe et al. 2016 ). As in ( Iranzo, Koonin, et al. 2016 ), gene–genome networks can be partitioned using heuristic methods of community detections. They can also be decomposed exactly as in ( Jaffe et al. 2016 ), avoiding a heuristic treatment of the graph. In this paper, we looked at twins and at articulation points ( fig. 1 , panel B). These patterns have indeed the interesting property to be uniquely defined (unlike communities that can strongly depend on the clustering algorithm that has been used), whereas uncovering already interesting biological phenomena. We applied this approach to 8,214 genomes, thereby extending a former analysis of gene sharing between cells and mobile elements, that identified genetic worlds ( Halary et al. 2010 ). On the one hand, the resulting bipartite web of life effectively generalizes conclusions from microbial evolution. It shows that the web of life is composed of phylogenetically distinct elements, which are genetically intertwined, according to detectable rules: genes are primarily shared between groups of closely related genomes (i.e., taxonomically consistent groups) and between groups of genomes with the same type (i.e., typologically consistent groups, for example, phages with phages and plasmids with plasmids). It also shows that transposases navigate across the branches of the web. The greatest influence on gene sharing was host type with many prokaryotic (cellular) and MGE (acellular) kinds characterized by exclusive gene contents. Moreover, using graph compression, we analyzed “gene externalization” ( Corel et al. 2016 ), a situation which occurs when a CHG is present on at least one extrachromosomal element and at least one chromosome. This observation is different from lateral gene transfer between two cells, because gene externalization occurs between otherwise completely unrelated genomes, that is, viruses, plasmids, and prokaryotes, which do not show a single last common ancestor. We unraveled that gene externalization was especially significant among Bacteria, and mainly driven by gene function, illustrating strong biases in the kind of genes that persist on multiple vectors in this kind of prokaryotes.",
"discussion": "Results and Discussion We initially constructed bipartite graphs from a data set of 382 prokaryotic genomes and 7,832 mobile element genomes. This family-based data set was carefully selected to avoid the sequencing bias in microbial genomics toward Bacteria. In addition, the resulting data set size is amenable to BLAST-based sequence-similarity analyses, and extends over a former study of gene sharing between cells and mobile elements (see supplementary fig. S1 , Supplementary Material online for a genome network updated with respect to Halary et al. 2010 ). In these bipartite graphs, the “top” nodes correspond to the genomes and the “bottom” nodes correspond to CHG, defined at various stringencies ( fig. 1 , panel A). The stringency parameters of minimum percentage of identity in sequences allowed us to focus, for example, on recent gene family transmissions (i.e., when two sequences could be aligned over ≥ 80% of their mutual length, and were ≥ 95% identical in sequence; ≥95% ID for short). Varying stringency parameters allowed us to consider a variety of evolutionary time scales (see Materials and Methods). Still, the criterion of ≥ 80% mutual cover, critical to identify homologous genes, typically filtered out information about partial similarity, that is, between recombinant gene forms, such as fused genes or remodeled genes ( Jachiet et al. 2014 ; Méheust et al. 2016 ). This means that our estimates of genetic sharing and the proportion of genes externalization presented below are conservative, restricted to the identification of full-sized (externalized) genes. An undirected edge connecting a top and a bottom node indicated that a member of a CHG was found in a genome. Thus, these graphs include simultaneously several levels of organization (showing genes and genomes), and several agents (showing chromosomes and plasmids or viruses). Hence, they provide information about the distributions of CHGs across a broad range of genomes. As in any inference of comparative genomics however, the distribution of homologs across taxa only approximates actual gene exchanges, possibly because of the size of our sample, but also since intermediate unknown players are likely (see, for instance, those discovered lately in Hug et al. 2016 ). Despite these limits, the structure of these graphs is already very informative. The network is amenable to standard structural analyses. The CHG sizes follow a power-law-like distribution ( supplementary fig. S2 -1, Supplementary Material online). The node degree distribution in these graphs differs according to the node type, consistently with ( Iranzo, Krupovic, et al. 2016 ). The degree of a genome represents the part of the genome that is shared at the given similarity level. The degree of a CHG represents the number of genomes in which a given CHG is found. The degree distribution of CHGs seems to display a power law behavior, whereas genomes show a different degree distribution. Viruses and plasmids display a subpower law with far less small degree nodes, and a two-mode bump likely inherited from the bimodal size distribution of both types of MGEs. This last feature can also be observed in the results reported by ( Iranzo, Krupovic, et al. 2016 ). This trend is moreover stable with the size of the data set ( supplementary fig. S2 -2, Supplementary Material online). Beyond these topological features, the network is also amenable to analyses that help understanding microbial evolution. Connected Component Analysis Reveals Groups of Genomes with Exclusive Gene Pools For each network at a given stringency threshold, we first enumerated all its connected components (CCs), that is, all sets of nodes for which there is always an interconnecting path. These CCs represent groups of genomes associated with an exclusive pool of CHGs, that is, a CHG found in a CC is by definition absent in any other CC. The robust recovery of multiple CCs (522 in the graph at ≥ 95% ID and 156 in the graph at ≥ 30% ID, also see supplementary fig. S1 , Supplementary Material online) is consistent with the genetic worlds identified in ( Halary et al. 2010 ), albeit with a now much larger data set and a different network approach ( supplementary fig. S1 , Supplementary Material online). The discrete nature of this graph suggests a discontinuity in vertical and horizontal transmission of full-sized genes between genomes belonging to different CCs. This barrier may reflect phylogenetic isolation, ecological isolation, the use of an alternative genetic code or quite simply the nonexhaustive data set of genes and genomes at our disposal. For example, the CC in supplementary figure S2 , Supplementary Material online, illustrates the case of the Spiroplasma phages, which are characterized by the alternative use of the codon UGA to encode Tryptophan instead of “STOP” (i.e., the Mycoplasma/Spiroplasma code). The taxonomic homogeneity of this CC suggests that these phages have been exclusively sharing a unique pool of genes, in effect privatized by their own lineage ( McInerney et al. 2011 ). Note that, conversely, members of a given CC do not necessarily directly share a CHG, meaning that even genomes belonging to the same CC are not necessarily connected by vertical or horizontal gene transmission. Indeed, our bipartite graphs also display a giant CC (gCC), encompassing 6,362 (i.e., 80.1%) genomes and 80,136 (99%) CHGs (at ≥ 30% ID) ( supplementary table S1 and fig. S4-1 , Supplementary Material online). This single gCC include genomes that have no homologous genes in common, yet participate in a giant network of gene sharing. The ability to reconstruct such a pattern is a significant advantage associated with the use of bipartite graphs. Twin CHG Are Likely Genetic Public Goods To understand the coinheritance of CHGs, and also to provide us with a tool for the study of phenotype evolution when phenotypes are not associated with a single monophyletic taxonomic group, we analyzed each CC, including the gCC, at a more fine-grained level, by enumerating all the twins of bottom nodes (BT) within these connected components ( fig. 1 , panel A). Twins are nodes with identical sets of neighbors in a graph. BTs represent CHGs that are exclusively present in exactly the same set of genomes. Therefore, a group of genes that are cotransmitted, vertically and/or horizontally (from a common ancestor or via LGT) within a club of genomes, will be detectable as a BT (above a sufficient similarity level, see supplementary fig. S3 , Supplementary Material online and below). We verified that the compositions of our BTs were significantly different from the ones expected from random networks (empirical adjusted p value = 4.76 × 10 −3 , on 1,165 simulations with permuted genome attributions for genes, see Materials and Methods) and robust (see proportions for different subsets of data in supplementary fig. S4 , Supplementary Material online). This observation confirmed that there is an evolutionary structure in the network, but it cannot be used to determine the respective strengths of the vertical and horizontal modes of inheritance. Detecting individual or sets of CHGs shared by many genomes with otherwise totally distinct gene contents (at a given similarity threshold) is essential to track long-distance horizontal gene transmission across the web of life. We demonstrated that this detection can be achieved via exact graph compression. We reduced the bipartite graph by grouping together BT nodes into bottom metanodes, and simplified it further by removing all BTs that were present in only one genome (see Materials and Methods). We did not further exploit here the information regarding the number of paralogs within the CHG contributing to the metanodes. Notably, the result of this graph reduction is unique and robust, that is, it does not depend from the order in which twins are merged. This merging produces a quotient, BT-free, bipartite graph with no loss of information due to this compression. It is then trivial to enumerate all bottom articulation points (BAPs) in such reduced graphs, that is, all nodes whose removal would increase the number of connected components. Although strictly topological, the notion of BAPs could in principle help to detect public genetic goods ( McInerney et al. 2011 ), that is, genetic material that is being shared by taxonomically distant genomes, which possibly benefit from the properties these shared genes confer, for some other reason than genealogy (i.e., genes coding for environmental adaptation or others hitch-hiking with them…). Around 16% of the articulation points in the network at ≥ 30% ID (and up to 71% at ≥ 90% ID) were proposed as horizontally transferred according to our horizontality test with majority as decision rule (Materials and Methods). We report for instance the case of the 3′-phosphoadenosine 5′-phosphosulfate sulfotransferase (PAPS reductase)/FAD synthetase CHG, shared by the Gram-positive bacterium Ruminococcus bromii and the Gram-negative bacterium Fibrobacter succinogenes , forming a BAP in our graph at ≥ 90% ID. This CHG encodes an enzyme with the rare ability to store two electrons without the need for cofactor or prosthetic groups, which likely enhances the success rate of transfer for this CHG in the rumen ( supplementary fig. S5 , Supplementary Material online). At a stringency of ≥ 90% ID, 56 BAP nodes (out of 811 BAP nodes) encompass transposases which, as the graph suggests ( supplementary tables S3 and S4 , Supplementary Material online), possess the capability to move across distantly related genomes ( Hooper et al. 2009 ). The Topology of the Web of Life Shows Patterns of Gene Transmission Simple graph patterns in a CHG-genome network subjected to exact decomposition are already sufficient to provide abundant biological information. Detecting recurrent patterns in the compressed bipartite graphs of this data set of 382 prokaryotic genomes and 7,832 mobile elements (3,613 viruses and 4,219 plasmids) confirmed prior knowledge about vertical or horizontal gene transmission, whereas extending these conclusions to a more comprehensive data set. More precisely, a single analysis of CC and BT analyses ( supplementary figs. S6-1 and S6-2 , Supplementary Material online, also supplementary tables S1–S7 , Supplementary Material online) generalize observations about several rules of gene transmission, that had been made on separate studies ( Beiko et al. 2005 ; Kunin et al. 2005 ; MacLeod et al. 2005 ; Hooper et al. 2009 ; Halary et al. 2010 ; Puigbo et al. 2010 ; Kloesges et al. 2011 ; Schliep et al. 2011 ; Smillie et al. 2011 ; Desnues et al. 2012 ; Tamminen et al. 2012 ; Busby et al. 2013 ; Yutin et al. 2013 ; Iranzo, Koonin, et al. 2016 ; Iranzo, Krupovic, et al. 2016 ; Jaffe et al. 2016 ; Popa et al. 2017 ). First, the vast majority of CCs and BTs were composed of genomes consistent by type of hosts, for all stringency thresholds. For example, at ≥ 90% ID, 94.26% of the CCs, and 92.01% of BTs showed gene sharing between genomes of the same type (i.e., either exclusively cellular, exclusively viral or exclusively plasmid, see fig. 2 ). In addition, not only were the vast majority of CCs and BTs consistent with genome type but the constituent genomes were also generally taxonomically consistent ( supplementary fig. S6 -1 and tables S1 and S2, Supplementary Material online), for all stringency thresholds. For example, at ≥ 90% ID, 78.5% of the CCs, and 99% of the BTs that contain prokaryotes (i.e., 19% of all BTs) showed gene sharing among members of the same phylum (as defined independently by the NCBI taxonomy). Overall, this strong taxonomic signal reflects the fact that distantly related genomes have rather different gene contents, which is consistent with the relatively independent evolution of various kinds of cellular organisms in the web of life ( Halary et al. 2010 ). We confirmed this disconnection of the living world (in terms of gene content) by plotting the distribution of CHGs across taxa using a heatmap, constructed on the sole criterion of ≥ 80% mutual coverage (see supplementary fig. S11 , Supplementary Material online), in order not to increase the differences between taxa, which could artifactually happen at high stringency thresholds when one considers that two genomes display different gene contents, whereas their seemingly different CHGs are simply divergent groups of related sequences belonging to common ancient gene families. This approach showed that genomes from different lineages use genuinely different sets of genes, and that very few CHGs are shared widely between prokaryotes (in agreement with Ku et al. 2015 ). This approach was also used for the identification of Exclusively Shared CHGs (ESCHG), see below).\n Fig . 2 Support composition of twins and CHGs. The taxonomy of the carriers of the genes in a CHG is called its support composition. For every similarity threshold (on the x -axis), we report how many CHGs (on the right) or twins of CHGs (on the left) are exclusively found in Archaea, in Bacteria, in both kinds of prokaryotes (“mixed prokaryotes”), in viruses only, in plasmids only, in both kinds of MGE (“mixed MGE”), or in both cellular and mobile elements (“externalized”). Second, within these major taxonomic lineages, genomic evolution is highly reticulated. We verified this by computing, for each group of closely related genomes in this data set, the number of CHGs that are shared by members of this group and exclusively by them. The size of these ESCHG amounts to the percentage of BTs (i.e., sets of exclusive CHGs) found exclusively in genomes belonging to the taxonomic group. This notion differs from that of core genome (at the given similarity level), since ESCHGs are not present outside this taxonomic group and the core genome can include CHGs that are also present in other groups from different lineages (e.g., housekeeping genes). Notably, categories featuring substantial relative amounts of core gene families are environmentally ( Thermococci , Halobacteria ) or metabolically ( Methanococci , Cyanobacteria ) specific lineages. The fraction of these ESCHG that are core is typically small ( fig. 3 ): at most ∼18% for the 15 Methanococci and the 15 Thermococci , and <5% for the 16 Cyanobacteria , and 25 Halobacteria contained in our data set.\n Fig . 3 ESCHG and core. The “shared pangenome” of a lineage is composed of all CHGs that are shared by at least two genomes and contain at least one member of the lineage. ( A ) For each taxonomic group, we report the percentage of CHGs forming the shared pangenome that are core exclusive, exclusive but noncore and not exclusive. ( B ) For each taxonomic group, we report the percentage of the shared pangenome that is comprised of core exclusive, and core but not exclusive CHGs. The total height of the bar represents the core itself. Thus, most BTs are not associated with all genomes of a taxonomic group, consistent with a high turnover of CHGs, at least of lineage-specific genes, within prokaryotic genomes. Jackknife analyses show that these trends are robust with respect to the size of the data set ( supplementary fig. S4 and tables S5–S7, Supplementary Material online). We also implemented a conservative horizontality test, which exploits the network information to determine which BTs have likely been horizontally transferred between cells (see Materials and Methods). About 80% of the taxonomically consistent BTs and 61% of the taxonomically inconsistent BTs at ≥70% ID were considered as laterally inherited for the majority decision rule ( supplementary fig. S7 , Supplementary Material online). These numbers represent a minimum estimate of horizontal gene transfer, since the absence of evidence for transfer for the other BTs does not mean that these other BTs were necessarily vertically inherited. They could also have been transferred between close relatives. Our estimates of transferred genes/genomes are consistent with some published estimates found in the literature, although possibly a bit more conservative. Overall, our analysis supports the generally admitted notion that horizontal gene transfer deeply impact microbial evolution ( Raymond et al. 2003 ; Zhaxybayeva et al. 2006 ; Kloesges et al. 2011 ; Popa et al. 2011 ; Koonin 2016 ), with rates that vary across genomes ( Koonin et al. 2001 ; Dagan et al. 2008 ; Kloesges et al. 2011 ). Thresholding our networks allow us to compare our results with reports on both recent HGTs, as well as on cumulative HGTs. For example, Kloesges et al. (2011) reported that 9.6% of the genes within a prokaryotic genome were recently acquired, Lawrence and Ochman (2002 ) reported 18% of genes recently acquired by HGT in E. coli , or Hernández-López et al. (2013) proposed that up to 25% of core genes were recently transferred in Rickettsiales. Consistently, we proposed that up to 15% of the genes of the tested genomes have been recently transferred (i.e., showing ≥ 90% similarity between donor and hosts genomes, supplementary fig. S7 -1, Supplementary Material online). At ≥ 30% similarity, cumulative effects of HGT become noticeable ( supplementary fig. S7 -1, Supplementary Material online), affecting in average 41.73% of the genomes and up to 68.34% of a genome ( Starkeya novella DSM 506). These values are in the same range that the ones suggested by other publications, that is, that on an average, at least 81% ± 15% of the genes in each prokaryotic genome were involved in HGT during their history ( Dagan et al. 2008 ), and see also ( Kloesges et al. 2011 ), who reported that 75% of the genes of a genome were on an average affected by one HGT, and ( Koonin 2015 ), who suggested that 60% of the information flux between prokaryotic genomes is not tree-like. Our values are however a bit higher than analyses by (Snel et al. 2002 ; Beiko et al. 2005 ; Kunin et al. 2005 ), that reported HGT rates varying from 20% to 39% of the prokaryotic gene families analyzed by phylogeny. We verified with an expanded data set (Materials and Methods) that the trends described earlier were robust. Namely, we observed the following: 1) taxonomically homogeneous CCs (only ≥90% ID) and taxonomically homogeneous BTs (33–66% of all BTs, depending on the identity threshold); 2) taxonomically heterogeneous CCs (the gCC and up to 19 CCs at ≥95% ID) and taxonomically heterogeneous BTs (7–38% of all BTs); 3) typologically homogeneous CCs (from 455 at ≥95% ID to 151 at ≥30% ID) and typologically homogeneous BTs (70–83% of all BTs); 4) typologically heterogeneous CCs (the gCC and up to 42 at ≥95% ID) and typologically heterogeneous BTs (16–29% of all BTs). We also detected a broad range of externalized genes in all these prokaryotic genomes ( supplementary fig. S5 -3 and S5-4, Supplementary Material online). However, detailed analysis of this broader data set was out of the scope of the present paper. Compressing Bipartite Graphs Detects Novel Instances of the Mobilization of Public Genetic Goods Within the likely transferred BTs, we focused next on some with potential adaptive content ( Karcagi et al. 2016 ). In the graph at ≥ 90% ID, our very discrete sampling of genomes contained 20 BTs distributed on genomes from different phyla. In the graph at ≥ 30% ID, including more ancient sharing events, there were 12,864 BTs (i.e., 30.9% of all BTs) grouping genomes from different phyla. Such taxonomically heterogeneous BTs point to candidate genetic public goods, transferred over large phylogenetic distances, that is, since these sequences are used by phylogenetically heterogeneous hosts, which was confirmed by our horizontality test ( supplementary fig. S8 , Supplementary Material online). For example, Twin 7227 is a CHG involved in cell wall—peptidoglycan—lysis. The protein is found in viruses and bacteria and is important in degrading the cell wall—either for the purposes of infecting a bacterium or for cell division. This kind of “cell puncturing device” is likely to enhance horizontal transfer. Twin 3034 is the LexA protein, which in purified form acts as a repressor of RecA and itself. This protein can function to reduce the level of recombination and SOS-mediated response from an organism ( Pant et al. 2016 ). The SOS response is triggered by DNA damage, as is RecA . Therefore the function of this twin seems to be to repress recombination and to stop DNA repair processes which might prevent the integration of a sequence into a genome. Other interesting examples stem from these analyses. Twin 7401 at ≥ 90% ID corresponds to a particular prokaryotic compartment involved in the carbon fixation from atmospheric CO 2 called the carboxysome ( Yeates et al. 2008 ), shared by taxonomically divergent bacteria (two Cyanobacteria and two Gammaproteobacteria). The carboxysome is also present as twin 69 (under a sufficiently divergent form as to make a different CHG): this time it is even an articulation point linking one Bacteroidetes, one Chloroflexi, and one Actinobacterium. We also find conspicuous plant nodule associated genes: twin 1436 is a nitrogenase subunit NifH forming a twin for a club of three nodule associated Alphaproteobacteria, and twin 7710 is an articulation point, with a dehydrogenase function, between one Acidobacterium ( Candidatus Solibacter usitatus ) and two nodule Alphaproteobacteria ( Methylocella silvestris and Mesorhizobium australicum ). The removal of the articulation point neatly separates the three according to taxonomy, and seems ecologically driven, since all three are soil-dwellers ( Chen et al. 2010 ; Challacombe et al. 2011 ). This hypothesis is supported by information on the isolation sites, retrieved from the GOLD database ( Mukherjee et al. 2017 ), that is, SE Australia for Candidatus Solibacter usitatus , W Australia for M. australicum , and Europe (Germany) for M. silvestris. This Acidobacterium has moreover a large number of genes associated with MGEs ( Challacombe and Kuske 2012 ; Fondi et al. 2016 ), and Mesorhizobium australicum harbors a laterally acquired 455.7-kb genomic island, indicating that these genomes are prone to acquire genes. Public goods however are not the only genes that can be shared so broadly. Twin 13016 is a toxin–antitoxin system, a famous “addiction” system (here shared between three plasmids and one phage). Both genes are needed in the genome in order to function. In general, the toxin is long-lived and the antitoxin is short-lived, and keeps the cell safe from the toxin by binding to it. When the genes are removed from the cell, then the short-lived antitoxin breaks down, leaving the toxin to kill the cell. This mechanism removes cells that have been cured of the toxin–antitoxin system, providing an advantage to those cells that have both genes ( Gerdes et al. 2005 ; Otsuka 2016 ). Maximum likelihood trees reconstructed a posteriori for each of these twins confirmed that the genes discussed here were likely involved in LGT, in agreement with our test of horizontality transfer ( supplementary fig. S8 , Supplementary Material online). Transposases Flood the Web of Life but Do Not Persist We also observed the diffusion of other so-called “selfish” genes. In general, transposases were broadly distributed over MGE and chromosomes, as expected according to, for example, Aziz et al. (2010) . Notably, although transposases are not limited to prophages, and all prophages do not encode a transposase, we verified that the number of transposases in prokaryotic genomes did not correlate with the number of inserted phages, confirming that these mobile elements had decoupled dynamics of chromosomal invasion (adjusted R 2 coefficient = −0.001), and that chromosomes in our data set are not prone to a general inclusion of these diverse types of MGEs. In the overall graph at ≥ 90% ID, 4.78% of the CCs and 8.03% of the BTs were annotated as containing a transposase, respectively. Interestingly, transposases were overrepresented in BTs mixing different types of genomes ( supplementary table S4 , Supplementary Material online), because some transposases travel across different host genomes ( Hooper et al. 2009 ). Homologous transposases are indeed known to be found and functional in different hosts, eventually from different domains of life (i.e., the piggyBac transposable element, isolated from a virus, operates in a diversity of eukaryotes; Johnson and Dowd 2014 ). Likewise, various unrelated studies of genomics have reported the presence of transposases on plasmids ( Jones-Dias et al. 2016 ; Dias et al. 2018 ) and viruses ( Sun et al. 2015 ; Wilson et al. 2017 ), occasionally with adaptive hitch-hiking genes ( Ahmad et al. 2015 ; Manageiro et al. 2015 ; Sabat et al. 2015 ; Aleksandrzak-Piekarczyk et al. 2016 ; Ageevets et al. 2017 ; Sun et al. 2017 ), disseminating the view that transposases are commonly found on these types of mobile genetic elements. Other works have highlighted the evolutionary interplay between transposases and different types of mobile elements (transpovirons, Koonin and Krupovic 2017 ; casposons, Krupovic et al. 2017 ; and retroviruses, Skala 2014 ). Thus our results, offering a systematic survey of the distribution of transposases across mobile elements, are compatible with this background knowledge. In our family-based data set (see supplementary tables S3 and S4 , Supplementary Material online), transposases were found in 7.21% (888 out of 12,321) of the BTs joining the same type of genomes, and in 17.94% (192 out of 1,070) of the BTs joining different types of genomes (e.g., any combination of virus, plasmids, or chromosomes). Likewise, transposases were found in 3.04% (14 out of 460) of the CCs joining the same type of genomes, and in 21.43% (6 out 28) of the CCs that joined different types of genomes. In about 1/6 of these BTs with heterogeneous phyla, other CHG possibly hitch-hiked with these transposases. This suggests that transposases are actively travelling across the web of life, possibly leveraging over the mobility of other MGEs, but they do not organize the web of life. Indeed, removing annotated transposases from the analyses does not substantially change the topology of the bipartite graph. We also tested that the persistence of the network structure was not due to the CHG that have hitch-hiked with the transposases by removing all the CHG that were associated with transposases (i.e., that share the same BT than transposases, in order to reduce the impact of genes hitch-hiking with transposases). This protocol did not affect the topology of the bipartite graphs ( supplementary fig. S6 -1 and S6-2 bottom rows, Supplementary Material online), consistent with our claim that transposases do not organize the web of life. Introgressive evolution has by contrast shaped mobile genetic elements—as can be seen for example in the sharing of very similar genes between viruses and plasmids within the mobilome network (i.e., 8 CCs [out of 488] and 107 BTs [out of 13,391] mixing viruses and plasmids at ≥ 90% ID). Gene Externalization Levels Are Remarkably High in the Microbial World Since our networks encode exact information about which genomes share which CHG, we were able to quantify the extent of “gene externalization,” that is, of sharing between chromosomes and extrachromosomal elements (e.g., when a given CHG is connected to two genomes of different kinds). The idea is that externalized genes are copied on different supports (i.e., chromosomes, plasmids, or virus). In that sense, copies of the same gene are encoded in different media (without necessary be lost from their original support). To continue this analogy, externalized genes can also be viewed as “remastered gene copies.” Gene externalization differs from LGT between chromosomes, although it can contribute to LGT when a gene from a chromosome is copied to a MGE and from that MGE to another chromosome (or to the same genome, in the case of autologs; Popa et al. 2017 ). The difference between gene externalization and LGT means that rules relating to gene externalization may differ from rules relating to LGT. In particular, gene externalization may be random and at a high rate, which would not be visible from LGT analyses, if the host recipient cell selects against the residency of some of the externalized genes (i.e., for example, informational genes may be more externalized than transferred). We observed an impressive proportion of externalized genes in the web of life ( fig. 4 and supplementary table S8 , Supplementary Material online).\n Fig . 4. Percentage of gene externalization at ≥ 30% ID for the 382 prokaryotes in our data set. The proportion of externalized genes from Bacteria (green triangles) is significantly higher (Student’s t -test, p value < 10 −16 ) than for Archaea (yellow dots), to the notable exception of Haloarchaea (red dots). It is moreover largely uncorrelated with genome size (regression lines with shaded confidence interval at 95%). More precisely, for our data set, Bacteria generally have higher externalization than Archaea (significant t -test for ≥ 30%, 40%, 50%, and 60% ID, see supplementary table S8 , Supplementary Material online). A notable exception to this rule is the Haloarchaea, which is likely explained by their chimeric nature ( Nelson-Sathi et al. 2012 ). Careful analyses of the genomes with highest externalization proportions (> 60% at ≥ 30% ID, see fig. 4 ) did not identify structural, ecological nor taxonomical commonalities between these genomes. They were all of high quality (with usually ≥ 11-fold sequence coverage) and their gene content had been carefully investigated ( Dunfield et al. 2003 ; Nandasena et al. 2006 ; Munk et al. 2011 ; Huo et al. 2012 ; Kappler et al. 2012 ). Some of these genomes have interesting metabolic or physiological properties, like a high trophic versatility in the sulfur-oxidizing α-proteobacterium Starkeya novella DSM 506 ( Kappler et al. 2001 , 2012 ; Wang et al. 2016 ), suggestive of LGT affecting that genome. Likewise, 18 putative horizontally transferred regions only had been described in the endophytic β-proteobacterium Herbaspirillum seropedicae SmR1 ( Pedrosa et al. 2011 ), as well as a plasmid-carried photosynthetic ability in the α-proteobacterium Rhodospirillum rubrum ATC 11170 ( Kuhl et al. 1984 ; Munk et al. 2011 ). The largest set of detected transferred genes concerned a genomic island of 455.7 kb in the root nodule α-proteobacterium Mesorhizobium australicum WSM2073, a symbiosis island from the original inoculant strain from the host legume ( Nandasena et al. 2006 ). This genomic island only amounts to 7.3% of the genome of Mesorhizobium australicum WSM 2073 ( Reeve et al. 2013 ). Thus, LGT, and some bacteriophages, transposases, genomic islands, and extra chromosomal small-sized plasmids had been described in some (but not all) of the genomes with high externalization proportion and their relatives. However, all these MGE inserted in chromosomes were found in considerably lower proportion than the proportion of externalized genes ( supplementary fig. S9 -1, Supplementary Material online). Indeed, there was almost no correlation between the externalization proportion and the proportion of inserted phages in these genomes (adjusted R 2 coefficient = 0.1392, for the number of externalized genes vs. the number of inserted MGEs), and in prokaryotic genomes in general (adjusted R 2 coefficient = 0.11, for the number of externalized genes vs. the number of inserted MGEs). Likewise, there was no correlation between the externalization proportion and the proportion of transposases present in the prokaryotic genomes (adjusted R 2 coefficient = 0.07, for the number of transposases vs. the number of externalized genes). Furthermore, externalized genes were scattered across the genomes of their carriers ( supplementary fig. S9 -2, Supplementary Material online). Therefore, the detection of high externalization proportions describes a substantial novel general phenomenon, which can affect very different genomes. Gene Externalization Is not random with Respect to Gene Function We also followed the dynamics of gene externalization for networks of decreasing stringency (assuming a molecular clock, from the most recent to the most ancient externalization events) in order to identify some of externalization rules and to test whether gene externalization is random with respect to gene function. The distributions of the COG categories associated with externalized genes were markedly different, and these differences persisted at different similarity thresholds ( fig. 5 , and also supplementary fig. S10 , Supplementary Material online). The “L” category was abundant among recently externalized genes, suggesting that transposases were among the recent CHG that have moved across different types of host genome. This finding is consistent with the reports of implications of transposons in the evolution of chimeric molecular structures (such as transpovirons, Desnues et al. 2012 and casposons, Béguin et al. 2016 ). However, full-sized genes from that “L” category do not tend to accumulate in their externalized form in genomes, as evidenced by the way their proportion dropped in graphs with lower % ID. This indicates that these transposases do not persist in their host genomes. By contrast, the proportion of externalized genes from the “M” (membrane biogenesis) and “T” (Signal transduction) categories was smaller for recent events than for events considered over a longer time period (i.e., genes from these categories tended to accumulate progressively in genomes as externalized). Other functional categories, such as “E” (Amino-acid metabolism and transport) and “P” (Inorganic ion metabolism and transport) presented more complex distributions.\n Fig . 5. Distribution of functional categories among externalized genes. Color bars represent the percentage of a given COG category among externalized genes above a given identity threshold (according to the color code on the right side of the figure). On the upper bar, COG categories are grouped by large functional groups (including “Poorly characterized,” which includes COG categories R and S). The “No”/0 class (on the left) refers to the genes for which no COG category was attributed). Note the very conspicuous peak for the “L” category at thresholds ≥ 90% and 95% ID. Due to gene externalization, the web of life appears, in its prokaryotic parts, as a collection of largely disconnected, isolated, prokaryotic strands, affected by introgression between close relatives, doubled by spider-webs of mobile elements ( supplementary fig. S1 , Supplementary Material online). One should therefore not be misled by the taxonomical consistency of CCs and BTs. Phylogenetically consistent prokaryotic groups are typically subjected to and sustained by processes that are not simple vertical descent with modification. Gene externalization, from cells to mobile genetic elements, and from mobile genetic elements to cells, may contribute to the high turnover of genes in genomes and their patchy distribution in prokaryotic lineages. The high levels of gene externalization that we report indicate that, collectively, genomes of MGE contain most (possibly all) CHGs from several individual bacterial genomes, which are present, dispersed through fragmented copies, in the unrelated genomes of viruses and plasmids. We predict that as more MGE genomes are sequenced, the percentage of externalized genes per prokaryotic genome will increase, albeit at different rates for different biological functions. Conclusion Our work complements the findings of other recent studies using networks. Cong et al. (2017) , for example, analyzed genetic exchange communities, by decomposing a k-mer based LGT network using notions of graph theory. They inferred functionally biased LGT between cells from various prokaryotic data sets, with up to 144 chromosomes. Our strategy was similar to this inspiring work, but aimed at a different goal. In contrast with ( Cong et al. 2017 ), we represented a diversity of sharings (vertically transmitted genes, horizontally acquired genes, or simply shared genes when the mode of inheritance could not be determined), between a diversity of hosts (between cells, between cells and MGE, and between MGE). In that regard, we did not attempt to reconstruct genetic exchange communities of prokaryotes, united by LGT, but rather to offer a broader picture of CHG distribution across microbial genomes. This was achieved by two means: first, we used a broader data set than Cong et al. (2017) , since a total of 8,214 genomes (7,832 MGEs and 382 chromosomes) were included in the network. Second, we used a different type of network, that is, a genome-CHG bipartite graph. In our analyses, we further focused on genes shared between extrachromosomal elements and chromosomes (externalized genes), because these types of sharings would not be necessarily detected by the identification of LGT between cells, whereas our work shows that the sharing of genes between chromosomal and extrachromosomal elements constitutes an important process of microbial communities, which is also functionally biased. Namely, homologous genes which are present, after selection, on both chromosomal and extrachromosomal elements have biased functional profiles, in particular a high turn-over of genes associated with the L category, because genes from this COG category, whereas overrepresented in the pools of conserved externalized genes are less abundant in the pools of divergent externalized genes than genes with, for example, the “M” and “T” functions. Thus, genes that are recently externalized are not enriched in metabolism, regulation, and transport, contrasting with the genes that Cong et al. reported as overrepresented in LGT within genetic exchange communities. Therefore, our work underscores that two distinct introgressive processes affect the evolution of microbial communities, that is, gene externalization and LGT. Our findings also strongly echoe with the remarkable study by Popa et al. (2017) . These authors used a bipartite CHG-genome network (in their case, directed between donors and hosts), with 3,982 prokaryotic genomes and phages genomes (including prophages of prokaryotic chromosomes, individualized as nodes in their network) to investigate lateral gene transfer by transduction. These authors report that transduction is mostly restricted between closely related donors and recipients, consistent with experimental observations and with our observation that viral genomes are generally peripheral on the genome network, that is, associated with specific cellular host lineages ( Halary et al. 2010 ), which is also confirmed in our present work with an expanded data set. Interestingly, Popa et al. accurately decomposed transduction events in two phases: the uptake of a gene from a donor chromosome into a phage, and the acquisition of a gene from a prophage into a recipient chromosome. According to our own terminology, each of these phases corresponds to an externalization event (one from a cell to a virus, and another from a virus to a cell, respectively). For that reason, the edges in their network correspond to polarized externalization events between prokaryotes and phages. Interestingly, Popa et al. (2017 ) stress that 9% of the transduction events they detected are autologs, that is, duplication of genes mediated by mobile DNA vectors. This autology is consistent with our claim that externalization introduces genetic redundancy (and possibly resilience) within microbial communities. Of note, our work further underscores that the process of externalization (i.e., the sharing of genes on different types of supports, extrachromomal, and chromosomal elements) is very general. We report that externalized genes amount to even higher proportions of the chromosomes (i.e., >60% in Starkeya novella , which are not due to inserted prophages) than the proportion of transferred genes detected by Popa et al. (who inferred 15,298 edges between 2,573 bacteria and 4,650 phages). This is logical, because only events of gene externalization 1) beginning in a chromosome and ending in a chromosome, 2) mediated by a phage, were described in their directed bipartite networks of chromosomes and phages ( Popa et al. 2017 ). By contrast, in our analysis, the first constraint is alleviated, and plasmids, rather than phages, appear as the main carriers of externalized genes (85–99% of externalized genes match on plasmids at ≥30% ID, see supplementary fig. S9 -3, Supplementary Material online). To summarize, we introduced an analysis of gene sharing between chromosomal and extrachromosomal elements from the microbial world based on a bipartite graph. This strategy allowed us to independently recover major trends of microbial evolution (i.e., the taxonomical and typological biases in the pattern of gene sharings, the existence of genetic worlds, in particular in networks built at low stringency thresholds, the promiscuity of transposases), to propose some novel HGT candidates, and to bring forward the general process of gene externalization. We thus show that the web of life is disconnected, with major prokaryotic kinds, largely, but not absolutely isolated from one another in terms of gene sharing, surrounded by spider-webs of mobile genetic elements, with which chromosomes share externalized genes. Moreover, transposases have recently run across this web of life. Consequently, bipartite graphs appear as a powerful way to study the processes of microbial life beyond classic taxonomy and customary genomic analyses, and we propose that beyond coding CHG, the use of bipartite graphs could be further generalized to small RNA families-genome networks and gene families-metagenome networks, and applied to even larger data sets to keep up with the impressive accumulation of molecular sequences. Finally, gene families-genome-metagenome tripartite graphs constitute an exciting horizon for expanded multilevel analyses."
} | 14,526 |
34977394 | PMC8685918 | pmc | 4,545 | {
"abstract": "Bio-manufacturing via microbial cell factory requires large promoter library for fine-tuned metabolic engineering. Ogataea polymorpha , one of the methylotrophic yeasts, possesses advantages in broad substrate spectrum, thermal-tolerance, and capacity to achieve high-density fermentation. However, a limited number of available promoters hinders the engineering of O. polymorpha for bio-productions. Here, we systematically characterized native promoters in O. polymorpha by both GFP fluorescence and fatty alcohol biosynthesis. Ten constitutive promoters (P PDH , P PYK , P FBA , P PGM , P GLK , P TRI , P GPI , P ADH1 , P TEF1 and P GCW14 ) were obtained with the activity range of 13%–130% of the common promoter P GAP (the promoter of glyceraldehyde-3-phosphate dehydrogenase), among which P PDH and P GCW14 were further verified by biosynthesis of fatty alcohol. Furthermore, the inducible promoters, including ethanol-induced P ICL1 , rhamnose-induced P LRA3 and P LRA4 , and a bidirectional promoter (P Mal -P Per ) that is strongly induced by sucrose, further expanded the promoter toolbox in O. polymorpha . Finally, a series of hybrid promoters were constructed via engineering upstream activation sequence (UAS), which increased the activity of native promoter P LRA3 by 4.7–10.4 times without obvious leakage expression. Therefore, this study provided a group of constitutive, inducible, and hybrid promoters for metabolic engineering of O. polymorpha , and also a feasible strategy for rationally regulating the promoter strength.",
"introduction": "1 Introduction Bio-manufacturing represents for a promising approach for sustainable supplying of chemicals with mild reaction conditions, low energy consumption [ 1 ]. Microbial cell factories with extensive metabolic engineering have been applied for productions of bulk chemicals [ 2 ] and natural products [ 3 , 4 ]. Construction of biosynthetic pathways requires expression of multiple genes, which is normally realized by different promoters with various strengths. Besides, fine-tuning metabolic flux, including overexpression and down-regulation of key genes [ 5 ], directed evolution of enzymes [ 6 ], cofactor engineering [ 7 ], and so on, reduce the toxic intermediates and enhance the production of target products. Consequently, the convenient and commonly used transcriptional regulation via a large promoter library guarantees a superior microbial cell factory for efficient productions [ [8] , [9] , [10] ]. Ogataea polymorpha (hereafter O. polymorpha ), a methylotrophic yeast, has a broad spectrum of substrates like glucose, xylose, glycerol, methanol, and high thermo-tolerance [ 11 ]. For example, high temperature (45 °C) fermentation enabled efficient ethanol production from xylose in O. polymorph [ 11 ]. In addition, the characteristics of O. polymorpha in post-translational modification and high-density fermentation make it a promising candidate to produce heterologous proteins [ 12 ]. However, there is limited reports on chemicals production in O. polymorpha , which may be partially attributed to the poor genetic manipulation tools and promoter library [ 13 ]. Recently, CRISPR/Cas9 based genome editing system was established for O. polymorpha [ [14] , [15] , [16] ]. While the promoter lack situation is still remaining and seriously hinders the extensive metabolic engineering of O. polymorpha [ 17 ]. Generally, promoters are classified into constitutive and inducible promoters. Constitutive promoters possess basically stable activities among different fermentative conditions. The strength of inducible, or repressive promoters are dynamic regulated by specific inducers or repressors. Promoters commonly used for gene expression in O. polymorpha includes the strong methanol-induced promoter P AOX1 (the promoter of alcohol oxidase I gene) and promoter P FMD (the promoter of formate dehydrogenase gene) [ 18 , 19 ], and the strong constitutive promoter P GAP (the promoter of glyceraldehyde 3-phosphate dehydrogenase gene) [ 20 ]. Obviously, these limited tools are far from enough for extensive metabolic engineering. Therefore, further screening other available promoters in O. polymorpha is essential, which should promote its potential as a chassis host in protein and chemicals production. In this study, green fluorescent protein (GFP) was used to characterize the promoter strength ( Fig. S1 ). A total of ten constitutive promoters were characterized, among which the promoters P GCW14 and P PDH were further verified by production of fatty alcohol. Additionally, multiple inducible promoters were evaluated, and the regulation of promoter activity was achieved by constructing tandem repeats of upstream activation sequence (UAS), generating hybrid promoters with greatly enhanced activities [ [21] , [22] , [23] ]. Overall, our results offered a promoter toolbox with distinguished activities, and a feasible strategy to control promoter activities in O. polymorpha , which will pave the way to adopt this superior host for extensive metabolic engineering in both fundamental and industrial applications.",
"discussion": "4 Discussion Availability of promoters is very essential for extensive metabolic engineering. This study identified and characterized three different types of native promoters (constitutive, inducible and hybrid) in O. polymorpha for further constructions of cell factory, and also provides a feasible strategy for promoter mining and control. We characterized several endogenous constitutive promoters from central metabolic pathways, which however was much weaker compared with the strong constitutive promoter P GAP (10%–60%). We here found the promoter strength was varied between glucose and methanol, which demonstrated that these promoters are not strictly constitutive, and may be related to cell growth status under various fermentative conditions. In addition, a strong constitutive promoter P GCW14 of a potential glycosylphosphatidylinositol (GPI) anchored protein, which is not related to specific pathway, was characterized to be a relatively constitutive promoter under different culture conditions. Interestingly, a homologous promoter P PpGCW14 from P. pastoris was also functional in O. polymorpha . This universal expression regulation among different yeasts may provide a reference strategy for further mining similar promoters with different intensities [ 33 ]. To further expand the promoter library in O. polymorpha , the inducible promoters were excavated, including ethanol-induced promoter P ICL1 , rhamnose-induced promoters P LRA3 and P LRA4 , and a bidirectional promoter P Mal -P Per . These inducible promoters demonstrated strict glucose repression, and even though in a mixture of glucose and inducers, the activity remained an extremely low level. Just like GAL system in S. cerevisiae [ 44 ], these inducible promoters may be regulated by other transcriptional factors, and their regulatory mechanisms are worthy of further exploration by a more refined stepwise truncation strategy [ 45 ]. We also found that the strength of the inducible promoter such as P LRA3 was relatively low. A UAS-tandem strategy was developed to increase the promoter strength of P LRA3 by 4.7–10.4 times without influencing the inducible feature. Compared with the site-directed mutagenesis approach [ 46 ], this rational design may directly obtain the hybrid promoter with a predictable manner [ 35 , 47 ]. We can expect that our hybrid promoter strategy can be applied to other large number of inducible promoters with low strengths. We also used various promoters P PDH and P GCW14 for regulating the biosynthesis of fatty alcohols. The positive relation between promoter strength and fatty alcohol production verified the practicability of promoter based pathway regulation. Larger number of promoters with various strengths can help to regulate the metabolic pathways with a precise manner, which should be helpful for optimization of metabolic network in cell factory construction [ 48 ]. In summary, we identified and evaluated three different types of promoters for providing sound biological elements in metabolic engineering of O. polymorpha , as well as provided feasible strategies for promoters mining and engineering."
