pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
30363846 | PMC6182589 | pmc | 9,541 | {
"abstract": "Spatially complex habitats provide refuge for prey and mediate many predator–prey interactions. Increasing anthropogenic pressures are eroding such habitats, reducing their complexity and potentially altering ecosystem stability on a global scale. Yet, we have only a rudimentary understanding of how structurally complex habitats create ecological refuges for most ecosystems. Better informed management decisions require an understanding of the mechanisms underpinning the provision of physical refuge and this may be linked to prey size, predator size and predator identity in priority habitats. We tested each of these factors empirically in a model biogenic reef system. Specifically, we tested whether mortality rates of blue mussels ( Mytilus edulis ) of different sizes differed among: (i) different forms of reef structural distribution (represented as ‘clumped’, ‘patchy’ and ‘sparse’); (ii) predator species identity (shore crab, Carcinus maenas and starfish , Asterias rubens ); and (iii) predator size. The survival rate of small mussels was greatest in the clumped experimental habitat and larger predators generally consumed more prey regardless of the structural organisation of treatment. Small mussels were protected from larger A. rubens but not from larger C. maenas in the clumped habitats. The distribution pattern of structural objects, therefore, may be considered a useful proxy for reef complexity when assessing predator–prey interactions, and optimal organisations should be considered based on both prey and predator sizes. These findings are essential to understand ecological processes underpinning predation rates in structurally complex habitats and to inform future restoration and ecological engineering practices.",
"introduction": "Introduction Habitat complexity plays a key role in mediating biotic interactions, such as predator–prey relationships (Heck and Crowder 1991 ; Warfe and Barmuta 2004 ; Klecka and Boukal 2014 ). Structurally complex habitats may provide refuge space for prey (Křivan 1998 ; O’Connor and Crowe 2008 ), thus modifying predator–prey dynamics (Beck 1995 ; Barrios-O’Neill et al. 2015 ), and lead to a cascade of indirect effects on multiple trophic levels (Grabowski and Kimbro 2005 ; Grabowski et al. 2008 ; O’Connor et al. 2013 ). Spatial refuges within complex habitats can be of particular importance for smaller individuals (Hacker and Steneck 1990 ; Strain et al. 2017 ), including recent recruits and juveniles, which are usually more vulnerable to predation than larger individuals (Gosselin and Chia 1995 ). Studies of habitat complexity often use different definitions of complexity or confound complexity with other habitat characteristics, such as surface area or heterogeneity (Beck 2000 ; Frost et al. 2005 ; Kovalenko et al. 2012 ; Loke et al. 2015 ), which can lead to misuses of these metrics for management purposes (Wedding et al. 2011 ). Habitat complexity per se is often used as an over-arching term that encompasses variation in several habitat ‘components’, e.g. density of specific habitat component such as pits, pneumatophores or crevices (McCoy and Bell 1991 ), which limits the application of the results of studies using generic or obtuse terminology. A more useful approach is to use only specific metrics of individual habitat components (Beck 2000 ). Many structurally important habitats, e.g. rainforests, saltmarshes, and aquatic biogenic reefs, are under threat from anthropogenic disturbances (Ellison et al. 2005 ; Airoldi et al. 2008 ; Silliman et al. 2009 ; Newbold et al. 2014 ; Firth et al. 2015 ). Biogenic reefs formed by bivalves play an essential role as ecosystem engineers (Geraldi et al. 2017 ) by: (i) promoting higher levels of biodiversity than surrounding local environments (Gutierrez et al. 2003 ; O’Connor and Crowe 2007 ); (ii) providing habitat that acts as a nursery for commercially important species (Kent et al. 2016 , 2017 ); (iii) stabilising sediments (Meadows et al. 1998 ); (iv) acting as natural wave barriers and protecting soft coastal habitat (Stone et al. 2005 ); and (v) contributing substantially to nutrient cycling (Kellogg et al. 2013 ). The loss of such biogenic habitat following disturbance events can lead to changes in many biotic interactions, which can impede the recovery of a system following further disturbances (Lotze et al. 2006 ; Bertness et al. 2015 ; Mrowicki et al. 2016 ). It is often assumed that biogenic reefs have a self-sustaining mechanism, such that once a reef is established, its complex structure provides refuge from predation, which facilitates recruitment (Bertness and Grosholz 1985 ; Nestlerode et al. 2007 ; Walles et al. 2015 , 2016 ) and maintains a healthy and stable reef system. When a reef is damaged, however, this process will be diminished (Lenihan 1999 ), which could de-stabilise a reef-dominated system by reducing the establishment of new recruits and subsequent reef re-formation (Barrios-O’Neill et al. 2017 ; Fariñas-Franco et al. 2018 ; Fariñas‐Franco and Roberts 2018 ), potentially leading to an alternative stable state which may be represented as a ‘degraded’ system lacking in complexity (Petraitis and Dudgeon 2004 ). It is essential to understand how predator–prey relationships interact with spatial complexity (Warfe et al. 2008 ; Hesterberg et al. 2017 ) so that we can comprehend how these processes are linked to refuge availability, which underpins biogenic reef formation and persistence. The density of reef-forming species can be used as a proxy to estimate the organisation (or spatial arrangement) of vertical objects in horizontal space (Bell et al. 1991 ). Density is a tractable measure that can be quantified and manipulated experimentally and, thus provides useful insights into the recruitment dynamics of reef-forming species (Carroll et al. 2015 ). The density of reef-formers, however, is often confounded with other factors, such as volume or abundance of individuals (e.g. Humphries et al. 2011a , b ). The aim of our study was to test whether different, but typical, spatial arrangements of habitat structure within an experimental reef system affected the survival of small mussels. Specifically, by ensuring that other components of habitat structure were constant and manipulating only density, we tested directly whether the size of interstitial spaces available affected overall mussel survival, and whether different spatial arrangements of this habitat provided better protection for small-sized mussels from differently sized predators (Toscano and Griffen 2014 ; Bartholomew et al. 2016 ). Additionally, we tested whether refuge efficacy differed between species of common predators with distinct methods of catching and killing prey (O’Connor et al. 2008 ; Farina et al. 2014 ). Small interstitial spaces which are typical of mussel beds may be beneficial for the smaller mussels, whilst being of limited use for larger mussels, which may become more vulnerable to predation from larger or different predator species (Enderlein et al. 2003 ; Calderwood et al. 2015b ). In two separate experiments, the effects of predation of two common benthic predators (the shore crab, Carcinus maenas, and the starfish, Asterias rubens ), on their shared prey (mussels) was quantified using artificial reefs that were designed to represent three different forms of habitat organisation. The size of the predators and of prey was also manipulated to test explicitly for size-dependent effects and to identify mechanisms that underpin predation in this system. Specifically, both experiments tested the hypotheses that: predation rates on mussels are dependent upon habitat organisation (‘sparse’, ‘patchy’ and ‘clumped’) with (1) mortality rate of small mussels being lowest in the ‘clumped’ habitat organisation; whereas (2) a ‘patchy organisation’, with heterogeneous sizes of refugia available, will provide generally more options for refuge, thus decreasing the mortality of larger individuals; and that (3) the size of predators will affect the mortality rates of their prey, in relation to accessibility to the interstitial spaces (Fig. 1 ), while (4) the ratio of small mussel mortality compared to total mortality will only differ with habitat spatial organisation. Fig. 1 Hypothetically predicted predator–prey relationships in ‘sparse density’ (dotted line), ‘patchy organisation’ (dashed line) and ‘clumped density’ (black line) treatments with predators of increasing size for: a total mussel mortality, and b small mussels",
"discussion": "Discussion In agreement with our initial hypotheses (Fig. 1 ), we found that in the presence of crabs, different habitat organisations affected mussel predation rates differently, and that mussel survival was greatest in the clumped density treatment. In contrast, we did not identify any effects of different habitat organisation on total mussel mortality in the presence of starfish. However, when predation effects were examined for small mussels only, clumped habitat organisation significantly increased survival of small mussel classes independent of predator species. There was no effect of organisation on the ratio of small mussels consumed compared to total mortality when crabs were the predators, while there was a slight decrease in this ratio with increasing organisation density when starfish were predators. In the present study, habitat organisation was manipulated only with regard to horizontal space by manipulating object (mussel mimic) density. Despite this simplification of variability in reef structure, which did not consider variations in three dimensions (Hesterberg et al. 2017 ), and used habitat mimics, it was a highly suitable proxy to test for effects of different interstitial space sizes among objects (Bartholomew and Burt 2015 ; Bartholomew et al. 2016 ) and was a suitable mediator of predation rates. Where habitat organisation was clumped, prey mortality, in general, was lower, suggesting that altering habitat organisation even in two dimensions reveals useful mechanistic insights with regard to refuge availability and refuge efficacy. The reduction in mortality owing to predation in habitats with clumped density organisation could be driven by the ability of prey to hide from predators and/or the inability of predators to reach inside the refuge to access prey (Klecka and Boukal 2014 ), rendering the task too difficult or energetically prohibitive (Dolmer 1998 ). Variability in the efficacy of refugia was prey size specific. Small mussels benefitted from the clumped-density organisation, where they found refuge. Mortality for this size class was lower in the clumped compared to the lower density organisation treatments, suggesting an effect of habitat organisation and not general prey size preference per se. These findings suggest that promoting a habitat with small available refuge spaces can significantly increase survival of small mussels in the presence of predators, and thus enhance their potential to increase reef sustainability and growth (van de Koppel et al. 2005 ; Commito et al. 2014 ; Folmer et al. 2014 ; Bertolini et al. 2017 ). The size of the predators affected mussel mortality positively and the lack of interactions showed that this occurred independently of habitat organisation. This result was in contrast to our third and fourth hypotheses, as we found that in general consumption rate, particularly the consumption of larger mussels, increased with predator size. Smaller predators in both experiments were small enough to use the interstitial spaces in the ‘clumped density’ organisation, and were observed doing so. In contrast to predictions, larger crabs were able to feed on small mussels in the clumped organisation. This did not occur when starfish were present because predator size did not affect their predation rates on mussels in the clumped organisation. This could be because of their different physical feeding methods, with crabs having strong claws (Vermeij 1977 ) and able to reach into a refuge space and pull a mussel out. Starfish might not be able to reach and secure a small mussel in a small space because of the bulk of their arms and relative weakness of a hydrostatic tube foot system leading to a poor capacity to grip prey (Dolmer 1998 ). Moreover, predators with different prey-detecting strategies (visual vs chemical vs tactile cue) may have different rates of prey encounter in complex habitats (Farina et al. 2014 ; Klecka and Boukal 2014 ). Larger predators consumed more mussels and tended to prefer larger sizes, whilst smaller predators were limited to consumption of smaller mussels, suggesting that larger mussels may be able to escape predation from smaller predators in all of the organisations tested here. This was consistent with results from studies of crab predation on oyster reefs where the presence of large crabs was the major determinant of mussel and oyster mortality (Toscano and Griffen 2012 ; Pickering et al. 2017 ). Thus, mussels that are excluded from refuge space and are of edible size may suffer from high mortality rates. While other experiments found that, A. rubens (Hummel et al. 2011 ) and C. maenas (Smallegange and Van Der Meer 2003 ) often prefer small prey items. We found that overall small mussel mortality also increased with increasing predator size, and the ratio of small mussels mortality compared to total mortality decreased with increasing predator size, suggesting that consumption of larger size classes is important for larger predator sizes. In natural reefs, refugia may be highly variable in size with suitable but limited refuge space for all size classes, however, for damaged reefs to recover it is important that refugia for small mussels is present so they can grow to regenerate a self-sustaining reef. Different predator species were found to have generally similar effects on mortality rates, but some differences were highlighted. For example, larger crabs were found to eat small mussels in the clumped density organisation. This was not observed in trials involving large starfish. The importance of considering predator assemblage composition has been highlighted for commercial mussel seeding operations (Calderwood et al. 2015a ) and this is reinforced by the present study. Clumped mussel habitat can also be beneficial for smaller predators which may hide therein (Thiel and Dernedde 1994 ), highlighting the importance of considering the ontogenetic and behavioural responses of predators (Pirtle et al. 2012 ). It is known, for example, that mussel reefs are nursery grounds for whelks (Kent et al. 2016 ) and crabs (Lindsey et al. 2006 ) and the size of small predators used in this experiment may spend more time sheltering from larger predators in refuge space afforded by a reef than actively feeding. This should be tested empirically. It is concluded that refuge efficacy in reducing mortality from predation is greatest in clumped density habitat organisation but is strongly dependent on prey size rather than predator size. This contrasts with previous research, which found that space size relative to predator width (Sp/Pr) was one of the most important predictors of survivorship, with prey survivorship decreasing sigmoidally with increasing Sp/Pr (Bartholomew et al. 2000 ). We recommend that refuge size should be evaluated in relation to prey size (Sp/Py) (Hacker and Steneck 1990 ; Bartholomew and Shine 2008 ; Bartholomew 2012 ), that habitats containing multi-sized interstitial spaces should be promoted to offer refuge to multiple size classes, and ultimately should be context-specific with regards to the identity of predators present in the system. Our findings demonstrate the use of spatial organisation as a measure of habitat complexity to explain the effects of predator–prey interactions (Almany 2004 ; Carroll et al. 2015 ; Hesterberg et al. 2017 ). Also, the density and size of mussels may be critical in off-setting the effects of one or more predator species each able to access and exploit different components of the mussel population (Garner and Litvaitis 2013 ). These findings have important consequences for ecological engineering projects (Firth et al. 2016 ) and the management of structures when the aim is to aid the reintroduction of species (e.g. canopy algae, Susini et al. 2007 ; Perkol-Finkel et al. 2012 ; or native oysters, Strain et al. 2017 ) while keeping in mind the role of biotic interactions (Ferrario et al. 2016 ; Gianni et al. 2018 ) to promote self-sustainability after initial restoration efforts."
} | 4,173 |
37110410 | PMC10144548 | pmc | 9,543 | {
"abstract": "Cyanobacterial harmful algal blooms (CyanoHABs) are longstanding aquatic hazards worldwide, of which the mechanism is not yet fully understood, i.e., the process in which cyanobacteria establish dominance over coexisting algae in the same eutrophic waters. The dominance of CyanoHABs represents a deviation from their low abundance under conventional evolution in the oligotrophic state, which has been the case since the origin of cyanobacteria on early Earth. To piece together a comprehensive mechanism of CyanoHABs, we revisit the origin and adaptive radiation of cyanobacteria in oligotrophic Earth, demonstrating ubiquitous adaptive radiation enabled by corresponding biological functions under various oligotrophic conditions. Next, we summarize the biological functions (ecophysiology) which drive CyanoHABs and ecological evidence to synthesize a working mechanism at the population level (the special mechanism) for CyanoHABs: CyanoHABs are the consequence of the synergistic interaction between superior cyanobacterial ecophysiology and elevated nutrients. Interestingly, these biological functions are not a result of positive selection by water eutrophication, but an adaptation to a longstanding oligotrophic state as all the genes in cyanobacteria are under strong negative selection. Last, to address the relative dominance of cyanobacteria over coexisting algae, we postulate a “general” mechanism of CyanoHABs at the community level from an energy and matter perspective: cyanobacteria are simpler life forms and thus have lower per capita nutrient demand for growth than coexisting eukaryotic algae. We prove this by comparing cyanobacteria and eukaryotic algae in cell size and structure, genome size, size of genome-scale metabolic networks, cell content, and finally the golden standard—field studies with nutrient supplementation in the same waters. To sum up, the comprehensive mechanism of CyanoHABs comprises a necessary condition, which is the general mechanism, and a sufficient condition, which is the special mechanism. One prominent prediction based on this tentative comprehensive mechanism is that eukaryotic algal blooms will coexist with or replace CyanoHABs if eutrophication continues and goes over the threshold nutrient levels for eukaryotic algae. This two-fold comprehensive mechanism awaits further theoretic and experimental testing and provides an important guide to control blooms of all algal species.",
"introduction": "1. Introduction CyanoHABs are one of the most profound environmental hazards in modern human history in terms of their global geographical scale [ 1 ], longstanding duration (over a century) [ 2 , 3 ], and tremendous economic loss [ 4 ]. The mere fact that they are still intensifying and expanding under global climate change [ 1 , 5 , 6 ] attests to the fact CyanoHABs are still not fully understood, such that their ecology remains “complicated and confusing” [ 7 ]. To demystify CyanoHABs, they need to be examined from an evolutionary ecological perspective, considering all the main factors: bloom-forming cyanobacteria, other coexisting algal species, and eutrophic conditions. From an evolutionary perspective, all cyanobacteria as individual populations evolve in oligotrophic aquatic environments prior to water eutrophication and form blooms almost immediately in response to elevated nutrient loading [ 8 , 9 ]; ecologically, CyanoHABs are a consequence of uncontrolled growth of cyanobacteria in phytoplankton communities where coexisting algae do not form blooms at the same levels of nutrients [ 9 , 10 , 11 ]. In light of this analysis, the overarching question—of why cyanobacteria, not the coexisting algae, form blooms in the same eutrophic waters—can be broken down into two, one constituting the sufficient condition and the other the necessary condition: (1) how cyanobacteria form blooms in the eutrophic waters at the population level, and (2) why do coexisting algae not form blooms in the same eutrophic waters at the community level? Both questions above boil down to the differential interplays between different life forms (cyanobacteria and coexisting algae) and environments in terms of energy and matter, leading to sustained and even saturated growth of cyanobacteria, but not of coexisting algae and thereby CyanoHABs. In this review, we first review the origin and evolution of cyanobacteria in oligotrophic conditions to provide the evolutionary ecological context for CyanoHABs, particularly in terms of the origin of relevant biological functions (ecophysiology). Next, we summarize the biological functions driving bloom formation, as a special mechanism of CyanoHABs at the population level. We then show that these functions are not a result of positive selection by water eutrophication, but of long adaptation to oligotrophic conditions prior to water eutrophication. Last, we propose a general mechanism of CyanoHABs at the community level to account for the fact that cyanobacteria but not coexisting algae form blooms in the same waters."
} | 1,256 |
33580191 | PMC7881103 | pmc | 9,544 | {
"abstract": "Collective behaviour in flocks, crowds, and swarms occurs throughout the biological world. Animal groups are generally assumed to be evolutionarily adapted to robustly achieve particular functions, so there is widespread interest in exploiting collective behaviour for bio-inspired engineering. However, this requires understanding the precise properties and function of groups, which remains a challenge. Here, we demonstrate that collective groups can be described in a thermodynamic framework. We define an appropriate set of state variables and extract an equation of state for laboratory midge swarms. We then drive swarms through “thermodynamic” cycles via external stimuli, and show that our equation of state holds throughout. Our findings demonstrate a new way of precisely quantifying the nature of collective groups and provide a cornerstone for potential future engineering design.",
"introduction": "Introduction Organisms on every size scale, from single-celled 1 to highly complex 2 , regularly come together in groups. In many cases, such aggregations are collective, in that the group as a whole displays properties and functionality distinct from those of its individual members or simply their linear sum 3 , 4 . It is generally assumed that since evolution has led so many different kinds of animals to behave collectively, the performance of collective groups at whatever task they seek to achieve ought to be well beyond the capabilities of a single individual 5 , while also being robust to uncertain natural environments 6 , 7 and operating without the need for top-down control 8 . For these reasons, there has been significant interest both in understanding how collectivity conveys these advantages 9 and how to exploit it in engineered systems 10 , 11 . Taking advantage of evolutionary adaptation for the design of such a bio-inspired artificial collective system requires both determining the interaction rules used by real animals and properly understanding the function of the group. Both of these tasks remain a challenge. Extracting interaction rules by observing group behaviour is a highly nontrivial inverse problem 12 that can typically only be solved by assuming a modelling framework a priori 13 , 14 . Appropriate model selection is made more difficult given that interactions may change in different contexts 7 , 8 , 15 . Even less work has been done to precisely determine the tasks optimized by collective behaviour. Assumptions about the purpose of group behaviour typically come from ecological reasoning 16 rather than quantitative empirical evidence 8 —and in some cases, such as hypothesized aerodynamic benefits conveyed to flocking birds, such reasoning has proved to be incorrect 17 , 18 . We argue that the essential nature of the group functionality is encoded in its properties—and therefore that understanding these properties both allows one to quantify the purpose of the collective behaviour and to predict the response of the group to environmental changes. As recent work has demonstrated 19 – 21 , a powerful way to characterize these properties is to borrow ideas from other areas of physics. For groups on the move such as human crowds, hydrodynamics is a natural choice, and empirically measured constitutive laws have allowed the formulation of equations of motion that accurately predict how crowds flow 20 . But for stationary groups such as insect swarms, where the group as a whole does not move even though its constituent individuals are continuously rearranging, thermodynamics is a more natural framework, as it allows one to precisely describe the state of the system irrespective of its net motion 22 . The most fundamental relationship for doing so is the equation of state, which links the state variables that describe the macroscopic properties of the system and encodes how they co-vary in response to environmental changes. Here, we formulate such an equation of state for laboratory swarms of the non-biting midge Chironomus riparius (Fig. 1 a). We define appropriate state variables, and empirically deduce their relationship by analysing a large data set of measured swarms 23 . Then, by applying a suitable sequence of external perturbations to the swarms, we show that we can drive them through a thermodynamic cycle in pressure–volume space throughout which our empirical equation of state holds. Figure 1 Swarm kinematics. ( a ) Trajectories (> 40 s long) of individual midges (each colour corresponding to a different midge) are individually convoluted but remain spatially localized over a ground-based swarm marker (black square). ( b ) Averaged spring constant \\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}$$\\langle k|N\\rangle$$\\end{document} ⟨ k | N ⟩ as function of the swarm size N (symbols). The black line is a power-law fit to the data. ( c ) Probability density function (PDF) of midge positions in the horizontal plane (blue) along with a Gaussian fit to the data (red). ( d ) PDF of midge positions in the vertical (gravity) direction (blue) and a Gaussian fit to the data (red). The deviation from Gaussianity in the vertical component of the position arises from the symmetry breaking due to the bottom floor of the experimental setup. ( e , f ) PDFs of the horizontal ( e ) and vertical ( f ) midge velocities (blue) along with Gaussian fits to the data (red).",
"discussion": "Discussion Our findings demonstrate the surprising efficacy of classical equilibrium thermodynamics for quantitatively characterizing and predicting collective behaviour in biology. Even though individual midges are certainly not in equilibrium and need not obey the same rules as, for example, particles in an ideal gas, we find that the collective behaviour of ensembles of these individuals is surprisingly simple. The existence of a well-defined equation of state for this system gives us a new way both of illuminating the purpose of collective behaviour, given that it encodes the nature of the collective state, and quantitatively distinguishing different kinds of animal groups that may have similar movement patterns but different functions 1 – 3 , 8 . Importantly, we note that this equation of state is not a swarm model per se, in that it does not make any detailed predictions about the dynamics of individuals. Rather, it gives us a quantitative way of analysing and interpreting swarm data at the macroscale. Finally, these results also provide a natural starting point for designing artificial collective systems by outlining a framework for adapting intuition and expertise gained from engineering thermodynamic systems to this new situation. This approach could, for example, be useful to guide the design of engineered drone swarms via machine learning techniques 32 and to provide a precise and quantifiable global description of the collective nature of swarms."
} | 1,764 |
38447200 | PMC11095224 | pmc | 9,545 | {
"abstract": "Abstract Actuators utilizing snap‐through instabilities are widely investigated for high‐performance fast actuators and shape reconfigurable structures owing to their rapid response and limited reliance on continuous energy input. However, prevailing approaches typically involve a combination of multiple bistable actuator units and achieving multistability within a single actuator unit still remains an open challenge. Here, a soft actuator is presented that uses shape memory alloy (SMA) and mixed‐mode elastic instabilities to achieve intrinsically multistable shape reconfiguration. The multistable actuator unit consists of six stable states, including two pure bending states and four bend‐twist states. The actuator is composed of a pre‐stretched elastic membrane placed between two elastomeric frames embedded with SMA coils. By controlling the sequence and duration of SMA activation, the actuator is capable of rapid transition between all six stable states within hundreds of milliseconds. Principles of energy minimization are used to identify actuation sequences for various types of stable state transitions. Bending and twisting angles corresponding to various prestretch ratios are recorded based on parameterizations of the actuator's geometry. To demonstrate its application in practical conditions, the multistable actuator is used to perform visual inspection in a confined space, light source tracking during photovoltaic energy harvesting, and agile crawling.",
"introduction": "1 Introduction Snap‐through instabilities have a central role in a wide range of bistable and multistable systems, from the venus flytrap [ \n \n 1 \n \n ] to shape reconfigurable metamaterials [ \n \n 2 \n \n ] and soft robot actuators. [ \n \n 3 \n , \n 4 \n \n ] For engineered systems, such structures can be designed to respond to a large variety of stimuli, including pneumatic/hydraulic pressure, [ \n \n 5 \n , \n 6 \n , \n 7 \n , \n 8 \n , \n 9 \n , \n 10 \n \n ] electric field, [ \n \n 11 \n , \n 12 \n , \n 13 \n , \n 14 \n \n ] magnetic field, [ \n \n 15 \n , \n 16 \n , \n 17 \n , \n 18 \n \n ] light, [ \n \n 19 \n , \n 20 \n , \n 21 \n \n ] solvents, [ \n \n 22 \n , \n 23 \n , \n 24 \n \n ] and humidity. [ \n \n 21 \n , \n 25 \n \n ] Bistable and multistable structures can be constructed from elastic beams, [ \n \n 17 \n \n ] plates and shells, [ \n \n 8 \n , \n 12 \n , \n 26 \n , \n 27 \n , \n 28 \n \n ] and thin‐walled balloons, [ \n \n 5 \n , \n 29 \n \n ] as well as from spring‐hinge structures [ \n \n 30 \n , \n 31 \n \n ] and stiffness‐tunable materials. [ \n \n 32 \n \n ] They can also be patterned using origami, [ \n \n 25 \n , \n 33 \n , \n 34 \n , \n 35 \n , \n 36 \n , \n 37 \n , \n 38 \n , \n 39 \n \n ] kirigami, [ \n \n 18 \n , \n 40 \n \n ] and other metamaterial structures. [ \n \n 2 \n , \n 26 \n , \n 41 \n \n ] Progress in the development of bistable and multistable structures has also led to a variety of breakthrough applications in the emerging field of soft robotics. These include mobile soft robots capable of rapid and multimodal locomotion, [ \n \n 10 \n , \n 31 \n , \n 42 \n , \n 43 \n , \n 44 \n \n ] grasping and manipulation, [ \n \n 6 \n , \n 16 \n \n ] soft valves for controlling fluid flow, [ \n \n 7 \n \n ] and soft logic devices. [ \n \n 15 \n , \n 25 \n , \n 45 \n , \n 46 \n \n ] Compared to traditional monostable counterparts, bistable, and multistable actuators embrace the following two unique advantages. First, snap‐through instability enables rapid response, as well as remarkable force output. After overcoming an initial energy barrier, the actuator automatically converges to a newly‐established stable state and can release a large amount of stored elastic potential energy. This can result in high‐performance actuators for fast locomotion [ \n \n 10 \n , \n 31 \n \n ] and jumping. [ \n \n 8 \n , \n 30 \n \n ] Second, once snapped, bistable and multistable actuators naturally maintain their configuration without the need for continuous energy input, something that is not viable in traditional monostable actuators. Consequently, it demonstrates a reliable solution for robot grasping, [ \n \n 6 \n , \n 16 \n \n ] dynamic changing in locomotion gait, [ \n \n 43 \n \n ] structure morphing, [ \n \n 32 \n , \n 37 \n \n ] gated logic devices [ \n \n 25 \n , \n 45 \n , \n 46 \n \n ] and electronics‐free‐control [ \n \n 47 \n \n ] in a rapid and energy‐efficient fashion. To further exploit the potential of snap‐through instability, researchers have endeavored to create multistable architectures that expand a system's total number of stable configurations and modes of deformation. This is typically achieved by directly combining multiple bistable units. For instance, researchers have proposed the creation of multistable origami arms capable of omnidirectional bending and twisting, [ \n \n 16 \n \n ] multistable origami logic circuits, [ \n \n 15 \n \n ] and multimodal deformation [ \n \n 39 \n \n ] by serially connecting multiple units of Kresling origami bistable modules. Furthermore, a multistable soft flapping‐wing swimmer with enhanced maneuverability has been presented through the parallel combination of two pneumatic bistable actuators. [ \n \n 10 \n \n ] Additionally, arrays of bistable shell units have been integrated to develop multistable thin‐walled domes [ \n \n 26 \n \n ] and multistable responsive surfaces. [ \n \n 24 \n \n ] Moreover, multistable metamaterials and structures typically require the spatial assembly of multiple unit cells in order to achieve more intricate shape changes. [ \n \n 22 \n , \n 41 \n \n ] Recently, researchers have combined kirigami with bistable dome structures, which leads to architectures with multiple stable states. [ \n \n 18 \n , \n 40 \n \n ] In addition, latching mechanisms and stiffness tuning have been adopted to develop multistable beams with latch‐able configurations. [ \n \n 32 \n \n ] Although these precedents offer several feasible approaches for achieving multistable actuators, the direct realization of intrinsic multistability within a single actuator unit still remains a challenging task. Here, we introduce an intrinsically multistable actuator unit that can adopt mixed‐mode snap‐through instabilities to realize six stable configurations, including two pure‐bend (B1 and B2) and four bend‐twist (T1 + , T1 − , T2 + , and T2 − ) stable states. The actuator consists of a prestretched elastic membrane sandwiched between two elastomeric frames that are each embedded with a pair of shape memory alloy (SMA) coils. By precisely controlling the activation sequence and duration of the SMA coils individually, the actuator performs reversible transitions among its six stable states within a few hundred milliseconds. To examine this multistable response, we establish a comprehensive transition diagram and corresponding working principles for the three major types of stable state transitions. Additionally, we parameterize the actuator's geometry and empirically examine the influence of the prestretch ratio on the actuator's geometry and deformation. Finally, we present several demonstrations that highlight the ability to utilize this multistable actuator in various practical applications. The first demonstration is of an actuator mounted with a miniaturized camera for visual inspection. The second involves a heliotropism‐inspired energy‐harvesting that a photovoltaic cell is mounted on the actuator and tracks a moving light source to harvest energy. The third involves a multistable actuator with directionally asymmetric frictional feet that is capable of transitioning between crawling and turning motions.",
"discussion": "3 Discussion and Conclusion This paper reports a soft actuator that is intrinsically multistable and capable of rapid transition between various configurations through the electrical activation of embedded nitinol coils. The actuator is composed of two elastomeric frames that are embedded with a pair of nitinol SMA coils and which are bonded to opposite sides of a prestretched elastic membrane. The actuator has six stable states: two pure‐bend states, named B1 and B2, and four bend‐twist states, named T1 + , T2 + , T1 − , and T2 − . Thanks to the combination of mixed‐mode snap‐through and SMA activation, the actuator is capable of rapid transitions among all six stable states at speeds on the order of hundreds of milliseconds. Among the various possible transitions, three representative types are highlighted in this work: i) transitions from a pure‐bend stable state to an opposing pure‐bend stable state; ii) transitions from a pure‐bend stable state to a bend‐twist state that manifests an opposite bending direction; iii) transitions from one bend‐twist state to another bend‐twist state, maintaining the same twist direction while modifying the bending direction to the opposite. By employing this control strategy, the system can effectively navigate between different stable states, enabling versatile and agile behavior. Furthermore, the actuator's geometry is parameterized, allowing for the measurement of documentation of bending and twisting angles in all stable states across different prestretch ratios. This parameterization facilitates the characterization of transitional behavior as well. A series of practical implementations are presented to demonstrate the potential applications of the multistable actuator. These applications encompass the visual inspection of confined spaces, the enhancement of energy harvesting based on principles of heliotropism, and a multimodal crawler with agile locomotion capabilities. By utilizing a single unit of this multistate actuator, rapid transitions and morphing can be achieved without the need for complex designs with multiple bistable actuator units. While promising, there are still areas in which the multistable actuator can be better understood or improved. First, the size of our multistable actuator is limited to the centimeter scale due to constraints imposed by the use of nitinol coils. In principle, it is possible to use other actuator technologies or to create a structure that is fully passive, as has been demonstrated in Movie S7 (Supporting Information) in which SMA is replaced with 3D‐printed elastic rods (Flexible 80A, Formlabs Inc.). The key element in our approach lies in providing multiple routes (bending and twisting) for the system to fulfill the “desire” of the prestretched membrane to contract. Consequently, the dimensions and geometry of the actuator can be freely scaled, while the actuation method should be modified correspondingly. For example, the actuator could potentially be reduced to the millimeter scale and triggered by the magnetic field. Moreover, the elastic potential energy profile shown in Figure 1B is just a schematic representation, while the theoretical computation of the potential energy still remains unsolved. While an analytic model has been proposed in Supporting Information, it is currently limited to predicting two pure‐bend stable states. This limitation arises from two key factors: i) The deformation of the frame in bend‐twist modes exhibits considerable complexity, surpassing the capabilities of our current analytic model in accurately assuming its geometry; ii) The prestretched membrane has a tendency to contract not only along its length but also its width when the structure bends, which causes the center of the frames to twist inwards – a deformation that is not currently accounted for in the analytic model. To overcome these limitations, a more complete computational simulation that accounts for the entirety of the nonlinear 3D deformation of the frames, coils, and membrane is required. One possible future research direction is using finite element analysis (FEA) to simulate the multistable actuator. However, to carefully and precisely simulate this complex structure, there are a few points of challenge to overcome. First, the frame consists of multiple materials (silicone elastomer and SMA coils) and has a complicated shape (channel for SMA coils and wavy surface pattern). This characteristic will result in discontinuity and nonlinearity of the system and would be computationally expensive to simulate. In addition, the mixed‐mode snap‐through involves both twisting and bending deformation, thus the correct convergence of the stable states is challenging. Moreover, the relaxation of the prestretched elastic membrane might also require consideration. Additionally, the merit of bistable and multistable actuators lies in their ability to undergo rapid transitions between stable states while maintaining stability without the need for additional energy input for control. However, the presence of external forces can impact the stable states, and in extreme occasions, the stable states can be lost. This could impede precise manipulation and interaction in practical applications. One potential solution is to employ a hybrid approach that incorporates snap‐through instability for large‐scale transitions, while locally controlling the system around the desired stable states when external forces are involved. Another possible approach is utilizing stiffness‐tuning and latching mechanisms to lock and secure the stable state. [ \n \n 32 \n \n ] By adjusting the stiffness of specific components to latch the actuator in a particular stable configuration, the actuator's stability and response to external forces could be improved."
} | 3,330 |
34417871 | PMC8379602 | pmc | 9,546 | {
"abstract": "The global marine environment has been impacted significantly by climate change. Ocean temperatures are rising, and the frequency, duration and intensity of marine heatwaves are increasing, particularly affecting coral reefs. Coral bleaching events are becoming more common, with less recovery time between events. Anomalous temperatures at the start of 2020 caused widespread bleaching across the Great Barrier Reef (GBR), extending to southern, previously less affected reefs such as One Tree Island. Here, nine video transects were conducted at One Tree Island, in the Capricorn Bunker Group, and analysed for community composition and diversity, and the extent of bleaching across taxa. Average live hard coral cover across the area was 11.62%, and almost half of this was identified as severely bleached. This bleaching event is concerning as it occurred in an area previously considered a potential refuge for corals and associated fauna from the risks of climate warming. Due to the global impacts of COVID-19 during 2020, this report provides one of potentially few monitoring efforts of coral bleaching. Supplementary information The online version contains supplementary material available at 10.1007/s10661-021-09330-5.",
"introduction": "Introduction The significance of coral reefs is unquestioned, both for ecosystem functioning (e.g. Glynn & Enochs, 2011 ; Lefcheck et al., 2019 ) and local human populations (e.g. Spalding et al., 2017 ; Januchowski‐Hartley et al., 2020 ). Due to climate change, marine heatwaves are increasing in frequency and duration (Oliver et al., 2018 ), with some lasting for years at a time (Di Lorenzo & Mantua, 2016 ). Such heatwaves are causing mass coral bleaching and mortality events (Hoegh-Guldberg, 2011 ; Couch et al., 2017 ), including among coral populations previously considered thermally tolerant (Le Nohaïc et al., 2017 ). The Great Barrier Reef (GBR) has experienced extensive bleaching, and the recent bleaching event in 2020, a non El Niño year, is the third widespread event since 2015 (Hughes et al., 2018 ). The most recent bleaching events affecting the GBR have occurred at regional scales, with the 2016 event most severely impacting reefs in the Far North and Northern GBR (Wolanski et al., 2017 ) and the 2017 event mostly affecting reefs in the North and Central GBR (Hughes et al., 2019 ). These events spared the reefs in the Southern GBR, although localised bleaching events have been recorded in earlier years (e.g. Booth & Beretta, 2002 ; Ortiz et al., 2009 ). Here, we report on bleaching on the Southern GBR at One Tree Island during 2020. Due to the worldwide COVID-19 pandemic and resulting travel restrictions, in situ observations of the recent GBR bleaching event have been limited. Many research and boating activities were stopped, preventing important quantitative surveys of bleaching impacts (Bates et al., 2020 ), and increasing the value of any records collected during this time. Here, we analyse video transects taken in February 2020, shortly before research activities were restricted, showing the severity of bleaching at One Tree Island in the Southern GBR.",
"discussion": "Discussion The observed severe bleaching in the One Tree Island lagoon followed months of regionally record-breaking temperatures. Mean air temperatures recorded between November 2019 and February 2020 at the nearby station of Seventeen Seventy were the hottest since records began in the 1980s (Commonwealth of Australia Bureau of Meteorology, 2020 ). Daily average water temperatures in the lagoon in February were up to 3.2 °C higher in 2020 than over the previous decade. Additionally, temperature anomalies (above 27.5 °C) during February 2020 were recorded every day (29 days) compared to 14 and 8 days in 2018 and 2019 respectively, or 0 days in the long-term average where data was available (2000–2009 and 2017–2018: Fig. 4 a). In the month following this data collection, sea surface temperatures sharply dropped from a daily average of approximately 29 to 26 °C (Fig. 4 b), likely providing relief for the affected colonies. However, coral recovery is a complex and variable process, which differs with species and colony size (Baird & Marshall, 2002 ) and is further influenced by physical characteristics, such as the frequency of cyclones or the presence of Crown of Thorns starfish. Coral recovery on the GBR has declined significantly in recent years (Ortiz et al., 2018 ) and may continue to decline in the future as the increasing frequency of bleaching events reduces recovery time between disturbances (Dietzel et al., 2020 ). Fig. 4 Temperature patterns at One Tree Island. a Data from 11 August 2019 to 20 June 2020. Data recorded at 2.3 m depth at 00:00 daily. Dashed vertical line indicates date when transects were recorded. Raw data downloaded from AIMS (data.aims.gov.au). b Temperature data from One Tree Island Third Lagoon for February 2020 (upper blue line) compared to long-term February averages for 2000–2009 and 2017–2018 (excluding 7 years where records were unavailable; lower black line). Error bars show standard deviation for long term data. Data from temperature loggers deployed by One Tree Island Research Station The Southern GBR has been largely unaffected by mass bleaching events in the past (Kennedy et al., 2018 ). This bleaching event is the first major bleaching to affect One Tree Island since 1998 where 10–30% of corals bleached (Berkelmans et al., 2004a , b ). Due to the relatively low bleaching incidence in the past, the southern GBR has been suggested as an area with potential to withstand temperature-related coral cover loss. This is in part due to restocking of coral larvae from further north, where individuals are adapted to higher temperatures (Matz et al., 2020 ). The present study has shown that these Southern reefs are still susceptible to widespread bleaching events. It will be important to closely follow the recovery of Southern GBR reefs, potentially looking at One Tree Island as a case study, in order to understand the potential of these reefs to act as a refuge under future climate scenarios."
} | 1,533 |
35629910 | PMC9146292 | pmc | 9,547 | {
"abstract": "Rapid proliferation of cyanobacteria in both benthic and suspended (planktonic) habitats is a major threat to environmental safety, as they produce nuisance compounds such as cytotoxins and off-flavors, which degrade the safety and quality of water supplies. Temperature and light irradiance are two of the key factors in regulating the occurrence of algal blooms and production of major off-flavors. However, the role of these factors in regulating the growth and metabolism is poorly explored for both benthic and planktonic cyanobacteria. To fill this gap, we studied the effects of light and temperature on the growth and metabolic profiling of both benthic ( Hapalosiphon sp. MRB220) and planktonic ( Planktothricoides sp. SR001) environmental species collected from a freshwater reservoir in Singapore. Moreover, this study is the first report on the metabolic profiling of cyanobacteria belonging to two different habitats in response to altered environmental conditions. The highest growth rate of both species was observed at the highest light intensity (100 μmol photons/m²/s) and at a temperature of 33 °C. Systematic metabolite profiling analysis suggested that temperature had a more profound effect on metabolome of the Hapalosiphon , whereas light had a greater effect in the case of Planktothricoides . Interestingly, Planktothricoides sp. SR001 showed a specialized adaptation mechanism via biosynthesis of arginine, and metabolism of cysteine and methionine to survive and withstand higher temperatures of 38 °C and higher. Hence, the mode of strategies for coping with different light and temperature conditions was correlated with the growth and alteration in metabolic activities for physiological and ecological adaptations in both species. In addition, we putatively identified a number of unique metabolites with a broad range of antimicrobial activities in both species in response to both light and temperature. These metabolites could play a role in the dominant behavior of these species in suppressing competition during bloom formation. Overall, this study elucidated novel insights into the effects of environmental factors on the growth, metabolism, and adaptation strategies of cyanobacteria from two different habitats, and could be useful in controlling their harmful effects on human health and environmental concerns.",
"introduction": "1. Introduction Cyanobacterial blooms in water bodies are an increasing worldwide threat to both human and aquatic organisms. Some of the cyanobacterial species multiply rapidly and form benthic or planktonic blooms due to an increase in unfavorable environmental conditions, such as water eutrophication due to urbanization and agriculture, global warming [ 1 , 2 , 3 , 4 ], and water pollution by herbicides [ 5 ] and industrial waste. These blooms have severe impacts on the ecosystem, such as changes in biodiversity, light, and oxygen concentration and the habitat of aquatic organisms. These benthic or planktonic blooms can produce a wide range of noxious secondary metabolites, including potent toxins and odors [ 6 ] such as microcystins (MCs) and cylindrospermopsin [ 7 , 8 ], which can influence terrestrial and aquatic organisms [ 9 ], as well as human health. Hence, it is essential to restrict the risks of cyanotoxins, as well as the production of off-flavor compounds such as geosmin and 2-methylisoborneol (2-MIB), to reduce these incidents’ occurrence due to both planktonic and benthic cyanobacteria. The off-flavor compounds produced by both planktonic or mats of benthic type cyanobacteria are a widespread odor problem in drinking water and fish production worldwide [ 10 , 11 ]. The muddy or earthy–musty odor in the water usually occurs due to the secretion of volatile metabolites such as geosmin and 2-MIB. Both 2-MIB and geosmin are terpenoids, and are synthetized by terpene synthases [ 12 ]. Temperature and light irradiance are the key factors in regulating the occurrence of algal blooms and production of both 2-MIB and geosmin through various upstream primary metabolic processes [ 11 , 13 ]. Genes involved in the biosynthesis of 2-MIB were found in cyanobacteria in [ 14 ], and the authors suggested that the transcription of 2-MIB synthesis genes was light-regulated. The production of the off-flavor compound 2-MIB in response to different light intensities and temperature has been experimentally verified in a free-living filamentous planktonic cyanobacterium —Planktothricoides sp. SR001—obtained from a freshwater reservoir in Singapore [ 14 ]. They also identified the genes associated with the production of this compound. Similarly, the benthic cyanobacterium Hapalosiphon sp. strain MRB 220 isolated from a benthic cyanobacterial mat gathered from a sediment sample of an urban freshwater water body in Singapore also demonstrated the capability of nitrogen fixation and production of the off-flavor compounds 2-MIB and geosmin [ 15 ]. Temperature is one of the most important factors in the growth and metabolism of cyanobacteria in the absence of nutrient shortage [ 16 ], as most of the metabolic reactions are temperature-dependent. It has been shown that cyanobacteria outcompeted other algae at elevated temperature, and their toxin content also increased with higher growth rates [ 17 , 18 , 19 ]. Changes in temperature often mediate production of secondary metabolites that are coupled with photosynthesis [ 20 , 21 ]. In addition, the higher production of these metabolites is also coupled with the elevated temperature, which is optimal for the growth of the cyanobacteria [ 22 ]. Similarly, light plays a key role in the growth and pigment accumulation of cyanobacteria. Changes in light irradiance significantly affect pigment composition that is linked to light-harvesting mechanisms [ 23 , 24 ]. These changes in pigmentation are mainly due to the variations of red-colored phycoerythrin and blue-colored phycocyanin [ 25 ], and any slight change in the light-harvesting mechanism may lead to changes in the morphology and physiology of cyanobacteria. Nevertheless, the effects of environmental factors on the growth and metabolic pathways of algal blooms are not fully understood yet. To date, the majority of the studies on cyanobacterial blooms have focused on the model organism Microcystis and the bloom-producing planktonic forms of cyanobacteria that are associated with toxicity, which are risky for the aquatic ecosystem [ 26 , 27 ]. However, studies on benthic cyanobacteria that produce the common toxin microcystin have been gaining attention in the scientific community [ 28 , 29 , 30 ]. Benthic and planktonic cyanobacteria are ecologically distinct organisms that bloom in different habitats and have different evolutionary histories, which could be associated with their specific metabolic machineries. Hapalosiphon sp. is a nitrogen-fixing filamentous cyanobacteria, whereas Planktothricoides sp. is a branched cyanobacteria without the ability to fix atmospheric nitrogen. The mode of survival strategies, off-flavor production, and the impact of environmental factors on the production of microcystin and their metabolic networks are not well understood yet. Hence, we aimed to explore the impact of environmental factors such as light and temperature on the growth behavior and primary metabolic networks of both Hapalosiphon sp. MRB220, a benthic strain isolated from the benthos in Singapore, and Planktothricoides sp. SR001, a planktonic strain isolated from a freshwater reservoir in Singapore. A number of studies so far were conducted mainly to characterize different toxin-producing cyanobacteria [ 31 , 32 ]. However, the study of the effects of environmental conditions that promote the proliferation of both benthic and planktonic freshwater cyanobacteria is very limited. The recent advancement in omics approaches provides insights associated with the physiological and biochemical characteristic linked to the proliferation of these cyanobacteria. Environmental metabolomics approaches have been utilized to assess the organism–environment interactions at the molecular level to improve the understanding of their mode of action [ 33 , 34 ]. Through this application, identification of different metabolites assists in detecting alterations in certain metabolic pathways under environmental stresses [ 35 ]. It also helps to identify the cellular activity of the organism in response to different contaminant-related stressors. Several metabolomics technologies have been developed to improve the measurement processes for specific groups, classes, and individual metabolites. This metabolomics approach has been employed to analyze different metabolic networks in cyanobacteria, such as inorganic carbon acclimation, glycogen biosynthesis processes, and structural diversity of metabolites [ 36 , 37 , 38 ]. Therefore, we aimed to utilize the metabolomics approach to elucidate the effects of environmental factors on the cellular/metabolic processes of both Hapalosiphon sp. and Planktothricoides sp. This will assist in improving our understanding of the mode of action and proliferation of these two cyanobacteria in response to light and temperature, which have structural and behavioral differences. Additionally, it may provide information on the production of multiple biofuels and bioproducts for biotechnology application, as well as the development of models for management of eutrophication problems.",
"discussion": "3. Discussion Alterations in environmental variables such as temperature, light intensity, partial pressure of water, high nutrients, or low turbidity could trigger an increase in the cyanobacterial bloom in both frequency and intensity, subsequently affecting metabolic activities for the production and release of algal off-flavors/toxins/pharmaceutical compounds [ 40 , 41 ]. The metabolic profiling of both species studied here was important, as it provided essential metabolism information linked to their growth and compound biosynthesis. 3.1. Effects of Different Intensities of Light on the Growth and Metabolic Profiling of Hapalosiphon sp. MBR220 and Planktothricoides sp. SR001 Our findings indicated that a relatively high light intensity (100 ± 3 μmol photons m −2 s −1 ) was necessary for the maximum growth of Hapalosiphon sp. ( Table 1 ). The effect of light intensity on the growth rate of other free-floating cyanobacteria such as Planktothrix sp. showed that a light intensity of 85 μmol photons m −2 s −1 was optimum for growth in laboratory conditions [ 42 ], whereas for the benthic Hapalosiphon sp ., the optimum light requirement was observed at 100 ± 3 μmol photons m −2 s −1 in laboratory conditions, which was slightly higher than the requirement for Planktothrix sp. The growth of Planktothricoides sp. was also maximized at a high light intensity (100 ± 3 μmol photons m −2 s −1 ), which could be correlated with their planktonic habitat, as they normally bloom at the surface of water where light intensities are stronger ( Table 1 ). The metabolic profiling suggested that light had some impact on the biosynthesis and metabolism of different amino acids (alanine, aspartate and glutamate, glycine, serine, threonine, arginine, and proline), leading to higher syntheses of proteins, nucleic acids, and regulatory molecules ( Figure 1 ). These amino acids play many roles in the defense system, such as osmotic regulators, free radical scavengers, and stress signal molecules [ 43 , 44 , 45 ]. Although it has been reported that environmental conditions affected the content of amino acids in cyanobacteria, the effects of light and temperature on amino acid content has not been sufficiently studied [ 46 , 47 ]. Glutamate plays an important role in nitrogen metabolism [ 48 ], chlorophyll biosynthesis [ 49 ], and the adaptation of cyanobacteria to higher light intensities. Asparagine can reabsorb the released free amino acids to reduce the toxic effect of ammonia [ 50 , 51 ]. Serine, threonine, and tyrosine might be involved in the metabolic reactions associated with photosynthesis [ 52 ]. The intensity and quality of light also activate various signal transduction pathways that regulate physiological adaptation [ 53 ]. The identification of riboflavin metabolism can be linked to the growth of Hapalosiphon sp., as it is linked to chlorophyll and porphyrin metabolism through the vitamin B12 pathway. Benthic cyanobacteria could produce a large amount of vitamin B complex and vitamin E [ 54 ], and release the excess quantity to their surrounding environment. It has been reported that nitrogen-fixing cyanobacteria excreted more vitamin B12 compared to non-nitrogen-fixers [ 55 ]. The identification of the riboflavin pathway also suggested that this species of benthic cyanobacteria can be used as a valuable source of vitamins for commercial purposes [ 56 ]. Our results showed that under high light conditions, Hapalosiphon sp. grew through primary metabolic pathways as well as with environmental pollutants as a growth substrate, which showed their roles in the degradation of industrial pollutants and in cleaning up contaminated environments. The presence of C5-branched dibasic acid metabolism could be associated with energy production during high light intensity. Overall, the effects of different light intensities on the metabolites of Hapalosiphon sp. suggested that primary energy-producing metabolic pathways and amino acid metabolism are the main metabolic pathways that could have an impact on their growth and survival, which was related to their benthic habitat ( Figure 1 ). Surprisingly, we did not identify any metabolite directly associated with secondary metabolic pathways. Light intensity showed a greater effect on the metabolic profiling of Planktothricoides sp. compared to that of benthic Hapalosiphon sp., as higher number of different amino acid biosynthesis processes and metabolites associated with different primary and secondary metabolic pathways were identified ( Figure 2 ). A higher number of secondary metabolites showed an enhancement of different secondary metabolic pathways in this species in response to different light intensities. Since these secondary metabolites are not vital for the survival of Hapalosiphon sp., these could be produced in response to high light intensities. The function involved could be for photoprotection, biodegradation, antioxidant activity, and defense against the response to abiotic and biotic stresses [ 57 ]. Carotenoids provide photoprotection of the intracellular molecules in cyanobacteria [ 58 , 59 ]. However, the content of carotenoids is strain-specific and depends on different environmental/culture conditions, particularly light intensity [ 60 , 61 ]. In addition, accumulation of b-carotene could lead to the biosynthesis of abscisic acid (ABA) to regulate the stress-response mechanism during abiotic stress responses ( Figure 2 ). Hence, microalgal ABA signaling may share some functions with higher plants, particularly in the stress-response mechanism. It was reported that extracellular ABA was produced in response to salt stress by the cyanobacteria Nostocl nuscorum, Trichormus uariabilisa, and Synech. Ococculeso poliens in a culture medium [ 62 ]. In addition, these cyanobacteria producing different terpenoids in response to different light intensities could be beneficial for synthetic production of pharmaceutical and industrial compounds. For instance, Hapalosiphon sp. has the ability to biodegrade amino benzoate and benzoate under varied light conditions. Interestingly, the identification of metabolites linked to streptomycin biosynthesis suggested that they could be used synthetically as a target for antibiotics. However, the metabolic activity was still not fully functional compared to that in the growth stage, which could be due to an insufficient light intensity compared to their growth in natural habitat. The effects of different light intensities on metabolite identification in both species, which differ in habitat and structure, clearly indicated that they responded to a change in environmental conditions according to their habitat. Higher light intensities can also be linked to the bloom-forming potential of the planktonic cyanobacterium Planktothricoides sp., which has more metabolic activity with metabolic partitioning for both growth and the production of secondary metabolites. These broad groups of secondary metabolites have beneficial uses, such as in pharmaceutical and industrial compounds, as well as in the biodegradation and production of vitamins ( Figure 2 ). In addition, different light intensities had a major impact on the photosynthetic efficiency of Planktothricoides sp., which was related to their habitat. These species possess gas vesicles, and these vesicles are more abundant when the light intensity is reduced. Hence, at a low light level, the growth rate became slow and led to a lower production of secondary metabolites. This buoyancy regulation has a lot of advantages in comparison with other phytoplankton species. 3.2. Effects of Different Temperatures on the Growth and Metabolic Profiling of Hapalosiphon sp. MBR220 and Planktothricoides sp. SR001 In our analysis, the growth rate of Hapalosiphon sp. was found to be the highest at 33 °C, similar to what we found for Planktothricoides sp. ( Table 2 ). Experimental evidence showed a positive correlation between cyanobacterial biomass levels and an increase in water temperatures [ 63 , 64 ]. It also was shown that the upper thermal limit (UTL) for the growth of different cyanobacteria in laboratory conditions was species-dependent, and varies from ~30 to 35 °C; in many species, the UTL is <33 °C [ 65 ]. However, the mechanism of this effect of temperature on increasing the algal bloom and off-flavor production is not known yet. This could be related to their habitat or adaptation to a changing environment, or a combination of both. Hence, a systematic understanding of the potential effects of both light and temperature through metabolite profiling would provide some insight into the metabolic activities of these cyanobacteria. The metabolic profiling of Hapalosiphon sp. suggested that at varied temperatures, they produced both primary and secondary metabolites ( Figure 3 ). Many of them played an important part in the defense mechanism to survive in complex environments, and provided resources to increase their competences regarding bloom development. This coincided with the information that an increase in temperature from mid to late summer caused massive blooms of a benthic cyanobacteria belonging to the genus Ostreopsis , which is harmful to human health in some urbanized temperate areas [ 65 ]. Temperature demonstrated a more profound effect than light intensity, with a higher number of amino acids involved in the biosynthesis/metabolism. The metabolism of phenylalanine, a precursor of secondary metabolites such as antitoxins, alkaloids, lignin, and flavonoids, in response to different temperature suggested that these metabolites might play a key role in regulating tolerance to temperature/heat stress. Hence, temperature could influence the expression of P450 enzymes in the thylakoid to participate in the phenylpropanoid biosynthesis. However, some of the metabolites associated with phenylpropanoid biosynthesis pathway could not be verified due to a lack of information in the cyanobacterial databases. The identification of tyrosine metabolism can be linked to aspartate metabolism and the synthesis of hormonelike molecules, as well as flagellar assembly (quorum sensing). It is also an important node in the cAMP signaling pathway. However, the presence of flagella is not known in cyanobacteria, yet they depend on the T4P for motility, a protein filament that extends from a membrane-spanning pore complex for mobility. This allows them to move toward a favorable environment from an unfavorable stress condition, such as intensity of light/high temperature. The cAMP signaling pathway is important, as it plays a key role in the regulation of various biological activities by controlling gene expression levels in respiration, light sensing, cell motility [ 66 ], and sensing with carbon acquisition [ 67 ]. In cyanobacteria, cAMP represents a high carbon signal [ 68 ]. During abiotic stress conditions, this signaling pathway plays a major role in the adaptation to the stress condition [ 69 ]. Interestingly, we identified a metabolite that was linked to lysine degradation, and this degradation could probably fuel this pathway for the biosynthesis of antibiotics, which was not observed in the response to different light intensities of this species. Pathways linked to antibiotics such as streptomycin were also observed in Hapalosiphon sp. in response to the temperature effect. Surprisingly, different temperature levels had some pronounced effects on the metabolic activities of Planktothricoides sp. ( Figure 4 ). Only a few metabolites linked to the biosynthesis of arginine and the metabolism of cysteine and methionine were identified. Arginine is a major storage and transport form of amino acid, the precursor for polyamines and nitric oxide (NO), and an essential metabolite for cellular and developmental processes. This amino acid can act as a signaling molecule that regulates essential cellular functions such as protein synthesis, apoptosis, and growth during critical conditions. It could also be acting as a twin-arginine translocation (Tat) pathway involved in the transport of proteins into and across the plasma and thylakoid membranes [ 70 ]. In cyanobacteria, the plasma and thylakoid membranes carry out very different activities, and arginine could play a key role in the transport of proteins. The metabolism of methionine is necessary for growth and development, as it acts as an initiator of protein synthesis ( Figure 4 ). In addition, it functions as an endogenous antioxidant and redox sensor during unfavorable conditions [ 71 , 72 ]. It also increases the survival rate by modulating autophagy, or by inducing mitochondrial function and antioxidant defense [ 73 ]. Likewise, cysteine inhibits several enzymes of amino acid synthesis; therefore, increasing cysteine concentrations could increase the levels of the inhibited enzymes. It acts as a limiting factor for glutathione biosynthesis, which can be especially crucial for cyanobacteria, which rely on both the sufficient sulfur supply from the growth media and on the protection of glutathione against ROS that are produced during photosynthesis. Hence, it has the power to protect against oxidative stress. Although the metabolic activities were restricted at certain temperatures, this species still had the potential to degrade aromatic compounds such as naphthalene, benzene, and xylene to provide substrates for growth. Even though the effects of high temperature restricted the growth and metabolic activities of Planktothricoides sp., they clearly suggested the mechanism of adaptation to higher temperatures by increasing the survival rate through protection and antioxidant defense. This behavior was related to the habitat of this species, as they normally grow in local reservoirs in Singapore where the temperature varies between 28 and 34° C. This cyanobacterium can sustain many types of chemical stresses due to its capability for metabolic adaptations, and can also be used to clean up ccontaminated environments. The metabolite analysis also coincided with the observation that the growth of Planktothricoides sp. declined at the highest temperature; i.e., 38 °C. The overall results suggested that different temperatures had more insightful effects on the metabolome of Hapalosiphon sp. compared to the effects of different light intensities. They survived higher light conditions through primary energy-producing metabolic pathways and amino acid metabolism. In contrast, the response of Planktothricoides sp. to different light intensities and temperatures was opposite to that of the Hapalosiphon sp., which could be related to their habitats, potentials for bloom formation, and structural differences. The light intensity had a more profound effect on the metabolome of this planktonic species compared to the effects of temperature. This suggested that the planktonic species has a higher tolerance capacity for high light intensities through carotenoid production, which protects the cells from photoprotection. The stress-responsive amino acid proline in Planktothricoides sp. could possibly play a role in scavenging reactive oxygen species (ROS) by stimulating an antioxidant defense mechanism in response to high light intensities. Similarly, ascorbate metabolism could be protecting the cyanobacterial cells from damage by ROS. The metabolism of methane suggests an alternative carbon source for their growth and metabolism [ 74 ]. Interestingly, they have the ability to sustain different types of chemical stresses through metabolic adaptation by degrading a number of aromatic compounds in response to high light intensities. Moreover, they showed the abilities to cope with imbalances in response to higher-temperature conditions through the accumulation of nitrogen storage compounds, acclimatize to higher temperature conditions through different survival and protection mechanisms, and degrade industrial pollutants. This is the first report on the metabolic profiling of two diverse cyanobacteria from different habitats in response to environmental changes. The information gathered from our analysis can be the basis for understanding the alterations in the metabolic activities of these two taxa for physiological and environmental adaptations. In addition, it also can assist in guiding biotechnological applications and strain designs. A more detailed analysis of endogenous and exogenous metabolites and their localizations for specific classes/groups will help in deciphering the variations in enzyme level and gene expression patterns associated with cellular responses to environmental stimuli and their combinations. This work on metabolic profiling suggested that a number of targeted metabolomics could be conducted to quantify metabolic changes under such environmental factors/stressors that would elucidate dominating metabolic pathways more precisely. There are also numerous toxic cyanopeptides produced by cyanobacteria besides the well-known microcystins [ 75 , 76 ]. Recent developments in advanced analytical techniques have facilitated the identification of those cyanopeptides. In the future, it would be interesting to identify the effects of light and temperature on different cyanopeptides. Additionally, a comprehensive understanding of the impact of environmental factors on cyanobacterial characteristics and an evaluation of the risk of cyanotoxin generation can help in managing the bloom control in and quality of water."
} | 6,759 |
33024198 | PMC7538881 | pmc | 9,549 | {
"abstract": "In recent years, most biofilm studies have focused on fundamental investigations using multispecies biofilm models developed preferentially in simulated naturally occurring low-nutrient medium than in artificial nutrient-rich medium. Because biofilm development under low-nutrient growth media is slow, natural media are often supplemented with an additional carbon source to increase the rate of biofilm formation. However, there are knowledge gaps in interpreting the effects of such supplementation on the resulting biofilm in terms of structure and microbial community composition. We investigated the effects of supplementation of a simulated freshwater medium with sodium citrate on the resulting structure, bacterial community composition, and microbial network interactions of an early-stage multispecies biofilm model. Qualitative and quantitative analyses of acquired confocal laser scanning microscopy data confirmed that sodium citrate supplementation distinctly increased biofilm biomass. Sequencing data revealed that the microbial community structure of biofilms grown in sodium citrate-supplemented conditions was characterized with increased relative abundance and dominance of Proteobacteria compared with that of biofilms grown in sodium citrate-free conditions. Our findings suggest that the supplementation of a low-nutrient medium with a carbon source in experiments involving multispecies biofilms may lead to structural and compositional biases of the microbial community, causing changes in biofilm phenotype.",
"conclusion": "Conclusion Our study revealed that adding sodium citrate, a carbon source, to an early-stage multispecies biofilm model significantly affects the biofilm 3D structure, characteristics, bacterial community composition, and network interconnection, thus generating a biased biofilm model for the investigation. The findings provide caveats for investigators intending to accelerate biofilm growth by adding a carbon source in their low-nutrient media for generating multispecies biofilm models.",
"introduction": "Introduction Although the use of in vitro monospecies biofilms and nutrient-rich media has been widely applied in the context of fundamental investigations, realistic multispecies biofilm models have gained much attention in recent years. Given that the microbial systems in natural and engineered environments usually comprise consortia of highly diverse microbial communities organized in the form of biofilms, investigations focusing on multispecies biofilms has emerged as a relevant topic for elucidating the behavior of targeted microorganisms within complex systems 1 , 2 .\n Compared to monospecies biofilm development, the establishment of multispecies biofilm model is a challenging feat, considering the existing complex biological processes linked to their high bacterial diversity 3 where various forms of interspecies interactions are involved 4 . Therefore, simulating multispecies biofilm within a laboratory setting has always been considered an arduous task, which explains why most published studies have opted to investigating monospecies biofilm models or, in the best case, biofilm models involving two to three different species. Nevertheless, one of the caveat of performing in vitro experiments is associated with the choice of growth medium, regardless of which type of monospecies or multispecies biofilm model is being investigated. In situations where the composition of the inoculum is known, an appropriate medium can be selected for optimal biofilm growth. For monospecies biofilms, past studies have shown that biofilm development outcomes are significantly affected by the selected growth medium 5 , 6 . In oral biofilm related research, where both monoculture and multispecies biofilm models are typically employed, growth media as well as specific nutrients were also shown to affect growth rate and biofilm composition outcomes 7 – 12 . More specifically, ecological plaque hypothesis (a hypothesis to explain the relationship between the dental plaque microflora and the host in health and disease, and to identify new strategies for disease prevention) proposed by Marsh assumed that the discrepancy in the microbial composition of oral biofilms impacted by both environmental factors and nutrient availability was found to contribute to the enrichment of disease-associated pathogens 13 . However, in experiments in which biofilm models constituting a plethora of microorganisms need to be investigated, the growth medium is generally obtained from a nonsterile natural or simulated source 14 , 15 . In experiments involving the use of nonsterile medium, the issue of repeatability or inconsistent biofilm outcomes may arise due to the dynamic chemical conditions of the natural source and the high diversity of initial microorganisms. For instance, the nutrient contents in freshwater environments can be affected by random factors such as weather alteration, seasonal shifts, or anthropogenic activities 16 , 17 . Moreover, the speed of biofilm formation may be low in freshwater environments because unlike nutrient-rich media, nonsterile natural media usually contain lower levels of essential nutrients. For example, in nutrient-rich medium, biofilm formation generally occurs within one week, whereas in the natural river medium, at least 15 days are required for the formation of a thin patchy biofilm 18 . One strategy commonly used to accelerate biofilm formation in nutrient-poor media is the addition of an organic carbon source. In biofilm experiments using nutrient poor media, sodium citrate or sodium acetate is occasionally supplemented as the sole carbon source for accelerating biofilm development in monospecies biofilm models 19 – 22 or multispecies consortia 23 – 26 . It is through carbon source supplementation that such studies were able to accelerate biofilm development processes needed for conducting timely quantitative and qualitative biofilm analyses, thereby providing an attractive means for experimental productivity. Though the addition of substrates, such as sodium citrate, is an ideal way to increase the rate of biofouling, biofilm formation, and biomass level in monospecies models, it is unclear whether such addition is similarly beneficial in multispecies models. In multispecies biofilm models, we hypothesized that the addition of sodium citrate may alter the multispecies community composition as well as its microbial succession. While sodium citrate is likely to accelerate biofilm development in multispecies biofilms, how substrate addition may affect experimental outcomes in terms of biofilm structure and microbial community composition remains unclear, thereby causing potential experimental biases. There is currently a plethora of various in vitro biofilm models used in the context of environmental or health-related research, however, the selection of either dynamic- or static-forms of systems will ultimately depend on the original conditions the selected model was meant to simulate. While, static biofilm harvesting systems are perhaps the most widely used in research due to their simplicity and other positive attributes 12 , 27 , 28 , they are not ideal in research experiments meant to simulate freshwater environments characterized by continual flow and shear conditions, as well as constant renewal of nutrients. In the present study, we sought to investigate the effects of sodium citrate supplementation on a previously described multispecies biofilm models, using a dynamic simulated irrigation water distribution system 29 . The structure and microbial community composition of biofilms in the absence and presence of sodium citrate were compared at different time points of their early-stage formation.",
"discussion": "Discussion Sodium citrate and sodium acetate are the most commonly used carbon sources for supplementing natural or artificial low-nutrient media to accelerate the formation of monospecies or multispecies biofilm in laboratory-based biofilm experiments 19 – 26 . However, the potential biased effects caused by supplementation with such carbon sources on the structural attributes, microbial community composition, and microbial interaction networks of biofilms remain unexplored. To close this knowledge gap, we investigated the effects of a commonly used carbon source, sodium citrate, on the structure, microbial community composition, and microbial interactions of an early-stage multispecies biofilm model previously described in the context of simulated irrigation water distribution systems 29 . The results of our analyses revealed that the addition of sodium citrate creates a biased biofilm model in terms of an inconsistent microbial composition that may affect the outcome of the resulting biofilm. With respect to biofilm biomass accumulation, as expected the supplementation of a nutrient-poor artificial medium with sodium citrate successfully increased the rate of biofilm formation, as indicated by the enhanced 3D biofilm structure. Notably, the total biovolume, mean thickness, and roughness parameters increased significantly in biofilms grown in sodium citrate-supplemented medium. This finding is in accordance with findings in previous studies in which supplementary carbon source could increase microbial biomass 40 , 41 . Bester et al. 40 revealed that compared to carbon-limited conditions, carbon-replete environments led to 40% increase in planktonic cell yields of Pseudomonas sp . biofilms. Another study showed that carbon sources such as glucose, sucrose and acetate presented a significant enhancement in microalgae growth 41 . In particular, comparison of the microbial community profiles of biofilms between sodium citrate-supplemented and sodium citrate-free conditions revealed that the sodium citrate supplementation resulted in experimental bias in the resulting biofilm phenotypes. Sodium citrate supplementation increased the abundance of the Proteobacteria phylum (70.0%) such that it became the most dominant phylum under sodium citrate conditions. This is a phylum of gram-negative bacteria that includes a wide variety of pathogenic genera, such as Salmonella , Vibrio , Helicobacter , Yersinia , Legionellales , and Pseudomonas 42 . Such a shift in the microbial community structure may be attributable to the fact that some genera under Proteobacteria favor citrate as their sole carbon source such as Salmonella 43 . This shift also caused significant changes in the microbial network dynamics in terms of increased co-exclusion interactions of Proteobacteria with other phyla, suggesting the potential nutritional advantage to Proteobacteria. Previous studies have described biofilms primarily composed of Proteobacteria, could influence the overall structural attributes, due to their distinct extracellular polymeric substances (EPS) and biofilm viscoelastic properties 44 . In this study, the thickness and roughness of biofilms grown under sodium citrate conditions varied toward the end of the experiment on day 21 and 28, which could be explained by the potential sloughing of the biofilms in the reactor. Sloughing have been demonstrated as one of the distinct models of biofilm dispersal, for biofilm suddenly sheds a portion of its structure especially at a later stage 45 – 47 . However, whether the sloughing was attributable to both the significant biofilm thickness and shear conditions within the reactor or to the intrinsic viscoelastic property of the biofilms allowing for effortless sloughing still remains unclear and should be investigated in the future. In sharp contrast, biofilms grown under sodium citrate-free conditions showed less susceptibility to sloughing under experimental shear conditions, indicating that the high abundance of Actinobacteria may have increased the cohesiveness of the biofilm. Previous investigations on Actinobacteria biofilms have revealed that these biofilms are highly adherent and may promote the recruitment of other cells to its structure. In one recent study, Micrococcus luteus , a species of the Actinobacteria phylum, was found to enhance the adhesion of other bacteria to the biofilm 48 . Two other species of the Actinobacteria phylum, Corynebacterium renale and C. pilosum , are also associated with increased adherence to various bacteria 49 . Future studies are clearly warranted to further elucidate the impact of microbial community structure on the viscoelastic property of multispecies biofilms. The observed differences in the microbial community profiles caused by sodium citrate supplementation in low-nutrient media may lead to experimental biases in studies focusing on the structural properties of biofilms or specific microbial communities. In the former case, supplementation with sodium citrate or another carbon source to accelerate the turnover of multispecies biofilm models for testing biofilm mitigation strategies could result in the design or engineering of antifouling features that may be inefficient in natural conditions. As an important feature in many types of biofilms including oral, environmental and medical biofilms 50 , the viscoelasticity property of biofilms serves a protective role from external stresses in the form of chemical or mechanical challenges 51 . More specifically, most biofilms in their natural habitats are exposed to external forces applied in a compressive, tensile or shear modes 51 . Given that biofilm EPS are characterized as a multi-component biological material 52 – 54 , biofilm developments is influenced in a time-dependent manner with series of shifts in elastic and viscous properties. Since biofilm viscoelasticity properties are closely related to biofilm structure and composition, the present study provides a caveat in which researchers should also consider the effects of nutrient supplementation when studying biofilm viscoelasticity and their related response to induced stresses; and by doing so, avoiding potential biased outcomes. In a microbial ecological context, the beneficial faster turnover of the biofilm model resulting from carbon source supplementation may cause a serious misrepresentation of the microbial community intended for fundamental investigations. In this study, sodium citrate supplementation significantly decreased the relative abundance of Actinobacteria among other phyla. Such biases may hamper accurate fundamental studies on biofilms, especially multispecies biofilm models, in specific simulated environmental conditions. Although it is understandable that the use of carbon source to accelerate the time-consuming biofilm formation in experimental models is essential, the results of this study suggest the importance of carefully considering the potential biases caused by carbon source supplementation, as they may significantly affect the experimental results. The potential biases in biofilms grown in supplemented media include shifts in the biofilm structure and microbial composition compared with those in biofilms grown in non-supplemented media."
} | 3,757 |
39157752 | PMC11328508 | pmc | 9,550 | {
"abstract": "Plastic material performance is strongly correlated to the polymer's molecular weight. Obtaining a sufficiently high molecular weight is therefore a key goal of polymerization processes. The most important polyester polyethylene terephthalate (PET) and the new polyethylene furanoate (PEF) require metal catalysts and time-consuming production processes to reach sufficiently high molecular weights. Metal catalysts, which are typically antimony or tin for polyesters, end up in the plastic products which may result in sustainability and ecological challenges. When the less reactive comonomer isosorbide is introduced to produce (partly) biobased materials with enhanced thermal properties, such as polyethylene- co -isosorbide furanoate (PEIF), reaching high enough molecular weight becomes even more challenging. This study presents an easily implementable approach to produce high molecular weight PET and PEF polyesters and their isosorbide copolyesters PEIT and PEIF by coupling lower molecular weight polymer chains by the reactive diguaiacyl oxalate (DGO) chain extender. DGO is so reactive, that the use of metal catalysts can be completely avoided and it helps avoiding an extra solid-state polymerization step. In addition, DGO distinguishes itself from typical chain extenders by its ability to be completely removed from the resulting polymer, thereby avoiding the inherent drawbacks associated with typical chain extenders.",
"conclusion": "Conclusions This research has shown that DGO can be effectively used as molecular weight booster to generate high molecular weight PET, PEF, PEIT and PEIF polyesters. Compared to commonly used chain extenders, DGO has the advantage that it can be removed from the polymer product. This means that apart from increasing the molecular weight, it has no influence on the polymer's properties. When synthesizing polyesters, applying DGO therefore provides increased molecular weight without the inherent drawbacks of commonly used chain extenders. For PEF, molecular weights were produced close to what is normally obtained after a 24-hour solid-state polymerization. For amorphous polyesters SSP is not even an option, which currently limits the maximum T g at which PEIT can be produced and addition of DGO to this process actually opens up a pathway to industrially produce such materials. As long as the polyester contains sufficient amounts of ethylene glycol monomers, DGO should work. DGO is very reactive, which opens the door to the production of polyesters such as PET and PEF without catalyst or with more benign, but less effective catalysts, even when a large amount of unreactive isosorbide is incorporated. This is of particular relevance due to the health risks associated with antimony, which is the preferred catalyst for PET production. 18 Consequently, the leaching of this element from PET into packaged products and the environment poses a significant concern. 14–17 Moreover, due to the utilization of antimony at very low concentrations, effectively recovering it from waste plastic becomes virtually impossible. 19,54–56 Also the future availability of Sb is an important issue, as reserves are running low. Furthermore DGO presents a workable alternative to the commonly used solid state polymerization, saving on time and energy. The simplicity of this method makes it scalable and production of the booster itself can also be scaled, as DGO can be effectively synthesized by the transesterification of dimethyl oxalate with guaiacol. Unlike the metal catalysts used for polymerization, the catalyst used for the production of DGO can be easily recovered and recycled. 57 Guaiacol is also non-toxic and has the potential to be derived from lignin, an abundant renewable resource. 58 DGO's high reactivity and easy removal from the polymer could also help solve current issues in mechanical recycling of polyesters. This is especially relevant for PET, due to its production scale, where the decrease in molecular weight due to thermal degradation with every recycle and the accompanying loss in quality severely limits closed-loop recycling. The addition of small amounts of DGO to such a melt can readily boost the molecular weight back to sufficient levels, making the recycled PET fit for purpose.",
"introduction": "Introduction With more than 70 million tons per year of polyethylene terephthalate (PET) manufactured worldwide, PET is the most important commercial polyester today. 1–3 PET finds application in numerous products, especially packaging and textile products such as bottles, sheets, carpets and clothes. 4 In Europe, PET accounts for roughly 16% of the plastic consumption. 5 Polyethylene furanoate (PEF) is considered as one of the most promising renewable future polyester and PEF is expected to become commercially available in 2024. 6–8 Compared to PET, PEF shows improved barrier-, thermal- and mechanical performance, allowing for thinner packaging with improved performance. 6 Melt polycondensation is the most common synthetic method for large-scale production of polyesters. 9 It is essentially a reversible esterification reaction with a relatively low equilibrium constant and relies on the removal of the condensation products (typically water) to reach sufficient molecular weight. Material performance of plastics is strongly correlated to the molecular weight of the polymer: insufficient molecular weight could lead to appliance failure. 10 As the molecular weight increases, the viscosity of the melt also increases, making it more difficult to remove the condensation products. Eventually, this becomes a limiting factor for the molecular weight. 11 To further enhance the removal of condensation products, higher temperature, longer reaction time and improved reactor designs are used. 12 However, these harsher conditions also contribute to unwanted side reactions, which change the physical properties of the material and will result in product coloration. Decarboxylation of PEF's monomer furan-2,5-dicarboxylic acid (FDCA) and acetaldehyde formation from the ethylene glycol moiety are examples of undesired side reactions often seen with the production of PEF and PET under harsher conditions. 9,13 Also metal catalysts are added to accelerate the polymerization process. The vast majority of global PET production (>90%) relies on antimony trioxide as the catalyst. However, concerns have been raised about the safety of antimony, given its classification as a heavy metal, particularly when it is utilized in products that come into contact with water, beverages, pharmaceuticals, and food items. 14–20 Additionally, PET fibres (typically containing Sb catalyst) have been reported in a wide range of human foods and beverages, including seafood, drinking water, beer, salt and sugar. 17 For PET, it is difficult to obtain a number average molecular weight ( M n ) higher than 20 kilodalton (kDa), equal to an intrinsic viscosity (IV) of ∼0.60 dL g −1 , without additional processes after the melt polymerization. 21 This molecular weight range is sufficient for applications demanding lower molecular weight, such as textiles. To obtain higher molecular weights necessary for bottle applications (0.73–0.85 dL g −1 ) and industrial yarns (>1.2 dL g −1 ) an additional solid-state polymerization (SSP) is typically required. 22,23 In SSP, polymer pellets are heated above the glass transition temperature ( T g ), but below the melting point ( T m ), while being rotated under a nitrogen flow or reduced pressure. Due to the low mobility of the polymer end groups and condensate in the solid state, it is a time-consuming process and therefore energy consuming and expensive. 24 Typically, for PET SSP takes more than 8 hours. 25 When less reactive diols, like isosorbide, are introduced to produce e.g. polyethylene- co -isosorbide terephthalate (PEIT) and polyethylene- co -isosorbide furanoate (PEIF), it becomes even more challenging to obtain sufficient molecular weight. 26–29 Isosorbide is a commercial biobased building block produced from glucose via sorbitol. 30 Incorporation of isosorbide is desired due to its positive effect on thermomechanical stability and mechanical performance, which enables the production of high-performance biobased materials. DURABIO and ECOZEN are examples of commercially available isosorbide-based engineering polymers and can be found in smart phone screens, car dashboards, sports drink bottles and food containers. 30–32 Isosorbide is less reactive due to its secondary alcohol groups and melt polycondensation becomes considerably more difficult with increasing isosorbide content. Additionally, the crystallinity is lost with isosorbide contents above 15 mol% (relative to total diol), which makes SSP impossible, as amorphous polymer pellets will stick and clump together at temperatures above their T g . 33 These limitations hamper the growth of the isosorbide-based polyesters market. An alternative route for producing high molecular weight polymers is by using a so-called chain extender after melt polycondensation. 22,25,34 Chain extenders are very reactive molecules that function as a coupling agent by reacting with the remaining functional chain ends to increase the molecular weight. When combined with melt polycondensation, only a small amount of chain extender is required, as already a considerable chain length can be obtained by melt polycondensation. In this manner the final stage and most difficult part of the polymerization can be easily overcome by the high reactivity of the chain extender, which leads to considerably less time and less harsh conditions to obtain high molecular weights. Additionally it can make SSP redundant. 22 The combination of these factors likely reduces the process costs significantly. Chain extenders are also attractive for recycling or up-cycling polymers, due to the low cost and easy use. 35,36 Various chain extenders have been used for polyesters, for example: bis-oxazolines, pyromellitic dianhydride, organic phosphites, di-isocyanates, di-epoxides, carbonyl biscaprolactam, diphenyl carbonate, diphenyl esters, bisketenimines, bislactams. 10,22,35,37–44 However, the use of chain extenders also comes with significant drawbacks; often, side reactions occur, such as crosslinking or chain scission, which changes the physical properties of the final polymer. After each chain extension, the end of life and recycling of the polymer becomes more complicated. 36 Furthermore, the chain extender residues that are incorporated into the polymer backbone often influence the properties of the resulting polymer and some chain extenders are considerably toxic, on their own, or as residue in the final polymer, which excludes their use in food-grade applications. 10,45 These drawbacks are probably the reason that chain extenders are rarely used today in standard commercial polyester production. Our previous research on isosorbide based polyesters revealed that diaryl oxalates are very effective in polyesterification reactions due to their high reactivity, caused by the leaving group properties of aryl groups and the fact that the carboxyl groups of oxalate are vicinal. 46 Especially the diguaiacyl ester (DGO) is highly reactive, due to the steric hindrance caused by the methoxy group in the ortho position, impeding the reverse reaction of guaiacol with the polymer chain. Another interesting characteristic of oxalate is its ability to create a six-membered ring when combined with ethylene glycol in the polymer, leading to the formation of ethylene oxalate (EO). 44,47,48 This opens a route to use the high reactivity of oxalate and at the same time remove it from the polymer structure by ring formation, see Fig. 1 . This shows potential for chain extender applications, free from the inherent drawbacks of commonly used chain extenders. Therefore this research is focused on the use of diguaiacyl oxalate as a traceless chain extender for catalyst free polyester. Fig. 1 Schematic illustration of the molecular weight boosting of PET. Starting at the bottom, a low molecular weight polymer of PET is reacted with a small amount of DGO to produce a high molecular weight PET. DGO is a very reactive difunctional molecule, able to link residual alcohol end-groups together by transesterification. Above a certain temperature, the resulting oxalate in the polymer chain readily forms ethylene oxalate, a six membered ring molecule with its ethylene glycol neighbor. All DGO components, guaiacol and EO, are then removed from the reactor at the process temperature and reduced pressure. For a more detailed reaction mechanism of the ethylene oxalate formation, see the ESI overview S1. †"
} | 3,181 |
32500893 | null | s2 | 9,551 | {
"abstract": "Covalent adaptable hydrogels (CAHs) reversibly adapt their structure in response to external stimuli, emerging as a new platform for biological applications. Due to the unique and complex nature of these materials, a characterization technique is needed to measure the rheology of these CAHs in biological processes. μ"
} | 79 |
29159286 | PMC5693563 | pmc | 9,552 | {
"abstract": "Mesostructuring and geometric constriction control the temperature-dependent thermal transport properties in granular matter.",
"introduction": "INTRODUCTION With increasing energy consumption and further miniaturization of electronic devices, a need for new, space-saving, and functional materials to manage heat arises. Recent examples report on the theory and realization of thermal memory ( 1 – 3 ), thermal rectification ( 4 – 6 ), dynamic insulation ( 7 , 8 ), phase change materials ( 9 ), thermal cloaking ( 10 ), and thermal switching materials ( 11 ). The experimental realization of many of these emerging applications is still a great challenge. One major limiting factor is given by the typical power-law temperature dependence of the thermal conductivity of most materials. For crystalline materials, one usually finds a power-law exponent of +3 up to the point where phonon-phonon scattering dominates (~10% of the Debye temperature). Beyond that point, a −1 to −3 exponent is found for increasing temperatures ( 12 ). Amorphous materials merely exhibit a monotonic increase across the entire temperature range, combined with commonly one or two plateau regimes ( 13 ). To pave the way toward advanced heat management devices and thermal logic circuits, tailor-made materials with non–power-law but well-controlled temperature-dependent properties are needed. For example for thermal diodes, nonlinearity is required ( 12 ), whereas abrupt changes with a small input of excess heat are necessary for the gate material of thermal transistors ( 14 ). State-of-the-art materials use a first-order phase transition either in their homogeneous bulk form ( 9 , 14 – 17 ) or in a heterogeneous blend ( 18 – 21 ) to manipulate the temperature-dependent thermal transport. In homogeneous bulk materials, the thermal properties are governed by the material composition, rendering it difficult to target a specific application. Composite materials provide a higher degree of flexibility, owing to the selection of certain material combinations. Quite importantly, the temperature-dependent properties of a material can additionally be strongly influenced by the underlying micro- and nanostructure ( 22 , 23 ). Prime examples are colloidal crystals, which have received much attention, predominantly within the field of photonics ( 24 – 27 ), phononics ( 25 , 28 , 29 ), or as template structures ( 30 – 32 ). Highly defined colloidal superstructures are accessible in a simple and scalable way by established fabrication methods ( 30 – 32 ). Colloidal crystals represent a significantly underexplored field with respect to their thermal transport properties. When going through the second-order phase transition, namely, the glass transition temperature of the constituting polymer, the increase in polymer mobility leads to a loss of the particulate nanostructure. Consequently, the thermal conductivity increases sharply ( 33 ). The versatile structural fabrication can be complemented by specific particle design to add further functionality to the colloidal ensemble. This allows to widely program the thermal transport properties to a specific need. Here, we demonstrate the vast potential of constriction-controlled thermal transport through particulate ensembles. We choose polymer colloidal crystals as a case study to specifically tune temperature-dependent thermal conductivity. We emphasize that this tuning is solely based on geometric constriction. Precisely, thermal conductivity is governed by the thermally induced changes of the nanosized interparticle contact area between adjacent particles in a close-packed colloidal superstructure. Figure 1 outlines the unique possibilities provided by constriction-controlled thermal transport. We demonstrate four key aspects, which are of paramount importance for future heat management devices and become accessible for the first time via our concept: (i) adjustment of the (second-order) phase transition to a desired temperature ( Fig. 1B ); (ii) tuning of the phase transition range ( Fig. 1C ); (iii) introduction of multiple discrete transition steps ( Fig. 1D ); and (iv) controlling of the degree of transition change ( Fig. 1E ). Fig. 1 Key aspects for heat management devices and their realization based on constriction-controlled thermal transport in colloidal assembly structures. ( A ) By exceeding T g , the thermal conductivity irreversibly increases based on the enlargements of contact points during particle sintering. ( B ) The transition temperature can be tailored by assembling the crystal from particles having different T g . ( C ) The random coassembly of equal-sized particles but different T g results in a broad transition. ( D ) Multiple transition steps can be introduced by a discrete layer-by-layer assembly. ( E ) The height of the transition steps is controllable by the thickness of the respective layer. We show how to program the described transition behavior of these assemblies by adjusting the thermal properties of the polymer particles and by selecting a suitable mesoscopic colloidal crystal architecture. Our system is based on copolymer particles consisting of n -butyl methacrylate ( n -BA) and methyl methacrylate (MMA). By adjusting the n -BA content of the particles, it is possible to control the glass transition temperature of the copolymer ( 26 ).\n\nIntroduction of multiple discrete transition steps To program distinct steps into the temperature-dependent thermal conductivity, we fabricated more intricate colloidal superstructures. Therefore, we used filtration, which easily allows fabricating layered colloidal ensembles. Filtration represents a much faster self-assembly method compared to the evaporation-induced self-assembly. However, this comes at the expense of the long-range crystalline order (fig. S5A). Nevertheless, filtration provides direct access to tailor-made colloidal superstructures in a layer-by-layer fashion. Thus, we fabricated multilayered, freestanding colloidal monoliths in which every layer consisted of particles with a predefined T g . We demonstrate the thermal transport properties of three particles of ~420-nm diameter with T g,1 = 61°C, T g,2 = 103°C, and T g,3 = 124°C. This introduces multiple transition steps of the thermal conductivity by a discrete sintering of the individual layers at the respective T g . The schematic structure for these monoliths is illustrated in Fig. 4A . Fig. 4 Introduction of multiple-step transitions. ( A ) Schematic illustration of the structure of a colloidal monolith consisting of one, two, and three particle layers where each layer has a different T g (blue, green, and red). ( B ) Temperature-dependent thermal conductivity of colloidal monoliths consisting of one, two, and three particle layers. On the basis of the discrete layer assembly, multiple step-like increases (dashed red lines) at the specific T g of the copolymer particle are observed. Error bars represent the SD derived from three individual measurements. Thermal diffusivity data can be found in fig. S3 (A and B). Closed symbols represent the heating cycle; open symbols represent the cooling cycle. The temperature-dependent thermal conductivity of colloidal assemblies consisting of one, two, and three particle layers is illustrated in Fig. 4B . In contrast to the randomly mixed binary colloidal crystal discussed above ( Fig. 3 ), the discrete layer assembly evokes distinct steps in the thermal conductivity profile. This is based on the sintering of the homogenous, individual layer at its corresponding T g . The unmolten layers remain in their pristine state. Exceeding the T g of the remaining layers results in a further, multistep increase of the effective thermal conductivity of the entire ensemble. Conceptually, an arbitrary number of distinct steps could be introduced in this fashion to a particulate material. We demonstrate a three-step material by layering three particle types. The respective transition temperatures coincide with the predetermined T g ( Fig. 4B , orange).",
"discussion": "DISCUSSION These four concepts show that colloidal assembly structures can control the temperature-dependent thermal transport properties with an unprecedented degree of flexibility. This capability becomes even more relevant because the fabrication method is scalable and can be flexibly adapted to other materials too. This allows introduction of further functional properties. Furthermore, the constriction-controlled thermal transition represents a purely solid-state transition, with no liquids involved. Although the polymer platform presented here does not allow a reversible adjustment of the thermal properties, we are convinced that this concept can be expanded to other material systems too. These may then provide the required reversibility for future applications. Our findings outline a general approach to specifically tailor the temperature-dependent thermal conductivity of a nanostructured material. We want to stress the high relevance of the interparticle contact points, which is the first ingredient to allow this impressive degree of flexibility. The ability to adjust the onset of the glass transition temperature of the polymer particles by simple chemical synthesis is the second key ingredient. Combining these two parameters in a tailor-made colloidal superstructure allowed us to show four key properties, which will be of relevance for future heat management devices: (i) adjustable onset temperature, (ii) width of transition, (iii) multistep transitions, and (iv) height of transition steps. However, one also has to consider the current shortcoming of this simple material composition, namely, the irreversibility of changes to the interparticle contact area. Nevertheless, we are convinced that this contribution will motivate more research on thermal transport through particulate structures. This may very likely lead to the availability of more functional particle compositions, which may circumvent the irreversibility of the polymer particle sintering. Furthermore, these may allow for the introduction of other external stimuli, such as pH, solvents, light, electric currents, or electric fields to trigger the necessary transition. Considering the inherent and well-known photonic and phononic properties of colloidal crystals and glasses adds even another dimension of functionality, which we did not elaborate on in this contribution. Thus, this concept paves the way toward a genuinely multiphysical and multifunctional heat management material."
} | 2,640 |
29160039 | PMC5743807 | pmc | 9,554 | {
"abstract": "Summary The methylotrophic yeast Pichia pastoris ( Komagataella spp.) is widely used as cell factory for recombinant protein production. In the past recent years, important breakthroughs in the systems‐level quantitative analysis of its physiology have been achieved. This wealth of information has allowed the development of genome‐scale metabolic models, which make new approaches possible for host cell and bioprocess engineering. Nevertheless, the predictive accuracy of the previous consensus model required to be upgraded and validated with new experimental data sets for P. pastoris growing on glycerol or methanol as sole carbon sources, two of the most relevant substrates for this cell factory. In this study, we have characterized P. pastoris growing in chemostat cultures using glycerol or methanol as sole carbon sources over a wide range of growth rates, thereby providing physiological data on the effect of growth rate and culture conditions on biomass macromolecular and elemental composition. In addition, these data sets were used to improve the performance of the P. pastoris consensus genomic‐scale metabolic model iMT 1026. Thereupon, new experimentally determined bounds, including the representation of biomass composition for these growth conditions, have been incorporated. As a result, here, we present version 3 (v3.0) of the consensus P. pastoris genome‐scale metabolic model as an update of the iMT 1026 model. The v3.0 model was validated for growth on glycerol and methanol as sole carbon sources, demonstrating improved prediction capabilities over an extended substrate range including two biotechnologically relevant carbon sources.",
"conclusion": "Conclusions In this study, we analysed the performance of P. pastoris growing in chemostat cultures using glycerol or methanol as single carbon source over a wide range of growth rates. The observed biomass composition changes in terms of protein and RNA content as a function of growth rate further supports the growth rate effect hypothesis on biomass composition; i.e. for both carbon sources, higher content of protein and RNA was observed at higher growth rates. Moreover, biomass composition also showed a strong dependence on carbon source, as protein content in biomass was higher in methanol‐grown cells. In addition, the carbon source has a significant impact on lipid and amino acid profiles. Overall, the information gathered on biomass composition at different growth rates and carbon sources allowed to calculate average biomass compositions for glycerol‐ and methanol‐grown biomass. This allowed us to extend the iMT1026 model with new biomass equations for growth on glycerol or methanol as sole carbon sources. Energetic maintenance requirements were estimated for the first time in P. pastoris in both carbon sources. Furthermore, the model was validated for the range of growth rates tested, and it accurately described the experimental physiological data. Minor discrepancies between experimental data and simulations were found for glycerol at lower growth rates, where a nonlinear behaviour of growth parameters has been reported due to a metabolic shift on metabolism that enables P. pastoris to reduce its maintenance energy requirements. Such discrepancies can be easily taken into account by decreasing the value of maintenance energy requirements included in the model. Experimental data derived from chemostat cultivations provide information for calculating carbon source‐specific energetic parameters. These values allow for significantly improving the precision of estimated macroscopic behaviour. Therefore, the recalibration of energetic parameters, both NGAME and GAME, may be used for extending the model to alternative carbon sources. Furthermore, the characterization of biomass and definition of condition‐specific biomass equations enhance model performance and accuracy and allow for a more precise and realistic calculation of metabolic flux distribution. In summary, the third version of iMT1026, v3.0, consensus model for P. pastoris, provides to the scientific community an improved metabolic engineering and analysis tool with expanded capabilities for predicting the metabolic phenotype in a broader range of conditions as well as an improved tool for future design of model‐based metabolic engineering of the P. pastoris cell factory.",
"introduction": "Introduction \n Pichia pastoris ( Komagataella spp.) has become one of the most commonly used hosts for recombinant protein production (Corchero et al ., 2013 ; Gasser et al ., 2013 ) including biopharmaceuticals (Martínez et al ., 2012 ; Walsh, 2014 ). Since 1995, the number of genes heterologously expressed in this yeast has steadily increased (Bill, 2014 ). The establishment of P. pastoris as widely used cell factory has been supported by the development of improved high cell density operational strategies (Cos et al ., 2006 ), synthetic biology tools, such as the availability of novel constitutive and inducible promoters (Prielhofer et al ., 2013 ; Weinhandl et al ., 2014 ), the application of novel genetic engineering techniques for its manipulation (Vogl et al ., 2013 ; Weninger et al ., 2015 ), as well as increased body of knowledge of P. pastoris at the genetic and physiological levels. Moreover, progress in synthetic biology of this yeast has also opened the door towards utilizing this yeast as whole‐cell biocatalyst for non‐native value‐added metabolite production (Pscheidt and Glieder, 2008 ; Heyland et al ., 2010 ; Cheng et al ., 2014 ; Geier et al ., 2015 ). At an industrial scale, reduced cost of raw materials is as important as high production yields for cost‐effective processes (Kroll et al ., 2010 ; Gustavsson and Lee, 2016 ). In addition, in order to optimize the metabolite biosynthesis process to obtain high yields, it is also important to select the most appropriate substrate (Goldman, 2010 ). In this context, glycerol is a by‐product in the conventional biodiesel production process and therefore represents an attractive opportunity for revalorization of an industrial waste stream, that is, for the development of a glycerol‐based integrated biorefinery concept (Kiss et al ., 2015 ). Indeed, glycerol is becoming an attractive carbon source in fermentation processes to produce high added value compounds (Johnson and Taconi, 2007 ; Yang et al ., 2012 ; Valerio et al ., 2015 ). Furthermore, the reduction degree of glycerol (4.67) is different from that of glucose (4.0), and therefore, higher yields of certain secondary metabolites can be obtained from this compound (da Silva et al ., 2009 ). Nonetheless, crude glycerol is far from being pure and contents several other compounds such as methanol (Posada et al ., 2012 ). Methanol is usually toxic for microbes with the exception of methylotrophic microorganisms. Thus, subsequent purification and refinement steps should be applied to the raw glycerol if it has to be used by non‐methylotrophic organisms. On the other hand, methanol is also an increasingly interesting C1 compound as building block for value‐added compound biosynthesis (Schrader et al ., 2009 ; Khosravi‐Darani et al ., 2013 ; Nguyen et al ., 2016 ). In this context, P. pastoris is able to efficiently use glycerol and/or methanol as energy and carbon sources (Solà et al ., 2007 ; Çelik et al ., 2008 ; Jordà et al ., 2014 ). In addition, the most extensively used promoters for heterologous gene expression in P. pastoris (namely, P GAP , constitutive and P AOX , inducible) are directly associated with glycerol and methanol metabolism (Cos et al ., 2006 ; Gasser et al ., 2013 ). Therefore, P. pastoris appears as an organism of high potential for the development of the glycerol biorefinery concept. Genome‐scale metabolic models (GSMM) allow to predict the phenotype of a microorganism in a range of conditions, including those derived from genetic modification (Oberhardt et al ., 2009 ; Kim et al ., 2012 ). This capability makes GSMM a powerful tool for the design of metabolic engineering strategies to enhance productivities or implementing new pathways (Cvijovic et al ., 2011 ; Gustavsson and Lee, 2016 ). Nevertheless, validation of GSMM for different conditions requires the availability of extensive cultivation data information describing its physiology. In addition, a wide range of information on biomass composition enables building specific biomass equations to accurately describe cell growth in each case (Dikicioglu et al ., 2015 ). Three independent GSMM for P. pastoris were initially published, namely iPP 668 (Chung et al ., 2010 ), PpaMBEL1254 (Sohn et al ., 2010 ) and iLC915 (Caspeta et al ., 2012 ). More recently, the consensus model iMT1026 has been published (Tomàs‐Gamisans et al ., 2016 ), integrating and upgrading the previous models. The consensus iMT1026 model showed a significant improvement in prediction accuracy and was validated for two sets of conditions: growth on glucose as a sole carbon source under different oxygen availability conditions and growth on different glycerol and methanol mixtures as carbon sources at different growth rates. However, given the impact of biomass composition on the model predictive accuracy in a context‐dependent manner (Dikicioglu et al ., 2015 ), this model was still not suitable for describing growth on glycerol or methanol as single carbon sources. This is because biomass composition equations take a major role on prediction reliability, and small changes in that composition, or using an inadequate biomass equation, may rend model calculations inaccurate (Dikicioglu et al ., 2015 ). Hence, GSMMs are in continuous evolution (e.g. for Saccharomyces cerevisiae (Aung et al ., 2013 )) usually involving error‐fixing steps related to poor or wrong gene annotation (Dikicioglu et al ., 2014 ), or extending GSMM capabilities for a broader range of cultivation conditions. In this work, we expand the iMT1026 model capabilities by implementing the capacity of accurately describing P. pastoris growth phenotype when using glycerol or methanol as sole carbon sources. A series of chemostat cultures were performed at a wide range of growth rates using glycerol or methanol as sole carbon sources in order to provide comprehensive physiological data sets needed to upgrade the model. This included quantitative analyses of the elemental and macromolecular biomass composition for each tested growth condition. This allowed to introduce new biomass reaction equations to the metabolic model specific for growth on glycerol or methanol. Furthermore, the new version of the model (v3.0) was validated for growth on these two substrates within the tested growth rate range.",
"discussion": "Results and discussion Physiological macroscopic parameters \n Pichia pastoris X‐33 strain was cultivated in carbon‐limited chemostat cultures at different dilution rates to characterize its physiology using different carbon sources. This information was used to estimate the energetic parameters and to calibrate the model for such carbon sources. Glycerol cultivations were carried out at different dilution rates (D): 0.035, 0.050, 0.065, 0.100, 0.130 and 0.160 h −1 . At 0.160 h −1 , the inflowing gas was supplied with an air:O 2 mixture (92.5:7.5) due to the higher O 2 demand and cell concentration. Due to this operational limitation, no higher dilution rates were tested, despite P. pastoris has been reported to grow at higher growth rates (Cos et al ., 2006 ). Methanol limiting chemostats were run at 0.035, 0.050, 0.065, 0.080, 0.100 and 0.130 h −1 . At 0.130 h −1 , bioreactor washed out. Biomass concentration, CO 2 production and O 2 consumption continuously decreased, and methanol accumulated. According to a chemostat washout kinetics (Doran, 1995 ), maximum growth rate on methanol was estimated to be between 0.11 and 0.12 h −1 , which is in agreement with previously reported values (Barrigon et al ., 2015 ). Based on the chemostat data, specific productivities and yields were calculated for each condition (Table 1 ). In both glycerol and methanol cultivation series, main growth parameters show a linear correlation with growth rate (μ). Table 1 Macroscopic growth parameters after the reconciliation procedure for glycerol and methanol cultivations at different dilution rates Carbon source μ SP (h −1 ) μ exp (h −1 ) q S (mmol · gDCW −1 · h −1 ) q O2 (mmol · gDCW −1 · h −1 ) q CO2 (mmol · gDCW −1 · h −1 ) q X (Cmmol · gDCW −1 · h −1 ) Y XS (g X · g S \n −1 ) RQ Glycerol 0.035 0.035 ± 0.001 −0.58 ± 0.05 −0.82 ± 0.13 0.53 ± 0.11 1.22 ± 0.04 0.65 ± 0.03 0.64 ± 0.03 0.050 0.049 ± 0.002 −0.85 ± 0.06 −1.26 ± 0.10 0.84 ± 0.08 1.70 ± 0.14 0.62 ± 0.05 0.67 ± 0.07 0.065 0.064 ± 5e‐4 −1.07 ± 0.01 −1.52 ± 0.02 1.00 ± 0.02 2.22 ± 0.02 0.64 ± 5e‐4 0.65 ± 5e‐4 0.100 0.094 ± 0.004 −1.52 ± 0.08 −2.04 ± 0.11 1.28 ± 0.08 3.27 ± 0.15 0.71 ± 0.04 0.63 ± 0.01 0.130 0.124 ± 0.001 −1.92 ± 0.08 −2.36 ± 0.15 1.41 ± 0.13 4.36 ± 0.24 0.71 ± 0.04 0.60 ± 0.07 0.160 0.154 ± 0.002 −2.41 ± 0.03 −2.92 ± 0.01 1.74 ± 1e‐3 5.47 ± 0.08 0.70 ± 2e‐3 0.60 ± 3e‐3 Average 0.67 ± 0.04 0.63 ± 0.03 Methanol 0.035 0.035 ± 0.001 −2.81 ± 0.16 −2.98 ± 0.22 1.59 ± 0.14 1.22 ± 0.02 0.38 ± 0.01 0.53 ± 0.01 0.050 0.049 ± 2e‐4 −3.88 ± 0.10 −4.07 ± 0.15 2.15 ± 0.10 1.73 ± 0.01 0.39 ± 0.01 0.53 ± 0.01 0.065 0.065 ± 0.001 −4.87 ± 0.22 −4.97 ± 0.29 2.55 ± 0.18 2.33 ± 0.04 0.41 ± 0.01 0.51 ± 0.01 0.080 0.084 ± 0.001 −6.23 ± 0.16 −6.36 ± 0.18 3.27 ± 0.12 2.96 ± 0.13 0.42 ± 0.02 0.51 ± 0.02 0.100 0.099 ± 0.001 −7.82 ± 0.28 −8.22 ± 0.37 4.34 ± 0.24 3.47 ± 0.04 0.40 ± 0.01 0.53 ± 0.01 Average 0.40 ± 0.01 0.52 ± 0.01 μ SP corresponds to the set point growth rate and μ exp , the measured experimental μ. John Wiley & Sons, Ltd Regarding biomass yields (Y XS ), there is a slight decrease at lower growth rates on both carbon sources, similarly as reported by Van Dijken et al . ( 1976 ) and Rebnegger et al . ( 2014 ). Despite this apparent correlation, there are no statistically significant differences within the tested range, and average Y XS and RQ can be calculated for the abovementioned range of growth rates. Average Y XS in methanol is 0.40 g X · g S \n −1 and is in accordance with yields previously reported for P. pastoris and other yeast (Hazeu and Donker, 1983 ). This value is considerably lower than 0.67 g X · g S \n −1 , the average Y xs for glycerol. The Y XS for glycerol ranged between 0.62 and 0.71 g X · g S \n −1 , similar to yields on this substrate reported for different Pichia species and other yeasts (Taccari et al ., 2012 ). Macromolecular and elemental biomass composition Growth rate‐dependent stoichiometry To investigate the potential impact of growth rate on biomass composition, samples of the cultures were taken for analysis of the biomass elemental and macromolecular composition at different dilution rates. In particular, we analysed the biomass composition at four different growth rates for glycerol (μ = 0.035, 0.065, 0.100 and 0.160 h −1 ) and three for methanol (μ = 0.035, 0.065 and 0.100 h −1 ). The experimental data sets and the calculated (reconciled) biomass composition are summarized in Fig. 1 and Table 2 . Figure 1 Comparison of the reconciled macromolecular composition of glycerol and methanol cultures at different growth rates. Table 2 Detailed reconciled elemental and macromolecular composition of cells grown on glycerol and methanol at different growth rates, and the averaged biomass composition used for defining the stoichiometric coefficients in iMT1026 v3.0. Values represent weight/weight % ± SD Glycerol Methanol Average glycerol a \n Average methanol a \n Glucose b \n D (h −1 ) 0.035 0.065 0.100 0.160 0.035 0.065 0.100 Protein 36.0 ± 6.7 41.2 ± 0.4 41.1 ± 2.9 43.3 ± 2.8 48.6 ± 1.0 50.7 ± 0.5 51.5 ± 2.1 41.0 ± 1.5 50.1 ± 0.8 37.0 ± 2.4 Carbohydrate 45.6 ± 7.5 37.0 ± 1.1 35.9 ± 3.1 33.9 ± 1.3 29.0 ± 2.8 28.9 ± 0.8 23.5 ± 1.3 35.9 ± 2.0 27.3 ± 1.5 36.9 ± 3.5 Lipid 1.3 ± 0.3 1.8 ± 0.4 2.2 ± 0.5 2.8 ± 3e‐2 2.2 ± 0.2 1.9 ± 0.1 2.4 ± 0.6 2.5 ± 0.4 2.0 ± 0.2 6.2 ± 3.3 RNA 6.0 ± 0.1 6.7 ± 1.1 7.6 ± 1.2 9.1 ± 2.6 5.7 ± 0.6 6.3 ± 1.3 7.1 ± 0.4 7.8 ± 0.6 7.0 ± 0.6 6.6 ± 0.7 DNA 0.19 ± 1e‐3 0.19 ± 0.01 0.18 ± 0.01 0.18 ± 0.01 0.19 ± 4e‐3 0.18 ± 4e‐3 0.18 ± 0.02 0.19 ± 0.01 0.18 ± 0.01 0.13 ± 0.01 SO 4 \n 0.28 ± 0.06 0.40 ± 1e‐3 0.45 ± 0.10 0.46 ± 0.11 0.63 ± 4e‐3 0.69 ± 0.05 0.66 ± 0.06 0.46 ± 0.08 0.63 ± 0.04 0.3 ± 0.3 H 2 O 5.8 ± 0.6 5.7 ± 0.2 6.6 ± 2.5 6.6 ± 0.2 8.4 ± 0.1 6.2 ± 1.8 8.1 ± 0.7 5.6 ± 0.7 7.2 ± 0.9 6.3 ± 2.4 Metals 5.3 ± 1.0 7.0 ± 0.1 6.0 ± 0.9 5.8 ± 1.1 5.3 ± 2.1 5.2 ± 2.0 6.6 ± 0.1 7.0 ± 0.6 6.3 ± 0.6 6.4 ± 0.4 C 42.3 ± 0.1 41.9 ± 0.1 41.9 ± 0.9 42.44 ± 1.00 42.2 ± 0.8 43.2 ± 0.1 41.8 ± 0.2 41.98 ± 0.27 42.28 ± 0.53 43.0 ± 1.4 H 6.3 ± 3e‐2 6.24 ± 2e‐3 6.3 ± 0.2 6.44 ± 0.08 6.6 ± 0.1 6.5 ± 0.2 6.5 ± 0.1 6.24 ± 0.06 6.43 ± 0.05 6.3 ± 0.2 N 7.4 ± 1.2 8.4 ± 0.1 8.7 ± 0.7 9.22 ± 0.24 9.7 ± 0.1 10.2 ± 0.1 10.4 ± 0.4 8.58 ± 0.31 10.06 ± 0.16 6.9 ± 0.4 O 37.8 ± 2.2 35.4 ± 4e‐2 36.0 ± 2.6 34.98 ± 0.40 35.2 ± 1.3 33.9 ± 1.6 33.4 ± 0.7 35.11 ± 0.58 33.83 ± 0.46 36.4 ± 1.4 S 0.25 ± 0.06 0.30 ± 1e‐3 0.25 ± 0.04 0.30 ± 0.04 0.41 ± 0.01 0.4 ± 0.02 0.43 ± 0.03 0.30 ± 0.03 0.41 ± 0.01 0.2 ± 0.1 Ashes 5.9 ± 1.0 7.7 ± 4e‐3 6.8 ± 1.1 6.6 ± 1.3 5.9 ± 2.0 5.9 ± 1.8 7.4 ± 0.1 7.8 ± 0.6 7.0 ± 0.6 7.1 ± 0.4 a Average compositions are weighted averages using 1/SD. b Corresponding to P. pastoris growing at D = 0.1 h −1 . Data taken from Carnicer et al . ( 2009 ) . \n John Wiley & Sons, Ltd Notably, the protein and RNA fractions positively correlate with growth rate in both glycerol‐ and methanol‐fed cultivation series. This increment on protein and RNA with increasing growth rates is at expenses of carbohydrate content. This trade‐off between RNA–protein and carbohydrate content has been widely reported in yeast species (Verduyn et al ., 1990 ; Verduyn, 1991 ), including in P. pastoris (Jordà et al ., 2014 ). The increase in protein fraction is consistent with the measured changes in the elemental composition: the nitrogen content is also higher at high growth rates (Table 2 ). Nonetheless, only the correlation of RNA and growth rate is statistically significant. This stoichiometric dependence of biomass components on growth rate can be described on the basis of the growth rate hypothesis (GRH). Essentially, GRH attributes this shift to the tight control of the expensive protein synthesis system (Henriksen et al ., 1996 ). At higher growth rates, cells need a higher ribosomal content to maintain the enzymatic machinery. The ribosomes are reported to consist of 53% RNA and 47% protein in Aspergillus niger (Hangeraaf and Muller, 2001 ). Thus, the increase in ribosome levels has a deep impact in overall cell protein and RNA content. As mentioned above, the increase in protein and RNA would be at expenses of the carbohydrate content. Biomass characterization in S. cerevisiae showed similar results, with a decrease in carbohydrate content at higher growth rates (Küenzi and Fiechter, 1972 ; Lange and Heijnen, 2001 ). At low growth rates, there is a larger fraction of carbon source not used for energy or cell machinery (protein/RNA) generation which is stored in the form of carbohydrates. As growth rate increases more and more, carbon source is derived towards energy and biosynthetic machinery generation at the expense of stored carbohydrates (Pejin and Razmovski, 1993 ). Regarding the lipid fraction, no statistically significant differences were found across the series of methanol biomass samples collected at different growth rates. Conversely, the cell lipid fraction shows a positive correlation with the growth rate in glycerol‐grown cells. Nevertheless, a negative correlation of lipid content with growth rates has been commonly reported (Meeuwse et al ., 2011 ; Rakicka et al ., 2015 ). Therefore, the positive correlation observed in our case may be attributable to the strong reduction in relative carbohydrate content, which seems to be not entirely compensated by the increase in RNA and protein content. Carbon source effects on biomass composition Besides the impact of the specific growth rate on biomass composition described above, other factors such as the carbon source are also known to have a significant influence (Jordà et al ., 2014 ). In our case, the effect of the carbon source can be appreciated in Fig. 1 : cells grown on methanol show a significantly higher protein fraction than those grown on glycerol. This protein fraction is also higher than the one described for glucose‐grown cells (Table 2 ). However, similar profiles are observed when comparing glycerol‐specific biomass composition to the original biomass composition for glucose‐grown cells (Table 2 ) previously reported by Carnicer et al . ( 2009 ). Indeed, none of the macromolecular components of the glucose‐grown biomass showed any significant difference with the glycerol‐grown biomass in terms of relative abundances. In contrast, growth on methanol has a higher impact on the relative abundance of macromolecules, mainly increasing the protein fraction. This effect was also observed by Jordà et al . ( 2014 ) in a study where P. pastoris was grown in chemostats using different glycerol:methanol mixtures as carbon source. The corresponding biomass composition analyses showed that protein content increased when the methanol/glycerol ratio was higher. Similarly, P. pastoris cells growing on a glucose:methanol mix in chemostat cultivations showed higher protein content than when growing on glucose as a sole carbon source under analogous conditions (Jordà et al ., 2012 ). Consequently, the increase in cell protein content seems to be directly related to methanol utilization and, more specifically, to the amount of enzymes needed for methanol assimilation (Rußmayer et al ., 2015 ). In fact, it is known that genes encoding for the methanol utilization pathway such as the alcohol oxidase ( AOX ) and dihydroxyacetone synthase ( DAS ), two major enzymes involved in the initial steps of methanol metabolism, are highly induced in the presence of methanol (Rußmayer et al ., 2015 ). They are reported to account for up to 10–20% of total protein in methylotrophic yeasts (Van Dijken et al ., 1976 ; Stewart et al ., 2001 ). This fact, together with the significant increase in the cell volume occupied by peroxisomes in methanol‐grown cells, may be a plausible explanation of the increase in cell protein content in these conditions (van der Klei et al ., 2006 ; Veenhuis and van der Klei, 2014 ). On the other hand, amino acid composition analysis of the cell proteome showed no significant differences when comparing cells grown at different growth rates for each substrate Table S1 (Appendix S1 ). However, the amino acid composition of biomass differed significantly for some amino acids when comparing glycerol‐ versus methanol‐grown cells (Table S1 ). In addition, the subset of amino acids showing significant differences of relative abundances in methanol‐grown cells (compared to the glycerol condition) was compared with the amino acid composition of enzymes related to the methanol metabolization pathway (Fig. 2 ). This analysis clearly reveals how the amino acid composition of the methanol metabolization enzymes affects the overall cell amino acid composition with respect to glycerol. Therefore, the higher protein fraction in biomass composition in methanol appears to be related to the increased content of methanol‐assimilating pathway enzymes. Figure 2 Comparison of average amino acid profiles from glycerol and methanol cultures in relation to amino acid abundance in the most abundant proteins in methanol metabolization. Amino acid abundance is presented as mol/mol %. Glycerol and methanol represent the average amino acid composition of glycerol and methanol cultivations respectively. Other variables correspond to the most abundant proteins in the presence of methanol: alcohol oxidase ( AOX 1), dihydroxyacetone synthase ( DAS 1), formate dehydrogenase ( FDH 1), formaldehyde dehydrogenase ( FLD ), catalase ( CTA 1). Glx and Asx represent the pair of Asp/Asn and Glu/Gln respectively. In terms of cell total lipid content, no statistically significant differences were found when comparing the average carbon source‐specific biomass compositions. In addition, there are neither differences with previously described lipid fractions for cells grown on glucose nor with those grown in glucose–methanol mixtures (Carnicer et al ., 2009 ; Jordà et al ., 2014 ). Nevertheless, there are significant differences in the lipid composition profile of cells depending on the carbon source (Fig. 3 , Table S2 in Appendix S1 ). Specifically, these differences are found in triacylglycerols (TAG), free fatty acids (FFA) and phosphatidic acid (PA). There is a higher content of TAG and PA at expenses of FFA in glycerol‐grown cells, whereas in methanol‐grown cells, FFA is the major lipid fraction, and TAG and PA are present only in trace amounts. Glycerol is a direct precursor for many lipids. In addition, the relative content of both TAG and PA, which are lipid molecules with a glycerol backbone, is increased in glycerol‐grown cells. Therefore, these differences seem to reflect the lower synthesis cost of TAG and PA from its direct precursor glycerol. Figure 3 Average lipid profile for biomass grown on glycerol (black) and methanol (grey). Triacylglycerols ( TAG ), free fatty acids ( FFA ), sterols ( STE ), cardiolipin ( CAR ), phosphatidic acid ( PA ), phosphatidylcholine ( PC ) and phosphatidylinositol/phosphatidylserine ( PI / PS ). When formulating a biomass equation for glycerol and methanol growth conditions, despite that certain biomass components appear to be correlated with biomass‐specific growth rate, statistical analyses do not show significant differences associated with growth rate. In contrast, statistically significant differences are found when comparing average glycerol and methanol biomass compositions. Consequently, new biomass equations have been formulated for growth on glycerol and methanol incorporating specific equations for each relevant macromolecule (proteins, lipids) as well as for the fractional contribution of each macromolecule to biomass. The coefficients for the biomass equations were directly extracted from the average carbon source‐specific compositions reported in Table 2 . Energetic parameters estimation Prior to model validation, energetic parameters have to be estimated in order to assure accurate predictions of cell performance. These parameters are the growth associated and the non‐growth associated maintenance energy (GAME and NGAME respectively). NGAME values differed significantly for glycerol and methanol growth conditions. On the one hand, growth on glycerol showed a NGAME of 2.51 mmol ATP · g DCW \n −1 · h −1 , which is similar to the corresponding value previously calculated for glucose growth conditions, 2.81 mmol ATP · g DCW \n −1 · h −1 (Rebnegger et al ., 2016 ; Tomàs‐Gamisans et al ., 2016 ). In contrast, the NGAME calculated for methanol growth is 0.44 mmol ATP · g DCW \n −1 · h −1 , i.e. much lower compared with the corresponding values calculated for the other carbon sources. For GAME estimation for growth on glycerol, physiological parameters corresponding to the μ = 0.035 h −1 condition were not considered, as a metabolic shift seems to change the phenotypic profile at this (and lower) growth rates (Rebnegger et al ., 2014 ). This can be directly inferred from the specific CO 2 production rate (q CO2 ) and specific O 2 consumption rate (q O2 ) observed at this growth rate, which do not follow the same linear trend as in the rest of measured range (Fig. 4 ). Hence, taking into account this consideration, GAME for glycerol was estimated to be 70.66 mmol ATP · g DCW \n −1 , that is, 2.4‐fold lower than for methanol (166.77 mmol ATP · g DCW \n −1 ). As mentioned above, there is an important change in protein composition in methanol‐grown cells compared to glycerol growth due to the high levels of enzymes associated with methanol metabolization. The metabolic overload resulting from the maintenance of this cell machinery could be one of the reasons for the higher GAME besides the fact that growth on highly reduced substrates such as methanol (reduction degree (RD) of 6) is known to be usually less efficient (higher energy dissipation and lower biomass yields) compared to glycerol (RD 4.67) or glucose (RD 4; Heijnen and Van Dijken, 1992 ). When compared to glucose culture conditions, GAME for glycerol growth is very similar to the 72 mmol ATP · g DCW \n −1 calculated for glucose growth in our previous study (Tomàs‐Gamisans et al . ( 2016 )). Figure 4 Evaluation of simulated and experimental macroscopic variables for the growth in glycerol and methanol. For each carbon source, the growth rate was constrained, and the absolute value of substrate uptake rate was minimized. A. Chemostats on glycerol. B. Chemostats on methanol. \n q S \n , \n glycerol/methanol (●), q CO \n 2 (■), q O \n 2 (▼); predicted q S \n , \n glycerol/methanol (dashed line), predicted q CO \n 2 (continuous line), predicted q O \n 2 (dotted line). Model validation The updated model, iMT1026 v3.0 (Appendix S3 and available at BioModels Database with model ID MODEL1612130000), integrating the new specific biomass equations for growth on glycerol and methanol as sole carbon sources, was used to estimate the main macroscopic growth parameters as described in \n Experimental procedures \n section. The version 3.0 of iMT1026 accurately predicts macroscopic growth parameters within the range of tested growth rates for both carbon sources (Fig. 4 ). Despite the great overall performance, model deviates from the experimental data by overestimating q O2 and q CO2 in the case of glycerol growth at 0.035 h −1 (Fig. 4 ). P. pastoris has been reported to reduce maintenance energy requirements at very low growth rates associated with metabolic adaptations and changes in gene expression (Rebnegger et al ., 2014 , 2016 ). To take into account this lower maintenance energy requirement, a series of additional simulations were carried out by constraining the NGAME at values lower than 2.51 mmol ATP g DCW \n −1 · h −1 (i.e. the default value set for glycerol‐grown cells) and maximizing growth at a given substrate uptake rate. In this way, iMT1026 v3.0 can be used to accurately predict the main macroscopic growth parameters for glycerol growth at 0.035 h −1 when NGAME is lowered (Fig. S1 in Appendix S1 ). In particular, values between 1 and 1.5 mmol ATP g DCW \n −1 · h −1 allow the best accuracy in predicting the experimental data at 0.035 h −1 , as shown in Fig. S1 According to these calculations, there is between a twofold and threefold reduction of NGAME at the lower growth rate range. Notably, these results are in agreement with Rebnegger et al . ( 2016 ), who reported threefold reduction in the maintenance requirements at low growth rates. Compared to iMT1026 v2.0, this new version improves the accuracy in the prediction of the main macroscopic variables for glycerol‐ or methanol‐grown biomass (Fig. 5 ) In addition, To demonstrate the importance of using accurate NGAME and GAME as well as precise condition‐specific biomass composition equations, a series of simulations were performed by changing each one of NGAME, GAME and biomass equations, and its overall accuracy was compared (Table S3 in Appendix S1 ). Results showed the best accuracy when all the parameters were adjusted to each specific carbon source. Thus, simulations using glycerol‐ or methanol‐specific biomass equations and estimated NGAME and GAME showed an overall deviation around 2%, while those simulations of glycerol and methanol cultivation data using the glucose‐specific biomass equation and energetic parameters resulted in average deviations of 7–10%. The isolated adjustment of one of the two energetic parameters to those calculated in iMT1026 v3.0 does not result in a significant increase in overall accuracy in all the cases. Despite improvements can be observed when predicting glycerol cultivation data with adjusted both GAME and NGAME parameters when methanol cultivation data are simulated, no such improvement is observed by changing only one of the two parameters. However, the setting of both GAME and NGAME values to those calculated specifically for glycerol and methanol has a positive effect on model prediction accuracy for both carbon sources, and model prediction deviations are reduced to 3–5%. Finally, an alternative approach was tested using the glucose‐specific biomass equation and recalibrating the GAME values, as described in ‘Energetic parameters calculation’ section in Experimental Procedures and according to the experimental values for glycerol and methanol cultivations. As reported in Table S3 , simulations performed with these recalibrated GAME values are able to reduce the deviation of estimated values from experimental data (2–4%), but discrepancy still remained over the deviation of iMT1026 v3.0 that uses specific energetic parameters and macromolecular biomass equations. Hence, the calculation of new non‐growth associated maintenance energy coefficient according to Pirt's equation (Pirt, 1982 ) and the subsequent recalibration of growth associated maintenance energy might be an alternative for adapting genome‐scale metabolic models to expand the model to other carbon sources. Thus even without having new carbon source‐specific biomass macromolecular compositions, a GSMMs could be used for simulations with alternative carbon sources. However, such approach implies a penalty in overall prediction accuracy; thus, it could be used assuming higher deviations (more than twofold higher in the glycerol example, Table S3 ). Moreover, despite achieving acceptable macroscopic parameter estimations (2–5%) without adapting the biomass composition, an inaccurate description of biomass composition may result in false predictions of gene or enzyme essentiality (Duarte et al ., 2004 ). In addition, metabolic flux distribution is sensitive to biomass composition (Dikicioglu et al ., 2015 ); therefore, a wrong or inaccurate biomass equation may result in the estimation of erroneous flux distributions. Therefore, adapting the biomass equations to the condition‐specific composition would be the more accurate approach that would reflect the in vivo flux distribution with greater accuracy. Figure 5 Performance of iMT 1026 v3.0 and iMT 1026 v2.0 models compared to experimental data for the glycerol and methanol cultivations at different growth rates. For the simulations, the specific substrate uptake rate was set as constraint, and biomass was maximized. In iMT1026 v3.0, the specific biomass equations, as well as new non‐growth associated maintenance energy values for glycerol and methanol, were enabled accordingly to the corresponding carbon source‐specific simulation."
} | 8,806 |
32547532 | PMC7270577 | pmc | 9,556 | {
"abstract": "Different soybean genotypes can differ in their tolerance toward aluminum stress depending on their rhizosphere-inhabiting microorganisms. However, there is limited understanding of the response of fungal communities to different aluminum concentrations across different genotypes. Here, we used metabarcoding of fungal ribosomal markers to assess the effects of aluminum stress on the rhizosphere fungal community of aluminum-tolerant and aluminum-sensitive soybean genotypes. Shifts in fungal community structure were related to changes in plant biomass, fungal abundance and soil chemical properties. Aluminum stress increased the difference in fungal community structure between tolerant and sensitive genotypes. Penicillium , Cladosporium and Talaromyces increased with increasing aluminum concentration. These taxa associated with the aluminum-tolerant genotypes were enriched at the highest aluminum concentration. Moreover, complexity of the co-occurrence network associated with the tolerant genotypes increased at the highest aluminum concentration. Collectively, increasing aluminum concentrations magnified the differences in fungal community structure between the two studied tolerant and sensitive soybean genotypes. This study highlights the possibility to focus on rhizosphere fungal communities as potential breeding target to produce crops that are more tolerant toward heavy metal stress or toxicity in general.",
"introduction": "Introduction Aluminum (Al) toxicity is one of the most widespread problem in acidic soils, affecting approximately 40% of the arable land worldwide ( Ma et al., 2001 ; Pierluigi et al., 2008 ). In acidic soils with pH values below five, insoluble forms of Al are turned into soluble Al 3+ ions ( Kinraide, 1991 ; Da Mota et al., 2008 ). Many studies have reported that Al 3+ with high phytotoxicity causes inhibition of nitrate reductase activity and disruption of nitrogen reduction and assimilation ( Zhao and Shen, 2018 ). Moreover, the absorption and utilization of other soil elements such as phosphorus, potassium and iron by plant roots are also affected by Al stress ( Pfeffer et al., 1986 ; Delhaize and Ryan, 1995 ; Pineros and Kochian, 2001 ). Therefore, increased concentration of soluble Al can lead to inhibition of plant root growth and, thus, reduction in crop yield ( Kochian, 1995 ; Kong et al., 1997 ; Exley, 2012 ; Riaz et al., 2018 ). There have been several reports on possible mechanisms of plants to increase tolerance toward high Al 3+ concentrations, among which the chelation of Al 3+ through organic acids such as malic acid, oxalic acid, or citric acid excreted by plant roots is considered to be one of the vital mechanisms ( Ma et al., 2001 ). The different levels of Al tolerance vary significantly among genotypes, largely because of different types and quantities of secreted organic acids ( Miyasaka et al., 1991 ; Wu et al., 2018 ). For instance, more organic acids are excreted by Al-tolerant (Al-T) soybean genotypes when compared with Al-sensitive (Al-S) genotypes, leading to chelation of more Al 3+ ( Yang et al., 2010 ). However, the increased amounts of organic acids not only regulate Al 3+ concentrations in soil ( Silva et al., 2004 ), but also shapes the microbial community composition at the root-soil interface through providing nutrients ( Jones et al., 1996 ; Bürgmann et al., 2005 ). Microorganisms greatly contribute to plant health and productivity ( Mendes et al., 2013 ; Li et al., 2014a , b ). When subjected to environmental stress, plants have the potential to recruit specific microbes to the root system to alleviate the stress ( Rodriguez et al., 2019 ). For example, some plant-growth-promoting bacteria (PGPB) in the soil such as Klebsiella , Serratia , and Enterobacter have the capacity to form Al 3+ -siderophore complexes and improve P-uptake efficiency to cope with Al stress ( Mora et al., 2017 ). Previous studies have investigated the structure of rhizosphere bacterial communities in Al-T and Al-S plants and suggested that Al-T genotypes recruit certain bacterial species that help mitigating Al toxicity ( Yang et al., 2012a ; Wang et al., 2013 ; Lian et al., 2019 ). However, these studies have focused on rhizosphere bacteria as key players, ignoring that fungal species play important roles in nutrient cycling and stress tolerance ( Kawai et al., 2000 ; Peltoniemi et al., 2012 ). For instance, some fungi, such as Penicillium and Aspergillus have been shown to improve Al-tolerance by producing organic acids and at the same time provisioning plants root with nitrogen and phosphorus to promote growth and increase vitality ( Kiers et al., 2011 ). In this study, differences in rhizosphere fungal community structure of cultivated soybean genotypes with different tolerance levels to Al were assessed using high-throughput DNA sequencing of the internal transcribed spacer (ITS) region, and correlated with plant growth and chemical soil properties. Based on the higher adaptability of Al-tolerant soybean genotypes to Al toxicity, we hypothesized that (1) fungal diversity of Al-T genotypes is higher when compared to Al-S genotypes, and (2) the response of fungal community structure to aluminum between Al-T and Al-S soybean genotypes is different, and these differences increase with increasing Al concentration.",
"discussion": "Discussion The aim of this study was to reveal the effects of Al stress on the rhizosphere fungal community structure of aluminum sensitive (Al-S) and tolerant (Al-T) soybean genotypes. The Al tolerant genotype harbored more abundant and structurally different fungal communities when compared to the sensitive genotype ( Figures 1 , 2 ). This finding supports the hypothesis that the plant, besides directly secreting more organic acid to chelate Al, might also recruit specific fungal species to the rhizosphere that themselves secret organic acids for Al detoxification ( Kochian, 1995 ; Ma et al., 1997 ; Yang et al., 2012b ). Thus, different soybean genotypes secrete different amounts and types of organic acids that cause different response to Al toxicity ( Ryan et al., 2001 ; Kochian et al., 2015 ). However, fungal alpha-diversity showed no difference among the treatments, which is in contrast with our first hypothesis. This result indicated that fungal diversity was stable in the rhizosphere and was not affected by Al stress and soybean genotypes in this study. Notably, Al stress tended to increase difference in fungal community composition between the tolerant and sensitive genotypes, which is in accordance to what has been observed for bacteria ( Lian et al., 2019 ). Al stress reduced fungal abundance indicating that Al inhibited the growth of soil fungi ( He et al., 2012 ). Moreover, Al stress altered the rhizosphere fungal community composition of both genotypes ( Figure 2 ), which is consistent with previous studies suggested that Al affects fungal community structure ( Vosátka et al., 1999 ; He et al., 2012 ). However, the observation that fungal community structure differed between the genotypes even without Al stress is in contrast with the study by Wang et al. (2009) reporting that fungal communities of three soybean genotypes were not different at the same growth stage. This discrepancy could be explained by the fact that different genotypes were investigated, different soil types were tested, and fungal communities were assessed using lower-resolution molecular methods. Several studies have shown that Al tolerant genotypes can secrete more organic acids to chelate more Al ions ( Kochian, 1995 ; Ma et al., 1997 ; Innes et al., 2004 ; Yang et al., 2012b ). Based on the indicator species analysis, we have revealed that certain fungal genera significantly associated with the tolerant genotypes at high Al concentration. The comparisons among the treatments has identified several fungal taxa that have increased in relative abundance in the rhizosphere of the tolerant plant, including Penicillium , Cladosporium and Talaromyces . Penicillium has previously been shown to be highly tolerant to Al stress and therefore reduce Al toxicity by secreting organic acids and increasing soil pH ( Zhang et al., 2002 ). Previous study has also reported that Penicillium can promote plant growth via increasing nutrient status of plants ( Vessey and Heisinger, 2001 ). Cladosporium has been shown tolerant toward heavy metals, and could transfer phosphorus to the plant and promote plant growth under phosphorus deficiency, thereby cope with the Al toxicity ( Bewley, 1980 ; Shao and Sun, 2007 ; Hiruma et al., 2016 ). Talaromyces , which is close to Penicillium and has initially been described as a sexual state of Penicillium species, are also known to be tolerant toward heavy metals ( Yilmaz et al., 2014 ; Nam et al., 2019 ). Thus, the tolerance of soybean to Al toxicity may be closely related to the presence of these species. A considerably large fraction of the community, representing 26% of the OTUs, responded significantly to Al addition and plant genotype ( Figure 4 ). It has been suggested that aluminum contamination and soybean genotypes can affect entire clades of the rhizosphere microbial community structure by changing fundamental factors such as nutrition availability ( Xu et al., 2009 ; Yang et al., 2012a ). It is worth noting that some OTUs, e.g., OTU490 ( Penicillium decumbens ) and OUT 240 ( Penicillium simplicissimum ) associated with Al-S genotype under low Al concentration were affiliated to the genus Penicillium . Considering that this genus was significantly increased with Al stress and showed a higher relative abundance under the tolerant genotypes, it might contribute to Al tolerance with both sensitive and tolerant genotypes, and this contribution might be different for the two genotype groups ( Figure 4 ). Soil chemical properties, such as available phosphorus, NH 4 + -N, NO 3 – -N, exchangeable H + , exchangeable Al 3+ , and pH, were significantly correlated with shifts in fungal community structure of both genotypes ( Table 4 and Supplementary Figure S2 ), and all these chemical properties were significantly associated with changing Al concentrations ( Supplementary Table S1 ). These results suggested that the impacts of Al stress on the fungal communities might be directly linked to the alteration of soil chemical properties and vice versa. Co-occurrence networks showed substantial structural differences between the Al tolerant and sensitive genotypes at both low and high Al concentrations. At the high Al concentrations, the fungal network of the tolerant genotypes revealed more negative correlations and lower modularity, which could be interpreted as representing increased inter-species competition according to network theory ( Saavedra et al., 2011 ; Fan et al., 2018 ). Moreover, most fungal OTUs in Al-T genotypes are connected by positive links are considered to be unstable; in such a network, fungal OTUs may generated co-fluctuations and positive feedback along with environmental changes ( Coyte et al., 2015 ; Vries et al., 2018 ). Besides, fungal hubs might also play an important role in mediating Al toxicity. For example, potential hub OTU48 was assigned to Aspergillus , which can produce organic acids that might alleviate Al toxicity by forming complexes with Al ( Kawai et al., 2000 ). In conclusion, aluminum stress had no effect on fungal diversity, but increased differences in fungal community structure between the sensitive and tolerant genotypes with increasing aluminum concentrations. Fungal genera such Penicillium , Cladosporium , and Talaromyces increased with increasing Al concentration and were enriched under the tolerant genotypes at high Al concentration. A more complex structure in fungal co-occurrence networks was found for the tolerant genotypes at high Al concentrations. However, to what extent these “enriched” fungal taxa have an impact on Al detoxification is not yet known and subject to future, more mechanistic experiments. These experiments also need to be carried out in different soil types and under different climatic conditions in order to evaluate the universality of the findings. This study highlights the possibility that rhizosphere fungi involved in Al detoxification can be used as breeding target."
