pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
27669057 | null | s2 | 2,822 | {
"abstract": "In this study, we exploit the excellent fouling resistance of polymer zwitterions and present electrospun nanofiber mats surface functionalized with poly(2-methacryloyloxyethyl phosphorylcholine) (polyMPC). This zwitterionic polymer coating maximizes the accessibility of the zwitterion to effectively limit biofouling on nanofiber membranes. Two facile, scalable methods yielded a coating on cellulose nanofibers: (i) a two-step sequential deposition featuring dopamine polymerization followed by the physioadsorption of polyMPC, and (ii) a one-step codeposition of polydopamine (PDA) with polyMPC. While the sequential and codeposited nanofiber mat assemblies have an equivalent average fiber diameter, hydrophilic contact angle, surface chemistry, and stability, the topography of nanofibers prepared by codeposition were smoother. Protein and microbial antifouling performance of the zwitterion modified nanofiber mats along with two controls, cellulose (unmodified) and PDA coated nanofiber mats were evaluated by dynamic protein fouling and prolonged bacterial exposure. Following 21 days of exposure to bovine serum albumin, the sequential nanofiber mats significantly resisted protein fouling, as indicated by their 95% flux recovery ratio in a water flux experiment, a 300% improvement over the cellulose nanofiber mats. When challenged with two model microbes Escherichia coli and Staphylococcus aureus for 24 h, both zwitterion modifications demonstrated superior fouling resistance by statistically reducing microbial attachment over the two controls. This study demonstrates that, by decorating the surfaces of chemically and mechanically robust cellulose nanofiber mats with polyMPC, we can generate high performance, free-standing nanofiber mats that hold potential in applications where antifouling materials are imperative, such as tissue engineering scaffolds and water purification technologies."
} | 478 |
26386069 | PMC4651090 | pmc | 2,823 | {
"abstract": "Lignin is a complex aromatic polymer found in plant cell walls that makes up 15 to 40% of plant biomass. The degradation of lignin substructures by bacteria is of emerging interest because it could provide renewable alternative feedstocks and intermediates for chemical manufacturing industries. We have isolated a bacterium, strain SG61-1L, that rapidly degrades all of the stereoisomers of one lignin substructure, guaiacylglycerol-β-guaiacyl ether (GGE), which contains a key β- O -4 linkage found in most intermonomer linkages in lignin. In an effort to understand the rapid degradation of GGE by this bacterium, we heterologously expressed and kinetically characterized a suite of dehydrogenase candidates for the first known step of GGE degradation. We identified a clade of active GGE dehydrogenases and also several other dehydrogenases outside this clade that were all able to oxidize GGE. Several candidates exhibited stereoselectivity toward the GGE stereoisomers, while others had higher levels of catalytic performance than previously described GGE dehydrogenases for all four stereoisomers, indicating a variety of potential applications for these enzymes in the manufacture of lignin-derived commodities.",
"introduction": "INTRODUCTION The lignin polymer, representing 15 to 40% of plant-derived biomass, is a potential renewable alternative to petrochemical feedstocks for chemical manufacturing industries, particularly with regard to aromatic compounds ( 1 ). One of the major hurdles for the utilization of lignin in this way is its recalcitrance; the polymeric structure of lignin coupled with the nature of its chemical bonds renders it highly resistant to chemical or biological degradation. Chemical degradation of lignin has been utilized by some industries, but it usually involves nonselective destruction of lignin via burning or treatment under highly alkaline conditions, which renders it less useful or of lower value to many downstream applications ( 2 ). The biological depolymerization of lignin should be a more selective and energy-efficient process and therefore potentially a cost-effective and