} | 2,075 |
35745983 | PMC9230667 | pmc | 4,546 | {
"abstract": "Superhydrophobic coatings are widely applied in various applications due to their water-repelling characteristics. However, producing a durable superhydrophobic coating with less harmful low surface materials and solvents remains a challenge. Therefore, the aim of this work is to study the effects of three different solvents in preparing a durable and less toxic superhydrophobic coating containing polydimethylsiloxane (PDMS), silica solution (SS), and epoxy resin (DGEBA). A simple sol-gel method was used to prepare a superhydrophobic coating, and a spray-coating technique was employed to apply the superhydrophobic coating on tile substrates. The coated tile substrates were characterized for water contact angle (WCA) and tilting angle (TA) measurements, Field-Emission Scanning Electron Microscopy (FESEM), Atomic Force Microscopy (AFM), and Fourier Transform Infrared Spectroscopy (FTIR). Among 3 types of solvent (acetone, hexane, and isopropanol), a tile sample coated with isopropanol-added solution acquires the highest water contact angle of 152 ± 2° with a tilting angle of 7 ± 2° and a surface roughness of 21.80 nm after UV curing for 24 h. The peel off test showed very good adherence of the isopropanol-added solution coating on tiles. A mechanism for reactions that occur in the best optimized solvent is proposed.",
"conclusion": "4. Conclusions In this work, the best superhydrophobic coating has been synthesized using PDMS, SS, and epoxy resin (DGEBA) with isopropanol as the solvent rather than hexane and acetone. Isopropanol, which is a polar solvent, poses dissolving-like behavior towards PDMS, SS, and epoxy and aids in the creation of multiscale roughness. The S3 coated sample has a water contact angle of 152 ± 2° after UV irradiation and showed good durability. Such a superhydrophobic coating can be utilized on tiles to act as a self-cleaning surface.",
"introduction": "1. Introduction A material is known to be superhydrophobic when it has a water contact angle of more than 150° and a negligible titling angle of less than 10° [ 1 , 2 ]. Superhydrophobic coating has gained more attention and lead to many applications such as anti-icing [ 3 , 4 ], anti-fogging [ 5 , 6 ], oil-water separation [ 7 , 8 ], anti-corrosion [ 9 , 10 , 11 ], self-cleaning [ 12 , 13 ], antibacterial [ 14 , 15 ] and biomedical applications [ 16 ]. Various methods that have been used for preparation of rough surfaces are layer by layer assembly [ 17 , 18 ], spray-coating [ 19 , 20 , 21 ], lithography [ 22 , 23 ], sol gel processing [ 24 , 25 ], electrochemical deposition [ 26 , 27 ] and chemical vapour deposition [ 28 ]. Among them, spray-coating is a fabrication process used for industrial applications due to its availability in commercial form and a simple procedure that uses inexpensive materials [ 2 , 29 ]. In regard to this, the increase in surface roughness and the reduction of surface energy are vital in forming a superhydrophobic surface [ 30 , 31 , 32 ]. A summary of findings on the choice of materials used for spray-coating is shown in Table 1 . As seen, the main precursors needed for superhydrophobic spraying techniques are solvent, low surface energy material and nanoparticles. The function of low surface energy material is to reduce the wettability of the surface up to 120°. Nanoparticles are to achieve appropriate roughness to trap the air to further reduce the wettability to achieve a water contact angle greater than 150° and a tilting angle smaller than 10°. The function of solvent is to reduce the viscosity of low surface energy material for easy application of superhydrophobic coating [ 33 ] on material surface. The emphasis of this work was on the solvent utilized to generate the superhydrophobic coating, as there have been many studies on nanoparticles and low surface energy materials. According to Luo et al. the dispersibility of nanoparticles in solvent influences the generation of even dispersion or aggregation, which controls the formation of a smooth or rough coating on the surface of a substrate [ 45 ]. To produce a well-dispersed or aggregated superhydrophobic coating, the polarity and relative permittivity of the solvent are the important criteria to be considered. For instance, inorganic components such as SiO 2 nanoparticles aggregate when non-polar solvents are used but disperse evenly in polar solvents. This happens because polar solvents stabilize silica dispersion through strong hydrogen bonding to silanol groups on the silica surface. Conversely, low-polarity solvents result in destabilization and gelation of silica particles via hydrogen bonding between adjacent silica particles [ 46 , 47 ]. The relative permittivity of the solvent is also a measure of solvent polarity; the lower the relative permittivity of the solvent, the lower the polarity of the mixture; the polarity of the mixture may eventually decrease to the point where it is no longer sufficient to sustain the dispersion of the polar silica nanoparticles, resulting in particle aggregation. Nonetheless, high aggregation conditions prior to the phase separation limit (gelation of SiO 2 nanoparticles) are desirable as they help to create multi-scale roughness [ 47 ]. Therefore, an appropriate solvent needs to be selected to form the required roughness to reduce the wettability behavior. A variety of solvents used by researchers to prepare superhydrophobic coating is tabulated in Table 2 . In 2018, Zhang et al. used xylene as solvent for superhydrophobic epoxy/PDMS nanocomposite coating fabrication [ 6 ]. Tetrahydrofuran was also reported as solvent to synthesize superhydrophobic wood surfaces, hydrophobic sol-gel coating and UV-cured superhydrophobic cotton fabric surfaces [ 25 , 48 , 49 ]. Saleem used toluene as solvent in the development of superhydrophobic surfaces [ 33 ]. Among these solvents, hexane is the most popular solvent used to produce superhydrophobic coating due to low permittivity nature as it helps to create multi-scale roughness surface to trap air and increase the WCA. [ 12 , 35 , 50 , 51 , 52 ] are among the author researchers fabricated superhydrophobic coating by using hexane as solvent. However, these solvents are toxic and hazardous towards organs through prolonged exposure [ 53 ]. Other solvents such as isopropanol and acetone that has less hazardous effect towards users are not explored probably due to the relatively high permittivity value, 17.9 and 20.7, respectively as compared to hexane (1.9) 54. Due to safety concern, these two solvent are still explored in this work because the use of hydroxyl terminated PDMS (ε = 2.30 − 2.80) would react with mild polar hydroxyl and carbonyl group of those solvents because of the like-dissolve-like concept [ 54 , 55 ]. This would sustain the dispersion of the polar silica nanoparticles, with certain degree of aggregation. Besides, to reduce the toxicity of the superhydrophobic coating produced in this work, a low surface energy material, PDMS was used. To value add, the SiO 2 nanoparticles that is required to increase the surface roughness for the superhydrophobic coating was extracted from palm oil fuel ash waste (POFA). A suitable mechanism was also proposed for the development of super-hydrophobic coating on the tiles.",
"discussion": "3. Results and Discussion 3.1. Effect of Solvent on Wettability Table 3 shows the water contact angle, tilting angle, surface energy and the roughness of tiles coated with S1, S2 and S3 that were prepared with different solvents. Water contact angles for S1 and S2 coated with acetone and hexane added solution are 88 ± 1° and 85° ± 1°, respectively. Therefore, S1 and S2 exhibit hydrophilic behavior. On the other hand, S3 that was coated with isopropanol-added solution has water contact angles of 149 ± 2°, showing hydrophobic behavior. The high surface energy of the S1 and S2 coatings, which is 15 times higher than the S3 with 2 J/m 2 , contributes for the difference in water contact angle. S1 and S2 may indeed be affected by the dispersion of PDMS and SS particles in different solvents, which will be explained in more detail later. Furthermore, as compared to S1 and S2, sample S3 has a higher roughness, which could be another aspect that improves the sample’s hydrophobicity. As for tilting angle, it was found S1 with 7.09 nm has a lower tilting angle (20°) as compared to S2 that has a roughness of 13.01 with a tilting angle of 46°. The findings are consistent with [ 29 , 60 ], which demonstrated that for hydrophilic samples, higher surface roughness results in a higher tilting angle. Besides, higher surface roughness will also result in a stronger water pining effect due to the absence of air pockets. This situation will result in the penetration of water droplets into the grooves. Thus, S1 and S2 are predicted to be in a Wenzel state, as the Wenzel equation states that roughness emphasizes the effect of surface chemistry. In other words, for hydrophilic surfaces, the higher the surface roughness, the more hydrophilic the surface is while for a hydrophobic surface, the higher the surface roughness, the more hydrophobic the surface will be. The wetting mechanisms of hydrophobic surfaces (S3) can also be determined based on their respective tilting angles. The tilting angle of 10 ± 2° for S3 suggests the surface is in a Cassie-Baxter state, which is further supported by the surface morphology and high surface roughness (18.57 nm), which render a hierarchical structure. Such morphology leads to the formation of air voids that help in water droplet suspension, which results in a low tilting angle. Apart from that, slight changes in water contact angle were also observed after the samples were cured under UV for 24 h ( Table 3 ). After UV-curing, the water contact angles of samples that were coated with hexane-added solution and isopropanol-added solution were increased by ~3°, while the water contact angle of sample that was coated with acetone-added solution was decreased by ~6°. The increase in water contact angle of S2-UV and S3-UV after UV curing may be attributed to the increase in grafting density of the PDMS polymer chain to form a stronger 3D network 36. Besides, UV-curing helps to cure the polymer phase that was disrupted by aggregated particles in oven curing, producing a coating with strong adhesion, mechanical reliability, and chemical resistance [ 51 , 61 , 62 ]. Even though the difference in WCA was not statistically significant, the effect of UV was still considered crucial as the WCA was increased to above 150°, causing S3-UV to be superhydrophobic. However, the decrease in water contact angle of S1-UV after UV curing may be attributed to surface oxidation, which is triggered by the functional carbonyl group (C=O) presence in acetone. This accelerates the degradation of long alkyl chains into smaller chains [ 63 , 64 ], thus increasing the wettability. 3.2. Effect of Solvent on Surface Morphology and Surface Roughness Surface roughness is another key contributor to the superhydrophobic characteristics of coated samples other than surface energy 35. Figure 1 shows the 3D AFM topographical images ( Figure 1 a–f), AFM line profile ( Figure 1 (a a –f a ), Figure 1 (c c ,f c )) and FESEM images ( Figure 1 (a b –f b )) of S1, S2 and S3 before and after UV treatment. From Figure 1 (a b ,b b ), it is observed that the surfaces of S1 and S2 before UV curing do not have obvious micro papillae structure. However, under AFM, the surface possesses certain roughness as indicated by the peaks and valleys in ( Figure 1 (a a )) and ( Figure 1 (b a )) corresponding to RMS of 7.09 nm and 13.01 nm, respectively. In comparison to S1 and S2, FESEM images of S3 before UV curing show that the surfaces are relatively rough ( Figure 1 (c b )). S3 that was coated with isopropanol-added solution has an RMS value of 18.57 nm. The high surface roughness in S3 is due to the dispersion of the coating solution, in which multiscale roughness is created. This sample has an appropriate degree of particle aggregation in the isopropanol-added solution, which induces hierarchical structure [ 47 ]. For samples cured with UV ( Figure 1 (d b ,e b ,f b )), a similar trend was observed as before UV curing ( Figure 1 (a b ,b b ,c b )), respectively. However, the distance between the hills for S1-UV after UV curing ( Figure 1 (d a )) is wider than that before UV curing ( Figure 1 (a a )), leading to a lower water contact angle after UV curing. Figure 1 (e a ) shows that S2-UV after UV curing has peaks and valleys with a lower contrast as compared to after UV curing, leading to a lower RMS roughness value of 11.57 nm. The S3-UV that was coated with isopropanol-added solution and UV-cured has the highest RMS roughness value of 21.80 nm. This is probably ascribed to the even distribution of peaks and valleys with spikes that renders a hierarchical structure in S3 after UV curing ( Figure 1 (f a ,f c )) as compared to before curing ( Figure 1 (c a ,c c )). In summary, the formation of hierarchical structures with appropriate roughness and distance between hills improves hydrophobicity. Those characteristics were achieved in the isopropanol-added solution that has been UV cured. 3.3. Effect of Solvent on Dispersion of Coating Solution The effects of solvents such as hexane, acetone, and isopropanol on the dispersion of superhydrophobic coating solutions were investigated. Figure 2 shows that agglomeration was evident in all three solutions, but that after shaking, the isopropanol-added solution became well-dispersed whereas the agglomeration in the other two solutions remained. The relative permittivity of the solvent utilized causes such circumstance. Hexane is a non-polar solvent with a relative permittivity of 1.90, making it the least polar. As a result, it failed to disperse SS particles, which are polar in nature with OH groups. Hexane’s relative permittivity is too low, resulting in SS particle gelation, rather than particle aggregation. It was proven by the fact that samples S2 coated with hexane-added solution have low water contact angles before and after UV curing due to the absence of hierarchical structure on their surface ( Figure 1 (b b ,e b )). On the other hand, acetone and isopropanol are polar solvents with a relative permittivity of 20.7 and 17.9, respectively. Hence, both solvents were able to disperse SS particles with a certain degree of aggregation, resulting in surface roughness with a maximum RMS value of 21.80 nm ( Table 3 ). In addition, the agglomeration in the isopropanol-added solution could be dispersed after it was shaken due to the like-dissolves-like concept. This dissolves-like concept indicates that solutes, which are PDMS, SS and epoxy in this work, will dissolve best in a solvent that has a similar chemical structure to them [ 65 ]. From Figure 3 A-a, OH-bonds were found in isopropanol but not in hexane ( Figure 3 B-a) and acetone ( Figure 3 C-a) Therefore, PDMS, SS, and epoxy dissolved best in isopropanol instead of acetone or hexane. 3.4. FTIR of Different Types of Solvent and Coating Solution FTIR spectroscopy was used to investigate functional groups in coating solutions of different types of solvent by analyzing their respective precursors. For isopropanol ( Figure 3 A-a), it can be seen that O-H stretching, and O-H bending are indicated at 3318 cm −1 and 951 cm −1 , respectively. At 2970 cm −1 , C-H stretching of CH 3 group is indicated. In addition, C-H bending vibration of CH 2 and CH 3 are shown at 1467 cm −1 and 1379 cm −1 , correspondingly. A peak noticed at 3338 cm −1 corresponds to O-H stretching, which is attributed to its hydroxyl-terminated structure of PDMS ( Figure 3 A-b). At 1640 cm −1 , O-C = O stretching is detected. Peaks that are noticed at 1259 cm −1, 1022 cm −1 and 795 cm −1 in Figure 3 A-b, representing Si-CH 3 , O-Si-O and Si-C bonds, respectively [ 36 , 66 , 67 ]. The finding of functional groups in isopropanol and PDMS matches with their respective chemical formulae of C 3 H 8 O and (C 2 H 6 OSi) n . For Mixture B ( Figure 3 A-c), PDMS, AMPS and DBTL were added into isopropanol. The reaction occurred in the preparation of Mixture B can be proven by the reduced intensity of O-H stretching at 3325 cm −1 and presence of C-N at 1200 cm −1 , as a result of the AMPS is grafted at the hydroxyl-end of PDMS [ 6 ]. C-H stretching and bending of CH 3 groups and O-H bending that were originally absent in PDMS spectra, were observed at 2969 cm −1 , 1378 cm −1 and 951 cm −1 , respectively, ascertaining the formation of methanol (CH 3 OH) as side-product in the proposed mechanism ( Figure 4 ). Figure 3 A-d shows FTIR spectrum of SS, in which O-H, Si-O and O-Si-O bonds are detected at 3318 cm −1 , 1261 cm −1 and 1081 cm −1 , respectively. A peak observed at 1640 cm −1 belongs to O-C = O stretching. In Figure 3 A-e, O-H stretching is also found in epoxy (DGEBA) at 3319 cm −1 . C-H stretching and rocking vibrations are noticed at 2967 cm −1 and 794 cm −1 . In addition, C-O-C stretching and C-O stretching of oxirane ring are also found at 1022 cm −1 and 1259 cm −1 [ 6 ]. Lastly, Figure 3 A-f shows the FTIR spectrum of isopropanol-coating. Both O-H stretching, and O-H out-of-plane bending are detected at 3338 cm −1 and 951 cm −1 , respectively. Besides, C-H stretching of CH 3 , C-H bending vibrations of CH 2 and CH 3 groups are also detected at 2970 cm −1 , 1467 cm −1 and 1379 cm −1 , correspondingly. At 1261 cm −1 , Si-CH 3 bond is indicated. Peaks that are observed at 1128 cm −1 and 815 cm −1 correspond to O-Si-O and Si-C bonds, respectively. From 3A-c to Figure 3 A-f, intensity of O-H stretching at ~3300 cm −1 increases, which may be attributed to the reaction occurred between Mixture B, epoxy and SS. Other than this, a strong and sharp peak observed at 951 cm −1 corresponds to high intensity of O-H out-of-plane bending may be caused by the presence of methanol as side-product [ 66 ]. Next, Figure 3 B shows FTIR spectra of hexane solution and its precursors. The functional groups of PDMS ( Figure 3 B-b), SS ( Figure 3 B-d) and epoxy ( Figure 3 B-e) and are similar to those in Figure 3 A. The precursor was included in Figure 3 B for comparison purpose. For hexane ( Figure 3 B-a), with chemical formula of C 6 H 14 , O-H bond is absent, while C-H stretching of CH 3 (2925 cm −1 ), C-H bending of CH 2 and CH 3 groups (1467 cm −1 and 1379 cm −1 ), as well as C-C stretching (1260 cm −1 ) are detected. In this case, reaction between PDMS and AMPS is verified as O-H bonds from PDMS are fully contributed in this reaction, causing depletion of O-H bonds and formation of C-N groups as observed at 1218 cm −1 in Mixture B ( Figure 3 B-c). Based on Figure 3 B-f, O-H bond is not detected after the addition of epoxy and SS, which may be due to the polarity of hexane that results in agglomeration of SS particles; thus, the solution is not well-dispersed [ 46 ]. FTIR spectra of acetone solution and its precursors are shown in Figure 3 C. The functional groups of PDMS ( Figure 3 C-b), SS ( Figure 3 C-d) and epoxy ( Figure 3 C-e) are similar to those in Figure 3 A. The precursor was included in Figure 3 C for comparison purpose. For acetone ( Figure 3 C-a), shows peaks correspond to C = O (1711 cm −1 ), C-H stretching of CH 2 (1423 cm −1 ) and CH 3 groups (1361 cm −1 ), together with C-C stretching (1220 cm −1 ), which are identical to its chemical formula of C 3 H 6 O [ 68 ]. For Mixture B ( Figure 3 C-c) of acetone-added solution, the presence of C-N groups at 1200 cm −1 and O-H stretching is also absent as O-H groups from PDMS and are utilized for bonding between AMPS and PDMS. In acetone-added coating solution ( Figure 3 C-f), intensity of O-H stretching noticed at 3479 cm −1 is low. This may be attributed to “like-dissolves-like” concept, leading to a higher degree of dispersion as compared to hexane-added solution but lower than that of isopropanol-added solution, as acetone is a polar solvent, but hydroxyl group is absent [ 65 ]. As seen, compared to S1 solution, S2 solution has similar peaks expect for peaks that correspond to O-H stretching and bending vibration. When S3 solution compared S1, an extra peak of C=O was observed while the Si-CH 3 and O-H bending vibration were absence. Absence of C-H stretching (~2900 cm −1 ) and Si-CH 3 bond (~1260 cm −1 ) in S3 is probably due to oxidation of C-H to form C=O bonds [ 69 , 70 ]. The oxidation is further affirmed by color of acetone-added solution, which is brownish in color as compared to other solutions ( Figure 2 ). 3.5. The Mechanism of Reaction of S3 Solution Since S3 with isopropanol solvent shows the best hydrophobic behavior, the mechanism of this solution is proposed. A mechanism for reactions occurred in the preparation of isopropanol-added solution is illustrated schematically in Figure 4 a–c, based on FTIR analysis. In the preparation of Mixture B ( Figure 3 A-c), Si-O-Si (refer A) bonds were formed upon the mixing of AMPS and PDMS with the presence of DBTL as catalyst, by hydrogen bonding with PDMS at the –OCH 3 ends of AMPS and methanol is formed as by-product ( Figure 4 a). Then the ring opening reaction of epoxy occur with AMPS-PDMS at the amine end and forming modified PDMS ( Figure 4 b, refer B). Then, the modified PDMS is attached to a central Si atom of SS particles, by forming hydrogen bonding and covalent bond after heating and release water ( Figure 4 c, refer C). 3.6. Peel of Test A peel-off test was performed on S3 coated tiles to determine the durability of the coating on the substrate. This test was repeated five times. In Figure 5 , the appearance of the tape surface and the water contact angle after the peel off test are presented. As seen, the surface of the tape is clear from any debris, indicating the coating’s resistance to separation from the substrate. This characteristic implies that the coating has a good adhered to the surface and forms a strong bond. As a result, there are no significant changes in the topography and roughness of the S3 sample before and after peeling, resulting in a WCA that is identical."
} | 5,547 |
22239666 | null | s2 | 4,548 | {
"abstract": "Biofilm formation in Vibrio cholerae is in part regulated by norspermidine, a polyamine synthesized by the enzyme carboxynorspermidine decarboxylase (NspC). The absence of norspermidine in the cell leads to a marked reduction in V. cholerae biofilm formation by an unknown mechanism. In this work, we show that overexpression of nspC results in large increases in biofilm formation and vps gene expression as well as a significant decrease in motility. Interestingly, increased NspC levels do not lead to increased concentrations of norspermidine in the cell. Our results show that NspC levels inversely regulate biofilm and motility and implicate the presence of an effective feedback mechanism maintaining norspermidine homeostasis in V. cholerae. Moreover, we provide evidence that NspC and the norspermidine sensor protein, NspS, provide independent and distinct inputs into the biofilm regulatory network."
} | 227 |
35995792 | PMC9395534 | pmc | 4,550 | {
"abstract": "Muconic acid is a bioprivileged molecule that can be converted into direct replacement chemicals for incumbent petrochemicals and performance-advantaged bioproducts. In this study, Pseudomonas putida KT2440 is engineered to convert glucose and xylose, the primary carbohydrates in lignocellulosic hydrolysates, to muconic acid using a model-guided strategy to maximize the theoretical yield. Using adaptive laboratory evolution (ALE) and metabolic engineering in a strain engineered to express the D-xylose isomerase pathway, we demonstrate that mutations in the heterologous D-xylose:H + symporter (XylE), increased expression of a major facilitator superfamily transporter (PP_2569), and overexpression of aroB encoding the native 3-dehydroquinate synthase, enable efficient muconic acid production from glucose and xylose simultaneously. Using the rationally engineered strain, we produce 33.7 g L −1 muconate at 0.18 g L −1 h −1 and a 46% molar yield (92% of the maximum theoretical yield). This engineering strategy is promising for the production of other shikimate pathway-derived compounds from lignocellulosic sugars.",
"introduction": "Introduction The development of economic processes for the production of biofuels and biochemicals from lignocellulose will be critical to help reduce anthropogenic greenhouse gas emissions associated with fossil fuel consumption 1 , 2 . Among the various areas of metabolic space that have been explored for biochemical production, molecules from microbial aromatic catabolic pathways exhibit substantial chemical diversity 3 , 4 . Of note, cis , cis -muconic acid (hereafter muconate) is a popular platform chemical from the catechol catabolic pathway that can be produced from lignin-derived aromatic compounds, carbohydrates and waste plastics-derived aromatic compounds 5 – 12 . Muconate is a bioprivileged molecule 13 that can be converted into either direct replacement chemicals, such as adipic acid and terephthalic acid 5 , 14 , 15 , or converted to performance-advantaged bioproducts 16 – 24 . Muconate production from carbohydrates is based on the shikimate pathway for aromatic amino acid biosynthesis and was first demonstrated in recombinant Escherichia coli 5 . Erythrose-4-phosphate (E4P) and phosphoenolpyruvate (PEP) are condensed to form 3-deoxy- d -arabinoheptulosonate 7-phosphate (DAHP), which is further converted to 3-dehydroshikimate (3-DHS), a key intermediate in the shikimate pathway. From 3-DHS, at least five pathways have been reported for muconate biosynthesis 25 – 30 . Among these pathways, one proceeds through the intermediate protocatechuate (PCA) via a 3-DHS dehydratase ( asbF ) and results in a higher maximum theoretical yield than the others, which proceed through shikimate via a shikimate dehydrogenase ( aroE ) (Fig. 1a ) 29 . Fig. 1 Muconate production from glucose and xylose. a Schematic of the overall metabolic engineering strategy. To utilize xylose, xylE , d -xylose isomerase ( xylA ), xylulokinase ( xylB ), transaldolase ( tal ), and transketolase ( tkt ) from E. coli were heterologously expressed. E4P and PEP were condensed to form DAHP via a feedback-resistant DAHP synthase ( aroG D146N ) 70 . To convert DAHP to muconate (MA), genes encoding a 3-DHS dehydratase ( asbF ) from Bacillus cereus , and a PCA decarboxylase ( aroY ) and its corresponding co-factor generating protein ( ecdB ), both from Enterobacter cloacae , were heterologously expressed. aroB and catechol 1,2-dioxygenase ( catA ) were overexpressed 3 , 32 . An engineered chorismate pyruvate-lyase from E. coli ( ubiC -C22) 40 was overexpressed to convert chorismate (CSA) to 4-hydroxybenzoate (4HB), which can be converted to PCA and MA. Deleted genes are shown in red. Glucose dehydrogenase ( gcd ) was deleted to prevent the formation of xylonate or gluconate. Glucose-6-phosphate isomerases pgi-1 and pgi-2 were each deleted previously 3 , 32 , but pgi-1 was restored in this study. Pyruvate kinases pykA and pykF were each deleted to reduce competition for PEP. To accumulate MA, pcaHG and catBC were deleted to prevent ring-opening of PCA and catabolism of MA, respectively. P phosphate, 2-KGn 2-ketogluconate, 2-KG-6-P 2-ketogluconate-6-P, G6P glucose-6-P, 6PG 6-phosphogluconate, KDPG 2-keto-3-deoxy-6-phosphogluconate, G3P glyceraldehyde-3-P, FBP fructose-1,6-P 2 , F6P fructose-6-P, S7P sedoheptulose-7-P, R5P ribose-5-P, Ri5P ribulose-5-P, 3PG 3-phosphoglycerate, CAT catechol, SA shikimate, S3P shikimate-3-phosphate, ICIT isocitrate, CIT citrate, AKG alpha-ketoglutarate, SUCC succinate, FUM fumarate, MAL malate, GLX glyoxylate, OAA oxaloacetate, AcCoA acetyl-Coenzyme A. b Metabolic modeling of the maximum theoretical muconate molar and carbon yields with or without pgi-1 . The blue lines represent the asbF pathway, the red lines represent the aroE pathway, solid lines represent % molar yield, and dashed lines represent % carbon yield. The gray areas represent the molar percentage of xylose consumed from 33 to 40% (with the balance glucose), mimicking the composition of corn stover hydrolysates. c Shake-flask cultivations of strain QP328 on glucose and xylose. % Molar yield was calculated as [mM muconate/mM (glucose + xylose) × 100], and % carbon yield was calculated as [mM muconate × 6/mM (glucose × 6 + xylose × 5) × 100]. Error bars represent the standard deviation of biological triplicates. Source data are provided as a Source Data file. Several previous efforts to produce muconate from sugars via asbF have disrupted the shikimate pathway by deleting aroE 10 , 11 , 31 . Deletion of aroE results in strains that are auxotrophic for essential aromatic amino acids, which is undesirable for a bioprocess 10 , 25 . Recently, Pseudomonas putida KT2440 (hereafter P. putida ) strains have been engineered to efficiently produce muconate from glucose via asbF 3 , 32 , 33 . Most recently, we reported the engineering of P. putida that achieved a titer of 22.0 g L −1 at 0.21 g L −1 h −1 and a 35.6% molar yield from glucose in a pH-controlled bioreactor 32 . To date, most efforts to produce muconate from carbohydrates have employed glucose as a substrate. However, the co-utilization of glucose and xylose—often are the two major carbohydrates in lignocellulose 2 — is crucial for the valorization of biomass hydrolysates. Co-utilization of glucose and xylose for muconate production has been studied in Escherichia coli 11 . In this previous work, xylose was metabolized to the TCA cycle to avoid carbon catabolite repression (CCR), thus limiting muconate yield, which motivates the development of other strategies toward this goal. Unlike E. coli , P. putida is natively unable to utilize xylose, which provides an opportunity to engineer optimal xylose pathways in the absence of CCR 34 – 37 . In this work, we seek to incorporate xylose utilization to achieve efficient muconate production from glucose and xylose in P. putida . To this end, we first delete hexR and engineer the D-xylose isomerase pathway into a strain previously engineered to produce muconate from glucose (Table 1 ). By combining metabolic modeling, rational strain engineering, adaptive laboratory evolution, and bioreactors cultivation, we identify successful strategies to improve muconate production from glucose and xylose. Finally, metabolomics is performed to infer the impact of the genetic modifications on metabolic flux. Table 1 Strains in this study Strain Genotype References CJ522 P. putida KT2440 Δ catRBC ::P tac : catA Δ pcaHG ::P tac : aroY : ecdB : asbF Δ pykA :: aroG-D146N : aroY : ecdB : asbF Δ pykF Δ ppc Δ pgi-1 Δ pgi-2 Δ gcd 3 JE3226 P. putida KT2440 Δ hsdR ::P tac :BxB1 int - attB Δ gcd Δ ampC ::P xylE * xylE : xylAB : tktA : talB 35 JE3692 JE3226 ∆ gcd::araE-araCDABE 35 QP328 CJ522 Δ hexR Δ ampC ::P xylE * xylE :P tac : xylAB : talB : tktA ∆ pgi-1 :: pgi-1 PP_1736-1737(intergenic)::Plac: ubiC -C22 This study QP478 QP328 xylE-A62V , A455V P PP_2569 G→A duplication of PP_5050–PP_5242 This study LC041 JE3226 ∆ pgi-1 This study LC061 QP478 restoration of G→A at P PP_2569 This study LC078 QP478 restoration of xylE-A455V This study LC093 QP478 restoration of xylE-A62V This study LC091 QP328 xylE-A62V, A455V This study LC092 QP328 P PP_2569 G→A This study LC100 LC091 P PP_2569 G→A This study LC111 JE3692 xylE-A62V , A455V This study LC147 LC100 Δ pykF ::P lac : gpmI This study LC150 LC100 Δ pykF ::P tac : maeB This study LC151 LC100 Δ pykF ::P tac : rpiA This study LC168 LC100 Δ pykF ::P tac : aroK : aroB This study LC171 QP478 ΔPP_5050–PP_5242 This study LC173 QP478 ΔPP_5084–PP_5242 This study LC199 LC100 Δ pykF ::P tac : aroK This study LC224 LC100 Δ pykF ::P tac : aroB This study LC345 JE3226 ∆ pgi-2 This study LC347 LC041 ∆ pgi-2 This study LC349 QP328 Δ pykF ::P tac : aroB This study * Represents mutation in promoter of xylE 35 .\n\nIntroducing the d -xylose isomerase pathway into muconate-producing P. putida Three xylose metabolic pathways were considered to enable the production of muconate from this substrate 36 , including the isomerase pathway in which xylose is metabolized to xylulose-5-P (X5P) in the pentose phosphate pathway (PPP) 38 , the Weimberg pathway that feeds xylose to the TCA cycle via α-ketoglutarate 38 , 39 , and the Dahms pathway 40 , which shares the initial three steps with the Weimberg pathway, after which α-ketoglutaric semialdehyde is converted by an aldolase into pyruvate and glycolaldehyde. Among these, the d -xylose isomerase pathway, in which xylose is metabolized via the d -xylose isomerase ( xylA ) and xylulokinase ( xylB ) to xylulose-5-phosphate (X5P), is ideal for achieving a high theoretical muconate yield since X5P can be further converted to E4P and subsequently enter the shikimate pathway (Fig. 1a ) 35 . We integrated the isomerase pathway into a strain previously engineered to produce muconate from glucose, CJ522 3 , by overexpressing codon-optimized versions of the E. coli \n d -xylose isomerase ( xylA ), xylulokinase ( xylB ), and d -xylose:H + symporter ( xylE ), together with a transaldolase ( tal ) and a transketolase ( tkt ) to improve carbon flux within the PPP (Fig. 1a ) 35 . We also deleted hexR , which encodes a transcriptional regulator that controls expression of genes important for sugar metabolism, since we had previously found this to improve the conversion of glucose to muconate 32 . Thompson et al. previously reported that employing both the asbF and aroE pathways can help to maximize net precursor assimilation and metabolite flux toward muconate 25 . Thus, an engineered chorismate pyruvate-lyase ( ubiC-C22 ) 41 with relieved product inhibition was integrated to enhance muconate production through the shikimate pathway via aroE (Fig. 1a ). We had previously deleted pgi-1 and pgi-2 , which encode redundant glucose-6-P isomerases, to disrupt the EDEMP cycle, a combination of the Entner-Doudoroff, gluconeogenic Embden-Meyerhoff-Pernass, and the pentose phosphate pathways 42 . The purpose of disrupting the EDEMP cycle is to prevent it from cycling to generate pyruvate independent of PEP during growth on glucose, which could enable the cell to redirect carbon toward growth at the expense of muconate production, despite deletion of the genes encoding the pyruvate kinases ( pykA , pykF ) and PEP carboxylase ( ppc ) 3 . This strategy is beneficial for muconate production from glucose as the sole carbon source, but in this case, deletion of pgi-1 and pgi-2 would decrease the maximum theoretical muconate yield of both asbF- and aroE- catalyzed muconate biosynthesis pathways when xylose is converted via the PPP (Fig. 1b ). Considering that the xylose fraction in the mixture of glucose and xylose (xylose/glucose+xylose%, moles) in corn stover hydrolysate ranges from 34 to 38% (Supplementary Fig. 1 ), the modeling predicted maximum theoretical yield of muconate with pgi-1 and pgi-2 deleted to be lower than if one or both are present (Fig. 1b ). To test the hypothesis that glucose-6-phosphate isomerase (encoded by pgi-1 and pgi-2 ) activity is necessary for xylose flux to enter the EDEMP cycle, we built strains based on JE3226 35 , a P. putida KT2440-dervied strain that was previously engineered to utilize xylose using the d -xylose isomerase pathway but is otherwise wild-type, generating strains LC041 (∆ pgi-1 ), LC345 (∆ pgi-2 ), LC347 (∆ pgi-1 ∆ pgi-2 ). In plate reader cultivation on M9 medium containing xylose, LC347 failed to grow, whereas both LC041 and LC345 demonstrated reduced growth rates and increased growth lags (Supplementary Fig. 2 ). LC345, with pgi-1 intact, exhibited a lower growth rate and longer growth lag compared to LC041, which contains only pgi-2 , suggesting that Pgi-1 contributes less to the overall glucose-6-phosphate isomerase activity than Pgi-2. Since the EDEMP cycle would be expected to compete with muconate biosynthesis and reduce the muconate yield, we thus restored pgi-1 to enable xylose flux into the EDEMP cycle and improve the maximum theoretical yield, generating strain QP328 (Fig. 1a and Table 1 ). Strain QP328 was cultivated in shake flasks on a mixture of glucose and xylose to examine their conversion to muconate. Although the xylose isomerase pathway has been shown to be efficient in wild-type P. putida 35 , the xylose utilization rate of QP328, however, was very low compared to that of glucose (Fig. 1c ). Since glucose and xylose can be utilized simultaneously in the P. putida KT2440 wild-type background upon introduction of the same xylose isomerase pathway 35 , we hypothesized that a bottleneck in xylose transport or metabolism was present in our muconate-producing strain.",
"discussion": "Discussion Technologies for the production of sustainable bio-based chemicals are needed to displace incumbent petrochemicals. Critical to this endeavor is the engineering of strains to convert lignocellulosic sugars such as glucose and xylose to product at industrially relevant titer, rate, and yield. In this work, the maximum theoretical molar yield of muconate from a mixture of glucose and xylose increased from ~40% with glucose alone to 50% when the xylose content in the mixture is between 33 and 40% (mol%), which is a relevant ratio in lignocellulosic hydrolysates (Fig. 1b , Supplementary Fig. 1 ). This was achieved by introducing the d -xylose isomerase pathway to supply E4P, and reintroducing pgi-1 to enable the EDEMP cycle. ALE was used to identify additional targets to engineer a strain that ultimately achieved a 46% yield on a mixture of glucose and xylose (Fig. 5c ), considerably higher than the 35.6% we had achieved previously with a strain engineered to convert glucose alone 32 . During ALE, mutations in xylE arose in all the selected isolates (Figs. 2 a and 3f ) that improved growth on xylose. All five mutations are in the transmembrane domains of the transporter (Fig. 3f ). Based on previous work in the same system 35 and our own data showing that xylose metabolism was inhibited in the muconate-producing strain QP328 (Figs. 1 c and 2b ) but not in the non-muconate-producing analog LC345 (Supplementary Fig. 2a ), we propose the mutations were a response to inhibitor(s) from the muconate-producing background strain. Further research to identify and characterize the potential inhibitors will be pursued in future work. Moreover, increased expression of PP_2569, a putative MFS transporter, enabled by a G→A point mutation in the promoter region, led to substantially higher muconate yield and lower biomass yield in LC100 (Fig. 2b, c ), and metabolomics analysis suggested a metabolic flux redirect from the EDEMP cycle to the shikimate pathway (Fig. 3a ). Intracellular and extracellular metabolomics analysis of strains LC091 and LC100 grown on xylose suggested PP_2569 may be able to export anthranilic acid, thereby reducing the intracellular accumulation of aromatic amino acids, which are known to inhibit native DAHP synthases. We also observed higher intensities of extracellular anthranilic acid in strains QP478 and LC224 compared to the unevolved strain QP328 (Fig. 6b ). Mechanistic studies with PP_2569 may be of utility for further engineering. ALE also resulted in a duplication of the genomic region from PP_5050–PP_5242. Within this region, we demonstrated that overexpression of aroB was necessary to reach high growth rates on xylose in LC224. In strain GB062, a strain previously engineered for improved conversion of glucose to muconate by deleting hexR in CJ522 3 , transcriptomics indicated that expression of aroB was already increased upon deletion of hexR 32 . In another study in which P. putida was engineered to produce PCA from glucose, overexpression of aroB did not contribute to improved production 45 . In strain LC100 cultivated on a mixture of glucose and xylose, however, AroB activity seemed to be rate limiting, since overexpression of aroB in LC224 improved growth and muconate production (Fig. 4d–i and Supplementary Fig. 10b, c ). This may indicate that with xylose entering the non-oxidative pentose pathway, the supply of E4P was enhanced, leading to increased flux of carbon into the shikimate pathway via the condensation of E4P and PEP to DAHP. The improved level of DAHP made the next reaction, catalyzed by AroB, rate limiting, where it was not before. Overall, the engineering strategy shown here to improve flux of carbon into and through the shikimate pathway could be leveraged to improve the production of other shikimate-derived products from glucose and xylose in P. putida . Our rationally engineered strain LC224 outperformed the evolved strain QP478 in growth on xylose (Fig. 4d ). One potential reason could be the redundancy, complexity, and burden of the large duplication in QP478. Such duplications are likely enabled by recombination within similar sequences at two or more locations within the genome such that duplication of certain regions is favored or limited, ultimately limiting the ability of evolution to arrive at ideal outcomes within laboratory time scales. Genome engineering, however, can be used to make precise changes. Indeed, overexpression of aroB alone in LC224 outperformed QP478 (Fig. 4d–f and Supplementary Fig. 10a, c ), which contains the entire PP_5050–PP_5242 duplication. This demonstrates the utility and power of ALE as a tool to identify targets for rational engineering. Overexpression of aroB substantially improved the sugar utilization of LC224 relative to LC100 (Figs. 2 d and 4h ). In the metabolomics analysis, the intracellular DAHP level of LC224 is lower than the other two strains (Fig. 6 ). This may suggest that DAHP accumulation has a negative effect on sugar metabolism that can be relieved by aroB overexpression. Moreover, aroB overexpression alone in QP328, generating LC349, led to slightly lower muconate yield compared to LC224 on mixture of glucose and xylose (Supplementary Fig. 11d–f ). Since restoring the G→A mutation in P PP_2569 in the QP478 strain did not have comparable effect to reverse engineering in LC091 in terms of varying the muconate yield (Fig. 2c and Supplementary Fig. 6g, h ), these results together may suggest that aroB overexpression (resulting from duplication) and upregulation of PP_2569 have a similar effect on reducing feedback inhibition of DAHP synthases, which led to improved flux into muconate biosynthetic pathway. Metabolomic analysis of our engineered strains provides early insights into future engineering efforts for further improving muconate production, beyond what we demonstrated here with LC224, which will be pursued in future studies. The quinate accumulation by this strain (Fig. 6b ) may suggest an approach to improve its performance by overexpressing aroQ or deleting quiA . As for shikimate accumulation, overexpression of aroB and aroK did not improve the performance of LC168 relative to LC224, the equivalent strain overexpressing aroB alone (Fig. 4h and Supplementary Fig. 10c, e ). This may suggest potential bottleneck(s) in the downstream steps of the pathway, especially considering the relatively high level of extracellular anthranilic acid (ANA) accumulated by QP478 (Fig. 6b ). Insufficient conversion of chorismate (CSA) to 4-hydroxybenzoate (4HB) might be one cause of anthranilic acid accumulation. The gene ubiC-C22 encoding chorismate pyruvate-lyase, which was previously engineered to reduce product inhibition, was driven in our strains by the relatively weak lac promoter, and a potential approach to accelerate muconate biosynthesis via shikimate in LC224 could be to overexpress aroK while simultaneously increasing the expression level of ubiC-C22 . Previously conducted techno-economic analysis for the conversion of glucose as well as glucose and pentose sugars 3 indicated that the minimum selling price (MSP) of muconate would decrease substantially with increased yield and rate. Our engineered strain GB271 produced muconate from glucose at a 36% yield and a rate of 0.21 g L −1 h −1 , which corresponds with an MSP around $3 kg −1 according to this model 32 . Here, LC224 achieved a nearly 50% yield at 0.28 g L −1 h −1 (Fig. 5b ). This would reduce the MSP to around $2.2 kg −1 , which is close to the $1.96 MSP previously predicted to be commercially viable 3 . The model suggests that the MSP can be further reduced by increasing the rate and yield. Considering the 50% molar yield is already at the theoretical maximum in our strain design, further rate increases will be key to MSP improvements. In conclusion, this work demonstrates an effective strategy for producing muconate from glucose and xylose using P. putida . Considering the promising yield and titer of muconate from glucose and xylose, our strain LC224 could also represent a promising platform strain for the production of other shikimate pathway-derived compounds."
} | 5,519 |
36557508 | PMC9782629 | pmc | 4,551 | {
"abstract": "Several species of plants and animals demonstrate an ability to resist the accumulation of contaminants natural to their environments. To explain this phenomenon, mechanisms that facilitate fouling resistance have to be deciphered. Along these lines, this study is focused on the correlation between drag reduction and fouling resistance for underwater surfaces. This was accomplished by means of a novel microtopography inspired by fish-scales and conceived as a series of asymmetric triangular microgrooves oriented in the spanwise direction. A parametric study involving Large Eddy simulations was carried out to determine the most effective dimensions of the riblets and the results obtained have indicated a 9.1% drag reduction with respect to a flat reference surface. Following this, functional samples were fabricated in acrylic by means of a multi-axis micromachining center and diamond tooling. Surface quality and form accuracy of the fabricated samples were assessed with an optical microscope and optical profilometer. Finally, the fouling resistance of the samples was assessed by subjecting them to a flow of contaminated water. The results demonstrate that a relationship exists between the relative size of the particle and the fouling resistance of the microstructured surface.",
"conclusion": "5. Summary and Conclusions Novel bioinspired asymmetric triangular riblet surfaces have been shown to exhibit vortex drag reduction and a delayed transition to turbulence through numerical simulations. Optimal dimensions are largely dependent on the flow conditions and past studies proposed that drag reduction performance increases with Reynolds number with a maximum achieved performance of 9.1% compared to a flat surface. The holistic approach presented in this study includes all phases of research and development focused on fouling resistant microtopographies. This represents a departure from past attempts that were focused on limited aspects related to fouling resistant surfaces. The framework presented in this work can be used as a foundation for future modeling studies to provide designs capable to provide the desired functionality and manufacturability at micro/nano-scales. To investigate the connection between drag reduction and fouling resistance, three samples were fabricated from PMMA. The results obtained suggest that the analyzed microstructures do not exhibit fouling resistant characteristics under the low Reynolds conditions yielded by the experimental setup used. Nonetheless, this study cannot be considered conclusive such that future experimental studies at higher Reynolds numbers will be conducted to further investigate the potential of the fabricated ATR samples. The following conclusions can be drawn from this study: Drag reduction was achieved for the full range of dimensions and Reynolds numbers considered in the numerical simulations While the experimental differences were rather small and therefore less conclusive, the trials performed suggested that the low Reynolds drag reduction is associated with smaller feature heights when compared to their higher Reynolds counterparts Drag reduction increases with the Reynolds number ATR structures can be fabricated through micromilling; their form accuracy was ±7 µm whereas the lowest areal roughness was S a = 85.2 nm The fouling resistance trials suggest that structures larger than the contaminating particles are more effective at reducing the overall settlement Future work will focus on the development of an experimental apparatus that enables a higher channel velocity in order to be able to match the Reynolds numbers considered in the numerical simulations. This will allow the experimental verification of the drag reduction as well as the investigation of fouling resistance in the drag reduction region. Furthermore, the parametric study will be expanded to determine optimal feature dimensions with regard to drag reduction performance.",
"introduction": "1. Introduction Many marine creatures such as mollusks, crabs, sharks, fish, and sea stars demonstrate varying degrees of protection against the accumulation of environment contaminants on their outer surfaces [ 1 , 2 , 3 ]. This ability is known as fouling resistance and has also been observed in several species of plants. For many of these creatures, this capability is crucial for their survival since they lack the ability to clean themselves. The accumulation of fouling on sharks and fish makes them more susceptible to skin diseases. Fouling also increases the hydrodynamic drag and this results in increased energy needs and/or reduced swimming speeds that lead to susceptibility to predators and food scarcity. Likewise, surface fouling on plants blocks the sunlight required to create the energy required for metabolic processes such as photosynthesis. Fouling also affects man-made systems. For instance, the underwater surfaces of ship hulls, buoys and offshore structures are susceptible to the accumulation of aquatic organisms (barnacles, algae, etc.). These organic accumulations roughen the surface of ships and this increases drag and thereby stress on the propulsion system, ultimately leading to increased fuel consumption rates. The aquatic growth also increases corrosion rates and thus reduces equipment lifetime. The annual global cost associated with fuel consumption, cleaning and maintenance is estimated to be more than $150 billion USD [ 4 , 5 ]. To combat fouling, ship hulls have been coated with tin-based biocide paints that provide protection for up to seven years. These special paints leach out compounds of tin that poison the nearby organisms and prevent their adhesion. Nonetheless, while being effective at hull fouling prevention, these paints proved to be hazardous to non-targeted marine species and eventually to humans that might be consuming them. By 2008, a global ban was issued on the use of any paint containing specific compounds of tin [ 6 ]. While alternative coatings were developed, they were reportedly being less effective while being associated with a certain level of toxicity to aquatic organisms. Because of this, alternative biocide-free fouling resistant solutions became an important avenue to be explored. Along these lines, much of the work conducted recently was focused on the understanding of the natural defense mechanisms employed by various members of the flora and fauna. So far, researchers have proposed five mechanisms capable to lead to fouling resistance: surface topography, mucous secretion, flexion, sloughing, and surface energy [ 2 ]. Each of these mechanisms can be employed on its own or in combination with others. For engineering applications, the most common fouling resistant mechanism is constituted by surface topography, typically because of its passive nature as well as its applicability across a wide range of materials. To date, sharkskin is typically regarded as one of the best models of fouling resistant topography. An adhesive film mimicking its microtopography is presently available on the market [ 7 , 8 ]. The skin of sharks is comprised of many tooth-like denticles each comprised of several microriblets that are positioned parallel with the direction of the flow. While the basic geometry of the microstructures does not vary much among shark species, the length of their scales is variable and could be anywhere between 30 µm and 300 µm [ 9 , 10 ]. Past research efforts have shown that bio-inspired shark skin topography can reduce the friction drag associated with turbulence by nearly 10%. Subsequent studies found that shark skin-inspired microriblets are capable of reducing surface fouling by as much as 86% [ 11 ]. It is believed that a complementary correlation exists between the fouling resistance of this surface and its ability to reduce drag, possibly because the rapidly moving water layer located in the close proximity of the structured surface carries away particles and organisms that would otherwise settle and adhere to the surface. Nevertheless, the bioinspired shark skin topography does not seem a decisive solution to surface fouling. In a long-term field trial of shark skin-inspired topography, the fouling deterrent effect was significantly diminished after six months [ 12 ]. This loss in performance prompts the need for further development, particularly with respect to microstructure geometries that are inspired by other types of native fouling resistant surfaces. Additionally, investigating the correlation between drag reduction and fouling resistance for alternative surfaces may reveal fundamental underlying mechanisms employed by nature. Along these lines, fish scales have also been associated with drag reduction in low Reynolds flow [ 13 ]. In this case, the proposed mechanism was a delayed onset of the turbulent flow and the increased drag associated with it. In this regard, Muthuramalingam et al. [ 13 ] demonstrated a 3% drag reduction for fish scales with respect to a flat witness surface. Although fish scales were not investigated to the same extent as shark skin structures, the preliminary studies of fish scales showed promising fouling resistant properties [ 14 ]. The hydrophobicity/oleophobicity demonstrated by biological fish scales is believed to contribute to their fouling resistance. In this context, it becomes increasingly clear that a better understanding of the correlation between surface microstructures and drag reduction is quintessential for generation of “man-made” surfaces whose fouling resistance would equal or exceed that of their natural counterparts. Building on this, the primary objective of this study was to examine the drag reducing mechanisms associated with the bioinspired ribletted microstructures along with their ability to resist fouling. For this purpose, a novel riblet geometry was analyzed as a possible candidate of a drag reduction surface topography therefore allowing for the investigation of the anticipated correlation leading to fouling resistance. Computational fluid dynamics (CFD) models of the topography were performed for a range of microriblet dimensions in an attempt to evaluate drag and establish functional geometries. Samples were then microfabricated in PMMA (polymethyl methacrylate) followed by the analysis of form accuracy and surface quality for surface microstructures. Finally, the fouling resistance of the samples was investigated by means of a gravity driven channel and contaminated flow."