} | 3,091 |
27309381 | PMC4911146 | pmc | 9,558 | {
"abstract": "Providing the neurobiological basis of information processing in higher animals, spiking neural networks must be able to learn a variety of complicated computations, including the generation of appropriate, possibly delayed reactions to inputs and the self-sustained generation of complex activity patterns, e.g. for locomotion. Many such computations require previous building of intrinsic world models. Here we show how spiking neural networks may solve these different tasks. Firstly, we derive constraints under which classes of spiking neural networks lend themselves to substrates of powerful general purpose computing. The networks contain dendritic or synaptic nonlinearities and have a constrained connectivity. We then combine such networks with learning rules for outputs or recurrent connections. We show that this allows to learn even difficult benchmark tasks such as the self-sustained generation of desired low-dimensional chaotic dynamics or memory-dependent computations. Furthermore, we show how spiking networks can build models of external world systems and use the acquired knowledge to control them.",
"introduction": "Introduction The understanding of neural network dynamics on the mesoscopic level of hundreds and thousands of neurons and their ability to learn highly complicated computations is a fundamental open challenge in neuroscience. For biological systems, such an understanding will allow to connect the microscopic level of single neurons and the macroscopic level of cognition and behavior. In artificial computing, it may allow to propose new, possibly more efficient computing schemes. Randomly connected mesoscopic networks can be a suitable substrate for computations [ 1 – 5 ], as they reflect the input in a complicated, nonlinear way and at the same time maintain, like a computational “reservoir”, fading memory of past inputs as well as of transformations and combinations of them. This includes the results of computations on current and past inputs. Simple readout neurons may then learn to extract the desired result; the computations are executed in real time, i.e. without the need to wait for convergence to an attractor (“reservoir computing”) [ 1 , 2 ]. Non-random and adaptive network connectivity can change performance [ 6 – 8 ]. Networks with higher computational power, in particular with the additional ability to learn self-sustained patterns of activity and persistent memory, require an output feedback or equivalent learning of their recurrent connections [ 2 , 3 ]. However, network modeling approaches achieving such universal (i.e. general purpose) computational capabilities so far concentrated on networks of continuous rate units [ 2 , 4 ], which do not take into account the characteristics that neurons in biological neural networks communicate via spikes. Indeed, the dynamics of spiking neural networks are discontinuous, usually highly chaotic, variable, and noisy. Readouts of such spiking networks show low signal-to-noise ratios. This hinders computations following the described principle in particular in presence of feedback or equivalent plastic recurrent connections, and has questioned it as model for computations in biological neural systems [ 9 – 11 ]. Here we first introduce a class of recurrent spiking neural networks that are suited as a substrate to learn universal computations. They are based on standard, established neuron models, take into account synaptic or dendritic nonlinearities and are required to respect some structural constraints regarding the connectivity of the network. To derive them we employ a precise spike coding scheme similar to ref. [ 12 ], which was introduced to approximate linear continuous dynamics. Thereafter we endow the introduced spiking networks with learning rules for either the output or the recurrent connection weights and show that this enables them to learn equally complicated, memory dependent computations as non-spiking continuous rate networks. The spiking networks we are using have only medium sizes, between tens and a few thousands of neurons, like networks of rate neurons employed for similar tasks. We demonstrate the capabilities of our networks by applying them to challenging learning problems which are of importance in biological contexts. In particular, we show how spiking neural networks can learn the self-sustained generation of complicated dynamical patterns, and how they can build world models, which allow to compute optimal actions to appropriately influence an environment.",
"discussion": "Discussion The characteristic means of communication between neurons in the nervous system are spikes. It is widely accepted that sequences of spikes form the basis of neural computations in higher animals. How computations are performed and learned is, however, largely unclear. Here we have derived continuous signal coding spiking neural networks (CSNs), a class of mesoscopic spiking neural networks that are a suitable substrate for computation. Together with plasticity rules for their output or recurrent connections, they are able to learn general, complicated computations by imitation learning (plastic CSNs, PCSNs ). Learning can be highly reliable and accurate already for comparably small networks of hundreds of neurons. The underlying principle is that the networks reflect the input in a complicated nonlinear way, generate nonlinear transformations of it and use fading memory such that the inputs and their pasts interfere with each other. This requires an overall nonlinear relaxation dynamics suitable for computations [ 2 ]. Such dynamics are different from standard spiking neural network dynamics, which are characterized by a high level of noise and short intrinsic memory [ 9 – 11 , 69 ]. To find spiking networks that generate appropriate dynamics, we use a linear decoding scheme for continuous signals encoded in the network dynamics as combinations of membrane potentials and synaptic currents. A specific coding scheme like this was introduced in refs. [ 12 , 37 ] to derive spiking networks encoding linear dynamics in an optimal way. We introduce spiking networks where the encoded signals have dynamics desirable for computation, i.e. a nonlinear, high-dimensional, low-noise, relaxational character as well as significant fading memory. We conclude that, since we use simple linear decoding, already the dynamics of the spiking networks must possess these properties. Using this approach, we study two types of CSNs: Networks with saturating synapses and networks with nonlinear dendrites. The CSNs with saturating synapses use a direct signal encoding; each neuron codes for one continuous variable. It requires spiking dynamics characterized by possibly intermittent phases of high rate spiking, or bursting, with inter-spike-intervals smaller than the synaptic time constants, which leads to a temporal averaging over spikes. Dynamics that appear externally similar to such dynamics were recently highlighted as a ‘second type of balanced state’ in networks of pulse-coupled, intrinsically oscillating model neurons [ 51 ]. Very recently [ 70 , 71 ] showed that networks whose spiking dynamics are temporally averaged due to slow synapses possess a phase transition from a fixed point to chaotic dynamics in the firing rates, like the corresponding rate models that they directly encode. In the analytical computations the spike coding was not specified [ 70 ] or assumed to be Poissonian [ 71 ]. Numerical simulations of leaky integrate-and-fire neurons in the chaotic rate regime can generate intermittent phases of rather regular high-rate spiking [ 70 ]. The networks might provide a suitable substrate for learning computations as well. However, since the chaotic rate dynamics have correlations on the time scale of the slow synapses its applicability is limited to learning tasks where only a short fading memory of the reservoir is needed. For example delayed reaction tasks as illustrated in Fig 4a-4c would not be possible. Interestingly, in our scheme a standard leaky integrate-and-fire neuron with saturating synapses appears as a special case with recovery current of amplitude zero. According to our analysis it can act as a leaky integrator with a leak of the same time constant as the synapses, λ x = λ s . In contrast, in presence of a recovery current, our networks with saturating synapses can encode slower dynamics on the order of seconds. After training the network, the time scales can be further extended. In the CSNs with nonlinear dendrites the entire neural population codes for a usually smaller number of continuous variables, avoiding high firing rates in sufficiently large networks. The networks generate irregular, low frequency spiking and simultaneously a noise-reduced encoding of nonlinear dynamics, the temporal averaging over spikes in the direct coding case is partially replaced by a spatial averaging over spike trains from many neurons. The population coding scheme and our derivations of CSNs with nonlinear dendrites generalize the predictive coding proposed in ref. [ 12 ] to nonlinear dynamics. The roles of our slow and fast connections are similar to those used there: In particular, redundancies in the spiking are eliminated by fast recurrent connections without synaptic filtering. We expect that these couplings can be replaced by fast connections that have small finite synaptic time constants, as shown for the networks of ref. [ 12 ] in ref. [ 72 ]. In contrast to previous work, in the CSNs with nonlinear dendrites we have linear and nonlinear slow couplings. The former contribute to coding precision and implement linear parts of the encoded dynamics, the latter implement the nonlinearities in the encoded dynamics. Further, in contrast to previous work, the spike coding networks provide only the substrate for learning of general dynamical systems by adapting their recurrent connections. Importantly, this implies (i) that the neurons do not have to adapt their nonlinearities to each nonlinear dynamical system that is to be learned (which would not seem biologically plausible) and (ii) that the CSNs do not have to provide a faithful approximation of the nonlinear dynamics Eqs ( 6 ),( 11 ), since a rough dynamical character (i.e. slow dynamics and the echo state property) is sufficient for serving as substrates. We note that refs. [ 73 , 74 ] suggested to use the differential equations that characterize dynamical systems to engineer spiking neural networks that encode the dynamics. The approach suggests an alternative derivation of spiking networks that may be suitable as substrate for learning computations. Their rate coding scheme, however, allows for redundancy and thus higher noise levels, and it generates high frequency spiking. In a future publication, B. DePasquale, M. Churchland, and L.F. Abbott will present an approach to train rate coding spiking neural networks, with continuous rate networks providing the target signals [ 75 ]. We will discuss the relation between our and this approach in a joint review [ 76 ]. A characteristic feature of our neuron models is that they take into account nonlinearities in the synapses or in the dendrites. On the one hand this is biologically plausible [ 13 , 19 – 21 ], on the other hand it is important for generating nonlinear computations. Our nonlinearities are such that the decoded continuous dynamics match those for typical networks of continuous rate neurons and provide a simple model for dendritic and synaptic saturation. However, the precise form of the neuron model and its nonlinearity is not important for our approaches: As long as the encoded dynamical system is suitable as a computational reservoir, the spiking system is a CSN and our learning schemes will work. As an example, a dendritic tree with multiple interacting compartments may be directly implemented in both the networks with saturating synapses and in the networks with nonlinear dendrites. A future task is to explore the computational capabilities of CSNs incorporating different and biologically more detailed features that lead to nonlinearities, e.g. neural refractory periods, dendritic trees with calcium and NMDA voltage dependent channels and/or standard types of short term synaptic plasticity. Inspired by animals’ needs to generate and predict continuous dynamics such as their own body and external world movements, we let our networks learn to approximate desired continuous dynamics. Since effector organs such as muscles and post-synaptic neurons react to weighted, possibly dendritically processed sums of post-synaptic currents, we interpret these sums as the relevant, continuous signal-approximating outputs of the network [ 39 ]. Importantly, this is not the same as Poissonian rate coding of a continuous signal: As a simple example, consider a single spiking neuron. In our scheme it will spike with constant inter-spike-intervals to encode a constant output. In Poissonian rate coding, the inter-spike-intervals will be random, exponentially distributed and many more spikes need to be sampled to decode the constant output (cf. Fig A in S1 Text ). The outputs and recurrent connections of CSNs can be learned by standard learning rules [ 4 , 41 ]. The weight changes depend on the product of the error and the synaptic or dendritic currents and may be interpreted as delta-rules with synapse- and time-dependent learning rates. PCSNs, with learning of recurrent weights or output feedback, show how spiking neural networks may learn internal models of complicated, self-sustained environmental dynamics. In our applications, we demonstrate that they can learn to generate and predict the dynamics in different depths, ranging from the learning of single stable patterns over the learning of chaotic dynamics to the learning of dynamics incorporating their reactions to external influences. The spiking networks we use have medium size, like networks with continuous neurons used in the literature [ 2 , 4 ]. CSNs with saturating synapses have, by construction, the same size as their non-spiking counterparts. In CSNs with nonlinear dendrites the spike load necessary to encode the continuous signals is distributed over the entire network. This leads to a trade-off between lower spiking frequency per neuron and larger network size (cf. Fig F in S1 Text ): The faster the neurons can spike the smaller the network may be to solve a given task. Previous work using spiking neurons as a reservoir to generate a high dimensional, nonlinear projection of a signal for computation, concentrated on networks without output feedback or equivalent task-specific learning of recurrent connectivity [ 1 , 50 , 77 ]. Such networks are commonly called “liquid state machines” [ 78 ]. By construction, they are unable to solve tasks like the generation of self-sustained activity and persistent memorizing of instructions; these require an effective output feedback, since the current output determines the desired future one: To compute the latter, the former must be made available to the network as an input. The implementation of spiking reservoir computers with feedback was hindered by the high level of noise in the relevant signals: The computations depend on the spike rate, the spike trains provide a too noisy approximation of this average signal and the noise is amplified in the feedback loop. While analytically considering feedback in networks of continuous rate neurons, ref. [ 3 ] showed examples of input-output tasks solved by spiking networks with a feedback circuit, the output signals are affected by a high level of noise. This concerns even output signals just keeping a constant value. We implemented similar tasks ( Fig 4d ), and find that our networks solve them very accurately due to their more efficient coding and the resulting comparably high signal-to-noise ratio. In contrast to previous work, our derivations systematically delineate spiking networks which are suitable for the computational principle with feedback or recurrent learning; the networks can accurately learn universal, complicated memory dependent computations as well as dynamical systems approximation, in particular the generation of self-sustained dynamics. In the control task, we show how a spiking neural network can learn an internal model of a dynamical system, which subsequently allows to control the system. We use a path integral approach, which has already previously been suggested as a theory for motor control in biological systems [ 79 , 80 ]. We apply it to learned world models, and to neural networks. Path integral control assumes that noise and control act in a similar way on the system [ 61 ]. This assumption is comparably weak and the path integral control method has been successfully applied in many robotics applications [ 81 – 83 ], where it was found to be superior to reinforcement learning and adaptive control methods. Continuous rate networks using recurrence, readouts, and feedback or equivalent recurrent learning, are versatile, powerful devices for nonlinear computations. This has inspired their use in manifold applications in science and engineering, such as control, forecasting and pattern recognition [ 26 ]. Our study has demonstrated that it is possible to obtain similar performance using spiking neural networks. Therewith, our study makes spiking neural networks available for similarly diverse, complex computations and supports the feasibility of the considered computational principle as a principle for information processing in the brain."
} | 4,400 |
32555223 | PMC7303173 | pmc | 9,559 | {
"abstract": "Robust methods for the characterisation of microbial biosignatures in geological matrices is critical for developing mineralogical biosignatures. Studying microbial fossils is fundamental for our understanding of the role microorganisms have played in elemental cycling in modern and ancient environments on Earth and potentially Mars. Here, we aim to understand what promotes the fossilisation of microorganisms after the initial stages of biomineralisation, committing bacteriomorphic structures to the geological record within iron-rich environments. Mineral encrusted cell envelope structures were routinely identified within a goethite-rich vein that cross-cut the saprolite (iron ore) of a weathered banded iron formation (BIF) system in the Quadrilátero Ferrífero, Brazil. The preservation of potential organic and mineralogical biosignatures associated with these fossils was characterised using the following high-resolution analytical techniques: scanning and transmission electron microscopy, focused ion beam scanning electron microscopy, nanoscale secondary ion mass spectrometry, synchrotron-based Fourier transform infrared spectroscopy and Raman spectroscopy. Electron microscopy demonstrated that mineral nucleation associated with a range of cell envelope structures typically followed the extant cell templates. These biologically-influenced iron-rich minerals are microcrystalline with minimal secondary growth. In contrast, intracellular mineralisation formed larger minerals that grew inward from the cell membrane to infill intracellular voids after cell death. A three dimensional reconstruction of encrusted cell envelopes in a fossilised biofilm suggests that microorganisms may be able to replicate, during the initial stages of mineralisation. Carbon and nitrogen signatures are preserved associated with the cell envelope structures; however, there were no conclusive mineralogical biosignatures associated with the mineralised cell envelopes highlighting the classical importance of morphology and elemental biosignatures in determining the biogenicity of bacteriomorphic structures.",
"conclusion": "Conclusions Iron oxide encrusted microfossils, preserved in near-surface environments by the extensive mineralisation of the cell envelope structures, prevent the degradation of the soft organic material of microorganisms. A remarkable variety of microfossil textures are preserved. Following cell envelope mineralisation, continued exposure to iron-rich solutions results in the permineralisation of microfossils, as iron oxide minerals grow inwards to infill intracellular voids. Microorganisms appear to be able to replicate even while partially mineralised. As microorganisms are fossilised, they can infill pore and fissure spaces within near-surface rocks. These iron oxide plateaus that contain abundant microbial fossils on Earth may also be attractive geomorphological targets for the search for life, or remnants of life (fossils) on Mars.",
"introduction": "Introduction In 1996, the geoscience community was challenged with a question: What evidence is required to prove the existence of life 1 ? Characterisation of the ALH84001 meteorite presented 20 nm diameter tubular nanofossil structures, indirect organic signatures and mineral formation anomalies as evidence for past life. Intense scrutiny of these controversial results has led to questions regarding the robustness of these signatures as evidence of life; however, the work has been instrumental in stimulating scientists to better understand microbial fossilisation and the array of microbial biosignatures that may be preserved 2 . The surface of Mars is extremely inhospitable for life as we know it, with records of life on Mars likely to be restricted to mineralogical biosignatures 3 . Therefore, in-depth characterisation of microbial fossils here on Earth is required to develop new biosignatures that may be preserved in the geological record, particularly mineralogical biosignatures. Low temperatures and pressures, little-to-no atmospheric protection from ionising radiation and oxidising geological conditions provide little optimism for finding extant life. Future Mars Rover missions are being designed to collect, encapsulate and store samples from below the surface (up to 2 m depth) for later collection and transportation back to Earth. The potential success of such rover-based missions requires identifying key near-surface environments on Earth conducive to microbial fossilisation 4 . This article responds to a call for additional work and understanding of microbial fossilisation in pore and fracture filling near-surface environments 4 . Here, we have correlated a range of high-resolution analytical techniques to characterise well-preserved iron oxide encrusted microbial cell envelopes fossilised in vein structures below the surface (~15 m depth) to aid in the search for robust microbial biosignature targets. Studying the mineralogy associated with microfossils assists in constraining the environments in which microorganisms existed and their role in altering the biogeochemistry of their local environment. The continued development of nano- and microscale analytical techniques provides scientists with an increasingly large toolkit to more effectively characterise bacteriomorphic structures and determine their biogenicity. This article offers insights into the effectiveness of various analytical methods when assessing biogenicity of bacteriomorphic structures and the development of robust biosignature targets. In addition, this manuscript serves a timely reminder that we, as a scientific community, must maintain high standards for what we accept as microbial fossils as set out by Westall 5 to avoid ambiguity in the literature.",
"discussion": "Discussion Here, we present abundant mineralised cell envelope structures fossilised within a goethite-rich vein structure that cross-cuts a weathering BIF 6 . These indisputable microbial fossils provide a superb opportunity to understand potential mineralogical biosignatures preserved during microbial fossilisation by authigenic minerals. Rock fissures, fractures and pore spaces present ideal locations for the preservation of microorganisms as elements in solution percolate throughout the weathering profile. The multivalent oxidation states of iron and its low solubility in circumneutral pH environments make iron-rich regions an attractive target for microbial fossilisation in near-surface environments, both on Earth and potentially Mars. Cell envelope structures reduce the stability of cations in solution, altering the mineral phases that would otherwise have precipitated abiotically 13 . If these biominerals resist recrystallisation, unique mineral phase distributions may be preserved as mineralogical biosignatures 3 , 14 . In this study, crystal orientation and crystal size of the microbially-influenced iron oxide minerals appears to be altered by the microbial cell envelopes (Figs. 1 and 2 ) but no clear mineralogical biosignatures were preserved (Figs. 5 – 7 ). Although characterisation techniques used here are not exhaustive, there was no clear mineralogical difference between iron oxides that precipitated around the cell envelope and neighbouring iron oxide minerals in the matrix (Figs. 6 – 7 ), despite the apparent enrichment of aluminium associated with the cell envelope (Fig. 3 ). Therefore, the organic-mineral complexes that resist recrystallisation during the initial stages of mineral formation 15 do not appear to be preserved in million-year-old fossils. Given the low-temperature formation and alteration of the iron oxide minerals 9 , the aliphatic hydrocarbon moieties preserved with the microfossils are likely to be replaced during low-temperature metasomatism by iron and aluminium in solution. Levett et al. 6 demonstrated that organic biosignatures associated with younger (ca. 2 Ma) permineralised microfossils in the overlying duricrust are completely replaced, likely due to the increased cycling of iron oxide minerals that occurs near the surface 16 . In this study, mineral precipitates associated with cell envelope structures are extremely fine-grained compared with the post-death intracellular mineral precipitates (Figs. 6 – 7 ). Cations binding to active sites on the cell envelope 17 appears to create multiple mineral nucleation sites, restricting crystal growth. In contrast, post-death mineral precipitates within the intracellular voids have fewer mineral nucleation sites and, therefore, are allowed to grow in a less restricted manner. Consistent with Cosmidis, et al . 18 , intracellular mineral precipitates always grow from the cell envelope inward to fill the intracellular void. The 3D characterisation of iron oxide encrusted microbial fossils, provides an important opportunity to produce orientated reconstructions useful in the search for fossilised bacteria or biofilm (Video 1). Though FIB-SEM was used in this study, synchrotron-based nanotomography now offers non-destructive 3D reconstructions with submicron spatial resolution 19 . These technical developments offer unparalleled opportunities to understand the mechanisms that contribute to microbial fossilisation. The 100 nm resolution of the 3D reconstruction produced for multiple sections of each microbial fossil. The microfossils examined here are typically cocci-shaped and approximate 1 μm in diameter. Sarcina -like multicellular packet structures (for example, Supplementary Fig. S3 ) are never observed in the fossilised biofilm characterised here; however, rarely, paired cells are preserved that share a cytoplasm in a single section (Fig. 2 ). These textures provide evidence that the fossilised microorganisms presented here may be able to replicate during the initial stages of biomineralisation as has been previously postulated 20 – 22 . Microorganisms fossilised by authigenic mineral nucleation, rather than the binding of sediments within the biofilm 23 , provide valuable insights into the environmental conditions in which the living microorganism existed. These biogenic minerals may also provide information on influence of the microorganisms on their surrounding environment. Previous experiments have demonstrated the precipitation of lepidocrocite associated with neutrophilic iron-oxidising microorganisms 24 and nitrate-dependent iron-oxidising bacteria 25 . Given the microfossil structures in this study are generally cocci-shaped, they are unlikely to represent sheath structures of classic Leptothrix -type neutrophilic iron-oxidising bacteria 26 , though filaments do exist (see Video 1; 38–40 s, top right-hand corner). In addition, the low nitrogen availability in these environments 27 suggests that microfossils are also unlikely to represent nitrate-dependant iron oxidisers. The apparent binding of aluminium with cell envelope structures, indicates that these microorganisms were likely to have been preserved by the passive nucleation of minerals on the cells’ surfaces 14 , 28 . The formation of lepidocrocite together with goethite (Fig. 5 ) suggests that the pH was between 5–7. Lepidocrocite forms preferentially to goethite under slightly slower oxidation rates of iron 29 , indicating reduced partial pressures of oxygen in pore spaces below the surface compared with atmospheric conditions. In this BIF weathering profile, many cells contribute to mineral nucleation by the passive interaction of cations with the net negative cell envelope 28 . Even amongst the cells that contribute to biomineralisation 30 , few are likely to achieve a state of ‘fossilisation’, whereby they are preserved in the geological record. Extensive mineralisation is required to achieve microbial fossilisation and preservation. Based on earlier findings 6 , aluminium binding irreversibly with cell envelope structures appears to play an important role in the preservation of organic biosignatures; however, synchrotron-based FT-IR analysis could not resolve aluminium-organic complexes. The preservation of microorganisms in the geological record is rare. As such, an abundance of well-preserved microfossils in any environment requires careful consideration. To resist the breakdown of cellular components, particularly cell envelope structures and potentially EPS, rapid and extensive mineralisation is required. The influence of water in microbial fossilisation within the lithosphere is also likely to be critical. While fine-grained, generally amorphous iron oxide precipitates readily nucleate on cell envelope structures in iron-rich aqueous environments 28 , for example, 2-line ferrihydrite 31 ; these cells are unlikely to be fossilised within water saturated environments. Therefore, following this initial stage of biomineralisation during a cells exposure to cation-rich solutions, periods of drying appear to be imperative for fossilisation. During dehydration, any remaining ions in solution (in this case, predominately iron and aluminium), would be concentrated, accelerating additional mineral nucleation on the cells’ surfaces. Alternating wet and dry periods may be required to promote additional mineralisation. In this scenario, additional metals in solution would be provided during wet periods, which may allow for the recrystallisation of existing iron oxide minerals and the additional precipitation of new iron oxide minerals 32 . During drying periods, newly mineralised microorganisms may be committed to the geological record, contributing to preservation of relatively large microfossil clusters. The organic compounds associated with the cell envelopes are likely to be preserved by the electrostatic-driven nucleation of aluminium and iron oxide minerals within relatively oxidising environments 6 , 7 , 28 . As the microfossils are continuously exposed to aluminium and iron-rich solutions, the mineralised cell envelopes appears to act as a filter; iron is allowed into the cell whereas aluminium is enriched around the cell envelope 6 , 33 . The structure of the cell envelop appears to restrict aluminium transfer into the cell 33 , possibly even after cell death. Aluminium may also continue to be enriched around the cell envelope as it preferentially precipitates with existing aluminium-substituted iron oxide minerals that have previously nucleated on the cells’ surfaces 29 . In this way, even after all the organic components of the cell envelope are replaced and the cell has been completely permineralised, aluminium enrichment around the cells may help to preserve the bacteriomorphic structure within the geologic record 6 , 34 . Appropriate sample preparation for high-resolution analytical work is critical. Many techniques require a polished surface to spatially resolve distinctions between minerals influenced by microorganisms compared with ‘abiotic’ mineral precipitates. To study microfossils, sample preparation using a FIB-SEM offers a number of benefits including, targeted preparation of localised regions of interest without introducing organic contaminants. As a destructive sample preparation technique, great care and skill is required when preparing microfossil lamella using a FIB-SEM; however, this sample preparation technique is highly versatile 10 . Samples can be made thin enough to be analysed using transmission X-ray and infrared sources but also robust enough for NanoSIMS, a destructive secondary ion technique. Therefore, FIB-SEM sample preparation allows for highly targeted, polished sample preparation and for correlation between several different analytical datasets, as demonstrated in this study. Ultrathin samples (~100 nm thick) can also be prepared for high-resolution transmission electron microscopy (HR-TEM) and scanning transmission X-ray microscopy. The redistribution of elements that contribute to near-surface microbial fossilisation in rock pore spaces and fissures is fundamental to targeting drill regions for the identification of microfossils on samples from Mars. Organic biosignatures and abundant microbial fossils preserved in iron-rich environments highlights the potential to target iron-rich regions on the Martian surface for the search of potential microbial biosignatures. In depth characterisation of indisputable microbial fossils combining a suite of nano- and microscale analytical techniques sets important benchmarks for the identification of biosignatures within the geological record 35 . Additional studies of natural microbial fossils in a variety of environments is required to understand potential biosignatures preserved in different environments and aim to develop new robust biosignatures, for example, mineralogical biosignatures."
} | 4,170 |
34578061 | PMC8473281 | pmc | 9,562 | {
"abstract": "The objective of this study was to replace elastomer crosslinking based on chemical covalent bonds by reversible systems under processing. One way is based on ionic bonds creation, which allows a physical crosslinking while keeping the process reversibility. However, due to the weak elasticity recovery of such a physical network after a long period of compression, the combination of both physical and chemical networks was studied. In that frame, an ethylene-propylene-diene terpolymer grafted with maleic anhydride (EPDM-g-MA) was crosslinked with metal salts and/or dicumyl peroxide (DCP). Thus, the influence of these two types of crosslinking networks and their combination were studied in detail in terms of compression set. The second part of this work was focused on the influence of different metallic salts (KOH, ZnAc 2 ) and the sensitivity to the water of the physical crosslinking network. Finally, the combination of ionic and covalent network allowed combining the processability and better mechanical properties in terms of recovery elasticity. KAc proved to be the best ionic candidate to avoid water degradation of the ionic network and then to preserve the elasticity recovery properties under aging.",
"conclusion": "4. Conclusions The objective of this work was to improve the mechanical properties of an EPDM-g-MA (more precisely, the compression set) while keeping the processability of the samples. In that frame, the influence of a covalent and an ionic-reversible network has been studied. Samples crosslinked with DCP have the expected recovery elasticity but to the detriment of the processability. On the other hand, the ionic networks can be processed but with limited recovery plasticity properties (CS = 80%). Combining both networks, an improvement of the recovery elasticity (lower compression set values) is observed while maintaining the elasticity recovery properties. Actually, there is a compromise between the processability and relevant mechanical properties (CS around 55%). This brings forth a cooperative effect between the two types of crosslinking. The influence of the cations (K + , Na + , and Zn 2+ ) is compared, and the monovalent ions seem to be the best candidates in order to improve the elastic properties. The water aging was also studied, and it was observed that potassium acetate combines both good elastic recovery properties and water resistance properties. Finally, a compression set of 55% is obtained while maintaining processing processability without any negative effect of water treatment.",
"introduction": "1. Introduction The challenge of obtaining elastomers’ reversible crosslinking to meet classical thermoplastics melt processing is elegant and of importance in terms of industrial applications and circular economy. This concept of reversibility is most of the time associated with the possibility of processing at high temperatures (160–250 °C) while maintaining elastic properties at moderate temperatures (−30 to 120 °C). These temperature ranges are indicative and depend on the systems and applications, but for the automotive market, significant elastic properties are currently required up to temperatures of 120 °C (ASTMD 395 standards, for example). In terms of elastic recovery and processability at high temperatures, Thermoplastic Elastomers (TPE) meet these requirements on the principle of reversible networks based on a chemical nature and structuration. Two types of TPE can be defined. First, some block copolymers present soft blocks with elastomeric behavior and rigid blocks that act like a thermoplastic phase. This specific composition and nature will lead to phase separation, which causes remarkable elastic properties at room temperature. The best known of these TPE is poly(styrene-butadiene-styrene block) block copolymer (SBS). It is composed of polybutadiene blocks ( T g ≈ −100 °C) and polystyrene blocks ( T g ≈ +100 °C). Below the glass transition of the polystyrene (PS), this material behaves as a crosslinked rubber [ 1 ] due to the widely separated phases at room temperature. Beyond the T g of the PS phase, the material can be a priori processed. However, this also leads to a loss in the elastic recovery properties [ 2 , 3 ]. Actually, thermoplastic elastomers have weak elastic properties at temperatures above 100 °C and specifically under long times of compression. Secondly, a polymer blend with a thermoplastic and an elastomer permit combining the properties of each component. Indeed, the idea is to associate a crosslinked elastomer phase to a thermoplastic phase in order to give the final material both elasticity and processing properties. This objective has been achieved using the dynamic crosslinking process [ 4 ], which consists of crosslinking a major elastomer phase concomitantly to its blending with a minor thermoplastic phase. Finally, the crosslinked elastomer is dispersed in the thermoplastic phase. This has led to the emergence of Thermoplastic Vulcanizated (TPV), which has been a real industrial success. Several TPV materials have been developed in recent years, but the most widespread are still based on reactive blends from polypropylene (PP) and ethylene-propylene-diene terpolymer (EPDM) [ 5 , 6 , 7 ]. Despite their success, these TPVs have some limitations due to sub-products of reactions such as organic volatile compounds. Another possibility could be to turn to reversible chemistries. The reversible reactions of Diels–Alder [ 8 , 9 ] have often been considered, and recently, the concept of “vitrimers” [ 10 , 11 , 12 ] has opened up new fields of research and potential applications. However, these chemistries are still not mature for industrial applications because of their cost and the steps they require (chemical modification, grafting, etc.). Ionomer-based polymer materials were developed several years ago from the pioneering works of Eisenberg [ 13 , 14 ]. These polymers contain a small number of ionic groups (up to 15%mol) pendant or incorporated in the backbone [ 15 ]. The ionic groups (in most studies based on maleic anhydride [ 16 , 17 , 18 ] or other carboxylate groups [ 19 , 20 , 21 ]) tend to form ionic aggregates, which act as physical crosslinks. The reversibility is related to the increased mobility of ion pairs within ionic clusters under high temperature and shear stress allowing melt processing. At the same time, elastic properties at lower temperature are expected by recovering the cluster morphology by phase separation. Several studies have shown that the neutralization of these ionomers with different salts can influence the mechanical properties. For example, Van der Mee et al. [ 17 ] showed the improvement of the compression set at room temperature on EPDM-g-MA (88% without salts versus 15% for a K ionomer and 22% for a Zn ionomer both neutralized at 100%). Choi et al. [ 22 ] also worked on EPDM-g-MA neutralized with ZnO. They showed the increase in the crosslink density with the quantity of zinc. The compression set at room temperature is also decreased at 72% with 5 phr of ZnO. Rousseaux et al. [ 23 ] showed the influence of the concentration in salt on polypropylene grafted with maleic anhydride (PP-g-MA). As a result, the shear elastic modulus of the ionomers neutralized with NaOH increases from 1 × 10 3 Pa for a neutralization degree (ND) of 10% to 2 × 10 5 for a ND of 70%. Spencer et al. [ 20 ] showed similar results from ethylene methacrylic acid copolymers. The tensile modulus increases from 0.05 GPa without neutralization to 0.45 GPa for the K ionomers and to 0.40 GPa for the Na ionomers (both neutralized at 40%). In both cases, a plateau is observed at this neutralization degree. This type of technology was also used to develop materials with good shape memory capability such as those of Salaeh et al. [ 24 ] or Wang et al. [ 25 ]. Unfortunately, elastic recovery properties are generally required at high temperatures, typically in the temperature range 80–120 °C. Under these test conditions, the use of the elastomers-based ionomer is not suitable or at least very restricted. This has already been pointed out by Mora-Barrantes et al. [ 21 ]. Actually, to meet this challenge, they then proposed the principle of a combined covalent crosslinking network, i.e., a radical formed network combined with an ionic network. They used a carboxylated nitrile rubber (XNBR) with magnesium oxide (MgO), which will neutralize the carboxylic groups and form the ionic network. They demonstrated that the appropriate combination of covalent and ionic crosslinking allowed controlling the structure of the network of ionic elastomers in order to obtain suitable properties for these materials. They managed to improve the tensile properties of the samples in comparison of the covalent systems. For example, in their system, the modulus at 300% deformations increases from 2.3 MPa with a covalent system to 12 MPa combining both ionic and covalent systems. Unfortunately, the properties are studied until 70 °C and do not answer to the problematic at very high temperature. Based on this same concept, the objective of the present work is to study the elastic recovery properties (compression set at T = 100 °C) and the processability of an EPDM-g-MA crosslinked by a twin network, i.e., a covalent network coupled with an ionomer network. We will study successively the covalent network and the ionic network before scrutinizing their combination and its impact on the expected physical properties. Different metallic salts will be studied as well as their sensitivity to hot water ( T = 80 °C) from the dependence the equilibrium modulus, the gel fraction, and the compression set.",
"discussion": "3. Results and Discussion 3.1. Covalent Crosslinking Figure 1 shows the dependence of the storage modulus on the peroxide concentration used to crosslink the EPDM-g-MA. As expected, it can be observed that the higher the peroxide concentration, the higher the storage modulus. This is consistent with the results of Le Hel et al. [ 27 ]. Actually, the regime of the permanent elasticity is not reached at the lowest frequencies, or in other words, a dissipative phenomenon still exists due to the low crosslinking degree and/or the imperfections of the chemical network. Furthermore, from a processability aspect, the samples can be transformed into a uniform film for the lowest DCP concentrations, 0.2 and 0.4 wt.%, respectively. For the highest DCP concentrations (0.8 and 1.5 wt.%), it was not possible to reach a homogeneous film anymore. Consequently, the concentration of 0.4 wt.%, which corresponds to a gel fraction of 0.88, appears to be the critical one in view of the processability of the crosslinked EPDM samples. From a quantitative point of view, the maximum crosslinking density obtained for a concentration of 1.5% DCP was calculated according to G e = ν R ⋅ R ⋅ T ⋅ G F 1 / 3 , where Ge is the equilibrium modulus, ν R is the crosslinking density, R is the gas constant, T is the absolute temperature, and GF is the gel fraction. From this equation, we obtain a crosslinking density of 50 mol/m 3 for 1.5 wt.% of DCP, which is in agreement with previous work on the crosslinking of EPDM [ 27 ]. For 0.2 wt.% DCP, the crosslinking density is less than 7 mol/m 3 . This calculation was done using G ′ at ω = 0.01 rad/s, so the crosslinking density is overestimated. The compression set aims to quantify the elasticity recovery at long times under a constant deformation. Actually, a compression set not equal to zero proves a dissipative phenomenon at long times. This dissipative behavior depends on different parameters such as the crosslinking density and the nature of the crosslinking chemistry used. It has been demonstrated that the higher the crosslinking density, the better the compression set [ 27 , 28 ]. In terms of chemistry, the radical chemistry leads to the formation of an imperfect network with the presence of dangling chain ends that are at the origin of this dissipative behavior [ 28 , 29 ]. Consequently, and as expected, the compression set ( Figure 2 ) decreases with increasing the gel fraction and so the crosslinking density. The dangling chain ends becomes shorter. The recovery elasticity is more pronounced with increasing the crosslinking density, which is proportional to the DCP concentration [ 27 ]. In the case of a radical crosslinking, we can go down to a compression set around 30%, but to the detriment of the sample processability. Finally, these results mean that a processable sample crosslinked with DCP cannot have a compression set lower than 80%. 3.2. Ionic Crosslinking First, we studied the potentiality of KOH to react with the grafted maleic anhydride (EPDM-g-MA) according the reaction scheme in Figure 3 . Note that the sample with a neutralization degree of 100% in KOH was transformed in the form of powder inside the mixer chamber. This means that the crosslinking is too important and the sample is not processable anymore. Based on these first results, the impact of the KOH concentration (with a maximum ND of 75%) on the evolution of the storage modulus of the ionic crosslinked EPDM-g-MA is reported in Figure 4 . As expected, the higher the KOH concentration, the higher the storage modulus. In fact, we can observe that at low frequencies, the storage modulus moves from 2 × 10 3 Pa without KOH up to 1 × 10 5 Pa for a ND of 75%. Generally speaking, a strong increase in the elasticity behavior can be observed. According to the model proposed by Eisenberg [ 13 , 31 ], the KOH is reacting with the maleic anhydride, and the ion pair tends to aggregate (inter- or intra-chain interaction), forming a physical network ( Figure 3 ). This physical structuration limits the polymers chains’ mobility in comparison with the bulk neat polymer. The ionic-rich domains act as crosslinks and points of reinforcement for the elastic properties. These observations are clearly in agreement with the results reported by Li et al. [ 30 ] dealing with thermoreversible K ionomers based on butyl rubber. Based on the reaction between the maleic anhydride and the KOH, they demonstrated that the reprocessability of their system is observable for a degree of neutralization up to 70% (75% in our case) and that the ionic domains are microphase separated as evidenced by X-ray diffraction (XRD) analysis. Once again, the higher the KOH concentration, the higher the gel fraction and so the crosslinking density ( Figure 5 ). We reached a maximum of 63% for a ND of 50%. Qualitatively, one can then expect a more pronounced dissipative effect compared with the chemical network and consequently poor recovery elasticity properties. Actually, the limit of the compression set that has been reached is 80% ( Figure 5 ). Van der Mee et al. [ 17 , 18 ] show some similar results on EPDM-g-MA neutralized with KAc. Indeed, a compression set of 100% has been measured at 100 °C for the EPMD-g-MA alone. For an ND at 100, the compressions set at 39%. The same tendencies are observed: the higher the metallic salt concentration, the lower the compression set values. They also observed that the samples with an ND-100 are not completely homogenous. This is similar to what was observed in this study, namely that the sample at ND-100 is not processable. Furthermore, the optimum KOH concentration corresponds to a neutralization degree of 50%. Over this concentration, the compression set remains constant and around 80%. This result means that an ionic network does not lead to relevant elastic recovery properties, even if the storage modulus values correspond to a high elastic behavior. We have to be aware that the complex shear modulus is measured under linear deformation ( γ < 0.05), whereas the compression set is measured under nonlinear condition ( ε 0 = 0.25). Consequently, under the nonlinear conditions of the compression set, the ionic network can be reorganized and rearranged (due to its ionic nature) losing a part of recoverable elasticity. However, the positive effect associated to this dynamic ionic system is that all the samples can make a film under large deformation. The ionic bonds allow chain mobility, which explains the capacity to be recycled. Finally, whatever the crosslinking route (ionic or radical networks), it is not possible to obtain significant compression set properties while maintaining the processing property of the EPDM samples. In fact, a recovery elasticity target could be a compression set lower than 50%; unfortunately, the results show that it is not possible to go lower than 80% while being processable, in our experimental conditions. 3.3. Combination of Covalent and Ionic Networks 3.3.1. KOH-Based Systems According to the works of Mora-Barrantes et al. [ 21 ], an improvement of the elasticity recovery from the combination of both covalent and ionic networks can be expected. Our main objective is to maintain the processable properties of the crosslinked samples, so we fixed the DCP concentration at 0.4 wt.%. As a result, Figure 6 shows the variation of the storage modulus for the covalent network (0.4 wt.% DCP, from Figure 1 ), the KOH-based ionic network (ND = 50 from Figure 4 ), and the combination of both networks. The cooperative effect of both networks can be pointed out. The storage modulus is higher with both crosslinking systems in comparison with the EPDM-g-MA alone or with only one crosslinking agent. For example, a crosslinking density of 67 mole/m 3 is measured with the double crosslinking ( Ge = 2 × 10 5 Pa). This is much higher compared with the crosslink density of the respective network ( ν = 11 mol/m 3 , Ge = 4 × 10 4 Pa for the covalent system and ν = 25 mol/m 3 , Ge = 8 × 10 4 for the KOH ionic system). Furthermore, the elastic recovery properties are considerably improved, as shown in Figure 7 . For example, for 0.4 wt.% of DCP, the compression set decreases from 80% to 60% with the combination of both networks. Furthermore, this cooperative effect is observed at the different concentrations of DCP. One explanation of such behavior has been addressed by Mora-Barrantes et al. [ 21 ] who studied the impact of both DCP and MgO on XNBR crosslinking. They explained that the covalent bonds reduce the mobility of the polymers chains and thus modify the aggregation of the ionic domain. As a consequence, the mixed systems are formed by more ionic crosslinked domains but they are smaller in size than for the pure ionic systems. This is demonstrated by the decrease in the aggregation number for the mixed systems [ 15 ] for a pure ionic system—4 phr of MgO—versus 6 for the mixed sample—4 phr of MgO and 2 phr of DCP). The changes in mechanical properties depend of various factors, but one of the most important is the network structure. The synergic effect between both types of crosslinking allowed an evolution of the network structure, which is beneficial for the compression set properties. Finally, with the twin crosslinking system, we managed to decrease until 60% of the compression set (versus 80% with only the covalent network or only the ionic network). 3.3.2. Impact of the Nature of Ions From these cooperative effects of combined networks, it is of importance to study the influence of the counterion nature and consequently the most appropriate one in terms of elasticity recovery. The effects of the combination of both crosslinking systems on the gel fraction, storage modulus, and compression set were then examined. First of all, Figure 8 shows the storage modulus of the EPDM-g-MA with 0.4 wt.% of DCP and neutralized at 100% with the different salts (except KOH, which is neutralized at 50%). For all ions used, no real difference can be observed between the four salts (KOH, KAc, ZnAc 2 and NaAc). The gel fraction ( Figure 9 a) in the range between 90 and 100% which clearly evidenced the crucial contribution of the covalent crosslinking. It must be pointed out that all samples can be processed. Furthermore, it can be noticed that the ZnAc 2 allowed obtaining the highest values of gel fraction (more than 90% against 63% for the KOH with a DN of 50%) of the sample gel fraction. Surprisingly, these differences observed on the GF are not observable on the storage modulus. These results highlight that a divalent cation (Zn 2+ ) is more favorable to reach high gel fraction than the use of monovalent ions (K + and Na + ). This can be actually related to the specific coordination behavior of the Zn 2+ cation with the dicarboxylate moieties of the EPDM-g-MA. This aspect was evidenced for example by Grady et al. [ 32 ], who depicted that the Zn 2+ cation would preferentially coordinate with two carboxylate groups from two distinct maleic anhydride species, which can thus lead to a high gel fraction. These results are consistent with the observations reported by Stevens et al. [ 33 ] and Bragrodia et al. [ 34 ]. These former authors also described the bivalent Zn 2+ cation as the more “covalent” character of the zinc. For the monovalent cation, the difference between the K + and Na + action is less pronounced in particular for the DN = 50, as for example, the size of the cation is in the same range (102 and 140 pm for respectively the sodium and potassium cations). Similarly, Wang [ 25 ] demonstrated a better shape memory effect of polyacrylate crosslinked by zinc salt. The nature of the counterion has to be considered. Likewise, the acetate counterion is favorable to increasing the gel fraction relative to the hydroxide when there is no combination with covalent crosslinking. This is coherent with the results shown by Wang et al. [ 25 ]. The acetate is the counterion that shows the best enhancement effect on tensile properties of Zn ionomers. The large anion group contributes to the dissociation of zinc salts, which can explain the best results observed with the acetate. Finally, we can observe that these differences are leveled when the combination covalent/ionic crosslinking reactions are used. However, the expected correlation between gel fraction and compression set was not observed ( Figure 9 a,b): namely, the higher the crosslinking rate, the better the compression set. In fact, ZnAc 2 has the best crosslinking rates, yet it has the poorest compression set. The same phenomenon was related by Van der Mee et al. [ 17 ], who compared the ionic crosslinking of an EPDM-g-MA with respectively the ZnAc 2 and KAc. Actually, they showed that the tensile properties and elasticity in particular of the compression set are better for the K ionomers than for the Zn ionomers due to the relatively weak microphase separation for the latter ionomers. This could also be explained by the characters “hard” and “soft” of the acids defined by Bagrodia et al. [ 34 ]. Indeed, Zn 2+ is at the limit between hard and soft acids, and K + and Na + are very hard acids. The carboxylate sites should associate more strongly with the potassium/sodium acetate than with the zinc acetate. It also has been demonstrated by Stevens et al. [ 33 ] that monovalent and divalent ions do not prefer the same number of ligands. Zn 2+ prefer larger clusters ( n = 3), while K + and Na + prefer smaller clusters with n = 2 ligands. So, the samples have different network structures, which can explain the differences observed in the elastic recovery properties compared with gel fractions. The cooperative effect between both types of crosslinking observed previously for the KOH/DCP-based system is confirmed here for all the other salt-based systems. The evolution of the network structure is beneficial for the compression set properties. However, according to the previous observations, the compression set for the system elaborated with the ZnAc 2 remains higher even with the contribution of the DCP covalent crosslinking (minimum of 70%). Finally, with the double crosslinking system, we managed to decrease until 55% of the compression set for the monovalent cation-based system (versus 80% with only the covalent network or only the ionic network). This confirms the major influence of the nature of the cation and its ability to enhance a micro-phase separation. All these samples are processable, so a compromise is obtained at 55% of the compression set. 3.3.3. Water Aging In terms of real and industrial applications, the question concerning the water resistance of these ionic bonds and their dissociation over time and under high temperature (80 °C) conditions is fundamental. Thus, the influence of the water (48 h of immersion in water at 80 °C), depending on the ions used for the neutralization, is shown in Figure 10 . We can notice that when the samples are ionically crosslinked, a decrease in the gel fraction is observed after the water treatment for all the metallic salts used ( Figure 10 a). This observation can be related to the well-known water solubility of these salts. Classically, at room temperature, the water solubility can be classified as from the less soluble to the most soluble: NaAc < ZnAc 2 < KOH < KAc. To illustrate, for the KAc-based ionic system with a DN = 100, the gel fraction decreases drastically from almost 80% to less than 60%. In the same conditions, the gel fraction is almost not changed for the use of NaAc. However, in the presence of both networks, this decrease is not noticeable. Actually, as discussed by Mora et al. [ 21 , 35 ], the dual crosslinking action leads to smaller ionic domains, which can be less accessible to water molecules. In that case, the gel fraction remains almost unchanged in our water action conditions. It can be noticed for all the salts used, except for NaAc, that the water treatment in our experimental conditions has no influence on the compression set ( Figure 10 b). On the contrary, with the NaAc, a degradation of the mechanical properties after water treatment is observed. The compression set increases by almost 20% after the water treatment with or without the covalent crosslinking with peroxide, even if this salt is the less water soluble of our series. Van deer Mee et al. [ 18 ] discussed specifically this water action in the case of EPDM-g-MA crosslinked by respectively ZnAc 2 and KAc. They described that the water absorption can induce two competing effects: either opening residual maleic anhydride to yield two carboxylic acid groups, giving the capability to form hydrogen bonds or the promotion of the acid–cation exchange relaxation mechanism as described by Vanhoorne and Register [ 36 ]. In this publication, the former authors in particular demonstrated that unneutralized acid groups substantially reduce the viscosity of Na ethylene-methacrylic acid (EMAA)-based ionomers through this acid–cation exchange mechanism or through “plasticization” of the ionic aggregates but did not affect the viscosity of the similar Zn ionomers. Some other studies [ 37 ] in the literature present the same influence of water on ionomeric ethylene with different salts and are consistent with our experiments. Indeed, the EMAA neutralized with zinc does not absorb water, which explains the constancy of the results on stiffness and the small water contents of the samples aged in water (0.22% for 90 days at 100% of relative humidity). On the other hand, the EMAA neutralized with sodium absorbs more water, which will disturb the aggregates and thus the properties (decrease in stiffness in the presence of water and 3.80% of water for 90 days at 100% of relative humidity). For the potassium, the effect of water is not very clear, because the samples tend to absorb water similarly to sodium, but we do not see any changes."
} | 6,917 |
31800619 | PMC6892483 | pmc | 9,565 | {
"abstract": "Host-mediated microbiome engineering (HMME) is a strategy that utilizes the host phenotype to indirectly select microbiomes though cyclic differentiation and propagation. In this experiment, the host phenotype of delayed onset of seedling water deficit stress symptoms was used to infer beneficial microbiome-host interactions over multiple generations. By utilizing a host-centric selection approach, microbiota are selected at a community level, therein using artificial selection to alter microbiomes through both ecological and evolutionary processes. After six rounds of artificial selection using host-mediated microbiome engineering (HMME), a microbial community was selected that mediated a 5-day delay in the onset of drought symptoms in wheat seedlings. Seedlings grown in potting medium inoculated with the engineered rhizosphere from the 6 th round of HMME produced significantly more biomass and root system length, dry weight, and surface area than plants grown in medium similarly mixed with autoclaved inoculum (negative control). The effect on plant water stress tolerance conferred by the inoculum was transferable at subsequent 10-fold and 100-fold dilutions in fresh non-autoclaved medium but was lost at 1000-fold dilution and was completely abolished by autoclaving, indicating the plant phenotype is mediated by microbial population dynamics. The results from 16S rRNA amplicon sequencing of the rhizosphere microbiomes at rounds 0, 3, and 6 revealed taxonomic increases in proteobacteria at the phylum level and betaproteobacteria at the class level. There were significant decreases in alpha diversity in round 6, divergence in speciation with beta diversity between round 0 and 6, and changes in overall community composition. This study demonstrates the potential of using the host as a selective marker to engineer microbiomes that mediate changes in the rhizosphere environment that improve plant adaptation to drought stress.",
"conclusion": "Conclusions In summary, we showed that HMME can be used to enhance seedling drought tolerance at least under our experimental conditions. The findings from this study support the hypothesis that host mediated microbiome engineering can be used to alter the root rhizosphere microbiome and demonstrate the potential of engineered microbiomes to mediate changes in the rhizosphere environment, effectuating improved plant adaptation to water stress. Future research to test these observations should include mechanistic studies that profile the plants under duress from water deficit, exopolysaccharide production characterization assays of the microbiomes from each generation, and shotgun metagenomic sequencing for hologenome characterization and the observation of plasmid exchange, horizontal gene transfer, and changes in allelic frequency.",
"introduction": "Introduction Host-mediated microbiome engineering (HMME) is a cycle-dependent strategy that indirectly selects microbiomes based on host phenotype ( Fig 1 ). For example, by directly selecting for increased seedling water stress tolerance, the host phenotype (e.g., delayed onset of seedling water deficit stress symptoms) is used to indirectly select for beneficial microbiome-host interactions over multiple generations using the same host germplasm. In this host-centric selection process, all microbiota are sub-selected at a community level, rather than on an individual basis [ 1 ]. This method allows microbiomes to change through both ecological (e.g., diversity, relative abundance) and evolutionary (e.g., extinction events, alterations in allele frequency, mutation, horizontal gene transfer) processes [ 2 ]. Previous research demonstrated that HMME can indirectly select microbiomes for enhanced growth under altered soil pH by utilizing above ground biomass as a selection marker in A . thaliana [ 1 ]. Similarly, HMME was used to cultivate microbiomes capable of altering flowering onset and leaf biomass [ 3 , 4 ]. In Brachypodium distachyon , results suggested HMME indirectly selected a rhizosphere microbiome that conferred salt-tolerance measured through the host phenotype [ 5 ]. 10.1371/journal.pone.0225933.g001 Fig 1 Concept of host mediated microbiome engineering (HMME). 1) An initial microbiome is inoculated. 2) Seeds are planted into well-watered conditions. 3) Emerging seedlings are then exposed to a drought stress by withholding watering. 4) When 90% of the plants display symptoms of water stress (wilting, leaf curling, etc.), the 5 best-performing plants are selected. Their rhizospheres (roots and planting medium) are amalgamated with autoclaved Metro-Mix 900 in a 1:10 ratio. The next round of selection is then initiated by planting seeds into the engineered planting medium. In the present study, we sought to use HMME to improve wheat ( Triticum aestivum ) seedling establishment under severe water stress associated with lack of rainfall, since seedling establishment is often the most vulnerable stage and may have large impacts on crop stand and yield [ 6 ]. The host phenotype used for screening was the delayed of onset of drought stress symptoms in wheat seedlings establishing under waster deficit conditions. The source of the original was obtained from the rhizospheres of perennial grasses collected from El Paso, TX, where the semi-arid environment provides a strong selective pressure for survival under nearly constant water deficit. The rationale for choosing the starting material was that perennial grasses growing vigorously under pervasive water stress conditions were likely to foster a microbiome capable of mediating drought stress. In this HMME experiment, seeds of wheat cultivar TAM 111 were planted into well-watered planting medium inoculated with the microbiome from the grassland rhizosphere soil. Water was then withheld until 90% of the seedlings showed symptoms of extreme drought stress (wilting to collapse). The plants displaying the least drought stress were selected and their rhizospheres (roots and planting medium) were then used as inoculum for subsequent selection cycles. Each of the subsequent selection cycles were similarly halted when 90% of the seedlings experienced collapse and again the rhizospheres of the best performing plants were used as inoculum. The cycling was terminated when there were no further improvements in time to wilting. Additional objectives of the study were to 1) determine whether changes in plant growth and development were associated with HMME-mediated improvements in water stress tolerance by comparing plants grown with non-autoclaved versus autoclaved inoculum from the final round of selection and 2) characterize changes in the taxonomic and functional diversity of the wheat seedling microbiomes during HMME selection rounds.",
"discussion": "Discussion To our knowledge, this study provides the first example of utilizing HMME to indirectly select a rhizosphere microbiome that increased the seedling drought tolerance during water deficit conditions, by directly selecting for a delay in the onset of drought symptoms in newly established wheat seedlings. A steady increase in the ability of plants to survive water deficit occurred from the start of the experiment over the 6 rounds of HMME selection. Drought stress symptom onset delayed from 10 to 15 days. The improvements in drought tolerance stabilized after 6 rounds of HMME selection (and thus 6 rounds of 1:10 dilutions), and was transferable in subsequent 1:100 dilution, but lost if diluted further. The effect on seedling water stress tolerance was abolished by autoclaving the HMME inoculum. These results indicate that the change in plant adaptation to drought stress was a function of microbial population dynamics. The single greatest challenge in this HMME experiment lied in the experimental design, as it is impossible to separate soil components with microbes completely. Ideally, the negative control for the HMME study would have same rhizosphere components except microbes but reproducing this negative control treatment is extremely difficult to produce due to the iterative temporal nature of the study. We instead used the autoclaved HMME soil as the negative control to compare with the HMME soil and determine the effect of selected microbial community in the HMME soil on the host plants. However, we could not exclude any unknown effects from autoclaving soil. Alternatively, we also conducted the experiment that the HMME soil was diluted with non-autoclave potting mix to verify the effect of selected microbiomes. Previous studies have also utilized a “low-line” treatment consisting of the worst-performing pots as a control for the HMME experiment [ 1 , 2 , 4 ]. During the experimental design stage of our experiment, we identified that incorporating a low-line HMME soil selected based on a low-scoring plant phenotype in our screening method also caused the accumulation of pathogens in the soil, as previous reports have suggested [ 1 , 4 , 16 ]. Picking the worst-performing pots to carry through the continuous rounds of the HMME process resulted in severe diseases or complete wilts of tested plants due to the generational recruitment of plant pathogens. We agree that the low-line microbiome would have been a logically idealistic control, but the low-line soil that easily recruits and becomes dominated by pathogenic microbes could not serve as a control in our HMME study. Given these limitations at preparing the control soils to compare the HMME soil with, we were able to determine the effect of the HMME soil by two different methods, the comparison with autoclaved HMME soil and the serial dilution assay of the HMME soil using non-autoclaved potting mix. The increase in seedling drought tolerance with HMME was accompanied by changes in seedling growth and development. Seedlings grown with HMME inoculum were larger in size and had more extensive root system development than seedlings grown with autoclaved HMME inoculum. The observed alterations in plant growth and root system development are consistent with plant adaptation for maintaining plant productivity under drought [ 17 ]. Previous research showed that increased root length and surface area contribute to increased soil exploration for available water [ 18 – 21 ]. It is unclear the extent to which the HMME microbiome may have contributed directly to plant growth and development via plant growth promoting activities and/or indirectly via modifications to the rhizosphere environment. For example, plant growth-promoting bacteria have been reported to contribute to plant growth and root development via a number of mechanisms including suppression of seedling disease (although not a factor in this experiment) [ 22 ], production of plant growth regulating compounds [ 23 – 25 ], assistance in nutrient uptake [ 26 ], and mediation of redox stress [ 18 ]. Additionally, previous research has demonstrated that extracellular polysaccharide production from beneficial microbiota provides significant indirect benefits to plant growth and development via improved soil structure and increased soil water retention [ 21 , 27 ]. These improvements coupled with higher respiration rates (and associated water release) by microbial communities selected for rapid growth and colonization of wheat seedling rhizospheres may lead to further improvements in water availability in the engineered microbiomes, therein enabling plants to avoid drought stress longer. Indeed, we observed (both visually and in terms of weight/volume) greater water retention in the HMME soil at the start and throughout the experiment. Each round of HMME was associated with changes in taxonomic diversity and composition. As expected, dilution with HMME rounds of selection resulted in a reduction in alpha diversity [ 28 ]. Comparison of beta diversity indicated potential host-mediated changes in rhizosphere populations with successive generations, showing increasing divergence or dissimilarity from the original community composition. Similar trends in species community structure associated with dilution were noted previously, although functional profiles of rhizosphere communities resulting from dilution overlapped more [ 28 ]. Explaining this trend, the authors hypothesized that enrichment processes in the rhizosphere were more likely to select microbes with particular functionalities than taxonomies. The observed changes in relative abundance in Actinobacteria and Betaproteobacteria with each round of selection ( S1 Fig ) is a unique observation not typically seen in other drought related microbiome research studies [ 23 , 25 , 29 – 31 ]. Reasoning for the differences in this observation are currently unknown but may lie in the comparison of the differences in the objective and experimental design of the study. Our study of the wheat seedling rhizosphere microbiome under drought conditions consisted of an iterative microbiome propagation and differentiation strategy focused on ascertaining an increased understanding of an optimized microbiome to withstand seedling drought through simultaneous manipulation of ecological and evolutionary pressures. Previous research on root microbiomes associated with water stress utilize different environmental conditions, different mechanisms for implementing water stress, different stages of growth and different plant models. The most notable difference of our experiment from previous research is the integration of a microbiome selection, transplantation and propagation-based strategy, focusing on the phenotype of the host as a selection marker. In another study, we conducted an analysis of predicted functional metagenomic changes using the bioinformatic software package Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt) [ 32 ]. Results from this analysis yielded statistically significant increases level 2 KEGG orthologs involved in metabolism, signal transduction, cell processes and signaling, and cell motility when comparing across the R0 to R6 [ 33 ]. Pairwise comparisons between R0 and R6 with 95% confidence revealed similar significant increases in the KEGG orthologs associated with metabolism, signal transduction, cell processes and signaling, and cell motility [ 33 ]. The average nearest sequenced taxon index (NTSI), which reflect the relatedness to the reference genomes, were all less than 0.15 (> 85% similarity), indicating acceptable inference data quality [ 32 ]. Although this inference is at too coarse of a scale to draw many conclusions, enrichment in motility and metabolism are consistent with selection for a microbiome that can colonize and proliferate quickly (e.g., within 10 to 15 days of germination) in the seedling rhizosphere as would be selected by our HMME protocol."
} | 3,695 |
39195029 | PMC11353967 | pmc | 9,566 | {
"abstract": "Covalent cross-linked hydrogels based on chitosan and poly(maleic acid- alt -vinyl acetate) were prepared as spherical beads. The structural modifications of the beads during the preparation steps (dropping in liquid nitrogen and lyophilization, thermal treatment, washing with water, and treatment with NaOH) were monitored by FT-IR spectroscopy. The hydrogel beads have a porous inner structure, as shown by SEM microscopy; moreover, they are stable in acidic and basic pH due to the covalent crosslinking. The swelling degree is strongly influenced by the pH since the beads possess ionizable amine and carboxylic groups. The binding capacity for Cu 2+ ions was examined in batch mode as a function of sorbent composition, pH, contact time, and the initial concentration of Cu 2+ . The kinetic data were well-fitted with the pseudo-second-order kinetic, while the sorption equilibrium data were better fitted with Langmuir and Sips isotherms. The maximum equilibrium sorption capacity was higher for the beads obtained with a 3:1 molar ratio between the maleic copolymer and chitosan (142.4 mg Cu 2+ g −1 ), compared with the beads obtained using a 1:1 molar ratio (103.7 mg Cu 2+ g −1 ). The beads show a high degree of reusability since no notable decrease in the sorption capacity was observed after five consecutive sorption/desorption cycles.",
"conclusion": "3. Conclusions Hydrogel beads based on CS and MA-VA copolymers were obtained as new sorbent materials that cumulate the properties of both natural polycation and synthetic polyanion. The obtained beads are porous, with pores sizes ranging between 5 and 40 µm, allowing fast swelling and high water uptake values. The polymeric network of the beads is stable at acidic and basic pHs due to the covalent crosslinking between the amine and carboxylic groups. The sorption of Cu 2+ from aqueous solution onto CS/MA-VA beads is higher at an initial pH between 4 and 5.2. The sorption kinetics data are better fitted with the PSO model, meaning that the rate-control mechanism is chemisorption. The sorption equilibrium data were better fitted by Langmuir and Sips isotherms. The maximum equilibrium sorption capacity was higher for the CS/MA-VA3 beads possessing higher amounts of carboxylic groups (142.4 mg Cu 2+ g −1 ), compared with the CS/MA-VA1 beads with a higher amount of amine groups (103.7 mg Cu 2+ g −1 ). This difference demonstrates that interaction between carboxylic groups and the metal cations brings a greater contribution than that given by the coordination between -NH 2 and –OH groups from CS and the metal ions. Desorption of Cu 2+ from the CS/MA-VA beads was easily performed with 0.1 M HCl, and then the adsorptive sites were regenerated with 0.1 M NaOH. The hydrogel beads kept their sorption capacity and physical integrity after five sorption/desorption cycles, proving their reusability. In conclusion, the obtained hydrogel beads with good adsorption capacity and reusability can be regarded as new sorbents for Cu 2+ retention from wastewater.",
"introduction": "1. Introduction Heavy metal ions are one of the pollutants that can cause serious environmental problems. Copper has been identified as a major heavy metal contaminant of waste waters because it is a by-product of mining, is widely used in industry (electroplating, paints, pigments, fuel, catalysts, batteries), and in agriculture (in fertilizer, pesticides) [ 1 , 2 ]. In high amounts, copper is toxic to plants, animals, and humans [ 2 ]. Like other heavy metal ions, it can be removed from wastewater by different techniques (ion exchange, membrane filtration, etc.) [ 1 ], with absorption being one of the most effective methods. Good sorbents should be cost-effective, environmentally friendly, easily regenerated, and have high adsorption capacity and fast kinetics [ 3 , 4 ]. Porous polymeric hydrogels can fulfill these criteria, so they are proposed for these applications [ 5 , 6 , 7 ]. Compared with larger hydrogels, spherical beads have a higher surface area, and better mass transfer and diffusion behavior. Compared with nano-size sorbents, the beads have the advantages of easy recovery and reusability, properties that are required in water treatment [ 8 ]. Chitosan (CS), a cationic polysaccharide derived from chitin, is well-known for its applications in heavy metal ion removal [ 3 , 9 , 10 ]. CS has the advantages of its abundant availability, low cost, biocompatibility, biodegradability, and, most importantly, its high metal ion adsorption capacity due to the presence of amino and hydroxyl groups. CS’ disadvantages, like solubility in acidic environments and low mechanical properties, can be overcome by physical and chemical modifications [ 6 , 9 ]. Chemical modifications of CS suppose the stabilization of the polymeric network by cross-linking and the addition of new functionalities to increase the sorption capacities [ 9 , 11 ]. Among other derivatizations, grafting carboxylic groups onto CS was used to enlarge the pH solubility domain and increase its sorption properties [ 9 , 12 ]. For example, the polymerization of acrylic acid in the presence of CS and cross-linkers [ 13 , 14 , 15 , 16 ] or the grafting of maleic acid on CS [ 17 , 18 ] was used to obtain materials with a high absorption capacity for metal ions. Copolymers of maleic anhydride with different comonomers (styrene, methyl vinyl ether, acrylic acid, etc.) are known as anti-scale agents, as phosphate substituents in detergents, or for their pharmaceutical applications [ 19 , 20 ]. In the aqueous environment, the hydrolysis of the anhydride cycle from the maleic copolymers leads to the formation of two adjacent carboxylic groups with different acidity constants [ 21 ]. The complexation of the two adjacent carboxylic groups with divalent metal ions [ 22 , 23 ] leads to the utilization of the maleic copolymers as anti-scale agents [ 24 ], but also for the absorption of the metal ions from wastewater [ 25 , 26 , 27 ]. By combining maleic acid copolymers with CS, new materials with high metal ion adsorption capacity can be obtained. In our previous work, different maleic copolymers were used in water purification [ 28 ]. Among them, poly(maleic acid-alt-vinyl acetate) (MA-VA) proved to be the optimal polymer for obtaining microspheres with the highest dye adsorption capacity. This is why the hydrophilic copolymer MA-VA is proposed in the present study to prepare hydrogels with applications in metal ion removal. CS and maleic copolymers form polyelectrolyte complexes in aqueous solutions, but the physical electrostatic interactions between the two weak polyelectrolytes are not stable over the entire pH domain [ 29 , 30 ]. Stable chemical interactions are required for reusable materials applied to metal ion absorption. Thus, hydrogels obtained by grafting poly (acrylamide-co-maleic acid) on chitosan by gamma irradiation were recently obtained and used for the removal of cobalt or europium ions [ 31 , 32 ]. In the present paper, the covalent crosslinking between CS and maleic copolymer is performed by an amidation reaction. The strategy used to prepare stable porous spherical beads was the dropping of the CS/MA-VA mixture solution into liquid nitrogen, followed by lyophilization. Thermal treatment of the dried beads will induce covalent bonds between the polymers. Two ratios between the polymers were used for the preparation of the beads to evaluate how the ratio between the amine groups of CS and the carboxylic groups of the maleic copolymer influences the absorption of Cu 2+ ions. The structural modifications during the preparation steps were monitored by FT-IR spectroscopy. The morphology of the beads and their swelling behavior were followed. The sorption capacity of the new materials for Cu 2+ removal was investigated as a function of pH, contact time, and the initial concentration of metal ions. For a better understanding of Cu 2+ sorption mechanisms, the sorption kinetics and isotherms were fitted with different models. The reusability of the CS/MA-VA beads was also evaluated. To our knowledge, this is the first time CS/MA-VA hydrogels were obtained and employed for metal ion removal.",
"discussion": "2. Results and Discussion 2.1. Preparation and Characterization of CS/MA-VA Beads CS and MA-VA aqueous solutions were mixed, both polymers being in a protonated state to avoid the formation of a polyelectrolyte complex. Then, the polymeric solution was dripped into liquid nitrogen, and the resulting beads were dried by lyophilization. Subsequently, the beads were cross-linked by the amidation reaction between the amine groups of CS and the carboxylic groups of MA-VA copolymers under thermal treatment ( Figure 1 ). The beads were washed with distilled water to remove the uncross-linked polymers, and then with alkaline water to dissociate the remaining carboxylic groups of the maleic copolymer. Following this treatment, the amine groups of CS chains are in a non-protonated state, the optimal state that can be involved in the chelation with metal cations [ 9 ]. FT-IR spectroscopy was used to elucidate the structure of the beads after each treatment. Figure 2 presents the FT-IR spectra of the beads after the first freeze-drying (A), thermal treatment (B), anhydride hydrolysis (C), and finally treatment with NaOH (D). In the spectrum of CS/MA-VA beads initially obtained ( Figure 2 A), the characteristic bands for both CS and MA-VA copolymers can be observed at: 1735 cm −1 (carbonyl groups from maleic acid and vinyl acetate units), 1631 cm −1 (amide I from CS), 1518 cm −1 (–NH 3 + groups from CS chloride), 1381 cm −1 (–CH 3 from vinyl acetate), 1241 and 1175 cm −1 (–C–O stretching from vinyl acetate), 1086 and 897 cm −1 (C–O vibration from CS) [ 33 , 34 , 35 , 36 ]. After the thermal treatment ( Figure 2 B), new bands appeared at 1851 and 1778 cm −1 because of the de-hydration and re-formation of the anhydride cycle in the maleic copolymer [ 36 ]. The amide bonds formed between the carboxylic groups of the maleic copolymer and the amine groups in CS are overlapped by the amide bonds in CS. After washing the beads with distilled water ( Figure 2 C), the anhydride cycles are hydrolyzed with the obtaining of carboxylic groups, and the bands from 1851 and 1778 cm −1 were not observed anymore. The appearance of a new band at 1577 cm −1 can be due to the ionization of some of the carboxylic acid groups. After treatment with NaOH and washing with distilled water ( Figure 2 D), a large band can be observed between 1680 and 1480 cm −1 attributed to the amide bonds, dissociated –COO − groups from the maleic copolymer, and also –NH 2 groups from CS. Two molar ratios between the CS and maleic copolymer were used to prepare the beads, as shown in Table 1 . In order to determine the cross-linking degree of CS, the ninhydrin test was used [ 37 ]. The content of free –NH 2 groups in CS was determined before and after thermal treatment ( Table 1 ). The difference between these two values is given by the amine groups involved in the amidation reactions that take place under the thermal treatment. The gel fractions were relatively high for both samples, showing that the bead structure is stabilized not only by ionic interaction but also by the covalent cross-linking between the CS and MA-VA copolymers. The beads have spherical shapes, with diameters in their dried state ranging between 2.4 and 3.8 mm. SEM images ( Figure 3 ) showed that both samples have a porous inner structure with lamellar and interconnected pores, characteristic of beads prepared in liquid nitrogen and dried by lyophilization [ 38 ]. Compared with CS/MA-VA1 beads with a mean pores diameter of 20.1 µm, the CS/MA-VA3 beads have smaller pores (mean diameter 12.6 µm) and a narrower pores distribution ( Figure 3 C,D). This is due to the higher cross-linking degree of CS/MA-VA3 beads through amide linkage and electrostatic interaction. It is widely recognized that a high swelling capacity is essential for a good sorbent [ 5 ]. However, a too-high swelling degree can make materials brittle and fragile [ 39 , 40 ]. This drawback can be overcome by an increase in the cross-linking density, which lowers the hydrophilicity of the polymer [ 39 ]. Cross-linking can enhance the resistance of the polymer against acid, alkali, and chemicals but reduce the swelling degree and the efficiency of the uptake of pollutants. Therefore, the swelling kinetics of CS/MA-VA beads were performed. As shown in Figure 4 A, both CS-MA-VA1 and CS-MA-VA3 beads absorb high amounts of water in the first few minutes, then gradually swell and reach equilibrium in about 6 h. However, the CS/MA-VA3 beads, even if they show a high degree of cross-linking, have the highest swelling ratio due to the large content of dissociated carboxylic groups at the pH value of distilled water (5.4). The influence of the pH on the swelling capacity at equilibrium was studied for CS/MA-VA1 beads and presented in Figure 4 B. It is known that the intrinsic dissociation constants for MA-VA copolymers in pure water are pK a 1 = 4.2, and pK a 2 = 7.3. Nevertheless, added salts or polycations can force the dissociation of the carboxylic groups, decreasing the pK a values (i.e., pK a 1 = 3.4, pK a 2 = 5.6 in the presence of 0.1 M LiCl) [ 20 ]. The protonation constant for CS is approximately 6.5 [ 41 ]. At a pH of around 2, the carboxylic groups of the maleic copolymer are not dissociated, but the amine groups from CS are protonated, which determines the swelling of the beads ( Figure 4 B). In this form, the beads with a higher content of CS (CS/MA-VA1) have the highest swelling degree. When the pH is between 3 and 4, half of the carboxylic groups and the amine groups of CS are ionized and interact with each other, leading to the collapsing of the beads (polyelectrolyte complexation). When the pH increases above 4.2, more and more carboxylic groups begin to dissociate. A higher number of dissociated carboxylic groups compared with the ionized –NH 3 + groups of CS determines the swelling of the beads. Even if the amine groups of CS are in the –NH 2 form at pH over 6.4, the high swelling of the beads is assured by the carboxylic groups of the maleic copolymer that are all dissociated at basic pH. In this form, the CS/MA-VA3 beads with the highest content of carboxylic groups show the highest swelling capacity. 2.2. Cu 2+ Sorption 2.2.1. Influence of Initial pH The pH value of solutions affects the charge of the adsorbent and the speciation of heavy metal ions, thus being one of the factors that influence sorption. The adsorption of Cu 2+ ions by the CS/MA-VA3 beads was studied in the acidic pH range of 2–5.2 ( Figure 5 ). The initial pH of Cu 2+ solution could not be further increased due to the precipitation of the formed Cu(OH) 2 . As seen from Figure 5 , the sorption capacity increases with the increase of the initial pH from 2.3 to 4, and remains constant at a pH between 4 and 5.2. Similar results could be seen in the literature for sorbents based on CS [ 42 , 43 ] or CS and anionic polymers containing carboxylic groups [ 14 , 31 , 44 ]. The complexation between Cu 2+ and CS was shown to take place at pH > 4 [ 45 ] or even at pH > 5.3 [ 46 ] when CS has non-protonated amine groups that are involved in the chelation with metal cation through the free electron doublet of nitrogen [ 9 ]. In addition, the increase in pH leads to the ionization of the polymeric carboxylic groups that will electrostatically attract the Cu 2+ cations from the solution. 2.2.2. Sorption Kinetics Figure 6 A presents the effect of contact time on the Cu 2+ retention and shows that the sorption equilibrium is achieved after 6 h. To elucidate the sorption kinetics, the experimental data were fitted by pseudo-first-order (Equation (1)), pseudo-second-order kinetics (Equation (2)), and the intra-particle diffusion Weber and Morris model (Equation (3)). These models are based on the following equations [ 47 , 48 ]: (1) q t = q e 1 − e − k 1 t \n (2) q t = k 2 q e 2 t / ( 1 + k 2 q e t ) \n (3) q t = k d i f t 1 / 2 + A \nwhere q t (mg g −1 ) and q e (mg g −1 ) are the masses of Cu 2+ adsorbed per gram of beads at time t and at equilibrium, respectively. k 1 (min −1 ), k 2 (g mg −1 min −1 ), and k d i f (g mg −1 min −0.5 ) represent the rate constants of pseudo-first-order, pseudo-second-order and intra-particle diffusion models, respectively. The constant A (mg g −1 ) is related to the diffusion resistance. The pseudo-first-order (PFO) model is applied at a high initial concentration of sorbate ( C 0 ) and when the adsorption is controlled by external and internal diffusion. The pseudo-second-order (PSO) model is applied at low C 0 and when the sorption is controlled by adsorption on the abundant active sites [ 41 ]. The non-linear fitting of the experimental data with the PFO and PSO models is shown in Figure 6 A, and the obtained kinetic parameters and the correlation coefficients (R 2 ) are presented in Table 2 . The higher values of R 2 show that the PSO model described the Cu 2+ sorption onto CS/MA-VA beads better than the PFO model. This means that the rate-control mechanism is chemisorption, in agreement with other reports of metal ion adsorption on chitosan-based materials with or without carboxylic groups [ 13 , 16 , 42 , 43 , 49 , 50 ]. In the case of CS/MA-VA beads, the chemisorption may be accomplished by chelating Cu 2+ ions with the –COO − , –NH 2 , and –OH groups. The q e values calculated with the PSO model agree with the experimental values ( Table 2 ). From the k 2 and q e values, it can be concluded that the adsorption is faster for the CS/MA-VA1 sample, but the adsorption capacity is higher for the beads containing a high amount of carboxylic groups (CS/MA-VA3). The intra-particle diffusion model, proposed by Weber and Morris, was further applied to identify the diffusion mechanism during the sorption process. Figure 6 B shows the representation of the Cu 2+ adsorption amount versus the square root of time ( q t vs. t 1 / 2 ). It is known that if this dependence is linear and passes through the origin, the intra-particle diffusion is the controlling process; otherwise, the adsorption is controlled by multiple processes [ 47 , 51 ]. For Cu 2+ sorption onto CS/MA-VA beads, three different slopes are required to fit the experimental data, indicating that the process involves three stages. Generally, the last process (with the lowest slope) is attributed to the equilibrium phase, where the low concentration of sorbate in solution and the fewer adsorption sites determine the slowing down of intraparticle diffusion [ 51 , 52 , 53 ]. The first two stages can be attributed to (1) external surface adsorption, (2) intraparticle diffusion [ 42 , 51 , 52 , 53 , 54 ], or (1) macropore diffusion, and (2) micropore diffusion [ 55 , 56 ]. Taking into account the high porosity of CS/MA-VA beads, demonstrated by SEM, and the fact that the initial adsorption phase takes place in the first 40 min when a high swelling rate of the beads was observed ( Figure 4 A), the explanation involving macropores and micropore diffusion seems more plausible for the sorption of Cu 2+ ions from aqueous solution in the first two stages. The values of rate parameters for the three stages are given in Table 2 , with k dif ,1 > k dif ,2 > k dif ,3 . 2.2.3. Sorption Isotherms The sorption at equilibrium as a function of Cu 2+ concentration at equilibrium was presented in Figure 7 . The experimental data were fitted with three isotherm models: Langmuir (Equation (4)), Freundlich (Equation (5)), and Sip (Equation (6)) [ 4 ].\n (4) q e = q m a x K L C e 1 + K L C e \n (5) q e = K F C e 1 / n \n (6) q e = q m a x K S C e 1 / n 1 + K S C e 1 / n \nwhere q e (mg g −1 ) is the Cu 2+ sorption at equilibrium, C e (mg L −1 ) is the concentration of the solute at equilibrium in liquid phase, q m a x (mg g −1 ) is the maximum adsorption capacity, K L (L mg −1 ) is the Langmuir equilibrium constant, K F ((mg g −1 )(L mg −1 ) 1/n ) is the Freundlich constant, K S (L mg −1 ) 1/n ) is the Sips constant, and n (dimensionless) is a constant in Freundlich and Sips models regarded as a measure of the system heterogeneity. It is known that the Langmuir model (based on the assumption that monolayer adsorption occurs on the homogeneous surface with energetically equivalent adsorption sites) is characterized by a plateau at high sorbent concentrations, while the Freundlich model (assuming multilayer adsorption on the energetic heterogeneous adsorption surface) better fits the experimental data in the moderate concentration range, while the Sips model combines the Langmuir and Freundlich models [ 4 ]. The non-linear fitting of the experimental data with these models ( Figure 7 ) led to the parameters presented in Table 3 . The high values of R 2 show that Langmuir and Sips isotherms describe very well the Cu 2+ sorption onto CS/MA-VA beads. The K L and K S values are higher for CS/MA-VA3 beads compared with CS/MA-VA1 beads, showing that the adsorption of Cu 2+ is more favorable for polymeric materials possessing a higher amount of carboxylic groups. This can also be seen from the shape of the isotherms: for CS/MA-VA1 beads, the isotherm has an L-shape (Langmuir type) characteristic of favorable adsorption, and for CS/MA-VA3 beads, the shape of the isotherm is more like the H-shape (high affinity) characteristic of strongly favorable sorption [ 4 , 57 ]. The Langmuir isotherm model predicts a maximum adsorption capacity ( q max ) of 103.7 mg Cu 2+ g −1 and 142.4 mg Cu 2+ g −1 for CS/MA-VA1 and CS/MA-VA3, respectively. The maximum adsorption capacities for Cu 2+ found in the literature for different materials based on maleic acid copolymers, or CS, are presented in Table 4 . From the analysis of these data, we can conclude that the q max value obtained for CS/MA-VA3 beads is generally higher than those obtained for the materials based on maleic acid copolymers [ 25 , 26 , 27 , 58 , 59 , 60 ] or native CS [ 50 , 61 , 62 , 63 , 64 , 65 ], and comparable with the materials containing CS together with other chelating groups (phosphate, carboxyl, L-arginine, amidoxime, xanthate, etc.) [ 15 , 16 , 17 , 18 , 49 , 53 , 66 , 67 , 68 , 69 , 70 , 71 , 72 ]. 2.2.4. Adsorption Thermodynamics The values of the enthalpy change (Δ H ), entropy change (Δ S ), and Gibbs free energy (Δ G ) were calculated using the equations: (7) l n K d = ∆ S R − ∆ H R · T \n (8) ∆ G = − R · T · l n K d \nwhere K d is the distribution constant at equilibrium ( K d = q e / c e ) , T is the temperature in Kelvin, and R is the ideal gas constant (8.314 J K −1 mol −1 ). Experiments were carried out for CS/MA-VA beads at 298, 308, and 318 K, and the thermodynamic parameters obtained by representing l n K d as a function of 1/ T are given in Table 5 . The negative values of Δ G illustrate spontaneous and favorable sorption at these temperatures. The sorption process is endothermic (Δ H > 0), meaning that temperature increases are favorable for the adsorption of Cu 2+ onto CS/MA-VA beads. This finding agrees with other thermodynamic studies of copper sorption onto CS-based hydrogels [ 16 , 72 ]. 2.2.5. Characterization of the Beads after Sorption The first indication of Cu 2+ sorption onto CS/MA-VA beads is the appearance of the blue color, as shown in Figure 8 . The mechanism of Cu sorption was studied by FT-IR spectroscopy. As shown in Figure 8 , there are some differences in the spectra of CS/MA-VA3 beads before and after Cu 2+ loading. The broad adsorption band from 3438 cm −1 attributed to the O-H and N-H bonds moved to 3432 cm −1 and broadened after Cu 2+ sorption, suggesting that the hydroxyl and amine groups from CS are involved in the interaction with the metal ion [ 15 , 17 , 73 ]. The adsorption bands at 1592 cm −1 and 1734 cm −1 assigned to the carbonyl from the dissociated –COO − and acidic –COOH groups, respectively, are shifted to lower wavenumbers (1587 and 1725 cm −1 , respectively) in the spectrum of the beads loaded with Cu 2+ . This fact proved that the carboxylic groups from the maleic acid copolymer are also involved in the electrostatic interaction and coordination with the divalent metal [ 17 , 73 , 74 ]. The compression tests showed that the mechanical properties of the beads were modified after Cu 2+ loading. The forces required to break the unloaded wet spherical beads were 1.25 ± 0.08 N and 1.72 ± 0.18 N for the CS/MA-VA1 and CS/MA-VA3 beads, respectively. In contrast, the beads after Cu 2+ sorption can be compressed with 40 N without breaking but only undergoing plastic deformation, showing that the chelation acts as a further crosslinking agent and increases the mechanical properties of the beads. The morphology of the beads surface and cross-section before and after copper sorption was studied by SEM. The surface of the CS/MA-VA beads was modified after the adsorption of Cu 2+ ( Figure 9 A–D). Due to the mass transfer of copper ions onto the beads, the surface appeared denser and smoother, and a reduction of the pores was observed. The adsorption of Cu 2+ in the volume of CS/MA-VA beads was confirmed by energy-dispersive X-ray analysis (EDX) in cross-section ( Figure 9 E–J). The EDX elemental images of Cu presented in Figure 9 F,I show that the metal was absorbed in the porous structure of the beads and was uniformly distributed. The mass percent of copper was 8.4 ± 0.9% and 12.3 ± 1.6% in the loaded CS/MA-VA1 and CS/MA-VA3 beads, respectively. Sulfur from Cu(SO 4 ) was also present in low amounts in the beads after sorption. 2.2.6. Reusability The reusability of the materials used for the sorption of metal ions from aqueous solutions is a crucial factor from a practical and economical point of view. Therefore, in the present study, the desorption of Cu 2+ ions from the CS/MA-VA beads was performed with 0.1 M HCl, and the beads were regenerated with 0.1 M NaOH to re-dissociate the carboxylic groups and deprotonate the amine groups. After regeneration, the beads were used for another sorption cycle, and the equilibrium sorption capacity during consecutive sorption/desorption cycles was determined. As Figure 10 A shows, the adsorption of Cu 2+ ions at equilibrium remained almost unchanged after the fifth cycle of sorption, both for CS/MA-VA1 and CS/MA-VA3. The chemical crosslinking between the CS and MA-VA copolymers allows the beads to be exposed to different pHs while maintaining their integrity after five sorption/desorption cycles. After desorption, the beads lose the blue color given by copper ( Figure 10 B)."
} | 6,691 |
39027108 | PMC11256198 | pmc | 9,568 | {
"abstract": "Overgrazing and climate change are the main causes of grassland degradation, and grazing exclusion is one of the most common measures for restoring degraded grasslands worldwide. Soil fungi can respond rapidly to environmental stresses, but the response of different grassland types to grazing control has not been uniformly determined. Three grassland types (temperate desert, temperate steppe grassland, and mountain meadow) that were closed for grazing exclusion for 9 years were used to study the effects of grazing exclusion on soil nutrients as well as fungal community structure in the three grassland types. The results showed that (1) in the 0–5 cm soil layer, grazing exclusion significantly affected the soil water content of the three grassland types ( P < 0.05), and the pH, total phosphorous (TP), and nitrogen-to-phosphorous ratio (N/P) changed significantly in all three grassland types ( P < 0.05). Significant changes in soil nutrients in the 5–10 cm soil layer after grazing exclusion occurred in the mountain meadow grasslands ( P < 0.05), but not in the temperate desert and temperate steppe grasslands. (2) For the different grassland types, Archaeorhizomycetes was most abundant in the montane meadows, and Dothideomycetes was most abundant in the temperate desert grasslands and was significantly more abundant than in the remaining two grassland types ( P < 0.05). Grazing exclusion led to insignificant changes in the dominant soil fungal phyla and α diversity, but significant changes in the β diversity of soil fungi ( P < 0.05). (3) Grazing exclusion areas have higher mean clustering coefficients and modularity classes than grazing areas. In particular, the highest modularity class is found in temperate steppe grassland grazing exclusion areas. (4) We also found that pH is the main driving factor affecting soil fungal community structure, that plant coverage is a key environmental factor affecting soil community composition, and that grazing exclusion indirectly affects soil fungal communities by affecting soil nutrients. The above results suggest that grazing exclusion may regulate microbial ecological processes by changing the soil fungal β diversity in the three grassland types. Grazing exclusion is not conducive to the recovery of soil nutrients in areas with mountain grassland but improves the stability of soil fungi in temperate steppe grassland. Therefore, the type of degraded grassland should be considered when formulating suitable restoration programmes when grazing exclusion measures are implemented. The results of this study provide new insights into the response of soil fungal communities to grazing exclusion, providing a theoretical basis for the management of degraded grassland restoration.",
"conclusion": "5 Conclusion The results of this study showed that grazing exclusion did not cause significant changes in the soil fungal community α diversity ( P > 0.05) but significantly altered the soil fungal β diversity ( P < 0.05). In addition, temperate grassland soil fungi are more stability to grazing exclusion. Soil pH was found to be a key factor influencing the abundance, diversity, and network complexity of soil fungal communities in the three grassland types. We observed that grazing exclusion indirectly affected soil fungal communities by influencing soil nutrients and that plant diversity was significantly positively and negatively correlated with fungal diversity and network complexity, respectively. The results of this study provide a deeper understanding of the soil fungal community structure of different grassland types in response to grazing exclusion and a theoretical basis for the healthy restoration of degraded grasslands according to local conditions.",
"introduction": "1 Introduction Grazing is a major grassland utilization strategy that comes with certain economic effects and environmental consequences (Yin et al., 2021 ), such as grassland degradation due to interactions with changing climatic conditions, slowing vegetation growth (Dlamini et al., 2016 ), altering the soil structure, and significantly affecting ecosystem services such as grassland windbreaks and sand stabilization, water retention, and carbon sequestration functions (Zhao et al., 2020 ). Grassland degradation has become an important ecological problem worldwide and has received increasing attention from ecologists (Bardgett et al., 2021 ; Wang et al., 2022c ). Grazing exclusion is an effective way to restore degraded grasslands by relieving grazing pressure and promoting the self-recovery of degraded grasslands (Liu et al., 2020 ; Sun et al., 2021 ). Most of the previous studies on the effects of grazing bans on soil microorganisms have focussed on soil bacteria (Wang et al., 2022a ). However, fungi, which are directly dependent on plant communities, are more sensitive to changes in soil nutrients and have stronger aboveground and belowground interactions (Millard and Singh, 2010 ). Therefore, research on the effects of grazing exclusion on soil fungal communities is highly important. Soil fungi, as the second most important group in the soil microbial community and decomposers in the ecosystem, play an important role in promoting the uptake of various nutrients by vegetation, improving soil structure, participating in the degradation of apoplastic matter, promoting the turnover of nutrients during cycling and other ecological processes (Tedersoo et al., 2014 ; Peay et al., 2016 ). In addition, soil fungal communities are affected by different environmental factors, can adapt dynamically to the environment, and their composition can reflect the ecological status of the soil (Li et al., 2011 ). Therefore, studying the effects of grazing bans on soil fungal communities and understanding the drivers that influence soil fungal communities are critical for improving our understanding of ecosystem restoration mechanisms. However, the restoration of degraded grasslands by grazing exclusion is controversial. For example, Zhang et al. ( 2018 ) reported that grazing exclusion leads to a decrease in fungal diversity in a study on semiarid grasslands. Studies on meadow grasslands (Kaurin et al., 2018 ) and temperate steppe grasslands (Ma et al., 2023 ) reported that grazing exclusion favored an increase in fungal diversity, whereas a study on Seriphidium transiliense desert grasslands (Li et al., 2023 ) reported that the response of soil MBC, MBN, and MBP to grazing exclusion was not significant. These results suggest that no consensus exists on the effect of grazing exclusion on grassland soil fungal communities, which may be because most of the previous studies focussed on one grassland type and few studies considered different habitat conditions and their effect on soil microbial responses. Whether these differences are caused by different biotic and abiotic factors in the context of a wider variety of biotic communities remains to be investigated. The Tian Shan Mountain Range is one of the seven major mountain systems in the world, with complex topography and remarkable geomorphological features, leading to obvious differences in climate, soil conditions, and vegetation cover in mountainous areas, with multiple types of desert–alpine meadows, which play a more important role in grassland ecosystems (Li et al., 2022 ) and is an ideal area for the study of different grassland types. Numerous studies (Asitaiken et al., 2021 ; Zhou et al., 2022 ) have shown that grazing exclusion promotes the recovery of vegetation and soil nutrients in degraded grasslands in the Tian Shan Mountains, but there is a lack of research on the effects of grazing exclusion on soil fungal communities. To address the above questions, this study selected three different grassland types (temperate desert grassland, temperate steppe grassland, and mountain meadow grassland) in the Tian Shan Mountains as the research object, studied the effects of grazing exclusion on soil nutrients and soil fungal communities in different grassland types, and proposed two scientific questions: (1) does grazing exclusion have a differential effect on soil fungal communities in the three grassland types? and (2) what drives changes in soil fungal diversity, and how does grazing exclusion affect soil fungal communities through changes in plant communities and soil nutrients?",
"discussion": "4 Discussion 4.1 Changes in soil nutrients under grazing exclusion In recent years, grassland degradation has increased, and grassland ecosystems have been severely damaged as a result of long-term overgrazing. In this study, the values of SOC and TP in the grazing exclusion areas of temperate desert, temperate steppe, and mountain meadow grassland were much lower than the mean values of soil SOC (29.51 g/kg) and TN (2.3 g/kg) in China (Zhang et al., 2022 ), which indicated that the nutrient contents of temperate desert grassland soils were lower and that the soils were poorer. A comparison of the mean values of soil TP (0.52–0.78 g/kg) in China revealed that, in comparison to temperate desert grassland soils, temperate steppe grassland soils had greater TN and lower TP. Some findings have shown that grazing exclusion promotes soil nutrient accumulation (Cheng et al., 2016 ) by reducing livestock foraging and trampling, which allows vegetation communities to grow and reproduce. This, in turn, increases carbon inputs due to an increase in aboveground biomass and litterfall (Du and Gao, 2021 ). It has also been shown that soil carbon loss is accelerated through soil microbial respiration due to the inputs of livestock feces and urine from grazing sample plots (Pang et al., 2018 ). Whilst Yuan et al. ( 2020 ) reported that grazing exclusion did not significantly affect organic carbon, the present study revealed reduced soil organic carbon in three grassland types, particularly for mountain meadow grasslands ( P < 0.05), possibly because the improvement in soil moisture increased the input of soil carbon (Hu et al., 2016 ). But we found a significant reduction in the soil moisture content of the three grassland types, which led to a decrease in soil carbon input. When the depletion amount was greater than the accumulation amount, it resulted in an overall decrease in soil organic carbon (Li et al., 2023 ). Soil is an important carrier for vegetation growth, and the distribution and content of elements such as nitrogen and phosphorus can directly affect its nutrient status. In this study, the TN and TP contents of the soil in the grazing exclusion area of the mountain meadow grassland were significantly lower than those in the grazing area, consistent with the findings of Zhang et al. ( 2019 ) but not with those of Du and Gao ( 2021 ). This may vary depending on the geographical location of the forbidden grassland, the degree of degradation of the grassland, the number of years it has been forbidden to graze, and the climatic conditions. Changes in C, N, and P contents during nutrient cycling are considered important factors for ecological stability (Li et al., 2018 ), and it was observed in this study that soil C/N decreased after grazing exclusion and that low C/N ratios accelerated microbial decomposition of organic matter and increased the rate of nitrogen mineralisation (Springob and Kirchmann, 2003 ). This suggests that the treatment increased microbial diversity, which in turn increased the rate of decomposition. Soil bulk density is an indicator of aeration that is positively correlated with density and is mainly affected by soil structure, grazing and trampling, and soil organic matter content. After grazing exclusion treatment completely eliminated the direct trampling by livestock and allowed the vegetation to recover, reduced compactness and increased pore space led to reduced soil bulk density of temperate desert and temperate steppe grasslands (Wang S. et al., 2018 ). Unlike the results of most studies in which grazing exclusion decreased soil pH (Yao et al., 2018 ; Ma et al., 2023 ), our results revealed that the treatment significantly increased soil pH in montane meadow grasslands compared to grazing areas, which is consistent with the findings of Zhang et al. ( 2017 ) on the response of soil pH to grazing exclusion in desertified grasslands. The reason may be because livestock in grazing areas excrete feces and urine as they forage, and the increased volume of livestock urine leads to an increase in the rate of cycling of soil ions, which increases the concentration of hydrogen ions in the top layer of the soil, resulting in a higher soil pH in grazing exclusion areas than in grazing areas (Woodbridge et al., 2014 ). Taken together, for mountain meadow grasslands, which are richer and more diverse in plant species, grazing exclusion is detrimental to the restoration of their grassland soils, but for temperate desert and temperate steppe grasslands, grazing exclusion improves the physical structure of grassland soils. 4.2 Effects of grazing exclusion on fungal communities and co-occurrence patterns Soil fungi act as decomposers in ecosystems, effectively breaking down organic matter and humus and participating in the C and N cycles (Lv et al., 2023 ). In this study, Dothideomycetes, Archaeorhizomycetes, and Sordariomycetes, of the Ascomycota, and Agaricomycetes of the Agaricomycetes, were the dominant fungal groups of the three grassland types. Although the fungal response to grazing exclusion differed, the dominant groups were more or less the same. Ascomycota and Basidiomycetes have been shown to be dominant groups of soil fungi (Wang et al., 2022d ), and studies on alpine meadows on the Tibetan Plateau have shown that the dominant community of soil fungi in degraded grasslands is Basidiomycetes (Li et al., 2021 ), which is consistent with the results of this paper. Additionally, Ascomycota is also found to be the dominant community of soil fungi in the globally sampled range (Tedersoo et al., 2014 ). Although both Ascomycota and Basidiomycetes have important roles in decomposing organic matter, their division of labor is different; Ascomycota usually decomposes decaying and complex organic matter in the soil, whilst Basidiomycetes mainly decomposes lignocellulose, which is difficult to degrade in apoplastic plant matter (Yao et al., 2017 ). In this study, grazing exclusion had no significant effect on the abundance or α diversity of the dominant soil fungi, possibly because the 9 years of treatment were short and the response of soil fungi to successional age was weak (Brown and Jumpponen, 2015 ); therefore, the effect of short-term grazing exclusion on the abundance and the α diversity of soil fungi was not significant. For example, Wang et al. ( 2019 ) studied 14 and 19 years of grazing exclusion in semiarid grasslands and reported that soil fungal diversity increased with increasing years of grazing restriction. β diversity analysis was used to compare differences in species composition between groups. The closer the distance between two groups on a coordinate plot, the more similar the composition of these two groups. Wang et al. ( 2023 ) reported that prolonged grazing exclusion altered the composition of fungal communities. According to the PCoA results, the soil fungal β diversity of all three grassland types was significantly altered under the grazing exclusion treatment, whereas the changes amongst the grassland types were not significant, consistent with the results of Chen et al. ( 2020 ). This suggests that grazing exclusion altered the composition of the soil fungal community and that the changes in the composition were closely related to microbial activities. Grazing exclusion could regulate microbial ecological processes by changing the fungal community composition rather than its abundance or diversity. This may be because plant nutrient uptake, amongst other factors, is strongly linked to soil fungi and plant communities. Grazing exclusion, on the other hand, affects the aboveground biomass (Lan et al., 2023 ) and changes the composition and structure of plant communities (Sigcha et al., 2018 ), amongst other factors. This leads to a change in plant nutrient requirements (Du and Gao, 2021 ), a change that is an important factor leading to changes in soil fungal community composition. Microbial interactions can form a complex network that enables the effective transfer of energy, matter, and information between microorganisms that contribute to ecosystem function (Faust and Raes, 2012 ). Co-occurrence network analyses are used to assess how numerous species aggregate into different ecological clusters and to reveal the interactions between them (Berry and Widder, 2014 ). Studies have shown that positive connections indicate mutual synergistic relationships between microorganisms, whilst negative connections indicate competitive relationships between microorganisms (Wang X. et al., 2018 ; Blanchet et al., 2020 ). The connections in the soil fungal co-occurrence network in this study were all dominated by positive correlations, which is consistent with the findings of Duan et al. ( 2021 ). This result may indicate that co-operative relationships between soil fungal communities work together to resist external disturbances when the soil fungal community is subjected to environmental stress (Hernandez et al., 2021 ). For example, the ability of fungal hyphae to find nutrients is enhanced when there is a shortage of substrate (de Boer et al., 2005 ). Moreover, the soil fungal community in the grazing area had a greater average degree and average path length than that of the treated area, indicating that the community had greater connectivity. This study revealed that the average path length of the network was short, and the rate of information transfer between the species of the soil fungal network was fast (Zhou et al., 2011 ). These findings indicate that the fungal network response speed gradually accelerated when the environment changed from grazing treatment to grazing exclusion treatment, making the soil fungal community more susceptible to environmental changes. The temperate steppe grassland exhibited low connectivity and high modularity characteristics under the grazing exclusion treatment, indicating that the soil fungi in the temperate steppe grassland exhibited high stability under the grazing exclusion treatment. This is because the rich plant source resources and improved soil environment created more ecological niches for microorganisms after grazing exclusion (Chen et al., 2020 ; Lin et al., 2021 ). This was evident in the fact that all three grassland types with grazing exclusion had more modules than the grazing soils. A greater number of modules indicates a higher complexity of the soil fungal community, implying that the fungi had a stronger ability to resist external disturbances (Wang et al., 2022b ). A greater diversity of modules, in turn, leads to a greater diversity of interactions (Wang et al., 2022a ). Taken together, exclusion treatment increased the stability of temperate steppe grasslands, increased the rate of information transfer between the soil fungal networks of the three grassland types, and increased the diversity of interactions. 4.3 Factors influencing soil fungal community composition and co-occurrence networks In most studies, changes in soil microbial diversity as well as co-occurrence networks are usually associated with environmental variables (Zhang et al., 2018 ; Jiao et al., 2022 ; Geng et al., 2023 ). Soil microorganisms are very sensitive to the environment in which they live and differences in grassland utilization, type of grazing livestock, vegetation composition, geography, climate, and soils can lead to changes in soil microorganisms (Yin et al., 2019 ), and differences in soil nutrients affect soil microbial habitats to varying degrees, leading to changes in microbial communities (Kaspari et al., 2017 ). Some studies have shown that soil fungal communities are closely related to soil nutrients (Wang et al., 2015 ). The SEM revealed that although grazing bans did not have a direct effect on fungal communities and their diversity, they could indirectly and positively affect them by altering soil nutrients and plant diversity. Plants affected fungal communities through their aboveground apoplastic matter, nutrients from their underground root system, and the carbon they provided (Cline et al., 2018 ; Zhang et al., 2019 ). Increased plant diversity resulted in increased formation of apoplastic material as well as underground root secretions, which led to increased soil fungal diversity (Thakur et al., 2015 ). Previous findings that the effect of soil on microbial diversity is more significant than that of plants (Shu et al., 2024 ) are not consistent with our results showing that plant diversity had the greatest total effect on soil fungal communities. The findings that there is a lag in the effect of plant communities on soil fungi (Zhang et al., 2018 ) are also at odds with our results, suggesting that 9 years of grazing exclusion may be sufficient for soil fungi to respond to plant diversity. Unlike fungal diversity, plant communities were significantly negatively correlated with soil fungal networks, whilst soil nutrients were positively correlated with fungal networks, consistent with previous studies (Chen et al., 2020 ). Soil fungal networks are affected by pH (Liu et al., 2023 ), phosphorus content (Li et al., 2020 ), and nitrogen content (Chen et al., 2020 ), and soil TN significantly affects soil fungal networks (Deng et al., 2020 ). Moreover, when the nitrogen content is low, soil fungal networks meet their needs through enhanced competition (Yuan et al., 2021 ). According to the RDA, soil TP was found to be the environmental factor influencing dominant soil fungal flora, possibly because soil microbial genetic structure requires more phosphorus, so the amount of phosphorus limits the abundance of dominant flora of the soil fungal community as well as the complexity of the network (Mori et al., 2018 ). In addition, soil phosphorus increases the effective soil nitrogen by facilitating nitrogen mineralisation, which in turn affects dominant soil fungi (Wang et al., 2022b ). Pommier et al. ( 2018 ) found a strong correlation between TN content and the diversity of dominant taxa in fungal communities in a long-term nitrogen addition experiment on European grasslands, which is similar to the results of this paper's study that found soil TN to influence the diversity of fungal communities. In our study, soil bulk density was the main driver of soil fungal changes, and moist and permeable soils allowed for a richer environment for microorganisms and thus greater heterogeneity of living environments, a result similar to that of Jiao et al. ( 2019 ). This study also revealed that both the abundance of fungal dominant classes and fungal diversity were affected by soil pH, suggesting that it is a key limiting factor affecting the abundance, diversity, and network complexity of soil fungi under grazed exclusion conditions."
} | 5,798 |
39297874 | PMC11412253 | pmc | 9,569 | {
"abstract": "Abstract \nA corrigendum of this article has been published full details can be found at 10.1099/mic.0.001528\n Model microbial communities are regularly used to test ecological and evolutionary theory as they are easy to manipulate and have fast generation times, allowing for large-scale, high-throughput experiments. A key assumption for most model microbial communities is that they stably coexist, but this is rarely tested experimentally. Here we report the (dis)assembly of a five-species microbial community from a metacommunity of soil microbes that can be used for future experiments. Using reciprocal invasion-from-rare experiments we show that all species can coexist and we demonstrate that the community is stable for a long time (~600 generations). Crucially for future work, we show that each species can be identified by their plate morphologies, even after >1 year in co-culture. We characterise pairwise species interactions and produce high-quality reference genomes for each species. This stable five-species community can be used to test key questions in microbial ecology and evolution.",
"introduction": "Introduction Controlled experiments on synthetic microbial communities have been used to test general ecological and evolutionary theories about - for instance - species coexistence [ 1 ], adaptive radiations [ 2 4 ], and ecosystem functioning [ 5 6 ]. The short generation times and large population sizes of bacteria make them ideal for testing ideas that are experimentally intractable in other model systems. Culturable bacteria are also easy to manipulate, allowing us to change species richness [ 5 7 8 ], abundances [ 9 10 ], and even interaction type [ 11 12 ] and strength [ 13 ] in a systematic and increasingly high throughput way. The majority of previous experimental work has been done on monocultures or in synthetic communities consisting of only a few species. However, an increasing number of studies have used more diverse microbial communities to study how species rich communities assemble and community dynamics through time to understand the maintenance of the high levels of microbial diversity seen in nature [ 8,14 17 ]. Experiments on both simple and complex model communities are key to understanding the processes underpinning community assembly and coexistence, and there is an inherent trade-off between more detailed mechanistic understanding and complexity [ 18 ]. A key assumption of this previous work is that the community members stably coexist, meaning that the densities of species in the system do not show long-term trends [ 19 ]. Modern coexistence theory presents the ability to invade from rare (negative frequency dependence) which tests that each species is able to increase from low density when in the presence of the rest of the community, as a key test of coexistence [ 19 20 ]. Experimentally establishing that model communities are stable and that species coexist is key to increasing the likelihood that the findings are relevant to natural communities (where high levels of diversity are maintained, composition is relatively stable through time [ 21 23 ], and where patterns of diversity cannot be explained solely through neutral processes [ 24 26 ]). Second, when a model community is not stable, it risks biasing our understanding of species interactions to strongly competitive and negative interactions. In contrast, natural communities are thought to be generally stable owing to ecological and evolutionary processes that diminish competition. Third, demonstrating stable coexistence and understanding the mechanisms of coexistence (e.g. equalising or stabilising [ 19 ]) would increase the generalisability of findings and allow future work to understand under what circumstances stability breaks down. There are several proposed methods to measure invasion growth rate [ 27 ], but they generally involve testing whether each species (e.g. species A) has a positive growth rate after invading into a resident community (containing all other community members other than species A) at equilibrium [ 28 ]. However, this becomes more complicated to test as species diversity increases. For example, following the removal of a species, the resident community may destabilise or convert to an alternative stable state [ 29 ]. Explicit tests of coexistence in model microbial communities have been done when communities are simple and contain only strains of one [ 30 32 ], two [ 33 ], or three species [ 7 ], but they remain scarce when species diversity goes beyond this [ 5 8 13 15 17 ], but see [ 34 ]). Most model microbial communities are performed in batch culture, where populations are regularly transferred into fresh microcosms [ 2 ], meaning that ‘equilibrium’ is never reached in a measurable sense. Consequently, microbiologists have used invasion-from-rare assays to test for negative frequency dependence and stable coexistence. In these assays, the community is inoculated into fresh media, but one species (the invader) is at 100-fold lower density than the other species (the resident community). If each species has a higher growth rate when rare, relative to the resident community (a relative invader fitness greater than 1) then the community shows negative frequency dependence and is likely to be stable [ 31 35 36 ]. Here, we report the characterisation of a stable five species community that can be used for future experiments in ecology and evolution. We cultured a community pool of 46 isolates retrieved from the same soil sample over several weeks in lab medium (1/64 Tryptic Soy Broth (TSB)) and the final community composition was determined. As outcomes were highly repeatable, we took a single replicate community and conducted reciprocal invasion experiments to check coexistence. In addition, we measured species interactions using spent media and co-culture assays, tracked long-term community dynamics, and generated high quality reference genomes. Finally, we discuss potential questions that could be answered with this - or any - stable community when so much information about its constituent species is known.",
"discussion": "Discussion We characterized a model five species community natural community that can be used in experiments in microbial ecology and evolution. The community contains species with mostly competitive interactions, but Variovorax sp. demonstrates exploitative (potentially cross-feeding) interactions with three of the four other species. We found that estimates of interactions from supernatant assays and co-culture assays agreed qualitatively, but not quantitatively, with estimates of interactions in co-culture being stronger on average than those in supernatant. This suggests that contact dependent interactions (e.g. T6SS and biofilm formation) may also be important, but it also may just reflect methodological differences in the approaches. We found evidence of indirect interactions, but they became weaker (and more likely to be buffering) as expected pairwise interaction increased. This meant that in the full five species community, indirect interactions were relatively weak. This community has numerous advantages as a model system for ecological and evolutionary studies. Sequencing has confirmed that each species has a distinct colony morphology that means it can be tracked and re-isolated from the other members of the community, and that these colony morphologies remain distinct through a year of long-term culture. This makes this community ideal for experimental evolution. Furthermore, we used invasion-from-rare assays to experimentally demonstrate species coexistence, and sequencing showed that the community is stable through long-term (>1 year) batch culture. We encourage - where possible - more explicit tests of coexistence in microbial community experiments. Our model community is composed of diverse bacterial species all isolated from the same soil sample. Even so, this community is not designed to mimic a soil community. Instead the aim was to create a community that we could show empirically to be stable that could then be used to investigate general questions about ecology and evolution. By using the complex media Tryptic Soy Broth (TSB) at a (relatively) low concentration (1/64), we likely created conditions that promote coexistence through both stabilising (increased niche differences meaning intraspecific competition was greater than interspecific competition ( Fig. 3a )) and equalising (diluted media reduced the fitness differences between species) mechanisms. The long-term stability of the community and the unique ability to re-isolate each species to test for evolutionary responses in phenotype make the community perfect for doing long-term experimental evolution experiments, a multi-species version of Lenski’s pioneering LTEE. The genome assemblies provide a reference database to track genetic changes through evolutionary time. A recent editorial considered what features would be desirable for a model community to be widely adopted. This included having a functional output for the community, balancing complexity with feasibility, reproducibility, the ability to study the community over time, and ensuring the community in some way represents a natural community [ 58 ]. Our community achieves most of these, but we believe coexistence is a key desirable feature for a model community, and is essential to many of the other features such as including a temporal aspect and reproducibility. Moreover, demonstrating stability means model communities can more easily be used to test ecological and evolutionary theory where the assumption is often that the species coexist and the community is stable. In addition, the ability to manipulate a model community and re-isolate individuals is an invaluable feature for measuring evolutionary changes. The utility of this five species community is demonstrated in the current and previous work by us, and its potential to test future ideas. For example, we have demonstrated how species become (mal)adapted after coevolutionary time [ 47 ] and how invaders and disturbances interact synergistically in their effect on resident diversity [ 59 ]. While the five species community as described only contains bacteria, we have also discovered three phages that each can infect one of the five species [ 60 ] and combined the model community with a broad host range plasmid which can be carried by all five species [ 61 ]. Moreover, we have used this community to demonstrate that fitness effects of plasmids shape the structure of bacteria–plasmid interaction networks [ 61 ], and that the cost of plasmid carriage can depend on community context [ 62 ] . Given its long-term stability and the distinct morphotypes of each species, this community is ideal to understand how microbial communities evolve and how this alters stability in the face of environmental stressors. Furthermore, its stability means this community is an ideal one to use to measure invasibility, or as a way to validate and improve Genome Scale Models (GSMs) that use genomic information to predict metabolic functioning and species interactions [ 63 ]. Environmental conditions such as primary substrate, pH, or temperature could be changed to see how well GSMs do under different scenarios."
} | 2,824 |
30404349 | PMC6189914 | pmc | 9,570 | {
"abstract": "We report a new flow control method for centrifugal microfluidic systems; CO 2 is released from on-board stored baking powder upon contact with an ancillary liquid. The elevated pressure generated drives the sample into a dead-end pneumatic chamber sealed by a dissolvable film (DF). This liquid incursion wets and dissolves the DF, thus opening the valve. The activation pressure of the DF valve can be tuned by the geometry of the channel upstream of the DF membrane. Through pneumatic coupling with properly dimensioned disc architecture, we established serial cascading of valves, even at a constant spin rate. Similarly, we demonstrate sequential actuation of valves by dividing the disc into a number of distinct pneumatic chambers (separated by DF membranes). Opening these DFs, typically through arrival of a liquid to that location on a disc, permits pressurization of these chambers. This barrier-based scheme provides robust and strictly ordered valve actuation, which is demonstrated by the automation of a multi-step/multi-reagent DNA-based hybridization assay.",
"introduction": "1. Introduction Increasingly over the past decade, centrifugal microfluidic systems [ 1 , 2 , 3 ] have been applied to a variety of application fields such as biomedical diagnostics [ 4 , 5 , 6 ], bioprocess monitoring [ 7 ] and environmental screening [ 8 , 9 , 10 ]. This “Lab-on-a-Disc” (LoaD) platform is particularly useful for near patient/point-of-care/point-of-use applications, deriving its advantages from the ease with which a sample can be processed without a need for pneumatic interfaces or external pumps. The cartridges have dimensions similar to commonly available optical data storage media such as compact discs TM (CDs). The comparatively simple instrumentation (often just a low-cost spindle motor) and the inherent capability to centrifuge samples such as blood [ 11 , 12 ] are other major benefits of the LoaD platform. However, as all liquids on-disc are subjected to the same centrifugal field, advanced valving schemes are required to automate a sequences of laboratory unit operations (LUOs) such as mixing, metering and reagent release [ 13 ]. On-disc valving can be broadly categorised into three sub-types: externally actuated, rotationally controlled, and event-triggered. Externally actuated valves can broadly be defined as those controlled by modules on the peripheral instrument. Such interactions can include provision of external pressure [ 8 ], heat to induce phase-changes [ 14 , 15 , 16 , 17 ] and physical manipulation [ 18 , 19 ]. While these approaches expand the functionality of the centrifugal platform and also the number of LUOs that can be automated on a single cartridge, they tend to compromise the simplicity of the associated instrument. The most common valves are opened by a change of the spin rate and are termed rotationally actuated. Typically, they are based around the interplay between the centrifugally induced hydrostatic pressure and the capillary force acting on liquid in microchannels. Counter pressure induced by compression of entrapped gas might also be used as one of the competing forces. The high-pass version of rotationally actuated valves, which actuate towards elevated spin rates, include capillary burst valves [ 20 , 21 , 22 , 23 , 24 ], dissolvable-film (DF) valves [ 25 ], burstable foils [ 26 ], elastomeric membranes [ 27 ], and dead-end pneumatic chambers [ 28 ]. Triggered by a reduction of the spin rate, the low-pass valves include hydrophilic siphons [ 7 , 29 , 30 ] and pneumatically primed siphons [ 12 , 31 , 32 , 33 , 34 ]. The primary drawback of all rotationally-actuated valves is fidelity and reproducibility of the rotationally induced burst pressures. The burst pressures are intimately linked to the geometry, topography and surface chemistry governing the shape and contact line of the meniscus. Hence, in order to reliably separate subsequent assay steps such as the release of on-disc stored reagents, frequency bands have to be reserved to account for manufacturing tolerances and defects. For rotationally actuated valves, this minimum spacing of burst frequencies tends to severely restrict the number of LUOs that might be implemented in series. Many efforts have been made to mitigate this drawback; for example, low-pass and high-pass have been combined [ 7 , 29 ] in series to good effect. In another approach, the release of liquid from a group of valves, triggered simultaneously by a decrease in spin rate, had been staggered through the use of high-resistance microchannels [ 33 ]. In a new, recently introduced valving class [ 35 , 36 ], the arrival of liquid in a designated location triggers the subsequent valving steps. In this so-called event-triggered flow control, the layout of the disc-based channel network, rather than changes in disc spin rate, fully determines the order in which valves actuate. Based on (water) dissolvable-film (DF) technology [ 35 , 36 ], these event-triggered valves can function akin to an electrical relay and can enable Boolean-like AND- and OR-conditional triggering [ 35 , 36 ]. As event-triggered valves operate essentially independent of the spin rate, the number of assay steps that can be automated on a disc is not restricted by the finite spin rate envelope. However, these valves can result in a relatively large reagent and sample loss due to the dead volume of these valves. Additionally, the intervals between valving steps are prescribed by dissolution of the film, which may notably deviate with reaction or incubation times of the bioassays. In this paper, we present a new mechanism for serial actuation of valves, which is largely independent of the spin rate. The CO 2 gas issued by household baking powder [ 37 , 38 ] stored on-disc is released by an ancillary liquid to pressurise a sealed microfluidic compartment. Two alternative valve types are demonstrated. In a first approach, which we refer to as “volume governed”, the order of serial actuation of on-disc DF-valves is established through tailoring the volume of their respective pneumatic chambers ( Figure 1 ). In the second approach, called “barrier governed”, we demonstrate how the disc can be divided into discrete chambers that are separated by dissolvable films. By sequential dissolution of these DFs, each section of the disc is pressurised in a sequence encoded by the disc architecture, resulting in ordered valve actuation. We characterise both flow control schemes in terms of their timing. We then present an automated disc, where a sample containing DNA is hybridised with an oligo-spotted substrate and then washed by three buffers.",
"discussion": "5. Discussion We presented a new mechanism for valve actuation through gas release from baking powder that is initially dry-stored on the centrifugal microfluidic disc. The most important advantage of this technology is the option for valving at rather random, e.g., constant spin rate, thus making flow control independent from external instrumentation. Furthermore, for the barrier-governed valves, the order of actuation is strictly defined by the disc architecture. This offers a similar performance to event-triggered valves while allowing simpler disc architecture and reducing dead volume, thus saving precious real estate. While there is some variation in the timing of burst valve actuations, we believe these are a result of manufacturing defects where the valve geometries are not repeatably defined, or, in some cases, where micro-leaks might affect the rate of pressurisation of the discs. Use of an alternative manufacturing method will certainly improve repeatability of the system. This chemically actuated valving mechanism will increase the exposure of the reagents to CO 2 ; this may, in turn, affect performance of certain bioassays. However, in most cases, the relatively short period of point-of-care applications will attenuate this effect. A second challenge is the need to hermetically seal the disc, which might lead to slightly more complex loading procedures. However, this drawback might be relieved through ergonomic design; furthermore, the operation of the disc above atmospheric pressure might, like a positive displacement filter hood, even reduce the risk of assay contamination from the surrounding environment."
} | 2,075 |
35685171 | PMC9126649 | pmc | 9,571 | {
"abstract": "Marine biofouling which interferes with normal marine operation and also causes huge economic loss has become a worldwide problem. With the development of society, there is an urgent need to develop non-toxic and efficient anti-fouling strategies. Capsaicin is an environmentally friendly antifouling agent, but controlling the stable release of capsaicin from the coating is still a challenge to be solved. To achieve long-lasting antifouling property, in this work, we incorporate a derivative of capsaicin N -(4-hydroxy-3-methoxybenzyl)acrylamide (HMBA) to prepare double network (DN) hydrogels and make HMBA a part of the polymer network. Polyvinyl alcohol (PVA) has good hydrophilicity, and as a soft and ductile network, acrylamide (AM) and HMBA can polymerize to generate a rigid and brittle network. By adjusting the content of HMBA in the DN hydrogels, we can obtain a PVA–PAH X hydrogel with high mechanical strength, low swelling rate, and excellent antifouling effect, which provides a feasible way for the practical application of a hydrogel coating in long-term marine antifouling.",
"conclusion": "4. Conclusions In summary, we have prepared DN hydrogels coating containing capsaicin analogs by the one-pot method based on the green strategy. The double network structure can ensure the high strength mechanical properties of the hydrogels and can lock HMBA well in the materials. Simultaneously, the PVA–PAH X has the properties of low swelling ratio, high hydrophilicity, and superoleophobicity. More importantly, the introduction of HMBA into the hydrogels coating has shown excellent antifouling effect. It is worth noting that the HMBA immobilized on the DN hydrogels can achieve a long-lasting antifouling effect rather than the release model. This work will contribute to the practical application of hydrogel materials in the field of marine antifouling.",
"introduction": "1. Introduction Marine biofouling refers to the phenomenon in which some microorganisms, plants, and animals adhere to the surface of a material when it is submerged in the marine environment. 1–3 For ships, it can increase the friction on their bottom surface and their energy consumption. 4–6 It also accelerates material corrosion, increases maintenance costs, 7 and may increase the risk of biological invasion in regional seas. 6 A variety of strategies have evolved to address this world problem. 8–10 For example, adding biocides to coatings to poison and kill organisms, 11 modifying surface micro-scale structures, 12 compounding low surface energy coatings, 13,14 constructing hydrophilic antifouling coatings such as PEG, amphoteric polymers, hydrogels and polysaccharide-based coatings, etc. , 15–20 are the common strategies to confront biofouling. Tributyltin-based compounds (TBT) incorporated into self-polishing coatings showed efficient antifouling properties and were applied after the 1960s. 21 However, TBT was found to induce deformations in fish and other species in the ocean, and as a result, coatings containing organotin were removed from the market in 2003 and banned in 2008. 22 Some countries have banned the addition of heavy metal salts such as copper, zinc and lead to antifouling paints. 23 Observing the potential hazards of organic biocides and heavy metals to the marine environment, there is a growing focus on green antifouling strategies such as natural and environmentally friendly antifouling agents. 24,25 Capsaicin has been reported in the literature to have excellent antifouling properties and has been developed for applications due to the low-risk assessment of the environment. 25–27 However, the difficulty of controlling the release rate of capsaicin when added to antifouling coatings has become a major problem and challenge. 28–30 Therefore, some studies have reported that analogs with more utility than capsaicin itself have become their substitutes. 31,32 As a capsaicin analog, HMBA is a monomer, which is similar to capsaicin in chemical structure, but it can polymerized with other monomers to synthesize antifouling materials, mostly for some membrane materials modification, 33–36 and antifouling copolymer coatings. 37–39 However, capsaicin analogs are mostly constructed as hydrophobic coatings due to a hydrophobic monomer, and hydrophobic surfaces in the ocean may have a significant affinity for organic pollutants thus causing biofouling. 3,40 Hydrophilic materials tend to be more attractive than hydrophobic materials in antifouling strategies. 41,42 Hydrogel material is a three-dimensional hydrophilic polymers network containing a certain amount of water. 43 Hydrogel material contains many hydrophilic groups, which will cause the surface of the material to capture water molecules to form a dense hydrated film, thereby achieving antifouling effects. To achieve a green antifouling strategy, in our previous work, the DN hydrogel with underwater superoleophobicity was prepared by the one-pot method. 44 Afterwards, hydrogels containing 3-(trimethoxysilyl)propylmethacrylate (TMSPMA) were prepared, and TMSPMA could be hydrolyzed on the surface of the gel into proto-silicate analogs to deceive algae to achieve specific anti-algae effects. 45 Hydrophilic/hydrophobic heterogeneous anti-biofouling hydrogels with well-regulated rehydration were also prepared. 46 All these works provide good ideas for hydrogel in the field of marine antifouling. However, the mechanical properties of the hydrogel material itself are characterized by high water content and poor mechanical properties. 47 Gong prepared a DN hydrogel similar to cartilage in 2003, which inspired people to develop in the direction of tough performance. 48,49 Moreover, the construction of DN hydrogel has also broadened the multifunctional development of hydrogel materials, which can meet the needs of specific functions by changing the components of different networks. 50 Inspired by the DN hydrogel to enhance its toughness 51 and hydrophilic–hydrophobic microzone help for antifouling. 52 In this work, PVA was selected as the first network, HMBA and AM generate polymer chains as a second network through polymerization reaction, DN hydrogel was constructed by one-pot method ( Fig. 1 ). The hydroxyl groups on the PVA polymer chains form intermolecular hydrogen bonds with the phenolic hydroxyl groups, amino groups, and carbonyl groups on the PAH chains, which can help immobilize HMBA in the gel. The formation of hydrogen bonds between polymer chains can also increase the degree of cross-linking of the gel, which further changes the properties of the hydrogel. We finally obtained a hydrogel with low swelling, high toughness, and an excellent antifouling effect. It is proved that adding polymerizable capsaicin to the hydrogel coating can be better for practical needs, so it is an ideal strategy for marine antifouling applications. Fig. 1 A schematic illustration of the preparation and features of the PVA–PAH X gel.",
"discussion": "3. Results and discussion 3.1 Characterization of the chemical structures In the present work, we analyzed the HMBA by 1 H-NMR and the identification results are shown in Fig. 2a , 1 H-NMR (DMSO, ppm): 8.85 (s, 1H, O–H), 8.45 (s, 1H, N–H), 6.70 (d, 2H, –CH \n \n\n<svg xmlns=\"http://www.w3.org/2000/svg\" version=\"1.0\" width=\"13.200000pt\" height=\"16.000000pt\" viewBox=\"0 0 13.200000 16.000000\" preserveAspectRatio=\"xMidYMid meet\"><metadata>\nCreated by potrace 1.16, written by Peter Selinger 2001-2019\n</metadata><g transform=\"translate(1.000000,15.000000) scale(0.017500,-0.017500)\" fill=\"currentColor\" stroke=\"none\"><path d=\"M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z\"/></g></svg>\n\n CH–aromatic), 6.23 (d, 2H, CH 2 ), 5.58 (t, 1H, CH ), 4.23 (d, 2H, CH 2 –NH), 3.73(s, O–CH 3 ). Fig. 2 (a) The 1 H NMR spectrum of HMBA, (b) the FT-IR spectrum of HMBA, PVA, PAH and PVA–PAH0.6. \n Fig. 2b shows the IR spectrum of HMBA, PVA, PAH, and PVA–PAH0.6 hydrogels. It can be regarded as the strong absorption peak caused by phenolic hydroxyl (OH–) on the benzene ring in 3325 cm −1 , and the corresponding absorption bands at 1654 cm −1 can be assigned to stretching of carbonyl (C O) in the acid amide of HMBA monomer. It also shows that two absorption peaks at 1597 cm −1 and 1523 cm −1 , which are the vibrations of the (C C) skeleton on the aromatic ring. From the above, it can be concluded that the HMBA monomer has been successfully synthesized. For PVA gel, the strong band at 3857–2985 cm −1 corresponds to the strong (–OH) band typical of intermolecular and intramolecular hydrogen bonding. The FTIR spectrum of PAH shows peaks appears near 1654 cm −1 corresponding to the secondary amide stretching vibration, the stretching vibration peak of the secondary amide-association state appears near 3421 cm −1 , the benzene ring skeleton vibration peak appears near 1652 cm −1 , 1487 cm −1 , and 1456 cm −1 , and the carbon–oxygen vibration peak of the phenyl hydroxyl group appears near 1247 cm −1 . These all indicate the existence of HMBA structural units in the polymer molecules. The associative stretch vibration peaks of primary amides appear at 3200–3350 cm −1 , indicating the existence of AM structural units in the polymer molecules. Similarly, for PVA–PAH0.6, peaks at 1652 cm −1 , 1487 cm −1 , and 1456 cm −1 can be detected with the benzene ring skeleton vibration, thus it can be judged that the gel PAH and PVA–PAH0.6 have been polymerized and formed. 3.2 Swelling properties of hydrogels Generally speaking, the degree of swelling of the hydrogels as coatings has a critical impact in practical applications. The swelling behavior of hydrogels is influenced by multiple factors, such as the crosslink density of the material itself, hydration, ion concentration in the aqueous environment, pH, etc. Fig. 3a shows the swelling behavior of different gels in deionized water with time. It can be seen that the hydrogels showed a large weight change in the first 3 h after being placed in water, except for the PVA gels which showed a sharp increase in swelling behavior, and PVA showed a special behavior of −1.52%. Immediately afterward, the swelling rate of each component gels showed a slow increase, and the swelling rate of PVA continued to show steady negative growth. Within 24 h, all hydrogels reached the peak of their swelling rate variation, and then the gels variation showed a slow decrease until equilibrium was reached. Among all the hydrogels, the swelling rate of PAH is the largest for the reason that PAH has the largest pores in SEM ( Fig. 6 ), in addition, PAH is rich in hydroxyl and amino to have a better affinity with water molecules. PVA–PAM also shows a faster rate of water absorption, probably because of PVA and AM composite, where the hydrophilic group can quickly absorb water, but the formation of intermolecular hydrogen bonds instead of inhibiting their own swelling. It can be seen by PVA–PAH0.6 and PVA–PAM that HMBA slows down the water absorption rate of the gel in DN, and the water absorption rate of the gel is slower as the amount of HMBA added increases. These are attributed to the fact that HMBA is a hydrophobic monomer. Fig. 3 (a) The swelling rate of the hydrogel changes with time. (b) Comparison of various gels (from left to right are PVA, PAH, PVA–PAM, PVA–PAH0.6, PVA–PAH1.2, PVA–PAH1.8) before and after swelling balance. From Fig. 3b , it can be seen that only PVA is collapsed in water, and the reason is that PVA forms a good homogeneous network because the mixture of DMSO and water is a good solvent, and when placed in deionized water, the hydrogen bonds bound to DMSO in PVA are reduced, from which a more dense hydrogen bond is formed in the gel, and thus the swelling rate of PVA shows negative growth. The appearance of the PAH shows a large increase. Since PAM itself has a strong water absorption capacity, it can be seen from SEM ( Fig. 6 ) that the pores of the gel are also the largest. On the other hand, the equilibrium swelling rate of DN gels decreased gradually with the increase of HMBA, because HMBA is a hydrophobic monomer. Equally important, with reference to the SEM images ( Fig. 6 ) of hydrogels, the swelling behavior of hydrogels is also related to the cross-link density. As the cross-link density increases, the cross-link points in the cross-link network increase, the chain segments between the cross-link points become shorter, and the micropores in the network structure become smaller, so the swelling ratio decreases. 3.3 Mechanical properties of hydrogels When hydrogels coatings are needed to meet practical needs in marine antifouling work, good mechanical properties are usually required to combat the complex marine environment, where the measured parameters are generally compressive strength and energy storage modulus. A universal testing machine was used to test the hydrogels after swelling equilibrium in water to examine the SN hydrogels and DN hydrogels performance tests, and also to examine the effect of HMBA addition on the mechanical properties of DN hydrogels. The measures we took were to subject the prepared SN gels PVA, PAH, and DN gels PVA–PAH X to uniaxial compression tests with increasing HMBA concentration and to test the hydrogels after swelling equilibrium under the condition of 80% compressive strain. From Fig. 4a , it can be seen that the compressive stress of DN hydrogel is higher than those of the SN hydrogels, and the recorded data show that the stress of the SN hydrogel PAH is the smallest in the compression process of the hydrogels, and the stress of the PVA are higher than those of PAH only. The parameters of DN hydrogels are significantly higher than those of SN hydrogels. With the increase of HMBA addition, the stress of DN hydrogels with high HMBA content is higher than that of low HMBA content. In the comparison of DN hydrogels, it can be seen that with the increase of HMBA content, the compressive stress ( σ ) of DN hydrogel at 80% strain gradually increases from 2.05 MPa to 3.21 MPa, and the mechanical properties are significantly higher than that of SN hydrogels PAH and PVA. The compressive stress ( σ ) of DN hydrogels gradually increased from 2.05 MPa to 3.21 MPa. In contrast, the storage modulus of the DN gels was much higher than any SN gels ( Fig. 4b ). Besides, the order size of the hydrogels G ′ is PVA–PAH1.8 > PVA–PAH1.2 > PVA–PAH0.6 > PVA–PAM > PVA > PAH, which indicated that DN hydrogels are highly elastic. The introduction of DN in the hydrogel can well improve the strength and toughness of the hydrogel, in which the proportion of HMBA in the PAH gel can be adjusted to achieve this purpose. Fig. 4 The effects of SN and DN with the addition of HMBA after equilibrium swelling on (a) uniaxial compression performance and (b) storage modulus in rheology are studied. Generally, the hydrogels are affected by the cross-linking of molecular chains during its construction. In the construction of PVA–PAH X hydrogels, the PVA network is physically cross-linked by hydrogen bonding. While the content of MBAA in the PAH network is constant, the change of HMBA content plays a decisive role in improving the mechanical strength of the gels, and with the increase of HMBA content, the mesh of PVA–PAH X becomes denser and the mechanical strength of the hydrogel gradually increases. \n Fig. 5a clearly shows that the SN gel PAH can be easily broken under extrusion conditions, PVA gel is a soft network although it will not break it is difficult to return to the original shape after a single extrusion, while the DN gel PVA–PAH0.6 still maintains its initial shape after extrusion, has good elasticity and toughness, and has excellent mechanical properties compared with any SN gels. For convenience, Fig. 5b shows the cutting of PVA–PAH0.6 alone was demonstrated and the gel remained intact under the forceful cutting of the steel ruler. Fig. 5 (a) The hydrogel compression process, compressed at a rate of 10 mm min −1 to a strain of 80%. (b) Cut to the lowest point in the middle of the PVA–PAH0.6 by means of a steel ruler. 3.4 Surface microstructure properties of hydrogels A hydrogel is a three-dimensional network-like structure, inside which is usually filled with a large amount of water. Hydrogels material are usually porous structures, which satisfy in the ocean will exchange a large amount of water flux with the external environment, where the size of the pores in the network affects the rate of water exchange between the hydrogels and the external environment and its properties. The surface structures of hydrogels observed by swept surface electron microscopy are shown in the figure below. It can be seen that the PVA gel shows a small pore arrangement with a pore diameter of 0.2–3 μm, from which it can be seen that although the pores of the PVA are more numerous, the distribution of the pore walls between the voids is thicker, probably because the solvent in the PVA is a mixture of DMSO and water in the process of freezing does not appear enough ice crystals inside the gel. It is easy to see that the pores of PAH are the largest, and their pore diameters are distributed in the range of 10–40 μm, while the pores of PVA–PAM are between those of PVA and PAH0.6, and the voids are about 1–8 μm, and the pores are more densely arranged with thinner pore walls. When HMBA was added, we can see that the pores of PAH0.6 did not change much compared with PVA–PAM, but the pore walls became thicker, which was due to the increase of HMBA. With the increase of HMBA content, the pores of PVA–PAH1.2 and PVA–PAH1.8 are obviously smaller, and the pore diameter is distributed in 0.3–4 μm, while the pore distribution of PVA–PAH1.8 with the most HMBA content is smaller and more uniform than PVA–PAH1.2, and the pore wall is also thicker. The pore size of PVA, PAH, and PVA–PAH X is positively correlated with the water content of the gels. The degree of cross-linking of hydrogels can also be seen in the figure. Comparing the DN gels, it can be seen that with a constant MBAA content, the increase of HMBA content can increase the intermolecular hydrogen bonding arrangement of polymer chains thus increasing the cross-linking density of the gels and increasing the cross-linking point, which makes the pores of DN gels smaller and makes the pore walls thicker, thus better explaining the swelling behavior and mechanical properties of the hydrogels. 3.5 The wettability of hydrogel The wettability of the hydrogel surface is also an important part of the antifouling performance test, and hydrogels have an important development prospect as a hydrophilic material to prevent biological adhesion in the ocean. For this reason, we tested the static contact angle of water droplets in the air and the static contact angle of oil droplets in water for each component gels, where 2 μL of deionized water was used for water droplets and 2 μL of 1,2-dichloroethane for oil droplets. As shown in Fig. 7a , the CA of PVA and PVA–PAM were 11.2° ± 1.5° and 11.4 ± 3.8°, respectively, with the smallest CA indicating good hydrophilicity. The CA of PAH was 23.2 ± 3.7° which also indicated that the PAH gel with capsaicin addition also showed hydrophilicity. With the increase of HMBA content in the gels, the PVA–PAH X also showed an increasing trend of CA size although they were hydrophilic. Overall, HMBA, a hydrophobic monomer, weakens the hydrophilic properties of the gels, but the gels still exhibit hydrophilic properties. Fig. 6 SEM images of the pore size comparison of SN gels and DN gels with different HMBA contents. Fig. 7 Static contact angle of hydrogel (a) CA of water in the air (b) CA of underwater oil. From Fig. 7b , it is easy to see that the CA of PVA and PVA–PAM are 147.5° ± 5.6° and 143.5° ± 7.7°, respectively, showing strong oleophobic, while the CA of PAH is 135.4° ± 13.1° with some hydrophobic molecules (HMBA). We can attribute this to the fact that the gel has many hydrophilic groups which absorbs a large amount of water, giving the body a good oleophobic property. It can also be seen that with the increase of HMBA addition in the PVA–PAH X , the CA also decreases and shows slight oleophilicity. We can explain that the HMBA monomer is uniformly distributed on the surface of the gels after reaching the swelling equilibrium in the gels system, and its components make the surface of the gels have slightly oleophilic. Surprisingly, the PVA network has excellent hydrophilicity, while the PAH network is obtained by copolymerization of hydrophilic AM monomer and hydrophobic HMBA monomer, which has hydrophilic. In the DN gels network, PVA–PAM without HMBA has hydrophilic and superoleophilic properties, and with the increase of HMBA content also has an oleophilic ability. It indicates that HMBA is distributed on the surface of PVA–PAH X , which can explain to some extent that capsaicin on the surface of DN gels is convenient to contact with marine adherent organisms to play an antifouling effect. 3.6 The antifouling properties of hydrogel To further explore the antifouling performance of hydrogels, we tested the adhesion ability of various gels to Nitzschia closterium and Navicula sp. respectively under fully simulated marine environment (reference experimental method). Biofouling counts are shown in Fig. 8a and b , and in general to increase the control we selected glass slice samples to increase the indication of the good activity of algae. Firstly, it can be observed that after three days of incubation period all substrates had biologically active algae attached on top. Fig. 8 Anti-diatom adhesion performance (a) and (b) fluorescence micrographs pictures of the densities of adhered Navicula sp. and Navicula (the ruler in the picture is 50 μm). (c) and (d) Statistical chart showing diatom adhesion ( p < 0.01). As shown in Fig. 8c , the glass as control is about 1004 diatoms per mm 2 , and the hydrogel PVA and PVA–PAM without HMBA are about 1360 diatoms per mm 2 and 1478 diatoms per mm 2 , respectively, with a large number of diatoms adhering to its surface, indicating poor antifouling properties. The number of diatoms was suppressed to some extent for the fraction with HMBA added, for example, 680 diatoms per mm 2 for PAH0.6, while for DN gel with different loading of HMBA, the number of diatoms showed a decrease with the increase of HMBA content. From Fig. 8b and d , the micrographs and diatom counts were observed. The highest amount of adherence was PVA–PAM and reached 368 diatoms per mm 2 which indicates that the DN gel without HMBA added adhered more diatoms, while AH had some inhibitory effect compared to the other components without HMBA added. Comparing the number of diatoms attached to PAH and PVA–PAH X , the HMBA content in PVA–PAH X is not as high as that of PAH, but it is better than the former in terms of anti-fouling and anti-algae effects. And the adhesion of the PVA network combined with the PAH network was reduced, which indicates that the composition of the double network helps HMBA to play a better antifouling ability. And with the increase of HMBA content, the antifouling effect was also significantly improved."