environmentally sustainable alternative to the chemical processes currently employed ( 2 ). However, to date, the identification of suitable biocatalysts involved in lignin depolymerization has proven difficult. Lignin has a polyaromatic structure with more than five different types of intermonomer linkages ( 3 ); the β- O -4 linkage (also known as the β-aryl ether linkage) represents 45 to 70% of these linkages ( 4 , 5 ). The first step in the biological degradation of the lignin polymer occurs nonspecifically through the action of laccases and extracellular peroxidase enzymes (manganese peroxidase, versatile peroxidase, and lignin peroxidase) via radical ion mechanisms ( 5 – 7 ). While fungi are thought to be the main contributors to lignin degradation, recent reports on bacteria and their enzymes suggest that certain bacterial strains may also play a role in lignin polymer or kraft lignin degradation ( 8 – 23 ). While the early steps involved in the biological degradation of the lignin polymer are nonspecific, a few bacteria have shown potential in directly and specifically degrading smaller units of the lignin polymer into industrially useful chemicals ( 4 , 24 – 31 ). In the 1980s, bacterial dehydrogenases from Pseudomonas spp. were isolated and found to catalyze C-alpha-alcohol oxidation of β-aryl ether- or diarylpropane-linked lignin dimers ( 28 , 32 ). In this case, the proteins responsible were not identified and the degradation pathways were not further explored. Subsequently, however, dehydrogenases catalyzing such reactions from another bacterium, Sphingomonas \n paucimobilis SYK-6, have been characterized. SYK-6 has been reported to degrade several different types of model lignin dimers ( 5 ), including a molecule with a β- O -4 linkage known as guaiacylglycerol-β-guaiacyl ether (GGE). The proposed pathway for the degradation of GGE in this organism begins with the oxidation of GGE via one of several C-alpha-dehydrogenases (LigD, LigO, LigL, or LigN) ( 33 ), followed by ether bond cleavage via one of several glutathione S -transferases (GSTs; LigE, LigF, or LigP) ( 34 , 35 ), and then glutathione removal via one of several GSTs (LigG and probably others that have yet to be identified) ( 34 , 36 ) ( Fig. 1A ). These reactions generate the end products guaiacol and β-hydroxypropiovanillone (HPV), the latter of which is eventually fed into the protocatechuate 4,5-cleavage pathway ( 4 , 5 ). In SYK-6, the biological conversion of GGE to HPV and guaiacol is performed by a series of stereoselective enzymes ( 33 , 36 , 37 ); there are two stereocenters in the GGE molecule and therefore four possible stereoisomers ( Fig. 2A ). Expression of the C-alpha-dehydrogenases LigD, LigO, LigL, and LigN in Escherichia coli and gene disruption experiments with SYK-6 showed that these enzymes are responsible for the first step in the pathway, collectively oxidizing all of the GGE stereoisomers but individually exhibiting preference for one or two of them ( 33 ). FIG 1 Bacterial growth and degradation of GGE. (A) Pathway for the transformation of GGE to HPV in SYK-6 and SG61-1L. (B) Metabolite formation and disappearance over the course of the GGE growth cell experiment in SG61-1L (red) and SYK-6 (blue). Each graph depicts a different metabolite (labeled in the upper right corner). (C) GGE degradation (left y axis) and bacterial growth (right y axis) over time for SYK-6 (left) and SG61-1L (right) during growth cell experiments with GGE as the sole carbon source compared with control experiments with no carbon source. The arrows indicate the time at which cultures of SG61-1L were pelleted and resuspended in fresh MM supplemented with GGE as described in Materials and Methods. FIG 2 GGE stereoisomer preferences during growth experiments. (A) The four GGE stereoisomers grouped according to enantiomer pairs (erythro and threo). (B) GGE stereoisomer degradation measured from growth cell experiments run on a chiral column (color coded by stereoisomer). While bacteria are able to degrade intermonomer linkages of lignin substructures, their reported degradation rates are either not known or quite low ( 28 , 31 , 38 , 39 ), and optimization of these transformation pathways (and the enzymes involved) is paramount for their industrial utilization. Here we report the isolation of a bacterium (SG61-1L) that is able to degrade a GGE model lignin dimer at a significantly higher rate than SYK-6. These results led to our identification of several enzyme candidates with robust activities for industrial transformation of a key intermediate in the biological breakdown of the lignin polymer.",