} | 2,616 |
32106516 | PMC7142476 | pmc | 4,553 | {
"abstract": "Iron-rich pelagic aggregates (iron snow) are hot spots for microbial interactions. Using iron snow isolates, we previously demonstrated that the iron-oxidizer Acidithrix sp. C25 triggers Acidiphilium sp. C61 aggregation by producing the infochemical 2-phenethylamine (PEA). Here, we showed slightly enhanced aggregate formation in the presence of PEA on different Acidiphilium spp. but not other iron-snow microorganisms, including Acidocella sp. C78 and Ferrovum sp. PN-J47. Next, we sequenced the Acidiphilium sp. C61 genome to reconstruct its metabolic potential. Pangenome analyses of Acidiphilium spp. genomes revealed the core genome contained 65 gene clusters associated with aggregation, including autoaggregation, motility, and biofilm formation. Screening the Acidiphilium sp. C61 genome revealed the presence of autotransporter, flagellar, and extracellular polymeric substances (EPS) production genes. RNA-seq analyses of Acidiphilium sp. C61 incubations (+/− 10 µM PEA) indicated genes involved in energy production, respiration, and genetic processing were the most upregulated differentially expressed genes in the presence of PEA. Additionally, genes involved in flagellar basal body synthesis were highly upregulated, whereas the expression pattern of biofilm formation-related genes was inconclusive. Our data shows aggregation is a common trait among Acidiphilium spp. and PEA stimulates the central cellular metabolism, potentially advantageous in aggregates rapidly falling through the water column.",
"conclusion": "5. Conclusions Aggregation appears to be a common mechanism in all Acidiphilium spp., since nearly 4% of their shared gene clusters are associated with mechanisms responsible for aggregation, including autoaggregation, motility (flagellar assembly, chemotaxis), and biofilm formation (exopolysaccharide biosynthesis and secretion). All genes associated with these mechanisms were transcribed under our incubation conditions; however, RNA-seq data did not show clear evidence that PEA affected aggregate formation directly. Inconsistent gene expression patterns relating to the formation and secretion of EPS and flagellar-based motility, despite enhanced aggregate formation with the addition of PEA, suggests this compound functions as an infochemical regulating other cellular mechanisms, and not aggregation mechanisms directly. In fact, Acidiphilium cells seem to retain motility within the aggregates. We did observe induced upregulation of glycolysis, the TCA cycle, oxidative phosphorylation, and synthesis of ribosomes, although these activities were not linked to enhanced growth. Degradation of polysaccharides appears to be a major function within the heterotrophic Alphaproteobacterial genus Acidiphilium , which is optimized by the complementarity of specific genes present in unique strains in addition to shared core functions.",
"introduction": "1. Introduction Pelagic aggregates, composed of microorganisms, phytoplankton, feces, detritus, and biominerals, are local hotspots for microbial interaction in nearly all aquatic habitats [ 1 , 2 , 3 ]. These snow-like aggregates are stabilized by a matrix of extracellular polymeric substances (EPS) and vary in size, ranging from micrometers to centimeters, depending on their residence time in the water column and the trophic state of the ecosystem [ 1 , 4 , 5 ]. Microbial colonization and coordinated group behavior within these pelagic aggregates are likely regulated by chemical signaling, including quorum sensing signaling molecules [ 6 , 7 ]; however, most chemical mediators involved in interspecies interaction are still unknown. Iron-rich pelagic aggregates (iron snow), analogous to the more organic-rich marine or freshwater aggregates, are characterized by lower chemical and microbial complexity [ 5 ]. Iron snow forms at the redoxcline of stratified iron-rich lakes, where the oxygen-rich epilimnion water meets ferrous iron (Fe 2+ ) of the anoxic hypolimnion [ 5 ]. Many of these lakes are acidic due to the inflow of protons in addition to Fe 2+ and sulfate (SO 4 2- ) from mine tailings [ 8 , 9 ]. Under acidic conditions, Fe 2+ is oxidized by microorganisms to ferric iron (Fe 3+ ), from which goethite and schwertmannite form via hydrolysis of Fe 3+ cations [ 10 , 11 ]. In lignite mine lakes, these biominerals form the main inorganic component of pelagic aggregates [ 12 , 13 ]. These aggregates are stabilized by the adsorption of other metals, nutrients, and organic matter. Iron snow is an attractive habitat for heterotrophic microorganisms, especially those capable of using Fe 3+ as an electron acceptor, such as Acidiphilium species [ 14 ]. Together, iron-oxidizing bacteria (FeOB) and iron-reducing bacteria (FeRB) can comprise up to 60% of the total microbial community found in iron snow aggregates [ 15 ]. To study the interactions between these iron-cycling bacteria, we isolated several key players from iron snow, including Acidiphilium , Acidocella , Acidithrix, and Ferrovum species [ 13 , 16 , 17 ]. The iron-oxidizing isolate Acidithrix sp. C25 forms large cell-mineral aggregates in the late stationary phase [ 13 ]. When co-cultured with the iron-reducing isolate Acidiphilium sp. C61, motile cells of Acidiphilium also form cell aggregates with similar morphology to iron snow. Comparative metabolomics identified the aggregation-inducing signal, 2-phenethylamine (PEA), which also induced faster growth of Acidiphilium sp. C61 [ 17 ]. PEA is a small molecule that exhibits an array of seemingly unrelated functions, including roles as a neurotransmitter and in food processing [ 18 ]. PEA was found in the brains of humans and other mammals [ 19 ] and reportedly has stimulatory effects, resulting in the release of biogenic amines [ 20 ]. In high concentrations, PEA can act as an anti-microbial against Escherichia coli on beef meat [ 18 ]. Bacteria can produce PEA via decarboxylation of phenylalanine or as a by-product of the tyrosine decarboxylase reaction [ 21 ]. PEA is capable of inhibiting both swarming and the expression of the flhDC gene cluster, which encodes a flagellar regulon that regulates flagellar motility in Proteus mirabilis [ 22 , 23 ]. Swarmer cell differentiation is dependent on specific environmental conditions, including the presence of a solid surface, inhibition of flagellar rotation, and density-based cell–cell signaling by extracellular signals [ 24 , 25 , 26 ]. However, swarming is not known to exist in Acidiphilium spp. and this flhDC gene cluster is absent in all sequenced Acidiphilium spp. genomes [ 17 ]. Therefore, the molecular mechanisms underlying PEA-induced aggregate formation in Acidiphilum spp. remain unknown. To broaden our understanding of chemical communication between iron-cycling bacteria shaping pelagic aggregates, we amended different Acidiphilium spp. and two other iron snow key players with PEA to see if this aggregation effect was isolate specific. We sequenced the genome of Acidphilium sp. C61 to gain more insights into the metabolic pathways and potential behaviors (e.g., motility, chemotaxis) of this organism. Furthermore, we performed comparative transcriptomics of Acidphilium sp. C61 amended with 10 μM PEA compared to cultures without PEA to elucidate the genetic mechanisms underlying aggregate formation.",
"discussion": "4. Discussion Bacteria of the heterotrophic alphaproteobacterial genus Acidiphilium are ubiquitous in acidic environments [ 63 ]. These heterotrophs are often isolated as contaminants from iron-oxidizing mixed cultures composed of acidophiles like Acidithiobacillus ferrooxidans [ 64 , 65 ] or species related to Ferrovum myxofaciens P3G [ 58 ]. In these iron-oxidizing mixed cultures, Acidiphilium spp. enhance the activities of these chemolithoautotrophs in bioleaching. In return, Acidiphilium spp. seem to benefit from their secreted metabolites and biomass remnants [ 28 , 66 ]. Acidiphilium spp. have been also directly isolated from acidic mine drainage waters and sediments [ 14 , 67 ] and from acidic hypersaline river sediments in Australia, where they can make up high relative fractions of the microbial community [ 68 ]. Independent of their original ecological niche, all seven Acidiphilium spp. analysed by pangenomics show high similarities regarding their functional genome organization. Not surprisingly, both strains isolated from the same lake share the highest number of accessory gene clusters (93 gene clusters), with most of them being related to hypothetical proteins except a few related to transporters. Genes encoding different mechanisms of aggregation were present in all seven genomes, i.e., genes involved in the synthesis and secretion of EPS, suggesting that these mechanisms of aggregation are common in Acidiphilium spp. Indeed, all three Acidiphilium isolates tested in this study were able to aggregate to some extent, even without PEA addition. This morphological feature observed in Acidiphilium isolates has been previously documented, for example, the salt-tolerant Acidiphilium strain, AusYE3-1, also forms flocs and alters cell shapes from rod-shaped or coccobacillus to filamentous structures when stressed under high salt concentrations [ 68 ]. Our study shows PEA enhanced aggregation of all Acidiphilium strains tested, but not of other acidophiles [ 15 , 17 ] also present in iron snow. However, the PEA enhanced aggregate formation of Acidiphilium sp. C61 was less pronounced ( Figure 2 a) compared to the high number of large macroscopic cell aggregates formed by cultures of Acidiphilium sp. C61 soon after isolation from iron snow [ 17 ]. In that previous study, increased growth in the presence of 10 µM PEA was also observed, which could not be repeated in our study, suggesting adaptations during extended laboratory incubation of Acidiphilium sp. C61. Based on our previous model [ 17 ], we anticipated that PEA induced gene expression changes would primarily be related to motility similar to its role in Proteus mirabilis [ 22 , 23 ]. However, the assembled genome of Acidiphilium sp. C61 lacks the flhDC gene cluster present in P. mirabilis , and flagellar motility was not negatively affected by PEA addition. Motility still seems to be essential for Acidiphilium sp. C61, as the six genes involved in flagella biosynthesis were even slightly upregulated. This finding agrees with the results of a metaproteomic approach, which detected many flagellin domain proteins from Acidiphilium spp. in iron snow samples [ 15 ]. Furthermore, chemotaxis sensor proteins were downregulated in the presence of PEA, enabling more smooth swimming. Thus, flagellar motility might help Acidiphilium sp. C61 join iron oxidizers, like Acidithrix sp. C25 in the growing aggregate, then again, there may not be sufficient time for the microorganism to switch from a pelagic to an attached lifestyle. Acidiphilium , Acidithrix , Acidocella, and Ferrovum spp. can make up 53% of the total bacterial community of aggregates formed in acidic lignite lakes [ 15 ]. In these shallow lakes, iron snow forms a continuous shower of iron minerals, (in)organic matter and microorganisms (∼10 8 –10 10 cells (g dry wt −1 )) rapidly falling through the water column to the sediment [ 5 , 69 , 70 ]. Thus, there is only a short lifespan of these pelagic aggregates, which consequently means there is only limited time for microbial-coordinated activities, and for energy and matter fluxes to occur within these aggregates. Although acyl-homoserine lactone (AHL) mediated gene regulation has been shown to influence EPS production and biofilm formation in many proteobacteria, including A. ferrooxidans [ 71 ], we could not find autoinducer synthesis or receptor genes linked to quorum sensing in the genome of Acidiphilium sp. C61. Thus, communication appears to occur via other interaction mechanisms mediated by diffusive exometabolites (infochemicals). Bacterial EPS is usually composed of a mixture of polysaccharides, proteins, lipids, and extracellular DNA (eDNA) [ 72 , 73 ]; however, the main constituents of EPS extracted from Acidiphilium strain 3.2Sup(5) are proteins and carbohydrates mostly composed of carboxylic, hydroxylic, and amino groups [ 74 ]. Although we observed the upregulation of several genes for exopolysaccharide precursor synthesis (e.g., UDP-glucose, UDP-galactose) and capsular polysaccharide exporters in the presence of PEA, the overall expression pattern of genes involved in polysaccharide synthesis, as well as autotransporters, were inconsistent. Thus, we cannot conclude that biofilm formation, in general, is enhanced in the presence of PEA, nor can we explicitly conclude the mechanisms involved in Acidiphilium sp. C61 biofilm formation. Similarly, we did not detect significantly enhanced eDNA concentrations, indicating eDNA is likely not a primary constituent of EPS secreted by Acidiphilium sp. C61 and Acidiphilium sp. C61 may prefer to aggregate with other cells over forming biofilms. The high surface area of the poorly crystalline iron mineral schwertmannite, which forms the inorganic matrix of iron snow [ 13 , 69 ], favors adsorption of organic matter that are ideal substrates for Acidiphilium spp. [ 14 , 15 ]. The above mentioned metaproteomic approach also identified Acidiphilium -related glucose uptake proteins in iron snow [ 15 ]. The genome of Acidiphilium sp. C61 contains ABC transporters for the uptake of ribose, fructose, and xylose ( Figure 4 ). In contrast to the genome of Acidiphilium sp. JA12-A1 that lives in co-culture with Ferrovum sp. JA12 [ 58 ], we did not find polysaccharide-hydrolyzing enzymes, such as β-glucosidases, or break down EPS or cell envelope polysaccharides from decaying cells endoglucanases in Acidiphilium sp. C61. However, glycoside hydrolase, alpha-amylase, beta-N-acetylhexosaminidase, and glucoamylase were present in all Acidiphilium spp. based on the pangenomic analysis (GC_1878, GC_1672, GC_1296, GC_1827) ( Table S2-3 ). In addition, Acidphilium sp. C61 possesses one more unique glycoside hydrolase (GC_6119), whereas another glycosidase (GC_1572) is present in the other six Acidiphilium strains. The capacity for polysaccharide degradation seems to be a common trait for Acidiphilium spp., but individual differences exist between the strains. Thus, these individual differences allow for niche differentiation and also ensures complementarity, since a diverse mixture of strains will colonize specific habitats. In general, sugar compounds appear to be the preferred carbon source for biomass production in all Acidiphilium sp. We identified full sets of genes of the pentose phosphate pathway, compensating for the incomplete glycolysis pathway, a complete tricarboxylic acid (TCA) cycle, and genes encoding all pathways necessary for the synthesis of proteinogenic amino acids, nucleotide, and fatty acid biosynthesis. Acidiphilium sp. C61 is capable of urea uptake, a unique trait among Acidiphilium sp. Thus, it can be characterized as a prototrophic cell, able to synthesize all the compounds needed for growth listed above without the need for a partner organism. Different Acidiphilium strains present in complex communities appears to release a diverse suite of glycoside hydrolases and glucosidases to utilize the organic substances secreted by other community members or derived from microbial cell decay. In return, Acidiphilium spp. provide the chemolithoautotrophs with elevated CO 2 concentrations locally, which is advantageous especially in low pH environments, such as acidic coal mining lakes. This type of interspecies carbon transfer has been previously described for acidophilic mixed cultures containing Acidiphilium cryptum and Acidithiobacillus ferrooxidans [ 75 ] and other mixed cultures derived from a pilot plant for remediation of acid mine drainage (AMD) containing Acidiphilium sp. JA12-A1 and an iron oxidizer related to Ferrovum myxofaciens P3G [ 76 ]. To our surprise, PEA did not preferentially affect one or more mechanisms of aggregate formation in Acidiphilium but induced upregulation of the central cellular metabolism by affecting more than 50% of the genes involved in glycolysis, the TCA cycle, and oxidative phosphorylation. Similarly, the synthesis of ribosomes, amino acid biosynthesis and transcription, as well as secretion systems, were stimulated. This broad range of affected upregulated genes points to a more general stimulatory mechanism of PEA, similar to its general role as a neurotransmitter [ 18 ] and stimulator for the release of biogenic amines in humans [ 20 ]. Thus, it is probable that these Acidiphilium cells are just more active in iron snow in the presence of the infochemical PEA released by Acidithrix sp. C25. After the formation of iron minerals at the oxic-anoxic interface, iron snow will reach anoxic conditions in the hypolimnion. Since Fe 3+ is energetically much more favorable as an electron acceptor at acidic compared to pH neutral conditions [ 77 ], the majority of the chemolithoautotrophic Fe 2+ oxidizers are also capable of Fe 3+ reduction, including Acidithrix sp. C25 [ 13 ]. These heterotrophic Acidiphilium spp., as well as other heterotrophic acidophiles, are also capable of Fe 3+ reduction even in the presence of oxygen [ 78 , 79 , 80 ]. Thus, single cells within the iron snow aggregates may begin to respire Fe 3+ in the redoxcline, even at low oxygen concentrations. Switching to this anaerobic metabolism requires activation, as genes responsible for Fe 3+ reduction in Acidiphilium spp. do not seem to be constitutively expressed [ 78 ]. Although the Fe 3+ reduction mechanism in Acidiphilium spp. has not yet been revealed in detail, different membrane-associated proteins potentially related to electron transport chain genes have been identified in iron snow, including OmpA/MotB domain proteins, TonB-dependent receptor, and ApcA [ 15 ]. Genome assembly of Acidiphilium sp. C61 reveals MsrQ that can bind to two b-type hemes via conserved histidine residues along with MsrP; these proteins form a methionine sulfoxide reductase operon functioning to repair oxidized periplasmic proteins [ 81 ]. Additionally, the cytosolic NAD(P)H flavin reductase (Fre) has been shown to function as a proficient electron donor to MsrQ moieties and the soluble dehydrogenase partner, in Escherichia coli , for example [ 81 ]. These findings suggest that Fre and MsrPQ might form a membrane-spanning two-component system for electron transfer ( Figure 4 ). Because MsrPQ is involved in oxidative stress response, specifically in the repair of oxidized periplasmic proteins, such as oxidized methionine residues, there is a potential role for the MsrPQ operon in the maintenance of the activated methyl cycle, which can be remotely linked to iron reduction via the transsulfuration pathway. We also identified a gene coding for an arsenate reductase (AcpC61_1183). Previous studies suggest that TetH or ArsH have the potential to mediate Fe 3+ reduction in acidophiles [ 62 , 82 , 83 ]. However, since we did not perform RNA-seq analysis of Acidiphilium sp. C61 under iron-reducing conditions, we do not know how PEA would affect its anaerobic metabolism."
} | 4,840 |
24309679 | null | s2 | 4,556 | {
"abstract": "Previously we showed how delay communication between globally coupled self-propelled agents causes new spatio-temporal patterns to arise when the delay coupling is fixed among all agents [1]. In this paper, we show how discrete, randomly distributed delays affect the dynamical patterns. In particular, we investigate how the standard deviation of the time delay distribution affects the stability of the different patterns as well as the switching probability between coherent states."
} | 121 |
38637667 | PMC11026385 | pmc | 4,559 | {
"abstract": "Dominant vegetation in many ecosystems is an integral component of structure and habitat. In many drylands, native shrubs function as foundation species that benefit other plants and animals. However, invasive exotic plant species can comprise a significant proportion of the vegetation. In Central California drylands, the facilitative shrub Ephedra californica and the invasive Bromus rubens are widely dispersed and common. Using comprehensive survey data structured by shrub and open gaps for the region, we compared network structure with and without this native shrub canopy and with and without the invasive brome. The presence of the invasive brome profoundly shifted the network measure of centrality in the microsites structured by a shrub canopy (centrality scores increased from 4.3 under shrubs without brome to 6.3, i.e. a relative increase of 42%). This strongly suggests that plant species such as brome can undermine the positive and stabilizing effects of native foundation plant species provided by shrubs in drylands by changing the frequency that the remaining species connect to one another. The net proportion of positive and negative associations was consistent across all microsites (approximately 50% with a total of 14% non-random co-occurrences on average) suggesting that these plant-plant networks are rewired but not more negative. Maintaining resilience in biodiversity thus needs to capitalize on protecting native shrubs whilst also controlling invasive grass species particularly when associated with shrubs.",
"conclusion": "Conclusions and implications Interactions with foundation and invasive species are a powerful tool to advance community assembly theory and inform management of habitat through vegetation in dryland ecosystems. Centrality metrics from ecological network theory directly informs species richness given the importance of foundational native and introduced keystone species. It is a strongly complementary measure because it examines composition differences and how they shift in response to invasion (and to the shrubs or keystone species if included in the data structure). This metric provides an alternative, new facet to summarize community structure that is appropriately represented within communities best simplified by foundation and keystone species concepts. It can be used to infer key levers, such as invasion at fine-scales, that can shape plant-plant associations and potential interactions 68 . Based on the present analyses, the restructuring of the plant community is evident. When connections within a community shift with invasion, it is a significant conservation concern. Other plant communities may therefore be vulnerable to the same degree of restructuring observed within this study if a pervasive enough exotic plant becomes established. Although not tested here, the long-term temporal changes that occur within the network can provide useful insights into the potential restructuring over time 69 . Both the relative proportion and overall density of the invasive species can likely alter the degree of centrality in the network. Research is needed to examine network stability and resilience in the face of increasing invasions and change. We also need to begin to include other measures of network connectivity and centrality. This is a novel approach, but the findings are nonetheless promising very broadly in invasion biology provided there are clear factors such as distinct microsites or key dominant plants to reasonably anchor analyses. Here, the presence of brome reduced species diversity, but it also had more subtle effects on the plant-plant interactions inferred from the network analysis. There was not a dramatic shift to more negative interactions, but instead, there was a shift to more tightly connected and thus less resilient networks of species 60 . Furthermore, brome increased centrality more under shrubs transforming their capacity as native, foundation species to support diverse and potentially resilient plant communities. Consequently, if native shrubs are to be leveraged for restoration and conservation as viable local sinks for regional biodiversity 70 , targeted control of invasive species such as Bromus rubens must selectively and aggressively reduce their presence at these fine-scale hotspots.",
"introduction": "Introduction The movement of species around the globe is a contemporary facet of anthropogenic global change. This introduces exotic species to communities comprised of species that have coexisted for millennia, and some of these exotic species will become invasive 1 – 3 . Invasive plant species can negatively influence almost every ecological dimension of pattern and process from plant-plant 4 , 5 , to plant-animal 6 , 7 , to ecosystem-level functions 8 , 9 . Declines in resident native diversity, function, and resilience to other exotic species are common negative outcomes 10 , 11 . As a counterpoint to some of the effects of invasive plant species, foundation native plant species are typically defined as plants that provide significant cover or structure in many systems including drylands 12 – 14 . This benefactor species can in turn provide structure for and benefits to other plants 15 , 16 or animals 17 . Consequently, a deeper understanding of the capacity for these foundation plant species to support diverse and potentially resilient communities regionally is critical 18 , 19 . Re-establishment of native vegetation is also a common goal in the restoration in many terrestrial habitats 20 . Foundation plant species can thus fulfill both a fundamental and applied purpose supporting the communities they occupy, but the facilitation effects must understood within the broader context of species interactions. Fundamentally, we must ascertain whether native foundation plant species can offset the negative ecological effects of invasive plants 21 either in terms of preserving a proportion of the native species 22 or in protecting some of the capacity for interactions between the remaining species locally or regionally to respond to additional changes 23 . Facilitation or positive plant-plant interactions will thus be a pivotal stabilization process provided native foundation species such as shrubs facilitate other natives 24 . Unfortunately, this is not always the case. Invasive exotic species such as annual plants can often take advantage of these native foundation species 25 – 27 . Facilitation by native shrubs can become a two-edged sword—as demonstrated in drylands 28 . We need to better understand whether the declines in richness of native plant species in this specific context, whilst clearly negative, are still protected in some capacity through their connections to one another in the remaining associational network 29 . Contrasting the connections between species with and without the effects of foundation species and with and without a highly invasive annual species will highlight the importance of moving beyond single species foci in considering whole-community assembly 30 . Here, we test the hypothesis that the centrality or frequency of associations between resident species within fine-scale ecological networks can be mediated by shrub facilitation in the face of a common, disruptive invasive plant species within a dryland region. Fine-scale plant ecological surveys provide a powerful lens into local interactions and association patterns. Censuses of diversity, abundance, and habitat structure provided by foundation and dominant plant species are key approaches to infer the importance of microclimate 31 , differences in distribution locally and regionally 32 , richness patterns 33 , and interactions with the environmental drivers of change including soil dynamics 34 . In drylands, i.e. arid and semi-arid grass and mixed shrub lands 35 , a simple two-phase model or categorization is often used to structure vegetation sampling, typically termed shrub/open microsite—but it also can be improved in future research by incorporating distances from shrubs 30 . This structured, simple paired sampling best nonetheless approximates the vegetation mosaic in dryland ecosystems for this region and particularly for sampling across distributed sites 36 . Distributed fine-scale sampling such as paired shrub-open sampling within a region is an effective experimental design for surveys testing for differences in distribution and diversity at scales relevant to plant-plant interactions and invasive species impacts on community assembly in deserts 37 , 38 . Here, we used this approach to test the generalized hypothesis proposed that a native shrub can mitigate some of the effects of a widespread invasive plant within the central drylands of California using a co-occurrence and interaction network framework to capture salient community dynamics.",
"discussion": "Discussion The implication of these robust findings is that individual species under shrubs invaded by brome experience relatively higher degrees of centrality within the networks or assemblages. This implies that these co-occurring species are more highly connected, likely generalist in association, and at least some of the species such as brome can function more fundamentally as keystone species 58 . Higher mean degree centrality is a very simple count of links for each species in each of these specific ecological contexts. This may seem like a positive ecological outcome in terms of resilience, but higher centrality can in some instances reduces resilience and capacity for a set of species to respond to change and disturbance because the species are more frequently associated in space or time 60 . Greater centrality in the network suggests some species are more important, and thus the entire assemblage is sensitive to the loss of one species. For example, in plant-pollinator systems with higher mean degrees of centrality (more frequent associations between individual species), the loss a single species of plant for floral resources or a pollinator species critical for plant reproduction can become important locally if either species is connected to many other species within the community 60 . Species redundancy can offset some of the losses in reliability of key ecological functions, but without experimentation to test for engineering capacities for each species, it is prudent to assume that more flexible networks are a desirable conservation outcome 64 . This same resilience principle is relevant here in plant-plant interaction networks. More links to each species under shrubs with brome suggests that loss of a species can result in a significant erosion of sets of ecological interactions—whether positive in supporting persistence through amelioration 65 or negative interactions through indirect competition that can promote higher annual plant diversity in drylands 66 . There was also a significant difference in the contrast between shrub and open microsites without brome, i.e., shrub-only networks were less connected with lower relative centrality (Fig. 3 , GLM estimated marginal means post hoc contrast, estimate = 0.366, p = 0.37). This further suggests that communities or assemblages of species without brome but facilitated by native shrubs are more resilient and thus able to respond to changes, i.e., fewer specific-species annual plant associations. Less associations per species under shrubs without brome means less to lose per species. Change capacity is supported best through opportunities to shift co-occurrences in annual plant communities 36 , 67 . Colloquially, all eggs are not in one or fewer baskets. Consequently, the compounded effects of brome in reducing native species diversity coupled with the increasing ‘keystoneness’ under shrubs, that generates more tightly connected species in the remaining communities, suggests that it can usurp the central role of shrubs as agents of positive change in drylands. Brome co-opted the positive effects of native foundation species and established new critical thresholds in diversity and associations that are less robust 10 . Strategies that balance dominant native species conservation and even restoration within these ecosystems must plan for the increased likelihood that sub-dominant exotics will become both particularly invasive into these local ecological opportunities and even undermine responsiveness of the remaining species to future change."
} | 3,098 |
23977959 | null | s2 | 4,560 | {
"abstract": "LigAB from Sphingomonas paucimobilis SYK-6 is the only structurally characterized dioxygenase of the largely uncharacterized superfamily of Type II extradiol dioxygenases (EDO). This enzyme catalyzes the oxidative ring-opening of protocatechuate (3,4-dihydroxybenzoic acid or PCA) in a pathway allowing the degradation of lignin derived aromatic compounds (LDACs). LigAB has also been shown to utilize two other LDACs from the same metabolic pathway as substrates, gallate, and 3-O-methyl gallate; however, kcat/KM had not been reported for any of these compounds. In order to assess the catalytic efficiency and get insights into the observed promiscuity of this enzyme, steady-state kinetic analyses were performed for LigAB with these and a library of related compounds. The dioxygenation of PCA by LigAB was highly efficient, with a kcat of 51 s(-1) and a kcat/KM of 4.26 × 10(6) M(-1)s(-1). LigAB demonstrated the ability to use a variety of catecholic molecules as substrates beyond the previously identified gallate and 3-O-methyl gallate, including 3,4-dihydroxybenzamide, homoprotocatechuate, catechol, and 3,4-dihydroxybenzonitrile. Interestingly, 3,4-dihydroxybenzamide (DHBAm) behaves in a manner similar to that of the preferred benzoic acid substrates, with a kcat/Km value only ∼4-fold lower than that for gallate and ∼10-fold higher than that for 3-O-methyl gallate. All of these most active substrates demonstrate mechanistic inactivation of LigAB. Additionally, DHBAm exhibits potent product inhibition that leads to an inactive enzyme, being more highly deactivating at lower substrate concentration, a phenomena that, to our knowledge, has not been reported for another dioxygenase substrate/product pair. These results provide valuable catalytic insight into the reactions catalyzed by LigAB and make it the first Type II EDO that is fully characterized both structurally and kinetically."
} | 477 |
37766973 | PMC10520884 | pmc | 4,561 | {
"abstract": "Summary Even though the discovery of lytic polysaccharide monooxygenases (LPMOs) has fundamentally shifted our understanding of biomass degradation, most of the current studies focused on their roles in carbohydrate oxidation. However, no study demonstrated if LPMO could directly participate to the process of lignin degradation in lignin-degrading microbes. This study showed that LPMO could synergize with lignin-degrading enzymes for efficient lignin degradation in white-rot fungi. The transcriptomics analysis of fungi Irpex lacteus and Dichomitus squalens during their lignocellulosic biomass degradation processes surprisingly highlighted that LPMOs co-regulated with lignin-degrading enzymes, indicating their more versatile roles in the redox network. Biochemical analysis further confirmed that the purified LPMO from I. lacteus CD2 could use diverse electron donors to produce H 2 O 2 , drive Fenton reaction, and synergize with manganese peroxidase for lignin oxidation. The results thus indicated that LPMO might uniquely leverage the redox network toward dynamic and efficient degradation of different cell wall components.",
"introduction": "Introduction The deconstruction of lignocellulosic biomass into its molecular building blocks is essential for global carbon cycle and sustainable production of renewable fuels, chemicals, and materials. Lytic polysaccharide monooxygenases (LPMOs) attracted broad interest for lignocellulose deconstruction due to its unique capacity to oxidize carbohydrates and to synergize with hydrolytic enzymes for biomass conversion to sugars. 1 , 2 LPMOs were originally characterized as glycoside hydrolase family 61 or carbohydrate-binding module 33. 3 Recent breakthroughs expanded the substrates of LPMO and discovered the oxidative activities of LPMOs on all major polysaccharides, including cellulose, 4 , 5 hemicellulose, 6 , 7 , 8 chitin, 3 , 9 and starch. 10 Current studies also suggested that LPMOs oxidized carbohydrate through recruiting oxygen to copper active site 11 , 12 , 13 and accepting electrons from a variety of external electron donors, including cellobiose dehydrogenase (CDH), 14 photosynthetic pigment, 15 and small molecular reductants. 1 , 3 , 5 , 16 , 17 Despite the progresses, it is still unclear if LPMOs oxidize the substrates beyond the carbohydrates in lignocellulose. Our recent work showed that the purified LPMO could supply H 2 O 2 for versatile peroxidase to derive its oxidation activity to lignin oxidation. 18 Meanwhile, we also demonstrated that LPMO participated in the degradation of lignin macromolecular and lignin-carbohydrate linkages in fungus Pleurotus ostreatus through enhancing the extracellular hydroquinone-quinone redox cycling. 19 Given the broad range of electron donors for LPMO and the extensive redox network involved, it will be highly interesting to further study the role of LPMO involved in degradation of other components of biomass, like lignin. It is also essential to understand how LPMOs can synergize with other enzymes in the redox network for biomass deconstruction. The fundamental understanding will not only better define the roles of LPMOs in biomass deconstruction but also reveal how the redox network involved in degrading different cell wall components synergistically. White-rot basidiomycete fungi have evolved with unique capacity to degrade all cell wall components, including the highly recalcitrant lignin, with synergistic enzyme and none-enzyme radical generation systems. 20 , 21 White-rot fungi provide ideal model systems to study the lignocellulosic degradation mechanisms. More interestingly, our recent study showed that LPMO gene family expansion correlated with lignin-depolymerizing enzymes in the 43 analyzed biomass-degrading microbes, suggesting that lignin degradation may be driving the evolution of LPMO family. 22 In order to unveil the role of LPMO in biomass-degrading redox network, we carried out comparative systems biology analysis of two white-rot fungus species, Irpex lacteus CD2 and Dichomitus squalens DSM 9615, with different lignocellulosic degradation patterns and enzymatic systems to study the composition and dynamics of extracellular redox networks. 23 , 24 With the systems-biological understanding, we further carried out in vitro study to approve that the purified LPMOs could drive Fenton reaction and synergize with manganese peroxidase for lignin oxidation. These studies provided a new insight to the role of LPMOs in biomass degradation and to understand the mechanisms of synergistic redox network for lignin and polysaccharide degradation.",
"discussion": "Discussion The discovery of LPMO represents a major breakthrough for understanding lignocellulosic biomass degradation mechanisms in nature. 1 , 3 , 35 , 36 LPMO has been found to be able to oxidize diverse carbohydrate substrates in biomass, and the roles of LPMO in redox network for biomass degradation are continuing to expand. Our previous study also revealed that the specific Il LPMO99 used in this study also showed traditional polysaccharides oxidation capability. 19 Recently, more and more researches have illustrated the linkages between lignin degradation and cellulose degradation through LPMO. 3 , 15 , 16 , 37 , 38 However, all these studies only demonstrated that the lignin degradation-derived aromatic compounds could serve as electron transfer for LPMO to enhance the polysaccharide oxidation. In this study, we have further demonstrated that LPMO is also involved in lignin degradation in white-rot fungi. The discovery was established by gene expression analysis and biochemical assay. We first found that LPMOs were expressed throughout the biomass degradation process and co-regulated with the redox enzymes related to lignin degradation in the white-rot fungi I. lacteus CD2 and D. squalens DSM 9615. Some of the LPMOs were significantly upregulated in I. lacteus CD2 at the point of time when lignin was significantly degraded. The gene expression analysis led to the hypothesis that LPMO might be involved in lignin degradation. The hypothesis was well validated in biochemical analysis, where we have established that LPMO could generate H 2 O 2 and promote the Fe 3+ reduction to drive Fenton reaction, supply H 2 O 2 to LMPs, and even synergize with MnP for lignin oxidation. The LPMO activities and enzyme synergies were well verified with both model compounds and enzymatic hydrolysis lignin substrates. The biochemical analysis was further substantiated by finding that gene family expansion of LPMO is more driven by the lignin degradation capacity in biomass-degrading fungi as shown in our recent study. 22 More interestingly, the lignin oxidation capacity of LPMO is only slightly affected by insoluble cellulose, which indicated that LPMO may play a more important role in lignin degradation in absence of soluble carbohydrate. The study suggested that LPMO may play a unique role to bridge the degradation of different components of plant cell wall (cellulose, hemicellulose, and lignin), where LPMO may promote lignin degradation in absence of soluble carbohydrate during early stage of wood degradation. Therefore, the study strongly suggested dual roles of LPMOs for both lignin and polysaccharides degradation depending on the substrate availability, co-expressed enzymes, and redox environment ( Figure 4 ). In other words, LPMO might uniquely harness the extracellular redox network with temporal and spatial precisions to specifically target lignin or carbohydrate at different stages of biomass degradation to maximize the efficiency. In the context of extracellular redox network, LPMO could be activated by direct accepting electrons from CDH 14 and some small molecular reductants like diphenols 1 and ascorbic acid 3 to carry out oxidation reactions. When activating in absence of cellulose, LPMO could produce H 2 O 2 by recruiting O 2 and accepting electron from the donors. 39 , 40 On one side, the H 2 O 2 produced by LPMO could be used to generate hydroxyl radicals via Fenton reaction with Fe 2+ for lignin and cellulose oxidation. Fe 2+ could be either reduced from Fe 3+ by CDH or LPMO-hydroquinone. As we demonstrated in this study, LPMO-hydroquinone could significantly promote the production of H 2 O 2 and Fe 2+ , which could in turn generate hydroxyl radicals to oxidize lignin. On the other side, the H 2 O 2 produced by LPMO could be used to activate the fungal lignin-modifying peroxidases for lignin degradation as we demonstrated in our recently publication. 18 Unlike the other H 2 O 2 -generating enzymes in GMC and CRO families, LPMOs can accept electrons from much broader donors to produce H 2 O 2 , which could make the fungi adaptable to various extracellular environments during lignocellulose degradation. Although a recent study showed that H 2 O 2 could be co-substrate of LPMOs, it only happens in the presence of carbohydrate substrates such as cellulose. 11 Our results showed that the H 2 O 2 production by LPMO was significantly inhibited by soluble polysaccharides, consistent with previous studies, 39 while the insoluble MCC did not significantly inhibit the H 2 O 2 production by LPMO. Considering that the native cellulose microfibrils are embedded in lignin and exist in crystalline structure rather than soluble polysaccharide, 41 many LPMOs could react in an environment with low-soluble polysaccharide. In addition, many LPMOs have no carbohydrate-binding modules, resulting in enzymes with much lower carbohydrate affinity. Therefore, the free LPMOs (unbound to cellulose) might generate H 2 O 2 for lignin degradation in the fungal extracellular redox environment and synergize with LMPs for lignin oxidation. Recent studies also suggested that the peroxidases have higher affinity to H 2 O 2 than LPMOs, and thus the peroxidases could compete the H 2 O 2 produced by LPMOs to affect the LPMOs’ activity on cellulose and drive the reaction to H 2 O 2 generation, even when the LPMOs are binding to cellulose. 11 Overall, the study has revealed a much more versatile role of LPMOs ( Figure 4 ), where the enzyme might be involved in lignin oxidation via Fenton reaction and synergizing with lignin-modifying peroxidases besides the oxidation of polysaccharides. Figure 4 The proposed extracellular redox network for lignin oxidation The free LPMO (non-binding to cellulose), glucose-methanol-choline oxidoreductases (GMCs) and copper radical oxidases (CROs) produce H 2 O 2 for Fenton reaction and lignin-modifying peroxidases (LMPs). The LPMO receives electron from cellobiose dehydrogenases (CDH) and small molecular reductants (SMRs) to initiate its active. The CDH will reduce Fe 3+ to Fe 2+ for Fenton reaction when it donates its electron to LPMO. Hydroquinone was oxidized into semiquinone, which was further oxidized by Fe 3+ -EDTA and lead to the production of quinone. In the process of oxidation of hydroquinone to quinone by LPMO, semiquinone promoted ferric iron reduction, generating in Fe 2+ . The hydroxyl radicals produced via Fenton reaction and other reactive radicals produced by LMPs and laccase will generate phenol compounds, which could serve as electron donor for LPMO and mediator for laccase. To our knowledge, this is the first study to show the role of LPMOs in lignin degradation from different aspects. The discovery indicated that LPMO’s role is much more dynamic depending on the redox environment, substrate availability, and co-regulated enzymes. It could be that LPMOs in wood-degrading enzymes acted together with peroxidases and Fenton reaction agents to break down lignin at early stage of biomass degradation, when less-soluble carbohydrate is available. With more crystalline cellulose broke down and more soluble polysaccharides available for hydrolysis, LPMOs synergize with hydrolase to break down the polysaccharides. The scheme makes perfect evolutionary sense in that the enzyme could explore the redox network in a very versatile way to maximize the efficiency of degrading different cell wall components. The study thus brought new perspectives of biomass degradation that a broader redox network involving LPMOs needs to be dissected for both lignin and carbohydrate degradation. The dynamics, synergy, and regulation of such network need to be further defined so that novel enzyme mixtures may be developed for holistic degradation of cell wall components including both lignin and cellulose. Limitations of the study We analyzed the transcriptome data to provide the genetic background of LPMOs in the role of biomass degradation, and verified the new function of LPMO in vitro through biochemical analysis, and how it synergizes with MnP for lignin oxidation through driving Fenton reaction. More research is still needed to identify whether other lignin degradation enzymes besides MnP interact with LPMOs in the process of biomass degradation, and how they could effectively use redox networks to efficiently degrade different cell wall components remains to be further explored. The biological functions of LPMO in lignocellulose still need to be further illustrated through in vivo biological experiments. In addition, since our previous studies showed that LPMO only could not oxidize the lignin model compounds, lignin dimers, or lignin polymers, 18 we did not include the LPMO alone as control in this study for its synergistic lignin degradation analysis with MnP."