} | 5,819 |
34045776 | null | s2 | 9,572 | {
"abstract": "Interfacing devices with cells and tissues requires new nanoscale tools that are both flexible and electrically active. We demonstrate the use of PEDOT:PSS conducting polymer nanowires for the local control of protein concentration in water and biological media. We use fluorescence microscopy to compare the localization of serum albumin in response to electric fields generated by narrow (760 nm) and wide (1.5 μm) nanowires. We show that proteins in deionized water can be manipulated over a surprisingly large micron length scale and that this distance is a function of nanowire diameter. In addition, white noise can be introduced during the electrochemical synthesis of the nanowire to induce branches into the nanowire allowing a single device to control multiple nanowires. An analysis of growth speed and current density suggests that branching is due to the Mullins-Sekerka instability, ultimately controlled by the roughness of the nanowire surface. These small, flexible, conductive, and biologically compatible PEDOT:PSS nanowires provide a new tool for the electrical control of biological systems."
} | 278 |
27534803 | PMC4992128 | pmc | 9,573 | {
"abstract": "Droughts are increasing in severity and frequency, yet the mechanisms that strengthen ecosystem resilience to this stress remain poorly understood. Here, we test whether positive interactions in the form of a mutualism between mussels and dominant cordgrass in salt marshes enhance ecosystem resistance to and recovery from drought. Surveys spanning 250 km of southeastern US coastline reveal spatially dispersed mussel mounds increased cordgrass survival during severe drought by 5- to 25-times. Surveys and mussel addition experiments indicate this positive effect of mussels on cordgrass was due to mounds enhancing water storage and reducing soil salinity stress. Observations and models then demonstrate that surviving cordgrass patches associated with mussels function as nuclei for vegetative re-growth and, despite covering only 0.1–12% of die-offs, markedly shorten marsh recovery periods. These results indicate that mutualisms, in supporting stress-resistant patches, can play a disproportionately large, keystone role in enhancing ecosystem resilience to climatic extremes.",
"discussion": "Discussion As the above results show, stress-resistant patches can have a disproportionately large effect on ecosystem recovery by virtue of the fundamental spatial processes they support. This finding reveals that: (1) a surprising level of ecosystem resilience can result even when only a limited area of dispersed patches of habitat-forming species remain after episodes of severe stress (that is, a system with low overall resistance), and (2) the time it takes for such ecosystems to recover and, hence, their resilience 8 may not be predicted simply by the size of the disturbance. Instead, to develop accurate predictions for the many systems where dominant, habitat-forming species exhibit vegetative growth and/or spatially limited propagule dispersal (for example, reef-building corals, oysters and mussels; meadow-forming prairie, marsh and sea-grasses; bed-forming marine and riverine macroalgae), ecologists, ecosystem managers and conservation scientists should detail the distribution of remnant patches acting as habitat growth nuclei and identify the mechanisms underpinning such patterns in survival. In salt marshes, our study reveals that these patterns in survival can be largely dictated by a keystone mutualism between cordgrass and mussels, classified as keystone because mussel impacts on ecosystem resistance and recovery are disproportionate to their low cover in marsh landscapes 27 28 . Importantly, recent work showing that mussels do not protect cordgrass against runaway consumption by herbivorous crabs in New England salt marshes 29 serves as a warning that while keystone mutualists may enhance resilience to certain stressors, they may not do so for others, or may have little effect in cases where stress levels are so high that the mutualist's buffering capacity is exceeded. From our central findings, we draw two main conclusions. First, positive interactions can indeed enhance ecosystem resistance to and recovery from large-scale disturbance as has been proposed, but not yet demonstrated in a real, non-theoretical system 13 30 . As mutualism-dependent ecosystems occur in all corners of the earth (for example, mangroves, seagrass meadows, coral reefs, peat bogs, boreal forests), we suspect that similar processes may enhance the stress resistance and accelerate the recovery of critical habitat-forming species globally. Second, we anticipate that managers may achieve impressive gains in ecosystem resilience through relatively little investment where they integrate keystone mutualisms and optimal patch distributions into conservation and restoration strategies 26 . To our knowledge, these analyses are the first to show the potential for spatially dispersed, stress-resistant patches to promote ecosystem resilience to a major disturbance, a finding that may help explain differences in resilience observed across ecosystems and enhance our ability to maintain ecosystems in the face of climate change."
} | 1,010 |
29404706 | PMC5799091 | pmc | 9,575 | {
"abstract": "Background Bioethanol obtained by fermenting cellulosic fraction of biomass holds promise for blending in petroleum. Cellulose hydrolysis yields glucose while hemicellulose hydrolysis predominantly yields xylose. Economic feasibility of bioethanol depends on complete utilization of biomass carbohydrates and an efficient co-fermenting organism is a prerequisite. While hexose fermentation capability of Saccharomyces cerevisiae is a boon, however, its inability to ferment pentose is a setback. Results Two xylose fermenting Kodamaea ohmeri strains were isolated from Lagenaria siceraria flowers through enrichment on xylose. They showed 61% glucose fermentation efficiency in fortified medium. Medium engineering with 0.1% yeast extract and peptone, stimulated co-fermentation potential of both strains yielding maximum ethanol 0.25 g g −1 on mixed sugars with ~ 50% fermentation efficiency. Strains were tolerant to inhibitors like 5-hydroxymethyl furfural, furfural and acetic acid. Both K. ohmeri strains grew well on biologically pretreated rice straw hydrolysates and produced ethanol. Conclusions This is the first report of native Kodamaea sp. exhibiting notable mixed substrate utilization and ethanol fermentation. K. ohmeri strains showed relevant traits like utilizing and co-fermenting mixed sugars, exhibiting excellent growth, inhibitor tolerance, and ethanol production on rice straw hydrolysates. Electronic supplementary material The online version of this article (10.1186/s13065-018-0375-8) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions Screening for microbes capable of co-fermentation is necessary for efficient conversion of lignocellulosic biomass into ethanol with enhanced productivity. There is a significant advancement in developing a robust microbial strain with co-fermentation potential as well as tolerance to inhibitors. K. ohmeri strains, studied here showed promising mixed sugar fermentation potential with enhanced xylose utilization. Strains were also tolerant to HMF, furfural, formic acid and could grow well in presence of acetic acid on prolonged incubation. The study emphasizes that this genus could provide robust native yeast strains with co-fermentation properties which can be evolved further. Lignocellulosic hydrolysates often generate unexpected results due to the presence of inhibitors, as they vary widely in nature [ 12 ]. These strains displayed efficient growth and ethanol production from biologically pretreated rice straw hydrolysates.",
"discussion": "Results and discussion Growth and characterization Lagenaria siceraria flowers are rich in pentose and hexose sugars and thus used as a source for isolating pentose assimilating K. ohmeri strains [ 31 ]. K. ohmeri strains were isolated and purified from L. siceraria flowers by enrichment on MXYP medium and maintained as glycerol stocks. Both the strains grew well on minimal medium with xylose as sole carbon source (Additional file 1 : Figure S1). They showed distinct opaque, butyroid, creamy, circular colony morphology with regular margins and raised elevation. Under phase contrast microscope, cells appeared ovoid and occurred singly (Additional file 1 : Figure S2). Scanning electron microscopy images showed shrunk cells with irregular margins indicating stress. Budding cells were also observed under scanning microscopy (Fig. 1 ). Fig. 1 Scanning electron micrographs of strain 5 and strain 6. Cells of strain 5 ( a ) and strain 6 ( b ) appear stressed due to growth on xylose under micro-aerophilic conditions. Budding cells are clearly visible in the electron micrographs Biochemical characterization showed that both strains could assimilate maltose, sucrose, galactose, cellobiose, raffinose, trehalose, glucose and xylose while inositol, dulcitol, lactose, melibiose were not assimilated and urease test was also negative (Additional file 1 : Table S1). Both strains were identified to be K. ohmeri upon partial sequencing. Strain 5 (GenBank Accession No. KT598022) showed 100% similarity with K. ohmeri while strain 6 (GenBank Accession No. KT598023) displayed 97% similarity. Phylogenetic tree constructed using Maximum-Likelihood [ 32 ] also showed their relationship with K. ohmeri (Fig. 2 ). Kodamaea genus was earlier placed under Pichia genus but was separated later due to considerable genetic distances as measured by partial sequences of 18S and 26S ribosomal RNA and only seven species were placed under the genus Kodamaea including K. anthophila , K. kakaduensis , K. ohmeri , K. laetipori , K. nitidulidarum, K. transpacifica, K. meredithae have been described [ 33 – 35 ]. Fig. 2 Phylogenetic tree of K. ohmeri strain 5 and strain 6 \n Attributes pertaining xylose metabolism Xylose reductase and xylitol dehydrogenase enzyme activities pertaining to xylose metabolism [ 36 , 37 ] were exhibited by both the strains but levels were low. The activities suggested the presence of xylose metabolizing pathway in these strains but levels were too low and their ratio predicted the flow of the pathway towards ethanol production. Specific activities (U mg −1 protein) of the strains were found to be 0.024, 0.2 (XR) for strain 5 and 6 respectively, while 0.011 and 0.015 (XDH) for strain 5 and strain 6 respectively. Fermentation and co-fermentation capabilities and effect of supplementation As evident from absorbance at 660 nm, both the strains grew well on minimal medium with xylose as sole carbon source and also on mixture of xylose and glucose and fermented them to ethanol (data not shown). On minimal medium containing salts, ethanol was produced by both strains with fermentation efficiency of ~ 25 and ~ 5% on glucose and xylose respectively (Fig. 3 ; Additional file 1 : Table S2). Ethanol was the major product of glucose and xylose fermentation though trace amounts of xylitol and acetic acid were also detected during mixed sugar fermentation. Higher ethanol yield of 0.31 g g −1 from glucose with fermentation efficiency of 61% was obtained when minimal medium w as supplemented with 1% yeast extract (YE) and 1% peptone (Table 3 ) without salts. In a study, d -arabitol production was observed as main product from glucose as the carbon source on rich medium (with 1% YE and 1% peptone) by K. ohmeri, and only trace amounts ethanol were observed. Production levels of polyols as fermentation products, largely depend on factors like proper ratio of nitrogen, carbon sources in the medium, original habitat of the fermenting organism and growth conditions [ 23 ]. Presence of salts in growth medium influence physiology, hamper growth and distress fermentation efficiency in yeasts while medium with organic supplements augment ethanol fermentation efficiency. Fig. 3 Xylose ( a ) and glucose ( b ) fermentation efficiency on minimal media with salts. Salts hamper the fermentation process as is visible from the lower fermentation efficiencies \n Table 3 Glucose utilization and ethanol yield of strain 5 and strain 6 Treatment Glucose (g L −1 ) Ethanol yield (g g −1 ) Strain 5 Time (h) 96 108 120 96 108 120 0.1% yeast extract + 0.1% peptone 90.40 ± 16.6 97.95 ± 1.8 88.90 ± 17.2 0.16 ± 0.04 0.28 ± 0.05 0.20 ± 0.09 0.5% yeast extract 99.97 ± 0.06 100 100 0.16 ± 0.09 0.25 ± 0.06 0.28 ± 0.12 1% yeast extract + 1% peptone 97.74 ± 3.71 99.91 ± 0.16 100 0.13 ± 0.02 0.12 ± 0.02 0.12 ± 0.02 Strain 6 0.1% yeast extract + 0.1% peptone 100 99.9 ± 0.17 100 0.12 0.14 ± 0.01 0.12 ± 0.02 0.5% yeast extract 99.18 ± 1.42 100 100 0.22 ± 0.05 0.24 ± 0.12 0.13 1% yeast extract + 1% peptone 100 99.83 ± 0.21 100 ± 0.01 0.24 ± 0.14 0.31 ± 0.10 0.20 ± 0.10 SE m (±) 1.61 0.23 1.71 0.02 0.02 0.02 CD @5% 4.46 0.64 4.73 0.05 0.06 0.05 Ethanol yield (g g −1 ) = {concentration of ethanol produced (g L −1 )/concentration of sugar consumed (g L −1 )} SE m standard error of mean, CD critical difference \n Supplementation with 0.1% YE and 0.1% peptone enhanced fermentation efficiency (to ~ 50%) (Table 4 ). Further enhancing level of supplementation in medium with higher concentrations of YE/peptone did not increase fermentation efficiency significantly. Studies have suggested significant role of cultivation media components to provide favorable conditions for growth and product formation [ 10 ]. Xylose consumption was also enhanced to ~ 40% during co-fermentation and highest ethanol yield was 0.25 g g −1 sugar consumed when Kodamaea were grown on 10% total mixed sugars (5% glucose + 5% xylose). Table 4 Effect of supplementation on sugar utilization and ethanol yield of K. ohmeri strains Treatment Xylose consumed (g L −1 ) Glucose consumed (g L −1 ) Ethanol yield (g g −1 ) Fermentation efficiency (%) Strain 5 Time (h) 96 108 120 96 108 120 96 108 120 96 108 120 0.1% (YE + P) 20.47 ± 1.9 20.4 ± 3.2 16.2 ± 3.05 37.1 37.9 37.9 0.22 0.21 ± 0.017 0.22 ± 0.033 44 ± 3.3 40.8 ± 6.5 43.6 ± 5.9 0.5% (YE) 20.3 ± 2.2 18.8 ± 0.57 14.3 ± 1.9 50 50 50 0.16 ± 0.02 0.17 ± 0.01 0.18 ± 0.01 32 ± 4.5 32.5 ± 1.5 34.9 ± 2.4 1% (YE + P) 17.7 ± 2.8 15 ± 7.4 13.3 ± 0.02 50 50 50 0.21 ± 0.03 0.19 ± 0.07 0.2 ± 0.001 41.3 ± 5.4 37.1 ± 14.6 39.7 ± 0.18 Strain 6 0.1% (YE + P) 11.7 ± 1.7 13 ± 4.3 14 ± 0.45 49.4 49.3 ± 0.19 50 0.25 ± 0.02 0.19 ± 0.05 0.2 ± 0.001 48.6 ± 3.13 38 ± 8.8 39.7 ± 0.25 0.5% (YE) 15.1 ± 9.13 18 ± 6.15 14.2 ± 3.8 50 50 50 0.2 ± 0.01 0.19 ± 0.06 0.2 ± 0.03 38.7 ± 18.1 38 ± 12.1 39.6 ± 6.2 1% (YE + P) 12.2 ± 0.67 66.6 ± 2.4 14.9 ± 0.12 50 50 50 0.25 ± .007 0.18 ± 0.03 0.18 ± 0.001 49.5 ± 1.5 36 ± 5.3 35.1 ± 0.28 SE m (±) 1.24 1.06 0.50 0.00 0.02 0.00 0.01 0.01 0.01 2.46 2.08 1.03 CD @5% 3.41 2.92 1.38 0.00 0.06 0.00 0.03 0.03 0.01 6.80 5.74 2.85 YE + P yeast extract + peptone \n Amongst most of the pentose utilizing yeasts only a few have been reported to produce ethanol as major product from pentose fermentation [ 38 ]. A mixed sugar fermenting yeast, Candida lignohabitans possessing remarkable capability to ferment both pentoses and hexoses, exhibited highest ethanol yield of 0.2 g g −1 on rich medium containing 1% yeast extract and 2% soya peptone with 2–5% carbon sources, while no ethanol was detected on minimal medium without supplementation. This might be due to the lower biomass accumulation on minimal medium [ 7 ]. In this study, K. ohmeri strain 6 exhibited high ethanol yield during mixed substrate fermentation with minimal supplementation. Insignificant increase in fermentation efficiency upon medium with higher supplementation suggested to avoid excessive nutrient supplementation as it favors biomass production [ 23 ]. Table 5 shows ethanol yields of related yeast strains. Zheng et al. [ 3 ] observed stimulating effect of supplementation on acetone-butanol fermentation using Clostridium saccharoperbutylacetonicum and stated that lower supplementation is cost effective and reduces overall production cost. Table 5 Ethanol yields of pentose fermenting strains Strain Fermentable sugar Ethanol yields (g g −1 ) Reference \n C. lignohabitans \n Glucose + xylose 0.2 [ 7 ] \n K. kakuduensis \n Glucose Traces [ 21 ] \n K. ohmeri \n Glucose Traces (by product) [ 23 ] K. ohmeri strain 5 Xylose + glucose 0.28 This study K. ohmeri strain 6 Xylose + glucose 0.31 This study \n Inhibitor tolerance Lignocellulosic biomass is pretreated to facilitate higher conversion of biomass polysaccharides to fermentable sugars such as glucose, xylose, arabinose etc. This process generates by-products which inhibit growth of microbes and obstruct fermentation process. In general, these inhibitors are classified into four groups including lignin degradation by-products (phenolics), sugar degradation by-products (HMF and furfural), and products derived from the structure of the biomass and heavy metal ions (chromium and nickel) [ 39 ]. Effect of most commonly found inhibitors like HMF, furfural, acetic acid and formic acid was determined on growth of K. ohmeri strains. Concentration ranges were selected based on yields commonly reported in literature and mostly encountered in biomass hydrolysates after different pretreatments [ 40 , 41 ]. Increasing concentrations of HMF and furfural reduced growth of both strains as compared to controls (Fig. 4 ). Furfural was inhibitory in initial growth stages but inhibition was gradually overcome upon prolonged growth after 96 h (Additional file 1 : Figure S3). This coincided with earlier observations that furfural can reduce growth rate above a certain concentration. It has been proved that furfural inhibits alcohol dehydrogenase (ADH) formation which lead to the accumulation of acetaldehyde intracellularly, causing enhanced lag phase of growth during which enzymes and co-enzymes are produced for the reduction of furfural [ 42 ]. HMF also posed similar threats on growth and ethanol productivity of K. ohmeri strains as growth was reduced and lesser biomass resulted in lesser ethanol production. K. ohmeri strains were found to be tolerant to HMF up to 3 g L −1 concentration while at 5 g L −1 concentration, growth was reduced. Fig. 4 Effect of hydroxy methyl furfural on strain 5 ( a ) and strain 6 ( b ). Growth pattern is similar to the control in case of strain 6 and 0.5–3.0 g L −1 concentration of the HMF is not inhibitory for the growth Effect of organic acids on growth of K. ohmeri strains was more pronounced. With formic acid (3–11 g L −1 ) and acetic acid (5–15 g L −1 ), growth was highly affected due to pH change, as optimum pH for yeast growth is 5–6. Formic and acetic acids at concentration used in these experiments reduced pH to 3 leading to reduction in biomass production. In case of acetic acid, there was a sudden rise in growth of both strain 5 and strain 6 after 48 and 72 h respectively. Acetic acid at concentrations up to 6 g L −1 did not cause any reduction in growth of the strains [ 40 ] (Fig. 5 ). Acetic acid works by lowering intracellular pH, which is neutralized by plasma membrane’s ATPase by pumping out protons from the cell, thereby, leading to the production of additional ATPs by increasing ethanol production under anaerobic conditions due to enhanced biomass formation. This might be the reason for sudden rise in growth after a certain period as observed in case of K. ohmeri strains. Effect of formic acid was more severe and growth of both strains was impeded. Major cause of decreased growth was assumed to be lowering of pH as inhibitory effect of formic acid was nullified when pH was adjusted to optimum (data not shown). This reduction was due to drop in extracellular pH which causes diffusion of undissociated acids inside the cell leading to reduction in intracellular pH [ 43 ]. ABE fermentation was repressed by the production of acetic acid produced as a byproduct when C. saccharoperbutylacetonicum was grown on eucalyptus hydrolysates [ 3 ]. Fig. 5 Effect of acetic acid over K. ohmeri strain 5 ( a ) and strain 6 ( b ). Strain 6 exhibits a sudden rise in efficiency after 48 h at a concentration of 5 g L −1 \n Growth and ethanol production by K. ohmeri strains from biomass hydrolysates Kodamaea ohmeri strains were evaluated for growth and ethanol production on biomass hydrolysates prepared from biologically pretreated rice straw. Total sugar content in the hydrolysates was ~ 1.3% (with 2% glucan loading and 57% saccharification efficiency). Growth on hydrolysates was comparable to the control (Fig. 6 ). Maximum sugar consumption and ethanol production occurred within 24 h. HPLC analyses of samples showed ethanol production and maximum ethanol level at 72 h by both the strains and it was ~ 2 and 1.3 g L −1 by strain 5 and strain 6 respectively (Table 6 ). Thus, these strains of K. ohmeri were able to grow and produce ethanol from paddy straw hydrolysates. Fig. 6 Sugar consumption (%) and growth of K. ohmeri strain 5 ( a ) and strain 6 ( b ) on biologically pretreated rice straw hydrolysate \n Table 6 Ethanol yields of K. ohmeri strains from rice straw biomass hydrolysates Ethanol produced (g L −1 ) 24 h 48 h 72 h 96 h K. ohmeri strain 5 (control) 0.98 ± 0.01 0.065 ± 0.003 0.059 ± 0.0005 0.015 ± 0.0025 K. ohmeri strain 5 (hydrolysate) 0.3 ± 0.001 0.35 ± 0.01 1.92 ± 0.04 0.001 ± 0.0015 K. ohmeri strain 6 (control) 1.07 ± 0.01 1.15 ± 0.05 0.71 ± 0.02 0.26 ± 0.02 K. ohmeri strain 6 (hydrolysate) 0.12 ± 0.03 0.78 ± 0.02 1.28 ± 0.01 0.04 ± 0.008"
} | 4,104 |
29773789 | PMC5958124 | pmc | 9,577 | {
"abstract": "Interfacing photosynthetic proteins specifically photosystem 1 (PS1) with electrodes enables light-induced charge separation processes for powering semiartificial photobiodevices with, however, limited long-term stability. Here, we present the in-depth evaluation of a PS1/Os-complex-modified redox polymer-based biocathode by means of scanning photoelectrochemical microscopy. Focalized local illumination of the bioelectrode and concomitant collection of H 2 O 2 at the closely positioned microelectrode provide evidence for the formation of partially reduced oxygen species under light conditions. Long-term evaluation of the photocathode at different O 2 concentrations as well as after incorporating catalase and superoxide dismutase reveals the particularly challenging issue of avoiding the generation of reactive species. Moreover, the evaluation of films prepared with inactivated PS1 and free chlorophyll points out additional possible pathways for the generation of oxygen radicals. To avoid degradation of PS1 during illumination and hence to enhance the long-term stability, the operation of biophotocathodes under anaerobic conditions is indispensable.",
"introduction": "Introduction The integration of isolated photosynthetic protein complexes such as photosystem 1 (PS1) or photosystem 2 (PS2) into semiartificial photoelectrochemical devices requires immobilization strategies that allow optimized electron transfer while ensuring its adequate stability. The design-engineered electron-transfer chains for an efficient coupling of photosynthetic proteins to electrode surfaces were demonstrated for PS1 1 – 10 , PS2 11 – 15 , and bacterial reaction centers 16 , 17 . However, light-induced damage of photosynthetic protein complexes in semiartificial assemblies causes fast degradation together with a drop in activity. Consequently, the potential applicability of biohybrid devices is considerably limited 18 , 19 . Environmental as well as light-induced stress in plants creates an imbalance between the produced reactive oxygen species (ROS) and scavenging antioxidant systems 20 . PS2 has been considered the primary target for photoinhibition 21 and much is known about the mechanism of damage and repair in vivo 21 – 26 . In contrast, PS1 presents a considerably longer life time in vivo 27 and has been considered to be less sensitive to light stress than PS2 22 , 28 . However, PS1 is similarly susceptible to environmental stress 29 , 30 , with ROS causing inactivation of the protein complex 22 , 23 , 31 , 32 . Moreover, since damaged PS1 is not repaired as PS2 33 , PS1 damage seems to be irreversible and may have a much stronger impact on the survival of plants than PS2 photoinhibition 23 . In photobioelectrochemical devices that utilize O 2 as terminal electron acceptor, the generation of ROS is inherently associated. Since the reduction of O 2 to H 2 O requires four electrons, the stepwise production of partially reduced intermediates prevails. The initial step in the formation of partially reduced oxygen species is the generation of the superoxide radical (O 2 •− ) by a one electron reduction of molecular oxygen. The cascade of reactions continues with dismutation of O 2 •− leading to the production of H 2 O 2 and the formation of the hydroxyl radical (HO • ) by a one electron reduction process 20 , 25 , 34 . Moreover, a kinetically hindered uptake of electrons from the F B − site of PS1 can result in back-reactions leading to the formation of chlorophyll in an excited triplet state, able to react with O 2 under formation of singlet oxygen, O 2 ( 1 Δ g ) 35 . All these highly reactive species can easily disrupt any biological assembly 36 , 37 . Many PS1-based semiartificial devices make use of redox mediators to enhance the electron-transfer rate from the photosynthetic protein complex to O 2 . Viologen derivatives exhibit sufficiently negative redox potentials and are widely used as redox mediators for the efficient electrocatalytic reduction of O 2 38 , 39 . However, the increased electron-transfer rates are also associated with an increased generation of partially reduced oxygen species. Particularly, methyl viologen (MV 2+ ) is known for its herbicidal activity in plants due to the production of partially reduced ROS, with H 2 O 2 as the major product in aqueous solutions, and peroxidation of lipid constituents eventually leading to necrosis 39 . MV 2+ effectively scavenges the high-energy electrons from PS1. A voltammetric study about MV 2+ -mediated reduction of O 2 suggested 39 , 40 that the generated methyl viologen radical cation (MV +• ) reacts efficiently with O 2 to generate the superoxide radical (Eq. ( 1 )): 1 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$${\\mathrm{MV}}^{ + \\bullet } + {\\mathrm{O}}_2 \\to {\\mathrm{MV}}^{2 + } + {\\mathrm{O}}_2^{ \\bullet - }\\quad {{k}}_{\\mathrm{f}} = 6 \\times 10^9{\\kern 1pt} {\\mathrm{M}}^{ - 1}{\\kern 1pt} {\\mathrm{s}}^{ - 1}$$\\end{document} MV + ∙ + O 2 → MV 2 + + O 2 ∙ - k f = 6 × 1 0 9 M - 1 s - 1 Subsequently, the generated \\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}$${\\mathrm{O}}_2^{ \\bullet - }$$\\end{document} O 2 ∙ - may dismutate or further react with MV +• with the eventual generation of H 2 O 2 (Eqs. ( 2 ) and ( 3 )): 2 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$2{\\mathrm{O}}_2^{ \\bullet - } + {\\mathrm{2H}}^ + \\to {\\mathrm{O}}_2 + {\\mathrm{H}}_{\\mathrm{2}}{\\mathrm{O}}_{\\mathrm{2}}\\quad {{k}}_{\\mathrm{f}} = 1.3 \\times 10^6{\\kern 1pt} {\\mathrm{M}}^{ - 1}{\\kern 1pt} {\\mathrm{s}}^{ - 1}$$\\end{document} 2 O 2 ∙ - + 2H + → O 2 + H 2 O 2 k f = 1.3 × 1 0 6 M - 1 s - 1 3 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$${\\mathrm{MV}}^{ + \\bullet } + {\\mathrm{O}}_2^{ \\bullet - } + {\\mathrm{2H}}^ + \\to {\\mathrm{MV}}^{2 + } + {\\mathrm{H}}_2{\\mathrm{O}}_2\\quad {k}_{\\mathrm{f}} = 6.5 \\times 10^8{\\kern 1pt} {\\mathrm{M}}^{ - 1}{\\kern 1pt} {\\mathrm{s}}^{ - 1}$$\\end{document} MV + ∙ + O 2 ∙ - + 2H + → MV 2 + + H 2 O 2 k f = 6.5 × 1 0 8 M - 1 s - 1 Therefore, the occurrence of H 2 O 2 can be considered as evidence for the generation of partially reduced oxygen species causing deleterious processes of PS1-based biophotocathodes under aerobic conditions. Our aim was to implement an analytical system for the simultaneous evaluation of O 2 consumption and associated H 2 O 2 production under irradiation and to gain deeper understanding on light-induced stress. As the integration of isolated photosynthetic protein complexes with electrodes decreases their stability and prevents biological repair mechanisms, the effect of reactive species for semiartificial devices is even more delicate. While competing charge transfer pathways and short circuits at the integrated chlorophyll molecules constituting antenna complexes further restrain the efficiency of PS2-based biodevices 41 , 42 , the immobilization of isolated PS1 at electrodes has shown as well a rather limited stability upon light irradiation 18 , 19 , 29 . However, the long-term stability of PS1-based photoelectrodes has been commonly underestimated and much less is known about inactivation processes and ROS generation at devices using isolated PS1 coupled to electrode surfaces. Particularly, under aerobic conditions and when MV 2+ is used as an electron acceptor, the fast electron transfer to O 2 is unequivocally causing the generation of partially reduced oxygen species 39 , leading to an eventual loss in PS1 activity. The deleterious effect of partially reduced oxygen species generated under light is therefore also of particular importance for PS1-based photobioelectrodes operated under aerobic conditions, an effect which until now has been suggested but not extensively evaluated because of the lack of experimental evidence. Scanning electrochemical microscopy (SECM) is a scanning probe technique for the visualization of local electrochemical activity. An accurately positioned microelectrode tip is used as a probe for the evaluation of electrochemical processes or collection of species of interest evolved at the sample. Particularly for the analysis of photoactive materials, scanning photoelectrochemical microscopy (SPECM) was suggested combining SECM with a light source for the controlled and focalized illumination of the sample 43 , 44 . We present an in-depth evaluation of a highly efficient photobiocathode 18 comprising isolated PS1 embedded within an Os-complex-modified polymer. SPECM is used for investigation of light-induced stress at the PS1-based photocathode and the concomitant generation of partially reduced oxygen species supposedly involved in PS1 degradation. Evidence for the formation of partially reduced oxygen species under illumination of PS1/redox polymer-modified photobiocathodes is demonstrated by in situ collection of H 2 O 2 at the SPECM tip. Simultaneous localized irradiation of the PS1-based photocathode and collection of H 2 O 2 at different O 2 concentrations and in the presence of scavenging enzymes, i.e., catalase (Cat) and superoxide dismutase (SOD), is presented. Moreover, by studying inactivated PS1 and free chlorophyll immobilized within the Os-complex-modified polymer, previous observations at PS2-based bioanodes concerning the light-induced generation of partially reduced oxygen intermediates at chlorophyll pigments were confirmed 41 , 42 .",
"discussion": "Discussion We have demonstrated the in situ detection of H 2 O 2 and the associated O 2 consumption at PS1/redox polymer-based photocathodes under light stress conditions. Performing the experiments over long periods of time while working with intermittent consecutive dark and light conditions were used to minimize variations in electron transfer and diffusion rates by potential local heating effects. Moreover, the use of intermittent illumination allows recording a clear baseline for photocurrent monitoring over extended periods of time, making long-term evaluation of the bioelectrodes under different conditions possible. PS1/Os-complex-modified redox polymer photocathodes working under aerobic conditions generate partially reduced oxygen species, which are responsible for the degradation of the PS1 protein complex and a drop in activity of the photoelectrodes with time. Incorporation of Cat and SOD in the PS1/Os-P biodevice demonstrates that suppression of partially reduced oxygen species is difficult due to the high reactivity and short diffusion time of partially reduced oxygen species. We have additionally shown that the formation of these partially reduced reactive species under light is not only associated with the main electron-transfer pathway following charge separation by PS1 upon light absorption. H 2 O 2 associated with O 2 consumption could be also demonstrated for thermally inactivated PS1 complexes and free Chl a , indicating the capability of O 2 reduction by chlorophyll pigments. Our findings provide further understanding about electron-transfer pathways at PS1-modified electrode surfaces and highlight the limited applicability of PS1-based photocathodes working under aerobic conditions, where the unavoidable formation of ROS is responsible for an overall short long-term stability. Hence, the obtained results are strongly encouraging the design of PS1-based biophotocathodes operating under strictly anaerobic conditions."
} | 3,054 |
35441658 | PMC9154287 | pmc | 9,579 | {
"abstract": "Abstract Motivation Bioproduction of value-added compounds is frequently achieved by utilizing enzymes from other species. However, expression of such heterologous enzymes can be detrimental due to unexpected interactions within the host cell. Recently, an alternative strategy emerged, which relies on recruiting side activities of host enzymes to establish new biosynthetic pathways. Although such low-level ‘underground’ enzyme activities are prevalent, it remains poorly explored whether they may serve as an important reservoir for pathway engineering. Results Here, we use genome-scale modeling to estimate the theoretical potential of underground reactions for engineering novel biosynthetic pathways in Escherichia coli . We found that biochemical reactions contributed by underground enzyme activities often enhance the in silico production of compounds with industrial importance, including several cases where underground activities are indispensable for production. Most of these new capabilities can be achieved by the addition of one or two underground reactions to the native network, suggesting that only a few side activities need to be enhanced during implementation. Remarkably, we find that the contribution of underground reactions to the production of value-added compounds is comparable to that of heterologous reactions, underscoring their biotechnological potential. Taken together, our genome-wide study demonstrates that exploiting underground enzyme activities could be a promising addition to the toolbox of industrial strain development. Availability and implementation The data and scripts underlying this article are available on GitHub at https://github.com/pappb/Kovacs-et-al-Underground-metabolism . Supplementary information \n Supplementary data are available at Bioinformatics online.",
"introduction": "1 Introduction Due to our strong dependency on fossil materials, there is a growing interest for the development of sustainable ways to replace them and to provide alternative approaches to produce chemical compounds with application in fields such as medicine (pharmaceuticals), cosmetics, materials (e.g. bioplastics) and food. One promising approach is the use of microbial cell factories to produce such value-added compounds. More specifically, the metabolic network of these species can be redesigned and engineered to optimize the production of the desired compounds ( Hadadi and Hatzimanikatis, 2015 ; Ko et al. , 2020 ; Nielsen and Keasling, 2016 ; Wang et al. , 2017 ; Yim et al. , 2011 ). Several metabolic engineering approaches have been developed to improve the production of value-added compounds ( Okano et al. , 2018 ; Pontrelli et al. , 2018 ). Up till now, the most widely applied approach is the introduction and expression of heterologous metabolic reactions (i.e. reactions from a different organism), which generate heterologous biosynthetic pathways that enable the production of non-native value-added compounds by the host organism ( Lechner et al. , 2016 ; Pickens et al. , 2011 ). Several computational approaches have been proposed to facilitate pathway engineering through heterologous reactions using genome-scale metabolic modeling ( Ko et al. , 2020 ; Wang et al. , 2017 ). Genome-scale metabolic models are available for the most common microorganisms used in metabolic engineering and they can accurately predict how the addition of heterologous enzymes affect metabolite production ( Pharkya et al. , 2004 ). Although the vast repertoire of available heterologous reactions provide a huge potential for value-added compound production, there are several limitations. First, heterologous enzymes often require specific cofactors which cannot be provided by the host organism hindering the biosynthetic pathway of its proper working ( Boynton et al. , 1996 ). Second, heterologous expression could lead to stress response due to protein overproduction or accumulation of toxic intermediates ( Gill et al. , 2000 ; Martin et al. , 2003 ). Third, microorganisms containing heterologous reactions are considered as GMO, whose commercial application, especially in food industry, relies on the juristic regulations of countries. One possible approach to overcome these limitations is to exploit what is known as the underground metabolism, i.e. the collection of enzyme side activities in a cell ( D’Ari and Casadesús, 1998 ; Fig. 1 ). In addition to their native activities, most enzymes display weak side activities by which substrates are turned into products, albeit at low rates, due to the limited substrate specificity of the enzyme. Although the physiological effect of such underground reactions is mostly neglected given their inefficient kinetics, they can often be enhanced by only a few mutations ( Aharoni et al. , 2005 ; Khersonsky and Tawfik, 2010 ; Notebaart et al. , 2014 ). Such genetic changes allow underground metabolism to contribute to microbial growth and adaptation to novel nutrient conditions ( Cam et al. , 2016 ; King et al. , 2017 ; Notebaart et al. , 2014 ; 2018 ). Although underground activities have been utilized for pathway development in case studies, no comprehensive work has yet explored the biotechnological potential of the underground metabolic network ( Notebaart et al. , 2018 ; Rosenberg and Commichau, 2019 ). In particular, it remains unexplored whether biochemical reactions catalyzed by enzyme side activities are as likely to contribute to new biosynthetic pathways toward value-added compounds as those catalyzed by heterologous enzymes. Fig. 1. Utilization of underground activities as an alternative to incorporating heterologous enzymes. The schematic figure depicts two strategies to enhance or establish the production of industrially relevant chemicals in microorganisms. The conventional strategy is to introduce one or more heterologous enzymes from a different organism (cell with long flagella, blue) to create a heterologous pathway capable of producing the value-added compound (lower part). A less well explored alternative is to amplify existing low-activity underground reactions (upper part, orange arrow) catalyzed by the endogenous enzymes of the organism (A color version of this figure appears in the online version of this article) Here, we aim to systematically assess the extent to which underground metabolism can increase the production of industrially important compounds. As genome-scale computational modeling of underground metabolism successfully predicts the potential to utilize new nutrient sources in Escherichia coli ( Notebaart et al. , 2014 ), we reasoned that a similar approach could be employed to characterize the theoretical potential of underground reactions to produce industrially relevant chemical compounds. To this end, we integrated experimentally reported and predicted underground reactions into a genome-scale metabolic reconstruction of E. coli ( Orth et al. , 2011 ) and characterized its biosynthetic properties across 64 different industrially important compounds spanning diverse applications, such as health, material sciences, chemical industry and food. Our computational analyses demonstrate that underground enzyme activities frequently enhance the production of value-added compounds with industrial importance, which is comparable to heterologous enzyme activities that enhance such value-added compounds. Notably, we find that activation of only a single underground reaction can already substantially increase the production of important industrial compounds, such as a precursor of bioplastics (3-hydroxypropanoate) and a potential biofuel (1-butanol). Overall, our work reveals that the underground metabolic network provides a promising metabolic engineering tool to improve the production of both native and non-native value-added compounds.",
"discussion": "4 Discussion By extending the native metabolic network of E. coli by underground reactions, we present a framework to further broaden the metabolic engineering application for the production of value-added compounds ( Pontrelli et al. , 2018 ). Our genome-scale computational analysis gave several new insights. First, we show that underground metabolism can enhance the production yield of numerous industrially relevant compounds and even enable the production of new compounds that cannot be produced by the native metabolic network. Second, we also demonstrate that production of a given value-added compound through underground metabolism often hinges on one or few underground reactions only. This implies that it would be sufficient to engineer only few enzyme side activities for any given target compound. Engineering a small number of enzymatic steps would likely benefit a successful metabolic engineering strategy, especially since it concerns weak-side activities of the enzyme. There are several ways to translate our predictions toward the production of engineered strains in vivo , which is clearly the next major step. First, mutations that increase the underground activity could be engineered via various genome editing techniques, such as CRISPR-Cas and MAGE-based methods ( Csörgő et al. , 2020 ; Jakočiūnas et al. , 2016 ; Wang et al. , 2009 ). Second, adaptive laboratory evolution has shown to be successful in increasing underground activities ( Guzmán et al. , 2019 ; Pontrelli et al. , 2018 ), and third, a combination of editing and evolution could be applied ( Pontrelli et al. , 2018 ; Wannier et al. , 2020 ). Perhaps the most important new insight provided by our study is that underground and heterologous reactions have similar theoretical potentials to contribute to the production of value-added compounds. Thus, underground enzyme activities may provide a complementary source of biochemical reactions for overproduction purposes. This is a notable result, because the use of heterologous genes might be unfavorable for applications that demand a GMO-free status, such as food fermentation products. In contrast, the use of underground reactions coupled with adaptive evolution does not involve the introduction of specific DNA from other organisms. As such, the second approach may contribute to applications that go beyond GMO. Moreover, in a preliminary analysis we found several cases where specific underground and heterologous activities are jointly required to improve the production yield of value-added compounds. This result suggests that combining the two reaction repertoires can be advantageous for industrial applications. Clearly, future works are needed to fully explore this possibility. Last, we expect that underground activities might have a limited potential compared to heterologous enzyme activities in one particular area of application: many industrially relevant metabolites are produced through secondary metabolism in plants, which are unlikely to be producible by side activities of microbial host enzymes. We note that our estimate of the industrial potential of underground reactions might be distorted by at least two phenomena. On the one hand, the applied modeling framework might overestimate the contribution of underground reactions to new pathways for several reasons. First, we used a simple FBA framework that ignores thermodynamic realizability and therefore some of the predicted pathways might be unrealistic under physiologically relevant metabolite concentrations. Genome-scale modeling methods that incorporate thermodynamics as well as enzyme constraints through kinetics could address this shortcoming ( Hoppe et al. , 2007 ; Salvy et al. , 2019 ; Sánchez et al. , 2017 ). Recent frameworks allow the incorporation of enzyme kinetics data as well as quantitative omics data into models, which could result in the prediction of biological relevant phenotypes through FBA ( Filippo et al. , 2022 ; Sánchez et al. , 2017 ). Second, underground enzyme activities should not interfere with the native metabolic network structure to become biochemically and physiologically functional. As such, it might be challenging to enhance underground activities in vivo, especially if extensive protein engineering is needed ( Porokhin et al. , 2021 ). Third, the FBA-based modeling framework might be over-optimistic and in reality, multiple genetic modifications are needed to achieve the desired production. For example, a single underground activity associated with the fucO gene is sufficient for efficient ethylene glycol production in silico. However, in addition to fucO overexpression, three other native and two heterologous enzymes had to be overexpressed and a gene deleted to achieve a high-flux pathway from glucose to ethylene glycol in vivo ( Pereira et al. , 2016 ). Similarly, to produce (S)-Propane-1,2-diol, a previous work overexpressed three genes, replaced an enzyme with a more efficient heterologous enzyme and disrupted pathways that alter the flux from the synthesis ( Clomburg and Gonzalez, 2011 ). In contrast, the addition of a single underground activity is sufficient to increase the predicted yield in our simulations. Therefore, hits from our simulations are potentially needed to be expanded with further pathway improvements, including deletion of genes that divert the flux from biosynthesis and modifications that alter the redox balance. Nevertheless, our method gives suggestions that open new ways to produce important chemicals after further refinement. On the other hand, our knowledge of underground activities is still rudimentary and there might be orders of magnitude more side activities than currently known that could potentially be recruited for new pathways. Notably, there are various promising recent reports to predict metabolic (side) reactions from cheminformatics, enzyme structures and machine learning ( Amin et al. , 2019 ; Carbonell et al. , 2014 ; Carbonell and Faulon, 2010 ; Koch et al. , 2017 ; Mou et al. , 2021 ; Robinson et al. , 2020 ). We therefore anticipate that future advances in machine learning and cheminformatics will further expand the known ‘metabolic reaction space’ of species, i.e. the total number of metabolic reactions that could potentially be active in a species ( Hafner et al. , 2021 ; Tyzack et al. , 2019 ). This ‘space’ could be exploited for the production of value-added compounds ( Campodonico et al. , 2014 ; Carbonell et al. , 2014 ). The present study focused on underground reactions as raw materials for building new pathways, however, promiscuous activities of enzymes in the existing metabolic network may also affect negatively the production of target compounds ( Kim and Copley, 2012 ). For example, it has been shown that promiscuous phosphatase activities redirect flux from a heterologous terpenoid biosynthetic pathway, hence decreasing its efficiency ( Wang et al. , 2018 ). A more complete knowledge of the repertoire of underground reactions, including those catalyzed by native and heterologous enzymes alike, would therefore be instrumental to avoid network disruptions arising from enzyme promiscuity ( Porokhin et al. , 2021 ). Our research focuses on public information of value-added compounds used in the industry, but this is likely an underestimate of the total complement of compounds of interest. Therefore, our approach could easily be adapted for the needs of the industry to incorporate their compound of interest. Moreover, we report that changes in the exact nutrient environment does not alter the predictions substantially, since the value-added compounds are produced from central metabolism. This may, however, change once the underground metabolic network is further extended in the future, as well as when additional value-added compounds are added to the network. Hence, our approach may also predict new industrially relevant cost-reducing environments to produce similar product yield or even increased yields. Our results pave the way for exploiting underground metabolism to produce novel strains and we anticipate that a growing interest in underground metabolism will go together with a rise of biotechnological applications."
} | 4,038 |
40136907 | PMC11942406 | pmc | 9,580 | {
"abstract": "Polyzwitterion (PZW) hydrogel has excellent marine anti-biofouling performance, but it is difficult to effectively work for a long time in natural seawater due to its weak mechanical strength. In this study, a new natural rubber (NR)-PZW composite hydrogel has been reported for long-term anti-biofouling by simply dispersing NR latex into the poly(sulfobetaine methacrylate) (PSBMA) hydrogel network. First of all, owing to the PZW hydrogel network having an anti-polyelectrolyte effect, this NR-PZW hydrogel can provide outstanding anti-biofouling performance, including broad-spectrum anti-bacteria, anti-algae, and anti-protein properties in marine environments. Furthermore, it has a composited natural rubber nanoparticle with a hydrophilic negatively charged outer protein membrane, which can uniformly disperse in the hydrogel to significantly improve its mechanical properties. Therefore, this composited hydrogel can provide not only highly enhanced tensile strength (0.52 MPa) but also ultra-high breaking elongation (738%), which can effectually resist harsh seawater environments. As a result, the NR-PZW composite hydrogel can achieve excellent anti-biofouling performance for more than 3 months within a real marine environment. This work can provide an excellent, robust polyzwitterionic hydrogel for long-term marine anti-biofouling, which will also inspire new strategies for anti-biofouling materials.",
"conclusion": "3. Conclusions In summary, by simply dispersing hydrophilic natural rubber nanoparticles into the PZW hydrogel network, a new NR-PZW hydrogel membrane has been presented for marine anti-biofouling. This hydrogel can be obtained simply by mixing the NR latex into the sulfobetaine methacrylate (SBMA) zwitterionic hydrogel precursor solution before free-radical polymerization. Firstly, owing to the PZW network with an excellent anti-polyelectrolyte effect, this hydrogel can form a dense ultra-hydrophilic water film on the hydrogel surface to efficiently resist bio-adhesion, which is due to excellent broad-spectrum anti-bacteria, anti-algae, and anti-protein properties in the marine environment. Furthermore, the NR nanoparticles of the NR latex with hydrophilic protein outer shell layer can uniformly disperse in the PZW hydrogel covalent-crosslinked network. The composited NR can further form strong supramolecular interaction between the carboxylate radical groups (negative charges) of its protein shell layer and the quaternary ammonium groups (positive charges) of the PZW polymeric chain, which can provide highly enhanced mechanical properties including tensile strength from 0.04 to 0.52 MPa and breaking elongation from 94% to 738%. Therefore, this NR-PZW couples the highly efficient marine anti-biofouling and robust mechanical performance for long-term usage. As a result, this NR-PZW composite hydrogel can achieve excellent anti-biofouling performance for more than 3 months within a real marine environment. This work can provide a promising long-term marine anti-biofouling material, which will also inspire new strategies for anti-biofouling materials.",
"introduction": "1. Introduction Marine biofouling severely damages underwater facilities and is caused by the attachment and invasion of marine organisms, including bacteria, seaweeds, shellfish, and barnacles [ 1 , 2 ]. Marine biofouling leads to more than USD 15 billion of economic loss per year, especially in shipping, aquaculture, and offshore oil/gas exploitation [ 3 ]. To deal with this problem, many anti-biofouling methods have been developed, mainly including physical elimination and chemical bio-killing techniques [ 4 , 5 ]. Physical elimination techniques [ 4 , 5 , 6 , 7 ] contain manual or mechanical removal of biological dirt, and recent ultraviolet or ultrasonic-assisted cleaning methods, which are low cost but the process is cumbersome and inefficient. Chemical bio-killing techniques [ 8 , 9 ] can sustain resistance to marine bio-adhesion over a long time via chemical surface coatings by killing the marine creatures, but these methods may damage the marine environment, although the bio-killing regents have evolved from high toxicity (such as Chlorine-based oxidizing biocides, tributyltin, and Zinc sulfide) to low toxicity (such as organic quaternary ammonium salt and inorganic cuprous oxide). In recent years, different from the traditional methods above, some new eco-friendly biofouling coatings [ 10 ] have emerged that can efficiently resist marine biofouling without discharging harmful substances into the marine environment. For example, through designing biomimetic micro-structure surfaces [ 11 , 12 , 13 ] or low surface energy coatings [ 14 , 15 , 16 ], marine organisms cannot adhere to the surface easily. Among them, the eco-friendly hydrogel-based anti-biofouling membrane is one of the hotspots in this marine anti-fouling field [ 17 , 18 ]. Hydrogel-based membranes commonly own soft and hydrophilic surfaces, which themselves have a certain degree of anti-bioadhesion and can be easily designed for further improvement of their anti-biofouling performance. Firstly, hydrogels with hydrophilic 3-dimensional networks can contain various nontoxic biomass bio-killing reagents [ 19 , 20 ], which can be control-released for long-term marine anti-biofouling. Furthermore, most recently, new kinds of eco-friendly polyzwitterionic (PZW) hydrogels [ 21 , 22 , 23 ] have been explored based on their excellent anti-polyelectrolyte effect [ 24 , 25 , 26 ], which not only can efficiently resist marine biofouling but also do not release harmful materials to damage the marine environment. However, these PZW hydrogels commonly cannot endure natural marine environments for a long time due to their naturally weak mechanical properties [ 23 , 24 ]. That is, although the PZW hydrogels own outstanding anti-biofouling performance for both micro-organisms and large organisms, their high content of water leads to relatively low mechanical strength, leading to constant invasion of marine organisms and intense external forces from waves/tides in harsh marine conditions. To enhance the mechanical properties of the PZW hydrogel for long-term usage in marine environments, double network (DN) structures [ 27 , 28 ], interpenetrating polymer network (IPN)/semi-IPN structures [ 29 , 30 , 31 ], and composited with nano-materials [ 32 , 33 ] have been introduced. Herein, a natural rubber (NR)-PZW composite anti-biofouling hydrogel membrane has been explored by simply mixing the NR latex into the sulfobetaine methacrylate (SBMA) zwitterionic hydrogel precursor solution, which was then free-radical polymerized at 65 °C ( Scheme 1 ). First of all, owing to the PZW hydrogel network with an excellent anti-polyelectrolyte effect [ 26 ], this NR-PZW hydrogel can form a dense ultra-hydrophilic water film on the hydrogel surface to efficiently resist bio-adhesion, which can provide excellent broad-spectrum anti-bacteria, anti-algae, and anti-protein properties in marine environments. Furthermore, thanks to the NR nanoparticles of the NR latex with a hydrophilic protein outer shell layer, there can be uniform dispersion in the PZW hydrogel covalent-crosslinked network. The composited NR nanoparticle can form a strong supramolecular interaction between the carboxylate radical groups (negative charges) of its protein shell layer and the quaternary ammonium groups (positive charges) of the PZW polymeric chain, which can provide highly improved mechanical properties including not only highly enhanced tensile strength from 0.04 to 0.52 MPa but also breaking elongation from 94% to 738% [ 34 , 35 ]. Therefore, this as-prepared NR-PZW couples highly efficient marine anti-biofouling and robust mechanical performance for long-term usage. As a result, this NR-PZW composite hydrogel can achieve excellent anti-biofouling performance for more than 3 months within a real marine environment. This work can provide a promising long-term marine anti-biofouling material, which will also inspire new strategies for anti-biofouling materials.",
"discussion": "2. Results and Discussion 2.1. Basic Characterizations of the NR-PZW Hydrogel Membrane The basic chemical component and the morphology of NR-PZW composite hydrogel were researched ( Figure 1 and Figure S1–S8 ). The differences in chemical structure between NR, PZW hydrogel, and NR-PZW hydrogel were evaluated via Fourier transform infrared spectroscopy (FT-IR) ( Figure 1 a). Firstly, the specific adsorption peaks of NR were tested based on the reported literature [ 34 , 35 ]. (1) Specific peaks of NR’s cis-polyisoprene. Three strong peaks: 2960 cm −1 of -CH 3 asymmetric stretching vibration, 2916 cm −1 of -CH 2 symmetric stretching vibration, and 2851 cm −1 of -CH 2 asymmetric stretching vibration; two strong peaks: 1448 cm −1 of antisymmetric deformation vibration peak of methylene, 1375 cm −1 of symmetric deformation vibration of methyl; strong single peak: 840 cm −1 of C-H out-of-plane deformation vibration on a cis-double-substituted carbon–carbon double bond. (2) Specific peaks of proteins on the surface of the NR nanoparticle. Weak single peak: 1629 cm −1 of amide I absorption band and a weak single peak of 1544 cm −1 of amide II absorption band. Secondly, the specific adsorption peaks of the PZW were measured: two strong peaks of the sulfonic group including 1033 cm −1 and 1064 cm −1 in the spectrum of PZW came from the SBMA zwitterionic monomer. Comparison of the characteristic adsorption peaks between the NR, PZW, and NR-PZW samples, the appearance of the specific PZW peaks, and the specific cis-polyisoprene peaks in the spectrum of the NR-PZW indicated that the NR-PZW hydrogel coupled the PZW polymer chain and the NR nanoparticle. Furthermore, the weak single peak (1544 cm −1 of amide II specific absorption band) in the spectrum of the NR-PZW further proved the dispersion of NR nanoparticles in the NR-PZW composited hydrogel. In addition, because the content of protein was low in the NR, both of the two specific peaks were weak. Because the content of protein was further reduced in the NR-PZW hydrogel, only the weak single peak (1544 cm −1 of amide II specific absorption band) can be found in the spectrum of the NR-PZW. The nanoparticle of the NR latex was observed by scanning electron microscopy (SEM) and the SEM-connected energy-dispersive spectrometer (EDS) mapping ( Figure 1 b and Figures S1 and S2 ): the NR nanoparticles were about 500 ± 300 nm of average diameter, and each of them had a wrinkled surface with a high surface area; the SEM-connected EDS mappings showed a clear but slight signal of S ( Figure 1 b) and N ( Figure S2 ), which confirmed the thin protein of the NR nanoparticle outer layer had a small amount of S (because the inner of NR nanoparticle was made of polyisoprene without S). Furthermore, the NR-PZW was compared with the NR sheet and the PZW hydrogel based on the SEM-connected EDS mappings ( Figure 1 c and Figures S3–S8 ). Compared to the slight S element signal intensity of the NR sheet and the ultrahigh S element signal intensity of the PZW hydrogel, the moderate intensity of the S signal of the NR-PZW hydrogel further confirmed that it was composited with NR and PZW. In addition, the cross-section SEM images of the NR-PZW ( Figure S7 ) showed a uniform microporous structure, and the related EDS mappings ( Figure S8 ) also manifested evenly the S signal, which further confirmed the NR nanoparticles were uniformly distributed in the entire volume of NR-PZW hydrogel. That is, NR and PZW can be evenly composited to form the NR-PZW hydrogel, which may explain why although the hydrogelation time was 3 h, the precursor solution state can transform the bulk hydrogel state with the continuous 3D cross-linked PZW network in no more than 10 min (although the reaction is still going on) and the NR nanoparticles (with lower density than water) can be evenly fixed in the cross-linked hydrogel network before the NR nanoparticles aggregating and floating upward. In order to further explore the composition of NR-PZW hydrogels, NR, PZW hydrogels, and NR-PZW hydrogels were compared by X-ray photoelectron spectroscopy (XPS) ( Figure 2 and Figure S9 ). In a comparison of the specific peak of the sulfonic group between the NR sheet, PZW hydrogel, and the NR-PZW hydrogel, there was no specific peak of the sulfonic group in the NR sheet spectrum, while both the PZW hydrogel and the NR-PZW hydrogel had a strong characteristic peak of S2p at (165–170 eV), while there was no signal at the corresponding position of the NR hydrogel spectrum ( Figure 2 a and Figure S9 ). This test result indicated that the NR-PZW hydrogel has the PZW hydrogel polymer chain, which had an S element. Furthermore, as shown in the more clear XPS peak fitting curve of NR-PZW hydrogel ( Figure 2 b), the S2p element signal came from the superposition of two sulfonic acid group signals including S2p 1/2 at 168.7 eV and S2p 3/2 at 167.4 eV, which further verified that the S element of NR-PZW hydrogel came from the sulfonic acid group of the PZW polymer chain. 2.2. Mechanical Properties of NR-PZW Hydrogel The mechanical performance of marine antifouling materials that can withstand wave/tidal erosion and marine biological invasion is an important indicator that must be considered in order to be effectively used in real marine environments for a long time. In particular, hydrogel materials naturally have weak mechanical properties, and it is difficult to maintain long-term integrity and highly efficient anti-bioadhesion in a real seawater environment. In this work, the mechanical properties of the NR-PZW hydrogel were evaluated ( Figure 3 and Figures S10–S15 , Table S1 ). A series of NR-PZW hydrogels with a constant 40 wt% of dry mass percentage and different mass ratios of PZW/NR have been systematically researched ( Figure 3 a and Figure S10 ) to explore the optimum formulation of the NR-PZW hydrogel for improvement of mechanical properties. With the increase in NR, from m (PZW) /m (NR) = 35/5 to 20/20, the mechanical properties (both the tensile strength and the breaking elongation) of NR-PZW hydrogel increased gradually and reached the peak value at m (PZW) /m (NR) = 20/20, from 0.07 MPa of tensile strength and 111% of breaking elongation to 0.52 MPa of tensile strength and 738% of breaking elongation. However, with a further increase of NR, from m (PZW) /m (NR) = 20/20 to 10/30, the tensile strength decreases gradually, and the breaking elongation remains almost unchanged. In addition, with the change in m (PZW) /m (NR) , the SEM images of different NR-PZW hydrogels were significantly different, which indicated the large difference in microstructures between different NR-PZW hydrogels ( Figure S11 ). Therefore, the mass ratio of m (PZW) /m (NR) = 20/20 can be selected to fabricate the optimum NR-PZW hydrogel for all experiments below without special instruction. Furthermore, the Zeta potential and electrophoretic mobility of the 20 wt% natural rubber aqueous dispersion showed −3.4 ± 0.27 eV and −44.3 ± 0.79 μm cm/V s, which manifested that the NR nanoparticles have relatively strong negative charges based on the carboxylate radical groups on their surface ( Figure S12 ). Therefore, this experiment result may explain that with the increase in NR, the anion–cation interaction between the carboxylate radical groups of the NR nanoparticle protein outer layer and the quaternary ammonium groups of the PZW hydrogel network, can be gradually enhanced to form the added supramolecular interaction of the original PZW hydrogel covalent-crosslinked network, which can significantly enhance the tensile strength; but the PZW largely decreased with a further increase in the NR, which can largely decrease the strength of the PZW hydrogel network and subsequent reduction tensile strength of the NR-PZW hydrogel. In addition, with the increase in NR, this strong supramolecular interaction can firmly and evenly fix the NR nanoparticles with excellent elasticity in the PZW network, which can achieve excellent stress dissipation to reduce stress concentration and, consequently, provide higher and higher tensile strengths. Furthermore, comparison of the mechanical properties of pure PZW hydrogel, that of the NR-PZW hydrogel [m (PZW) /m (NR) = 20/20], were significantly enhanced from 0.04 MPa of tensile strength and 94% of breaking elongation to 0.52 MPa of tensile strength and 738% of breaking elongation, which may be due to the composited NR with high tensile strength and high elasticity ( Figure 3 b) and can be further confirmed by the difference of SEM images ( Figure 3 c) and mechanical properties ( Figures S13 and S14 ) between them. In addition, a piece of NR-PZW hydrogel with a thickness of 1.5 mm and a width of 1.5 cm was used to lift a bottle of 500 g water ( Figure S15 ), which demonstrated excellent mechanical properties and indicated that it can be used as an anti-biofouling material for long-term usage in the harsh real marine environment. 2.3. Swelling–Deswelling Performance of NR-PZW Hydrogel The swelling–deswelling performance of NR-PZW hydrogel was researched in fresh water and seawater ( Figure 4 and Figure S16, Table S2 ). Although common polyelectrolyte-based hydrogels are super-hydrophilic and can provide a dense H 2 O film on their surface for anti-bioadhesion in fresh water, their hydrophilicity can be largely reduced from when they are in pure water to when they are in seawater with high salinity (about 3.5 wt%), and their size will significantly deswell. That is, common polyelectrolyte-based hydrogels cannot achieve a surface-dense H 2 O film in seawater, which cannot provide highly efficient anti-biofouling. This polyzwitterionic NR-PZW hydrogel, different from polyelectrolyte-based hydrogels, can show better hydrophilicity form in fresh water than in seawater, owing to the anti-polyelectrolyte effect, which can provide a denser H 2 O film and better anti-biofouling performance in seawater than in pure water. From pure water to seawater, the size of the NR-PZW hydrogel largely changed by 50% from 2 cm to 3 cm ( Figure 4 a), and the swelling ratio changed from 210% to 605% ( Figure 4 b), calculated by Formula (1), which confirmed the NR-PZW hydrogel membrane in seawater can provide better hydrophilicity and a denser water film on its surface than in fresh water and can provide excellent anti-biofouling in theory. That is, this dramatic change can be explained thanks to the poly(sulfobetaine methacrylate) (PSBMA) polymeric network with the synergy of its positive quaternary ammonium groups and negative sulfonic acid groups, which provide excellent anti-polyelectrolyte effects in seawater with high-salinity. In contrast, the NR-polyacrylamide (NR-PAM) composite hydrogel showed significant volume shrinkage from pure water to seawater (its size shrank 20% from 2 cm to 1.6 cm and its swelling ratio decreased from 338% to 214%) ( Figure 4 b and Figure S16 ), which may be because it had composited charged NR nanoparticles, similar to the polyelectrolyte. Furthermore, the hydrophilicity of the NR-PZW can be evaluated via comparison of the seawater contact angle change between NR, NR-PZW, and PZW ( Figure S17 ): the NR had relatively poor hydrophilicity with contact angles of 80.3° and 77.6° after 0.5 s and 1.0 s, respectively, while the PZW had excellent hydrophilicity with contact angles of only 24.8° and 21.2° after 0.5 s and 1.0 s, respectively. The NR-PZW had a contact angle of 46.7° and 35.4° after 0.5 s and 1.0 s, respectively, which manifested that the NR-PZW can still maintain excellent hydrophilicity in seawater, although composited with relatively poor hydrophilic NR. 2.4. Anti-Biofouling Properties of NR-PZW Hydrogel Marine biofouling is mainly caused by various marine organisms, such as bacteria, microorganisms, and many other large marine organisms, and the biofouling process is complex and long term [ 1 , 2 , 3 ]. Generally, when marine equipment or systems without anti-biofouling properties are immersed in the ocean, surface fouling (especially in biofouling) is severe. Firstly, organic matter such as protein, nucleic acid, and polysaccharide will adhere to the surface of the material within a few minutes to form a layer of nutrient base film, and then attract bacteria and microalgae that live on it and form a microbial mucous membrane in about 1 day. After several days of development, it is enough to attract the reproduction of algal spores and the feeding of small organisms. Finally, after several months of continuous biological fouling, large algae and shellfish will take root on the surface of the material and attract large organisms to feed. Once the Marine biological fouling process enters the third step, it will cause irreversible damage to the material, so an important strategy for designing anti-biofouling materials is to delay the formation of the nutrient base membrane and microbial mucosa as much as possible. Hydrogels with high hydrophilicity will form a water film on the surface (the PZW hydrogel can form a denser water film than common hydrogel) when they are in the water environment, which shows unique advantages in this respect. The anti-bioadhesion properties of the NR-PZW hydrogels were researched including anti-bacteria, anti-algae, and anti-protein ( Figure 5 and Figure 6 ) properties, to systematically evaluate the marine anti-biofouling performance of the NR-PZW hydrogel. The anti-biofouling properties of NR-PZW hydrogels were researched within a simulated marine environment based on real seawater ( Figure 5 ). A series of experiments were designed for the evaluation. Firstly, there were representative Gram-positive bacterium ( Escherichia coli ), representative Gram-negative bacterium ( Staphylococcus aureus ), and unique marine bacterium ( Vibrio alginolyticus ), which were selected to test the anti-bacterial adhesion properties of NR-PZW hydrogel. The preparation procedure of the NR-PAM hydrogel as the control group was the same as that of the NR-PZW hydrogel, except that the zwitterionic monomer SBMA was replaced by acrylamide (AM). In this work, a staining method was selected to test the anti-bacterial adhesion performance of hydrogels: two groups of hydrogel samples were incubated with E. coli , S. aureus, and V. alginolyticus , respectively, and then the surface of the samples was washed with sterilized seawater three times to simulate the scouring effect of ocean waves. Finally, the bacterial adhesion of the samples after staining with green fluorescent nucleic acid stain (SYTO-9) was observed by laser confocal microscopy. In addition, in order to better simulate the marine environment, a real seawater-based culture medium was specially used for bacteria cultivation. As can be seen from the confocal laser photos, the surface of the NR-PAM hydrogel as the control group showed serious E. coli adhesion, while only a small number of bacteria appeared on the surface of the NR-PZW hydrogel, and the calculated inhibition ratio of E. coli reached 86.3% ( Figure 5 a,d). Similar effects were seen in confocal laser images of S. aureus and V. alginolyticus, where NR-PZW hydrogel showed even better inhibition ratios of 94.4% and 95.7% against these two bacteria, respectively, indicating that it was a highly effective broad-spectrum anti-bacterial adhesion material in the ocean ( Figure 5 b–d). The excellent anti-bacterial adhesion performance of NR-PZW hydrogels may be attributed to the super-hydrophilic charged groups of the PZW network, which can produce strong hydration with water molecules to form a dense hydration water surface layer on the NR-PZW hydrogel to resist bacterial adhesion. In addition, NR-PZW hydrogels become more hydrophilic in seawater because of the anti-polyelectrolyte effect, and bacteria need to pass through the dense water layer, which can highly efficiently resist the adhesion of bacteria and most bacteria can be washed away by waves in the real marine environment. On the contrary, the NR-PAM hydrogel without an anti-polyelectrolyte effect cannot form a dense water film in seawater; therefore, it just showed relatively weak anti-bacterial properties ( Figure 5 ). The anti-algae and anti-protein performance of the NR-PZW hydrogels was also researched within the natural marine environment ( Figure 6 ). Microalgae is one of the main components of microbial mucosa formed in the second stage of marine biological fouling, so the anti-algae adhesion performance of hydrogels is a necessary index to investigate. In this work, chlorella incubated in a seawater medium was selected to test the anti-algal adhesion performance of the NR-PZW hydrogel. The hydrogel samples were taken out after being immersed in chlorella (it was selected as a representative of algae) solution for 3 days. The surface of the samples was washed with seawater 3 times, and then the adhesion of chlorella was observed under a microscope ( Figure 6 a). Microscopic photos showed that the NR-PAM hydrogel as the control group had serious algal adhesion, and the coverage ratio of chlorella on its surface was 7.7%, calculated by ImageJ (version 1.8.0) software; however, the surface of the NR-PZW hydrogel only had a small number of chlorella, and the corresponding coverage ratio was only 0.9% ( Figure 6 b). The anti-alga adhesion performance of the NR-PZW hydrogel was 88.3% higher than that of the control group, indicating that the dense hydration layer formed by the hydration of charged groups and water molecules on the PZW network can effectively resist the adhesion of microalgae. Protein is one of the most important nutrients in the ocean for bacteria and microalgae, which can adhere to the surface of materials in a very short time to form a nutrient base film, which is the basic condition for the occurrence of marine biological fouling. By inhibiting the adhesion of organic matter, the development of subsequent fouling of materials can be effectively slowed down and the service life can be extended. In this work, fluorescein isothiocyanate (FITC)-labeled bovine serum albumin (BSA) was selected as a representative to test the anti-protein adhesion performance of NR-PZW hydrogel. The anti-adhesion of the NR-PZW hydrogel resisting on the BSA was evaluated by fluorescence intensity observed on the surface of the hydrogel under a fluorescence microscope, within a BSA dispersion for 24 h. As shown in Figure 6 c, NR-PAM hydrogel as the control group showed bright green strong fluorescence, indicating the presence of a large amount of BSA. In contrast, very weak fluorescence on the surface of NR-PZW hydrogels indicated very little BSA adherence. The NR-PZW hydrogel wrapped in a dense hydration layer can effectively resist protein adhesion, and its excellent marine anti-biofouling performance was proved by combining the previous anti-bacterial adhesion and anti-algal adhesion results. 2.5. Anti-Biofouling Properties of NR-PZW Hydrogel in Real Marine Environment Anti-biofouling materials face more complex adhesion types, gnawing by large organisms, and wave impact in the real marine environment, and these effects are difficult to simulate in the laboratory. In order to test the anti-biofouling properties of NR-PZW hydrogel in practical applications, a real marine test was conducted in the sea area near Haikou for 3 months. A total of four groups of samples of PZW hydrogel, NR, NR-PAM hydrogel, and NR-PZW hydrogel were used for real marine experiments. Except for PZW hydrogel, the sampling frequency was once every 10 days, and the sampling frequency of other samples was once every month. As can be seen from Figure 7 a, NR was attached with a large amount of dirt and algae thriving on its surface after 3 months of marine testing. The smeared area was calculated by ImageJ software as high as 50.7% ( Figure 7 b). This was due to the frost-spraying effect of NR after prolonged immersion in seawater, where hydrophilic substances such as proteins migrated to the surface to form a nutrient-rich thin layer, causing NR, which already lacked anti-biofouling properties, to attract microbial aggregation and growth. As a blank sample, NR-PAM hydrogel was also very dirty and visible to the naked eye with algae and large colonies, with a stain area of 22.2%, indicating that the anti-biofouling performance of ordinary hydrogels in the ocean was limited. In contrast, the surface of NR-PZW hydrogel had only a few slight discoloration areas, and the dirt area was only 0.4%, which proved that the dense hydration layer on its surface could effectively resist the adhesion of various organisms and limited the occurrence of subsequent biofouling. In addition, the NR-PZW hydrogel remained intact after three months of sea shock, and its mechanical properties only slightly decreased from 0.52 MPa to 0.46 MPa ( Figure 7 c). On the contrary, although PZW hydrogel had excellent anti-biofouling properties, its weak mechanical properties made it break in a short time and disappear within 1 month ( Figure S18 ). As shown in the comprehensive performance comparison diagram ( Figure 7 d), the NR-PZW hydrogel can integrate the highly enhanced mechanical properties of NR and the excellent anti-biofouling performance of the PZW hydrogel together. Therefore, this NR-PZW hydrogel can provide excellent anti-biofouling behavior for relatively long-term usage in real marine environments, which is outstanding among existing hydrogel-based anti-biofouling materials ( Table 1 ) [ 10 , 22 , 23 , 29 , 36 , 37 , 38 , 39 , 40 , 41 ]."
} | 7,422 |
39730367 | PMC11680704 | pmc | 9,581 | {
"abstract": "Spiking Neural Networks (SNNs) stand as the third generation of Artificial Neural Networks (ANNs), mirroring the functionality of the mammalian brain more closely than their predecessors. Their computational units, spiking neurons, characterized by Ordinary Differential Equations (ODEs), allow for dynamic system representation, with spikes serving as the medium for asynchronous communication among neurons. Due to their inherent ability to capture input dynamics, SNNs hold great promise for deep networks in Reinforcement Learning (RL) tasks. Deep RL (DRL), and in particular Proximal Policy Optimization (PPO) has been proven to be valuable for training robots due to the difficulty in creating comprehensive offline datasets that capture all environmental features. DRL combined with SNNs offers a compelling solution for tasks characterized by temporal complexity. In this work, we study the effectiveness of SNNs on DRL tasks leveraging a novel framework we developed for training SNNs with PPO in the Isaac Gym simulator implemented using the skrl library. Thanks to its significantly faster training speed compared to available SNN DRL tools, the framework allowed us to: (i) Perform an effective exploration of SNN configurations for DRL robotic tasks; (ii) Compare SNNs and ANNs for various network configurations such as the number of layers and neurons. Our work demonstrates that in DRL tasks the optimal SNN topology has a lower number of layers than ANN and we highlight how the state-of-art SNN architectures used in complex RL tasks, such as Ant , SNNs have difficulties fully leveraging deeper layers. Finally, we applied the best topology identified thanks to our Isaac Gym-based framework on Ant-v4 benchmark running on MuJoCo simulator, exhibiting a performance improvement by a factor of 4.4 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$\\times$$\\end{document} × over the state-of-art SNN trained on the same task.",
"conclusion": "Conclusions In our study, we developed an SNN framework tailored for Deep Reinforcement Learning (DRL) using skrl and Isaac Gym, significantly reducing training time with respect to available tools. This framework facilitated a comprehensive comparison between ANNs and SNNs across two distinct RL tasks: Cartpole and Ant . Our findings revealed that: i) SNNs perform better with shallow networks compared to ANNs; ii) SNN is able to exploit deeper layers in simpler tasks, such as Cartpole and Multimodal Tracking. However, in more complex tasks, SNNs struggle to train these deeper layers, resulting in inferior performance compared to ANNs. We hypothesize that this behavior is linked to the surrogate gradient function, which approximates the gradient—resulting in greater inaccuracies as the SNN depth increases. This issue, along with coding strategies critical for effectively feeding information to SNNs, will be the focus of future investigation. Additionally, we observed that in simpler tasks such as Cartpole , SNNs achieved higher performance than ANNs with the same number of training parameters. This behavior is likely due to the greater expressiveness of spiking neurons compared to artificial neurons. Finally, leveraging insights gained from these experiments, we applied learned lessons to the Ant-v4 task, surpassing state-of-the-art algorithms by a factor of 4.4 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$\\times$$\\end{document} × .",
"introduction": "Introduction Spiking Neural Networks (SNNs) are often regarded as the third generation of Artificial Neural Networks (ANNs) because their functionality closely resembles that of the mammalian brain compared to previous generations. Also, it has been established that SNNs offer greater expressiveness compared to conventional ANNs 1 . Being their behavior based on a system of Ordinary Differential Equations (ODEs) which depict spiking neurons as dynamic systems, SNNs inherit temporal dimensionality, with neuron dynamics capable of encoding information in membrane potentials. Also, their event-based nature, where spikes serve as the medium for communication between neurons, potentially leads to low inference latency, which is suitable for real-time tasks. These features are particularly appealing in Reinforcement Learning (RL) tasks for robotics, in which an agent has to learn dynamically through direct interaction with their surroundings in real-time 2 , 3 . In this context, SNNs emerge as a promising solution for efficiently tackling such tasks 4 . Common benchmarks employed for this purpose include Cartpole and Ant . In these environments, a neural network learns to control the robot joints to achieve a specific objective. Previous studies focusing on SNN for robotic tasks adopted a conversion of the inputs from the continuum domain to the spiking one 5 , 6 before feeding the network. From this point of view, an alternative approach that we consider in this work is to feed directly the continuous input to the network and let the first layer of the SNN convert it into spikes, thus letting the network learn the spike representation. Within DRL, two primary algorithm families exist: off-policy and on-policy 2 . Off-policy algorithms leverage data collected from previous neural network policies (i.e. representing past experience). In contrast, on-policy algorithms base their learning solely on data generated by the current policy being evaluated 2 . For off-policy algorithms, every sample is randomly taken from a buffer of collected samples, called the experience memory. As a consequence, there is no temporal correlation between consecutive samples and the network should be reset at every sample. On the other side, on-policy algorithms use temporally consecutive samples. For this reason, to leverage SNN time correlation capabilities, with on-policy algorithms the network is not reset between samples. As the field of SNNs in DRL is relatively recent, most of the work done uses off-policy methods not exploiting the full potential of SNNs 7 – 20 . Another relevant aspect to be considered when bringing SNN to DRL is the training method. To circumvent the non-differentiable nature of spikes, conversion-based methods have been proposed, which involve converting pre-trained ANNs into SNNs with identical architectures 21 – 23 . However, these methods typically require a high number of timesteps to attain comparable performance, leading to significant inference latency and energy consumption, making them unattractive for real-time robotic tasks. Conversely, direct training methods employ Surrogate Gradients (SG) and lead to lower latency inference compared to conversion-based methods 24 – 26 . By substituting the all-or-nothing gradients of the spike activity function with various shapes of SG, they allow for gradient backpropagation within a broader range of membrane potentials. At the same time, numerous efforts 25 , 27 – 30 have been undertaken to mitigate the gradient vanishing or explosion problems typical of SNN direct training, to enhance its performance. As a consequence, direct training enables competitive performance under low latency, making them mature to be applied in real-time DRL tasks. As a result, to explore the potential of SNN for DRL, direct SNN training using the on-policy method is desirable. Also, training efficiency is key to performing an effective network exploration and comparing SNN performance with the one achievable with ANN, so as to promote further SNN exploration and optimization. Today, simulator acceleration and multi-environment learning are key enabling technologies for DRL 31 , 32 . However, SNN training and inference in such an efficient framework are missing. In this work, we filled this gap by developing a framework called SpikeGym 33 for training SNNs using the Proximal Policy Optimization (PPO) 34 algorithm, implemented with Isaac Gym 32 on top of skrl 35 . This framework takes full advantage of the Isaac Gym features, namely full GPU pipeline and multi-environment training. Thanks to our framework, training an SNN agent in Ant (Isaac Gym) takes about 7 min instead of 3 h and 20 min as required by State-of-the-Art (SoA) implementation based on Ant-v4 (MuJoCo), enabling a suitable exploration of the behavior of the agent. To replicate just one specific use-case (Ant) from the two analyzed in this study, which involved 2400 Deep Reinforcement Learning (DRL) training sessions (30 different deep neural network configurations, each replicated 20 times, across two phases of training and inference, and using both ANN and SNN) with state-of-the-art technologies 5 it would have required 333 days of simulation time on a GPU-accelerated workstation. However, the proposed framework reduced this time to just 11 and a half days, rendering the study more manageable and facilitating replication. Further, by applying SNNs to both Cartpole and Ant tasks in the Isaac Gym simulator, we demonstrate that SNNs exhibit greater robustness than ANNs in Cartpole task. However, deeper SNN architectures (with more than one layer) experience performance drops of up to 5 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$\\times$$\\end{document} × when using more hidden layers (2–4) compared to a single layer. This limitation becomes particularly apparent in the Ant task, where ANNs outperform SNNs due to the task demand for deeper layer utilization, which SNNs struggle to leverage effectively even using a state-of-art direct training method 36 . To assess the generality of this behavior beyond DRL tasks, we explored the application of SNNs in a self-supervised learning task involving the reconstruction of multimodal signals. Our findings align with observations from DRL tasks, indicating that SNNs face difficulties with deeper layer utilization in complex scenarios. To the best of our knowledge, this is the first study directly comparing the performance of directly trained SNN and ANN in DRL, showing the inefficiencies in training deep SNN networks. Finally, we apply the learning rules found in our experiments to the Ant-v4 benchmark in MuJoCo 37 showing an increase in performance of 4.4 \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$$\\times$$\\end{document} × compared to the state-of-the-art spiking network proposed for the same task 5 . In summary, our work includes the following key contributions: We introduce a new framework called SpikeGym, which is built on IsaacGym and skrl, enabling the training of SNNs using the state-of-the-art reinforcement learning algorithm PPO in minutes rather than hours by state-of-the-art frameworks. We conduct a performance comparison between several network configurations trained with the PPO algorithm in two well-known environments: Cartpole and Ant . Our results demonstrate that SNN can perform complex tasks but with lower performance than ANN. The reason is that they struggle to fully leverage the advantages of deeper layers. To assess the generality of the effect of the number of layers on the SNN performance, we further investigate this behavior in a multimodal tracking task, where the same pattern persists. Finally, we apply our insights to the Ant-v4 environment and compare our results with state-of-the-art, demonstrating that our framework and approach outperform the current state-of-the-art by a factor of 4.4. We believe that our study and the framework that we make openly available to the community will stimulate researchers to further explore and address the current advantages and limits of SNNs in DRL tasks, thereby expanding the application potential of neuromorphic computing in real-world tasks."
} | 3,129 |
21847112 | PMC3265365 | pmc | 9,583 | {
"abstract": "Nanoparticles at fluid interfaces are central to a rapidly increasing range of cutting-edge applications, including drug delivery, uptake through biological membranes, emulsion stabilization and the fabrication of nanocomposites. Understanding nanoscale wetting is a challenging issue, still unresolved for individual nanoparticles, and is essential in designing nanoparticle-building blocks with controlled surface properties. The core information about the structural and thermodynamic properties of particles at fluid interfaces is enclosed in the three-phase contact angle θ . Here we present a novel in situ method, on the basis of freeze-fracture shadow-casting cryo-scanning electron microscopy, that allows the measurement of contact angles of individual nanoparticles with 10 nm diameter, and thus greatly surpasses the current state of the art. We study hydrophilic and hydrophobic, organic and inorganic nanoparticles, demonstrating general applicability to systems of fundamental and applied nanotechnological interest. Significant heterogeneity in the wetting of nanoparticles is observed.",
"discussion": "Discussion The key result that can be extracted from the data presented in Figures 5 and 6 and summarized in Table 1 is that the spread in the contact angle distributions is greater than the error in the single-particle measurement, highlighting the fact that the wide distributions are not due limitations in the resolution of the method for the contact angle measurement. Large-scale heterogeneity of the interface could be the cause of the broad distributions, but contact angles measured on different samples as well as on the same interface on mm 2 scales showed no systematic variations. Moreover, the inset to Figure 4c highlights the fact that two neighbouring 200 nm amidine latex particles of identical size show significantly different contact angles, underlining the fact that the heterogeneity of the wetting properties probably derives from the distribution of surface properties of the nanoscale materials themselves. Broadening of θ distributions at the nanoscale can also stem from the presence of Brownian motion. NPs are trapped at the interface with an energy Δ E , as discussed previously, but, for small objects, thermal motion can be significant enough to displace them from their individual equilibrium position relative to the interface. The FreSCa method takes a snapshot of the particles at the interface and therefore intrinsically a distribution of heights (and thus of contact angles) is measured. As reported in the literature, the shape of the potential well around the equilibrium position at the interface is quadratic with the displacement 11 14 15 , and the probability distribution of finding a particle at position z from the bottom of the well will be proportional to . Using the full expression for Δ E(z) and neglecting line tension contributions 11 , the width of the z -displacement distribution is , where . For the smallest investigated particles with r =5 nm particle at the water / n-decane interface, this corresponds to a width of the displacement distribution of 2.6% of the particle radius that translates into a±3° error in measuring θ for a NP with an equilibrium contact angle of 150°, which is significantly smaller than the observed distribution width. For larger particles, this effect becomes rapidly negligible. Polydispersity for small nanoparticles can also introduce further spreading of the contact angle distributions due to line tension effects. In fact, line tension contributions to θ depend on particle size and become increasingly more relevant at the nanoscale 15 . The unprecedented resolution of our method allows for measuring directly the line tension of individual NPs. Figure 6b shows the values of the line tension τ vs. r for the nominally 20 nm particles at the water / n-decane interface; our data indeed follow a linear scaling with the size as from theoretical predictions 15 and the values we measured are consistent with previous findings 11 (more details in the Supplementary Note 4 ). Summarizing, broad θ distributions are found for the same NP material, with increasing spread of the contact angle values for smaller NPs. For the reasons discussed above, this reflects the fact that the wetting behaviour of colloidal particles becomes heterogeneous at the micro- and nano-scale where the details of surface properties (for example, chemistry, shape and topography) make relatively larger contributions to the average bulk material properties. For instance, as recently demonstrated, interparticle interactions at liquid interfaces strongly depend on the individual particles chosen for the measurement 34 . This demonstrates the need for true single-particle characterization that has already been undertaken by numerical simulations 35 36 37 38 but up to now has been missing experimentally. In conclusion, we have described a new characterization method on the basis of freeze-fracture and shadow-casting cryo-SEM that enabled us to measure directly for the first time three-phase contact angles in-situ at diverse water –oil interfaces for individual nanoparticles as small as 10 nm. Our technique places itself at the forefront of the currently available characterization methods in terms of capability, resolution and flexibility. Key fundamental questions, including line tension, surface roughness and homogeneity of surface chemistry, demand characterization of wetting properties on the single-NP level that can be met by this technique. Finally, we foresee the extension of our approach to the investigation of interfacial assembly and wetting properties of many more systems, including anisotropic particles, for which orientation at the interface can be directly measured with our technique, and composite, core-shell nanoparticles. This will help improve the current basic understanding and applications of NPs for which interfacial control is essential."
} | 1,500 |
39965638 | PMC11835496 | pmc | 9,584 | {
"abstract": "Efficient energy harvesting, conversion and recycling technologies are crucial for addressing the challenges faced by modern societies and the global economy. The potential of harnessing mid-infrared (mid-IR) thermal radiation as a pervasive and readily available energy source has so far not been fully exploited, particularly through bioinspiration. In this article, by reviewing existing photon-based strategies and the efficiency of natural systems in harnessing light and thermal radiation, I highlight the promising role of bioinspiration in enhancing energy capture, conversion and recycling. Natural photonic structures found in various organisms, including insects, birds and plants, exhibit sophisticated optical properties that can be leveraged for energy-efficient applications. These developments pave the way for future research and innovation in bioinspired energy solutions. Ultimately, they contribute to the pursuit of a sustainable and environmentally conscious future by harnessing the beauty of nature’s designs to meet humankind’s energy needs.",
"conclusion": "4 . Conclusions Photonic structures occurring in the integuments of living organisms such as arthropods, birds and plants are very sophisticated optical devices that give rise to various striking optical effects, including UV, visible and IR radiation management and absorption enhancement. They occur in biological tissues encompassing butterfly wings, beetle elytra, bird feathers, spider cuticle, viper skin, as well as plant leaves and petals. These phenomena are often crucial for the survival of animal and plant species. This review article also showcased the promising potential of bioinspiration in the field of energy capture and conversion. Exploiting light trapping, impedance matching or antireflection observed in natural structures is indeed highly interesting, given the development of bioinspired energy-efficient applications such as PV and TPV cells, TEG, artificial photosynthesis and photocatalysis. With the optimization of the efficiency of such applications, these advances inspire future research and innovation in the field of bioinspired energy solutions. Ultimately, this research paves the way for a more sustainable and environmentally conscious future by harnessing the beauty of nature’s designs to meet humanity’s energy needs.",
"introduction": "1 . Introduction Addressing the significant challenges faced by modern global society and the world economy, the advancement of efficient energy harvesting and recycling technologies [ 1 – 3 ] stands as a prominent area of research on a global scale. Mid-infrared (mid-IR) thermal radiation, namely, with a wavelength ranging from 3 to 8 μm, represents a pervasive and readily available energy source. This is not only due to the long illumination of some parts of the Earth by the Sun but also because many machinery, engines and industrial processes dissipate energy in the form of heat radiation, distinct from thermal conduction or convection mechanisms. While the primary energy source may vary in its environmental impact, the recycling of this ‘wasted’ energy presents a sustainable approach to converting radiative heat losses into diverse forms of energy. Numerous mechanical components found in machinery, engines, industrial processes and even household systems generate mid-IR thermal emissions at moderately elevated temperatures, typically ranging from 150°C to 950°C. These emissions are an intrinsic by-product of the regular functioning of these components and constitute an unavoidable energy loss. The prospect of harnessing this radiative heat loss is compelling, as it offers the opportunity to transform it into electrical power, effectively enabling devices to utilize their own recycled radiative heat loss for enhanced functionality. Photon-based strategies have already played a crucial role in harnessing solar energy, enhancing the performance of energy conversion devices [ 4 – 12 ]. For instance, devices designed for solar light trapping have effectively increased the efficiency of photovoltaic (PV) cells and thermal photovoltaic (TPV) cells. Similar photonic devices are instrumental in augmenting the efficiency of solar thermal panels, or in energy harvesting for thermoelectric generators (TEG), artificial photosynthesis, and photocatalysis. In nature, numerous biological organisms have developed highly efficient mechanisms to harness thermal radiation, a crucial adaptation for their survival. Over millions of years of evolution, these natural systems have honed specialized characteristics to maximize their radiation harvesting abilities [ 13 , 14 ]. Consequently, certain structures within their integuments have become increasingly inspiring for the development, design and production of energy-efficient materials [ 14 – 17 ]. Bioinspiration emerges as a powerful and promising strategy in this context. Natural photonic structures found in various animals, including insects, birds and fish, are examples of effective thermal radiation collectors [ 15 , 17 , 18 ]. In addition, this type of structure exhibits a diverse array of properties, such as structural colours (resulting from light interference in nanostructures) [ 13 , 14 , 19 , 20 ], antireflection features [ 16 , 18 , 21 ], thermoregulation mechanisms [ 22 – 26 ], light-trapping capabilities [ 27 – 30 ] and enhanced light-extraction methods [ 31 ]. These properties emerge from the interaction between radiation and structures composed of biopolymers such as chitin, keratin, collagen or cellulose, sometimes in combination with pores. The existence of these naturally occurring radiation management systems challenges the human imagination. While human beings have access to a wide range of materials, human designs sometimes fall short in complexity compared with these remarkable natural structures. Identifying and comprehending these natural photonic devices not only expands human understanding but also empowers engineers and materials scientists to conceptualize new ideas and explore potential technological applications through bioinspired principles [ 14 – 17 ]. These exciting possibilities have captivated the attention of researchers worldwide. Despite the development of artificial intelligence, bioinspiration remains a guiding force in the quest for novel technological applications. The convergence of both approaches holds promise for unprecedented advancements in this field. In this article, I first review previously investigated cases of photonic structures enhancing electromagnetic-wave absorption (also known as structural absorption) in natural organisms across the ultraviolet (UV), visible and infrared (IR) range. This is because the dimensions of a visible light absorber occurring in nature may be adjusted to another range such as IR through a bioinspiration approach due to the scalability of Maxwell’s equations. 1 Finally, I review examples of bioinspired IR absorbers from the literature."
} | 1,741 |
35573202 | PMC9097702 | pmc | 9,585 | {
"abstract": "Summary The DNA origami technique is used to construct custom-shaped nanostructures that can be used as components of two-dimensional crystalline structures with user-defined structural patterns. Here, we designed an Mg 2+ -responsive hexagonal 3D DNA origami block with self-shape-complementary ruggedness on the sides. Hexagonal DNA origami blocks were electrostatically adsorbed onto a fluidic lipid bilayer membrane surface to ensure lateral diffusion. A subsequent increase in the Mg 2+ concentration in the surrounding environment induced the self-assembly of the origami blocks into lattices with prescribed geometries based on a self-complementary shape fit. High-speed atomic force microscopy (HS-AFM) images revealed dynamic events involved in the self-assembly process, including edge reorganization, defect splitting, diffusion, and filling, which provide a glimpse into how the lattice structures are self-improved.",
"conclusion": "Conclusion Herein, we designed an Mg 2+ -responsive hexagonal 3D DNA origami block with self-complementary features and demonstrated its ability to form LB-SAS in lattices with prescribed structural patterns. The primary advantage of the self-shape-complementary design employed here is that it can inactivate connections at arbitrary sides by simply extending the polyT tails from blunt ends. This enables the construction of lattices with different patterns from a single 3D DNA origami block design. To satisfy the conditions that allow both the surface diffusion of the origami blocks and their Mg 2+ -induced assembly, we employed an artificial lipid bilayer membrane as a soft and flat surface to serve as an alternative to the naked mica surface. The lipid bilayer-adsorbed origami blocks retained their surface mobility and self-assembled into lattices with the desired geometries, in a process involving edge reorganization, defect splitting, diffusion, and filling, in response to the changes in Mg 2+ concentration in the surrounding environment. In our protocol for DNA origami preparation, four times the excess amount of staple strands are mixed against a scaffold DNA (p8064, 0.75 USD per pmol). Because each staple DNA costs 0.12 USD per base at a molar yield of 10 nmol and 7596 bases of the scaffold were folded into a designed shape with staples, total cost of the origami per pmol is 1.1 USD (i.e., 1.8 × 10 −12 USD per origami). Given that the area occupied by a single hexagonal DNA block is about 1.5 × 10 −11 cm 2 (1500 nm 2 ), the cost of the DNA required to assemble the honeycomb lattice is estimated to be 0.12 USD per cm 2 , which is comparable to the cost required for the mica-assisted self-assembly of conventional 2D DNA origami lattices ( Xin et al., 2021 ). This cost can be lowered, owing to the development of methods for the mass production of DNA strands ( Praetorius et al., 2017 ) and the further improvement of DNA synthesis technology. Lipid membrane-interacting DNA nanostructures have recently attracted great attention in the fields of synthetic biology ( Czogalla et al., 2016 ) and molecular robotics ( Hagiya et al., 2014 ) for their potential to mimic the structures and functions of membrane proteins ( Burns et al., 2013 , 2016 ; Krishnan et al., 2016 ; Langecker et al., 2012 ; Thomsen et al., 2019 ), membrane-binding proteins ( Czogalla et al., 2015 ; Franquelim et al., 2018 ; Grome et al., 2018 ; Khmelinskaia et al., 2018 ), and membrane-cytoskeleton networks ( Kocabey et al., 2015 ; Kurokawa et al., 2017 ). We believe that our approach to assembling lattices with 3D features matches with this direction, not just for its application as a scaffold for the periodic placement of protein molecules and metal nanoparticles. Other possible applications may result from the use of nanospaces compartmentalized by origami blocks. It has been implicated that reaction kinetics of various biological molecules in the nanoconfinement spaces differ from those in bulk solutions ( Kuchler et al., 2016 ). DNA origami nanotechnology has enabled the construction of a designed 3D nanospace, in which a single to several molecules of interest are enclosed and applied to reveal the effects of nanoconfinement on enzymatic reactions ( Grossi et al., 2017 , Zhao et al., 2016 ) and the structural transitions of DNA ( Jonchhe et al., 2018 , Shrestha et al., 2017 ). The porous crystalline DNA origami structures can be regarded as arrays of such nanoconfinement spaces, which provide a tool for multiplexed single-molecule analysis ( Sakamoto et al., 2020 ) that will enable the simultaneous monitoring of reactions occurring in individual nanopores.",
"introduction": "Introduction Molecular self-assembly has attracted considerable attention as a method for the construction of novel supramolecular architectures. The scaffolded DNA origami method has enabled the one-pot preparation of almost arbitrarily shaped two-dimensional (2D) ( Rothemund, 2006 ) and three-dimensional (3D) DNA nanostructures ( Andersen et al., 2009 ; Dietz et al., 2009 ; Douglas et al., 2009a ; Han et al., 2011 ; Kuzuya and Komiyama, 2009 ), which can be further used as components of higher-order architectures that self-assemble based on intermolecular interactions between DNA ends, such as sticky-ended cohesion ( Liu et al., 2011 ; Rajendran et al., 2011 ; Tikhomirov et al., 2017 ; Zhang et al., 2018 ; Zhao et al., 2011 ) and blunt-ended stacking ( Wagenbauer et al., 2017 ; Woo and Rothemund, 2011 ). Among the various assembly approaches, mica-surface-assisted self-assembly (mica-SAS) is a promising method to obtain two-dimensionally ordered arrays of 2D DNA origamis ( Aghebat Rafat et al., 2014 ; Kielar et al., 2018 ; Ramakrishnan et al., 2016 ; Woo and Rothemund, 2014 ; Xin et al., 2021 ). One key to the success of mica-SAS is the setting of appropriate adsorption conditions (i.e., not too weak and not too strong) to ensure the surface mobility of the component DNA origami on the mica substrate. In general, DNA origami structures are prepared in a buffer solution containing 10–20 mM Mg 2+ in which they are strongly adsorbed onto mica surfaces and do not exhibit two-dimensional diffusion. Therefore, for the mica-SAS of DNA origami crystals, 200–700 mM NaCl was added to weaken the electrostatic interaction between DNA and the mica surface ( Aghebat Rafat et al., 2014 ; Woo and Rothemund, 2014 ). This strategy has been well established for the construction of monolayered crystalline structures from 2D DNA origami, but cannot be simply applied to those from 3D DNA origami, especially those that require high Mg 2+ concentrations to connect with each other. Representative examples of such structures include the 3D DNA origami blocks with shape-complementary recession-protrusion patterns proposed by Gerling et al. (2015) . These structures can be multimerized based on shape-fitting upon increasing the Mg 2+ concentration, to neutralize the electrostatic repulsion between the shape-complementary interfaces. However, this type of assembly mechanism is unsuitable for conventional mica-SAS as increasing Mg 2+ also strengthens the adsorption of DNA origami blocks onto the mica surface, making them immobile. Increasing the Na + concentration is also unfeasible in this case due to its inhibitory effect on the multivalent cation-induced DNA-DNA interactions ( Hibino et al., 2006 ). To reconcile these conflicts, we employed the lipid bilayer-surface-assisted self-assembly (LB-SAS) ( Avakyan et al., 2017 ; Kempter et al., 2019 ; Kocabey et al., 2015 ; Suzuki et al., 2015 ) to assemble Mg 2+ -responsive 3D DNA origami blocks into monolayered crystalline structures. Artificial lipid bilayers, especially glass- or mica-supported lipid bilayers (SLBs), provide a flat but dynamic surface with the desired physicochemical properties (such as fluidity and surface charge) by tuning their lipid compositions, which enables us to realize conditions to ensure the lateral mobility of assembly components. In this study, we designed a 3D DNA origami block whose sides exhibited self-shape-complementary ruggedness. The electrostatic adsorption of DNA origami blocks onto a fluidic lipid bilayer membrane allows their diffusion on the surface. The subsequent increase in Mg 2+ promotes their self-assembly into a lattice with a designed geometry through homomultimerization based on a self-complementary shape-fit. The utilization of block-shaped components with prescribed thickness expands the constructible structures, whose thickness has been limited to ∼2 nm in the conventional assemblies of sheet-like 2D DNA origami structures, via a surface-assisted approach. Our study provides a general approach for the construction of custom DNA origami lattices with the desired assembly patterns and nanoporous structures that will serve not only as a nano-micro scaffold for molecular patterning but also as an array of nanocompartmentalized spaces.",
"discussion": "Results and discussions Design of a hexagonal DNA origami block As the self-assembly component, we designed a hexagonal 3D DNA origami using caDNAno2 ( Figure 1 A, see also Figure S1 ) ( Douglas et al., 2009b ). The sides of the hexagonal 3D origami (3D-Hx) exhibited shape-complementary ruggedness ( Video S1 ), which imposes self-complementary shape-fitting to allow assembly into a homomultimeric two-dimensional lattice ( Gerling et al., 2015 ). The 3D-Hx was prepared by annealing a mixture of 8064 nt scaffold DNA (p8064) and staple strands in folding buffer (5 mM Tris-HCl [pH 8.0], 15 mM MgCl 2 , and 5 mM EDTA), and then purified using a glycerol gradient ( Lin et al., 2013 ) ( STAR Methods , Figure S2 ). Atomic force microscopy (AFM) imaging ( Figure 1 B) revealed the monodispersed nature of the purified 3D DNA origami blocks, with a size of 53 ± 4 nm (mean ± S.D., N = 85), which is in good agreement with the expected size. Figure 1 Three-dimensional hexagonal DNA origami blocks (3D-Hx) with self-complementary ruggedness (protrusion-recession patterns) on the sides (A) Schematic illustration of 3D-Hx. Adjacent sides of 3D-Hx showing upside-down patterns of the ruggedness. See also Video S1 for the structural design. (B) Atomic force microscopy (AFM) image of the purified 3D-Hx. Scale bars: 200 nm. \n Video S1. Computer graphic model of a hexagonal 3D DNA origami block (3D-Hx) The adjacent sides of 3D-Hx show upside-down patterns of shape-complementary ruggedness related to Figure 1. \n 2D self-assembly on the mica-supported lipid bilayer membrane The purified 3D-Hx was self-assembled into honeycomb lattices on a mica-supported 1,2-dioleoyl- sn -glycero-3-phosphocholine (DOPC) bilayer. In this experiment, 3D-Hx was first adsorbed onto the DOPC bilayer in folding buffer (5 mM Tris-HCl [pH 8.0], 15 mM MgCl 2 , and 1 mM EDTA). The concentration of Mg 2+ was then increased to reduce electrostatic repulsions among the negatively charged DNA origami interfaces, leading to shape-complementary fittings among the ruggedness of the 3D-Hxs, which were further stabilized via blunt-ended interactions ( Figure 2 A). Although several defects remained in the assembled lattices, high-magnification AFM images revealed the periodic features of the honeycomb lattice, comprising internal cavities with a depth of 11 ± 1 nm and a distance between opposite sides of 27 ± 2 nm (N = 133) ( Figures 2 B and 2C, see also Figure S3 ). Figure 2 Lipid bilayer-assisted self-assembly of 3D-Hx into a honeycomb lattice (A) Schematic of the experimental procedure. 3D-Hx is deposited onto the mica-supported lipid bilayer membrane (mica-SLB). The connection among 3D-Hx blocks are then induced by increasing Mg 2+ concentration of the surrounding environment (buffer solution). (B) AFM image of the honeycomb lattice assembled on the mica-SLB. Scale bars: 200 nm. (C) Cross-sectional profile along the line a-b in (B). Errors represent standard deviations. Dynamic events involved in lattice formation Dynamic events, such as boundary reorganization, defect diffusion, and defect filling, occurring on an SLB were successfully monitored using time-lapse AFM imaging. Figure 3 A shows the representative sequential images obtained after increasing the [Mg 2+ ] to 50 mM ( Video S2 ). The dynamic association and dissociation of 3D-Hx monomers and oligomers frequently occur around the edges of the lattices (See also another example in Video S3 ). It should be noted that the area occupied by 3D-Hxs in the image area gradually increased with fluctuations in its value ( Figure S4 ), indicating that diffusing 3D-Hxs entered and exited the area, some of which were incorporated into the already existing lattices. In particular, the main lattice island in the scanning area ( Figures 3 A, 0 s) increased with the incorporation of 3D-Hxs. Intriguingly, 3D-Hx monomers and oligomers dissociated from adjacent lattices were also incorporated into this lattice. Figure 3 Dynamic events involved in lattice formation (A) Lattice growth monitored using high-speed AFM (HS-AFM). Region surrounded by dashed yellow line in the initial frame exhibited dynamic growth. Scale bar: 200 nm. Images were recorded at a scan rate of 0.5 frames per second (fps). For the complete movie, see Video S2 . (B) Defect splitting and defect diffusion monitored using HS-AFM. The defect indicated by yellow circle (0 s) split into a single defect (orange dashed circle, 75 s) and a twin defect (light blue dashed circle, 75 s). Scale bar: 100 nm. Images were recorded at a scan rate of 0.2 fps. For the complete movie, see Video S4 . (C) Defect filling was monitored using HS-AFM. The first and second monomers jumped into the defect and are indicated by dashed yellow and red circles, respectively. Scale bar: 100 nm. Images were recorded at a scan rate of 0.5 fps. For the complete movie, see Video S5 . \n Video S2. Assembly processes monitored using high-speed atomic force microscopy Image size: 800 × 800 nm. The original scan rate was 0.5 fps. The movie is played at ten times the original speed related to Figure 3. \n \n Video S3. Assembly processes monitored using high-speed atomic force microscopy Image size: 425 × 425 nm. The original scan rate was 0.2 fps. The movie is played at ten times the original speed related to Figure 3. \n Dissociation and re-association of the components occur not only at the edges of the lattices but also at the defects in the assembled lattice, resulting in defect diffusion. In the example shown in Figure 3 B (see also Video S4 ), a defect equivalent of triangularly arranged 3 hexagonal blocks was observed in the initial frame. This defect split into a single defect and a twin defect, the latter of which moved from the left to the right of the image area, while the single defect remained at the same position. Figure 3 C depicts another dynamic event, in which a large defect was reduced by incorporating 3D-Hx monomers. In the initial frame of this example ( Figure 3 C, see also Video S5 ), a large defect equivalent to seven hexagonal blocks was observed. The first monomer, which remained unadsorbed on the SLB surface, jumped into this defect, attempting to make a connection with the upper edge of the defect, but soon moved to the lower edge to make a tentative connection, indicating that the connection at the two sides was not sufficiently strong for the incorporation of the monomer into the already assembled lattice at this Mg 2+ concentration (50 mM). After the arrival of the second monomer, the first monomer again moved to the upper side and was incorporated into the lattice with the second monomer. Both the first and second monomers made connections on three of the six sides in the last frame. These observations provide a glimpse into how defects arising in the assembly process become smaller. \n Video S4. Defect splitting and diffusion captured using high-speed atomic force microscopy Image size: 400 × 400 nm. The original scan rate was 0.2 fps. The movie is played at twenty times the original speed related to Figure 3. \n \n Video S5. Defect-filling captured using high-speed atomic force microscopy Image size: 400 × 400 nm. The original scan rate was 0.5 fps. The movie is played at ten times the original speed related to Figure 3. \n Self-assembly into depleted lattice and its surface modification One of the unique features of this system is that we can inactivate the designated sides by disabling blunt-ended interactions through the addition of polyT tails at the ends of the protruding parts of the ruggedness. This simple customizability enabled us to realize different assembly modes of 3D-Hx. Here, ruggedness at the three symmetrical sides of the 3D-Hx was inactivated by staple extension with polyT tails ( Figure 4 A S5 ). Because the adjacent sides of our 3D-Hx origami showed upside-down patterns of ruggedness ( Video S1 , Figure S1 ), self-assembly of this 3-symmetric-side-inactivated hexagonal block (3D-Hx-135) resulted in a “depleted” honeycomb lattice in which face-down and face-up blocks are alternately laid out ( Figure 4 B). AFM images of the assembled lattices revealed structural features in which pores of two different sizes were arranged ( Figure 4 C and S6 ). To address the facing orientation of the 3D-Hx-135 blocks in the depleted honeycomb lattice, we prepared a functionalized block, with one of the faces having three biotin moieties, and performed post-self-assembly modification ( Figure 4 D). The depleted lattice was first prepared on the bilayer from one-sided biotinylated blocks. After lattice formation was confirmed, streptavidin was loaded onto the same sample. With this post-assembly treatment, only the blocks facing upwards could bind streptavidin. Figure 4 E and 4F shows a representative AFM image after the post-assembly modification, where every other block in the lattice was decorated, demonstrating facing-up (modified) blocks and facing-down (unmodified) blocks alternately arranged in the depleted lattice, as expected. Figure 4 Lipid bilayer-assisted self-assembly of 3D-Hx into a depleted honeycomb lattice (A) Schematic of the variation of the 3D-Hx whose ruggedness at three symmetric sides are inactivated. Active ruggedness and inactivated ruggedness are colored with orange and light orange, respectively. For convenience, the obverse and reverse of the 3D-Hx are indicated by different colors. (B) Schematic illustration of the depleted honeycomb lattice. (C) AFM image of the depleted honeycomb lattice assembled on the mica-SLB. Scale bars: 200 nm. (D) Schematic of the experimental procedure of the post-assembly modification with streptavidin molecules. (E) AFM images of the streptavidin-bound depleted honeycomb lattice. Scale bar: 100 nm. (F) Surface plot of the streptavidin-bound depleted honeycomb lattice. Conclusion Herein, we designed an Mg 2+ -responsive hexagonal 3D DNA origami block with self-complementary features and demonstrated its ability to form LB-SAS in lattices with prescribed structural patterns. The primary advantage of the self-shape-complementary design employed here is that it can inactivate connections at arbitrary sides by simply extending the polyT tails from blunt ends. This enables the construction of lattices with different patterns from a single 3D DNA origami block design. To satisfy the conditions that allow both the surface diffusion of the origami blocks and their Mg 2+ -induced assembly, we employed an artificial lipid bilayer membrane as a soft and flat surface to serve as an alternative to the naked mica surface. The lipid bilayer-adsorbed origami blocks retained their surface mobility and self-assembled into lattices with the desired geometries, in a process involving edge reorganization, defect splitting, diffusion, and filling, in response to the changes in Mg 2+ concentration in the surrounding environment. In our protocol for DNA origami preparation, four times the excess amount of staple strands are mixed against a scaffold DNA (p8064, 0.75 USD per pmol). Because each staple DNA costs 0.12 USD per base at a molar yield of 10 nmol and 7596 bases of the scaffold were folded into a designed shape with staples, total cost of the origami per pmol is 1.1 USD (i.e., 1.8 × 10 −12 USD per origami). Given that the area occupied by a single hexagonal DNA block is about 1.5 × 10 −11 cm 2 (1500 nm 2 ), the cost of the DNA required to assemble the honeycomb lattice is estimated to be 0.12 USD per cm 2 , which is comparable to the cost required for the mica-assisted self-assembly of conventional 2D DNA origami lattices ( Xin et al., 2021 ). This cost can be lowered, owing to the development of methods for the mass production of DNA strands ( Praetorius et al., 2017 ) and the further improvement of DNA synthesis technology. Lipid membrane-interacting DNA nanostructures have recently attracted great attention in the fields of synthetic biology ( Czogalla et al., 2016 ) and molecular robotics ( Hagiya et al., 2014 ) for their potential to mimic the structures and functions of membrane proteins ( Burns et al., 2013 , 2016 ; Krishnan et al., 2016 ; Langecker et al., 2012 ; Thomsen et al., 2019 ), membrane-binding proteins ( Czogalla et al., 2015 ; Franquelim et al., 2018 ; Grome et al., 2018 ; Khmelinskaia et al., 2018 ), and membrane-cytoskeleton networks ( Kocabey et al., 2015 ; Kurokawa et al., 2017 ). We believe that our approach to assembling lattices with 3D features matches with this direction, not just for its application as a scaffold for the periodic placement of protein molecules and metal nanoparticles. Other possible applications may result from the use of nanospaces compartmentalized by origami blocks. It has been implicated that reaction kinetics of various biological molecules in the nanoconfinement spaces differ from those in bulk solutions ( Kuchler et al., 2016 ). DNA origami nanotechnology has enabled the construction of a designed 3D nanospace, in which a single to several molecules of interest are enclosed and applied to reveal the effects of nanoconfinement on enzymatic reactions ( Grossi et al., 2017 , Zhao et al., 2016 ) and the structural transitions of DNA ( Jonchhe et al., 2018 , Shrestha et al., 2017 ). The porous crystalline DNA origami structures can be regarded as arrays of such nanoconfinement spaces, which provide a tool for multiplexed single-molecule analysis ( Sakamoto et al., 2020 ) that will enable the simultaneous monitoring of reactions occurring in individual nanopores. Limitations of the study One of the primary limitations of this study is that in our origami design and surface-assisted approach, self-assembly fundamentally occurs in 2D, although the assembly component itself is a 3D object. However, extending the assembling process into 2nd, 3rd, or more layers could be investigated in the future by designing interactions between the top and bottom faces of the origami block. Introducing specific interactions among multiple different components, which should lead to the surface-assisted self-assembly of lattices with predetermined shapes and sizes, will also be challenging. Our 3D-Hx can be regarded as a modular structure because we can prevent an arbitral side from connecting with others by disabling blunt-end stacking via staple extension with polyT. However, the current design has a symmetric appearance and thus does not allow us to define the combination of sides to be connected, as only one type of shape-complementary ruggedness is employed. This drawback should be addressed by designing specific pairs of complementary ruggedness patterns at specific pairs of sides. Such strategy could be applied not only to our hexagonal blocks but also to other 3D structures, such as cubes and cuboids. The implementation of photoresponsive moieties ( Asanuma et al., 2007 ; Kamiya and Asanuma, 2014 ; Yoshimura and Fujimoto, 2008 ) to control the blunt-ended stacking ( Gerling and Dietz, 2019 ; Willner et al., 2017 ) is also a fascinating direction, which would enable in situ switching of the assembly mode of the DNA origami component, thus leading to controlled reorganization of the lattice geometry from honeycomb to depleted, and vice versa ."
} | 6,058 |
35078994 | PMC8789899 | pmc | 9,586 | {
"abstract": "In bacteria and other microorganisms, the cells within a population often show extreme phenotypic variation. Different species use different mechanisms to determine how distinct phenotypes are allocated between individuals, including coordinated, random, and genetic determination. However, it is not clear if this diversity in mechanisms is adaptive—arising because different mechanisms are favoured in different environments—or is merely the result of non-adaptive artifacts of evolution. We use theoretical models to analyse the relative advantages of the two dominant mechanisms to divide labour between reproductives and helpers in microorganisms. We show that coordinated specialisation is more likely to evolve over random specialisation in well-mixed groups when: (i) social groups are small; (ii) helping is more “essential”; and (iii) there is a low metabolic cost to coordination. We find analogous results when we allow for spatial structure with a more detailed model of cellular filaments. More generally, this work shows how diversity in the mechanisms to produce phenotypic heterogeneity could have arisen as adaptations to different environments.",
"introduction": "Introduction Different species use different mechanisms to produce adaptive phenotypic heterogeneity (Fig. 1 ) 1 – 5 . In some cases, there is coordination across individuals to determine which individual will perform which role ( coordinated specialisation ) 1 , 6 . This coordination could use signals, cues, or a developmental programme to provide information about the phenotypes adopted by other individuals in the group 7 . For example, when honey bee workers feed royal jelly to larvae to produce reproductive queens (Fig. 1a ), or when the local density of a signalling molecule determines whether cyanobacteria cells develop into sterile nitrogen-fixing heterocysts (Fig. 1b ) 8 – 10 . In other cases, each individual adopts a helper phenotype with a certain probability, independently and without knowledge of the phenotypes adopted by other individuals ( random specialisation ) 2 , 5 , 11 , 12 . For example, in Salmonella enterica co-infections, random biochemical fluctuations within each cell’s cytoplasm are used to determine whether the cell sacrifices itself to trigger an inflammatory response that eliminates competitor species (Fig. 1d ) 12 , 13 . In yet further cases, the phenotype is influenced by the individual’s genotype (genetic control). For instance, in some ant societies, whether individuals develop into queens, major or minor workers can be determined, in part, by their genes (Fig. 1c ) 3 , 14 – 16 . Across the tree of life, some species employ one mechanism to produce phenotypic heterogeneity whereas in other species mixed forms exist with a combination of coordinated specialisation, random specialisation, or genetic control 3 , 15 , 17 – 22 . Fig. 1 Different mechanisms to produce phenotypic heterogeneity in nature. a In honey bee hives ( Apis mellifera ), larvae develop as sterile workers unless they are fed large amounts of royal jelly by adult workers (coordinated specialisation) 8 (Photo by Waugsberg via the Wikimedia Commons). b In cyanobacteria filaments ( A. circinalis ), some individuals develop into sterile nitrogen fixers (lighter/yellow, round cells) if the amount of nitrogen fixed by their neighbours is insufficient (coordinated specialisation). This leads to a precise allocation of labour, with nitrogen-fixing cells distributed at fixed intervals along the filament 9 (Picture by Dr. Imre Oldal via the Wikimedia Commons, cropped). c In the army ant ( Eciton Burchelli ), whether individual ants become a major or minor worker has a genetic component (genetic control) 16 (Photo by Alex Wild via the Wikimedia Commons, cropped.). d In S. enterica infections (serovar Typhymurium), each cell amplifies intracellular noise to determine whether it will self-sacrifice and trigger an inflammatory response that eliminates competing strains (random specialisation) 13 (Photo by Rocky Mountain Laboratories, NIAID NIH via Wikimedia Commons). We lack general evolutionary explanations for why different species use different mechanisms to produce phenotypic heterogeneity 2 , 3 , 23 , 24 . Previous work has focused on the non-reproductive division of labour in the social insects, and the proximate mechanisms that lead to different worker castes 6 , 16 , 25 – 29 . However, the focus in that literature is on a different question—how different proximate mechanisms can produce coordinated specialisation—rather than the broader question of whether coordinated specialisation should be favoured over random specialisation or genetic control in the first place. It is with the reproductive division of labour that these three very different mechanisms have been observed in different species and for which there is an absence of evolutionary explanations 2 , 3 , 23 , 24 , 30 . Reproductive division of labour in bacteria and other microbes offers an excellent opportunity for studying why different mechanisms to produce phenotypic heterogeneity are favoured in different species 1 , 2 . Reproductive division of labour occurs when social groups are composed of more cooperative ‘helpers’ who gain indirect fitness benefits by the aid they provide to less cooperative ‘reproductives’. Across microbes, the two primary mechanisms used to produce reproductive division of labour are coordinated and random specialisation (Fig. 2 ). Furthermore, while the form of cooperation and life histories of microbes share many similarities, they also vary in factors that could influence the evolution of division of labour, such as social group size 31 , 32 . Fig. 2 Mechanisms to produce reproductive division of labour in clonal groups. We examine the relative advantages and disadvantages of the two key mechanisms to produce reproductive division of labour in social microorganisms 1 , 5 , 11 , 46 . a Random specialisation occurs when cells randomly specialise into helpers or reproductives independently of one another. This can occur when a genetic feedback circuit is used to amplify small molecular fluctuations in the cytoplasm of each cell (phenotypic noise) 4 , 11 , 12 , 78 – 80 . b Coordinated specialisation occurs when cells interact with one another, and share (or gain) phenotypic information while they are differentiating. This could occur through the secretion and detection of extracellular molecules (signals or cues), or with a shared developmental programme (epigenetics) 1 , 2 , 25 . We develop theoretical models to examine whether the relative advantages of random and coordinated specialisation can depend upon social or environmental conditions. Our aim is to use the reproductive division of labour in microbes as a ‘test system’ to address the broader question of whether evolutionary models can explain the diversity in the mechanisms that produce phenotypic heterogeneity more broadly. We show that coordinated specialisation is more likely to evolve over random specialisation in well-mixed groups when: (i) social groups are small; (ii) helping is more “essential”; and (iii) there is a low metabolic cost to coordination. We find the same qualitative results with deliberately simple models that are designed to capture the essence of the problem and with more detailed models that allow for spatial structure.",
"discussion": "Discussion Our analyses provide a theoretical framework to help explain why different species of microorganisms use different mechanisms to divide labour 2 . Coordinated division of labour is more likely to be favoured when: (i) social groups are small; (ii) helping is more “essential”; and (iii) there is a low metabolic cost to coordination. While testing our predictions with a formal comparative analysis would require data from more species, our predictions can help to understand the mechanisms that have evolved in well-studied examples. There are many reasons why coordinated specialisation was favoured to evolve in cyanobacteria filaments. First, cyanobacteria only divide labour when fixed N 2 is growth-limiting and so the relative importance of cooperation is high (low \\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}$$\\phi$$\\end{document} ϕ and high \\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}$$h/b$$\\end{document} h / b ) 9 , 58 , 63 . Second, the fixed nitrogen produced by helpers diffuses along the filament, preferentially aiding nearby reproductives and so the effective social group size is small (low \\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}$$\\eta$$\\end{document} η and small \\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}$$n$$\\end{document} n ) 9 , 49 , 67 . Third, the initial costs of coordination may have been quite small as new cells could use the local level of fixed N 2 as a cue (low \\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}$$\\eta$$\\end{document} η ) 68 . Finally, cyanobacteria filaments have a rigid spatial structure with local benefits from cooperation and thus random specialisation could have led to the accumulation of large sterile clumps, which is a very inefficient distribution of phenotypes (high functional or stochastic cost; Fig. 6 ). Colonies of Volvox carteri and Dictyostelium discoideum use coordination to divide labour, despite the fact that these groups are composed of large numbers of cells (high \\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}$$n$$\\end{document} n ; on the order of 1000 s of cells or more) 20 , 69 – 71 . This highlights that no single factor can fully explain empirical patterns, and that further factors not captured by simple models might be relevant in specific cases. For instance, colonies of Volvox carteri require a specific spatial distribution of flagella beaters across the group, which may create a strong selection pressure for coordination, analogous to the avoidance of clumps in cyanobacteria filaments. Furthermore, in some cases, details of the mechanism of division of labour are still not well understood. For instance, it is possible that there is also an initially random component to pre-stalk specialisation in Dictyostelium 70 . There are multiple reasons why random specialisation would have been favoured to evolve in other well-studied species. In Salmonella enterica , the self-sacrificing helper cells provide a competitive advantage that eliminates other microbes but is not “essential” to the replication of Salmonella cells (lower \\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}$$h/b$$\\end{document} h / b ) 12 , 13 . Further, the benefits of cooperation are provided to all cells in the co-infection ( \\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}$${{\\eta }}=1$$\\end{document} η = 1 ) and so the effective social group size is reasonably large (higher \\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}$$n$$\\end{document} n ). Finally, Salmonella pathogens do not have a rigid spatial structure and so there is no scope for the accumulation of growing helper clumps as for cyanobacteria filaments. In Bacillus subtilis , a subset of cells become helpers that produce and secrete protein degrading proteases 44 . However, these helper cells are not sterile and so the consequence of deviating from the optimal caste ratios is reduced (Supplementary Methods B.4 ). To conclude, most previous work on phenotypic heterogeneity has tended to be either mechanistic, focusing on how different phenotypes are produced (caste determination), or evolutionary, focusing on why heterogeneity is favoured in the first place 1 – 6 , 8 , 11 , 15 , 23 – 28 , 30 , 49 , 70 , 72 – 77 . We have used evolutionary models to address the broader question of why different mechanisms are used in different species 2 , 3 , 12 , 23 – 25 . Focusing on the reproductive division of labour in microorganisms, we have shown that coordinated specialisation is more likely to be favoured over random specialisation in small groups, when relative coordination costs are low, and when there are larger fitness costs to deviating from optimal caste ratios. We have also shown how these patterns can hold in groups with spatial structure, where there can be a large pressure for an even distribution of phenotypes. These results identify social and environmental factors that could help to explain the distribution of mechanisms to produce phenotypic heterogeneity that has been observed in bacteria, other microbes, and beyond. Aside from microorganisms, our results also suggest a hypothesis for why random caste determination has not been widely observed in animal societies. During the initial evolution of complex animal societies, group sizes were likely to be small and the relative costs of coordination might have been minor compared to each individual’s day-to-day organismal metabolic expenditure."
} | 3,696 |
39909903 | PMC11799125 | pmc | 9,587 | {
"abstract": "Spontaneous afforestation following land abandonment has been increasingly recognized as a nature-based solution to mitigate climate change and provide measurable benefits to biodiversity. However, afforestation effects on biodiversity, particularly on soil microbial communities, are still poorly characterized, with most previous studies focusing on artificial plantations rather than forest rewilding dynamics. Here, we assessed changes in topsoil physical–chemical properties and related dynamics of bacterial and fungal community composition and structure following spontaneous afforestation of abandoned grasslands in Northeast Italy over the last 70 years. With a space-for-time approach, we selected four chronosequences representing different successional stages: grassland, early (2000–2020), intermediate (1978–2000), and late (1954–1978). Results showed that spontaneous afforestation progressively reduced topsoil pH and total phosphorus (P), while soil organic carbon (SOC), nitrogen (N), and C:N ratio increased. Correspondingly, the overall α-diversity of the fungal community, assessed by ITS DNA metabarcoding, progressively decreased after an initial increase from grassland conditions, following substrate acidification and trophic specialization. Bacterial diversity, assessed by 16S DNA metabarcoding, was highest at the initial stages, then progressively decreased at later stages, likely limited by lower organic matter quality. Shifts of fungal community composition included an increase of ectomycorrhizal Basidiomycota linked to topsoil’s higher SOC, N, and C:N ratio. Differently, bacterial community composition responded substantially to pH, with topsoil acidity favoring Proteobacteria (Pseudomonadota) and Acidobacteria (Acidobacteriota) at the late afforestation stages. Our findings provide a first contribution to clarify how fungi and bacteria respond to spontaneous afforestation. This is particularly relevant in the context of climate change mitigation, considering the fundamental role of microorganisms in shaping soil carbon storage dynamics. Supplementary Information The online version contains supplementary material available at 10.1007/s00248-025-02500-9.",
"conclusion": "Conclusion This study highlighted that natural afforestation following grassland abandonment significantly alters topsoil physicochemical properties and soil microbes, at the core community scale. The process led to progressive enrichment of SOC and nutrients and soil acidification driven by increased plant litter input and changes in organic matter quality. Soil content of K and P varied based on environmental factors such as forest type and afforestation age. Bacterial communities were decisively influenced by pH, with acidic conditions favoring Proteobacteria (Pseudomonadota) and Acidobacteriota while limiting other bacteria, hence resulting in a progressive decline of α diversity at a late afforestation stage. Fungal communities, after an initial enhancement compared to grassland conditions, showed declining α diversity, with ectomycorrhizal Basidiomycota prevailing at later stages. These findings provide a first contribution to the understanding of microbial responses to spontaneous afforestation, a topic interrelated with biogeochemical cycles, and then of utmost relevance in the frame of nature-based solutions for climate change mitigation.",
"introduction": "Introduction Forests are of paramount importance at the global scale for climate change mitigation by sequestration and storage of atmospheric carbon (C) in above- and below-ground stocks [ 1 , 2 ]. In Europe, such mitigation potential is particularly relevant due to afforestation following land abandonment, driven in the last 70 years by the progressive depopulation of mountain areas in Southern Europe, by the Common Agricultural Policy in EU Member States, and by the end of the communist regime in Central and Eastern Europe [ 3 , 4 ]. The reduction of net CO 2 emissions is the central goal of the European Green Deal, aiming for climatic neutrality by 2050 [ 5 ]. This raises the research question of whether forests can provide a cost-effective and natural toolset for enhancing C storage at the ecosystem level [ 6 – 8 ]. Moreover, afforestation as a nature-based solution (NbS) for climate change mitigation should also support biodiversity [ 9 ], particularly soil biodiversity, its composition, and functions. In fact, understanding microbial dynamics with spontaneous afforestation in the long term is of utmost importance to detect possible trade-offs between biodiversity conservation and climate change mitigation, under a land management perspective [ 10 , 11 ]. Afforestation, as a land-use change process, progressively alters soil physicochemical properties, increasing acidity, carbon-to-nitrogen ratio (C:N), and nutrient content [ 12 , 13 ]. Depending on forest type, this process could also alter litter composition and input rates [ 14 ], affecting soil organic C quality [ 15 ] and sequestration dynamics [ 16 , 17 ]. These changes impact soil microbial community diversity and structure [ 18 , 19 ], which play a crucial role in nutrient cycling, C sequestration, and food webs [ 20 , 21 ]. Despite its significance, the effects of spontaneous afforestation on soil properties and microbial communities remain understudied, especially in the long term. Many previous studies have focused on artificial tree plantations, which often disrupt soil biota, particularly during the preparation and planting phases [ 22 ]. For instance, Pinus tabuliformis plantations on grasslands reduced fungal diversity but increased ectomycorrhizal (ECM) abundance [ 23 ], with nitrogen (N) availability playing a key role in driving microbial community changes. Similarly, available phosphorous (P), cation exchange capacity, and C:N ratio positively influence bacterial diversity in coniferous plantations [ 24 ]. These factors, alongside pH and other environmental variables, drive shifts in bacterial community composition. Moreover, in long-term tree plantations P and sulfur (S) availability shaped more distinctly bacterial than fungal communities [ 24 ]. Finally, in a recent study conducted in an artificial forest of Pinus armandii , bacterial communities were more responsive to afforestation than fungal ones [ 25 ]. Most of the previous studies dealing with microbial community shifts were limited to high-level taxonomic ranks, such as phylum level, where results often vary depending on the context. For example, in areas afforested by Robinia pseudoacacia and Caragana korshinskii , afforestation shifted the bacterial communities from Actinobacteriota- to Proteobacteria (Pseudomonadota)-dominated [ 26 ], while in afforestation by Taxodium in the Yangtse river basin, an increasing abundance for Nitrospirae was reported [ 27 ]. In different conditions, Acidobacteria (Acidobacteriota) can dominate soil communities, driven more by pH than by any other environmental factor [ 28 ]. The effect of single soil properties on microbial diversity is complex, especially over long-term afforestation dynamics, where stochastic processes in microbial community assembly are enhanced [ 29 ]. Conflicting findings on soil nutrient status, particularly for S and potassium (K), further complicate our understanding of afforestation impacts. Some studies suggested S increases in afforested areas due to dry deposition trapped by the forest canopy [ 30 ], whereas others indicated that S content rises due to contribution from the above-ground vegetation [ 31 , 32 ]. Potassium (K) levels also vary across studies. Some research points to a decrease in soil K following afforestation, especially with Pinus and Eucalyptus plantations [ 12 , 33 ], while others showed that afforestation of grasslands can enhance soil K [ 16 , 34 ]. Overall, the effects of afforestation on soil nutrient content, and the resulting impact on bacterial and fungal communities, remain understudied and context-dependent and require careful consideration of local conditions, such as tree species, soil types, and geological factors, in forest management practices aimed to climate mitigation and biodiversity conservation. In this study, we selected a typical mountain area, in Northeast Italy, that underwent depopulation and land abandonment in the recent decades, leading to the spontaneous transformation of grasslands into mixed deciduous woodlands [ 35 ]. Using a space-for-time approach, we selected four afforestation stages in four replicated chronosequences spanning 70 years. In each stage, we sampled topsoil DNA for metabarcoding and analyzed associated physicochemical properties to test the following hypotheses: (i) Bacteria and fungi respond differently to changes in soil properties, depending on different trophic specialization; (ii) topsoil properties change following litter input quality and increasing canopy cover, towards high C:N ratio, more acidic pH, and low environmental variability at local scale, thus leading to less rich, but evenly distributed microbial communities.",
"discussion": "Discussion Effect of Afforestation on Soil Properties Our findings clearly showed that spontaneous grassland afforestation affects soil properties such as pH, SOC, and nutrient content. Specifically, soils in the later stages of afforestation, compared to those in grasslands, were more acidic, with increased SOC, N, C:N, total S, and decreased P. A decline in soil pH is a well-documented outcome of afforestation, primarily due to the changes in litter fall occurring along the successional dynamics [ 59 ]. In addition to a progressive increase in the amount of litter yearly deposited to the forest floor, at least during the first decades of afforestation [ 60 ], the chemical quality of litter materials becomes progressively more acidic, as compared to that of grasses that persist the initial successional stages. Moreover, it was highlighted that afforestation globally leads to soil acidification and can significantly enhance soil nutrient availability [ 59 ], although the dynamics of N and P can vary depending on factors such as ecozone, forest type, tree species, and climate [ 61 ]. In this study, the significant reduction in soil P following afforestation is a known feature of the temperate forest zone [ 61 ]. Indeed, in temperate regions, and particularly at low temperatures, plants may preferentially use existing labile P forms over newly mineralized ones, ultimately consuming the available P pool [ 62 ]. Additionally, trees tend to have a higher demand for P than grasses, often resulting in a decrease in soil P content following grassland afforestation [ 63 ]. Resource Availability and Trophic Specialization Limit Fungal Community Diversity Substantial shifts of topsoil chemical properties influenced the composition and diversity of soil microbial communities, but with different responses for bacteria and fungi along the afforestation process. Fungal community diversity progressively decreased during plant successional dynamics, following grassland afforestation. These findings suggest that afforestation can lead to progressively more homogeneous edaphic conditions at the late stage, hence to a narrower range of niches, which in turn could drive to a reduction of the total number of taxa, increasing local extinction by competitive exclusion, and hence to a more even resource distribution by niche partitioning among the survivors. In other words, the increasing acidity and recalcitrance of litter inputs [ 59 ] could reduce the total pool of fungal species to those capable to either feed upon intrinsically limiting substrate or to undertake trophic interactions with tree roots, as in the case of mycorrhizal fungi, which we found thriving at the late afforestation stage. Accordingly, the progressive increase in SOC and nutrient content could result in increased overall resource availability, reducing interspecific competition at the late stages and enhancing the relative abundance of all the (trophically competent) fungi. Previous observations are only partially consistent with our explanation, as papers on soil microbiome changes along afforestation dynamics refer to remarkably different conditions with respect to our study in terms of ecoregion, substrate type, tree species, and time span (e.g., [ 23 – 25 , 29 , 64 , 65 ]). In fact, contrary to our results, some previous studies [ 23 , 64 , 65 ] found bacterial and fungal richness increases following artificial reforestation in different areas of China. In particular, Wang et al. [ 23 ] related their findings to large tree root growth enhancing soil porosity, which fosters the accumulation of root exudates, and soil aeration ultimately boosts microbial metabolism. Differently, consistent with our findings, Panico et al. [ 66 ] reported that the increase of SOC, N, and C:N ratio creates favorable conditions for the growth of fungal communities capable of decomposing recalcitrant organic matter. Resource Quality, More than Quantity, Controls Soil Bacterial Community Diversity Bacterial communities underwent remarkable changes especially at the late afforestation stage, with significant compositional differences between different stages and sites. Such a pattern indicates that bacterial community diversity is not directly enhanced by overall resource availability, in terms of amounts of soil organic matter and nutrients (as we observed at the late afforestation stage). Rather, it is reasonable to infer that the highest bacterial diversity at early-intermediate afforestation stages is sustained by the wider range of quality of available resources, in turn derived from the molecular diversity of the litter input [ 67 ]. Indeed, at early and intermediate successional stages, the highest plant diversity can be maintained by the persistence of some of the open-habitat species occurring at the pre-afforestation stage, as well as the establishment of the first individuals of the late-successional forest species, in addition to generalist species [ 68 ]. A high plant diversity, not (only) taxonomically but in terms of leaf traits, should also correspond to the large molecular diversity of the leaf litter input [ 69 ] and hence of the dissolved organic matter (DOM) that leaches into the soil from the litter layer. In this respect, it is well known that microbial life in the soil is fuelled by dissolved organic matter that leaches from the litter layer [ 70 ] and that the soil bacterial community specializes towards the molecular type of litter source and its state of decomposition [ 71 , 72 ]. However, the causal role of soil pH cannot be excluded, at least because it directly affects the molecular diversity of SOC, as differentially influencing the decomposition rates of different C pools. At this stage, we are aware that our explanatory hypothesis is somehow speculative, as we did not characterize soil organic matter in our soil samples in terms of molecular diversity. However, this specific issue will be addressed in a further study, with the molecular quality of soil organic matter assessed by 13C-CPMAS-NMR [ 73 , 74 ]. Compositional Shifts of Microbial Communities Following Afforestation Core community composition distinctly differed between grassland and late-stage sites, while the early and intermediate afforestation stages exhibited more compositionally similar assemblages. This directly reflects the progressivity in the changes in topsoil properties. In our research, Proteobacteria, Acidobacteria, and Verrucomicrobiota were more abundant in afforested soils compared to grassland, and their prevalence increased with the afforestation stage, which is consistent with some previous findings from the forest soil environment [ 75 , 76 ]. Among the topsoil properties more associated with bacteria, pH, SOC, and N were the most predictive factors influencing the overall community composition and structure, as well as controlling the abundance of single taxonomic groups, as in the cases of Verrucomicrobia, Bacteroidetes, and Acidobacteria, indicating that lower-rank taxa within these groups likely share similar ecological traits [ 77 ]. Similarly, oligotrophic (low nutrient, r -strategy) or copiotrophic (high nutrient, K-strategy) traits may be deeply conserved among soil bacteria, as emerging at high taxonomic rank [ 78 ]. Overall, the observed changes in composition likely reflect the response of the core community to broader ecosystem succession and vegetation shifts, as plant inputs (leaf litter, root biomass, and exudates) vary by vegetation type and have a strong influence on the microbial response [ 71 ]. For example, we observed an increase in Alphaproteobacteria and a decrease in Actinobacteria, which aligns with a transition from grassland to forest soils [ 79 ]. Moreover, during the afforestation, low soil pH and C:N ratio explained significant fractions of the compositional variance of the fungal communities, also showing significant associations with single taxonomic groups. This finding highlights the dependence of soil fungal development on specific substrate types in terms of molecular quality, as C:N is a well-known indicator of soil organic matter quality and decomposability [ 80 ] and therefore a critical factor for niche-driven fungal community assembly. In more detail, the predominant fungal phyla in our dataset were Ascomycota and Basidiomycota, which are the largest groups in both grassland and forest ecosystems [ 81 , 82 ]. Ascomycota were consistently abundant before and after the grassland-to-forest transition. Their stability irrespective of the above-ground condition could be due to the stress-resistant and highly competitive nature of most Ascomycota [ 83 ]. On the other hand, Basidiomycota have been frequently reported as the dominant fungal group in forest soils [ 84 , 85 ]. In our study, their relative abundance was highest at the late stages. Basidiomycetes fungi are effective decomposers, particularly thriving on dead wood or litter. In fact, the lignin-rich environment of forest soils likely explains the high abundance of Basidiomycota in these sites [ 86 ]. These fungi are capable of degrading plant and animal residues, and some species even cause wood decay [ 23 ]. Additionally, certain Basidiomycota are known to withstand extreme environmental conditions like high temperatures, cold, drought, and UV radiation [ 81 ]. Given that early-stage afforested sites lack mature tree cover, which normally buffers against fluctuating conditions and reduces direct UV exposure, these fungi may have a competitive edge in such challenging environments. Among more specific taxa, in addition to saprotrophs, we observed increases in tree-associated ectomycorrhizae (e.g., Boletales, Cantharalleles, and Russulales [ 87 ]). These are well-known ECM responsive to P and N [ 88 ]. As such, in temperate deciduous forest soils, they can play important roles in nutrient cycling, often making up a similar or even greater fraction of the fungal community compared to saprotrophs [ 89 ]. We acknowledge that our results may possibly be regarded as preliminary, as the intrinsic limited sample size prevented a proper assessment of microbiome compositional variation at all the explored spatial levels. Particularly, while afforestation stage-dependency and between-site variability were disentangled by our PERMANOVA model, further within-site heterogeneity at a small spatial scale was not clarified. However, we considered it of major interest to address larger-scale effects, also considering that these broader studies are ongoing, aiming to clarify the benefits of natural afforestation and associated biodiversity patterns at an international scale."