"discussion": "DISCUSSION Here we report the isolation of SG61-1L, a bacterium that is able to degrade the GGE model lignin dimer at a much higher rate than the previously characterized GGE-degrading bacterium SYK-6. In order to understand the molecular basis for its activity and investigate the biotechnological potential of the enzymes involved, we have proceeded to characterize a suite of its dehydrogenases that are prime candidates for the first step in β-aryl ether degradation, the NAD(P) + -dependent oxidation of GGE to MPHPV. Our analyses suggest that the four SYK-6 enzymes are kinetically inferior to two of the SG61-1L enzymes, 724 and 2550, as catalysts for all four stereoisomers. This result may, in part, explain the higher rate of GGE transformation observed in SG61-1L. C-alpha-dehydrogenases that were previously purified from Pseudomonas spp. have K m values for GGE (represented by a mixture of stereoisomers) as low as 11 to 12 μM ( k cat values were not reported) ( 28 , 32 ), but the genes encoding these proteins were not identified; interestingly, the monomeric molar mass of the C-alpha-dehydrogenase from Pseudomonas sp. strain GU5 was estimated to be approximately 52,000 kDa ( 28 ), which is quite different from the average monomeric molar mass range of all of the characterized proteins in this work (approximately 30,000 kDa), suggesting that perhaps another, as-yet-unidentified, type of C-alpha-dehydrogenase in this organism can oxidize GGE. Most of the enzymes with GGE dehydrogenase activity characterized herein fall into a single clade, and it is likely that this clade in general is a good predictor of GGE dehydrogenase activity. It is also apparent that the sequences of GGE dehydrogenases in this clade have diverged significantly from other functionally annotated dehydrogenases ( Fig. 4 ). Three additional uncharacterized SYK-6 genes are also present in the GGE dehydrogenase clade, but it seems unlikely that they play a larger role in GGE oxidation than those already characterized. This notion is based on a previous experiment with a mutant SYK-6 bacterium containing knockouts of three GGE dehydrogenase genes (those for LigD, LigN, and LigL) that result in nearly complete loss of the ability to oxidize GGE compared to that of the wild type ( 33 ). Under the selection pressures imposed on bacteria in an environment that is rich in lignin-derived substructures, it is possible that dehydrogenases outside the GGE dehydrogenase clade and native to other metabolic pathways (such as 3344 and 474) have evolved some ability to contribute to GGE oxidation. For example, 3344 and 3730 localize to a larger clade of the phylogenetic tree that contains a functionally verified levodione reductase from Corynebacterium \n aquaticum M-13; 3344 and 3730 show 38 and 43% amino acid identity to this levodione reductase, respectively, indicating that 3344 may have evolved from a levodione reductase (perhaps 3730) encoded in the SG61-1L genome. This idea that biological degradation occurs through multiple unrelated dehydrogenases native to different pathways is reminiscent of a recent report on the oxidation of dehydrodiconiferyl aldehyde in SYK-6, which is proposed to occur through the action of several aldehyde dehydrogenases ( 49 ). SYK-6 and SG61-1L were both isolated from pulp and paper mill waste sites ( 50 ), but unlike SYK-6, SG61-1L was specifically selected through rounds of enrichment culturing for the ability to utilize GGE as a sole carbon source. This difference is perhaps reflected in the poorer kinetic parameters of the SYK-6 GGE dehydrogenases than the two highest-performing SG61-1L GGE dehydrogenases. There is no obvious correlation between phylogeny and stereoselectivity within the GGE dehydrogenase clade. For