} | 3,374 |
29099490 | PMC5739024 | pmc | 4,564 | {
"abstract": "The rock-hosted subseafloor crustal aquifer harbors a reservoir of microbial life that may influence global marine biogeochemical cycles. Here we utilized metagenomic libraries of crustal fluid samples from North Pond, located on the flanks of the Mid-Atlantic Ridge, a site with cold, oxic subseafloor fluid circulation within the upper basement to query microbial diversity. Twenty-one samples were collected during a 2-year period to examine potential microbial metabolism and community dynamics. We observed minor changes in the geochemical signatures over the 2 years, yet the microbial community present in the crustal fluids underwent large shifts in the dominant taxonomic groups. An analysis of 195 metagenome-assembled genomes (MAGs) were generated from the data set and revealed a connection between litho- and autotrophic processes, linking carbon fixation to the oxidation of sulfide, sulfur, thiosulfate, hydrogen, and ferrous iron in members of the Proteobacteria , specifically the Alpha -, Gamma - and Zetaproteobacteria , the Epsilonbacteraeota and the Planctomycetes . Despite oxic conditions, analysis of the MAGs indicated that members of the microbial community were poised to exploit hypoxic or anoxic conditions through the use of microaerobic cytochromes, such as cbb 3 - and bd -type cytochromes, and alternative electron acceptors, like nitrate and sulfate. Temporal and spatial trends from the MAGs revealed a high degree of functional redundancy that did not correlate with the shifting microbial community membership, suggesting functional stability in mediating subseafloor biogeochemical cycles. Collectively, the repeated sampling at multiple sites, together with the successful binning of hundreds of genomes, provides an unprecedented data set for investigation of microbial communities in the cold, oxic crustal aquifer.",
"conclusion": "Concluding remarks The microbial community in the crustal fluids of North Pond is temporally and spatially dynamic. The putative genomes extracted from our time series reveal a microbial community capable of impacting subseafloor biogeochemical cycles for carbon, nitrogen, sulfur and iron. These potential functions are redundant as community membership varies in time and space, suggesting that the communities present in both boreholes are poised to utilize the redox potential of the oceanic crust by exploiting reduced sulfur compounds and ferrous iron to drive autotrophic growth. Further research will elucidate the extent to which these organisms drive global biogeochemical processes. Data availability This project has been deposited at DDBJ/ENA/GenBank under the BioProject accession no. PRJNA391950, drafts of metagenome-assembled genomes are available with accession no. NVQK00000000-NVXW00000000, and raw sequence reads are available with accession no. SRX3143886-SRX3143902. Raw sequence reads from Meyer et al. (2016) constituting the metagenomic samples from 2012, are available under the BioProject accession no. PRJNA280201. Additional files have been provided and are available through figshare ( https://figshare.com/s/939160bb2d4156022558 ), such as: all primary and secondary contigs; MAGs and bins not analyzed as part of this research; and, all files described as Supplementary Data 1–17 .",
"introduction": "Introduction The largest actively flowing aquifer system on Earth is circulating through oceanic crust underlying the oceans and sediments ( Sclater et al. , 1980 ; Stein and Stein, 1994 ; Johnson and Pruis, 2003 ). The movement of water through the aquifer serves as a vital conduit for exchange of both microorganisms and nutrients between the ocean basins and the subseafloor and offers a route by which organisms can extract energy from the fluids and rocks beneath the seafloor ( Orcutt et al. , 2013 ; Meyer et al. , 2016 ). Our understanding of life within the marine crustal aquifer has largely been shaped by studies of anaerobic and thermophilic organisms in warm ridge flank environments ( Cowen et al. , 2003 ; Huber et al. , 2006 ; Jungbluth et al. , 2013 , 2016 ) and crustal-source basalts exposed at the seafloor ( Lysnes et al. , 2004 ; Mason et al. , 2009 ; Santelli et al. , 2009 ; Lee et al. , 2015 ). However, much of the microbial interaction with the crustal aquifer occurs within the seafloor at sites where cold, oxygenated deep ocean waters circulate through basaltic crust, entering and exiting through seafloor exposures ( Fisher and Wheat, 2010 ; Edwards et al. , 2012 ; Wheat et al. , 2017 ). Therefore, despite advancing knowledge about microbial life in the subseafloor, our understanding is limited relative to which microorganisms live in the rocky oceanic crust, what hydrogeologic processes control subsurface fluid circulation, how these organisms harness energy in this environment, and the overall contribution to marine biogeochemical cycles is limited. To more effectively study these prevalent ocean environments, several subseafloor observatories, termed circulation obviation retrofit kits (CORKs; Davis et al. , 1992 ; Wheat et al. , 2011 ), have been deployed in oceanic crust in part to allow for sampling and monitoring of the crustal aquifer ( Wheat et al. , 2011 ). Two CORK observatories are installed at the well-studied site North Pond, an isolated sediment basin (8 km × 15 km, ~4484 m water depth), just west of the Mid-Atlantic ridge on 7–8 million years old crust (22°45′ N, 46°05′ W; Edwards et al. , 2010 ). At North Pond, seawater circulates between the crust and the deep ocean through the exposed ridge flanks, while sediments within the basin act as an impermeable barrier that prevents seawater exchange. Previous studies have sought to constrain the microbial community and its activity within the basaltic aquifer at North Pond. Measurements of carbon fixation activity on basalts recovered by ocean drilling ( Orcutt et al. , 2015 ) were unable to detect quantifiable rates of activity at in situ temperatures (4°C), while additions of nitrate and ammonia to crustal rocks stimulated microbial growth ( Zhang et al. , 2016 ). Modeling of the subsurface at North Pond suggests that hydrogen and ferrous iron likely have an important role in maintaining microbial biomass, with ferrous iron estimated to support ~10% of the microbial biomass ( Bach, 2016 ). In support of this hypothesis, a Marinobacter isolate capable of iron oxidation was enriched from North Pond basalts ( Zhang et al. , 2016 ). PCR-based assessments of the microbial community associated with the basalts from North Pond have shown that Gammaproteobacteria are the dominant phylogenetic group ( Jørgensen and Zhao, 2016 ), while the presence of genes involved in the carbon fixation through the Calvin-Benson-Bassham cycle are more common than the reverse citric acid cycle ( Orcutt et al. , 2015 ). Additional work examining the crustal fluids from the aquifer at North Pond has shown that the geochemistry of the fluids is nearly identical to the deep Atlantic bottom water (DABW), indicating a short residence time for seawater within the crustal aquifer at North Pond ( Meyer et al. , 2016 ). However, basaltic formation fluids within the aquifer have concentrations of dissolved oxygen, silica and dissolved organic carbon that are different than those of the deep bottom water ( Meyer et al. , 2016 ), and assessment of the crustal fluid microbial community through 16S rRNA gene and transcript sequencing, stable-isotope incubations, and metagenomics revealed that the aquifer community was active with a distinct community structure from bottom water. The community also had the capacity to perform both autotrophy and heterotrophy ( Meyer et al. , 2016 ), with low rates of activity detected using nanocalorimetry ( Robador et al. , 2016 ). Together, these initial studies show a diverse and distinct microbial community living in the oligotrophic, oxic, basaltic crustal aquifer at North Pond with relatively low levels of metabolic activity. However, little is known about the metabolic potential and community dynamics in this understudied environment. Here, we present genomic reconstruction of North Pond crustal fluid samples collected over a span of two years, providing 21 samples for a detailed examination of potential microbial metabolism and community interactions within this subseafloor aquifer. Our high-resolution analysis of hundreds of genomes reveals a temporally and spatially dynamic microbial community and provides new insights into microbially-mediated biogeochemical cycling within the crustal aquifer.",
"discussion": "Discussion Despite being the largest actively flowing aquifer on Earth, our understanding of microbial communities and their role in biogeochemical cycling in subseafloor crustal fluids is largely unknown. The bulk of our understanding is from studies of fluids from warm environments, including the Juan de Fuca Ridge flank in the NE Pacific Ocean and hydrothermal vents around the globe ( Takai and Horikoshi, 1999 ; Huber et al. , 2002 ; Reveillaud et al. , 2016 ). These environments are characterized by high temperature (25–80 °C), low-oxygen fluids that are usually dominated by mesophilic and (hyper)thermophilic microorganisms with microaerobic and anaerobic metabolisms ( Cowen et al. , 2003 ; Huber et al. , 2006 ; Jungbluth et al. , 2013 ; 2016 ). This is in contrast to North Pond, which represents a common, but understudied type of ridge flank region, where circulating fluids are cold (4–15 °C) and oxygenated ( Edwards et al. , 2012 ; Meyer et al. , 2016 ). Previous work at North Pond showed that the fluids in the basaltic crust have similar chemistry to the oceanic bottom water, but that the microbial community has a distinct population structure with potential for both heterotrophic and autotrophic activity ( Meyer et al. , 2016 ). Using the increased temporal and spatial sampling offered by our metagenomic time series at North Pond, we verified that the microbial community composition of the crustal fluid samples is fundamentally different from the DABW, and extended this finding to microbial communities and their genomic functional potential using MAGs ( Figures 2 and 5 , Supplementary Data 10 ). Further, we also found that the microbial communities within the crustal fluids show shifts in the dominant phyla (and proteobacterial classes) over time within a single hole and between the two holes ( Figure 2 ). Gammaproteobacteria are dominant in 10 of the crustal fluid samples, but several other phylogenetic groups, Alpha - and Deltaproteobacteria , Epsilonbacteraeota and Bacteroidetes , are abundant in other samples. The initial samples were collected in 2012, approximately six months after the holes were drilled and the CORK systems were installed, therefore it is possible that the observed shifts are due to the holes returning to a natural state after the perturbation of drilling, during which surface water is pumped into the borehole to clear cuttings, inevitably pumping surface waters into the formation. Such shifts in subseafloor crustal fluid community structure have been documented in samples collected shortly after drilling and for several years afterwards on the flanks of the Juan de Fuca Ridge, a younger, warmer crustal system ( Jungbluth et al. , 2012 , 2016 ), highlighting the importance of time series for understanding such ecosystems and potential stresses. However, the magnitude of chemical shifts observed in discrete samples collected in 2012 and 2014 suggests only minor changes in geochemistry, including a decrease in dissolved oxygen concentrations and increase in dissolved silica concentrations at all four sampling horizons. Increases in dissolved silica may result from either diffusive exchange with sediment pore waters or water-rock reactions at low temperatures, whereas the decrease in oxygen concentrations indicates continued consumption of oxygen ( Ziebis et al. , 2012 ; Meyer et al. , 2016 ), such as that inferred from a similar cool ridge flank setting at Dorado Outcrop ( Wheat et al. , 2017 ). The high-resolution analysis, provided by the relative abundance of the reconstructed genomes, reveals that the microbial communities of U1382A, U1383C, and the DABW are composed of distinct MAGs ( Figure 3 ). Importantly, genomes from the DABW form a cohesive group of organisms that were not present (or had a limited presence) in the crustal fluids, and conversely none of the crustal-originating genomes were detected in the DABW. From these results, it is clear that the genomes we reconstructed represent residential subseafloor bacteria and archaea from North Pond crustal fluids, thus allowing for detailed examination of microbial metabolic functions and community dynamics and interactions within the North Pond crustal habitat. It is important to note, however, that the reconstructed genomes only represent a subset of the total microbial community from any one of the metagenomic samples, thus we can only interpret results from the observed community members ( Supplementary Data 4 ). It is likely, though, that due to the dynamics of assembly and binning that these genomes represent many of the most abundant organisms in the environment. Carbon fixation Previous results from North Pond samples in 2012 showed lower concentrations of dissolved organic carbon in the crustal fluids compared to seawater, as well as the potential for carbon fixation, with higher potential rates of autotrophy in the crust compared to seawater, especially at warmer temperatures (25°C) and deeper in the crust ( Meyer et al. , 2016 ). In addition, limited metagenomic analysis of three samples from 2012 showed the presence of some genes associated with carbon fixation ( Meyer et al. , 2016 ). Our assessment of genomes for the presence of genes representative of autotrophic carbon fixation resulted in the identification of two carbon fixation pathways: the CBB cycle and the reverse citric acid (rTCA) cycle ( Table 2 ). All instances of the rTCA cycle were identified within the Epsilonbacteraeota , and the CBB cycle was identified in several different groups, including the Alpha -, Gamma -, and Zetaproteobacteria , as well as the Planctomycetes . Each of the genomes with potential for carbon fixation was also analyzed for pathways that could provide a lithotrophic source of reducing potential necessary for carbon fixation ( Table 2 ). Results indicate that the most prevalent electron source identified amongst the putative carbon fixing genomes was sulfide, but several other electron sources were also identified, including thiosulfate, ferrous iron, sulfur, and hydrogen. These electron sources are likely coupled to the reduction of oxygen, as all but one of the genomes with predicted carbon fixation possess aerobic or microaerobic terminal oxidases. Possible additional terminal electron acceptors include nitrate and the intermediates of denitrification, as all but two of the carbon fixation genomes possess components of the denitrification or DNRA pathways ( Supplementary Figure 4 ). While a majority of the genomes with carbon fixation potential are linked to the oxidation of sulfur compounds, a group of genomes have the potential to utilize both H 2 and Fe 2+ to drive biomass production in support of the model proposed by Bach (2016) . These putative energy couples are congruent with the hypothesis of subseafloor microbial communities that can take advantage of the redox gradient created by the presence of reduced material in volcanic-derived basalt rocks and the oxygenated aquifer fluids ( Bach and Edwards, 2003 ; Bach, 2016 ). Hydrogen sulfide and iron species have not been detected in the crustal fluids at North Pond ( Meyer et al. , 2016 ), but the oxidation of the iron in sulfide complexes in crustal rocks (via biotic or abiotic process) would increase access to sulfide compounds for microorganisms ( Barco et al. , 2017 ) and for the abiotic oxidation of sulfide to thiosulfate ( Moses et al. , 1987 ). In this manner, it would be possible to sustain carbon fixation through multiple lithotrophic pathways, which are likely important due to the oligotrophic nature of the crustal fluids. This is similar to the prevailing theory in regards to terrestrial crustal systems ( Hallbeck and Pedersen, 2008 ), where lithoautotrophic growth in microorganisms via the CBB cycle has been found in deep terrestrial aquifers in the Fennoscandian shield ( Wu et al. , 2015 ). Genomic evidence for the prevalence of hypoxic conditions All measurements at North Pond show that the aquifer fluids at North Pond are oxygenated, with O 2 concentrations equal to or slightly less (185–244 μ m ) than that of the DABW (~250 μ m ; Table 1 ; Meyer et al. , 2016 ). Therefore, it was unexpected to find that many of the North Pond genomes had genes that suggest hypoxic or potentially anoxic conditions. More than half of the genomes (56%) had terminal c -type cytochromes for both aerobic ( aa 3 - and bo -type) and microaerobic ( cbb 3 - and bd -type) metabolisms, with an additional 13% of genomes only possessing the microaerobic cytochromes ( Supplementary Figure 4 ). There was substantial evidence that the organisms in this environment were capable of the reduction of nitrate via both dissimilatory nitrate reduction to ammonia (DNRA; 36%) and denitrification (36% Supplementary Figure 4 ). Further, NORP6 possessed the canonical sulfite reductase, necessary for the anaerobic conversion of sulfite to sulfide ( Supplementary Figure 4 ). The role that these genes, commonly associated with anaerobic metabolisms, play in the environment is unclear. It is possible that, similar to sub-oxic microenvironments encountered in the oxic surface ocean ( Ploug et al. , 1997 ), the subseafloor hosts microenvironments in which anaerobic metabolisms are ecologically viable. Like the surface ocean, one possible source of such microenvironments may be organic-rich particles, that can be readily colonized by heterotrophic microorganisms. In 2012, samples collected from North Pond crustal fluids showed a high heterogeneity of particles as detected on GFF filters ( Meyer et al. , 2016 ). Another possibility may be that the complex and fractured structure of the crustal aquifer provides both oxic and sub-oxic conditions. For example, hydrogeological studies of the Juan de Fuca Ridge flank indicated that fluid flow through the crust likely only occurs through small, discrete channels, restricted to a small volume (<1%) of the crust ( Fisher and Becker, 2000 ). Consequently fluid flow would be highly channelized through a small volume of the crustal rock. While measurements at North Pond CORKs show abundant oxygen, it is possible there are regions where fluid flow slows down and fluids could become stagnant, and anaerobic metabolisms may be more significant to the community as oxygen is consumed by heterotrophic activity or abiotic reactions. However, such stagnant fluids would likely not be indicative of the large crustal flow. Overall, the lack of an appreciable signal in the geochemical data may be the result of the extremely low biomass (~10 4 cells ml −1 ) and relatively recent entrainment of the formation fluids, especially in U1382A. Variable inter- and intra-borehole metabolic diversity The microbial community observed in U1382A can be effectively assigned to seven ecological units with distinct occurrence patterns ( Figure 4 ). These ecological units generally progress in sequential order, though several genomes within an ecological unit were detected in multiple time points, with up to 11 months between samples (TP2 vs TP7). This re-occurrence of members of the community suggest that there is mechanism for organisms to persist in the aquifer, either locally or transported from elsewhere within the subseafloor. Patterns may also be related to local geochemical conditions, where growth, and thus relative abundance, is tied to specific metabolic processes. Despite these changes in community structure over time, the genomes that are present in the ecological units are functionally redundant, with various metabolisms related to carbon fixation and nitrogen and sulfur cycling present in each of the measured time points ( Figure 4 ). While the ecological units as a whole are functionally redundant, the fraction of the observed community capable of a specific metabolic potential shifts over the course of the time series ( Figures 5a–c ). Shifts in genomes capable of nitrate reduction (DNRA and complete denitrification) and sulfur oxidation (thiosulfate oxidation and sulfur redox) processes were positively correlated, suggesting that these metabolic pairs are linked to the same environmental change. Further, shifts in the fraction of the community capable of sulfide oxidation is linked to a microbial community structure that overlaps TP1 and TP3-6, while thiosulfate oxidation is linked to overlaps in TP2 and TP7-8 ( Figures 5a–c ; Supplementary Figure 9 ). This suggests that changes in availability of sulfide and thiosulfate are responsible for the changes in microbial community structure, or conversely, that microbial community metabolic potential impacts the availability of sulfide and thiosulfate. In comparing U1382A and U1383C, several large, cohesive microbial groups were present in both boreholes ( Figure 3 ), with organisms more abundant in U1383C clustering together, to the exclusion of organisms more abundant in U1382A. However, it was common for a group of MAGs to be more abundant in one hole and also have a reduced or minimal abundance in the other hole ( Figure 3 ). While this result suggests there is some connectivity between the two subseafloor environments sampled by the CORKs, it is also clear that there are distinct, dominant populations within each hole, likewise there are distinct chemical signatures in both. However, the variation in community structure does not result in differences in metabolic potential, with functional redundancy in all queried processes, except for nitrogen fixation ( Figure 6 ). This functional redundancy is further reflected in the fraction of the observed microbial community capable of participating in each metabolic step, with no statistically significant difference between the boreholes, except for ammonia oxidation ( Figure 6 ). These results indicate that the observed differences in community structure are not related to carbon fixation or nitrogen and sulfur cycling, and are likely governed by environmental parameters that structure spatially distinct communities with a high degree of functional redundancy. A top–down control on community structure could be susceptibility to viral predation ( Nigro et al. , 2017 ), while a bottom-up control may involve limits in trace nutrients or vitamin availability. Continued analysis of these data and future sampling efforts will help to elucidate the extent of these controls on the microbial community."
} | 5,762 |
25322701 | PMC4200407 | pmc | 4,565 | {
"abstract": "A coupling process of anaerobic methanogenesis and electromethanogenesis was proposed to treat high organic load rate (OLR) wastewater. During the start-up stage, acetate removal efficiency of the electric-biological reactor (R1) reached the maximization about 19 percentage points higher than that of the control anaerobic reactor without electrodes (R2), and CH 4 production rate of R1 also increased about 24.9% at the same time, while additional electric input was 1/1.17 of the extra obtained energy from methane. Coulombic efficiency and current recorded showed that anodic oxidation contributed a dominant part in degrading acetate when the metabolism of methanogens was low during the start-up stage. Along with prolonging operating time, aceticlastic methanogenesis gradually replaced anodic oxidation to become the main pathway of degrading acetate. When the methanogens were inhibited under the acidic conditions, anodic oxidation began to become the main pathway of acetate decomposition again, which ensured the reactor to maintain a stable performance. FISH analysis confirmed that the electric field imposed could enrich the H 2 /H + -utilizing methanogens around the cathode to help for reducing the acidity. This study demonstrated that an anaerobic digester with a pair of electrodes inserted to form a coupling system could enhance methanogenesis and reduce adverse impacts.",
"discussion": "Results and Discussion Comparison of acetate removal and CH 4 production during the start-up stage In order to assess the effects of electrodes on anaerobic methanogenesis during the start-up stage, the electric-biological reactor (R1) and the control reactors (R2 and R3) were operated continuously for 58 days experiments and the results are showed in Fig. 1 . From Fig. 1A , the acetate removal efficiency of R1 increased gradually from 25.2 ± 2.1% to 63.3 ± 2.3%. As compared with R1, the acetate removal efficiency of R2 only increased from 26.1 ± 1.7% to 55.4 ± 3.2% and the acetate removal efficiency of R3 only increased from 24.7 ± 2.2% to 54.2 ± 2.3%. Especially, the acetate removal rate of R1 at day 30 had nearly reached the maximum removal efficiency, about more than 19 percentage points (amount to OLR: 2.4 Kg COD/L·d −1 ) higher than R2 and R3 at the same time. It implied that a faster startup and higher removal efficiency were achieved in R1 with addition of bioelectrochemical system. From Fig. 1B , the CH 4 production rate of R1 gradually increased from 31.9 ± 1.2 mL/h to 66.8 ± 2.7 mL/h. Comparatively, the CH 4 production rate of R2 increased only from 31.8 ± 1.6 mL/h to 53.5 ± 1.2 mL/h. At the same time, the CH 4 production rate of R3 increased only from 30.7 ± 1.9 mL/h to 52.7 ± 2.1 mL/h. Remarkably, during the 58 days experiments in the start-up stage, the average acetate removal efficiency and the average methane production rate of R3 was 39.1% ± 9.7% and 40.2 ± 7.4 mL/h respectively. Comparatively, the average acetate removal efficiency and the average methane production rate of R2 was 40.9 ± 9.9% and 41.3 ± 7.8 mL/h respectively. The statistical analysis of the three reactors is listed in Table S1 and Table S2 . These results showed that both acetate removal and methane production in R3 only had less than 5% differences as compared with those in R2, and the correlation coefficient of the two reactors was higher than 0.99 and the P value based on two tailed student t-test (n = 58) was also higher than 0.05. Therefore, it reasonably demonstrated that the electrodes themselves had no significant effects on the performances of the anaerobic system in the acetate removal and methane production, which could be ignored. The lower CH 4 production of R2 was similar to the results of Hao et al. 25 26 who reported that the high initial acetate concentration resulting in the accumulation of organic acids would (>50 mM) inhibited the activity of aceticlastic methanogenesis during the start-up stage. The results indicated that the electrodes might compensate the low rate of methanogenesis during the start-up stage. Remarkably, the only difference between the two reactors (R1 and R2) was the additional electrochemical system. Therefore, it was reasonably assumed that more decomposition of acetate of R1 could be ascribed to the role of anodic oxidation according to the reaction of CH 3 COO − + 2H 2 O = 2CO 2 + 7H + + 8e − , and the extra CH 4 production of R1 could be due to the cathodic reduction based on the reaction of CO 2 + 8H + + 8e − = CH 4 + 2H 2 O. To further clarify this assumption of the role of additional electrochemical system, anodic Coulombic efficiency and current of R1 had been measured and recorded in Fig. 1C . Theoretically, anodic Coulombic efficiency is a parameter to assess the fraction of electrons available from acetate that ends up as electrical current 21 . Therefore, anodic Coulombic efficiency could be reasonably used to calculate and distinguish the contribution of anodic oxidation and aceticlastic methanogenesis in the acetate removal. From Fig. 1C , the current increased from 0.379 ± 0.012 A to 0.434 ± 0.008 A during the initial 24 days, indicating that both anodic oxidation and cathodic reduction were enhanced which drove the more electron transfer produced from anode to cathode. In this stage, anodic Coulombic efficiency was more than 50% although it appeared a significant decreased trend, implying that anodic oxidation was the main pathway to degrade acetate in the initial start-up stage because aceticlastic methanogenesis was weak. From day 24 to day 58, the change of current was in relatively steady stage, slightly ranging from 0.409 ± 0.012 A to 0.434 ± 0.011A, but anodic Coulombic efficiency still reduced about 13 percentage points (decrease from 45.0% to 32.0%). The results indicated that the percentage of acetate decomposition by anodic oxidation in the total acetate decomposition decreased. In other words, aceticlastic methanogenesis was gradually acclimated to compete with anodic oxidation for acetate decomposition. Considering that the acetate removal efficiency and the CH 4 production rate still kept increasing, it suggested that aceticlastic methanogenesis became the main pathway to degrade acetate and produce CH 4 . At this time, aceticlastic methanogenesis replaced anodic oxidation to obtain more substrates which would decrease anodic oxidation and cathodic methanogenesis. These electrochemical parameters were well in agreement with the performance of the reactor shown in Fig. 1A and Fig. 1B . During the start-up stage, the average acetate removal efficiency of R1 and R2 was 52.7 ± 11.3% and 41.0 ± 9.9% respectively shown in Table S3 (see Supplementary material ). The difference of acetate removal efficiency between R1 and R2 was 11.7%. The average anodic Coulombic efficiency was 45.0 ± 12.9%. The acetate removal efficiency through anodic oxidation of R1 was 23.7% (23.7% = 45.0% × 52.7%). This calculated result was obviously higher than the difference of acetate removal rate between R1 and R2 and demonstrated that more decomposition of acetate of R1 as compared with R2 should be ascribed to the role of anodic oxidation according to the reaction of CH 3 COO − + 2H 2 O = 2CO 2 + 7H + + 8e − . The more electrons was produced through anodic oxidation, the more methane would be formed according to the reaction of CO 2 + 8H + + 8e − = CH 4 + 2H 2 O. The average acetate removal efficiency of direct methanogenesis (aceticlastic methanogenesis) of R1 was 29.0% ([100% − 45.0%] × 52.7% = 29%). It was assumed that the acetate removal through direct methanogenesis of R1 had a same conversion efficiency of 54.6% with R2 shown in Table S3 (see Supplementary material ). The methane production rate through direct methanogenesis of R1 was 30.1 mL/h (30.1 mL/h = 29.0% × 3000 mg/L [influent]/59 × 10 3 mg/mol × 22.4 × 10 3 mL/mol/6 h × 54.6%). The average methane production rate of R1 was 59.8 mL/h. Therefore, the methane production rate through cathodic reduction of CO 2 into CH 4 was 28.7 mL/h. The difference of methane production rate between R1 and R2 was about 11.7 mL/h (11.7 mL/h = 52.7 mL/h − 41.0 mL/h), and this result was obviously lower than that of cathodic reduction of CO 2 into CH 4 . It reasonably implied that the extra CH 4 production of R1 should be due to the role of cathodic reduction based on the reaction of CO 2 + 8H + + 8e − = CH 4 + 2H 2 O. The cathode potential and the potential difference between anode and cathode of R1 were recorded during the start-up stage (shown in Fig. S1 ). From this figure, the potential difference increased from 0.749 ± 0.002 V to 0.807 ± 0.003 V (vs Ag/AgCl electrode) in the initial 28 days, and then decreased from 0.807 ± 0.003 V to 0.759 ± 0.002 V. The average cathode potential of R1 was −1.081 ± 0.016 V (vs Ag/AgCl electrode) which was significantly lower than the theoretical potential of cathodic reduction of CO 2 into CH 4 (−0.44 V NHE) and also lower than the needed cathode potential (−0.7 V) for the significant methane production reported by Cheng et al. 14 Especially, during the overall start-up 58 days, the electric energy supply or consumption calculated was 543.2 J/h according to the following formula 37 : W E = (IE ap − I 2 R)Δt, where E ap is the average potential difference between anode and cathode (0.779 ± 0.023 V) according to Fig. S1 , I is the average current (0.418 ± 0.003 A), Δt is per unit time (3600 s) and R is the external resistor (1 Ω). This energy supply was less than the energy harvest from the extra increased CH 4 production. The extra increased CH 4 production of R1 as compared with that of R2 was averagely 17.5 mL/h. It meant that the extra obtained energy from CH 4 was 635.9 J/h (635.9 J/h = 17.5 × 10 −3 L/h/24.5 L/mol × 890.31 × 10 3 J/mol), about 1.17 times of the electric energy supply, where 24.5 L/mol was the molar volume of the gas at normal temperatures and pressures and 890.31 × 10 3 J/mol is the energy content of methane based on the heat of combustion. Effects of different anode potentials on the acetate removal and CH 4 production of R1 To further study the effects of different anode potentials on acetate removal and CH 4 production in R1, the anode potential was in turn increased from −400 to −350, −300, and −250 mV (vs Ag/AgCl). Table 1 shows the acetate removal efficiency, CH 4 production rate and anodic Coulombic efficiency of R1 at different anode potentials and Fig. 2 shows the change of current. With the increase of anode potential from −400 mV to −250 mV, the current increased from 0.142 ± 0.008 A to 0.473 ± 0.013 A, as well as anodic Coulombic efficiency increased from 18.6 ± 3.1% to 38.1 ± 1.7%. The increased anodic Coulombic efficiency meant that the contribution of anodic oxidation to acetate decomposition was raised. This result was consistent that the acetate removal efficiency increased from 52.9 ± 2.1% at −400 mV to 77.1 ± 3.3% at −250 mV (shown in Table 1 ). Actually, the amount of acetate removal increased from 1629.3 mg/L (1629.3 mg/L = 52.9% × 3080 mg/L [influent]) to 2374.7 mg/L (2374.7 mg/L = 77.1% × 3080 mg/L [influent]). The increased amount of acetate removal was 745.4 mg/L. At the same time, according to the increased anodic Coulombic efficiency shown in Table 1 , the increased amount of acetate removal by anodic oxidation was 601.7 mg/L (601.7 mg/L = 38.1% × 2374.7 mg/L − 18.6% × 1629.3 mg/L). It meant that about 81% of increased acetate removal was resulted from the increase of potential from −400 mV to −250 mV. The increase of anode potential might accelerate the electron transport rate, facilitating electrogens to consume more substrates 21 . Therefore, the enhanced acetate decomposition was observed with increase of anodic oxidation. Theoretically, when more substrates were degraded by electrogens, less substrate was available for aceticlastic methanogens. It would directly reduce the CH 4 production from aceticlastic methanogenesis. Reversely, when increasing anode potential from −400 mV to −250 mV, the CH 4 production significantly increased from 55.9 ± 3.3 mL/h to 77.7 ± 5.4 mL/h (shown in Table 1 ). It was reasonably ascribed to the role of cathodic reduction of CO 2 into CH 4 . To further clarify this deduction, it was assumed that the increase of acetate removal by direct methanogenesis was completely converted to methane. Therefore, the increased methane production rate by direct methanogenesis was about 9.1 mL/h (9.1 mL/h = [745.4 mg/L − 601.7 mg/L]/59 × 10 3 mg/mol × 22.4 L/mol/6 h). Actually, according to the Table 1 , with the increase of anode potential from −400 mV to −250 mV, the increased methane rate was 21.8 mL/h (21.8 mL/h = 77.7 mL/h − 55.9 mL/h). Therefore, the contribution of cathodic reduction of CO 2 to the increased methane production was higher than 60%. This result implied that the cathodic reduction of CO 2 contributed quite a large part of the increased methane production with the increase of anode potential. Together with the above results, this bioelectrochemical enhancement of methanogenesis would be potentially applied to improve the performance of anaerobic digester by gradually increasing anode potential or apply voltage when the treatment efficiency was low. Effects of acidic conditions on the performance of R1 and R2 In order to clarify the contribution of bioelectrochemical system to methanogenesis under the acidity accumulated conditions, the electric-biological reactor (R1) and the control reactor (R2) were operated with the influent pH gradually dropped from 7.0 to 5.0 during the 48 days experiments. Here, the anode potential of R1 was maintained at −250 mV (vs Ag/AgCl). Methanogens would be inhibited at acidic conditions (pH < 6.2) as reported by Kotsyurbenko et al. 27 who observed that lower pH extended the lag phase for methanogenesis. From Fig. 3A,B , with the influent pH decreased from 7.0 to 5.5, the acetate decomposition and CH 4 production of R2 appeared an obvious decreasing trend, at which the acetate removal efficiency dropped from 85.2 ± 2.7% to 34.3 ± 1.5% and CH 4 production rate dropped from 92.8 ± 2.6 mL/h to 35.3 ± 3.1 mL/h. Comparatively, R1 was less affected by the acidic pHs. The acetate removal efficiency of R1 was about 9 percentage points (amount to OLR: 1.2 Kg COD/L·d −1 ) higher than that of R2 at influent pH 6.2 and about 20 percentage points (amount to OLR: 2.6 Kg COD/L·d −1 ) higher than that of R2 at influent pH 5.5, while the average CH 4 production rate of R1 was about 15 mL/h higher than R2 at influent pH 6.2 and about 25 mL/h higher than R2 at influent pH 5.5. When the influent pH further decreased to 5.0, the acetate removal efficiency of R2 was only 5%–9% and nearly no CH 4 produced (shown in Fig. 3A,B ), while acetate removal efficiency of R1 was about 30.5 ± 2.1% and CH 4 production rate was 14.9 ± 2.8 mL/h. Commonly, in an anaerobic system of feeding with acetate, the acetate decomposition would partially neutralize organic acids. A good performance of CH 4 production in R1 was partially due to the more acetate decomposition. This consideration had been further verified by changes of the effluent pH shown in Fig. 3C . With the influent pH decreased from 7.0 to 5.0, the effluent pH of R1 was still maintained at a near-neutral pH (>6.0). Comparatively, the effluent pH of R2 was less than 5.5, causing the activity of methanogens still in a low level. Fig. 3D shows the change of anodic Coulombic efficiency and current of R1. With the influent pH decreased from 7.0 to 5.0, the average current of R1 decreased from 0.451 ± 0.012 A to 0.302 ± 0.008 A. Compared with the decrease of methane production, acidic pH had less effect on the anodic oxidation. This assumption had been documented in many literatures. Chae et al. 17 obtained an extra 10% hydrogen yield through suppressing methanogens using acidic feeding. Similarly, Liang et al. 28 found that the optimum pH for anodic oxidation in a BES was 4.5. These indicated that anodic oxidation (exoelectrogens) is more accommodative than methanogens in the low pH conditions. With the influent pH decreased from 7.0 to 5.0, the acetate removal efficiency of R1 decreased from 85.6 ± 1.8% to 30.5 ± 2.1%. The amount of acetate removal decreased from 2670.7 mg/L (2670.7 mg/L = 85.6% × 3120 mg/L [influent]) to 976.0 mg/L (976.0 mg/L = 30.5% × 3200 mg/L [influent]). Therefore, the decreased acetate removal of R1 was 1694.7 mg/L. During this time, the anodic Coulombic efficiency increased from 28.1% to 62.3% (shown in Fig. 3D ). The decreased acetate removal by anodic oxidation was 142.3 mg/L (142.3 mg/L = 28.1% × 2670.7 mg/L − 62.3% × 976.0 mg/L) accounted for 20.0% (20.0% = 142.3 mg/L/[2670.7 mg/L × 28.1%]) of total acetate removal by anodic oxidation at pH 7.0, while the decreased acetate removal by methanogens was 1552.4 mg/L (1552.4 mg/L = 1694.7 mg/L − 142.3 mg/L) accounted for 80.8% (80.8% = 1552.4 mg/L/{[100.0% − 28.1%] × 2670.7 mg/L}) of total acetate removal by methanogens at pH 7.0. Thus, it was demonstrated that the decrease of acetate removal was caused by both methanogens and anodic oxidation but acidic impacts had a less effect on anodic oxidation than methanogens. The portion of anodic oxidation to acetate decomposition increased and anodic oxidation hereby gradually replaced aceticlastic methanogensis to become the main pathway. This role of anodic oxidation helped the reactor maintain relatively stable performance when aceticlastic methanogenesis got stressed due to the low pHs. Commonly, the H + consumption through hydrogenotrophic methanogens played an important role to make the anaerobic reactor adaptable for acidic impact 27 29 30 . Hydrogenotrophic methanogenesis coupling with anodic oxidation might be a major reason for the better performance in R1. One hand, the cathodic hydrogentrophic methanogens accepted the electron produced from anode to drive acetate oxidation in the anode to happen. In other words, acetate could not be anodically decomposed until the electron produced was accepted by cathodic acceptors such as hydrogentrophic methanogens. On the other hand, anodic oxidation reduced the acidity to gradually create a favorable condition for aceticlastic methanogens. Considering the relationship between the electron and hydrogentrophic methanogens, it was assumed that the electrochemical function was likely to facilitate the cathodic hydrogentrophic methanogens. To make clear the assumption above, FISH was used to determine the relative abundance of hydrogenotrophic methanogens in the archaea community of R1 and R2 (shown in Fig. 4 ). From Fig. 4a,b , according to the analysis of using Image-Pro Plus 6.0, the relative abundance of hydrogenotrophic methanogens at the bottom of R1 was 56.25%, about 30 percentage points higher than that of R2 about 26.83%. This finding could explain the difference of methane production between R1 and R2 under the acidic pHs. Cheng et al 14 enriched a high abundance of hydrogen-utilizing methanogens from a mixed culture as the biocathode to produce methane. Villano et al. 31 reported that H 2 -utilizing methanogens in the cathode were critical for methane production. These indicated that anaerobic methanogenesis coupled with a pair of electrodes could improve the H 2 -utilizing methanogenesis. To clarify this deduction, the biofilm attached to the cathode of R1 was collected to determine the abundance of hydrogenotrophic methanogens used FISH analysis. From Fig. 4c , according to the analysis using Image-Pro Plus 6.0, the relative abundance of hydrogenotrophic methanogens of biofilm attached to the cathode was 85.01%. It was implied that the dominant methanogenic microbial community was hydrogentrophic methanogens around the cathode and this result was well in agreement with the report by Cheng et al. 14 who found that Methanobacterium accounted for 86.7% of the total cells in the cathode. The relative abundance of hydrogenotrophic methanogens around the cathode was obviously higher than that at the bottom of the reactor and also much higher than that in R2. These results demonstrated that the additional electrochemical system could enrich the hydrogen-utilizing methanogens around the cathode to serve as electron acceptors to drive acetate oxidation in the anode and to reduce the acidity. This enhancement of hydrogenotrophic methanogenesis might be an important reason for the stable performance of this coupling system at acidic pHs. Thus, anodic oxidation coupled with hydrogenotrophic methanogenesis created a favorable environment for aceticlastic methanogenesis. This further enhanced aceticlastic methanogenesis to accelerate the acetate decomposition and methane production."
} | 5,212 |
30377376 | PMC6235447 | pmc | 4,567 | {
"abstract": "Functional profiling from metagenomic or metatranscriptomic (“meta’omic”) sequencing provides insight into the molecular activities of microbial communities. These analyses are typically carried out using comprehensive search of sequencing reads, which is time-consuming, prone to spurious mapping, and often limited to community-level quantification. We developed a tiered meta’omic search strategy (HUMAnN2) which enables fast, accurate, and species-resolved functional profiling of host-associated and environmental communities. HUMAnN2 identifies a community’s known species, aligns reads to their pangenomes, performs translated search on unclassified reads, and finally quantifies gene families and pathways. Relative to pure translated search, HUMAnN2 is 3x faster and produces more accurate gene family profiles (89% vs. 67%). We apply HUMAnN2 to clinal variation in marine metabolism, ecological contribution patterns among human microbiome pathways, variation in species’ genomic vs. transcriptional contributions, and strain profiling. Finally, we introduce “contributional diversity” to explain patterns of ecological assembly across different microbial community types.",
"introduction": "INTRODUCTION Profiling microbial community function from metagenomic and metatranscriptomic (“meta’omic”) sequencing is a critically important challenge in culture-independent microbial ecology. It has the potential to characterize the extensive biochemical “dark matter” observed in many communities 1 , as well as to link specific molecular activities to environmental 2 and health-associated 3 phenotypes. In contrast with taxonomic profiling, functional profiling aims to quantify the gene and metabolic pathway content contributed by known and uncharacterized community members 4 . While taxonomic profiling can be performed on a maximally informative subset of meta’omic sequencing reads 5 , 6 , comprehensive functional profiling must consider all reads and the vast space of genes from which they may derive, thus adding considerable analytical complexity. Several previous methods exist for functional profiling of metagenomes 7 – 9 , a subset of which have been applied to metatranscriptomes 10 – 13 . These include HUMAnN 14 , which we developed during the Human Microbiome Project (HMP) 15 for host- and environmentally-associated meta’omic functional profiling. Like later methods, HUMAnN interprets translated search of meta’omic sequencing reads to reconstruct metabolic functions. While existing methods benefit from recent methodological advances in translated search 16 – 18 , they remain considerably slower than nucleotide-level analyses. In addition, while some functional profiling methods incorporate taxonomic concepts for database refinement 7 or targeted quantification 9 , most are limited to reporting community-level abundances (not per-organism contributions). Similarly, functional profiling lags behind efforts in strain-level analysis of microbial communities 19 – 21 , despite a growing appreciation for strain-variable functions within species. To integrate taxonomic information with microbial community functional profiles, and to limit the bottleneck imposed by translated search, we developed HUMAnN2 as a next-generation meta’omic functional profiling method. HUMAnN2 represents both a completely new methodology and implementation: incorporating a tiered approach with accelerated nucleotide-level, translated search, and pathway reconstruction components. With these, HUMAnN2 exceeds the accuracy and performance of current pure translated search strategies. Moreover, gene and pathway abundances quantified by HUMAnN2 are automatically stratified into contributions from known and uncharacterized species. This provides previously inaccessible detail in interpreting host-associated and environmental community meta’omes.",
"discussion": "DISCUSSION HUMAnN2 is a new approach for functional profiling of meta’omically sequenced microbial communities. The method introduces a novel tiered search algorithm that provides exceptionally accurate profiles for characterized members of microbial communities, with fallback to translated search for uncharacterized members. These tiers operate jointly in far less time than traditional pure translated search. Moreover, tiered search provides taxonomic stratification of microbial functions at the species-level, thus quantifying the community abundance of functions while simultaneously assigning them to specific contributors. The utility of tiered search will only improve as reference catalogs continue to rapidly expand. In addition, tiered search facilitates this expansion by identifying unclassified meta’omic sequencing reads for external assembly of novel genes. HUMAnN2’s functional stratifications enable discussion of “contributional diversity”: an analog of community-level diversity for individual microbial functions. Community-level function is often more conserved than community composition 15 , 39 – 41 , consistent with a functional repertoire “defining” a niche and satisfied by different microbial assemblages. Contributional diversity adds another means by which this feature of functional ecology may be understood 1 , in that, while some functions do appear to be distributed evenly across community members, others are more restricted. Similarly, modern “multi’omic” analyses of microbial communities distinguish between community functional potential (encoding by genomes) and functional activity (gene or protein expression) 39 , 42 , 43 . Contributional diversity reveals another way in which these measurements can differ: for example, broadly encoded functions that are expressed dominantly by one or a few species. Functional meta-analysis 44 of diverse, new and existing meta’omic profiles are among the future biological areas opened up by the HUMAnN2 methodology, with the potential to reveal i) novel microbial community biochemistry and signaling, ii) determination of these functions’ source species and contributional diversity patterns and, in multi’omic datasets, iii) species-resolved deviations between functional potential and activity. In the human microbiome, HUMAnN2 provides the opportunity to generate testable hypotheses regarding specific species- (or strain-) level functions associated with health-related differences in community-level function. To support these future discoveries, the method is implemented as open source, fully documented software, packaged with demonstration data and training materials, and supports an active user community, accessible via http://huttenhower.sph.harvard.edu/humann2 ."
} | 1,654 |
30617344 | PMC6353673 | pmc | 4,569 | {
"abstract": "Coral reefs are declining dramatically and losing species richness, but the impact of declining biodiversity on coral well-being remains inadequately understood. Here, we demonstrate that lower coral species richness alone can suppress growth and survivorship of multiple species of corals ( Porites cylindrica , Pocillopora damicornis , and Acropora millepora ) under field conditions on a degraded, macroalgal dominated reef. Our findings highlight the positive role of biodiversity in the function of coral reefs, and suggest that loss of coral species richness may trigger a negative feedback that causes further ecosystem decline."
} | 159 |
24138057 | null | s2 | 4,571 | {
"abstract": "Chitosan, a load-bearing biomacromolecule found in the exoskeletons of crustaceans and insects, is a promising biopolymer for the replacement of synthetic plastic compounds. Here, surface interactions mediated by chitosan in aqueous solutions, including the effects of pH and contact time, were investigated using a surface forces apparatus (SFA). Chitosan films showed an adhesion to mica for all tested pH ranges (3.0-8.5), achieving a maximum value at pH 3.0 after a contact time of 1 h (Wad ~ 6.4 mJ/m(2)). We also found weak or no cohesion between two opposing chitosan layers on mica in aqueous buffer until the critical contact time for maximum adhesion (chitosan-mica) was reached. Strong cohesion (Wco ~ 8.5 mJ/m(2)) between the films was measured with increasing contact times up to 1 h at pH 3.0, which is equivalent to ~60% of the strongest, previously reported, mussel underwater adhesion. Such time-dependent adhesion properties are most likely related to molecular or molecular group reorientations and interdigitations. At high pH (8.5), the solubility of chitosan changes drastically, causing the chitosan-chitosan (cohesion) interaction to be repulsive at all separation distances and contact times. The strong contact time and pH-dependent chitosan-chitosan cohesion and adhesion properties provide new insight into the development of chitosan-based load-bearing materials."
} | 348 |
20711520 | null | s2 | 4,572 | {
"abstract": "Researchers studying complex cognition have grown increasingly interested in mapping symbolic cognitive architectures onto subsymbolic brain models. Such a mapping seems essential for understanding cognition under all but the most extreme viewpoints (namely, that cognition consists exclusively of digitally implemented rules; or instead, involves no rules whatsoever). Making this mapping reduces to specifying an interface between symbolic and subsymbolic descriptions of brain activity. To that end, we propose parameterization techniques for building cognitive models as programmable, structured, recurrent neural networks. Feedback strength in these models determines whether their components implement classically subsymbolic neural network functions (e.g., pattern recognition), or instead, logical rules and digital memory. These techniques support the implementation of limited production systems. Though inherently sequential and symbolic, these neural production systems can exploit principles of parallel, analog processing from decision-making models in psychology and neuroscience to explain the effects of brain damage on problem solving behavior."