} | 4,947 |
33589803 | PMC7610556 | pmc | 9,588 | {
"abstract": "Altruism between close relatives can be easily explained. However, paradoxes arise when organisms divert altruism towards more-distantly-related recipients. In some social insects, workers ‘drift’ extensively between colonies and help raise less-related foreign brood, seemingly reducing inclusive fitness. Since being highlighted by W. D. Hamilton, three hypotheses (bet-hedging, indirect reciprocity, and diminishing returns to cooperation) have been proposed for this surprising behaviour. Here we show using inclusive fitness theory that bet-hedging and indirect reciprocity could only drive cooperative drifting under improbable conditions. However, diminishing returns to cooperation create a simple context in which sharing workers is adaptive. Using a longitudinal dataset comprising over a quarter of a million nest-cell observations, we quantify cooperative payoffs in the Neotropical wasp Polistes canadensis , where drifting occurs at high levels. As the worker-to-brood ratio rises in a worker’s home colony, the predicted marginal benefit of a worker for expected colony productivity diminishes. Helping on related colonies can allow effort to be focused on related brood that are more in need of care. Finally, we use simulations to show that cooperative drifting evolves under diminishing returns when dispersal is local, allowing altruists to focus their efforts on related recipients. Our results indicate the power of nonlinear fitness effects to shape social organisation, and suggest that models of eusocial evolution should be extended to include neglected social interactions within colony networks.",
"discussion": "Discussion Established accounts of the evolution of eusociality assume actors must choose either to stay as helpers or leave as reproductives 8 , 28 . Our results suggest that diminishing returns may drive altruists to diversify their help across recipients: workers in some primitively-eusocial societies may increase inclusive fitness by providing altruism to recipients beyond their home colony. Under positive kinship, spatial kin clustering, and diminishing returns 17 , 29 , worker investments can evolve to become diffusible public goods. Our model predicts the conditions under which we expect cooperative drifting to have evolved ( Equation 2 , Fig. 1f ). Intuitively, drifting is more likely when there are stronger diminishing returns (higher T), a larger difference in workforce between nests (smaller g), increased total workforce (higher ψ), and a greater capacity to target kin (higher d ♀ and d ♂ ). For simplicity in Equation 2 , we assume that all colonies have the same sex ratio, but between-colony sex-ratio variation suggests an additional factor: a colony producing mainly brothers has a reduced worker relatedness to the brood, at which point switching colony may be rational for a worker. In short, drifting offers a simple route to boost inclusive fitness when neighbouring colonies differ in parameters that determine the value of a worker. Differences in worker and brood number arise easily among P. canadensis colonies ( Fig. 3a ), which are subject to several sources of stochasticity. These include fluctuations in worker number due to the high attrition rate of foraging workers 12 , frequent loss of brood to parasitoids, presumed loss of brood due to disease (based on workers’ hygienic removal of larvae), episodes of queen replacement, and so on. Fluctuations in brood cohort size translate into fluctuations in workforce size once the brood pupate. Since Michener 19 highlighted diminishing returns in hymenopteran societies in 1964, a number of studies across ants, bees, and wasps have revealed declines in per-capita productivity with rising group size (e.g. 18 , 30 – 33 ). This so-called ‘reproductivity effect’ has not proved universal (e.g. 34 – 36 ), but its frequent occurrence leads to ‘Michener’s paradox’: why do apparently partly-redundant helpers exist 26 , 30 ? Previous tests of the reproductivity effect have used snapshots of per-capita productivity. By contrast, we provide a prediction of plausible ranges for the payoffs of cooperation in a primitively-eusocial insect using colony dynamics. Diminishing returns exist, but – through cooperative drifting – workers can mitigate redundancy arising from stochastic variation in worker-to-brood ratios between colonies. The extent of drifting across primitively-eusocial insects remains to be explored 5 , 12 . However, the relatively high levels of drifting observed in Neotropical species such as P. canadensis contrast with, for example, the European wasp P. dominula , which also forms dense colony aggregations 13 but shows high aggression towards neighbours. This difference in social organization may be due to differences in the intensity of diminishing returns (e.g., due to food availability or parasite density). Higher stochastic predation of workers in some species may undermine workers’ abilities to track need across nests. Alternatively, drifting may be more likely in the tropics: unlike temperate species in which foundresses establish nests synchronously in the spring, tropical species often establish nests throughout the year 25 , and so nests may be more likely to differ in worker-to-brood ratio. Tropical species may also experience less uncertainty in neighbour relatedness, since nests are more commonly founded by local dispersal from parent nests (simulated in Fig. 4e ), although kin spatial structure can be reestablished in temperate species by natal philopatry of spring foundresses 37 . Direct comparisons between species with and without cooperative drifting are needed. Cooperative drifting has also emerged among complex eusocial species. Ant ‘supercolonies’ exist when nests with multiple queens (polygyny) exchange workers (polydomy) 6 , 38 . Supercoloniality results in remarkably low-relatedness cooperation, and remains a theoretical challenge. The evolution of supercoloniality is likely to involve informational constraints preventing nepotism 6 , although some positive relatedness may be maintained by cryptic kin structure 39 . Our results are relevant here: the initial drivers of low-relatedness cooperation are unlikely to have been either bet-hedging by risk-spreading at the expense of the expectation of inclusive fitness ( Equation 1 ; Fig. 1d ) or the reciprocity scenario proposed by Ref. 5 ( Fig. 1e ). In principle, diminishing returns may initially have favoured partial diversion of altruism to more-distantly-related colonies. However, supercoloniality and primitively-eusocial cooperative drifting are not completely analogous. Supercoloniality may have been a relatively small step for ants that had already evolved high within-colony polygyny – and consequently reduced relatedness 40 – for other reasons. Unlike primitively-eusocial wasps, the first step to explaining cooperative drifting in ants is explaining polygyny 41 . Manipulating colony networks by adjusting worker-to-brood ratio (ψ) may offer tests of whether wasps make strategic adjustments to investments (y). Empirical studies are needed to identify whether host workers discern cooperative drifters and adjust acceptance thresholds (m) adaptively 42 , 43 according to need. Future theoretical work could assess the tension between selfish and cooperative drifting in determining the acceptance of foreign workers. Wasps with high resource-holding potential may exploit the relaxation of nest boundaries to drift for direct fitness (e.g., joining dominance hierarchies on multiple nests to maximise chance of nest inheritance). Models of the mechanisms individual workers might use to distribute their effort would be useful, potentially inspired by resource-use models in foraging theory 44 . Nonlinear payoffs exert strong effects on social evolution: diminishing returns can limit the tragedy of the commons 45 , promote polymorphic equilibria 46 , and increase sharing in reproductive skew games 47 . However, the extent to which diminishing returns shape investment patterns remains little quantified – despite clear theoretical predictions. A tempting explanation for divestment across recipients is that actors help different recipients in proportion to relatedness (an idea known as the ‘proportional altruism’ model 48 ). This argument commits the ‘gamblers’ fallacy’ 49 : instead, it is always best to invest in the single recipient who carries the highest inclusive fitness returns at any one time 50 . To explain altruism towards more-distant relatives, in the 1980s Altmann 49 , Weigel 17 , and Schulman and Rubenstein 29 highlighted diminishing returns to investment by a single individual. Here, we have considered diminishing returns to investment by multiple individuals. In both cases, diminishing returns provide a simple explanation for helping multiple recipients, which awaits empirical study in many species. Our results indicate the power of nonlinear fitness effects to shape social organisation, and suggest that models of eusocial evolution should be extended to include neglected social interactions within colony networks. Author contributions. PK, SS, and ANR planned field data collection, and PK and PB collected field data. PK and ADH conducted modelling. PK, NJW, and ANR conducted statistical analysis, and PK, SS, and ANR interpreted the results. PK drafted the manuscript and all authors contributed to its development."
} | 2,358 |
24736783 | PMC4013631 | pmc | 9,594 | {
"abstract": "Quorum sensing (QS) has been recognized as a general phenomenon in microorganisms and plays an important role in many pathogenic bacteria. In this report, we used the Agrobacterium tumefaciens biosensor strain NT1 to rapidly screen for autoinducer-quenching inhibitors from bacteria. After initial screening 5389 isolates obtained from land and beach soil, 53 putative positive strains were identified. A confirmatory bioassay was carried out after concentrating the putative positive culture supernatant, and 22 strains were confirmed to have anti-LasR activity. Finally, we determined the strain JM2, which could completely inhibit biofilm formation of Pseudomonas aeruginosa PAO1, belonged to the genus Pseudomonas by analysis of 16S rDNA. Partially purified inhibitor factor(s) F5 derived from culture supernatants specifically inhibited LasR-controlled elastase and protease in wild type P. aeruginosa PAO1 by 68% and 73%, respectively, without significantly affecting growth; the rhl -controlled pyocyanin and rhamnolipids were inhibited by 54% and 52% in the presence of 100 μg/mL of F5. The swarming motility and biofilm of PAO1 were also inhibited by F5. Real time RT-PCR on samples from 100 μg/mL F5-treated P. aeruginosa showed downregulation of autoinducer synthase (LasRI and rhlI ) and cognate receptor ( lasR and rhlR ) genes by 50%, 28%, 48%, and 29%, respectively. These results provide compelling evidence that the F5 inhibitor(s) interferes with the las system and significantly inhibits biofilm formation.",
"conclusion": "4. Conclusions JM2 belonged to the genus Pseudomonas by analysis of 16S rDNA. The partially purified inhibitor factor(s) F5 from culture extraction could completely inhibit biofilm formation of P. aeruginosa PAO1, and inhibit other LasR-controlled virulence factors without significantly affecting growth. Real time RT-PCR result showed F5-treated P. aeruginosa PAO1 had downregulation of autoinducer synthase (LasRI and rhlI) and cognate receptor (lasR and rhlR) genes by 50%, 28%, 48%, and 29%, respectively. The evidence indicated that the F5 inhibitor(s) interfered with the las system and significantly inhibited the relative virulence.",
"introduction": "1. Introduction Pseudomonas aeruginosa is one of the most difficult pathogens to treat clinically, and infects vulnerable patients including those with postoperative immune suppression. In patients with cystic fibrosis (CF), P. aeruginosa causes lung disease or death. This pathogen exhibits intrinsic resistance to many structurally unrelated antibiotics [ 1 ]. Quorum sensing (QS) is a population density-dependent regulatory system that regulates the secretion of pathogenic virulence factors and biofilm formation in P. aeruginosa . It is thought that interfering with QS will inhibit microbial pathogenicity [ 2 , 3 ]. In P. aeruginosa , there are three interconnected QS systems: the las , rhl , and pqs systems [ 4 – 6 ]. The major signal molecules involved in these three QS systems are 3OC12-homoserine lactone, C4-homoserine lactone, and 2-heptyl-3-hydroxy-4-quinolone (PQS), respectively [ 6 , 7 ]. Among them, the las QS system is at the top of the QS hierarchy, and regulates the rhl and pqs QS systems [ 8 ]. N -(3-oxododecanoyl)- l -homoserine lactone (3OC12-HSL, OdDHL) is produced by the LasI AHL synthase in the las system. Once OdDHL reaches a critical threshold concentration, it binds to transcriptional regulatory protein LasR. Dimers of OdDHL-LasR then bind to target promoters and upregulate the expression of downstream genes such as protease and elastase genes. The rhl system consists of N -butanoyl- l -homoserine lactone (C4-HSL, BHL), the cognate receptor RhlR, and the synthase RhlI. Virulence factors such as pyocyanin and rhamnolipid are mainly regulated by the rhl system. The las and rhl systems control a complicated regulatory network involving several hundred genes [ 9 ]. Infections of P. aeruginosa are of great concern because of its increasing resistance towards conventional antibiotics. QS in P. aeruginosa acts as a global regulator of almost all virulence factors, including biofilm formation [ 10 ]. As the QS system of P. aeruginosa directly relates to its pathogenesis, targeting the QS systems will provide an improved strategy to combat drug resistance in this organism. Small molecule chemicals called quorum sensing inhibitors (QSIs) can selectively act on the receptors at the cell surface of bacteria, or directly penetrate the cell membrane to interact with the enzymes or proteins of various signal transduction cascades, eventually interfering with pathogenicity. Recently, there have been reports of QSIs specific for Pseudomonas aeruginosa . It was reported that bromohalogenated furanones from marine red alga Delisea pulchra effectively suppressed P. aeruginosa biofilm formation by interfering with QS [ 1 ]. Patulin and penicillic acid from Penicillium spp can enhance P. aeruginosa biofilm sensitivity to tobramycin, and activate neutrophilic granulocytes to remove the bacteria in a mouse model of P. aeruginosa infection [ 11 ]. A variety of bioactive agents, both natural and synthetic, were recently reported to have significant anti-biofilm activity against Gram positive and negative bacteria [ 12 , 13 ]. One synthesized QSI molecule, N -decanoyl- l -homoserine benzyl ester, was found to down-regulate total protease and elastase activities, as well as the production of rhamnolipid, without affecting bacterial growth. It had synergistic interactions with several antibiotics, including tobramycin, gentamycin, cefepime, and meropenem [ 14 ]. In previous reports, we constructed a rapid plate method to screen for QSIs from bacteria, using two biosensor strains, Agrobacterium tumefaciens NT1 for OdDHL inhibitors and Chromobacterium violaceum CV026 for BHL inhibitors [ 15 , 16 ]. The purple pigment violacein in Chromobacterium violaceum CV026 (Kmr cviI::mini-Tn5) is inducible by AHL with N -acyl side chains from C4 to C8 in length. An isolate identified as a Pseudomonas sp. was capable of inhibiting violacein production according to the C. violaceum CV026 bioassay. A more highly purified preparation (4 μg/mL) from concentrated culture supernatants of this isolate specifically inhibited rhl -controlled pyocyanin and rhamnolipid production in wild type P. aeruginosa PAO1 by 49%, without significantly affecting growth. The inhibitor reduced protease activity by about 46% but had no effect on biofilm in PAO1 [ 17 ]. QS is a key mechanism that regulates several aspect of biofilm development, including adhesion, motility, maturation, and dispersal [ 18 – 20 ]. In searching for novel quorum-quenching bacteria from soil samples, we focused on screening the las QS system, and obtained an isolate that strongly inactivated autoinducing activity and reduced the PAO1 biofilm formation. The compound produced by this isolate could potentially be a biological control for biofilm infection. A. tumefaciens NT1 (traR, tra::lacZ749) displays the broadest sensitivity to AHLs at the lowest concentrations, and senses AHL with N -acyl side chains in long length from 6 to 12 carbons as well as three unsubstituted AHLs [ 21 , 22 ]; we used A. tumefaciens NT1 as the reporter strain for las system inhibitor isolation in this study. A description of a novel autoinducer-quenching strain is presented here, including the anti-LasR fragment from culture supernatant extract, and its inhibition of biofilm formation and QS dependent virulence factors. We also describe its phylogenetic position based on 16S rRNA gene sequence information. At present, a therapy that efficiently targets bacterial biofilm does not exist, since biofilms are inherently resistant to conventional antibiotics. The threat of resistance development with these drug candidates is uncommon, as they attenuate only the virulence factors and not the growth of the pathogen [ 8 , 10 , 14 ]. In the present study, we targeted the las system of P. aeruginosa and studied its inhibition upon exposure to bioactives from one bacterium (JM2). This study also emphasizes the potential of JM2 to produce bioactive agents with anti-LasR and anti-biofilm properties that are novel drug candidates.",
"discussion": "2. Results and Discussion 2.1. Isolation of the Anti-LasR Strain 2.1.1. Detection of Anti-LasR on Solid Medium For bacterial screening, the test isolates from soil were first inoculated using sterile toothpicks onto their appropriate media and incubated overnight at 28 °C [ 16 ]. In the initial plate screening after incubation overnight, a blue color in the indicator bacteria NT1 occurred without inhibitors. Many test isolates grew well and had a blue-colored background, which indicated that there was no particular compound inhibiting the action of exogenously added OdDHL. In some instances, a small, cloudy, and colorless circle appeared around the test bacteria, indicating that las system repressors were being produced by the test strain. Figure 1 shows the colorless circle of bacteria on the NT1 plate ( Figure 1A ) designated JM2. Among more than 5000 bacteria isolates from land and beach, we obtained 53 isolates that showed a very obvious, colorless, opaque halo in the NT1 assay. Being cloudy indicated the growth of the reporter stain was not affected, and being colorless indicated that the production of β-galactosidase in NT1 was not successfully induced, therefore indicating that the QS signaling had been disturbed. Several possible reasons existed for how the QS signaling was disrupted: (1) the generation of QS autoinducer signals was inhibited; (2) the QS signals were metabolized or degraded; or (3) the signal molecules were prevented from activating the LasR transcription. The latter possibility was the focus of this study; we chose to explore whether the signal inhibitor chemicals produced by the test bacteria functioned in this capacity. We chose to study test bacteria with a colorless, but viable circular zone produced on the lawn of NT1, because these test strains had no effect on the growth of the reporter strain, which indicated no antibiotic products were being secreted by the test isolates. Growth inhibition would produce a clear halo versus a cloudy halo, while QS inhibitors will permit growth but inhibit only the hydrolysis of X-gal, showing a colorless, non-transparent zone around the target strain. 2.1.2. Further Bioassay of the Putative Strains Out of more than 5000 bacterial strains screened preliminarily for anti-LasR activity, 53 strains showed different levels of QS inhibition. We isolated these 53 putative anti - LasR strains for supernatant extraction and evaluation with further bioassays. Twenty-two isolate extracts demonstrated anti-LasR activity. One strain extract had stronger inhibition than all other isolates; this isolate was purified further on YEB agar plates by streaking, and the purified species was designed as JM2. As shown in Figure 1B , there was no colored colony at the blank agar slice. The length of colored colonies at the agar slice of extract of JM2 and 10 μM OdDHL was shorter than that of 10 μM OdDHL alone, demonstrating strong inhibition of QS. The length of blue colonies is inversely proportional to the autoinducer-quenching activity. At the same time, we found that methanol had no effect on QS, and the JM2 extract had no effect on reporter strain growth (data not shown). Therefore, we determined that the active compounds produced by JM2 were not antibiotics. The negative effect on blue pigment production on the bioassay plate was not caused by inhibition of growth, but by disruption of QS signaling systems ( Figure 1B ). 2.2. Bioassay Guided Fractionation of JM2 Methanol Extract Following the screening of fractionation effectiveness, compounds were tested in a concentration dependent manner. Figure 1B shows the results of these experiments in which varying concentrations of JM2 methanol extract were used. The anti-LasR component activity was more pronounced in a concentration gradient of bioactive fraction F5 (0.2–1.0 mg/mL). Fraction F5 showed anti-QS activity even at 10 μg/mL, as shown by a 72% ( p < 0.001) reduction in blue pigment production by the NT1 reporter strain ( Figure 1C ). 2.3. Construction of lasR Deletion Mutant (lastR − ) The deletion vector was first introduced into DH5α and subsequently into S17-1 by transformation because DH5α has a high transformation efficiency, while S17-1 has a high conjugation efficiency. Transconjugants that contain the deletion vector in the PAO1 chromosome were recovered from the gentamicin-containing MM agar, and streaked onto MM agar containing 10% sucrose. Sucrose is detrimental to the suicide vector due to the SacB region, thus, causing the vector region to be excised from the chromosome. The sucrose-resistant colonies were then screened using colony PCR to identify the deletion mutants. The resulting positive PCR results were subsequently reconfirmed by repeating PCR on the possible mutants. Sequencing of the possible mutant was carried out to confirm the lasR − mutant when the control samples failed. 2.4. Effect of F5 on the Production of Virulence Factors in P. aeruginosa PAO1 2.4.1. Effect of F5 on Pyocyanin, Elastase, Rhamnolipids and Protease Production in P. aeruginosa PAO1 and LasR − Mutant P. aeruginosa virulence factors were reduced by the bioactive fraction F5 in PAO1; significant reduction ( p < 0.001) in pyocyanin (54%), elastase (68%), rhamnolipids (52%), and protease (73%) production in the presence of 100 μg/mL of F5 was observed. In lasR − mutant, these virulence factors dropped strongly and even added F5 ( Figure 2 ). All assays were done in triplicate, and the values were expressed as mean ± SD. 2.4.2. Effect of F5 on Biofilm Formation The PAO1 strain was able to develop biofilms on a plastic surface, with OD 570 values ranging from 0.6 to 0.67, but lasR − mutant was not ( Figure 3A ). The effects of F5 on P. aeruginosa biofilm formation were further evaluated using a crystal violet-based biomass-staining assay. At the concentration of 100 μg/mL, F5 caused a 95% decrease in the ability of PAO1 to form biofilms, relative to PAO1 grown without F5 ( Figure 3A ). With lower concentrations (20 μg/mL) of F5 there was a slight inhibitory effect on biofilm formation ( Figure 3A ); F5 at 50 μg/mL caused a larger decrease in biofilm formation (82% decrease) ( Figure 3B ). All assays were done in triplicate, and the values were expressed as mean ± SD (Dunnett’s test, p < 0.01). We performed a growth curve analysis for P. aeruginosa PAO1 cells exposed to 100 μg/mL of F5, and these cells showed no lag in growth as compared with control cultures containing only media and the solvent (methanol) used for compound resuspension ( Figure 3B ). These data suggest that F5 affects biofilm formation independent of cell growth and propagation. 2.4.3. Effect of F5 on Swarming Assay Microorganisms with motility are able to move to find more favorable or less hazardous niches for colonizing and persisting in a given environment. Swarming is a special kind of motility observed on semi-solid surfaces, defined as surface translocation dependent on extensive flagellar motion and cell-to-cell connections. It has an important role in the avoidance of harmful environments, and in colony formation. Swarming motility in P. aeruginosa requires QS because it uses rhamnolipids that act as a biosurfactant. Rhamnolipid production is dependent on and regulated by QS systems. The results shown in Figure 4 indicate that F5 is capable of inhibiting swarming motility in PAO1 cells grown in medium. The observed difference in colony diameters ( Figure 4 ) shows the effect of F5 on the P. aeruginosa QS system. When F5 was added to the culture medium, no effect on microbial growth was observed at the assayed concentrations. Swarming was limited at these concentrations, but not fully inhibited. Figure 4 shows images of swarming cells taken after 24 h of culture. PAO1 colonies have edges with fan-shaped or finger-shaped protrusions, and, under our experimental conditions, colony diameters of 2.2 to 2.5 cm. After culturing the cells with the addition of 50 μg/mL F5 for 24 h, the swarming ability of PAO1 was significantly weakened. The colony edges, as compared with the control group, appeared smooth and nearly circular, and the colony size shrunk to a diameter of 0.8 to 1.1 cm. 2.5. Phylogenetic Analysis To establish the phylogenetic position of the isolate JM2, the 16S rDNA gene from JM2 was sequenced. The accession number in NCBI BankIt is ID 1666536 ACCESSION BSeq#1. The JM2 16S rDNA sequences were aligned with published GenBank 16S rDNA sequences, and homologous comparison results showed that the JM2 16S rDNA sequence shared 98.8333% homology with 16S rDNA from the marine bacterium Pseudomonas pachastrellae ( Figure 5 ), and was more than 95% identical to 16S rDNA from bacteria in the Pseudomonas genus. Therefore, JM2 is most closely related to Pseudomonas pachastrellae KMM330 (T). 2.6. Real Time RT-PCR Real time RT-PCR showed a 50%, 48%, 21%, and 25% reduction in the expression of lasI , lasR , rhlI and rhlR genes, respectively, upon treatment with 100 μg/mL of F5 ( Figure 6 ). Our results reveal a new class of bacterial biofilm inhibitor, and further support an approach to biofilm inhibition via inhibition of the las QS system. Although much success in drug discovery has been reported using chemical molecules from plants, the likelihood of finding a novel drug from this collection of molecules is extremely low. This prompted us to embark on a study using microbes with various bioactive potentials. Studies are underway to explore the specific mechanism of action of the F5 compound(s) in biofilm inhibition, and to further identify the molecular structure(s) and activity. We found that the isolated strain JM2, which was belonged to the genus Pseudomonas, can efficiently interfere with the quorum-sensing signaling circuit in wild type P. Aeruginosa PAO1. A partially purified the inhibitor factor(s) F5 derived from the culture supernatants specifically inhibited las -controlled biofilm, elastase and protease in PAO1. Real-time polymerase chain reaction analysis showed that F5 downregulated the transcriptions of autoinducer synthase ( lasI and rhlI ) and their cognate receptor ( lasR and rhlR ) genes, which resulted in attenuation of QS-regulated virulence activities, such as biofilm formation, and secretion of protease, elastase and pyocyanin. The protease and elastase are las-controlled virulence factors [ 23 ], pyocyanin and rhamnolipid is controlled by rhl system [ 24 ]. The reduction data of elastase (69%), protease production (73%), pyocyanin (53%), rhamnolipids (51%) showed that F5 have stronger effect on lasR than on rhlR. Further, the PAO1 las system contributes greatly to biofilm formation substantial down-regulated transcription of lasR and lasI may be directly responsible for the reduction of PAO1 biofilms [ 25 , 26 ]. These results suggest that F5 from JM2 may possess lasR inhibitory activity against the virulence of PAO1."
} | 4,813 |
34555324 | null | s2 | 9,596 | {
"abstract": "Genome-scale models of metabolism (GEMs) are key computational tools for the systems-level study of metabolic networks. Here, we describe the \"GEM life cycle,\" which we subdivide into four stages: inception, maturation, specialization, and amalgamation. We show how different types of GEM reconstruction workflows fit in each stage and proceed to highlight two fundamental bottlenecks for GEM quality improvement: GEM maturation and content removal. We identify common characteristics contributing to increasing quality of maturing GEMs drawing from past independent GEM maturation efforts. We then shed some much-needed light on the latent and unrecognized but pervasive issue of content removal, demonstrating the substantial effects of model pruning on its solution space. Finally, we propose a novel framework for content removal and associated confidence-level assignment which will help guide future GEM development efforts, reduce duplication of effort across groups, potentially aid automated reconstruction platforms, and boost the reproducibility of model development."
} | 269 |
27152511 | PMC4859483 | pmc | 9,600 | {
"abstract": "The structure and function of microbial communities is deeply influenced by the physical and chemical architecture of the local microenvironment and the abundance of its community members. The complexity of this natural parameter space has made characterization of the key drivers of community development difficult. In order to facilitate these characterizations, we have developed a microwell platform designed to screen microbial growth and interactions across a wide variety of physical and initial conditions. Assembly of microbial communities into microwells was achieved using a novel biofabrication method that exploits well feature sizes for control of innoculum levels. Wells with incrementally smaller size features created populations with increasingly larger variations in inoculum levels. This allowed for reproducible growth measurement in large (20 μm diameter) wells, and screening for favorable growth conditions in small (5, 10 μm diameter) wells. We demonstrate the utility of this approach for screening and discovery using 5 μm wells to assemble P . aeruginosa colonies across a broad distribution of innoculum levels, and identify those conditions that promote the highest probability of survivial and growth under spatial confinement. Multi-member community assembly was also characterized to demonstrate the broad potential of this platform for studying the role of member abundance on microbial competition, mutualism and community succession.",
"conclusion": "Conclusions A high-throughput platform for measuring the growth of independent microbial populations in three-dimensional, microscale landscapes with controlled physical and chemical features facilitates exploring the complex parameter space that influences microbial community development. Central to this methodology is the ability to isolate microbial cells precisely into wells using a parylene-based lift-off technique, combined with the capability of assembling initial populations with tunable dispersity by controlling the geometric features of the wells. We have demonstrated that seeded populations of bacteria can be trapped in three-dimensional microwells under appropriate environmental conditions, allowing for dynamic growth or decay measurements across a large number of independent populations using a simple, quantitative fluorescence readout. The ability to tune the initial population dispersity using microwell diameter and depth is attractive because it allows for the study of community behavior under different environmental and initial conditions. Seeding in wells with diameters significantly greater than the size of individual cells drives homogenous population assembly ( CV well ≤ 0.20), enabling a ‘high-statistics’ approach to monitoring the growth or decay of replicate populations. Conversely, seeding into wells with diameters that approach the size scale of individual bacterium drives heterogeneous population assembly, enabling one to screen a large parameter space in order to identify cellular combinations and environments that are conducive or inhibitory to community growth and proliferation under prescribed conditions. Future work is aimed at screening interactions using high-dispersion population assembly with multi-species microbial communities in order to uncover symbiotic, mutualistic, and pathogenic relationships.",
"introduction": "Introduction Microbial communities impact our lives in dramatic ways. Taking on the role of both friend and foe, these communities shape our environment and ecosystems, fuel business and agriculture, and simultaneously support and perplex our healthcare system. Most communities are as dynamic as they are diverse, continuously adapting in their composition, organization and function to survive and thrive in changing environmental landscapes. Within the community, microbes exchange materials, energy and information via metabolite transfer, diffusive chemical signaling (e.g. quorum sensing) and contact-mediated interactions (e.g. protein secretion) [ 1 – 3 ]. As a result, complex behaviors such as robustness (e.g. antibiotic resistance), efficient resource utilization, mutualism and enhanced biosynthetic capacity emerge. Community development is dramatically influenced by the chemical and physical landscape of the local environment [ 4 ]. Systems such as soil, packed bed reactors and medical device surfaces contain spatially-confined micro-environments or niches, where the path length for diffusive chemical signaling is significantly altered and where micro and nanoscale topology influence surface attachment events that impact community structure and function [ 5 , 6 ]. Conventional physiological assays of single species and multi-member cultures, carried out in large fluid volumes or over solid media, fail to address the impact of spatial organization and environmental heterogeneity on community development and function. Recent advancements in the capacity to characterize communities with improved spatial and temporal resolution have fueled a growing appreciation of the role that composition and physical architecture play on community dynamics. Advances in omics and sequencing technologies have been combined with the ability to image communities at the cellular and subcellular scales to provide a systems-level understanding of natural community function and adaptation [ 7 ]. Concurrently, a growing capacity to manipulate material and chemical environments across length scales, using micro and nanofabrication, has enabled studies that uncover the impact of hierarchical and heterogeneous structure on early-stages of community development and function [ 8 ]. For example, manipulation of surface architecture and spatial confinement at the length scale of single cells has been used to dramatically alter early colonization and self-regulated quorum signaling [ 9 , 10 ]. At the community level, studies into the role of confinement and microscale chemical heterogeneity has provided new insights into competition, cooperation, antibiotic resistance and succession [ 11 – 14 ]. Despite these advances, much is unknown about how microbial community members develop with respect to their neighbors and within their respective physical and chemical landscapes. This has inhibited our ability to assemble artificial communities of natural isolates or design synthetic communities of engineered organisms that exhibit the robustness and functional capacity of natural communities. These limitations are due, in large part, to our inability to systematically screen the vast parameter space of factors that influence ecological succession and function [ 15 ]. Member abundance, confinement, niche connectivity, and nutrient exchange are just a few of the factors that can influence such processes [ 16 ]. Discovery necessitates the development of screening tools that can rapidly and systematically explore this space in order to identify those conditions and combinations of community members that promote or inhibit heterogeneous growth, coordinate function and drive emergent behavior. Toward these goals, several platforms have been developed that utilize micro- and nano-fabrication techniques to enable the study of uncultivable bacterial species, the study of competition or synergy in small bacterial communities [ 17 ], and studies of the effects of spatial structure and chemical environment on colonization [ 11 , 18 ]. Several approaches have been taken that confine small numbers or single microbial cells in small volumes, enabling these individual microbes or small colonies to be observed microscopically. Microfluidic droplet generators that encapsulate bacteria are a common approach [ 19 – 24 ]. Also used are arrays of microwells [ 25 ], like the million-well growth chip designed by Ingham et al. that was applied for use as a high-throughput screening tool [ 26 ]. Another interesting approach utilizes a ‘cell docking’ method that delivers microbes to arrays of different diameter wells, using microfluidic channels, potentially allowing different aspects of confinement to be analysed [ 25 ]. Here, we detail the development of a discovery platform that combines the high-throughput nature of the droplet technologies and the million-well growth chip mentioned above [ 26 ] with the flexibility in size and environmental control afforded by microfabricated wells described by Park et al. [ 25 ] This multi-diameter microwell array is designed to screen unique microbial communities for growth across a large range of population and environmental parameters. We describe a microfabrication approach that uses a polymer dry lift-off method [ 27 – 29 ] to assemble microbial populations into silicon microwells with controlled surface chemistry and physical features ( Fig 1 A and 1 B ). By varying the length scale of the well diameter ( Fig 1 C ), we show that homogenous populations assemble into large (10 1 –10 2 μm diameter) wells, while highly heterogeneous populations assemble into small (10 0 –10 1 μm diameter) wells. Leveraging the high initial population dispersion driven by assembly into small wells, we screen unique Pseudomonas aeruginosa colonies to identify conditions that are either conducive or inhibitive to growth in spatially-confined environments. Finally, seeding with a binary system was characterized to demonstrate the utility of this platform for pairing interacting cells together in a controlled or randomized fashion. These results demonstrate a new, high-throughput methodology for screening population and environmental parameters for microbial growth. 10.1371/journal.pone.0155080.g001 Fig 1 Microwell array fabrication and design. (A) Microwell fabrication process: (i,ii) Positive photoresist is patterned over parylene-coated silicon wafers using conventional photolithography. (iii) Dry etching is then used to etch parylene and then silicon to the desired well depth. (iv) The well surface is then modified with a protein layer then (v) a solution of bacterial cells. (vi) Parylene is removed from the substrate and (vii) the substrate is contacted with agar-coated coverslips loaded with the desired chemical media. (B) Dry lift-off procedure involving peel-off of the parylene mask (step vi). (C) Layout of a combinatorial microwell array substrate.",
"discussion": "Results and Discussion Characterization of the Physicochemical Microwell Environment The microwell fabrication process was designed to allow for systematic control of both the physical structures and chemical surface features of the well interface. The well features were physically and chemically characterized prior to bacterial seeding. SEM images of low and high-aspect ratio wells demonstrate the capability to tune the level of spatial confinement into which the bacterial communities are assembled ( Fig 2 A and 2 B ). Wells contain walls with a periodic ribbed structure and a smooth floor, characteristic of the Bosch etching process ( Fig 2 A \n insert ). Well surfaces can also be modified with desired organic or biological components using liquid or vapor deposition over the entire stencil ( Fig 1 A , step iv). During this functionalization step, the silicon well surface is rich in Si-OH groups, making it amenable to direct contact with aqueous-phase solutions and reactive to organic silane-reagents. To demonstrate modification of the well surface, a solution of fluorescently labeled, adhesive lectin protein (wheat germ agglutinin-AlexaFluor 488, WGA-A488) was incubated over the entire substrate. After washing and removal of the parylene mask, fluorescent images show the protein exclusively coated within the well surface, with higher intensity levels noted at the well edges due to the three-dimensional wall structure ( Fig 2 C and 2 D ). The variation in protein density between wells within an array was 16%. This level of variation is likely due to protein aggregation that occurs across the well surface, and is comparable to the variation previously noted using microcontact printing methods (11–13%) [ 33 ]. Further improvements in protein uniformity can likely be obtained using printing buffers that minimize aggregation [ 27 ]. Background regions appear unmodified, which was expected as proteins do not diffuse through the parylene layer. 10.1371/journal.pone.0155080.g002 Fig 2 Characterization of the physicochemical features of microwell arrays. (A) SEM image of an array of 2μm diameter microwells and of an individual well after substrate cleavage (inset). (B) SEM images of a 20μm diameter microwell array. (C) Fluorescent image of microwells after functionalization with WGA-A488 and dry lift-off. The arrow highlights a portion of un-peeled parylene, also containing adsorbed WGA-A488. (D) Fluorescent line plot of WGA-A488 coated wells corresponding to the red line in 2C. Seeding Behavior of Bacterial Cells Into the Microwell Array Control of the physical and chemical features of the microwell interface can be used to bias the bacterial populations isolated during the seeding step. It is well known that the attachment of bacterial cells to a solid interface depends on topological and chemical surface features, and numerous reports have used micro and nanofabrication strategies to control these properties, either to promote or inhibit bacteria attachment and interactions [ 34 – 36 ]. Here, substrates containing arrays of wells with diameters ranging from 5 to 1000 μm were used to investigate the effect of diameter on the population distribution of seeded Escherichia coli cells expressing green fluorescent protein (GFP). False color fluorescent images generated from cells isolated in wells after seeding and lift-off reflect the cell densities in the wells ( Fig 3 A ). As evident, cells are seeded exclusively within the well boundaries. Magnified images of representative 4×4 arrays qualitatively demonstrate that arrays of smaller wells contain increasingly heterogeneous numbers of cells within initial populations. 10.1371/journal.pone.0155080.g003 Fig 3 The distribution of bacteria seeded in microwell arrays is guided by well diameter. (A) Mosaic 10X false-color fluorescent image of a combinatorial microwell array after seeding E . coli -GFP at OD 600 = 0.3 and dry lift-off to remove background cells. The false color scale denotes fluorescent signal intensities indicative of cell densities. (B) Averaged well fluorescence intensities ± standard deviation measured from individual wells within each array (black line) and CV array (red dashed line), the standard deviation divided by the average fluorescent signal for each well diameter. To quantify the trends shown in Fig 3A , the average fluorescence intensity and the variation in signal between replicate wells within the same array ( CV array ) were measured for each diameter ( Fig 3 B ). Here, average well intensity reflects the average number of cells contained within an array, while CV array reflects the variation in initial cell populations present in wells across an array. Populations assembled within larger wells (≥80 μm diameter) allowed for reproducible cell populations ( CV array ≤ 0.2) to distribute in the well, likely due to minimal crowding or interference from the sidewalls. As well diameters decreased below 80 μm, the average fluorescent signal also decreased, despite larger surface-area to volume ratios for cell attachment. This can be attributed to higher levels of spatial confinement, rendering a higher fraction of well binding sites inaccessible to cells. Also, stochastic interactions between single cells and wells become more pronounced at smaller diameters, causing larger CV array values. Similar trends have been noted while seeding mammalian cells into microscale wells [ 37 ], and also while encapsulating molecular systems into nanoscale compartments, such as DNA probe functionalization in nanowells for digital PCR systems [ 38 ]. To further characterize the population distribution in seeded wells, E . coli -GFP was seeded into 5 μm diameter wells over a two order of magnitude concentration range (OD 600 = 0.01 to 1.0, Fig 4 A ). The corresponding frequency histograms ( Fig 4 B ) indicate that initial well populations follow a Poisson distribution, suggesting that seeding occurs as independent, random events with frequencies proportional to the concentration of cells present in bulk solution. To mathematically describe the relation between seeding concentration and cell distribution within these wells, the data was fit according to the probability distribution function:\n P ( x ) = A · λ x x ! · e - λ ; x = 0 , 1 , … , n (2) \nwhere x represents the number of occurrences (cells captured per well), A is an amplitude parameter, λ represents the mean value, n represents the total number of cells present, and P ( x ) represents the probability of capturing x number of cells in a well. 10.1371/journal.pone.0155080.g004 Fig 4 Bacterial well populations follow a Poisson distribution. (A) 20X false color fluorescent images of 5 μm diameter wells seeded with E . coli -GFP at OD 600 = 0.01, 0.1, and 1.0. (B) Probability distributions for cell populations at the varied seeding concentrations. Diamonds represent data and solid lines represent a Poisson distribution fit to the data according to eq 2 . Seeding at an OD 600 of 0.01, 0.1, and 1.0 resulted in a λ value of 1.9, 6.2, and 68.6, respectively, and an A value of 1.65, 1.35, and 1.00, respectively. These findings suggest that well dimensions can be used in combination with cell seeding concentration to tune the distribution of initial microbial populations isolated within an array. Control of this distribution is attractive, because it enables parallel monitoring of either a homogenous or heterogeneous assortment of initial populations; both are situations that are informative for characterizing microbial community development. Further optimization of the seeding protocol will allow for improved control of well populations. Currently, spatial correlations can often be identified when inspecting arrays spanning larger areas (~2 mm 2 ), likely caused by drying artifacts that occurred on the surface before the lift-off step. However, analysis of a large number of wells minimizes this effect. Finally, while the characterizations were made here with a model E . coli –GFP system, seeding other species may result in changes to population distributions, driven by differences in microbial traits (e.g. motility, extracellular matrix composition, cell-surface and cell-cell affinities). However, it can be expected that similar transitions to highly heterogeneous population assembly will occur for any species as the size of the well approaches the scale of individual cells. Moreover, quantifying deviations from Poisson-like distributions of microbes that result during the seeding process may be used to screen for microbe biases or affinities for, or against, particular surfaces or other microbes. Trapping and Growth of P . aeruginosa in Microwell Arrays After seeding, the chemical environment within the wells was controlled by sealing the microwell substrate with an agar-coated coverslip that had been treated with the desired chemical media ( Fig 1 A , step vii). This provided a physical barrier to trap motile cells within the well structures. The successful trapping of motile bacterial populations was demonstrated using P . aeruginosa modified to express GFP, which showed both surface-attached and un-attached populations confined within the wells during real-time monitoring ( S1 Video ). The motility observed from un-attached cells suggested that a large fraction of cells remained viable through the seeding and trapping process. Similar observations were made while monitoring bacterial cell populations confined in smaller well volumes ( S2 Video ). After trapping, a small number of cells were also found in background regions, which was caused by cell removal from wells during contact of the seeded substrate with the agar coverslip. Background cells appeared to be trapped between the agar and the silicon interface and typically showed no motility over time. While these cells also have the potential for growth [ 39 ], they are few relative to the number that remain trapped within the well volume and usually cause no interference with the test sites. Due to the static nature of media in the wells the after trapping step, it is expected that mass transfer across the arrays will be diffusion-limited. However, uniform concentration of oxygen and nutrients across the arrays is expected due to the small array area (200 x 200 μm). For example, the diffusivity of oxygen in a 1.5% (w/v) agar gel at 30°C is D O2,agar = 2.70 x 10 −5 cm 2 /s [ 40 ], and the characteristic time-scale for diffusion across the length of this array is ~10 s, well below the time scale required for growth (hours). After well sealing with agar-coated coverslips, P . aeruginosa was monitored for growth in wells with diameters between 5 and 20 μm, where the initial population distribution ( CV well ) was shown to be highly dependent on well diameter ( Fig 3B ). Here, Luria-Bertani (LB) media was added in the agar coverslip coating to promote growth. Growth was observed by monitoring well arrays with time-lapse fluorescence microscopy and then quantified in terms of cell volume fraction, the measured cell volume normalized to the overall volume of the well. This metric was determined using a correlation curve relating the average well fluorescence intensity (A.U.), obtained from the epi-fluorescence microscope system, to cell volume, quantified with a confocal microscope ( S3 Fig ). At the initial time point, broader distributions of P . aeruginosa populations were noted in wells with smaller diameters ( CV 5μm well = 0.68 ± 0.08; CV 20μm well = 0.28 ± 0.04, n = 4), consistent with the trends noted previously ( Fig 3B ). The fluorescent signals measured from wells during incubation indicated that growth occurred over a 10 hr period. Cells in 20 μm diameter wells had a 4 hr lag phase followed by 4 hr growth period until the cell volume reached the final volume of the well ( Fig 5 A , S3 Video ). A small over-estimation in cell volume, indicated by cell volume fractions slightly higher than 1.0, was noted at late growth times, and was likely caused by increases in cell density as cells fill the entire well volume. Similar effects have been noted while monitoring P . aeruginosa growth in other confined systems [ 14 ]. After growth, fluorescence intensity remained stable for at least 40 hr. Upon inspection of these wells under fluorescence, bright cellular aggregates appeared to be present, but no cellular motility could be detected within the cell mass. An example of 20 μm diameter wells after growth is shown in real-time playback ( S4 Video ). The lack of cellular motility here is in stark contrast to that seen immediately after seeding ( S1 and S2 Videos), where high cellular-motility is easily observed. This suggests the possible establishment of biofilms within the wells during the growth period. Also apparent is the presence of an outlier well, which occurred in the 20 μm diameter arrays with a frequency of 5–10% (n = 4). However, the relatively low variation in initial cell populations assembled under these seeding conditions allowed for directed, repeatable growth trajectories, which enables averages of trajectories to be determined. This demonstrates the utility of this platform as a ‘high-statistics’ method for monitoring growth kinetics of replicate, microscale bacterial populations in controlled microenvironments. Cells in 5 and 10 μm diameter microwell arrays showed strikingly different growth behavior. In 5 μm diameter wells, a wide variety of trajectories were measured ( Fig 5 B , S5 Video ), ranging from an increase in signal intensity due to growth (< 4 hr to stationary phase), to population decay and extinction, indicated by a decaying fluorescence signal. The decay in fluorescent signal is due to cell lysis and extracellular GFP diffusion, which was previously shown using 60X time-lapse fluorescent microscopy to monitor individual cells during lysis [ 41 ]. Similar behavior was also noted in 10 μm diameter wells, and replicate experiments consistently showed this behavior to persist within smaller (5, 10 μm diameter) wells, whereas 20 μm diameter wells provided reproducible growth outcomes ( S4 Fig ). 10.1371/journal.pone.0155080.g005 Fig 5 P . aeruginosa growth trajectories in 5 and 20 μm diameter microwell arrays. (A) Top: False-color fluorescent images of growth in 20 μm diameter arrays. Bottom: Corresponding growth trajectories. The dashed red trajectory indicates growth in an outlier well. (B) Top: False-color fluorescent images of growth in 5 μm diameter arrays. Solid black trajectories denote wells where growth and colonization occurred, dashed red trajectories denote wells where decay and extinction occurred. Data is representative of 4 independent growth experiments. We hypothesized that the variation in growth response in the 5 and 10 μm diameter wells could be attributed to the broader distributions of initial populations seeded into the wells. To investigate this further, we compared the initial and final (t = 24 hrs incubation) cell volumes from 5 and 10 μm wells (n = 4 experiments), generating hundreds of independent growth trials ( Fig 6 A and 6 B ). Additionally, the probability of well colonization, defined as the percentage of wells showing a significant increase in cell volume after incubation, was computed ( Fig 6 C and 6 D ). As evident, the probability for colonization was highly dependent on initial conditions. In the case of the 10 μm diameter wells, colonization was possible when wells were inoculated up to a volume fraction of 0.3. Within this region, an inoculation range where the probability of colonization was the highest appears at cell volume fractions between 0.1 and 0.15. Comparable trends were found within the 5 μm wells, where the probability of colonization was highest at cell volume fractions between 0.01 and 0.1 and diminished between 0.1 and 0.3. In both cases, colonization did not occur when initial well volumes were greater than 0.3. It is likely that decay occurs at this inoculum level due to over-consumption of resources during the early stages of growth, since nutrient exchange is limited in the confined environments. Although additional analysis is required to test this hypothesis, these findings point towards the importance of spatial confinement in community development. In addition, these results demonstrate the first successful application of this platform to screen hundreds of unique, independent bacterial populations, driven by highly heterogeneous population assembly in small wells, for the discovery of populations and environmental conditions that influence colony growth. 10.1371/journal.pone.0155080.g006 Fig 6 Growth of P . aeruginosa in confined volumes depends on inoculum levels. (A) Scatter plots of initial and final (t = 24 hrs) cell volume fraction in 10 μm diameter wells and (B) 5 μm diameter wells. Growth-decay line deciphers wells that increased or decreased in cell numbers over the incubation period. (C) Probability of well colonization with initial volume fraction of seeded cells for in 10 μm diameter and (D) 5 μm diameter wells. Data was taken from n = 256 wells for 10 μm diameter arrays and n = 840 wells for 5 μm diameter arrays from 4 independent growth experiments. Assembly of Multi-Component Bacterial Communities In addition to single-species colonies, dynamic interactions within multi-species populations can be investigated using this platform. Microbial communities are often shaped by cooperative, competitive, or pathogenic interactions between different species, and recently several pair-wise interactions have been shown to be critical in driving community phenotype [ 42 ]. However, the vast majority of interactions occurring within poly-microbial communities are unknown, but likely depend on the relative abundance of interacting members present [ 16 ]. The high-throughput nature inherent to this method makes it attractive for characterizing inter-species interactions after assembling either homogenous or heterogeneous populations, similar to the single-species systems previously described. Here, the assembly of a model, two-component E . coli system constitutively expressing mCherry or GFP into large (40 μm diameter) wells promoting homogenous population assembly, or small (2 μm diameter) wells promoting heterogeneous population assembly was examined. In large wells, assembled E . coli -mCherry and GFP populations had a reproducible mCherry (red) to GFP (green) signal ratio of 0.39 ± 0.09 ( Fig 7 A ), reflective of the ratio at which the pair was mixed together in solution, demonstrating pairing at low dispersity. In stark contrast, these cells were paired with high dispersity in small wells, providing a highly heterogeneous distribution of initial populations, as noted by a variety of unique GFP and mCherry signatures ( Fig 7 B ). A scatter plot of GFP-mCherry signal intensities generated from individual wells in each case contrasts the differences between initial population dispersions ( Fig 7 C ). This finding demonstrates the potential use of this platform to measure growth and interactions between replicate, multi-member populations, or to screen unique combinations of interacting pairs in a manner similar to the single-component systems previously described. This unlocks a new, transformative approach for studying fitness, competition, mutualism, or pathogenicity across a vast parameter space using a single substrate. 10.1371/journal.pone.0155080.g007 Fig 7 Multi-member bacterial communities can be assembled at low or high dispersion. (A) Low dispersion pairing: Seeding a 1:9 mixture of E . coli -mCherry (red) and E . coli -GFP (green) at an overall OD 600 of 0.4 into 40 μm diameter microwell arrays. (B) High dispersion pairing: Seeding a 1:1 mixture of E . coli -mCherry and E . coli -GFP into 2 μm diameter arrays at an overall OD 600 of 1.0. (C) Scatter plot of GFP and mCherry signals after low or high dispersion pairing."
} | 7,583 |
38940183 | PMC11211831 | pmc | 9,601 | {
"abstract": "Abstract Summary Shotgun metagenomics allows for direct analysis of microbial community genetics, but scalable computational methods for the recovery of bacterial strain genomes from microbiomes remains a key challenge. We introduce Floria, a novel method designed for rapid and accurate recovery of strain haplotypes from short and long-read metagenome sequencing data, based on minimum error correction (MEC) read clustering and a strain-preserving network flow model. Floria can function as a standalone haplotyping method, outputting alleles and reads that co-occur on the same strain, as well as an end-to-end read-to-assembly pipeline (Floria-PL) for strain-level assembly. Benchmarking evaluations on synthetic metagenomes show that Floria is > 3 × faster and recovers 21% more strain content than base-level assembly methods (Strainberry) while being over an order of magnitude faster when only phasing is required. Applying Floria to a set of 109 deeply sequenced nanopore metagenomes took <20 min on average per sample and identified several species that have consistent strain heterogeneity. Applying Floria’s short-read haplotyping to a longitudinal gut metagenomics dataset revealed a dynamic multi-strain Anaerostipes hadrus community with frequent strain loss and emergence events over 636 days. With Floria, accurate haplotyping of metagenomic datasets takes mere minutes on standard workstations, paving the way for extensive strain-level metagenomic analyses. Availability and implementation Floria is available at https://github.com/bluenote-1577/floria , and the Floria-PL pipeline is available at https://github.com/jsgounot/Floria_analysis_workflow along with code for reproducing the benchmarks.",
"conclusion": "4 Conclusion The importance of strain-level analyses, combined with ever-increasing amounts of sequencing data, implies that accurate and efficient methods for unraveling strains will be of interest for the foreseeable future. In response, we developed Floria, a metagenome haplotype phasing tool, along with an associated assembly pipeline that can rapidly extract microbial haplotypes for downstream analysis. Floria’s phasing and assembly capabilities provide a path forward for large-scale analyses of diverse metagenomic data with even short and noisy long-reads. For future work, hybrid short and noisy long-read approaches for haplotyping to leverage the throughput and accuracy of short reads would be interesting. However, it should be noted that the accuracy of noisy long-reads is continuously improving. Another potential improvement to Floria is considering indels and structural variations in the phasing process, as currently only SNPs are used.",
"introduction": "1 Introduction To accurately assess the full genetic potential of microbial communities and capture their evolutionary and ecological dynamics, it is often necessary to resolve genomes at the strain level ( Goyal et al. 2022 ). This is because there can be significant phenotypic variation between different strains of the same species ( Pierce and Bernstein 2016 , van Opijnen et al. 2016 ), e.g. Escherichia coli being either pathogenetic ( Leimbach et al. 2013 ) or probiotic ( Sonnenborn 2016 ) in a strain-dependent manner. Thus, disambiguating strains within human microbiomes has important implications for human health and precision medicine ( Fukuda et al. 2011 , Federici et al. 2022 ). Yet, many widely used metagenomic workflows ( Uritskiy et al. 2018 , Moss et al. 2020 ) are not designed to recover multiple highly similar strain genomes. This often results in assemblies where less abundant strains are not recovered, potentially missing out on ecologically and medically important genetic features of the microbial diversity present. Computational strain recovery comes in different forms. Strain haplotyping (or phasing ) is the recovery of alleles, such as SNPs, that co-occur along the same chromosome ( Browning and Browning 2011 ), or in the case of haploid microbes, the same strain in a community (haplotype) ( Nicholls et al. 2021 ). We are interested in using read overlap information to link alleles along the same strain haplotype. In this case, by clustering reads into strain-level clusters, one can also recover the sequence of alleles for each strain ( Bonizzoni et al. 2016 ). On the other hand, strain-level assembly is the recovery of all genomic content for each strain by base-level de novo assembly ( Feng et al. 2022 , Benoit et al. 2024 ). An orthogonal approach is strain-level profiling , where reference genomes from a database are used to identify corresponding strains from a metagenome ( Francis et al. 2013 , Dilthey et al. 2019 , van Dijk et al. 2022 ). Profiling assumes the existence of a reference genome for each strain in the population and does not reconstruct haplotypes explicitly. In this work, we will focus instead on phasing and assembly. Short reads are limited in their ability to resolve between-strain similarities required to construct high-contiguity strain-level assemblies, but the advent of long-read Oxford Nanopore and PacBio sequencing has opened the door for more complete strain resolution. In particular, PacBio HiFi-based assembly methods leverage highly accurate, but more expensive, long reads for strain-level assemblies ( Feng et al. 2022 , Benoit et al. 2024 ), but HiFi reads are often not available for population-scale cohorts due to its cost. Therefore, methods that work well on long-reads with higher error rates, as well as short-reads, are still desirable despite their inherent limitations. For nanopore reads, strain-level assembly is a challenging task at lower coverages or when using older, less accurate chemistries. Even with the Oxford Nanopore’s newest R10.4 flow cell, 40 × coverage is still recommended ( Sereika et al. 2022 ). Strain-level assembly is also computationally expensive for all technologies. In contrast, because phasing requires resolution of only a sparse subset of alleles, it is more feasible for sequencing data with lower accuracy, while being more computationally efficient than assembly based approaches. Thus, phasing is a very useful task when strain-level assembly may give low-quality contigs or is too time-consuming. However, most existing haplotyping methods have only been designed and tested for short-reads [e.g. DESMAN ( Quince et al. 2017 ), ConStrains ( Luo et al. 2015 ), Gretel ( Nicholls et al. 2021 ), EVORhA ( Pulido-Tamayo et al. 2015 ), BHap ( Li et al. 2019 )]. A few long-read strain-level assemblers with built-in haplotyping capabilities exist, but they have yet to be applied to large cohorts with deep metagenome sequencing data [e.g. Strainberry ( Vicedomini et al. 2021 ), Strainy ( Kazantseva et al. 2023 )]. In this work, we introduce Floria , a novel strain haplotyping algorithm that can take both long and short-read metagenomic datasets as input. With Floria, the haplotype phasing task takes only minutes on a standard workstation, an order of magnitude faster than assembly, enabling large-scale recovery of microbial haplotypes. Additionally, our package Floria-PL uses the Floria haplotyper to provide a one-command strain-level assembly pipeline for complex metagenomes that is more accurate and several times faster than existing methods."
} | 1,841 |
22660323 | null | s2 | 9,602 | {
"abstract": "Programmed self-assembly of strands of nucleic acid has proved highly effective for creating a wide range of structures with desired shapes. A particularly successful implementation is DNA origami, in which a long scaffold strand is folded by hundreds of short auxiliary strands into a complex shape. Modular strategies are in principle simpler and more versatile and have been used to assemble DNA or RNA tiles into periodic and algorithmic two-dimensional lattices, extended ribbons and tubes, three-dimensional crystals, polyhedra and simple finite two-dimensional shapes. But creating finite yet complex shapes from a large number of uniquely addressable tiles remains challenging. Here we solve this problem with the simplest tile form, a 'single-stranded tile' (SST) that consists of a 42-base strand of DNA composed entirely of concatenated sticky ends and that binds to four local neighbours during self-assembly. Although ribbons and tubes with controlled circumferences have been created using the SST approach, we extend it to assemble complex two-dimensional shapes and tubes from hundreds (in some cases more than one thousand) distinct tiles. Our main design feature is a self-assembled rectangle that serves as a molecular canvas, with each of its constituent SST strands--folded into a 3 nm-by-7 nm tile and attached to four neighbouring tiles--acting as a pixel. A desired shape, drawn on the canvas, is then produced by one-pot annealing of all those strands that correspond to pixels covered by the target shape; the remaining strands are excluded. We implement the strategy with a master strand collection that corresponds to a 310-pixel canvas, and then use appropriate strand subsets to construct 107 distinct and complex two-dimensional shapes, thereby establishing SST assembly as a simple, modular and robust framework for constructing nanostructures with prescribed shapes from short synthetic DNA strands."
} | 483 |
26331158 | null | s2 | 9,604 | {
"abstract": "Two label-free molecular imaging techniques, confocal Raman microscopy (CRM) and secondary ion mass spectrometry (SIMS), are combined for in situ characterization of the spatiotemporal distributions of quinolone metabolites and signaling molecules in communities of the pathogenic bacterium Pseudomonas aeruginosa. Dramatic molecular differences are observed between planktonic and biofilm modes of growth for these bacteria. We observe patterned aggregation and a high abundance of N-oxide quinolines in early biofilms and swarm zones of P. aeruginosa, while the concentrations of these secreted components in planktonic cells and agar plate colonies are below CRM and SIMS detection limits. CRM, in conjunction with principal component analysis (PCA) is used to distinguish between the two co-localized isomeric analyte pairs 4-hydroxy-2-heptylquinoline-N-oxide (HQNO)/2-heptyl-3-hydroxyquinolone (PQS) and 4-hydroxy-2-nonylquinoline-N-oxide (NQNO)/2-nonyl-hydroxyquinolone (C9-PQS) based on differences in their vibrational fingerprints, illustrating how the technique can be used to guide tandem-MS and tandem-MS imaging analysis. Because N-oxide quinolines are ubiquitous and expressed early in biofilms, these analytes may be fundamentally important for early biofilm formation and the growth and organization of P. aeruginosa microbial communities. This study underscores the advantages of using multimodal molecular imaging to study complex biological systems."
} | 367 |
36144045 | PMC9501207 | pmc | 9,606 | {
"abstract": "Recent advances in precision manufacturing technology and a thorough understanding of the properties of piezoelectric materials have made it possible for researchers to develop innovative microrobotic systems, which draw more attention to the challenges of utilizing microrobots in areas that are inaccessible to ordinary robots. This review paper provides an overview of the recent advances in the application of piezoelectric materials in microrobots. The challenges of microrobots in the direction of autonomy are categorized into four sections: mechanisms, power, sensing, and control. In each section, innovative research ideas are presented to inspire researchers in their prospective microrobot designs according to specific applications. Novel mechanisms for the mobility of piezoelectric microrobots are reviewed and described. Additionally, as the piezoelectric micro-actuators require high-voltage electronics and onboard power supplies, we review ways of energy harvesting technology and lightweight micro-sensing mechanisms that contain piezoelectric devices to provide feedback, facilitating the use of control strategies to achieve the autonomous untethered movement of microrobots.",
"conclusion": "6. Conclusions In this review, we summarize the recent advances in the application of piezoelectric materials in developing microrobots. A variety of mechanisms are presented with the details of the weight and speed they can reach. The microrobotic field is advancing toward making autonomous untethered microrobots where one challenge is to meet the high voltage requirement for piezoelectric actuators. Here, we review the methods of untethered mobility with power harvesting that would benefit the design of next-generation microrobots. Furthermore, for nature-inspired autonomy, the microrobots require adequate sensing and actuation to formulate a sense of the environment and incorporate control strategies to stabilize the system. Adding a sensor to the microrobot can significantly increase the weight of the robot. Therefore, we conduct reviews of practical approaches along with piezoelectric sensing and control methods to achieve stable orientations and accurate trajectories. As wireless power transmission technologies advance and the sensors become lighter and more accurate, by overcoming the design and fabrication challenges, untethered autonomous piezoelectric microrobots will gain more popularity.",
"introduction": "1. Introduction In recent years, there has been an increasing effort to utilize piezoelectric materials in the development of microrobotic systems. With the growing number of mechanisms that scientists use to develop microrobots, there is a great challenge in designing lighter and more efficient microrobots to achieve a certain level of autonomy. Important reviews on microrobotics have been published with different focuses, such as micro-scale flapping-wing robots [ 1 ], biohybrid microrobots [ 2 ], light-powered microswimmers [ 3 ] and drug delivery microrobots [ 4 ], etc. Here, we summarize the recent application of piezoelectric materials for the development of microrobots in the direction of autonomy based on the areas corresponding to its basic challenges, including mechanism, power, sensing, and control. Piezoelectric materials are widely employed in precision motion due to their distinctive advantages such as quick response, high displacement resolution, high stiffness, high actuating force, and little heat generation [ 5 , 6 ]. These features make piezoelectric materials good candidates for developing the actuating module of microrobots. In 2006, Anton and Sodano [ 7 ] reviewed the literature (2003–2006) on power harvesting using piezoelectric materials for self-powered wireless sensor applications, and they updated their review with Safaei [ 8 ] to include the literature from 2008 to 2018. Moreover, Mahapatra et al. [ 9 ] reviewed the nanostructures of piezoelectric materials, manufacturing methods, and material-specific underpinning concepts. The application of piezoelectric actuators is discussed more specifically in areas such as medical and robotics engineering by Uchino [ 10 ]. In 2018, Shevtsov et al. [ 11 ] discussed the mathematical modeling, experimental techniques, and computer algorithms for piezoelectric generators. They included the particular effects of piezoceramics, such as the flexoelectric effect, and methods for defect identification. As piezoelectric materials development advanced, computational methods were proposed to contain certain phenomena, including rate-dependent switching in the micromechanical 3D finite element model [ 12 ]. The mechanism by which piezoelectric microrobots achieve mobility, as well as the environment in which they are expected to maneuver, are important in the development of microrobots. Each mechanism has specific characteristics suitable for an objective environment. Ambulatory locomotion gives the advantages of mobility on rough surfaces as opposed to the traditional wheeled mechanism [ 13 ]. Moreover, increasing the number of legs enables the system to be more robust due to the actuation failure [ 14 ]. The inchworm mechanism can gain control of the friction force by exploiting the squeeze film effect [ 15 , 16 ]. To create biologically-inspired flapping-wing microrobots like insects for exploration purposes, high-density actuation power [ 17 ] is required, which can be developed using piezoelectric materials. Additionally, amphibious microrobots are designed to conform to the multi-environment [ 18 ]. The key roles in the operation of the microrobot are the power source to achieve mobility and the way it is transferred to the microrobot. The piezoelectric materials that are used in microrobots require high input voltages, which can reach as high as 220 V [ 19 ], creating challenges for power transmission. As a promising strategy, there are a variety of methods based on energy harvesting in the direction of the wireless functionality of microrobots relying on piezoelectric actuation [ 20 ]. The sensing capabilities involve microrobots or piezo-based devices dealing with their environments to achieve autonomy like their biological counterparts, such as tactile sensing similar to that of nature-inspired insects [ 21 ]. Additionally, piezoelectric sensing is investigated in a range of fields such as detecting cracks [ 22 ] and human health monitoring [ 23 ], which could be used to inspire ideas for microrobotic applications. The control strategy is essential for the microrobots to achieve stability and follow trajectories. Due to their lightweight and miniature sizes, microrobots are more sensitive to environmental disturbances. Researchers have attempted to address these challenges by adding dampers [ 24 ], taking into account the disturbances [ 25 ], and using adaptive, model-free MIMO, nonlinear, and spiking neural network control strategies [ 26 , 27 , 28 , 29 , 30 ]. Moreover, the augmentation of accurate sensors to the system helps enhance the stability control of the microrobot. In this review, we intend to give an overview of the latest advances in the field of piezoelectric microrobots focusing on their innovation and limiting factors. Therefore, we present the up-to-date applications of piezoelectric mechanisms to the development of microrobots with a focus on power, sensing, and control, so as to recognize the challenges that need more consideration and contribute to the understanding, design, and fabrication of piezoelectric microrobots."
} | 1,878 |
38807528 | PMC11338568 | pmc | 9,607 | {
"abstract": "Human crowds display various self-organized collective behaviours, such as the spontaneous formation of unidirectional lanes in bidirectional pedestrian flows. In addition, parts of pedestrians’ footsteps are known to be spontaneously synchronized in one-dimensional, single-file crowds. However, footstep synchronization in crowds with more freedom of movement remains unclear. We conducted experiments on bidirectional pedestrian flows (24 pedestrians in each group) and examined the relationship between collective footsteps and self-organized lane formation. Unlike in previous studies, pedestrians did not spontaneously synchronize their footsteps unless following external auditory cues. Moreover, footstep synchronization generated by external cues disturbed the flexibility of pedestrians’ lateral movements and increased the structural instability of spatial organization. These results imply that, without external cues, pedestrians marching out of step with each other can efficiently self-organize into robust structures. Understanding how asynchronous individuals contribute to ordered collective behaviour might bring innovative perspectives to research fields concerned with self-organizing systems.",
"introduction": "1 . \n Introduction Highly organized collective behaviour exhibited by a large number of people, as seen in a marching band, impresses spectators. Such disciplined behaviour is orchestrated through a predetermined plan and/or guided by an external conductor from a global perspective. However, collective behaviour can also organize spontaneously in nature. Witnessing massive flocks of birds and schools of fish can be an amazing sight [ 1 ]. It is also fascinating that many people can come and go on crowded streets of cities worldwide without colliding. In these behaviours, there are no predetermined plans or external conductors and individuals behave relatively freely. Instead of being driven by external global forces, they self-organize through local interactions among group members [ 2 ]. Human crowds, in particular, have attracted the attention of researchers in various fields, with the aim of helping to manage mass events and daily pedestrian transportation [ 3 ]. Some collective patterns of organization bring about beneficial results to the group, such as lane formation, where unidirectional lanes are spontaneously formed in bidirectional pedestrian flows in crowded streets or crossings, which increases the efficiency of traffic flow [ 4 – 7 ]. On the other hand, if people in a crowd create collective patterns of motion in other ways, such as crowd turbulence, they become uncontrollable and can lead to serious disasters [ 8 , 9 ]. Although providing traffic information and guidance is important to proactively prevent crowd disasters, it is difficult to restrict and control the movements of people who have happened to gather together [ 3 ]. To accomplish crowd management that facilitates traits of pedestrian’s movements that contribute to functional self-organization, it is essential to understand the mechanisms underlying collective human behaviour. Some recent experiments have suggested that some of the footsteps of pedestrians in a group are synchronized, some of which are in-phase and some are anti-phase. For example, their footsteps can spontaneously synchronize [ 10 – 14 ] even without indirect interactions among pedestrians via a wobbling walkway, as in the case of London Millennium Bridge, where pedestrians fell into step via vibrations of the suspension footbridge [ 15 – 17 ]. Most previous studies on spontaneous footstep synchronization in human crowds have adopted a one-dimensional approach known as the single-file experiment [ 12 – 14 ]. In this set-up, pedestrians walk in the same direction along a narrow, circular-shaped corridor, which prohibits lateral movements and overtaking and enables researchers to extract elementary forms of interactions between consecutive pedestrians. The footstep synchronization observed in this manner can serve as an optimization strategy, where simultaneous movements of each pedestrian’s same-side feet efficiently exploit limited spatial resources in a jammed situation, reducing collisions and enhancing overall flow [ 12 – 14 ]. The assumption of functional footstep synchronization has also been supported by various single-file experiments where pedestrians walk while listening to or following external sounds (e.g. music or a sound at a constant tempo) [ 18 – 20 ]. However, in the one-dimensional approach, the stability of the single-file structure is guaranteed by a boundary condition (i.e. a narrow corridor with borders). In contrast, most pedestrians’ motions in daily life are not restricted to one dimension, and pedestrians can spontaneously generate ordered structures in the absence of external forces, such as those observed with lane formation phenomena [ 4 – 7 ]. In addition to making adjustments to preceding and following individuals, pedestrians commonly use lateral movements under two-dimensional conditions [ 4 , 6 ]. Pedestrians dynamically modify their configurations to avoid collisions with oncoming pedestrians and to overtake others. It is possible that footstep synchronization may be an optimal strategy to exploit spatial resources with a preceding individual only when the single-file structure remains static. To further investigate the role of collective footsteps in human crowds, it is essential to investigate pedestrian behaviour under conditions that allow them to move with more freedom of movement than available in one-dimensional experiments. In this study, we conducted experiments to examine lane formation phenomena in a crossing scenario by means of tracking pedestrians’ positions in two dimensions and recording their foot movements. Lane formation can be an ideal system to investigate collective footsteps in two dimensions because pedestrians spontaneously segregate into multiple lanes, which are not fixed static structures, and lateral movements dynamically contribute to self-organization. In fact, previous experiments revealed the intrinsic instability of bidirectional flow organization owing to overtaking behaviour and lateral movements [ 4 ]. Furthermore, in another experiment, during the development of lanes, lateral explorations from the direct straight path to the destination inevitably occurred as pedestrians passed through a crowd, avoided oncoming pedestrians, and thereby achieved lane formation [ 6 ]. In these previous experiments, pedestrians did not follow any external cues to temporally adjust their movements, and moving at their own timing may have facilitated such lateral exploration. In other words, aligning steps with external cues might restrict lateral fluctuations in a two-dimensional scenario. To address this possibility, we set an experimental condition in which participants were asked to align their steps with external auditory cues and compared the results with those of another condition without any temporal cues. We find that pedestrians do not spontaneously align their footsteps with each other unless there is an external conductor. Moreover, although we observed that the external cues could produce an organized structure, they also yield potential instability, which was characterized by an increased number of pedestrian lanes formed, a shorter time/distance to potential collisions and a larger rotation of pedestrians’ shoulders. Furthermore, we find that aligning steps with external cues decreases the lateral exploratory motion of pedestrians, suggesting a relation to the width of generated lanes. These findings shed light on the importance of asynchronous motions enabling exploratory lateral behaviour and its contributions to robustly self-organizing human crowds. We also discuss the relation between asynchronous motions and theoretical models that include discrete positional updates.",
"discussion": "3 . \n Discussion In this study, we addressed whether footstep synchronization occurs in pedestrian crowds with relatively more freedom of movement and contributes to spatial self-organization. To this end, we conducted bidirectional flow experiments. One condition had external auditory stimuli to be followed by pedestrians and the other did not. Footstep synchronization was only observed in the former condition. This synchronization driven by the external auditory cues increased the number of lanes formed, causing structural instability and potential collision risks. It also restricted the pedestrians’ usual exploratory lateral movements, especially in the middle of the lane formation process. These results imply that the pedestrians with no external conductors did not spontaneously synchronize; instead, they were able to deviate laterally, which appeared to promote robust formation of groups. In contrast, the footstep synchronization driven by the external cues disturbed the process of robust lane formation. In previous studies, pedestrians have been shown to synchronize their foot motions in single-file experiments [ 12 – 14 ], where the mechanism for synchronization was attributed to pedestrians stepping in unison on the same line to efficiently exploit spatial resources and avoid collisions with preceding pedestrians [ 14 ]. However, a prerequisite for this mechanism to function (i.e. the single-file structure guaranteed by the boundary condition) was not met under our experimental conditions, which allowed for more freedom of movement. In our scenario, pedestrians can easily move laterally to avoid collisions. Moreover, overtaking behaviour and the presence of oncoming pedestrians also lead to lateral movements, distracting pedestrians from solely focusing on preceding pedestrians. In addition, pairs of pedestrians in a non-crowded open space spontaneously synchronize their stepping, presumably through auditory feedback of steps [ 10 , 11 ]. However, it would be difficult for pedestrians to hear the sound of steps in our crowded experimental conditions. Finally, it is possible that when pedestrians are moving toward each other, asynchronous stepping would facilitate smooth collision avoidance, because pedestrians who move later find it easier to adjust their movements relative to those who moved first [ 25 ]. Thus, it seems rather natural to consider that pedestrians in a crowd with more freedom of movement take steps asynchronously. Note that the degree of the synchronization in our experiment did not increase with time ( figure 3 ). Thus, the lack of synchronization in the experiment without an external auditory cue is most likely not a result of the amount of time allotted (i.e. a longer period of time would not have changed the results). Most footstep synchronization in crowds is likely to be induced by indirect interactions such as the pedestrians’ interactions observed in the crowd incident on Millennium Bridge [ 16 ]. Our study revealed that the temporal pattern of stepping motions influenced the spatial motions of pedestrians. At the individual level, we observed that the variability of lateral fluctuations of an individual’s trajectory from the direct path to the destination decreased when pedestrians followed the external auditory cue ( figure 5 ). At the collective level, the external cue increased the number of generated lanes ( figure 4 a \n ). Lateral movements can play an important role in the process of pattern formation. In the condition with no cues, moving laterally enables pedestrians that are more distant from each other in the same group to come closer, which contributes to formation of wider lanes. In the condition with external auditory cues, there is less lateral movement, allowing pedestrians to cut between oncoming pedestrians if there is enough space, thereby creating more (thinner) lanes. Moreover, analyses of time/distance to potential collisions and shoulder rotation ( figure 4 b,c \n and electronic supplementary material, figure S4) support the idea that the width of lanes corresponds to the robustness of the generated structure. This result is consistent with a previous theoretical study showing that a larger number of lanes (i.e. narrower lanes) has a shorter lifetime because of the larger fraction of pedestrians at contact surfaces between lanes moving in the opposite direction [ 21 ]. Our findings, therefore, highlight the possibility that external cues disturb the process of lane formation and emphasize the importance of exploratory lateral movements in promoting robust self-organization in human crowds. In addition, pedestrians spontaneously coordinated their steps in terms of duration (or frequency). Indeed, without external cues, the differences in duration between two pedestrians in actual pairs were significantly smaller than those in random pairs ( figure 2 c \n , top). Although additional data are required, this coordination may occur because two paired pedestrians in the same group have to adjust their movements when merging into a single lane so that they are not cut off by oncoming pedestrians. On the other hand, with external cues, the differences in durations of actual pairs were relatively narrowly distributed, but there was no significant difference in the durations of actual and random pairs ( figure 2 c \n , bottom). This simply indicates that pedestrians followed external cues well, and hence, step frequencies did not vary much for any individual in the crowd. Human locomotion is basically updated by each discrete footstep owing to the biomechanics of the bipedal gait [ 26 ]. This can be likened to discrete positional updates in computational models. Most previous models of human crowd behaviour assume synchronous position updates, in which all pedestrians simultaneously update their positions [ 27 , 28 ] (but see [ 29 , 30 ]). However, because we observed asynchronous movements among pedestrians and their influence on self-organization in human crowds, it would be valuable to incorporate characteristics of asynchronous behaviour in computational models in pedestrian crowds. For collective behaviours of other animals, various theoretical models with asynchronous positional updates have been proposed [ 31 – 38 ]. For example, asynchrony in position updates has been suggested to allow anisotropy to emerge in interactions among individuals and to generate inherent noise, which drives autonomous motion in groups without external noise, as has been observed in real animal groups [ 33 , 34 ]. Moreover, in some theoretical work on non-human animal collective behaviour, asynchrony plays important roles for anticipatory interactions among individuals [ 35 – 38 ], and these are also fundamental in pedestrian interactions [ 7 ]. We expect that our results may provide quantitative support to asynchronous models that incorporate anticipation. Enhanced human crowd models that integrate asynchronous behaviours encompassing lateral movements will offer a more accurate depiction and prediction of pedestrian flows. There are, however, limitations to this work. In our study, experiments were conducted on bidirectional flow to investigate the relationship between footstep synchronization and emergent pattern formation. Therefore, we did not test whether synchronization occurs in the unidirectional flow of a single group. Moreover, while we adopted a previously used experimental setting for bidirectional flows [ 5 – 7 ], we did not verify the dependence of footstep synchronization with the density of crowds. Also, we only employed a single tempo as the external cue. In future experiments, it will be important to explore the effect of different external cues with various tempos. For example, in a single-file experiment, it is known that a slower-paced external cue can improve the flow [ 18 ], but it is unknown whether this is true in crowds with more freedom of movement. Furthermore, our experiments focused on comparatively short-term bidirectional flow, similar to that observed at busy crossings. However, there can be long-term bidirectional flow, for example, on streets with many shops, where the structural instability of lanes manifests [ 4 ] and lateral movements potentially have more influence than short-term bidirectional flow."
} | 4,061 |
29192930 | null | s2 | 9,608 | {
"abstract": "We report here that a dense liquid formed by spontaneous condensation, also known as simple coacervation, of a single mussel foot protein-3S-mimicking peptide exhibits properties critical for underwater adhesion. A structurally homogeneous coacervate is deposited on underwater surfaces as micrometer-thick layers, and, after compression, displays orders of magnitude higher underwater adhesion at 2 N m"
} | 100 |
37163596 | PMC10171810 | pmc | 9,610 | {
"abstract": "Earth’s life-sustaining oceans harbor diverse bacterial communities that display varying composition across time and space. While particular patterns of variation have been linked to a range of factors, unifying rules are lacking, preventing the prediction of future changes. Here, analyzing the distribution of fast- and slow-growing bacteria in ocean datasets spanning seasons, latitude, and depth, we show that higher seawater temperatures universally favor slower-growing taxa, in agreement with theoretical predictions of how temperature-dependent growth rates differentially modulate the impact of mortality on species abundances. Changes in bacterial community structure promoted by temperature are independent of variations in nutrients along spatial and temporal gradients. Our results help explain why slow growers dominate at the ocean surface, during summer, and near the tropics and provide a framework to understand how bacterial communities will change in a warmer world.",
"introduction": "INTRODUCTION Oceans cover 70% of the surface of our planet and are home to myriad bacterial species that are responsible for many of Earth’s crucial biogeochemical functions ( 1 , 2 ), including fixing carbon and nitrogen, recycling nutrients and dissolved organic matter, and degrading biomass. Surveys of the ocean microbiome over broad spatial and temporal scales have revealed repeatable patterns across seasons ( 3 – 6 ), latitude ( 7 ), and depth ( 8 ), which have been linked to a variety of environmental factors, including temperature ( 3 , 4 , 8 , 9 ), water mixing ( 10 , 11 ), nutrient availability ( 12 ), light ( 13 ), and the occurrence of phytoplankton blooms ( 6 ). Nevertheless, general principles underlying the compositional turnover of vital marine bacterial communities are lacking, impairing our ability to predict the response of ocean systems to future environmental changes. The distribution of fast- and slow-growing taxa is a powerful trait-based description of the structure of a bacterial community ( 14 , 15 ), just as it is for multicellular organisms ranging from plants ( 16 ) to corals ( 17 , 18 ). Maximum growth rates are a key component of the diverse life history strategies exhibited by micro- and macro-organisms ( 19 – 23 ) and determine the ability of a species to survive and compete in a given environment ( 24 ). In aquatic bacteria, maximum growth rate differentiates fast-growing copiotrophs, found in nutrient-rich waters, from slow-growing but efficient oligotrophs, which can grow in nutrient-poor environments ( 25 – 29 ). While the role of nutrients in the distribution of fast- and slow-growing bacteria is an area of active study ( 12 , 30 – 32 ), the effects of temperature have been comparatively less explored, despite ample evidence that temperature is a fundamental driver of marine communities ( 3 , 4 , 8 , 33 ). Here, we analyze marine datasets of free-living bacteria from around the world, where temperature varies across seasons, latitude, and depth. We show that higher temperatures are associated with an increase in the relative abundance of slower-growing taxa, with ribosomal RNA (rRNA) gene operon copy number used as a proxy for maximum growth rate. We show that temperature, more than other environmental variables such as nutrients, exerts an overriding effect on community structure and that this effect is robust when controlling for the effects of other variables. The result that slower-growing bacteria are favored by warmer temperatures agrees with predictions that we derive from simple ecological models incorporating bacterial growth as a function of temperature, competition, and mortality. These models show that differences in growth rates matter the most in cold temperatures, where mortality is a greater burden for slower-growing taxa. In warmer temperatures, increased growth rates of all bacteria reduce the differences in the impact of mortality, allowing slower-growing taxa to become more abundant. This concept, which we term mortality burden, illuminates a principle for how temperature determines not only individual microbial niches but also structures multispecies communities.",
"discussion": "DISCUSSION We discovered a macroecological pattern in the distribution of fast- and slow-growing bacteria in natural marine communities along environmental gradients of temperature. This pattern can be explained by increasing temperature favoring slower-growing taxa when competing against faster-growing ones, regardless of whether competition is for nutrients or through direct interference, and independently of growth-efficiency or growth-competitive ability trade-offs. In a system like the ocean where mortality exerts a strong control over bacterial communities, increasing temperatures, by increasing growth rates of all species when within their thermal optima, have the effect of decreasing the mortality burden more for the slower-growing taxa compared to the faster-growing ones, ultimately reducing the competitive abilities of the latter. This result shows that the concept of fast versus slow growth is a powerful way to describe the structure of bacterial communities and illuminates a unifying principle explaining the compositional turnover of marine bacterial communities across seasons, latitude, and depth. Our proposed rule of higher temperatures favoring slower growers is potentially applicable to any community subject to growth and mortality and could underlie some previously observed patterns. Among common cyanobacteria in the ocean, the slow-growing Prochlorococcus dominates near the equator but decreases in abundance toward the poles, where the comparatively fast-growing Synechococcus persists ( 50 ) [see ( 13 , 19 , 51 ) for cyanobacteria growth rate estimates]. Similarly, in a 16-year time series of observations of a phytoplankton community at a nearshore site on the Northeast U.S. Shelf, picoeukaryotes with comparatively faster growth rates were outnumbered by Synechococcus populations during warmer months ( 52 ). Last, in mid-latitude forest habitats, a 20-year-long in situ soil warming experiment resulted in a decrease in the MCN of bacterial communities that was attributed to an assumed decline in nutrients ( 53 , 54 ). Another study measured a massive reduction in RNA content after incubating a soil bacterial community for 1 week at increased temperature compared to an ambient control, despite no difference in nutrients between conditions ( 55 ). The authors attributed the decrease in RNA to an intracellular reduction in ribosome concentration, but a reduction in MCN of the soil community is also consistent with these results. In all of the above examples, we propose that the direct influence of temperature on ecological community composition presents a new interpretation for the observed patterns. While this study has focused on free-living bacteria (as defined by filter fractions between 0.2 and 3 μm), an important question is how temperature affects the dynamics of particle-associated communities (see fig. S13 and table S1), where spatial processes might be important [see ( 56 , 57 ) for theoretical predictions about the role of nutrient concentrations in spatially extended settings]. Another question is whether our conclusions depend on the method for estimating growth rates with rRNA copy number. This method is supported by a positive correlation between copy number and maximum growth rate that has been vetted for many bacterial taxa, including marine ones ( 34 – 36 ); however, the estimation of 16 S rRNA gene operon copies is still limited to taxa for which the full genome has been recovered. To test whether our results could extend beyond this method, we used another method based on codon usage bias ( 58 ), which yielded an overall negative relationship between temperature and weighted mean growth rate (MGR; fig. S14). 16 S rRNA gene copy number and codon usage bias are features encoded in genomes that do not change over ecological time scales and therefore are not expected to change with temperature across seasons or latitudinal gradient. Nevertheless, both copy number and codon usage bias might be under selection over evolutionary time scales, and their distribution across spatial and temporal gradients might reflect evolutionary processes. Together, our study does not exclude the role of other processes in the spatial and temporal turnover of the ocean microbiome but shows that the direct ecological effects of temperature on the distribution of growth strategies have been generally overlooked. Our prediction that increasing temperature favors slower-growing bacteria does not depend on any trade-offs between growth rate and competitive ability. In the terminology of modern coexistence theory ( 59 , 60 ), increasing temperature can have an equalizing effect on species competition, because, as the mortality burden decreases more for the slower grower than for the faster grower, the fitness difference between the two species can also decrease (Supplementary Text). While this result is robustly supported by both the LV competition model and a simple consumer resource model (fig. S4), additional and potentially relevant modeling complications could change our prediction. For example, we assume that increasing temperature increases all species’ growth rates, because we assume the optimum temperature for most marine bacteria to be near 30°C or higher than most in situ temperatures. However, optimum temperature has been shown to decrease in low-nutrient conditions, a result that has consequences for nitrate-limited phytoplankton in a warming ocean with lower anticipated nitrate levels ( 33 ). Whether a reduction in nitrate could have cascading effects upon carbon-limited heterotrophic bacteria needs investigation. As this study has shown, temperature can affect not only individual species but also interactions and thereby entire communities. If a combination of lower nutrients and warming caused ambient ocean temperature to exceed the bacterial optimum, our prediction would reverse, as faster-growing bacteria would be expected to be favored under deteriorating conditions ( 61 ). Connecting unifying rules to biogeographical drivers of bacterial community composition throughout the global ocean is an important challenge, especially considering the effects of climate change ( 62 , 63 ). Many studies show that temperature sets the biogeography of marine bacterial species by imposing constraints on their ability to grow and thereby predict future compositional changes based on the thermal limits of each species ( 33 ). Our study, comprising a simple model and seven datasets, substantially expands upon these predictions by highlighting that, over the broad thermal range of the ocean, temperature has a generic effect on community composition by determining the distribution of fast- and slow-growing taxa. We have presented macroecological patterns in the distribution of growth strategies in bacterial communities across ocean temperature gradients, revealing that slow growers are consistently more abundant around the tropics, during summer, and at the surface of the ocean, in agreement with our theoretical predictions. Overall, our results emphasize that temperature plays a direct role in structuring bacterial community composition and global biogeography of marine bacteria and suggest that warming ocean temperatures may lead to increases in the abundance of slower-growing taxa."
} | 2,877 |
32110877 | PMC7084717 | pmc | 9,611 | {
"abstract": "In this paper, we survey recent advances in the self-assembly processes of novel functional platforms for nanomaterials and biomaterials applications. We provide an organized overview, by analyzing the main factors that influence the formation of organic nanostructured systems, while putting into evidence the main challenges, limitations and emerging approaches in the various fields of nanotechology and biotechnology. We outline how the building blocks properties, the mutual and cooperative interactions, as well as the initial spatial configuration (and environment conditions) play a fundamental role in the construction of efficient nanostructured materials with desired functional properties. The insertion of functional endgroups (such as polymers, peptides or DNA) within the nanostructured units has enormously increased the complexity of morphologies and functions that can be designed in the fabrication of bio-inspired materials capable of mimicking biological activity. However, unwanted or uncontrollable effects originating from unexpected thermodynamic perturbations or complex cooperative interactions interfere at the molecular level with the designed assembly process. Correction and harmonization of unwanted processes is one of the major challenges of the next decades and requires a deeper knowledge and understanding of the key factors that drive the formation of nanomaterials. Self-assembly of nanomaterials still remains a central topic of current research located at the interface between material science and engineering, biotechnology and nanomedicine, and it will continue to stimulate the renewed interest of biologist, physicists and materials engineers by combining the principles of molecular self-assembly with the concept of supramolecular chemistry.",
"conclusion": "8. Conclusions and Future Perspectives We highlighted recent advances of the self-assembly processes employed for the development of new functional platforms for nanomaterials and biomaterials applications. The building block properties (size, morphology, surface functionality), the mutual and cooperative interactions, as well as the initial spatial configuration (and environment conditions) play a crucial role in the design of efficient nanostructured materials and have a great impact on the evolution towards the desired functional products. The organic materials (building blocks) present better characteristics and properties to match the physico-chemical condition encountered in biological tissues and represent the best examples of biocompatible nanostructures. However, despite the variety of smart biomaterials developed in recent years, their effective therapeutic use is often hindered by the complexity of the encountered biological environments that strongly influence the effective functionality of the nanomaterial. Therefore, the identification of the fundamental factors involved in the hierarchical self-assembly of nanostructures represents the initial step that drives towards the development of novel protocols for the construction of novel advanced functional materials.",
"introduction": "1. Introduction Materials self-assembly is a key strategy for the design and fabrication of nanostructured systems and has become a fundamental approach for the construction of advanced materials and their application in the fields of nanomaterials and biotechnology [ 1 , 2 , 3 ]. In nanomaterials self-assembly, the mutual interaction between disordered building blocks drive a material system toward the spontaneous formation of more ordered (or more organized) nano-structured systems. Main examples of the self-assembly method can be found in biomaterials, where the interaction of various macromolecular components and the integration of their actions allow the occurrence of highly specific functions of biological interest [ 4 , 5 , 6 ]. For example, the folding of a polypeptide chain within a protein or the nucleic acids conformational changes into a variety of functional forms are important examples of self-assembly processes involved in many biological functions [ 7 ]. The construction and optimization of functional materials at the atomic and molecular level require a structural control and investigation at the nano-scale and have stimulated a parallel development of instrumentation techniques and observational methods on those size scales. Moreover, the development of multifunctional nanostructures and biomaterials using the recent concepts of nanoarchitectonics utilizes advanced processes for the self-assembly, such as supramolecular chemistry and host–guest processes [ 7 , 8 , 9 ]. The reversibility of the noncovalent forces allows a dynamic switching of the structure and morphology of the nanostructures in response to various (internal and external) stimuli, thus providing additional flexibility for the design and fabrication of versatile smart materials and functional nano-devices [ 10 , 11 , 12 ]. Finally, the new structure and properties obtained by the precise engineering of chemical structures allow the modulation and control of morphology and the efficient use of noncovalent forces (and structure directing interactions) with the introduction of chirality, signal processing and recognition processes [ 7 , 13 ]. In this review we highlight the recent development of the self-assembly approaches with a focus on the main properties and recent applications. We present an organized overview, by analyzing the main parameters that sensitively influence the design of organic nanostructured systems, while putting into evidence challenges, limitations and emerging approaches in the various fields of nanotechology and biotechnology."
} | 1,415 |
30965752 | PMC6418889 | pmc | 9,613 | {
"abstract": "With the integration and miniaturization of electronic devices, thermal management has become a crucial issue that strongly affects their performance, reliability, and lifetime. One of the current interests in polymer-based composites is thermal conductive composites that dissipate the thermal energy produced by electronic, optoelectronic, and photonic devices and systems. Ultrahigh thermal conductivity makes graphene the most promising filler for thermal conductive composites. This article reviews the mechanisms of thermal conduction, the recent advances, and the influencing factors on graphene-polymer composites (GPC). In the end, we also discuss the applications of GPC in thermal engineering. This article summarizes the research on graphene-polymer thermal conductive composites in recent years and provides guidance on the preparation of composites with high thermal conductivity.",
"conclusion": "6. Conclusions In this paper, we have reviewed the graphene-polymer thermal conductive material in recent years. The thermal conductive mechanisms in graphene, polymers, and their composites have been discussed. The recent advances on thermal conductivity of graphene-polymer composites have also been reviewed. Furthermore, we have discussed the factors influencing the thermal conductivity of graphene-polymer composites, such as the characteristics, the loading, the orientation of graphene, and the interface. Finally, the applications of thermal conductive graphene-polymer composites have been demonstrated. This review reveals the relationship between thermal conductive mechanisms and properties and also provides guidance on the preparation of composites with high thermal conductivity.",
"introduction": "1. Introduction Thermal management has become a crucial issue in the modern electronics industry as electronic devices have become more integrated and miniaturized. The power required for some processor modules can reach 250 W in a high-performance computer, leading to heat loads as large as 1 kW in this system [ 1 ]. If the heat can-not be dissipated promptly, the lifetime and the efficiency of the system could be reduced, or even breakdown. In this situation, materials with high thermal conductivity are strongly needed to dissipate the heat and solve the problem [ 2 ]. Polymers have a lot of advantages, such as being lightweight, low cost, easy to process, and exhibiting good corrosion resistance. However, most polymers are heat insulators and have a thermal conductivity between 0.1 and 0.5 W m −1 K −1 [ 3 ], which is due to their amorphous state. There are three kinds of carriers in solids to transport energy: phonons, electrons, and photons [ 4 ]. Phonons are quantized modes of vibration in a rigid crystal lattice, which is the fundamental mechanism of heat conduction in most polymers. Polymers in amorphous state are usually considered to have lots of defects that contribute to numerous phonon scatting, leading to a low thermal conductivity [ 5 ]. In past years, a lot of works have studied thermal conductive polymer-based composites. Many different materials with high thermal conductivity have been used as fillers to improve the thermal conductivity of composites, such as boron nitride (BN) [ 6 , 7 , 8 , 9 ], carbon nanotubes (CNTs) [ 10 , 11 , 12 , 13 , 14 ], aluminum oxide [ 15 , 16 , 17 ], diamond [ 18 , 19 , 20 , 21 ], and graphene [ 22 , 23 ]. Graphene has attracted great attention because of its unique two dimensional (2D) structure and novel properties, such as the zero-gap band structure, high electron mobility, and high thermal conductivity [ 24 ]. Balandin and his co-workers reported a measurement of the thermal conductivity of suspended single-layer graphene around 5000 W m −1 K −1 , which was one of the highest thermal conductivities of currently known materials [ 25 ]. Although there are lots of reviews on the thermal conductivity of polymer-based composites, system summaries on thermal conductive graphene-polymer composites are rare [ 2 , 3 , 4 ]. In this situation, it is necessary to review the advances in the thermal conductivity of graphene-polymer composites. In this article, we review the advances in thermal conductivity of graphene-polymer composites in recent years. Special attention is given to the mechanism, the properties, and the influence factors of graphene-polymer composites. Additionally, we discuss the applications of thermal conductive graphene-polymer composites."
} | 1,108 |
33596827 | PMC7890633 | pmc | 9,614 | {
"abstract": "Background Straw retention is a substitute for chemical fertilizers, which effectively maintain organic matter and improve microbial communities on agricultural land. The purpose of this study was to provide sufficient information on soil fungal community networks and their functions in response to straw retention. Hence, we used quantitative real-time PCR (qRT-PCR), Illumina MiSeq (ITS rRNA) and FUNGuild to examine ITS rRNA gene populations, soil fungal succession and their functions under control (CK) and sugarcane straw retention (SR) treatments at different soil layers (0–10, 10–20, 20–30, and 30–40 cm) in fallow fields. Result The result showed that SR significantly enhanced ITS rRNA gene copy number and Shannon index at 0–10 cm soil depth. Fungi abundance, OTUs number and ACE index decreased with the increasing soil depth. The ANOSIM analysis revealed that the fungal community of SR significantly differed from that of CK. Similarly, significant difference was also observed between topsoil (0–20 cm) and subsoil (20–40 cm). Compared with CK, SR decreased the relative abundance of the pathogen, while increased the proportion of saprotroph. Regarding soil depth, pathogen relative abundance in topsoil was lower than that in subsoil. Besides, both sugarcane straw retention and soil depths (topsoil and subsoil) significantly altered the co-occurrence patterns and fungal keystone taxa closely related to straw decomposition. Furthermore, both SR and topsoil had higher average clustering coefficients (aveCC), negative edges and varied modularity. Conclusions Overall, straw retention improved α-diversity, network structure and fungal community, while reduced soil pathogenic microbes across the entire soil profile. Thus, retaining straw to improve fungal composition, community stability and their functions, in addition to reducing soil-borne pathogens, can be an essential agronomic practice in developing a sustainable agricultural system. Supplementary Information The online version contains supplementary material available at 10.1186/s12866-021-02115-3.",
"conclusion": "Conclusions This study, we demonstrated that the fungal community composition, function, and co-occurrence pattern changed significantly in response to straw retention throughout the soil profile. The straw retention increased the diversity and abundance of fungi in 0–10 cm soil depth. Both straw retention and topsoil had a decreasing effect on the abundance of pathogens. Straw retention and depth of soil influenced the keystone taxa. Overall, these findings enhance our understanding of fungal metabolic functions and networks under straw retention in different soil profiles.",
"discussion": "Discussion Research shown that straw retention can alter soil microbial distribution throughout the soil profile [ 40 ]. Similarly, in this study showed that straw retention improved fungal abundance, especially in topsoil, while it decreased exponentially with increasing soil depth. The previous study has shown that fungi dominated litter-C decomposition, and fungal community composition varied within different soil profiles and controlled resource availability [ 41 ]. It is well documented that different organic materials, especially wheat straw, farm manure [ 42 ], and cow manure compost, change microbial biomass and agricultural land activity [ 2 ]. A similar study showed that straw retention positively impacts cucumber seedlings growth by increasing soil microbial biomass and changing soil microbial community structure [ 43 ]. Increased biodiversity can promote the stability of ecosystems and enhance the mix of basic microbial functions and activities [ 44 ]. Alpha fungus diversity decreased with increasing soil depth [ 45 ], which is supported by our results that straw retention decreased fungal diversity at a soil depth of 0–30 cm than 0–10 cm. Compared with CK, Fungal richness showed no obvious change in response to straw retention, while soil depth was the main driving force to change fungal diversity, which is consistent with previous studies [ 41 , 46 , 47 ]. Fungi play a key role in decomposing recalcitrant substrates [ 48 , 49 ], and their abundance changed under the combined effect of treatment and soil depth . Ascomycota plays a key role in the decomposition of organic substrates [ 30 , 32 , 50 ] and is found to be the main phyla of fungi. Furthermore, SR improved the relative abundance of Ascomycota at different soil depths. Straw retention significantly improved Basidiomycota abundance, which is consistent with previous findings [ 51 ]. As an important decomposer, Basidiomycota produces enzymes (e.g, peroxide) to degrade recalcitrant plant compounds, such as cellulose and lignin [ 50 ]. A previous study showed that straw retention could increase carton content and cellulase activity [ 18 ]. Funguild analysis revealed that straw retention not only altered saprotroph (dung saprotroph-plant saprotroph-wood saprotroph) but also suppressed pathogenic (fungal parasites) (Fig. 4 a). A similar phenomenon was detected in the composting of Chinese herb residues [ 52 ]. Many studies have confirmed that saprotrophs are involved in the decomposition process [ 53 ], and pathogenic fungi commonly acquire nutrients for invading host cells, so they are known to pose a threat to other fungal community members [ 54 ]. Thus, the result showed that straw retention can improve soil nutrient cycles and health. Many investigations indicated that between plant roots and a diverse array of mutualistic endophytic symbionts enhance crop quality, health, and soil nutrition [ 55 , 56 ]. For example, AM can protect plant root and improve plant nutrient absorption capacity [ 57 , 58 ]. Meanwhile, its variability depends on soil depth [ 59 ], same as our result. Pathogens in subsoil were higher than topsoil, which harms other fungal community members This finding corresponds to a previous study, in which the relative abundance of soil pathogenic fungi increased with increasing soil depth [ 45 ]. The ecological network of biological communities has been extensively studied in animal and plant ecology and has recently received microbial ecology attention. However, current research provides insights into the effect of straw retention on changes in fungal communities, with a focus on soil depth. Our findings revealed that the fungal community network in straw retention and topsoil revealed negative edges and modularity network (Table S7 ). If the degree of modularity of the two networks exceeds 0.4, it indicates that these networks are modular [ 60 ]. Many studies have shown that the existence of modularity and negative interactions enhance the stability of disturbed networks [ 61 – 64 ]. The AveCC of SR treatment and topsoil were higher than CK and subsoil, showing that there were more potential connections and small-world behavior. In a small-world network, more OTUs could be available to all other OTUs via a comparatively short path [ 65 ]. On the contrary, the more connected the network is, the more it can contribute to effective and efficient carbon utilization [ 65 ]. Betweennes centrality scores indicates how often a node is found on the shortest path between two nodes in the network to connect it to each other, the higher number, the more connected they are. Network analysis showed that Ascomycota was identified as the main phylum for straw retention and soil depth, indicating that they played an important role in maintaining the function and structure of the ecological community. Keystone taxa were correlated with the C and N cycle in the CK and SR treatments. In the CK treatment, the number of Pleosporales was higher in the CK than SR. Species of Pleosporales occurred in various habitats, that can be epiphytes, endophytes or parasites of living leaves or stems, hyperparasites on fungi or insects, lichenized, or are saprobes of dead plant stems, leaves or bark) [ 66 ]. While Hypocreales order was present in SR treatment. Sordariomycetes are soft-rot fungi, which are well known to effectively decompose organic substrates such as cellulose, cellobiose and lignin [ 67 ]. These results showed that the keystone taxa were involved in various carbon and nitrogen substrates, such as TN, TC/TN, DOC, and DOC/DON, P cycle and AP utilization. This finding is consistent with previous studies documenting that different soils can support different fungal flora [ 68 ]. Hypocreales , which belongs to the patotroph, were also reported to be negatively correlated with TN and DOC [ 69 ]. A previous study indicated that environmental factors, such as TN, DOC, and DOC/DON were unfavorable conditions for Hypocreales , and also postulated that excess nutrients decreased chlamydospore production [ 70 ]. In a related study, Glomerales , which belongs to arbuscular mycorrhiza fungi (AMF) [ 71 ], was associated with a high amount of available P, which in turn boosted plant growth [ 58 ]. Additionally, Auriculariales and Sordariales are generally considered saprophytic fungi [ 72 ], which stimulate the decomposition of organic substrates by saprotrophic fungi [ 73 ]."
} | 2,285 |
34634928 | PMC8515829 | pmc | 9,617 | {
"abstract": "ABSTRACT Wolbachia are endosymbiont bacteria known to infect arthropods causing different effects, such as cytoplasmic incompatibility and pathogen blocking in Aedes aegypti . Although several Wolbachia strains have been studied, there is little knowledge regarding the relationship between this bacterium and their hosts, particularly on their obligate endosymbiont nature and its pathogen blocking ability. Motivated by the potential applications on disease control, we developed a genome-scale model of two Wolbachia strains: w Mel and the strongest Dengue blocking strain known to date: w MelPop. The obtained metabolic reconstructions exhibit an energy metabolism relying mainly on amino acids and lipid transport to support cell growth that is consistent with altered lipid and cholesterol metabolism in Wolbachia -infected mosquitoes. The obtained metabolic reconstruction was then coupled with a reconstructed mosquito model to retrieve a symbiotic genome-scale model accounting for 1,636 genes and 6,408 reactions of the Aedes aegypti - Wolbachia interaction system. Simulation of an arboviral infection in the obtained novel symbiotic model represents a metabolic scenario characterized by pathogen blocking in higher titer Wolbachia strains, showing that pathogen blocking by Wolbachia infection is consistent with competition for lipid and amino acid resources between arbovirus and this endosymbiotic bacteria.",
"conclusion": "Conclusions. In this work, we have presented a novel approach for studying Wolbachia 's metabolic interaction with Aedes aegypti . By shifting the paradigm of analysis of metabolic networks from the focus on known reactions to identifying gaps as potential interaction points between the endosymbiont and its host, we have developed a methodology that allows us to propose effective coupling points between the metabolic networks of both organisms. Based on this new methodology, we have developed the first eukaryote-bacterium endosymbiotic genome-scale model. By means of this model, we were able to discover and study key metabolites that must be trafficked between both organisms in order to sustain Wolbachia while providing its host with pathogen-blocking capabilities. Due to its endosymbiotic nature, Wolbachia metabolism is characterized by incomplete pathways. On the other hand, Wolbachia -infected A. aegypti cells exhibit an altered amino acid and lipid metabolism caused by the survival requirements of the endosymbiont. Associated with this altered lipid metabolic profile, we propose that Wolbachia uses cholesterol instead of lipid A as a main membrane component, which is consistent with reported results on perturbed cholesterol metabolism observed in infected mosquitoes. Wolbachia -mediated pathogen blocking has been successfully used as a strategy for disease control, which has motivated an increasing number of studies regarding the underlying mechanisms that result in the pathogen-blocking phenotype. One of the widely accepted conclusions is that common effects between both arboviral and Wolbachia effects on host cellular metabolism, such as induced stress as a consequence of depletion of intracellular metabolites and highly dependency of cholesterol for their replication, are key to understanding this complex process. Based on our analysis using our novel systems biology approach, pathogen blocking was associated with competition for resources, mainly amino acids, cholesterol, and lipids. While cholesterol is essential for viral entry, assembly, and secretion, amino acids are key for blocking viral synthesis at the intracellular level, since our results demonstrate that virus production is highly affected by amino acid availability and consumption. A more detailed sampling approach study showed that competition for protein-derived resources results in high pathogen blocking capabilities at higher Wolbachia titers. This novel approach for studying endosymbiotic systems has allowed us to understand the metabolic interactions between Wolbachia and its host and to improve knowledge on the relevant metabolites involved in its survival and pathogen-blocking capabilities, in agreement with previously reported experimental results. We believe that this new paradigm for the analysis of coupled metabolic networks holds the key for unveiling the mechanisms behind complex symbiotic relationships.",
"introduction": "INTRODUCTION Wolbachia are obligate intracellular symbionts, members of the Rickettsiales group, known to infect over 65% of all insects species, mainly arthropods, developing diverse interactions with their hosts, such as supplementation with vitamins ( 1 , 2 ), cytoplasmic incompatibility ( 3 , 4 ), pathogenic interactions ( 5 , 6 ), and pathogen blocking ( 7 , 8 ), among others. Due to their pathogen-blocking abilities, these endosymbionts have been of special interests in disease control. In particular, the nonpathogenic strain w Mel, originally found in Drosophila melanogaster , has been used for infecting Aedes aegypti , obtaining mosquitoes that exhibited almost no transmission of Dengue and Zika viruses ( 7 , 8 ). However, the mechanisms behind Wolbachia -mediated pathogen blocking have not been fully characterized. Two main hypotheses have been proposed for explaining this phenomenon, one associated with an improved immunological response as a consequence of Wolbachia infection and the other one related to competition for host cell resources in the synthesis of key building blocks for this endosymbiont and arboviruses ( 9 ). We propose that a thorough analysis of Wolbachia 's metabolism could unveil key aspects of the metabolic relationship between this endosymbiont and its host and how their interactions influence the pathogen-blocking response in Aedes aegypti . This representation of Wolbachia metabolic capabilities will be given by genome-scale models (GSMs). Genome-scale models have emerged as a powerful tool for studying cellular metabolism based on their genome annotation ( 10 ). In particular, recent applications of GSMs for studying endosymbiotic organisms in insects have shown the potential of this approach ( 11 , 12 ), these analysis focused on the interactions between endosymbionts found in Cinara cedri ( 12 ) and Bemisia tabaci ( 11 ) finding that the coexistence of endosymbionts leads to an additional reduction of genetic information and a fragile metabolic network ( 12 ). However, none of these efforts has explicitly considered the whole metabolism of the host insect and how it is affected by its interaction with these endosymbiotic bacteria using a GSM approach, but only the immediate extracellular environment of the bacteria. As a result of its adaptation to depend on another organism for its survival, Wolbachia pipientis has a reduced genome size compared to similar nonendosymbiotic organisms. This leads to a small and rather incomplete metabolic network, as has been observed previously for other endosymbiotic bacteria ( 11 – 14 ). We hypothesized that the analysis of the metabolic gaps in the curation stage of the metabolic reconstruction could reveal potential candidates that explain the mutualistic relationship between Wolbachia and Aedes aegypti . In this work, we present three novel metabolic reconstructions developed for the study of Wolbachia pipientis and Aedes aegypti endosymbiosis from a systems biology perspective: a genome-scale model for W. pipientis and A. aegypti and the first model that represents the endosymbiosis between these two organisms. We propose that the metabolic analysis of the interactions between these organisms holds key features for explaining Wolbachia 's mediated pathogen blocking and that this blocking effect is a direct consequence of competition for the host resources.",
"discussion": "RESULTS AND DISCUSSION Metabolic reconstruction for the symbiotic bacteria Wolbachia . Two Wolbachia pipientis strains were selected for this work based on data availability and known pathogen-blocking capabilities: the Drosophila melanogaster endosymbionts w Mel and w MelPop. The obtained genome-scale models were used as a base to reconstruct the Wolbachia pipientis genome-scale model (iNJ644). This model includes 644 genes and 790 reactions, of which 220 are orphan reactions predicted to be present in the model despite the fact that some genes associated with this reaction are absent from Wolbachia genomes ( 13 ). These reactions are added to ensure that a functional model, capable of representing the synthesis of all the required components for cell growth, is obtained in the reconstruction process ( 10 ). Validation is performed to test if the obtained models can accurately represent reported and experimental data available for this endosymbiont. Information regarding Wolbachia ( 15 , 16 ) and Rickettsia 's metabolism ( 14 ) was gathered from the literature to compare metabolic features predicted by this metabolic reconstruction with the ones reported for these organisms ( Table 1 ). TABLE 1 Wolbachia pipientis genome-scale model (iNJ644) validation a No. Test Strain Result 1 Amino acid transport (Pro, Asp/Glu, Ala) w Mel + 2 Amino acid metabolism (Gly, Glu, Gln, Pro, Ser, Thr) w Mel + 3 Inability to produce LPS w Mel + 4 Complete pentose phosphate pathway w Mel + 5 Absence of ADP-ATP exchanger protein w Mel − 6 Threonine degradation pathway w Mel + 7 Riboflavin biosynthesis w Mel + 8 Complete TCA cycle w Mel + 9 Glycolysis starting from fructose 1,6 BP w Mel + 10 Lethal inhibition of MurA w Bm ++ 11 Nonlethal inhibition of DdlA w Bm −− a Validation tests based on metabolic features reported for Wolbachia pipientis \n w Mel and w Bm ( Brugia malayi ). Present metabolic feature, +; absent metabolic feature, −; consistent with experimental observation, ++; inconsistent with experimental observation, −−. We tested if Wolbachia 's preliminary model includes metabolic pathways previously reported to be present on Wolbachia pipientis , such as glycolysis starting from fructose 1,3 bisphosphate, a complete pentose phosphate pathway, TCA (tricarboxylic acid) cycle, an active amino acid metabolism, including transport of amino acids (proline, aspartate, glutamate and alanine), and catabolism of glycine, glutamate, glutamine, proline, serine, and threonine; the inability to produce lipopolysaccharides; and the absence of an ADP-ATP exchanger protein ( 15 ). Our analysis of the obtained metabolic network supports these affirmations, finding a complete glycolysis from fructose-6P toward phosphoenolpyruvate and a partially complete TCA cycle. The peptidoglycan synthesis pathway in Wolbachia and Chlamydia has been reported to be functional while growing inside their hosts. Two key points of this metabolic pathway were tested based on studies in peptidoglycan synthesis in Wolbachia , finding that inhibition of the first step of peptidoglycan synthesis catalyzed by the murA (UDP- N -acetylglucosamine 1-carbovinyltransferase) gene is lethal in Wolbachia w Bm infecting C6/36 cells ( 16 ). Additionally, they tested the inhibition of Ddla ( d -alanine d -alanine ligase A) by d -cycloserine, finding that it does not have a negative effect on lipid II biosynthesis in Wolbachia cells. Our Wolbachia reconstruction can replicate the lethal effect of phosphomycin in silico . The effect of high concentrations of this inhibitor is represented as a gene knockout, resulting in no cell growth in the performed simulations. iNJ644 knockout analysis predicts that inhibition of ddlA by d -cycloserine produces a nongrowth phenotype. ddlA is associated with the conversion of two alanines into one alanine dipeptide (Ala-Ala), which is a crucial step in lipid II synthesis. An additional blast search (coverage above 85% and E-values lower than 0.001) showed that the Wolbachia genome includes genes with high homology to the Ala-Ala transporters found in Escherichia coli , suggesting that Wolbachia could get this metabolite from its host to be consistent with experimental data ( 17 , 18 ). As a consequence, transport of the alanine dipeptide was added to the Wolbachia metabolic reconstruction. A final curated model was obtained and analyzed to find metabolic candidates that explain the obligate intracellular character of this bacteria. Genome-scale model for Aedes aegypti . To identify the key metabolic features that link Wolbachia 's metabolism with its host's, the metabolic gaps present in Wolbachia 's metabolic network were studied. To analyze Wolbachia 's pathway gaps in their metabolic context, we reconstructed a metabolic model for Aedes aegypti using an ortholog-based approach. Ortholog search between Aedes aegypti and Homo sapiens retrieved 5,327 ortholog groups between both organisms. This information was complemented with the gene associations present in the human metabolic reconstruction Recon 2.2 ( 19 ) to obtain the first Aedes aegypti metabolic model. This metabolic reconstruction accounts for 991 genes associated with 2,735 reactions and was then curated to include all the specific metabolic features reported for A. aegypti . Since a human metabolic reconstruction was used as a template, known insect and A. aegypti metabolic features were added prior to validation of the obtained model. In particular, Diptera and specifically A. aegypti insects are unable to synthesize sterols from acetate ( 20 , 21 ); hence, they acquire cholesterol from their diet ( 22 ). Particularly in Aedes aegypti , cholesterol transport is mediated by sterol carrier protein 2 (SCP2) ( 23 – 25 ). Overexpression of SCP2 has been found to increase incorporation of cholesterol ( 26 ), while its knockdown has led to a reduced uptake of cholesterol in female mosquito adults ( 27 ). Addition of SCP2 cholesterol transport is associated with the gene identifiers AAEL026044 and AAEL025252 ( 27 , 28 ). Additionally, a mosquito-specific pathway for urea disposal proposed by Scaraffia et al. ( 29 ), also known as the ureide pathway, was integrated into the mosquito metabolic reconstruction. It considers a series of reactions in which uric acid is transformed into allantoin by urate oxidase (UO), to allantoic acid by allantoinase (ALN), and to ureidoglycolate by alllantoicase (ALLC) ( 29 ). This pathway considers a final spontaneous reaction in which ureidoglycolate is transformed to glyoxylate and urea (urease) ( 5 ). In Diptera , thioredoxin, as opposed to glutathione, is used as a redox buffer. Particularly for Anopheles gambiae , it has been reported that alternative splicing of a thioredoxin reductase gene ( trdR ) leads to mitochondrial and cytoplasmic variants that keep redox homeostasis in this organism ( 30 ). Based on the metabolic requirements of Aedes aegypti , a cytoplasmic variant of the trdR gene was added. Experimental data regarding A. aegypti metabolism, growth media, and gene deletions were retrieved from the literature. Validation of this curated model was performed by simulating previous culture conditions in mosquito cell lines, metabolic observations, and gene knockdowns represented in the obtained reconstruction ( Table 2 ). We tested if the obtained mosquito model is able to replicate lipid and sterol usage in Aedes aegypti cells ( 31 ), finding that growth in cholesterol, phosphatidylcholine, and phosphatidylethanolamine is accurately represented by this metabolic reconstruction, as is the use of palmitate, stereate, and oleate for sustaining cellular growth ( 31 ). TABLE 2 Validation of the Aedes aegypti model a Test Result Medium conditions Growth in cholesterol, phosphatidylcholine, and phosphatidylethanolamine ++ Unable to grow on palmitic acid, stearic acid, or oleic acids ++ Growth in ergosterol, zymosterol − Growth in sphingomyelin, β-carotene, or α-tocopherol acetate − Growth in proline as energy substrate ++ Gene deletions Sterol carrier protein (AeSCP2) knockdown resulted in higher mortality ++ Thioredoxin reductase (Trdx1) is necessary for survival ++ Inactivation of glutamine synthase (AeGS) is lethal ++ Blood fed mosquitoes with alanine transferase (ALT) is not lethal ++ Knockdown of xanthine dehydrogenase (XDH-1) is lethal to blood-fed mosquitoes − Arginase (AR) silencing is not lethal ++ Urate oxidase (UO) silencing is not lethal ++ AR and urate oxidase (UP) silencing is not lethal ++ Nitric oxide synthase silencing is not lethal ++ a Validation tests based on metabolic features reported for Aedes aegytpi . Consistent with experimental observation, ++; inconsistent with experimental observation, −. The obtained metabolic reconstruction is unable to represent cellular growth based on ergosterol and zymosterol as lipid sources despite the fact that this has been extensively reported for this organism ( 31 – 33 ) and for the mosquito Culex pipiens ( 34 ). An analysis of the sterol pathway and search of similar sequences revealed no potential candidates in the Aedes aegypti genome that could lead to metabolization of these alternate sterol sources toward cholesterol and, hence, cell growth. On the other hand, this mosquito genome-scale model is able to represent cellular maintenance using proline as an energy source. The obtained cell growth is slower than the one obtained for glucose or trehalose as carbon sources but higher than the one reported with a lack of sterols in the media, which is consistent with proline being used only in extreme nutrient deprivation cases in mosquito flight response ( 35 ). Although there are several studies where gene deletions or silencing have been used as a means of controlling mosquito population, most of these targets are associated with cellular processes that cannot be represented by a metabolic reconstruction, e.g., olfactory sensors ( 36 ), blood feeding ( 37 ), reproduction ( 38 ), or apoptosis ( 39 – 43 ), among others ( 44 – 49 ). Urea formation in Aedes aegypti is achieved by two mechanisms: argininolysis and uricolysis. Gene silencing of arginase, urate oxidase, and nitric oxide synthase was revealed not to be lethal in this organism, consistent with gene deletion simulations ( 50 ). However, despite the fact that xanthine dehydrogenase 1 (XDH1) deletion has been proven lethal in blood-fed mosquitoes ( 51 ), gene deletion simulations for XDH1 and XDH2 were not able to replicate this behavior in silico . Since it has also been reported that this deletion had no effect on sugar-fed mosquitoes ( 51 ), we believe that there should be an unidentified process, currently not represented in our model, associated with blood metabolization where XDH1 is essential. On the other hand, the obtained metabolic reconstruction can replicate the lethal effect of sterol carrier protein ( AeSCP2 ) knockdown in Aedes aegypti mosquitoes ( 23 , 52 ) and the essentiality of thioredoxin reductase in Anopheles gambiae ( 30 ) and of glutamine synthase in the mosquito-derived cell line C6/36 ( 53 , 54 ). Overall our metabolic model was able to replicate nearly 80% (11/14) of the tested conditions in which gene deletions were associated with metabolic functions that are represented in genome-scale models. The obtained model was subsequently coupled with our Wolbachia pipientis metabolic reconstruction to shed light on the metabolic processes associated with bacterial and insect symbiosis. Both the presented models and their curation process are described in detail in file S1 at https://github.com/natJimenez/symbioticModelAnalysis . A novel symbiotic model approach for studying Wolbachia -mosquito metabolic interactions. A symbiotic model was obtained by integration of the Wolbachia and Aedes aegypti metabolic model. Further curation was performed to connect Wolbachia precursors to the mosquito metabolic network ( Fig. 1 ). We analyzed each metabolic requirement and its association with metabolic gaps in the Wolbachia model. Flux balance analysis (FBA) simulations showed that this endosymbiont requires the import of isopentenyl diphosphate (IPP) for biomass synthesis. These predictions are consistent with metabolite exchange predictions made for Rickettsia ( 14 ), obligate intracellular bacteria that belong to the same order as Wolbachia , where uptake of these metabolites is predicted to be essential for peptidoglycan and lipopolysaccharide biosynthesis ( 55 ). FIG 1 Wolbachia pipientis metabolism and its interactions with Aedes aegytpi. Key interaction points with Aedes aegypti are highlighted in red, complete metabolic pathways are represented as empty boxes, and dotted lines represent connections between metabolites with intermediate steps lumped. Only connections relevant to metabolic pathways of interest are presented. Adapted from reference 14 with permission. PLP, pyridoxal 5′-phosphate; ThDP, thiamine diphosphate; PDC, pyruvate dehydrogenase complex; TCA, tricarboxylic acid; MEP/DOXP, mevalonate-independent 2-C-methyl- d -erythritol 4-phosphate/1-deoxy- d -xylulose 5-phosphate pathway; THF, tetrahydrofolate; GSH, glutathione; OXPHOS, oxidative phosphorylation; g3p, glycerate 3 phosphate; PE, phosphatidylethanolamine; PS, phosphatidylserine; PG, phosphatidylglycerol; FPP, farnesyl diphosphate; IPP, isopentenyl phosphate; DMAPP, dimethylallyl pyrophosphate. Wolbachia 's metabolic model includes the exchange of metabolites that are either absent from the mosquito metabolic network or cannot be produced by Aedes aegypti in order to be retrieved by Wolbachia . In particular, lanosterol was predicted to be imported by this endosymbiont, but it is generally absent from the Aedes aegypti intracellular environment unless directly consumed by the mosquito. This metabolite is required by this symbiotic bacteria for farnesyl diphosphate (FPP) synthesis, a compound known to be imported in Rickettsia ( 14 ); hence, an additional FPP transport reaction was added to the Wolbachia model. Additionally, the lipid A synthesis pathway is completely absent from our Wolbachia metabolic reconstruction, as was previously reported for w Mel ( 15 ). Based on this information, we suggest that Wolbachia uses cholesterol instead of lipid A for its cell wall composition, as has been reported for other closely related endosymbiotic bacteria ( 56 ). Gap analysis of the symbiotic model unveils new features in Wolbachia 's lipid metabolism. The Wolbachia genome-scale model includes reactions that allow synthesis of phosphatidylglycerol, phosphatidylserine, and phosphatidylethanolamine from acyl-coenzyme A (CoA), unlike what has been published previously for w Mel ( 15 ) ( Fig. 2 ). However, the metabolic representation does not allow the synthesis of other membrane components, such as phosphatidylcholine and cardiolipin, making Wolbachia highly dependent on the host's intracellular membranes for cell growth. FIG 2 Wolbachia pipientis glycerophospholipid metabolism. Wolbachia can synthesize phosphatidylglycerolphosphate, phosphatidylserine, and phosphatidylethanolamine. Based on its association with intracellular membranes ( 57 , 58 ), phospholipid import in the Wolbachia model was represented as a transport reaction that preserves the stoichiometric coefficients of the components present in the mosquito membrane. An analysis of A. aegypti and W. pipientis membrane composition stoichiometry showed that there is a surplus of cholesterol in Wolbachia that is not required for membrane synthesis. We speculate that this excess of cholesterol is drained in the form of the lipid droplets observed in Wolbachia -infected A. aegypti cells and that this imbalance is associated with the reported cholesterol-altered metabolism observed in mosquito-infected cells ( 59 – 61 ). Wolbachia symbiosis affects amino acid and cholesterol metabolism in A. aegypti . Wolbachia and A. aegypti interaction was explored by analyzing 20,000 feasible solutions of the obtained model without imposing an optimality criterion, as illustrated in file S3 at https://github.com/natJimenez/symbioticModelAnalysis . Additionally, we analyzed the Pareto front (dashed lines), where the optimal use of resources toward Wolbachia and mosquito biomass synthesis is presented. These results evidence that, in an optimal distribution of resources, there is a trade-off between the growth rates of both organisms. This can be derived from the negative slope of the optimal use of the resource curve (Pareto front) for cell growth in this symbiotic system. Since the value of this slope is −2.4 [μ Aedes aegypti /μ Wolbachia ], producing a mosquito cell requires more nutrients than producing a Wolbachia cell. For instance, in the scenario of optimal use of resources, a 0.04 mosquito cell growth rate is associated with nearly double this value for Wolbachia growth (file S3 at the GitHub link above). The sampled fluxes exhibit a distribution with a stable state where low mosquito growth rates and a nonzero minimum Wolbachia growth rate are observed, given by Wolbachia riboflavin supplementation to A. aegypti . FBA simulations in a riboflavin-deprived environment show that although riboflavin supplementation is essential, higher growth rates of Wolbachia result in nutrient depletion and decreased mosquito biomass synthesis. The obtained results show that most of the feasible solutions were obtained around duplication times of 100 h for Wolbachia and between 0 and 400 h for Aedes aegypti . By analyzing duplication times of each organism (file S3 at https://github.com/natJimenez/symbioticModelAnalysis ), we estimate that there is a median of 3.45 Wolbachia cells per mosquito. Given that quantification of Wolbachia density inside the mosquito cell line Aa23 was estimated as between 300 and 1,200 bacteria per mosquito cell ( 60 ), we propose that Wolbachia pressure for remaining in A. aegypti is mainly due to regulatory processes rather than metabolic interactions. Results show that Wolbachia relies mainly on amino acids for supporting its cell growth ( Fig. 3 ); these amino acids are transported into the mosquito- Wolbachia endosymbiotic system and then consumed by Wolbachia . Tryptophan, valine, methionine, and leucine in particular have been reported as crucial for remediating low fecundity in Wolbachia -infected mosquitoes due to host competition for amino acids ( 62 ). Additionally, it has been reported that the presence of Wolbachia can reduce total cholesterol levels in mosquitoes up to 25%, which is consistent with the 20% total cholesterol consumed obtained in flux balance analysis simulations ( Fig. 3 ). FIG 3 Wolbachia pipientis and Aedes aegypti metabolic interactions. Flux balance analysis simulations are performed considering variations in the ratio between Wolbachia and its host based on values obtained by sampling of the symbiotic model (file S3 at the GitHub link in the article text above). Carbon sources, amino acids, and lipids consumed by higher (dark blue) and lower (light blue) infecting Wolbachia densities. Lipids and lipid droplets are represented in dark orange and light orange, respectively. The obtained results suggest that at least 10% of the total cholesterol present in mosquito cells is in the form of lipid droplets. This value is lower than the 25% Wolbachia -induced reduction of total cholesterol reported in Aedes aegypti mosquitoes ( 62 ) and the nearly 10% reduction of total cholesterol observed in the mosquito-derived cell line Aag2 infected with w MelPop ( 59 ). This is in agreement with the composition of mosquito cells considered in this study being closer to the one reported for mosquito-derived cell lines than to the variety of cell compositions found in whole insects. Geoghegan et al. ( 59 ) studied Aedes aegypti 's response to Wolbachia infection by analyzing proteomic data of Aag2 cells, showing that they exhibit an altered cholesterol metabolism similar to the responses associated with Niemann-Pick disease. We hypothesize that this alteration of sterol homeostasis could be triggered by cholesterol accumulation due to metabolites transported but not required for Wolbachia membrane synthesis. Elevation of esterified cholesterol has been associated with Wolbachia infection ( 59 , 60 , 62 ). Since our results show that these intracellular bacteria do not have genes known to interact with cholesterol- or sterol-derived metabolites, we propose that esterification is performed by Aedes aegypti rather than Wolbachia for sterols to be stored as lipid droplets. This would be the initial response of the metabolic cascade associated with intracellular cholesterol accumulation, which is followed by downregulation of LDL receptor and fatty acid synthase ( 59 ). Wolbachia 's pathogen blocking explained as competition for limited resources. One of the main features of interest in Wolbachia -infected mosquitoes is pathogen blocking. Several hypotheses have been proposed for explaining the underlying mechanisms resulting in protection against arboviral infection granted by Wolbachia infection. Some of these mechanisms are associated with an immunological boost in mosquito cells due to the presence of this endosymbiont, and others are related to competition for host cell resources. Recent findings have shifted the interest toward the latter, with amino acids and lipids being the proposed bottlenecks for viral replication ( 59 , 62 , 63 ). In this work, we propose that genome-scale models are suitable for exploring the hypothesis that can be directly represented by chemical reactions, such as competition for host resources. Viral replication is represented in the endosymbiotic model as a metabolic reaction comprising the RNA, amino acid, and lipid composition of the Dengue virus (DENV) ( 64 , 65 ). Flux balance analysis was performed to simulate viral infection in scenarios characterized by different Wolbachia and A. aegypti growth rates ( Fig. 4 ). The obtained results indicate that this system shows pathogen blocking at high Wolbachia cellular densities, consistent with previously reported studies on pathogen blocking by 10 different Wolbachia strains, where higher titers were consistently found to exhibit stronger pathogen-blocking properties ( 66 ). FIG 4 Wolbachia -mediated pathogen blocking: amino acid and lipid composition was derived from ZIKV and the West Nile virus. (a) Determination of key Wolbachia components in the pathogen-blocking phenotype. Simulations used different Wolbachia compositions comprising the nucleotide, amino acid, lipid fraction, and its effects on viral reproduction. (b) FBA simulations were performed using maximization of virus synthesis as their objective function, showing that higher Wolbachia pipientis densities are detrimental for viral synthesis. Additionally, we tested which component of Wolbachia 's biomass has a higher influence on pathogen blocking. Different artificial Wolbachia biomass compositions, including only amino acids, nucleotides, or lipids, were simulated in order to test which component is critical for pathogen blocking ( Fig. 4 ), showing that amino acids are crucial for this phenotype. Amino acid depletion has also been linked to host cell responses to viral infections in the absence of Wolbachia infection. It has been reported that phosphorylation of eukaryotic initiation factor 2 alpha (eIF2α) as a consequence of a diminished amino acid pool leads to viral replication arrest ( 67 ). This increased demand for protein precursors contributes to the observed unfolded protein response (UPR) in wMelPop -infected organisms and is consistent with Wolbachia 's location near the endoplasmic reticulum in order to ensure a constant supply of amino acids to support its energy metabolism ( 68 ). Recently this fact has been debated, since Fattouh et al. ( 58 ) found no UPR associated with Wolbachia w Mel infection, stating that this response is associated with the pathogenic behavior of wMelPop rather than Wolbachia itself. A previous analysis of metabolic networks for w Mel and w MelPop found that both strains are almost identical ( 55 ); in this approach, pathogenicity is represented by higher biomass production. Consistent with this, higher titers of Wolbachia are associated with a stronger pathogen-blocking phenotype. Recently, Koh et al. ( 69 ) compared perturbation on lipid profiles of mosquitoes infected with DENV and w Mel, finding that the overlap between them was not enough to suggest direct competition for lipids in a pathogen-blocking scenario. Instead, they proposed that Wolbachia pipientis perturbs lipid metabolism in a way that is detrimental for viral synthesis. Based on their results, they performed knockdown of key genes in lipid metabolism, particularly cardiolipin synthase (CRLS), which had negative effects on both bacterial growth and viral production. A comparison between Wolbachia pipientis , Aedes aegypti , and DENV lipid composition in our model shows that although Wolbachia and DENV share requirements for phosphatidylserine and phosphatidylethanolamine, most of their compositions differ. Even so, Wolbachia can synthesize both of these phospholipids from glycerol-3P ( Fig. 2 ). Koh et al. found two lipid classes that were enriched in a DENV infection but depleted in a dual Wolbachia -DENV infection: sphingomyelin and cardiolipin. In our model, sphingomyelin is considered part of the lipid fraction of DENV, which is synthesized from serine and palmitoyl-CoA in Aedes aegypti . Since sphingomyelin is not required for biomass synthesis for both mosquito and Wolbachia , an enrichment of this metabolite in a DENV infection scenario is consistent with our model. Cardiolipin, on the other hand, is part of both Wolbachia pipientis and A. aegypti composition but not of the DENV lipid fraction. Their enrichment on the DENV infection scenario is explained as a mechanism to stop apoptosis instead of being directly destined for viral production, consistent with our analysis of viral composition. Depletion of cardiolipin in a dual infection could be explained by the inability of Wolbachia pipientis to synthesize this lipid, posing an additional sink in this scenario that is detrimental for viral synthesis. Arboviral replication is a complex process that requires hijacking several metabolic processes in the mosquito cell, which alters membrane composition, drains intracellular nucleotides and energy, and subsequently modifies vesicle formation to ensure viral secretion ( 9 ). In particular, cholesterol has been reported to be of extreme importance in pathogen blocking ( 59 , 61 ). Caragata et al. ( 61 ) showed that in D. melanogaster , higher dietary cholesterol resulted in reduced Wolbachia -mediated pathogen blocking against Drosophila C virus (DCV). In Aedes aegypti , esterified cholesterol solubilization leads to a 100-fold increase in DENV genome copies in the presence of w MelPop ( 59 ). We propose that although cholesterol is essential for viral entry, vesicle trafficking, viral assembly, and secretion ( 70 ), amino acid consumption by Wolbachia is crucial for blocking viral synthesis at the intracellular level. This is consistent with findings made by Caragata et al. ( 61 ), where sterol supplementation was not able to completely eliminate the protective effect of this endosymbiont. We have found that the addition of cholesterol to the stoichiometric composition of the simulated arboviral particle has no significant effects on viral replication inhibition by Wolbachia . This is a significant contribution toward unveiling the mechanism of Wolbachia -derived pathogen blocking. Conclusions. In this work, we have presented a novel approach for studying Wolbachia 's metabolic interaction with Aedes aegypti . By shifting the paradigm of analysis of metabolic networks from the focus on known reactions to identifying gaps as potential interaction points between the endosymbiont and its host, we have developed a methodology that allows us to propose effective coupling points between the metabolic networks of both organisms. Based on this new methodology, we have developed the first eukaryote-bacterium endosymbiotic genome-scale model. By means of this model, we were able to discover and study key metabolites that must be trafficked between both organisms in order to sustain Wolbachia while providing its host with pathogen-blocking capabilities. Due to its endosymbiotic nature, Wolbachia metabolism is characterized by incomplete pathways. On the other hand, Wolbachia -infected A. aegypti cells exhibit an altered amino acid and lipid metabolism caused by the survival requirements of the endosymbiont. Associated with this altered lipid metabolic profile, we propose that Wolbachia uses cholesterol instead of lipid A as a main membrane component, which is consistent with reported results on perturbed cholesterol metabolism observed in infected mosquitoes. Wolbachia -mediated pathogen blocking has been successfully used as a strategy for disease control, which has motivated an increasing number of studies regarding the underlying mechanisms that result in the pathogen-blocking phenotype. One of the widely accepted conclusions is that common effects between both arboviral and Wolbachia effects on host cellular metabolism, such as induced stress as a consequence of depletion of intracellular metabolites and highly dependency of cholesterol for their replication, are key to understanding this complex process. Based on our analysis using our novel systems biology approach, pathogen blocking was associated with competition for resources, mainly amino acids, cholesterol, and lipids. While cholesterol is essential for viral entry, assembly, and secretion, amino acids are key for blocking viral synthesis at the intracellular level, since our results demonstrate that virus production is highly affected by amino acid availability and consumption. A more detailed sampling approach study showed that competition for protein-derived resources results in high pathogen blocking capabilities at higher Wolbachia titers. This novel approach for studying endosymbiotic systems has allowed us to understand the metabolic interactions between Wolbachia and its host and to improve knowledge on the relevant metabolites involved in its survival and pathogen-blocking capabilities, in agreement with previously reported experimental results. We believe that this new paradigm for the analysis of coupled metabolic networks holds the key for unveiling the mechanisms behind complex symbiotic relationships."