example, 1498 from SG61-1L and LigN from SYK-6 show the highest sequence identity (67%) of any pair in the GGE dehydrogenase clade, yet 1498 was able to oxidize only the (αS,βR)- and (αS,βS)-GGE stereoisomers while LigN could oxidize all four. Similarly, 2550 from SG61-1L and LigL from SYK-6 also cluster together in this clade ( Fig. 4 ) and were both able to oxidize all four GGE stereoisomers, but each enzyme displayed a different pattern of stereospecificity ( Table 3 ). We suspect that empirical structural data will be necessary to identify the sequence motifs guiding stereoselectivity preferences among these GGE dehydrogenases. To the best of our knowledge, SG61-1L is one of the fastest known bacterial degraders of a lignin substructure; previously reported rates of bacterial degradation for model lignin dimers are quite low ( 38 , 39 ), albeit various conditions were used and the results may therefore not be directly comparable. We report here that SG61-1L can degrade GGE much faster than SYK-6 can under the same experimental conditions and that in SYK-6, transformation of MPHPV to α-glutathionyl-β-HPV (GS-HPV) appears to be a rate-limiting step. Following GGE oxidation to MPHPV, the enzymes that perform the next two reactions (the β-etherase reaction to generate GS-HPV, followed by glutathione removal to generate achiral HPV) are GSTs. They have been identified as LigF-LigE-LigP and LigG, respectively, in SYK-6, and homologs have been functionally verified in closely related sphingomonads ( 34 – 37 ), including N . aromaticivorans DSM12444, which also contains genes present in the GGE dehydrogenase clade. There are several uncharacterized GSTs in the SG61-1L genome, two of which have the highest sequence identity to LigP (63 to 79%) and five of which have the highest sequence identity to LigF (34 to 59%). Curiously, none of the SG61-1L-encoded GSTs have high sequence identity to LigG (<17%). Thus, there are several candidates in SG61-1L for the first GST step, but the second apparently involves an enzyme not closely related to LigG. Given that the LigG reaction in SYK-6 is specific to only one of the two GS-HPV stereoisomers ( 36 ), it has also been suggested that there is at least one other, as-yet-uncharacterized, GST with activity for the other GS-HPV stereoisomer in SYK-6. Future work will involve the characterization of these later steps in the pathway in order to articulate mechanisms by which SG61-1L proceeds through GGE degradation at a relatively high rate. There is an emerging interest in the development of lignin-degrading enzymes for industrial purposes, especially for integration into processes generating replacements for petroleum-based commodities. Selective depolymerization of lignin or lignin components could generate high-value, complex aromatics, including catechols, resorcinols, keto acids, and polyhydroxy aromatics ( 2 , 51 ). Vanillin, for example, has been produced from lignocelluloses in a Rhodococcus jostii RHA-1 mutant ( 52 ). Vanillin also has potential to be produced on a large scale via the biodegradation of lignin substructures (such as β-aryl ether-linked lignin dimers) by engineered versions of bacteria such as SG61-1L and SYK-6 or by free enzymes driving the relevant reactions derived from them ( 5 , 53 ). Thus far, we have characterized a group of dehydrogenases from SG61-1L that are involved in the first step of a lignin degradation pathway and could be used to produce a high-value chemical such as vanillin. The priorities of a high k cat versus a low K m and broad versus narrow stereoselectivity for a given enzyme will vary, depending on the process, but the suite of dehydrogenases from SG61-1L that we have characterized provides a wide range of options across these criteria. Any further protein engineering required to make promising enzymes more fit for a purpose can also now be guided by our findings on the phylogenetic distribution of GGE dehydrogenase activity. SG61-1L is also a very promising source of catalysts for downstream steps in the GGE degradation pathway, particularly the β-etherase step, which results in cleavage of the highly stable aryl-aryl ether bond of MPHPV to produce GS-HPV."