} | 290 |
27242685 | PMC4862312 | pmc | 4,573 | {
"abstract": "Hydrothermal sulfide chimneys located along the global system of oceanic spreading centers are habitats for microbial life during active venting. Hydrothermally extinct, or inactive, sulfide deposits also host microbial communities at globally distributed sites. The main goal of this study is to describe Fe transformation pathways, through precipitation and oxidation-reduction (redox) reactions, and examine transformation products for signatures of biological activity using Fe mineralogy and stable isotope approaches. The study includes active and inactive sulfides from the East Pacific Rise 9°50′N vent field. First, the mineralogy of Fe(III)-bearing precipitates is investigated using microprobe X-ray absorption spectroscopy (μXAS) and X-ray diffraction (μXRD). Second, laser-ablation (LA) and micro-drilling (MD) are used to obtain spatially-resolved Fe stable isotope analysis by multicollector-inductively coupled plasma-mass spectrometry (MC-ICP-MS). Eight Fe-bearing minerals representing three mineralogical classes are present in the samples: oxyhydroxides, secondary phyllosilicates, and sulfides. For Fe oxyhydroxides within chimney walls and layers of Si-rich material, enrichments in both heavy and light Fe isotopes relative to pyrite are observed, yielding a range of δ 57 Fe values up to 6‰. Overall, several pathways for Fe transformation are observed. Pathway 1 is characterized by precipitation of primary sulfide minerals from Fe(II) aq -rich fluids in zones of mixing between vent fluids and seawater. Pathway 2 is also consistent with zones of mixing but involves precipitation of sulfide minerals from Fe(II) aq generated by Fe(III) reduction. Pathway 3 is direct oxidation of Fe(II) aq from hydrothermal fluids to form Fe(III) precipitates. Finally, Pathway 4 involves oxidative alteration of pre-existing sulfide minerals to form Fe(III). The Fe mineralogy and isotope data do not support or refute a unique biological role in sulfide alteration. The findings reveal a dynamic range of Fe transformation pathways consistent with a continuum of micro-environments having variable redox conditions. These micro-environments likely support redox cycling of Fe and S and are consistent with culture-dependent and -independent assessments of microbial physiology and genetic diversity of hydrothermal sulfide deposits.",
"conclusion": "Conclusions One of the main objectives of the present study is to determine whether alteration materials associated with seafloor sulfide deposits possess mineralogical or stable isotope biosignatures. In essence, we want to know the reactants, mechanisms, and products for reactions involving Fe. In light of the microbial community data (Sylvan et al., 2012 ; Toner et al., 2013 ), textural complexity noted in hand specimens (Rouxel et al., 2008a ), and spatial scale of mineralogical variability in petrographic sections (this work), we chose a spatially resolved spectroscopic and isotopic approach to the study of the EPR sulfides. The spectroscopic approach provides the reaction products (and in some cases reactants) through measurement of the chemical form of Fe, as well as the identity of co-located elements: while the isotopic approach provides the source-Fe reactant(s) and the extent of the reaction. The application of these complementary tools at the 10–50 micron spatial scale certainly must have averaged over finer submicron geochemical information. However, the spatial scale of investigation did allow us to describe the diversity of mineral forms and isotopic signatures intrinsic to these sulfides, and demonstrates the synergy of spectroscopic and isotopic approaches. Culture-independent microbiology results show that microbial communities of the EPR change when chimney structures become hydrothermally extinct (Sylvan et al., 2012 ). Once extinct, the EPR sulfide deposits host bacterial communities similar in composition and structure to mid-ocean ridge sulfides from far distant sites, such as the Indian Ocean (Toner et al., 2013 ). The presence of bacterial groups known for Fe and S cycling in EPR sulfide samples (e.g., β-Proteobacteria Gallionella sp. and ε-Proteobacteria Sulfurimonas sp.) indicates that microorganisms should be able to mediate the alteration of sulfide mineral substrates. As a result of S and/or Fe oxidation processes, Fe(III) reaction products are expected, and Fe(III)-bearing phases are abundant in the EPR sulfide deposits examined in this study. Overall, these findings provide evidence to support the idea that inactive sulfide deposits are more than passive surfaces for microbial attachment. However, only the genetic potential, not the metabolic capabilities, of microbial community members are measured by the sequencing approaches used to-date for this sample set. While our analysis of Fe mineralogy and isotope values do not support or refute a unique biological role in Fe(III) oxyhydroxide precipitation during sulfide alteration, our findings do reveal complex reaction pathways of Fe—precipitation/dissolution and oxidation/reduction—in seafloor sulfide deposits. The variety of Fe transformation pathways we observe is consistent with the development of micro-environments within the sulfide deposits. By defining four pathways of Fe transformation, we can propose several micro-environments. We observe zones of mixing between vent fluids and seawater with both reducing (Pathways 1 and 2) and oxidizing conditions (Pathways 2 and 4). These micro-environments are consistent with the diverse genetic potential, and correspondingly wide range of potential metabolisms, observed in organisms cultured from sulfide deposits. The reducing-to-oxidizing range of mixing zones, in particular, could support micro-environments favorable to Fe and S oxidation and reduction within a single chimney deposit. In contrast to the dynamic multi-step pathways revealed by Fe isotopes, the Fe XAS observations for these samples are consistent with stabilization of poorly ordered Fe(III) oxyhydroxide phases when Si and As are present. At first glance, it is surprising that multi-stage Fe oxidation and precipitation processes would preserve poorly ordered, and presumably, metastable Fe(III) oxyhydroxide phases. However, laboratory experiments have demonstrated that Fe atom replacement within Fe(III) oxyhydroxides, such as goethite, can produce the same mineral structure and particle morphology in the presence of Fe(II) aq (Handler et al., 2009 ). While there are many time-dependent factors that our measurements cannot assess—temperature, flow-rate, time-resolved vent fluid chemistry—the dynamic history supported by the Fe isotopes and static history supported by the mineralogy leads us to conclude that inorganic ligands (As, Si), and possibly biological material (Toner et al., 2009b , 2012 ), stabilize poorly ordered Fe(III) oxyhydroxide phases in these deposits. In addition, ligands appear to be retained by the deposit throughout a complex set of Fe transformation events and dynamic re-working of the Fe-bearing alteration products. It is possible that this mineral-stabilization outcome is driven by a decrease in permeability of the sulfide deposits as a function of time. Low-flow conditions could lead to the retention of ligands in pore spaces and preservation metastable phases despite continued alteration of the Fe-bearing minerals. The physical and chemical environments created by seafloor sulfide deposits do not appear to provide unique biosignatures. However, they do appear to be promising recorders of micro-environmental conditions present during hydrothermal activity.",
"introduction": "Introduction Seafloor hydrothermal activity at oceanic spreading centers is one of the fundamental processes controlling the exchange of heat and chemical species between seawater and ocean rocks (Edmond et al., 1979 ; Stein and Stein, 1994 ; Elderfield and Schultz, 1996 ; Wheat et al., 2004 ). The altered rock and mineral deposits created by hydrothermal circulation are known to harbor microbial communities with ecological and functional characteristics corresponding to the chemistry of the host substrate (Santelli et al., 2008 ; Orcutt et al., 2011 ; Sylvan et al., 2012 ; Lever et al., 2013 ; Toner et al., 2013 ). In addition, it has been demonstrated that microorganisms interact with their rock/mineral environment by promoting mineral dissolution and precipitation (Holden and Adams, 2003 ; Houghton, 2007 ; Pagé et al., 2008 ; Templeton et al., 2009 ; Houghton and Seyfried, 2010 ). Investigations of the physiological and phylogenetic diversity of rock-hosted prokaryotes using both culture-based and molecular approaches show their important ecological roles in biogeochemical cycles of carbon (C), sulfur (S), nitrogen (N), and iron (Fe) (Reysenbach and Cady, 2001 ; Slobodkin et al., 2001 ; Edwards et al., 2003a ; Byrne et al., 2009 ; Yamamoto and Takai, 2011 ). The development of microbial habitats within hydrothermal chimney deposits is a combination of physical (temperature, porosity), chemical (dissolved and mineral), and biological factors (biofilms, mineral alteration). It is well-established that steep temperature and geochemical gradients form within the walls of actively venting chimneys (Tivey, 1995 ). Corresponding changes in microbial communities along physical and chemical gradients have been demonstrated at various levels of spatial resolution (Karl, 1995 ; Schrenk et al., 2003 ; Nakagawa et al., 2005 ; Pagé et al., 2008 ; Takai et al., 2008 ; Callac et al., 2015 ). The development of micro-environments within chimneys could explain the diverse genetic potential and wide range of metabolisms observed in organisms cultured from sulfide deposits. Analytical tools able to measure chemical, mineralogical, and isotopic information on the micron spatial scale are available, and the geoscience community has begun to apply them in concert (Marcus et al., 2015 ). Through the combination of these analytical approaches, one can now define the properties of micro-environments and gain the information needed to interpret micro-habitats or micro-niches within rock and mineral substrates. In mid-ocean ridge (MOR) hydrothermal systems, Fe is a fundamental element (Emerson, 2016 ). Iron deserves special attention when considering the biogeochemistry of mid-ocean ridges because it: (1) is abundant in most vent fluids; (2) has dynamic solubility properties and precipitates with S to form part of the physical structure of sulfide deposits; (3) has dynamic oxidation-reduction (redox) properties and the ability to set and record redox conditions within fluid flow paths; and (4) is a potential substrate for microbial energy (Fe reduction) and respiration (Fe oxidation). In mid-ocean ridge systems, such as the East Pacific Rise at 9–10°N or Juan de Fuca Ridge Main Endeavor Field, the breadth of possible biogeochemical roles for Fe is fully populated. Hydrothermal sulfides are observed as active and inactive chimneys, massive sulfide deposits, and particles in hydrothermal plumes settling from the water column (Feely et al., 1992 , 1994 ; Hannington et al., 1995 ; Tivey, 2007 ; Rouxel et al., 2008a ; Toner et al., 2009a ; Fouquet et al., 2010 ; Yucel et al., 2011 ; Breier et al., 2012 ). Iron (oxyhydr)oxides crusts form through the alteration of Fe-rich basalts and sulfides, and as microbial mats associated with diffuse venting (Alt, 1988 ; Mills and Elderfield, 1995 ; Wheat et al., 2000 ; Boyd and Scott, 2001 ; Emerson and Moyer, 2002 ; Edwards et al., 2011 ). The oxidizing conditions at the seafloor create a driving force for conversion of hydrothermally derived Fe(II)aq, as well as ferrous Fe in basalt and sulfide minerals, to Fe(III)-bearing minerals through chemical and biological means (Feely et al., 1994 ; Field and Sherrell, 2000 ; Edwards et al., 2003b ; Toner et al., 2009b ). In addition to the mineral forms of Fe, complexes with particulate and dissolved organic matter occur (Bennett et al., 2008 ; Toner et al., 2009a ; Breier et al., 2012 ; Hawkes et al., 2013 ) and microorganisms take up hydrothermal Fe in plumes by a variety of mechanisms (Li et al., 2014 ). Despite our understanding of the potential (bio)geochemical pathways for Fe transformation in hydrothermal systems, the actual Fe pools and most important mechanisms of transformation are difficult to measure and are an area of active research (Saito et al., 2013 ; Fitzsimmons et al., 2014 ; Resing et al., 2015 ). The strategic application of analytical tools over a range of spatial scales, from dissolved Fe species over a 1000 km to particulate Fe species within submicron aggregates, presents a way forward. In this contribution, we investigate the μm- to mm-scale chemistry of active and extinct hydrothermal sulfide deposits using micro-probe X-ray diffraction (μXRD), micro-probe Fe 1s X-ray absorption spectroscopy (μXAS), and spatially resolved Fe stable isotopes (laser ablation and micro-drilling). The main goal of this study is to describe the Fe transformation pathways in natural sulfide mineral deposits at the seafloor, and examine the products of alteration for indications or markers of biological activity. The research is an extension of incubation studies that subjected polished hydrothermal sulfide samples to seafloor conditions over a known period of time (Edwards et al., 2003b ; Toner et al., 2009b ) and Fe isotope studies of seafloor hydrothermal vents (Rouxel et al., 2004 , 2008a ). For our study of natural sulfide mineral alteration, we chose the East Pacific Rise (EPR) 9–10°N area because hydrothermally inactive sulfide deposits from this location are known to host microbial communities with the genetic potential to alter Fe- and S-bearing minerals through redox reactions (Sylvan et al., 2012 ; Toner et al., 2013 ). Our investigation reveals a suite of complex Fe transformation pathways, each of which can be partial or complete. We find that metastable Fe oxyhydroxide minerals persist in the samples despite strong evidence for dynamic Fe transformations, and that Fe isotopic fractionation creates very light isotopic signatures (δ 57 Fe values as low as –7‰) for Fe oxyhydroxides in some samples but isotopically heavy in others. The most important outcome of this work is a record of micro-environments within active and inactive sulfide deposits. These micro-environments are consistent with the diverse genetic potential, and correspondingly wide range of potential metabolisms, observed in organisms cultured from sulfide deposits. Micro-environments favorable to microbial Fe and S oxidation and reduction are supported by mineral and isotopic signatures. However, we did not identify any uniquely biological signatures and attribute this outcome to the complex interplay between biotic and abiotic reactions.",
"discussion": "Discussion Iron isotope systematics Significant variability of Fe isotope composition in (oxyhydr)oxide precipitates is expected when Fe(II) is partially oxidized in conditions with slow rates of oxidation. Microbiological experiments have shown that Fe isotope fractionations are produced during dissimilatory Fe(III) reduction (Beard et al., 2003b ; Icopini et al., 2004 ; Johnson and Beard, 2005 ) and anaerobic photosynthetic Fe(II) oxidation (Croal et al., 2004 ; Swanner et al., 2015 ). Iron isotopes can also be fractionated by abiotic Fe(II) oxidation, precipitation of ferric hydroxides (Bullen et al., 2001 ; Beard et al., 2010 ), and by sorption of aqueous Fe(II) onto Fe oxyhydroxides (Icopini et al., 2004 ; Wu et al., 2012 ). The largest equilibrium isotope fractionations (~4.5‰ in 57 Fe/ 54 Fe ratios) observed and theoretically calculated are for coexisting Fe(III) and Fe(II) aqueous species (Welch et al., 2003 ; Anbar et al., 2005 ). Hence, in view of the variety of fractionating processes during Fe redox changes, it is not surprising that marked variations in Fe-isotope composition have been observed in Fe-rich marine environments (Severmann et al., 2006 ; Rouxel et al., 2008b ). Large variations are also reported in banded iron formations (Johnson et al., 2003 ; Rouxel et al., 2005 ; Dauphas and Rouxel, 2006 ; Planavsky et al., 2012 ), hydrothermal fluids and precipitates (Sharma et al., 2001 ; Beard et al., 2003a ; Rouxel et al., 2003 , 2004 ), and ancient seafloor hydrothermal Fe-Si deposits (Planavsky et al., 2012 ; Moeller et al., 2014 ). Iron (oxyhydr)oxide particles within near-vent (buoyant) hydrothermal plumes at the Rainbow hydrothermal field (Mid Atlantic Ridge) have variable δ 57 Fe values (0.15 to 1.65‰) relative to the original vent fluid, consistent with fractionation during partial oxidation of Fe(II) aq to Fe(III) aq in seawater (Severmann et al., 2004 ). In contrast, Fe (oxyhydr)oxide-rich sediments precipitated from non-buoyant hydrothermal plumes have δ 57 Fe values that are indistinguishable from that of high-temperature hydrothermal fluids (Severmann et al., 2004 ). In a more recent study, low-temperature Fe (oxyhydr)oxide deposits from the Jan Mayen hydrothermal vent fields Norwegian-Greenland Sea yielded very light δ 57 Fe values down to −2.8‰ (Moeller et al., 2014 ). These values are indistinguishable from low-temperature hydrothermal fluids from which they precipitated, suggesting that hydrothermal vent fluid underwent significant partial Fe(II) oxidation below seafloor leading to isotopically light Fe values for Fe(II), as previously proposed (Rouxel et al., 2003 ). In our study, δ 57 Fe values for Fe oxyhydroxides of the chimney wall are both heavier and lighter than the coeval pyrite or hydrothermal fluids. We propose that a reservoir effect is created during partial Fe(II) oxidation and can explain the diversity of values: (1) heavier values are consistent with small extent of Fe(II) oxidation; (2) lighter values are consistent with a greater extent of Fe(II) oxidation; and (3) lightest values are consistent with loss of Fe oxyhydroxides along a flow path (open system Fe oxidation) (Dauphas and Rouxel, 2006 ; Moeller et al., 2014 ). Extremely light Fe isotope values can be attained through Rayleigh-type fractionation and could explain the δ 57 Fe values, as low as −7‰, we observe for Fe oxyhydroxides in sample EPR-4059-M4 (Figure 10C ). Such partial Fe(II) oxidation probably required micro-aerobic conditions in which the rate of Fe(II) oxidation is slow enough to generate high variability of δ 57 Fe values. Considering that such conditions are also favorable for the growth of Fe-oxidizing microorganisms (Druschel et al., 2008 ), the involvement of microorganisms in the Fe transformation pathways is likely. However, biotic Fe oxidation is not demonstrated in the present case because abiotic Fe oxidation can also produce these Fe isotope fractionation factors. Iron transformation pathways in sulfide mineral deposits Summary of transformation pathways The main goal of this study is to describe the Fe transformation pathways in natural sulfide mineral deposits at the seafloor, and examine the transformation products for indications or markers of biological activity. The hydrothermally inactive sulfide deposits of the EPR 9–10°N are known to host microbial communities with the genetic potential to alter Fe- and S-bearing minerals through redox reactions (Sylvan et al., 2012 ; Toner et al., 2013 ). Our investigation of the EPR sulfide deposits indicates that the alteration products are generated by at least four different pathways. Each pathway can be complete and quantitative , leading to an isotopic signature similar to the source Fe, or partial , leading to isotopic fractionation:\n (Pathway 1) direct precipitation of Fe(II) aq , to form primary sulfide minerals, from hydrothermal fluids in zones of mixing between vent fluids and seawater; (Pathway 2) precipitation of Fe(II) aq , to form sulfide minerals, from Fe(III) reduction in zones of mixing between vent fluids and seawater; (Pathway 3) direct oxidation of Fe(II) aq from hydrothermal fluids, to form Fe(III) precipitates, in zones of mixing between vent fluids and seawater; and (Pathway 4) oxidative alteration of pre-existing sulfide minerals to form Fe(III) precipitates. Pathways created by vent fluid-seawater mixing In several sulfide deposits at the EPR, we observe layered zones of Fe(III)-, Si-, and sulfide-rich precipitates that are consistent with cooling vent fluids and low redox potential (e.g., Figures 1A,D,C , 5 ). The precipitation of pyrite in the exterior chimney wall is probably a late-stage phenomenon reflecting a different pathway of mineral formation than the coarse-grained to euhedral pyrite composing the chimney wall. In many cases, we observe colloform and spherulitic pyrite textures on the chimney exterior in association with amorphous silica and often lining former worm tubes. Fine-grained to colloform pyrite with minor sphalerite may occur along fossil worm tubes (formed probably by Alvinella ) (Little et al., 2007 ; Rouxel et al., 2008a ). As discussed by Xu and Scott, 2005 , spherulitic and colloform textures of pyrite reflect rapid crystallization (i.e., disequilibrium) with cooling caused by mixing between hot vent fluid and cold ambient seawater. Overall, for our samples, the texture and morphology of these pyrite-Si-oxyhydroxide zones is highly variable; we speculate that this is caused primarily by the wide range of fluid flow and redox regimes created in actively venting chimneys. For many of our samples, pyrite lined fossil worm tubes exhibit δ 34 S and δ 57 Fe values (4.3 to 3.0 and −1.8 to −1.2‰, respectively) typical of bulk pyrite from the chimney wall (Rouxel et al., 2008b ), which is consistent with direct precipitation from hydrothermal fluids (Pathway 1). In addition to Pathway 1, the presence of pyrite within layered zones of Fe(III)-, Si-, and sulfide-rich precipitates could be caused by reductive processes in the mixing zone (Pathway 2). For example, pyrite forming at low–temperature and in an open system in contact with sulfate-rich seawater creates the potential for microbial sulfate and Fe(III) reduction. This possibility has been proposed from studies of microbial diversity in hydrothermal chimneys (Callac et al., 2015 ), as well as metabolic energetic calculations (Amend et al., 2011 ). Sulfur isotope studies show the contribution of microbial sulfate reduction and sulfide formation at mid-oceanic ridges (Peters et al., 2010 ); however, a robust S isotope biosignature in hydrothermal chimney environments has not been demonstrated. Our results provide strong evidence for secondary and late-stage pyrite formation along the outside wall of the chimneys due to vent fluid-seawater mixing. This interpretation, based on isotope data, Fe speciation, and mineral textures, is also consistent with the observed enrichment in silica. In general, silica precipitation in seafloor hydrothermal chimneys requires some degree of conductive cooling due to the solubility of amorphous Si in simple mixing between hydrothermal fluids and seawater (Hannington et al., 1995 ). Iron oxyhydroxides may also form in low-oxygen environments during hydrothermal-fluid mixing, although the expected slow rate of Fe oxidation in dilute hydrothermal fluids may limit the accumulation of Fe oxyhydroxides in the absence of microbial mediation. Localized Si and As enrichments within the Fe oxyhydroxides is common for our samples (Figures 1A,C,D , 5A–D ). In many cases, the layered Fe(III)-, Si-, and sulfide-rich zones exhibit δ 57 Fe values that are consistently lighter than hydrothermal vent fluids (Figure 10 ). This observation cannot be explained by either partial or quantitative Fe(II) oxidation because the resulting Fe(III) should be enriched in heavy Fe isotopes relative to Fe(II), regardless of the mechanism and extent of Fe(II) oxidation. Therefore, it is likely that these Fe oxyhydroxides formed through a combination of quantitative oxidation of vent fluid Fe(II) (Pathway 3) and alteration (i.e., oxidation) of pyrite characterized by much lighter δ 57 Fe values than vent fluids (Pathway 4). In sample EPR-4053-M3-reg1 (Figure 2 ), δ 57 Fe values are consistent with either partial oxidation of pyrite ( partial Pathway 4) or direct and quantitative oxidation of late-stage vent fluid Fe(II) (Pathway 3). The δ 57 Fe values of the Fe oxyhydroxides range from −0.32 to −1.59‰ which is heavier than associated pyrite by up to 2.3‰ but also within the range expected for hydrothermal vent fluids, especially low-temperature or late-stage vent fluids (Rouxel et al., 2008a ; Moeller et al., 2014 ). In contrast, sample EPR-4059-M4-reg2 (Figure 7B ) exhibits very light Fe isotope values for the Fe oxyhydroxides, which are up to 4.7‰ lighter than associated pyrite. These data cannot be explained by simple mixing between Fe(III) derived from local sulfide (pyrite) oxidation and Fe(II) from vent fluid. During partial Fe(II) oxidation, the remaining Fe(II) in solution is expected to become isotopically lighter due to the precipitation of isotopically heavy Fe-oxyhydroxide ( partial Pathway 3). This mechanism, often referred to as a reservoir effect, has been shown to lead to isotopically light Fe(II) in sediment porewater (δ 57 Fe down to −7.38‰) (Rouxel et al., 2008a ) and likely occurs within cavities of hydrothermal chimneys. The fact that isotopically light Fe oxyhydroxides occur in association with significant Si enrichment (e.g., EPR-4059-M4-reg4; Figure 4B ) and filling cavities of the chimney wall (EPR-4059-M4-reg2; Figure 7B ) is consistent with this assumption. In this case, Fe(II) released during pyrite alteration (Pathway 4) or late-stage hydrothermal fluid undergoing partial oxidation in the chimney wall ( partial Pathway 3), leads to the formation of isotopically light Fe(II) diffusing out of the chimney wall and ultimately precipitating as Fe oxyhydroxide. Pathways created by complete and direct oxidation In contrast to the Fe transformation pathways discussed in section Pathways Created by Vent Fluid-Seawater Mixing, the two massive sulfide deposits exhibit a simpler set of Fe reactions. For the EPR-4057-M2-reg1, the δ 57 Fe values are very close to Fe in unaltered pyrite (−1.93 ± 0.20 vs. −1.69 ± 0.11‰), and are consistent with direct and complete oxidation of pyrite-Fe to goethite-Fe (Pathway 4; Figure 4A ). While the incipient Fe oxyhydroxide phase is not known in this case, a short-term (2 months) chimney incubation study at the Juan de Fuca Ridge indicates the possibility of a biogenic-like Fe oxyhydroxide incipient phase (Toner et al., 2009b ). The fact that the Fe oxyhydroxide is primarily goethite, with no co-located Si or As, is suggestive of transformation of the initial precipitation products to the stable goethite phase over time in ambient deep ocean waters."
} | 6,738 |
32108224 | PMC7144361 | pmc | 4,574 | {
"abstract": "Abstract In order to develop successful strategies for coral reef preservation, it is critical that the biology of both host corals and symbiotic algae are investigated. In the Ryukyu Archipelago, which encompasses many islands spread over ∼500 km of the Pacific Ocean, four major populations of the coral Acropora digitifera have been studied using whole-genome shotgun (WGS) sequence analysis (Shinzato C, Mungpakdee S, Arakaki N, Satoh N. 2015. Genome-wide single-nucleotide polymorphism (SNP) analysis explains coral diversity and recovery in the Ryukyu Archipelago. Sci Rep. 5:18211.). In contrast, the diversity of the symbiotic dinoflagellates associated with these A. digitifera populations is unknown. It is therefore unclear if these two core components of the coral holobiont share a common evolutionary history. This issue can be addressed for the symbiotic algal populations by studying the organelle genomes of their mitochondria and plastids. Here, we analyzed WGS data from ∼150 adult A. digitifera , and by mapping reads to the available reference genome sequences, we extracted 2,250 sequences representing 15 organelle genes of Symbiodiniaceae. Molecular phylogenetic analyses of these mitochondrial and plastid gene sets revealed that A. digitifera from the southern Yaeyama islands harbor a different Symbiodiniaceae population than the islands of Okinawa and Kerama in the north, indicating that the distribution of symbiont populations partially matches that of the four host populations. Interestingly, we found that numerous SNPs correspond to known RNA-edited sites in 14 of the Symbiodiniaceae organelle genes, with mitochondrial genes showing a stronger correspondence than plastid genes. These results suggest a possible correlation between RNA editing and SNPs in the two organelle genomes of symbiotic dinoflagellates.",
"introduction": "Introduction Symbiotic dinoflagellates in the family Symbiodiniaceae (previously known as the genus Symbiodinium ) live together with many host organisms in coral reefs, including corals, sea anemones, bivalves, sponges, acoels, and forminiferans, in addition to existing as free-living cells ( Hirose et al. 2008 ; Yamashita and Koike 2013 ; Pochon et al. 2014 ; Lajeunesse et al. 2018 ; González-Pech et al. 2019 ). Over the last 20 years, molecular phylogenetic analyses of the nuclear ribosomal DNA (rDNA) have revealed the high genetic diversity of Symbiodiniaceae ( Rowan and Powers 1992 ; Coffroth and Santos 2005 ; Pochon et al. 2014 ). Population genetic analyses of Symbiodiniaceae have relied on comparisons of the internal transcribed spacer regions (ITS1 and ITS2) of nuclear rDNA, noncoding regions associated with the plastid psbA gene, and microsatellites ( Thornhill et al. 2014 ). The existence of dozens of Symbiodiniaceae species has been suggested by phylogenetic analysis of these noncoding sequence data. However, in spite of the existence of draft genomes from Symbiodiniaceae ( Shoguchi et al. 2013 ; Lin et al. 2015 ; Aranda et al. 2016 ; Liu et al. 2018 ; Shoguchi et al. 2018 ), whole-genome shotgun (WGS) sequence data have not yet been used for population genomic analysis of these algae. Genome sequence data from the coral Acropora digitifera ( Shinzato et al. 2011, 2015 ) have been used as a reference to study single-nucleotide polymorphisms (SNPs) in WGS reads from 155 coral individuals. These data revealed the population structure of A. digitifera in the Ryukyu Archipelago, Japan ( Shinzato et al. 2015 ). Four major clusters or populations were found in this study: Okinawa (OK), Kerama (KR), Yaeyama-North (YN), and Yaeyama-South (YS) ( fig. 1 ). There is approximately a distance of 500 km between the northern (OK and KR) and southern islands (Yaeyama) that were sampled ( fig. 1 ). In addition, the previous genome-wide population genetic analysis of A. digitifera showed that these four populations had limited connectivity, particularly between OK and KR ( Shinzato et al. 2015 ). Because KR is often considered a source for OK population recovery, this result provides an important cautionary note with regard to local conservation efforts. Namely, the transplantation of KR corals to OK coasts may not always be appropriate to facilitate the recovery of OK wild corals. The fertilized eggs of the coral A. digitifera do not have the symbiotic dinoflagellates. Acropora digitifera acquires symbiotic algae horizontally (acquired from the seawater environment) when they are in the planula larval stage ( Harii et al. 2009 ). In the population genetic study of Shinzato et al. (2015) , coral branches, including symbiotic dinoflagellates, were sampled for WGS analysis. It is therefore likely that Symbiodiniaceae genomes remain in these samples, which can also be analyzed to gain a perspective on symbiont distribution within the host populations. In particular, high copy number organelle genomes provide an ideal target for such an approach.\n Fig . 1. —Schematic diagram showing the sampling sites used for the population analysis of the coral Acropora digitifera in the Ryukyu Archipelago, Japan. Using whole-genome SNP analysis of A. digitifera , four clusters comprising Okinawa (OK), Kerama (KR), Yaeyama-North (YN), and Yaeyama-South (YS) were identified by Shinzato et al. (2015) . The inset (top left) indicates phylogenetic relationships among the clusters based on an inferred tree of A. digitifera populations. The numbers in the parentheses indicate the coral sample numbers at each location. The information for Okinawa prefecture in white boxes was obtained from the National Land Numerical Information System ( http://nlftp.mlit.go.jp/ksj/gmlold/index.html ; last accessed January 17, 2019). The following islands, excluding the sampling locations, are omitted. Hd, Hedo; Ik, Ikei; Irm, Uehara; IS, Oohama; Isy, Kabira; KrA, Geruma; KrC, Yakabi; KrD, Aka; KrE, Zamami; Mz, Manza; Od, Ohdo; Ss, Sesoko. Among Symbiodiniaceae, mitochondrial (mt) and plastid (pt) genomes from Breviolum minutum (previously known as Symbiodinium minutum ) are available as reference sequences ( Mungpakdee et al. 2014 ; Shoguchi et al. 2015 ). Transcriptome analyses showed that all organelle protein-coding genes undergo RNA editing. This process is a posttranscriptional modification that is mediated by specific enzymes ( Takenaka et al. 2013 ). It has been reported that pt RNA editing in land plants exhibits site-specific sensitivity for temperature and is inhibited by high temperature ( Karcher and Bock 2002 ). The temperature sensitivity of RNA editing may also be related to the diversity of the organelle response to a changing environment in the Symbiodiniaceae, but this issue is poorly understood. In this study, we analyzed Symbiodiniaceae organelle genome data from ∼150 individuals of A. digitifera . Our study posed two major questions: 1) do the phylogenies of organelle genes in the Symbiodiniaceae recapitulate host relationships that show the presence of local populations, and 2) do SNPs among the Symbiodiniaceae organelles have a potential relationship with RNA-editing events.",
"discussion": "Results and Discussion Diversity of Symbiodiniaceae Organelle Sequences in A. digitifera We used B. minutum as the reference to extract Symbiodiniaceae organelle sequences from the Illumina database (DRA003938) derived from 155 A. digitifera holobionts ( fig. 1 ). Five samples with low coverage of protein-coding sequences were removed for each organelle genome analysis, leaving 150 samples for downstream analysis (see Materials and Methods). As a result, we recovered 450 representative sequences from 3 genes in the mt genome and 1,800 representative sequences from 12 genes in the pt genome ( supplementary data sets S1 and S2, Supplementary Material online). To determine whether the evolutionary histories of organelle genes in the Symbiodiniaceae populations recapitulate host relationships, representative sequences from each compartment were used to build phylogenies. We hypothesized that the four clusters found in the population analysis of the host A. digitifera were also present in the associated Symbiodiniaceae populations ( fig. 1 ). The phylogenetic tree inferred from the three mt genes showed the clustering of the specimens from YN with high bootstrap support ( fig. 2 , bottom left). However, we failed to detect the four clusters identified in the analysis of the host coral. Thus, the mitochondrial sequence data supported only the presence of the YN group in the associated Symbiodiniaceae populations. In addition, sequences from seven YN individuals and three YS individuals exhibited a long branch ( fig. 2 ). This suggests that A. digitifera from the Yaeyama islands (YN and YS) harbor a different Symbiodiniaceae population from those of Okinawa Island (OK) and the Kerama islands (KR), in addition to a common population in the Ryukyu Archipelago.\n Fig . 2. —Maximum likelihood trees inferred from organelle genes of Symbiodiniaceae populations. Only nodes with ≥70% bootstrap support are indicated in the tree. Three concatenated mt genes were used to reconstruct the tree on the left side of the figure. Six YN samples are clustered with moderate high bootstrap support (72%). Some Symbiodiniaceae sequences from the Yaeyama area have long branches. A total of 12 concatenated pt genes were used to reconstruct the tree on the right side of the figure. Yaeyama Symbiodiniaceae sequences with long branches also correspond to the samples with long branches in the mt gene tree (Irm3, Irm4, Irm9, Irm27, Isy15, IS1, IS3, and IS7). Twelve protein-coding genes ( psbA , psbB , psbC , psbD , psbE , psbI , petB , petD , psaA , psaB , atpA , and atpB ) are encoded in Symbiodiniaceae pt DNA. These have plasmid-like structures referred to as minicircles ( Zhang et al. 1999 ) that each encodes a single-gene (1.8–3.3 kb) in B. minutum ( Mungpakdee et al. 2014 ). A ML tree of plastid genes showed that some of the sequences from YN and YS individuals cluster with high bootstrap support ( fig. 2 ). The four host clusters, KR, OK, YN, and YS (inset of fig. 1 ), were also absent from the pt data, although seven populations from YN are clustered with an OK population. Many of the Yaeyama samples (YN and YS) that had a long branch in mt gene trees also had a long branch in pt gene trees ( fig. 2 , middle right). The populations of YN and YS were clustered with 90% bootstrap support. These results suggest that some of the southern A. digitifera individuals maintain different Symbiodiniaceae populations from the remaining corals. The holobionts from the southern islands may be more diversified than those of the northern islands in the Ryukyu Archipelago. Our analysis of Symbiodiniaceae populations using organelle genomes suggests that the A. digitifera clusters in the southern region may contain a different, locally adapted population of symbiotic algae. To validate the presence of a different Symbiodiniaceae, we studied ITS2 sequences in the WGS data ( supplementary table 1 , Supplementary Material online). The detected ITS types supporting the majority belong to the genus Cladocopium (clade C type in previous classification). Interestingly, the Durusdinium (clade D in previous classification) were found only in the WGS data of the Yaeyama samples, supporting the presence of different populations in the southern region of the Ryukyu Archipelago. Therefore, future studies should focus on both coral and Symbiodiniaceae populations to understand the establishment of coral reefs in different areas. Diversity of Organelle Genes and Possible RNA Editing Sites RNA editing has been analyzed in detail for transcripts from dinoflagellate mt and pt genes ( Lin et al. 2002 ; Zauner et al. 2004 ; Zhang et al. 2008 ; Klinger et al. 2018 ). The conservation patterns of edited sites from mt mRNAs have been studied among core dinoflagellates, including the basally diverging Amphidinium and the Symbiodiniaceae ( Zhang et al. 2008 ). A recent report has discussed the dynamics and evolution of RNA editing in dinoflagellate plastid genomes using a large data set of dinoflagellates ( Klinger et al. 2018 ). To examine the relationship between SNPs and RNA editing sites, we used 2,250 sequences ( supplementary data sets S1 and S2, Supplementary Material online) from each of the three mt genes and 12 pt genes from 150 samples recovered from the Symbiodiniaceae organelle genomes. By aligning sequences from each of the three mt genes and 12 pt genes from the 150 samples, we determined the percentages of SNPs in the genes ( fig. 3 A ; table 1 ; supplementary fig. S1 , Supplementary Material online). Even though the SNP percentages in petB (6.4%) (42/657) and petD (5.2%) (25/477) were slightly lower than those in the other pt genes ( table 1 ; supplementary fig. S1 , Supplementary Material online), the total SNP percentage of pt genes (9.0%; 1,260/13,959) was higher than that in mt genes (5.6%; 185/3,288) ( fig. 3 B ).\n Fig . 3. —The relationship between SNPs and possible RNA editing (pRNAe) sites. ( A ) Alignment of a region of the cob sequence in mt DNA is shown with the RNA editing sites (highlighted in red) of the reference sequences. The correspondence between SNP and pRNAe is marked with yellow arrowheads. ( B ) SNPs from 3,288 sites in mt genes (left) and from 13,959 sites in pt genes (right) were identified by comparing 150 coral holobiont samples. The numbers in parentheses show the percentage of SNPs and pRNAe. The SNP percentage in pt genes was higher than in mt genes. Comparison with pRNAe shows that many pRNAe sites in mt genes correspond to SNP sites. The numbers in square brackets indicate nonsynonymous (ns) and synonymous substitution (ss) SNPs, respectively. The numbers on each gene are shown in supplementary figure S1 , Supplementary Material online. ( C ) Hypothesis for the relationship between nonsynonymous and synonymous SNPs in organelle genomes of the Symbidiniaceae. As an example, a region of the psaA alignment is shown and indicates that stretches of ∼20 nucleotides in sites upstream of edited sites may provide a specific sequence context recognized by editing activity ( Takenaka et al. 2013 ). Table 1 Correspondence between Detected SNP Sites and Potential RNA Editing Sites Gene Analyzed Site No. of SNPs (%) No. of pRNAe a No. of Correspondences Mitochondria cox1 1,455 57 (3.9) 29 26 cox3 771 56 (7.3) 24 24 cob 1,062 72 (1.8) 19 18 Plastid psbA 1,029 103 (10.0) 3 2 psbB 1,500 137 (9.1) 30 6 psbC 1,359 143 (10.5) 26 7 psbD 1,074 90 (8.4) 8 4 psbE 234 17 (7.3) 9 2 psbI 108 16 (14.8) 3 0 petB 657 42 (6.4) 23 3 petD 477 25 (5.2) 33 7 psaA 2,022 153 (7.6) 100 17 psaB 2,094 194 (9.3) 79 12 atpA 1,434 143 (10.0) 43 9 atpB 1,971 197 (10.0) 49 12 a Mungpakdee et al. (2014). We did not have transcriptome data from the Symbiodiniaceae populations, therefore, we studied the data from B. mimutum ( Mungpakdee et al. 2014 ; Shoguchi et al. 2015 ) and defined the known edited sites as possible RNA editing sites (pRNAe) in our data. By comparing SNPs and pRNAe in the three mt genes, we found that 68 sites were shared between these two data sets ( fig. 3 B ). These account for 36.8% (68/185) of the SNP sites and 94.4% (68/72) of the pRNAe sites ( fig. 3 B ; supplementary fig. S1 , Supplementary Material online). The SNPs of the shared sites potentially contain the nucleotides prior to and after RNA editing ( fig. 3 A ), suggesting that gain (or loss) of RNA editing sites may cause the polymorphism. Within the pt data, the percentages of SNP sites and pRNAe sites were 9.0% (1260/13,959) and 2.9% (406/13,959), respectively, with the shared sites totaling 81 ( fig. 3 B ). The reason for the lower percentage of shared sites in pt mRNA than those in mt mRNA may be explained by a recent report that suggested individual RNA editing sites in dinoflagellate plastids are species-specific and not highly conserved ( Klinger et al. 2018 ). Alternatively, SNPs among RNA editing sites may be low in the pt genome of the Symbiodiniaceae. Finally, to characterize the high SNP percentage of pt genes, we classified the sites into SNPs with nonsynonymous or synonymous substitutions (ns or ss) ( fig. 3 B and supplementary fig. S1 , Supplementary Material online). The sites of synonymous SNPs (809) exceeded those of nonsynonymous SNPs (451) in pt genes, although synonymous sites were less than nonsynonymous sites in mt genes ( fig. 3 B ; supplementary fig. S1 , Supplementary Material online). In land plant organelles, mRNA editing relies on cis-binding sites for trans-acting editing-site-specific proteins encoded in the nucleus ( Lynch et al. 2006 ). We hypothesize that some of the nonsynonymous SNPs correspond to RNA editing sites, and that the potential cis-binding sites may relate to the presence of many synonymous SNPs in pt genes ( fig. 3 C ). The simultaneous sequencing of genomes and transcriptomes from single-Symbiodiniaceae cells is needed to better understand pt SNP and the pRNAe data. In summary, we analyzed the genetic diversity of two organelle genomes from Symbiodiniaceae hosted by four A. digitifera populations. Our results show that corals in the southern sites (YN and YS) contain a different Symbiodiniaceae population from those in the north (OK and KR). Some of the same algal symbionts are, however, shared by these areas ( fig. 2 ). This suggests the presence of complex relationships among the southern holobiont populations. Many of the SNP sites in the mt DNA from the symbiotic dinoflagellates correspond to known RNA editing sites ( fig. 3 ). The sharing of these sites is apparently at a lower percentage in pt genes (6.4%) than in mt genes (36.8%) when using hypothetical RNA editing data. Future studies of the relationship between local climate change and the diversity of organelle genome sequences (including RNA editing) may provide critical insights into environmental adaptability among Symbiodiniaceae populations ( Baker 2003 ; Hidaka 2016 )."