} | 9,701 |
33231450 | null | s2 | 9,618 | {
"abstract": "The development of non-natural photoenzymatic systems has reinvigorated the study of photoinduced electron transfer (ET) within protein active sites, providing new and unique platforms for understanding how biological environments affect photochemical processes. In this work, we use ultrafast spectroscopy to compare the photoinduced electron transfer in known photoenzymes. 12-Oxophytodienoate reductase 1 (OPR1) is compared to Old Yellow Enzyme 1 (OYE1) and morphinone reductase (MR). The latter enzymes are structurally homologous to OPR1. We find that slight differences in the amino acid composition of the active sites of these proteins determine their distinct electron-transfer dynamics. Our work suggests that the inside of a protein active site is a complex/heterogeneous dielectric network where genetically programmed heterogeneity near the site of biological ET can significantly affect the presence and lifetime of various intermediate states. Our work motivates additional tunability of Old Yellow Enzyme active-site reorganization energy and electron-transfer energetics that could be leveraged for photoenzymatic redox approaches."
} | 287 |
39697486 | PMC11654632 | pmc | 9,621 | {
"abstract": "Highlights • Microbial genetic resources are key for integral and global biodiversity of ecosystems worldwide. • The application of microbial genetic resources in agri-food and industrial sectors is indispensable for human advancement. • High-impact research on microbial genetic resources aims to enhance plant growth, improve soil quality, and address key aspects for sustainable agriculture."
} | 98 |
35548815 | PMC9086568 | pmc | 9,622 | {
"abstract": "Laminar-flow microbial fuel cells (LFMFCs) utilize the co-laminar flow feature in the microchannel as a virtual barrier to separate the anolyte and catholyte. However, for LFMFCs reported before, syringe pumps were always used to drive the fluid and form the co-laminar flow of anolyte and catholyte in the microchannel, reducing the net power output and the efficiency of the whole system. In this study, a laminar-flow microbial fuel cell (LFMFC) without any additional power supply is proposed. The LFMFC is successfully started-up after inoculation for 90 h. The anode biofilm distribution becomes sparser along the flow direction due to the thicker boundary layer and unfavorable crossover from the catholyte downstream. Moreover, the LFMFC delivers a maximum volumetric power density of 3200 W m −3 , which is higher than that of previous LFMFCs without membranes. Considering the practical application of LFMFC as a power source, the cell voltage responses to different conditions are further investigated. When the external resistance is switched between 1000 Ω and 4000 Ω, it takes the LFMFC 10 minutes to reach a stable voltage output. However, the voltage response to the intermittent supply takes 1 h to reach a stable value. Additionally, short-term cold storage has little effect on bacterial metabolic activity and cell voltage.",
"conclusion": "4. Conclusions A membrane-free, gravity-driven LFMFC was proposed. After inoculation for 90 h, the LFMFC was successfully started-up, and uneven biofilm distribution was observed due to the entrance effect and crossover of the catholyte. Owing to the large specific surface area of the carbon cloth anode and low internal resistance in the LFMFC, a maximum volumetric power density of 3200 W m −3 was delivered. Moreover, voltage response tests showed that the LFMFC performance could recover rapidly after changing the external resistance or short-term cold storage, while it took the voltage more time to stabilize if the reactant supplies were stopped. This study demonstrates that the proposed LFMFC could operate successfully without additional power supply.",
"introduction": "1. Introduction Microbial fuel cells (MFCs) are bio-electrochemical systems that harness microbial metabolism to convert waste energy into bioelectricity. 1,2 In recent decades, efforts have been made to develop MFCs to obtain clean and renewable energy. Generally, MFCs can be classified as macro-sized or micro-sized (μMFCs) according to the volume of the reactors and the electrode surface areas. Owing to their low cost, intrinsic small scale, easy assembly and precise control, μMFCs have received considerable attention as environmental monitors, medical kits, bacterial screening, and biosensors. 2–4 Conventional μMFCs are directly scaled down from macro dual-chamber MFCs, which consist of an anode and a cathode chamber and a proton exchange membrane (PEM) to separate them. However, μMFCs with membranes delivered a relatively poor performance due to high internal resistance, which is greatly affected by anode materials. 5 Although gold is highly conductive and compatible with microfabrication methods as anodes in μMFCs, 6–9 poor interactions between anode bacteria and the gold electrode lead to high internal resistance. 7,10 Compared with gold, carbon materials exhibit better contact with bacteria and smaller internal resistance, and a lower cost for μMFCs. Qian et al. designed a micro-MFC with sub-5 μL chamber with carbon cloth electrodes, which generated a power density of 62.5 W m −3 . The total internal resistance was reduced from 30 kΩ (gold anode) to 16 kΩ, which was still high for the μMFC. 11 Jiang et al. proposed a flow-through μMFC using 3D graphene foam as an anode, which had an estimated internal resistance of 7.3 kΩ. 12 Recently, micro MFCs without membranes, also called laminar-flow microbial fuel cells (LFMFCs), have attracted much attention, which contain a virtual barrier controlled by the co-laminar flow to separate the anolyte and catholyte. 13–17 The membrane-less structure can reduce both the fabrication cost and the internal resistance of the fuel cell. The first μ-scale membraneless microbial fuel cell was proposed by Li et al. , which used Shewanella oneidensis and produced a maximum current density of 25.42 mA m −2 . 18 After that, owing to their short hydraulic retention time and fast response, membrane-less bio-electrochemical systems were used to detect microbial electrochemical activity 19,20 and bacterial growth and respiration. 14 Moreover, in our previous study, a non-uniform biofilm distribution along the microchannel in the graphite electrode-based LFMFC was observed. 15 To enhance the biofilm distribution and performance, LFMFCs with different geometries and multiple anolyte inlets were further investigated. 16,17 The internal resistance was as low as 1092 Ω for the LFMFC with multiple inlets. However, for the LFMFCs in previous studies, syringe pumps were used to drive the fluid and form a co-laminar flow between the anolyte and catholyte in the microchannel, reducing the net power output of the whole system and limiting their miniaturization. Therefore, for practical application, it is essential to eliminate the external power supply to membrane-less LFMFCs. In this study, a gravity-driven, membrane-less LFMFC without any external power supply was proposed and fabricated. After successful inoculation with mixed bacteria, the biofilm morphology was characterized using a scanning electron microscope (SEM). Discharging tests under different external resistances and electrochemical impedance spectroscopy (EIS) tests were conducted to estimate the cell performance and the internal resistance, respectively. Finally, the effects of the external resistance, intermittent reactant supply, and cold storage on the voltage output were investigated.",
"discussion": "3. Results and discussion 3.1 Visualization of co-laminar flow in the LFMFC The co-laminar flow pattern of the anolyte and catholyte in the LFMFC was captured after assembling the device, as shown in Fig. 2 . The fluid–fluid interface between the anolyte and catholyte was clear in the glass fiber, which is a porous material to ensure co-laminar flow and separate two streams. 21 Moreover, the width of mixing region increased gradually along with the flow direction in the microchannel owing to transverse diffusion. This can be explained by the following scaling law 1 where δ x represents the maximum width near the top and bottom walls, D is the diffusion coefficient, H the channel height, z the distance the fluid flows downstream and U the average velocity. Overall, co-laminar anolyte and catholyte flow is successfully formed in the proposed LFMFC. Fig. 2 The co-laminar flow pattern of the anolyte and catholyte in the LFMFC. The green liquid is catholyte and the colorless one is anolyte. 3.2 Start-up process Connecting with an external resistance of 4000 Ω, the circuit voltage of the LFMFC was recorded during the start-up process, as shown in Fig. S2. † A 30 h lag period appeared, then the voltage went up rapidly and finally reached a stable value of 0.556 V after inoculation for 90 h. This indicates that the bacteria have been successfully inoculated on the anode surface. The start-up process of the proposed LFMFC is faster than that of the reported laminar flow-based microfluidic bio-electrochemical system with gold anodes (about 10 days). 19 This is mainly ascribed to the better biocompatibility and porous structure of carbon cloth, which benefits bacterial-adhesion and the electron transfer between the bacteria and anode. 22,23 The surface morphologies of the anode biofilm along the flow directions were captured by SEMs, as shown in Fig. 3 . The biofilm successfully formed on the surface of the carbon cloth. The densest and most uniform biofilm distribution is observed at the beginning of microchannel, while less biomass attached to the carbon cloth at the middle of the microchannel. The biofilm was sparse at the end of the microchannel. This result is consistent with our previous studies. 15 The decreasing biomass along the flow direction is mainly due to the entrance effect and the mixing of the anolyte and the catholyte. Due to the entrance effect, the boundary layer thickness increases gradually along the flow direction, leading to a high mass transfer resistance and further a few bacteria attachment. Moreover, the mixing zone of the anolyte and catholyte broadens along the flow direction ( eqn (1) ). Once the catholyte transport crossover to the anode at the end of the microchannel, the catholyte at the anode could act as electron acceptors directly and inhibit biofilm formation. Therefore, fewer bacteria could attach to the anode surface at the downstream of the microchannel. Fig. 3 Surface morphologies of the anode biofilm (a) and (d) at the beginning, (b) and (e) middle, and (c) and (f) end of the microchannel. 3.3 Internal resistance Employing carbon-based materials is an effective way to reduce the internal resistance for μMFCs because of their large surface area and functional organic groups benefiting electroactive bacteria. 5,24 EIS test was carried out to analyze the internal resistance of the LFMFC as shown in Fig. 4(a) . The internal resistance of the LFMFC includes the ohmic resistance, charge transfer resistance ( R ct ) and diffusion resistance ( R d ). For the membraneless MFC, the ohmic resistance depends on the solution resistance ( R s ). Through fitting the Nyquist plot, the results were obtained as follows: R s = 53.53 Ω, R ct = 1523.7 Ω and R d = 0.0046 Ω. The total internal resistance of the LFMFC is 1577 Ω. Therefore, the charge transfer resistance dominates the internal resistance of the LFMFC, which is in agreement with our previous results. 14 The internal resistance of the proposed LFMFC is much lower than that of the μMFC with physical membrane because of the removal of ion exchange membrane and high surface to volume of the carbon electrodes. 10,25,26 Fig. 4 (a) Nyquist plot of the LFMFC. The equivalent circuit model is shown in the inset. The Warburg impedance describes the diffusion resistance at anode ( R d ). CPE is the constant phase element related to the double layer capacitance. The subscripts a and c stand for anode and cathode, respectively. (b) The polarization and power density curves of the LFMFC. 3.4 Cell performance The polarization and volumetric power density curves were collected to evaluate the LFMFC performance by varying external resistances and the result is shown in Fig. 4(b) . The maximum power density is 3.2 ± 0.1 mW cm −3 (3200 ± 100 W m −3 ) at a current density of 7.6 mA cm −3 , which is higher than those of previous membrane-less laminar-flow microbial fuel cells (Table S1 † ). Compared with membrane-less LFMFCs with graphite or carbon anode as shown in Table S1, † the higher performance can be attributed to the large specific surface area of the carbon cloth and low internal resistance in this LFMFC. The current density decreased from 9.6 mA cm −3 to 8.8 mA cm −3 as the cell voltage decreased from 0.23 V to 0 V, which is referred to the “overshoot” phenomenon caused by the mass transport limitation at the anode. Under this condition, the anode biofilm was unable to provide a sufficient current for the electron acceptor. 27,28 The results demonstrate that the proposed LFMFC could operate successfully without additional power supply. 3.5 Voltage responses to different conditions A promising application of the LFMFC is using it as a miniature power source. Therefore, it is necessary to study the cell voltage responses to different conditions, including external resistance, intermittent reactant supply, and cold storage. 3.5.1 Response to the external resistance External resistances of 4000 Ω and 1000 Ω were selected to vary the operating conditions of the LFMFC. The voltage corresponding to each external resistance was held for approximately 10 minutes until it reached a steady state. Fig. 5 shows the voltage response during six discharge cycles (varying from 1000 Ω to 4000 Ω). The voltage increased to 0.5 V immediately when the external resistance changed from 1000 Ω to 4000 Ω; it then gradually stabilized after 10 minutes. When the resistance was changed from 4000 Ω to 1000 Ω, the voltage suddenly decreased and then took 10 minutes to stabilize, as shown in Fig. 5 . Moreover, the steady cell voltage value was almost the same during all six cycles, which indicates the repeatability of the voltage response to different external resistances. Fig. 5 Voltage responses during six cycles of changing the external resistance from 1000 Ω to 4000 Ω. 3.5.2 Response to intermittent reactant supply Considering that the proposed LFMFC is driven by gravity, if the driving force is insufficient, the problem of a discontinuous reactant supply may occur. Therefore, the voltage response to an intermittent reactant supply is also investigated. During the test, the external resistance was fixed at 4000 Ω. Two water-stop clamps were used to stop the anolyte and catholyte and the result is shown in Fig. 6 . When the reactant supply was stopped, the voltage decreased suddenly and then continued to decrease slowly to 0.02 V, which could be ascribed to the consumption of reactant reserved in the LFMFC. Once the reactant supply restarted, the voltage increased rapidly to 0.45 V and then continued to gradually increase. After two cycles of stopping the supply, the voltage recovered to 0.5 V. This suggests that the LFMFC could provide sufficient power after the reactant supply returns to normal after stopping for 1 h. Nevertheless, compared to its response to the external resistance, the voltage response to the intermittent supply requires a longer time (1 h) to stabilize. Fig. 6 Voltage response to intermittent reactant supply after stopping the supply of both the anolyte and catholyte for 1 h. 3.5.3 Response after cold storage For storage or transportation as a miniature power source, the LFMFC could be stored in a refrigerator when it is in not running. Therefore, the voltage evolution of the LFMFC was recorded after storage at 5 °C for various durations (1 h, 3 h, 9 h, 27 h), as shown in Fig. 7 . The anolyte and catholyte were drained from the LFMFC prior to storage. During cold storage, the LFMFC was disconnected from the data acquisition tool and was not supplied with any electrolyte. The cell voltage recovered immediately to its value prior to cold storage (0.5 V) after 1 h, 3 h and 9 h of storage. This suggests that short-term cold storage has little effect on bacterial metabolic activity and cell voltage output. However, the voltage recovered slowly, and the stable value was slightly lower than 0.5 V after 27 h of cold storage. This could be attributed to the irreversible loss of metabolic activity during storage at a low-temperature for a long time. 29 Fig. 7 Voltage evolution of the LFMFC after cold storage at 5 °C for various durations (1 h, 3 h, 9 h and 27 h)."
} | 3,769 |
36726680 | PMC9886061 | pmc | 9,623 | {
"abstract": "Barley is a major cereal crop for temperate climates, and a diploid genetic model for polyploid wheat. Cereal straw biomass is an attractive source of feedstock for green technologies but lignin, a key determinant of feedstock recalcitrance, complicates bio-conversion processes. However, manipulating lignin content to improve the conversion process could negatively affect agronomic traits. An alternative approach is to manipulate lignin composition which influences the physical and chemical properties of straw. This study validates the function of a barley ferulate 5-hydroxylase gene and demonstrates that its downregulation using the RNA-interference approach substantially impacts lignin composition. We identified five barley genes having putative ferulate 5-hydroxylase activity. Downregulation of HvF5H1 substantially reduced the lignin syringyl/guaiacyl (S/G) ratio in straw while the lignin content, straw mechanical properties, plant growth habit, and grain characteristics all remained unaffected. Metabolic profiling revealed significant changes in the abundance of 173 features in the HvF5H1 -RNAi lines. The drastic changes in the lignin polymer of transgenic lines highlight the plasticity of barley lignification processes and the associated potential for manipulating and improving lignocellulosic biomass as a feedstock for green technologies. On the other hand, our results highlight some differences between the lignin biosynthetic pathway in barley, a temperate climate grass, and the warm climate grass, rice, and underscore potential diversity in the lignin biosynthetic pathways in grasses.",
"introduction": "1 Introduction Lignocellulosic feedstocks are attractive commodities for producing a variety of bio-products due to their renewable nature and smaller environmental footprint compared to petrochemicals. However, conversion of lignocellulosic biomass into high value bio-products and second-generation biofuels is challenging, partly due to the presence, heterogeneity, and structural complexity of lignin. This has led to a focus on researching the biosynthesis, properties, and manipulation of lignin to increase both fundamental understanding, and to explore opportunities for biomass improvement. Lignin is produced via the phenylpropanoid pathway which starts with the enzyme phenylalanine ammonia-lyase (PAL) and ultimately synthesizes the H ( p -hydroxyphenyl), G (guaiacyl), and S (syringyl) monolignol units. Lignification begins when these monolignols are oxidised by cell-wall laccases and/or peroxidases resulting in radicals that are combinatorially coupled through ether or carbon-carbon bonds leading to assembly of lignin polymers. The outcome of the polymerization is heavily influenced by the type and amount of phenolic monomers that are translocated into the cell wall and that are competent for free-radical polymerization (for review, see Tobimatsu et al., 2013 ). Grass lignin is mainly made up of S and G units, with H units constituting a small but significant percentage. Lignin composition influences the physical and chemical properties of biomass feedstocks and has been one of the key targets of cell wall engineering studies ( Halpin, 2019 ). Recent research has highlighted some of the unique features of the lignin pathway in grasses compared to dicots. In dicots, lignin biosynthesis starts with the deamination of phenylalanine (Phe) into cinnamic acid and subsequent conversion to p -coumaric acid ( p CA) via cinnamate 4-hydroxylase (C4H) activity. Grasses are able to synthesize p CA from Phe or tyrosine (Tyr) using bifunctional phenylalanine/tyrosine ammonia-lyases (PTAL), with the TAL activity contributing to nearly half of the cell wall total lignin in Brachypodium ( Barros et al., 2016 ). The metabolites derived from Tyr seem to be preferentially incorporated into S lignin units and into cell-wall-bound p CA, another characteristic feature of grass cell walls ( Withers et al., 2012 ; Barros et al., 2016 ). Some of that wall-bound p CA is attached to lignin through the involvement of p -coumaroyl-CoA:monolignol acyltransferase (PMT) in the acylation of a proportion of monolignols to produce p -coumaroylated S monolignol conjugates ( Marita et al., 2014 ; Petrik et al., 2014 ). The conventional lignin biosynthetic pathway enzyme, ferulate 5-hydroxylase (F5H, also known as coniferaldehyde 5-hydroxylase or Cald5H), is essential to S lignin biosynthesis in dicots, but relatively few studies have examined its role in grasses. Recent work in rice has suggested that F5H may only catalyse production of the nonacylated portion of lignins in grasses ( Takeda et al., 2019a ). This has led to the hypothesis that a grass specific lignin pathway may exist for the production of p -coumaroylated S lignin which is independent of the co-existing conventional lignin biosynthesis pathway ( Takeda et al., 2019a ; Barros and Dixon, 2020 ). However, the genes and enzymes of the hypothetical grass specific lignin pathway remain to be identified and it is still unclear whether this pathway, if it exists, is a feature of lignin biosynthesis unique to rice or common to other grasses. Consequently, the role of F5H in dictating the overall S lignin content of grasses is still an open question. The proportions of G and S units in lignin critically influence the structure of the polymer and the ease of enzymatic biomass processing to release sugars (saccharification) for downstream uses. In the conventional S lignin biosynthetic pathway, the C5 position on the G-lignin precursor units (coniferaldehyde or coniferyl alcohol) is hydroxylated and methylated by F5H and caffeic acid O -methyltransferase (COMT) enzymes, respectively. F5H knockout plants of Arabidopsis lack S-lignins ( Chapple et al., 1992 ). The availability of C5 on G-lignin precursors can lead to pretreatment-recalcitrant β–5 or 5–5’ linkages within lignin and trigger inter-chain connections during lignin polymerization, such that G-rich lignins are more chemically complicated with larger molecular weights and higher melting points compared to S rich lignins ( Vanholme et al., 2010 ; Ralph et al., 2019 ). Structurally, a lignin polymer that is extreme in its content of S units, methoxylated at both C3 and C5, is mainly composed of more homogeneous linear chains, which are shorter relative to H- or G-rich lignins and have a high proportion of chemically-labile β– O –4 bonds ( Stewart et al., 2009 ; Vanholme et al., 2010 ). Many studies in various dicot species in which F5H expression has been manipulated illustrate that reduced F5H expression reduces S lignin units and can lower sugar release after biomass pretreatment, whereas increased F5H expression increases S lignin units and may improve sugar release (e.g., Ciesielski et al., 2014 ; Fan et al., 2020 ). However, the few studies in grasses that link F5H expression and S lignin content to saccharification efficiency have yielded contradictory results. Research performed on Brachypodium yielded results consistent with those from dicots ( Sibout et al., 2017 ), whereas work on switchgrass revealed no change to saccharification on F5H up- or down-regulation ( Wu et al., 2019 ), and work on sugarcane surprisingly indicated that reduced F5H expression and reduced S units in lignin might increase sugar release upon saccharification after mild acid pretreatment ( Bewg et al., 2016 ). The roles of F5H and S lignin content in influencing saccharification and biomass processing in grasses after various pretreatments therefore remains an open question and it is possible that not all grasses behave in the same way. In order to clarify both the role of F5H in influencing the S lignin content of grass lignins, and to explore the impact of F5H downregulation on saccharification efficiency, research is needed in a wider variety of grasses. Barley ( Hordeum vulgare ) is the fourth largest cereal crop globally and a close relative of wheat ( Triticum aestivum ), the second most important global crop. Together, barley and wheat crops are the greatest source of straw biomass in temperate regions. In this study, we identified five F5H genes in barley and by RNAi downregulation we validated that one of these genes, HvF5H1 , plays a dominant role in S lignin production and in determining the lignin S/G ratio in straw. Subsequently, we studied the impact of F5H downregulation on saccharification recalcitrance, agronomic traits, and straw mechanical properties. Our data suggest differences in overall S lignin pathways between barley and what has been reported for rice and highlights potential diversity in lignin biosynthesis in grasses with implications for straw improvement strategies.",
"discussion": "3 Discussion This study demonstrates that expression of HvF5H1 plays a major role in determining the proportions of S and G units in lignin in barley culm tissues. Transgenic plants with downregulated HvF5H1 expression have substantially reduced frequency of S units in culm lignin and a correlatively high enrichment in G units compared to control plants. In the RNAi line with the largest shift in lignin composition, S units represented only 19% of the thioacidolysis released units as compared to 60% in the azygote control plants, representing a relative decrease of 68% in S units. Correspondingly, G units increased from 37% of the units released from azygote plants to 78% of the released units in RNAi plants, a relative increase of 111%. Concomitant with the %S decrease and %G increase, the thioacidolysis yield was reduced, which is consistent with the higher proclivity of G units to be involved in resistant interunit bonds such as phenylcoumaran or biphenyl linkages that thioacidolysis does not release monomers from. Our 2D NMR analysis independently confirmed the thioacidolysis lignin data revealing a 73% reduction in S units and a 95% increase in G units overall. Unlike thioacidolysis which releases monomers from lignin units only involved in β-aryl ether units (linked by their characteristic β–O–4-ether bonds), 2D-NMR on unfractionated cell walls can provide information on additional lignin linkages and showed that, although the proportion of β-ether units decreased from 93% in the wild type to 83% in the RNAi line, phenylcoumaran linkages proportionally increased significantly, from 4% in the wild type to 15% in the RNAi line. There was little change in resinol linkages in the RNAi line. These changes to lignin on HvF5H1 suppression are very consistent with what has previously been seen in dicot species (e.g., Meyer et al., 1998 ; Reddy et al., 2005 ) and converse changes (i.e., an increase in the proportion of S units) have been seen when F5H is over-expressed (e.g., Meyer et al., 1998 ; Franke et al., 2000 ; Stewart et al., 2009 ). Our analyses also showed significant reductions in ester-linked p CA in two HvF5H1 RNAi lines. Mild alkaline hydrolysis indicated a 28 and 41% reduction in the amount of ester-linked p CA in HvF5H 1-RNAi lines T and B respectively. As much of the cell wall p CA in grasses is ester-linked to S units in the lignin, a reduction in esterified p CA when S units are reduced is not unexpected, although the proportion of p CA that is ester-linked to arabinoxylans would be retained. Again, 2D-NMR confirmed a p CA reduction of approximately 40% suggesting that the impact of HvF5H1 suppression is not restricted to the nonacylated portion of lignin as in rice ( Takeda et al., 2019a ), but indicates that the p -coumaroylated S lignin has been reduced. In contrast to the many studies on F5H manipulation in dicots, only a few studies address the impact of F5H suppression in grasses. In sugarcane, F5H was suppressed by RNAi to varying degrees although 16-94% of control expression remained in four transgenic lines analysed ( Bewg et al., 2016 ). Nevertheless, the best transgenic line had an altered S:G ratio of 48:52 compared to 61:39 in controls, i.e., a 21% decrease in the proportion of S units and a 33% increase in the proportion of G units, changes completely consistent with but less extreme than what we observed in barley. The NMR analysis also detected phenylcoumaran (β–5-linked) units in this transgenic plant but not in the controls ( Bewg et al., 2016 ). Similarly, in F5H -RNAi suppressed switchgrass ( Wu et al., 2019 ), plants with approximately 55% reduction in F5H expression, by our calculations had an altered S:G ratio of about 32:68 compared to 46:54 in controls, i.e., a 30% decrease in the proportion of S units and a 26% increase in the proportion of G units. In an initial study in rice, F5H1 suppression by RNAi eliminated over 90% F5H1 expression but S:G ratio determined via thioacidolysis was only reduced from 36:64 in control to 26:74 in the RNAi lines, i.e., a relative decrease of S units by 28% according to Takeda et al. (2017) . To determine whether the high proportion of remaining S units was due to residual F5H1 activity, or to the activity of alternative unidentified hydroxylases, the same research group knocked out rice F5H1 using CRISPR/Cas9 ( Takeda et al., 2019a ). Surprisingly, plants with presumed total knock-out of F5H1 still produced considerable amounts of S units in the culm, to approximately 62–70% of the level of wild-type plants. This is very unexpected as an F5H knock-out mutant in Arabidopsis, the fah1-2 mutant, produces no S lignin ( Chapple et al., 1992 ). Moreover, DFRC lignin analysis of the rice F5H1 CRISPR knock-out indicated that the changes in G and S units were restricted to the non- p -coumaroylated lignin units and units with p CA attached were almost unchanged ( Takeda et al., 2019a ). This discovery prompted the authors and others to propose that rice and possibly grasses in general must have two parallel pathways for making S lignin, only one of which requires F5H1 whereas monolignol units to which p CA is attached are produced by an undiscovered alternative pathway ( Takeda et al., 2019a ; Barros and Dixon, 2020 ). Our results throw some of the details of this hypothesis into question at least for barley. Only HvF5H1 was downregulated in our experiments and transcriptomic data from RNAi plants confirmed that expression of other HvF5Hs did not change. Nevertheless, we achieved a substantial decrease of 68-73% in S units using RNAi compared to in rice where there were only 32% and 38% S unit reductions using RNAi and CRISPR knock-outs respectively. The approximately 30% of S units remaining in our barley plants could be explained by the low level of HvF5H1 expression that remains. The levels of esterified p CA in cell walls was also clearly reduced in barley by approximately 30–40% consistent with the reduction in S units and the fact that most p CA esterified to grass lignin is found on S units (although a significant proportion is found esterified to arabinoxylan). Consequently, although we cannot conclude that an alternative S lignin pathway does not exist in barley, our results can be explained based on the conventional lignin pathway alone. Taking into consideration the scale of intermolecular bond alterations, the high impact of HvF5H1 RNAi downregulation on S and G lignin, and the significant reduction of p CA, HvF5H1 is a major gene for S lignin biosynthesis in barley and the conventional lignin pathway clearly dominates. The changes in lignin in our HvF5H1 -RNAi plants appear to have no impact on saccharification. This is consistent with most other work in the literature that either shows no change, or a reduction in saccharification, in F5H -suppressed plants, and an increase in saccharification when F5H is overexpressed and S lignin proportion increased. Several papers demonstrate that the predominantly G-lignin in the Arabidopsis fah1 F5H mutant impedes saccharification compared to wild-type lignin after liquid hot water pretreatment ( Li et al., 2010 ) or maleic acid treatment ( Ciesielski et al., 2014 ), although weak alkaline pretreatment reveals no change to saccharification ( Smith et al., 2022 ). On the other hand, Atf5h1 T-DNA insertion mutants allelic to fah1 did show an improved saccharification when no pretreatment was used ( van Acker et al., 2013 ). In Populus tomentosa , downregulating F5H expression by manipulating miR6443, a microRNA that regulates S lignin biosynthesis, increases G units and reduces S units in lignin and lowers sugar yield after saccharification with both alkaline or acidic pretreatments ( Fan et al., 2020 ). Conversely, increasing F5H expression increases S lignin and saccharification in Arabidopsis after alkaline pretreatment ( Ciesielski et al., 2014 ; Smith et al., 2022 ), or maleic acid ( Sakamoto et al., 2020 ), or liquid hot water pretreatment ( Li et al., 2010 ) and in poplar after pretreatments with alkali ( Lapierre et al., 2021 ) or acid/alkali ( Fan et al., 2020 ). In the rice F5H1 CRISPR knock-out lines, saccharification efficiency was lowered compared to wild type only when using hot water pretreatment and unchanged with dilute acidic or alkaline pretreatments ( Takeda et al., 2019b ), whereas in switchgrass up- or down-regulation of F5H did not affect saccharification ( Wu et al., 2019 ). In Brachypodium, F5H overexpression increased S lignin and saccharification without pretreatment ( Sibout et al., 2017 ). In sugarcane, one F5H-suppressed line apparently had increased sugar release after mild acidic pretreatment but others did not ( Bewg et al., 2016 ) so the relevance of this observation which conflicts with the rest of the literature is unclear. Our data showing that barley HvF5H1- RNAi plants have increased resistant phenylcoumaran units and reduced esterified p CA (which generally occurs as free-phenolic pendent groups on lignin ( Ralph et al., 1994 ; Lu and Ralph, 1999 )) suggests a more condensed or more p CA-decorated lignin would not necessarily have any beneficial effect on saccharification. Incidentally, in artificial model systems where other factors could be controlled, lignin S/G composition had little effect on wall degradability ( Grabber et al., 1997 ; Grabber et al., 2009 ). Downregulation of HvF5H1 via RNAi led to changes in the abundance of 173 features. The majority of metabolite features have unknown identities. As detailed in \n Table S2 \n , 78 compounds were less abundant in the HvF5H1 -RNAi plants. Predictably, S lignin-related compounds such as sinapyl alcohol and syringic acid 4- O -hexoside were among the known compounds shown to be decreased, which is consistent with the reduction of S lignin. The α‐oxidized β‐ether oligomer of sinapyl alcohol, Sox(8-O-4)S, was 150-fold less abundant in the HvF5H1 -RNAi plants, followed by several small chains of G-S lignols such as G(8-O-4)S(8-8)S. Sox(8-O-4)S was previously shown also to be reduced 10-fold in abundance in barley HvCOMT -RNAi plants ( Daly et al., 2019 ). On the other hand, the abundance of 95 features increased significantly in the HvF5H1 -RNAi plants. Although the identity of the majority of these features is unknown, we notably found that di- or tri-lignols composed of only G units were increased 10 to 73-fold in the HvF5H1 -RNAi lines, consistent with the increased abundance of G units in lignin and reduced proportion of S units. This is very different from HvCOMT -RNAi lines in which the greatest increased abundance detected involved 5‐hydroxyconiferyl alcohol, the product of F5H activity and substrate for subsequent COMT activity, and caffeoyl alcohol, another supposed substrate for COMT ( Daly et al., 2019 ). Although the G-unit-enriched lignin of F5H -suppressed plants may not be improved in digestibility, it may be useful for other purposes; for example the more highly-branched and less oxygenated lignin has a higher fuel-value and improved combustion properties. However, if we are to improve the lignocellulosic biomass of food crops it is crucial that changes in cell wall characteristics do not negatively affect agronomic traits. In this study, downregulation of HvF5H1 did not lead to changes in grain characteristics or other agronomic traits and the mechanical characteristics of straw in the HvF5H1 -RNAi lines were similar to the controls. Some previous studies have explored the relationship between lignin structure and biomechanical properties but firm conclusions have proved elusive. In poplar supressed in the lignin biosynthesis gene CAD, lignin content was associated with mechanical stiffness ( Özparpucu et al., 2017 ), but the pronounced changes in lignin composition did not alter wood tensile properties ( Özparpucu et al., 2018 ). However, in wheat, lignin content was not associated with straw breaking force but a lower proportion of S lignin appeared to increase it ( Muhammad et al., 2020 ). In Arabidopsis fah1 F5H mutants, stiffness was increased compared to wild type in apical and middle stem segments, but flexibility was unaltered between fah1 and wild type plants ( Ménard et al., 2022 ). Differences in stem architecture between dicots and grasses and indeed between individual species are likely to complicate any simplistic generalized interpretation of the effects of lignin characteristics on plant mechanical properties since even different morphotypes of tracheary element have different lignin chemistries and mechanical properties ( Ménard et al., 2022 ). In our work, it is promising to see that radical changes to lignin composition and structure in HvF5H1 suppressed plants is not accompanied by any change to straw strength. Thus, F5H may represent a good target for manipulating lignin composition while maintaining crop health. Indeed, F5H suppression might improve some agronomic properties as illustrated recently in Brassica napus in which four genes for F5H were knocked out by CRISPR/Cas 9 and this apparently improved resistance to Sclerotinia stem rot ( Cao et al., 2022 ). Our work highlights once again that, despite striking similarities among grasses in terms of their cell wall composition, there may be important differences between species in their lignin biosynthetic pathways. For example, we previously showed that the key lignin biosynthetic gene caffeoyl shikimate esterase (CSE), originally identified in Arabidopsis, only seems to have bone fide orthologues in some grasses including rice and switchgrass, but is apparently absent in barley, Brachypodium, and maize ( Vanholme et al., 2013 ; Ha et al., 2016 ). Here we show that the dominant route to S lignin biosynthesis in barley proceeds though the conventional lignin pathway and F5H, and that the proposed alternative ‘grass specific’ pathway that apparently exists in rice, is either absent or only a minor route in barley."
} | 5,706 |
35257660 | PMC8903832 | pmc | 9,624 | {
"abstract": "Collective migration—the directed, coordinated motion of many self-propelled agents—is a fascinating emergent behavior exhibited by active matter with functional implications for biological systems. However, how migration can persist when a population is confronted with perturbations is poorly understood. Here, we address this gap in knowledge through studies of bacteria that migrate via directed motion, or chemotaxis, in response to a self-generated nutrient gradient. We find that bacterial populations autonomously smooth out large-scale perturbations in their overall morphology, enabling the cells to continue to migrate together. This smoothing process arises from spatial variations in the ability of cells to sense and respond to the local nutrient gradient—revealing a population-scale consequence of the manner in which individual cells transduce external signals. Altogether, our work provides insights to predict, and potentially control, the collective migration and morphology of cellular populations and diverse other forms of active matter.",
"introduction": "Introduction The flocking of birds, schooling of fish, herding of animals, and procession of human crowds are all familiar examples of collective migration. This phenomenon also manifests at smaller scales, such as in populations of cells and dispersions of synthetic self-propelled particles. In addition to being a fascinating example of emergent behavior, collective migration can be critically important—enabling populations to follow cues that would be undetectable to isolated individuals ( Camley, 2018 ), escape from harmful conditions and colonize new terrain ( Cremer et al., 2019 ), and coexist ( Gude et al., 2020 ). Thus, diverse studies have sought to understand the mechanisms by which collective migration can arise. Less well understood, however, is how collective migration persists after a population is confronted with perturbations. These can be external, stemming from heterogeneities in the environment ( Sándor et al., 2017 ; Morin et al., 2016 ; Wong et al., 2014 ; Chepizhko and Peruani, 2013 ; Chepizhko et al., 2013 ; Chepizhko and Peruani, 2015 ; Toner et al., 2018 ; Maitra, 2020 ), or internal, stemming from differences in the behavior of individuals ( Yllanes et al., 2017 ; Bera and Sood, 2020 ; Alirezaeizanjani et al., 2020 ). Mechanisms by which such perturbations can disrupt collective migration are well documented. Indeed, in some cases, perturbations can abolish coordinated motion throughout the population entirely ( Sándor et al., 2017 ; Morin et al., 2016 ; Yllanes et al., 2017 ; Bera and Sood, 2020 ; Chepizhko and Peruani, 2013 ; Chepizhko et al., 2013 ; Chepizhko and Peruani, 2015 ; Toner et al., 2018 ). In other cases, perturbations couple to the active motion of the population to destabilize its leading edge, producing large-scale disruptions to its morphology ( Wong et al., 2014 ; Alert and Trepat, 2020 ; Alert et al., 2019 ; Driscoll et al., 2016 ; Doostmohammadi et al., 2016 ; Williamson and Salbreux, 2018 ; Miles et al., 2019 ). Indeed, for one of the simplest cases of collective migration—via chemotaxis, the biased motion of cells up a chemical gradient—morphological instabilities can occur due to the disruptive influence of hydrodynamic ( Subramanian et al., 2011 ; Lushi et al., 2012 ; Lushi et al., 2018 ) or chemical-mediated ( Ben Amar and Bianca, 2016 ; Ben Amar, 2016 ; Funaki et al., 2006 ; Brenner et al., 1998 ; Mimura and Tsujikawa, 1996 ; Stark, 2018 ) interactions between cells. By contrast, mechanisms by which migrating populations can withstand perturbations have scarcely been examined. Here, we demonstrate a mechanism by which collectively migrating populations of Escherichia coli autonomously smooth out large-scale perturbations in their overall morphology. We show that chemotaxis in response to a self-generated nutrient gradient provides both the driving force for collective migration and the primary smoothing mechanism for these bacterial populations. Using experiments on 3D-printed populations with defined morphologies, we characterize the dependence of this active smoothing on the wavelength of the perturbation and on the ability of cells to migrate. Furthermore, using continuum simulations, we show that the limited ability of cells to sense and respond to a nutrient gradient causes them to migrate at different velocities at different positions along a front—ultimately driving smoothing of the overall population and enabling continued collective migration. Our work thus reveals how cellular signal transduction enables a population to withstand large-scale perturbations and provides a framework to predict and control chemotactic smoothing for active matter in general.",
"discussion": "Discussion By combining experiments and simulations, this study elucidates a mechanism by which collectively migrating populations can smooth out large-scale perturbations in their overall morphology. We focus on the canonical example of chemotactic migration, in which coherent fronts of cells move in response to a self-generated nutrient gradient. The smoothing of these fronts underlies the utility of standard agar-based assays for chemotaxis, in which bacteria spread outward in smooth, circular rings from a dense inoculum ( Wolfe and Berg, 1989 ; Tittsler and Sandholzer, 1936 ; Croze et al., 2011 ; Cremer et al., 2019 )—despite the presence of irregularities in the initial inoculum that are inevitably introduced by human error. To our knowledge, the robustness of the front morphology to such perturbations has never been examined or quantitatively explained; as a result, previous studies have only focused on the migration of the smooth fronts that ultimately result ( Adler, 1966 ; Lauffenburger, 1991 ; Keller and Segel, 1971 ; Cremer et al., 2019 ; Fu et al., 2018 ; Saragosti et al., 2011 ; Bhattacharjee et al., 2021 ; Bai et al., 2021 ). Our work now provides an explanation for why perturbed fronts smooth out. It therefore provides a counterpoint to previous studies investigating the ability of perturbations to instead disrupt collective migration ( Sándor et al., 2017 ; Morin et al., 2016 ; Yllanes et al., 2017 ; Bera and Sood, 2020 ; Chepizhko and Peruani, 2013 ; Chepizhko et al., 2013 ; Chepizhko and Peruani, 2015 ; Toner et al., 2018 ; Wong et al., 2014 ; Alert and Trepat, 2020 ; Alert et al., 2019 ; Driscoll et al., 2016 ; Doostmohammadi et al., 2016 ; Williamson and Salbreux, 2018 ; Miles et al., 2019 ; Subramanian et al., 2011 ; Lushi et al., 2012 ; Lushi et al., 2018 ; Ben Amar and Bianca, 2016 ; Ben Amar, 2016 ; Funaki et al., 2006 ; Brenner et al., 1998 ; Mimura and Tsujikawa, 1996 ; Stark, 2018 ). It also complements recent theoretical work describing how chemotaxis can stabilize the hydrodynamic instabilities that arise in unconfined populations of self-propelled particles ( Nejad and Najafi, 2019 ). The 3D printing platform provides a unique way to tune the shape of the initial perturbation, as well as the extent to which cellular migration is hindered. Our experiments using this approach reveal that the dynamics of smoothing are regulated by both the undulation wavelength and the ease with which cells migrate. The continuum simulations recapitulate the essential features of this behavior and shed light on the underlying mechanism. We find that even though cells in peaks of an undulated front experience a stronger driving force given by the local nutrient gradient, the higher nutrient levels they are exposed to saturate their cell-surface receptors, and hence they exhibit a weaker chemotactic response than cells in valleys. That is, while variations in the nutrient gradient along the leading edge of a front act to amplify undulations, variations in the ability of cells to sense and respond to this gradient dominate and instead smooth out the undulation. Importantly, this mechanism of smoothing is distinct from diffusion, which is typically responsible for the smoothing of traveling waves in reaction-diffusion systems—and in our case, is much too slow to drive smoothing. Conditions for chemotactic smoothing to arise While our study utilizes a specific form of the sensing function f ( c ) established for E. coli ( Cremer et al., 2019 ; Fu et al., 2018 ), the phenomenon of chemotactic smoothing can manifest more generally. Specifically, our description of smoothing requires that (i) convex regions of a population are exposed to more nutrient c than concave regions, and (ii) f ( c ) is monotonically increasing and concave, with f ′ ′ ( c ) < 0 ; when these conditions are satisfied, the chemotactic response is weaker at convex regions than at concave ones, thereby promoting smoothing (as indicated in Figure 4B ). The first requirement is frequently satisfied for collective migration in general; for example, in chemotactic migration, nutrient concentration c decreases from the outward boundaries into the population over a length scale given by the interplay between nutrient diffusion and consumption. This first requirement is also satisfied by many other forms of active matter that rely on other modes of sensing to collectively migrate, for which c would generically represent the stimulus being sensed. Documented examples include durotactic cell groups ( Roca-Cusachs et al., 2013 ; Sunyer et al., 2016 ; Alert and Casademunt, 2019 ), phoretic active colloids ( Illien et al., 2017 ; Liebchen and Löwen, 2018 Stark, 2018 ), and phototactic robots ( Mijalkov et al., 2016 ; Palagi and Fischer, 2018 )—systems for which migration is directed toward regions of larger c , and therefore convex regions are more likely to be exposed to larger c . The second requirement is also satisfied for diverse active matter systems; in the context of chemotaxis, specific examples include other bacteria ( Menolascina et al., 2017 ), enzymes ( Jee et al., 2018 ; Agudo-Canalejo et al., 2018 ; Mohajerani et al., 2018 ), aggregating amoeba cells ( Keller and Segel, 1970 ), and mammalian cell groups during development, immune response, and disease ( Camley, 2018 ; Iglesias and Devreotes, 2008 ; Theveneau et al., 2010 ; McLennan et al., 2012 ; Malet-Engra et al., 2015 ; Puliafito et al., 2015 ; Tweedy et al., 2020 ). This second requirement is again also satisfied for active matter that collectively migrates using other sensing mechanisms, for which sensing has been documented to increase and eventually saturate with the stimulus, be it the stiffness of the underlying surface ( Roca-Cusachs et al., 2013 ; Sunyer et al., 2016 ; Alert and Casademunt, 2019 ), temperature ( Illien et al., 2017 ; Liebchen and Löwen, 2018 ), or light intensity ( Mijalkov et al., 2016 ; Palagi and Fischer, 2018 ). Thus, exploring the physics described here in diverse other forms of active matter will be a useful direction for future work. As a final illustration of the necessity of the sensing function f ( c ) to be concave, f ′ ′ ( c ) < 0 , we repeat our analysis but instead consider a strictly linear f ( c ) = c / c lin , which does not saturate. We choose c lin = ( 1 / c - - 1 / c + ) - 1 so that the linear f ( c ) matches our original logarithmic f ( c ) at small c . With this linear sensing function, the chemotactic response is independent of concentration, f ′ ( c ) = 1 / c lin , and the condition of concavity is violated: f ′ ′ ( c ) = 0 . We therefore expect chemotactic smoothing to not occur. Consistent with our expectation, repeating the analysis underlying Figure 4C but for the strictly linear f ( c ) yields fronts for which valleys no longer move faster than peaks. Instead, as shown in Figure 4—figure supplement 2 , the profile of chemotactic velocity is now inverted with respect to that of the bottom panel in Figure 4C . Hence, the front does not smooth. Overall, this sample computation illustrates a way of modifying f ( c ) that abrogates sensing saturation and hence would prevent chemotactic smoothing. Broader implications of chemotactic smoothing The chemotactic smoothing process described here is autonomous, arising without any external intervention. Instead, it is a population-scale consequence of the limitations in cellular signal transduction—motivating future studies of other population-scale effects, beyond smoothing, that may emerge from individual behaviors. Indeed, while studies of chemotaxis typically focus on the role of the external nutrient gradient in driving cellular migration, our work highlights the distinct and pivotal role played by the cellular chemotactic response function in regulating migration and large-scale population morphology more broadly. Our work therefore contributes a new factor to be considered in descriptions of morphogenesis, which thus far have focused on the role of other factors—such as differential forcing by signaling gradients, differential proliferation, intercellular mechanics, substrate interactions, and osmotic stresses ( McLennan et al., 2012 ; Fujikawa and Matsushita, 1989 ; Bonachela et al., 2011 ; Nadell et al., 2010 ; Farrell et al., 2013 ; Trinschek et al., 2018 ; Allen and Waclaw, 2019 ; Beroz et al., 2018 ; Fei et al., 2020 ; Yan et al., 2019 ; Yan et al., 2017 ; Copenhagen et al., 2020 ; Smith et al., 2017 ; Ghosh et al., 2015 ; Zhang et al., 2021 )—in regulating the overall morphology of cellular communities and active matter in general."
} | 3,382 |
11574056 | PMC56898 | pmc | 9,625 | {
"abstract": "Background Most organisms have developed ways to recognize and interact with other species. Symbiotic interactions range from pathogenic to mutualistic. Some molecular mechanisms of interspecific interaction are well understood, but many remain to be discovered. Expressed sequence tags (ESTs) from cultures of interacting symbionts can help identify transcripts that regulate symbiosis, but present a unique challenge for functional analysis. Given a sequence expressed in an interaction between two symbionts, the challenge is to determine from which organism the transcript originated. For high-throughput sequencing from interaction cultures, a reliable computational approach is needed. Previous investigations into GC nucleotide content and comparative similarity searching provide provisional solutions, but a comparative lexical analysis, which uses a likelihood-ratio test of hexamer counts, is more powerful. Results Validation with genes whose origin and function are known yielded 94% accuracy. Microbial (non-plant) transcripts comprised 75% of a Phytophthora sojae -infected soybean ( Glycine max cv Harasoy) library, contrasted with 15% or less in root tissue libraries of Medicago truncatula from axenic, Phytophthora medicaginis -infected, mycorrhizal, and rhizobacterial treatments. Mycorrhizal libraries contained about 23% microbial transcripts; an axenic plant library contained a similar proportion of putative microbial transcripts. Conclusions Comparative lexical analysis offers numerous advantages over alternative approaches. Many of the transcripts isolated from mixed cultures were of unknown function, suggesting specificity to symbiotic metabolism and therefore candidates likely to be interesting for further functional investigation. Future investigations will determine whether the abundance of non-plant transcripts in a pure plant library indicates procedural artifacts, horizontally transferred genes, or other phenomena.",
"discussion": "Discussion Clearly, the word-counting approach provides a reliable solution to the problem of source identification with known confidence, and has several significant advantages. The reliability of the method is best justified in terms of the favorable validation test results, and is further corroborated by agreement with an analysis of GC content. In test cases where the correct answer is known a priori , results were correct within error rates expected from overlap in training sets. (Recall that α = 0.088 for comparisons between plants and stramenopiles, and α = 0.052 for comparisons between plants and bacteria.) Unlike GC content, the problem is clearly resolved by word counting with a threshold value of t = 0, and with statistical rigor, because false-positive and false-negative rates for a set of comparisons are readily computed from cumulative distributions of dissimilarity between two training sets. Optimal statistical power (minimal false-negative rate) is ensured when using a likelihood-ratio test statistic, as demonstrated by the Pearson-Neyman Theorem [ 28 ]. Further, word counting need not be trained only for the species being compared. Rather, it is sufficient that the training set be related to, but not necessarily congeners of, the species from which sequences are being compared. Sequences from several species of the genus Phytophthora were correctly distinguished from plant and bacterial sequences, and three genes from Agrobacterium tumefaciens were correctly identified as representing a bacterial sequence. However, several caveats warrant prudence. Transcribed sequences that do not encode proteins, but rather catalytic single-stranded RNAs such as transfer and ribosomal RNAs [ 32 ], should be treated independently because they are more highly conserved across taxa than messenger RNAs. Also, filtering or trimming of low-complexity repeat regions, such as poly(A) or poly(T) tracts, is helpful because comparison results can be influenced by the abundance of a single hexamer. Early in our investigations, using one set of training sequences obtained from directionally cloned P. infestans cDNAs produced results that were difficult to interpret. It eventually became clear that, as the P. infestans sequences were all single-pass reads from the 5' end of a clone generated with the T3 primer, few sequences complementary to the 3' end of the mRNA sequence were present in the training set. This meant that the hexamer AAAAAA was common, but the hexamer TTTTTT scarce. Large amounts of the poly(T) hexamer would be expected when sequencing reverse complements of mRNAs obtained from 3' sequences generated with the T7 primer. Both poly(A) and poly(T) regions were present among plant training sequences. As a result, any sequence that contained a poly(T) tract tended to resemble the plant sequences. Further, because the error rates for an inference depend on the degree to which calibration curves overlap, the best results are obtained where overlap is minimal. Despite these caveats, word counting presents a viable solution to the problem. The P. sojae -infected G. max library provides a clear example of contrast in both hexamer composition and GC content, resulting in readily diagnosed origins. Not every case is this simple. For clear separation between the two species to appear, the two must differ in composition and a detectable proportion of transcripts from each species must be present in the library. To be detectable, the proportion of transcripts present from a particular species must be greater than the error rate obtained from calibration curves. Though these criteria are true for the infected G. max library ( t < 0 for <25% of 927 transcripts), they do not appear to be true for the M. truncatula libraries we analyzed ( t < 0 for 80–99% of 890–3,017 transcripts). In the P. medicaginis interaction library, we might expect the same bimodal distribution as seen with P. sojae . However, the two libraries were prepared in different ways. The P. sojae -infected library was prepared two days after infection, using a susceptible plant host strain, so as to maximize the number of pathogen transcripts present in the host tissue [ 18 ]. Further, G. max hypocotyl tissues were infected directly with a zoospore suspension. In contrast, the P. medicaginis -infected library was prepared ten days after infection and individual plants varied in their degree of susceptibility (C. Vance, unpublished data). Plants were also inoculated in a different manner: ground mycelia were dissolved in sterile water and incubated, and the resulting inoculum was pipetted onto the soil surface, rather than the plant. These differences in how tissues were cultured prior to library preparation could have produced the disparate abundance of plant transcripts, though both libraries were prepared from plant tissues infected with Phytophthora . For mycorrhizal root libraries, we might explain the relative lack of symbiont sequences as resulting simply from a relative lack of transcripts in the host tissue. Most of the biomass in mycorrhizal roots is plant biomass [ 33 ]. We might therefore expect that most of the transcripts therein originate from the plant host. Confounding this result, the error rates in this comparison are the greatest among all the comparisons we performed, most likely because the evolutionary distance between fungi (zygomycetes and chytridiomycetes) and plants is the least among comparisons [ 34 ]. Also, zygomycete protein-coding sequences are rare in GenBank, which resulted in a small training set for these fungi, and may have amplified any biases. The high false-negative rate probably led to a failure to detect some symbiont transcripts. In nodulating root libraries, we do not expect to observe an abundance of bacterial transcripts, because bacteria generally do not form polyadenylated mRNAs [ 35 ]. As the protocols used to extract and purify mRNAs from tissue lysate for the libraries cited in this study all relied on the presence of polyadenylation sites, we generally do not expect to find bacterial transcripts. The abundance of putative microbial symbiont transcripts among sequences from a pure plant root library is difficult to interpret. The predicted portion of microbial transcripts was greater in the axenic root-hair enriched library than in mixed cultures. Error rates were greatest for comparisons between training sets from plant and pooled zygomycete and chytridiomycete sequences. Other than providing an 87% confidence level, the 13% false-positive rate does not completely explain why about 15% of root-hair enriched transcripts resemble fungal hexamer composition more closely than plants, and warrants further study. Care had been taken to avoid contaminating plant tissue cultures by culturing seedlings in covered plates. Because of concern that ethylene accumulation in covered plates could improperly stimulate nodulation-related gene expression, seedlings were treated with Ag 2 SO 4 , an inhibitor of the plants' response to ethylene [ 6 ]. Inhibition of the ethylene response could have resulted in synthesis of transcripts that are uncharacteristic of plant roots. Analysis of another axenic root-hair enriched library, particularly one provided a carbon source to identify potential contaminants, and not treated with an inhibitor of ethylene response, would be an informative test. These observations warrant further experimental scrutiny. The transcripts identified as most and least like plant or symbiont might also be studied in more detail as candidate participants in symbiosis. Symbiotic interactions, whether pathogenic or mutualistic, present novel challenges to both plant hosts and the biologists who study them. Computational approaches, in concert with experimental verification, can help resolve these challenges."
} | 2,461 |
35456872 | PMC9027962 | pmc | 9,629 | {
"abstract": "Soils in the high jungle region of Peru continuously face erosion due to heavy rain, which leads to significant nutrient losses. Leguminous plants may provide a sustainable solution to this problem due to their ability to fix atmospheric nitrogen with the help of symbiotic rhizospheric microbes that reside in their root nodules and help restore soil fertility. The aim of this study was to isolate native rhizobial strains that can form functional nodules in red kidney beans to help improve their growth, development, and yield in field conditions. Rhizobium strains were isolated from soil samples collected from coffee fields using bean plants as trap hosts. The strain RZC12 was selected because it showed good root nodule promotion and a number of PGPR (plant-growth-promoting rhizobacteria) attributes. In the field, bean plants inoculated with the strain RZC12 and co-cultivated with coffee plants produced approximately 21 nodules per plant, whereas control plants produced an average of 1 nodule each. The inoculation with RZC12 significantly increased plant length (72.7%), number of leaves (58.8%), fresh shoot weight (85.5%), dry shoot weight (78%), fresh root weight (85.7%), and dry root weight (82.5%), compared with the control. The dry pod weight produced by the plants inoculated with RZC12 was 3.8 g, whereas the control plants produced 2.36 g of pods. In conclusion, RZC12 is a promising strain that can be used in field conditions to improve the overall productivity of red kidney beans.",
"conclusion": "5. Conclusions The rhizobial authentication assay demonstrated that rhizobia strains did not develop in the strongly acidic soils of Villa Rica. In vitro and in vivo results of rhizobial effectivity obtained in the laboratory were further validated in the field. RZC12 was found to be an effective root-nodulating rhizobia strain, with the ability to promote plant growth of beans. Improvement in agronomic growth parameters, such the fresh and dry weight of plants and pods, is an important indicator for crop yield and productivity. In this study, strain RZC12 demonstrated its ability to increase crop yield by almost doubling the production of pods in field conditions.",
"introduction": "1. Introduction Over the last 50 years, the human population has rapidly increased, placing enormous pressure on the agricultural sector to increase its productivity. It is estimated that, by the year 2050, 70% more (from the current levels) food grains and other food products will be needed to feed the entire global population [ 1 ]. Recently, farmers have widely adapted the use of fertilizers and other agrochemical products to overcome soil nutritional deficiencies, meet the increasing demand for food products, and increase their revenue [ 2 , 3 ]. Among the various soil nutrients needed for optimum plant growth, nitrogen is considered to be the most important and most limiting. Proper supply management and practices for efficient usage of this particular macronutrient are necessary to improve crop productivity and yield, especially in agricultural fields suffering from its deficiency [ 1 , 4 , 5 ]. However, the extensive use of chemical fertilizers to achieve high agricultural productivity is a major contributor to soil, water, and atmospheric pollution. As is widely acknowledged, the most important challenge of this century is to increase the productivity and yield of major staple crops in an environmentally friendly, sustainable way with minimum water wastage, while utilizing the resources at our disposal to their maximum potential [ 6 , 7 , 8 ]. Soil erosion due to excessive rainfall is a very common natural phenomenon in the high-altitude forests of Peru. This phenomenon causes soil impoverishment and subsequently leads to reduced soil fertility. Sustainable agriculture, therefore, is a much-needed approach. This challenge gives agricultural lands an opportunity to recover from nutrient losses and helps restore soil fertility. Growing legumes improves overall soil quality and increases fertility through biological nitrogen fixation [ 9 , 10 ]. The major atmospheric-nitrogen-fixating organisms known are of the genus Rhizobium . The members of this genus live in a symbiotic association with plants and play an important role in increasing crop yield, especially in nitrogen-deficient soil. Rhizobium –legume symbiosis has been extensively studied and explored to improve soil fertility. This association benefits the legume crops and the crops co-cultivated with them [ 11 , 12 , 13 , 14 ]. Microbes from the genus Rhizobium are known to have the ability to capture atmospheric nitrogen by reducing it enzymatically into ammonia within the root nodules and make it readily available to the plant for their growth and development. Some rhizobia are also reported to exhibit plant-growth-promoting features, such as inorganic phosphate solubilization, indole acetic acid (IAA), cytokinins, gibberellins, and iron transportation, among others. Phaseolus vulgaris var. red kidney was introduced into Peru by the Bean Program in the 1960s after extensive field trials were carried out in the Urubamba Valley in Cusco, Peru. This plant species is now being cultivated in more Central American countries, suggesting that it has adapted to warm climates [ 15 , 16 ]. At the same time, coffee is among the most important economic crops primarily cultivated in the high jungle region of Peru. Coffee is a long-term plant, and it only begins to generate profits for farmers after three years of cultivation, when the berries are produced. In this context, the cultivation of Phaseolus vulgaris var. red kidney bean inoculated with native strains of Rhizobium , grown together with coffee plants, gives farmers a sustainable alternative to increase their income, while helping to overcome soil nitrogen deficiency. The aim of our study was to isolate and characterize native Rhizobium strains from the functional nodules of P. vulgaris var. red kidney bean plants cultivated along with coffee plants in the Chanchamayo region of Peru and to develop them as potential bio-inoculants to increase the yield of bean plants.",
"discussion": "4. Discussion Legumes can form a positive symbiotic relationship with nitrogen-fixing soil bacteria called rhizobia. The rhizobium–legume is a well-known model system that requires a complex signal exchange between both organisms [ 31 ]. Functional nitrogen-fixing root nodules are commonly red or pink. The term “functional” here denotes their ability to actively fix nitrogen for plant usage. The characteristic red or pink color appears due to the presence of leghemoglobin. Nitrogenase is another important enzyme for nitrogen fixation activity and the sustainability of the symbiotic association between the plant and rhizobia, but it is extremely sensitive to environmental oxygen [ 32 , 33 , 34 , 35 ]. Leghemoglobin provides root nodules with the ability to protect the nitrogenase from oxygen. Sometimes, this kind of symbiosis does not show good results in the absence of functional nodules. Furthermore, nodulation failure in acidic soil conditions is a common phenomenon and is observed in soils with pH values less than 5 [ 36 ]. This finding explains the failure of root nodulation in plants cultivated in soil samples A2 and B2 (Villa Rica), which have very low pH values. Low soil pH was found to reduce nodule numbers on legumes such as common bean, lentil, pea, and soybean by more than 90% and nodule dry weight by more than 50%. However, some legumes species, such as Lupinus spp. and Mimosa spp., were found to exhibit nodulation under acidic soil conditions [ 37 ]. On the other hand, plants sown in rhizospheric soil B1 showed better growth and development compared to the others. This may be explained by the high number of functional nodules found in their roots compared to the plants sown in the other soil samples. Nodules’ development is influenced by soil pH and organic matter content. Some rhizobia strains are also known to secrete certain plant growth hormones, such as IAA, which has a positive effect on plant growth and development. IAA also plays an important role in the formation and development of root nodules. In addition, Rhizobium species with phosphate solubilization capability can also effectively release phosphorous, another important macronutrient for plants, from complex inorganic and organic compound pools and therefore have the potential to improve plant yield and reduce fertilizer requirements [ 38 ]. Rhizobia strains with the abilities to produce IAA and solubilize phosphate can improve the growth and quality of a variety of crops [ 39 ]. These PGPR attributes are important in intercalated crops, such as the ones included in this study. All strains in the present study could solubilize phosphate and secrete IAA. Among the strains tested, RZC12 and RZC17 showed the best ability to produce IAA (48.9 and 41.4 ppm, respectively). The efficacy of the rhizobial strains infecting nitrogen-fixing nodules was determined using the authentication test. Finding other non-rhizobial endophytes inside the root nodules is not uncommon; therefore, plant authentication assays were carried out to distinguish these from the rhizobial strains [ 40 , 41 ]. Non-nodulating rhizobia could also be found in the nodules. These rhizobia probably lost one of the symbiotic plasmid-containing genes necessary for their ability of root nodulation [ 42 ]. All eleven strains isolated in this study showed positive results in the authentication tests and so were assigned to the genus Rhizobia . Sometimes, symbiotic associations between the plant and rhizobia remain ineffective, with no significant benefit for the plant. A greater production of nodules means a greater productivity of the plant, but not in all cases. In a study carried out on beans, the inoculation of R. tropici gave a higher nodule dry weight compared to R. ethyli , while this last strain promoted a significant increase in the dry weight of pods [ 43 ]. The capacity for N 2 fixation can be assessed by comparing yields of inoculated plants with the +N controls as well as with the commercial inoculant strains. Strains can be ranked by comparing yield as a percentage of that achieved by the +N treatment or by the best strain or by the commercial strain, as required [ 44 ]. Numerous studies have shown the positive effects of inoculating native rhizobia strains in legumes compared to uninoculated controls under field conditions [ 45 ]. In our study, the majority of the strains showed an increase in shoot and root weights compared with the control. The most effective strain was RZC12, which showed a significant increase in fresh root (105%) and shoot (96%) weight and a tendency to improve root (127%) and shoot (100%) dry weight. Positive results were shown in a similar experiment using Rhizobium spp. and Bacillus spp. strains as inoculants. Growth parameters, such as plant height, fresh and dry aerial weight, and number of flower buds, resulted in a significantly high value for common bean grain yield, compared to the untreated control [ 46 ] In vivo plant experiments conducted under laboratory conditions showed that strains RZC12 and RZC17 could significantly increase the plant shoot length, which could be linked with their ability to fix atmospheric nitrogen and other PGPR traits [ 44 ]. Among the two strains, RZC12 could induce the formation of the maximum number of root nodules and exhibited better PGPR abilities, which steered us to select this strain for further field studies. Rhizobium strains are not only N-fixers in symbiosis with legumes but are also major promoters of plant growth due to their production of phytohormones and solubilization of phosphates [ 47 ]. Numerous studies have linked the nitrogen-fixing activity of rhizobia strains with the improvement in plant yields, measured as an increase in the production of biomass dry weight [ 48 , 49 ]. This study highlights the effectiveness of RZC12 to nodulate red kidney bean plants and significantly increase most of the agronomical parameters evaluated, including shoot/root fresh and dry weight, numbers of pods per plant, and dry pod weight per plant. These results are not surprising because it is widely known that symbiotic root rhizobia have the ability to fix high amounts of atmospheric nitrogen and make it readily available to plants for their growth requirements [ 50 ]. Moreover, in the case of legume crops, pods are also considered marketable products for the farmers and so are included in the estimations of overall bean crop yield [ 50 , 51 ]. This study demonstrated that the application of RZC12 can improve the yield of red kidney bean crops and can be used as a part of the biological management of soil fields linked to co-cultures or rotation systems."
} | 3,210 |
32063827 | PMC6999159 | pmc | 9,632 | {
"abstract": "Stochastic gradient descent requires that training samples be drawn from a uniformly random distribution of the data. For a deployed system that must learn online from an uncontrolled and unknown environment, the ordering of input samples often fails to meet this criterion, making lifelong learning a difficult challenge. We exploit the locality of the unsupervised Spike Timing Dependent Plasticity (STDP) learning rule to target local representations in a Spiking Neural Network (SNN) to adapt to novel information while protecting essential information in the remainder of the SNN from catastrophic forgetting. In our Controlled Forgetting Networks (CFNs), novel information triggers stimulated firing and heterogeneously modulated plasticity, inspired by biological dopamine signals, to cause rapid and isolated adaptation in the synapses of neurons associated with outlier information. This targeting controls the forgetting process in a way that reduces the degradation of accuracy for older tasks while learning new tasks. Our experimental results on the MNIST dataset validate the capability of CFNs to learn successfully over time from an unknown, changing environment, achieving 95.24% accuracy, which we believe is the best unsupervised accuracy ever achieved by a fixed-size, single-layer SNN on a completely disjoint MNIST dataset.",
"conclusion": "4.5. Conclusion We presented a biologically-inspired dopaminergic modulation of synaptic plasticity to exploit STDP locality. Trained stimulation during the presentation of novel inputs allows the system to quickly perform isolated adaptation to new information while preserving useful information from previous tasks. This method of controlled forgetting successfully achieves lifelong learning. Our Controlled Forgetting Networks show only a slight reduction in accuracy when given the worst possible class ordering, i.e., completely sequential without revisiting previous classes, while successfully avoiding catastrophic forgetting.",
"introduction": "1. Introduction Artificial neural networks have enabled computing systems to successfully perform tasks previously out of reach for traditional computing, such as image and audio classification. These networks, however, are typically trained offline and do not update during deployed inference. One of the current obstacles preventing fully autonomous, unsupervised learning in dynamic environments while maintaining efficiency is the stability-plasticity dilemma , or the challenge of ensuring that the system can continue to quickly and successfully learn from and adapt to its current environment while simultaneously retaining and applying essential knowledge from previous environments (Grossberg, 1987 ). There have been a handful of terms used in literature to describe the process of learning from data that is temporally distributed inhomogeneously, such as the terms incremental learning, sequential learning, continual learning, and lifelong learning. In this work, we will use the term “lifelong learning.” Lifelong learning is the process of successfully learning from new data while retaining useful knowledge from previously encountered data that is statistically different, often with the goal of sequentially learning differing tasks while retaining the capability to perform previously learned tasks without requiring retraining on data for older tasks. When traditional artificial neural networks are presented with changing data distributions, more rigid parameters interfere with adaption, while more flexibility causes the system to fail to retain important older information, a problem called catastrophic interference or catastrophic forgetting . Biological neuronal systems dont seem to suffer from this dilemma. We take inspiration from the brain to help overcome this obstacle. To avoid catastrophic forgetting, important information from older data must be protected while new information is learned from novel data. Non-local learning rules may not provide such isolation. Localized learning, on the other hand, may provide the desired segmentation while also being able to perform unsupervised learning, which is critical for lifelong learning in unknown environments. Spike Timing Dependent Plasticity (STDP) is a localized biological Hebbian learning process where a synaptic weight's adjustment is a function of the timing of the spikes , or firing events, of its locally connected pre- and post-synaptic neurons. Spiking Neural Networks (SNNs), which have been explored for their potential energy advantages due to sparse computing (Han et al., 2018 ), have been shown to perform successful unsupervised clustering tasks with STDP (Diehl and Cook, 2015 ). However, even though STDP learning is localized, it is still susceptible to catastrophic forgetting because the algorithms that employ STDP are traditionally designed for randomized input ordering. Certain features, such as homeostasis, attempt to distribute the effect of input groupings globally in order to benefit from the full network. Without a temporally uniform distribution of classes, traditional STDP algorithms still lose important older information, which is either replaced by or corrupted with information from newer samples (Allred and Roy, 2016 ). We present a new learning paradigm, inspired by the dopamine signals in mammalian brains that non-uniformly, or heterogeneously modulate synaptic plasticity. We create Controlled Forgetting Networks (CFNs) that address the stability-plasticity dilemma with rapid/local learning from new information, rather than the traditional gradual/global approach to learning. Our approach allows fixed-size CFNs to successfully perform unsupervised learning of sequentially presented tasks without catastrophically forgetting older tasks. Many recent papers have tackled the challenge of lifelong learning without catastrophic forgetting, but they are not designed to target the goal of this paper, which is autonomous learning on a deployed neuromorphic system. This goal requires real-time unsupervised learning, energy efficiency, and fixed network resources. Wysoski et al. ( 2006 ), Srivastava et al. ( 2013 ), Wang et al. ( 2014 ), Wang et al. ( 2015 ), Rusu et al. ( 2016 ), Fernando et al. ( 2017 ), Kirkpatrick et al. ( 2017 ), Lee et al. ( 2017 ), Aljundi et al. ( 2018 ), Li and Hoiem ( 2018 ), Bashivan et al. ( 2019 ) and Du et al. ( 2019 ) all employ supervised or reinforcement learning methods, in some way provide the network with the knowledge of when a task change occurs, or provide access to previous samples for retraining. For example, the work by Aljundi et al. ( 2018 ) requires that the system be allowed a parameter-“importance update” period on the older task(s) before proceeding to a new task. Similarly, Panda et al. ( 2018 ) requires that samples from earlier distributions be presented in disproportionately larger quantities than later distributions to avoid catastrophic forgetting, which would require knowledge of a task change. Additionally, Srivastava et al. ( 2013 ), Rusu et al. ( 2016 ), Fernando et al. ( 2017 ), Kirkpatrick et al. ( 2017 ), Lee et al. ( 2017 ), Li and Hoiem ( 2018 ) and Rios and Itti ( 2018 ) are also not applicable to localized learning rules that may be employed on spiking networks. And Wysoski et al. ( 2006 ), Dhoble et al. ( 2012 ), and Wang et al. ( 2017 ) are morphological systems that do not work with static-sized networks, which would exclude them from direct mapping onto physical hardware implementations.",
"discussion": "4. Discussion In this section, using a qualitative analysis we discuss reasons why the non-dopamine SNNs failed at lifelong learning in the disjoint scenario and how the CFNs avoided those failures. We also discuss the expected sequential penalty and graceful degradation of accuracy. 4.1. A Qualitative Analysis In these fully-connected one-layer SNNs, each neurons weight vector can be viewed as a reference vector that captures a specific input representation, ideally successfully generalized. As such, we may qualitatively observe the success of dopaminergic learning over time by viewing these representations. For a better visual demonstration of the disjoint scenario, we show the weights of the networks for the first four digits ‘0' through ‘3' in Figure 10 , with 100 neurons arranged in a 10x10 grid. Figure 10 Grid view of the weight vectors of reference neurons over time, showing the first four digits, learning ‘0' through ‘3' for (A) the proposed CFN, (B) a non-dopamine SNN without homeostasis, and (C) a non-dopamine SNN with homeostasis, each with 400 neurons, although only the 100 top-firing neurons are shown for space. For the CFN, digits highlighted in dashed green are examples of successfully learned generalized representations. Digits highlighted in dotted orange are examples of outlier representations. Digits highlighted in solid blue are examples of representations preserved from previous tasks. Also shown is (D) another non-dopamine SNN with homeostasis, but with reduced learning on each digit, showing catastrophic interference between classes causing corruption. Note that in the CFN case ( Figure 10A ) there are two very distinct categories of representations. The digit representations that appear to have a more consistent pixel intensity and a more consistent line width and curvature are generalized representations refined by many similar samples in a cluster. On the other hand, the digit representations that appear less defined and with more irregularity in pixel intensity are outlier representations from only one or a few samples. Notice that the digit representations that are preserved from one task to another are the useful generalizations rather than the outliers, which on the other hand are the first to be overwritten when space for a new task is required. In addition, the representations that are preserved from previous tasks experience very little and infrequent corruption during later learning stages. The dopamine signals are able to successfully replace old information with new information without interference and while maintaining accuracy because of the targeted localization. In contrast, we can visually see the failure of the non-dopamine SNNs in the disjoint scenario. In the network without homeostasis ( Figure 10B ) we see that only a few neurons experienced any learning. Without homeostasis the neurons that fired first migrated closer to the input distributions and dominated the firing activity. Even when the input distribution changed between tasks, the already used neurons were closer to the new distributions than the unused neurons with random weight vectors. Continuing the reuse the same neurons caused the SNN to overwrite and forget previous tasks. Next, in the network with homeostatic adaptive thresholding ( Figure 10C ), we see a better use of network resources from the distributed firing activity. But without targeted dopaminergic modulation homeostasis distributes the learning for a new task over all the neurons previously used in earlier tasks. Even when the learning per-digit is reduced ( Figure 10D ), the activity for the new tasks are still globally distributed by the adaptive thresholding, causing corruption between tasks. The CFNs with dopaminergic learning avoid globally distributing firing activity during a single task by not having traditional homeostatic adaptive thresholding. In addition, the CFNs avoid continuing to reuse the same neurons by proactively identifying novel data and targeting specific neurons to learn the novel data, preserving essential information from previous tasks. We note that for the failed networks where older classes are entirely overwritten by new classes, the networks still report some, albeit poor, accuracy for the forgotten tasks. This is because the varied intra-class distributions can still be somewhat useful at differentiating inter-class distributions. For this purpose, the accuracy comparisons to the SNNs with random weights are essential at identifying catastrophic forgetting, indicating that around 40–50% is a failure baseline for unsupervised learning using SNNs of these sizes on the MNIST dataset. 4.2. The Expected “Sequential Penalty” We see that the CFNs in the disjoint scenario perform on par with the interleaved scenario, averaging only a 1.04% accuracy reduction across all sizes. This penalty is expected due to sequentializing the tasks. In fact, such a penalty may be impossible to completely avoid, as the interleaved scenario provides more information to the network throughout training by providing all distributions up front, whereas the disjoint scenario never provides an opportunity to temporally overlap learning of different distributions. Even so, the sequential penalty for the CFNs is minimal, and may be acceptable given the systems avoidance of catastrophic failure in the disjoint scenario. In fact, even with this penalty, the 6400 neuron CFN achieves a respectable 95.24% test accuracy after lifelong learning, which we believe is the best unsupervised accuracy ever achieved by a fixed-size, single-layer SNN on a completely disjoint MNIST dataset. The CFNs in the disjoint scenario even outperform (Diehl and Cook, 2015 ) in all cases for which they provide results, even though that work is in the interleaved scenario. 4.3. Graceful Degradation Instead of Catastrophic Forgetting Controlled forgetting allows the network to gracefully degrade its accuracy in exchange for the ability to learn new tasks with limited resources, rather than failing. The true success of a lifelong learning system is shown not just by the final accuracy, but also by its performance throughout the training process and across training tasks. Notice how in Figure 8 while the system expectedly performs better for some tasks rather than others, there is no single task for which the system fails; i.e., the sequential penalty is spread between tasks. In fact, the lifelong system performs best at the same tasks (digits ‘0,’ ‘1,’ and ‘6’) and worst at the same tasks (digits ‘8’ and ‘9’) that the offline/non-lifelong system does. We believe that this type of approach with modulated plasticity and targeted stimulation can be useful for allowing deployed systems to gracefully adapt to changing environments rather than failing to adapt or requiring frequent offline retraining. 4.4. Future Work We expect that a deeper network will improve accuracy beyond that of these results and allow for learning of more complicated datasets. As mentioned earlier, in a deeper network, it may be that only the last few layers would require lifelong learning, performing a readout from a liquid state machine or a fixed feed forward network sufficiently pre-trained on low-level representations. We also plan to evaluate this method on time-encoded signals to improve sparsity and energy efficiency. Further, we hope to explore other dopaminergic weight adjustment policies that have a higher time-dependence or weight policies with habituation, such as in Panda et al. ( 2018 ), in order to allow for operation in an environment of changing priorities, and not just temporally separated tasks. 4.5. Conclusion We presented a biologically-inspired dopaminergic modulation of synaptic plasticity to exploit STDP locality. Trained stimulation during the presentation of novel inputs allows the system to quickly perform isolated adaptation to new information while preserving useful information from previous tasks. This method of controlled forgetting successfully achieves lifelong learning. Our Controlled Forgetting Networks show only a slight reduction in accuracy when given the worst possible class ordering, i.e., completely sequential without revisiting previous classes, while successfully avoiding catastrophic forgetting."
} | 3,958 |
30317133 | null | s2 | 9,635 | {
"abstract": "Learning in physical neural systems must rely on learning rules that are local in both space and time. Optimal learning in deep neural architectures requires that non-local information be available to the deep synapses. Thus, in general, optimal learning in physical neural systems requires the presence of a deep learning channel to communicate non-local information to deep synapses, in a direction opposite to the forward propagation of the activities. Theoretical arguments suggest that for circular autoencoders, an important class of neural architectures where the output layer is identical to the input layer, alternative algorithms may exist that enable local learning without the need for additional learning channels, by using the forward activation channel as the deep learning channel. Here we systematically identify, classify, and study several such local learning algorithms, based on the general idea of recirculating information from the output layer to the hidden layers. We show through simulations and mathematical derivations that these algorithms are robust and converge to critical points of the global error function. In most cases, we show that these recirculation algorithms are very similar to an adaptive form of random backpropagation, where each hidden layer receives a linearly transformed, slowly-varying, version of the output error."
} | 341 |
31700097 | PMC6838118 | pmc | 9,637 | {
"abstract": "Microbes in subsurface coal seams are responsible for the conversion of the organic matter in coal to methane, resulting in vast reserves of coal seam gas. This process is important from both environmental and economic perspectives as coal seam gas is rapidly becoming a popular fuel source worldwide and is a less carbon intensive fuel than coal. Despite the importance of this process, little is known about the roles of individual bacterial taxa in the microbial communities carrying out this process. Of particular interest is the role of members of the genus Pseudomonas , a typically aerobic taxa which is ubiquitous in coal seam microbial communities worldwide and which has been shown to be abundant at early time points in studies of ecological succession on coal. The current study performed aerobic isolations of coal seam microbial taxa generating ten facultative anaerobic isolates from three coal seam formation waters across eastern Australia. Subsequent genomic sequencing and phenotypic analysis revealed a range of ecological strategies and roles for these facultative anaerobes in biomass recycling, suggesting that this group of organisms is involved in the degradation of accumulated biomass in coal seams, funnelling nutrients back into the microbial communities degrading coal to methane.",
"introduction": "Introduction Globally, coal represents a key fuel, accounting for almost ~25% of the world’s energy consumption (International Energy Agency). The use of coal for power generation, however, is associated with significant environmental and health impacts 1 , 2 . As a cleaner alternative, methane derived from coal, known as coal seam gas (CSG) or coal bed methane has become an increasingly important ‘bridge’ fuel for a global transition to renewables. With the discovery that significant portions of the worlds CSG are produced through microbial degradation of the organic matter in coal to methane, has come an increasing interest in understanding the microbial communities involved in this process, with the aim of stimulating CSG production from coal reserves 3 , 4 . Efforts to understand the process of biological coal degradation have focussed primarily on either (1) understanding what nutrients to add to stimulate microbial communities to degrade organic matter in coal to methane, and/or; (2) characterising the microbial community compositions on coal and associated formation waters either in situ on or in laboratory grown microcosms (reviewed in Ritter et al ., 2015). These studies have typically used 16S rRNA gene based community profiling. This approach has been informative in characterising the members of coal-associated microbial communities, including which taxa appear to be ubiquitous in coal systems and those which appear to be dependent upon local conditions such as coal rank, depth and temperature. This type of research has enabled some generalisations to be made about the composition of these communities: all contain ubiquitous methanogenic archaea, a small number of very abundant bacterial taxa from the Proteobactera or Actinobacteria; and less frequently Firmicutes, and a long tail of rare organisms from a range of phyla 3 , 5 , 6 . One key limitation of this approach is that assigned functions of any these uncultured organisms are obtained from related, cultured taxa. While this is useful for some groups such as the methanogenic archaea, where metabolic capabilities are largely constrained (see also 7 , 8 ), it is considerably less reliable in the majority of bacterial groups where metabolic functions are not always taxonomically conserved. In order to address this, some effort to assign functions to coal seam microbes has been attempted through genome reconstructions from metagenomic sequences 9 – 11 . This approach has generated some insights into the metabolisms of members of these environments, notably the discovery of a novel methanogenic taxon 10 , however, as many genes in the process are unknown the effectiveness of this method is limited. Finally, classical isolation into either axenic or gnotobiotic culture can be used to elucidate the metabolic roles of individual organisms. Unlike metagenomics-based approaches, isolation has the advantage of being able to experimentally test the metabolic and physiological characteristics of the isolated taxa through growth studies, though a key drawback remains that many microbial species are recalcitrant to isolation. In 2016, Vick et al . identified the presence of several early coal-colonists in a Sydney Basin formation water. One of the highly abundant OTUs (Operational Taxonomic Unit) observed in this 16S rRNA amplicon sequence based study was taxonomically identified as a pseudomonad (OTU_9), and its presence during early colonisation suggested it may play some role in degrading organic matter in coal. Interestingly, Pseudomonas species have been previously identified from almost all surveys of coal seams conducted to date 6 , 12 – 25 and pseudomonads isolated from coals in the past have been reported for the production of biosurfactants 26 and lignin degradation phenotypes 27 . Additionally, a previous study has noted the prevalence of seemingly aerobic taxa and metabolisms in deep anoxic hydrocarbon environments 28 . In order to try to assign functions to these coal-associated pseudomonads, and functionally related facultative aerobic taxa, a culturing effort was undertaken to bring facultatively aerobic organisms, including that corresponding to OTU_9, into axenic culture. This isolation was followed by genomic and phenotypic characterisation to uncover the metabolic and ecological roles played by these organisms in coal seams. As these organisms are facultative aerobes, the present study employed oxic conditions and a complex medium to obtain axenic cultures.",
"discussion": "Discussion Pseudomonas represents a ubiquitous bacterial genus observed in almost all surveyed coal seam microbial communities. Despite this, the ecological and metabolic roles played by these pseudomonads and related facultatively aerobic taxa in coal seams remains poorly understood. The current study aimed to isolate representatives of these pseudomonads from multiple coal seam environments and to characterise their capabilities both genomically and phenotypically, in order to better understand their roles in this environment. Aerobic isolation from three coal seam formation waters on peptide rich media successfully generated two pseudomonad isolates from different species clades. This isolation effort also resulted in the generation of eight additional bacterial isolates from the actinobacterial genera Tessaracoccus and Actinotalea and the proteobacterial genera Thauera , Marinobacter , Chelatococcus , Citrobacter and Vibrio . As a phylogenetically diverse range of bacterial isolates were obtained it was decided to pursue genomic and phenotypic characterisation of all bacterial isolates to investigate putative ecological functions for this set of facultatively anaerobic, heterotrophic bacteria. The cellular morphology of all isolates was examined through scanning electron microscopy, carbon metabolisms characterised through Biolog phenotype screening and genomes sequenced and annotated to examine ecophysiologically relevant genes and genetic elements. The roles of microbes in coal seam microbial communities are typically thought of in terms of their carbon catabolism, where the organic components of coal are sequentially degraded to methane in a linear fashion by a succession of microbial species. The ecological functions in a microbial community, however, are more complex and include a range of community functions including primary and secondary degradation, predation, scavenging and biomass recycling, amongst others. In fields of macro-ecology these ecological roles are often interrogated in terms of life strategy categories such as the competitive, stress tolerant and ruderal (CSR) classification used to classify the ecological strategies of plants during succession 42 . These lifestyle classifications and the catabolisms and phenotypes associated with them are, in microbial systems, dependent upon the type of environment and how the energy and nutrient resources are introduced into the environment 43 . The coal seam constitutes an endogenous heterotrophic environment in that it is a closed system without regular microbial, nutrient or energy inputs from external sources 43 . In this type of environment initial colonising ruderal species are heterotrophic organisms which modify the physicochemistry of the environment through their metabolic processes until the environment is modified to a point where slower growing, stress tolerating organisms or those with specialised metabolic capabilities are able to compete and proliferate 43 . Examination of the genotypic and phenotypic analysis of isolate from the current study suggests the presence of three broad ecological life strategies. The first resembles a ruderal life strategy and is shared by Citrobacter SUR-1, Citrobacter BOW-7 and Pseudomonas BOW-2, characterised by rapid growth on common, labile biomolecules with few genes associated with specialised metabolisms. The second life strategy appeared to be an opportunotrophic lifestyle as described by Singer et al . 44 . This was characterised by genes associated with diverse metabolisms suitable for survival across a wide range of different environments and was observed in Thauera SYD-3, Marinobacter SUR-4 and Pseudomonas SYD-2. The third life strategy group represents the specialists. This group is broader in function and includes different specialist strategies but shares the common trait of having narrower metabolisms with a greater genomic investment in a targeted catabolic process. Isolates which could be considered to have specialist life strategies include Chelatococcus BOW-1 and Actinotalea SUR-A1. In addition to isolates that had characteristics aligning with known ecological strategies, two isolates ( Tessaracoccus SUR-6 and Vibrio SUR-5) had life strategies that did not appear to conform to known models based on the genotypic and phenotypic characterisation performed here (Fig. 7 ). From our analyses of the isolates and their genomes described here, it is not believed that any of these isolates partakes in a lifestyle or strategies centred around direct biodegradation of coal compounds in situ . Figure 7 Conceptual model of the ecologically relevant capabilities of bacterial isolates determined through genomic and phenotypic analysis. The first life strategy group constituted isolates displaying a ruderal life strategy. Comparison of the isolates from the current study to OTUs observed in a previous study of succession in coal seams, using the CSMB OTU reference set 6 , shows that Citrobacter SUR-1, Citrobacter BOW-7 and Pseudomonas BOW-2 matched to OTUs which, in a previous study, showed initial rapid proliferation, quickly dominating the communities, before declining in abundance at later time-points 45 . This pattern of growth is characteristic of ruderals which rapidly proliferate on the most labile compounds available in an environment before being replaced by microbes with more specialised metabolisms once this labile resource is exhausted. In a previous study, these OTUs matching Citrobacter SUR-1, Citrobacter BOW-7 and Pseudomonas BOW-2 have also showed a very high level of variation in relative abundance across replicate microcosms compared to other taxa in the community. This high variation in abundance is due to competition by different ruderal taxa in other replicates as success as a ruderal is often decided by small stochastic affects during inoculation 46 . The phenotypic and genotypic features presented in the current study support this hypothesis of a ruderal lifestyle and suggest catabolic niches supporting them. These putatively ruderal isolates all showed growth on a wide variety of amino acids, peptides and carboxylic acids (Fig. 3 ). The genomes of these organisms also contained large numbers of peptidase genes and membrane transport genes for amino acids, sugars and carboxylic acid transporters (Supplementary Material 1 ). This collection of genes and phenotypes suggests that growth on cellular debris, particularly proteinaceous material, may drive the initial ruderal response in coal seam microbial communities. This cellular debris likely originates from cell death due to stress associated with the inoculation event but in native coal environments cellular debris could be liberated by chemical or physical perturbations to the coal seams or the ingress of meteoric waters. The second life strategy type observed amongst our isolates was an opportunotrophic lifestyle, sensu Singer et al . 44 . Isolates conforming to this grouping ( Thauera SYD-3, Marinobacter SUR-4 and Pseudomonas SYD-2) have been observed in a previous study of succession in coal seams 45 . In this previous study these opportunotrophic organisms remained at low abundances during the succession process 45 . These were also the only isolates to have considerable numbers of genes involved in aromatic hydrocarbon degradation (Fig. 4 ). Interestingly, in all of these organisms the pathways proceed through the metabolic intermediate catechol, an intermediate indicative of aerobic catabolism 47 . As all the eastern Australian coal seams sampled in the current study are highly anoxic and chemically reduced it is unlikely that these isolates are using these pathways for aromatics degradation in the coal seam under the conditions from which they were isolated. It should be noted, however, that these aromatic molecules are common in these coal seams and these may become available in the more oxic regions of aquifers into which coal seam formation waters may migrate. This observation of genes involved in aerobic aromatics catabolism, along with the facultative anaerobic growth patterns of these isolates, suggests that many of these organisms have diverse metabolic potentials allowing for persistence and growth across a large range of environments, scavenging diverse carbon compounds when environmental conditions allow it. A lifestyle in line with that described for the prototypical opportunotroph Marinobacter aquaeolei 44 , 48 . It is noteworthy that two pseudomonads with distinctly different metabolic and ecological lifestyles were observed in this study. This likely reflects the wide metabolic and lifestyle diversity in the genus which spans parasitism to mutualism 49 – 51 and free-living lifestyles 52 . Amongst the isolates to be aligned with the specialist life strategy group is Chelatococcus BOW-1. Chelatococcus BOW-1 displays very high number of diverse membrane transport genes, particularly ABC transporters involved in the uptake of amino acids, sugars, and polyamines. This factor, coupled with its growth on a wide range of carbon sources and high peptidase gene number but non-ruderal succession observed in previous studies and paucity of genes involved in aromatic hydrocarbon degradation suggests that this organism likely specialises in scavenging low quantities of diverse cellular material for nutrients and energy. This lifestyle would allow this organism to survive in diverse environments so long as other organisms were present to provide it with biomass to recycle. Similarly, Actinotalea SUR-A1 showed a distinctive ecophysiological profile setting it apart from the other isolates examined. Actinotalea SUR-A1, in contrast to Chelatococcus BOW-1 showed a reduced metabolic diversity compared to other isolates having a low number of transporters and growing on very few carbon compounds. Actinotalea SUR-A1 did, however, display a large number of carbohydrate active enzymes including all components involved in the formation of a cellulosome 53 . This suggests Actinotalea SUR-A1 may be responsible for complex polysaccharide degradation in the coal seam, presumably from biofilm or extra-cellular material as has been previously reported to be produced by coal seam microbial communities 24 , 45 . The genome of Actinotalea SUR-A1 also displayed a number of traits involved in cellular defence including a large number of CRISPR elements and a large number of daunorubicin and other multidrug efflux pumps often associated with antibiotic production in Actinobacteria 54 . Together with the observed production of spores by this isolate (Fig. 1 ), these results indicate that Actinotalea SUR-A1 follows a stress tolerator lifestyle, relying on a specialised metabolism based around carbohydrate catabolism rather than a broader opportunitrophic lifestyle, although sharing metabolism of common biomolecules as well as the ability for environmental persistence through sporulation. A small number of isolates from the current study ( Tessaracoccus SUR-6 and Vibrio SUR-5) did not appear to conform to any clear life strategy models based on the genotypic and phenotypic characterisations performed in the current study. It may be that these organisms are also involved in biomass recycling albeit on a smaller range of cellular biomolecules than the other scavenger and opportunotroph organisms. Alternatively, it may be that these organisms are involved in novel or unexamined catabolic niche processes in coal or other subsurface waters. From a broader perspective, some notable differences were observed in the numbers of insertion sequence and CRISPR elements between the different isolate genomes. One interesting observation was the large difference in the number of insertion sequence elements observed in the closely related Citrobacter isolates BOW-7 and SUR-1 (Fig. 6 ). This increase in insertion sequence number in Citrobacter sp. SUR-1 may indicate recent adaptations as insertion sequences have been implicated in genomic shuffling and reorganisation 55 . CRISPR elements have previously been reported as unexpectedly common in a metagenomic exploration of a deep fractured shale environment and have been suggested to be a common feature of the terrestrial subsurface 56 . The current study supports these findings with observations of CRISPR sequences in six out of ten isolate genomes examined. If this high abundance of viral particles in subsurface environments is indeed the case, then viral predation may be a contributing factor in controlling biomass turnover, cell density and species distributions in subsurface coal environments, as it has been shown to be in other microbial environments 57 . This could have implications for subsequent methane production by methanogenic coal seam communities and warrants further attention as a potential limiting factor for methane production rates from coal. Much of the literature in the field of coal seam microbiology is focused on improving coal seam gas production rates through nutrient and microbial amendments 3 . This has meant that studies have primarily focused on the microbial compositions of formation waters and coal solids, considering less the continuous nature of aquifers and whether those organisms that live within them require flexibility as they may find themselves in different physical and chemical environments along the course of the aquifer. Further, it would be valuable to be able to distinguish those taxa that are subsurface generalists from those that are coal seam specialists. A concept that has recently been explored by Barnhart and colleagues 58 . Based on their metabolisms as facultative anaerobes and their genomic potential for recycling of cellular biomass, it is likely that the isolates from the current study belong to a group of subsurface generalised organisms rather than coal seam specialists. It should be noted, however, that the current observations are made from genome sequences and expression of these genes under in situ conditions remains undetermined. Examination of expression of these genes under anaerobic conditions mimicking the coal seam environment may shed further light on the lifestyles of these microbes. In summary, bacteria in coal seams are thought to carry out the majority of steps involved in coal degradation, with methanogenic archaea responsible for only the very final stages of conversion of acetate, CO 2 or other simple methylated compounds to methane 3 . These initial stages of degradation, carried out by the bacteria taxa in the community, are thought to be the rate limiting step in the conversion of coal to methane and so identifying and characterising the bacterial taxa responsible for these processes is important for understanding the microbial coal to methane conversion process 4 . The current study reports on the successful isolation of two pseudomonads and eight other bacterial isolates, from eastern Australian coal seam formation waters. Phenotypic and genomic characterisation of these bacterial facultative anaerobes from the coal seam identified metabolisms supporting ruderal, opportunitrophic and specialist lifestyles centred around recycling carbohydrates and proteinaceous material in the coal seam. This study represents an initial step in assigning ecological and metabolic functions to the bacterial taxa observed in coal seams and utilised a protein rich isolation media and aerobic isolation conditions to target members of the genus Pseudomonas , which have been identified as ubiquitous members of coal seam microbial communities. Future work should focus on isolation strategies to target other metabolisms involved in the conversion of organic matter in the coal to methane, these include anaerobic hydrocarbon degraders to identify taxa responsible for breakdown of hydrocarbons in coal or sulphate reducing bacteria which constitute a large proportion of coal seam microbial communities in eastern Australian coal seam formation waters despite the absence of elemental sulphur and sulphate from these environments and for whom a definitive catabolic role is not yet known."
} | 5,514 |
19762644 | null | s2 | 9,640 | {
"abstract": "Metabolic pathways have traditionally been described in terms of biochemical reactions and metabolites. With the use of structural genomics and systems biology, we generated a three-dimensional reconstruction of the central metabolic network of the bacterium Thermotoga maritima. The network encompassed 478 proteins, of which 120 were determined by experiment and 358 were modeled. Structural analysis revealed that proteins forming the network are dominated by a small number (only 182) of basic shapes (folds) performing diverse but mostly related functions. Most of these folds are already present in the essential core (approximately 30%) of the network, and its expansion by nonessential proteins is achieved with relatively few additional folds. Thus, integration of structural data with networks analysis generates insight into the function, mechanism, and evolution of biological networks."
} | 224 |
25872137 | null | s2 | 9,641 | {
"abstract": "We investigated the impact of below-ground and above-ground environmental heterogeneity on the ecology and evolution of a natural plant-pathogen interaction. We combined field measurements and a reciprocal inoculation experiment to investigate the potential for natural variation in abiotic and biotic factors to mediate infection outcomes in the association between the fungal pathogen Melampsora lini and its wild flax host, Linum marginale, where pathogen strains and plant lines originated from two ecologically distinct habitat types that occur in close proximity ('bog' and 'hill'). The two habitat types differed strikingly in soil moisture and soil microbiota. Infection outcomes for different host-pathogen combinations were strongly affected by the habitat of origin of the plant lines and pathogen strains, the soil environment and their interactions. Our results suggested that tradeoffs play a key role in explaining the evolutionary divergence in interaction traits among the two habitat types. Overall, we demonstrate that soil heterogeneity, by mediating infection outcomes and evolutionary divergence, can contribute to the maintenance of variation in resistance and pathogenicity within a natural host-pathogen metapopulation."
} | 311 |
39888988 | PMC11784860 | pmc | 9,643 | {
"abstract": "Pneumatic soft robots are promising in diverse applications while they typically require additional electronics or components for pressure control. Fusing pneumatic actuation and control capabilities into a simple soft module remains challenging. Here, we present a class of bistable fabric mechanisms (BFMs) that merge soft bistable actuators and valves for electronics-free autonomous robots. The BFMs comprise two bonding fabric chambers with embedded tubes, where the straightening of one chamber compels the other to buckle for the bistability of the structure and the switching of the tube kinking. Our BFMs can facilitate fast bending actuation (more than 1166° s −1 ), on/off and continuous pressure regulation, pneumatic logic computations, and autonomous oscillating actuation (up to 4.6 Hz). We further demonstrate the capabilities of BFMs for diverse robotic applications powered by one constant-pressure air supply: a soft gripper for dynamic grasping and a soft crawler for autonomous jumping. Our BFM development showcases unique features and huge potential in advancing entirely soft, electronics-free autonomous robots.",
"introduction": "INTRODUCTION Pneumatic soft robots have drawn extensive attention owing to their safety, adaptability, simplicity, practicability, and low cost ( 1 ). Benefiting from these advantages, they have shown tremendous promise and diverse applications, such as biomimetic robots ( 2 – 6 ), soft grippers ( 7 – 10 ), and wearable devices ( 11 – 15 ). However, pneumatic soft robots, especially those composed of multiple actuators, typically require several hard electronic valves and control modules for pressure control ( 16 ), which limits their integration of fully soft components. Moreover, their applications may be hindered in environments where electronics are prone to failure, such as underwater or high-radiation zones. To achieve electronics-free autonomy in pneumatic soft robots, researchers have predominantly focused on assembling multiple classes of soft actuators and control modules. During the past decade, various soft actuators have been developed using elastomers, polymers, and fabrics, which enable a range of actuation motions including extension, contraction, bending, twisting, and hybrid movements ( 17 – 23 ). On the other hand, soft control modules have been reported through the development of pneumatic inverters ( 24 – 26 ), soft valves ( 27 , 28 ), membranes with slits ( 29 , 30 ), and narrow tubes with viscous flow ( 31 , 32 ). These soft control modules have been successfully used to design pneumatic logic gates, oscillators, nonvolatile memory storage, and pressure regulators ( 28 , 33 – 40 ), generating pressure signal sequences analogous to electronic circuit currents. In this sense, electronics-free soft robots can be conveniently realized through their combinations ( 41 , 42 ). However, these developments usually depend on different components for actuation and control, resulting in cumbersome fabrication and difficulties of integration. In tackling these challenges, one promising strategy is to integrate actuation and control capabilities into a unified design ( 28 , 38 , 43 , 44 ). As a promising attempt, Lee et al. ( 38 ) used the inherent interactions between the buckling-sheet ring oscillator and the surroundings to achieve multimodal locomotion robots. Besides, Jiao et al. ( 43 ) presented soft origami LEGO with actuation, computation, and sensing capabilities to develop intelligent soft turtle robots. Recently, Decker et al. ( 28 ) introduced programmable soft valves with inner piston actuators for linear actuation of untethered robots. Despite the accumulative results, it remains challenging to fuse the actuation and control capabilities into a simple and efficient soft module. Alternatively, snap-through bistability has been widely used to develop high-performance soft bistable actuators or valves, but not both simultaneously ( 45 ). They typically combine separate pneumatic actuators with bistable structures ( 46 – 49 ), or couple tube kinking with bistable membrane deformation ( 27 , 50 ). While directly combining existing bistable structures, pneumatic actuators, and tubes might offer a solution, it generally requires complicated configurations. Accordingly, the key challenge lies in the design of simple soft bistable structures and actuators, whose deformation can be inherently coupled with pneumatic components for control. Here, we present a class of bistable fabric mechanisms (BFMs) that merge soft bistable actuators and valves by partially bonding two fabric chambers and embedding tubes. Controlling the pressures in the two chambers, the BFMs exhibit bistability with tunable energy barriers and can transition between stable states rapidly. In each stable state, one chamber straightens while the other buckles, causing the corresponding embedded tubes to be unkinked or kinked. We demonstrate that our BFMs can be configured as bistable actuators with fast bending actuation (more than 1166° s −1 ), pneumatic circuit switches (PCSs) for interaction control, and pneumatic logic gates for autonomous control. Notably, a single BFM can multiplex the actuator and valve functions to achieve autonomous oscillating actuation (up to 4.6 Hz). Further, we develop a soft gripper capable of detecting objects and performing dynamic grasping and a soft crawler that jumps forward continuously (6.6 ± 0.8 cm s −1 ) after a press on its tail. Both are fully soft and electronics-free and achieve intelligent control under a constant-pressure air supply.",
"discussion": "DISCUSSION Bistable structures have been successfully used in developing soft actuators with fast actuation or soft valves for electronics-free control. However, achieving them both in a unified and compact design remains elusive. Herein, we present a class of BFMs with embedded pneumatic control components, capable of functioning as various soft actuators and control modules, including bistable actuators, PCSs, and pneumatic logic gates. We also demonstrate that our BFMs can multiplex their actuation and control capabilities to develop intelligent oscillators, which directly produce oscillatory bending motions up to 4.6 Hz. Therefore, the BFMs can be used as modules to construct electronics-free soft autonomous robots, such as the gripper and the crawler developed in this work. Notably, these robots primarily contain fabric and tube materials and achieve intelligent operation powered by one constant air supply. Different from most pneumatic soft bistable actuators ( 46 – 49 ), our BFMs are fully soft, simple, and compact, seamlessly integrating pneumatic actuators and bistable structures. Compared with existing soft control modules (table S1), our BFMs demonstrate not only comprehensive pneumatic control capabilities but also direct actuation capabilities. On the basis of our experiments, the BFMs can operate at pressures up to 150 kPa, which meets the requirements for most pneumatic soft robots. In addition, our BFMs are lightweight, low-cost, foldable, scalable, and easily fabricated with heat-sealing/pressing methods. To highlight these advantages, our developed gripper and crawler are fully soft and primarily use BFMs for actuation and control. However, we should mention that the BFMs require an auxiliary reference pressure ( P ref ), as fabric chambers exhibit almost no bending stiffness when unpressurized. The reference pressure, although increasing complexity, also enhances tunability. Although the air supply in this work still needs traditional air pumps and pressure regulators, these components could be replaced with soft power devices in the future. Moving forward, many opportunities exist in exploring the applications of BFMs. For example, the proportional PCS can be integrated with human joint (such as finger, wrist, and elbow) to enable real-time control of pneumatic soft robots or assistive wearable devices based on users’ motions. In summary, our BFM-based modules and robots take a step toward entirely soft, self-contained, electronics-free, and autonomous robotic systems. Moreover, we anticipate that our design method of partially bonding chambers and embedding tubes will inspire future research in multistable, reconfigurable, and intelligent structures, actuators, and robots."
} | 2,082 |
26528323 | PMC4604266 | pmc | 9,646 | {
"abstract": "The consequences of global change for the maintenance of species diversity will depend on the sum of each species responses to the environment and on the interactions among them. A wide ecological literature supports that these species-specific responses can arise from factors related to life strategies, evolutionary history and intraspecific variation, and also from environmental variation in space and time. In the light of recent advances from coexistence theory combined with mechanistic explanations of diversity maintenance, we discuss how global change drivers can influence species coexistence. We revise the importance of both competition and facilitation for understanding coexistence in different ecosystems, address the influence of phylogenetic relatedness, functional traits, phenotypic plasticity and intraspecific variability, and discuss lessons learnt from invasion ecology. While most previous studies have focused their efforts on disentangling the mechanisms that maintain the biological diversity in species-rich ecosystems such as tropical forests, grasslands and coral reefs, we argue that much can be learnt from pauci-specific communities where functional variability within each species, together with demographic and stochastic processes becomes key to understand species interactions and eventually community responses to global change.",
"conclusion": "Conclusion There is an urgent need to understand how different drivers of global change differentially but simultaneously impact ecosystems and which are the corresponding magnitude and direction of the changes in species interactions and coexistence. Recent developments of ecological theories are improving the forecast of these changes but more empirical data are needed for a solid theory of the mechanisms driving species coexistence. There are three main empirical approaches to the study of community assembly: experimental manipulations of the abiotic or biotic environment, assessments of trait-phylogeny-environment relationships, and quantification of frequency-dependent selection and population growth. Each approach alone is not strong enough to reveal which niche axes and which traits determine the outcome of competition, the extent of facilitation and the eventual structure and dynamics of the community. Thus, only the combination of these three approaches can significantly contribute both to conceptual ecology and to guidelines for ecosystem management under global change ( HilleRisLambers et al., 2012 ). Nonetheless, the combination of the three in a single research project requires an enormous effort that sometimes is unjustified. The degree of resolution would depend on the research aim. For instance, if the question is related to how species are precisely responding to a combination of different global change drivers (e.g., an increase in precipitation or aridity, an increase in nitrogen deposition, or an increase in grazing) then to study how these drivers affect species fitness could be enough. However, if the question relates to how specific species responses translate to community dynamics, then it is also necessary to study niche differences among species to know the outcome of species interactions. While the amazing richness of ecosystems like tropical forests have attracted fruitful research and theories on species coexistence, there is much that can be learnt from pauci-specific communities where the value of each single species is large and where the functional variability within each species becomes key to understand species interactions and eventual community responses to global change. In both research and conservation activities, we have to move from species coexistence to the coexistence of genotypes, paying more attention to the functional variability existing within each species.",
"introduction": "Introduction Species composition of a local community is the result of several processes and factors that act at different scales, none of them being mutually exclusive. This encompasses from features and processes that act at global and regional scales, such as randomness, historical patterns of speciation, extinction, migration as well as dispersal processes, to abiotic factors (physical constraints of the environment) and biotic interactions (both positive and negative) that act at local scale. These factors, known as hierarchical filters, act from broad to fine spatial scales to impose rules on community assembly ( Götzenberger et al., 2012 ). There are numerous theories about these filters and the coexistence mechanisms involved in the composition of species in a community. In this article we focus on those acting at local scales (Figure 1 ), but we also refer to broader scales and the corresponding interactions since they are key to understand regional and global species diversity. FIGURE 1 Influence of intraspecific variability in the filtering of potential species integrating a community. (A) classical community assembly theory without taking into account intraspecific variability and (B) community assembly theory incorporating intraspecific variability. Species with mean trait values matching the abiotic requirements and being either ecologically different or capable of tolerating competition will contribute to the eventual community. By incorporating intraspecific variability, more species will pass biotic and abiotic filters because they are able to adjust by phenotypic plasticity or simply because they are genetically variable so more species could join the community in (B) than in (A). Each shape represents a species and each color represents a given trait value within a species. Dashed lines represent abiotic and biotic filters. Biological diversity is about species interactions inter alia, and it is commonly limited by competitive exclusion and sometimes fueled by positive relationships. Competitive exclusion has a crucial role in structuring communities and has therefore prompted intensive ecological research over decades ( Pianka and Horn, 2005 ). Competition has both an evolutionary and an ecological role since it increases diversity through speciation ( Brännström et al., 2012 ) and regulates species diversity through species interactions ( Chesson, 2000 ). Classical coexistence theories establish that each species inhabits a particular niche, involving a given combination of abiotic and biotic factors, where it outcompetes the rest of the species in the local pool (i.e., niche theory; Grinnell, 1917 ; Gause, 1934 ). Under this premise, niche overlap penalizes worse competitors, which results in their exclusion from a community, and supports that species coexist by being functionally different and by exploiting different niches ( Hutchinson, 1959 ). If true, the total number of species in an ecosystem is thought to be proportional to the total range of the environment divided by the niche breadth of the species ( MacArthur and Levins, 1967 ). In contrast, neutral theory ( Hubbell, 2001 ) assumes that individuals and species are ecologically interchangeable and therefore equivalent in their competitive ability, i.e., none of the species shows an advantage or disadvantage over the others. According to the neutral theory, random processes, stochastic events, and equivalence between opposite forces are the drivers of population dynamics and species coexistence ( Bell, 2000 ; Hubbell, 2001 , 2005 ; Götzenberger et al., 2012 ). However, these theoretical frameworks seem insufficient to explain species coexistence in many natural ecosystems and numerous discrepancies have been found between theoretical predictions from classic niche theory and empirical studies ( Nathan et al., 2013 ). Here we review the theory about the mechanisms underlying the maintenance of species coexistence. Although conclusions and main concepts apply to all sort of living organisms, we have placed special focus on plant communities and, hence, on plant species coexistence and diversity. We give special attention to concepts like competition, facilitation, ecological differences among species, intraspecific variability and environmental heterogeneity. In each section, we discuss how global change may affect species coexistence through modifications in important biotic and abiotic factors. The consideration of all global change factors potentially affecting coexistence would largely exceed the limits of this article so we have focused on the best studied ones and on those illustrating different responses and cascade effects on community dynamics and species interactions. We include an analysis of biological invasions, as a large and unique ecological and evolutionary experiment of coexistence. Also, we encompass the particular case of species coexistence in pauci-specific systems, which complement the better studied cases of tropical, hyperdiverse systems."
} | 2,216 |
35457128 | PMC9025071 | pmc | 9,650 | {
"abstract": "Rapid population and economic growth, excessive use of fossil fuels, and climate change have contributed to a serious turn towards environmental management and sustainability. The agricultural sector is a big contributor to (lignocellulosic) waste, which accumulates in landfills and ultimately gets burned, polluting the environment. In response to the current climate-change crisis, policymakers and researchers are, respectively, encouraging and seeking ways of creating value-added products from generated waste. Recently, agricultural waste has been regularly appearing in articles communicating the production of a range of carbon and polymeric materials worldwide. The extraction of cellulose nanocrystals (CNCs) and carbon quantum dots (CQDs) from biomass waste partially occupies some of the waste-recycling and management space. Further, the new materials generated from this waste promise to be effective and competitive in emerging markets. This short review summarizes recent work in the area of CNCs and CQDs synthesised from biomass waste. Synthesis methods, properties, and prospective application of these materials are summarized. Current challenges and the benefits of using biomass waste are also discussed.",
"conclusion": "5. Conclusions and Outlook In recent years, LCB waste has been utilized to prepare CNCs with different attributes and applications. This choice of a feedstock is renewable, green, and affordable. Several modifications of the pre-treatment and extraction stages continue to be explored, with the aim of attaining CNCs with desired attributes (including their production at a commercial scale). In addition to the applications of CNCs in pharmaceuticals, medicine, composite materials, energy, and packaging, their application as a precursor in synthesis of carbon nanomaterials is gaining momentum. This is in line with developing new greener methods of material synthesis as well as finding cheaper ways of producing smart materials. CQDs seem to be much easier to fabricate from LCB waste sources when compared to other types of carbon nanomaterials. Several advantages and disadvantages exist in both the fields of CNC extraction and CQD fabrication from LCB waste. The popular chemical treatment method of LCB in order to obtain CNCs is lengthy and requires the continuous use of strong acids in order to completely separate the amorphous content from the cellulose. This is not environmentally friendly. The use of strong acids and bases also implies that large amounts of water will be used to purify the CNCs. Compared to other extraction methods, chemical methods are cheaper; however, process efficiencies should be evaluated. The application of LCB waste in the fabrication of carbon nanomaterials is a promising field with the potential to transform the agricultural sector as we know it. CQDs are versatile materials with the potential to replace many toxic, heavy-metal-based optoelectronic devices. Based on the current research trends, we can predict that CQDs will be popular in the market in the near future. However, large variations in the properties of CQDs as the result of differences in feedstocks means that CQDs can be optimised for a specific application per batch. Further studies on large-scale synthesis of CQDs are yet to be explored.",
"introduction": "1. Introduction The increasing demand for food as a result of population growth has resulted in an increase in agricultural production, which has consequently led to the acceleration of agricultural waste generation. This is accompanied by increased energy demands globally, depletion of fossil fuels, and climate change. Growing research interest has emerged concerning the use of biomass waste material to produce value-added products, due to its potential to form inexpensive and environmentally friendly materials without conflicting with food stock [ 1 , 2 ] Lignocellulosic biomass (LCB) is highly considered as a viable source for renewable energy and an important factor in sustainable economies. The three major building components of LCB are cellulose, hemicellulose, and lignin, with varying percentage composition. These components together with many other products can be extracted as primary or secondary products from LCB, as shown in Figure 1 . Studies on the use of biomass waste for the fabrication of carbon-based materials have emerged recently, such as the use of corncob residue for the fabrication of: porous carbon materials for supercapacitor electrodes [ 3 ], hollow spherical carbon materials for supercapacitors [ 4 ]), carbon nanosheets for lithium–sulphur batteries [ 5 ], carbon nanospheres for use as a high-capacity anode for reversible Li-ion batteries [ 6 ], and carbon quantum dots for metal ion detection [ 7 ]. Carbon quantum dots (CQDs) are the newest members of the carbon family. Since their discovery in 2004 by Xu et al. [ 8 ] and in 2006 by Sun et al. [ 9 ], they have gradually become a rising star in the ‘carbon nanomaterials’ family. CQDs are a subclass of zero-dimensional nanoparticles that consist of a carbon core and constitute different functional groups at the surface [ 10 ]. They are characterised by quasi-spherical morphology composed mainly of amorphous carbon with sp 2 -hybridised structure and a size less than 10 nm [ 11 ]. They exhibit attractive properties such as tuneable photoluminescence, functionalizability, dispersibility, multicolour emission associated with excitation, biocompatibility, size-dependent optical properties, facile synthesis, and low toxicity as compared to their counterparts (semiconductor quantum dots (QDs)) [ 12 ]. These extraordinary features make them suitable for potential applications in sensors, catalysis, healthcare, and energy storage devices [ 13 ]. In this review, we look into the most recent developments in the extraction of CNCs as well as the fabrication of CQDs from LCB waste."
} | 1,468 |
38545610 | PMC10966399 | pmc | 9,651 | {
"abstract": "Coral reefs are increasingly impacted by climate-induced warming events. However, there is limited empirical evidence on the variation in the response of shallow coral reef communities to thermal stress across depths. Here, we assess depth-dependent changes in coral reef benthic communities following successive marine heatwaves from 2015 to 2017 across a 5–25 m depth gradient in the remote Chagos Archipelago, Central Indian Ocean. Our analyses show an overall decline in hard and soft coral cover and an increase in crustose coralline algae, sponge and reef pavement following successive marine heatwaves on the remote reef system. Our findings indicate that the changes in benthic communities in response to elevated seawater temperatures varied across depths. We found greater changes in benthic group cover at shallow depths (5–15 m) compared with deeper zones (15–25 m). The loss of hard coral cover was better predicted by initial thermal stress, while the loss of soft coral was associated with repeated thermal stress following successive warming events. Our study shows that benthic communities extending to 25 m depth were impacted by successive marine heatwaves, supporting concerns about the resilience of shallow coral reef communities to increasingly severe climate-driven warming events.",
"conclusion": "5. Conclusion It has been suggested that reefs at greater depths may escape the effects of climate-induced bleaching events [ 18 , 19 , 161 ]. By surveying shallow reefs, we show the loss of hard and soft coral cover post-heatwaves between 15 and 25 m. While less than the observed changes at shallower 5–15 m depths, our results highlight that benthic communities down to 25 m depth can be impacted by severe heatwaves. This is supported by an increasing number of studies that show reefs at greater depths, including mesophotic reefs (30–150 m) are not immune to thermal stress with widespread evidence of bleaching impacts and changes in community composition [ 9 , 20 , 22 , 24 , 26 , 30 , 104 ]. The decreasing change in cover of benthic groups with depth highlights the importance of surveying multiple depth gradients when explaining the rate at which communities can be altered after large-scale disturbance events. The effects of initial and repeated thermal stress across depths show that even remote reefs that are protected from direct anthropogenic impacts are not resistant to the impacts of elevated seawater temperatures and subsequent changes in community dynamics and zonation patterns. Given the predicted increase in the frequency and severity of climate-driven marine heatwaves, our findings support concerns about the ability of contemporary benthic communities across the depth range of shallow coral reefs (<30 m) worldwide to resist and recover from recurrent thermal stress.",
"introduction": "1 . \n Introduction Climate-induced thermal stress is one of the main drivers of change in coral reef benthic communities [ 1 – 3 ]. However, the degree to which disturbance events affect ecological communities is not uniform across space and varies along environmental gradients [ 4 – 6 ]. Coral reefs show significant spatial and temporal heterogeneity in community structure post-disturbance [ 7 – 9 ]. While most assessments of the extent of coral bleaching and consequent degradation and succession processes are carried out in shallow depths (2–10 m) [ 7 , 10 – 13 ], the response of shallow benthic communities to thermal stress over depth ranges extending to 30 m is poorly described [ 9 , 14 ]. As light, temperature [ 15 ] and waves decrease with increasing depth, and the availability of organic resources increases at depths [ 16 , 17 ], it is expected that the response of benthic communities to thermal stress will dampen [ 18 ]. However, contrary to the deep reef refuge hypothesis, which suggests that reefs at greater depths could escape the effects of climate-induced bleaching events [ 18 , 19 ], increasing evidence of bleaching and mortality within shallow (2–27 m) and mesophotic depth zones (≥ 30 m) [ 8 , 14 , 20 – 27 ] underlines the necessity to assess communities at larger depth gradients. Additionally, the limited overlap of species between shallow and deep reefs [ 28 , 29 ] and significant genetic divergence between species across depths indicate that deeper populations may not always provide viable propagules to repopulate shallow reefs [ 30 , 31 ]. These observations have important implications for coral reef assessments that evaluate the ecological response of shallow reef organisms to increasingly frequent thermal stress events, which are predicted to occur annually within the twenty-first century [ 32 ]. In 2014–2017, the world observed the first back-to-back coral bleaching events that were documented over 3 years [ 33 – 35 ]. The recurring heatwaves caused coral bleaching on 80% of coral reefs globally and the mass mortality of corals [ 1 , 11 , 35 , 36 ]. Thermal anomalies such as the 2014–2017 warming events have raised global concern about the resilience and persistence of tropical reef systems [ 1 , 35 ]. Chronic disturbances can impede recovery and induce significant degradation in communities [ 37 ]. As the increased intensity and frequency of ocean-warming events [ 38 – 40 ] shorten the recovery window for benthic communities [ 41 – 43 ], they can lead to potentially irreversible ecological changes [ 44 , 45 ]. Large-scale bleaching-induced coral mortality on coral reefs alters the dynamic interactions between benthic groups, causing significant shifts in community assemblages [ 46 , 47 ]. While the loss of hard corals can cause bare reef pavement to dominate the substrate [ 48 , 49 ], ecological regime shifts on coral reefs can generate non-coral-dominated states primarily characterized by benthic taxa such as algae (including crustose coralline algae (CCA), turf and macroalgae), soft corals and sponges [ 47 , 50 – 54 ]. Opportunistic and faster growing in nature, these alternative benthic groups competitively replace hard corals, which can lead to a loss of complexity [ 55 , 56 ], reduce reef carbonate budgets [ 57 , 58 ] and alter reef-associated fish communities [ 59 , 60 ], resulting in a decline in provision of important ecosystem services [ 61 , 62 ]. In some instances, coral community composition is altered despite recovering to pre-disturbance coral abundance [ 7 , 63 , 64 ]. While climate-driven community changes may result in degraded and less resilient reefs [ 23 , 65 ], the depth variability in changes in benthic community assemblages is poorly understood. Coral reef benthic communities are highly heterogeneous [ 66 , 67 ] and naturally show diverse vertical zonation patterns [ 28 , 68 , 69 ] as a result of the interaction between biophysical processes that concomitantly vary across depths [ 17 , 70 ]. During a marine heatwave, modifications in physiologically important factors such as the combination of sustained elevated seawater temperatures and high solar radiation are often the main causes of bleaching and degradation on reefs [ 18 , 71 , 72 ]. Additionally, the proportion of change in benthic communities following a marine heatwave may be due to the assemblage composition [ 11 ]. Studies show that coral species have different thermal tolerance [ 73 – 75 ], and the susceptibility of species within the same genus to climate-induced bleaching can vary significantly across depths [ 14 ]. In addition, imports of nutrient-rich and cold water from deep-water upwelling and internal waves onto shallow reef systems can confer resilience to change in ecological communities during ocean-warming events [ 76 , 77 ]. High spatial variation is therefore expected to occur in the response of benthic lifeforms to thermal stress within the water column [ 14 , 78 , 79 ]. Where post-bleaching coral reef assessments incorporate multiple depth zones, they can estimate depth-dependent mortality, identify surviving populations with the potential to repopulate and provide a better understanding of the trajectory of community changes across the extent of shallow reef systems (0–30 m) [ 14 , 25 , 27 , 78 , 80 ]. Here, we use benthic community surveys from shallow forereefs of the remote Chagos Archipelago, Central Indian Ocean, before and after successive marine heatwaves in 2015–2017. We examine the effects of recurrent severe thermal stress on benthic communities across a depth gradient of 5–25 m. In 2015–2017, the marine heatwave caused severe coral bleaching and mortality in the Chagos Archipelago, with sustained declines in coral cover due to two consecutive years of elevated temperature stress [ 41 ]. Specifically, we assess the change in percentage cover of five broad benthic groups and examine spatial variation in the change in composition of those groups across depths at four atolls in the archipelago. We also examine differences between the initial and repeated effects of thermal stress on the observed changes in benthic composition.",
"discussion": "4. Discussion It is evident that intense ocean-warming events such as the successive heatwaves in 2014–2017 [ 36 , 115 ] are changing coral-dominated reefs to alternative configurations, with severe impacts on ecosystem functioning [ 11 ]. However, there is limited understanding of the changes that occur in benthic communities from recurring warming events at depths greater than 10 m [ 1 , 60 , 101 , 116 ]. We observed changes in benthic communities on the remote Central Indian Ocean reefs of the Chagos Archipelago following successive heatwaves in 2015–2017 at depths down to 25 m. Our findings indicate greater changes in benthic group cover at shallower depths (5–15 m) relative to deeper zones (15–25 m). However, the changes in benthic communities in response to elevated seawater temperatures were variable at different depths. 4.1. Variation in ecological response to thermal stress across depth Changes in communities occur as a result of the complex interactions between their habitat, physiological traits and the nature of disturbance events [ 117 ]. Here, we show that the change in benthic communities decreased with increasing depth. This pattern [ 14 , 78 ] may relate to biophysical forcing, such as light, temperature, salinity [ 15 , 118 , 119 ] and wave and current regimes [ 16 , 120 ] that simultaneously vary across depth. During a global warming event, compositional changes observed in benthic communities occur as a result of thermal stress-related coral bleaching, which is caused by increased seawater temperature and light irradiance. As light intensity and temperature naturally decline with increasing depth [ 15 ], lower levels of bleaching and mortality are frequently observed at depth [ 8 , 14 , 24 , 25 , 27 ], which could account for the lower change in soft and hard coral cover at 15–25 m compared with the shallower communities at 5–15 m. In addition to the combination of the natural attenuation of light and temperature with increasing depth, the presence of internal waves may explain the smaller change in benthic community cover at deeper zones relative to shallower parts of the reefs. Significant water temperature fluctuations with increasing depth suggest strong internal wave activity in the Chagos Archipelago [ 121 ]. Internal waves which drive deep-water upwelling can significantly decrease thermal stress and mitigate the response of benthic organisms to bleaching events at depth [ 122 ]. By cooling temperatures and bringing in allochthonous nutrients from the deep, upwelling can increase productivity in shallow reef systems [ 120 ]. When autotrophy is compromised during a warming event [ 123 ], upwelling can promote heterotrophy and the survival of mixotrophic organisms such as hard and soft corals [ 17 , 77 , 124 , 125 ]. Despite the variation in the change of benthic groups across depths, we observe a consistent pattern of loss in hard coral and soft coral cover, followed by an increase in CCA, reef pavement and sponge cover across all depth zones. The increase in sponge and CCA cover can also be associated with the increase in the proportion of reef pavement after the marine heatwaves. Mass bleaching events have been shown to induce widespread mortality in hard and soft coral communities [ 40 , 126 ] creating vacant space [ 127 ]. Recent studies in the Chagos Archipelago have shown a higher proportion of reef pavement and boring sponge cover on shallow exposed sites after the 2015–2017 marine heatwaves compared with more sheltered sites [ 48 ]. Higher wave energy at shallow depths can increase the susceptibility of benthic groups such as dead hard and soft corals to physical damage and dislodgement [ 128 ], resulting in vacant bare reef pavement and promoting the growth of rapidly colonizing and wave-tolerant organisms like sponge and CCA [ 129 , 130 ]. 4.2. Variation in ecological response to initial and repeated thermal stress 4.2.1. Change in hard coral cover following bleaching events There were variable and modest responses in benthic communities to initial and repeated thermal stress across depth zones. Recent studies show similar trends of little to no significant interaction between depth and thermal stress in explaining post-heatwave trajectories in benthic communities across depth gradients [ 8 , 26 , 131 ]. Among all surveyed benthic groups, only hard coral cover had a consistent negative effect of initial thermal stress across all depth zones. This suggests that the initial effect of thermal stress in 2015 had the largest influence on the observed decline in hard coral cover, potentially due to mortality of coral assemblages with low tolerance to thermal stress [ 101 , 117 ]. In contrast, the cumulative effect of repeated thermal stress from 2015 to 2017 was associated with a smaller decline in hard coral cover. Previous studies in Chagos support these findings with comparable patterns reporting higher coral bleaching and mortality in 2015 despite higher exposure to thermal stress in 2016 [ 41 ]. Similar trends of lower bleaching and mortality rates in the second year of consecutive coral bleaching events were reported on the Great Barrier Reef [ 116 ] and in the Coral Sea [ 101 ], even though higher thermal stress was recorded in the second year. There are several mechanisms which could drive successive bleaching events resulting in lower coral mortality. Over the duration of a warming event, less susceptible hard corals can resist, adapt and recover from thermal stress [ 132 , 133 ]. For example, exposure to the initial warming event in 2015 may have created thermal preconditioning for the remaining live hard coral community to develop resistance to subsequent bleaching [ 134 , 135 ]. In addition, natural association with [ 136 , 137 ] or shifting to thermally tolerant endosymbionts during stress [ 138 , 139 ] can decrease bleaching susceptibility and lower coral cover decline during subsequent warming events. 4.2.2. Change in other benthic groups following bleaching events Unlike the hard coral community, the remaining benthic groups showed weak and variable responses to both initial and repeated thermal stress across depth zones. For example, repeated thermal stress increased soft coral cover loss at 20–25 m, whereas no effects were observed on change in soft coral cover at shallower depths. Studies which have looked at the impact of recurring thermal stress on soft coral communities in the Western and Central Pacific also show similar trends of loss related to repeated bleaching and eventual die-offs [ 88 , 126 ]. The impact of thermal stress at 20–25 m may relate to lower thermal variability at depths making deeper soft coral communities less resilient to chronic warming [ 88 ]. In addition, soft coral communities at depth could be made up of more vulnerable populations like Sarcophyton colonies, which are known to degrade completely after recurring bleaching events, compared with Lobophytum and Sinularia, which are more resistant to bleaching [ 88 , 126 ]. Both initial and repeated thermal stress decreased the recovery in sponge cover following the 2015–2017 bleaching event. This effect was most pronounced at the shallowest depths (5–10 m). The increase in sponge cover that we found is probably driven by an increase in encrusting boring sponges [ 90 , 140 ], also observed following the major 1998 bleaching event in the Chagos Archipelago [ 141 ]. These bio-eroding sponges can host Symbiodinium spp., which are resistant to bleaching [ 142 ]. This association with thermally tolerant holobionts helps the sponges spread rapidly on stressed and dead corals during warming events [ 143 ]. CCA was the benthic group least influenced by initial and repeated thermal stress. Repeated thermal stress marginally reduced the recovery of CCA cover at one atoll (Peros Banhos), and minimal to no effect of initial thermal stress was observed on the increase in CCA cover. Recent studies show that CCA has high thresholds for elevated water temperatures [ 89 ]. While an acute thermal stress event can significantly reduce photosynthetic rates in CCA, it also has the potential to acclimatize to chronically elevated water temperatures and maintain photosynthesis and calcification [ 144 ]. In addition to being highly tolerant to heat stress, the increase in CCA cover may relate to their high dispersal rates and ability to spread rapidly on various substrates such as available bare reef substrate, dead coral colonies and coral rubble post-bleaching [ 145 ]. There were variable effects of both initial and repeated thermal stress on the increase in available reef pavement. This is probably linked to the loss of previously dominant hard and soft coral cover following the 2015–2017 marine heatwaves on these reefs. The overall gain in available reef pavement also coincides with an increase in fish community herbivory about 2 years following the warming events [ 146 ]. By sustaining high levels of grazing on endolithic and epilithic algae following bleaching, herbivores can maintain a high proportion of bare reef pavement [ 146 , 147 ]. 4.3. Benthic community reorganization following thermal stress The reorganization of coral reef benthic communities following a disturbance event depends on the dynamic interaction between the nature (scale and severity) of the disturbance event and the reef community composition [ 148 ]. Marine heatwaves have been shown to cause widespread bleaching and mortality in scleractinian corals [ 1 , 123 ]. This subsequently modifies the network of interactions between the wider benthic community and can cause coral-dominated reefs to shift to reefs with novel alternative configurations [ 1 , 33 , 40 , 100 , 149 ]. During the 2015–2017 thermal stress events, reefs within the Indian Ocean region that experienced high to extreme levels of bleaching [ 149 – 151 ] altered to configurations dominated by epilithic algal matrix (e.g. Seychelles, Aldabra atoll [ 140 ]; Kenya, Zanzibar [ 13 ]) and rubble beds (e.g. Maldives [ 10 ]). Despite high bleaching and mortality rates during the successive marine heatwaves in the Chagos Archipelago [ 41 ], there is no evidence suggesting that changes in benthic groups are indicative of a regime shift from a hard coral-dominated community (electronic supplementary material, figures S8 and S9). This potential resilience to regime shifts following bleaching in the Chagos Archipelago may be attributed to (i) the isolation of the reefs from direct anthropogenic activities (e.g. pollution, eutrophication, sedimentation) [ 101 ]; (ii) local upwelling [ 121 ], which may reduce thermal stress and deliver nutrients required by benthic communities resulting in a less severe change in community structure following the warming events [ 77 , 152 ]; and (iii) high fish biomass [ 153 , 154 ] with strong top-down control on algae (macroalgae cover less than 7%) maintained by a high herbivore density post-bleaching in the archipelago [ 146 ], which may reduce the successional dominance of algae on the reefs post-disturbance and which will also provide vacant space for coral recruitment and recovery [ 155 ]. The Chagos Archipelago also has one of the most diverse coral communities in the Indian Ocean region, which has been associated with resilience and recovery following previous severe bleaching events in the archipelago [ 156 ] as well as the Great Barrier Reef, Moorea and Jamaica [ 65 ]. Benthic community recovery is usually assessed as the return of cover to pre-disturbance levels as well as the reassembly to similar pre-disturbance taxa relative abundances [ 14 , 148 , 157 ]. Recovery to pre-disturbance configurations can take 8–13 years [ 148 , 157 ]. However, as the return time between climate-induced warming events becomes shorter, the ability of coral reefs to recover to pre-bleaching levels may become compromised [ 1 , 40 , 43 ]. Despite significant bleaching and mortality of key coral species such as Acropora , Pachyseris , Echinopora , Isopora and Galaxea during the back-to-back bleaching events in the Chagos Archipelago [ 21 , 41 , 158 ], our data show a lack of a shift in dominance 1–2 years post-bleaching. Additionally, Lange et al . [ 159 ] recently indicated high recovery potential with new coral recruitment across all morphotypes (branching, massive, encrusting, tabular Acropora , branching Acropora ). However, the recovery trends shown by Lange et al . [ 159 ] in the Chagos Archipelago only describe reefs at 8–10 m. This and several post-bleaching coral reef studies similarly focus on changes occurring shallower than 10 m [ 1 , 10 , 40 , 101 , 159 , 160 ], limiting our understanding of how deeper reefs respond to recurring marine heatwaves."
} | 5,436 |
39899715 | PMC11831133 | pmc | 9,652 | {
"abstract": "Significance This paper describes co-zorbs, which are spherical aggregates of bacteria that move and transport other bacteria. Flavobacterium johnsoniae cells aggregate and collect diverse bacterial species in a manner reminiscent of phagocytes, creating an organized core–shell structure we refer to as co-zorbs. The discovery of co-zorbs introduces a unique type of bacterial movement and transport involving cooperation among bacterial species. Co-zorbs have the potential for engineering microbial systems for biotechnology applications and for managing the spread of bacterial pathogens in their hosts.",
"discussion": "Discussion In this study, we introduce a bacterial behavior, designated co-zorbing, in which F. johnsoniae zorbs, or motile biofilms, encapsulate diverse bacterial species to enable collective transport in a spatially organized structure. This process illustrates cooperative motility of a biofilm structure, interspecies interactions, and spatial organization, providing insights into the complexity of interspecies interactions among bacteria. The ability of F. johnsoniae to form co-zorbs with both Gram-positive and Gram-negative bacteria, despite their diverse characteristics, and to form tri-zorbs when coinoculated with two companion species underscores the versatility and variety of zorb interactions. The formation of co-zorbs inside zebrafish indicates their potential for mediating biological interactions with a host. Future studies might dissect further the biophysical requirements enabling co-zorb formation in vivo (i.e., adhesion substrates). Collective behavior is observed throughout the biological world. Birds flock, fish school, and ants form colonies, and at the microscopic level, animal cells coordinate movement during morphogenesis, tissue remodeling, and cancer progression ( 21 , 22 ). In prokaryotes, collective behavior is often associated with surfaces and involves diverse forms of motility, including swarming, twitching, gliding, and swimming ( 23 ). F. johnsoniae is a gliding bacterium that uses motility adhesins secreted by T9SS to move across surfaces ( 24 ). F. johnsoniae exhibits collective motility by forming vortex patterns under low-nutrient conditions, and deletion of the gene encoding the motility adhesin, SprB, impairs gliding motility, disrupting these patterns ( 25 ). Other members of the phylum Bacteroidota, such as C. gingivalis , also use gliding motility to swarm in counterclockwise vortex patterns and exhibit cargo transport of nonmotile species within the oral microbiome ( 7 ). Similarly, members of the phylum Myxococcota coordinate swarming for predation and aggregate into multicellular fruiting bodies under nutrient-deprived conditions, a process mediated by Type IV pilus machinery or focal adhesion complexes ( 26 , 27 ). These examples highlight the importance of motility and cell–surface adhesins in facilitating collective behaviors under nutrient-limited conditions. Zorbs and co-zorbs add to the pantheon of group strategies that bacteria employ to navigate their environments. The inability of F. johnsoniae to transport secondary species in the absence of SprA further supports the past findings indicating the role of gliding motility, with motility adhesins likely contributing to the movement and spatial organization of co-zorbs and tri-zorbs. Additionally, we speculate that the adhesion affinity of F. johnsoniae may differ among bacterial species ( E. coli , S. aureus, and P. aeruginosa ) resulting in the unique spatial organization observed in tri-zorbs. Motility and biofilm formation are typically used by bacteria to adapt to very different environmental conditions, and consequently, there are few examples of motile biofilms, such as zorbs ( 28 – 30 ). We demonstrate the movement of single-species and multispecies biofilms, as well as the bulk localization and transport of diverse species by F. johnsoniae . Several collective behaviors of F. johnsoniae are inhibited by high glucose levels ( 25 , 31 – 33 ), including zorb formation ( SI Appendix, Fig. S5 ), suggesting that such collective motility may be a strategic response to nutrient deprivation. For example, colony spreading in F. johnsoniae is inhibited by N-acetylglucosamine, a structural component of the bacterial cell wall peptidoglycan ( 25 ), which serves as a carbon source for F. johnsoniae under low nutrient conditions ( 34 ). Perhaps F. johnsoniae, well-known for its ability to degrade complex polymers ( 35 , 36 ), uses co-zorbing as a nutrient-acquisition strategy, feeding on cell components of the encapsulated species while providing it with rapid, safe transit across large distances inside the zorb. Conversely, it remains possible that co-zorb formation could be nonadaptive in nature and that biophysical forces may contribute to its formation and spatial organization. Thus, future work will focus on dissecting various genetic, biochemical and biophysical factors (e.g., cell–cell adhesion or cell–substrate adhesion) that drive co-zorb formation both in vitro and in vivo and will explore the potential impact of co-zorbs on bacterial spread and colonization in animals, on plant surfaces, and in other environments. Although the benefits of forming co-zorbs remain to be determined, the genetic tractability of F. johnsoniae makes it a candidate for engineering for practical use. It might serve as a tool to encapsulate other bacteria, including S. aureus -MRSA, and act as a scavenger in a manner analogous to that of phagocytes. Co-zorbing therefore presents a unique means of bacterial translocation, extends our knowledge of microbial community behaviors, and opens innovative avenues for biotechnological and medical applications."