} | 3,698 |
24031729 | PMC3768782 | pmc | 2,825 | {
"abstract": "Polyhydroxyalkanoates (PHA) are biodegradable and biocompatible green thermoplastics, synthesized by wide variety of bacteria as an intracellular carbon and energy storage intermediate. They are used as an alternative to nonrenewable petroleum derived plastics. The current interest in these biopolyesters is stimulated by the search for cost-effective capitalized production. This paper attempts to achieve maximized production rate from recombinant system using inexpensive substrate. Molasses from agro-industrial waste was used to produce PHA from recombinant E.coli in batch culture. PHA yield in molasses (3.06g/L ± 0.05‒75.5%) was higher than that of sucrose (2.5g/L ± 0.05 - 65.1%). Properties of the polymer produced from molasses and sucrose were analyzed by DSC, TGA, DTA, GC/MS, TLC and optical rotation studies. The findings suggested that molasses enhanced PHA production in recombinant E.coli .",
"introduction": "INTRODUCTION Polyhydroxyalkanoates (PHAs) are optically active polyesters of natural origin ( 15 ). It is synthesized in microorganisms as intracellular carbon reserve material during the excess of carbon under nitrogen limiting conditions. It can be utilized by micro-organism as a reserved material at the time of its need ( 13 ). As it has material properties (Molecular weight, Melting temperature, Glass transition temperature) similar to synthetic polymers ( 24 ), it is trusted to triumph over the problems and harmful effects of plastic wastes ( 3 , 16 , 22 ). Owing to these features, PHAs have drawn much attention for numerous industrial applications. A major limitation to achieve marketable production of PHAs is their higher price than synthetic fossil fuel based plastics. The high production cost is the most important barrier hindering PHA to compete the market with conventional synthetic polymers. For instance, Zeneca Bio Products (Billingham, UK) produced approximately 1,000 tons per year of PHB/V copolymer at ca. US $ 16/kg. The price of conventional petrochemical plastics is less than US $ 1/kg ( 11 , 23 ). To achieve successful commercialization of PHA, economic production system must be sort out. As the cost of raw material (substrate, bacterial strain) is one of the major factors influencing the economy of production ( 18 , 20 ), the present study is concentrated on the cost effective production of PHA by combining the use of inexpensive substrate (molasses) for the growth of the potential strain. Information on Physio - chemical material property is obligatory for the possible usage of polymer. For example, PHA that is semi-crystalline, with low melting point (T m ), high elongation to break can be used for biomedical applications. Based on these above said concerns, the present study is focused on to achieve lower production cost using molasses as substrate for recombinant strain and characterization of material properties of PHA to ensure quality product for medical application.",
"discussion": "RESULTS AND DISCUSSION Biosynthesis of polymer The influence of substrate (molasses and sucrose) on the growth and PHA production by recombinant E.coli harboring phaC1 of Pseudomonas .sp.LDC-5 was investigated. The presence of bright spherical PHA granules in the cells were clearly demonstrated by phase contrast light microscope (OLYMPUS-DP12-CX41) ( Figure 1 ). The PHA yield was increased when molasses (3.06g/L± 0.05- 75.5%) was used as carbon source than with sucrose (2.5g/L ± 0.05 - 65.1%). Figure 2 illustrates the kinetics of the production. Figure 1 Phase contrast microscopic view of recombinant E.coli harboring partial phaC1 gene a. Recombinant E.coli grown for 12hin Sucrose b. Recombinant E.coli grown for 72hin Sucrose c. Recombinant E.coli grown for 12hin Molasses d. Recombinant E.coli grown for 72hin Molasses Figure 2 Growth kinetics of recombinant E.coli a. PHA production per Litre by recombinant E.coli grown using Sucrose b. PHA production per Litre by recombinant E.coli using Molasses CDW – Cell Dry Weight ; PHA – Polyhydroxyalkanoate; RCW – Residual Cell Weight The enhanced PHA production in molasses could be attributed by the composition of the substrate. The C:N ratio has led to unbalanced nutrient condition, which in turn had led to inhibition of TCA cycle enzymes such as citrate synthase and isocitrate dehydrogenase and consequently have slowed down the TCA cycle. As a