} | 4,536 |
34308069 | PMC8296599 | pmc | 4,575 | {
"abstract": "Thermal sprayed aluminum\ncoatings are widely scalable to corrosion\nprotection of the offshore\nsteel structure. However, the corrosion rate of the Al coating increases\nconsiderably due to the severe marine environment. It has remained\na challenge to improve the corrosion resistance and protective ability\nof Al coatings. The superhydrophobic surface provides a potential\nway to improve the corrosion resistance of metal materials. Hence,\nthe development of superhydrophobic Al coatings with superior corrosion\nresistance is of great interest. In this work, the feasibility of\nthe preparation of superhydrophobic Al coatings on a steel substrate\nwas explored. First, Al coatings were prepared onto the steel substrate\nby the arc-spraying process, followed by ultrasonic etching with 0.1\nM NaOH solution, and afterward passivated using 1% fluorosilanes.\nThe effects of the etching time on morphology, contact angle, and\ncorrosion resistance of the Al coatings were evaluated. The schematic\nmodel of the fluorosilane passivation process on the Al coating surface\nwas provided. The micro/nanoscale surface structure of the low-surface-energy\nfluorosilanes promotes the wetting angle of 153.4° and a rolling\nangle to 6.6°, denoting the superhydrophobic properties. The\nsuperhydrophobic Al coating surface displays excellent self-cleaning\nperformance due to its weak adhesion to water droplets. The corrosion\ncurrent density of the superhydrophobic Al coating (1.36 × 10 –8 A cm –2 ) is 2 orders of magnitude\nlower than that of the as-sprayed Al coating (1.18 × 10 –6 A cm –2 ). Similarly, the charge-transfer resistance\nis found to be 12 times larger for the superhydrophobic Al coating\nand the corresponding corrosion inhibition efficiency reaches 98.9%.\nThe superhydrophobic Al coating displays superior corrosion resistance\nand promising applications in a marine corrosion environment.",
"conclusion": "3 Conclusions In summary, a superhydrophobic surface\nwas prepared onto the arc-sprayed\nAl coating by chemical etching using NaOH solution followed by passivation\nwith FAS-17 ethanol solution. The typical micro/nanoscale structures\nare presented on the surface of the superhydrophobic Al coatings.\nThe wetting angle and RA of the superhydrophobic Al coating are 153.4\nand 6.6°, respectively. The superhydrophobic Al coating surface\ndisplays excellent self-cleaning performance. The electrochemical\nresults show that the corrosion current density of the superhydrophobic\nAl coating (1.36 × 10 –8 A cm –2 ) is 2 orders of magnitude lower than that of the as-sprayed Al coating\n(1.18 × 10 –6 A cm –2 ), and\nthe corresponding corrosion inhibition efficiency of the superhydrophobic\nsurface reaches 98.9%. At the low frequency, the moduli of impedance\n| Z | values of the as-sprayed Al coating and the superhydrophobic\nAl coating are 65.3 and 584.4 kΩ cm 2 , respectively.\nCompared with the as-sprayed Al coating, the charge-transfer resistance\nis found to be 12 times larger for the superhydrophobic Al coating.\nThe superhydrophobic Al coating demonstrates superior corrosion resistance\nproperties and it has potential application prospect in a marine corrosion\nenvironment.",
"introduction": "1 Introduction The\ncorrosion protection of the offshore steel structure is a significantly\nimportant issue. Protective coatings provide a positive method to\nextend the life span of offshore steel structures. 1 − 3 Among them,\nAl coatings are widely scalable to corrosion protection of steel due\nto low cost, nontoxicity, and cathodic electrochemical protection. 4 − 6 However, under violent temperature fluctuations and high humidity\nconditions in a marine environment, the corrosion rate of the Al coating\nincreases considerably and it can be only applied in a limited range.\nIn this regard, it is still a very challenging and vital topic to\nimprove the corrosion resistance and protective ability of Al coatings\nin the severe marine environment. Superhydrophobic surfaces\nhave come to be a hot issue and gained\nremarkably increasing interest from engineers and scientists during\nthe past decades. 7 − 11 These surfaces, with unique structure and function such as large\nwater contact angles (CA > 150°) and little sticking to water\ndrops, have been used to solve many thorny problems, such as corrosion\nprevention, 12 , 13 anti-icing, 14 antisplashing, 15 self-cleaning, 16 , 17 drag reduction, 18 and so on. Generally\nspeaking, the micro/nanostructure and low-surface-free energy are\nthe key players in the generation of superhydrophobic surfaces. 19 , 20 The chemical etching process is a facile method to fabricate high\nrough micro/nanoscale structures by the chemical reaction on the metal\nsurface. 21 , 22 Saleema et al. designed a one-step technique\nto fabricate a superhydrophobic surface with CA as high as ∼166°\nby immersing the AA6061 alloy in NaOH and fluoroalkyl-silane (FAS-17)\nmixture solution. 23 The results indicate\nthat there is no significant difference in corrosion performance between\nthe superhydrophobic surface and the hydrophilic surface. However,\nEscobar et al. produced a superhydrophobic surface on a pure Al plate\nvia ethanol hydrochloric acid etching and lauric acid modification. 24 The modulus of impedance of the superhydrophobic\nAl alloy surfaces prepared using NaOH solution etching and ethanolic\nstearic acid (SA) passivation was 70 times higher than that of the\nAA6061-Al alloy. 25 After being etched in\nCuCl 2 solution and modified by SA, the superhydrophobic\nAl surfaces with excellent corrosion resistance and reparability were\nobtained by Zhan et al. 26 Abbasi et al.\nreported a highly stable superhydrophobic 6020-Al alloy surface with\nexcellent anticorrosion by a combination of shot peening-etching treatment\nand silane modification processes. 27 Zhang\net al. declared a method combination of droplet etching and modification\nto prepare a superhydrophobic Al surface with a CA of 156°. 28 More recently, Guo et al. prepared a superhydrophobic\n7055-Al alloy surface with a CA of 167.3° using MgCl 2 solution etching and then modified by perfluorooctyltriethoxysilane. 13 The corrosion current density of the superhydrophobic\nsample dropped by surpass 2 orders than that of immensity of the bare\nAl sample, and meanwhile, the corrosion inhibition efficiency was\n99.67 in 3.5% NaCl solution. These results show that the corrosion\nresistance of Al alloys can be enhanced remarkably by superhydrophobic\nmanipulation. Despite notable progress for the superhydrophobic Al\nalloy substrate, the preparation and protective ability of the superhydrophobic\nAl coating on steel have remained elusive in marine corrosion protection. Thus, the purpose of this work was to investigate the possibility\nof the formation of superhydrophobic Al coatings on the steel substrate\nand evaluate their corrosion resistance and self-cleaning performance.\nThe superhydrophobic Al coatings were prepared by ultrasonic etching\nwith NaOH solution and modification by FAS-17 ethanol solution. The\nimpressions of the etching time and passivation on surface morphology\nand wettability of the coatings were analyzed. To better compare the\nproperties of the as-sprayed coating and superhydrophobic coating,\nself-cleaning and electrochemical measurements were carried out.",
"discussion": "2 Results and Discussion 2.1 Morphology of the Coatings Figure 1 shows the\nsurface\nmorphology and wettability of the Al coatings as a function of the\netching time in NaOH solution. Figure 1 a shows the top-view SEM image of the as-sprayed Al\ncoating. The single flattened particle presents a smooth surface.\nThere are some microscale irregular protrusions on the coating surface.\nThe average water CA of the as-sprayed Al coating is 139.5°,\nindicating the hydrophobic surface by the arc-spraying process. To\nobtain a superhydrophobic surface, a rough micro/nanostructure was\nbuilt by pretreatment of chemical etching the as-received Al coating. Figure 1 b–d reveals\nthe surface morphology of the coatings with an etching time of 1,\n5, and 7 min, respectively. Undoubtedly, an increasing number of corrosion\npits with micro/nanoscale hierarchical structures is present on the\nsurface of Al coating as a function of etching time. It promotes the\nprogressively rough etching surfaces of the coating. After being modified\nby FAS-17, the CA of three coatings increases to 141.4, 148.5, and\n153.4°, respectively, demonstrating a transition from hydrophobic\nto superhydrophobic surface. When the etching time reaches 7 min,\nthe coating shows superhydrophobicity. With further increase in etching\ntime to 10 min, as shown in Figure 1 e, the number density of the micro/nanoscale corrosion\npits decreases dramatically and connects on the whole coating surface,\nindicating the falling of the roughness. The CA of the coating is\n150.4°. Comparing with the coating etched for 7 min, the CA decreases\nslightly but it still shows superhydrophobicity. However, excessive\netching will bring about a decrease significant in the coating thickness,\nwhich is detrimental to the coating. Figure 1 SEM images of the coatings: (a) as-sprayed\nAl coating, (b) coating\nafter etching of 1, (c) 5, (d) 7, and (e) 10 min, and (a1)–(e1)\nand (a2)–(e2) corresponding magnification images. The insert\nin (a2)–(e2) referring to the water CA s after being modified. Figure 2 exhibits\nthe 3D morphologies and roughness of the coatings with the etching\ntime. The coatings have a number of micro/nano irregular protrusions,\nwhich makes the coatings hydrophobic. The surface roughness of the\nas-sprayed Al coating shows the lowest value of 0.78 ± 0.26 μm\namong the tested samples, as shown in Figure 2 a. In order to obtain a superhydrophobic\nsurface, chemical etching was used to promote the surface roughness\nof the coating. Figure 2 b,c exhibits the variation of 3D morphology and roughness of the\ncoating with an etching time of 3 and 7 min, respectively. The roughness\nof the Al coating is increasing as a function of etching time, which\nimproves the hydrophobicity of the coating. When the etching time\nis 7 min, the roughness of the coating reaches the maximum value of\n1.10 ± 0.28 μm, and the irregular protrusion on the coating\nsurface is the most prominent. With further increase in the etching\ntime to 10 min, as shown in Figure 2 d, the surface roughness of the coating decreases slightly\nto 1.06 ± 0.45 μm. Therefore, excessive etching will bring\nabout the decrease in coating roughness, which is unfavorable to the\ncoating. Figure 2 3D images of the coatings: (a) as-sprayed Al coating and (b) coating\nafter etching of 3, (c) 7, and (d) 10 min. Figure 3 depicts\nmore details of the coating with an etching time of 7 min. It can\nbe seen from Figure 3 a,b that numerous irregular protrudes and micron-scale corrosion\npits are distributed on the coating surface, which improves the roughness\nof the coating. Meanwhile, a large number of nanoscale corrosion pits\nare also detected on protrusions, as shown in Figure 3 c,d. Inspired by the lotus leaf, its superhydrophobicity\nmostly comes from the microscale papillae structure and a large number\nof nanoscale cylindrical protrudes. The hierarchical micro/nanoscale\nporous structure plays a positive role in surface wettability of the\ncoating. According to the Cassie Baxter equation 8 1 where θ r and θ represent\nthe apparent CA and intrinsic CA, respectively. f 1 and f 2 are the surface fraction\nof droplets in contact with the solid surface and air ( f 1 + f 2 = 1), respectively.\nWhen the etching time increases to 7 min, the maximum CA of 153.4°\nreveals a water droplet 14% contact with the coating surface while\nleaving the remaining 86% with air. Therefore, the droplet looks like\na sphere standing on the Al coating. Figure 3 SEM images of the coating after etching\nof 7 min: (a) 500×,\n(b) 1000×, (c) 2000×, and (d) 5000×. 2.2 XPS Analysis of the Superhydrophobic Surface X-ray photoelectron spectroscopy (XPS) was used to analyze the\ndifference between the as-sprayed Al coating and the superhydrophobic\nAl coatings. Figure 4 a illustrates the survey spectra of the coatings. The C 1s, O 1s,\nand Al 2p peaks are detected for both coatings. The intensity of these\npeaks of the as-sprayed Al coating is slightly stronger than that\nof the superhydrophobic Al coating. What is more, there is a strong\nF 1s peak accompanied with a small peak of Si 2p on the surface of\nthe superhydrophobic coating. That is to say, numbers of FAS-17 molecules\nwere adsorbed on the NaOH-etched Al coating throughout the modification\nprocess. Figure 4 XPS spectra of the coatings: (a) survey spectra, (b) F 1s, (c)\nSi 2p, (d) Al 2p, (e) O 1s, and (f) C 1s of the superhydrophobic coating. Figure 4 b–f\nshows the XPS high-resolution spectra of the superhydrophobic coating. Figure 4 b,c shows the spectra\nof F 1s and Si 2p that are composed of a strong peak at 688.8 and\n104.1 eV, respectively. The Al 2p spectrum shows only a peak at 75.7\neV because of the bonding of Al–O, as seen in Figure 4 d. Figure 4 e depicts the O 1s peak that is composed\nof three strong peaks having a binding energy of 532.4, 533.3, and\n534.1 eV, corresponding to Al–O–Si, O–C, and\nSi–O–Si bond, respectively. 29 The presence of the −Si–O–Al– component\nindicates that the removal of C 2 H 5 from the\nFAS-17 molecules may give rise to the formation of −Si–O–Al–\nbond at the substrate via the process of hydrolysis. 23 −Si–O–Al– bonds induce the\nstrong adhesion between the Al surface and FAS-17 molecules and it\nis the main contribution of the mechanical stability of the superhydrophobic. Figure 4 f shows the spectrum\nof the C 1s peak. The limited peaks at 288.6, 291.7, and 294.1 eV\nare attributing to the −CF 2 –CF 3 , −CF 2 , and −CF 3 groups of FAS-17,\nrespectively. The successful derivatization of the ultralow surface\nenergy terminal group of −CF 2 and −CF 3 on the rough Al coating surface is the key to improve the\nrepellent capability and superhydrophobicity. In addition, the two\nstrong peaks at 285.3 and 286.3 eV correspond to the boding of C–C\nand C–O, respectively. The high-resolution XPS peak analysis\non the C 1s, O 1s, F 1s, Si 2p, and Al 2p validates the presence of\nAl–O–Si–CF 2 –CF 3 and\n−Si–O–Si–. These results are regarded\nas evidence that the low-surface-energy fluoroalkylsilane film was\nsuccessfully assembled on the Al coating surface, which is consistent\nwith the results in reference reported by Sarkar. 30 From the XPS analysis, the mechanism for the management\nwith NaOH\nand FAS-17 molecules can be inferred. Figure 5 proposes a schematic presentation of the\nreaction mechanisms leading to the superhydrophobic coating. After\netching in NaOH solution followed by FAS-17 modification, aluminum\nhydroxide and the integration of CF 2 functional group are\nformed on the surface of the coating. During the hydrolysis process,\nthe C 2 H 5 component is erased from FAS-17 molecules.\nThe Si bonds with O in the surface and the C–F functional groups\nare oriented outward from the surface. The growing amount of FAS-17\nmolecules adhered to the surface with the low-surface-energy C–F\nfunctional groups provides favorable ways for the formation of superhydrophobic\nproperties. 31 , 32 Figure 5 Schematic diagram of the formation of\nthe superhydrophobic Al coating. 2.3 Superhydrophobic Property To further\nanalyze the water adhesion of the coatings with different processing,\nthe CA measurements and rolling angle (RA) measurements of the coatings\nwere conducted, just as shown in Figure 6 . For the as-sprayed Al coating, the CA is\n139.4° and RA is 82.9°. After being modified by FAS-17,\nthe CA and RA of the as-sprayed Al coating are 141.4 and 33.4, respectively.\nCompared with the as-sprayed Al coating, the CA of the modified coating\nincreases slightly, indicating a decrease in surface energy. As the\netching time increases from 1 to 10 min, the CA increases at first,\nreaches a local maximum, and then decreases. The RA displays an inverse\ntrend. When the etching time is 7 min, the CA of the coating reaches\nthe maximum value of 153.4°and RA is 5.8°. After etching,\nthe coating surface shows a micro/nanoscale structure. The surface\nfree energy of the coating is merely 0.82 mN/m, which is much lower\nthan the surface tension of water (72.1 mN/m 32 ). Therefore, the droplets cannot spread on the coating surface,\nbut it can contact with the irregular multivoid and the gap filled\nwith air, which prevents droplets to wet the surface. 33 Figure 6 shows the deionized water, blue ink, and red ink (10 μL) spread\nslightly on the hydrophobic Al coating. The dripped droplets are suspended\non the protuberances and contact indirectly with the coating. It further\ndemonstrates an extremely high repelling water property ( Figure 7 ). Figure 6 CA and RA of the coatings\nwith different processes. Figure 7 Schematic\ndiagram of droplets on the surface of a superhydrophobic\nAl coating: (a) schematic diagram of droplets on the coating surface\nand (b) physical diagram of droplets on the coating surface. Figure 8 exhibits\na dynamic droplet-bouncing experiment. During the entire bouncing\nprocess, the droplets drop freely and a high-speed camera was used\nto record the droplet (2 μL) impacting the surface of the coating. Figure 8 a,b shows the free\nfalling process of the droplet. In the initial state ( Figure 8 a), there is a certain distance\nbetween the coating and the droplet. The droplet manifests a natural\nappearance due to self-gravity. As the bouncing time prolongs to 4\nms, the droplet is right impacting the coating surface and comes into\nbeing the maximum deformation ( Figure 8 b). Thereafter, the retractable droplet begins to rebound\nfully upward ( Figure 8 c–f). Moreover, the droplet can bounce elastically several\ntimes before falling down the surface ( Figure 8 g) and there is no water vestigial on the\ncoating. The dynamic droplet-bouncing experiment shows that the irregular\nmicro/nanoscale porous structure prepared by the proper process has\nan extraordinarily weak water adhesion and drag resistance. So far,\na straightforward and low-cost process for preparing a superhydrophobic\nAl coating has been successfully developed. Figure 8 Sequence of snapshots\nof droplets (2 μL) impacting the superhydrophobic\nAl coating surface: (a) 0, (b) 4, (c) 8, (d) 12, (e) 16, (f) 20, and\n(g) 24 ms. 2.4 Self-Cleaning\nBehaviors of the Coatings Figure 9 demonstrates\nthe jet flow experiment on the superhydrophobic Al coating. It can\nbe seen that the coating surface presents good water repellency and\nhigh stability under high-speed water flow. Figure 9 Diagram of water droplet\njet experimental (figure reference scale:\nthe samples in a size of 50 mm × 25 mm × 8 mm). Figure 10 shows\nthe antifouling performance tests of the coating. First, the coating\nwas placed in the glass dish with a slight inclination of 10°\n( Figure 10 a,b), and\nthen, the blue or red ink was slowly dropped onto the coating surface.\nThe droplets rolled on the slightly inclined coating surface without\nleaving any trace on the coating ( Figure 10 (a1–3),(b1–3)). Therefore,\nthe coating has excellent antifouling properties. This is mainly due\nto the interaction of low-energy materials and surface micro/nanoscale\nstructure on the coating surface. When the droplet is on the surface,\nthe actual contacting area is very small. Therefore, the droplet is\napproximately spherical and rolls easily on the coating surface. Figure 10 Antifouling\nperformance of the superhydrophobic Al coating surface\n(figure reference scale: the samples in a size of 50 mm × 25\nmm × 8 mm): (a) blue ink and (b) red ink. Inspired by the repelling dust behaviors of the lotus leaf, the\nself-cleaning behaviors of the superhydrophobic Al coating were performed,\nas shown in Figure 11 . First, a film of dirt was well-distributed on the superhydrophobic\nAl coating surface, as seen in Figure 11 (a1). As the droplets roll over the slightly\nsloped coating surface, they can remove dust along the rolling track\n( Figure 11 (a2)). Subsequently,\nas the droplets were dripped down consecutively, a resembling self-cleaning\nbehavior was detected until all the covered dust was swept away, as\nshown in Figure 11 (a3),(a4) ( Video S1 ). When the coating\nsurface is shrouded by sand ( Figure 11 (b1)) and chalk ash ( Figure 11 (c1)), similar results were obtained, as\nseen in Figure 11 (b4),(c4)\n( Videos S2 and S3 ). Consequently, the superhydrophobic Al coating displays superior\nself-cleaning performance. Figure 11 Self-cleaning behaviors of the superhydrophobic\nAl coating surface\n(figure reference scale: Al coating samples in a size of 25 mm ×\n15 mm × 8 mm): (a) dirt, (b) sand, and (c) chalk dust. 2.5 Corrosion Resistance of\nthe Coatings For the sake of studying the effect of different\netching times on\nthe corrosion resistance of the coating surface, electrochemical tests\nin 3.5 wt % NaCl solution were conducted on the coating samples modified\nby FAS-17 with different etching times (1, 3, 5, 7, 9, and 11 min). Figure 12 a shows the polarization\ncurves of the coatings. The corrosion potential ( E corr ) and corrosion current density ( I corr ) of the coatings originated from the potentiodynamic\npolarization curve are listed in Table 1 . With prolonging the etching time, I corr of the coating decreases at first and then increases\nslightly. Compared with the as-sprayed Al coating, the I corr decreases by 2 orders of magnitude with an etching\ntime of 7 min. When the etching time further increases from 9 to 11\nmin, the I corr of the coating increases\nslightly. However, it is still lower than that of the as-sprayed Al\ncoating. The higher corrosion potential and lower corrosion current\ndensity indicate that the coating has superior corrosion resistance. Figure 12 (a)\nPotentiodynamic polarization curve, (b) Nyquist curve, (c)\nBode modulus diagram, and (d) Bode phase angle diagram of the coatings\nwith different etching times. Table 1 Electrochemical Parameters of the\nCoatings with Different Etching Times the etching\ntime of the coatings (min) E corr (V) I corr (×10 –6 A cm –2 ) 1 –0.71 4.29 3 –0.73 0.848 5 –0.71 0.836 7 –0.62 0.0136 9 –0.68 0.0388 11 –0.64 0.0524 Figure 12 b–d\nplots Nyquist and Bode curves of the coatings. With the increase in\nthe etching time, the impedance arc radius of the coatings first increases\nand follows a slight decrease. It is well-known that at a lower frequency,\na higher Z-modulus shows better corrosion resistance to metallic substrates. 34 , 35 Under a low frequency, with prolonging the etching time, the Z modulus\nof the coating shows an increasing trend and Bode phase angles in\nthe Bode diagram peak also show a similar trend. 36 Figure 13 describes\nthe electrochemical corrosion behaviors of the as-sprayed Al coating,\nthe Al coating etched with NaOH solution for 7 min, the as-sprayed\nAl coating modified by FAS-17, and the superhydrophobic Al coating.\nThe polarization diagrams for the coatings are depicted in Figure 13 a, while Table 2 summarizes the relevant\nparameters. For the as-sprayed Al coating, the E corr and I corr are −0.78\nV and 1.18 × 10 –6 A cm –2 ,\nrespectively. After being modified by FAS-17, E corr of the coating shifts positively from −0.78 to\n−0.69 V while I corr enlarges slightly\nto 1.27 × 10 –6 A cm –2 . After\nbeing etched in NaOH solution, the Al coating has the lowest E corr and the highest I corr , indicating the worst corrosion resistance among the tested\nsamples. In comparison, the superhydrophobic coating holds superior\nanticorrosion behavior among the tested samples. It has the largest E corr and smallest I corr in the coatings. The I corr of the superhydrophobic\nAl coating is 2 orders of magnitude lower than that of other coatings,\nrevealing excellent corrosion resistance. The corresponding corrosion\ninhibition efficiency (η p ) can be calculated using\nthe equation 26 2 where icorr 0 and i corr are corrosion current density of the as-sprayed\ncoating\nand the superhydrophobic coating, respectively. According to Formula 2 , the η p of the superhydrophobic coating is 98.9%, which further confirms\nits excellent corrosion resistance. The main reasons are that the\netched as-sprayed coating surface has an irregular multivoid structure\nat the micro/nanolevel. After FAS-17 modification, the surface freedom\nof the etched coating surface significantly decreases from 13.07 to\n0.82 mN/m. When the coating is soaked in NaCl solution, the air film\nis formed between the solution and the hydrophobic coating. The air\nfilms will reduce the real contact area between the NaCl solution\nand the coating, which postpones the penetration of Cl – into the Al coating. 35 Therefore, the\ncorrosion resistance of the coating is significantly improved. Figure 13 (a) Polarization\ncurves, (b) Nyquist plots, (c) Bode modulus diagrams,\nand (d) Bode phase angle diagrams of the coatings. Table 2 Corrosion Potential and Corrosion\nCurrent Density of the Coatings the tested\nsamples E corr (V) I corr (×10 –6 A cm –2 ) the as-sprayed Al\ncoating –0.78 1.18 the modified Al coating –0.69 1.27 the etched Al coating –0.82 5.27 the superhydrophobic Al\ncoating –0.62 0.0136 Figure 13 b illustrates\nthe Nyquist plots of the coatings with different processes. All of\nthe coatings show a trend to capacitive semiarcs up to frequencies\nof about 1 Hz. The diameter of the semicircle represents a higher\ncharge-transfer resistance ( R ct ) and a\nlower corrosion current density and correlated with the mechanism\nof superhydrophobic film resistance. The superhydrophobic Al coating\nhas the largest capacitive semiarc among the tested samples, as seen\nin Figure 13 b. The\nlarger impedance value of the superhydrophobic Al coating denotes\nthat the superhydrophobic surface is more resistant against corrosion. Figure 13 c plots\nthe Bode modulus diagram of the coatings. It can be observed that\nthe as-sprayed Al coating has a | Z | value of 19.37\nΩ cm 2 at a high frequency of 10 4 Hz, while\nthe superhydrophobic Al coating exhibited a | Z | value\nof 38.88 Ω cm 2 , which is almost two orders of the\nas-sprayed Al coating at the same frequency. Similarly, at a low frequency\nof 0.1 Hz, the | Z | value of the as-sprayed Al coating\nis 65.3kΩ cm 2 . In contrast, it was as high as 584.4\nkΩ cm 2 on the superhydrophobic Al coating. Generally\nspeaking, the high-frequency AC impedance indicates the response of\nthe coatings with the solution, while at a low frequency, it reflects R ct and the double-layer capacitance. 37 The higher | Z | value in the\nlow-frequency region shows a better barrier in the coating. 38 Compared with the as-sprayed Al coating, the\n| Z | value of the superhydrophobic Al coating is close\nto one order of magnitude larger at low frequencies. On the basis\nof the analysis of Bode modulus curves, the superhydrophobic Al coating\nhas a better anticorrosion in comparison with the as-sprayed Al coating.\nThe Bode phase angle diagrams of the coatings present a shoulder followed\nby two time constants, as shown in Figure 13 c. The first time constant provided by the\nshoulder is relevant to the properties of the coating. Another time\nconstant at the low-frequency region is related to the corrosion behavior\nof the substrate. From Figure 13 c, the superhydrophobic Al coating exhibits a time\nconstant at a lower frequency and a higher phase angle of 78.4°\nthan other coatings, suggesting a better barrier performance. In order to obtain impedance parameters such as resistances and\ncapacitances, two well-known equivalent circuits are chosen using\nthe ZSimpwin software, as shown in Figure 14 . The relative parameters\nare listed in Table 3 . Figure 14 a shows\nthe equivalent electrical circuit for the as-sprayed Al coating, the\netched Al coating, and the superhydrophobic Al coating. Figure 14 Electrical\nequivalent circuits for EIS of the coatings: (a) R(Q(R(QR)))\nand (b) R(Q(R(Q(RW)))). Table 3 Fitting\nCircuit Parameters for EIS\nof the Coatings sample R s Ω·cm2 Q c × 10 –6 S cm –2 s n n c R c Ω·cm 2 Q dl × 10 –6 S cm –2 s n n dl R ct × 10 5 Ω cm 2 W × 10 –3 S cm –2 s 5 χ 2 ×10 –4 the as-sprayed Al coating 10.41 7.065 0.8334 5058 7.795 0.5577 1.216 3.53 the modified Al coating 5.893 1.187 0.8067 6398 6.810 0.2454 6.720 6.066 5.50 the etched Al coating 15.10 4.268 0.8689 21250 9.632 0.4784 4.208 2.66 the superhydrophobic Al\ncoating 14.19 1.110 0.9205 30940 9.358 0.6589 14.74 3.60 Figure 14 b summarizes\nthe equivalent circuit of the modified Al coating due to the two semicircles\nobserved on the Nyquist plot. In the circuits, R s is the solution resistance, R c is the coating resistance, and R ct is\nthe charge-transfer resistance. Due to the inhomogeneity of the electrode\nsurface, the frequency response characteristics of the double-layer\ncapacitor are inconsistent with those of the pure capacitor. 39 To obtain a better fitting result, the constant\nphase angle element is used to replace the ideal capacitance in the\nequivalent circuit. 40 Q c and Q dl represent the coating\ncapacitance and the double-layer capacitance, respectively. Due to\nthe obvious dielectric difference between the as-sprayed Al coating\nand the substrate, there are two time constants on the fitting circuit\ndiagram, namely, R c and Q c are parallel, which corresponds to the dielectric property\nof the as-sprayed Al coating and the first time constant. Another\nparallel subcircuit, R ct and Q dl , means the dielectric properties of the coating-substrate\ninterface, giving a second time constant. In addition, for the modified\nAl coating, since the charge transfer is affected by the semi-infinite\ndiffusion process, Warburg impedance ( W ) appears\nin the equivalent circuit. From Table 3 , the R ct value of the\nsuperhydrophobic Al coating is about 12 times that of the as-sprayed\nAl coating. The large R ct value denotes\nthat the charge-transfer process at the interface between the coating\nand the substrate is more difficult. 39 Therefore,\nthe superhydrophobic Al coating dominates a superior corrosion resistance. In addition, Table 4 summarizes the E corr and I corr of the superhydrophobic surfaces prepared by the\nchemical etching method of this work and some previous works from\nthe references. It is found that the superhydrophobic Al coating investigated\nin this work not only has a lower I corr but also manifests a higher E corr than\nthe other superhydrophobic surfaces on the Al substrate and 6061-Al\nalloy. Therefore, the developed superhydrophobic Al coating has superior\ncorrosion resistance. It can provide valuable guidance for the protection\nof the marine engineering steel structure. Table 4 Corrosion\nResistance Property of the\nSuperhydrophobic Surfaces from This Work and Literature Studies matrix Methods low surface\nenergy materials E corr (V) I corr (μA/cm 2 ) NaCl\nsolution (wt %) time (h) refs Al foils chemical\netching (CuCl 2 ) SA –1.38 6.10 3.5 1 ( 26 ) Al plates chemical etching (NaOH) ZnAl–LDH–La –0.76 0.0674 3.5 1 ( 41 ) 6061-Al alloy chemical\netching (NaOH) SA –0.58 0.0350 3.5 1 ( 25 ) 6061-Al alloy chemical\netching (HCl) FAS-17 –0.74 0.205 3.5 1 ( 42 ) Al coating chemical etching (NaOH) FAS-17 –0.62 0.0136 3.5 1 this work"
} | 7,820 |
31689680 | PMC6964152 | pmc | 4,576 | {
"abstract": "Complementary Learning Systems (CLS) theory suggests that the brain uses a ’neocortical’ and a ’hippocampal’ learning system to achieve complex behaviour. These two systems are complementary in that the ’neocortical’ system relies on slow learning of distributed representations while the ’hippocampal’ system relies on fast learning of pattern-separated representations. Both of these systems project to the striatum, which is a key neural structure in the brain’s implementation of Reinforcement Learning (RL). Current deep RL approaches share similarities with a ’neocortical’ system because they slowly learn distributed representations through backpropagation in Deep Neural Networks (DNNs). An ongoing criticism of such approaches is that they are data inefficient and lack flexibility. CLS theory suggests that the addition of a ’hippocampal’ system could address these criticisms. In the present study we propose a novel algorithm known as Complementary Temporal Difference Learning (CTDL), which combines a DNN with a Self-Organizing Map (SOM) to obtain the benefits of both a ’neocortical’ and a ’hippocampal’ system. Key features of CTDL include the use of Temporal Difference (TD) error to update a SOM and the combination of a SOM and DNN to calculate action values. We evaluate CTDL on Grid World, Cart–Pole and Continuous Mountain Car tasks and show several benefits over the classic Deep Q-Network (DQN) approach. These results demonstrate (1) the utility of complementary learning systems for the evaluation of actions, (2) that the TD error signal is a useful form of communication between the two systems and (3) that our approach extends to both discrete and continuous state and action spaces.",
"conclusion": "5 Conclusions Taken together, we believe that our work highlights CTDL as a promising avenue for achieving complex, human-like behaviour and exploring RL within the brain. Having a ‘neocortical’ and ‘hippocampal’ learning system operating in parallel conferred a learning advantage over a single ‘neocortical’ system. This advantage was attributable to two main properties of CTDL. Firstly, both the ‘neocortical’ and ‘hippocampal’ system contributed to the calculation of action values for decision-making, with the arbitration between the two dependent on the memory content of the ‘hippocampal’ system. Secondly, the contents of the ‘hippocampal’ system were dynamically updated using the TD error from the ‘neocortical’ system. This allowed the ‘hippocampal’ system to target regions of the state space that the ‘neocortical’ system was poor at evaluating or that violated generalizations made by the ‘neocortical’ system. These key properties of CTDL represent promising avenues for future research both computationally and empirically. From a computational perspective, it will be interesting to explore how embedding functions can be utilized to reflect the fact that the hippocampus receives latent representations from cortical areas as input. This may be a key component for scaling up CTDL to complex problems with high dimensional state spaces. With respect to future empirical work, CTDL can be used to make predictions about which tasks should utilize the hippocampus and which regions of the state space should be encoded by it. CTDL also predicts that TD errors should promote the formation of episodic memories in the hippocampus and so we highlight this as a key area for further investigation and clarification.",
"introduction": "1 Introduction Reinforcement Learning (RL) ( Sutton & Barto, 1998 ) represents a computational framework for modelling complex reward-driven behaviour in both artificial and biological agents. For cognitive scientists it is of continuing interest to explore how RL theory maps onto neural structures in the brain ( Lee et al., 2012 , Niv, 2009 ). One of the most influential findings in this regard is the encoding of Temporal Difference (TD) error by phasic midbrain dopaminergic neurons ( Schultz, 2016 , Schultz et al., 1997 ). One of the major projection sites of these neurons is the striatum ( Bray and Doherty, 2007 , Doherty et al., 2003 , Mcclure et al., 2003 ) and it has been proposed that the striatum is responsible for evaluating states and actions for decision making ( Houk et al., 1995 , Roesch et al., 2009 , Schultz, 1998 , Schultz et al., 1992 , Setlow et al., 2003 ). Interestingly, the striatum receives inputs from both cortical areas and the hippocampus, suggesting that it is responsible for evaluating different forms of information. Complementary Learning Systems (CLS) theory posits that the neocortex and hippocampus have complementary properties that allow for complex behaviour ( Kumaran et al., 2016 , McClelland et al., 1995 ). More specifically, the hippocampus relies on fast learning of conjunctive, pattern-separated memories. These memories then support the learning of a second system, the neocortex, which slowly learns distributed representations that support generalization across features and experiences. The purpose of the present study is to explore how the brain’s RL machinery might utilize these opposing properties to achieve complex behaviour. Much of the recent success of RL has been due to the combination of classical RL approaches with the function approximation properties of Deep Neural Networks (DNNs), known as deep RL ( François-lavet et al., 2018 ). Typically in deep RL, the action-value function Q ( s , a ) is represented using a DNN that takes the state s t as input and outputs the corresponding action values for that state. Despite impressive results, such as super human-level performance on Atari video games ( Mnih et al., 2015 ), deep RL approaches are often criticized for being data inefficient and adapting poorly to changes in the input distribution ( Lake, Ullman, Tenenbaum, & Gershman, 2017 ). From a CLS perspective, the DNNs used in deep RL can be seen as sharing many commonalities with a ‘neocortical’ learning system. In particular, both the neocortex and DNNs rely on small learning rates and distributed representations for efficient generalization. CLS theory suggests that the addition of a ‘hippocampal’ learning system to deep RL approaches may improve our understanding of how RL is implemented in the brain and address the aforementioned criticisms of deep RL. Ideally an agent should be able to utilize the advantages of a ‘neocortical’ system (distributed representations and generalization) and a ‘hippocampal’ system (fast learning and pattern-separation) to perform complex reward-driven behaviour. Indeed, many theoretical advantages have been proposed for the use of hippocampal episodic information in RL. In particular, it has been suggested that episodic information can be used to approximate value functions, increase data efficiency and reconcile long-range dependencies ( Gershman & Daw, 2017 ). An alternative to using a DNN to represent the action-value function is to represent it in a tabular manner ( Sutton & Barto, 1998 ). Such an approach is more in line with a hippocampal learning system as experiences are stored in a pattern-separated manner and larger learning rates can be used. Importantly, the tabular case means that every action value is stored as its own memory, which eliminates the potential for interference. This is in contrast to DNNs that naturally suffer from interference due to their distributed representations. However, as the number of states and/or actions increases, the tabular case will require more experience to encounter each action-value and more computational resources to store the values. The distributed representations of DNNs then become advantageous because they allow for efficient generalization over the state space. In an ideal scenario a DNN would be responsible for generalization over certain areas of the state space while a tabular method would store pattern-separated memories that are crucial to behaviour and that violate the generalizations of the network. Previous work in deep RL has often touched upon CLS theory and the benefits of a hippocampal learning system. Indeed, one of the most influential deep RL approaches, the Deep Q-Network (DQN) ( Mnih et al., 2015 ), utilizes a secondary system that tentatively mirrors a hippocampal learning system. More specifically, the DQN has a table that stores past experiences in a pattern-separated manner and then uses them to train a DNN in an interleaved fashion. This process is tentatively compared to ‘replay’; a biological phenomenon that appears to replay information from the hippocampus to the neocortex in biological agents ( Olafsdottir, Bush, & Barry, 2018 ). However, despite the DQN having what appears to be two complementary learning systems, the decision making (calculation of Q values) is ultimately based on the predictions of the DNN, which learns slowly via distributed representations. More recently, research in deep RL has begun to demonstrate some of the advantages of an explicit ‘hippocampal’ learning system that evaluates states and actions ( Botvinick et al., 2019 ). Most notably ( Blundell, Pritzel, & Rae, 2016 ) proposed an algorithm called ‘model-free episodic control’, which consisted of a table containing the maximum return (sum of discounted rewards) for each state–action pair experienced. The memory requirements for this table were kept constant by removing the least recently updated table entry once the size limit had been reached. Each observation from the environment was projected by an embedding function (either a random projection or a variational autoencoder) to a state value and actions were selected based on a k-nearest neighbours method, which allowed for some degree of generalization to novel states. Blundell et al. (2016) tested this approach on the Arcade Learning Environment (Atari) ( Bellemare, Naddaf, Veness, & Bowling, 2013 ) and Labyrinth ( Mnih et al., 2016 ), which both require the use of visual information to learn an optimal policy. The results of these simulations showed that model-free episodic control was significantly more data efficient than other classical deep RL approaches, suggesting that episodic information is important for fast learning. While taking a first step towards highlighting the benefits of a ‘hippocampal’ learning system that utilizes fast learning of pattern-separated information, the work of Blundell et al. (2016) has several notable drawbacks. Firstly, the table recorded the maximum return from any given episode and used this to inform the policy of the agent. This naturally cannot handle stochastic environments, where the expected return is the important quantity and not the maximum return of an individual episode. Secondly, this approach is likely to be highly inflexible. For example if a state–action pair suddenly becomes highly aversive then the entry in the table will not be updated because only the maximum value is stored. A third criticism is that the approach relies on the full return for each state–action pair and this is only possible when the task has distinct finite episodes. Some of these criticisms have been addressed in subsequent work, for example Pritzel et al. (2017) propose a fully differentiable version of ‘model-free episodic control’ that learns the embedding function in an online fashion using N-step Q-learning. The above issue not withstanding, what is most pertinent to the present study is that ‘model-free episodic control’, and its various derivatives, do not rely on two complementary learning systems that operate in parallel to evaluate actions. The embedding function may be tentatively compared to a ‘neocortical’ learning system but it operates before the ‘hippocampal’ learning system and as a result only the output of the ‘hippocampal’ learning system is used to evaluate action values. This means that any advantages that may be conferred from the additional predictions of a ‘neocortical’ learning system are lost. In essence, the aforementioned approaches cannot arbitrate between the predictions of a ‘neocortical’ and a ‘hippocampal’ learning system, but are instead restricted to using episodic predictions. This is inconsistent with the finding that the striatum receives inputs from both cortical areas and the hippocampus and needs to arbitrate between the two ( Pennartz, Ito, Verschure, Battaglia, & Robbins, 2011 ). With these criticisms in mind, we present a novel method for imbuing a deep RL agent with both a ‘neocortical’ and a ‘hippocampal’ learning system so that it benefits from both types of learning system. Most importantly these two systems: (1) learn in parallel, (2) communicate with each other using a biologically plausible signal, and (3) both make action value predictions. We represent the ‘neocortical’ system as a DNN and the ‘hippocampal’ system as a Self-Organizing Map (SOM). Importantly, the size of the SOM is significantly smaller than the state space experienced by the agent so as to replicate the restricted computational resources experienced by biological agents. The SOM is tasked with storing pattern-separated memories of states that the DNN is poor at evaluating. To achieve this we use the TD error from a DNN in order to train the SOM. Critically, this novel CLS approach demonstrates how the TD error of a ‘cortical’ system can be used to inform a ‘hippocampal’ system about when and what memories should be stored, with both systems contributing to the evaluation of action-values. This allows the agent to utilize the benefits of both a neocortical and hippocampal learning system for action selection. We call our novel algorithm Complementary Temporal Difference Learning (CTDL) and demonstrate that it can improve the performance and robustness of a deep RL agent on Grid World, Cart–Pole and Continuous Mountain Car tasks.",
"discussion": "4 Discussion According to CLS theory, the brain relies on two main learning systems to achieve complex behaviour; a ‘neocortical’ system that relies on the slow learning of distributed representations and a ‘hippocampal’ system that relies on the fast learning of pattern-separated representations. Both of these systems project to the striatum, which is believed to be a key structure in the evaluation of states and actions for RL ( Houk et al., 1995 , Roesch et al., 2009 , Schultz, 1998 , Schultz et al., 1992 , Setlow et al., 2003 ). Current deep RL approaches have made great advances in modelling complex behaviour, with DNNs sharing several similarities with a ‘neocortical’ learning system. However these approaches tend to suffer from poor data efficiency and general inflexibility ( Lake et al., 2017 ). The purpose of the present study was to explore how a ‘neocortical’ and ‘hippocampal’ learning system could interact within an RL framework and whether CLS theory could alleviate some of the criticisms of deep RL. Our novel approach, termed CTDL, used a DNN as a ‘neocortical’ learning system and a SOM as a ‘hippocampal’ learning system. Importantly the DNN used a small learning rate and distributed representations while the SOM used a larger learning rate and pattern-separated representations. Our approach is novel in that the SOM contributes to action value computation by storing action values independently from the DNN and uses the TD error produced by the DNN to update its state representations. More specifically, the TD error produced by the DNN is used to dynamically set the learning rate and standard deviation of the neighbourhood function of the SOM in an online manner. This allows the SOM to store memories of states that the DNN is poor at predicting the value of and use them for decision-making and learning. Importantly the size of the SOM is smaller than the state space encountered by the agent and so it requires less memory resources than the purely tabular case. We compared the performance of CTDL to a standard DQN on a random set of 2D grid worlds. CTDL out-performed the DQN on the majority of grid worlds, suggesting that the inclusion of a ‘hippocampal’ learning system is beneficial and confirming the predictions of CLS theory. Removal of replay between the SOM and DNN appeared to have marginal impact upon the performance of CTDL suggesting that the SOMs contribution to the calculation of action values is the predominant benefit of CTDL. Future work should explore how information from the SOM may be replayed to the DNN in a more principled fashion (e.g. Mattar and Daw (2018) ) instead of random sampling. We proposed that the SOM was able to contribute to the calculation of the action values in a targeted manner by using the TD error of the DNN to encode states that the DNN was poor at evaluating. We provided evidence of this by demonstrating that the removal of the TD signal between the DNN and SOM had a negative impact upon the performance of CTDL. Our interpretation of these results is that, particularly early on in learning, the DNN is able to represent generalizations of the state space while the SOM is able to represent violations of these generalizations. In combination these two systems can then be used to formulate policies in both a general and specialized manner. We tested this hypothesis by presenting CTDL and DQN with a grid world consisting of a general rule and two other grid worlds consisting of violations of this rule. As our interpretation predicted, CTDL out-performed the DQN when violations of the general rule were present, presumably because the SOM was able to store states that were useful for circumnavigating these violations. This hypothesis was further supported by a simulation that ran both CTDL and DQN on sequential grid worlds. CTDL appeared to be better equipped to deal with the change in environment compared to the DQN. In addition, the SOM component of CTDL encoded states close to the change in the environment, providing further evidence of its ability to represent violations of predictions. This ability of the SOM to encode violations of the generalizations made by the DNN has interesting parallels to imaging work in rodents demonstrating that CA3 neurons appear to encode decision points in T-mazes that are different from the rodents current position ( Johnson & Redish, 2007 ). Such decision points can be viewed as obstacles or important deviations from the animals general direction and we therefore predict that they would be encoded by the SOM component of CTDL. In the future, application of CTDL to other reinforcement learning tasks may provide testable predictions about the regions of the state space that should be encoded by the hippocampus. In addition, the fact that the SOM encode states close to obstacles in order to account for changes in the environment suggests that the hippocampus may be important for adapting to changes in the environment and is consistent with recent studies that have implicated the hippocampus in reversal learning ( Dong et al., 2013 , Vila-Ballo et al., 2017 ). To investigate the generality of CTDL we also applied it to the Cart–Pole and Continuous Mountain Car problems. The Cart–Pole problem is fundamentally different to the grid world problem because the state space observed by the agent is continuous. We found that in comparison to the DQN, the learning of CTDL was more gradual but also more robust. This is perhaps a surprising result given that the DQN has a perfect memory of the last 100,000 state transitions whereas CTDL has no such memory. Indeed, one would expect the SOM component of CTDL to have less of an effect in continuous state spaces because generalization from function approximation becomes more important and the probability of re-visiting the same states decreases. With this being said, Blundell et al. (2016) demonstrated that even when the probability of re-visiting the same state is low, episodic information can still be useful for improving learning. Generalization of episodic information in CTDL is likely controlled by the temperature parameter τ η that scales the euclidean distance between the states and the weights of the SOM units. The Continuous Mountain Car problem consists of both a continuous state and action space. In order to apply deep RL to the Continuous Mountain Car problem we used an A2C architecture, with two separate DNNs representing an actor and critic respectively. As with the original implementation of CTDL, we augmented A2C with a ‘hippocampal’ learning system in the form of a SOM and termed the resulting algorithm CTDL A 2 C . CTDL A 2 C both outperformed the standard A2C approach and demonstrated more robust learning with no substantial decreases in performance. A defining feature of the Continuous Mountain Car problem is that the agent will learn not to move unless it experiences the positive reward of the target location and then utilizes this information efficiently. It is possible that the addition of a learning system that quickly learns pattern-separated representations helps to alleviate this problem by storing rare and surprising events in memory and incorporating them into value estimates, rather than taking a purely statistical approach. More generally, these results demonstrate the applicability of CTDL to continuous control problems and further highlight the benefits of using TD error to inform the storage of episodic information. The reduced benefit of CTDL on the Cart–Pole problem compared to the Grid World and Continuous Mountain Car problems may allude to interesting differences in task requirements. In particular, both the Grid World and Continuous Mountain Car problems appear to rely on rare discrete events that are highly informative for learning a policy e.g. both tasks involve a goal location. In comparison, the Cart–Pole task relies on a range of rewarded events or states to inform the policy and so the utilization of episodic information may be less valuable. From a biological perspective, it is perhaps unsurprising that CTDL performs better on Grid World problems given that they represent spatial navigation tasks which are thought to heavily recruit the hippocampus in biological agents ( Burgess, Maguire, & Keefe, 2002 ). In comparison, the Cart–Pole problem can be seen as a feedback-based motor control task which involves learning systems such as the cerebellum in addition to any cortical-hippocampal contributions. CTDL may therefore represent a useful empirical tool for predicting the utilization of hippocampal function in biological agents during RL tasks. Future work will need to investigate whether the increased robustness and performance of CTDL in continuous state and action spaces is a general property that extends to more complex domains. In particular, it would be of interest to run CTDL on maze problems such as ViZDoom ( Kempka, Wydmuch, Runc, Toczek, & Ja, 2016 ), which are rich in visual information. Indeed, deep RL approaches using convolutional neural networks are at the forefront of RL research and these could be easily incorporated into the CTDL approach. In the case of ViZDoom, each state is represented by a high-dimensional image and so the generalization capabilities of a DNN are crucial. From a biological perspective, it is worth noting that the hippocampus operates on cortical inputs that provide latent representations for episodic memory. This is captured in ‘model-free episodic control’, which relies on an embedding function to construct the state representation for episodic memory ( Blundell et al., 2016 , Pritzel et al., 2017 ). An embedding function therefore represents a biologically plausible method of scaling CTDL up to complex visual problems such as VizDoom. The embedding function could be pre-trained in an unsupervised manner or sampled from the DNN component of CTDL. We leave this interesting avenue of research to future work. In addition to relatively low complexity, one consistent feature of the tasks presented in the present study was a low degree of stochasticity. As with discrete state spaces, low stochasticity means that events re-occur with high probability and the episodic component of CTDL can exploit this. It is likely that in more stochastic environments the benefits of CTDL will be reduced as the DNN is required to generalize over several outcomes. It is therefore an open question how well CTDL will perform on tasks that have a high degree of stochasticity, which are also supposedly harder for biological agents. One interesting element of CTDL that was not explored in the present study was the temporal evolution of pattern-separated representations in the SOM. Logically as the DNN improves its ability to evaluate the optimal value function its TD errors should reduce in magnitude and therefore free up the SOM to represent other episodic memories. If part of the environment changes then a new episodic memory will form based on the new TD error and it will remain in episodic memory until the ‘neocortical’ learning system has learnt to incorporate it. CTDL therefore suggests that the transfer of information from the hippocampus to neocortex is based to the ‘need’ for an episodic memory as encoded by TD errors. This can be viewed as a form of ‘consolidation’ whereby memories stored in the hippocampus are consolidated to the neocortex over time ( Olafsdottir et al., 2018 ). As a concluding remark, we have only demonstrated the benefits of a ‘hippocampal’ learning system from a purely model-free perspective. A growing body of research however, is implicating the hippocampus in what has historically been considered model-based behaviour. For example, it has been proposed that the hippocampus encodes a predictive representation of future state occupancies given the current state of the agent ( Stachenfeld, Botvinick, & Gershman, 2017 ). This predictive representation has been termed the Successor Representation (SR), and has been shown to imbue agents with model-based behaviour using model-free RL mechanisms in a range of re-evaluation tasks ( Russek, Momennejad, Botvinick, Gershman, & Daw, 2017 ). Therefore the inclusion of a ‘hippocampal’ learning system may have additional benefits above and beyond those demonstrated by CTDL."
} | 6,544 |
26577813 | PMC4649621 | pmc | 4,578 | {
"abstract": "The bright and iridescent blue color from Morpho butterfly wings has attracted worldwide attentions to explore its mysterious nature for long time. Although the physics of structural color by the nanophotonic structures built on the wing scales has been well established, replications of the wing structure by standard top-down lithography still remains a challenge. This paper reports a technical breakthrough to mimic the blue color of Morpho butterfly wings, by developing a novel nanofabrication process, based on electron beam lithography combined with alternate PMMA/LOR development/dissolution, for photonic structures with aligned lamellae multilayers in colorless polymers. The relationship between the coloration and geometric dimensions as well as shapes is systematically analyzed by solving Maxwell’s Equations with a finite domain time difference simulator. Careful characterization of the mimicked blue by spectral measurements under both normal and oblique angles are carried out. Structural color in blue reflected by the fabricated wing scales, is demonstrated and further extended to green as an application exercise of the new technique. The effects of the regularity in the replicas on coloration are analyzed. In principle, this approach establishes a starting point for mimicking structural colors beyond the blue in Morpho butterfly wings.",
"discussion": "Discussions Owing to the limitation of fabrication technique, there are still a number of differences in the structures between the fabricated and real wing scales, leading to deviations in coloration. First, the ridge pillars are formed by different refractive index materials (PMMA and LOR), which causes the reduction of reflectance by 10–20% with some dips on the peaks arising from resonant transmission modes in the PMMA/LOR pillar; Second, the artificial scales have strong regularity in the structure including periodical ridge grating with the same height and flat top. As the result, red component by the ridge grating is added to the total color by 10%, leading to slight redshift with oblique viewing angels. Furthermore, the angle independence of blue color is narrowed down to roughly ±16°. Third, the material in the artificial scale is polymer based and totally colorless. In this issue, no observable effect is detected, indicating that the blue color in this work is entirely caused by the lamellae structure free-standing on the scale without the need of pigment. Finally, the total periods are 7 with 15 layers, which is much less than the real ones. This is believed to be the main reason for low reflectivity in the artificial wing scale. Nevertheless, the observed coloring property by the mimicked scales provide us with invaluable experimental evidences in the interpretation of the blue iridescence seen from real butterfly wings. More importantly, it points out the direction in technical development to improve the process and the design by creating necessary irregularity in real scales. For example, the surface of the substrate can be pre-treated to create random mesas with the heights in the order of 50 nm. Fluorine based dry etch in plasma can naturally file the top corners of each tree in the ridge grating to form Christmas-tree like shape. Therefore, we believe, with further improvements on the nanolithography process innovated in this work, the artificial butterfly wings should mimic real ones closely, leading to promising applications in the future."
} | 865 |
32968619 | null | s2 | 4,580 | {
"abstract": "Although microbes competing for simple substrates are well-known to obey the ecological competitive exclusion principle, little is known regarding how complex substrates influence the ecology of microbial communities. The vast structural diversity of polysaccharides makes them ideal substrates for cooperative microbial degradation. Potential mechanisms for division of metabolic labor in microbial communities consuming polysaccharides are 1) complementary differences in gene content, 2) alternate regulation of polysaccharide degradation genes, 3) subtle differences in hydrolytic enzyme functionality, and 4) specialization in transport and consumption of hydrolysis products. Engineering division of labor in polysaccharide degradation using these mechanisms as control points may improve our ability to engineer microbiomes for improved productivity and stability in diverse environments."
} | 223 |
26648912 | PMC4664643 | pmc | 4,581 | {
"abstract": "Many definitions of resilience have been proffered for natural and engineered ecosystems, but a conceptual consensus on resilience in microbial communities is still lacking. We argue that the disconnect largely results from the wide variance in microbial community complexity, which range from compositionally simple synthetic consortia to complex natural communities, and divergence between the typical practical outcomes emphasized by ecologists and engineers. Viewing microbial communities as elasto-plastic systems that undergo both recoverable and unrecoverable transitions, we argue that this gap between the engineering and ecological definitions of resilience stems from their respective emphases on elastic and plastic deformation, respectively. We propose that the two concepts may be fundamentally united around the resilience of function rather than state in microbial communities and the regularity in the relationship between environmental variation and a community's functional response. Furthermore, we posit that functional resilience is an intrinsic property of microbial communities and suggest that state changes in response to environmental variation may be a key mechanism driving functional resilience in microbial communities.",
"conclusion": "Conclusions and future research needs Moving toward an integrated framework for understanding microbial community resilience, we propose reconciling concepts of engineering and ecological resilience through (1) consideration of microbial communities as systems that undergo both elastic and plastic deformation, and (2) defining resilience as the rate of recovery of a function of interest. Refocusing on the system's fundamental characteristics (such as the community-level functions and community-environment relationships) not only minimizes conceptual variation across different resilience definitions, but also provides a deeper understanding of the intrinsic community properties. In parallel, from a practical point of view, it is also of great importance to develop rational methods for quantifying microbial community resilience and predicting approaching tipping points. Future research will need to address several important, unresolved issues—primarily, the identification of fundamental mechanisms responsible for microbial community resilience. For example, redundancy, diversity, and modularity are frequently advanced as mechanisms for robustness in complex systems (Kitano, 2004 ), but in some cases, and particularly in structurally simple consortia, they may not be directly related to resilience. The question remains: what unifying mechanisms impart resilience across both structurally simple and complex microbial communities? While the concept of networked buffering offers a potential mechanism (Whitacre and Bender, 2010 ; Konopka et al., 2014 ), rigorous analysis of microbial communities has yet to be performed. Another issue is the possible occurrence of trade-offs between system robustness (or resilience) and performance (Kitano, 2007 ) or trade-offs between robustness with respect to distinct perturbations (e.g., the conservation principle as discussed by Doyle and colleagues; Carlson and Doyle, 1999 ; Csete and Doyle, 2002 ). One major question is to what degree do resilience mechanisms identified for microbial communities overcome such trade-offs? Finally, the principles for structural organization of microbial communities as robust networks need to be further examined, as little is known about the general topological characteristics of microbial association networks and their relationships to resilience. Critical questions include: how does the compartmentalization of genes into a network of species affect the structural and higher-order properties of microbial communities; and are microbial community properties better understood as networks of species or networks of genes? Future research focusing on these issues will significantly advance our capability for the design, prediction, and control of microbial communities and maintenance of the critical ecosystem services they provide. 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 Microorganisms collectively exceed the biomass of all macrobiota on the planet (Whitman et al., 1998 ). Communities of microbes control the biogeochemical cycles upon which all macrobiota depend (Falkowski et al., 2008 ; Strom, 2008 ; Nazaries et al., 2013 ), and the role of microbial communities in shaping human health and physiology is also increasingly appreciated (Song et al., 2014a ; Braundmeier et al., 2015 ; Lone et al., 2015 ; Sassone-Corsi and Raffatellu, 2015 ). Although natural microbial communities continually respond to perturbations (Konopka et al., 2014 ), their functioning can exhibit remarkable stability over time (Fuhrman et al., 2015 ), even under extreme environmental variation (Shade et al., 2012b ; Lindemann et al., 2013 ). Comprehending the processes governing the responses of microbial communities to perturbation is critical both to ecologists concerned with predicting effects on ecosystem function (Hooper et al., 2005 ) and engineers designing communities for stable biotechnological processes (Lucas et al., 2015 ). Though there is widespread interest in factors driving microbial community stability, the conceptual bases of stability measures, like resilience, are poorly defined. A report by the Community and Regional Resilience Institute (CARRI Report, 2013 ) summarized 47 definitions of resilience used in diverse scientific areas including engineering, ecology, sociology, economics, and psychology. Ambiguity is found even within disciplines: in ecology, resilience has been discussed alongside, and sometimes interchangeably with, ~70 other terms describing various stability measures (e.g., resistance, sustainability, and vulnerability; Grimm and Wissel, 1997 ). The conceptual variability in ecology surrounding stability, and resilience in particular, likely stems from system-specificity. Microbial communities span orders of magnitude in the diversity of their interacting components, from experimental or engineered systems to diverse natural communities. The metrics employed to evaluate each system's stability are largely idiosyncratic. The diversity of environments in which communities are investigated, the large array of functions of interest, and the range of research objectives concerning stability beg the question of whether a single definition of resilience can be universally applied across systems and scales for microbial communities. Seeking an integrated concept applicable to all microbial communities, we herein compare engineering and ecological resilience and reconcile them by arguing that resilience is an intrinsic property of complex adaptive systems which, after perturbation, recover their system-level functions and interactions with the environment, rather than their endogenous state."