} | 1,441 |
39257784 | PMC11383685 | pmc | 9,653 | {
"abstract": "Biofilms are three-dimensional structures containing one or more bacterial species embedded in extracellular polymeric substances. Although most biofilms are stationary, Flavobacterium johnsoniae forms a motile spherical biofilm called a zorb, which is propelled by its base cells and contains a polysaccharide core. Here, we report formation of spatially organized, motile, multispecies biofilms, designated “co-zorbs,” that are distinguished by a core-shell structure. F. johnsoniae forms zorbs whose cells collect other bacterial species and transport them to the zorb core, forming a co-zorb. Live imaging revealed that co-zorbs also form in zebrafish, thereby demonstrating a new type of bacterial movement in vivo. This discovery opens new avenues for understanding community behaviors, the role of biofilms in bulk bacterial transport, and collective strategies for microbial success in various environments.",
"introduction": "Introduction Like a bustling city where a variety of people live together, interact with one another, and shape the environment, diverse microorganisms converge to create intricate structures called biofilms. Within these multispecies biofilms, microbes embedded in an extracellular matrix interact, influencing the spatial architecture and functionality of the community ( 1 ). These interactions give rise to emergent properties that are absent in single-species biofilms, profoundly impacting the overall behavior and characteristics of the biofilm ( 2 ). The significance of multispecies biofilms across applications in health care, agriculture, and environmental management highlights the need to understand the complex interspecies interactions in biofilms, which is crucial to address global challenges as diverse as food poisoning, biofouling in industrial infrastructure, and antibiotic resistance in chronic infections ( 3 ). Biofilms in the environment are typically stationary structures. We recently discovered that Flavobacterium johnsoniae , a Gram-negative bacterium commonly found in soil and freshwater, uses cell features involved in gliding motility and colonization to form spherical biofilm-like microcolonies called zorbs ( 4 , 5 ). These zorbs are unique as they are propelled by their base cells without pili or flagella and contain an extracellular polysaccharide core. There are only a few examples in which motility is observed in interspecies biofilms. For example, Pseudomonas aeruginosa shifts from collective to single-cell motility, initiating exploratory movements upon sensing Staphylococcus aureus biofilms ( 6 ), Capnocytophaga gingivalis transports non-motile bacteria as cargo along its cell length, shaping the spatial organization of polymicrobial communities ( 7 ), and Candida albicans and Streptococcus mutans demonstrate a “forward-leaping motion” in interkingdom biofilms ( 8 ). These behaviors depend on either single-cell motility, hitchhiking, or interspecies expansion, in which individual cells carrying non-motile bacteria move, grow, and spread across surfaces to form interspecies or multispecies biofilms. In contrast, our study presents a new mechanism of bacterial movement in which motile F. johnsoniae biofilms transport both motile and non-motile bacterial species in a core-shell structure. In this unique collective movement, a group of F. johnsoniae cells surround and collect other bacterial species and transports them as a multispecies biofilm structure across surfaces, including inside zebrafish larvae. Co-zorbing challenges the notion that multispecies biofilms are stationary and suggests a fertile area for future research in community ecology and infectious disease.",
"discussion": "Discussion In this study, we introduce a new bacterial behavior, designated co-zorbing, in which F. johnsoniae zorbs, or motile biofilms, encapsulate diverse bacterial species to enable collective transport in a spatially organized structure. This process illustrates cooperative motility of a biofilm structure, interspecies interactions, and spatial organization, providing new insights into the complexity of interspecies interactions among bacteria. The ability of F. johnsoniae to form co-zorbs with both Gram-positive and Gram-negative bacteria, despite their diverse characteristics, and to form tri-zorbs when co-inoculated with two companion species underscores the versatility and variety of zorb interactions. The formation of co-zorbs inside zebrafish indicates their potential for mediating biological interactions with a host. Collective behavior is observed throughout the biological world. Birds flock, fish school, and ants form colonies, and at the microscopic level, animal cells coordinate movement during morphogenesis, tissue remodeling, and cancer progression ( 18 , 19 ). In prokaryotes, collective behavior is often associated with surfaces and involves diverse forms of motility, including swarming, twitching, gliding, and swimming ( 20 ). F. johnsoniae is a gliding bacterium that uses the Type IX Secretion System (T9SS) and gliding motility apparatus to move across surfaces, exhibiting collective motility in the formation of vortex patterns under low-nutrient conditions ( 21 , 22 ). Other members of the phylum Bacteroidota, such as C. gingivalis , swarm in counterclockwise vortex patterns and exhibit cargo transport of non-motile species within the oral microbiome ( 7 ). Members of the phylum Myxococcota coordinate swarming for predation and aggregate into multicellular fruiting bodies under nutrient-deprived conditions ( 23 ). Zorbs and co-zorbs add to the pantheon of group strategies that bacteria employ to navigate their environments. Motility and biofilm formation are typically used by bacteria to adapt to very different environmental conditions, and consequently, there are few examples of motile biofilms, such as zorbs ( 24 – 26 ). We demonstrate the movement of single-species and multi-species biofilms, as well as the bulk localization and transport of diverse species by F. johnsoniae . Several collective behaviors of F. johnsoniae are inhibited by high glucose levels ( 22 , 27 – 29 ), including zorb formation ( Fig. S4 ), suggesting that such collective motility may be a strategic response to nutrient deprivation. For example, colony spreading in F. johnsoniae is inhibited by N-acetylglucosamine, a structural component of the bacterial cell wall peptidoglycan ( 22 ), which serves as a carbon source for F. johnsoniae under low nutrient conditions ( 30 ). Perhaps F. johnsoniae , well-known for its ability to degrade complex polymers ( 31 , 32 ), uses co-zorbing as a nutrient-acquisition strategy, feeding on cell components of the encapsulated species while providing it with rapid, safe transit across large distances inside the zorb. The benefits of forming co-zorbs remain to be determined. In particular, the impact of co-zorbs on bacterial spread and colonization in animals, on plant surfaces, or in other environments is a fertile area for future investigation. The genetic tractability of F. johnsoniae makes it a candidate for engineering for practical use. It might serve as a tool to encapsulate other bacteria, including S. aureus -MRSA, and act as a scavenger in a manner analogous to that of phagocytes. Co-zorbing therefore presents a new means of bacterial translocation, extends our knowledge of microbial community behaviors, and opens innovative avenues for biotechnological and medical applications."
} | 1,868 |
32045053 | PMC7116209 | pmc | 9,655 | {
"abstract": "In 2013, the “biofabrication window” was introduced to reflect the processing challenge for the fields of biofabrication and bioprinting. At that time, the lack of printable materials that could serve as cell-laden bioinks, as well as the limitations of printing and assembly methods, presented a major constraint. However, recent developments have now resulted in the availability of a plethora of bioinks, new printing approaches, and the technological advancement of established techniques. Nevertheless, it remains largely unknown which materials and technical parameters are essential for the fabrication of intrinsically hierarchical cell–material constructs that truly mimic biologically functional tissue. In order to achieve this, it is urged that the field now shift its focus from materials and technologies toward the biological development of the resulting constructs. Therefore, herein, the recent material and technological advances since the introduction of the biofabrication window are briefly summarized, i.e., approaches how to generate shape, to then focus the discussion on how to acquire the biological function within this context. In particular, a vision of how biological function can evolve from the possibility to determine shape is outlined.",
"conclusion": "4 Concluding Remarks and Future Perspectives The recent progress in hydrogel design together with the development of new bioprinting strategies, have introduced effective solutions to extend the biofabrication window, reducing the need to compromise on the use of materials that display satisfactory structural properties, but provide a nonoptimal environment for cells to thrive. Despite this remarkable progress, we are only beginning to tap into the potential of biofabrication in aiding the reconstruction of fully functional living engineered tissues. Importantly, while our ability to precisely mimic architectural facets of living tissues increases, the extent of resolution that is required in order to achieve fully functional biofabricated constructs, remains unknown. Salient advances in the creation of tissues that exerted native functions in vivo have been achieved with precisely patterned, yet relatively simple architectures. [ 18 , 28 , 182 ] At the same time, an increasing number of studies is emerging, suggesting improved maturation of biofabricated tissues mimicking native structures, particularly when multiple cell types are distributed in precise areas of the produced constructs. [ 53 , 233 , 281 ] Bioprinting itself may help answer such fundamental yet elusive questions. High-accuracy and multimaterial printing can permit the generation of models with increasing degree of complexity, in which different combinations of geometrical cues, physicochemical properties and relative positioning of cells and materials can be combined. In order to achieve this, the influence of each variable should be isolated and studied, possibly even converging the field of biofabrication with automated high-throughput analysis, materiomics, and artificial intelligence-driven approaches that are already being introduced for biomaterial research. [ 295 – 297 ] Such systematic research will provide key insights for the design of the next generation of bioprinted constructs, eventually clearing an important step toward clinical translation. Additionally, although most of the expectations of bioprinting are associated with the generation of constructs that can copy or simulate the function of human tissues, alternative nonphysiological elements derived from other disciplines in biology, physics and engineering could also be envisioned and introduced, as recently suggested with the bioprinting of stimuli-responsive materials, nonmammalian cells as sources of metabolites or constructs designed by deterministic chaos principles as platforms to study cell–cell interactions. [ 298 – 301 ] Next generation bioinks, in addition to design criteria centered on rheology and printability, will need to be inspired by substantial input from advances in cell biology and biotechnology. Fundamental lessons learned from embryonic development, mechanobiology, cell differentiation and repair in species with regenerative capacity superior to that of humans, [ 302 , 303 ] will be paramount to guide bioink development, as well as to instruct which architectures and cell patterns to print to boost maturation. Given the dynamic and multifaceted events that determine development and progress of living tissues, other important criteria will be endowing bioprinted structures with the ability to evolve over time and provide different stimuli to the printed cells, for instance via the incorporation of materials in which biological signals and growth factors can be patterned and released on demand. [ 304 ] Together with knowledge from developmental biology, the emerging role of organoids as self-developing, miniature functional units of living tissues, and in general of microtissue analogues created by bottom-up assembly of stem cells can provide new opportunities for biofabrication. One possible avenue is to use preformed organoids and embryoid bodies that can be led to self-assemble and thus to produce tissues with high cell content by jamming them into a mold. These dense cellular structures can then be used as suspended bath to print vasculature within such engineered constructs to ensure organoid viability. [ 305 ] Furthermore, hybrid printing strategies can be envisioned, in which part of the construct architecture is imposed by the printing process, and at the same time also rely on the ability of stem cells to self-organize into polarized and heterocellular constructs. This can help to recreate microscale functional groups of cells that can otherwise not easily be resolved with current printing techniques. Certainly, progress in basic biology is necessary and should be given priority in the near future to fulfill the goal of creating transplantable tissues. However, there is also room and need for further technological developments in additive manufacturing approaches. Interestingly, alongside extrusion-based bioprinting, several new and already existing biofabrication techniques are starting to gain more relevance in the field. In the quest to capture tissue complexity, future work can benefit from converging bioprinting with technologies that exhibit such complementary advantages. For instance, methods to manipulate cell suspensions (and even single cells), such as inkjet printing, as well as technologies that resolve features at the nano- and micro-scale, like two photon polymerization and MEW, permit high resolution, [ 306 , 307 ] but struggle to create large constructs with clinically relevant sizes. The combination with extrusion printing can, however, aid the generation of larger engineered tissues with features spanning across different dimensional scales to mimic the hierarchal composition of native tissues or to introduce regions with unique cell patterns. Likewise, tissues generated through the bioassembly of tissue spheroids, which are typically soft and prone to deform due to cell generated forces, can benefit from extrusion printing technologies to provide mechanically reinforcing frameworks, [ 293 ] that could possibly facilitate surgical handling and implantation. Additionally, extrusion technologies are limited in the geometries that can be achieved by the use of printed fibers as building blocks and by the need of supports. Recently, lightbased biofabrication technologies, such as stereolithography and digital light projection-based bioprinting have gathered attention for their ability to print convoluted porous geometries, typical of native tissues, and to replicate functional vasculature and microchannels within bioprinted constructs, [ 6 , 308 ] a step often indicated as major bottleneck toward clinical translation. Therefore, more research related to bioinks for light-based printing, as well as the integration with other biofabrication approaches can also be envisioned toward the generation of complex, functional tissues. Moreover, in the perspective of the generation of functional and transplantable tissue and organs, scaling up and regulatory concerns regarding bioprinted cells remain important topics to be tackled in the near future. Likewise, bioprinted constructs used as in vitro models and drug testing platforms will also need to undergo proper validation and clarification from regulatory bodies will be required to determine the exact criteria these models have to meet in order to be used as a complement to, or more desirably, as replacement for animal experimentation. In view of a shifting focus on the biology of printing, detailed investigations of the impact of printing processes on cells, beyond the assessment of viability alone, will be required, especially for the creation of large grafts that require long printing times. During long printing processes cells are exposed to nonoptimal conditions (i.e., shear stresses and depletion of nutrients). Approaches like printing in support baths could permit supplementation of nutrients in the support material, possibly alleviating the impact of long fabrication times. Further research, can integrate novel technological solutions for the fast creation of living constructs. The recent introduction of volumetric bioprinting (VBP) based on visible light optical tomography, led to the creation of centimeter scale anatomically shaped constructs in less than 30 s, [ 309 ] opening new potential avenues for printing with minimal stress on embedded cells. In summary, adequate bioactivity and biofunctionality of printed cells and biofabricated tissues have to be addressed, as postprinting maturation is necessary for the generation of a functional tissue analogue. Accurate bioink design, as well as utilization of advanced cell culture platforms including cocultures, organoids and bioactive cues during the biofabrication process, are fundamental to achieve this objective. This requires the thorough characterization of cell behavior in the long term to ensure the preservation of superior cell functionalities beyond just viability. Biofabrication strategies are reaching a new cornerstone, maturing toward creating tissue grafts with envisioned applications in pharmaceutical industries and clinical therapies. Bioink design should reflect this yearning by investigating beyond the simple proof of concept to show feasibility of printing. Cell functionality is necessary both for in vitro models and to move forward toward the demonstration of applicability of bioprinted constructs as biomedical devices that can eventually be used as clinical solution to repair damaged tissues, and bring researchers a step closer toward the ambitious goal of organ bioprinting.",
"introduction": "1 Introduction The functionality of living tissues is intimately linked to their intricate and highly specialized architecture. Tissues and organs are composed of multiple types of cells and extracellular matrix (ECM) components and, with few exceptions, are infiltrated with vascular and neural networks. The hierarchical spatial arrangement of these elements is paramount to how they interact with each other and, thus, closely orchestrates several processes during embryonic development, [ 1 ] in healthy tissue homeostasis, as well as during tissue regeneration. [ 2 ] Strategies to generate cell–material constructs that ultimately yield a healthy and mature functional tissue remains a major challenge in the field of regenerative medicine. With the introduction of additive manufacturing, technologies became available to design and fabricate 3D material scaffolds with unprecedented shape and precision. Although, many of these technologies are potentially harmful for active biological components, including living cells, biological matter can successfully be used as building blocks for the generation of living 3D objects. This constitutes the research field of biofabrication, [ 3 ] a revolutionizing toolkit for regenerative medicine that allows cells, biomaterials and bioactive moieties to be precisely combined and patterned into 3D constructs through automated, cell-friendly fabrication methods, such as bioprinting and bioassembly. [ 4 ] Indeed, modified extrusion, [ 5 ] (stereo)lithographic, [ 6 ] inkjet, [ 7 ] and laser printing methods [ 8 ] are now available for the processing of living cells and can be applied to recreate anatomical parts using medical images as blueprints. [ 9 ] Within the past decade there has been a particular focus on techniques and materials compatible with extrusion-based printing due to the high versatility in printing multiple materials, the relative low cost and easy access to the required hardware for this technology. [ 10 ] Taken together, biofabrication has gained significant momentum and provides a powerful approach to tackle major hurdles in the generation of engineered living tissues. As building materials in biofabrication processes, two different types of printable “inks” can generally be distinguished. [ 11 ] Firstly, materials that are used to print acellular structures, on which cells are seeded or that can also be used as surgical tools or implants after fabrication, are termed biomaterial inks. Many different materials, including thermoplasts and metal powders, can be processed using a variety of technologies and the process parameters are only restricted by technology and the respective material. Secondly, the printable formulations that contain living cells are termed bioinks. [ 3 ] This simple but clear distinction severely limits the number of suitable fabrication technologies as the process must ensure viability of the embedded cells. Therefore, the printing process needs to be performed under physiological conditions and in an aseptic environment. Bioinks are generally aqueous formulations with adjusted rheology that can provide a highly hydrated environment for the encapsulated cells. [ 11 ] For this, often formulations of hydrogel precursors are applied that can be crosslinked postfabrication. An important challenge in this context is, however, the counterdirectional effect of polymer concentration on shape fidelity and cytocompatibility. In other words, low polymer content in the bioink results in soft, loosely crosslinked hydrogels after printing that are beneficial to support cell survival, migration and bioactivity, but such formulations do typically result in structures with poor shape fidelity. In contrast, bioinks with higher polymer content and thus higher viscosity can more rapidly be crosslinked into stiffer gels after printing and are in general preferred for the accurate 3D placement and postprinting shape retention, [ 12 ] resulting in a printed object that faithfully reproduces its original computer design. However, this comes with the cost of reduced ability of the cells to spread, migrate and colonize the hydrogel matrix with newly synthesized ECM. Additionally, high crosslinking density or polymer content can also hinder the ability of the embedded cells to remodel the hydrogel matrix over time, a process necessary for tissue maturation. The simultaneous need for these opposing requirements led to the conceptualization of the biofabrication window, the range of material properties suitable both for printability with high shape fidelity and for the support of cell function ( Figure 1 ). [ 12 ] This concept has since then widely been appreciated in the biofabrication community. [ 13 – 17 ] Strategies to extend this biofabrication window and allow for printing with good shape fidelity under cytocompatible conditions with as little material content as possible have been an important focus of recent research in the field. Bioprinted structures are currently being studied as potential transplantable grafts for tissue restoration, [ 18 ] as advanced in vitro models to aid the testing of drugs and as potential alternatives to animal experimentation. [ 19 – 21 ] These are used to study tissue development and disease [ 22 ] and as components integrated within organ-on-chip devices. [ 23 ] While the development of new bioinks and adaptations of existing printing technologies is an important part of current biofabrication research, novel strategies have emerged, introducing alternative approaches to push the boundaries of the biofabrication window. This allows for the fabrication of larger, more sophisticated structures even when softer hydrogels are used and led to significant advances in the generation of 3D constructs displaying salient features of native tissues, such as those of bone, [ 24 ] skin, [ 25 ] cartilage, [ 26 ] cardiac muscle, [ 27 ] thyroid, [ 28 ] and liver tissue. [ 29 ] Despite these promising examples, studies that clearly demonstrate the advantage of bioprinting in achieving 3D cell–material constructs that exhibit, at least to a certain extent, functional characteristics of living tissues are not widespread. In the present review, we summarize key strategies that have expanded the biofabrication window and that lead to improved control over shape. Building on such advances in material science, the main focus here is on the current and future steps toward mimicking salient functionalities of living tissues, through the creation of hierarchically structured constructs, in particular when using bioinks as building blocks for extrusionbased bioprinting. For this, the impact of bioprinted constructs with preformed spatial organization to facilitate tissue morphogenesis will be critically discussed. We highlight recent and upcoming developments in biofabrication that could influence the next generation of engineered tissues. Finally, we urge that future strategies embrace biological (developmental) processes and integrate them with bioprinting technologies to yield constructs with biological function toward the ambitious goal of printing functional tissues or even entire organs."
} | 4,510 |
32461810 | PMC7240055 | pmc | 9,658 | {
"abstract": "Graphical abstract",
"conclusion": "Concluding remarks The major global challenge of these days is to produce more yields from crops with less use of fertilizer and agrochemical inputs on limited land. Here, we discussed about this emerging field of rhizosphere microbial engineering which offers influential and exciting opportunities to fill these knowledge gaps and endow with possible answers. By exogenous inoculation of particular microbes or beneficial microbiome at large it is possible to alter the structure of the microbial community to increase disease resistance in plants and uptake of specific nutrients. In this regards, the development of so-called “microbiome-driven cropping systems” might result in the next revolution in agriculture, resulting in a more sustainable system for plant production. Furthermore, the application of multiomics approaches coupled with genome editing techniques like CRISPR for enhancing nutritional status, disease resistance and crop yield will result in the progress of Non-GMO or non-genetically modified crops with desired yield and will help in future for achieving zero hunger goal for continuously increasing human population. Future research studies will utilize synthetic biology approaches, to make full use of positive plant-microbiome interactions and employ a combination of both approaches to improve the productivity of major food and bioenergy crops under environmental stress conditions, at the same time, granting for an increased drawdown of atmospheric CO 2 to stabilize carbon pools in the soil.",
"introduction": "Introduction To feed the growing human population of 7.6 billion to an estimated 9.5–10 billion by 2050, will be a major challenge for the scientists across the globe. Recently, crop production is facing severe threat due to various abiotic and biotic stresses as well as limited land availability. In nature, plants are exposed to trillions of microbes that colonize and occupy different chambers or compartments of the plant like rhizosphere, rhizoplane, endosphere and phyllosphere, hence considered as a secondary genome of plant [1] , [2] . Several studies have been conducted in the greenhouse, field and in laboratory in order to minimize input cost and to provide beneficial services to the plants ( Table 1 ). The plants and its microbiome are therefore, reported to function as metaorganism or holobiont [3] , [4] . The roots of crop plants creates an interface between the plant and the soil environment, thus establishing an enormous reservoir of microbial community [5] , [6] . Rhizosphere is the narrow zone of the plant roots surface and is of paramount importance for providing various ecosystem services, like cycling of nutrients and uptake of carbon [7] , [8] . To maximize the microbiome functions, we have to understand the biochemical and molecular determinant around the roots or the rhizosphere that governs the selective microbial enrichment [9] , [10] , [11] . Earlier, carbohydrates were recognized as the molecular determinants in the rhizosphere, but the studies validated that amino acids act as chemical determinants present in the rhizosphere [12] . Additionally, various flavonoids and secondary plant metabolites were considered as key drivers for the successive establishment of the host specific microbial population in the rhizospheric zone [13] , [14] , [15] . However, it’s not clear that these microbes are interacting with some plants either in positive or in a negative way as diversity of these microbes are different in different plants. Strong published evidences, showed that these plant inhabiting microbes are potential biofertilizers and biocontrol agents and can be used for sustainable crop production [16] , [17] . Studies conducted by different researchers unravel the understanding of the mechanism of beneficial microbiome for enhancing plant health and performance under different stress conditions [1] , [2] , [4] , [18] , [19] , [20] , [21] , [22] . These studies were based on the cultivable microbial diversity, whereas the uncultivable microbes have rarely been explored and there is an urgent need to explore the potential of these unseen microbial diversity [1] , [23] . Table 1 Pyrosequencing analysis of taxonomic composition of microbes from different compartments of host plants (Rhizosphere, Endosphere, Rhizoplane). S. No. Plant/crop Rhizosphere Endosphere Rhizoplane Sequencing technique used Dominant species References 1. Para grass ( Urochloa mutica ) +++ 16S rRNA Bacillus , Chloroflexi, Microcoleus Clostridium , Caldilinea , [153] 2. Wheat plants ( Triticum aestivum ) +++ 16S rRNA Achromobacter , Clostridia , Cellulomonas , Bacillus , Gallionella , Herbaspirillum , Pseudomonas , Rhizobium , Xanthomonas , Sinorhizobium , Burkholderia , Pantoea , Enterobacter, Geobacter , Stenotrophomonas , Nocardia , Mycobacterium , Microbacterium [33] 3. Maize ( Zea mays L.) +++ 16S rRNA variable gene (V4–V5) Acidobacteria, Gemmatimonas Rhodoferax [154] 4. Taxus cuspidate var. Nana +++ 16S rRNA Actinobacteria, Chloroflexi [155] 5. Aloe vera ( Aloe barbadensis ) +++ 16S rRNA variable gene (V3–V4) Proteobacteria, Firmicutes, Actinobacteria, Bacteriodetes [156] 6. Rice ( Oryzae sativa ) +++ 16S rRNA gene sequencing Geodermatophilus , Actinokineospora , Actinoplanes , Streptomyces , Kocuria [157] 7. Triticum aestivum (Wheat) +++ +++ 16S rRNA gene sequencing Bacillus , Acetobacter, Stenotrophomonas [158] 8. Triticum aestivum (Wheat) +++ 16S rRNA gene sequencing Azoarcus , Balneimonas , Bradyrhizobium , Gemmatimonas, Lysobacter , Methylobacterium , Mesorhizobium , Microvirga , Rubellimicrobium , Rhodoplanes , Skermanella [159] 9. Soybean ( Glycine max ) +++ 16S rRNA gene sequencing Bacillus , Bradyrhizobium rhizobium , Stenotrophomonas , Streptomyces [160] 10. Lettuce ( Lactuca sativa ) +++ 16S rRNA gene sequencing Alkanindiges , Sphingomonas , Burkholderia , Novosphingobium , Sphingobium [161] 11. Salix (Willow) +++ 16S rRNA gene sequencing Pseudomonas, Sphingomonas yanoikuyae , Staphylococcus haemolyticus , Microbacterium oleivorans , Janthinobacterium lividum, Stenotrophomonas , Micrococcus luteus, Pantoea, Sphingomonas, Delftia [162] 12. Arabidopsis thaliana (Thale cress) +++ 16S rRNA gene sequencing Arthrobacter , Kineosporiaceae, Flavobacterium , Massilia [163] 13. Arabidopsis thaliana (Thale cress) +++ +++ +++ 16S rRNA. variable gene (V5–V6) Acidobacteria, Planctomycetes, Proteobacteria, Actinobacteria, Bacteroidetes [164] 14. Pennisetum +++ BOX-PCR ,16S rRNA and nifH sequences Azospirillum brasilense , Gluconacetobacterdi azotrophicus, Gluconacetobacter liquefaciens, Gluconacetobacter sacchari, Burkholderia silvatlantica, , Klebsiella sp., Enterobacter cloacae and Enterobacteroryzae [165] 15. Oryza sativa (Cultivated Rice) +++ Metaproteogenomic approach Actinobacteria, Proteobacteria [166] 16. Populus deltoides (Poplar) +++ +++ Acidobacteria, Proteobacteria [167] 17. Sugarcane +++ 16S rRNA gene sequencing Citrobacter , Enterobacter , Pantoea , Klebsiella , Erwinia, Brevibacillus , Staphylococcus , Curtobacterium , Pseudomonas sp. [168] 18. Poplar ( Populus deltoides ) +++ Shotgun metagenomics P. putida [169] 19. Avena fatua (wild oat) +++ 16S rRNA microarray (Phylochip) Actinobacteria, Firmicutes, Proteobacteria [170] Recent researches proved the use of beneficial microbiome in improving the crop yield and health of plants grown under limited conditions. Although, more research are needed on individual crops growing under stressed conditions to harness full microbiome potential. Moreover, the global climate change includes unpredicted weather pattern and elevated temperature which affects the overall functioning of ecosystem and rhizosphere biology, through direct and indirect mechanism. Therefore, the diversity of microbes present near the rhizosphere zone plays a pivotal role in enhancing plant growth by facilitating the acquisition of nutrition, providing defense against pest and pathogens, and helping plant to tolerate different types of abiotic and biotic stresses. Various types of abiotic stress include drought, salinity and high temperature that causes several negative impacts such as a major economic loss in crop productivity by reducing water absorption, nutrient acquisition, disease susceptibility and disturbing hormonal balance and also by affecting photosynthetic capacity of the plant [24] . However, still these beneficial microbes are not utilized on a full scale as only about 1–5% of the microbes present on the earth are cultivable remaining 95–99% of microbes are uncultivable [23] . U nderstanding of plant microbe interaction has been a foremost area of research for several years. Recently, the advancement of high-throughput sequencing and NGS approaches has provided new insight into how these microbial communities are affected by different environmental factors and the crop genotype had made an entire catalog of the pathogens associated with specific crops, [25] , [26] . In case of plant disease a intricate interaction between a pathogen and the host plant, and the resistance/susceptibility response can involve many components [27] . Genome editing technologies like CRISPR/Cas9 have rapidly progressed and become essential genetic tools used for developing pathogen stress tolerance in plants [28] . Many studies conducted by different scientists have shown the importance of omics approaches to find out the uncultivable microbial flora however taxonomic and functional study of plant microbial flora is limited and rarely emphasized in detail. The rationale of this review is to decipher the role of cultivable and uncultivable microbial community associated with rhizosphere for maintaining growth and development of the plant, including the concept of shaping plant microbiome for sustainable crop production. Present review also highlighted the omics approaches, strategies for engineering rhizosphere microbiome of the plant and modern advancement made for the protection of plant by using CRISPR/Cas9 technology in some model crops plant in response to diseases caused by various microbes. Schematic flow of development of strategies for analyzing plant microbiome from different compartments and use of Omics approach for understanding of cultivable and uncultivable microbiome for plant growth promotion is shown here in Fig. 1 . Fig. 1 Schematic flow of development of strategies for analyzing plant microbiome from different compartments and use of Omics approach for understanding of cultivable and uncultivable microbiome for plant growth promotion."
} | 2,669 |
29938091 | PMC6010792 | pmc | 9,659 | {
"abstract": "Abstract Foraging behavior is a critical adaptation by insects to obtain appropriate nutrients from the environment for development and fitness. Bumble bees ( Bombus spp.) form annual colonies which must rapidly increase their worker populations to support rearing reproductive individuals before the end of the season. Therefore, colony growth and reproduction should be dependent on the quality and quantity of pollen resources in the surrounding landscape. Our previous research found that B. impatiens foraging preferences to different plant species were shaped by pollen protein:lipid nutritional ratios (P:L), with foragers preferring pollen species with a ~5:1 P:L ratio. In this study, we placed B. impatiens colonies in three different habitats (forest, forest edge, and valley) to determine whether pollen nutritional quality collected by the colonies differed between areas that may differ in resource abundance and diversity. We found that habitat did not influence the collected pollen nutritional quality, with colonies in all three habitats collecting pollen averaging a 4:1 P:L ratio. Furthermore, there was no difference in the nutritional quality of the pollen collected by colonies that successfully reared reproductives and those that did not. We found however, that “nutritional intake,” calculated as the colony‐level intake rate of nutrient quantities (protein, lipid, and sugar), was strongly related to colony growth and reproductive output. Therefore, we conclude that B. impatiens colony performance is a function of the abundance of nutritionally appropriate floral resources in the surrounding landscape. Because we did not comprehensively evaluate the nutrition provided by the plant communities in each habitat, it remains to be determined how B. impatiens polylectic foraging strategies helps them select among the available pollen nutritional landscape in a variety of plant communities to obtain a balance of key macronutrients.",
"conclusion": "5 CONCLUSIONS In this study, we expected that differences in habitat would lead to differences in colony performance and success, which would be associated with variation in nutritional intake quantity or quality. Our data indicate that B. impatiens collected similar pollen nutritional resources (in terms of the macronutrient ratios) in all three habitats where we placed colonies, although time of the season influenced these variables (Kämper et al., 2016 ; Kriesell et al., 2016 ). However, there were clear differences in colony foraging rates, growth, and reproduction in these different habitats. Furthermore, we found that colony development strongly correlated with foraging resource quality and availability, as reflected by foraging rates, the amount of pollen collected, and the total macronutrient quantity the foragers brought in to the colony. Thus, poor performance was associated with reduced collection of nutritionally suitable floral resources from the habitat. This reduction in foraging effort may have resulted from fewer floral resources in the landscape or workers having to travel farther to find appropriate resources, both of which could have reduced the foraging force (Dornhaus & Chittka, 2001 , 2004 ; Génissel et al., 2002 ; Kitaoka & Nieh, 2008 ; Pope & Jha, 2018 ) and nutritional intake of colonies. Future studies should analyze the nutritional quality of both bee‐collected pollen and the local plant communities to determine how bumble bees—and indeed, other bee species—selectively versus opportunistically forage among a variety of pollen nutritional landscapes to meet their nutritional requirements.",
"introduction": "1 INTRODUCTION An appropriate quality and quantity of macronutrients are essential for development and reproduction of every organism (Behmer, 2009 ; Behmer & Joern, 2008 ). These nutritional needs are hypothesized to strongly influence foraging behavior, ensuring that an animal obtains required macronutrient (carbohydrate, protein, and lipid) intake from varied environments where the nutritional qualities of resources may differ (Jensen, Mayntz, Toft, Raubenheimer, & Simpson, 2011 ; Mayntz, Raubenheimer, Salomon, Toft, & Simpson, 2005 ; Raubenheimer, Mayntz, Simpson, & Tøft, 2007 ; Raubenheimer & Simpson, 1999 ; Simpson & Raubenheimer, 1993 , 2012 ). Foraging bees obtain all their nutrients from floral pollen and nectar, and these floral resources vary among plant species in quality, quantity, and availability throughout space and time (Nicolson, Nepi, & Pacini, 2007 ; Petanidou, Kallimanis, Tzanopoulos, Sgardelis, & Pantis, 2008 ; Roulston & Cane, 2000 ; Willmer & Stone, 2004 ). Therefore, foraging bees must select among these resources to support their own homeostasis and reproduction, and provide nutrients for larvae confined to brood cells (Brodschneider & Crailsheim, 2010 ; Cane, 2016 ; Nicolson et al., 2007 ; Roulston & Cane, 2000 ). Bumble bees produce annual colonies, initiated by a single foundress queen, that ultimately comprise several hundred individuals (dependent on species) before reaching a “switching point” where the colony produces the next generation of reproductives (gynes and males) by the end of the growing season (Cnaani, Schmid‐Hempel, & Schmidt, 2002 ; Crone & Williams, 2016 ; Duchateau & Velthuis, 1988 ; Goulson, 2010 ; Williams, Regetz, & Kremen, 2012 ). Thus, these species must have continual access to quality floral resources for months to continuously grow the colony to a reproductive stage. Global declines in bee populations, including bumble bees, have been linked to habitat degradation, including agricultural intensification, that reduces floral abundance and diversity (Biesmeijer et al., 2006 ; Goulson, Nicholls, Botías, & Rotheray, 2015 ) and the loss of key host‐plant species (Carvell et al., 2006 ). However, studies examining how variation in landscape influences bee health have focused on managed Apis mellifera honey bees (Otto, Roth, Carlson, & Smart, 2016 ; Requier et al., 2015 ; Smart, Pettis, Euliss, & Spivak, 2016 ; Smart, Pettis, Rice, Browning, & Spivak, 2016 ; Sponsler & Johnson, 2015 ). It is therefore not necessarily accurate to extrapolate the results of these studies to solitary bees or bumble bees, which have substantially smaller, annual colonies. Only a handful of studies have monitored how landscape and floral resource availability influence dynamics of bumble bee colonies (Elliott, 2009b ; Goulson, Hughes, Derwent, & Stout, 2002 ; Kämper et al., 2016 ; Lanterman & Goodell, 2017 ; Westphal, Steffan‐Dewenter, & Tscharntke, 2009 ; Williams et al., 2012 ). Colony growth appears to be strongly correlated with early‐season resource availability (Westphal et al., 2009 ; Williams et al., 2012 ), although growth has not been found to be a predictor of reproductive output (Crone & Williams, 2016 ; Goulson et al., 2002 ; Williams et al., 2012 ). In landscapes with higher diversity of floral resources, B. terrestris colonies grew more quickly and larger, yet did not differ in their ultimate output of reproductive individuals (Goulson et al., 2002 ). However, in B. impatiens , both colony growth and reproductive output were increased by floral diversity (Lanterman & Goodell, 2017 ). We expect that landscapes with differing degrees of floral resource diversity and abundance would lead to differences in pollen nutritional quality and quantity available, and ultimately differences in colony fitness. However, thus far the influence of landscape on colony‐level nutritional intake has not been explicitly examined. Mounting evidence suggests that bumble bees (Hymenoptera: Apidae: Bombus spp.), especially B. impatiens and B. terrestris , show foraging preferences for plant species based on nutritional quality of pollen (Cardoza, Harris, & Grozinger, 2012 ; Hanley, Franco, Pichon, Darvill, & Goulson, 2008 ; Kitaoka & Nieh, 2008 ; Kriesell, Hilpert, & Leonhardt, 2016 ; Leonhardt & Blüthgen, 2011 ; Ruedenauer, Spaethe, & Leonhardt, 2016 ). Evolutionarily, this preference aligns with a goal of providing optimal resources for their brood, because suboptimal pollen quality can lead to reproductive deficit, egg cannibalism, and larval ejection (Génissel, Aupinel, Bressac, Tasei, & Chevrier, 2002 ; Tasei & Aupinel, 2008 ). In the laboratory, bumble bees prefer pollen diets with higher protein concentrations (Kitaoka & Nieh, 2008 ; Konzmann & Lunau, 2014 ; Ruedenauer, Spaethe, & Leonhardt, 2015 ; Ruedenauer et al., 2016 ), and these preferences extend to the field among plant species or within the same species (Cardoza et al., 2012 ; Hanley et al., 2008 ). Furthermore, bumble bee colonies will increase their foraging efforts to higher quality pollen (or nectar), or reduce foraging efforts to low‐quality pollen, even if no alternative is available (Dornhaus & Chittka, 2001 , 2004 ; Kitaoka & Nieh, 2008 ). Our previous research revealed that B. impatiens , when collecting pollen for their colony in an enclosed outdoor foraging‐arena, preferred host‐plant species with pollen of high protein:lipid, or P:L ratios (~5:1 P:L, which was the maximum for the plant species in this study; Vaudo, Patch, Mortensen, Tooker, & Grozinger, 2016 ). Notably, foragers nearly ignored plant species offering the lowest P:L pollen (0.72:1 P:L), even when abundant pollen was available for collection (Vaudo, Patch et al., 2016 ). Additionally, in the laboratory in the absence of external floral cues and brood, B. impatiens maintained these P:L preferences among pollen from different species and exhibited preferences of 5:1–10:1 P:L from nutritionally modified pollens (Vaudo, Patch et al., 2016 ). Insect nutritional preferences exhibited in controlled settings may reflect the optimum that would be collected in the field. However, in different landscapes of floral abundance and diversity, bees may differ in their ability to meet their optimal nutritional needs and therefore affect colony fitness. We tested whether bumble bees differ in their nutritional intake and colony growth in different habitats of a typical Pennsylvanian agricultural landscape—agricultural valley, field edge, and forest—that we expected to vary in floral diversity and abundance. We expected that field edges provide higher diversity of host‐plant species that should provide bumble bees nutritionally rich and season‐long forage availability and result in highest colony growth (Kammerer, Biddinger, Rajotte, & Mortensen, 2016 ). In contrast, forest summer floral resource diversity and abundance are lower, and agricultural land should have reduced floral diversity (punctuated by blooming of only a few crop species), both of which may result in lower colony growth and ability to obtain optimum nutrition (Goulson et al., 2002 ; Williams et al., 2012 ). We (1) determined whether nutritional quality of pollen collected by colonies varied among these three habitats; and (2) determined whether behavioral and nutritional factors (foraging rates, pollen quantity, and pollen quality) related to total nutritional intake rates influenced colony growth and reproduction. Overall, these data provide critical information integrating nutritional intake of bumble bees with colony behavioral dynamics, growth, and fitness.",
"discussion": "4 DISCUSSION Our study is the first to test the nutritional quality of pollen collected by Bombus impatiens over time while nesting in different habitats. Further, the study appears to be the first to incorporate the nutritional value of pollen into predictions of colony fitness. Our approach began by evaluating the nutritional quality of pollen collected, then we determined what environmental, behavioral, and colony factors lead to colony growth, and finally, we created a metric of nutritional intake to integrate these factors and predict how the rate of nutrient consumption affects colony growth dynamics and reproductive output. 4.1 Pollen nutrition Protein, lipid, carbohydrate, and P:L values of pollen collected by colonies did not differ between habitats (Table 1 ), contrary to our expectation that floral diversity would differ between habitats (increasing at the field edge habitat) and lead to differences among the nutrients collected by colonies. The pollen collected in each habitat averaged a ~4:1 P:L ratio. This is similar finding to our previous research wherein semi‐field and laboratory choice assays B. impatiens preferred pollen with a ratio of 5:1 (Vaudo, Patch et al., 2016 ). In the current study, the distribution of pollen across the colonies and weeks was both above and below 5:1 (Figure 2 a), indicating substantial variation in the P:L ratios of available pollen in the landscape. Among the diversity of pollen nutrients available throughout the landscape, bumble bees in each habitat were still able to converge on an average P:L ratio that resembled our predictions from our controlled experiments (Vaudo, Patch et al., 2016 ). These results do not definitively answer the question of whether B. impatiens selectively foraged for a 4:1 P:L ratio or if they passively collected what was available in the landscape, which would require that the floral plant communities in these different landscapes all exhibited an average of 4:1 P:L ratios. Our hypothesis that bumble bees selectively forage for nutrition is supported by choice assays and empirical surveys. When fed modified pollen or synthetic diets, B. impatiens preferred and survived best on 5:1–10:1 P:L diets (Vaudo, Patch et al., 2016 ; Vaudo, Stabler et al., 2016 ). But high pollen P:L ratios such as 10:1 may be rare or in low abundance in the field (see the following references for pollen nutritional concentrations: Roulston & Cane, 2000 ; Roulston, Cane, & Buchmann, 2000 ; note that pollen P:L ratios so far have only been calculated in Vaudo, Patch et al., 2016 ). In an empirical survey of the nutritional content of pollen of 68 bee‐pollinated plant species, 52 had P:L ratios below 4:1 P:L (unpublished data). This suggests that pollen P:L ratios greater than 4:1 pollen may be relatively uncommon in the landscape. Several studies have suggested that bumble bees do selectively forage in the field: They tend to visit plant species with higher pollen protein content (Hanley et al., 2008 ) and exhibit foraging behavior to obtain quality (nutrient content or abundant) pollen including traveling further for higher quality resources (Jha & Kremen, 2013 ; Pope & Jha, 2018 ; Ruedenauer et al., 2016 ). Additionally, when foraging in the same landscapes, bumble bees collect pollen higher in protein content than honey bee colonies (Leonhardt & Blüthgen, 2011 ). Nevertheless, it is not possible to definitely test the hypothesis that B. impatiens selectively forages for particular P:L ratios in floral communities in the field until one assesses the actual distribution of pollen nutrition and abundance across all host‐plant species in the landscape and compares it against the nutritional content of pollen brought back to colonies. Protein‐to‐lipid ratios of collected pollen varied over the season, yet these were driven by variation in protein content while lipid concentrations of corbiculate loads remained surprisingly consistent (5.5 ± 1.1% lipid; Figure 2 b). Similarly, the protein content of collected pollen varied with the temperature and time of day at which foragers were foraging while lipid content did not (Figure 3 ). These results are similar to our previous findings that B. impatiens collected more preferred pollen earlier in the day and then moved to lower quality resources (Vaudo, Patch, Mortensen, Grozinger, & Tooker, 2014 ; Vaudo, Patch et al., 2016 ). The trend for bees to collect higher protein content pollen on warmer days is also consistent with the finding that bumble bees are more likely to collect pollen in dry and warm conditions (Peat & Goulson, 2005 ) and perhaps are more selective for pollen quality under such conditions. Therefore, although pollen resources available to B. impatiens appear environmentally driven, through seasonal and daily phenology, their consistent collection of pollen lipid concentration is striking. As noted above, whether this consistency results from bumble bees actively collecting from quality (high protein content, low lipid content, or abundant) patches of particular floral resources (Jha & Kremen, 2013 ; Ruedenauer et al., 2016 ), or simply due to average lipid content from random collection, remains to be determined. 4.2 Nutrition, behavior, growth, and reproduction Although nutritional quality of pollen across all colonies and habitats did not differ, there were still differences in levels of colony growth between habitats. In particular, Valley colonies outperformed Edge and Forest colonies (Table 2 ), contrary to our expectations that colonies nesting along the field edge would have season‐long diverse pollen resources leading to increased colony growth and reproductive output. This suggests that Valley (agricultural and residential land) colonies may have had easier access to more preferred pollen patches, increasing foraging rates and worker populations in the agricultural habitat, which is consistent with other studies (Requier et al., 2015 ; Sponsler & Johnson, 2015 ) indicating that agricultural landscapes—which themselves contain a great deal of edge habitat and weedy plants—can provide abundant and diverse resources for bees. Additionally, proximity to forested land, and perhaps the lower floral diversity in this region, may have led to lower colony growth and reproduction in Forest and Edge habitats (Lanterman & Goodell, 2017 ). We attempted to separate colonies such that they may not overlap in foraging ranges or resources, yet they may have collected from similar patches of flowers. Foragers from colonies at different sites may have foraged further distances to reach these patches however (Pope & Jha, 2018 ), perhaps causing energetic stress and differences in colony growth. Indeed, there were significant differences between sites in all categories of growth, reproduction, and foraging, and therefore, broad classification of habitats alone did account for differences in colony health (Table 2 ; see “Case Study” in Supporting Information and Figure S1 ). For instance, Valley 2 colonies were placed in an agricultural area where hedgerows, neighborhoods, and flowering crops likely provided more consistent floral resource availability and diversity leading to increased health (Alaux et al., 2016 ; Goulson et al., 2002 ). Edge 4 colonies were placed in the only perceivable area with wildflower abundance along the forest edge, possibly explaining the success of these colonies compared to other Edge sites. In contrast, Valley 4 colonies were placed in wheat and corn fields without any obvious wildflower habitat. Therefore, assessing floral resources in proximity to colonies (Williams et al., 2012 ) would likely be more predictive of resource diversity, abundance, and quality than assumptions based on general habitat type (Lonsdorf et al., 2009 ; Requier et al., 2015 ). We therefore chose to determine what nutritional and behavioral factors were most predictive of colony reproductive growth and fitness independent of habitat using PCA (Figure 4 ). Pollen nutrition was not correlated to any measure of colony growth, again because there were no differences in colony‐level nutritional quality of pollen. Colony foraging rates, corbiculate pollen mass, colony biomass, and total population were all correlated with colony reproductive success (Figure 4 ), supporting the model that (1) nutritional quantity—likely determined by floral abundance—is critical for colony development (Williams et al., 2012 ) and (2) increasing colony size is critical for colonies to switch to producing new reproductives (Westphal et al., 2009 ). We developed a “nutritional intake” metric that integrates nutrition and behavior to evaluate the colony‐level rate of nutrient consumption how this influences growth and reproduction. We evaluated nutritional intake separately for protein, lipid, and carbohydrates to determine whether any one nutrient was more predictive of colony growth than total nutrients (and to be used as a tool for evaluating colony growth in future studies). For all three macronutrients, colonies exhibiting higher nutritional intake were able to outgrow other colonies and produce higher numbers of reproductives (Figures 5 , 6 ). Colony #18, which was the only colony to produce gynes (all the other reproductive colonies produced males), had the highest nutritional intake among all colonies for protein, lipid, and sugar exemplifying the use of this metric. These results suggest a positive feedback loop between environmental availability of resources (nutritional quantity) and pollen foraging rates and capacity (Dornhaus, Brockmann, & Chittka, 2003 ; Dornhaus & Chittka, 2001 , 2004 , 2005 ; Hendriksma & Shafir, 2016 ; Kitaoka & Nieh, 2008 ): Better resources lead to higher worker populations and increased foraging (Amsalem, Grozinger, Padilla, & Hefetz, 2015 ), which led to higher colony growth and reproduction. Thus, B. impatiens foragers appear to have collected pollen that at least met the minimum nutritional requirements of developing larvae in all cases, but the quantity of these pollen resources influenced the colony growth rate and ability to transition to their reproductive phase (Goulson et al., 2002 ; Jha & Kremen, 2013 ; Kämper et al., 2016 ; Kriesell et al., 2016 ; Lanterman & Goodell, 2017 ; Ruedenauer et al., 2016 ; Vaudo, Patch et al., 2016 ; Westphal et al., 2009 ; Williams et al., 2012 ). Further, our results suggest that B. impatiens can exhibit plasticity in colony growth and behavioral dynamics that accommodate nesting‐site specific differences in floral resources, allowing colonies to reach reproductive maturity despite their immediate resource base ( Supporting Information ; Figure S1 ). Further studies are needed to (1) determine the interaction of nutritional quality and quantity that yields maximum colony growth and (2) determine the interacting environmental and nutritional factors that contribute colony reproductive (male to gyne ratio) output."
} | 5,587 |
36459292 | PMC9718865 | pmc | 9,660 | {
"abstract": "Microbial carbon use efficiency (CUE)—the balance between microbial growth and respiration—strongly impacts microbial mediated soil carbon storage and is sensitive to many well-studied abiotic environmental factors. However, surprisingly, little work has examined how biotic interactions in soil may impact CUE. Here, we review the theoretical and empirical lines of evidence exploring how biotic interactions affect CUE through the lens of life history strategies. Fundamentally, the CUE of a microbial population is constrained by population density and carrying capacity, which, when reached, causes species to grow more quickly and less efficiently. When microbes engage in interspecific competition, they accelerate growth rates to acquire limited resources and release secondary chemicals toxic to competitors. Such processes are not anabolic and thus constrain CUE. In turn, antagonists may activate one of a number of stress responses that also do not involve biomass production, potentially further reducing CUE. In contrast, facilitation can increase CUE by expanding species realized niches, mitigating environmental stress and reducing production costs of extracellular enzymes. Microbial interactions at higher trophic levels also influence CUE. For instance, predation on microbes can positively or negatively impact CUE by changing microbial density and the outcomes of interspecific competition. Finally, we discuss how plants select for more or less efficient microbes under different contexts. In short, this review demonstrates the potential for biotic interactions to be a strong regulator of microbial CUE and additionally provides a blueprint for future research to address key knowledge gaps of ecological and applied importance for carbon sequestration.",
"conclusion": "Conclusion and Further Considerations Biotic interactions influence CUE by affecting the allocation of carbon to different metabolic processes, changing environmental conditions, and inducing shifts in microbial community composition. However, a huge variety of biotic interactions take place in soil, making it challenging to predict CUE without having a complete picture of their cumulative impacts. While we broadly show that competition is a “negative” interaction that reduces CUE, we also present evidence that indirect competition can, in some cases, positively impact CUE. At the same time, facilitation is a “positive” interaction that is generally expected to increase CUE, especially when microbial population densities are low. Although plant–microbe interactions are often facilitative, such interactions can reduce CUE in the rhizosphere and potentially more widely in ectomycorrhizal fungal-dominated systems. Furthermore, we know that biotic interactions can have complex cascading effects, with some interactions generating new interactions (see Fig. 2 ). For example, indirect competition can increase the probability of direct competition and facilitation can decrease the probability of direct competition or predation. Importantly, cascading effects of biotic interactions are widely known to introduce apparent stochasticity to microbial communities, which may be difficult to predict [ 108 ]. Future research must ultimately focus on teasing such apparent complexity apart. In conclusion, by reviewing literature spanning the fields of microbiology, evolution, and botany, we analysed how soil microbial CUE is affected by a key set of biotic interactions. This is important because while there is a concerted effort to understand the abiotic factors regulating microbial CUE, surprisingly few studies have addressed the additional role of biotic interactions in soil ecosystems. Of course, studying soil microbial interactions comes with substantial technical challenges. Several methods are used to measure CUE and they differ in meaningful ways depending on whether the CUE is measured at the population, community or ecosystem scale, or in cultures, mesocosms, or actual field soil (see [ 109 ]). Creating generalizable CUE frameworks is therefore challenging when working across different scales and media. While much of the evidence presented here considering biotic interactions and CUE yields more questions than answers, this review was written to organize a path forwards. By summarizing the key theoretical biotic interactions that should affect CUE and pairing this with available supporting evidence, we were able to provide concrete research suggestions for the future. Despite addressing a new area of CUE research, we argue that understanding how biotic interactions shape microbial CUE is important not only for conceptually, but also for managing natural systems, such as to identify strategies in agricultural soils that favour biotic interactions known to increase CUE (see [ 43 ]). In our opinion, considering biotic interactions alongside abiotic drivers of CUE will ultimately improve both mechanistic insights and predictive power in ecosystem ecology and management.",
"introduction": "Introduction Soil microbes are major actors in the terrestrial carbon cycle [ 1 ]. Microbial products (e.g. necromass, proteins, DNA) commonly comprise 10–80% of the total soil organic carbon (SOC) stock [ 2 ], and the formation and stabilization of these products are a key determinant of ecosystem carbon sequestration. At the same time, microbial activity accounts for approximately 60% of global soil carbon dioxide (CO 2 ) emissions, making microbes an important component of the terrestrial carbon balance [ 3 ]. Two fundamental processes influence microbial SOC formation and depletion: growth, which produces biomass that may eventually become SOC; and respiration, which releases SOC as CO 2 . The balance between microbial respiration and growth is termed microbial carbon use efficiency (CUE; a.k.a. growth efficiency or yield), and is specifically defined as the proportion of assimilated carbon used for building new biomass relative to that lost through respiration and the activity of endogenous metabolism [ 4 , 5 ]. CUE is one of the few explicit microbial variables in SOC cycling models [ 6 ], so accurately predicting it is therefore of considerable interest. Yet, our ability to do so is limited by an incomplete understanding of the factors affecting CUE. Here, we argue that biotic variables, such as competition and facilitation, constrain CUE above and beyond abiotic controls and should be explicitly included in the next phase of CUE research. While the effects of abiotic factors on CUE have been extensively tested, e.g. in [ 7 ], much less is known about how biotic factors, such as competition, facilitation, predation, and plant–microbe interactions, affect CUE. This review aims to guide researchers towards a coherent research direction by linking microbial biotic interactions to microbial physiology with implications for SOC cycling. We draw from multiple lines of theoretical and empirical evidence from the evolution and botanical literature, which collectively suggest that biotic interactions should strongly affect CUE through differences in life history strategies. The most relevant life history frameworks include K- vs. r - selection [ 8 ], the competition-stress-ruderal (C-S-R) life history axes [ 9 ], and a reimagined microbial C-S-R that considers growth yield (Y), resource acquisition (A), and stress tolerance (S)—referred to as the Y-A-S framework [ 10 ]. While all of these life history frameworks directly or indirectly consider biotic interactions in the context of competition for resources, we argue that greater development and additional explicit biotic life history traits could be integrated into these frameworks to better predict CUE. CUE is an emergent property of multiple abiotic and biotic factors that are difficult to disentangle. However, by isolating the contributing effects of specific biotic interactions on CUE, it should be possible to disentangle the mechanisms by which microbes respond to abiotic changes. Doing so could thus allow us to predict microbial physiological performance under changing and/or novel environmental conditions. There is a wide range of biotic interactions occurring among soil microbes, including competition, predation, facilitation, and mutualisms among microbes and with higher trophic level organisms [ 11 ]. Hence, CUE may be impacted by many different types of biotic interactions simultaneously. Because microbiomes are typically hyper-diverse in the sense of diversity and functionality, the combined sum of positive (facilitation), neutral (commensal) and negative interactions (competition) will drive CUE at the aggregate community level [ 12 ]. Competition relates to negative interactions that deplete a population through the activity of antagonists onto protagonists [ 13 ]. It can occur among the same species (intraspecific competition) or among different species (interspecific competition), and can be direct (i.e. interference), indirect (i.e. exploitative), or predator-mediated [ 14 ]—though these forms of competition often overlap among microbes [ 15 , 16 ]. Because competitive interactions commonly require life history strategies that promote fast growth and extensive investments in resource acquisition by antagonists and stress response by protagonists, we hypothesise that competition causes microbes to grow less efficiently. Facilitation relates to myriad positive interactions between organisms that benefit at least one organism and cause no harm to either organism [ 12 , 13 ]. Facilitation can favour high CUE by (1) ameliorating abiotic stress; (2) creating novel habitats to promote niche partitioning; (3) increasing habitat complexity and heterogeneity; (4) sharing services like producing common goods and (5) increasing the availability of otherwise inaccessible resources [ 12 ]. Mutualism is a specialized form of facilitation that benefits both species, such as the exchange of services commonly observed between host plants and rhizobia or mycorrhizal fungi [ 13 ]. Because facilitation regularly promotes the exchange of resources and ameliorates stress, we anticipate that positive biotic interactions promote efficient microbial growth and may increase CUE at the community level if the sum of facilitation is greater than the combined sum of commensal plus competitive interactions. To address these two expectations, we draw on theoretical and empirical lines of evidence to determine how biotic interactions (competition, predation, facilitation, and interactions with plants) drive certain microbial life history strategies, with a particular focus on implications for CUE and SOC cycling. While the focus of this review is on CUE, we also critically evaluate the current state of knowledge on microbial species interactions and life history characteristics relevant to CUE. First, we address intraspecific interactions and density-dependent feedbacks to provide a foundation for understanding relationships between microbial growth and biotic interactions (“ CUE Fundamentally Depends On Density Dependence And Carrying Capacity ”) section. We then discuss the roles of interspecific direct competition (“ Interspecific Direct Competition Induces Metabolic Costs Leading to Low CUE ”) section, interspecific indirect competition (“ Interspecific Indirect Competition And Its Coincidence With Environmental Heterogeneity ”) section, facilitation (“ Facilitation Among Microbes Promotes Coexistence and Increases CUE ”) section, predation (“ Predation Influences Microbial Density and Competitive Outcomes ”) section, and plants (“ Effects of Plant Community and Plant-Microbe Interactions on CUE ”) section on microbial CUE as well the influence of spatial separation among soil organisms (“ Effect of Spatial Heterogeneity on CUE ”) section—a unique but important characteristic of soil systems."
} | 2,962 |
26324258 | PMC4555042 | pmc | 9,661 | {
"abstract": "Iron plaque is a strong adsorbent on rice roots, acting as a barrier to prevent metal uptake by rice. However, the role of root iron plaque microbes in governing metal redox cycling and metal bioavailability is unknown. In this study, the microbial community structure on the iron plaque of rice roots from an arsenic-contaminated paddy soil was explored using high-throughput next-generation sequencing. The microbial composition and diversity of the root iron plaque were significantly different from those of the bulk and rhizosphere soils. Using the aoxB gene as an identifying marker, we determined that the arsenite-oxidizing microbiota on the iron plaque was dominated by Acidovorax and Hydrogenophaga -affiliated bacteria. More importantly, the abundance of arsenite-oxidizing bacteria (AsOB) on the root iron plaque was significantly negatively correlated with the arsenic concentration in the rice root, straw and grain, indicating that the microbes on the iron plaque, particularly the AsOB, were actively catalyzing arsenic transformation and greatly influencing metal uptake by rice. This exploratory research represents a preliminary examination of the microbial community structure of the root iron plaque formed under arsenic pollution and emphasizes the importance of the root iron plaque environment in arsenic biogeochemical cycling compared with the soil-rhizosphere biotope.",
"discussion": "Discussion It is well-accepted that biotic and abiotic As(III) pathways coexist in soil 22 . Biogeochemical processes affecting the behavior of As in the soil environment have been a matter of considerable research interest to determine the chemical or biological factors controlling As(III) oxidation under oxic or anoxic conditions. As(III) is much more slowly oxidized by atmospheric O 2 than by other components in the soil environment, such as minerals and microorganisms 23 . Owing to their high reactivity at low concentrations and poorly crystalline structures with high surface areas, manganese oxide minerals are thought to be the most important oxidants of abiotic As(III) in nature 24 . In particular, after Fe, Mn is the second most abundant metal element in the root plaques of aquatic plants 25 . Thus, the Mn-induced pathway should be the major abiotic As(III) oxidation route on rice root plaques, although direct experimental evidence for this phenomenon has been lacking. The photocatalytic oxidation of As(III) to As(V) on ferrihydrite is another abiotic pathway for As transformation in nature and has been well researched in high-light water environments 26 . However, photo-induced As(III) oxidation is much less effective in the rhizosphere and deep soil than in the surface layer of soil because of the darkness of such biospheres. Furthermore, a recent report on the redox transformation of arsenic by Fe(II)-activated goethite indicates that As(III) oxidation may occur in the process of Fe(III) oxyhydroxide reduction at the rice root-plaque interface 27 . Kinetic experiments have shown that the rates of the reaction between As(III) and most chemical oxidants are impacted by pH, Eh, adsorbing surfaces, organic matter, and key inorganic substances 28 . During rice growth, the geochemical condition of the paddy soil is influenced by anthropogenic activities, and subsequent changes in the concentrations of O 2 and Fe species in soil affects abiotic As(III) oxidation on root plaques. Recent microbiological evidence suggests that As(III) is readily oxidized to As(V) by a large diversity of microorganisms under aerobic or anaerobic conditions 6 . Bacterial oxidation of As(III) is typically slower than oxidation via Mn-oxides, but a detailed understanding of the composition and patterns of As(III)-oxidizing microorganisms in rice root plaques is required before we can identify the most important bacterial species and guide the exploration of the potential of these microbes in mitigating health risks associated with arsenic in rice. Because of its capacity to gradually accumulate metals, root iron plaque has a profound influence on metal uptake and translocation in wetland plants 13 14 . Iron plaque is thought to be generated by the excretion of oxygen and oxidants into the rhizosphere by physiologically active rice roots, leading to the oxidization of Fe(II) and the precipitation of Fe(III) 15 . The critical geochemical factors controlling iron plaque formation include radial oxygen loss (ROL), ferrous ion availability, pH and organic carbon 15 29 . However, recent studies have revealed that microbial processes are also associated with iron plaque formation on plant roots 16 17 . Bacteria such as Sideroxydans paludicola and Sagittaria australis have been proposed to actively contribute to the formation of iron plaque on plant roots 16 17 . The data collected in this study provide the first overview of the iron plaque microbiota under As contamination. We discovered that the root iron plaque of rice was enriched with Pseudomonadales, Burkholderiales, Sphingomonadales and Rhizobiales ( Fig. 2 ). This detailed information about the community structure of microbes on iron plaque will aid in understanding and predicting Fe-As element biogeochemical cycling in the micro-biosphere. The UniFrac PCoA clustering results suggested that the microbial community structure of the iron plaque was significantly distinct from that of the bulk and rhizosphere soils ( Fig. 1 ). Accumulating evidence suggests that long-term As exposure permanently alters the microbial community structures of the bulk and rhizosphere soils 16 17 30 31 by decreasing the alkaline phosphatase, arylsulfatase, protease and urease activities of soil microorganisms 32 . Contamination was observed to impact microbial diversity and species richness in As-contaminated soils 33 . In our study, the microbial community diversity was significantly negatively correlated with the As contents of the bulk soil, rhizosphere soil and iron plaque ( Supplementary Fig. S2 ), indicating that As reduces microbial diversity not only in the soil-rhizosphere ecosystem but also at the soil-root interface. In addition, microbial diversity was significantly lower in the iron plaque than in the bulk and rhizosphere soils ( Supplementary Fig. S1 ). First, this observation may be a result of the high iron content and micro-aerobic conditions of the root iron plaque. Second, during Fe(II) oxidation on the iron plaque, the protons that are generated and released reduce the pH at the root surface 34 , thus likely decreasing the phylogenetic diversity and structure of the microbial communities. A comparative analysis of the heavily metal-contaminated bulk and rhizosphere soils of the metal-hyperaccumulating plant Thlaspi caerulescens revealed similar microbial community structures and diversities 35 . This result is consistent with our study of the microbial communities of As-contaminated bulk and rhizosphere soils. In this study, the distances within the microbial community of the rhizosphere soil (with an average unweighted UniFrac distance of 0.31) were shorter than those of the bulk soil and iron plaque communities, suggesting that under As pollution, the microbial community in the rhizosphere was less diverse than those of the bulk soil and iron plaque. Much attention has been paid to microbial As oxidation in aquatic or soil environments. However, the role of AsOB at the soil-plant interface (root iron plaque) has not been explored. As a phylogenetic maker, the aoxB gene is widely used for studying the abundance and taxonomy of AsOB in nature. It has been estimated that there are as many as 1.7 × 10 7 copies of aoxB -related genes per g of As-amended soil 33 ; because AsOB typically contain only one or two copies of the aoxB gene in their genomes 36 , this finding represents approximately 10 7 AsOB per g of As-contaminated soil. In our study, the number of aoxB gene copies retrieved from 1 g of wet roots ranged from 1.08 × 10 7 to 1.78 × 10 8 . The high number of aoxB gene copies indicates the As(III)-oxidizing potential of the microorganisms on the rice root plaque. Because of the difficulty in identifying As species on the root plaque, further studies should be performed to verify the microbes that mediate As(III)-oxidizing processes on the root plaque using X-ray absorption near edge structure (XANES) technology combined with molecular biology. The rice rhizosphere is highly oxygenated, and most AsOB isolated from soil are aerobes that utilize O 2 as an electron acceptor 6 . Thus, aerobic microbial oxidation of As(III) may occur readily and rapidly on the root plaque. Under anaerobic conditions, the As(V) attached to soil minerals is readily reduced to As(III) 37 , increasing As mobility in the soil biotope. Then, the reduced As(III) may move to the rhizosphere and become oxidized on the root plaque, resulting in decreased As bioavailability for the rice root. Moreover, arsenite and arsenate are taken up into plant roots by different mechanisms. Arsenate is taken up by phosphate transporters, whereas arsenite is taken up by rice roots mainly through the Si uptake pathway 38 . Thus, the As(III)-oxidizing bacteria on the root plaque may impact the pathway of As uptake by the rice root. Therefore, aerobic As(III) oxidation on the root plaque is more effective than that in soil under anaerobic flooding conditions. Thus, it is evident that biological As(III) oxidation occurring at the root plaque interface is closely intertwined with As uptake by rice. Compared with aerobic As(III)-oxidizing microorganisms, little is known about anoxic As(III)-oxidizers. Although oxidation with O 2 is more favorable based on biochemical energetic considerations, alternative oxidants with lower reduction potentials are feasible for the oxidation of As(III). , which has a higher electrochemical potential than As(III) under standard conditions, can be an alternate electron acceptor to support the oxidation of As(III) to As(V) by denitrifying bacteria under anoxic conditions. Evidence is growing that the anoxic oxidation of arsenite linked to nitrate reduction is feasible in continuous bioreactors 39 , sludges and sediments 40 . However, As(III) oxidation coupled to nitrate reduction has not been reported in paddy soil. In the denitrification zone (always below 3 mm in soil depth), nitrate is produced by ammonium transformation during fertilization with chemical fertilizers such as ammonium sulfate or organic fertilizers such as urea and straw 41 . In such an oxygen-depleted zone, nitrate replaces oxygen as the major electron acceptor for As(III) oxidation, and this zone also provides space for microorganisms capable of arsenite oxidation coupled to nitrate reduction to survive in paddy soil. It has been shown that the addition of nitrate results in decreased As accumulation by rice, suggesting a possible link between nitrate reduction and As(III) oxidation in paddy soil 42 . From an arsenic-contaminated paddy soil, a newly anaerobic, autotrophic As(III)-oxidizing bacterium was isolated that also exhibited the ability to reduce nitrate 12 . Furthermore, our recent data shows that As(III) oxidation is promoted by nitrate addition (data not published). Heterotrophic As(III) oxidation is generally assumed to be a detoxification process in which the microorganisms do not obtain energy from the oxidation of As(III) 8 . Thus, the anoxic oxidation of As(III) linked to nitrate reduction should be a mechanism for energy generation coupled with metal detoxification. In future, the microbial composition of nitrate reducers and As(III) oxidizers should be explored using the narG and aoxB genes as phylogenetic makers, respectively, to identify the underlying microbial mechanisms. Additionally, the relevance of As(III) oxidation and nitrate reduction in the root plaque system remains unknown and must be determined through future studies. Recently, it has been shown that the Fe(III) oxides generated by nitrate-dependent, Fe(II)-oxidizing bacteria are strongly adsorbing for As(V) 43 . Thus, the simultaneous microbial oxidation of Fe(II) and As(III) facilitated by nitrate may be a significant process leading to the formation of particulate ferric-oxide and As(V), resulting in immobilized As in the form of As(V) adsorbed onto biogenic Fe(III) (hydr)oxides with reduced mobility and toxicity in the rice root plaque. We identified Acidovorax, Hydrogenophaga and Sinorhizobium as the major genera of the AsOB on the iron plaque of paddy rice ( Fig. 4 ). The most abundant aoxB sequences were affiliated with Acidovorax sp. strain NO1 and Albidiferax ferrireducens T118, which belong to the Comamonadaceae family. Acidovorax sp. strain NO1 is a facultative, anaerobic, arsenite-oxidizing and nitrate-reducing bacterium that was isolated from gold mine soil 44 , whereas Albidiferax ferrireducens (formerly known as Rhodoferax ferrireducens ) strain T118 is a dissimilatory iron-reducing bacterium used as an acetate electron acceptor 45 . The Acidovorax genus harbors many typical strains of neutrophilic FeOB, such as Acidovorax sp. strains BoFeN1 and 2AN and Acidovorax ebreus strain TPSY 46 47 48 . These results indicate that the biological process of As oxidation is tightly coupled with iron cycling; in fact, Fe-As coupling oxidation may occur at the same genus level or even on the strain level, although experimental evidence is lacking to support the hypothesis that neutrophilic FeOB are capable of As(III) oxidation. This lack of evidence is largely because research on microbial As(III) oxidation in neutral pH environments has focused on sediment or surface water ecosystems rather than anaerobic or micro-aerobic soil or even As-contaminated soil ecosystems. Prior studies have only revealed that the redox cycling of iron affects the release, transport, immobilization and bioavailability of As in paddy soils 5 . The relationship between microbial dissimilatory Fe(III) reduction and As(V) reduction has been well studied in model organisms such as Shewanella putrefaciens strain CN-32 49 . However, the coupling of Fe(II) oxidation with As(III) oxidation in a soil environment is not as well understood. Our research on iron plaque suggests that Acidovorax -affiliated strains may be the ideal microorganism for investigating the role of bacterially induced, As(III)- and Fe(II)-coupled oxidation in controlling As mobility in soil ecosystems. Due to the higher As and Fe concentrations of rice iron plaque in As-polluted paddy soil, more effort is needed to isolate strains capable of As(III) and Fe(II) oxidation from such biospheres. A synthesis of the available data suggests that rhizo-bacteria reduce metal uptake in metal-tolerant plants by accumulating heavy metals in the soil in a plant-unavailable form. However, many metal-resistant bacteria can promote the uptake of heavy metals in plants by increasing water-soluble metals in the soil solution 50 . The above evidence suggests that the type of microbes and the metal uptake capacity of plants are two factors in plant-microbe interactions that are affected by metal contamination. There is no doubt that the biogeochemical cycling of As in bulk soil or the rhizosphere interface affects As mobility in soil and its bioavailability to the plant 19 . However, element accumulation and speciation in the iron plaque at the root surface is pivotal in understanding the transfer of nutrients or contaminants into rice roots. In addition, As distribution and speciation near rice roots is influenced by iron plaque and the redox conditions of the soil matrix 51 . To address the effects of AsOB on As uptake in rice, we should consider the iron plaque because of the strong association of Fe-As element cycling with root plaque and because iron plaque is the last barrier to toxic elements in rice roots. The higher dependence of the As content in rice roots on the abundance of AsOB on iron plaque may be a result of the co-evolution of the plant and microbes under metal stresses. The oxygen released from the rice roots affects the As-oxidizing population density and activity and also promotes the formation of iron plaque, which adversely leads to As immobilization on the root surface and reduces As bioavailability to the rice plant. The oxidized species, As(V), is much more strongly adsorbed by the iron plaque than As(III), which eventually reduces As bioavailability 52 . This process is in agreement with the finding that much of the adsorbed As on the rice iron plaque appears to be arsenate 13 , although both arsenate and arsenite were found to be present in association with the iron plaque of other wetland plants 53 . The formation of root iron plaque is promoted by the release of oxygen and oxidants into the rhizosphere 13 . However, recent studies have indicated that the microbial community associated with the root iron plaques of wetland plants are enriched in iron-oxidizing bacteria 16 54 , suggesting the key role of this activity in the root plaque microbial community. Therefore, As(III) may be transferred from soil particles or solution to the rice root plaque and oxidized to As(V) by microorganisms, which thus contributes an additional strategy for improving As immobilization and mitigating As contamination in rice. Our study demonstrates a significant correlation between the abundance of As(III)-oxidizing bacteria and As content in rice root, suggesting another pathway for decreasing As bioavailability through the rice root plaque microbial community."
} | 4,404 |
20190750 | null | s2 | 9,662 | {
"abstract": "In nature, sophisticated functional materials are created through hierarchical self-assembly of simple nanoscale motifs. In the laboratory, much progress has been made in the controlled assembly of molecules into one-, two- and three-dimensional artificial nanostructures, but bridging from the nanoscale to the macroscale to create useful macroscopic materials remains a challenge. Here we show a scalable self-assembly approach to making free-standing films from amyloid protein fibrils. The films were well ordered and highly rigid, with a Young's modulus of up to 5-7 GPa, which is comparable to the highest values for proteinaceous materials found in nature. We show that the self-organizing protein scaffolds can align otherwise unstructured components (such as fluorophores) within the macroscopic films. Multiscale self-assembly that relies on highly specific biomolecular interactions is an attractive path for realizing new multifunctional materials built from the bottom up."
} | 246 |
19802352 | null | s2 | 9,665 | {
"abstract": "No abstract available"
} | 5 |
32936809 | PMC7494125 | pmc | 9,666 | {
"abstract": "In the California Current Ecosystem, El Niño acts as a natural phenomenon that is partially representative of climate change impacts on marine bacteria at timescales relevant to microbial communities. Between 2014–2016, the North Pacific warm anomaly (a.k.a., the “blob”) and an El Niño event resulted in prolonged ocean warming in the Southern California Bight (SCB). To determine whether this “marine heatwave” resulted in shifts in microbial populations, we sequenced the rpo C1 gene from the biogeochemically important picocyanobacteria Prochlorococcus and Synechococcus at 434 time points from 2009–2018 in the MICRO time series at Newport Beach, CA. Across the time series, we observed an increase in the abundance of Prochlorococcus relative to Synechococcus as well as elevated frequencies of ecotypes commonly associated with low-nutrient and high-temperature conditions. The relationships between environmental and ecotype trends appeared to operate on differing temporal scales. In contrast to ecotype trends, most microdiverse populations were static and possibly reflect local habitat conditions. The only exceptions were microdiversity from Prochlorococcous HLI and Synechococcus Clade II that shifted in response to the 2015 El Niño event. Overall, Prochlorococcus and Synechococcus populations did not return to their pre-heatwave composition by the end of this study. This research demonstrates that extended warming in the SCB can result in persistent changes in key microbial populations.",
"introduction": "Introduction Ocean warming may be driving an areal increase of the globe’s oligotrophic gyres through increased stratification and weaker nutrient entrainment [ 1 , 2 ]. Marine bacterioplankton may be sensitive to this global ocean change. To date, ocean time series efforts have shown that marine bacteria are highly responsive to seasonal environmental cycles [ 3 – 6 ] and display a slow decay in community similarity on an annual to multi-annual scales [ 7 , 8 ]. Paleo-oceanographic evidence suggests that climatic tipping points can lead to relatively rapid shifts in plankton community composition, including cyanobacteria [ 9 ]. However, across systems, seasonal patterns in bacterial community composition have generally been resilient to discrete, or “pulse,” disturbances such as storms and short-term anthropogenic impacts [ 10 – 13 ]. Moreover, quantitative measurements of the response of marine bacteria to warming experiments in situ are rare [ 14 ], particularly in comparison to the range of warming experiments in soils and other ecosystems [ 15 ]. In contrast, a multitude of studies have shown that eukaryotic marine phytoplankton and zooplankton show significant changes in community structure in response to longer-term climatic warming oscillations on time scales from El Niño to the Atlantic Multidecadal Oscillation [ 16 – 18 ]. Thus, empirical observations of bacterioplankton responses to in situ ocean warming are both limited and critical for understanding how ocean ecosystems respond to climate change. Given the rapid generation times of marine bacteria, El Niño is a natural laboratory that can be used to understand bacterioplankton responses to multiannual-scale ocean warming and climate change impacts in the North Pacific Ocean. From 2014 through 2016, the Warm Anomaly and a significant El Niño event resulted in a prolonged “marine heatwave” across the Eastern North Pacific Ocean [ 19 , 20 ]. In the Southern California Bight (SCB), these events resulted in suppressed primary production but little to no change in C export [ 21 , 22 ]. Stratification and lower nutrient supply led to reduced particulate organic matter concentrations with concurrent high carbon:phosphorus (C:P) and nitrogen:phosphorus (N:P) ratios [ 23 ]. Previous studies of plankton communities in the SCB revealed increasing cyanobacteria cell abundance [ 5 ]. In addition, zooplankton populations showed significant changes in composition at the species level [ 24 ], suggesting that warming has a significant effect on planktonic communities in this region. Prior to the 2014–16 events, an analysis of hydrographic data (1984–2012) from the California Cooperative Oceanic Fisheries Investigations (CalCOFI) program revealed declining inorganic N:P ratios at coastal locations in the SCB [ 25 ]. Based on these observations, we predicted that prolonged warming, decreased nutrient availability, and shifting N:P ratios due to the 2014–2016 climate anomalies would also result in a shift in marine cyanobacteria populations. The highly abundant, widely distributed, and biogeochemically significant cyanobacteria Prochlorococcus and Synechococcus represent an important model system that can be used to determine the impact of changing environmental conditions on population-specific microbial diversity. Both genera are characterized by genomic traits associated with light, temperature, and nutrient optima that drive variable, clade-specific biogeographic patterns of abundance at high levels of phylogenetic similarity [ 26 – 30 ]. Thus, the phylogenetic diversity of these genera (i.e., ecotypes [ 31 ]) can be readily associated with subtle ocean changes due to our extensive knowledge of their respective trait distributions [ 32 ]. Previous time series studies of Synechococcus and Prochlorococcus have demonstrated stable interannual patterns of relative ecotype abundance as well as seasonal switching in either ecotype or sub-ecotype taxonomic patterns occurring at very rapid time scales [ 33 – 37 ]. In the SCB, cold-water ecotypes Synechococcus Clade I and Clade IV have been shown to dominate most of the year, with intermittent incursions of the warm-water ecotype Clade II [ 33 , 37 , 38 ]. Thus, we predict that (i) cyanobacteria will be sensitive to climatically-forced environmental changes and (ii) shifts in diversity at specific phylogenetic levels and their associated traits will reveal how bacterioplankton experience these environmental changes. Here, we use cyanobacterial populations to test whether microbial populations across differing levels of phylogenomic similarity show significant community changes in response to a prolonged warming period. To characterize microbial populations, we sequenced the highly variable cyanobacterial rpo C1 gene [ 39 ] from weekly-to-monthly samples between 2009 and 2018 at the Newport Pier MICRO time series. We specifically predicted that (i) the abundance of Prochlorococcus relative to Synechococcus would increase in response to the El Niño conditions, (ii) warming would result in an increase in the relative abundance of the high-temperature ecotypes of Prochlorococcus and Synechococcus , and (iii) changes in nutrient availability would result in systematic shifts in the composition of microdiverse populations over seasonal and annual time scales. This work has important implications for understanding how the link between phylogenetic trait distributions and microbial populations both dictates response to and reveals the impact of future anthropogenic ocean warming.",
"discussion": "Discussion Our data suggests that long-term warming in the Southern California Bight initiated a significant change in picocyanobacteria populations that persisted for 2–5 years ( Fig 1 ). In contrast, zooplankton communities typically return to pre-El Niño states within 1–2 years [ 24 ]. Although “alternative stable states” have been documented in macroecology [ 50 , 51 ], few studies have reported persistent changes in microbial communities [ 52 , 53 ]. Stable, long-term shifts in microbial composition have generally been driven by changes in conditions linked to deeply conserved traits such as shifts from iron-reduction to sulfate-reduction [ 54 ], oxic to anoxic conditions [ 55 ], or increased photoinhibition [ 56 ]. In the marine environment, multi-centennial changes in temperature and nutrient availability may have led to shifts in the abundance of N 2 -fixing cyanobacteria in the North Pacific Subtropical Gyre (NPSG) [ 9 ]. Moreover, a polarity reversal of the Pacific Decadal Oscillation in 1976 may have caused increased mixed layer stratification and shoaling, decreased nutrient availability, and a regime shift from a eukaryote- to a prokaryote-dominated system in the central Pacific Ocean [ 57 ]. In the MICRO time series, the marine heatwave of 2014–2016 and concurrent 2015 El Niño resulted in an increase in ecotypes associated with warmer conditions including Prochlorococcus HLI, LLI, and HLII as well as Synechococcus Clade II ( Fig 1 ). Moreover, the HLI and Clade II microdiverse haplotypes demonstrated persistent, altered community composition in response to the 2015 El Niño disturbance ( Fig 4A ). Here, similar processes as those observed in the NPSG may be operating concurrently to drive community composition changes in the SCB. Given the systematic and persistent changes in Cyanobacteria populations in the SCB, what are the ecosystem processes driving this shift in microbial communities? Both Prochlorococcus as a whole, which is more abundant in high temperature biomes than Synechococcus [ 58 ], and the Synechococcus high temperature Clade II increased in relative abundance across the entire time series ( Fig 1 and S2 Fig ). In addition, it is well-documented that 20°C represents a temperature threshold above which Prochlorococcus HLII outpaces HLI ecotype growth [ 27 , 59 ]. At MICRO, in situ temperatures were above 20°C for greater than 20.8% of dates in 2014, 2015, 2017, and 2018 and less than 14.2% of dates in 2010–2013 and 2016. Thus, the increased abundance of all Prochlorococcus , Prochlorococcus HLII, and Synechococcus Clade II may all reflect this warming. However, the Prochlorococcus / Synechococccus ratio as well as HLII relative abundance peaked in the winter ( Fig 1 and S2 Fig ). Additionally, the within-genus HLII monthly trends were negatively correlated with temperature ( Fig 2 ). These results may indicate an important role of regional warming and advection of both oligotrophic water and microbial communities into the SCB [ 5 , 23 ]. The 2015 El Niño event was one of the strongest on record in terms of broad-scale regional warming, even though reductions in upwelling and upwelling-favorable winds were relatively weak in comparison to previous, strong El Niño events such as 1982–1983 and 1997–1998 [ 48 ]. In addition, the SCB represents a transition zone between the southward California Current (CC) and the northward California Undercurrent (CUC). Thus, relatively small changes in source waters for the SCB may significantly change microbial populations. An increase in the northward CUC signal in SCB waters was observed between 1984–2012 [ 25 ]. The CC is characterized by cold, low-salinity, nutrient-poor “young” temperate water, whereas the CUC has high-salinity, nutrient-rich “old” equatorial waters [ 60 ]. Therefore, changes in water mass may result in the influx of oligotrophic communities with a higher abundance of Prochlorococcus cells from the HLII ecotype. Comparison of microdiverse HLII and Clade II populations at MICRO and in the NPSG supports the conclusion that these populations are phylogenomically linked ( Fig 5 ). We hypothesize that an increased percentage of high-temperature days locally but also a larger-scale regional warming trend that promotes the advection of oligotrophic populations into the SCB, contribute to a shifting picocyanobacteria community at MICRO. There are several important caveats to our conclusions including the application of multiple sequencing platforms and temporal co-variance between environmental factors. The sequencing platform was partially associated with year of sampling and thus confounded with the shift in Prochlorococcus/Synechococcus sequence read ratios. However, the relationship between the sequence read ratio and the cell count-ratio of Prochlorococcus/Synechococcus was not dependent on sequencing platform ( S3 Fig ). Moreover, out of the six instances of samples collected with 48 hours but sequenced on different platforms, only one date showed significantly different ecotype frequencies between 454 and Illumina ( S4 Fig ). Previous studies have similarly shown that the ratio of major marine microbial taxa is consistent across genomic platforms [ 61 ]. In addition, a large number of highly conserved SNPs in the rpo C1 consensus sequences were observed throughout the time series, regardless of sequencing platform ( Fig 3 ). Rapid changes in sequencing platforms and technology, both in terms of read length and depth of coverage, are likely to continue into the future. Our analysis suggests that conserved SNPs can be integrated across platforms. The strong seasonal co-variance of multiple environmental factors also makes it challenging to separate the effects of temperature and nutrient concentrations on picocyanobacteria populations. However, our multi-year sampling regime partially addressed this issue as temperature and nutrient concentrations were less correlated at this time-scale. Overall, temperature disturbances including the El Niño event, the partial interannual decoupling of environmental trends, and the prevalence of stable microdiversity across the 9 years of the MICRO time series has made it an excellent natural laboratory to study the effects of climatic shifts on microbial populations. The MICRO picocyanobacteria time series illustrates how we can “bi-directionally” link shifts in microbial diversity and environmental conditions to develop a deeper understanding of the impact of global changes. Past studies have revealed a rich understanding of the phylogenetic conservatism of response traits in microbial populations [ 62 – 64 ]. One of the main trait differences between the Prochlorococcus and Synechococcus genera is cell size, which results in a competitive advantage of Prochlorococcus under warm, low nutrient conditions [ 65 ]. Within the genera, it is also well-established that different light, temperature, and iron response traits are linked to specific ecotypes [ 27 – 29 , 66 ]. In contrast, responses to biotic interactions, subtle temperature shifts, and nutrient supply ratios have been associated with microdiverse clades [ 30 , 49 , 67 , 68 ]. Prochlorococcus and Synechococcus responses to environmental change at MICRO generally conformed to the phylogenomically predicted response at the genus and ecotype level ( Fig 6 ). Prochlorococcus likely increased relative to Synechcococcus across the time series due to traits conferring a competitive advantage in warm, oligotrophic conditions. Similarly, on annual scales, HLI, HLII, and Clade II likely increased over Synechococcus Clade I and Clade IV due to traits related to high temperatures and low nutrient concentrations [ 28 ]. However, several responses to environmental change did not conform to our knowledge of expected response traits. Prochlorococus HLII and Synechococcus Clade IV showed a seasonally negative and positive correlation with temperature, respectively, and thus a divergent trend compared to past studies. We attribute these unexpected observations to advection of genotypes from the broader eastern Pacific Ocean region that is also experienced warming. Overall, we saw limited changes at the microdiversity level, suggesting that the effect of shifts in environmental conditions or biotic interactions [ 37 , 49 , 69 ] were muted for most haplotypes here. However, some of these processes likely elicited the large responses in HLI and Clade II haplotypes observed in 2014–2015 ( Fig 4 ), which supports our conclusion of a persistent shift in picocyanobacteria populations. It is too early to state whether this shift is indicative of a new ecosystem state in terms of nutrients, temperature, and cyanobacteria ( Fig 6 ), but future time series efforts should continue to monitor climatic changes in the SCB. Global climate change is expected to have a complex impact on marine microorganisms as a result of nonlinear biophysical interactions between environmental conditions [ 70 , 71 ]. Shifts in microbial populations and their traits can act as a “biosensor” and allow us to develop a stronger understanding of how climatic changes affect ecosystem functioning. 10.1371/journal.pone.0238405.g006 Fig 6 Conceptual diagram of picocyanobacterial community shifts at the MICRO time series from 2009–2018. Synechococcus cells are outlined in purple and Prochlorococcus cells are outlined in green. Blue cell shading indicates cold temperature-adapted ecotypes and yellow cell shading indicates warm temperature-adapted ecotypes. The blue line indicates the nitrate trend (N) and the red line indicates the temperature trend (T)."
} | 4,206 |
25685260 | PMC4322651 | pmc | 9,668 | {
"abstract": "Root nodule bacteria are free-living soil bacteria, belonging to diverse genera within the Alphaproteobacteria and Betaproteobacteria , that have the capacity to form nitrogen-fixing symbioses with legumes. The symbiosis is specific and is governed by signaling molecules produced from both host and bacteria. Sequencing of several model RNB genomes has provided valuable insights into the genetic basis of symbiosis. However, the small number of sequenced RNB genomes available does not currently reflect the phylogenetic diversity of RNB, or the variety of mechanisms that lead to symbiosis in different legume hosts. This prevents a broad understanding of symbiotic interactions and the factors that govern the biogeography of host-microbe symbioses. Here, we outline a proposal to expand the number of sequenced RNB strains, which aims to capture this phylogenetic and biogeographic diversity. Through the Vavilov centers of diversity (Proposal ID: 231) and GEBA-RNB (Proposal ID: 882) projects we will sequence 107 RNB strains, isolated from diverse legume hosts in various geographic locations around the world. The nominated strains belong to nine of the 16 currently validly described RNB genera. They include 13 type strains, as well as elite inoculant strains of high commercial importance. These projects will strongly support systematic sequence-based studies of RNB and contribute to our understanding of the effects of biogeography on the evolution of different species of RNB, as well as the mechanisms that determine the specificity and effectiveness of nodulation and symbiotic nitrogen fixation by RNB with diverse legume hosts.",
"conclusion": "Conclusion The legume-RNB symbiosis is one of the best-studied associations between microbes and eukaryotes, due to the economic and ecological importance of symbiotic nitrogen fixation. Targeting RNB for sequencing on the basis of firstly, phylogenetic diversity and secondly, isolation from taxonomically distinct host legumes growing in diverse biomes offers significant benefits. Previous RNB sequencing projects have tended to focus on a narrow range of model organisms. By setting a goal of maximizing the phylogenetic diversity of sequenced RNB strains, these projects, in keeping with the other members of the GEBA family of projects, aid the development of a phylogenetically balanced genomic representation of the microbial tree of life and allow for the large-scale discovery of novel rhizobial genes and functions. The chosen RNB strains are available to the global research community and are stored in culture collections that are dedicated to long-term storage and distribution. A wealth of experimental data and metadata is available for each strain, which will inform analyses to identify genes and gene sets that correlate with rhizobial adaptation to diverse biomes, to the nodule environments found in taxonomically distinct legume hosts and to the effectiveness of nitrogen fixation within these nodules. Moreover, the legume-RNB symbiosis is an excellent model system to study plant-bacterial associations, including symbiotic signaling, cell differentiation and the mechanisms of endocytosis. The sequenced RNB genomes will not only provide a greater understanding of legume-RNB associations, but can be used to gain insights into the evolution of N2-fixing symbioses and microbe-eukaryote interactions.",
"introduction": "Introduction The importance of the research Legumes, with around 20,000 species and over 700 genera, are the third largest flowering plant family and are found on all continents (except Antarctica). They are major components of most of the world’s vegetation types and have important roles in agriculture as both pastures and pulses [ 1 , 2 ]. Most legumes are able to form dinitrogen-fixing symbioses with soil bacteria, collectively known as root nodule bacteria or rhizobia. RNB infection elicits the organogenesis of a unique structure, the nodule, which forms on the root (or less commonly, the stem) of the host plant. The mode of infection and the morphology and structure of the resulting nodule varies within the different legume tribes and has phylogenetic significance [ 3 , 4 ]. Following infection, RNB migrate to the nodule primordium, are endocytosed within the host cell and differentiate into N 2 -fixing bacteroids. The availability of utilizable nitrogen is the critical determinant for plant productivity. Legume-RNB symbiotic nitrogen fixation is a vital source of N in both natural and agricultural ecosystems. Based on different estimates, the total annual input of biologically fixed N ranges from 139 to 175 million tons, 35 to 44 million tons of which is attributed to RNB-legume associations growing on arable land, with those in permanent pastures accounting for another 45 million tons of N. N 2 -fixation by legume pastures and crops provides 65% of the N currently utilized in agricultural production [ 5 , 6 ]. The economic value of legumes on the farm is estimated at $30 billion annually, including $22 billion in the value of legume crops and $8 billion in the value of N 2 -fixation. Increasing the efficiency of the legume-RNB symbiosis has been projected to have an annual US benefit of $1,067 million, while transferring SNF technology to cereals and totally eliminating chemical N fertilization of the major crops will have an annual US benefit of $4,484 million [ 7 ]. Incorporating SNF in agricultural systems also reduces energy consumption, compared with systems that rely on chemical N-input. Every ton of manufactured N-fertilizer requires 873 m 3 of natural gas and ultimately releases ~2 tons of CO 2 into the air [ 8 ]. Furthermore, >50% of US N-fertilizer is imported, which further increases the energy cost of chemical N fertilizer. SNF has the potential to reduce the application of manufactured N-fertilizer by ~160 million tons pa, equating to a reduction of 270 million tons of coal or equivalent fossil fuel consumed in the production process. As well as energy cost savings, this reduces CO 2 greenhouse gas emissions. Legume- and forage-based rotations also reduce CO 2 emission by maintaining high levels of soil organic matter, thus enhancing both soil fertility and carbon storage in soil [ 9 ]. There are additional significant environmental costs to the use of N fertilizer: agriculturally based increases in reactive N are substantial and widespread, and lead to losses of biological diversity, compromised air and water quality, and threats to human health [ 10 ]. Microbial nitrification and denitrification of soil N are major contributors to emissions of the potent greenhouse gas and air pollutant, nitrous oxide, from agricultural soils [ 5 ]. Emission of N 2 O is in direct proportion to the amount of fertilizer applied. In addition, fertilizer N not recovered by the crop rapidly enters surface and groundwater pools, leading to drinking water contamination, and eutrophication and hypoxia in aquatic ecosystems [ 8 ]. The global increase in population is predicted to double demand for agricultural production by 2050 [ 11 ]. To meet this demand without incurring the high and unsustainable costs associated with the increased use of chemical N-fertilizer, the N 2 -fixing potential of the legume-RNB symbiosis must be maximized. Achieving this target will require a greater understanding of the molecular mechanisms that govern specificity and effectiveness of N 2 -fixation in diverse RNB-legume symbioses. Genome sequencing of RNB strains has revolutionized our understanding of the bacterial functional genomics that underpin symbiotic interactions and N 2 -fixation. However, previous RNB sequencing projects have not reflected the phylogenetic and biogeographic diversity of RNB or the variety of mechanisms that lead to symbiosis in different legume hosts. As a result, the insights gained into SNF have been limited to a small group of symbioses and there has not yet been a systematic effort to remedy this narrow focus. Here, we outline proposals for two sequencing projects to be undertaken at the DoE Joint Genome Institute that aim to expand the number of sequenced RNB strains in order to capture this phylogenetic and biogeographic diversity. Through the Vavilov centers of diversity (Proposal ID: 231) and GEBA-RNB (Proposal ID: 882) projects we will sequence 107 RNB strains isolated from diverse legume hosts in various geographic locations in over 30 countries around the world. The sequenced strains belong to nine of the 16 validly described RNB genera and have been isolated from 69 different legume species, representing 39 taxonomically diverse genera, growing in diverse biomes. These proposals will provide unprecedented perspectives on the evolution, ecology and biogeography of legume-RNB symbioses, as no rhizobial sequencing project so far has attempted to relate extensive genomic characterization of RNB strains to comprehensive metadata and thereby identify correlations between the genomes of rhizobial strains, their symbiotic associations with specific legume hosts, and the environmental parameters of their habitats."
} | 2,276 |
36552272 | PMC9775066 | pmc | 9,670 | {
"abstract": "Simple Summary To fulfill the food demand of the enormously growing population, different synthetic pesticides and fertilizers are used to grow crops. These synthetic products pose ill effects on humans and the environment. In recent times, the trend has shifted towards developing and utilizing bioproducts that are eco-friendly and sustainable to use in agriculture. They enhance productivity and restore equilibrium naturally in agroecological systems. In this regard, plant growth-promoting rhizobacteria (PGPR) facilitate crop production in multiple ways. This review deals with the limitations and challenges of conventional pesticides following the different microbes used as bioproducts along with how Bacillus is one of the promising PGPR used in sustainable agriculture. Bacillus spp. improves crop growth in both direct and indirect ways through nitrogen fixation, P and K solubilization, phytohormones production, quorum quenching, biofilm formation, and lytic enzymes production. Moreover, Bacillus spp. boost plant resistance towards the notorious phytopathogens. As Bacillus spp. is eco-friendly, promotes plant growth, confers resistance against diseases, improves soil fertility, non-toxic, naturally occurring microbe, and supports sustainable agriculture, there is a need to explore the potential of native Bacillus spp. and to use them in bioproduct development to support sustainable agriculture. Abstract Food security will be a substantial issue in the near future due to the expeditiously growing global population. The current trend in the agriculture industry entails the extravagant use of synthesized pesticides and fertilizers, making sustainability a difficult challenge. Land degradation, lower production, and vulnerability to both abiotic and biotic stresses are problems caused by the usage of these pesticides and fertilizers. The major goal of sustainable agriculture is to ameliorate productivity and reduce pests and disease prevalence to such a degree that prevents large-scale damage to crops. Agriculture is a composite interrelation among plants, microbes, and soil. Plant microbes play a major role in growth promotion and improve soil fertility as well. Bacillus spp. produces an extensive range of bio-chemicals that assist in plant disease control, promote plant development, and make them suitable for agricultural uses. Bacillus spp. support plant growth by N fixation, P and K solubilization, and phytohormone synthesis, in addition to being the most propitious biocontrol agent. Moreover, Bacilli excrete extracellular metabolites, including antibiotics, lytic enzymes, and siderophores, and demonstrate antagonistic activity against phytopathogens. Bacillus spp. boosts plant resistance toward pathogens by inducing systemic resistance (ISR). The most effective microbial insecticide against insects and pests in agriculture is Bacillus thuringiensis ( Bt ). Additionally, the incorporation of toxin genes in genetically modified crops increases resistance to insects and pests. There is a constant increase in the identified Bacillus species as potential biocontrol agents. Moreover, they have been involved in the biosynthesis of metallic nanoparticles. The main objective of this review article is to display the uses and application of Bacillus specie as a promising biopesticide in sustainable agriculture. Bacillus spp. strains that are antagonistic and promote plant yield attributes could be valuable in developing novel formulations to lead the way toward sustainable agriculture.",
"conclusion": "11. Conclusions Pesticides have been proven to be a promising agent to fulfill the food demand of the growing population. However, these hazardous pesticides have caused human health problems, development of pest resistance, narrowing of biodiversity, and environmental challenges, raising concerns about the pesticides’ safety. Thus, the need to reduce reliance on these synthetic pesticides is pertinent. The application of PGPR is an auspicious solution for eco-friendly agriculture. Bacillus spp. have been elucidated as growth promoters in sustainable agriculture through both direct and indirect mechanisms. The N 2 -fixation, P and K Solubilization, phytohormones production by Bacillus strains, moreover synthesis of antibiotics, production of lytic enzymes, and ISR are direct and indirect mechanisms, respectively, and all these action mechanisms of Bacillus are supportive in the growth promotion of plants, pest resistance, and circumventing of disease. Some of the Bacillus spp. have been documented as promising biocontrol agents. Food production and its accessibility always are an overwhelming priority to feed the world’s population. So, the best route is to be cautious about chemical-based pesticides. Biopesticides have long been attracting global attraction due to their safer strategy than conventional pesticides. Considering the importance of sustainable agriculture [ 173 , 174 , 175 , 176 ], Bacillus spp.-based bioproducts could be a promising addition to sustainable agriculture as there is a limited product range available. There is a dire need to explore the potential of Bacillus spp. in combination with other compatible microbial agents to increase PGP activity and quality food production.",
"introduction": "1. Introduction A range of plant diseases are caused by a variety of pathogenic microorganisms, such as fungi, bacteria, viruses, nematodes, and protozoa, which drastically lower agricultural production and cause significant yield losses [ 1 ]. Pathogenic diseases are responsible for between 20–40% of crop productivity losses [ 2 ]. Numerous methods have been employed to stop the spread of plant diseases, such as the use of pesticides, crop rotation, less susceptible crops, and other management techniques, but due to the resistance to pesticides and the persistence of soil-borne pathogens, their effectiveness is generally low [ 3 ]. Additionally, overusing chemically synthesized fertilizers has not only detrimental impacts on the biosphere but also affects the functioning of the ecosystem and diminishes the sustainability of agriculture [ 4 ]. Eco-friendly options for managing plant diseases and boosting crop yields are now being researched and advocated as part of an integrated crop management system—ICMS [ 5 ]. Biological control, which is a crucial component of ICMS, is described as the deployment of beneficial microorganisms to lessen the detrimental impacts of phytopathogens and encourage advantageous plant responses [ 6 ]. One of the most researched biocontrol agents, as biopesticides, is the Bacillus species, which inhibits phytopathogens by the mechanisms of competition and antagonism [ 7 ]. Various microorganisms, including Hypericum gramineum , Pseudomonas fluorescence , and Streptomyces species, have been identified as biocontrol agents [ 8 ]. Bacillus species have emerged as an important biological control agent because of their ability to produce antibiotics and tough and resistant endospores to control a range of phytopathogens [ 9 ]. Plant growth-promoting attributes have been reported in a variety of Bacillus spp., including B. velezensis , B. subtilis , B. macerans , B. circulans , B. azotofixans , B. coagulans , and others [ 10 ]. Phosphate solubilization, nitrogen fixation, production of siderophores, phytohormones, production of antimicrobial compounds, and systemically induced disease resistance are a few of the direct and indirect ways through which Bacillus spp. promote plant growth [ 11 ]. Antagonistically important species of the genus Bacillus are growing quickly. Abiotic stress resistance, rapid replication, and a broad spectrum of biocontrol capabilities are all characteristics of Bacillus spp. Volatile organic chemicals produced by B. subtilis are required for stimulating plant development and activating defense mechanisms in plants by boosting the induced systemic resistance (ISR) in plants [ 12 ]. Various crops, including tomato, potato, cucumber, maize, common [ 11 ] bean, soybean, sunflower, wheat, pepper, and many others, have shown positive impacts of Bacillus spp. on growth and crop yield [ 13 ]."
} | 2,052 |
33071611 | PMC7521031 | pmc | 9,671 | {
"abstract": "Plant-microbe interactions are both symbiotic and antagonistic, and the knowledge of both these interactions is equally important for the progress of agricultural practice and produce. This review gives an insight into the recent advances that have been made in the plant-microbe interaction study in the post-genomic era and the application of those for enhancing agricultural production. Adoption of next-generation sequencing (NGS) and marker assisted selection of resistant genes in plants, equipped with cloning and recombination techniques, has progressed the techniques for the development of resistant plant varieties by leaps and bounds. Genome-wide association studies (GWAS) of both plants and microbes have made the selection of desirable traits in plants and manipulation of the genomes of both plants and microbes effortless and less time-consuming. Stress tolerance in plants has been shown to be accentuated by association of certain microorganisms with the plant, the study and application of the same have helped develop stress-resistant varieties of crops. Beneficial microbes associated with plants are being extensively used for the development of microbial consortia that can be applied directly to the plants or the soil. Next-generation sequencing approaches have made it possible to identify the function of microbes associated in the plant microbiome that are both culturable and non-culturable, thus opening up new doors and possibilities for the use of these huge resources of microbes that can have a potential impact on agriculture.",
"conclusion": "Conclusion, Future Prospects and Challenges Beneficial microbial-plant interaction can lead to promising solutions for environmentally sustainable farming. Moreover, plant-microbe interaction has played a vital role in developing the biofertilizer, biocontrol and bioremediation agents in sustainable agriculture. Although there is plenty of literature on plant-microbe interaction, the molecular mechanism underlying genes function and signal transduction during beneficial and pathogenic interaction are lacking. Therefore, understanding the genetic basis of plant-microbe relationship with the next-generation sequencing technology along with various ‘omics’ technologies will be the emerging tool to provide extensive and in-depth knowledge on the biological phenomenon to improve plant health, disease control, improve food quality and enhance plant stress (both biotic and abiotic) management [ 115 , 116 ]. The near future faces many challenges in this area of research that need to be addressed for an integrated understanding of plant-pathogen interactions. These challenges are mainly, identification of key factors involved in such interaction during plant immune responses, detection and effective management of new emerging and re-emerging plant pathogens and development of pathogen-resistant crops. In the post-genomic era, understanding the mechanism of plant microbe-interaction could help mitigate these challenges, thus enhancing sustainable agriculture. Genomics tools mentioned in this review, such as GWAS, will continue to provide us novel disease resistance or defense-related genes that can be incorporated into crops through biotechnological approaches, which will become increasingly popular in the next few years and will further advance our understanding towards crop advancement.",
"introduction": "1 Introduction The population of the world in December 2019, as reported by the United Nations through the Worldometer, is 7.8 billion, and it is increasing exponentially. Therefore, there is a dire need to meet the increasing demand for food, with enhancement in agricultural practices. This needs to be done in harmony with the ecological balance, which, in turn, translates to a reduction in the use of chemical fertilizers and pesticides. Agricultural research thus needs to focus on alternative options to enhance food production. Molecular study of plant-microbe interactions can be a better alternative for sustainable agriculture [ 1 ]. Plants live together with different microorganisms that survive in the rhizosphere below the ground and above in the phyllosphere [ 2 , 3 ]. They are present within the plants as endophytes, as epiphytes attached to plant surface, and around the roots in the surrounding soil. These microorganisms may have positive, neutral or harmful effects on the health and development of plants [ 4 , 5 ]. The mechanism behind plant-microbe interaction is still not completely known and there are many questions, which need to be answered. These queries are about plant immune response, signaling pathway (in both plants and microorganisms), beneficial and harmful interactions between plants and microorganisms, etc . These queries will help us understand the whole mechanism of plant-microbe interaction and also help to identify those microorganisms which can be used in the near future to increase crop yield [ 6 ]. In agriculture, the association of microbes with plants serves as a catalyst to spontaneously improve yield [ 7 , 8 ]. Current farming activities, which rely heavily on the intensive use of high yielding agrochemicals, often cause environmental hazards [ 9 ]. The consequences of drastic global climate change, diminishing agricultural lands, rapid urbanization, and widespread use of agrochemicals, have had a devastating impact on crop production and ecology around the world, thus prioritizing the need for eco-friendly and sustainable development in agriculture [ 10 ]. An important strategy related to climate-smart agricultural practices is to harness the role of microorganisms in increasing plant nutrient quality and crop yield [ 11 ]. Beneficial plant-microbe interactions include Plant Growth Promotion (PGP), biotic and abiotic stress protection by plant immune system priming or activating plant defense mechanisms; variable ecosystem adaptation, mycorrhizal symbiosis, nutrient uptake, and plant-accessible transfer of inaccessible nutrient sources have been summarized previously [ 12 ]. Additionally, emerging plant diseases pose a serious threat to the world's agricultural industries, food safety, and plant species survival. Therefore, it is important to be able to quickly identify a new phytopathogen and explain- “What factor is responsible?, How it has evolved?, How do they interact with plant systems?”, to answer these questions, it is necessary to mine and annotate the genes involved in the plant-microbe associations from the genomes of both partners. In this regard, DNA and RNA, genomics data analysis, transcriptomics, metagenomics, metabolomics, NGS techniques, and proteomics methods have proved to be valuable tools for exploring plant-microbe interactions and their associations [ 13 , 14 ]. The key proteins, that are involved in the growth and development of plants, and stress tolerance (both biotic and abiotic), play a vital role in the maintenance of cellular functions in the plants by controlling physiological and biochemical pathways [ 15 , 16 ]. Recent research in the post-genomic era showed that modern “omics” technologies emerged as an important tool for the discovery of new genes responsible for encoding a functional protein, which will be helpful in many crop advancement programs [ 17 ]. In this present review, we have discussed the importance of plant-microbe interaction for crop improvement and stress management, focusing specifically on the advantages of NGS, along with Genome-Wide Association (GWA) mapping."
} | 1,874 |
25098921 | PMC4236595 | pmc | 9,672 | {
"abstract": "Background Selenium (Se) is an essential trace element in most organisms but has to be carefully handled since there is a thin line between beneficial and toxic concentrations. Many bacteria have the ability to reduce selenite (Se(IV)) and (or) selenate (Se(VI)) to red elemental selenium that is less toxic. Results A strictly aerobic bacterium, Comamonas testosteroni S44, previously isolated from metal(loid)-contaminated soil in southern China, reduced Se(IV) to red selenium nanoparticles (SeNPs) with sizes ranging from 100 to 200 nm. Both energy dispersive X-ray Spectroscopy (EDX or EDS) and EDS Elemental Mapping showed no element Se and SeNPs were produced inside cells whereas Se(IV) was reduced to red-colored selenium in the cytoplasmic fraction in presence of NADPH. Tungstate inhibited Se(VI) but not Se(IV) reduction, indicating the Se(IV)-reducing determinant does not contain molybdenum as co-factor. Strain S44 was resistant to multiple heavy and transition metal(loid)s such as Se(IV), As(III), Cu(II), and Cd(II) with minimal inhibitory concentrations (MIC) of 100 mM, 20 mM, 4 mM, and 0.5 mM, respectively. Disruption of iscR encoding a transcriptional regulator negatively impacted cellular growth and subsequent resistance to multiple heavy metal(loid)s. Conclusions C. testosteroni S44 could be very useful for bioremediation in heavy metal(loid) polluted soils due to the ability to both reduce toxic Se(VI) and Se(IV) to non-toxic Se (0) under aerobic conditions and to tolerate multiple heavy and transition metals. IscR appears to be an activator to regulate genes involved in resistance to heavy or transition metal(loid)s but not for genes responsible for Se(IV) reduction.",
"conclusion": "Conclusion A strict aerobic bacterium, C. testosteroni S44, reduced Se(VI) and Se(IV) to red SeNPs with sizes ranging from 100 to 200 nm. The cytoplasmic fraction strongly reduced Se(IV) to red-colored selenium in the presence of NADPH but no SeNPs were observed in cells. Possibly, Se(IV) was reduced in the cytoplasm and then transported out of the cell where the SeNPs were formed.",
"discussion": "Discussion C. testosteroni S44 reduced soluble Se(IV) into insoluble and thus non-toxic SeNPs outside of cells under aerobic condition as indicated by SEM/TEM-EDX and EDS Mapping analyses. It should thus be possible to synthesize SeNPs by imitating the biological process in industrial nanomaterial manufacturing [ 30 ]. Diseases caused by high content of Se in soils have been confirmed for the Chinese provinces Hubei and Shaanxi and Indian Punjab [ 1 ],[ 4 ]. In general, the variation of Se level in humans and animals are correlated to both Se excess and deficiency through the food chain [ 20 ]. Plants took up less water-soluble Se oxyanions from soil when bacteria reduced Se(IV) to organic Se and element selenium [ 31 ]. High levels of Se are commonly associated with concurrent contamination by other heavy and/or transition metals. Therefore, C. testosteroni S44 could be very useful for bioremediation of heavy metal(loid) polluted soils because it has adapted to a metal(loid)-contaminated environment. Considering the fact that only a partial reduction of Se(IV) to Se(0) could be achieved (Figure 2 ), it would be better in Se bioremediation if C. testosteroni S44 was applied to the contaminated site together with other more efficient Se(IV)-reducing bacteria. In some bacterial strains, elemental SeNPs were observed both inside and outside of cells [ 12 ],[ 21 ],[ 32 ],[ 33 ] whereas in other bacteria nanoparticles were only observed outside of cells [ 20 ]. We did not detect Se(IV) by HPLC-HG-AFS in cellular fractions (data not shown) although elemental Se less than 0.1 μM meets the demand of bacteria for synthesis of selenocysteine [ 34 ]. We could not observe SeNPs produced inside of cells at log phase and stationary phase by TEM, EDX and EDS Elemental Mapping (Figures 3 , 4 and Additional file 1 : Figure S1) although SeNPs were easily observed by TEM in many bacterial cells [ 12 ],[ 21 ],[ 32 ]. In contrast, we only observed a large number of SeNPs appearing outside of cells (Figure 1 ). The cytoplasmic fraction showed the strongest Se(IV) reducing ability contrasting with a weak reducing ability in periplasmic and membrane fractions after addition of NADPH (Figure 6 ) or NADH (Data not shown). Accordingly, the process of Se(IV) reduction appears to be an NADPH- or NADH-dependent pathway and indicates two possible pathways. One possibility is that Se(IV) did not enter the cytoplasm of strain S44 or only trace levels of Se(IV) were present in the cytoplasm. The Se(IV)-reducing determinant might have initially been assembled in the cytoplasm and then transferred across cytoplasmic and outer membrane. The Se(IV)-reducing determinant would then be only active outside of cells in vivo [ 21 ]. Another possibility, and more likely at that, is that Se(IV) was reduced to Se(0) in the cytoplasm and then Se(0) was pumped out of the cells where small SeNPs aggregated into bigger particles. In many cases, the big and smooth-surface nanoparticles occurred outside of cells [ 20 ],[ 21 ],[ 32 ]. Here, a large number of SeNPs ranging from 100–200 nm were observed by SEM (Figure 1 ) and further confirmed by EDX (Figure 3 A). In our experiment it was obvious that small selenium particles aggregated into bigger particles as observed by TEM (Figure 3 and Additional file 1 : Figure S1). This was different from previous TEM images of a homogeneous density of SeNPs [ 20 ],[ 21 ],[ 32 ]. In addition, this was not impacted by sample preparation because other strains produced big and homogeneous nanoparticles outside of cells using the same sample preparation and TEM observation technique (Data not shown). Previous studies confirmed small particles having low negative charges to have a propensity to come together and form aggregates [ 12 ]. In addition, proteins and/or other biomolecules such as polysaccharides and fatty acid may play a key role in controlling selenium nanoparticle size and the morphology of the resultant SeNPs [ 30 ]. The bulk of the Se(VI) and Se(IV) reduction to Se(0) was reported to occur on or outside the envelope [ 21 ]. This is very different from the reported mechanism where selenium was bound to the assembling protein SefA and then formed nanoparticles which were exported from cells [ 35 ]. In most reported cases, Se(VI) reduction occurred under anaerobic condition [ 36 ]-[ 38 ]. C. testosteroni S44 has a weak ability to reduce Se(VI) into red-colored selenium under aerobic condition (Figure 5 B). The Se(VI) reductase complex was identified as a periplasmic Mo-containing enzyme in T. selenatis [ 38 ],[ 39 ] and B. selenatarsenatis [ 40 ]. The Se(VI)-reducing determinant of C. testosteroni S44 also is most likely a Mo-enzyme because tungstate inhibited Se(VI) reduction (Figure 5 B). In contrast, the Se(IV)-reducing determinant did not appear to contain Mo because tungstate did not inhibit Se(IV) reduction. Accordingly, Se(VI) reduction is a distinct activity different from Se(IV) reduction. Iron-sulfur (Fe-S) clusters are cofactors for many proteins across all three domains of life. Fe-S proteins function in a number of cellular processes, including electron transfer, gene regulation, photosynthesis and nitrogen fixation, anti-oxidative and iron stress among others [ 28 ],[ 29 ],[ 41 ]. The genomic organization of iscRSUA-hscBA-fdx , the operon encoding the housekeeping Fe-S biogenesis system (Isc), is conserved in many β- and γ-proteobacteria [ 27 ]. IscR (Isc regulator) regulates expression of the Isc pathway by modulating intracellular iron homeostasis via a negative feedback mechanism based on the cellular Fe-S demand in P. aeruginosa and E. coli [ 42 ],[ 43 ] and can also increase the expression of another operon, sufABCDSE, involved in synthesis of Fe-S clusters in E. coli [ 28 ],[ 29 ],[ 41 ]. IscR is part of the large Rrf2 family of winged helix-turn-helix (wHTH) transcription factors [ 44 ]. We could not find a suf operon on the genome of C. testosteroni S44, this is similar to genome of Pseudomonas spp. that is also lacking a suf operon [ 43 ]. As a result, only iscRSUA-hscBA-fdx encoding proteins are used for Fe-S cluster synthesis in C. testosteroni S44. In addition, IscR is a global regulator that regulates functions not only involved in Fe-S biogenesis but also directly or indirectly controlling the expression of ~40 genes in E. coli [ 28 ],[ 41 ]. Recently, it was shown that the highly conserved three cysteine residues (Cys92, Cys98, and Cys104) and His107 of IscR were essential for [2Fe-2S] cluster ligation [ 45 ]. [2Fe-2S]-IscR binds both type 1 and type 2 motifs from hya promoter, thereby exhibiting metal-dependent regulation of DNA binding specific for IscR [ 46 ]. The corresponding cluster ligands are Cys92, Cys98, Cys105 and His108 in IscR from C. testosteroni S44. The insertion sites of Tn5 mutants, iscR-280 and iscR-327, were close to bases encoding those four ligands. Moreover, the insertion site of iscR-327 was located next to the bases encoding His108 located at residues forming a helix involved in dimerization (residues 103–123 in E. coli ) of IscR [ 46 ], therefore disturbing the formation of IscR dimers. In contrast, the insertion site of iscR-513 is located at the tail end of iscR (537 bp full length) and the insertion site in iscS + 30 is located at the gap between iscR and iscS (Figure 7 ). As a result, the formation and function of IscR were more strongly disturbed in iscR-280 and especially in iscR-327, resulting in slower growth and less resistance than iscR-513 to heavy metal(loid)s (Figures 7 and 8 ). The insertional mutants iscR-513 and iscS + 30 would still produce a functional IscR regulator (albeit truncated at the C-terminus in iscR-513) but expression of subsequent genes of the operon would be significantly lower due to polar effects of an insertion by transposon Tn5. Those results are consistent with the result of a ∆iscR mutant that was 40- to 50-fold less resistant to organic hydroperoxides (tBOOH and CuOOH) in P. aeruginosa [ 43 ]. Therefore, IscR aids cellular growth and resistance to heavy metals not only by regulating expression of the iscSUA-hscBA-fdx operon, but probably also by directly or indirectly regulating expression of other genes [ 28 ] in C. testosteroni S44. C. testosteroni S44 was isolated from an antimony mine and contained resistance determinants to various metal(loid)s [ 26 ]. Due to a large number of genes encoding putative metal(loid) resistance proteins [ 26 ], C. testosteroni S44 is thought to be able to quickly pump heavy or transition metals and metalloids out of the cell or transform them into a less toxic species thereby becoming very resistant. This interpretation is consistent with the high MIC for Se(IV) and the postulated quick Se(0) secretion from the cytoplasm across the cell envelope to the outside of cells. Although C. testosteroni S44 was resistant to high level of heavy metals, it did not reduce Se(IV) efficiently. It is therefore possible C. testosteroni S44 evolved a balanced state between resistance of Se oxyanions and reduction (detoxification)."
} | 2,818 |
30853879 | PMC6395404 | pmc | 9,673 | {
"abstract": "In this work, we perform analysis of detection and counting of cars using a low-power IBM TrueNorth Neurosynaptic System. For our evaluation we looked at a publicly-available dataset that has overhead imagery of cars with context present in the image. The trained neural network for image analysis was deployed on the NS16e system using IBM's EEDN training framework. Through multiple experiments we identify the architectural bottlenecks present in TrueNorth system that does not let us deploy large neural network structures. Following these experiments we propose changes to CNN model to circumvent these architectural bottlenecks. The results of these evaluations have been compared with caffe-based implementations of standard neural networks that were deployed on a Titan-X GPU. Results showed that TrueNorth can detect cars from the dataset with 97.60% accuracy and can be used to accurately count the number of cars in the image with 69.04% accuracy. The car detection accuracy and car count (–/+ 2 error margin) accuracy are comparable to high-precision neural networks like AlexNet, GoogLeNet, and ResCeption, but show a manifold improvement in power consumption.",
"introduction": "1. Introduction Neural networks today are achieving state-of-the-art performance in competitions across a range of fields. Recent advances in deep learning (LeCun et al., 2015 ) have motivated the development of neural hardware substrates that are tailored to implementing deep networks with extremely low power and efficiency for a variety of embedded systems applications. Hardware that mimics the computational capabilities of a human brain through spiking neural networks has been shown to be not only extremely energy-efficient, but also capable of scaling up to large neural networks. Examples include the IBM TrueNorth Neurosynaptic System (Merolla et al., 2014 ), SpiNNaker (Furber et al., 2014 ), and the BrainScaleS project (Schemmel et al., 2008 ), all of which mimic the computational behavior of spiking neurons and can also be used to deploy deep neural networks. One of the major challenges that these spiking neural network-based platforms faced was deploying convolutional neural networks (CNNs) on spiking neurons. This issue was addressed in the recent work from Cao et al. ( 2015 ) and Esser et al. ( 2016 ), and Eta Compute (Moore, 2018 ). The authors in Esser et al. ( 2016 ) have proposed an algorithm named energy-efficient deep neuromorphic networks (EEDN) to map CNNs on TrueNorth. EEDN networks achieved at or near state of the art accuracy when compared with traditional 32-bit precision neural networks on standard benchmarks and they operated at a much higher throughput (Frames Per Second) per watt. These promising results show potential for deploying spiking neural network based platforms for a variety of applications where battery life and power consumption are primary concerns. Such applications include video surveillance, UAV surveillance, aerial image analysis, etc. Prior work such as Esser et al. ( 2015 , 2016 ), Wen et al. ( 2016 ), Rueckauer et al. ( 2017 ), and Sengupta et al. ( 2018 ) have discussed about how to efficiently train neural network models so that the inference neural network can be easily mapped onto low precision hardware such as TrueNorth without any loss in output accuracy. But these prior works have only done the evaluations against small object recognition datasets such as MNIST, CIFAR-10, and CIFAR-100. Prior work never listed out the challenges that might occur when mapping large CNN or DNN structures on TrueNorth for bigger datasets with large annotated images. For bigger datasets resource limitations and the CNN model limitations that TrueNorth can support start becoming a bottleneck. In this paper we evaluate the challenges related to deployment of EEDN trained neural network on TrueNorth hardware. Discussions that have been reported in this article are meant to complement the opportunities and challenges for spiking neural network hardware that have been reported in Pfeiffer and Pfeil ( 2018 ). The evaluations have been done against publicly-available dataset of overhead aerial images of cars that was proposed by Mundhenk et al. ( 2016 ) (Henceforth referred as COWC dataset). Examples from COWC dataset have been shown in Figure 1 . As the neural network structures start becoming more complex, we have to keep in mind limited number of TrueNorth (Henceforth referred as TN) cores that are available and design a neural network structure so that we can obtain benefits by using hardware substrates more judiciously. This paper presents design decisions that a developer would have to make to design a neural network for the TrueNorth NS16e system (Sawada et al., 2016 ) that is shown in Figure 2 . The goal of this work is to present how knowledge of hardware architecture affects the decisions and parameter choices made while training and deploying neural networks on TrueNorth. These observations can assist us in maximizing the benefits of TrueNorth's available hardware computational resources. Figure 1 Sample images from COWC dataset (Mundhenk et al., 2016 ). Images are 192-by-192 pixels. For detection, (A,B) , the model's goal is to detect whether a car is present in the center 48-by-48 pixels or not. Even though there are cars present in (B) , the label has been set to false because there is no car in the center 48-by-48 pixels of the image. For the counting task, (C) , the goal is to count the exact number of cars present in an image. The example shown in the figure has the label value “13,” since there are 13 cars in the image. Figure 2 (A) NS16e hardware system that was developed by IBM (Image from Shah, 2016 ). (B) Single neurosynaptic core which forms the computational block of the TrueNorth chips with the details presented in Cassidy et al. ( 2013 ) and Nere ( 2013 ). Contributions of the research proposed in this paper are: Evaluate TrueNorth deployed CNNs for counting and detection tasks on COWC dataset (Mundhenk et al., 2016 ). Resources consumed by AlexNet (Krizhevsky et al., 2012 ) and VGG-16 (Simonyan and Zisserman, 2014 ) neural networks when deployed on NS16e hardware (Sawada et al., 2016 ). Identifying the architectural bottlenecks of these CNN structures and proposed changes to the CNN structure so that it could be deployed on NS16e hardware. Analysis of change in resource consumption and output accuracy based on the prior works such as, network-in-network structure (Lin et al., 2013 ), MobileNets (Howard et al., 2017 ), and YOLO (Redmon and Farhadi, 2016 ) neural network models. Discussions presented in section 4 outline the opportunities that are present in SNN hardware that can address the challenges present in TrueNorth architecture and EEDN training algorithm.",
"discussion": "2.2.3.6. Discussion on fully convolutional neural network of VGG-16 As presented in section 2.2.2, one of the challenges that users might face when mapping standard neural network structures onto TrueNorth is that currently the proposed hardware architecture does not support convolutional layer to fully connected layer connections. Similar to modified AlexNet model, while mapping VGG-16 onto TrueNorth, the CNN features are downsampled all the way down to a one-by-one convolution using strided convolutions. The downsampling has been performed by having a convolutional layer that has convolution window of size 7 x 7 pixels and a stride of 7, (as shown in Figure 8A ) or by having a convolutional layer that has convolution window of size 6 x 6 pixels and a stride of 6 (as shown in Figure 8B ).\n\n4. Discussion 4.1. Summary In this paper we described four design decisions that a designer would have to address to deploy CNN structures on a neurosynaptic system such as IBM TrueNorth. These decisions are very important if the goal is to perform tasks such as detection and counting in a hardware constrained environment. Section 2.2 introduced the need to have a systematic approach for proposing neural network designs that can be mapped onto TrueNorth. Here we discussed how we can leverage prior work that have been proposed for CNN design and extend those ideas to EEDN based CNN models for TrueNorth. We showed that if a standard VGG-16 CNN model is modified systematically, while keeping in mind the architectural bottlenecks that are present in NS16e, hardware resource requirements can be reduced by 3x (refer to Figures 9 , 10 ). Similarly, we discussed in Table 1 that with systematic approach to mapping CNNs on TrueNorth, the accuracy could be improved by 8% for detection based task and by 20% for counting based task when compared to having a naive ternary-weight AlexNet implementation on NS16e. Results presented in Table 2 show that EEDN trained neural network can have similar accuracy as full precision AlexNet. It is important for us to consider how many TN cores are performing relevant computations. The analysis presented in Figure 14 shows that it is extremely important for users to consider the trade-off between the hardware resources that is available for mapping the neural network, and the input image size and feature counts of initial layers, to achieve the desired test accuracy. Section 3.4 analyzes the cost of the deployed neural network on TN hardware. As per the results presented in Table 3 , the EEDN-trained neural network when deployed on TN hardware has test accuracy that is comparable to high-precision neural networks like AlexNet, GoogLeNet, and ResCeption, but shows a manifold improvement in FPS per watt. 4.2. Extending This Work to Other Benchmarks and Neuromorphic Chips As neuromorphic computing is becoming more promising, it is important for researchers to understand the challenges that came up in TrueNorth architecture/algorithm and address these issues in future neuromorphic computing architectures/algorithms. First, it is important for us to have a new set of benchmarks and datasets that can be used to evaluate neuromorphic hardware for bigger CNN models or that require us to estimate continuous numbers such as regression problems. There have been benchmarks that were proposed keeping in mind SNN algorithms, viz., N-MNIST (Orchard et al., 2015 ) and CIFAR-10 DVS (Li et al., 2017 ), but both of these benchmarks have very small image sizes and both of these benchmarks can solved using classification models. Problems that require us to estimate continuous numbers bring out the architectural limitations that might arise if the goal is to predict large range of numbers. On the other hand, benchmarks from domains such as Micro-Aerial Vehicles (Ma et al., 2013 ) and video surveillance would be very interesting for the SNN community because these small drones already have SNN controllers in them (Clawson et al., 2016 ). Having video surveillance dataset from MAVs, will help us realize potential of SNNs to be deployed in energy-constrained environments. Evaluating the hardware with bigger CNN models will help us understand the architectural limitations that are present in the hardware and it will also motivate researchers to investigate better algorithms for hardware/software co-design on neural networks. Second, it is critical to investigate the fan-out limitations of architectures such as TrueNorth, so that neural networks can also support connections between convolutional and fully-connected layers. Even though there have been prior research that have proposed algorithms to train inception neural networks or residual networks for SNN hardware (Rueckauer et al., 2017 ; Sengupta et al., 2018 ), the current architectural limitations related to fan-out in SNN hardware such as TrueNorth, do not support such skip connection based CNNs. Concurrently, CNN structures such as MobileNets (Howard et al., 2017 ) have shown to significantly reduce the memory accesses and computations for embedded platforms. To the best of author's knowledge, currently there is no research that has successfully trained ternary quantized model for depthwise separable filters, which is a critical part of MobileNets. Prior work done in Holesovsky and Maki ( 2018 ) have attempted to train a depthwise separable CNN with ternary weights and activation, but reported a significant drop in accuracy when compared to the same CNN structure that was trained with single precision weights and activation. Third, it is important to address the architecture bottlenecks present between the CPU/FPGA hybrid system and the neuromorphic chips, otherwise, a considerable amount of computation resources may end up getting used up to handle these interactions, as shown in CNN baseline example of Figure 14 . Another direction that researchers can potentially investigate is improving the speed of deployed neural networks by analyzing the bottleneck present during inter-chip communication on a scaled-up hardware such as NS16e system. Finally, as neural network models become deeper and wider, there will be a considerable amount of communication happening between neurons mapped onto different chips. This bottleneck could be addressed by having a better placement algorithm for multi-chip placement which would constrain group neurons that communicate a lot with each other to a single chip, unlike the work proposed in Akopyan et al. ( 2015 ) where the goal of the placement algorithm is to minimize the wire-length of placed neurons. Or, researchers can propose a new interconnect architecture for inter-chip communication that could handle high backpressure of spikes that get delivered from one neuromorphic chip to another. Pruning may not always be the best approach to address hardware constraints while DNN training. As presented in Yazdani et al. ( 2018 ) even though pruning may give correct test accuracy, the inference confidence score reduces significantly. Researchers from hardware community have proposed pruning algorithms to reduce the size of bigger CNNs for hardware deployment (Han et al., 2015 ; Iandola et al., 2016 ). At present EEDN trained CNN models are highly sparse due to ternary weight representation, having more aggressive, such as pruning away TN cores for deep learning model, pruning technique may result in further drop in test accuracy. Therefore, rethinking the placement strategy for deep learning models on SNN may be an important step forward to address the issue of hardware constraints."
} | 3,606 |
28378285 | PMC5380647 | pmc | 9,674 | {
"abstract": "Studies proved that addition of nitrate in rumen could lead to reduction of methane emission. The mechanism of this function was involved in the competition effect of nitrate on hydrogen consumption and the inhibitory effect of generated nitrite on methanogen proliferation. The present study investigated an alternative mechanism that denitrifying anaerobic methane oxidizing (DAMO) bacteria, DAMO archaea and anammox bacteria may co-exist in rumen, therefore, more methane can be oxidized when addition of nitrate. Ruminal batch culture model was used to test the effects of addition of 5 mM NaNO 3 , 4 mM NH 4 Cl, or both into the culture substrate on methane production, fermentation patterns, and population of methanogens, NC10 and anaerobic methanotrophic-2d (ANME-2d). Our results showed that NC10 in the ruminal culture was detected by polymerase chain reaction (PCR) when using NC10 special primer sets, and addition of nitrate reduced methane production and the relative proportions of methanogen, whereas increased the relative proportion of NC10. A combined addition of ammonia salt and nitrate did not show further inhibitory effect on methane production but accelerated nitrate removal. We did not detect DAMO archaea in ruminal culture by real-time PCR when using ANME-2d special primer sets. The present study may encourage researchers to pay more attention to methane oxidation performed by anaerobic methanotroph when studying the strategies of inhibiting ruminal methane emission. Electronic supplementary material The online version of this article (doi:10.1186/s13568-017-0377-2) contains supplementary material, which is available to authorized users.",
"introduction": "Introduction There are increasing evidences showing the significant inhibitory effect of nitrate on methane production in vitro and methane emission from the rumen in vivo (Zhou et al. 2012 ; Patra and Yu 2014 ; Elzaiat et al. 2014 ; Olijhoek et al. 2016 ). An acceptable mechanism of this effect is that the reduction of nitrate or nitrite consumes hydrogen, which reduces hydrogen available for methane formation in the rumen (Ao and Emeritus 2008 ; Nolan et al. 2010 ). Nitrite is the intermediate product in nitrate reduction process. A detrimental problem of dietary supplementation of nitrate to ruminants is that nitrite could be accumulated in the rumen and then absorbed into the circulating system through the rumen wall, and results in a change of hemoglobin to methemoglobin (Sar et al. 2004 ), which is incapable of carrying oxygen and causes mild to severe methaemoglobinaemia. The toxic character of nitrite to ruminants leads to the controversy of nitrate supplementation, even though there are increasing proofs that nitrite would not accumulate to the toxic level in the rumen of sheep acclimated to nitrate supplementation. Guo et al. ( 2009 ) reported that the percentage of nitrogen to the total gas production was markedly increased when addition of nitrate into the ruminal fermentation system. In another study, the product of ammonia after an administration of nitrate was much lower than expected (Tillman et al. 1965 ), which was unlikely to be the major cause of the disappearance of nitrate (Ao and Emeritus 2008 ), indicating that there should be other metabolic pathways of nitrate in rumen. Research in the field of environmental protection has shown that nitrate can be reduced to nitrite by ANME-2d, and nitrite can be subsequently converted to nitrogen by ‘ M. oxyfera ’, which belongs to NC10 phylum. To date, NC10 bacteria are the only known bacteria that can anaerobically oxidize methane; all other anaerobic methanotrophs are archaea (He et al. 2016 ). ANME-2d and ‘ M. oxyfera ’ are two anaerobic methanotrophs and exist ubiquitously in the anaerobic environment abundant in methane, such as freshwater sludge, landfill and wastewater (Wu et al. 2012 ). The denitrification process of nitrate and nitrite coupled to methane oxidation is named as denitrifying anaerobic methane oxidation (DAMO), and believed as a potential process to remove nitrate and nitrite in wastewater even the DAMO rate is very low in vitro culture. It had been reported that DAMO can be extremely accelerated in an anaerobic ammonium oxidation reactor because nitrite produced from nitrate reduction can be jointly reduced to nitrogen when ammonium is oxidized to nitrogen by anammox (Chen et al. 2015 ). Therefore, nitrite produced from nitrate has two pathways to be converted to nitrogen in the mixture reactor of both DAMO and anaerobic ammonium oxidation when provided with nitrate, ammonium salt and methane simultaneously, and nitrite reduced from nitrate in the mixed reactor can be cleaned more rapidly than in the reactor with DAMO system alone. Actually, anammox also ubiquitously exists. The combined pathways of DAMO and anaerobic ammonium oxidation in the sludge of freshwater were proposed by Haroon et al. ( 2013 ). The ruminal microorganisms can adapt to addition of nitrate (Alaboudi and Jones 1986 ; Zhou et al. 2012 ), anaerobic oxidation of methane has been detected in the rumen (Valdés et al. 1996 ; Kajikawa et al. 2003 ) and an inclusion of nitrate in ruminal fermentation in vitro decreases methane production while increases nitrogen production (Guo et al. 2009 ). These facts together led to a generation of our research hypothesis that DAMO bacteria, DAMO archaea and anammox may exist in rumen. The present study tested this hypothesis by investigating the effect of addition of both nitrate and NH 4 Cl on methane production and nitrite accumulation compared with an addition of either nitrate or NH 4 Cl in in vitro ruminal fermentation, meanwhile anammox, DAMO bacteria and archaea were also detected by PCR.",
"discussion": "Discussion We found in the current experiment that an addition of 5 mM nitrate, alone or with ammonium salt, decreased methane production in the ruminal culture and changed the fermentative pattern with an increase of acetate percentage, but decrease of butyrate and propionate percentages. The total VFA production and the degradability of DMD were unaffected by treatments. The effect of nitrate on methane production in ruminal culture in the present study was in line with other reports (Zhou et al. 2012 ; Patra and Yu 2014 ; Elzaiat et al. 2014 ; Olijhoek et al. 2016 ). The inhibitory effect of nitrate on methane production in rumen is believed via a mechanism that the reduction process of nitrate and nitrite to NH 4 competes with CO 2 for hydrogen (Zijderveld et al. 2010 ; Patra and Yu 2014 ). The increased NH 4 concentration in response to addition of nitrate in our study supported this mechanism. Another mechanism of the inhibitory effect of nitrate on methane production is the toxic function of nitrite (the first intermediate of nitrate reduction) on methanogens (Božic et al. 2009 ; Zhou et al. 2011 ) and could be confirmed by the result in the current study that the relative population of methanogens in total bacteria was significantly decreased by nitrate addition. However, we noticed that the nitrite concentration after 24 h incubation was unchanged in response to nitrate addition. It was reported that nitrate/nitrite in the rumen fluid was detectable at early stage with an administration of nitrate into the rumen, but undetectable after about 8 h (Wang et al. 1961 ). Therefore, it was suspected that nitrite concentration in the ruminal culture might be increased at the beginning of the experiment as a result of addition of nitrate, but the differences in nitrite concentration among the treatments diminished after 24 h incubation because nitrite could be further reduced. If so, the inhibition of nitrite on methanogen proliferation might occur at early stage, but this inhibition had a sustained reductive effect on the number of methanogens in the ruminal culture. It was noticed in this study that nitrate was accumulated in response to nitrate addition in the ruminal culture, and the nitrate concentration in nitrate group was 1.2 mM higher than that in control group after 24 h incubation. This result is inconsistent with previous in vivo study (Wang et al. 1961 ). Ao and Emeritus ( 2008 ) proposed the possible reasons for rapid clearance of nitrate/nitrite in rumen, including absorption of nitrate into the host’s blood. The reason why there is a difference in the change of nitrate concentration between the present study and other in vivo experiments is probably because that there was no absorptive pathway for nitrate in an in vitro culture system. However, when nitrate was added with ammonium salt together, no increase in nitrate concentration was observed and indicated that more nitrate was removed compared with addition of nitrate alone. Alaboudi and Jones ( 1986 ) reported that some rumen bacteria could acclimate to nitrate addition. In addition, there were evidences proving that methane oxidation was coupled to nitrate or nitrite reduction by DAMO archaea and bacteria in sludge of lake or river (Haroon et al. 2013 ; Chen et al. 2015 ), and the removal of nitrite could be accelerated after DAMO was coupled to the anammox bacterial system (Chen et al. 2015 ). The bacteria that can reduce nitrite to nitrogen in this mixed system include NC10 and anammox bacteria (Hu et al. 2015 ). Rumen is a methane enrichment habitat and is connected with the external environment through the digestive tract, and DAMO bacteria and anammox bacteria may also exist in rumen. The detectable NC10 in ruminal culture in this study partly supported this hypothesis. The fact that addition of nitrate increased the NC10 population indicated that NC10 was one of the bacteria that can acclimate to nitrate addition. A combination of nitrate and ammonium salt in the present study resulted in more nitrate removal but no more methane reduction. A possible explanation was that anammox bacteria in the rumen culture could consume more nitrite as a result of addition of ammonia and hence accelerated nitrate reduction, but the nitrite reduction pathway performed by anammox didn’t increase the oxidized methane, therefore, there is no change in methane production compared with addition of nitrate alone. However, anammox bacteria in ruminal culture were not detected by common or real-time PCR in the current study. But Candidatus Kuenenia, an identified anammox bacterium, was found in rumen fluid when the ruminal bacterial flora was investigated by high through-put sequencing of the 16s rRNA gene (unpublished). It’s worth noting that Candidatus Kuenenia is the anammox species present in the enrichment culture of fresh water sediment (Haroon et al. 2013 ). The reason for no detectable anammox by anammox special primers PCR in this study was possibly because of its low abundance in rumen fluid. Actually, only ten sequences among the sequenced 27, 736 PCR products belong to Candidatus Kuenenia (Additional file 1 ). The fact that ANME-2d was not detected by PCR when using ANME-2d special primers in our study suggested that ANME-2d, the DAMO archaea, may not exist in rumen fluid. In conclusion, NC10 could be detected in rumen fluid by NC10 special primers PCR. The present study indicated addition of nitrate in the culture reduced methane production not only by inhibiting methanogens but also by stimulating NC10 population. A combination of nitrate and ammonia salt had no further inhibitory effect on methane production, but could accelerate nitrate removal in vitro. ANME-2d wasn’t detectable in rumen fluid by ANME-2d special primers PCR."