result, the acetyl co-A routed to PHA biosynthesis ( 2 ). Figure 3 depicts that the production cost using molasses is relatively lower compared to sucrose. Figure 3 Economic analysis: Production of one gram of PHA TLC identification of PHA PHA components were identified with the characteristic appearance of yellowish green colour spot in the TLC plates. The retention factor (R f ) was 0.58 for both the samples in accordance with reports of Paul et al (2004). DSC analysis of PHA Melting temperature (T m ) is determined from endothermal peaks of the DSC thermograms. PHA recovered from the strain cultured with molasses had T m of 129.59 ° C ( Figure 4a ) is indeed lower than that with sucrose (T m 152 ° C) ( Figure 4b ). T m is influenced by the length of side chain and functional groups present ( 5 ). Lower T m indicates that the side chains are longer. Molasses had incorporated longer side chains in comparison with sucrose. The range of T m reported in this study suits fabrication of product as per ASTM: D 882–91 test ( 21 ). Figure 4 Thermal analysis: Melting temperature determination from DSC thermogram a. PHA produced from Recombinant E.coli by fermentation of molasses b. PHA produced from Recombinant E.coli by fermentation of sucrose TGA & DTA analysis of PHA Thermal stability of PHA is important for their melt processing. The temperature at 5 % weight loss termed T d(5%) was employed to evaluate polymer thermal stability. PHA samples recovered from recombinant strain cultured with molasses had T d(5%) of 240 o C ( Figure 5a ). The PHA sample from the strain cultured with sucrose had T d(5%) of 310 o C ( Figure 5b ). T d(5%) was much higher than T m reflecting thermal permanence. This validates the improved avenues for polymer processing ( 21 ). Figure 5 TGA and DTA thermogram for determination of Decomposition and curing Temperature a. PHA produced from Recombinant E.coli by fermentation of molasses Figure 5 TGA and DTA thermogram for determination of Decomposition and curing Temperature b. PHA produced from Recombinant E.coli by fermentation of sucrose DTA helps to determine the heat of reaction of decomposition process. If cross linking reactions occurred during degradation of PHA, an exothermic peak would be detected in DTA thermogram. The temperature at which cross linking occurs is the curing temperature. The curing temperature for cost effectively produced PHA is around 431 o C. This stands to be a valid property as major barrier for commercial application of PHA, was their thermal instability due to lack of ability to crosslink ( 10 ). Examination of Optical rotation The optical rotation of methyl ester of PHA from molasses as well as sucrose grown cells was negative (-1) which indicates that R form is enantiomerically excess in accordance with earlier reports ( 6 ). GC-MS analysis of PHA GC-MS analysis helps in elucidating the structure of components. The key compounds of concern were identified based on their retention peak. PHA from recombinant E.coli cultured with molasses significantly contained C 9 H 20 O 3 (Propane – 1, 1 triethoxy-) and C 12 H 26 (Dodecane) ( Figure 6 ). These compounds signify that the monomer chains were of biodegradable polyester family ( 11 ). Characteristic fragment at m/z 103 suggested the presence of hydroxyl group of carbon 3 formed by cleavage of alpha to the hydroxylated carbon ( 12 ). This had made apparent that financial prudence effort had no way compromised the eminence of artifact. Figure 6 GC/MS analysis of PHA produced from Recombinant E.coli by fermentation of molasses PHA recovered from sucrose grown cultures had significant compounds like C 12 H 24 O 3 (Dodecanoic acid 3-hydroxyl-) and C 14 H 30 (Tetradecane) ( Figure 7 ).These are biodegradable by virtue of hydrolysable ester bonds ( 4 ). The components identified had served as evident for the verity that nutrient and culture conditions modify the side chain length. Figure 7 GC/MS analysis of PHA produced from Recombinant E.coli by fermentation of sucrose PHA production by recombinant E.coli harboring phaC1 gene was studied using molasses as nutrient substrate, indeed to lower the production cost. Its effect on properties was tracked and it’s now evident that molasses as a substrate did not adversely affect material properties and in fact had led to betterment. This study will lead to economic PHA production appropriate for large scale intended for commercialization."