} | 1,774 |
27114888 | PMC4841221 | pmc | 4,582 | {
"abstract": "On Hawaiian reefs, the fast-growing, invasive algae Gracilaria salicornia overgrows coral heads, restricting water flow and light, thereby smothering corals. Field data shows hypoxic conditions (dissolved oxygen (DO 2 ) < 2 mg/L) occurring underneath algal mats at night, and concurrent bleaching and partial tissue loss of shaded corals. To analyze the impact of nighttime oxygen-deprivation on coral health, this study evaluated changes in coral metabolism through the exposure of corals to chronic hypoxic conditions and subsequent analyses of lactate, octopine, alanopine, and strombine dehydrogenase activities, critical enzymes employed through anaerobic respiration. Following treatments, lactate and octopine dehydrogenase activities were found to have no significant response in activities with treatment and time. However, corals subjected to chronic nighttime hypoxia were found to exhibit significant increases in alanopine dehydrogenase activity after three days of exposure and strombine dehydrogenase activity starting after one overnight exposure cycle. These findings provide new insights into coral metabolic shifts in extremely low-oxygen environments and point to ADH and SDH assays as tools for quantifying the impact of hypoxia on coral health.",
"conclusion": "Conclusions Our findings indicate that, M. capitata increasingly relies upon ADH and SDH for anaerobic catabolism under low-oxygen conditions. This demonstrates that prolonged or repeated exposure to anoxia drives corals to rely upon less efficient anaerobic means of energy production. Under extended periods of low-oxygen exposure this may lead to energy deficits, which could result in increased susceptibility of corals to acute stress events, further leading to bleaching and tissue loss. Future efforts will focus on the application of these methods toward the analysis of oxygen deprivation on corals of different genera, elucidating possible variations in enzymatic stress response. These findings point to ADH and SDH activity assays as suitable biomarkers for rapid analyses of hypoxia-induced stress in corals with use in future analyses of environmental impacts on corals and coral reefs.",
"introduction": "Introduction Global coral health Coral reefs are important cultural, ecological, and economic resources, providing critical marine habitats for many invertebrates, fish, and algae species ( McClanahan, Polunin & Done, 2002 ; Hughes et al., 2003 ). Their complex structure provides marine life with food, suitable habitats for growth, and protection from predators, while also acting as natural barriers that buffer adjacent coastlines from coastal erosion ( McManus, 1997 ; White, Vogt & Arin, 2000 ; Cesar, Burke & Pet-soede, 2003 ; Bishop et al., 2011 ). However, despite their value, coral reefs have been devastated by increased anthropogenic impacts and changing abiotic environmental factors that continue to overwhelm these ecosystems ( McClanahan, Polunin & Done, 2002 ; Hughes et al., 2003 ; Fabricius, 2005 ; De’ath et al., 2012 ; Bahr, Jokiel & Rodgers, 2015 ). Destructive fishing practices, terrestrial run-off and pollution, sewage effluent infusion, coupled with intensifying storms, regional warming events, and invasive species impacts, have led to major losses in scleractinian coral cover, fish abundance, and decreased ability of reefs to support local human populations ( Stimson, Larned & Conklin, 2001 ; Fox et al., 2003 ; Fabricius, 2005 ; Dailer et al., 2010 ). These continued pressures have resulted in mass coral bleaching events, and have the ability to disrupt spawning patterns and trigger widespread coral mortality ( Downs et al., 2002 ; Levitan et al., 2014 ; Bahr, Jokiel & Rodgers, 2015 ; Paxton et al., 2015 ). As a result, mass losses of coral cover and subsequent shifts in the ecosystem balance has the potential to further reduce the abundance and diversity of fish and invertebrate species ( Jones et al., 2004 ). Historically, instances of mass coral mortality in combination to altered environmental conditions have led to phase shifts in coral reef structure, wherein coral reefs have shifted from a coral to an algae dominated state ( McCook, 1999 ; Stimson, Larned & Conklin, 2001 ). Although herbivores, such as urchins and fish, can potentially limit the expansion of algal cover ( Westbrook et al., 2015 ), aggressive invasive algae and herbivore grazing preferences for native species have had a profound effect on coral reefs, especially in the Hawaiian Islands where such algae have been found to overgrow coral colonies and negatively impact coral health ( McCook, Jompa & Diaz-Pulido, 2001 ; Stimson, Larned & Conklin, 2001 ; Smith et al., 2004 ). These impacts include algal overgrowth creating oxygen and solar radiation-poor environments for corals, which induce bleaching through photoinhibition and reduced photorespiration ( Martinez, Smith & Richmond, 2012 ). Due to their distribution and lack of motility, corals are vulnerable to hypoxia-inducing circumstances including algal overgrowth, eutrophication, and sedimentation ( Martinez, Smith & Richmond, 2012 ). Studies have been conducted documenting the negative impacts of oxygen deprivation on coral health ( Stimson, Larned & Conklin, 2001 ; Smith et al., 2006 ; Fabricius, 2011 ; Jokiel et al., 2014 ); however, these studies relied upon proxies analyzing long-term effects and percent coral mortality. As such, there is a need to further understand fine-scale impacts to coral health on a molecular level in order to monitor changes in health to reveal negative impacts prior to mortality. In order to address this need, in this study we have developed enzyme assays, which can be used to quickly quantify hypoxic metabolic stress in the coral Montipora capitata , a major reef-building coral in Hawaii. 10.7717/peerj.1956/fig-1 Figure 1 Typical cellular respiration pathway in eukaryotic cells. Red box denotes anaerobic respiratory pathway of interest. 10.7717/peerj.1956/fig-2 Figure 2 Pyruvate metabolism pathways. Conversion of pyruvate to lactate by lactate dehydrogenase (LDH, 1), pyruvate and alanine to alaopine by alanopine dehydrogenase (ADH, 2), pyruvate and arginine to octopine by octopine dehydrogenase (ODH, 3), and pyruvate and glycine to strombine by strombine dehydrogenase (SDH, 4). Coral metabolic biochemistry Under low oxygen conditions, organisms shift from aerobic to anaerobic metabolism of glucose, allowing for the continuation of energy production under hypoxic to anoxic conditions ( Fig. 1 ) ( Nelson & Cox, 2008 ). Rather than entering the pyruvate dehydrogenase enzyme complex following glycolysis, pyruvate can be instead converted to lactate by the enzyme lactate dehydrogenase (LDH) ( Fig. 2 ). While this reaction replenishes stores of NAD + , the required co-factor for catabolism of glucose molecules, it is accompanied by a significant decrease in net energy production from 38 moles to 2 moles of ATP per mole of glucose ( Nelson & Cox, 2008 ). Marine organisms, specifically polychaetes and cnidarians, have been shown to survive periods of environmental hypoxia (dissolved oxygen (DO 2 ) ≤ 2.0 mg/mL) for five or more days ( Schottler, 1982 ; Martinez, Smith & Richmond, 2012 ). However, reduced capacity for ATP production under hypoxia puts marine invertebrates at a greater disadvantage in energy production if they must rely upon anaerobic respiration for longer periods of time ( Livingstone, 1991 ). Many marine invertebrates, such as crustaceans and echinoderms, have been found to rely upon LDH and lactate production as a means of maintaining metabolism during oxygen stress ( De Zwaan & Putzer, 1985 ). However, the activities of opine dehydrogenases (OpDH) have also been discovered in many marine invertebrate tissues ( Plaxton & Storey, 1982 ; De Zwaan & Putzer, 1985 ; Livingstone, 1991 ; Sato et al., 1993 ; Lee, Lee & Pan, 2011 ). This enzyme suite, consisting of enzymes such as, alanopine dehydrogenase (ADH), octopine dehydrogenase (ODH), and strombine dehydrogenase (SDH) ( Fig. 2 ), has been characterized in a wide variety of organisms and has been found, in most cases, to be the favored pathway in anaerobic respiration over LDH ( Livingstone, 1991 ; Lee, Lee & Pan, 2011 ). Further, in many marine invertebrates, such as molluscs, polychaetes, cnidarians, and poriferans, OpDH activities have been found to be significantly higher than that of LDH ( Sato et al., 1993 ). Nevertheless, information regarding the activity and presence of these enzymes characterizing cellular anaerobic respiration is still limited. Though studies have been conducted analyzing the presence and production of LDH and OpDHs in the sea anemone Diadumene leucolena , little to no other published data exist describing the responses of LDH and OpDHs to oxygen deprivation in scleractinian corals ( Ellington, 1977 ; Ellington, 1979 ). Therefore, we sought to determine the activity of enzymes associated with anaerobic respiration, ADH, ODH, SDH, and LDH, to better characterize anaerobic metabolism in corals under prolonged oxygen deprivation, such that these enzymes may serve as proxies for rapidly characterizing hypoxic stress in coral. Due to its wide-spread distribution and observations of invasive algal overgrowth and oxygen deprivation of M. capitata in Kaneohe Bay, Oahu, Hawaii, this species was chosen as the model coral for this study.",
"discussion": "Discussion Increased activity of ADH and SDH enzymes in M. capitata exposed to prolonged durations of hypoxic conditions (>12 h) mimic other recorded increases in gastropod species ( Dando, 1981 ; Fields & Hochachka, 1981 ; Plaxton & Storey, 1982 ; Eberlee, Storey & Storey, 1983 ; Lee, Lee & Pan, 2011 ), as well as the sea anemone Diadumene leucolena ( Ellington, 1979 ). Though sequential significant increases in SDH and ADH activity were not found with increasing time intervals, treatment values in Figs. 6 and 7 demonstrate general trends of increasing activity with increasing duration of treatment. These results suggest that under prolonged intervals of anoxia, M. capitata becomes increasingly dependent on the activity of ADH and SDH to mediate anaerobic metabolism. While ODH and LDH activities have been monitored in the sea anemone Bunodosoma cavernata and various ctenophores and poriferans, their activity was barely detectable in M. capitata during this study ( Livingstone, 1991 ; Sato et al., 1993 ). Although it is well documented that ODH and LDH are important in the processing of glycolytic products under anaerobic conditions ( Sato et al., 1993 ; Lee, Lee & Pan, 2011 ), our lack of evidence does not mean that ODH and LDH do not exist in coral. The longest analyzed hypoxic treatment employed in this study was 3 days. However, it has been found that under chronic exposure to multiple weeks of hypoxia, other pathways for anaerobic respiration can be activated, becoming a significant pathway for metabolic turn over ( De Zwaan & Putzer, 1985 ). Additional studies propose that ADH and SDH act as the primary and anaerobic metabolic responses pathways during hypoxia, while ODH is important for chronic hypoxia exposure ( De Zwaan & Putzer, 1985 ). The significant positive relationship between exposure length and enzyme activity of ADH and SDH in M. capitata supports the implementation of these enzymes as biomarkers for rapid analysis of hypoxia-induced stress in corals. However, further research is necessary to provide additional evidence to confirm the presence of these enzymes through Western Immunoblotting or other molecular techniques. Furthermore, future studies monitoring the effect of long-term oxygen deprivation are required to fully understand variation between acute and chronic coral metabolic responses. Additionally, investigation of endogenous substrate and co-factor concentrations for these target enzymes within tissues through further colorimetric assays can help elucidate changes in the overall pathways and provide a better understanding of homeostatic health and subtler molecular alterations resulting from hypoxic exposure. Management strategies have been employed to mitigate the impacts from stressors, such as overgrowth by invasive algae ( Kittinger et al., 2013 ; Westbrook et al., 2015 ). However, these efforts occur once a stressor has already inflicted a significant amount of damage to a coral reef ( Stimson, Larned & Conklin, 2001 ; Conklin & Smith, 2005 ; Wolanski, Martinez & Richmond, 2009 ). Through the expansion of the cache of available biomarkers, we can better characterize stress levels in coral before bleaching and tissue loss/death occur. With this knowledge, we can actively sample corals that are found in environments that could be considered ‘less than optimal’ and specifically analyze the expression and activity of enzymes that should be up-regulated; ADH and SDH in low-oxygen environments, for example. Through this process, it would then be possible to proactively target coral reefs that are under ‘stress loads of concern,’ subsequently addressing the specific factor or factors eliciting the response. If explicitly looking to aid corals affected by invasive algal overgrowth, by understanding how severely increased the response of these enzymes are across a reef or several reefs, environmental managers will then be more efficiently able to prioritize efforts for the removal of invasive algae and/or the implementation of preventative measures, such as seeding reefs with herbivores, to improve environmental quality. Calls for such integration and collaboration between research science and coral reef management programs have been addressed in various papers ( Hughes et al., 2003 ; West & Salm, 2003 ; Aswani et al., 2015 ; Westbrook et al., 2015 ) and have led to successful reef management campaigns, such as that by The Nature Conservancy of Hawaii through their ‘Super Sucker’ and urchin bio-control programs. Further, many new studies leave room for the expansion of the analyses of coral response to factors such as hypoxia through the employment of newly described Acropora proteome data ( Dunlap et al., 2013 ). This will allow for the targeting of other enzymes of interest, such as hypoxia-inducible factors (HIF) 1 and 2, for use in rapid coral stress response analyses, complementing current findings, and expanding upon available biomarkers for use in assessing coral health in reef management programs."
} | 3,617 |
35541092 | PMC9083089 | pmc | 4,583 | {
"abstract": "In this work, inspired by some typical creatures from nature with superhydrophobic surfaces, a bio-inspired antifogging PDMS is designed and fabricated successfully using UV lithography and a template method. First, we fabricated an SU-8 layer with a bio-inspired micro-pillared array (MPA) using traditional UV lithography. Then, it was used as a template to fabricate a PDMS film (PF). After that, it was chemically modified with SiO 2 coatings. It was found that the PF coupled with sprayed SiO 2 coatings and a MPA have a higher water contact angle (CA) of 158° and a lower contact angle hysteresis (CAH) of less than 2°. Water drops can be separated from this bio-inspired PDMS surface within 86.8 ms. More importantly, this film’s antifogging property is superior, with a recovery time of less than 13 s, which is significantly superior to that of the flat PF and the PF with the MPA. Afterwards, FTIR was applied to analyse the surface chemistry features and suggested that the bio-inspired PF has extremely low surface tension. So, it can be confirmed that an excellent superhydrophobic antifogging property has been achieved on the surface of the PF. Meanwhile, the microscopic and macroscopic dynamic movement behaviour of the fog drops was further observed. Then, the underlying antifogging mechanism was also revealed. These properties mainly benefit from the coupling effect of intermolecular attraction of droplets, chemical compositions (nanometre roughness SiO 2 ) and the physical structures (MPA). The investigations offer a promising way to handily design and fabricate multiscale hierarchical structures on polymers and other materials. More importantly, these findings suggest great potential value for specific antifogging applications in display devices, transport, agricultural greenhouses, food packaging and solar products, especially in continuous harsh fogging conditions.",
"conclusion": "3. Conclusions In summary, an antifogging PDMS, inspired by some typical creatures from nature with superhydrophobic surfaces, was designed and fabricated successfully via traditional UV lithography combined with a soft replication and subsequent spray coating technique. First, the dimensional uniformity and quality of the bio-inspired PF was characterized using FESEM. It was confirmed that this bio-inspired PF had a coupling surface structure integrated MPA and functionalized SiO 2 coating. FTIR results indicated that both the functional groups of –CH 2 and –CH 3 existed on the surfaces of the bio-inspired PF samples. It not only increased its hydrophobicity dramatically, but also decreased its water adhesion property, ensuring the PF surface achieved superhydrophobic antifogging performance. Meanwhile, a set of optimized models were generated to illustrate the fabrication process. Moreover, the final antifogging behaviors were also revealed. The antifogging properties of PF coupled with sprayed SiO 2 coating and the MPA were characterized experimentally using a spray simulation system and optical CA measuring devices. The time-lapse transmittance measurements demonstrated that the bio-inspired PF also possessed superior fogging recovery property (less than 13 s). It confirmed the reliable optical performance of this advanced antifogging material in practical outdoor conditions. Furthermore, dynamic antifogging behaviors of the bio-inspired PF coupled with sprayed SiO 2 coatings and the MPA were also observed carefully. It can be confirmed that the bio-inspired PF possesses excellent superhydrophobicity and antifogging behaviours. It is anticipated that the findings reported here provide direct guidance for the future design of superhydrophobic antifogging materials on polymers and other material substrates, and suggest great potential value for specific antifogging applications, such as solar cell panels and window buildings.",
"introduction": "1. Introduction Fog formation and accumulation on the surfaces of equipment, such as eyeglasses, windshields, goggles, lenses and display devices in analytical and medical instruments, are known to cause serious economic and safety problems. 1–3 The fundamental principle of antifogging materials is to regulate the interaction between water drops and the solid surface via surface chemical composition as well as the rough features’ size and geometry to ensure appropriate wettability. 4,5 Antifogging surfaces with hydrophilic or even superhydrophilic wetting behaviour have drawn wide attention due to their ability to significantly reduce light scattering by only allowing fog droplets to condensate in a film-like form. 6–9 However, under harsh fogging conditions, these surfaces may exhibit frost formation or excess and inhomogeneous water condensation, which would cause irreversible catastrophic results, such as ceasing the operation, impairing the efficiency or even paralyzing the entire system, especially when considering applications in aircrafts, wind turbines, high-voltage power transmission, telecommunications equipment and heat exchangers. 1 Superhydrophobic-induced antifogging behaviour not only can improve the evaporation rate of fog because of its high CA to the tiny water droplets, 10 but can also induce tiny condensed droplets to merge with each other easily and then shed from the surface. This prevents moisture or microscale fog droplets from nucleating on a surface and so that the surface remains dry. 11–13 Bio-inspired micro-/nanopatterned structures combined with a variety of material substrates can improve the water repellency performance, even leading to the enhancement of the antifogging ability. 14–17 Compared with these materials, polymer materials have lots of peculiar attributes, such as low cost, good deformability and ease of fabrication and so they have broader application prospects. 18,19 Polydimethylsiloxane (PDMS) is typical example of these materials. 20,21 PDMS is inherently water repellent and one of the frequently-used surface modifiers to create superhydrophobic surfaces. 22 Introducing different surface textures such as microwell arrays, 23 femtosecond parallel arrays 24 or other 3D pattern dependent structures 25 into PDMS surfaces can create some chemistry/topography-combined superhydrophobic surfaces. However, a plausible issue has recently arisen. Superhydrophobicity is not the only criteria for generating high-performance antifogging or even anti-ice surfaces. 26 Besides, most of the methods are not scalable for industrial level. In fact, PDMS microscale pillar arrays can achieve higher CAs over 150° without further coatings or treatment steps, 27,28 but their antifogging ability is not obvious because of the good adhesion to droplets of PDMS itself. Some investigations on antifogging function involving PDMS have been reported, for example, the involved PDMS layer was treated with O 2 plasma to convert into highly porous silica films, 29 or as a “seed layer” by photochemical oxidation, 30 the common purpose was to result in a superhydrophilic antifogging layer. In addition, Zheng and co-workers 11 first designed a composite micro/nanostructure surface using a polyvinylidene difluoride polymer as the substrate, showing excellent antifogging and icing-delay properties. Next they presented a series of surfaces combined with nanohairs and micropillar arrays using PDMS as a negative replica and then epoxy as the substrate, demonstrating the excellent anti-icing abilities of the surface. 15 This provides the inspiration for our work as, to the best of our knowledge, the antifogging performance using PDMS directly as the substrate has rarely been characterized in detail. Herein, we designed the surface asperities to take the form of a regular micro-pillared array (MPA) using a PDMS film (PF) in combination with a silicon dioxide (SiO 2 ) modification. In this fabrication, an SU-8 mold of negative well arrays was obtained using traditional UV lithography, then a soft replication method was adopted to obtain a PF with a MPA as the substrate. Subsequent SiO 2 nanometer coatings were sprayed on the surface of the PF with the MPA. Depositing a layer of SiO 2 on the surface of the PF with the MPA using a spray coating technique makes PDMS with a superhydrophobic antifogging property. The dimensional uniformity and quality of the as-prepared PF was characterized with the help of field emission scanning electronic microscopy (FESEM). Fourier transform infrared spectroscopy (FTIR) results indicated that both the functional groups of –CH 2 and –CH 3 existed on the surfaces of the bio-inspired PF samples, which not only increases their hydrophobicity dramatically, but also decreases their water adhesion performance. So, the superhydrophobic antifogging performance of the bio-inspired PF is ensured. Meanwhile, a set of optimized models were generated to illustrate the fabrication process. Moreover, the final antifogging behaviors were also revealed. The antifogging properties of the PF were characterized experimentally using a spray simulation system and an optical CA measuring device. The time-lapse transmittance measurements demonstrated that the as-prepared PF possessed a superior fogging recovery property because it can reach a plateau in far less time (<13 s). It also suggested a reliable optical performance in practical outdoor conditions. Furthermore, the dynamic antifogging behaviors of the PF coupled with SiO 2 coatings and the MPA were observed carefully, verifying that the MPA can get dry and the fog drops can drop from the as-modified PF. It was confirmed that PF coupled with sprayed SiO 2 coatings and the MPA possesses excellent superhydrophobicity and antifogging behaviors.",
"discussion": "2. Results and discussion In this work, we designed a matrix of geometries made of periodic structures on a PDMS substrate. The overall process of preparation is shown in Fig. 1 . Briefly, a clean glass slide was spin-coated with a negative photoresist SU-8, the spinning speed determined the thickness of the SU-8 coating; a photomask containing the circle-shaped arrays was utilized in traditional UV lithography; the unexposed SU-8 was flushed off in the developer, leaving the circular micro-hole arrays standing on the glass slide; a mixture of the PDMS pre-polymer and curing agent in a 10 : 1 mixture (by weight) was degassed using a vacuum chamber and carefully poured onto the SU-8 masters; after curing at 80 °C for 1 h, the PDMS sample was gently peeled from the mold; the commercial SiO 2 coating agent was sprayed on top of the as-prepared PDMS. As illustrated in Fig. 2a , the side length of the pillar ( L ) is about 5 μm, the pitch between neighboring pillars ( P ) is about 7.5 μm and the nominal height of the pillar ( H ) is about 6 μm. The quality and uniformity of the pillars were inspected using FESEM. The FESEM sample was titled at 45° to reveal the actual structures. Fig. 2b shows the PDMS coating has a negligible effect on the global structure of the MPA, though the individual micro-pillar was coated with nanoscale SiO 2 ( Fig. 2c ). The FESEM results indicate that the as-prepared PF possessed a rough surface and contained many bumps. Fig. 1 The fabrication process of the bio-inspired PDMS coupled with sprayed SiO 2 coatings and the MPA. (a) A clean glass slide was spin-coated with a negative photoresist SU-8. (b) A photomask containing the circle-shaped arrays was utilized during the process of UV lithography. (c) With the circular micro-hole arrays (CMHAs) standing on the glass slide, PDMS was carefully poured onto the SU-8 masters and then gently peeled from the mold. (d) The commercial SiO 2 coating agent was sprayed on top of the MPA surface. Fig. 2 (a) The top view (left) and side view (right) of the MPA. This pattern was repeated periodically across the PDMS surface. (b) The FESEM image of the fabricated SiO 2 -sprayed MPA under low-magnification, demonstrating that the coated PDMS has a negligible effect on the global structure of the MPA. (c) The X-ray diffraction (XRD) spectrum of the coating and high-magnification FESEM image shows the nanoscale SiO 2 (insert). The wettability properties of the flat PF, PF with the MPA and PF coupled with sprayed SiO 2 coatings and the MPA were examined separately via an optical CA measuring device based on a sessile drop technique. An average of five measurements on each sample is effective. Fig. 3 shows the water CA of the three PFs was linearly increased and the hydrophobicity of flat PF and PF with the MPA was compared, exhibiting water CAs of 114° and 133°, respectively. The PF coupled with sprayed SiO 2 coatings and the MPA was found to be superhydrophobic, showing an average water CA of 158°. This indicated that the change in surface structure results from the MPA significantly increasing the water CA, rendering the surface superhydrophobic. The superhydrophobic property was elucidated using the Cassie–Baxter model, 31 suggesting that the microstructure on a low-surface-energy material significantly improved the superhydrophobicity of the surface. Fig. 3 CA and CAH measurements of three PF surfaces. The error bars denote standard deviations, which were obtained from distinct measurements on the three different PFs and at least at five different locations on each. The effective CA of the droplets can be regulated by the equation cos θ c = f s cos θ s + f v cos θ v , where θ s and θ v are the CA of the liquid contacting with solid and vapor parts and f s and f v are the area fractions of the solid and vapor on the surface. If rough structures on a surface can generate entrapped air pockets, in such circumstance, f s + f v = 1 and θ v = 180°. θ c can be calculated by the following equation, 1 cos θ c = f s (cos θ s + 1) − 1 In addition, a superhydrophobic surface with low adhesion to droplets is a crucial index of antifogging materials. It is found that the water adhesion on both the flat PF and the PF with the MPA was high. Therefore, our starting surface was hydrophobic with high adhesion. Amazingly, the PF coupled with sprayed SiO 2 coatings and the MPA had a low CAH of less than 2°. From a thermodynamic point of view, eliminating a liquid from its solid substrate requires the energy to overcome the adhesion. 32 The basic relation between the work of adhesion and surface wettability is given by the Dupre–Yong equation, 33 2 W e = γ lv (1 + cos θ c ) where W e is the work of adhesion at the equilibrium state and γ lv is the surface tension of the liquid–vapor interfaces. At a large static CA, it requires a small amount of work to remove droplets. When the SiO 2 coatings were applied on the top of the MPA using the spraying technology, their morphology changed dramatically as shown in Fig. 2b . Specifically, the pillars retained their microscale geometrical characteristics and also exhibited nanometer roughness due to the presence of the SiO 2 particles. Moreover, the increase in the CA of the SiO 2 -sprayed MPA was followed by a noticeable decrease in the water adhesion. Since the water drops roll off easily on the patterned surface, we can assume that the water drops stay on the top of the MPA without penetrating the gap between the neighboring pillars. Thus, by simply spraying SiO 2 on the PF with the MPA, we created “non-sticky” superhydrophobic surfaces (see ESI Video S1 † ). The very low water adhesion is due to the SiO 2 coatings in combination with the geometrical features of the rough surface, which further verified the superhydrophobic antifogging effect of both adding SiO 2 and the creation of the MPA on the PDMS surface. In summary, on one hand, a foundation of hydrophobic SiO 2 nanometer coatings can achieve the superhydrophobic PF chemically. On the other hand, the MPA further amplified the hydrophobic effect to realize the superhydrophobic effect (CA = 158° and CAH = 2°) physically, which played a crucial role in achieving the antifogging property. The water droplet bounce behaviors on the as-prepared PF were recorded with the help of a high-speed video camera when a 15.6 μL water droplet was dropped from a height of 54.8 mm (see ESI Video S2 † ). The dropping height was determined by the maximum height avoiding droplet fragmentation upon impact with the surface, ensuring maximum droplet momentum. This droplet volume (≈16 μL) was found to be optimum as the droplet could be replicated easily and fell under its own weight when dropped from a 23 gauge dispensing tip. As shown in Fig. 4 , the water droplet deformed quickly after contact with the as-prepared PF. The initial impacting velocity of the droplet was 0.94 ms −1 . At 3.1 ms, the spherical water droplet reached a disc-like form. Then, the water drop began to bounce twice and finally completely separated from surface of the as-prepared PF within 86.8 ms, which further illustrates the excellent superhydrophobic and low adhesion properties. Fig. 4 Bounce dynamics of a water droplet impacting with the PF coupled with sprayed SiO 2 coatings and the MPA surface. Furthermore, since fogging results in a certain degree of transmittance loss, we quantified the response of the flat PF, PF with the MPA and PF coupled with sprayed SiO 2 coatings and the MPA to fogging at regular intervals ( T t ) until the original transmittance ( T max ) was restored. In order to characterize the antifogging recovery property, the variation trends of the time-lapse transmittance measurements of the three PFs were performed after being sprayed by the generated fog. For this purpose, we built a spray simulation system to characterize their antifogging properties. 34 The PF was fixed by a clip which was adjusted to be perpendicular to the light beam. As shown in Fig. 5a , recovery from fogging was much faster for the PF coupled with sprayed SiO 2 coatings and the MPA, with T max / T t reaching a plateau in far less time than the others. It was confirmed that the PF coupled with sprayed SiO 2 coating and the MPA possesses a superior ability for antifogging recovery, especially in wet and humid environments. 35 Fig. 5 (a) Antifogging of the three PFs is quantified by performing time-lapse transmittance measurements ( T t ) to determine the required time to restore their original optical properties ( T max ). The PF coupled with sprayed SiO 2 coatings and the MPA shows a significantly faster recovery from fogging. (b) Energy-dispersive X-ray spectroscopy (EDS) spectra of the PF coupled sprayed SiO 2 coatings and the MPA. (c) FTIR spectra of the commercial SiO 2 coating agent. (d) FTIR spectra of the PF with the MPA (black) and the PF coupled with sprayed SiO 2 coatings and the MPA (purple). In order to clarify the reasons that the PF coupled with sprayed SiO 2 coating and the MPA possessed the superhydrophobic-antifogging function we investigated the chemical composition of the as-prepared PF surface. Fig. 5b shows the EDS spectra of the as-prepared PF and the results indicated that the fabricated PF is composed of three elements, carbon (C), silicon (Si) and oxygen (O). FTIR spectra of the commercial SiO 2 coating agent, the PF with the MPA and the PF coupled with sprayed SiO 2 coatings and the MPA were obtained. The FTIR spectrum of the commercial SiO 2 coating agent is shown in Fig. 5c . The peak at 810 cm −1 is due to Si–O–Si symmetric stretching, and the Si–O–Si asymmetric vibration is at 1082 cm −1 . The peaks at 1264 and 2960 cm −1 correspond to the symmetric bending vibration of Si–CH 3 and symmetric stretching vibration of Si–CH 3 , respectively. The peaks at 850 and 1405 cm −1 are due to the Si–C bending and Si–C stretching vibrations, respectively. The Si–CH 2 stretching band is at 2850 cm −1 and the peaks at 1389 and 1460 cm −1 are due to the bending vibration of Si–CH 2 . 36 The FTIR spectrum of the PF with the MPA and the PF coupled with sprayed SiO 2 coatings and the MPA is shown in Fig. 5d . The peak at 1089 is attributed to the Si–O–Si stretching vibration. The peaks at around 1263 and 803 cm −1 are assigned to the Si–C groups. Other characteristic peaks in the spectrum are assigned to the –CH, –CH 2 and –CH 3 groups of the polymer backbone (2964 cm −1 , 2904 cm −1 and 1415 cm −1 , respectively). It could be observed that the FTIR spectrum of the modified PF is consistent with the FTIR spectrum of the untreated PF. 37 The above results not only suggest the commercial SiO 2 coating agent without any impurities, but also the as-prepared PF, was enriched with extreme superhydrophobicity and low adhesion, due to the hydrophobic functional groups (–CH 2 and –CH 3 ) of the SiO 2 coating. To examine the antifogging property more intuitively, a 3D ultra-depth stereoscopic microscope was used to observe the micro-dynamic behaviour of fog drops on the PF coupled with sprayed SiO 2 coatings and the MPA surface (see ESI Video S3 † ). First, many individual fog drops with a spherical appearance were occurring on the top of the micropillars. As time went on, we found some tiny fog drops began to merge with each other and form new fog drops. As shown in Fig. 6 , fog drops A–E were growing smoothly ( t = 20 s). Subsequently, fog drops B and C coalesced into larger fog drop F ( t = 30 s), fog drops D and E coalesced into larger fog drop H ( t = 45 s) and fog drops A and F coalesced into larger droplet I ( t = 55 s). Then, fog drops H and I merged with fog drop G and formed fog drop J ( t = 90 s). This indicated that the condensed fog drops can be in Cassie’s state. We theorised that a de-wetting transition phenomenon may occur on the surface, 38 then the released surface energy can propel the fog drop jumping or self-removal from the surface. 39,40 Furthermore, when this flying fog drop touched another constrained fog drop, a transition would again be stimulated. Due to the occurrence of the de-wetting transitions and the self-removal phenomenon, the MPA can get dry and the fog drops can drop from the PF by means of the low surface adhesion. Fig. 6 (a) Optical images show the micro-dynamic behaviour of the fog drops movement on the PF coupled with sprayed SiO 2 coatings and the MPA surface. From t = 0 s to t = 90 s, as the fog drops (A–E) grow larger gradually ( t = 20 s), they will merge with each other, fog drops B and C coalesce into larger fog drop F ( t = 30 s), fog drops D and E coalesce into larger fog drop H ( t = 45 s) and fog drops A and F coalesce into larger droplet I ( t = 55 s). Then, fog drops H and I merge with fog drop G and form fog drop J ( t = 90 s). (b) Time evolution of the diameter of an individual fog drop during the merge process (blue triangles). The inserts correspond to t = 20, 30, 45, 55 and 90 s. (c) Additional details are displayed with the assistance of schematic diagrams. In order to verify our hypothesis about antifogging behaviour on the PF coupled with sprayed SiO 2 coatings and the MPA surface, the macro-dynamic process of fog drop movement was recorded using a 3D ultra-depth stereoscopic microscope (see ESI Video S4 † ). As shown in Fig. 7a , the fog spray was applied to the surfaces, some tiny fog drops initially condensed on the surface and subsequently the fog drops became gradually larger. As time went on, we amazingly found the same phenomenon as Fig. 6 , that some tiny fog drops began to merge with each other and form larger fog drops (see the same color circles from t = 90 s to t = 196 s). In addition, when the fog drops grew to a certain size, the same phenomenon occurred as in ESI Video S5, † that these fog drops began to roll off suddenly. It is interesting that this roll-off performance removed some circumjacent tiny fog drops, sweeping the surface clean and keeping the area dry (see purple circle at t = 236 s). This is consistent with our previous conjecture and demonstrated the excellent antifogging property of the as-prepared PF surface. To evaluate the antifogging ability of the PF coupled with sprayed SiO 2 coatings and the MPA surface, we estimated the percentage of dry areas versus time, as shown in Fig. 7b , at ∼50 s, the percentage was lower, but the percentage increased suddenly after ∼200 s, and the percentage was maintained at ∼83% from ∼250 to 500 s on the whole. This demonstrated the superior antifogging property of the PF coupled with sprayed SiO 2 coatings and the MPA surface. Fig. 7 (a) Optical images show the macro-dynamic process of the fog drops movement on the PF coupled with sprayed SiO 2 coatings and the MPA surface. From t = 90 s to t = 236 s, with water condensation, some tiny fog drops began to merge with each other and form larger fog drops (the same color circles from t = 90 s to t = 196 s). As the fog drops reach a certain size, the fog drops begin to roll off suddenly and take away some surrounding tiny fog drops (purple circle at t = 236 s). (b) The percentage of dry areas versus time. At ∼50 s, the percentage is lower, but the percentage increases suddenly after ∼200 s, and the percentage was maintained at ∼83% from ∼250 to 500 s on the whole. In order to further reveal the internal antifogging mechanism of the PF coupled with sprayed SiO 2 coating and the MPA, one possible reasonable explanation for these findings was that the synergistic effect of the droplets intermolecular attraction, chemical compositions and the MPA was the key factor in realizing the superhydrophobic antifogging property. On one hand, the low surface energy methylated (–CH 3 and –CH 2 ) components resulted from the SiO 2 nanometer coatings, which further increased the PF surface hydrophobicity and dramatically decreased the PF surface adhesion to water droplets. Indeed, the sliding angle for the water drops occurred for a CAH of less than 2°. Since the water drops rolled off easily on the PF surface, we can assume that the water drops stayed on the top of the MPA without penetrating the interpillar areas, and then induced tiny condensed droplets to merge with each other easily until shed from the surface. Moreover, the evaporation rate of fog drops was also improved. 41 Physically, the fog drops would remain repulsive to the MPA due to the surface forces having sufficient magnitude to suspend liquid against the downward pull of gravity (or other body forces) (see ESI Videos S6 and S7 † ). As a whole, the water molecules were affected by a repulsion force of the material itself ( F 1 ), intermolecular attraction ( F 2 ), a repulsion force of micro-pillared arrays ( F 3 ), surface tension ( σ ) and their own gravity ( G ) ( Fig. 8 ). Fig. 8 The antifogging behaviors of the PF coupled with sprayed SiO 2 coating and the MPA. Here, F 1 is the repulsion force of the material itself, F 2 is the intermolecular attraction, F 3 is the repulsion force of the micro-pillared arrays, σ is surface tension and G is gravity. On the other hand, it is a universal strategy to construct a superhydrophobic surface by creating surface roughness onto a low surface energy material. Interestingly, the SiO 2 -sprayed MPA played a significant role in amplifying the PF’s intrinsic hydrophobicity, which dramatically increased the surface roughness. Specifically, the pillars retained their microscale geometrical characteristics and also exhibited nanometer roughness due to the presence of the SiO 2 coating. The combination of the micro/nano-roughness as well as the well-known water-repellent chemical properties of the PDMS made the patterned surfaces superhydrophobic. Previous researchers have reported that the relationships between CA and roughness ratio in two different wettability states were quantitatively described as the Wenzel model and the Cassie–Baxter model 42 which indicated that the true CA of a flat hydrophobic surface would be lower with the increase in surface asperities. The very low water adhesion was due to the inherent property of the SiO 2 coatings in combination with the geometrical features of the MPA. Apparently, it was the hierarchical amplification effect of SiO 2 -sprayed MPA that brought big rewards for the achievement of the superhydrophobic antifogging surface. In addition, with the increase in the hydrophobic specific surface that arose from the SiO 2 -sprayed MPA, the PF surface free energy was obviously reduced. According to energy minimization theory, 43 once a droplet coalesces with the adjacent droplets, the released surface energy will overcome droplet adhesion, which may induce the fog drops from a Wenzel state to a Cassie–Baxter state, 31 so that coalescing fog droplets can self-remove from PF surfaces. Consequently, the transmittance of the PF coupled with sprayed SiO 2 coatings and the MPA would recover to the initial state. These were exactly consistent with the results of the transmittance spectra."