} | 2,891 |
27278828 | PMC4899690 | pmc | 9,675 | {
"abstract": "The realization of wearable electronic devices with extremely thin and flexible form factors has been a major technological challenge. While substrates typically limit the thickness of thin-film electronic devices, they are usually necessary for their fabrication and functionality. Here we report on ultra-thin organic transistors and integrated circuits using device components whose substrates that have been removed. The fabricated organic circuits with total device thicknesses down to 350 nm have electrical performance levels close to those fabricated on conventional flexible substrates. Moreover, they exhibit excellent mechanical robustness, whereby their static and dynamic electrical characteristics do not change even under 50% compressive strain. Tests using systematically applied compressive strains reveal that these free-standing organic transistors possess anisotropic mechanical stability, and a strain model for a multilayer stack can be used to describe the strain in this sort of ultra-thin device. These results show the feasibility of ultimate-thin organic electronic devices using free-standing constructions.",
"conclusion": "Conclusion The fabrication technology described here removes the substrate layer from organic electronic devices to make them ultra-thin and ultra-flexible with total thicknesses of less than one micron. The anisotropic dependency of the devices on the strain directions indicates that these devices require a rigorous strain model that takes in to account the thickness and Young’s modulus of each layer in order to calculate the mechanical strain applied to them. The rigid gold layer may be an important factor affecting the mechanical stability of the devices. Replacement of rigid metal materials with soft materials having smaller Young’s moduli, such as conducting polymers, would improve the mechanical flexibility of these ultra-thin organic electronic devices. Furthermore, such ultra-thin electronics can be stacked on top of one another, which could change the device integration strategy from that of the conventional lateral layout."
} | 519 |
38466879 | PMC10990143 | pmc | 9,676 | {
"abstract": "Significance Deforestation alters aboveground biodiversity and ecosystem services worldwide. Yet, the impacts of deforestation on soil biodiversity, and its associated ecosystem services, remain virtually unknown. Our global synthesis indicates that deforestation of native forest impacts soil biodiversity and the capacity to support ecosystem services. Conversion of native forests to managed ecosystems resulted in soils with reduced capacity to support soil-borne plant pathogen regulation, plant–soil symbiosis, carbon storage, nutrient cycling, and organic matter decomposition. Soil biodiversity and functions were most negatively affected when native forests were converted to cropland and in warmer and wetter ecosystems. Our work highlights the fundamental importance of avoiding soil degradation caused by deforestation to conserve soils and the services they provide for the next generations.",
"conclusion": "Conclusion In brief, our findings are of high significance because they provide consistent evidence that deforestation of native forest threatens global soil biodiversity and its capacity to provide ecosystem services. The conversion of native forests to managed ecosystems consistently resulted in soils with higher bacterial diversity and more homogeneous fungal communities dominated by pathogens, and these changes are associated with reductions in functions that support important ecosystem services, including carbon storage, nutrient cycling, and organic matter decomposition ( Fig. 6 ). Soil biodiversity and functions are most negatively affected when native forests are converted to croplands and in warmer and wetter ecosystems. When considering abiotic factors, the changes of microbial diversity and fungal guilds in response to deforestation were mainly influenced by soil pH and total P content. Deforestation of native forests, driven by agricultural expansion and land intensification, continues to occur at a rapid pace, especially in tropical regions and developing countries ( 3 , 7 ). Thus, governments and decision-makers should develop and follow conservation strategies to avoid soil degradation caused by deforestation of native forests. Meanwhile, restoring biodiversity and soil ecosystem services in managed ecosystems is also an important strategy to alleviate the conflicts between human and nature brought about by deforestation and achieve global Sustainable Development Goals ( 8 , 31 ). Fig. 6. Conceptual model illustrating the impacts of native forest conversion to other land use types on soil properties, microbial community, and functions.",
"discussion": "Results and Discussion Deforestation Led to Critical Reductions in Soil Ecosystem Services. Our work revealed that the conversion from native forests to plantations, grasslands, and croplands has critical impacts on soil properties and results in the reduction of key ecosystem services, including soil C storage and nutrient cycling ( Fig. 2 A and SI Appendix , Fig. S3 ). Generally, deforestation caused major declines (30% on average across sites) in soil organic C ( Fig. 2 A ). This soil organic C loss was substantial (≈24%) when forests were converted to tree plantations or grasslands. These grasslands are, of course, relatively young grasslands which may have a reduced capacity to capture carbon in their soils compared with older and well-developed grasslands ( 14 , 20 ). Importantly, loss of soil C was especially strong after forest conversion to croplands, which reduced by 48%. Decreases in net primary productivity with forest conversion are the most important reason for soil organic C losses due to reductions in C input from plant material ( 6 , 21 ). The increase of soil erosion with the decrease of plant cover in managed ecosystems also accelerates soil C loss ( 1 , 22 ). The decline in soil organic C content after forest conversion was equally strong across all biomes, suggesting that deforestation impact on soil C sequestration is a global problem. Fig. 2. Effects of native forest conversion on ( A ) soil properties/nutrient cycling (biogeochemistry) and ( B ) proxies of soil organic matter decomposition. Values are effect size ±95% CI. F-Plant, F-Gras, and F-Crop represent native forest conversion to plantation, to grassland, and to cropland, respectively. TropF, SubTroF, and TempF represent tropical forest, subtropical forest, and temperate forest, respectively. The sample size in each category is given at the Left . The closed symbols indicate significant effects, and the open symbols indicate nonsignificant effects. The difference between categories is significant if P < 0.05. The decreasing trend of soil total nitrogen (N) after forest conversion was also substantial (23%) ( Fig. 2 A ). Deforestation resulted in a substantial decline in soil C:N, with particularly strong reductions when forest was converted to cropland ( Fig. 2 A ). Thus, native forest conversion led to a greater decline in soil C storage than in soil N, which is likely related to the input of N fertilizers in managed ecosystems and the decrease of lignified and recalcitrant C in litter caused by the shift from woody plants to grasses after deforestation ( 12 , 13 ). The reduction of C:N ratio with deforestation was also associated with a major shift in the structure and functions of the soil microbial community, increasing the growth of fast-turnover bacteria, but negatively affecting taxa capable of degrading complex organic compounds ( Figs. 2 and 3 ) ( 13 , 21 ). Fig. 3. Effects of native forest conversion on soil biodiversity: ( A ) microbial diversity and ( B ) fungal guilds. Values are mean effect size ± 95% CI. F-Plant, F-Gras, and F-Crop represent native forest conversion to plantation, to grassland, and to cropland, respectively. TropF, SubTroF, and TempF represent tropical forest, subtropical forest, and temperate forest, respectively. The sample size in each category is given at the Left . The closed symbols indicate significant effects, and the open symbols indicate nonsignificant effects. The difference between categories is significant if P < 0.05. Native forest conversion to managed systems also increased soil pH, especially when converted to croplands and grasslands, and in tropical biomes ( SI Appendix , Fig. S3 ). Meanwhile, soil phosphorus (P) content increased with deforestation ( Fig. 2 A and SI Appendix , Fig. S3 ), which is mainly due to the intensive use of P fertilizers in managed systems. A reduction in soil acidity, as observed in our analysis, is known to release insoluble-P from mineral complexes, and is another likely mechanism for the increase of soil available P after deforestation ( 23 ). In general, reductions in soil organic C storage, changes in soil properties and fertilizer inputs after forest conversion to managed ecosystems lead to changes in soil C:N:P ( 22 ), which are linked to the impacts on microbial diversity and ecosystem functions, such as plant–soil symbiosis and pathogen control. Native forest conversion to managed systems also limited biologically driven processes involved in soil organic matter decomposition. The functional rates related to organic matter decomposition and C, N and P mineralization, including soil microbial respiration, β-D-glucosidase, phenoloxidase, invertase, N-acetylglucosaminidase, urease, and phosphatase activities, all significantly decreased after deforestation ( Fig. 2 B ). Moreover, native forest conversion to plantation limited the rate of soil functioning to a similar extent as does conversion to grassland and cropland. This result extends and validates previous assessments that compositionally simpler plantations used for wood and nonwood products are much less effective than complex native forests in maintaining soil ecosystem services ( 1 ). It is worth noting that, the reduction of C inputs from litterfall, rhizodeposition, and fine root turnover due to decreased ecosystem productivity after deforestation would also presumably support less microbial taxa to decompose organic matter. A previous study has shown a 28% decline in litterfall input from native forest to plantation alone ( 21 ) but did not examine the effects of more land use–intensive conversions to cropland or grassland. Thus, the declines in both soil C content and functional capacity related to organic matter decomposition of managed ecosystems indicate that the degradation of soil ecosystem functions after deforestation is difficult to restore. Environmental Context Associated with Forest Conversion Type and Changes in Soil Abiotic Factors Explained Shifts in Microbial Communities Following Deforestation. Native forests conversion to managed ecosystems also had critical impacts on soil microbial diversity ( Fig. 3 and SI Appendix , Figs. S4–S6 ). Environmental context linked with forest conversion type and associated changes in abiotic factors explained the influence of deforestation on soil microbial structure and functions ( SI Appendix , Figs. S5 and S6 ). First, our analyses revealed that deforestation resulted in a significant increase in bacterial diversity ( Fig. 3 A and SI Appendix , Fig. S4 ), suggesting that managed ecosystems following anthropogenic disturbance support the growth of bacteria with rapid turnover ( 12 , 14 ). We further showed that conversion of native forests to croplands and in tropical biomes resulted in the greatest increase in bacterial richness ( Fig. 3 A ). Using multiple linear regression and model selection, we found that these responses were associated with parallel changes in key abiotic factors such as an increase in soil pH ( Fig. 4 A and SI Appendix , Fig. S5 and Table S2 ). In fact, soil pH was the most important predictor of soil bacterial richness after deforestation based on the sum of Akaike weights ( SI Appendix , Fig. S5 A ). The increase of soil pH is caused by decreased vegetation biomass and bedrock rejuvenation, but also by the agricultural practice of liming, which is common for croplands, plantations, and to some extent in grasslands. Increasing soil pH may be especially important during the deforestation of conifers, which are known to acidify soils. Most bacterial species prefer neutral soils and have a narrow range of pH adaptation ( 24 ), and thus their diversity increases with pH after forest conversion. Moreover, intensive management practices (e.g., tillage, fertilization, and irrigation) used in croplands are also known to increase bacterial richness ( 12 , 14 ). Fig. 4. Mixed effects meta-regression analyses for the relationships between response ratios (RRs) of key soil properties and microbial diversity. RR is calculated from the natural logarithm-transformed ratio of treatment (converted ecosystem) to control (native forest). The gray area represents the 95% CI. Panels A – C represent regressions between different microbial and soil attributes. The effects of forest conversion on fungal richness varied with the conversion types ( Fig. 3 A ). Native forests converted to grasslands had a less negative effect on fungal richness than to other land use types. This can be attributed to the fact that grasslands are mostly used for grazing and typically have high vegetation diversity relative to other conversion types. Also, manures produced by livestock, which function as organic fertilizer, can expand the functional niche for fungi ( 25 ), thereby enhancing their richness. It has been reported that P-deficient soils in natural ecosystems often support the growth of microbial taxa with the ability to decompose organic matter and obtain P at the expense of other microorganisms possessing different functions ( 26 , 27 ). The change of fungal richness after native forest conversion was positively associated with soil total P ( Fig. 4 B and SI Appendix , Fig. S5 B and Table S2 ), indicating that the mitigation of P-deficiency in managed systems alleviates its restriction for microbial taxa with diverse functions. However, this does not inherently represent a positive phenomenon in terms of ecosystem function, as higher fungal richness caused by increased soil P content after deforestation is more likely to increase the diversity of pathogens rather than beneficial taxa, with implications for the functioning of these ecosystems ( 17 ). Our results also showed that deforestation led to a decline in fungal community dissimilarity, although the extent depended on forest conversion type ( Fig. 3 A and SI Appendix , Fig. S5 B ). Both fungal and bacterial community dissimilarities were lower in croplands compared to native forests ( Fig. 3 A ), indicating that agricultural intervention led to homogenization of soil microbial communities ( 14 ). The homogeneity of communities in managed ecosystems is likely a result of the loss of endemic microorganisms from native forests and/or an increase in the ranges of existing taxa, which alters the delivery of soil ecosystem services ( 6 , 14 ). In general, after forest conversion, ecosystems were dominated by bacteria and more homogeneous fungal communities challenging the conservation of native soils globally. Effects of native forest conversion on soil biodiversity also translated into important taxonomic changes. The bacterial communities of managed ecosystems included a larger proportion of Bacteroidetes, Gemmatimonadetes, Firmicutes, Chloroflexi, and Nitrospriae but had a reduced abundance of Proteobacteria, Acidobacteria, and Verrucomicrobia ( SI Appendix , Fig. S7 ). The fungal community after deforestation supported a larger proportion of Ascomycota at the expense of decreasing Basidiomycota abundance ( SI Appendix , Fig. S8 ). These shifts of microbial taxa after native forest conversion were also related to the increase in soil pH and total P ( SI Appendix , Fig. S9 ). Changes of community composition can also help explain differences in soil ecosystem functions and services. For example, the decline of Proteobacteria with deforestation, which contains a variety of beneficial bacteria that promote plant growth and protection against diseases, could negatively affect ecosystem productivity ( 28 ). Our study also revealed that climate is an important factor influencing the response of soils to the forest conversion ( SI Appendix , Fig. S6 and Table S3 ). Mean annual temperature (MAT) and aridity index (AI) were negatively correlated with the change of fungal richness, but positively correlated with the change of bacterial richness in response to forest conversion ( SI Appendix , Fig. S6 ). These changes were related to the decline in fungal richness, and the increase in bacterial richness in response to forest conversion, which were most pronounced in tropical biomes ( Fig. 3 A ). These results indicated that deforestation has a greater negative effect on fungal diversity in warmer and wetter ecosystems, but supports the growth of fast-turnover bacteria, thereby contributing to a shift from fungi-dominated to bacteria-dominated microbial communities ( 12 ). Fungi exhibited lower richness and higher community dissimilarity in warmer and wetter native ecosystems, which accelerates the loss of fungal endemic species after deforestation ( 6 , 8 , 13 ). In contrast, the survivability for bacteria experiencing higher temperatures, greater precipitation, and environmental disturbance suggests that they are much more likely to dominate managed ecosystems in warmer and wetter areas ( 13 ). Moreover, MAT was negatively correlated with the response of organic matter decomposition, including N-acetylglucosaminidase (chitin degradation), urease (urea hydrolysis), and phosphatase (P mineralization) to deforestation ( SI Appendix , Fig. S6 ). This suggested that climate can largely regulate the response of soil functions to deforestation. Deforestation Promotes Soil Fungal Plant Pathogens and Reduces Fungal Symbionts. Our analyses provided further evidence that conversion from native forests to managed systems negatively impacted the capacity of ecosystems to support plant–soil symbiosis, with a significant decline in the proportion of symbiotic fungi ( Fig. 3 B ) and increase in the proportion of soil-borne plant fungal pathogens. This shift of fungal guilds from symbiont-dominated (e.g., plant–soil symbiosis service) to soil-borne plant pathogen-dominated (e.g., less plant pathogen control) was influenced by land use, with the most pronounced shift occurring after conversion to croplands ( Fig. 3 B and SI Appendix , Fig. S5 C ). The decrease of vegetation diversity after deforestation, especially when converted to croplands, where monocultures prevail, results in the losses of symbiotic fungal species that have strong host specificity and limited functional breadth ( 15 ). In contrast, the increased density of specific hosts in managed systems may promote the colonization and accumulation of host-dependent fungal pathogens ( 16 , 17 ). The increase in the abundance of soil-borne plant pathogens strongly increases the risk of host-specific disease, which is detrimental to ecosystem health and limits productivity, especially in croplands ( 16 ). Considering soil properties, P input in managed ecosystems is the most important factor that negatively affects plant–soil symbiosis ( Fig. 4 C and SI Appendix , Fig. S5 C ). This outcome may be attributed to high P availability, which decreases plant reliance on symbiotic fungi and increases direct uptake of P by plant roots, resulting in a weakened symbiotic relationship that reduces ecosystem stability and stress resistance ( 27 ). We also observed an increase in the abundance of saprotrophs (e.g., decomposers) under native forest conversion ( Fig. 3 B ). There is a competitive relationship between decomposers and symbiotic fungi, especially in native forest ecosystems where ectomycorrhizal fungi with decomposition ability dominate the community ( 29 ). Thus, the weakening of plant–soil symbiosis after deforestation contributes to a relative increase in decomposers. Nevertheless, the increase in decomposer abundance under deforestation does not imply an increase in decomposing capacity of organic matter, because the saprophytic taxa supported by disturbed and nutrient-rich environments may not have a greater decomposition ability than those in native forest ecosystems ( 30 ). Taken together, these results reveal that native forest conversion to managed ecosystems, in particular to croplands, weakens the capacity for plant–soil symbiosis and increases the abundance of pathogens, which poses a long term threat to ecosystem health and functioning. Linking Soil Biodiversity and Functions Following Forest Conversion. Through mixed-effects linear regression analysis, we found that changes in soil biodiversity following native forest conversion to managed ecosystems influenced the response of soil functional rates related to organic matter decomposition ( Fig. 5 and SI Appendix , Fig. S10 ). The effect of deforestation on soil organic matter decomposition was determined by evaluating the changes in eight microbial ecosystem functions (i.e., linked with the decomposition of soil organic matter) using a random-effect model in each paired-site observation (native forest vs. deforestation) ( Fig. 2 B ) ( 18 ). We show that the response of soil organic matter decomposition to deforestation was negatively correlated with fungal richness ( Fig. 5 A ), and that functions associated with chitin degradation (N-acetylglucosaminidase), urea hydrolysis (urease) and P mineralization (phosphatase) were negatively related to bacterial richness ( SI Appendix , Table S4 ). This is likely attributed to dramatic changes of microbial community structure after deforestation, specifically through the loss of critical functional taxa ( 14 , 18 ). For example, the response of organic matter decomposition to deforestation was positively correlated with the abundance of symbiotic fungi and negatively correlated with the abundance of fungal pathogens ( Fig. 5 B and C ). This indicated that the shift in fungal guilds from symbiont-dominated to pathogen-dominated following deforestation reduced the functional rate of soil organic matter decomposition. The weakening of plant–soil symbiosis reduces the ability of key fungal taxa to secrete extracellular enzymes to decompose organic matter or acquire N and P to promote plant growth. In contrast, the accumulation of pathogens threatens ecosystem productivity and ultimately fed back to the decline of soil organic matter content and ecosystem functions ( 15 , 16 ). These results suggest that changes in microbial diversity and the loss of key taxa after native forest conversion negatively affect organic matter decomposition, leading to more abiotic-driven soil with reduced functionality. Fig. 5. Mixed effects meta-regression analyses for the relationships between the response ratios (RRs) of microbial diversity and soil organic matter decomposition after native forest conversion. RR is calculated from the natural logarithm-transformed ratio of treatment (converted ecosystem) to control (native forest). RR of OM decomposition is the overall RR of eight functions associated with organic matter decomposition in each observation. OM, organic matter. The gray area represents the 95% CI.Panels A – C represent regressions between different microbial and soil attributes. Spatial and Temporal Influence of Forest Conversion on Soil Biodiversity and Ecosystem Services. We then attempted to better understand the influence of land use conversion age on soil biodiversity and functionality. To such an end, we categorized the available information in our dataset into three land use age ranges: ≤10 y, 10 to 30 y, and ≥30 y. The results revealed that the observed microbial community shift, from fungal-dominated to bacterial-dominated communities with a decline of plant–soil symbiosis, decreased with the increase of stand age, and recovered to levels similar to native forest after long-term (≥30 y) conversion ( SI Appendix , Fig. S11 ). However, soil organic C content and C:N were lower throughout the entire land use age range after conversion ( SI Appendix , Fig. S12 ). The homogenization of fungal communities and the shift of fungal guilds from symbiotic-dominated to saprophytic- and pathogenic-dominated were exaggerated with the increase of land use age from native forest to cropland ( SI Appendix , Fig. S11 ). In the medium to long term (>10 y) of converted cropland, the proportion of fungal pathogens was five times higher than that in native forest. The decrease of soil C:N and the accumulation of P and potassium also increased with the increase of land use age of cropland ( SI Appendix , Fig. S12 ). These results suggested that intensive management of cropland has far greater adverse effects on soil functioning and health in the long term after conversion than other land use types. Finally, we would like to acknowledge the limitations of our study. It is known that the space for time substitution method has some confounding factors caused by environmental variation across sampling sites that could affect the validity of the results. To mitigate these potential variables in our global analysis, the literature included in the meta-dataset were field experiments that maintained a strict paired-site design. Managed ecosystems and native forest control plots were always adjacent to each other with the same climate (MAT and AI) and soil conditions (of the same soil classification). While there is potential for some factors to differ between native forests and converted ecosystems, such as slope and aspect, their impact on soil microbial communities relative to those resulting from drastic changes in land use is presumed to be relatively small."
} | 5,957 |
32653776 | null | s2 | 9,677 | {
"abstract": "The biogeography of the mammalian intestine is remarkable in that a vast microbial consortium exists inside the organism, surrounded by intestinal epithelial cells. The microbiome and the intestinal epithelium have developed a complex network of interactions that maintain intestinal homeostasis. We now recognize that functions of the epithelium are compartmentalized in specific intestinal epithelial cell subtypes. Furthermore, we are beginning to understand the ways in which microbes and their metabolic products impact the specific epithelial subsets. Here, we survey the mechanisms utilized by the microbiome to regulate intestinal epithelial function, and inversely, how different epithelial cell subtypes cooperate in regulating the microbiome."
} | 188 |
34581467 | null | s2 | 9,678 | {
"abstract": "Many bacteria can migrate from a free-living, planktonic state to an attached, biofilm existence. One factor regulating this transition in the facultative plant pathogen Agrobacterium tumefaciens is the ExoR-ChvG-ChvI system. Periplasmic ExoR regulates the activity of the ChvG-ChvI two-component system in response to environmental stress, most notably low pH. ChvI impacts hundreds of genes, including those required for type VI secretion, virulence, biofilm formation, and flagellar motility. Previous studies revealed that activated ChvG-ChvI represses expression of most of class II and class III flagellar biogenesis genes, but not the master motility regulator genes visN, visR, and rem. In this study, we characterized the integration of the ExoR-ChvG-ChvI and VisNR-Rem pathways. We isolated motile suppressors of the non-motile ΔexoR mutant and thereby identified the previously unannotated mirA gene encoding a 76 amino acid protein. We report that the MirA protein interacts directly with the Rem DNA-binding domain, sequestering Rem and preventing motility gene activation. The ChvG-ChvI pathway activates mirA expression and elevated mirA is sufficient to block motility. This study reveals how the ExoR-ChvG-ChvI pathway prevents flagellar motility in A. tumefaciens. MirA is also conserved among other members of the Rhizobiales suggesting similar mechanisms of motility regulation."
} | 349 |
38564528 | PMC10986988 | pmc | 9,685 | {
"abstract": "Bacterial communities directly influence ecological processes in the ocean, and depth has a major influence due to the changeover in primary energy sources between the sunlit photic zone and dark ocean. Here, we examine the abundance and diversity of bacteria in Monterey Bay depth profiles collected from the surface to just above the sediments (e.g., 2000 m). Bacterial abundance in these Pacific Ocean samples decreased by >1 order of magnitude, from 1.22 ±0.69 ×10 6 cells ml -1 in the variable photic zone to 1.44 ± 0.25 ×10 5 and 6.71 ± 1.23 ×10 4 cells ml -1 in the mesopelagic and bathypelagic, respectively. V1-V2 16S rRNA gene profiling showed diversity increased sharply between the photic and mesopelagic zones. Weighted Gene Correlation Network Analysis clustered co-occurring bacterial amplicon sequence variants (ASVs) into seven subnetwork modules, of which five strongly correlated with depth-related factors. Within surface-associated modules there was a clear distinction between a ‘copiotrophic’ module, correlating with chlorophyll and dominated by e.g., Flavobacteriales and Rhodobacteraceae, and an ‘oligotrophic’ module dominated by diverse Oceanospirillales (such as uncultured JL-ETNP-Y6, SAR86) and Pelagibacterales. Phylogenetic reconstructions of Pelagibacterales and SAR324 using full-length 16S rRNA gene data revealed several additional subclades, expanding known microdiversity within these abundant lineages, including new Pelagibacterales subclades Ia.B, Id, and IIc, which comprised 4–10% of amplicons depending on the subclade and depth zone. SAR324 and Oceanospirillales dominated in the mesopelagic, with SAR324 clade II exhibiting its highest relative abundances (17±4%) in the lower mesopelagic (300–750 m). The two newly-identified SAR324 clades showed highest relative abundances in the photic zone (clade III), while clade IV was extremely low in relative abundance, but present across dark ocean depths. Hierarchical clustering placed microbial communities from 900 m samples with those from the bathypelagic, where Marinimicrobia was distinctively relatively abundant. The patterns resolved herein, through high resolution and statistical replication, establish baselines for marine bacterial abundance and taxonomic distributions across the Monterey Bay water column, against which future change can be assessed.",
"conclusion": "Conclusions Our in-depth descriptions, identification of significant statistical WGCNA-based modules, and evolutionary analyses revealed patterns in vertical distributions of bacterial groups throughout the water column in Monterey Bay. Together with decades of biological oceanographical measurements [ 52 ], the data herein provide a baseline from which to study long-term changes to the overall health, nutrient cycling, and ecological change in the Monterey Bay. The microbial community composition is regulated by a combination of the co-occurrence of other microbes and the physical and chemical environment. Our phylogenetic reconstructions demonstrate evolutionary diversity that is not fully distinguished by the V1-V2 ASV analysis within the highly abundant, well-defined Pelagibacterales, as well as SAR324 bacteria, for which the ecological ramifications await discovery. Overall, these results emphasize the importance of identifying evolutionary diversity and how it might connect to ecological differences. The results further stress that the organisms represented as being most abundant by amplicon analyses are largely unrepresented in culture collections, and hence have still unknown ecologies. Our study, alongside ongoing and future efforts, provides a path for understanding the niches of these organisms, and eventually implications of their distinctive distributions and how they relate to ecosystem function.",
"introduction": "Introduction Patterns in marine bacterial community structure are observable spatially (e.g., latitude, longitude, depth) and temporally [ 1 – 4 ]. Depth and co-associated parameters are major factors in microbial community variations occurring between the surface and deep ocean [ 5 , 6 ]. A multi-depth view provides the opportunity to identify differences in community structure, potentially highlighting the diversity associated with distinct ecological niches influenced by the environmental conditions at each depth. Pelagibacterales (commonly referred to as SAR11) is an alphaproteobacterial lineage for which extensive characterization is available. Pelagibacterales cells are heterotrophic and have limited metabolic flexibility because of genome streamlining and subsequent loss of adaptive genes [ 7 , 8 ], requiring pyruvates, amino acids (or their respective precursors) as well as hydroxymethyl-2-methylpyrimidine for growth [ 9 , 10 ]. Pelagibacterales are highly abundant in the oceanic water column, comprising on average about one-third of bacterial cells and up to 50% at times in surface waters. Their presence extends to depths below the photic zone, where they can make up to 25% of bacterial cells [ 11 , 12 ]. Distinct ecotypes or clades exhibit specific annual cycles in depth-related distributions that correlate with seasonal mixing, blooms, and stratification in the photic versus upper-mesopelagic, defined in that study as 0–120 m and 160–300 m, respectively [ 13 – 15 ]. One particular Pelagibacterales clade, referred to as SAR11 clade II, is considered abundant and diverse on sinking particles at deeper depth, while other SAR11 clades and subclades exhibit differences in relative abundance over geographical regions and seasons [ 13 , 16 – 19 ]. Environmentally-mediated selection is considered crucial in shaping the biogeography of SAR11 subclades [ 20 ]. Like Pelagibacterales, SAR324 [ 21 ]—formally also known as Marine Group B [ 22 ] and part of Deltaproteobacteria, but now considered their own phylum [ 23 ]—are considered ubiquitous in the ocean, and particularly abundant below the photic zone [ 24 – 26 ]. This uncultivated group likely represents organisms with varied genomic content, corresponding potentially to considerable metabolic flexibility, and ecological niches. Some SAR324 genomes contain ribulose-1, 5-bisphosphate carboxylase-oxygenase and sulfur oxidation genes, suggesting the potential for autotrophic CO 2 fixation coupled with reduced sulfur compound oxidation [ 27 – 29 ]. SAR324 also appears capable of photoheterotrophy and alkane oxidation especially in surface layers, capabilities that potentially underpin the ubiquity of SAR324 as a whole [ 29 , 30 ]. Understanding the distributions of different groups within SAR324, paired with their apparent metabolic capacity, will allow identification and hypotheses on the ecological niches occupied by SAR324 taxa. Here, we examine the distribution of bacterial taxa and transitions in diversity throughout the Monterey Bay water column in the eastern North Pacific. Specifically, we sampled depths from 2 to 1,978 m during three expeditions conducted over two seasons, during two years, and investigated bacterial communities using V1-V2 16S rRNA gene amplicon analyses and flow cytometry. Our molecular analyses of the bacterial community used a two-part approach combining a Weighted Gene Correlation Network Analysis (WGCNA) of ASVs with phylogenetic analyses of two of the primary bacterial groups observed (Pelagibacterales and SAR324). These analyses advance our understanding of how different clades within bacterial groups are distributed throughout the water column and reveal variations in marine microbial community composition over space and time in a Pacific marine canyon ecosystem.",
"discussion": "Results and discussion Oceanographic conditions Expeditions were performed in Monterey Bay, located in the eastern North Pacific and encompassing a submarine canyon system with the deepest sites in the outer bay and shallowest in the inner canyon zone ( Fig 1 ). Profiles from stations sampled more than once were similar between years, apart from the photic zone ( S1 Table , S1 Fig ). The latter is known to be highly dynamic, with varying influences of both the eastern boundary current and upwelling extent [ 52 , 53 ]. During our study, the photic zone ranged from between 25 m (Stn. 1978m; May 2016) and 150 m (Stn. 1820m; Sept 2015), and subsurface chlorophyll maxima were generally not observed, indicating well-mixed surface waters ( Fig 1C , S1 Table ). High variability was also reflected in chlorophyll- a concentrations, which ranged from a minimum 0.84 mg m -3 to maximum 13.78 mg m -3 in surface waters between years ( S1 Table ), similar to prior studies [ 32 , 52 ]. Water temperatures ranged from 12.13 to 17.26°C at the surface, and from 2.05°C (Stn. 1978m) to 6.09°C (600 m; Stn. 633m) near the seafloor ( Fig 1C , S1 Table ). Stations with bottom depths >1500 m were consistently colder near the seafloor (2.05–2.38°C) than the others (3.72–6.09°C). Salinity ranged from 33.3 to 34.6, increasing with depth, while oxygen decreased ( S1 Table ). Phosphate, silicate, and nitrate concentrations generally increased with depth but exhibited different gradient patterns. Photic zone phosphate concentrations ranged from 0.36 to 2.45 μM; however, regardless of station bottom depth, concentrations near the seafloor ranged from 2.65 to 3.63 μM across all samplings and were generally between 2.27 and 3.94 μM in the dark ocean ( S1 Table ). Silicate increased by over two orders of magnitude between the surface and the deepest sites, with the steepest increase in the upper mesopelagic ranges ( S1 Table ). A subsurface nitrite maximum was observed that shifted in depth, with the shallowest being 15 m (Stn. 1018m; Sept 2016) and deepest 70 m (Stn. 633m; Sept 2015) ( Fig 1C and S1 Table ). Together these oceanographic conditions indicate that the study area is broadly representative of relatively deep waters of the eastern North Pacific and its adjacent upwelling areas, and indeed of eastern boundary upwelling ecosystems in general [ 54 ]. Sampling and analysis of the entire water column, and similar profiles (especially in the deep samples), across two time-points sets the stage for understanding how these conditions impact microbial communities. These collections capture communities associated with relatively variable productivity at the surface, apparently more stable, deep waters, and variable transitional depths in between. Specifically, the state of relatively high phytoplankton productivity and production of labile organic carbon in the surface layers fuels rapid growth of heterotrophic bacteria. As this material sinks through the water column, organic carbon is consumed, while in parallel temperature and light decrease (light decreasing to zero below the photic zone). At the same time inorganic nutrients increase due to remineralization processes, and at the deepest depths bacteria with metabolisms suited for cold, low and/or recalcitrant organic carbon flourish [ 55 ]. To date, the degree to which these communities vary between broadly similar profiles in such an eastern boundary current system is poorly characterized, especially relative to systems such as the sub-tropical Atlantic and tropical central Pacific [ 56 ]. Bacterial cell abundance To establish bacterial abundances, both cyanobacteria ( Prochlorococcus and Synechococcus ) and heterotrophic bacteria (i.e., non-pigmented) were enumerated using either natural autofluorescence characteristics or DNA staining, respectively, as well as scatter properties as ascertained by flow cytometry ( Fig 2A, 2B and S1 Table ). The abundance of heterotrophic bacteria varied 10-fold in the photic zone between stations and periods sampled ( Fig 2B and S1 Table ), such that their cell abundances ranged from 3.06×10 5 cells ml -1 (50 m; Stn. 1978m; May 2016) to 3.05×10 6 cells ml -1 (5 m; Stn. 633m; September 2016). Between the photic zone (on average 1.22×10 6 ±6.92×10 5 cells ml -1 ) and the deepest depths analyzed (on average 6.71×10 4 ±1.23×10 4 cells ml -1 ), heterotrophic bacterial abundances decreased by 1.5 orders of magnitude. Closest to shore (Stns. 633m and 1018m; Fig 1A and S1 Table ), we observed significantly higher cell abundances overlying the seafloor (1.35×10 5 ±4.32×10 4 cells ml -1 ) compared to offshore stations with deeper bottom depths (6.51×10 4 ±1.23×10 4 cells ml -1 ; t -test: P <0.001). Abundance also increased rapidly moving upwards through dark waters approaching the photic zone at Stns. 633m and 1018m ( S1 Table ). These vertical trends are similar to those reported at the San Pedro Time-series, SPOT, which is also located in the eastern North Pacific (~480 km from Monterey Bay) and has a bottom depth of ~900 m [ 3 ]. The rate of reduction in cell concentration was slower at depth than in shallower samples at our deeper stations ( Fig 2B ). Reduced abundances of heterotrophic bacteria at deeper depths have been established in regional studies [ 55 ] and more global circumnavigations [ 2 ] as well as meta-analyses [ 55 , 57 ]. This decrease in cellular abundance reflects differences in organic material available for growth, which is more abundant (and more labile) in the surface due to the activities of phytoplankton; differences in heterotrophic bacterial abundances also correspond to decreases in bacterial productivity with depth [ 55 ]. Here, cyanobacteria were present only above 150 m and Synechococcus was maximally 7.24×10 4 cells ml -1 in the photic zone (2 m; Stn. 1978m; September 2016), while the Prochlorococcus maximum cell abundance was 1.08×10 4 cells ml -1 (5 m; Stn. 1820m; September 2015; Fig 2A and S1 Table ). Further, Prochlorococcus was not always detected, and is generally found to be low in abundance or undetectable in the inner bay, whereas Synechococcus appears to be omnipresent and more abundant [ 31 , 58 , 59 ]. A t-test revealed a weak correlation of heterotrophic bacterial cell abundances with Prochlorococcus (0.47, P <0.0001) and a stronger correlation with Synechococcus (0.82, P <0.0001). 10.1371/journal.pone.0298139.g002 Fig 2 Abundance and diversity of bacteria and cyanobacteria across depth and sample periods. (A) Cell abundance of Prochlorococcus (closed symbols) and Synechococcus (open symbols) in the upper 200 m as enumerated by flow cytometry; cyanobacterial cells were not detected below this depth. (B) Heterotrophic bacterial cell abundance (i.e., non-pigmented; analyzed by flow cytometry after SYBR Green I staining) decreased with depth by 1.5 orders of magnitude. Replicates from individual sampling dates revealed a variance of 16% (n = 8) that can be attributed to methodology. (C) Shannon diversity indices for all sequenced samples and depth profiles as calculated from V1-V2 16S rRNA gene ASV data. September 2015 (squares), May 2016 (triangles), and September 2016 (circles) sampling dates are indicated, while stations are represented by color for all panels. Horizontal dotted lines (plots B and C) represent the bottom depth for the respective stations. Bacterial diversity To examine bacterial diversity throughout the water column, we sequenced and analyzed an average of 183,099±76,142 16S rRNA gene amplicons per sample, which resulted in a total of 57,827 ASVs including singletons (44,260 excluding singletons). The V1-V2 16S rRNA variable region used has previously been implemented to resolve taxonomic groups of marine bacteria [ 36 , 43 ], including diversity within Pelagibacterales [ 17 ]. First, we performed rarefaction analyses on the amplicons and ASVs generated for each sample for which computation of the slope on a sliding window indicated saturation for all samples (final slope- mean: 0.00918; median: 0.00819), based on Ocean Sampling Day (OSD) cutoffs [ 60 ]. Samples from Stn.1978m at 1800 m depth and Stn. 1018m at 300 m depth, although outliers with final slopes of only 0.034 and 0.025, were still saturated ( S1A and S1B Fig ). We also compared the Shannon Diversity index computed for each sample with the number of amplicons per sample and found no correlation between these two factors ( S2 Fig ). Taking the whole set together, the Shannon Diversity indices showed clear depth-related changes in bacterial diversity levels, with alpha diversity increasing with depth, such that it was consistently higher below the photic zone, and generally increased across photic zone depths below the very surface. The sharpest increase in bacterial diversity occurred near the base of the photic zone and remained relatively consistent across depths to the deepest point of sampling ( Fig 2C ). Prior studies of vertical trends typically involve amplicon clustering at the 97% OTU level, different from the 100% ASV level used herein, and larger increments between sampling depths. Nevertheless, similar trends were reported for the photic zone to 450–750 m and 1500 m in the northern Gulf of Mexico (V4 16S rRNA gene amplicons, 97% clustering) [ 61 ], and from the photic zone to 500 m and 2000 m in the Mediterranean Sea (V6 16S rRNA gene amplicons, 97% clustering) [ 62 ]. In contrast, a study of four depths in the western and eastern tropical North Pacific, and western subarctic North Pacific, found the highest bacterial diversity generally between 100 and 500 m, and lower diversity in the surface (10 m) and 2000 m, when using V3-V4 16S rRNA gene amplicons clustered at 97% [ 63 ]. Highest bacterial diversity in the mesopelagic has also been reported in two equatorial Pacific and one North Pacific gyre stations (V4-V6 16S rRNA gene amplicons, 97% clustering) with sampling at multiple depths in each depth zone and similar increments between sampled depths as in our study [ 5 ]. In addition to these regional studies, molecular diversity estimates of prokaryotes in more global expeditions, such as Tara Oceans, with sampling to 1000 m, and Malaspina, with a bottom sample at 4000 m, show that diversity increases with depth, although at low vertical resolution in the dark ocean [ 2 , 64 , 65 ]. The observed increase in bacterial diversity below the photic zone implies that the number of niches realized by prokaryotes is larger at depth. The mechanisms behind this are still unclear, but have been proposed to connect to the wide diversity of metabolic roles related to nitrogen, sulfur, and carbon cycling and other processes carried out by bacteria in the deep sea. Some of these metabolic roles may be more prevalent in the deep sea compared to the surface. Additionally, the presence of particles in our sampling may also play a role in the observed patterns of diversity, as particles have been typically reported to contain a higher diversity of bacteria [ 66 – 68 ], and this phenomenon may increase with depth [ 12 ]. Overall, the types and shifts in the proportion of labile versus recalcitrant organic material associated with particles is a function of depth, such that the presence of more diverse organic substrates in deeper waters likely results in the greater variety of bacterial taxa. Bacterial community composition delineates depth zones Classical studies have categorized different parts of the water column based on the resolution of sampling possible–and have identified four major zones: the epipelagic, often termed the photic zone, where sunlight is sufficient to support primary production (0–200 m), the mesopelagic (200 to 1000 m), and the bathypelagic (1000 m to 100 m above the seafloor) [ 69 ]. The bathypelagic is sometimes further delineated to define the abyssopelagic, depths below 4000 m, that are not addressed herein. Higher resolution studies based on 16S rRNA gene amplicon analyses have introduced more nuance to the classical depth zone ‘cutoffs’. For example, multi-depth vertical profiling (for multiple years) in the subtropical North Atlantic demonstrates that the top boundary of the upper mesopelagic is often in the vicinity of 160 m rather than 200 m, with important seasonal nuances [ 13 ]. Here, we focused on repeat sampling, with greater depth resolution than typically reported, and sample collection distributed through depths that would be classically partitioned as photic, mesopelagic, and bathypelagic. We first examined how bacterial community structure connects to the classically predefined zones by clustering samples based on the bacterial community. Specifically, we used the 1,000 most relatively abundant ASVs and Bray-Curtis dissimilarities, to perform hierarchical clustering and statistical testing. Clustering clearly delineated a ‘bathypelagic’ zone that incorporated all samples from 900 m (i.e., shallower than some strict definitions [ 69 ] and below (900 to 1980 m, the latter being at our deepest site), and a broad photic zone set (the majority of samples from 100 m and shallower) ( Fig 3 ). Within the latter were two primary clusters that revealed upper (primarily samples from 2 to 30 m) and lower photic zone sets (primarily 35 to 70 m). Adjacent to the nominal bathypelagic cluster were mesopelagic samples (150 to 750 m). These formed two clusters, one containing lower mesopelagic samples (300 to 750), and the other containing upper mesopelagic samples (150 to 250), with the latter also including some samples from shallower depths, likely reflecting the fact that the photic zone in Monterey Bay is highly variable in its depth extent, and at times very shallow [ 70 ]. Additionally, in May 2016 there was a distinctive upper photic cluster (2–25 m), with a second similar but still distinct community in the lower photic zone. Site-related or seasonal differences did not stand out within mesopelagic samples, which is in contrast to waters in which the entire water column can be subjected to vertical mixing like in the eastern Mediterranean and subtropical North Atlantic [ 71 , 72 ]. However, the delineation between upper (again, 150 to 250 m, with occasional shallower samples) and lower (300 to 750 m) mesopelagic were statistically supported, and samples from >750 m depth were excluded from the mesopelagic cluster altogether ( S3 Fig ). While we do not have the sampling resolution to examine where the shift to bathypelagic communities happens (i.e., between 750 m and 900 m), our findings clearly place 900 m communities, for these Monterey Bay profiles during our time of sampling, with depths ≥1000 m which have been classically defined as the bathypelagic. 10.1371/journal.pone.0298139.g003 Fig 3 Community and taxonomic patterns in bacterial communities along the depth gradient. The right side shows hierarchical clustering performed on Bray-Curtis dissimilarities of bacterial communities based on the 1000 most relatively abundant ASVs of bacteria (both non-photosynthetic and cyanobacteria (i.e., photosynthetic); see S4 Fig for non-photosynthetic bacteria only). Stations (color) and sampling dates (shape) are indicated as is the depth sampled (m). The left side shows relative abundance of the 50 most relatively abundant ASVs (columns) per sample from the five main depth zones as defined herein. Ordering along the X-axis is by phyla (indicated by horizontal gray bars at the base of figure) and rows represent individual water samples–corresponding to those immediately adjacent (i.e., to the right). ASV relative abundances, sequences, and taxonomic classification are detailed in S3 Table . The four groups with the overall highest relative abundances are indicated (boxes overlaying heat map) and were alphaproteobacterial ASVs assigned to the Rhodobacteraceae and Pelagibacterales Ia in the photic zone, whereas in mesopelagic and bathypelagic samples ASVs from SAR324 and Gammaproteobacteria Oceanospirillales exhibited the highest relative abundances. Taxonomic assignment for ASVs is based on the SILVA database. With respect to shifts in patterns of bacterial relative abundances, most variability was connected to photic zone variability, as above, and such shifts were not evident in the mesopelagic or bathypelagic over time. For example, increased relative abundances of alphaproteobacterial class Rhodobacteraceae ASVs influenced the composition-based clustering of May 2016 upper photic (2–25 m) samples, making up 24.2±9.3% of this cluster ( Fig 3 ). Additionally, Bacteroidetes show higher relative abundances in the photic zone, with means of 19.0±6.8%, 9.0±3.2%, and 8.4±1.2% in the photic, mesopelagic, and bathypelagic (henceforth depths ≥900 m), respectively ( Fig 3 ). The highest relative abundance of Bacteroidetes ASVs was observed in May 2016 upper photic samples (23.8±5.1%) coincident with the increased relative abundance of Rhodobacteraceae. Alphaproteobacteria dominated photic amplicons, with Pelagibacterales subclade Ia and Rhodobacteraceae ASVs exhibiting the highest relative amplicon abundances (20.0±6.8% and 16.4±8.2%, respectively) ( Fig 3 ). Contributions from these taxa decreased in the mesopelagic (13.0±2.5% and 2.0±2.7%, respectively) and bathypelagic (7.1±1.1% and 0.4±0.2%, respectively), where SAR324 and the gammaproteobacterial order Oceanospirillales dominated (mesopelagic: 13.6±4.4% and 16.8±4.1%, respectively; bathypelagic: 13.6±2.2% and 16.4±2.5%, respectively). Marinimicrobia (also commonly referred to as SAR406) were mostly present in the bathypelagic (13.6±1.6%), with lower relative abundances in the mesopelagic (8.2±3.3%) and photic (1.8±1.6%) zones. Actinobacteria relative abundances increased slightly through the water column (3.3±1.6% photic, 6.6±1.1% mesopelagic, 7.2±0.9% bathypelagic), and Nitrospinae were lowest in the photic zone (0.9±1.2%) with similar mesopelagic (4.2±0.6%) and bathypelagic (3.2±0.4%) contributions. Verrucomicrobia exhibited similar abundances from the photic (5.0±2.5%), to the mesopelagic (3.3±0.8%) and bathypelagic (4.7±1.0%; Fig 3 ). Strong vertical gradients in abundant bacterial ASVs We next examined the 50 most abundant ASVs of each of the three major depth zones as identified herein, i.e. photic (0 to 100 m), mesotrophic (150 to 750 m), and bathypelagic (900 to 2000, the latter being the rounded depth of our deepest site ( S3 Table ; in sum n = 148 ASVs, with some being present in more than one depth zone). ASVs belonging to specific lineages often exhibited depth-related patterns. For example, four of the Oceanospirillales ASVs had higher relative abundances in the mesopelagic (3.7±0.8%, 1.9±0.4%, 1.6±0.6%, 1.3±0.5%) and bathypelagic (3.1±0.3%, 2.8±0.4%, 2.3±0.4%, 1.4±0.3%) than in the photic zone (0.7±0.9%, 0.5±0.6%, 0.7±0.9%, 0.5±0.7%). Just one Oceanospirillales ASV had >1% relative abundance in the photic zone (1.6±0.7%) and exhibited lower relative abundance in the mesopelagic (0.6±0.6%) and bathypelagic (0.02±0.02%) depths. The remaining 19 Oceanospirillales ASVs were present at <0.5% at all depths. Likewise, four distinct Actinobacteria ASVs were observed in the bathypelagic (1.7±0.5%, 0.9±0.3%, 0.7±0.2%, 0.5±0.1%) that exhibited lower relative abundance at photic (0.1±0.1%, 0.1±0.1%, 0.0±0.1%, 0.0±0.0%) and mesopelagic (0.2±0.1%, 0.1±0.1%, 0.1±0.0%, 0.1±0.0%) depths. Three other Actinobacteria from the mesopelagic (0.7±0.5%, 0.6±0.3%, 0.6±0.3%) exhibited lower relative abundance at photic (0.4±0.5%, 0.1±0.1%, 0.1±0.1%) and bathypelagic (0.0±0.0%, 0.1±0.1%, 0.1±0.1%) depths. The two groups highlighted here (Oceanospirillales and Actinobacteria) are widespread in the ocean and have been shown to be abundant throughout the water column [ 12 , 73 , 74 ]. While the functional differences between the sub-species represented by the ASVs assigned to the Oceanospirillales and Actinobacteria are not yet known, these results highlight that even once more genomic information becomes known, there will likely be subtle differences in the ecosystem roles and impacts of different strains/sub-species. Co-occurrences of bacterial ASVs and correspondence with habitat Two other types of analyses were performed to examine bacterial communities in the context of their habitat. The first was an NMDS, which showed similar trends as seen using hierarchical clustering, wherein samples from depth zones generally grouped together, and the 900 m samples aligned in the first dimension with other bathypelagic samples ( Fig 4A ). Additionally, after taxonomic characterization, WGCNA was used to identify highly correlated ASVs across samples from the full dataset and relate the resulting modules (or subnetworks) of co-occurring bacterial taxa to environmental parameters [ 47 , 75 ]. Seven major modules were identified based on the co-occurrence of ASVs, with ASV numbers ranging from 10 to 389 per module ( Fig 4B and S4 Table ). Six modules showed varying degrees of correlation with changes in environmental conditions, specifically temperature, salinity, chlorophyll, fluorescence, phosphate, nitrate, nitrite, and oxygen. Silicate was excluded from this analysis as almost all of the samples from May 2016 did not measure silicate, and due to this missing data too many samples would have been excluded by WGCNA. One module (Module 7) did not connect with the environmental data analyzed and consisted of only 10 ASVs, including ASVs classified as Rhodobacteraceae, Flavobacteriaceae, and was the only module to include Cyclobacteriaceae (4 ASVs). For four modules (Modules 1 through 4, see below), the environmental parameter correlations followed depth-related vertical trends in the water column. 10.1371/journal.pone.0298139.g004 Fig 4 Bacterial community and ASV relationships with habitat. (A) NMDS exploring the relationship between communities within samples from the five refined depth zones (see Fig 3 ) and environmental parameters. (B) Pearson correlations relating to untransformed environmental physiochemical parameters to the eigengene (first principal component of the abundance matrix) of each WGCNA module. Positive relationships are in red and negative relationships are in blue (as indicated by the color bar). The top number is the Pearson correlation value and asterisks in parenthesis represent the P -value for each relationship. Bolded numbers are significant correlations ( P <0.01; *, <0.01; **, <1x10 -5 ; ***, <1x10 -10 ). Relationships were examined between the 7 modules and temperature (°C), salinity (PSU), chlorophyll (mg m -3 ), fluorescence, phosphate (μM), nitrate (μM), nitrite (μM), and oxygen (mg L -1 ). Pie charts show the taxonomic groups within each of the 7 modules of co-occurring bacterial ASVs that emerged from WGCNA based on taxonomy assigned using the SILVA database ( S4 Table ). Two WGCNA modules and their members exhibited similar statistically significant correlations ( P <0.05) with respect to the included environmental parameters. Specifically, Module 1 and Module 2 were inversely related to temperature, chlorophyll, fluorescence, nitrite, and oxygen, while positively related to salinity, phosphate, and nitrate ( Fig 4B and S1 Table ). The negative correlation to chlorophyll and fluorescence was less strong than to the other environmental parameters, and more strongly negative for Module 2 than for Module 1. This indicates that the taxa making up these modules are relatively well-adapted to the deep ocean, which has similar relationships to these parameters, as well as other attributes not quantified herein, such as low availability of labile organic carbon and relatively low productivity. The actual niches of the diverse individual taxa cannot be strictly defined from our limited environmental data, which may covary with other parameters. Nevertheless, the slight difference between Modules 1 and 2 suggests that distinct sets of environmental parameters (or covarying factors) correlate with the different sets of bacterial taxa. These distinctions require further investigation to understand the functional implications. We then restricted our analysis to the ASVs most highly correlated with the co-occurring ASVs in a module based on Module Membership (MM) > 0.80; where an MM value close to 1 indicates an ASV was highly correlated with its module [ 47 ]. Module 1 consisted of 167 highly correlated ASVs and Module 2 consisted of 303 such ASVs, corresponding to 92% and 78% of the total ASVs in these modules, respectively. Module 1’s ASVs were diverse, with 15% of ASVs being classified as Marinimicrobia, 11% SAR324, 10% Pelagibacterales, 12% Bacteroidetes, 11% Verrucomicrobia Arctic97B-4 marine group, 8% Chloroflexi SAR202, and 5% Acidimicrobiales Sva0996 marine group ( Fig 4B ). Module 2 differed with Pelagibacterales being more dominant than in Module 1, comprising 35% of the ASVs, while also containing 12% Marinimicrobia, 5% SAR324, and 4% Nitrospina ASVs. Between Module 1 and Module 2, there were 31 taxonomic groups that had considerable representation in both modules, but consisting of distinct ASVs (within those groups, S4 Table ), including Pelagibacterales, Marinimicrobia, Bacteroidetes, SAR324, Oceanospirillales and Salinisphaerales, and Verrucomicrobia Arctic97B-4 marine group. Thus, overall these two modules are diverse and primarily associated with deep waters. Given that they have similar correlations to environmental comparisons (except chlorophyll and fluorescence), this indicates potential interactions or intra-connectedness between the microbes may be driving the differences in the microbial network. Module 3 exhibited opposing patterns to those of Modules 1 and 2, as it was significantly negatively related to salinity, phosphate, and nitrate and positively correlated to temperature, chlorophyll, fluorescence, nitrite, and oxygen ( Fig 4B ). These relationships point to general characteristics of the surface ocean, including presence of phytoplankton and their exudates, as having importance in shaping the taxa in this Module. There were 188 highly correlated ASVs of Module 3 (MM > 0.80; 64% of Module 3 ASVs). The greatest percentage of ASVs in the complete Module 3 set was assigned to Flavobacteriales (39%), including a predominance of Cryomorphaceae (12% of ASVs) and Flavobacteriaceae (21% of ASVs; S4 Table ). Rhodobacteraceae were also abundant, making up 17% of the Module 3 ASVs. Pelagibacterales made up 6% of Module 3 ASVs, while there were no ASVs classified as SAR324. Module 6 was significantly positively correlated to temperature, nitrite, and oxygen and inversely correlated with salinity, phosphate, and nitrate—a similar correlation pattern to Module 3 ( Fig 4 ), but, in contrast, did not have a significant relationship with chlorophyll. These results suggest this module comprises taxa preferring the conditions of surface seawater in which there are variable amounts of phytoplankton (and/or or potential types), such as oligotrophic periods like summer or winter. There were 86 highly correlated ASVs in Module 6 (85%), 24% classified as Oceanospirillales, 16% as Pelagibacterales, 9% as SAR324, 9% as Rhodobacterales, and 7% as Rhizobiales and Rhodospirillales. At the phylum level there were clear differences from Module 3, and even within phyla differences could be observed, such as Oceanospirillales formed only 5% of ASVs in Module 3 versus 24% in Module 6 ( S4 Table ). The correlations of Module 4 were much weaker and significant only with chlorophyll (and fluorescence) and oxygen, both negatively, as well as nitrate (positively; Fig 4B ), and not correlated to temperature and salinity. This suggests that the taxa comprising Module 4 are capable of thriving across a range of temperatures and salinities, and adapted to conditions characterized by moderately low oxygen, and nitrate. This pattern aligns with the environmental conditions typically found in mid-water, where mixing of different water masses with highly variable temperature and salinity occurs. There were 90 highly correlated ASVs (93%), 41% classified as Pelagibacterales, 16% as Flavobacteriales, 10% as SAR324, and 8% as Nitrospinaceae. Module 5 showed significant but weak positive correlations with temperature and nitrite, and negative correlations with salinity and chlorophyll (and fluorescence; Fig 4 ). No correlation was observed with oxygen, nitrate, and phosphate, suggesting that the taxa in this module are associated with conditions in low-productivity mid waters. The most highly correlated ASVs of Module 5 (79 ASVs, 81%) were diverse and included 20% ASVs classified as Nitrospinaceae, 15% as Salinisphaeraceae, 10% as Oceanospirillales, 5% as SAR324, and 3% as Pelagibacterales. Overall, the co-occurrence patterns herein suggest the presence of a limited number of distinct microbial community ‘states’ in relation to a set of relevant environmental parameters. The functional characteristics of these communities may vary but should ultimately be possible to identify. Clearly, the power to define the niches based on the environmental parameters quantified herein is limited, and numerous other attributes may co-occur that could help distinguish the modules recovered more clearly. These include the amount and composition of organic carbon, and a large number of other elements or compounds (such as vitamins [ 9 ]). The modular and co-occurrence patterns observed here must be interpreted with the understanding that within each of these modules there may be numerous other states, overwhelmed by sharp gradients of the parameters from the entire dataset. Within these potential states, beyond the scope of our study, there may be other statistically significant microbial co-occurrence patterns that can be distinguished, which may represent microbial interactions. Our analysis cannot distinguish biological microbe-to-microbe co-occurrence patterns (ASV-to-ASV) from those that are ‘simply’ co-occurring in connection to environmental conditions (ASV-to-environmental parameter). Recently, several studies have revealed that microbe-to-microbe co-occurrence patterns in the ocean, potentially representing biological interactions, can be ‘disentangled’ from those that reflect co-occurrence due to correlations to environmental parameters [ 76 , 77 ]. Understanding how taxa within the different modules interact competitively or synergistically, and impact ecosystem processes, is a major frontier for study. Phylogenetic analysis of Pelagibacterales and SAR324 Thus far our analyses pointed to important roles for Pelagibacterales across the water column, and for SAR324 especially in deeper waters, though also occasionally in the photic zone, with both exhibiting numerous ASVs co-associated with other taxa. Most analyses have relied on SILVA based placement for taxonomic assignment, and we now sought to connect our observations to evolutionary aspects of these two diverse bacterial lineages by characterizing our amplicon sequences using phylogenetic methods. To this end, we performed phylogenetic reconstructions on both these groups that incorporated near full-length 16S rRNA gene sequences obtained from aquatic environments from previously unrecognized members, alongside known reference sequences. For the Pelagibacterales, the resulting tree topology ( Fig 5 and S4 Fig ) had similarities to prior phylogenetic analyses [ 14 , 17 , 73 , 78 – 80 ]. However, three supported, previously unrecognized subclades were exposed, and disrupted prior branching structures: Ia.B, Id, and IIc. To ensure that new subclades did not result from possible PCR-related errors, we required that each new subclade contains sequences from at least two different samples. Clade I has three previously identified 16S subclades: Ia and Ib, consistently found in the photic zone [ 13 , 20 ] and Ic, mostly found in the mesopelagic and bathypelagic [ 14 , 80 ]. We delineated two additional clade I subclades, named here Ia.B and Id, the former is sister to subclade Ia and the latter branches as sister to a group comprising Ia, Ia.B, and Ib. The third subclade (IIc) recognized herein clusters together with IIb. In our reconstruction, subclade IIa branched as sister to a statistically supported group containing all clade I subclades and subclades IIb and IIc. Compared to clades identified in prior publications, clade IV and clade V (subclades Va and Vb) exhibit significant genomic divergence at both the nucleotide and protein levels [ 81 , 82 ], and they branched basally to the above groups as initially described [ 17 , 78 ]. Information on the distributions and apparent niches of the known and newly identified clades are discussed below ( Distributions of phylogenetically resolved Pelagibacterales) . 10.1371/journal.pone.0298139.g005 Fig 5 Phylogeny and distributions of Pelagibacterales. The 16S rRNA gene phylogenetic reconstruction utilized full-length sequences and identifies 10 recognized subclades and 3 previously unrecognized subclades, all with bootstrap support >90% (1,000 replicates) except subclade IIb (86% support; see also S4 Fig ). Note that dashed lines indicate LCA placements at unsupported nodes, some with relatively few amplicons. For example, that demarked ‘Clade V’—with a dashed line comprised >0.01% of amplicons . Some others with dashed lines were notable, but lacked full length sequences and therefore references sequences were not in the reconstruction. Also shown is the relative abundance of Pelagibacterales subclades as percent of all Pelagibacterales ASVs in each sample for the Monterey Bay Stations and depths sampled. Pelagibacter amplicons were first retrieved using PhyloAssigner and a global 16S rRNA gene reference tree [ 17 ]. These amplicons were then run in PhyloAssigner using the above Pelagibacterales reference tree. Heat map columns represent individual samples ordered by hierarchical clustering based on the Bray-Curtis similarities of the Pelagibacterales community composition. White in the heat map indicates not detected. Stations (color) and sampling dates (shape) are indicated as is sample depth (m) for each column. We also performed a 16S rRNA gene phylogenetic reconstruction of SAR324 ( Fig 6 and S5 Fig ). The resulting tree topology corresponds to previously published topologies [ 27 , 25 , 30 , 83 , 84 ]. The original three clades of SAR324 based on ITS reconstruction (Brown & Donachie, 2007) were also resolved here with the 16S rRNA gene. We identified two putative basal clades potentially affiliated with this lineage, named here clades III and IV. Our reconstruction places clade III outside the three original SAR324 clades. Clade IV contains sequences included in previous phylogenetic reconstructions, but never as a distinctive or supported clade [ 30 , 83 ]; distributional data helps to identify the respective niches of these clades (see below). 10.1371/journal.pone.0298139.g006 Fig 6 SAR324 16S rRNA gene reference tree identifies novel clades and distributions. Shown is the ML reconstruction with percent relative abundance of SAR324 clade members out of total SAR324 amplicons at the study sites shown to the right of the tree topology. All clades retained support >90% apart from clades II and IV (82 and 88%, respectively; see also S5 Fig ). Amplicons were phylogenetically placed on the SAR324 reconstruction developed herein using PhyloAssigner [ 17 ] after initial assignment and retrieval using a global 16S rRNA gene reference tree [ 17 ]. Heat map columns represent individual samples ordered by hierarchical clustering based on the Bray-Curtis similarities of the SAR324 composition. White in the heat map indicates not detected. Stations (color) and sampling dates (shape) are indicated as is sample depth (m) for each column. Distributions of phylogenetically resolved Pelagibacterales The Pelagibacterales are known for high diversity [ 78 ], something that has been noted in many studies. High microdiversity has been further highlighted at SPOT, where the most temporally common taxa, such as Pelagibacterales, were reported to have the highest microdiversity, quantified therein as the number of 100% ASVs within a given 99% OTU [ 85 ]. Here, the 16S rRNA gene Pelagibacterales phylogenetic reconstruction allowed placement of Pelagibacterales amplicons from our study region and examination of community structure as well as subclade relative contributions to the Pelagibacterales collective. Subclade Ia had the highest relative abundance among all subclades and dominated the photic zone Pelagibacterales community (77±13%). This observation is in line with a recent study that found subclade Ia to be the most abundant Pelagibacterales subclade in the surface ocean [ 18 ]. Subclade Ib was represented by 27±4% and 24±2% of mesopelagic and bathypelagic Pelagibacterales amplicons, respectively. In the Red Sea subclade 1b is present in both mesopelagic and bathypelagic waters [ 79 ] and at the Bermuda Atlantic Time-series Study (BATS), it is usually found in mixed water during late spring and early summer, while Ic is found in deeper water [ 17 , 80 ]. We observed similar patterns, with subclade Ic exhibiting high relative abundances in the bathypelagic (14±1%), but less in the mesopelagic (4±3%) and the photic zone (<1%). The newly identified subclades exhibited relative abundances that rivaled subclade Ic, with subclade Ia.B having the highest relative abundances among Pelagibacterales in the mesopelagic (10±2%) and bathypelagic (10±1%). Subclade Id exhibited a lower presence than subclade Ia.B, with their highest relative abundance in the mesopelagic (1±0%), with a decrease in relative abundance in the bathypelagic (0.7±0.3%) and photic zone (0.2±0.2%). This suggests that these two new subclades are adapted to the cold, low organic material availability environments seen at deeper depths. Even with statistical support of the Pelagibacterales topology, which reflects those taxa captured thus far by culturing or full-length sequencing, some amplicons were classified as being related to the Last Common Ancestor (LCA) of clade I (indicated with dashed line inserted in phylogenetic tree topology, Fig 5 ) in the photic (0.2±0.2%), mesopelagic (1±0.3%), and bathypelagic (1±0.3%) zones. Further efforts to characterize these taxa by retrieving their full length 16S rRNA gene sequences would facilitate studies of the affiliation and diversity of these novel ASVs. Pelagibacterales subclade IIb dominated the mesopelagic (30±8%) and bathypelagic (40±3%) and was less relatively abundant in the photic zone (4±5%), while subclade IIc was detectable at ≤1% relative abundance in all three zones, suggesting this rare type occupies a niche that is conserved across all depths. The similar relative abundances of subclade IIb in the mesopelagic and the bathypelagic connects to the upper mesopelagic depth distributions that have been reported at BATS during spring [ 13 , 86 ] and also in the deeper cold Arctic Ocean [ 19 ]. Subclade IIa relative abundance was an order of magnitude lower (8±3%) than that of subclade Ia in the photic zone and decreased further to 5±1% and 2±1% of Pelagibacterales in the mesopelagic and bathypelagic, respectively. Our analysis of IIa sequences supports the further breakdown into two subclades [ 87 ], with subclade IIa.B exhibiting higher relative abundances in the photic zone (5±4%), decreasing in relative abundance in the mesopelagic (2±2%) and bathypelagic (0.4±0.2%). Subclade IIa.A was found at low abundances throughout the water column, with its highest relative abundance in the mesopelagic (0.6±0.4%) and even still being detectable (at <0.2% relative abundance) in the photic zone as well as the bathypelagic. While there are fewer IIa.A sequences in our dataset compared to other Pelagibacterales, their relative abundance was generally higher at depths where oxygen values are lowest within the water column, in agreement with reports of IIa.A higher relative abundances in Oxygen Minimum Zones (OMZs) [ 87 ]. A small number of amplicons were classified as being basal to clade I/II in the photic (0.1±0.2%), mesopelagic (1±0.3%), and bathypelagic (1±0.2%) ( Fig 5 ). While clades III, IV, and V were among the least relatively abundant in our samples, they did align with specific environmental conditions encountered in our sample set–or were not encountered at all, in accordance with prior knowledge of habitats. Subclade IIIa, which is usually abundant in brackish water [ 88 ], formed 0.05±0.07% of photic zone Pelagibacterales amplicons and was not detected in the mesopelagic and bathypelagic. Subclade IIIb (LD12) is a known freshwater subclade and was not detected in any samples. In contrast, Clade IV was detected throughout the water column, generally at ≤1% relative abundance, but, compared to its relative contributions in the photic zone and bathypelagic, was more important in the mesopelagic. Clade V showed slightly higher contributions, with Subclade Va showing <0.1% relative abundance in the mesopelagic and bathypelagic, but was an order of magnitude higher in the photic zone (still <1%). Subclade Vb at 3±1% and 4±0% in the mesopelagic and bathypelagic, respectively, and <1% in the photic zone. Adding to the established distribution of subclade Vb in the surface ocean [ 7 ], our study found subclade Vb at all depths throughout the water column. Overall, the broad similarity of our phylogenetics-based analysis of SAR11 ASVs and their relative abundances suggests conserved niches for the clades across and between ocean basins, a trait that is ripe to be systematically evaluated using phylogenetic approaches, such as those presented herein. Patterns in phylogenetically resolved SAR324 clades Analysis of our 16S rRNA gene amplicons against the SAR324 reference tree, using PhyloAssigner [ 17 ], augments previously described vertical distributions. While considered ubiquitous below the photic zone [ 24 – 27 ], SAR324 has been described in many marine environments such as the bottom of the mixed layer in the Sargasso Sea and northeastern subarctic Pacific Ocean [ 21 ], throughout the water column in the North Pacific Subtropical Gyre [ 25 , 29 , 84 ], the Saanich Inlet OMZ [ 89 ], and coastal polar waters during winter and deep polar waters year-round [ 24 ], as well as in marine hydrothermal systems [ 30 , 83 ]. A phylogenetic examination of the clade abundances of SAR324 amplicons identified clade II as the dominant clade across all depths (photic: 87±10%; mesopelagic: 91±2%; bathypelagic: 90±1%) ( Fig 6 ), as indicated from the original observations that distinguished Clade II from the North Pacific and other locations [ 28 ]. Clade I was found predominantly in the mesopelagic (3±2%), with lower contributions in the photic zone (2±1%) and bathypelagic (<0.1%). Low Clade I contributions in the photic zone are consistent with results from BATS [ 90 ]. Clade SAR276 decreased in relative abundance with depth ( Fig 6 ), making up 6±8% of SAR324 in the photic, 2±1% in the mesopelagic, and <0.1% in the bathypelagic; a preference for photic zone depths is consistent with previous results from the North Pacific and other marine regions [ 25 ], while SAR276 was relatively stable (2–3%) across the photic and mesopelagic at BATS where it was the only clade with significant abundance in the photic zone. Clades III and IV occurred at very low abundances throughout the water column. Clade III was mostly found in the photic zone (3±4%) with low detection (<0.01%) in the mesopelagic and bathypelagic. Clade III has a potentially photic zone distribution, consisting of sequences from the photic zone as well as being detected in our photic zone samples, though was also detected in a few samples from the mesopelagic and bathypelagic. Clade IV consists of sequences from sediment and subsurface environments [ 91 – 94 ]. In our dataset, clade IV was found in the photic (0.1±0.7%) and bathypelagic (0.2±0.1%) with minimal detection in the mesopelagic (0.07±0.06%). Our observation that there are distinct differences in depth distributions of SAR324 groups is similar to observations of depth differences observed based on average nucleotide identity of partial SAR324 genomes (classified as groups ’A’ to ‘E’) from the North Pacific Subtropical Gyre [ 29 ], which were related to functional differences. Similar to the SAR11 analyses above, this indicates remarkable niche similarity within and between ocean basins at least as experience by members of the various SAR324 clades. While the newly identified subclades III and IV are not yet connected to functional information, their distributional data should help targeting in future studies, given that Clade III is found primarily in the Monterey Bay photic zone and Clade IV is seen throughout the water column. Additionally, it will be important to resolve (as possible) the 16S rRNA phylogeny in connection to SAR324 phylogenomics and genome taxonomy. Finally, even with statistical support of the SAR324 topology, a notable number of amplicons were classified as being related to the Last Common Ancestor of SAR324 (“basal”; based on reference data availability), especially in the mesopelagic (7±1%) and bathypelagic (10±1%). Full length 16S rRNA gene sequences, or genome sequences, would help to resolve their affiliation and diversity. Here, hierarchical clustering shifted from the general depth-related consistency seen between the overall community composition ( Fig 3 ) and the SAR11 group ( Fig 5 ), indicating that greater partitioning of clade II likely exists but has not been phylogenetically resolved yet. Additional full-length sequences or increased genomic data from phylogenomic analyses will be important for resolving whether these are most closely related to the SAR324 Last Common Ancestor or belong to existing clades."
} | 13,415 |
24586995 | PMC3931815 | pmc | 9,688 | {
"abstract": "The creation of a synthetic microbe that can harvest energy from sunlight to drive its metabolic processes is an attractive approach to the economically viable biosynthetic production of target compounds. Our aim is to design and engineer a genetically tractable non-photosynthetic microbe to produce light-harvesting molecules. Previously we created a modular, multienzyme system for the heterologous production of intermediates of the bacteriochlorophyll (BChl) pathway in E. coli . In this report we extend this pathway to include a substrate promiscuous 8-vinyl reductase that can accept multiple intermediates of BChl biosynthesis. We present an informative comparative analysis of homologues of 8-vinyl reductase from the model photosynthetic organisms Rhodobacter sphaeroides and Chlorobaculum tepidum . The first purification of the enzymes leads to their detailed biochemical and biophysical characterization. The data obtained reveal that the two 8-vinyl reductases are substrate promiscuous, capable of reducing the C8-vinyl group of Mg protoporphyrin IX, Mg protoporphyrin IX methylester, and divinyl protochlorophyllide. However, activity is dependent upon the presence of chelated Mg 2+ in the porphyrin ring, with no activity against non-Mg 2+ chelated intermediates observed. Additionally, CD analyses reveal that the two 8-vinyl reductases appear to bind the same substrate in a different fashion. Furthermore, we discover that the different rates of reaction of the two 8-vinyl reductases both in vitro , and in vivo as part of our engineered system, results in the suitability of only one of the homologues for our BChl pathway in E. coli . Our results offer the first insights into the different functionalities of homologous 8-vinyl reductases. This study also takes us one step closer to the creation of a nonphotosynthetic microbe that is capable of harvesting energy from sunlight for the biosynthesis of molecules of choice.",
"introduction": "Introduction Sunlight is an abundant and sustainable energy source that is captured by photosynthetic organisms and converted into chemical energy for growth and survival. Utilization of the photosynthetic machineries of light harvesting organisms plays an important role in the bioproduction of fuels and chemicals [1] – [4] . Engineering light capture and conversion into genetically tractable, nonphotosynthetic and robust microorganisms already used for industrial processes represents an alternative approach [5] , [6] . Such designer microbes could be engineered to synthesize a range of valuable and novel compounds from inexpensive carbon sources where light-energy drives otherwise expensive synthetic reactions [7] . The first steps on the path towards engineering a nonphotosynthetic microorganism able to harvest light-energy are to install either simple light-driven proton pumps [8] or more powerful photosynthetic reaction centers [5] , [6] . Both systems require functional assembly of a biosynthetic pathway for carotenoid-derived pigments, and reaction centers also require (bacterio)chlorophyll ((B)Chl) pigments for function. While engineering of carotenoid pathways into various hosts has been shown [9] , [10] , complete reconstruction of a BChl biosynthetic pathway remains to be demonstrated, a formidable task owing to the complexities of the reaction pathway and enzymes involved ( Fig. 1 ). 10.1371/journal.pone.0089734.g001 Figure 1 Engineered pathway design for the heterologous production of BChl in the non-photosynthetic host E. coli . Using succinyl-CoA and glycine as precursor molecules, expression of the heme pathway enzymes HemA-F in E. coli results in production of P IX as the common intermediate of the heme and BChl biosynthetic pathways. Addition of the BChl enzymes magnesium chelatase (BchHID) and methyltransferase (BchM) yields MgP IX and MgP IX ME in E. coli \n [11] , [15] . Subsequent steps have not yet been functionally assembled in a heterologous system and depending on the enzymes substrate specificities, the order in which the enzymes operate may differ from the depicted pathway. Briefly, formation of the characteristic fifth E ring of chlorophylls is catalyzed by two unrelated and yet to be biochemically characterized cyclases AcsF (aerobic) [20] or BchE (anaerobic) [19] . The D pyrrole ring is reduced either by a light-dependent, nitrogenase-like (LPOR, three-subunit enzyme BchLNB) or a light-independent (DPOR) protochlorophyllide reductase; both enzymes have been biochemically characterized [23] – [26] . Reduction of the C8-vinyl group of BChl intermediates is catalyzed by the NADPH-dependent reductase BciA [27] , [30] investigated in this study. Seven additional enzymatic steps are required for production of Bchl a \n [14] . As a first step towards this goal, we created a modular system for the high level production of porphyrins, including protoporphyrin IX (P IX ), by assembling genes involved in heme biosynthesis (HemA-F) in E. coli \n [11] . P IX is the common intermediate between the heme and BChl biosynthetic pathways [12] – [14] and is committed to Bchl biosynthesis upon insertion of a central Mg 2+ catalyzed by a multi-subunit magnesium (Mg-) chelatase enzyme complex BchHID (homologues of H, namely S and T, are present in some bacteria like the green sulfur bacterium Chlorobaculum tepidum ) [15] . The chelatase subunit BchH interacts with the SAM-dependent methyltransferase BchM, which methylates MgP IX at the C13-carboxyl group, resulting in MgP IX monomethyl ester (MgP IX ME) [16] – [18] . Co-expression of BchSID and BchM from C. tepidum in our P IX overproducing E. coli strain resulted in high level production of P IX , P IX ME, MgP IX and MgP IX ME [15] . Detailed in vitro studies provided insights into enzyme interactions and kinetics and revealed that BchM also methylates P IX , resulting in the accumulation of the “dead-end” product P IX ME, which cannot be chelated by BchSID [15] . Following chelation and methylation of P IX , the characteristic fifth ring of the chlorin molecule is formed under anaerobic conditions by the radical-SAM cyclase BchE, or under aerobic conditions by AcsF, producing divinyl protochlorophyllide (DVP) [17] , [19] , [20] . Reduction of the D pyrrole ring of DVP to produce chlorophyllide is either catalyzed by a light-independent, nitrogenase like (DPOR) or by a light-dependent (LPOR) protochlorophyllide reductase [21] – [26] . An NADPH-dependent reduction of the C8-vinyl group to an ethyl group by 8-vinyl reductase BciA results in chlorophyllide a \n [27] – [30] . Beyond this, another seven polypeptides are required to complete the biosynthesis of bacteriochlorophyll a \n [14] . Fig. 1 shows the upper part of the BChl pathway; depending on the substrate specifities of the biosynthetic enzymes, the order in which they operate may differ from the sequence shown. While some of these additional enzymes have been functionally expressed and biochemically characterized in vitro (e.g. LPOR and DPOR [23] , [25] , [31] ), other steps of this complex pathway have only been elucidated by gene knockouts/deletions, complementation, and mutational studies [14] , [32] , [33] . Many of these enzymes form complexes, catalyze novel reactions and may interact with yet to be identified protein partners [34] , making biochemical studies as well as heterologous pathway reconstitution particularly challenging. In our quest towards recombinant BChl biosynthesis, we report the extension of the BChl biosynthetic pathway in E. coli to include an 8-vinyl reductase. Recent studies have indicated that various homologues of the 8-vinyl reductase BciA are substrate promiscuous in vivo and can reduce the C8-vinyl group of different intermediates of the BChl pathway [35] , [36] . We hypothesized that including an 8-vinyl reductase as the next step in our pathway would result in reduction of the C8-vinyl group of multiple BChl intermediates that do not have the fifth ring of divinyl protocholorophyllide (DVP) ( Fig. 1 ), thereby possibly removing barriers to the efficient turnover of P IX in our engineered system. We demonstrate co-expression of the heme biosynthetic pathway in conjunction with C. tepidum BchSID and BchM with two separate homologues of BciA from Rhodobacter sphaeroides ( RS BciA) [30] and C. tepidum ( CT BciA) [27] . We discovered that while CT BciA is capable of reducing the C8-vinyl group of several different intermediates in the BChl pathway, RS BciA is surprisingly completely inactive in our recombinant system. We therefore conducted a full purification and in vitro characterization of the two BciA homologues to elucidate possible mechanisms for their different activities. Results show that both RS BciA and CT BciA are substrate promiscuous in vitro , however, the two enzymes exhibit very different catalytic turnover efficiencies. Biophysical characterization suggests that these differences may be related to different mechanisms of substrate binding. This study provides useful insights for BChl pathway design and another enzymatic step in the complex pathways leading to (B)Chls.",
"discussion": "Discussion 8-vinyl reductases are widely distributed in photosynthetic organisms. Putative 8-vinyl reductases have been identified in several different domains of life ( Fig. S1 ), where they are suspected to catalyze a key functionalization of porphyrin molecules in (B)Chl biosynthesis [59] . Homologous enzymes exist across different species [36] , as well as within individual species [60] , highlighting the evolutionary selective advantage afforded by the catalytic role of 8-vinyl reductases. Recent studies revealed that homologous plant 8-vinyl reductases are substrate promiscuous, capable of reducing the C8-vinyl group of different pathway intermediates [36] . Data in this report demonstrates that substrate promiscuity is not only limited to plant 8-vinyl reductases, but is also a characteristic of 8-vinyl reductases from photosynthetic bacteria. Here, 8-vinyl reductases did not show a particular preference for any pathway intermediate, but activity was entirely dependent upon presence of the chelated Mg 2+ . These data are in agreement with previous studies, in which the presence of a chelated divalent metal ion was essential [54] , [55] . It is not fully understood why the presence of a chelated Mg 2+ in the porphyrin ring is essential for 8-vinyl reductase activity in this study. It is likely that the metal ion is required for the correct orientation of the porphyrin molecule in the active site or to sterically align the molecule in close proximity with the essential NADPH cofactor [56] . Our data provides some insights into the binding of the Mg-porphyrin substrate, which may affect molecular recognition and therefore catalytic activity. These data provide the first evidence that the mode of binding of substrate(s) varies between homologous 8-vinyl reductases. Whether this difference in binding confers a selective advantage to the host, or whether it provides the enzyme with a means to regulate reaction efficiency with differing substrates [36] , and therefore pathway flux, remains open to question. No data exists to show which residues are involved in binding the substrate or those which are involved in catalysis, which could explain the different modes of binding. Prior to this work, 8-vinyl reductase had not been purified, and limited biochemical data had been published [27] , therefore the mechanistic details of the enzyme remained elusive. While this study has shed some light on the reduction of the C8-vinyl group of BChl intermediates by different 8-vinyl reductases, an in-depth analysis of the step-by-step catalytic process is still required. Detailed comparative biochemical and structural characterization of homologous 8-vinyl reductases would provide the information needed for a full understanding of the reaction mechanism and the substrate recognition of this enzyme. The purpose of this study was to extend the engineered BChl biosynthetic pathway in E. coli \n [5] . It was necessary to insert a downstream enzyme that could accept the multiple products of our existing Mg chelatase-methyl transferase system [15] . In our endeavors to select a suitable candidate for this role, we discovered that CT BciA is capable of reducing the C8-vinyl group of several substrates when expressed as part of our engineered pathway. This is a significant step toward our goal of recreating the full BChl pathway in E. coli . One of the challenges of building a pathway engineered microbe is maintaining balance in pathway flux. Often, a slow catalytic rate of one or more enzymes can result in pathway bottlenecks, or unwanted side reactions can lead to inefficient use of metabolically expensive molecules [61] . We have not yet eliminated the potential for the production of non-Mg-chelated “dead end” porphyrins, although it is likely that the presence of mvP IX ME actually results from loss of the Mg ion from mvMgP IX ME upon extraction from the cell. Nonetheless, the addition of a substrate promiscuous 8-vinyl reductase to our system does provide a shuttle for the Mg-chelated intermediates, resulting in a reduction in the preference to produce P IX ME, and altering the pathway balance to produce equal ratios of the mono-vinyl Mg-chelated porphyrins ( Table 1 ). We were surprised to find that the closely related homologue RS BciA was inactive in our pathway. Furthermore, the in vitro catalytic rate of RS BciA was too slow to be relevant for our engineering purposes, despite extensive attempts to optimize reaction conditions. It may be that RS BciA requires an unknown species-specific cofactor, chaperone, interacting partner or shuttling enzyme for efficient function, which is the case for other enzymes involved in BChl biosynthesis [18] , [62] , [63] . It has been suggested by others that BchJ is not a 8-vinyl reductase as previously indicated [50] , but that it functions as a carrier or shuttle for porphyrin intermediates [18] . However, we found that coexpression of RS BchJ had no effect on RS BciA activity. Future studies beyond gene-knockouts could elucidate the exact nature of BciA behavior in R. sphaeroides , and could clarify whether this particular 8-vinyl reductase is capable of acting alone or whether its activity is upregulated in the presence of certain other members of the pathway or under different reaction conditions [16] . Very recently, an anaerobic 8-vinyl reductase (BciB) from the green sulfur bacterium Chloroherpeton thalassium was characterized. BciB requires two [4Fe-4S] clusters, FAD, and a reductant such as ferredoxin or sodium dithionite to reduce the C-8 vinyl group of DVP [64] . This study reveals the mechanistic diversity of 8-vinyl reductases, and highlights the importance of in vitro characterization for a full appreciation of optimal conditions for catalysis. Furthermore, the presence of a not-yet-identified 8-vinyl reductase in R. sphaeroides cannot be ruled out, as we are only just gaining insights into the sequence diversity of this class of enzymes [60] . Relying upon sequence analyses and the relatively limited biochemical data that existed for 8-vinyl reductases was not sufficient for the strategic design of our engineered BChl system in E. coli . We were not able to predict that CT BciA would be active in our engineered pathway and that RS BciA would be inactive in the same engineered pathway. The unsuitability of RS BciA for our purposes only became truly apparent upon our own detailed biochemical and biophysical characterization of the two enzymes. In some ways, this study serves as a good example to underline the fact that the strategic and streamlined design and engineering of metabolic pathways is heavily dependent upon having a detailed knowledge of the catalytic mechanism and/or three dimensional structure of the enzyme(s) in question [65] . Sometimes this data is not available to the pathway engineer, in which case it becomes necessary to characterize the enzymes involved, and gather the detailed information needed to optimize the system for a particular purpose. Pathway engineering is not a straightforward process of building a chain of enzymes to make a product, rather it is the intricate design, creation and polishing of a living system to entice it to carry out a completely new activity. In conclusion, this study provides an in-depth characterization of two 8-vinyl reductases from two photosynthetic organisms, and gives insights into the potential diversity of function with regards to substrate promiscuity and binding of substrates. This study brings us a step closer to the realization of the creation of an industrially relevant synthetic microbe that can use sunlight as a cheap source of energy to drive the biosynthesis of valuable and designer target molecules."
} | 4,247 |
22925136 | null | s2 | 9,689 | {
"abstract": "Biofilms are core to a range of biological processes, including the bioremediation of environmental contaminants. Within a biofilm population, cells with diverse genotypes and phenotypes coexist, suggesting that distinct metabolic pathways may be expressed based on the local environmental conditions in a biofilm. However, metabolic responses to local environmental conditions in a metabolically active biofilm interacting with environmental contaminants have never been quantitatively elucidated. In this study, we monitored the spatiotemporal metabolic responses of metabolically active Shewanella oneidensis MR-1 biofilms to U(VI) (uranyl, UO(2)(2+)) and Cr(VI) (chromate, CrO(4) (2-)) using non-invasive nuclear magnetic resonance imaging (MRI) and spectroscopy (MRS) approaches to obtain insights into adaptation in biofilms during biofilm-contaminant interactions. While overall biomass distribution was not significantly altered upon exposure to U(VI) or Cr(VI), MRI and spatial mapping of the diffusion revealed localized changes in the water diffusion coefficients in the biofilms, suggesting significant contaminant-induced changes in structural or hydrodynamic properties during bioremediation. Finally, we quantitatively demonstrated that the metabolic responses of biofilms to contaminant exposure are spatially stratified, implying that adaptation in biofilms is custom-developed based on local microenvironments."
} | 357 |
28900114 | PMC5595911 | pmc | 9,690 | {
"abstract": "Artificial bio-based scaffolds offer broad applications in bioinspired chemistry, nanomedicine, and material science. One current challenge is to understand how the programmed self-assembly of biomolecules at the nanometre level can dictate the emergence of new functional properties at the mesoscopic scale. Here we report a general approach to design genetically encoded protein-based scaffolds with modular biochemical and magnetic functions. By combining chemically induced dimerization strategies and biomineralisation, we engineered ferritin nanocages to nucleate and manipulate microtubule structures upon magnetic actuation. Triggering the self-assembly of engineered ferritins into micrometric scaffolds mimics the function of centrosomes, the microtubule organizing centres of cells, and provides unique magnetic and self-organizing properties. We anticipate that our approach could be transposed to control various biological processes and extend to broader applications in biotechnology or material chemistry.",
"conclusion": "Conclusion Overall, we have demonstrated that engineered ferritin nanocages can be designed as biochemical and magnetic scaffolds to nucleate and manipulate mesoscopic bioassemblies in vitro . As magnetic interactions can be contactless, remote controlled, and can deeply penetrate into thick materials; there is a strong interest in using tailored and functionalized magnetic nanoparticles to actuate biological processes upon magnetic actuation 62 – 64 . Integrating such magnetic properties to bio-based scaffolds could open numerous perspectives in nanomedicine, biotechnology, or material chemistry. Our study demonstrates that protein-based magnetic scaffolds can be used to spatiotemporally manipulate biomolecules with magnetic actuation. Our scaffold combines the advantages of being genetically encoded with a synthetic approach, ensuring their modular biofunctionalisation, biocompatibility, inducible self-assembly properties, which overall provides novel opportunities alternative to the use of synthetic magnetic nanoparticles. Our study, aiming to assemble an artificial and magnetic centrosome, raises interesting questions about the impact of multiscale assemblies on functional properties. For instance, we show that the mode of spatial organization of TPX2-ferritins, at the single nanocage level or assembled into micrometric scaffolds, influences the centring properties during microtubule self-organization, which eventually can impact the polarity of the overall cytoskeletal organization. This suggests that the fundamental property of the centring of microtubule asters may also depend on the local organization of the aster pole centre. Furthermore, our modular approach is generic and shows how artificial protein-based organelles can be engineered with emergent functional properties, suggesting its extension to mimic other protein-based organelles or to artificially trigger signalling pathways. Finally, the transposition of our strategy into living cells will be powerful to target and manipulate specific proteins and cellular organelles by combining chemically induced dimerization and magnetic manipulation 23 , 63 , 65 . A fully-genetically encoded strategy, avoiding in vitro biomineralisation and injection into cells, will require novel approaches to catalyse enhanced magnetic phases within cells. This will be necessary to overcome the limited magnetic properties of in vivo expressed ferritins that is a bottleneck for efficient magnetic manipulation 66 , 67 . Beyond the development of new methods for spatiotemporal control, our strategy could also benefit other applications including biomedical imaging, biosensing, drug delivery, or the design of stimuli-responsive materials.",
"introduction": "Introduction In living systems, proteins self-organize into macromolecular assemblies at various length scales to ensure the coordination of numerous biological functions in space and time 1 , 2 . For instance, at the nanometer scale, protein scaffolds are central to trigger by proximity the activation of proteins involved in signal transduction 3 , 4 . At the micrometric scales, multimeric interactions or repetitive interacting domains drive the organization of numerous functional structures, such as cytoskeleton fibres or organelles, with specific functional properties usually not found at the single molecule level 2 , 5 , 6 . Numerous studies are now engaged to establish a clear link between biological multiscale assemblies and emergent functional properties. From this perspective the development of bio-based nanomaterials, produced from the programmed assembly of biomolecules as DNA, RNA, and proteins, offers novel tools to analyse and control the spatiotemporal properties of molecular and cellular processes, but also to engineer novel synthetic functionalities 7 , 8 . For instance, DNA-based scaffolds, which provide very precise biomolecule spatial positioning, have been used to elucidate biophysical mechanisms underlying cytoskeleton motor activity 9 – 11 or as fluorescent biosensors to probe the internal environment of living cells 12 , 13 . Complementary, pioneer studies have demonstrated how synthetic protein scaffolds can modulate the cooperativity of ensembles of molecular motors 14 , and artificially control metabolic flux 15 or signalling pathways 16 – 18 . In this study we present a synthetic protein scaffold that combines specific features found in natural systems, such as multimeric interactions and multiscale assemblies, with novel properties provided by an artificial approach, such as stimulus-triggered assembly 19 and magnetic control 20 – 27 . In this regard, our synthetic protein scaffold, bioengineered from ferritin nanocages, recapitulates several remarkable characteristics: (i) upon chemical stimulation it self-organizes into micrometric structures in vitro , (ii) it can be functionalized with regulatory proteins to trigger specific biochemical responses, and (iii) it displays magnetic properties that could be useful for detection or for magnetic actuation. To illustrate the versatility of our synthetic scaffold, we devised an in vitro assay to study specific features of cytoskeleton spatial organizations, by focusing on the nucleation and the magnetic manipulation of microtubule structures. In particular, the generation of 3D micrometric scaffolds from single functionalized and biomineralised ferritins enables us to mimic Microtubule Organizing Centres, such as the centrosome, and to examine an emergent function of these resulting artificial organelles: the centring property, which is essential to define the polarity of cells 28 , 29 . The iron storage ferritin is a protein that assembles into a nanocage composed of 24-subunits. The capacity of ferritin to catalyse the precipitation of inorganic condensed phases within its internal cavity allows living organisms to control the availability of iron 30 . Several studies have reported modifications of the ferritin cage surface by non-covalent interactions in response to electrostatic interactions 31 , 32 or metal coordination 33 . The oligomeric state of the ferritins has been further exploited to generate controlled multi-scale assemblies 31 , 32 , 34 – 36 . On the other hand, the catalytic activity of the ferritin has been used for developing novel contrast agents in living organisms (magnetic resonance imaging, electron microscopy), nanoheaters for hyperthermia, nanoprobes for biosensing and cell markers, and magnetic actuators for gene expression control (magnetogenetics) 37 – 46 . In this study, we exploit both the multivalent and catalytic properties of ferritin. Our first goal was to engineer ferritin nanocages as building blocks for the production of inducible micrometric protein scaffolds sharing a specific biochemical activity and magnetic properties. To do so, we have designed a strategy for the controlled functionalization of the nanocage surface, by multivalent protein-protein interactions. We have genetically modified the ferritin monomer to use chemically inducible dimerization strategies based on the heterodimerization of FKBP and FRB (Fig. 1 ) 47 – 49 . Then, the catalytic activity of ferritin, by synthesising monodispersed ferric condensed nanoparticles within its cavity, provides specific magnetic properties to the scaffold 43 , 50 – 52 . This modular strategy permits the targeting of proteins of interest at the nanocage surface (Fig. 1a ), but also the formation of 3D clusters of ferritins by triggering multimeric interactions between FKBP- and FRB-ferritin cages and magnetically manipulating them, upon biomineralisation of the participating ferritins (Fig. 1b,c ). Figure 1 Genetically encoded protein scaffolds with modular biochemical and magnetic functions. ( a , b ) Schematic of the modular approach to functionalize and organize ferritin nanocages. ( a ) Chemically inducible dimerization strategy based on the heterodimerization of FKBP and FRB is used to target a protein of interest (POI) at the surface of the ferritin nanocages by addition of Rapamycin. ( b ) The formation of ferritin clusters is triggered by multimeric interactions between FKBP- and FRB-ferritins in presence of Rapamycin. ( c ) Illustration of the functionalization and magnetic manipulation of biomineralised ferritin clusters to control a microtubule assembly in space and time. \n We next examined how ferritin nanocages can serve as biochemical scaffolds to control specific biological processes using Xenopus egg extracts, a powerful cell-free system well suited to study the morphogenetic properties of cytoskeleton networks 53 – 56 . We showed that the nucleation of microtubule fibres and formation into asters can be triggered by individual ferritin nanocages when TPX2, a nucleating promoting factor, is recruited to their surface. Moreover, when organized into micrometric clusters, the TPX2-ferritin scaffolds are integrated into the microtubule asters and localized at their pole centre, which confers the ability to manipulate the aster position with magnetic forces. Our observations also suggest the emergence of different biophysical and self-organizing properties of microtubule asters, depending on the scale of ferritin organization, from single cages (nanometre) to clusters (micrometre). Finally, these results illustrate how TPX2-ferritin scaffolds share some similarities with Microtubule Organizing Centres, found in eukaryotic cells, and suggest their utilization as artificial magnetic centrosomes to manipulate microtubule-based structures."
} | 2,651 |
36790331 | PMC9979639 | pmc | 9,694 | {
"abstract": "Organic waste streams can be converted into high-value\nplatform\nchemicals such as medium-chain carboxylic acids (MCCAs) using mixed\nmicrobial communities via chain elongation. However, the heterogeneity\nof waste streams and the use of complex microbial communities can\nlead to undesirable reactions, thus decreasing process efficiency.\nWe explored suppressing excessive ethanol oxidation to acetate (EEO)\nby increasing the hydrogen partial pressure (P H2 ) through\nhydrogenotrophic methanogenesis inhibition by periodically adding\n2-bromoethanesulfonate (2-BES) to an MCCA-producing bioreactor to\nreach 10 mM of 2-BES upon addition. The bioreactor was fed with pretreated\nfood waste and brewery waste containing high concentrations of short-chain\ncarboxylic acids and ethanol, respectively. While 2-BES addition initially\nreduced EEO, some methanogens ( Methanobrevibacter spp.) persisted and resistant populations were selected over time.\nBesides changing the methanogenic community structure, adding 2-BES\nalso changed the bacterial community structure due to its impact on\nP H2 . While we demonstrated that P H2 could be\nmanipulated using 2-BES to control EEO, methods that do not require\nthe addition of a chemical inhibitor should be explored to maintain\noptimum P H2 for long-term suppression of EEO.",
"introduction": "Introduction Chain elongation of short-chain carboxylic\nacids (SCCAs, C1-C5)\nis an emerging anaerobic biotechnology to produce medium-chain carboxylic\nacids (MCCAs, C6-C12). It involves the stepwise elongation of the\ncarbon chain of SCCAs to MCCAs by two carbons via the reverse β\noxidation pathway. 1 The two-carbon acetyl\ngroup added to SCCAs is derived from ethanol, lactate, or other reduced\ncompounds. Microbial-based chain elongation processes have produced\nMCCAs such as caproate (C6), enanthate (C7), and caprylate (C8). 2 , 3 MCCAs have many industrial and agricultural applications. They can\nbe converted into longer-chain liquid fuels, or used directly as livestock\nfeed additives, antimicrobial agents, corrosion inhibitors, and plant\ngrowth promoters, or as building blocks for producing lubricants,\nfragrances, and dyes. 1 The field\nof waste management has been transitioning from landfill\ndisposal and incineration to utilizing sustainable biotechnologies\nto recover biofuels and biochemicals from organic waste streams. The\nproduction of MCCAs from waste streams using mixed microbial communities\nvia chain elongation 2 − 8 is one example consistent with this trend. One of the challenges\nof mixed-culture fermentation of waste streams is to control competing\nbiochemical pathways that have the potential to take place due to\nthe high diversity of the microbial communities, their broad metabolic\ncapacity, and the heterogeneity of most waste streams. Methanogenesis,\nsulfate reduction to sulfide, excessive ethanol oxidation to acetate\n(EEO), carboxylic acid oxidation, and the acrylate pathway (i.e.,\npropionate formation from lactate during lactate-mediated chain elongation)\nare some of the competing pathways that can affect chain elongation\nefficiency. 9 − 11 In the ethanol-driven chain elongation process, for\nevery six molecules of ethanol, one molecule of ethanol is anaerobically\noxidized into acetate to harvest one ATP via substrate-level phosphorylation. 1 Oxidation of ethanol to acetate at a proportion\nhigher than one out of six molecules can occur along with the chain\nelongation process and has been termed EEO. 10 It is important to suppress EEO to ensure efficient use of ethanol,\nespecially if costly synthetic ethanol is used. Diversion of ethanol\ntoward EEO can reduce the amount of acetyl-CoA available for chain\nelongation of SCCAs. Furthermore, acetate produced through EEO acidifies\nthe medium leading to higher alkalinity consumption. EEO can be beneficial\nwhen the acetate produced via EEO is subsequently used in chain elongation\nand is referred to as ethanol upgrading. 10 However, ethanol upgrading leads to inefficient use of ethanol. 10 For example, ethanol upgrading to MCCAs via\nEEO consumes three moles of ethanol for every mole of caproate produced,\nwhereas the reverse β oxidation pathway requires 2.4 moles of\nethanol to elongate acetate to one mole of caproate. EEO has\nbeen identified in several studies as an undesirable reaction, 4 , 6 , 10 and thus an adequate control\nstrategy needs to be developed. Anaerobic ethanol oxidation to acetate\nhas a positive standard Gibbs free energy of 49.6 kJ mole –1 reaction ( Table S1, eq S1 ). This reaction\nis energetically feasible only when the partial pressure of hydrogen\n(P H2 ) is low. The need for H 2 removal results\nin a syntrophic association between the H 2 -producing ethanol\noxidizers and hydrogenotrophic methanogens ( Table S1, eqs S1–S3 ) or other H 2 consumers. 12 Some studies have limited EEO by controlling\nthe CO 2 loading rate, which indirectly controls P H2 via hydrogenotrophic methanogenesis by limiting CO 2 availability. 4 , 10 Another way to control EEO is to inhibit hydrogenotrophic methanogenesis,\none of the major H 2 -consuming pathways in anaerobic systems,\nby adding methanogenic inhibitors. The most widely used methanogenic\ninhibitor in various applications is 2-bromoethanesulfonate (2-BES), 13 a structural analog of coenzyme M (CoM), the\nmethyl carrier in the final step of methanogenesis. 2-BES and other\nmethanogenic inhibitors have been used in previous chain elongation\nstudies. 7 , 14 − 18 However, these studies focused on the effect of such\ninhibitors on the suppression of methane production from acetate,\nthus preventing the consumption of acetate, an MCCA precursor, and\ngiving a competitive advantage to chain elongating microorganisms.\nAs 2-BES also inhibits hydrogenotrophic methanogenesis, it can influence\nmetabolic pathways affected by the P H2 . A study conducted\nto investigate the effects of 2-BES and chloroform on anaerobic bacterial\ncommunities showed that the use of 2-BES affected the growth of syntrophic\nbacteria (e.g., Syntrophomonas and Syntrophobacter ) and homoacetogenic bacteria (e.g., Moorella ) due\nto the accumulation of H 2 . 19 Similarly, it can be expected that EEO can be altered by 2-BES addition\ndue to its inhibition of hydrogenotrophic methanogens and the subsequent\nthermodynamic inhibition caused by high P H2 . The\nobjective of this study was to evaluate the effect of P H2 on EEO using 2-BES to inhibit H 2 consumption\nby hydrogenotrophic methanogens during MCCA production from pretreated\nfood waste and brewery waste. The inhibition was evaluated by monitoring\nmethane production, P H2 , ethanol consumption, and acetate\nproduction as well as by monitoring long-term changes in bacterial\nand archaeal population dynamics due to 2-BES addition. The knowledge\nobtained from this study can help assess the impact of P H2 in controlling undesirable reactions such as EEO during chain elongation\nand enhance substrate utilization toward MCCA production.",
"discussion": "Results and Discussion 2-BES Temporarily Suppressed Excessive Ethanol Oxidation to\nAcetate During the 339 days of operation, MCCAs were produced\nat an average rate of 4.4 ± 1.6 mmole L –1 d –1 with a maximum volumetric production rate of 9.1\nmmole L –1 d –1 ( Figure S1 ). Caproate was the dominant MCCA produced, comprising\n62.7 ± 8.7% (on a carbon basis) of the total MCCAs, while enanthate\nand caprylate constituted on average 30.5 ± 8.5% and 6.8 ±\n3.7% of the total MCCAs produced, respectively. Neither ethanol nor\nSCCAs were completely consumed, suggesting that their concentrations\nwere not limiting. Ethanol, constituting on average 75% of the total\ninfluent COD, was oxidized to acetate. The acetate accumulated in\nthe system reaching a maximum concentration of 156.6 mM on Day 9 ( Figures 1 a and S2 ), and it was not further elongated into MCCAs\ndespite sufficient ethanol availability. Roghair et al. 10 demonstrated that acetate derived from EEO could\nbe involved in chain elongation, so it is unclear why acetate continued\nto accumulate in our study. However, it should be noted that there\nwere several differences in operational conditions (pH 6.8 vs pH 5.5,\ntemperature 30 vs 37 °C, HRT of 17 h vs HRT 2–4 days)\nand inoculum source (granular and suspended chain elongation sludge\nvs rumen content) between the Roghair et al. 10 paper and our study that could have affected the microbial community\nand thus acetate chain elongation. Figure 1 Effluent acetate, net acetate, and influent\nacetate concentrations\n(a), hydrogen partial pressure (P H2 ) and net acetate concentration\n(secondary y -axis) (b), and daily gas composition\nafter 2-bromoethanesulfonate (2-BES) addition (c) over time in the\nbioreactor. The dashed lines represent 2-BES additions. Since EEO becomes thermodynamically unfavorable\nat high P H2 , 4 2-BES was added\nto suppress H 2 -consuming methanogens and thus inhibit EEO.\nP H2 in the\nbioreactor headspace averaged 2.7 × 10 –3 ±\n2.9 × 10 –3 atm from Days 0 to 229 before 2-BES\nwas added. This P H2 is still higher than the P H2 required for SCCAs (1.45 × 10 –4 atm for acetate,\n6.65 × 10 –6 atm for butyrate) and MCCAs (2.52\n× 10 –6 atm for caproate) oxidation via β\noxidation. 5 The observed increase in P H2 to levels as high as 0.44 atm on Day 251 ( Figure 1 b) after 2-BES addition suggested\nthat H 2 consumption by hydrogenotrophic methanogens was\ninhibited. This increase in P H2 made EEO thermodynamically\nunfavorable leading to a decrease in acetate concentration ( Figure 1 b) and a reduction\nin ethanol consumption. MCCAs were consistently produced, and their\nproduction did not appear to be affected negatively by this high P H2 ( Figure S1 ). In fact, the total\nMCCA volumetric production rate significantly increased from 4.6 ±\n1.8 to 5.5 ± 1.2 mmole L –1 d –1 after 2-BES addition ( p = 5.8 × 10 –3 , Figure S1 ). As ethanol was diverted\nfrom EEO, the higher availability of ethanol might have favored higher\nMCCA production. Therefore, our results demonstrated that chain elongation\nstill happened at a P H2 sufficiently high to suppress EEO\nand that maintaining a certain P H2 in a chain elongation\nsystem may be an effective EEO control strategy. As 2-BES gradually\nwashed out of the system ( Figure S3 ), P H2 decreased and the net acetate concentration\nincreased, indicating reduced inhibition. The P H2 again\nincreased after additional 2-BES was introduced, with a corresponding\nreduction in the net acetate concentration. This trend continued until\nDay 268, after which P H2 decreased despite six more 2-BES\nadditions indicating that the 2-BES-induced inhibition was short-lived.\nP H2 significantly decreased ( p = 6.78\n× 10 –5 ) from average values of 0.16 ±\n0.15 atm from Days 230 to 268 to 4.6 × 10 –3 ± 2.3 × 10 –3 atm from Days 269 to 339.\nThe corresponding net acetate concentrations also increased from an\naverage of 35.5 ± 21.9 mM (Days 230–268) to an average\nof 78.3 ± 17.8 mM (Days 269–339). While the average P H2 from Days 269 to 339 was slightly higher than the average\nP H2 before the 2-BES addition had started, P H2 was not high enough to suppress EEO. Although Grootscholten et al. 4 reported that a P H2 above 0.03 atm\nwas needed to control EEO, we observed EEO suppression at a P H2 higher than 0.02 atm. Besides P H2 , the thermodynamic\nfeasibility of the EEO reaction is also affected by in situ conditions\nsuch as pH and temperature. Theoretical thermodynamic calculations\nshow that the higher pH of 6.5–7.0 used by Grootscholten et\nal. 4 may explain the higher P H2 required for EEO inhibition in their study, compared to our study,\nwhich used a pH of 5.5; the slight difference in temperature between\nthe two studies did not have an impact ( Figure S4 ). There is little information available on the microorganisms\nresponsible for EEO during chain elongation, so further work is needed\nto study how metabolic triggers for EEO play out for different populations\ninvolved. Methanobrevibacter Dominated Despite the Addition\nof a Methanogenic Inhibitor We monitored 16S rRNA to study\nshort-term changes in microbial activity induced by 2-BES. Even though\nusing this approach to estimate activity has biases, 38 our use of both 16S rRNA and 16S rRNA gene sequencing data\nand comparing trends over time provides helpful insights into the\nmicrobial community responses to 2-BES additions. Using 16S rRNA sequencing\ndata, we determined that more than one-fifth (22.0 ± 5.2%) of\nthe active microbial community was composed of the phylum Euryarchaeota and was dominated by hydrogenotrophic methanogens.\nAceticlastic methanogens were not detected, indicating that they were\ninhibited by the low bioreactor pH of 5.5. Another study has also\nshown that aceticlastic methanogens are more sensitive to lower pH\nthan hydrogenotrophic methanogens and may also be inhibited to a greater\nextent by undissociated SCCAs and MCCAs. 5 16S rRNA and 16S rRNA gene sequencing data indicated that Methanobrevibacter was the dominant methanogenic genus at\neach sampling time point ( Figure 2 ). The relative activity (as determined by 16S rRNA\nsequencing) of methanogens was higher than their relative abundance\n(as determined by 16S rRNA gene sequencing) for each time point. Over\nthe period from Day 0 to Day 229, i.e., before 2-BES addition, the\nrelative abundance of Methanobrevibacter spp. averaged\n9.6 ± 3.4%, while their relative activity averaged 19.7 ±\n5.7%. Figure 2 Relative abundance (a) and activity (b) of methanogens identified\nto the genus level in the bioreactor samples over time using the amplicon\nsequence variant (ASV)-based approach. Red dashed lines represent\nthe start and end of wasting bioreactor content on Days 20 and 82,\nrespectively, for controlling solids retention time, and black dashed\nlines represent 2-bromoethanesulfonate (2-BES) additions. After the first 2-BES addition on Day 230, the\nrelative abundance\nof methanogens decreased from 9.5% on Day 228 to 1.9% on Day 234 ( Figure 2 a). The decrease\nin relative activity was even more pronounced (from 21.8% on Day 228\nto 4.1% on Day 234, Figure 2 b). Consistent with these observations, methane production\ndecreased after 2-BES addition ( Figure 1 c) and the average methane yield decreased significantly\nfrom 7.4 ± 2.4 to 5.3 ± 3.3% of the total sCOD fed (p =\n1.8 × 10 –3 ) ( Figure S5 ). The relative abundance of methanogens remained low for several\nweeks, but their relative activity increased substantially soon after\nthe first 2-BES addition ( Figure 2 ). These microbial data confirm that the periodic addition\nof 2-BES was ineffective in inhibiting methanogens over time, as suggested\nby the decrease in P H2 levels ( Figure 1 b). These long-term trends are further supported\nby the finding that there was no significant change in the average\nrelative abundance ( p = 0.744) and relative activity\n( p = 0.23) of methanogens before and after the start\nof 2-BES addition on Day 230. The archaeal diversity (both Shannon\nindex and Pielou’s\nevenness) and richness (observed ASVs) were compared before and after\nthe start of 2-BES addition ( Figure S6 ).\nThe mean number of archaeal ASVs decreased significantly ( p = 1.6 × 10 –6 ) from 23 ± 5\nto 12 ± 3 after 2-BES addition, and a similar decrease was observed\nfor the active archaeal ASVs ( p = 1.9 × 10 –2 ). The mean Shannon index and Pielou’s evenness\nof the total and active archaeal community also consistently decreased\nafter 2-BES addition, but the decrease was not always statistically\nsignificant. The archaeal community structures based on both 16S rRNA\ngene and 16S rRNA sequencing distinctly differed before and after\n2-BES addition, as shown by the Bray–Curtis dissimilarity analysis\n( Figure S7 ; 78 and 71% dissimilarity, respectively).\nHigh ANOSIM R values of 0.69 ( p = 0.001) and 0.47\n( p = 0.001) also indicated significant changes in\narchaeal community composition and activity, respectively, due to\n2-BES addition. Hydrogenotrophic Methanogenesis Was the Major H 2 -Consuming\nPathway Regardless of the effectiveness of 2-BES addition,\ninhibiting methanogens may not be sufficient to control the P H2 as there are other H 2 sinks besides hydrogenotrophic\nmethanogenesis in anaerobic processes. For example, H 2 can\nbe used by sulfate-reducing microorganisms or by homoacetogens for\nacetate production. While sulfate was not detected in the influent\n(data not reported), the sulfonate moiety of 2-BES can also serve\nas an electron acceptor for sulfate-reducing bacteria and thus support\ntheir growth. 39 However, sulfate-reducing\nbacteria (e.g., Desulfovibrio spp.) were present\nat a relative abundance and activity of less than 0.1% both before\nand after 2-BES addition. Similarly, homoacetogenesis was not observed\nin the bioreactor (discussed in detail below). Since the P H2 increased with a simultaneous decrease in methane production ( Figure 1 c) and relative abundance\nand activity of methanogens ( Figure 2 ) after the first few 2-BES additions, methanogenesis\nappeared to be the major pathway for H 2 consumption and\ndirectly affected EEO. As 2-BES inhibition of methanogens was short-lived,\nthe low P H2 in the bioreactor toward the end of the experiment\nagain favored EEO. Periodic Addition of 2-BES Selected for Resistant Methanogens The effective inhibitory concentration of 2-BES differs (10–50\nmM) for different methanogens and environmental conditions. 13 For example, aceticlastic methanogens are more\nsusceptible to 2-BES inhibition than hydrogenotrophic methanogens. 22 , 40 Several studies have reported the presence of methanogens after\nthe addition of 2-BES, 14 , 15 which may be due to differences\nin cell envelopes resulting in the varying ability to uptake inhibitors\nand differences in CoM transport rates. 13 Some methanogens can adapt to 2-BES through a loss of cell permeability\nto 2-BES and the selection of 2-BES resistant strains. 41 2-BES is a structural analog of CoM, the\nmethyl carrier in the final step of methanogenesis, which catalyzes\nthe reduction of the methyl group to methane by methyl CoM reductase. 13 Methanogens that can synthesize CoM do not depend\non external CoM and likely are more resistant to 2-BES. 42 For example, Methanobrevibacter\nsmithii can synthesize CoM, whereas Methanobrevibacter ruminantium M1 requires an external\nsource of CoM for growth. 43 , 44 In our study, Methanobrevibacter spp. represented the highest fraction\nof the total and active archaeal community throughout the period with\n2-BES addition ( Figure 2 ), possibly indicating that they could synthesize CoM, making them\nresistant to 2-BES. However, there is no direct evidence to support\nthis hypothesis. ASVs 5 and 29 were the two dominant Methanobrevibacter ASVs observed after 2-BES addition started.\nAs discussed in our\nprevious publication 3 and shown in Figure 3 , different Methanobrevibacter populations were prevalent during periods\nwith different SRTs. Specifically, Methanobrevibacter ASV 5 became more prevalent after the SRT was reduced on Day 20\n( Figure 3 ). These results\nsuggest that ASV 5 has a faster growth rate than other Methanobrevibacter populations (such as ASVs 20 and 12), allowing its growth and retention\nwhen operated at a short SRT. ASV 5 reappeared after 2-BES was added\n( Figure 3 ), suggesting\nthat the higher growth rate of this population combined with the likely\nability to synthesize CoM conferred resistance toward 2-BES and allowed\nits growth. In addition to ASV 5, Methanobrevibacter ASV 29 appeared to be resistant to 2-BES ( Figure 3 ). ASV 29, which was not detected in most\nsamples before 2-BES addition, started appearing after Day 230 when\n2-BES was added. The relative abundance and activity of ASV 29 increased\nfrom 0.8 ± 0.6 and 1.9 ± 1.5% during Days 234–262\nto 6.9 ± 2.8% and 4.7 ± 1.0% during Days 270–339,\nrespectively. This increase in relative abundance and activity of\nASV 29 aligns with the observation that P H2 remained low\ndespite frequent 2-BES additions ( Figure 1 b), showing decreased inhibition of some\nmethanogens. The SIMPER analysis also indicated that ASVs 5, 12, 20,\nand 29 contributed to most of the differences (>63%) observed between\nthe active archaeal community before and after 2-BES addition ( Figure S7b ). Therefore, the periodic addition\nof 2-BES likely provided a selective pressure to allow 2-BES-resistant\npopulations of Methanobrevibacter (i.e., ASVs 5 and\n29) to become abundant over time. Figure 3 Relative abundance (a) and activity (b)\nof methanogen amplicon\nsequence variants (ASVs) in the bioreactor samples over time. Only Methanobrevibacter ASVs discussed in the text are shown,\nwhile the remaining methanogen ASVs are grouped in the “Others”\ncategory. The four Methanobrevibacter ASVs shown\ncomprised 51.0–97.5% and 60.8–97.6% of the total 16S\nrRNA gene and 16S rRNA sequences, respectively. The red dashed lines\nrepresent the start and end of bioreactor content wasting on Days\n20 and 82, respectively, for controlling solids retention time (9.7\n± 5.8 days), and the black dashed lines represent 2-bromoethanesulfonate\n(2-BES) additions. Note that the data up to Day 229 in this figure\nwere previously reported 3 and that any\nobserved reactor performance data could not explain the disparity\nin the Day 270 data. The sequence data were also analyzed with an OTU-based\napproach\nusing mothur ( Figures S8 and S9 ). Both\nOTU- and ASV-based approaches produced similar trends in changes of\nrelative abundance and relative activity due to 2-BES addition ( Figures 2 vs S8 and Figures 3 vs S9 ). However, the OTU\nmethod indicated that a single Methanobrevibacter OTU, i.e., OTU 6, was the primary population able to grow in the\npresence of 2-BES ( Figure S9 ), in contrast\nto the ASV method, which indicated that two major Methanobrevibacter populations (ASVs 5 and 29) became dominant after 2-BES addition.\nComparing the two methods shows that the ASV method could differentiate\nsequence variants down to a single nucleotide difference, thus providing\nimproved taxonomic resolution. 45 The OTU\napproach clusters 16S rRNA gene sequences with 97% similarity into\nthe same OTU, and representative OTU sequences are compared with sequences\nin a reference database for taxonomic identification. The increased\nresolution provided by the ASV-based approach captured changes in\nmethanogen community structure in response to the 2-BES inhibition\nnot picked up by the OTU-based approach. At the same time, the ASV\nmethod has some disadvantages, including the inability to effectively\ndiscriminate between PCR bias or sequencing error and real biological\nvariation. Furthermore, many microorganisms harbor multiple rrn operons, and the 16S rRNA gene sequence diversity increases\nwith increasing rrn operon copy numbers. 46 In the case of intragenomic heterogeneity, multiple\nASVs can arise from a single population harboring multiple rRNA gene\ncopies leading to more ASVs than populations present in a community. 45 , 47 Methanobrevibacter spp. have two or three 16S rRNA\ngene copies and intragenomic sequence variation could lead to multiple\nASVs from the same Methanobrevibacter population. 48 So, it is possible that DADA2 assigned divergent\ncopies of the 16S rRNA gene that belonged to one Methanobrevibacter population into ASV 5 and ASV 29. Comparing metagenome-assembled\ngenomes of Methanobrevibacter spp. before and after\n2-BES addition would resolve this uncertainty, but this analysis is\nbeyond the scope of this study. Syntrophic Ethanol Oxidation to Acetate Was Primarily Responsible\nfor Acetate Production The high relative abundance and activity\nof Methanobrevibacter populations indicated that\na favorable ecological niche was created that supported their growth\nand activity even at low pH and during exposure to 2-BES. Some studies\nhave pointed toward the versatility of methanogens in substrate utilization\nfor methanogenesis. 49 For example, alcohols\nsuch as ethanol can be utilized for growth and methane production. 50 − 52 Bryant et al. 53 found a syntrophic association\nbetween an H 2 -producing ethanol oxidizer and an H 2 -utilizing microorganism such that ethanol oxidation was coupled\nwith interspecies H 2 transfer for methane production. Some\nstudies have reported that species of methanogens, such as M. ethanolicus and M. organophilum , can directly convert ethanol to methane and acetate. 29 , 30 Two moles of ethanol were oxidized to two moles of acetate for every\nmole of methane formed ( Table S1, eq S4 ). Phylogenetic analysis showed that the dominant Methanobrevibacter populations, ASVs 5 and 29, clustered with Methanobrevibacter\nwolinii strain SH, ASV 20 with Methanobrevibacter\nboviskoreani JH1, and ASV 12 with Methanobrevibacter sp. AbM4 ( Figure 4 ). Closely related Methanobrevibacter strains (e.g., M. wolinii strain DSM 11976 T and M. boviskoreani strain DSM 25824 T ) have\nthe genes to utilize ethanol for methanogenesis 50 and Methanobrevibacter sp. AbM4 is capable\nof growth without H 2 but in the presence of methanol/ethanol. 28 Figure 4 also shows that Methanobrevibacter ASVs\nwere phylogenetically relatively closely related to Methanosphaera sp. WGK6, which are methanogenic ethanol oxidizers found in the\nforegut of macropodids (e.g., kangaroos). 31 While these results suggest the possible involvement of methanogens\nin ethanol metabolism, obtaining additional evidence to verify the\nrole of Methanobrevibacter in EEO observed in our\nbioreactor was beyond the scope of this study. Future research should\nfocus on using quantitative PCR or multi-omics tools to retrieve genome-level\ninformation to confirm if Methanobrevibacter populations\nharbor any genes required for EEO. Figure 4 Phylogenetic tree of 16S rRNA gene sequences\nof most abundant methanogenic\namplicon sequence variant (ASVs, in red). Methanopyrus\nkandleri was used as the outgroup. GenBank accession\nnumbers are given in parentheses. Reference sequences are shown in\nblack. Methanogens previously identified as ethanol oxidizers or capable\nof growth in the presence of ethanol are shown in blue. The numbers\nat the nodes of the branch indicate bootstrap values. The scale bar\nof 0.05 represents 5% substitutions per nucleotide base pair. Acetogens can also carry out ethanol oxidation\nusing CO 2 as an electron acceptor with no thermodynamic\nrestriction ( Table S1, eq S5 ). ASVs belonging\nto the genus Acetobacter were consistently present\nthroughout bioreactor\noperation but at low relative abundance (1.4 ± 1.2%) and relative\nactivity (1.4 ± 1.4%). Acetobacter is a typical\nacetic acid bacterium characterized by its ability to convert ethanol\nto acetate in the presence of oxygen. 54 While acetic acid bacteria such as Acetobacter are\nthought to be strict aerobes, their ability to use electron acceptors\nother than oxygen suggests that they may be metabolically active under\nanaerobic conditions. 55 However, the relative\nabundance and activity of Acetobacter were not significantly\ncorrelated to acetate production in the bioreactor, which further\nindicates the involvement of other microbial populations in EEO. Homoacetogens can also produce acetate from CO 2 and\nH 2 ( Table S1, eq S6 ); however,\nmethanogens generally have a higher affinity for H 2 than\nhomoacetogens making H 2 consumption by methanogenesis more\ncompetitive than reductive homoacetogeneis. 56 An ASV that shared 98% similarity with Eubacterium\naggregans , a homoacetogenic bacterium, 57 was observed at a very low relative abundance\nand relative activity of less than 0.01%. Other homoacetogens, such\nas Acetobacterium spp. ( e.g ., Acetobacterium carbinolicum ( 58 )), which can combine ethanol oxidation to acetate with concomitant\nacetate formation from carbon dioxide, were not observed in the bioreactor.\nFurthermore, the Gibbs free energy of hydrogenotrophic methanogenesis\nwas exergonic over the bioreactor operating period, whereas homoacetogenesis\nwas endergonic for part of the operating time ( Figure S10 ), consistent with the low relative abundance and\nactivity of homoacetogens. If 2-BES had favored homoacetogens, the\nacetate yield should have improved. However, net acetate production\ndecreased after 2-BES addition showing that acetate production via\nhomoacetogenesis was not feasible in the bioreactor. These observations\nconfirm that syntrophic ethanol oxidation to acetate ( Table S1, eq S3 ), which is suppressed when methanogens\ndo not consume H 2 , was the most favorable pathway for acetate\nproduction under the bioreactor conditions. However, identifying the\nmicrobial groups responsible for EEO was not possible within the scope\nof this study. Implications for Ethanol Chain Elongation with Mixed Microbial\nCommunities While controlling competing metabolic processes\nsuch as EEO is challenging when heterogeneous waste streams are fed\nto a mixed community bioreactor, it is important to limit inefficient\nsubstrate usage and optimize MCCA yield and selectivity. Hydrogenotrophic\nmethanogenesis was the critical process leading to H 2 consumption\nas other H 2 sinks, such as homoacetogenesis and sulfate\nreduction, were limited under the bioreactor conditions. The addition\nof the methanogenic inhibitor 2-BES limited the activity of H 2 -consuming methanogens and minimized EEO due to thermodynamic\ninhibition caused by high P H2 . We observed that EEO was\nlimited under P H2 higher than 0.02 atm. However, the addition\nof 2-BES did not provide long-term EEO suppression. The periodic addition\nof 2-BES created a selective environment for the microbial community\nmaking 2-BES inhibition ineffective. It would be valuable to evaluate\nthe effect of the continuous addition of 2-BES on EEO and the microbial\ncommunity dynamics. The high cost of chemical additives could most\nlikely increase the operating cost of MCCA production, making the\naddition of methanogenic inhibitors challenging to scale up. Besides\nusing chemical inhibitors, other strategies such as decreasing the\nSRT to wash out slow-growing methanogens, heat-shock pretreatment\nof the inoculum, and maintaining low pH could be evaluated to test\nits effectiveness for long-term methanogenesis inhibition. While this\nstudy did not address the cost and environmental impact of using a\nchemical inhibitor, our approach can be used to evaluate whether controlling\nP H2 is a reliable operational strategy to ensure long-term\ninhibition of EEO. Methanobrevibacter was dominant\nthroughout the operational period, even under 2-BES-inhibited conditions.\nFuture studies should evaluate whether methanogens such as Methanobrevibacter have other roles in ethanol chain elongation\nbesides methanogenesis. Finally, future research needs to identify\nthe microbial populations involved in EEO and study their growth characteristics\nto devise alternate operational strategies to control EEO."
} | 7,743 |
24747284 | null | s2 | 9,697 | {
"abstract": "Organic chemists and metabolic engineers use largely orthogonal technologies to access small molecules like pharmaceuticals and commodity chemicals. As the use of biological catalysts and engineered organisms for chemical production grows, it is becoming increasingly evident that future efforts for chemical manufacture will benefit from the integration and unified expansion of these two fields. This review will discuss approaches that combine chemical and biological synthesis for small molecule production. We highlight recent advances in combining enzymatic and non-enzymatic catalysis in vitro, discuss the application of design principles from organic chemistry for engineering non-biological reactivity into enzymes, and describe the development of biocompatible chemistry that can be interfaced with microbial metabolism."
} | 207 |
20526380 | null | s2 | 9,698 | {
"abstract": "We characterized the planular-zooxanthellae symbiosis of the coral Pocillopora damicornis using criteria that are familiar in studies on corals. Similar to adult corals, planulae exhibited photoacclimation, as changes in symbiont chlorophyll a (chl a); changes in the light-saturation constant for photosynthesis (I(k)); and, at insufficient light, fewer zooxanthellae, decreased respiration, increased weight loss, and increased sensitivity to photoinhibition. Numbers of zooxanthellae in newly-released planulae varied by at least three-fold within broods. Planulae with low versus high numbers of zooxanthellae (termed pale versus dark planulae, respectively) did not differ in symbiont chl-a content, I(k), or biomass-specific rate of dark respiration. Pale planulae had lower rates of photosynthesis, but this difference vanished after three weeks, when zooxanthellar numbers increased by 225% in pale planulae and by 31% in dark planulae. Numbers of zooxanthellae also increased significantly in planulae cultured in ammonium-enriched seawater; ammonium also apparently prevented weight loss and induced settlement. Approximately 70% of photosynthetically-fixed carbon (labeled using (14)C) apparently was translocated from the zooxanthellae to their host. A comparison of planulae cultured at 0.3% versus 11% sunlight suggested that photosynthesis provided ~ 31% of the energy utilized by the latter. Overall, we conclude that the physiology of symbiosis in planulae of P. damicornis is broadly similar to symbiosis physiology in adult corals."