} | 7,225 |
32733397 | PMC7360803 | pmc | 4,584 | {
"abstract": "Methane seeps are widespread seafloor ecosystems shaped by complex physicochemical-biological interactions over geological timescales, and seep microbiomes play a vital role in global biogeochemical cycling of key elements on Earth. However, the mechanisms underlying the coexistence of methane-cycling microbial communities remain largely elusive. Here, high-resolution sediment incubation experiments revealed a cryptic methane cycle in the South China Sea (SCS) methane seep ecosystem, showing the coexistence of sulfate (SO 4 2– )- or iron (Fe)-dependent anaerobic oxidation of methane (AOM) and methylotrophic methanogenesis. This previously unrecognized methane cycling is not discernible from geochemical profiles due to high net methane consumption. High-throughput sequencing and Catalyzed Reporter Deposition-Fluorescence in situ Hybridization (CARD-FISH) results suggested that anaerobic methane-oxidizing archaea (ANME)-2 and -3 coupled to sulfate-reducing bacteria (SRB) carried out SO 4 2– -AOM, and alternative ANME-2 and -3 solely or coupled to iron-reducing bacteria (IRB) might participate in Fe-AOM in sulfate-depleted environments. This finding suggested that ANME could alter AOM metabolic pathways according to geochemical changes. Furthermore, the majority of methylotrophic methanogens belonged to Methanimicrococcus , and hydrogenotrophic and acetoclastic methanogens were likely inhibited by sulfate or iron respiration. Fe-AOM and methylotrophic methanogenesis are overlooked potential sources and sinks of methane in methane seep ecosystems, thus influencing methane budgets and even the global carbon budget in the ocean.",
"introduction": "Introduction Methane seeps are methane-dependent chemosynthetic ecosystems ( Paull et al., 1984 ) that occur widely in the marine environment, and are considered some of the richest benthic ecosystems on the seabed ( Valentine, 2011 ). The microbially mediated methane cycle dominates methane seeps ( Boetius et al., 2000 ; Knittel and Boetius, 2009 ) and has an important impact on the global carbon cycle ( Boetius and Wenzhöfer, 2013 ; Offre et al., 2013 ). Methane is produced from the degradation of organic matter by methanogens in deep sediment ( Reeburgh, 2007 ). The produced methane continuously seeps or erupts from the sedimentary subsurface to the seabed in the form of methane-rich fluids, and more than 90% of the methane is consumed by anaerobic methane-oxidizing archaea ( Valentine, 2011 ), forming an efficient methane biofilter that prevents its diffusion into the seawater. Previous studies believed that in the ocean, the majority of methane is oxidized anaerobically by anaerobic methane-oxidizing archaea (ANME) coupled with sulfate reduction ( Boetius et al., 2000 ). However, various chemical compounds are thermodynamically more favorable electron acceptors than sulfate for catalyzing anaerobic methane oxidation, such as nitrite ( Raghoebarsing et al., 2006 ), nitrate ( Haroon et al., 2013 ), ferric iron, and manganese ( Ettwig et al., 2016 ; Cai et al., 2018 ). For instance, biogeochemical profiling evidences indicated the widespread presence of Fe-anaerobic oxidation of methane (AOM) in the ocean, such as in Argentine Basin ( Riedinger et al., 2014 ), Alaskan Beaufort Sea ( Treude et al., 2014 ), North Sea Helgoland mud ( Oni et al., 2015 ), Baltic Sea ( Egger et al., 2017 ), and Mediterranean Sea ( Vigderovich et al., 2019 ). Targeted enrichment with ferrihydrite provides strong evidence for AOM coupled with iron reduction, and ANME-1, Methanococcoides /ANME-3 ( Beal et al., 2009 ), ANME-2a and -2c ( Scheller et al., 2016 ), ANME-2d ( Methanoperedens nitroreducens ) ( Ettwig et al., 2016 ; Shen et al., 2019 ), Candidatus Methanoperedens ferrireducens ( Cai et al., 2018 ), and Methanosarcina acetivorans ( Yan et al., 2018 ) might be involved in Fe-AOM. However, the Fe-AOM microorganism physiology and its contribution to the methane consumption remain poorly understood in deep sea ( Scheller et al., 2016 ; He et al., 2018 ). It is still possible that other unknown microorganisms can perform metal-AOM. For instance, Bar-Or et al. (2017) highlighted the essential role of methanogens and methanotrophic bacteria in the process of Fe-AOM. In addition, in methane seep ecosystems, acetoclastic and hydrogenotrophic methanogenesis are often considered the dominant methanogenic processes below the Sulfate-methane transition zone (SMTZ) ( Zhe et al., 2018 ). Thermodynamic laws indicate that sulfate-reducing bacteria (SRB) and iron-reducing bacteria (IRB) outcompete methanogens for hydrogen and acetate substrates, thus inhibiting the acetoclastic/hydrogenotrophic methanogenesis pathways ( Reeburgh, 2007 ; Reiche et al., 2010 ; Zhou et al., 2014 ; Zhuang et al., 2018 ). However, the co-occurrence of sulfate reduction, AOM, and methanogenesis has been demonstrated in marine sediments ( Sivan et al., 2014 ). This is also partly supported by the coexistence of ANME and methanogenic archaea in Sonora Margin shallow sediments (0–20 cm) ( Vigneron et al., 2015 ), Peruvian Margin sulfate-reducing zone (0–25 cm) ( Maltby et al., 2016 ), Mediterranean Sea shallow sediment (0–20 cm) ( Sela-Adler et al., 2017 ), and Aarhus Bay surface sediment ( Xiao et al., 2017 , 2018 ). Then, it is becoming clear that methylotrophic methanogenesis plays a major role in marine sediments, especially surficial sediments ( Zhuang et al., 2016 , 2018 ; Xiao et al., 2017 ). Methylotrophic methanogenesis using non-competitive substrates such as (methanol, methylamines, or methyl sulfide and so on) could co-occur with AOM and sulfate reduction ( Vigneron et al., 2015 ; Zhuang et al., 2016 ). In the presence of sulfate and iron oxides, methanogens may circumvent competition by utilizing ‘non-competitive’ methylated substrates. These non-competitive substrates are ubiquitous in the marine environment, and originate from the degradation of substances such as betaine, choline, lignins, pectin, and creatine, or from the bacterial reduction of trimethylamine oxide ( Oremland et al., 1982 ). Hence, methylotrophic methanogenesis has been suggested to occur in all major oceans ( Valentine, 2011 ; Chronopoulou et al., 2017 ). However, in methane seep ecosystems, this process is not easily discernible from geochemical profiles due to the overall high net methane consumption in methane seeps. Thus, research on methylotrophic methanogenesis in methane seep is relatively rare, and the contribution of methylated substrates to methanogenesis processes remains elusive ( Chronopoulou et al., 2017 ). The northern continental slope of the South China Sea (SCS) has various sites of methane seepage, covering a wide range of water depths (200–3000 m). The Jiaolong and Haima methane seeps are the two still-active seep sites in SCS. Previous studies on the seeps in SCS have led to many significant advancements including new insight into the methane seep structure ( Liu et al., 2018 ), mineral carbon/oxygen/sulfur isotopes ( Feng et al., 2018 ), macrobenthos ( Dong and Li, 2015 ; Gong et al., 2015 ), and microbial lipid biomarkers ( Guan et al., 2016 , 2018 ). In addition, the microbial distribution and diversity in SCS methane seeps were also investigated ( Jiang et al., 2007 ; Zhang et al., 2012 ), and a 16S rRNA gene-based survey indicated the presence of SRB and ANME. However, the biogeochemical evidences and potential activity of methane-cycling microorganisms and their niche differentiation patterns are largely unknown. In this study, the Jiaolong methane seep was chosen as our research object. In 2018, a remotely operated underwater vehicle, The Remotely Operated Platform for Ocean Science (ROPOS), found a new methane seepage site in this area. Combining high-throughput sequencing, CARD-FISH, and enrichment culture methods with pore water biogeochemistry, we investigated the microbially driven metabolic processes of methane production and consumption. The coexistence of sulfate- or iron-dependent AOM and methylotrophic methanogenesis was found, and the potential rates of these processes were assessed. These results are of great importance to the understanding of the biogeochemical processes in global methane seep ecosystems.",
"discussion": "Discussion Niches and Diversity of Microbes in the Jiaolong Methane Seep We compared the archaeal and bacterial diversity of the Jiaolong methane seep with those of 23 globally distributed methane seeps, and found that the microbial richness of the Jiaolong methane seep was at a medium level ( Supplementary Figure S3 ). However, microbial abundance data showed that the Jiaolong methane seep hosted higher biomass (5.8 × 10 6 –2.0 × 10 8 cells g –1 for archaea,1.1 × 10 6 –2.0 × 10 9 cells g –1 for bacteria) than other methane seeps in the SCS, such as the Haima methane seep (2.8 × 10 4 –3.4 × 10 6 cells g –1 for archaea, 4.5 × 10 5 –7.4 × 10 6 cells g –1 for bacteria) ( Niu et al., 2017 ), the GMGS2 gas hydrate station (1.3 × 10 4 –2.7 × 10 6 cells g –1 for archaea, 3.8 × 10 4 –1.0 × 10 7 cells g –1 for bacteria) ( Cui et al., 2019 ), and the Haiyang 4 hydrate station (with total cells of 10 5 –10 6 cells g –1 ) (the microbial abundance was converted from 16S rDNA qPCR data) ( Zhang et al., 2012 ). It was found that in the Jiaolong methane seep, the main microbes were SEEP-SRB1 coupled to ANME-2 and ANME-3, and a few SEEP-SRB4 coupled to ANME-2 and ANME-3. No ANME-1, SEEP-SRB2, or SEEP-SRB3 were observed in the sediments. In the surface layer, the dominant sequence cluster belonged mainly to Sulfurovum , Methyloprofundus , Sulfurimonas, Nitrosopumilus , and ANME-2a/b. In contrast, in the deepest sediments, the dominant sequence cluster belonged mainly to ANME-2, ANME-3, and Methanimicrococcus . The microbial diversity obtained in this study obviously differed from those in previous studies at methane seeps in the northern SCS. In June 2013, researchers found that in the Jiaolong methane seep, the majority of the microbial inhabitants at the surface layers (0–6 cm) were Sulfurimonas, Sulfurovum , and ANME-1, while SEEP-SRB1, ANME-1, and ANME-2 dominated the deepest layers (8–14 cm). The percentage of ANME-3 was the lowest in all layers, and SEEP-SRB3 and SEEP-SRB4 were not detected ( Wu et al., 2018 ). At the Haiyang 4 hydrate station, only ANME-1 was detected, and the bacterial groups Chloroflexi and JS1 (Atribacteria) were dominant ( Zhang et al., 2012 ). In the GMGS2 gas hydrate station, the majority groups were ANME-1b, ANME-2c, and bacterial group Desulfobacteraceae ( Cui et al., 2019 ). In Haima active methane seep ecosystems in the southwestern SCS, ANME-2a/b was predominant in the upper and middle layers of the SMTZ, whereas ANME-1b outnumbered ANME-2 below the SMTZ, and ANME-3 was absent ( Niu et al., 2017 ). Methane seeps are island-like habitats, harboring distinct microbial communities ( Ruff et al., 2015 ). The seep communities comprise bacteria and archaea that occur worldwide but are locally selected by the environment. For example, in situ temperature, methane concentration ( Knittel et al., 2005 ), oxygen concentration ( Meulepas et al., 2009 ), and sulfate concentration ( Yanagawa et al., 2011 ) can significantly affect the ANME species. A total of >1800 ANME sequences have been reported by far including three types of ANME (ANME-1, -2, and -3) across physiochemically contrasting ecological niches. Among them, ANME-1 and ANME-2 are the most widely distributed in the world and tend to be coupled with syntrophic SEEP-SRB1 bacteria. ANME-1 preferentially grows in hydrogen sulfide-rich and sulfate-depleted environments, while ANME-2 is closely associated with sulfate concentration ( Yanagawa et al., 2011 ) and preferentially grows in sulfate-rich areas. ANME-3 is distributed mainly in methane-seeping mud volcanoes and in some methane seeps ( Niemann et al., 2006 ). Furthermore, in marine sediments, an ecological niche separation occurs where ANME-2a/b dominates the upper layers and ANME-2c and/or ANME-1 outcompetes in deeper zones ( Timmers et al., 2017 ). Our results showed that geochemistry could be the primary force shaping the niche differentiation of functional microbial populations associated with methane-cycling in marine environment. Iron-Mediated Anaerobic Oxidation of Methane The results of incubation experiments ( Table 2 ) showed both SO 4 2– -AOM and Fe-AOM occurred in Jiaolong methane seep. Fe-AOM activity appeared below SMTZ at a depth of 8–10 cm, and potential net rates ranged from 20.40 to 37.99 nmol cm –3 day –1 . Compared with net SO 4 2– -AOM rates ranging from 10.39 to 900.20 nmol cm –3 day –1 , net Fe-AOM rates were 10 times lower. Our study is the first to discover Fe-AOM in the deep sea methane seeps of the SCS, and potential net rates are higher than those in freshwater and coastal sediments. The potential rate of Fe-AOM was 16.44 nmol cm –3 day –1 in Eel River Basin ( Beal et al., 2009 ), 3.61 nmol cm –3 day –1 in brackish coastal sediments ( Egger et al., 2015 ), 3.89 nmol cm –3 day –1 in coastal Georgia ( Segarra et al., 2013 ), and 3.45 nmol cm –3 day –1 in deep lake sediment cores ( Sivan et al., 2011 ). The higher rate of Fe-AOM in this study may result from the sufficient supply of iron oxide and massive methane flux. The slope of the northern SCS is one of the world’s most active areas of modern marine sedimentary processes ( Huang and Wang, 2007 ; Luan et al., 2019 ). A large amount of river-borne terrigenous sediment input leads to exceptionally high amount of iron oxides in the sediments of the northern SCS ( Zhang et al., 2007 ; Liu Z. F. et al., 2016 ; Liu et al., 2018 ). Previous research had suggested that different kinds of iron oxides could serve as electron acceptors for Fe-AOM ( Bar-Or et al., 2017 ). In addition, a large amount of unconsumed methane was released into the bottom seawater (22.23 μM, unpublished data), implying a high methane flux in this area. Thus, the combination of these factors probably stimulated the enhanced rate of Fe-AOM processes in the Jiaolong methane seep. To the best of our knowledge, there is no representative pure culture of Fe-AOM microorganisms from the marine sediment ( Liang et al., 2019 ). But some microorganisms were suspected of being related to metal-AOM in various earlier studies, which suggested that ANME-1, ANME-3 ( Beal et al., 2009 ), ANME-2a, 2c ( Scheller et al., 2016 ), ANME-2d ( Methanoperedens nitroreducens ) ( Ettwig et al., 2016 ; Shen et al., 2019 ), Candidatus Methanoperedens ferrireducens ( Cai et al., 2018 ), or Methanosarcina acetivorans ( Yan et al., 2018 ) might be involved in Fe-AOM. Furthermore, it is still possible that other unknown microorganisms perform metal-AOM. Bar-Or et al. (2017) findings highlight the essential role and participation of methanogens archaea and methanotrophic bacteria in the process of Fe-AOM. In our study, sequencing data showed that the methanotrophic bacteria (such as Candidatus Methylomirabilis oxyfera , Methylobacter, Methylosarcina, Methylomonas , and Methylococcus ) were not detected. Under the condition that 20 mM BES was added to inhibit methanogens archaea in the incubation experiments, the Fe-AOM activity appeared below SMTZ. These evidences excluded the potential participation of methanogenic archaea and methanotrophic bacteria in the process of Fe-AOM. CARD-FISH data showed that there was no significant change in the number of ANME at the depth of 6–12 cm, while the number of SRB decreased rapidly, and some ANME-2 and ANME-3 were not coupled with SRB ( Figures 6H,I ). These results were similar to those from methane seep enrichment samples of the Eel River Basin and Santa Monica Basin, which contained high abundances of ANME-2a and ANME-3 and could decouple the AOM process from SRB activities when metal compounds were added ( Beal et al., 2009 ; Scheller et al., 2016 ). So we speculated that ANME-2 and ANME-3 were probably involved in Fe-AOM. Different kinds of IRB were also detected in the Jiaolong seep sediments, such as Shewanella (0.01–0.07%), Geobacter (0.004–0.02%), Pseudomonas (0.01–0.1%), and Desulfuromonas (0.11–0.92%). Interestingly, the abundance of IRB was relatively higher in the 12–14 cm depth. To date, two potential ways of Fe-AOM were described in previous studies. (1) ANME oxidizes methane and transfer electrons directly to soluble metal ions or complexes, or solid metal oxides ( Ettwig et al., 2016 ; Scheller et al., 2016 ); (2) ANME should be partnered with metal-reducing microorganisms to perform metal-AOM, in a way similar to the ANME-SRB consortia ( Fu et al., 2016 ; He et al., 2018 ). It is worth further exploring whether ANME-2 or ANME-3 should be alone or coupled with IRB to perform Fe-AOM process in the Jiaolong methane seep. As the next step, we would like to use 14 C-CH 4 to enrich and cultivate methane-oxidizing microbial populations from the samples. Stable-isotope probing of active AOM would likely provide more hints on AOM metabolisms. Furthermore, nitrite/nitrate-dependent AOM activity had not been detected in Jiaolong methane seep. Generally, nitrite-dependent AOM is performed by the NC10 bacteria related to Candidatus Methylomirabilis oxyfera , and nitrate-dependent AOM is performed by the ANME-2d ( Ettwig et al., 2016 ). In our study, no sequences of Candidatus Methylomirabilis oxyfera and ANME-2d were detected. These results suggested that Fe-AOM was probably the dominant non-sulfate AOM pathway in the Jiaolong methane seep. Methylotrophic Methanogenesis Generally, there are three major methanogenic pathways: the acetoclastic, hydrogenotrophic, and methylotrophic pathways ( Zhuang et al., 2018 ). Hydrogenotrophic and acetoclastic methanogenesis are often considered the primary pathways in marine deep sediment ( Zhe et al., 2018 ). Based on thermodynamic laws, SRB and IRB outcompete methanogens for both acetate and hydrogen. Therefore, these methanogens are inhibited during active sulfate reduction and iron reduction ( Reeburgh, 2007 ; Reiche et al., 2010 ; Zhou et al., 2014 ; Zhuang et al., 2018 ). In our study, hydrogenotrophic and acetoclastic methanogenesis were not detected due to active sulfate or iron reducing respiration in the Jiaolong methane seep. However, methylotrophic methanogenesis using non-competitive substrates (methanol) appeared below the 4–6 cm sediment layer, and coexisted with SO 4 2– -AOM and Fe-AOM. Most methanogens clustered with the genus Methanimicrococcus . It is an obligatory methylotrophic methanogen, that is, it utilizes only non-competitive substrates, such as methanol or methylated compounds ( Sprenger et al., 2000 ; Zeng et al., 2007 ; Niemann et al., 2009 ; Wang et al., 2019 ). Non-competitive substrates, such as methanol, trimethylamine, methylamines, dimethylsulfide, and dimethylsulfoniopropionate are ubiquitous in the marine environment ( Oremland et al., 1982 ). To date, research on the methylotrophic methanogenesis process has been intensively investigated on non-seep sediments, such as those in the Peruvian Margin (with methanol as the substrate, 0.6–1.95 nmol cm –3 d –1 ) ( Maltby et al., 2016 ), Aarhus Bay (with methanol or trimethylamine as the substrate, 0.83–1.11 nmol cm –3 d –1 ) ( Xiao et al., 2017 , 2018 ), and the Western Mediterranean Sea (with methanol as the substrate, 0.03 nmol cm –3 d –1 ) ( Zhuang et al., 2018 ). This process is poorly understood in deep sea methane seep ecosystems due to the technical challenges to discern methane production against the overall high background of net methane consumption in methane seeps ecosystem. To the best of our knowledge, there is only one report in the Sonora Margin cold seep showing methane production by Methanococcoides burtonii on non-competitive substrates (with trimethylamine as the substrate, 180∼560 pmol cm –3 day –1 ) ( Vigneron et al., 2015 ) in shallow sediments above the SMTZ. The rate of methylotrophic methanogenesis in the Jiaolong methane seep reached a maximum of 5.85 nmol cm –3 day –1 , which was higher than those observed in other areas. Future study is warranted to elucidate the thermal kinetics underlying this microbially mediated process. In summary, the incubation experiments revealed the coexistence of sulfate-driven AOM, iron-driven AOM, and methylotrophic methanogenesis in Jiaolong methane seep sediments of the northern SCS where terrigenous sediments rich in iron oxide are imported in large quantities. Fe-AOM and methylotrophic methanogenesis are overlooked potential sources and sinks of methane in SCS methane cycle. Globally, large amounts of iron [∼730 Tg/year] from rivers are transported to ocean continental margins ( Jickells et al., 2005 ), and methane seeps are common along continental margins in areas of high primary productivity and tectonic activity. How this methane cycle, which is affected by the large input of iron oxides, will influence the global carbon cycle is worthwhile to study further."
} | 5,274 |
39154166 | PMC11330620 | pmc | 4,585 | {
"abstract": "The cell and molecular bases of arbuscular mycorrhizal (AM) symbiosis, a crucial plant-fungal interaction for nutrient acquisition, have been extensively investigated by coupling traditional RNA sequencing techniques of roots sampled in bulk, with methods to capture subsets of cells such as laser microdissection. These approaches have revealed central regulators of this complex relationship, yet the requisite level of detail to effectively untangle the intricacies of temporal and spatial development remains elusive. The recent adoption of single-cell RNA sequencing (scRNA-seq) techniques in plant research is revolutionizing our ability to dissect the intricate transcriptional profiles of plant-microbe interactions, offering unparalleled insights into the diversity and dynamics of individual cells during symbiosis. The isolation of plant cells is particularly challenging due to the presence of cell walls, leading plant researchers to widely adopt nuclei isolation methods. Despite the increased resolution that single-cell analyses offer, it also comes at the cost of spatial perspective, hence, it is necessary the integration of these approaches with spatial transcriptomics to obtain a comprehensive overview. To date, few single-cell studies on plant-microbe interactions have been published, most of which provide high-resolution cell atlases that will become crucial for fully deciphering symbiotic interactions and addressing future questions. In AM symbiosis research, key processes such as the mutual recognition of partners during arbuscule development within cortical cells, or arbuscule senescence and degeneration, remain poorly understood, and these advancements are expected to shed light on these processes and contribute to a deeper understanding of this plant-fungal interaction.",
"conclusion": "Conclusions Decades of research on AM symbiosis and the efforts of an always growing scientific community have allowed us to disentangle many aspects of the intricate plant-fungal interactions. AM symbiosis is unique for the intimate contact of their partners and for its features which are surprisingly constant notwithstanding the huge biodiversity of the organisms involved [ 111 ]. This stability probably mirrors the antiquity of this alliance, suggesting that plant speciation occurred after AM establishment. The process of AM fungi accompanying most plants during evolution might have caused a strong link between organ development, especially of roots, and AM fungal presence. It is evident that our understanding across multiple fields in biology, including plant research, has deeply advanced since we have gained access to single-cell resolution. Particularly, the ever expanding sequencing technologies could shed some light on the profound cellular heterogeneity inherent in this intimate plant-fungal interaction, offering new insights into the dynamic complexities of symbiotic relationships across evolution and within different ecosystems. The recent publication of the first AM symbiosis study using snRNAseq and ST, in which the authors generated an atlas showing plant and fungal transcriptomes simultaneously throughout their symbiotic relationship (72), is just the starting point in disentangling the molecular bases of this complex process with single-cell resolution. Certainly, these new technologies combined with the emerging ones in the coming years may address the questions that still lack answers and, in turn, promote the generation of new hypotheses. \n Fig. 1 Root colonization by AM fungi. Plants recognize fungal molecules, such as short-chain chitin oligomers (COs) or Myc factors, found in AM fungi exudates; in turn, plant strigolactones (SLs) induce spore germination and hyphal branching. Upon initial contact with the root epidermis, tip-growing hyphae differentiate, forming a structure known as a hyphopodium. Subsequently, the fungus penetrates the root guided by a plant structure called the pre-penetration apparatus (PPA). Once inside the root, the hypha traverses intercellular spaces of the inner cortex root cells and, following PPA formation, enters the cell and initiates the trunk formation. Then, young arbuscules are produced, which expand within the cortical cell through continuous branching until mature, occupying the entire space and engaging in mutualistic nutrient exchanges. These mature functional arbuscules persist for several days before undergoing senescence. Created with BioRender.com \n \n Fig. 2 Bulk versus sc/sn RNA sequencing in mycorrhizal roots. Bulk RNA sequencing offers an average gene expression profile of all the root and fungal cells. On the other hand, single-cell and single-nucleus RNA sequencing offer a distinct transcriptome profile for each individual plant and fungal cell, enabling the grouping of the different cell populations into physiological or functional clusters. Created with BioRender.com \n \n Table 1 An overview of single-cell research studies in plant-microbe interactions ordered by publication time. scRNA-seq = single-cell RNA sequencing, snRNA-seq = single nucleus RNA sequencing, snATAC-seq = assay for transposase-accessible chromatin using single nucleus sequencing, MERFISH = multiplexed error-robust fluorescence in situ hybridization Title Citation Main technique/s Results Organisms Spatial co-transcriptomics reveals discrete stages of the arbuscular mycorrhizal symbiosis [ 76 ] snRNA-seq and spatial transcriptomics Construction of the first spatially-resolved single-cell resolution integrated map of a multi-kingdom symbiotic interaction. First single-cell resolution study of the AM symbiosis. \n Medicago truncatula/ Rhizophagus irregularis \n The single-cell transcriptome program of nodule development cellular lineages in Medicago truncatula [ 60 ] scRNA-seq Description of cell-type-specific transcriptome response of Medicago truncatula roots to rhizobia during early nodule development. Medicago truncatula / Sinorhizobium meliloti Spatial metatranscriptomics resolves host–bacteria–fungi interactomes [ 80 ] Spatial metatranscriptomics Detection of spatial interaction between the plant and its colonizing microorganisms (55 μm of resolution). Study of bacterial and fungal hotspots and the plant’s response. Arabidopsis thaliana / Pseudomonas syringae Single-cell analysis identifies genes facilitating rhizobium infection in Lotus japonicus [ 74 ] scRNA-seq Generation of wild type and mutant L. japonicus seedlings cell atlas after rhizobial infection. Identification of candidate nodulation genes. Lotus japonicus / Mesorhizobium loti Cell-type-specific responses to fungal infection in plants revealed by single-cell transcriptomics [ 112 ] scRNA-seq Generation of Arabidopsis cell atlas during fungal infection. Description of cell-type specific gene expression, like high expression of immune receptors in vasculature cells and coordinated expression changes of abscisic acid in the guard cells. Report of a robust correlation between cell response and its proximity to the invading fungal hyphae. Arabidopsis thaliana / Colletotrichum higginsianum Single-nucleus transcriptomes reveal spatiotemporal symbiotic perception and early response in Medicago [ 75 ] snRNA-seq Time-course analysis of symbiotic perception, focusing on early cell-type specific responses to nodulation. Medicago truncatula / Sinorhizobium meliloti Cell specialization and coordination in Arabidopsis leaves upon pathogenic attack revealed by scRNA-seq [ 77 ] scRNA-seq Analysis of the cellular heterogeneity within an infected leaf. Description of spatial and temporal dynamics of immune and susceptible cell-clusters. Arabidopsis thaliana / Pseudomonas syringae Single-cell profiling of Arabidopsis leaves to Pseudomonas syringae infection [ 78 ] scRNA-seq, transcriptional reporter fusion Analysis of the cellular heterogeneity within an infected leaf. Description of spatial and temporal dynamics of immune and susceptible cell-clusters. Arabidopsis thaliana / Pseudomonas syringae Single-cell RNA sequencing profiles reveal cell type-specific transcriptional regulation networks conditioning fungal invasion in maize roots [ 113 ] scRNA-seq Study of the immune regulatory networks in major cell types of maize root tips in response to fungal infection. Maize ( Zea mays )/ Fusarium verticillioides Time-resolved single-cell and spatial gene regulatory atlas of plants under pathogen attack [ 79 ] snMultiome (snRNA-seq + snATAC-seq) and spatial transcriptomics (MERFISH) Generation of a spatiotemporal cell map of Arabidopsis leaf infected by virulent and avirulent bacteria, involving transcription factors, putative cis-regulatory elements and target genes associated with disease and immunity. Identification of new cell populations involved in immune response. Discovery of immune response heterogeneity according to pathogen distribution. \n Arabidopsis thaliana/ Pseudomonas syringae \n A high-resolution transcriptomic atlas depicting nitrogen fixation and nodule development in soybean [ 72 ] snRNA-seq, transcriptional reporter fusion Generation of cell atlas of soybean roots and determinate nodules. Comparison of determinate and indeterminate nodules. Description of GmCRE1 role in nodule formation and nitrogen fixation. Soybean ( Glycine max )/ Sinorhizobium fredii Integrated single-nucleus and spatial transcriptomics captures transitional states in soybean nodule maturation [ 114 ] snRNA-seq and spatial transcriptomics Generation of cell atlas of soybean roots and determinate nodules. Identification of rare-cell subtypes with important roles in nodule function and development. Soybean ( Glycine max )/ Bradyrhizobium diazoefficiens Single-cell resolution transcriptome atlases of soybean root organs reveal new regulatory programs controlling the nodulation process [ 73 ] snRNA-seq Generation of cell atlas of soybean roots and determinate nodules. Identification of different sub-populations of B. diazoeciens -infected cells in the nodule. Characterization of GmFWL3 protein. Soybean ( Glycine max )/ Bradyrhizobium diazoefficiens Single-cell transcriptomic analyses reveal cellular and molecular patterns of rubber tree response to early powdery mildew infection [ 115 ] scRNA-seq, transcriptional reporter fusion Study of the distinct gene expression patterns of cell clusters under powdery mildew infection. Report of HbCNL2 gene as disease-resistance gene. Rubber tree ( Hevea brasiliensis )/ Oidium heveae Cell-specific pathways recruited for symbiotic nodulation in the Medicago truncatula legume [ 71 ] snRNA-seq Analysis of the transcriptomic response of Medicago root cells to rhizobial infection. \n Medicago truncatula/ Ensifer meliloti \n Differentiation trajectories and biofunctions of symbiotic and un-symbiotic fate cells in root nodules of Medicago truncatula [ 70 ] scRNA-seq, transcriptional reporter fusion Generation of single-cell transcriptome map of indeterminate root nodules. \n Medicago truncatula/ Sinorhizobium meliloti \n Development of a single-cell atlas for woodland strawberry ( Fragaria vesca ) leaves during early Botrytis cinerea infection using single cell RNA-seq [ 69 ] scRNA-seq Generation of the first single-cell atlas of the plant pathogenic invasion process. Woodland strawberry ( Fragaria vesca )/ Botrytis cinerea"
} | 2,834 |
32911685 | PMC7566005 | pmc | 4,586 | {
"abstract": "An understanding of how fertilization influences endophytes is crucial for sustainable agriculture, since the manipulation of the plant microbiome could affect plant fitness and productivity. This study was focused on the response of microbial communities in the soil and tubers to the regular application of manure (MF; 330 kg N/ha), sewage sludge (SF; 330 and SF3x; 990 kg N/ha), and chemical fertilizer (NPK; 330-90-300 kg N-P-K/ha). Unfertilized soil was used as a control (CF), and the experiment was set up at two distinct sites. All fertilization treatments significantly altered the prokaryotic and fungal communities in soil, whereas the influence of fertilization on the community of endophytes differed for each site. At the site with cambisol, prokaryotic and fungal endophytes were significantly shifted by MF and SF3 treatments. At the site with chernozem, neither the prokaryotic nor fungal endophytic communities were significantly associated with fertilization treatments. Fertilization significantly increased the relative abundance of the plant-beneficial bacteria Stenotrophomonas , Sphingomonas and the arbuscular mycorrhizal fungi. In tubers, the relative abundance of Fusarium was lower in MF-treated soil compared to CF. Although fertilization treatments clearly influenced the soil and endophytic community structure, we did not find any indication of human pathogens being transmitted into tubers via organic fertilizers.",
"introduction": "1. Introduction The application of fertilizers to agronomical soil is a worldwide practice used for improving crop yield. Chemical fertilizers are believed to be the cause of the improvement in crop production by up to 50% during the 20th century [ 1 ]. Simultaneously, due to the constant production of organic waste material, attention has been paid to the re-use of this waste in agriculture instead of (or together with) using chemical fertilizers [ 2 ]. Therefore, interest in how the long-term application of fertilizers influence the soil microbiome has increased over the past few decades. The influence of fertilization on endophytes of cultivated plants, and especially the relation between organic fertilizers and the occurrence of pathogens has remained much less investigated. Soil microbial diversity is an important biological factor for the assessment of soil health, soil quality and ability to suppress diseases [ 3 , 4 ]. Soil microorganisms are actively involved in the cycling of nutrients, and have an impact on the dynamics of nutrient turnover [ 5 ]. They participate in the decomposition of soil organic matter [ 6 ], which releases nutrients, making them available for plants. Soil microorganisms also influence the soil formation [ 7 ] and soil quality [ 8 ]. In turn, changes in environmental conditions or soil properties can directly influence the microbial community and its functioning. For instance, fertilizers (chemical or organic) introduce required nutrients into soil [ 9 ], alter soil pH [ 10 ], water holding capacity, soil texture or cation exchange capacity [ 11 ]. All these modifications of soil physicochemical properties shape the microbial community structure [ 12 ] and metabolic activity [ 13 ], which is subsequently reflected in crop yields [ 14 ]. It is not only the soil microorganisms that play a crucial role in agronomy; a similarly important role can be ascribed to the endophytes (for a review, see [ 15 ]). The microbial community of endophytes colonize inter and intracellular spaces of all plants [ 16 ]. Some endophytes live with their host-plant in a close mutualistic relationship, providing their host plant with a wide range of benefits, while the plants provide them with a protected environment and nutrients [ 17 ]. The endophytic microbes can produce substances with antimicrobial or insecticidal effects, phytohormones altering the plant growth, iron chelators, siderophores and organic acids solubilizing phosphate complexes, or they are able to fix the atmospheric nitrogen [ 15 , 16 , 18 ] and modulate the growth of roots [ 19 ]. The presence of specific beneficial endophytes can also lead to enhancement of the nutritional composition of planted crops and their vitality [ 20 ]. Thus, the alteration of endophytic communities due to the changes in soil properties (e.g., via fertilization) can have broad consequences on host plant health and growth [ 21 , 22 ] as well as on post-harvest storage stability [ 23 ]. The first study on the impact of agricultural practices on endophytes [ 24 ] showed a significant association between fertilization and endophytic community structure. High-nitrogen fertilization was found to increase the abundance of methanogenic archaea in plant roots, while the application of low-N fertilizer caused increases in various functional genes for nutrient metabolism in the endophyte community [ 21 ], and organic fertilizers increased the number of diazotrophic endophytes [ 25 ]. Unfortunately, there was also evidence of antibiotic-resistant bacteria transfer from manure, an example of organic fertilizer, into plant tissues [ 26 ]. This direct transfer has raised concerns about the safety of using organic fertilizers, since they are often found to harbor human pathogens [ 27 , 28 ]. In this study, we evaluated the effect of 21 years of regular fertilization on prokaryotic and fungal communities in bulk soil and stem tubers of Solanum tuberosum L. (potatoes), one of the top five crops produced worldwide [ 29 ]. The influence of manure, sewage sludge (at two different application rates) and NPK fertilizers was studied, and the effects of these treatments were compared to each other and to the unfertilized (control) soil. The experimental fields were established at two geographical locations with different soil and climate characteristics. We hypothesized that the site characteristics will be the main driver of both soil and endophytic community structures, but we expected a significant association between fertilization and the microbial community in both soil and potatoes. We suppose that fertilized soils, due to their better macronutrient properties [ 11 ], could enhance plant-beneficial genera. On the other hand, the addition of sludge and manure can bring into the soil pathogens, which can be further transfer into the plant tissues. Therefore, the benefits of fertilization will be assessed.",
"discussion": "4. Discussion This study was focused on the influence of 21 years of chemical and organic fertilization on soil and endophytic microbial communities. The experiment was established in two geochemically and geographically distinct sites, hence, with different environmental conditions and edaphic characteristics ( Table 1 ). The diversity of soil prokaryotes and fungi did not significantly differ between any of the fertilized soils and the CF. Only the soil fungal diversity was significantly higher with the MF treatment compared to SF, probably due to the introduction of a wide range of biopolymers promoting fungal diversity, as was reported previously [ 46 , 47 ]. Higher fungal diversity positively affects the ability of soil to suppress diseases [ 46 ], indicating that MF treatment is more likely to suppress soil-borne pathogens than other treated soils, as suggested by the lower relative abundance of the common soil-borne pathogen Fusarium [ 3 ] in tubers from MF-treated soil compared to control ( Figure S3 ). In fact, the disease suppression does not generally require the complete eradication of the pathogen, while the establishment of a healthy and diverse microbiome can reduce the chances of infection or stimulate plant defenses [ 48 ], which might have been the case with MF treatment. Soil microbial diversity has not always been reported to be significantly associated with fertilizer application, but to the best of our knowledge, changes in soil characteristics caused by fertilization have been reflected in the structure of the soil microbial community [ 13 , 49 ]. In this study, the soil community structure of both prokaryotic and fungal communities were significantly associated with fertilization treatments ( Figure 3 , Table 2 ), which is in agreement with previous findings; these findings reported either an indirect influence of fertilization on the soil community structure through the alteration of physicochemical soil properties [ 10 , 11 ], or direct influence through the introduction of new species into the soil by the application of organic fertilizers [ 27 ]. The relative abundance of 20 bacterial genera and seven fungal genera was identified in this study to be increased by at least one fertilization treatment (MF, SF, SF3x and/or NPK) compared to CF. Interestingly, all 20 of the bacteria were enriched in SF-treated soil ( Table 2 ). Among the differently abundant genera, the genera Turicibacter , Rhodanobacter , Flavobacterium , Clostridium , Pseudoxanthomonas and Coprothermobacter have been previously isolated from sewage sludge or sewage sludge-treated soils [ 28 , 50 , 51 , 52 ], suggesting their possible direct transmission from the fertilizer to the soil. Several other differently abundant genera, such as Hydrogenophaga and Flavobacterium , were previously associated with the bioremediation of polluted soils [ 53 , 54 , 55 ], which corresponds to the findings that persistent organic pollutants, micropollutants and heavy metals are commonly found in sewage sludge [ 56 , 57 ], including the one used in this study [ 58 ]. Most of all, beneficial plant-associated genera were found to be significantly enriched in the SF and SF3x treatment, including Stenotrophomonas and Sphingomonas . Stenotrophomonas promotes the growth of plants and helps to control the abundance of fungal phytopathogenic fungi in soil through the production of chitinases [ 59 , 60 ]. Sphingomonas was found to be a keystone genus in healthy soils, and was also significantly associated with plant pathogen suppression [ 48 ]. Several plant-beneficial fungi were also identified to be significantly enriched by some of the fertilization treatments. Funneliformis and Diversispora (of the phylum Glomeromycota), which were enriched in MF, NPK and SF-treated soils, are arbuscular mycorrhizal fungi (AMF) [ 61 ], which enhance the solubility and availability of a wide range of nutrients, improve the soil structure, increase water uptake for plants and give them protection [ 62 , 63 ]. Their abundance in soil was previously associated with the fertilization regime and a higher uptake of macro- and micronutrients in plants [ 47 ]. SF-treated soil was also associated with a higher relative abundance of Mortierella , which is a phosphate-solubilizing fungal saprotroph whose presence positively influences the colonization of soil by AMF, and which was associated with higher plant weight [ 64 ]. An increased relative abundance of this genus associated with fertilization has already been reported [ 13 ]. Basidiobolus , a fungal genus enriched in MF- and SF-treated soils, is well-known for the production of chitinase, and therefore, Basidiobolus has been suggested to be used as a biocontrol for phytopathogenic fungi [ 65 , 66 ]. Only one genus among the differently abundant genera, Conocybe , which was enriched in NPK soils, was found to be potentially pathogenic [ 67 ]. In comparison with fertilized soils, the relative abundance of several prokaryotic and fungal genera was higher in CF ( Figures S1 and S2 ). Prokaryotes, such as Brevundimonas and Limnohabitans , were mostly reported to be ubiquitous [ 68 , 69 ], whereas Massilia and Phenylobacterium were associated with biodegradation activity [ 30 ]. Several of these fungal genera are among agronomically promising taxa. Coprinellus is a fungal genus that was found to suppress rot diseases of vegetables [ 70 ]. Idriella is a biocontrol agent of take-all diseases of wheat and barley, and its presence in soil reduces the severity of the crop damage [ 71 , 72 ]. In our experiment, the potato was rotated with spring barley and winter wheat, thus, the lower relative abundance of Idriella in fertilized soils may potentially increase the risk of crop damage unless the lower relative abundance of Idriella is associated with the promotion of other plant-beneficial genera. On the other hand, CF treatment was found to contain a significantly higher relative abundance of Ceratocystis , a genus that includes many plant pathogens, such as Ceratocystis fimbriata , which causes rot of sweet potatoes [ 73 ], when compared to SF3x treatment. Since our results showed that fertilization significantly influences the structure of soil microbial communities with an effect on the relative abundance of several beneficial microbes, we hypothesized that the application of fertilizer would also have an impact on the microbial community of endophytes in tubers. Understanding how fertilization influences endophytes is crucial for sustainable agriculture, since the manipulation of the plant microbiome could affect the plant benefits associated with the activity of endophytes [ 15 , 18 ]. As our results showed, site characteristics, fertilization and their interaction significantly shaped both soil and endophytic community structures. Whereas site characteristics had a higher influence on soil microbial communities than fertilization regimes, the endophytes were more influenced by fertilization and the interaction of site characteristics and fertilization ( Table 2 , R 2 values), yet the specific influence of fertilization varied between the sites ( Figure 3 ). In Humpolec, the prokaryotic and fungal endophytic communities significantly differed in their responses to the fertilization treatments, but the communities were not significantly associated with the fertilization treatment in Suchdol ( Figure 3 ). This difference can be ascribed to the soil characteristics in each locality. Whereas the soil in Humpolec is a cambisol, which is one of the most widespread soil types [ 74 ], Suchdol has chernozem, which is one of the most fertile soils [ 74 ], and, compared to cambisol, has a higher content of oxidizable carbon and higher pH ( Table 1 ) and in general higher microbial biomass and diversity [ 75 ]. Since plants recruit the endophytes to obtain growth and health benefits [ 76 ], we assume that the establishment of endophytic communities in this highly fertile and microbically diverse soil, is much less prone to changes brought about by fertilization. The significant relationship between endophytic community structure and the fertilization regime was more pronounced for prokaryotes ( P adj ≤ 0.001) than for fungi ( P adj ≤ 0.1). Prokaryotic community structure significantly differed in MF and SF3x treatments compared to the CF in both sites. Prokaryotic endophytes were previously found to be significantly altered by the fertilization regime [ 24 , 69 ]. Specifically, methanogenic archaea increased with a higher application of N [ 21 ], or organic fertilizers enhance the number of diazotrophic bacteria [ 25 ]. The possible reason why MF and SF3x treatments were associated with major shifts in prokaryotic endophytic communities might not be only the new sources of nutrients, but primarily the transfer of these fertilizer-borne taxa into the soil [ 27 , 28 ] and their penetration into the endosphere. These inoculated microbes could be already present in the soil, or new taxa could be introduced. Although not all organic fertilizer-borne taxa can be recruited as plant endophytes, they can still influence the indigenous endophytic community structure [ 77 ]. Differential analysis allowed us to identify endophytes, whose relative abundance differed with fertilization. Four genera were found to be relatively enriched in tubers grown in fertilized soils ( Table 3 ). Compared to CF, Xylophilus was enriched with MF and SF3x treatments, Duganella with NPK and SF3x and Acinetobacter with NPK treatment. Whereas Acinetobacter and Duganella are plant-growth promoting bacteria [ 78 , 79 ], Xylophilus is a phytopathogen. However, this bacterial genus only causes disease in grapevines [ 80 ] and its role in Solanum tuberosum L. remains unclear. The increased use of organic fertilizers in agriculture has raised concerns about their safety [ 81 ]. Organic fertilizers can also introduce a range of human and animal pathogens into soil [ 82 ]. For instance, animal manures were found to increase the abundance of Escherichia coli O157:H7 [ 83 , 84 ] or Listeria monocytogenes in cultivated vegetables [ 85 ]. Such findings would imply a potential threat to human health. In this study, 22 ASVs were assigned to one or more potential human pathogens. However, the majority of the potential human pathogens identified were found in soil and tubers under all treatments, or they were also present in CF-soil or CF-tubers. Their occurrence in CF implies that they are part of indigenous microbial communities regardless of fertilization. Appropriate treatment of organic wastes, composting of animal manure and application timing can reduce the potential risk of microbial contamination [ 84 ]. Hence, our results do not indicate that either stabilized sewage sludge (at 55 °C) or manure produced by proper composting, both applied in the autumn, would pose a significant threat to human health. In summary, our study contributes to the understanding of how microbial diversity and community structure in the soil and endosphere of Solanum tuberosum L. respond to 21 years of regular fertilization. Our results showed that while the soil prokaryotic and fungal communities are influenced by fertilization treatments, the effect of fertilization on endophytic communities is site-specific. Furthermore, the application of either chemical or organic fertilizers influenced the relative abundance of several plant-beneficial microbes in soil. In our study, we did not record the transfer of pathogenic microorganisms if properly stabilized manure and sewage sludge are used. What remains to be investigated is the succession of endophytic communities in fertilizer-treated soils over time, and there is a need to broaden the results to other types of soil and host-plant models."
} | 4,569 |
39372026 | PMC11447871 | pmc | 4,588 | {
"abstract": "Brazil is one of\nthe world’s leading producers\nof staple\nfoods and bioethanol. Lignocellulosic residual sources have been proposed\nas a promising feedstock for 2G bioethanol and to reduce competition\nbetween food and fuels. This work aims to discuss residual biomass\nfrom Brazilian agriculture as lignocellulosic feedstock for 2G bioethanol\nproduction as bagasse, stalk, stem, and peels, using biorefining concepts\nto increase ethanol yields. Herein, we focused on biomass chemical\ncharacteristics, pretreatment, microorganisms, and optimization of\nprocess parameters that define ethanol yields for bench-scale fermentation.\nAlthough several techniques, such as carbon capture, linking enzymes\nto supports, and a consortium of microorganisms, emerge as future\nalternatives in bioethanol synthesis, these technologies entail necessary\noptimization efforts before commercial availability. Overcoming these\nchallenges is essential to linking technological innovation to synthesizing\nenvironmentally friendly fuels and searching other biomass wastes\nfor 2G bioethanol to increase the biofuel industry’s potential.\nThus, this work is the first to discuss underutilized lignocellulosic\nfeedstock from other agrifoods beyond sugar cane or corn, such as\nbabassu, tobacco, cassava, orange, cotton, soybean, potatoes, and\nrice. Residual biomasses combined with optimized pretreatment and\nmixed fermentation increase hydrolysis efficiency, fermentation, and\npurification. Therefore, more than a product with a high added value,\nbioethanol synthesis from Brazilian residual biomass prevents waste\nproduction.",
"conclusion": "6 Conclusions and Prospects Corn meal\nand edible parts of sugar cane as a fermentative juice\nsource for synthesizing 1G bioethanol are applied on a large scale\nin Brazil. Regarding 2G bioethanol, challenges must be overcome to\nincrease its energy efficiency. Despite this, Brazil produces a range\nof lignocellulosic biomass from the agribusiness sector. Such residues\nand waste are even discarded in landfills. Therefore, the redirection\nof these residues for the synthesis of 2G bioethanol could significantly\nincrease its commercial availability, added value, and environmentally\nfriendly processes. Several lignocellulosic residues produced\nin Brazil (i.e., from\nAmazonian babassu coconut starch, tobacco, cassava, orange, cotton,\nand potatoes) can benefit the energy matrix. Its combination with\noptimized processing parameters (pretreatment strategies) and fermenting\n(mixed culture) can increase the hydrolysis efficiency of lignocellulosic\nfeedstock and then fermentation as well as the purification steps\nof produced ethanol. For example, the continuous insertion of substrates\nor other nutrients can favor contamination inside a reactor, causing\na decrease in the bioethanol yield. The composition of babassu cake\nmakes it an excellent raw material for solid-state fermentation in\nthe bioethanol process, especially for the cultivation of filamentous\nfungi, enabling the production of complex multienzymes containing\namylases, xylanases, cellulases, and proteases, which are capable\nof hydrolyzing starch granules crude. However, researchers should\ndirect future efforts to overcome challenges\ntoward pretreatment strategies for lignocellulosic residues rich in\naldohexoses series of carbohydrates (i.e., coffee straw and grounds\nafter consumption, rich in mannose). After that, studying the feasibility\nof other biomass residues can further increase bioethanol synthesis\nin the fuel sector, reducing environmental damage and improving the\nmanagement of industrial processes. Other strategies based on\nthe biorefinery concept of bringing 2G\nbioethanol into a hybrid unit should also be considered to produce\ndifferent biofuels from food waste, as accomplished in Brazil with\nbiomass gasification followed by syngas fermentation using acetogenic\nbacteria. In this context, lipids can be separated and converted into\nbiodiesel by transesterification before the bioethanol production\nprocess; defatted lipid biomass can be converted into biomethane by\nanaerobic digestion; and dried residue from enzymatic hydrolysis can\nbe transformed into other valued bioproducts for animal feed. Moreover,\nexploring the sustainability aspects of Brazilian agroindustrial waste\nvalorization platforms through mixed culture fermentation and microbial\nconsortia is crucial. A microbial consortium might improve the hydrolysis\nof the lignocellulosic material, once competition between microorganisms\ncan generate more extracellular enzymes. Finally, the industrial\napplication of 2G bioethanol amplifies\nthe commercialization of biofuels due to the high demand for environmentally\nfriendly products with economic potential for the industry.",
"introduction": "1 Introduction Brazil is essentially\nself-sufficient in food commodities, with\nmany crops grown and harvested to produce food or livestock (i.e.,\nfood and feed crops, fiber and oil crops, and industrial crops). 1 Brazil’s 2022/23 crops are expected to\nreach up to 312.4 million tons of soy, cotton, corn, rice, and other\nproducts in national grain production. 2 , 3 Moreover, Brazil\nhas emerged as a global supplier of a wide range of crops and basic\nfoodstuffs such as soybean, soybean derivatives, corn, herbal cotton,\nsugar cane, coffee, cassava, citrus, cacao, grains, and ethanol since\nthe early 21st century. 3 Moreover, Brazilian\nsugar cane ethanol production started in the 1970s due to the oil\ncrisis in search of alternative fuels. 4 − 6 In 2019, more than 45%\nof the energy power supply in Brazil came from renewable sources. 7 Biofuels represent 25% of transport fuels in\nBrazil, with the highest presence of bioethanol representing 49% of\nthe energy of combined gasoline and ethanol. 7 Although sugar cane 8 , 9 and corn 10 − 12 are major conventional\ncrops used in first-generation (1G) bioethanol, such crops cannot\nachieve world demand for bioethanol production due to their primary\nvalue as food and feed crops. 9 , 11 , 13 Therefore, since 2010 several Brazilian institutions and research\ngroups have described efforts toward cellulose-based sources as agroindustrial\nwastes as promising feedstock for bioethanol production, 14 once cellulose is the most abundant agroindustrial\nbiomass residue available. 15 , 16 Regarding biomass\nresidues, Brazil has different sources of agrifoods\ngenerating biomass residues, such as sugar cane bagasse, rice straw,\nand corn cob/corn leaf. 14 All of these\nenergy sources are considered industrial waste due to their application\nin the synthesis of high-value-added products. Therefore, applying\nsuch residual biomasses on a bioethanol industrial and commercial\nscale is a challenge. 17 Bioethanol\nis one of the most promising fuels on the market, partially\ncapable of replacing fossil fuels, requiring 68% less energy production\nthan high-octane gasoline. 18 , 19 The production commonly\nuses sugar cane, sweet potato, beetroot, and cereals, competing with\nthe food industry. Thus, other synthesis methods have been studied,\nsuch as hydrolyzing bagasse to sugars and the use of residues in the\nfermentation process. 19 − 22 Therefore, novel and efficient bioprocesses for underutilized biomass\nare at the forefront of biotechnological research and industrial application. The residues for bioethanol production require a high amount of\ncarbohydrates, microorganisms with good viability for propagation,\nand specific enzymes to break these carbohydrates into simpler molecules. 20 , 23 Brazil presents a high amount of food waste and microorganisms that\nopen many possibilities for optimizing the fermentation to generate\nindustrial bioethanol. 24 Therefore, beyond\nthe product synthesis with high added value, biomasses residual waste\nmay not be discarded. 21 , 25 − 28 The synthesis occurs through\nthe fermentation reaction of sugars converted into ethanol. Fermentation\ndepends on several factors such as time, temperature, specificity\nof metabolites produced by microorganisms, pH, and contaminants. 26 , 27 , 29 After pretreatment, the\nresidue undergoes hydrolysis by chemical\nor biological routes. Chemical hydrolysis can be achieved by diluted\nor concentrated sulfuric acid. 29 , 30 The thermal stability\nand specificity of enzymes such as amylases, cellulases, pectinases,\nand glycosidases 29 , 31 influence enzymatic hydrolysis.\nThe use of enzymes is encouraged by milder temperature and pH conditions\nand lower consumption of water vapor, making the process more industrially\napplicable. 20 , 32 Furthermore, a recent bibliometric\nanalysis showed higher efforts over the last 10 years. 33 Saccharomyces cerevisiae is a microorganism commonly used in lignocellulosic hydrolysates 34 for fermentation, a process that occurs at temperatures\nof 30 °C until the ethanol concentration is >10% in the fermentation\nbroth. Other microorganisms such as Xanthomonas axonopodis , Candida parapsilosis , and Trichoderma harzianum can also be cultivated and\nused in fermentation for producing cellulase enzymes that break down\ncarbohydrates, generating bioethanol with purity higher than 97.5%. 35 , 36 Once Brazil alone was the most significant sugar cane producer,\nsugar cane bagasse 37 − 40 was extensively reviewed and proposed as potential agricultural\nwaste in second-generation (2G) bioethanol due to high contents of\nfermentable sugars in biomasses (cellulose, lignin, and hemicellulose).\nHowever, the offer of bagasse and straw for bioethanol production\non an industrial scale has also been prospected to be insufficient,\ndemanding alternative lignocellulosic materials with potential in\n2G ethanol production. 40 Recently, researchers\noverviewed literature to design and propose an integrated soybean–sugar\ncane biorefinery with a high potential for implementation in Brazil. 41 However, since 1998, the increased possibility\nof other potential Brazilian agrifoods (i.e., babassu mesocarp) was\nprospected to produce low-cost ethanol compared to sugar cane or conventional\nstarch-rich material. The study showed the net profitability of ethanol\nproduction of 40% for babassu coconut as opposed to 10% for sugar\ncane. 42 Babassu wastes received very little\nattention in bioethanol production after two decades despite its feasibility,\nhigh essential density, and suitable lignin content in babassu kernel\nnut for sustainable bioenergy production later commented on. 43 Brazil has several biomasses that could\nbe applied to the synthesis\nof 1G bioethanol, such as papaya, banana, babassu, and potato, which\ncompete with the food industry. On the other hand, the third-generation\n(3G) ethanol, obtained from algae, stood out by its greater photosynthetic\nefficiency and less competition for agricultural resources, but its\nproduction on a commercial scale still requires more technological\nadvances. In that sense, the residues of those crops, such as banana\npeel, coffee peel, potato peel, papaya peel, corn cob, and dry corn\nleaf peels, among countless residues that are commonly discarded,\ncan be used for 2G bioethanol synthesis. Therefore, it provides a\nreagent not used by the food industry to synthesize a high-added value\nproduct for a greener future: 2G bioethanol, which has potential as\na clean and renewable transportation fuel. Moreover, by blending it\nwith gasoline, 2G bioethanol reduces conventional fuels’ carbon\nfootprint and allows countries to meet stringent emission standards. Although several reports are available on review analysis of bioethanol\nproduction from biomasses, they focused on sugar cane bagasse, 9 , 14 , 44 corn residues, 10 , 13 or food wastes rich in carbohydrates. 45 Moreover, until now, although some assessment on other lignocellulosic 46 was performed, some of the critical factors\nhave not yet been discussed, such as (i) reaction conditions at pretreatment\nof biomass waste to maximize bioethanol conversion efficiency; (ii)\nyields obtained by the alcoholic fermentation process; and (iii) the\nimpact on current Brazilian fuel production. Therefore, this Review\nfocuses on recent studies with different biomass residues available\nin Brazil for 2G ethanol. Besides, the analysis of 1G and 2G ethanol\nin different regions of Brazil using sugar cane and processing residues\nto increase its industrial production leads to an economic improvement\nand good environmental impact in Brazil and the global energy matrix. Once the use of agricultural wastes to produce bioethanol does\nnot compromise food security, contributes to waste management, prevents\nenvironmental degradation, and ensures energy security, this work\ncan potentially impact not only locally but globally, improving manufacturing\nactivities, enhancing farming and other food production related to\nagricultural waste generation, renewable fuel consumption, and emission\nof toxic gases."
} | 3,201 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.