} | 387 |
38531061 | PMC11165533 | pmc | 9,705 | {
"abstract": "Abstract Multi‐functional actuation systems involve the mechanical integration of multiple actuation and sensor devices with external energy sources. The intricate combination makes it difficult to meet the requirements of lightweight. Hence, polypyrrole@graphene‐bacterial cellulose (PPy@G‐BC) films are proposed to construct multi‐responsive and bilayer actuators integrated with multi‐mode self‐powered sensing function. The PPy@G‐BC film not only exhibits good photo‐thermoelectric (PTE) properties but also possesses good hydrophilicity and high Young's modulus. Thus, the PPy@G‐BC films are used as active layers in multi‐responsive bilayer actuators integrated with self‐powered sensing functions. Here, two types of multi‐functional actuators integrated with self‐powered sensing functions is designed. One is a light‐driven actuator that realizes the self‐powered temperature sensing function through the PTE effect. Assisted by a machine learning algorithm, the self‐powered bionic hand can realize intelligent gesture recognition with an accuracy rate of 96.8%. The other is humidity‐driven actuators integrated a zinc‐air battery, which can realize self‐powered humidity sensing. Based on the above advantages, these two multi‐functional actuators are ingeniously integrated into a single device, which can simultaneously perform self‐powered temperature/humidity sensing while grasping objects. The highly integrated design enables the efficient utilization of environmental energy sources and complementary synergistic monitoring of multiple physical properties without increasing system complexity.",
"conclusion": "3 Conclusion In summary, the free‐standing PPy@G‐BC film with a multilayer network structure was easily fabricated by a low‐cost method combining vacuum filtration and in‐situ polymerization. The strong physical and chemical cross‐linking between graphene, BC, and PPy endows the PPy@G‐BC film with excellent tensile strength (93.53 MPa) and Young's modulus (5.55 GPa), which is an important basis for the actuation deformation. Moreover, the PPy@G‐BC film possesses good PTE properties as a photo‐thermal layer (Seebeck coefficient of 42.8 µV K −1 ) and good hydrophilicity as a humidity‐sensitive layer (WAC of 37.9°). Based on these excellent comprehensive properties of the PPy@G‐BC film, two types of actuators with self‐powered and multi‐mode sensing functions were developed. One is the light‐driven PPy@G‐BC/BOPP actuator that deforms based on the asymmetric thermal expansion effect. The PPy@G‐BC/BOPP actuator can spontaneously generate thermoelectric voltage, thus reflecting its own deformation state and temperature changes in real‐time. A self‐powered bionic hand was designed based on this light‐driven actuator, which can perform various complex gestures. With the assistance of SVM algorithms, various gestures made by this bionic hand can be accurately recognized with an accuracy of 96.8%. The other is the humidity‐driven PPy@G‐BC/BOPP actuator that deforms based on the asymmetric humidity expansion effect. The PPy@G‐BC/BOPP actuator realizes self‐powered monitoring of its own deformation state and humidity by in situ integrating a flexible zinc‐air battery. The actuation deformation and self‐powered sensing functions of the two aforementioned actuators are in good consistency without interfering with each other. Finally, with the complementary synergy of these two actuators, the intelligent gripper can not only move objects under the dual stimulation of light and humidity but also realize the multi‐mode (temperature/humidity) sensing function. The ingenious strategy of simultaneously realizing multi‐responsive actuation and self‐powered sensing in a single device will advance the fields of flexible electronics, soft robots, human‐machine interaction, and environmental monitoring.",
"introduction": "1 Introduction The continuous development and innovation of artificial intelligence technology has promoted the wide application of smart actuators with excellent actuation performance in fields such as flexible electronics, [ \n \n 1 \n , \n 2 \n , \n 3 \n \n ] soft robots, [ \n \n 4 \n , \n 5 \n , \n 6 \n \n ] and biomimetic applications. [ \n \n 7 \n , \n 8 \n , \n 9 \n \n ] Due to the increasingly complex tasks required to be performed by the new generation of intelligent robots, the integration of various practical functions on traditional actuators is an inevitable path for the future development of smart actuators. Some researchers have attempted to integrate sensing functions on flexible actuators to monitor the external stimuli received in real time by collecting feedback electrical signals. [ \n \n 10 \n , \n 11 \n , \n 12 \n , \n 13 \n \n ] However, the portability and monitoring duration of these actuation systems integrated with sensing functions inevitably deteriorates because of the energy consumption associated with complex external power supplies and connecting wires. As a novel micro‐power generator, the triboelectric nanogenerator (TENG) can convert the charge in the tribolayer into an electrical signal through a contact‐separation mode. [ \n \n 14 \n , \n 15 \n \n ] TENG's in‐depth research provides a new solution to construct all‐in‐one actuators with both actuation and sensing functions. As a result, there have been several studies reported on integrating TENG inside or outside of photo‐thermal actuators, pneumatic actuators, and vapor actuators for the preparation of mechanical grippers with self‐powered sensing functions. [ \n \n 16 \n , \n 17 \n , \n 18 \n \n ] For instance, Wang et al. designed a polyethylene terephthalate‐carbon black ink‐polydimethylsiloxane actuator with excellent robustness and contact feedback that can simulate a frog's tongue and mechanical gripper to bend and output triboelectric voltage under light‐induced. [ \n \n 16 \n \n ] Chen et al. prepared a sensorized pneumatic gripper using silicone rubber and conductive sponge material, which can recognize the size and weight of the gripped object through self‐powered voltage. [ \n \n 17 \n \n ] However, the application range of TENG‐based self‐powered actuator devices is restricted due to their high preparation costs and the necessity for dynamic contact to generate electrical signals. In contrast, the photo‐thermoelectric generator (PTEG) is a passive energy conversion device working based on the Seebeck effect, which can convert thermal energy from the environment into electrical energy contactless under the condition of temperature difference. Several researchers have combined PTEG in situ on photo‐thermal actuators to prepare flexible photo‐thermal actuators with self‐powered sensing functions. [ \n \n 19 \n , \n 20 \n , \n 21 \n \n ] For example, Chen et al. prepared a graphite/paper/thermochromic dye photo‐thermal material by the pencil‐on‐paper method and applied it to a light‐driven actuator with an integrated self‐powered/visual dual‐mode sensing function and rewritable display function. [ \n \n 19 \n \n ] Weng et al. proposed a Ti 3 C 2 T X ‐based composite modified by bamboo nanofibers, based on which a light/electro‐driven actuator was developed to accomplish Marangoni floating and self‐powered temperature sensing. [ \n \n 21 \n \n ] Benefiting from the wireless actuation characteristics of the light source, the entire energy conversion process can be easily controlled without external force. In addition, PTEGs share similar actuation and energy utilization modes with light‐driven actuators based on the deformation of the asymmetric photo‐thermal expansion mechanism. [ \n \n 22 \n \n ] Nevertheless, none of the above studies were able to realize both multi‐responsive actuation and multi‐mode self‐powered sensing functions in a single device. Integrating multi‐responsive actuation and self‐powered sensing in a single device not only allows the device to more fully utilize the energy resources in the environment (e.g., light energy, moisture, etc.) but also enables complementary monitoring of multiple physical properties of itself (e.g., temperature, water content, etc.) without increasing the complexity of the device. Recently, the flexible zinc‐air battery consisting of flexible electrodes and solid electrolytes has gradually become a reliable and promising power device due to its high energy density, environmental friendliness, and inexpensive cost. [ \n \n 23 \n , \n 24 \n \n ] We notice that some reported humidity bilayer actuators possess humidity‐sensitive layers with high specific surface area and excellent electrical conductivity, which are well suited as electrodes for zinc‐air batteries. [ \n \n 22 \n , \n 25 \n , \n 26 \n \n ] Also, in situ integration of zinc‐air batteries in actuators has not been reported, as far as we know. Therefore, a novel actuator with a self‐powered sensing function could potentially be created by integrating the zinc‐air battery in situ on the bilayer actuator. Herein, the free‐standing polypyrrole@graphene‐bacterial cellulose (PPy@G‐BC) films were prepared by using a low‐cost method combining vacuum filtration and in‐situ polymerization. Through extensive hydrogen bonding, the G‐BC films with a multilayer network structure were constructed by stacking hydrophilic BC nanofibers and graphene nanosheets with excellent photo‐thermal conversion capabilities. In addition, a large number of conductive polypyrrole nanoparticles were uniformly filled into these network pores of G‐BC film through in situ polymerization. The prepared PPy@G‐BC film was laminated with a biaxially oriented polypropylene (BOPP) film via acrylic ester to construct a light/humidity‐driven bilayer actuator. Hydrophobic BOPP film possesses a larger coefficient of thermal expansion (CTE) than the PPy@G‐BC film. [ \n \n 27 \n \n ] Thanks to the excellent photo‐thermal conversion and thermoelectric properties, the PPy@G‐BC layer spontaneously converts the temperature difference of the actuator into a thermoelectric signal through the Seebeck effect during the asymmetric thermal expansion with the BOPP layer. The thermoelectric signal generated by the PPy@G‐BC/BOPP actuator can realize self‐powered temperature sensing. With the assistance of a machine learning algorithm, a self‐powered bionic hand designed based on PPy@G‐BC/BOPP actuators can accurately differentiate the spontaneously generated thermoelectric signals to realize gesture recognition. Furthermore, the highly conductive PPy@G‐BC layer also provides a good air electrode platform for the in situ integration of flexible zinc‐air batteries. The PPy@G‐BC/BOPP actuator with an in‐situ integrated zinc‐air battery can reflect the environmental humidity and its water content change in real‐time. The current signal change of the PPy@G‐BC/BOPP actuator results from the humidity resistance effect of the PPy@G‐BC film. Notably, the self‐powered electrical signals of these two PPy@G‐BC/BOPP actuators are in good consistency with their deformation state and physical properties. Finally, the two aforementioned multi‐functional actuators were highly integrated into a single intelligent gripper device, realizing multi‐responsive actuation and self‐powered multi‐mode sensing function, as shown in Scheme \n 1 \n . The intelligent gripper can not only move objects steadily but also monitor the physical properties (temperature, humidity, and water content) and deformation state of the actuators in a complementary and synergistic way. The innovative design of integrating multi‐responsive actuation and multi‐mode self‐powered sensing in a single flexible electronic device is expected to assist intelligent soft robots in complex and changing environments to better perceive and perform their tasks. Scheme 1 Intelligent gripper based on the multi‐responsive PPy@G‐BC/BOPP actuators integrated with self‐powered and multi‐mode sensing functions. Background image is designed by Midjourney.",
"discussion": "2 Result and Discussion 2.1 Fabrication and Characterization of PPy@G‐BC Film Graphene is an ultrathin 2D layered material that is composed of carbon atoms in sp 2 hybridization (Figure S1 , Supporting Information). Thanks to its excellent properties in photo‐thermal conversion, thermoelectricity, and electricity, graphene has become one of the most promising materials in the fields of actuators, thermoelectric generators, and nanogenerators. [ \n \n 28 \n , \n 29 \n , \n 30 \n \n ] However, pure graphene film generally lacks the abilities of free‐standing and hydrophilicity, which makes it difficult to meet the requirements of multi‐functional and multi‐responsive actuators. [ \n \n 20 \n \n ] The preparation process of PPy@G‐BC film is schematically shown in Figure \n 1 a . As a natural biomass material, there are a large number of oxygen‐containing functional groups on the surface of the BC nanofibers that can generate strong hydrogen bonding with water molecules. [ \n \n 26 \n \n ] What's more, as shown in Figure S2 (Supporting Information), the tightly cross‐linking porous network structure also endows BC nanofibers with satisfactory strength and toughness. [ \n \n 31 \n \n ] Therefore, BC nanofibers with excellent hydrophilic and mechanical properties were selected as the substrate. Uniform G‐BC dispersion was obtained by dispersing BC nanofibers and graphene nanosheets in deionized water by magnetic stirring and ultrasonic dispersion methods. Then, assisted by the vacuum filtration method, graphene nanosheets and BC nanofibers were alternately stacked together through extensive hydrogen bonding interactions and van der Waals’ forces, thereby forming a free‐standing film. The prepared G‐BC films exhibited a gray color (Figure S3a , Supporting Information). The TEM images depicted in Figure S4 (Supporting Information) illustrate G‐BC films at various magnifications, revealing the tight combination of graphene nanosheets with BC nanofibers. Besides, from the SEM images of the surface and cross‐section of the G‐BC film shown in Figure S5a–d (Supporting Information), it can be seen that the BC nanofibers tightly wrapped the graphene nanosheets, constructing a stable multilayer network structure through surface interaction forces. Although the mechanical properties and hydrophilicity of G‐BC film gain improvement compared to pure graphene film, the insulating properties of BC nanofibers will inevitably deteriorate the electrical conductivity of graphene. To address this issue, an in‐situ polymerization method was employed to grow highly conductive PPy nanoparticles on the G‐BC film. As shown in Figure 1b,c , the Py monomers grow in situ into pearl‐like PPy nanoparticles under the effect of oxidizing agents, which then uniformly attach to the surfaces and gaps of the G‐BC porous network structure by π‐π stacking and hydrogen bonding interactions. Compared to the G‐BC film, the resulting PPy@G‐BC film exhibits a darker black color (Figure S3b , Supporting Information), indicating successful polymerization. Notably, the prepared PPy@G‐BC film shows excellent flexibility (Figure S6a , Supporting Information). More experimental details are described in the Experimental Section. Since the BOPP film possesses excellent mechanical properties, the PPy@G‐BC/BOPP actuator with a bilayer structure also exhibits exceptional flexibility (Figure S6b , Supporting Information). The SEM image of the cross‐section of the PPy@G‐BC/BOPP actuator is shown in Figure 1d . It can be seen that the BOPP film is bonded to the PPy@G‐BC film by the adhesion of acrylic ester. In order to investigate the bonding force between the bilayers of the PPy@G‐BC/BOPP actuator, a peeling test was performed, and the test results are shown in Figure S7 (Supporting Information). The peeling force between the BOPP layer and the PPy@G‐BC layer was measured to be up to 10.12 MPa, indicating that there is an excellent bonding force between these two layers. The pressure‐sensitive acrylic ester (≈8 µm in thickness) is able to bond the PPy@G‐BC layer to the BOPP layer by molecular forces under pressure, thus providing a strong interaction force. As a result, the heat converted by the PPy@G‐BC layer can be quickly transferred to the BOPP layer. The thickness of the PPy@G‐BC film is ≈26 µm, and the thickness of the PPy@G‐BC/BOPP actuator is ≈64 µm. Figure 1 Fabrication and characterization of the PPy@G‐BC film. a) Schematic diagram of the fabrication process of the PPy@G‐BC film. b) TEM image of the PPy@G‐BC film. c) SEM image of the surface of PPy@G‐BC film. d) Cross‐section SEM image of the PPy@G‐BC/BOPP actuator. e) The XRD spectra of BC film, G‐BC film, and PPy@G‐BC film. f) The FTIR spectra of BC film, G‐BC film, and PPy@G‐BC film. g) Mechanical properties of BC film, G‐BC film, and PPy@G‐BC film (sample size = 5). XRD was first used to analyze the crystal structures of BC film, G‐BC film, and PPy@G‐BC film, as shown in Figure 1e . The prepared BC film exhibits three typical diffraction peaks at 14.8°, 17.04°, and 22.9°, which correspond to the (1 1 ¯ 0), (110), and (200) cellulose I crystal planes, respectively. [ \n \n 32 \n \n ] The G‐BC film exhibits a strong diffraction peak at 26.4° and a weak diffraction peak at 54.5°, corresponding to the (002) and (004) crystal planes of graphene. [ \n \n 33 \n \n ] The above result demonstrates that graphene maintains a good lattice structure and lamellar spatial arrangement after the introduction of BC nanofibers. After the polymerization of PPy, no obvious new diffraction peaks were observed in the PPy@G‐BC film. However, the characteristic broad peak at 26.4° is significantly enhanced, which is due to the superposition of the diffraction peaks of the introduced PPy with graphene. [ \n \n 34 \n , \n 35 \n \n ] The FTIR spectra of BC film, G‐BC film, and PPy@G‐BC film are shown in Figure 1f . Three typical characteristic peaks at 3478, 2836, and 1641 cm −1 appeared in the BC film, which corresponded to the stretching vibration of hydrogen bonding induced by the O‐H group, the asymmetric stretching vibration of C‐H in the parabolic ring, and the O‐H bending vibration, respectively. [ \n \n 36 \n \n ] After the introduction of graphene nanosheets, a diffraction peak appears at 1673 cm −1 in the G‐BC film, corresponding to the C = C stretching vibration on the benzene‐like ring structure in graphene. [ \n \n 37 \n \n ] In addition, the diffraction peak corresponding to the O‐H stretching vibration in the G‐BC film is red‐shifted from 3478 cm −1 to a shorter wavenumber of 3451 cm −1 , compared to the BC film. This is because the strong hydrogen bonding interaction between graphene nanosheets and BC nanofibers leads to the movement of the stretching vibration bands of O‐H, which are susceptible to hydrogen bonding interaction. [ \n \n 38 \n , \n 39 \n \n ] Typical characteristic peaks of PPy can be observed at 3420 cm −1 (N‐H stretching vibration) and 1588 cm −1 (asymmetric C‐N ring‐stretching vibration) for the PPy@G‐BC film, which indicates that PPy nanoparticles successfully grew on the G‐BC film. [ \n \n 40 \n \n ] Then, the mechanical properties of BC film, G‐BC film, and PPy@G‐BC film were also characterized, as shown in Figure 1g and Figure S8 (Supporting Information). The graphene doping can lead to a Young's modulus of 4.96 GPa for the G‐BC film, which is 2.4 times higher than that of the pure BC film (2.05 GPa). The phenomenon may be attributed to the fact that graphene nanosheets can enhance the connectivity of the BC network structure and hinder the twisting and buckling of BC nanofibers. After polymerization of PPy nanoparticles, the mechanical properties of PPy@G‐BC film were also enhanced compared to G‐BC film, with its Young's modulus and tensile stress enhanced by ≈1.12 times (4.96–5.55 GPa) and 1.08 times (86.41–93.53 MPa), respectively. Compared to the similar flexible materials in the field of multifunctional actuators, the mechanical properties (especially Young's modulus and tensile stress) of the PPy@G‐BC film exhibit a high level, as shown in Table S1 (Supporting Information). This is because PPy nanoparticles can form tight adsorptions on the surface and in the gaps between BC nanofibers and graphene nanosheets through π–π stacking and hydrogen bonding interactions. [ \n \n 35 \n , \n 41 \n \n ] As a result, PPy nanoparticles act as mechanically enhanced fillers to strengthen the mutual attraction between the nanomaterials and fill the structural gaps as well, which endows the PPy@G‐BC film with a compact internal structure. Furthermore, to evaluate the mechanical stability of the PPy@G‐BC film, it was fixed on a displacement platform by BOPP film for 300 bending cycle tests (Figure S9 , Supporting Information). It is found that the internal resistance of the PPy@G‐BC film only increased from 30.29 to 31.83 Ω after 300 cycles without significant surface damage. The test result demonstrates the excellent overall mechanical stability of the PPy@G‐BC film. Benefiting from the above properties, PPy@G‐BC film is expected to serve as a multi‐functional layer for broad applications in the field of multi‐functional actuators. 2.2 Thermoelectric Properties of PPy@G‐BC Film To investigate the thermoelectric properties of PPy@G‐BC film, the heat platform and the cold platform were used as heat source and cold source, respectively ( Figure \n 2 a ). In this way, a spatial temperature difference can be created. It is worth noting that high‐temperature‐resistant polyimide tapes were used to secure the electrodes at both ends of the PPy@G‐BC film. It can prevent the bending deformation of PPy@G‐BC film due to high temperature from affecting the accuracy of the thermoelectric property test. The left (1 cm) and right (1 cm) ends of the PPy@G‐BC film, fixed on the heat and cold platforms, served as the hot and cold ends of the thermoelectric generator. Meanwhile, the middle part (1 cm) was suspended. Specific experimental details and test structures are given in the Experimental Section and Figure S10 (Supporting Information). When the left heat platform starts to heat up, the hot and cold ends of the PPy@G‐BC film with almost the same initial temperature start to generate a temperature difference Δ T . According to the Seebeck effect, the current carriers (electric holes) in the PPy@G‐BC film will move from the hot end to the cold end, thus generating an output voltage. The Seebeck coefficient is one of the most important indicators used to characterize the thermoelectric conversion efficiency of a material. As shown in Figure 2b , nine different temperature segments were set on the heating platform to create the temperature gradient, which is necessary for the Seebeck coefficient test. Each temperature segment lasted for ≈300 s. The curve of the open‐circuit voltage V \n oc of PPy@G‐BC film with temperature difference Δ T is shown in Figure 2c . Apparently, V \n oc follows the same trend as the temperature gradient, showing a good linear correlation. V \n oc can reach 2.417 mV at a Δ T of 60.3 K. The association between V \n oc and Δ T is illustrated in Figure 2d . According to the formula of the Seebeck coefficient: S = V \n oc /Δ T ( S is the Seebeck coefficient of the thermoelectric generator), the Seebeck coefficient of PPy@G‐BC is calculated as 39.9 µV K −1 . Under the same temperature gradient (upper panel of Figure 2c ), the measured short‐circuit current I \n sc (Figure S11 , Supporting Information) of the PPy@G‐BC film also exhibited good consistency with the trend of Δ T . The above results reveal the excellent thermoelectric properties of PPy@G‐BC film, indicating it as an ideal flexible thermoelectric material. Apart from high Seebeck coefficients, ideal thermoelectric materials also require excellent electrical properties, which can directly affect the thermoelectric performance. [ \n \n 42 \n \n ] Therefore, the electrical properties of the PPy@G‐BC film were characterized under different temperature differences, as shown in Figure 2e . The test results indicate that the PPy@G‐BC film possesses excellent electrical conductivity, which increases with the temperature gradient from 37.73 to 49.07 S cm −1 . In addition, the highest power factor of the PPy@G‐BC film can reach 7.88 µW m −1 K −2 , indicating its good energy utilization when used as a thermoelectric generator. Then, to investigate whether the in‐situ polymerization of PPy would affect the thermoelectric properties, the thermoelectric and electrical properties of G‐BC films were also tested under different temperature differences (Figures S12 and S13 , Supporting Information). The Seebeck coefficient, electrical conductivity, and power factor of the G‐BC film were tested to be 19.4 µV K −1 , 5.85 S cm −1 , and 0.23 µW m −1 K −2 , which were much lower than the corresponding properties of the PPy@G‐BC film. It is obvious that both the thermoelectric and electrical properties of PPy@G‐BC films are substantially improved after the introduction of highly conductive PPy nanoparticles. The enhancement of the thermoelectric properties of PPy@G‐BC film is mainly attributed to three factors. First, the introduction of highly conductive PPy nanoparticles can improve the electrical conductivity of the PPy@G‐BC film, thus accelerating the carrier migration rate. Second, PPy nanoparticles can effectively reduce the thermal conductivity of the PPy@G‐BC film, as shown in Figure S14a (Supporting Information). According to Note S3 (Supporting Information), the reduction of thermal conductivity can enhance the ZT value of the PPy@G‐BC film, which is a quality factor that measures the comprehensive thermoelectric properties of a material. Third, as a good thermoelectric material (≈14.5 µV K −1 ), the extensive interface combination of PPy nanoparticles significantly enhanced the Seebeck coefficient of the film from 19.4 µV K −1 to 39.9 µV K −1 . [ \n \n 43 \n \n ] Consequently, the ZT values of the PPy@G‐BC film obtained ≈60 times enhancement compared to the G‐BC film, indicating that PPy nanoparticles can substantially improve the comprehensive thermoelectric properties of the film (Figure S14b , Supporting Information). Furthermore, as can be seen from Table S2 (Supporting Information), the comprehensive thermoelectric properties of the PPy@G‐BC films are at an intermediate level among the same type of thermoelectric materials, which is sufficient for self‐powered thermoelectric sensing applications. Figure 2 TE property of PPy@G‐BC film. a) Schematic diagram of the TE test. b) Infrared thermal images of the PPy@G‐BC film under different temperature differences. c) Δ T and V \n oc of the PPy@G‐BC film heated by a hot plate. d) The V \n oc of the PPy@G‐BC film as a function of Δ T . e) Electrical and thermoelectric properties of the PPy@G‐BC film under different temperature differences. f) Schematic diagram of the PPy@G‐BC film used as a thermoelectrical touch sensor. g) The V \n oc of the PPy@G‐BC touch sensor under multiple touches. h) Schematic diagram of the “self‐powered piano” formed by multiple PPy@G‐BC films. Image is designed by Midjourney. i) The V \n oc of the “self‐powered piano”, and the inset shows the corresponding optical photo. Scale bar: 2 cm. j) The V \n oc and Δ T of the PPy@G‐BC film as a water temperature sensor, and the inset showing the corresponding optical photos. Scale bar: 2 cm. k) Infrared thermal images of the beaker integrated with the PPy@G‐BC water temperature sensor. To evaluate the self‐powered performance and temperature sensitivity of the PPy@G‐BC film in practical applications, the film was employed as a thermoelectric touch sensor to detect finger touch movements directly (Figure 2f ). As shown in Figure 2g , the PPy@G‐BC touch sensor can respond quickly to successive finger touch actions by generating thermoelectric voltage spontaneously. The result demonstrates the excellent thermoelectric properties and temperature sensitivity of the PPy@G‐BC touch sensor, which can sensitively detect temperature differences and quickly recover to the initial state. Next, an interesting “self‐powered piano” formed by connecting multiple PPy@G‐BC films in series was developed, as shown in Figure 2h . Due to the exceptional temperature sensitivity of each PPy@G‐BC film, the “self‐powered piano” can spontaneously generate stepped voltages in various touch modes. This allows for the simulation of musical scales, as shown in Figure 2i . Finally, the PPy@G‐BC film was attached directly to a beaker for utilization as a water temperature sensor. As can be seen in Figure 2j,k , after adding cold and hot water to the beaker successively, the PPy@G‐BC water temperature sensor can rapidly output the corresponding voltage signal with the temperature difference. Notably, such a water temperature sensing strategy can not only detect the time of adding cold or hot water but also reflect the stable water temperature. The result indicates that PPy@G‐BC film can be used as a practical water temperature sensor with high sensitivity. In addition, two tests were performed to evaluate the stability and durability of the PPy@G‐BC film for thermoelectric applications (Figure S15 , Supporting Information). As can be seen from the results, the PPy@G‐BC film not only exhibits good stability in the long‐term thermoelectric performance test of 3600 s, but also stably outputs thermoelectric signals in the repeated thermostatic touch test. Therefore, the PPy@G‐BC film was demonstrated to be a flexible thermoelectric material with excellent temperature sensitivity, stability, and durability in thermoelectric applications. 2.3 Light‐Driven PPy@G‐BC/BOPP Actuator Integrated with Self‐Powered Temperature Sensing Function In recent years, PPy and graphene have been widely applied to PTEGs and light‐driven actuators due to their excellent photo‐thermal conversion and thermoelectric properties. [ \n \n 44 \n , \n 45 \n , \n 46 \n , \n 47 \n \n ] To overcome the limitations of the photo‐thermal conversion properties of a single material, graphene and PPy were combined to achieve more efficient photo‐thermal conversion. As shown in Figure S16 (Supporting Information), the PTE performance of the PPy@G‐BC/BOPP actuator was tested using copper tape as a photomask. The actuator was secured by a BOPP film on a U‐shaped glass frame. Near‐infrared (NIR) light was used as a light source. Specific experimental details and test structures are shown in the Experimental Section and Figure S16 (Supporting Information). The PPy@G‐BC/BOPP actuator was irradiated under NIR light with different light power densities, and the corresponding infrared thermal images are shown in Figure S17a (Supporting Information). Based on the PTE effect, the PPy@G‐BC/BOPP actuator can convert the temperature difference into output voltage, as shown in Figure S17b (Supporting Information). With the increase in light power density, the maximum Δ T and maximum V \n oc show the same increasing trend. The calculated Seebeck coefficient of the PPy@G‐BC/BOPP actuator is 42.8 µV K −1 , which is similar to the result obtained with the hot platform (Figure S17c , Supporting Information). In addition, the PPy@G‐BC/BOPP actuator maintained stable photo‐thermal conversion capability and self‐powered output capability under 300 cycle tests, as shown in Figure S17d (Supporting Information). The above results indicate that the PPy@G‐BC/BOPP actuator possesses excellent PTE properties. It is known that flexible light‐driven bilayer actuators are novel devices that can remotely realize mechanical actuation based on the photo‐thermal effect and asymmetric thermal expansion. As a result, the PPy@G‐BC/BOPP actuator, with a bilayer structure and excellent PTE properties, is ideally suited for self‐powered sensing in the form of a light‐driven actuator. As shown in Figure \n 3 a , a test structure that can simultaneously measure the light‐driven and self‐powered sensing properties of the PPy@G‐BC/BOPP actuator was constructed. In order to evaluate the light‐driven properties, a portion (3 cm) of the PPy@G‐BC/BOPP actuator was fixed to a U‐shaped glass frame. While the remaining actuation portion (1.5 cm) was allowed to bend freely under NIR light irradiation. Furthermore, to test the self‐powered sensing properties, one copper electrode was adhered 0.5 cm away from the actuation portion, while the other copper electrode was adhered uppermost for collecting the output thermoelectric signal. Specific experimental details and test structure are given in the Experimental Section and Figure S19 (Supporting Information). The infrared thermal images of the PPy@G‐BC/BOPP actuator under different light power densities are shown in Figure 3b . From these images, it can be seen that the bending angle and the surface temperature of the actuation portion of the PPy@G‐BC/BOPP actuator gradually increase with the enhancement of the light power density. This substantial light‐driven bending is mainly attributed to a combination of two mechanisms. The first mechanism is the thermal‐dehydration effect. Benefiting from its excellent hydrophilicity, the PPy@G‐BC layer can store water molecules in the internal multilayer network structure. When the PPy@G‐BC layer effectively converts light energy into heat energy, the water molecules stored inside will evaporate due to the temperature increase. As a result, the PPy@G‐BC layer will shrink after the water molecules desorb. The second mechanism is the asymmetric thermal expansion effect. Since the PPy@G‐BC/BOPP actuator has a compact bilayer structure, the BOPP layer with a larger CTE will exhibit greater volume expansion than the PPy@G‐BC layer when heated. The volume change mismatch between these two layers will result in internal stresses inside the actuator. The CTE difference is mainly attributed to their different material compositions and structures. The stable multilayer network structure formed by hydrogen bonding and Π‐Π stacking restricts the thermal expansion movement inside the PPy@G‐BC layer upon heating, thus exhibiting a small CTE. In contrast, the polypropylene molecules inside the BOPP layer move more freely during thermal expansion due to the loose structure and weak interaction forces, resulting in a larger CTE. Consequently, a mismatch volume change is caused between the bilayer structure, leading to a substantial bending of the PPy@G‐BC/BOPP actuator toward the BOPP layer. Figure 3 Light‐driven actuation and self‐powered sensing performances of the PPy@G‐BC/BOPP actuator. a) Schematic diagram of the light‐driven actuation test. b) Infrared thermal images of the PPy@G‐BC/BOPP actuator under different light power densities. Scale bar: 1 cm. c) Δ T , V \n oc , and curvature of the PPy@G‐BC/BOPP actuator under different light power densities. d) V \n oc and curvature of the PPy@G‐BC/BOPP actuator as a function of Δ T when irradiated by NIR light. The curvatures were calculated according to Figure S18 and Note S1 (Supporting Information). e) Optical photos and corresponding infrared thermal images of the “Panda” robot from an indifferent state to a happy state. f) Optical photo and corresponding infrared thermal image of the “athlete” robot from a straight state to a push‐up state. In order to investigate the correlation between the light‐driven actuation and PTE properties of the PPy@G‐BC/BOPP actuator, Δ T , V \n oc , and curvature of the PPy@G‐BC/BOPP actuator were simultaneously recorded under different light power densities (Figure S20 , Supporting Information). As can be seen, the trends of Δ T , V \n oc , and curvature of the PPy@G‐BC/BOPP actuator under different light power intensities maintain good consistency. The maximum Δ T , maximum V \n oc, and maximum curvature of the PPy@G‐BC/BOPP actuator under different light power densities are exhibited in Figure 3c (data extracted from Figure S20 , Supporting Information). It can be seen that the maximum V \n oc and maximum curvature of the PPy@G‐BC/BOPP actuator increased synchronously with the increase in Δ T . At a light power density of 300 mW cm −2 , the PPy@G‐BC/BOPP actuator achieves a maximum curvature of 1.18 cm −1 . Additionally, the maximum V \n oc and maximum Δ T values are 1.127 mV and 27.4 K, respectively. The fitted Sebeck coefficient of the PPy@G‐BC/BOPP actuator is 40.9 µV K −1 , which is similar to the result of the previous thermoelectric and PTE properties tests (Figure 2d ). The above results indicate that the light‐driven deformation hardly interferes with the PTE effect. In other words, the actuation function and the self‐powered sensing function of the PPy@G‐BC/BOPP actuator can be simultaneously effective and unaffected by each other. In addition, a good linear relationship between V \n oc and curvature of the PPy@G‐BC/BOPP actuator (with a slope of 1.08 mV cm −1 ) is demonstrated in Figure S21 (Supporting Information). Based on the excellent linearity, real‐time knowledge of the temperature and deformation state of the PPy@G‐BC/BOPP actuator can be achieved by monitoring the self‐powered output voltage. Besides the thermoelectric output performance, the photo‐thermal deformation capability is also an important indicator for a light‐driven actuator. Therefore, two flexible photo‐thermal actuation applications have been developed using PPy@G‐BC/BOPP actuators. First, a “panda” robot was created using a PPy@G‐BC/BOPP actuator as a mouth, as shown in Figure 3e . Without light, the PPy@G‐BC/BOPP actuator remained flat, which caused the “Panda” robot to show an indifferent state. As soon as exposed to NIR light, the PPy@G‐BC/BOPP actuator bent drastically due to its excellent photo‐thermal conversion ability, making the “Panda” robot show a happy state. Further, an “athlete” robot was created with PPy@G‐BC/BOPP actuators, which acted as its arms and torso, respectively (Figure 3f ). In the absence of light, the “athlete” robot remained flat. Upon activation of the NIR light, the “athlete” robot swiftly executes a push‐up motion, thanks to the rapid photo‐thermal conversion capability of the PPy@G‐BC/BOPP actuators. More importantly, the deformation repeatability of the PPy@G‐BC/BOPP actuator affects its accuracy and reliability for self‐powered sensing applications. Hence, 300 cycles of NIR irradiation (light power density of 200 mW cm −2 ) were performed on the PPy@G‐BC/BOPP actuator. The cyclic test results are displayed in Figure S22 (Supporting Information). After 300 irradiations, there was no significant degradation in the actuation performance of the PPy@G‐BC/BOPP actuator. The maximum curvature at the 300th cycle only decreased by 4.24% compared to the first irradiation. The above results proved that the PPy@G‐BC/BOPP actuator has a long light‐driven service life and stable performance. In the future, PPy@G‐BC/BOPP actuators with sensitive and stable self‐powered sensing and remote actuation capabilities are expected to advance the development of next‐generation multi‐functional actuators. 2.4 Self‐Powered Bionic Hand for Intelligent Gesture Recognition Gesture is a commonly used communication method in daily life that can convey information and express emotions. Inspired by the human five‐fingered hand, a self‐powered bionic hand was designed based on the aforementioned excellent PTE properties of the PPy@G‐BC/BOPP actuator. Five PPy@G‐BC/BOPP actuators were used as bionic fingers by attaching to a palm‐shaped photomask, and the specific dimension of the bionic hand is shown in Figure S23 (Supporting Information). However, the thermoelectric voltage generated by these actuators under different forms of irradiation tends to be complex and variable, which undoubtedly reduces the accuracy of remote gesture recognition. To overcome this problem, we attempted to find a more accurate way to recognize the gestures of the bionic hand. Support vector machine (SVM) algorithm, a classical algorithm in machine learning, is known for its ability to achieve high recognition accuracy with a little amount of training samples. As a result, it is widely used in data analysis and pattern recognition. Therefore, we decided to combine the self‐powered bionic hand with this machine‐learning algorithm to construct an intelligent gesture recognition system. A schematic flow diagram of intelligent gesture recognition is shown in Figure \n 4 a , which incorporates thermoelectric signal collection, data processing, algorithm model training, and gesture recognition results. First, the bionic hand was configured to perform gestures indicating different meanings by using various forms of irradiation while collecting thermoelectric signals generated by these processes. As shown in Figure 4b , when the NIR light was not switched on, the five bionic fingers stayed flat because there was no temperature difference on the surfaces, while the output voltage signals also remained almost unchanged. At this point, the bionic hand exhibited a “Stop” gesture. When the irradiation of NIR light stimulated the bionic forefinger, middle finger, and ring finger, these three bionic fingers were bent at the same time, forming the “Awesome” gesture (Figure 4c ). Meanwhile, the self‐powered output voltage signals generated by these three bionic fingers based on the PTE effect followed the same trend as the gesture. In addition, to accomplish other complex gestures such as “Shooting” (left panel of Figure 4d ), NIR light was used to irradiate the bionic middle finger, ring finger, and little thumb. And the corresponding electrical signals are shown in the right panel of Figure 4d . Finally, two gestures were designed: the “three‐pointer” gesture with the ring finger and pinky bent (Figure 4e ), and the “resolution” gesture with a clenched fist (Figure 4f ). The bionic hand also spontaneously generates thermoelectric voltages corresponding to these gestures. Figure 4 The self‐powered bionic hand for intelligent gesture recognition. a) Schematic flow diagram of gesture recognition. Optical photos and output voltage of the bionic hand with different gestures: b) initial state, c) “Awesome” gesture, d) “Shooting” gesture, e) “Three‐pointer” gesture, f) “Resolution” gesture. Scale bar: 2 cm. g) The typical structure of the machine learning algorithm model: SVM. h) Variation chart of loss and accuracy with the training cycle. i) Confusion matrix of gesture recognition results. j) The accuracy rate of the algorithm model to recognize the gesture data set. For these five gestures, 50 cycles of thermoelectric signal data were collected, thus creating a database. Then, it was carefully ensured that each sample contained sufficient information. And the collected thermoelectric signals were used as sample features, including the maximum, variance, minimum, deviation, peak, and standard deviation of the total data. As some interference signals occur during the collection of thermoelectric signals, an SVM model was constructed to further process the data, as shown in Figure 4g . Briefly, the principle of this model is to normalize and classify the data, cross‐validate the sample data features after extracting them through the kernel function, and finally pass them to the output layer corresponding to different gestures. As illustrated in Figure 4h , this algorithmic model requires only 65 iterations to obtain a high classification accuracy, which proves its feasibility and accuracy. As can be seen from the confusion matrix for different gestures in Figure 4i , the SVM algorithm has a high recognition rate for the thermoelectric signals of the bionic hand. Specifically, the prediction accuracies of the test set data and the training set data can respectively reach 95.4% and 96.8% (Figure 4j ). To summarize, the combination of the self‐powered bionic hand based on PPy@G‐BC/BOPP actuators and the SVM algorithmic model can accurately achieve intelligent gesture recognition, which has a wide range of application prospects in the fields of human‐computer interaction and intelligent recognition. 2.5 Humidity‐Driven of PPy@G‐BC/BOPP Actuator As a hydrophilic biomass substrate, BC is highly sensitive to humidity, thanks to the numerous oxygen‐containing groups in its nanofiber network. Therefore, it is reasonable to assume that a PPy@G‐BC/BOPP actuator based on BC nanofibers can be an excellent humidity‐driven actuator. As shown in Figure \n 5 a , a U‐shaped PPy@G‐BC/BOPP actuator with a length of 5 cm was prepared. Specific experimental details and device dimensions are shown in the Experimental Section and Figure S24 (Supporting Information). To investigate the humidity‐driven capability of the PPy@G‐BC/BOPP actuator, it was placed in a humidity‐controlled chamber with a relative humidity (RH) of 25% and humidified. As can be seen in the optical photos of Figure 5b , the actuator exhibited substantial bending actuation during humidification, with RH ranging from 25% to 90%. When the RH was 90%, the maximum bending curvature of the PPy@G‐BC/BOPP actuator was up to 1.26 cm −1 , demonstrating its excellent humidity‐driven capability. The mechanism of humidity‐driven actuation of the PPy@G‐BC/BOPP actuator can be explained by the asymmetric humidity expansion effect between the PPy@G‐BC layer and the BOPP layer (Figure 5a ). First, as can be seen from the hydrophilicity test in Figure S25a–c (Supporting Information), the PPy@G‐BC film exhibits a small water contact angle (WCA) of ≈37.9°, which is attributed to the large number of hydrophilic functional groups (e.g., ‐OH and ‐COOH) inside. Not only that, the multilayer compact structure inside the PPy@G‐BC film also contributes to the absorption and storage of water molecules. Although the concave‐convex surface structure formed by the polymerization of PPy nanoparticles resulted in a small deterioration in the hydrophilicity of the PPy@G‐BC film compared to the G‐BC film (WCA of 33.7°), the proper hydrophilicity can prevent the actuator from excessive humidity damage. In contrast, the large WCA of 71.6° for BOPP film proves it to be a highly hydrophobic polymer material, which is inert to external humidity changes. Hence, there is a huge asymmetry in the water absorption capacity between the PPy@G‐BC layer and the BOPP layer. The above analysis can be proved by the results of the water absorption tests in Figure S25d (Supporting Information). The mass change of the PPy@G‐BC film is over 20% when the RH increases from 25% to 90%, while that of the BOPP film is only 4%. Therefore, when the environmental humidity increases, the physical morphology of the BOPP layer remains almost unchanged, while the PPy@G‐BC layer absorbs water and expands. As a result, the PPy@G‐BC layer and the BOPP layer exhibit asymmetric humidity volume expansion in a humid environment. The expansion difference generates internal stresses that cause the actuator to bend toward the BOPP layer with less volume change. Figure 5 Humidity‐driven actuation and humidity‐sensitive properties of PPy@G‐BC/BOPP actuator. a) Schematic diagram of the humidity actuation. b) Optical photos of the humidity‐driven actuation of the PPy@G‐BC/BOPP actuator. Scale bar: 1 cm. c) Relative resistance change rate, curvature, and mass change of the PPy@G‐BC/BOPP actuator as a function of RH. Further, the relative resistance change rate, curvature, and mass change of the PPy@G‐BC/BOPP actuator under different humidity conditions were comprehensively analyzed to investigate the humidity sensitivity of the PPy@G‐BC/BOPP actuator. The corresponding curves of relative resistance change rate, curvature, and mass change are recorded in Figure 5c . In an environment with an initial RH of 25%, the PPy@G‐BC/BOPP actuator stayed flat while its resistance remained almost constant. As the curvature of the PPy@G‐BC/BOPP actuator increased to 1.26 cm −1 , its maximum relative resistance change rate reached 18.7% at the RH of 90%. It is because when the PPy@G‐BC layer absorbs water and expands, the electron transfer efficiency inside will be hindered by the water molecules with higher resistivity, which in turn increases the overall resistance. The trends of curvature, relative resistance change rate, and mass change of the PPy@G‐BC/BOPP actuator exhibit good consistency with the change of RH. Therefore, the water content change of the PPy@G‐BC/BOPP actuator can be monitored by analyzing the relative resistance change rate. 2.6 Self‐Powered Humidity Sensing Function of PPy@G‐BC/BOPP Actuator Although the water content of the PPy@G‐BC/BOPP actuator can be monitored by analyzing the relative resistance change rate, the complex external power supply and connecting wires would inevitably complicate the sensing system. With the development of flexible electronic devices, flexible zinc‐air batteries have become a prospective renewable energy solution due to their high energy density, low cost, and environmental friendliness. [ \n \n 23 \n , \n 48 \n \n ] Inspired by this, we attempted to integrate a flexible zinc‐air battery in situ on the PPy@G‐BC layer as an energy module of the PPy@G‐BC/BOPP actuator, as shown in Figure \n 6 a . Thanks to the excellent conductivity of the PPy@G‐BC layer, the in‐situ integration process of the zinc‐air battery does not require any wires, showing a high degree of integration and flexibility. Specific experimental details and device structures are described in the Experimental Section and Figure S26 (Supporting Information). The flexible zinc‐air battery is a battery that generates electrical energy through redox reactions. It has a semi‐open structure composed primarily of a zinc anode, an air cathode (PPy@G‐BC layer), and an alkaline solid electrolyte (KOH/PVA gel electrolyte). The specific electrochemical reactions during the discharge process are detailed in Note S2 (Supporting Information). Next, a 6 M KOH/PVA gel was used as a solid‐state electrolyte to test the performance of PPy@G‐BC film in a solid‐state zinc‐air battery, as shown in Figure 6b–d . The results indicate that the solid‐state zinc‐air battery also possesses stable output performance, where the maximum open‐circuit voltage, maximum current density, and maximum power density can respectively reach 1.17 V, 2.79 mA cm −2 , and 1.47 mW cm −2 . Figure 6 Humidity actuation and self‐powered humidity sensing of PPy@G‐BC/BOPP actuator. a) Schematic diagram of the structure of the PPy@G‐BC/BOPP actuator in situ integrated with zinc‐air battery. b) The open‐circuit voltage generated by the zinc‐air battery based on PPy@G‐BC film. c) The current density of the zinc‐air battery based on PPy@G‐BC film. d) The power density of the zinc‐air battery based on PPy@G‐BC film. e) The relative current change rate, curvature, and mass change of the integrated PPy@G‐BC/BOPP actuator. f) Optical photos of the deformation of the PPy@G‐BC/BOPP actuator under light/humidity dual stimulation. Scale bar: 1 cm. g) Performance of PPy@G‐BC/BOPP actuator under light/humidity dual stimulation. As can be seen in Figure S27 (Supporting Information), the humidity‐driven PPy@G‐BC/BOPP actuator integrated with a zinc‐air battery can still perform substantial humidity‐driven deformation. The maximum relative current change rate and the maximum curvature of the PPy@G‐BC/BOPP actuator can reach 10.76% and 1.05 cm −1 , respectively (Figure 6e ). Notably, the water content remains well consistent with the curvature but has a symmetric relationship with the current rate of change. This phenomenon is because as the output voltage of the integrated zinc‐air battery remains constant, the overall resistance of the device increases due to the water absorption in the PPy@G‐BC/BOPP actuator. As a result, the current change rate increased nonlinearly. We further investigated the capability of PPy@G‐BC/BOPP actuators to detect external environmental changes under light/humidity dual stimulation. Optical photos of the experimental process are shown in Figure 6f . The PPy@G‐BC/BOPP actuator was humidified to different bending levels and then dried to a straight state through NIR light while collecting current signals during the process. It can be seen from Figure 6g that the relative current change rate of the PPy@G‐BC/BOPP actuator varies synchronously with the deformation generated by humidification and irradiation. In addition, the PPy@G‐BC/BOPP actuator was also tested with cyclic light/humidity dual stimulation, as shown in Figure S28 (Supporting Information). As can be seen, the PPy@G‐BC/BOPP actuator exhibited stable light/humidity‐driven deformation performance and self‐powered sensing performance in the cyclic test, which indicates its combination of flexibility and durability. The above results indicate that the PPy@G‐BC/BOPP actuator integrated with a zinc‐air battery has excellent self‐powered humidity sensing capabilities. The novel self‐powered device with a multi‐mode sensing function will provide new ideas for the development of the environmental monitoring field. 2.7 Intelligent Gripper Integrated with Multi‐Mode and Self‐Powered Sensing Function Under different environmental stimuli, the light/humidity‐driven actuators based on PPy@G‐BC film have the capability of spontaneously outputting electrical signals during the actuation deformation process. Based on these advantages, a PPy@G‐BC/BOPP actuator with a self‐powered temperature sensing function and a PPy@G‐BC/BOPP actuator with a self‐powered humidity sensing function were highly integrated into a single intelligent gripper ( Figure \n 7 a ). The structure of the intelligent gripper is schematically shown in Figure S29 (Supporting Information). The strip‐shaped PPy@G‐BC/BOPP actuator and the U‐shaped PPy@G‐BC/BOPP actuator were placed face‐to‐face. The specific preparation method of the intelligent gripper is detailed in the Experimental Section. The optical photos of the movement process of the intelligent gripper are shown in Figure 7b . As shown in Figure 7c , when there was no external stimulus, the intelligent gripper did not exhibit any deformation or electrical signal changes (Figure 7b(i) ). After applying the humidity stimulus, the actuators on both sides of the intelligent gripper started to bend toward the object (Figure 7b(ii) ) and grasped the object to perform lifting and lowering operations (Figure 7b(iii),(iv) ). Notably, the weight of the grasped object is approximately twice the weight of the actuation parts of these two actuators. During the process, the current generated by the PPy@G‐BC/BOPP actuator decreased rapidly (blue part of the upper panel in Figure 7c ). Notably, the voltage generated by the PPy@G‐BC/BOPP actuator also decreased slightly (blue region of the middle panel in Figure 7c ). It is because the PPy@G‐BC/BOPP actuator sensed a small temperature drop caused by the humidity stimulus. When the humidity stimulus stopped and the NIR light stimulus was simultaneously turned on, the actuation portions of the intelligent gripper gradually heated up by the irradiation. Finally, the intelligent gripper slowly reverted to its initial deformation state, and the object was released (Figure 7b(v) ). During the release process, the current generated by the PPy@G‐BC/BOPP actuator increased slowly due to the water loss of the actuator under irradiation (red region of the upper panel in Figure 7c ). The voltage generated by the PPy@G‐BC/BOPP actuator significantly increased due to the heating of the actuator (red region of the middle panel in Figure 7c ). From the above results, it can be found that there is a good correspondence between the working state of the intelligent gripper, the electrical signals, and the temperature difference change (the bottom panel in Figure 7c ). It indicates that the intelligent gripper can remotely realize self‐powered and multi‐mode (temperature/humidity) sensing while grasping the object. In summary, the ingenious strategy of integrating multi‐responsive actuation, self‐powered, and multi‐mode sensing allows the full utilization of multiple environmental stimuli as well as the complementary synergistic monitoring of multiple physical properties. Figure 7 The intelligent gripper with self‐powered multi‐mode sensing function. a) Schematic diagram of the intelligent gripper gripping an object while realizing self‐powered multi‐mode sensing. b) A series of optical photos of the intelligent gripper during the operating process. Scale bar: 1 cm. c) Relative current change rate, output voltage, and temperature difference of the PPy@G‐BC/BOPP actuators in the gripper during the operating process (corresponding to Figure 7(b) )."
} | 14,184 |
34440588 | PMC8398003 | pmc | 9,706 | {
"abstract": "The future of long-duration spaceflight missions will place our vehicles and crew outside of the comfort of low-Earth orbit. Luxuries of quick resupply and frequent crew changes will not be available. Future missions will have to be adapted to low resource environments and be suited to use resources at their destinations to complete the latter parts of the mission. This includes the production of food, oxygen, and return fuel for human flight. In this chapter, we performed a review of the current literature, and offer a vision for the implementation of cyanobacteria-based bio-regenerative life support systems and in situ resource utilization during long duration expeditions, using the Moon and Mars for examples. Much work has been done to understand the nutritional benefits of cyanobacteria and their ability to survive in extreme environments like what is expected on other celestial objects. Fuel production is still in its infancy, but cyanobacterial production of methane is a promising front. In this chapter, we put forth a vision of a three-stage reactor system for regolith processing, nutritional and atmospheric production, and biofuel production as well as diving into what that system will look like during flight and a discussion on containment considerations.",
"conclusion": "9. Conclusions BLSS/ISRU systems will be a key component of any long-duration space mission in the future, a continuation of the effort to make exploration spaceflight more feasible and affordable. Current physiochemical systems are effective in their implementation but do not provide the sustainable reusability that a long-duration mission would require. ISRU-based regolith processing with physiochemical systems would require either high temperatures associated untenable power draws or with highly caustic and non-replenishable reagents. A well-designed BLSS system for both ARV and ISRU can potentially allow the needed renewable sustainability without the major drawbacks of regolith processing. A cyanobacterial-driven photosynthetic reactor may be the optimal choice to provide this reusability. When properly scaled and grown in the right conditions, cyanobacterial bioreactors can produce enough resources to support a four-person crew. Additional bioreactor modules, with proper storage mechanisms and defined enhancement procedures, can allow production to be scaled up for an increased crew size. Though many bioreactor photosynthetic species have high concentrations of proteins, work has been done to show that nitrogen starvation yields higher concentrations of carbohydrates, thereby molding the organism’s composition to be closer to human dietary needs. Oxygen production is effective and can be scaled up to include production of excess liquid oxygen for use as the oxidizer for the fuel of the descent/ascent vehicle. When looking for an ISRU-based return fuel, good fortune favors the additional property of cyanobacteria as a potential methane producer, a rocket-fuel with notably high I sp . While production of biomethane still needs to be refined, vehicle engineering would also need to be modified based on ISRU capabilities, as there has not been widespread adoption of LOX/methane engines in the Lunar/Mars DRM to date. Finally, bringing another new life form, in addition to humans, to a planetary location where a primary mission of the humans is to search for evidence of extant or extinct life makes it imperative to protect that planet from forward contamination. Sterilization and containment procedures are key to preventing the microorganisms we bring from Earth from creating a new ecosystem on another planetary body. While challenges remain in the field of planetary exploration BLSS/ISRU, as proposed designs still need to be put through rigorous field testing, the future for this evolving technology looks as bright as a Martian sunrise.",
"introduction": "1. Introduction When planning long-distance spaceflight missions, it becomes critical to create ways to reduce IMLEO (initial mass in low earth orbit) while also ensuring that the systems that increase IMLEO are reliable and have enough redundancies to ensure the success of the mission. Multiple probes and rovers have already been sent to Mars, but future missions will add new complications in keeping humans alive and returning them home safely, requiring food, oxygen, carbon dioxide scrubbing, and propellants for the transit vehicle and DAV (Descent/Ascent Vehicle). The current systems utilized on board spacecraft are physical-chemical, relying on both renewable and nonrenewable resources that are limited and require occasional resupply [ 1 ]. The current systems will not support a long-duration mission outside of LEO (Low Earth Orbit), and resupply to Mars, for example, will be both long and expensive. BLSS/ISRU (Bioregenerative Life Support System/In-Situ Resource Utilization) attempts to tackle these problems. Lunar and Martian regolith contain many useful elements and compounds for survival, but at the moment, they do not exist in a form that satisfies our life support needs. The Martian atmosphere contains 95% carbon dioxide and is very thin and inhospitable for humans and broad forms of life. Cyanobacteria are ancient photosynthetic organisms on Earth that are believed to be responsible for terraforming and oxygenating the planet to support higher orders of life. They are very effective sources of oxygen production, produce many useful compounds, and are used throughout the world as a nutritional supplement for their high protein content and wide resumé of vitamins, minerals, and antioxidants. Such a versatile organism has the potential to revolutionize the way we operate life support systems in space. Inflight BLSS would need to provide oxygen and food for the crew for interplanetary travel, thereby reducing dependence on foodstuffs and oxygen stores taken from Earth. For Planetary ISRU, a three-stage system is being proposed. Stage 1 will be responsible for bioweathering regolith by siderophilic cyanobacteria to free up non-organic elements and create organic compounds for photobioreactor growth. Stage 2 will involve a photobioreactor with species of cyanobacteria that will be responsible for production of oxygen, fixation of carbon dioxide, and accumulation of biomass for use in human consumption and fuel for subsequent operations. Stage 3 will involve a third bioreactor responsible for the creation of biofuels (methane) for use in the DAV."
} | 1,615 |
24023835 | PMC3759446 | pmc | 9,707 | {
"abstract": "We present a dynamical systems analysis of a decision-making mechanism inspired by collective choice in house-hunting honeybee swarms, revealing the crucial role of cross-inhibitory ‘stop-signalling’ in improving the decision-making capabilities. We show that strength of cross-inhibition is a decision-parameter influencing how decisions depend both on the difference in value and on the mean value of the alternatives; this is in contrast to many previous mechanistic models of decision-making, which are typically sensitive to decision accuracy rather than the value of the option chosen. The strength of cross-inhibition determines when deadlock over similarly valued alternatives is maintained or broken, as a function of the mean value; thus, changes in cross-inhibition strength allow adaptive time-dependent decision-making strategies. Cross-inhibition also tunes the minimum difference between alternatives required for reliable discrimination, in a manner similar to Weber's law of just-noticeable difference. Finally, cross-inhibition tunes the speed-accuracy trade-off realised when differences in the values of the alternatives are sufficiently large to matter. We propose that the model, and the significant role of the values of the alternatives, may describe other decision-making systems, including intracellular regulatory circuits, and simple neural circuits, and may provide guidance in the design of decision-making algorithms for artificial systems, particularly those functioning without centralised control.",
"introduction": "Introduction Animals constantly make decisions, yet decision-making mechanisms and their evolution are still poorly understood in many cases. Recent years have seen a convergence of several research fields aiming to improve our understanding of general decision-making principles. Behavioural ecologists have argued for the need to combine the traditional study of animal behaviour through the lens of optimality arguments [1] , with an increased understanding of the mechanisms underlying behaviour and their evolution [2] . At the same time psychologists and neuroscientists, who focus on understanding the mechanistic bases of behaviour, are increasingly focussing attention on how these mechanisms can implement optimal behaviour ( e.g. \n [3] - [5] ). Behavioural ecologists in the burgeoning subfield of collective animal behaviour are also interested in mechanisms, in terms of interaction rules and patterns, that generate sophisticated group decisions [6] . Some researchers have noted the parallels between these apparently disparate fields, by observing that the interaction patterns of neurons in brain circuits and animals in groups appear to be very similar [7] - [10] , and also that tools and concepts from psychology and neuroscience may usefully be imported into the study of collective animal behaviour [11] , [12] . These ideas have been made concrete in modelling studies where, for example, optimality analyses from neuroscience [9] or decision-making tests from psychology [8] have been applied to models of collective decision-making by social insect colonies of ants and honeybees, and in experimental studies where the parallels have successfully guided the search for decision-making mechanisms in honeybees [13] , [14] . In this paper we present a comprehensive analysis of our previous empirically-motivated model of decision-making by house-hunting honeybees swarms [13] , and argue that its decision-making properties may in turn guide the study of decision-making systems at other levels of biological complexity, up to individual brains, and down to intracellular decision-making circuits, as well as inform the design of artificial, decentralized decision-making systems. Our previous analysis showed that the particular pattern of `stop-signalling' observed in swarms allows them to adaptively avoid deadlock by choosing randomly when presented with two potential nest sites of equal quality, and to converge on choosing the best of two potential nest sites when there is a sufficiently large difference in their quality [13] , [14] . Here, we show further aspects of value-sensitive decision-making that arise from cross-inhibitory stop-signalling. We analyse a model whose decision-dynamics are characterised by fast attraction to a one-dimensional decision manifold, followed by slower time-evolution along this manifold. We leverage a time-scale separation to reveal how the strength of cross-inhibition critically determines the decision-system response to both the difference in value and the mean value of the two alternatives. These analytic results considerably extend our previous initial analysis of this model's decision dynamics [13] . We show that stronger cross-inhibition yields a greater minimum difference in value required for discrimination between the alternatives. When the difference in value is below this minimum, the alternatives are treated as equal or nearly equal, and the cross-inhibition determines whether or not the alternatives are of sufficiently high value to warrant breaking decision deadlock. A stronger cross-inhibition increases the minimum mean value of the alternatives above which a decision deadlock is broken and the system randomly chooses one of the alternatives. When the (nearly) equal alternatives have mean value below the minimum mean value threshold, deadlock is maintained, allowing for the arrival of information on other, possibly more valuable, alternatives. We show that cross-inhibition strength determines the minimum detectable difference in the value of alternatives, as a function of their mean value, in a manner similar to Weber's law as arising from psychological studies. We further show that for decisions over alternatives that do differ sufficiently in quality, that the stochastic decision dynamics exhibit a speed-accuracy trade-off in decision-making that depends critically on the difference in value and mean value of the alternatives, with dependence controlled by the strength of the cross-inhibition. The speed-accuracy trade-off is qualitatively similar to the statistically-optimal trade-off of the drift-diffusion model of decision-making, although we present evidence that decision-making does not achieve optimality under the parameterisations we consider here.",
"discussion": "Discussion Although motivated by and presented in terms of decision-making by house-hunting honeybee swarms, our model exhibits a number of beneficial decision-making qualities that we might expect other organisms to exhibit. At the heart of our analysis is the observation that, in a choice, an animal is typically rewarded by the value of the chosen alternative, rather than whether or not it chose the best. In particular the model decision-maker displays a sensitivity to the absolute as well as the relative value of the alternatives under consideration; this enables it to wait for information on better alternatives to arise when considering equally poor alternatives, but to spontaneously choose one equal alternative at random when both are good enough relative to a crucial decision-making parameter, the rate of stop-signalling, or cross-inhibition, . The decision-maker exhibits other properties observed in psychological studies, such as speed-accuracy trade-offs, and Weber's law of just-noticeable difference. The increasing rate of cross-inhibition may also improve the energetic costs of decision-making, although possibly at the expense of decision accuracy (as discussed in Text S1 and Figure S6 ). Our investigation has focussed on analytic treatment of the noise-free equations and stochastic simulations of speed-accuracy trade-offs and decision dynamics for binary decision-problems. Much work remains to be done in extending these analyses, for example to decisions over more than two alternatives. Having suggested that our model might describe adaptive decision-making in general, what are the prospects for finding similar decision-making networks in other species? The form of the model equations is that of chemical reaction kinetics, in which interactions between chemical species are described by `mass action' terms. Therefore, there is the potential for intra-cellular regulatory networks to implement these decision-dynamics quite easily, for example in deciding for which of a number of available substances to activate the associated metabolic pathway. Evidence that single-cells can, for example, implement Bayesian-estimation through intra-cellular signalling [31] , or exhibit Weber's law in gene regulatory pathways [32] , [33] indicates that such decision-making at the cell level is entirely plausible. Mutual inhibition also features in models of transcription in cell-fate decisions [24] . Another obvious class of decision models that invite comparison are those developed to describe neural networks for decision-making in simple perceptual decision tasks, such as those that take place in the primate visual cortex. A variety of accumulator models have been studied for their ability to fit experimental data, as well as implement optimal decision strategies ( e.g. \n [4] ). Optimal parameterisation of many such models requires evidence-accumulating pathways to interact [4] , which the cross-inhibition mechanism in our model also implements. While optimality analyses in these models do take account of variable rewards for correct choices ( e.g. \n [34] , [35] ), they do not typically account for the fact that in real animals incorrect choices over, for example, food items still result in a reward, albeit one that is not the best available. Recently however, there is increasing interest in combining ideas from psychophysics, such as the Drift-Diffusion Model (DDM) [19] , with the study of value integration processes ( e.g. \n [36] , [37] ). Many accumulator models, like the classic DDM, also struggle with the correct choice when presented with zero net evidence, such as when choosing between two stimuli of equal average magnitude, and thus cross decision thresholds only through random drift. When choice of either alternative would result in an equal reward, such behaviour is clearly sub-optimal. Proposals to deal with this include implementing `urgency signals' or collapsing decision thresholds over time [38] , [39] , and the use of time-dependent sensory gain, and asymmetric inhibition between evidence pathways [38] . Our model differs from these proposed mechanisms, in that it spontaneously exhibits behaviour like that of an unstable Ornstein-Uhlenbeck process in order to break deadlock, according to the value of the alternatives under consideration and the strength of cross-inhibition. Our non-linear model differs from the linear formulation of accumulator models underlying many analyses ( e.g. \n [4] ). The non-linear interaction terms of our model can, however capture neural activation dynamics; the logistic activation curve for neural populations in an accumulator model, used in [23] , are qualitatively similar to `activation patterns' in the stop-signalling model, and [23] derives behaviour qualitatively similar, although not identical, to the stable-deadlock and deadlock-breaking results presented above. It is not unreasonable to expect convergent evolution to arrive at the same simple solution to the problem of value-dependent decision-making, in systems as diverse as single cells, honeybee swarms, and vertebrate nervous systems."
} | 2,866 |
35517590 | PMC9059506 | pmc | 9,708 | {
"abstract": "We developed a method to fabricate a superomniphobic gold electrode by synthesizing hierarchical gold clusters on a gold substrate and treating the surface with low surface energy materials. The reduction of gold ions was repeated several times, causing the gold microparticles to grow in random directions and form hierarchical gold clusters. Treatment of the gold structures with perfluorothiol resulted in a superhydrophobic surface that also exhibited superoleophobicity for oils and liquids with surface tensions as low as 25.6 mN. The resulting electrode was not contaminated by hydrophilic and hydrophobic liquids, and by analyzing the current–voltage characteristics of the electrode with a PEDOT:PSS solution droplet, the electrode was found to be waterproof.",
"conclusion": "Conclusions We fabricated a superomniphobic gold electrode by creating hierarchical gold clusters and nanoflakes on a gold electrode via a simple wet-chemical reduction reaction. Hierarchical gold clusters and gold nanoflakes were produced when the reduction reaction was repeated with a high hydroxylamine concentration whereas only gold nanoflakes were produced with a low hydroxylamine concentration. After the formation of a self-assembled monolayer of PFT on the hierarchical gold cluster-grown surface, the contact angles for both water and hexadecane were more than 150°. In addition, it was demonstrated that the hierarchical gold cluster structure could be patterned on a flexible substrate and act as liquid-repellent electrodes; the negative effect of water on the performance of the device was drastically reduced. The non-wetting and non-fouling properties of the omniphobic metal electrodes may be useful making wearable electronic sensing devices less vulnerable to sweat contamination.",
"introduction": "Introduction Techniques that can control and manipulate the wettability of liquids of varying surface tension have attracted much attention because of their wide range of potential applications in chemical shielding, 1 non-wetting transparent films, 2,3 oil transportation, 4,5 and membrane technologies. 6,7 It is straightforward to fabricate conventional superhydrophobic surfaces that repel high surface tension liquids like water; a low energy material coating on a surface with appropriate roughness is usually sufficient. However, more sophisticated surface designs must be fabricated to obtain a superoleophobic surface that repels low surface tension liquids like organic solvents. Although there are numerous examples of natural superhydrophobic surfaces including lotus leaves 8,9 and water strider legs, 10 there are limited examples of natural superoleophobic surfaces; one notable example is the skin of springtails. 11 Because a surface that repels lower surface tension liquids generally repels higher surface tension liquids as well, superoleophobic surfaces are also superhydrophobic with a few exceptions. 12–14 A surface exhibiting both superhydrophobic and superoleophobic characteristics is referred to as superomniphobic. To achieve superomniphobic characteristics, a surface should possess reentrant or hierarchical textures. Various shapes of reentrant structures mimicking springtail's skins were fabricated by advanced lithography. 15–17 In the design of reentrant structures, the local texture angle should be smaller than the equilibrium contact angle. Krupenkin et al. fabricated reentrant silicon nanonails and honeycombs with overhang, 18 and Kim et al. fabricated doubly reentrant texture on silicon wafers. 19 Their work demonstrated that even without a low energy material coating, the hydrophilic silicon surface repelled almost all liquids. Although the reentrant and doubly reentrant textures are very successful in the preparation of omniphobic surfaces, their applications are limited because their fabrication requires the use of bulky and expensive clean-room-based lithography systems. Hierarchical textures can be an efficient alternative to reentrant textures. A hierarchically structured surface reduces the solid–liquid contact area by trapping air at multiple scales; it enhances the stability of the Cassie–Baxter state which yields higher contact angles as compared to a single-scale structured surface. 20–22 Various methods including spray coating, 23,24 multiple spin coating, 25,26 and chemical etching 27,28 have been devised to fabricate hierarchical textures on different substrates like polymers, 29 glasses, 30 and metal wires. 31 However, until now, there have been very few studies on the creation of micro/nano hierarchical structures using wet-chemical methods to obtain metallic omniphobic surfaces. Also, considering that electrode contamination results in electronic device malfunction, especially in wearable sensing devices, it is important to develop a fabrication technique that creates metal electrodes with protective and omniphobic properties. In this study, we fabricated a superomniphobic surface with hierarchical structures by using a gold reduction reaction. Using a high concentration of a reducing agent, gold ions were reduced to microparticle clusters with nanoroughness that stacked onto each other, forming hierarchical gold clusters. After perfluorination using thiolated molecules, the surface with the hierarchical gold cluster structure showed apparent contact angles higher than 150° for both water and hexadecane. Specifically, the fabricated hierarchical gold cluster surface retained a non-wetting Cassie state for liquids having a surface tension as low as 25.6 mN m −1 . Finally, we patterned a desired shape of the hierarchical gold cluster structure on polydimethylsiloxane (PDMS) and applied it to the fabrication of a flexible and multifunctional electrode exhibiting ultra-water repellency.",
"discussion": "Results and discussion The gold film deposited on the silicon wafer displayed a water contact angle of 117° after PFT treatment ( Fig. 1a ), indicating that it was not superhydrophobic but rather hydrophobic; since the surface did not have enough roughness to support an air interlayer between the water droplet and the surface. However, when the gold ion reduction was carried out, gold structures created enough roughness to produce superhydrophobic properties. A series of experiments were performed to investigate the influence of the NH 2 OH/HAuCl 4 concentration ratio on the gold structures. As described in the experimental section, the concentration of HAuCl 4 was fixed at 18.3 mM in this study. Fig. 1b–d shows SEM images of gold nanostructures created by a reduction reaction using 15.7 mM NH 2 OH; the reduction was carried out 1 ( Fig. 1b ), 2 ( Fig. 1c ), and 3 ( Fig. 1d ) times, and each reduction cycle was conducted for 10 min. Since the reduction of gold ions by hydroxylamine was dramatically catalyzed on the gold surface, 32,33 the reduction reaction with a low concentration of the reducing agent induced growth on the pre-deposited gold substrate rather than nucleation of new particles in the bulk solution; therefore, a nanosized flake structure was formed selectively on the pre-deposited gold surface. When the number of reduction cycles was increased from one cycle to three cycles, the size of the nanoflakes increased and the water contact angle after PFT treatment increased to 175° as shown in the bottom panel of Fig. 1d . However, the nanoflake-grown substrates were readily wetted by liquids with low surface tension. Fig. 1 (a) SEM image of gold film (upper panel) and optical microscopy image of a water droplet on the gold film after PFT treatment (lower panel). SEM images of gold nanoflake structures fabricated with (b) 1 cycle, (c) 2 cycles, and (d) 3 cycles of the reduction process (upper panel). The water contact angles on the gold nanoflake-grown surfaces after PFT treatment were 117°, 141°, 166°, and 175°, respectively (lower panel). When the concentration of the reducing agent was high (192 mM), the reduction rate of Au 3+ ions to the gold film increased rapidly, and hierarchical gold clusters formed simultaneously on the nanoflake surface (inset of Fig. 2a ). Fig. 2 shows the morphology of the fabricated hierarchical gold clusters after repeating the reduction process several times. As the number of reduction cycles increased, the number of branches per gold cluster increased, and the hierarchical structures reached a height of 14.6 ± 4.4 μm for the 8 cycle sample. Each reduction cycle was conducted for 10 min. The gold microparticle clusters grown via this bottom-up synthetic process were not broken by simple washing, N 2 gas flow, and sonication. Note that the hierarchical gold clusters exhibited a highly crystalline structure with a (111) preferential orientation (see Fig. S1 in ESI † ). After the resulting gold surface was chemically modified with PFT, it exhibited superomniphobic characteristics. In contrast, the surface prepared by the single reduction reaction for 60 min exhibited only superhydrophobicity (see Fig. S2 in ESI † ). Fig. 2 Top (upper panel) and cross-sectional (lower panel) SEM images of hierarchical gold clusters fabricated by (a) 2 cycles, (b) 4 cycles, (c) 6 cycles, and (d) 8 cycles reduction cycles on the nanoflake substrate. The inset of Fig. 2a shows the magnified SEM image of the boxed region. \n Fig. 3a shows the trend in water contact angle and contact angle hysteresis as a function of the number of reduction reactions. Even when the reduction process was performed only once, the water contact angle was greater than 170°. Since the water contact angle was already sufficiently high, it did not increase significantly even when several successive reduction steps were performed. However, the contact angle hysteresis, which corresponds to the degree of adhesion between the surface and the water droplet, became smaller and was less than 1° for the 6 and 8 cycle samples. The water repellency of the hierarchical gold cluster was also verified by observing a single droplet bounce on the gold surface ( Fig. 3b ). When the water droplet was dropped from a height of 2 cm, it completely left the surface without wetting. Fig. 3 Variations in (a) water contact angle (black square) and contact angle hysteresis (red circle), and (c) hexadecane contact angle as a function of the number of reduction cycles. Time-lapsed photographs of (b) water and (d) hexadecane droplets dropped from a height of 2 cm. To demonstrate the omniphobicity of the hierarchical gold cluster, the change in the contact angle for hexadecane was measured for different numbers of reduction cycles. Fig. 3c shows that the smooth fluorinated gold surface was wet by hexadecane with a surface tension of 27.47 mN m −1 and exhibited a contact angle of 73.2°. However, after two cycle of reduction, the contact angle of hexadecane increased sharply to about 126.3° and then increased gradually with the number of reduction cycles, reaching 158.6° for the sample with 8 cycles. Fig. 3d shows that the hexadecane droplet released from a height of 2 cm was not fully separated from the surface after collision, because the hexadecane droplet with low surface tension wets the hierarchical gold cluster structure more than the water droplet with high surface tension (see the maximum spread diameters of a water droplet at 5.6 ms and a hexadecane droplet at 4.2 ms). Therefore, less energy was available to transform into the kinetic energy required for the droplet to bounce back completely. A photograph of various liquid droplets, each with a different surface tension, on the hierarchical gold cluster is shown in Fig. 4a . The omniphobicity was verified based on how each liquid droplet beaded up. Fig. 4b–g shows the ability of the hierarchical gold cluster structure to maintain a range of low surface tension liquids in a non-wetting Cassie state. Starting with 1,3-propanediol that has a surface tension of 46.2 mN m −1 , contact angle measurements were conducted on several liquids with smaller surface tensions. Superomniphobicity was found for six low surface tension liquids, 1,3-propanediol, nitromethane, xylene, toluene, tetradecane, and dodecene; the measured contact angles (sliding angles) were 163.9 ± 1.8° (3°), 163.0 ± 1.9° (6°), 156.3 ± 0.76° (22°), 160.3 ± 1.9° (40°), 154.4 ± 3.3° (33°), and 152.4 ± 2.1° (90°), respectively. The superomniphobicity of the hierarchical gold cluster surface can be explained by the surface's unique morphology which possesses multiple scale roughness that drastically reduces the contact area between the solid and the liquid. Fig. 4 (a) Photograph of the hierarchical gold cluster surface with several types of liquids. Optical microscopy images of (b) 1,3-propanediol (46.2 mN m −1 ), (c) nitromethane (36.8 mN m −1 ), (c) xylene (30.1 mN m −1 ), (d) toluene (28.4 mN m −1 ), (e) hexadecane (27.47 mN m −1 ), (f) tetradecane (26.56 mN m −1 ), and (g) dodecene (25.6 mN m −1 ). (h) Schematic and SEM image of hierarchical gold clusters. The inset shows a magnified image of a single microparticle. Experiments were conducted to determine whether the hierarchical gold cluster surface could be used as a waterproof electrode. Fig. 5a shows the image of the hierarchical gold cluster electrode fabricated on a flexible PDMS substrate. Parallel electrodes with 3 mm separation were deposited and electrical contact was made via the electrode pads on both ends. Negligible change in the water contact angle was observed after the substrate recovered from bending, indicating that the hierarchical gold cluster structure did not fracture after deformation ( Fig. 5b ). The applicability of the omniphobic gold electrodes was investigated by measuring the current–voltage ( I – V ) characteristics of the electrodes when a droplet of 0.12% PEDOT:PSS solution was placed on and between the electrodes ( Fig. 5c ). Fig. 5 (a) Photograph of hierarchical gold cluster-grown electrodes fabricated on a flexible PDMS substrate. (b) Change in the water contact angle of the omniphobic superhydrophobic electrode before and after bending. (c) Schematic of the experiment to evaluate the effect of a droplet of PEDOT:PSS solution on the conductivity. (d) The corresponding I – V characteristic curves of the electrodes with (red circles) and without (black squares) PFT treatment. \n Fig. 5d shows that the current was about 0.74 ± 0.11 μA at the applied voltage of 0.5 V for a PEDOT:PSS solution droplet deposited on the electrode without PFT treatment. However, once the electrode was subjected to hydrophobic treatment, the current was reduced by nearly five orders of magnitude (8.74 ± 3.21 pA) because of the very small contact area between the electrode and solution. Note that the current is affected not only by the resistance of the electrolyte but also by the actual contact area between the electrode and the electrolyte solution when an external voltage is applied. In addition, the water-sliding angle of the fabricated hierarchical gold cluster structure was very small (<5°), thus making it easy to remove water droplets from the electrode."
} | 3,770 |
27877794 | PMC5099831 | pmc | 9,709 | {
"abstract": "Further to prior development in enhancing structural health using smart materials, an innovative class of materials characterized by the ability to feel senses like humans, i.e. ‘nervous materials’, is discussed. Designed at all scales, these materials will enhance personnel and public safety, and secure greater reliability of products. Materials may fail suddenly, but any system wishes that failure is known in good time and delayed until safe conditions are reached. Nervous materials are expected to be the solution to this statement. This new class of materials is based on the novel concept of materials capable of feeling multiple structural and external stimuli, e.g. stress, force, pressure and temperature, while feeding information back to a controller for appropriate real-time action. The strain–stress state is developed in real time with the identified and characterized source of stimulus, with optimized time response to retrieve initial specified conditions, e.g. shape and strength. Sensors are volumetrically embedded and distributed, emulating the human nervous system. Immediate applications are in aircraft, cars, nuclear energy and robotics. Such materials will reduce maintenance costs, detect initial failures and delay them with self-healing. This article reviews the common aspects and challenges surrounding this new class of materials with types of sensors to be embedded seamlessly or inherently, including appropriate embedding manufacturing techniques with modeling and simulation methods.",
"conclusion": "5. Concluding remarks and future directions This review investigates a new class of materials with a key property of being nervous, i.e. emulating human feeling of surrounding environmental parameters. With multi-sensing and delaying failure capabilities, this innovative material will see tremendous applications in small- and large-scale designs. The review has discussed several challenges encountered during this investigation. This includes suitable sensors, sensor embedding techniques, maintaining sensor integrity during embedding, the volumetric distribution of sensors to secure optimal coverage within the system under monitoring without compromising the material integrity, collection of data, and, power supply routes. Mechanisms of local self-healing within the material depending on the mode of failures, or inherent actuation techniques to at least delay any failure are of great importance. It goes without saying that cost will initially be a major factor. Although most of the issues associated with FBG sensors in composite materials were solved, there are remaining issues still to be addressed, such as the connection aspect and curing cycle monitoring. The connection aspect is vital for the industrialization of the process; and this can be achieved through either free-space connection or designing a dedicated connector. Full control and monitoring of the curing cycle may bring about more benefits [ 19 ]. Promising results were reported on the possibility of embedding polymer optical fiber gratings [ 22 ] in composite materials at temperatures less than 70 °C [ 20 , 21 ]. Research in this particular area should be focused on extending the use of polymer FBGs in high temperature environments. The embedding process of FBG sensors within metallic structures for structural health monitoring is in its infancy; and it is believed to be a cutting-edge research topic. FBG sensor-embedded materials are critical in many applications including the machining tools, aerospace and automotive industries [ 46 ]. Maintaining the integrity of the optical sensors by protecting them from high temperature during the embedding process and achieving higher sensitivity of the optical sensor are the critical issues to be addressed [ 44 ]. The multi-physics nature of the ultrasonic embedding makes it highly non-linear, requiring the coupling of several models together to successfully simulate the phenomenon of embedding fibers using an ultrasonic technique. This process requires heavily computational models and therefore techniques must be used to reduce the computational effort without compromising the accuracy of the solution to study the stresses, strains and integrity of the nervous material. Furthermore, in order to find the location of the external stimuli on a nervous material, algorithms need to be developed for various geometries that can accurately deduce the location based on the readings from embedded sensors. In summary, future research efforts should be centered on two important issues, i.e. embedding processes and protection of fibers while maintaining their sensing capability. In addition to exploring the possibility of introducing innovative processing techniques such as spark plasma sintering, there is a strong need to optimize the processing parameters of existing methods such as ultrasonic consolidation and laser-based layered manufacturing. The possibility of extending the use of polymer optical fiber gratings in harsh environments should be explored. Self-healing mechanisms could be embedded together in proximity to the sensors to correct for any possible failures, and would delay failures for a limited period, to avoid an unexpected instant accident. This will have a huge impact on equipment and public safety.",
"introduction": "1. Introduction The introduction to nervous materials requires an understanding of the complexities of the human sensory and nervous system. The latter is a living organism that senses the environment and responds appropriately. This system is exceedingly complex, with a large number of components and an extremely large number of interactions between these components. The system is inherently seamless. The human body contains an incredible sensory system, information transfer mechanism, and sensory information processing unit (brain), shown in figure 1 . The sensations in a human body occur when external stimuli interact with sensory receptors. Most of the sensory receptors are distributed in the skin and tissues of the body. Hence, the skin can feel pain, temperature, pressure, and other stimuli via these nerve cells in tissues. The sensory information is transferred by means of electrical signals because the neuron is sufficiently composed of multiple analogue inputs (called dendrites ) and a digital output (called an axon ) like an electrical device (figure 1 ). The sensory information converges on spinal nerves then enters the brain as trains of action potentials traveling along individual sensory neurons to evaluate the stimuli. Sundaresan and Shultz [ 1 ] commented on the purpose of structural health monitoring: ‘for the purpose of structural health monitoring, we can consider the following properties of a neural system:\n (1) information proceeds through the system through different levels; (2) information is processed at each level; the level of processing is higher the farther away from the input; (3) information is synthesized at each level, and a decision is made as to what is sent forward to the next level; (4) the system has mechanisms for noise suppression. There is also interconnection between sensory cortices’. Figure 1. The biological neural system: (a) the nervous system of a human, (b) the visual cortex, (c) a simplified illustration of the biological nervous system. Hence, nervous materials will have a surface embedded network of sensors that will not affect the integrity of the material but monitor its behavior and possibly react to correct or delay any possible failure. This will be classified as a new class of materials with various possible names, e.g. nervous or sensorial materials. In essence, such a material can ‘feel’ its environment and react to correct or delay failure using embedded actuators. With an ever-changing environment, the world is becoming complex, requiring more stringent sets of smart design to ensure personnel and public safety through products at all scales, e.g. the high integrity of aircraft, civil infrastructures, nuclear energy and high-tech aerospace/aircraft structures are just a few examples. Two major sets of design requirements are targeted: damage tolerance to preclude structural safety problems and durability to preclude expensive maintenance and repairs. This also means a need for a reactive system to any environmental change, accidental effect or expected failure. Early detection and correction of the structural damage or deterioration prior to local failure can prevent catastrophic failures; hence, a new class of materials is required with specific features and characteristics. This new generation of materials is required to have a nervous character and be part of a smart design, avoiding standard material configuration as well as dense deployment of sensors and actuators with narrow coverage and difficult-to-access locations. Active research and systems are currently being undertaken to deliver some output [ 2 – 4 ]. We expect to build a human-like nervous system in the core material used for frames and structures such as aircraft, cars and other sensitive applications able to capture the stress or ‘pain’ and possible failure of multi-parameters and immediately feedback to an intelligent controller for instant decision making. The nervous material, as a unit panel or block design, will have the ability to monitor large regions of the structure by connecting to each other with sensor clusters. Due to this advantage, it will be potentially possible to:\n (1) reduce maintenance cost, (2) perform an early failure detection in real time, (3) monitor a product that has a large and/or sensitive structure e.g. aircraft or nuclear power plant, and (4) compensate for any deformation and prevent failure (self-healing). Self-healing as a feature for short or long periods has become a hot research topic due to its importance. The concept of smart/intelligent materials and structures can be considered as a step in the general evolution of man-made objects as shown in figure 2 [ 3 ]. This schematic shows a continuous trend from simple materials to complex materials in man-made production. A simple material, which is a homogeneous material, made of natural properties, becomes complex by developing the technology of multi-materials (in particular, composite materials) that allow for the design of new structures with functional properties adapted to specific applications. Figure 2. General evolution of materials/structures used by people, and the place of smart structures. Reprinted from [ 3 ], copyright 1989, with permission from Elsevier. Recently, significant progress has been achieved in the area of structural health monitoring [ 5 ]. Materials with health monitoring systems exhibit several advantages including monitoring all aspects related to damages, loads, conditions, etc. However, these materials have limitations such as robust, effective, self-healing and reliable monitoring characteristics. Vision using structural health monitoring has two aspects: an intelligent sensing network through the structure being monitored, or embedded intelligence in the materials composing the structure. New structures that are made for specific needs should adapt to changing environmental conditions. This requires making them sensitive, controllable and active. Therefore, materials/structures are getting ‘smarter’ with a variety of functional properties. In other words, such materials could also be composed of multi-functional materials that are integrated with sensors, actuators, electronics and communication systems to decide actions at any environmental change. The structural conditions and residual life of a structure will be monitored and predicted by sensors and intelligent diagnostics. Moreover, an embedded sensory network and nervous system would enable structures to have intelligent communication. Due to the intelligent communication system, the structural conditions will be monitored in real time. Because of this advantage, the structure will be more adaptive to the changing environment. Hence, damage will be mitigated and catastrophe prevented with embedded sensors, actuators and multi-functional materials. For example, unmanned aerial vehicles (UAVs) must function reliably after long-term storage, requiring a nervous system integral to the airplane, which will add both complexity and cost. Sensors developed for one purpose can often be adapted to serve other sensory functions. In some cases, they can also serve as actuators. Sensor types and their embedding possibility, as well as embedding techniques and their related challenges, are discussed next with further concluding remarks."
} | 3,168 |
24692632 | PMC3977351 | pmc | 9,711 | {
"abstract": "ABSTRACT “ Candidatus Synechococcus spongiarum” is a cyanobacterial symbiont widely distributed in sponges, but its functions at the genome level remain unknown. Here, we obtained the draft genome (1.66 Mbp, 90% estimated genome recovery) of “ Ca. Synechococcus spongiarum” strain SH4 inhabiting the Red Sea sponge Carteriospongia foliascens . Phylogenomic analysis revealed a high dissimilarity between SH4 and free-living cyanobacterial strains. Essential functions, such as photosynthesis, the citric acid cycle, and DNA replication, were detected in SH4. Eukaryoticlike domains that play important roles in sponge-symbiont interactions were identified exclusively in the symbiont. However, SH4 could not biosynthesize methionine and polyamines and had lost partial genes encoding low-molecular-weight peptides of the photosynthesis complex, antioxidant enzymes, DNA repair enzymes, and proteins involved in resistance to environmental toxins and in biosynthesis of capsular and extracellular polysaccharides. These genetic modifications imply that “ Ca. Synechococcus spongiarum” SH4 represents a low-light-adapted cyanobacterial symbiont and has undergone genome streamlining to adapt to the sponge’s mild intercellular environment.",
"introduction": "INTRODUCTION As one of the oldest, most primitive metazoans, sponges are distributed globally and play important ecological roles ( 1 – 3 ). The association of symbiotic microbes with the sponge was identified several decades ago ( 4 ). Since then, sponge-associated microbial communities and their diversity have been studied extensively ( 5 ). Pyrosequencing techniques further facilitated the investigation of sponge-associated microbes ( 6 , 7 ). A recent study reported up to 32 bacterial phyla and candidate phyla in sponges ( 7 ). To some extent, sponge-associated microbial communities showed sponge-species specificity and tropical-subtropical dissimilarity ( 6 , 7 ). In addition, dissimilarity in the composition of microbial communities between sponges with high and low microbial abundance has been identified ( 8 ). Diverse symbiotic microbes in sponges function in nitrogen fixation, nitrification, photosynthesis, and sulfate reduction and affect the health, ecological distribution, and evolutionary processes of the host ( 5 , 9 ). However, most sponge symbionts are unculturable and some fall into sponge-specific clusters ( 10 ), which makes it difficult to understand their functions and symbiotic mechanisms at the genome level. To date, researchers have only obtained one complete genome of sponge symbiotic microbes that belong to the psychrophilic crenarcheon Cenarchaeum symbiosum ( 11 ). On the other hand, a draft genome of an uncultured deltaproteobacterium in association with the sponge Cymbastela concentrica was extracted from metagenomic data using the tetranucleotide frequency method ( 12 ). The single-cell method has also been used to study the genomes of poribacterial symbionts in marine sponges ( 13 ). However, our knowledge of sponge symbionts at the genome level remains limited. Cyanobacteria represent one of the most common members of the sponge-associated microbial communities and are considered to play important roles in photosynthesis, nitrogen fixation, UV protection, and defensive toxin production ( 5 , 14 ). Identified cyanobacterial sponge symbionts belong to Synechocystis , Aphanocapsa , Anabaena , Oscillatoria , and “ Candidatus Synechococcus spongiarum” ( 5 ). “ Ca. Synechococcus spongiarum,” proposed by Usher et al. ( 15 ), was found in at least 40 sponge species and represents the largest sponge-specific cluster to date ( 10 ). Electron and fluorescence micrographs of “ Ca. Synechococcus spongiarum” symbionts revealed their presence in the intercellular environment of host sponges and provided evidence for their interaction with sponge amebocytes ( 16 ). In addition, vertical transmission of “ Ca. Synechococcus spongiarum” from parents to offspring has been reported ( 17 ). Although the genetic differentiation of “ Ca. Synechococcus spongiarum” is considered to be very low among populations from different host species or geographical regions according to the similarity of the 16S rRNA genes, their internal transcribed spacer (ITS) region displays high variations ( 18 ). The functional properties of this highly prevalent sponge symbiont and its symbiotic interaction with sponges remain unclear. Marine picocyanobacteria of the genera Prochlorococcus and Synechococcus overwhelmingly dominate the picophytoplankton of the world ocean and contribute vitally to global primary production ( 19 , 20 ). The features that distinguish “ Ca. Synechococcus spongiarum” from free-living picocyanobacteria and the mechanism underlying its ability to adapt to the symbiotic partnership are still unclear. Genomic analyses may provide answers to these questions. According to our previous data, “ Ca. Synechococcus spongiarum” is highly abundant in the sponge Carteriospongia foliascens , collected from the Red Sea, and represents the dominant cyanobacterial symbiont (see Fig. S1 in the supplemental material), thus permitting the extraction of its genome from the microbial community. The development of bioinformatics also makes it feasible to obtain genomic sequences of uncultured bacteria from multiple metagenomes by genome binning based on differential coverage and tetranucleotide frequency ( 12 , 21 ). Here, we report a draft genome of “ Ca. Synechococcus spongiarum” strain SH4 extracted from metagenomic data. Using the extracted genome, we examined the evolutionary relationship and functional dissimilarity of “ Ca. Synechococcus spongiarum” SH4 with closely related free-living cyanobacterial strains and so pave the way to understand its adaptive mechanism to the sponge-symbiont partnership.",
"discussion": "RESULTS AND DISCUSSION Genome binning. Metagenomic DNA of the microbial community in the Red Sea sponge C. foliascens was subjected to 454 pyrosequencing, and metagenomic reads were assembled using GS De Novo Assembler (Newbler). A full-length cyanobacterial 16S rRNA gene was predicted from the assembled metagenomic contigs and completely matched with the dominant cyanobacteria (represented by operational taxonomic unit 1691 [OTU1691]; see Fig. S1 in the supplemental material) in the sponge C. foliascens . A phylogenetic tree based on the predicted 16S rRNA gene indicated that this symbiont is distantly related to free-living Synechococcus and Prochlorococcus species and that it groups together with sequences from “ Ca. Synechococcus spongiarum” derived from various sponge species ( 15 , 18 ) and shares more than 99% identity with them ( Fig. 1 ). This cyanobacterial symbiont in the sponge C. foliascens was designated “ Candidatus Synechococcus spongiarum” strain SH4. Analysis of 16S rRNA genes in 454 metagenomic reads revealed a consistently high abundance of the symbiont “ Ca. Synechococcus spongiarum” SH4 (see Fig. S2 in the supplemental material). Among the 16S rRNA reads, 67% (158/236) were assigned to Cyanobacteria , of which 155 reads matched the 16S rRNA gene of SH4 with more than 99% identity. These reads were thus sorted to “ Ca. Synechococcus spongiarum” SH4. The high abundance of SH4 in the metagenome data implied that the genome coverage of SH4 in the assembled contigs (average coverage, ~28×) was much higher than those of other sponge-associated microbes, which facilitated distinguishing contigs of SH4 from the others. FIG 1 Neighbor-joining phylogenetic tree of strain SH4 and closely related cyanobacterial strains based on partial 16S rRNA genes (1,100 bp). CSS, cluster of “ Ca . Synechococcus spongiarum.” Bootstrap values based on 1,000 replications are shown as percentages at branch angles. Scale bar indicates 1% estimated sequence divergence. Using a combination of genome coverage and tetranucleotide frequency patterns and taking the GC content and essential genes into account ( 21 ), we extracted the genome of “ Ca. Synechococcus spongiarum” SH4 from the assembled metagenomic contigs ( Table 1 ; see Fig. S3 in the supplemental material). The draft genome of SH4 was 1.6 Mbp in length, with a GC content of 63.4%. A total of 96 out of 106 single-copy, essential genes were identified in the draft genome, assuming a genome recovery of 90%. Although the genome recovery was not effectively complete, the draft genome is good enough to permit analysis of the functional properties of the sponge symbiont “ Ca. Synechococcus spongiarum.” To validate the occurrence of genome reduction, the genes of interest that were lacking in the SH4 draft genome were checked again in the remaining assembled metagenomic contigs with lengths longer than 500 bp and 454 metagenomic coverage higher than 8×. TABLE 1 General features and functional comparison of “ Candidatus Synechococcus spongiarum” SH4 and related picocyanobacterial strains Genome feature or gene function Value for indicated taxon a 1 2 3 4 5 6 7 8 Genome recovery ~90% F D D F F D F Genome size (Mbp) 1.66 2.22 2.58 3.04 2.61 1.75 2.83 3.34 No. of essential genes 96 106 104 106 105 106 105 106 % GC content 63.4 60.8 64.5 65.4 52.5 36.4 68.7 68.7 % Coding density 87.5 94.2 92.7 88.4 87.2 89.3 87.6 89.5 No. of genes in KEGG pathways Photosynthesis 47 60 64 61 64 53 57 59 Antenna proteins 24 28 18 23 28 8 21 17 TCA cycles 10 9 8 11 8 11 11 11 DNA replication 13 13 13 14 13 13 15 14 DNA repair 31 38 41 47 38 36 40 41 Cysteine/methionine metabolism 10 21 22 24 19 17 21 24 No. of genes in SEED/subsystems Response to oxidative stress 12 28 26 32 26 21 29 31 Resistance to antibiotics and toxins 8 22 22 28 23 9 30 36 CPS and EPS biosynthesis 3 13 13 16 10 8 19 17 Gram-negative cell wall 7 14 12 14 9 10 20 17 Peptidoglycan biosynthesis 16 15 15 17 15 15 17 16 Eukaryotic-like domain Fibronectin type III domain 2 0 0 1 0 0 0 0 AR 17 0 0 0 0 0 0 0 TPR 3 9 7 17 1 6 12 14 LRR 9 0 0 0 0 2 0 0 a Taxa: 1, “ Ca. Synechococcus spongiarum” SH4; 2, Synechococcus sp. strain RCC307; 3, Synechococcus sp. strain RS9917; 4, Synechococcus sp. strain WH 5701; 5, Synechococcus sp. strain CC9311; 6, P. marinus CCMP137; 7, Cyanobium sp. strain PCC 7001; 8, Cyanobium gracile PCC 6307. Genome recovery: F, finished genome; D, draft genome. Phylogenomic inference. The explosive growth of genomic data has provided conserved marker genes as alternatives to 16S rRNA genes for phylogenetic inference ( 22 ). Here, a phylogenomic tree based on 31 concatenated marker genes ( Fig. 2a ) revealed the evolutionary distinction of SH4 from picocyanobacteria of the genera Synechococcus , Prochlorococcus , and Cyanobium cluster Synechococcus and supported its evolutionary divergence from free-living cyanobacteria ( 18 ). The bipartition point where SH4 branched from other picocyanobacteria suggested that SH4 was an independent cyanobacterial lineage that had adapted to the symbiotic lifestyle for a long period of time. This is in accord with previous findings demonstrating that these symbionts are vertically inherited from the parent sponges ( 17 ). Average nucleotide identity (ANI) and tetranucleotide frequency are powerful tools for comparison analysis of genome composition ( 23 ). SH4 showed higher ANIs and tetranucleotide frequency similarity with Synechococcus and Cyanobium cluster Synechococcus than to Prochlorococcus ( Fig. 2b ). In addition, SH4 represented one of the cyanobacterial strains with the highest GC contents (>60%) and was more similar to Synechococcus and Cyanobium cluster Synechococcus than to Prochlorococcus ( Fig. 2c ). These results indicated that SH4 was more closely related to free-living Synechococcus than to Prochlorococcus . Accordingly, closely related strains affiliated with Synechococcus and Cyanobium cluster Synechococcus were selected for the genome-level functional comparison with SH4 ( Table 1 ). The low-light-adapted cyanobacterium Prochlorococcus marinus strain SS120 (CCMP1375 in Table 1 ), with a nearly minimal oxyphototrophic genome, was also included in the comparison analysis. FIG 2 Phylogenomic inference of “ Ca . Synechococcus spongiarum” SH4. (a) Maximum-likelihood phylogenomic tree using 31 concatenated conserved proteins of SH4 and cyanobacterial strains. The branches were colored according to the bootstrap values, ranging from purple (80%) to red (100%). Scale bar indicates 10% estimated sequence divergence. (b) ANI values plus correlation indexes of tetranucleotide frequency between SH4 and closely related Synechococcus , Prochlorococcus , Cyanobium cluster Synechococcus , and other cyanobacterial strains revealed differences in their genome composition. (c) GC content and genome size of SH4 and other cyanobacterial strains. Functional features at the genomic level. The numbers of genes present in selected pathways of “ Ca. Synechococcus spongiarum” SH4 and free-living picocyanobacteria strains are presented in Table 1 ; these data suggest the near completeness of key functional pathways, including photosynthesis, the citric acid cycle (tricarboxylic acid [TCA] cycle), DNA replication, and peptidoglycan biosynthesis. As a sponge symbiont, SH4 also displayed unique symbiotic features, with the highlight being genome streamlining following the loss of unnecessary genes in several pathways ( Table 1 ). Previous studies have revealed the distribution and evolutionary divergence of the sponge symbiont “ Ca. Synechococcus spongiarum” using molecular ecology techniques ( 18 ). Although “ Ca. Synechococcus spongiarum” was thought to enhance host metabolism and ecological fitness by providing a photosynthesis-derived carbon source, its functional properties at the genome level remain unclear. Here, the extracted genome provided direct insights into the carbon and energy metabolism of “ Ca. Synechococcus spongiarum” SH4 and its symbiotic adaptation to the sponge host. Host-symbiont interaction. Proteins with ankyrin repeats (ARs), leucine-rich repeats (LRRs), and fibronectin type III domains were enriched in SH4 but rare in free-living cyanobacteria ( Table 1 ). Proteins with eukaryoticlike domains, such as ARs, tetratricopeptide repeats (TPRs), LRRs, NHL repeats (PF01436), and fibronectin type III, have been reported to be enriched in sponge symbiotic microbes ( 12 , 24 ) and were suggested to modulate host behavior by interfering with eukaryotic protein-protein interactions. Specially, AR proteins from sponge symbionts modulate amoebal phagocytosis and might help symbionts escape digestion by the sponge host ( 25 ). The enrichment of these domains in SH4 is consistent with the role of “ Ca. Synechococcus spongiarum” as a sponge symbiont. Interestingly, the number of TPR proteins was lower in SH4 than in several free-living cyanobacterial strains ( Table 1 ). The TPR motif was originally identified in yeast ( 26 ) but was recently found in a wide variety of prokaryotes and suggested to be involved in numerous cell processes ( 27 ). The hypothesis that TPR functions as a symbiotic factor in the sponge-symbiont interaction requires further careful evaluation. Amino acid metabolism. Although obligatory symbiotic bacteria tend to lose essential metabolic pathways that are required for free-living organisms, especially those responsible for amino acid metabolism ( 28 , 29 ), the numbers of genes in most of the amino acid metabolism pathways of “ Ca. Synechococcus spongiarum” SH4 were similar to the numbers in free-living cyanobacterial strains (see Fig. S4 in the supplemental material). Surprisingly, the cysteine and methionine metabolism pathway was dramatically reduced ( Table 1 ). In-depth analysis showed that there was no enzyme for the de novo biosynthesis of the methionine precursor (homocysteine), although methionine synthase ( metH ) was present in the SH4 draft genome ( Fig. 3 ). In addition to de novo biosynthesis, the methionine salvage pathway is an important metabolic pathway for maintaining the concentration of l -methionine in bacteria ( 30 ). Key genes in the methionine salvage pathway, including S -adenosylmethionine decarboxylase ( speD ), spermidine synthase ( speE ), methylthioribose-1-phosphate isomerase ( mtnA ), and 1,2-dihydroxy-3-keto-5-methylthiopentene dioxygenase ( mtnD ), were lost ( Fig. 3 ). The lack of de novo biosynthesis and salvage of methionine suggested that the essential methionine might be provided by exogenous sources. In the methionine salvage pathway, SpeD and SpeE are two enzymes responsible for the biosynthesis of spermidine, a prevalent polyamine found in bacteria ( 31 ). Previous studies have shown that polyamines in bacteria play important roles in optimal cell growth, signaling cell differentiation, DNA protection, biofilm formation, and antibiotic resistance ( 31 ). High-performance liquid chromatograph analysis revealed that polyamines were widely distributed in cyanobacteria, which indicates their important roles for the cyanobacteria ( 32 ). SH4 lives in the sponge intercellular environment ( 18 ) and likely acquires exogenous polyamines therein. “ Ca. Synechococcus spongiarum” has been shown to interact with host amebocytes ( 18 ), a mobile cell responsible for food digestion. Amebocytes might digest food and release nutrients to satisfy the necessities, such as polyamines and methionine, for the cyanobacterial symbiont. Since other symbiotic bacteria were also found in the sponge host ( Fig. S1 ), they might be an alternative source of these essential chemicals for “ Ca. Synechococcus spongiarum.” FIG 3 Schematic overview of the methionine metabolism pathway in “ Ca . Synechococcus spongiarum” SH4. Red and green labels indicate the absence and presence, respectively, of that enzyme in SH4. Photosynthetic system. “ Ca. Synechococcus spongiarum” has been reported to contain phycocyanin and phycoerythrin ( 15 ). Here, we showed that “ Ca. Synechococcus spongiarum” SH4 contains genes encoding all three types of antenna proteins (phycocyanin, phycoerythrin, and allophycocyanin) ( Fig. 4 ), which suggests that this symbiont absorbs a wide spectrum of light for photosynthesis. The eight subunits of F-type ATPase were identified. Compared to free-living cyanobacterial strains, however, genes in photosystem II (PSII), including psbP , psbI , psbK , psbM , and psbY , were missing ( Fig. 4 ). In further analysis, these genes could not be detected in the entire assemblage of metagenomic contigs or raw pyrosequencing reads. PsbP, together with PsbO, PsbQ, PsbU, and PsbV, form the oxygen-evolving complex (OEC) of PSII in cyanobacteria. This complex oxidizes water to provide protons for photosystem I (PSI). Synechocystis sp. strain PCC 6803 mutants with inactive PsbP exhibit reduced photoautotrophic growth ( 33 ). PsbI, PsbK, PsbM, and PsbY are low-molecular-weight peptides involved in the assembly, stabilization, dimerization, and photoprotection of the photosynthetic center of PSII ( 34 ). The loss of PsbP and low-molecular-weight peptides implied that the PSII complex of “ Ca. Synechococcus spongiarum” SH4 is less stable than those of free-living strains and may represent a low-light-adapted photosynthetic system ( 35 ). Several genes encoding low-molecular-weight peptides in the PSI complex and cytochrome b 6 /f were also not found in SH4 ( Fig. 4 ). The abnormalities observed in the photosynthetic system of “ Ca. Synechococcus spongiarum” SH4 might represent a protective mechanism against damage caused by a high dosage of photosynthesis-derived oxygen and/or oxidative stress to the sponge host ( 36 ). FIG 4 Abundance of photosynthetic genes in “ Ca . Synechococcus spongiarum” SH4 and related cyanobacteria strains based on KEGG orthology annotation. The identities of the cyanobacterial strains are described in Table 1 . Resistance to oxidative stress. Reactive oxygen species (ROS) are by-products of aerobic metabolism and can cause intracellular oxidative damage. ROS generated by the photosynthetic electron transport chain pose a significant threat to photosynthetic organisms, such as cyanobacteria ( 36 ). The ability to rapidly perceive ROS and initiate antioxidant defense is crucial for the survival of these organisms. In cyanobacterial strains, antioxidant enzymes play important roles in resistance to oxidative stress ( 36 ). However, “ Ca. Synechococcus spongiarum” SH4, as a photosynthetic microbe, has lost several antioxidant enzymes, including superoxide dismutase (SOD), glutathione peroxidase (GPX), and DNA-binding protein (Dps) ( Fig. 5 ; see Table S1 in the supplemental material). P. marinus strain CCMP1375, a low-light-adapted picocyanobacteria with a nearly minimal oxyphototrophic genome ( 35 ), also lacks SOD and Dps ( Fig. 5 ) and escapes from oxidative damage through living at the bottom of the illuminated layer ( 35 ). The observed loss of antioxidant enzymes consistently confirmed that SH4 is a low-light-adapted organism. However, different from the antioxidant mechanism used by P. marinus CCMP1375 ( 35 ), “ Ca. Synechococcus spongiarum” SH4 lives in the mild intercellular environment of the sponge host ( 16 ), which provides the barrier of the sponge body to forbid too much sunlight arriving at the cyanobacterial symbiont, and thus it avoids the oxidative damage caused by a highly efficient photosynthesis process. FIG 5 Genes of “ Ca . Synechococcus spongiarum” SH4 and related cyanobacterial strains important for resistance to oxidative stress, antibiotics, and environmental toxins based on SEED/Subsystems annotation. See details in Tables S1 and S2 . The identities of the cyanobacterial strains are described in Table 1 . Resistance to antibiotics and toxic compounds. In addition to the loss of genes encoding antioxidant enzymes, there was also a dramatic reduction of genes involved in resistance to antibiotics and environmental toxins in “ Ca. Synechococcus spongiarum” SH4 ( Fig. 5 ; see Table S2 in the supplemental material). There was a depletion of genes encoding proteins involved in arsenic resistance, multidrug resistance efflux pumps, integrase, beta-lactamase, and the negative regulator of beta-lactamase. Genes encoding cobalt-zinc-cadmium and methicillin resistance were also dramatically reduced in “ Ca. Synechococcus spongiarum” SH4 compared to their occurrence in other free-living cyanobacterial strains. Interestingly, a large fraction of these genes were also lost in the genome of Prochlorococcus marinus CCMP1375. Cyanobacteria are a large and highly diverse group of photosynthetic prokaryotes which can adapt to various habitats, including those containing natural and artificial antibiotics and heavy metals ( 37 ). Resistance to these toxins in open water is important for the survival of these organisms ( 38 ). However, “ Ca. Synechococcus spongiarum,” which inhabits the mild intercellular environment of the sponge host ( 16 ), can evade these toxins via the barriers of the sponge host. Accordingly, genes involved in resistance to environmental factors are not required and were lost during the evolutionary development of this symbiont. Cell wall and capsule composition. Similar to the case for closely related free-living cyanobacterial strains, most of the genes responsible for the biosynthesis of peptidoglycan and lipopolysaccharide (LPS) could be found in SH4 ( Table 1 ). The presence of these genes allows the formation of a rigid cell wall and is consistent with the characteristic spiral thylakoid membrane of “ Ca. Synechococcus spongiarum” observed by transmission electron microscopy ( 16 ). Interestingly, genes responsible for the biosynthesis of capsular polysaccharide (CPS) and extracellular polysaccharides (EPS) were almost completely lost ( Table 1 ). CPS and EPS are extracellular products of a wide range of microorganisms, including cyanobacteria ( 39 ), and play important roles, such as protection against environmental stresses ( 40 ), biofilm formation ( 41 ), and survival against phagocytosis or antibiotics ( 42 , 43 ). The absence of CPS and EPS further indicated that SH4 has a low resistance to environmental stresses. However, this characteristic is likely to diminish the barrier between the symbiont and sponge cells, thus benefiting sponge-symbiont interactions and nutrient exchange. DNA replication and repair. Just like other reported bacterial symbionts ( 28 ), “ Ca. Synechococcus spongiarum” SH4 retained the same set of genes for DNA replication as are found in closely related free-living cyanobacterial strains, but genes for DNA repair capabilities are limited ( Table 1 ). Although the base excision repair and nucleotide repair pathway were found in SH4, the exonuclease Exo VII complex in the mismatch repair pathway, the exonuclease V complex (RecBCD) in the homologous recombination pathway, and ATP-dependent DNA ligase were not detected ( Table 1 ; see Table S3 in the supplemental material). There are also reports of a reduction of DNA repair genes in the symbiotic cyanobacterium UCYN-A ( 44 ) and in the cyanobacterium-originating inclusions termed chromatophores ( 28 ). The absence of DNA repair genes in “ Ca. Synechococcus spongiarum” SH4, similar to many cases of bacterial symbionts with extraordinarily small genomes ( 45 ), likely facilitates the evolution of the genome to adapt to the symbiotic partnership. Overview of the possible lifestyle of “ Ca. Synechococcus spongiarum” SH4. Based on the analysis of the extracted draft genome and its comparison with those of free-living picocyanobacteria, we proposed schematic functional features and adaptive schemes of “ Ca. Synechococcus spongiarum” SH4 to the sponge-symbiont partnership ( Fig. 6 ). Proteins with eukaryoticlike domains were identified in SH4, which is consistent with the role of “ Ca. Synechococcus spongiarum” as a sponge symbiont. Genes involved in the biosynthesis of methionine and spermidine were lost, suggesting that “ Ca. Synechococcus spongiarum” depends on the sponge host and/or the other sponge symbionts for essential nutrients and chemical factors. The presence of a functional photosynthesis pathway in this symbiont guarantees a steady carbon supply to the host and ensures its ecological success ( 5 ). However, the photosynthetic system in SH4 might be unstable and has a low efficiency due to the reduction of PsbP in the OEC complex and the loss of several low-molecular-weight peptides. Furthermore, SH4 should have a low resistance to oxidative stress because of the loss of several antioxidant enzymes. These features indicate that “ Ca. Synechococcus spongiarum,” similar to Prochlorococcus marinus SS120, should be a low-light-adapted picocyanobacterium but uses the alternative strategy of symbiosis for adaptation to low light ( 35 ). “ Ca. Synechococcus spongiarum” also had a low resistance to environmental antibiotics and toxins, which may be further compromised by the defect in the biosynthesis of CPS and EPS. These features force the symbiont to inhabit the mild intercellular environment of the host. However, the defect in the biosynthesis of CPS and EPS may represent a mechanism used to diminish the barrier between symbiont and sponge cells to benefit sponge-symbiont interactions and nutrient exchange. In addition, the loss of DNA repair genes may play roles in facilitating the genome evolution of “ Ca. Synechococcus spongiarum” SH4 to adapt to the sponge-symbiont partnership. FIG 6 Schematic of mode of life of the sponge symbiont “ Ca . Synechococcus spongiarum.” The schematic figure was deduced from the genomic analysis of the draft genome of strain SH4. Summary. Picocyanobacteria in the genera Prochlorococcus and Synechococcus numerically dominate the picophytoplankton communities of the world’s ocean ( 19 , 20 ). During their adaptation to open ocean and coastal environments, these organisms have overcome various environmental stresses ( 35 , 46 ). In contrast to free-living picocyanobacteria, “ Ca. Synechococcus spongiarum” is a sponge symbiont. The exclusive detection of “ Ca. Synechococcus spongiarum” in sponges ( 18 ), their vertical transmission between generations ( 17 ), and their large phylogenomic dissimilarity to free-living picocyanobacteria ( Fig. 2 ) suggest an intimate symbiotic relationship between “ Ca. Synechococcus spongiarum” and the sponge host. The draft genome of “ Ca. Synechococcus spongiarum” SH4 provided further insight into the adaptive mechanism of this intercellular symbiont to live in the sponge host ( Fig. 6 ). Although the draft genome is estimated to have a recovery of 90% and is not effectively completed, the absence of certain genes has been confirmed through searching against the entire assemblage of metagenomic contigs. However, the incomplete genome precludes the detection of several other potentially symbiotic features, such as transposase-driven genome rearrangement and horizontal gene transfer ( 45 ). The recovery of an effectively complete genome by combining a metagenome and a single-cell-derived genome, perhaps even using the bacterial artificial chromosome (BAC) library method, should further elucidate this symbiotic partnership. According to the observed intimate interdependence of the symbiont with the sponge and the phylogenomic dissimilarity with free-living picocyanobacteria, our study suggests that the symbiotic partnership between “ Ca. Synechococcus spongiarum” and the sponge has been established for a long time. However, this conclusion is inconsistent with the widespread distribution but low genetic differentiation of “ Ca. Synechococcus spongiarum” ( 47 ). Although cryptic diversity of “ Ca. Synechococcus spongiarum” among sponges has been suggested based on the variation in ITS regions, which supports the potential cryptic genetic differentiation ( 18 ), additional evidence is required to confirm the intraspecies diversity and divergence of this sponge symbiont. Due to the high abundance of “ Ca. Synechococcus spongiarum” in the sponge C . foliascens , binning genomes of these symbionts from multiple individuals of this sponge species located in a single and/or different geographical sites will improve our knowledge about their genetic diversity and differentiation."
} | 7,620 |
29474495 | PMC5825115 | pmc | 9,712 | {
"abstract": "The biobased production of adipic acid, a precursor in the production of nylon, is of great interest in order to replace the current petrochemical production route. Glucose-rich lignocellulosic raw materials have high potential to replace the petrochemical raw material. A number of metabolic pathways have been proposed for the microbial conversion of glucose to adipic acid, but achieved yields and titers remain to be improved before industrial applications are feasible. One proposed pathway starts with lysine, an essential metabolite industrially produced from glucose by microorganisms. However, the drawback of this pathway is that several reactions are involved where there is no known efficient enzyme. By changing the order of the enzymatic reactions, we were able to identify an alternative pathway with one unknown enzyme less compared to the original pathway. One of the reactions lacking known enzymes is the reduction of the unsaturated α,β bond of 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid. To identify the necessary enzymes, we selected N -ethylmaleimide reductase from Escherichia coli and Old Yellow Enzyme 1 from Saccharomyces pastorianus . Despite successful in silico docking studies, where both target substrates could fit in the enzyme pockets, and hydrogen bonds with catalytic residues of both enzymes were predicted, no in vitro activity was observed. We hypothesize that the lack of activity is due to a difference in electron withdrawing potential between the naturally reduced aldehyde and the carboxylate groups of our target substrates. Suggestions for protein engineering to induce the reactions are discussed, as well as the advantages and disadvantages of the two metabolic pathways from lysine. We have highlighted bottlenecks associated with the lysine pathways, and proposed ways of addressing them.",
"conclusion": "Conclusions We have identified an alternative pathway for converting lysine to adipic acid, with a reduced number of unknown enzymes and maintained, balanced redox potential. In addition, we have identified several parameters important for the achievement of the enzymatic reduction of unsaturated α,β bonds of 6-amino- trans -2-hexenoic acid and trans- 2-hexenedioic acid. We showed theoretically that the substrates could fit in the enzymatic pocket, and that increasing the electron withdrawing potential by protein engineering to create additional hydrogen bonds between both oxygens in the carboxylate group should improve the catalytic capability. We also suggest that residues that have high electron withdrawing potential for the creation of hydrogen bonds should be utilized in protein engineering strategies. We believe that this study has provided important knowledge that will be useful in realizing a metabolic pathway for biobased adipic acid production.",
"introduction": "Introduction The biobased production of adipic acid could be used to replace the current petrochemical-based production route and thus contribute to the more sustainable production of this platform chemical, which is used primarily for the production of nylon. Lignocellulosic materials from forestry and agricultural activities are interesting biobased raw materials for the replacement of oil-based raw material. The sugars in lignocellulose biomass, including glucose and other monosaccharides, can be released after pretreatment and hydrolysis, followed by microbial conversion, using metabolically engineered microorganisms, to convert glucose to adipic acid. Adipic acid is not naturally produced microbially to any great extent, and engineering microorganisms with an efficient metabolic pathway for the conversion of glucose into adipic acid presents a considerable challenge. In recent years, several metabolic pathways have been proposed, and some have been demonstrated to be functional [ 1 – 6 ]. However, titers and yields are far from being industrially relevant. The choice of an efficient metabolic pathway together with efficient enzymes for the targeted metabolic pathway are important for improving the titers and yields of adipic acid. One possible pathway, that we have studied, is based on lysine conversion [ 1 ] ( Fig 1 ). The theoretical maximum yield to adipic acid from glucose via the lysine pathway is reported to be between 40–50% [ 1 ]. Lysine is a ubiquitous metabolite that can be naturally synthesized from glucose by most microorganisms and it is also produced industrially by microorganisms at large scale [ 7 , 8 ]. In the first enzymatic reaction step, the NH 2 group located on the α-carbon is removed from lysine via a carbon-nitrogen lyase, forming 6-amino- trans- 2-hexenoic acid. In the second enzymatic reaction, the unsaturated α,β bond of 6-amino- trans- 2-hexenoic acid is reduced by an oxidoreductase, utilizing the cofactor NAD(P)H, to form 6-aminocaproic acid. In the third reaction, the amino group of 6-aminocaproic acid is transferred to α-ketoglutaric acid by a transaminase to form adipic acid semialdehyde and L-glutamic acid. In the fourth and final reaction, adipic acid semialdehyde is converted into adipic acid via an oxidoreductase utilizing NAD(P) + as cofactor. The main strength of this pathway, compared to many of the other proposed pathways for the production of adipic acid, is that the metabolic pathway from lysine to adipic acid is neutral in terms of redox potential, hence the overall intracellular redox balance is not affected, reducing the risk of lowering the overall yield due to the production of undesirable by-products. While lysine biosynthesis consumes 4 moles of NADPH [ 9 , 10 ], NADPH is not additionally consumed from converting lysine to adipic acid ( Fig 1 ). Synthetic pathways, that are redox neutral, have been successfully introduced into lysine-overproducing strains for direct production of the lysine-derived value-added chemicals from glucose [ 11 – 14 ]. Direct production of 1,5-diaminopentane from glucose using engineered Corynebacterium glutamicum with as high as 50% molar yield [ 15 ] has been reported, while the maximum theoretical yield for lysine from glucose is reported to be 75% molar yield [ 16 ]. Such pathways have also been introduced into microorganisms for conversion of extracellularly supplemented lysine [ 17 , 18 ] with as high as 99.9% molar yield [ 19 ]. However, the main challenge in this adipic acid pathway ( Fig 1 ) is that for three of the four metabolic reactions there is no known enzyme that can perform the required reaction efficiently. It is thus necessary to identify efficient enzymes, or to find an alternative pathway with known efficient enzymes. 10.1371/journal.pone.0193503.g001 Fig 1 Metabolic pathway for the conversion of lysine to adipic acid proposed by Burgard A. et al. [ 1 ]. The reaction number R05099 according to the Kyoto Encyclopedia of Genes and Genomes is given for the known reaction. For known enzymes, the E.C. numbers according to the Braunschweig Enzyme Database are given. akg = alpha-ketoglutaric acid, Glu-L = L-glutamic acid. The primary aim of the present study was to investigate the possibilities and limitations of the lysine pathway bearing in mind that the pathway includes several unknown enzymes. Firstly, efforts were made to identify an alternative pathways from lysine to adipic acid with the aim of reducing the number of unknown enzymes. Efforts were also made to identify enzymes capable of reducing the intermediates with unsaturated α,β bonds in the pathways namely 6-amino- trans -2-hexenoic acid (in the pathway shown in Fig 1 ) and trans -2-hexenedioic acid (in the alternative pathway found). Enoate reductases are capable of reducing the unsaturated α,β bonds in a range of substrates [ 20 , 21 ]. However, enoate reductases are sensitive to oxygen, due to the presence of iron-sulfur clusters, and the pathway from lysine requires aeration for maximal yield [ 1 ]. Therefore, enoate reductases were not considered in the present study. Instead the enzyme N -ethylmaleimide reductase (NemA) from Escherichia coli was chosen since it has been reported to convert 6-amino- trans -2-hexenoic acid, an intermediate of the original pathway [ 1 ], to 6-aminocaproic acid, although with a very low yield and productivity (<0.5% mol product per mol substrate after 48 hours of incubation) [ 22 ]. While the natural function of NemA is yet to be illustrated, it is known to confer N -ethylmaleimide resistance to E . coli [ 23 ]. We also chose the closely related Old Yellow Enzyme 1 (Oye1) from Saccharomyces pastorianus [ 24 ], which the natural function is also yet to be illustrated despite extensive studies for decades [ 25 ]. The enzyme Oye1 is known to reduce unsaturated α,β bonds in a broad range of substrates, including aldehydes and ketones [ 26 ], but its potential when acting on carboxylic acids is less promising [ 27 – 29 ]. Therefore, the secondary aim of this study was to carefully investigate the possibilities and limitations of reducing the unsaturated α,β bonds of the carboxylic acid intermediates 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid. A combined approach, in which docking of the enzymes and the substrates in silico was combined with experimental work in vitro gave us a detailed understanding of the interactions between the enzymes and the substrates, and helped to identify the most important obstacles that must be overcome in order to efficiently reduce the unsaturated α,β bonds of 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid.",
"discussion": "Discussion In the present work, we have investigated if and how the pathway proposed by Burgard et al . [ 1 ] for the conversion of lysine into adipic acid could be realized. The main challenge associated with this pathway is the lack of efficient enzymes for three of the four enzymatic reactions. To circumvent this problem, we proposed an alternative pathway, in which the order of the chemical reactions was changed. In doing this, we were able to reduce the number of unknown enzymatic steps, and include an additional substrate for each type of chemical reaction, without affecting the balanced redox potential of the pathway ( Fig 2 ). We also set out to identify important factors in the efficient reduction of the unsaturated α,β bonds of the pathway-specific intermediates 6-amino- trans- 2-hexenoic acid and trans- 2-hexenedioic acid, under aerobic conditions. Aerobic conditions are required to maximize the yields of the selected pathways [ 1 ], however efficient reduction of unsaturated α,β bonds in carboxylates under aerobic conditions has previously proven difficult [ 27 – 29 ]. Our aim was to gain detailed knowledge on the reaction, and the enzymes NemA and Oye1 were investigated with regard to their potential to reduce the unsaturated α,β bond of 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid both theoretically and experimentally. According to the results obtained from the in silico modelling, both target substrates, 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid, and the carboxylic substrate trans -2-hexenoic acid could fit in the enzymatic pocket of both Oye1 and NemA. In the case of Oye1 two hydrogen bonds were observed with one of the oxygen atoms in the carboxylate group with the known substrate-binding residues Asn194 and His191 [ 25 ]. NemA also created two hydrogen bonds, but separately, with one hydrogen bond to each of the oxygens of the carboxylate group, one with residue His182 and the other with His185 ( Fig 4 ). Despite the difference in hydrogen bonds, the distance between the N 5 of FMN and the β-carbon of both target substrates and trans -2-hexenoic acid, as well as the angle formed by N 10 /N 5 and the β-carbon were within the range necessary for the reaction to take place with both proteins [ 49 , 52 ], indicating a reaction favoring binding. Despite the positive binding predicted by in silico modelling, no activity was observed in vitro on either of the target substrates or the carboxylic substrate trans -2-hexenoic. These results are consistent with those in a recent study in which in vitro results for trans -2-hexenoic acid were negative for YqjM, an Oye family member, despite a successful in silico docking experiment [ 59 ]. While this does not explain the negative in vitro observations, it shows that successful docking results do not necessarily mean successful in vitro results. The fact that no activity was observed experimentally may be partly due to a lack of insight into whether hydride transfer to the β-carbon actually takes place, and if it does, how rapid the reaction proceeds. For the reaction to occur, the unsaturated α,β bond must be activated. Aldehydes and ketones have electron withdrawing potential which, upon hydrogen bonding with the enzyme, will increase further and activate the unsaturated α,β bond ( Fig 7 ). Carboxylic acids, on the other hand, do not have any electron withdrawing potential at neutral pH, and are thus unable to activate the unsaturated α,β bond ( Fig 8A ). The difference in electron withdrawing potential was not considered in the present docking experiments, but we attempted to induce electron withdrawing potential of the carboxylate group by protonating the acid by lowering the pH. However, the enzymes lost their catalytic activity completely at pH below the pKa of the target substrates and trans -2-hexenoic acid. An alternative way of increasing the electron withdrawing potential is by creating additional hydrogen bonds to the oxygens of the carboxylate group. If both oxygens of the carboxylate group form two hydrogen bonds each with the enzyme, this could induce an electron withdrawing potential, and the unsaturated α,β bond would be activated, favoring the reaction ( Fig 11 ). In order to test this hypothesis, in-depth in silico studies should be carried out to identify protein engineering strategies to create such hydrogen bonds in the enzymes NemA and Oye1. The engineering strategy should also take into account differences in the electron withdrawing potentials of the catalytic residues. For instance, positively charged residues, such as protonated histidine, are more likely to attract electrons and favor reaction than polar residues, such as asparagine. 10.1371/journal.pone.0193503.g011 Fig 11 Proposed mechanism for reduction of deprotonated carboxylic acid by Oye1. Engineering of the enzyme by substitution of the native residues with putative X and Y residues could lead to the formation of hydrogen bonds between the enzyme binding pocket and both oxygens of the carboxylate group. Upon hydrogen bonding to the enzyme (hashed lines) electrons from the double bond are shifted towards the catalytic residues Asn194 and His191 for one of the oxygens, and to the residues X and Y for the other oxygen (dotted lines), thereby creating a partial positive charge on the β-carbon (δ + ) of the substrate, which activates the double bond, making it prone to attack. When the double bond is activated the transfer of a hydride from the flavin N 5 to the β-carbon of the substrate and protonation from Tyr196 can occur, resulting in hexanoic acid as the final product. The movement of electrons involved in the hydride attack and protonation are indicated by the curved arrows. For simplicity, only the mechanism for Oye1 is shown. Although the results from docking studies suggested that the target substrates 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid, together with trans -2-hexenoic acid, could fit in the active site and create hydrogen bonds with both NemA and Oye1, despite the fact no enzymatic activity could be observed, we suggest, based on computed LIE data, that these substrates have better interaction with water than the catalytic pocket. If the trans -2-hexenoic acid does not stay in the pocket long enough, catalysis might not take place on the enoate moiety. Since trans -2-hexenoic acid preferentially interact with water than the enzyme, it spends more time in the solution than in the catalytic pocket. While the aldehyde moiety in trans -2-hexenal is stabilized by hydrogen bonds within the catalytic pocket, the binding residues might not be efficiently stabilizing the negative charge on enoate on trans -2-hexenoic acid. Thus, engineering the catalytic pocket for further stabilizing the enoate moiety will be necessary. Another possible reason why 6-amino- trans -2-hexenoic acid was not reduced, could be that this substrate spontaneously forms an internal ring structure leading to the formation of 2-pyrrolidineacetic acid upon warming. The compound 2-pyrrolidineacetic acid does not contain an unsaturated α,β bond and hence cannot serve as a substrate for NemA or Oye1. The finding that this substrate can form 2-pyrrolidineacetic acid upon warming, whereas trans -2-hexenedioic acid does not, favors our suggested alternative pathway rather than the one proposed by Burgard et al . ( Fig 2 ). According to the in silico study, trans -2-hexenedioic acid can bind to the enzymes with either of its two carboxylate groups. Which position favors the reaction depends on which of the two carboxylates groups binds to the enzyme. The possibility of trans -2-hexenedioic acid binding with the catalytic pocket via either of its carboxylate groups, where one is in a suitable position for reaction whereas the other is not, will reduce the rate of the reaction, but not prevent it. However, the low reaction rate of the alternative pathway for the production of adipic acid from lysine must be addressed. Despite our negative in vitro results, the previously reported activity of NemA on 6-amino- trans -2-hexenoic acid for reduction of the unsaturated α,β bond to yield 6-aminocaproic acid [ 22 ] indicates that NemA may be an interesting enzyme to target. However, the extremely low yield achieved (< 0.5% moles of product per mole substrate) after 48 hours of incubation, despite employing an assay with regeneration of NADPH, brings into question the feasibility of the enzyme for in vivo reactions. Nevertheless, NemA may be a suitable target for protein engineering, as discussed in this paper. Protein engineering of NemA could result in an enzyme able to catalyze the challenging reactions; namely the reduction of the unsaturated α,β bonds of 6-amino- trans -2-hexenoic acid and trans -2-hexenedioic acid, which are important steps towards the fulfillment of biobased production of adipic acid. The α,β-reduction we aimed for here intrinsically requires removal of the α-amino group both in lysine and amino adipic acid. This is one of the difficult steps in the pathway especially because it simultaneously requires removal of non-acidic hydrogen at the β-position. Thermodynamics also does not favor releasing α-amino groups ( S2 Table ). Our targeted reaction has so far only been shown to be possible for certain aromatic amino acids, e.g., tyrosine or phenylalanine, where the β-position hydrogen can be abstracted [ 60 ]. Thus, it might be necessary to design a different pathway consisting of transamination of the α-amino group and subsequent reduction of the ketone group followed by dehydration to reach enoate moiety. Moreover, the reduction reaction discussed in the present study highly resembles part of the chain elongation step in the fatty acid biosynthesis and reverse β-oxidation. During fatty acid biosynthesis, the acyl moieties are bound to acyl carrier protein (ACP) [ 61 ] while they are bound to coenzyme A during reverse β-oxidation [ 4 , 5 ]. Hence, the oxidoreductases under the enzyme classification 1.3.1. acting on enote moieties are active on ACP- or coenzyme A- activated substrates. ACP- or coenzyme A-independent oxidoreductases in the category are mostly oxygen sensitive [ 62 ]. Thus, in order to reduce the α,β-double bond in the enoate moieties, it might be necessary to activate the substrates with coenzyme A and subsequently release coenzyme A after the reduction has taken place."
} | 5,011 |
39191782 | PMC11350166 | pmc | 9,717 | {
"abstract": "Processing spatiotemporal data sources with both high spatial dimension and rich temporal information is a ubiquitous need in machine intelligence. Recurrent neural networks in the machine learning domain and bio-inspired spiking neural networks in the neuromorphic computing domain are two promising candidate models for dealing with spatiotemporal data via extrinsic dynamics and intrinsic dynamics, respectively. Nevertheless, these networks have disparate modeling paradigms, which leads to different performance results, making it hard for them to cover diverse data sources and performance requirements in practice. Constructing a unified modeling framework that can effectively and adaptively process variable spatiotemporal data in different situations remains quite challenging. In this work, we propose hybrid spatiotemporal neural networks created by combining the recurrent neural networks and spiking neural networks under a unified surrogate gradient learning framework and a Hessian-aware neuron selection method. By flexibly tuning the ratio between two types of neurons, the hybrid model demonstrates better adaptive ability in balancing different performance metrics, including accuracy, robustness, and efficiency on several typical benchmarks, and generally outperforms conventional single-paradigm recurrent neural networks and spiking neural networks. Furthermore, we evidence the great potential of the proposed network with a robotic task in varying environments. With our proof of concept, the proposed hybrid model provides a generic modeling route to process spatiotemporal data sources in the open world.",
"introduction": "Introduction It is not an exaggeration to say that the hungry demand for data learning is at the core of the ongoing age of intelligence. Being able to effectively process multi-scale complex spatiotemporal information is important for many real-world applications, such as handling video in self-driving cars, interpreting written text in mobile reading apps, and managing various types of sensor data in outdoor robots. However, it remains a significant challenge to process such complex data accurately, reliably, and efficiently, particularly in varying environments with different performance requirements. In mainstream machine learning, non-spiking recurrent neural networks (RNNs) serve as a pivotal model for processing spatiotemporal data. Unlike traditional feedforward architectures, RNNs incorporate recurrent connections into standard artificial neural network (ANN) models, enabling them to capture temporal patterns. While RNNs are extensively employed in diverse applications like speech recognition 1 , language modeling 2 , and state control 3 , they cannot learn long temporal dependency due to the gradient vanishing during backpropagation (BP) learning 4 . To address this issue, variants such as long short-term memory (LSTM) networks 5 have been developed. These advanced models, equipped with additional gated units, excel at capturing long-term temporal dependencies but come at the cost of increased computational complexity. Concurrently, there is growing interest in neuromorphic computing as an alternative pathway for developing intelligent models that are both computationally efficient and biologically inspired. Spiking neural networks (SNNs), regarded as the third generation of neural networks 6 , are the most famous family of neuromorphic models. The behaviors of each spiking neuron are described by the nonlinear dynamics of the membrane potential and the binary spiking mechanism for the communication between adjacent neurons 7 . Distinct from the extrinsic dynamics of RNNs induced by external recurrence, the dynamics in SNNs intrinsically exist within each neuron. The intra-neuron temporal dynamics and the spatial dataflow through the network make SNNs well-suited for processing spatiotemporal data. Up to now, SNNs have been extensively used for spike stream processing 8 , speech recognition 9 , ECG signal analysis 10 , state control 11 , and so forth. Machine-learning-oriented RNNs use intense matrix multiplications for computation and continuous activations for inter-neuron communication; on the contrary, neuromorphic-computing-oriented SNNs use sparse matrix accumulations for computation and binary spikes for inter-neuron communication. Compared to the continuous activation state space of RNNs, the spike states of SNNs usually evolve in a discrete space. With these distinctions, RNNs have been evidenced to achieve higher accuracy on conventional continuous data sources (e.g., speech signals and language texts) while SNNs are more suited for discrete data sources 12 such as the event stream collected by dynamic vision sensors (DVS) 13 . Owing to the natural filtering effect of the membrane potential leakage along with the spike firing and reset mechanisms of spiking neurons, SNNs have demonstrated strong robustness against variations in temporal resolution 12 and adversarial attack 14 . In addition, owing to the binary format of spikes and the sparsity of spiking activities, the computational cost of an SNN model can be much lower than its non-spiking counterpart under the same network structure 12 , 15 . Based on the above analyses, it can be seen that RNNs and SNNs present different performance results due to the disparate modeling paradigms. However, in practical scenarios, the type of data sources varies, e.g., continuous data or discrete data, and the performance requirement may also be highly diverse. For example, the high functional accuracy attracts the most attention from cloud users, while the low computational cost is more important for energy-restricted edge devices. Furthermore, for many core components in a system, how to guarantee high robustness against internal noise or external attack becomes the primary design consideration. Even though we can build a specific model to accomplish each task, it would be inefficient because researchers cannot directly apply the experiences accumulated in the modeling exploration when the task changes. To escape the one-task-one-model dilemma, a unified modeling framework to realize adaptive accuracy, robustness, and efficiency is highly expected for processing spatiotemporal data in various scenarios. Here we report a unified modeling framework that creates hybrid spatiotemporal neural networks (HSTNNs) by synergistically combining RNNs and SNNs for processing spatiotemporal data sources. To make the hybrid model learnable, our work builds on a unified learning methodology, backpropagation through time (BPTT) augmented with a surrogate function, which works for both RNNs and SNNs and thus opening the possibility for hybridization. Furthermore, we exploit a classical pruning method 16 , 17 to realize neuron selection from RNN and SNN populations and further develop a neuron-aware three-stage hybridization solution to create HSTNNs. It leverages the Hessian gradient information and enables automatic learning of a hybrid structure during the training phase. On several typical spatiotemporal dataset benchmarks, HSTNNs demonstrate better adaptive ability in balancing different performance metrics in terms of accuracy, robustness, and efficiency by tuning the configuration between two types of neurons, and usually outperform conventional single-paradigm RNNs and SNNs. With a robotic place recognition task, we evidence the great potential of HSTNNs in varying environments. Overall, the proposed HSTNNs provide an attractive way to adaptively process variable spatiotemporal data sources in the open world.",
"discussion": "Discussion We presented a generic hybridization approach that can maintain and integrate the complementary features of RNNs and SNNs, promising a unified effective way to process different types of spatiotemporal data. We observed that RNNs and SNNs have shown divergent performances across six distinct types of tasks. By leveraging their complementary features through our hybrid models, we demonstrated that the HSTNNs not only surpass single-paradigm models in comprehensive performance but also exhibit superior robustness against noise, frame loss, and adversarial attack. Furthermore, the adaptability of HSTNNs to diverse environmental conditions was evidenced in the robot place recognition task. The flexible hybrid paradigm yielded optimal recognition accuracy in a variety of lighting conditions, indicating its potential for handling the complexity and variability of real-world applications. Even though HSTNNs integrate two types of neurons, they can be deployed on emerging neuromorphic chips with the hybrid architecture for efficient execution. Interestingly, HSTNNs exhibit intriguing similarities to the coding strategies and integration of continuous and spiking activities observed in the human brain for information processing. The brain, renowned as a hybrid learning system, employs diverse types of neuron populations and a range of coding schemes to tackle complex spatiotemporal tasks. HSTNNs can actually achieve similar functionality by leveraging different coding strategies of SNNs and RNNs. We observed that the interaction between continuous and spike-based neural activities can alter the spiking activities of the spiking neuron population (see Supplementary Figs. 5 , 6 ). This amalgamation of various neuronal dynamics and coding strategies in HSTNNs embraces the diversity and richness of the brain’s own computational strategies, an aspect that has been underscored by contemporary neuroscientific research 35 , 36 . These parallels not only highlight the relevance and potential of HSTNNs for handling diverse real-world applications but also provide hints to understanding the design of more robust and adaptable systems for artificial intelligence. How to determine an optimal SNN ratio in HSTNNs and thereby achieve the balance between task performance and the computational cost is a crucial but open issue. This optimal SNN ratio is highly context-dependent and varies according to specific user requirements, as the weights assigned to accuracy and the cost differ across different environments. We show a heuristic method in Supplementary Information to address this challenge and provide an optimal SNN ratio automatically searched in specific tasks. By formulating the optimization problem and employing approximation optimization methods such as subgradient descent, we demonstrate in Supplementary Fig. 7 the feasibility of finding the optimal solution of the SNN ratio. It allows users to customize the model based on their specific needs and achieve an optimal trade-off between accuracy and cost. The practical implications of this optimization process are significant, as it facilitates real-world applications of HSTNNs in practical environments for effective and efficient processing of spatiotemporal data. Hybrid neural network models gain more and more interests from different fields due to the rapid development of neuroscience and the breakthrough of deep learning 23 , 31 . Several advantages of layer-wise hybridization have been demonstrated in references 23 , 24 . These layer-wise hybridization approaches focus on integrating non-recurrent ANN and SNN modules using layer-wise strategies, providing efficient solutions for practical applications such as optical flow estimation and high-speed tracking tasks. In contrast, the proposed neuron-wise hybridization approach in this work offers a finer-grained method of hybridization, enabling real-time interaction of the coding and computational features of different types of neurons. Moreover, the neuron selection strategy employed in the Selection stage represents a more general solution that encompasses layer-wise hybridization as a specific case. We demonstrate in Supplementary Fig. 4 that the proposed hybridization model can be applied to optical flow estimation, yielding comparable results to those of specially designed single-paradigm networks. The proposed HSTNN is a very initial effort to bridge dynamic models in machine learning and neuromorphic computing. As aforementioned, HSTNNs have presented great potential in task performance, model robustness, and computational cost, which provides a flexible trade-off to satisfy variable environments and user requirements under a unified modeling and learning framework. The models showcased in this work are relatively simple, offering considerable scope for further enhancement. For instance, HSTNNs could be enhanced with advanced transformer architectures and create deeper and larger models, enabling the processing of more complex spatiotemporal data. Furthermore, intelligent machines equipped with neuromorphic chips can incorporate HSTNNs to process spatiotemporal information collected by various sensors such as cameras, microphones, electroencephalogram electrodes, and so forth. We look forward to inspiring more investigations for taking complementary features and advantages of computer-science-oriented models and neuroscience-oriented models."
} | 3,264 |
25628615 | PMC4290590 | pmc | 9,719 | {
"abstract": "Plant tissues host a variety of fungi. One important group is the dark septate endophytes (DSEs) that colonize plant roots and form characteristic intracellular structures – melanized hyphae and microsclerotia. The DSE associations are common and frequently observed in various biomes and plant taxa. Reviews suggest that the proportion of plant species colonized by DSE equal that colonized by AM and microscopic studies show that the proportion of the root system colonized by fungi DSE can equal, or even exceed, the colonization by AM fungi. Despite the high frequency and suspected ecological importance, the effects of DSE colonization on plant growth and performance have remained unclear. Here, we draw from over a decade of experimentation with the obscure DSE symbiosis and synthesize across large bodies of published and unpublished data from Arabidopsis thaliana and Allium porrum model systems as well as from experiments that use native plants to better resolve the host responses to DSE colonization. The data indicate similar distribution of host responses in model and native plant studies, validating the use of model plants for tractable dissection of DSE symbioses. The available data also permit empirical testing of the environmental modulation of host responses to DSE colonization and refining the “ mutualism-parasitism-continuum ” paradigm for DSE symbioses. These data highlight the context dependency of the DSE symbioses: not only plant species but also ecotypes vary in their responses to populations of conspecific DSE fungi – environmental conditions further shift the host responses similar to those predicted based on the mutualism-parasitism-continuum paradigm. The model systems provide several established avenues of inquiry that permit more detailed molecular and functional dissection of fungal endophyte symbioses, identifying thus likely mechanisms that may underlie the observed host responses to endophyte colonization.",
"conclusion": "CONCLUSION Here we present arguments based on host growth responses and the potential for molecular dissection of an obscure endophyte symbiosis to better elucidate the ecological and molecular drivers underlying host responses to poorly known fungal symbionts. Our extensive experiments with model and non-model plants indicate a distribution of host responses to colonization and led to a proposal of a null model that permits testing hypotheses on host responses to a population of endophytic fungi as well as generating easily testable hypotheses on the shifts in these responses under altered environmental conditions. We further highlight examples of recent studies that have identified molecular cues and mechanisms underlying the host responses to fungal symbionts and vice-versa. It is the combination of the power of simple model systems and the ground-truthing those conclusions in relevant native plant systems that are likely to best elucidate the drivers and mechanisms of obscure and poorly understood symbioses. The findings of these studies can be coupled with deep interrogations of host and fungal transcriptomes to elucidate the mechanisms that underlay the observed host growth responses.",
"introduction": "INTRODUCTION Dark septate endophyte (DSE) fungi colonize plant roots and form characteristic structures – melanized hyphae and microsclerotia – and often have variable effects on plant growth. This inter- and intraspecific variability in host responses has been hypothesized to be central to plant community structuring by mycorrhizal fungi ( Wilson and Hartnett, 1998 ; Hartnett and Wilson, 1999 ; van der Heijden, 2002 ). Similarly, the variability in host responses to DSE fungi may promote selection mosaics proposed for ectomycorrhizal symbioses ( Piculell et al., 2008 ). An issue that has remained under continuous debate is whether the DSE symbiosis should be considered beneficial to the host plant or rather as a weak parasitism ( Jumpponen, 2001 ; Addy et al., 2005 ; Mandyam and Jumpponen, 2005 ; Alberton et al., 2010 ; Newsham, 2011 ; Mayerhofer et al., 2013 ). The general host responses to DSE fungi have remained difficult to discern, partly because of their wide variability, partly because of independent small studies that draw conclusions based on a limited number of fungal individuals. Here we aim to synthesize various bodies of data to better resolve the host responses to the colonization by these abundant fungi as well as to discern some abiotic controls that may lead to shifts in these observed host responses. Results from studies that use model and native plant systems provide unique empirical insights into the variability in host responses to DSE fungi drawn from populations of conspecific fungi. We argue that these data permit empirical evaluation of the “ mutualism-parasitism-continuum ” paradigm ( Johnson et al., 1997 ; Saikkonen et al., 1998 ). We conclude by describing a general neutral null-hypothesis of host responses to fungal symbionts applicable beyond the DSE symbiosis. The mutualism–parasitism paradigm has been used as a general framework to understand the mycorrhizal symbioses that have – similarly to DSE symbioses – been considered variable when observed in different hosts or compared under different abiotic conditions."
} | 1,316 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.