} | 2,217 |
39624404 | PMC11609164 | pmc | 2,826 | {
"abstract": "In this manuscript, we investigate the memristor-based implementation of neuronal ion channels in a mathematical model and an experimental circuit for a neuronal oscillator. We used a FitzHugh-Nagumo equation system describing neuronal excitability. Non-linearities introduced by the voltage-gated ion channels were modeled using memristive devices. We implemented three basic neuronal excitability modes including the excitable mode corresponding to a single spike generation, self-oscillation stable limit cycle mode with periodic spike trains and bistability between a fixed point and a limit cycle. We also found the spike-burst activity of mathematical and experimental models under certain system parameters. Modeling synaptic transmission, we simulated postsynaptic response triggered by periodic pulse stimulation. We found that due to the charge accumulation effect in the memristive device, the electronic synapse implemented a qualitatively bio-plausible synapse with a potentiation effect with increasing amplitude of the response triggered by a spike sequence.",
"conclusion": "5 Conclusion We proposed a mathematical and experimental model that simulates neuronal excitability and synaptic potentiation. The model was implemented by counter-parallel connection of memristive devices with different electrode compositions based on Au / Ta / ZrO 2 ( Y 2 O 3 )/ Pt / Ti / glass and Au / Ru / ZrO 2 ( Y 2 O 3 )/ Pt / Ti / glass of the mFHN generator circuit. The memristive devices had reliable characteristics possessing stable and gradual bipolar type resistive switching. To describe the model, we proposed a three-dimensional system of non-linear equations that describes three dynamic modes corresponding to excitable, self-oscillatory, and bistable neuronal dynamics. We have discovered that the system can demonstrate burst activity at a certain value of the ε parameter and the inbound external rectangular signal. A hardware implementation of a postsynaptic neuron model based on two counter-parallel memristive devices was developed. We verified computational modeling results with physical prototype memristive neurons in various modes: oscillations and multiple spike activity. We found that in addition, the simulation of neuronal excitability, using two counter-parallel memristive devices, can also model the effect of synaptic potentiation. The effect synaptic potentiation is manifested in the increase of the current amplitude in response to a series of pulse stimulations with constant amplitude. In other words, the memristive circuit proposed in the study can reproduce synaptic plasticity which is the basic feature of connectivity in the brain's neuronal circuit responsible for the generation of higher cognitive functions.",
"introduction": "1 Introduction Neuromorphic systems reproducing neuronal circuits and functions of the brain have attracted growing attention of researchers from different fields of science and technology. Spiking neuronal networks employ memristive devices to implement neuronal and synaptic components. Engineering of spiking neuronal networks and corresponding processing functions nowadays look as one of the most intriguing directions in neuromorphic system development (Makarov et al., 2022 ; Dalgaty et al., 2024 ). Memristors are electronic components based on the resistive switching (RS) effect (Chua, 2019 ), possessing at least two stable states that differ in their own resistance—a state with high (HRS) and low (LRS) electrical resistance. Devices utilizing this effect retain the acquired information in the form of resistance, the evolution of which is characterized by the restructuring of the atomic structure in thin insulating (dielectric) layers of nanometer-scale thickness under the stimulus of an electric field. Employing memristors as elements in electronic circuits have opened wide possibilities of designing non-linear oscillators with a variety of complex dynamical modes including chaos and multistability (Minati et al., 2020 ; Gokyildirim et al., 2022 ; Wang et al., 2022 ; Boudjerida et al., 2023 ; Chen et al., 2019 ; Spagnolo et al., 2022 ; Corinto and Forti, 2017 ). Possibility of complex non-linear dynamics and unique (biomimmetic) electro-physical properties with energetic efficiency has made memristors to be the most promising candidates for constructing biologically plausible neuron models and neuromorphic computations (John et al., 2022 ; Indiveri et al., 2013 ; Pisarev et al., 2020 ; Shchanikov et al., 2021 ). Specifically, memristors were used to simulate the dynamics of voltage-gated ion channels of neuron membrane. Implementation of rather simple potassium channels was discussed in Najem et al. ( 2018 ), Thomas ( 2013 ), Yi et al. ( 2018 ), and Gonzalez-Raya et al. ( 2020 ). More complex Hodgkin-Huxley neuron model employing both sodium and potassium channels was realized in Lv et al. ( 2016 ), Jeong et al. ( 2016 ), Sah et al. ( 2014 ), Gonzalez-Raya et al. ( 2019 ), and Hu and Liu ( 2019 ). Our recent study was reported on memristor-based implementation of FitzHugh—Nagumo (FHN) spiking neuron model that can reproduce both excitable and oscillatory neuronal dynamics (Kipelkin et al., 2023 ). Synaptic plasticity is one of the fundamental properties of living neuronal systems responsible for basic cognitive functions of the brain such as learning and memory (Sun et al., 2024 ; Kotaleski and Blackwell, 2010 ). Signals are transmitted between neurons via special biological devices called synapses. The strength of the synaptic connection is defined by complex chemical molecular transformations that occurs in both presynaptic (transmitter) neuron and postsynaptic (receiver) neuron (Lynch, 2004 ). Specifically, long-term changes in the connection strength are localized mostly in the postsynaptic neuron. When a spike is transmitted, the postsynaptic membrane, similarly to basic neuronal excitability, opens its ion channels and ions, particularly Na + and K + , cross the membrane generating postsynaptic potentials (Nadler, 2012 ). Interestingly, that is, the synapse is transmitting a series of; each consequent spike may induce voltage responses of variable amplitudes. If each next spike generates a stronger response of increasing amplitude, then the synaptic \n potentiation takes place (Vyazovskiy et al., 2008 ). So, the synaptic connection amplifies its strength. In the opposite case, there is a synaptic depression. The type of synaptic plasticity is defined by the neuron type and also by the dynamical characteristics of the transmitted signals, for example, frequency of spike and/or relative phase of the spike occurrences. For neuron models describing ion channels, information processing and encoding can be described by the dynamics of the action potential. However, due to the stochasticity and sensitivity of memristors, which affect dynamic processes, more in-depth mathematical and experimental studies are needed to control the behavior of devices and prevent undesirable effects through operational control. Based on the above sources and existing problems, we investigated a 3D model of neuronal excitability, implemented by two memristor-based FHN generator circuit. We employed memristive devices with different electrode compositions Au / Ta / ZrO 2 ( Y 2 O 3 )/ Pt / Ti / glass and Au / Ru / ZrO 2 ( Y 2 O 3 )/ Pt / Ti / glass to mimic ion channels. By extending the equations and modifying the scheme proposed earlier (Kipelkin et al., 2023 ), we presented mathematical and experimental investigation of the model. Experimental data obtained as a result of hardware measurements qualitatively confirm the computational modeling. We also analyzed how the electronic neuron responded on a spike sequence. Similarly to a postsynaptic neuron in real neuronal networks, our memristor-based device demonstrated synaptic potentiation when each next spike induced the response of growing amplitude.",
"discussion": "4 Discussion To highlight the potential of our study, we present a comparative table that evaluates our proposed model against (Huang et al., 2021 ) (The model 1) and (Nabil et al., 2022 ) (The model 2), both qualitatively and quantitatively. We compared the following model parameters ( Table 1 ): - the structural composition of the insulator used in the memristive devices; - mathematical model of a neuron describing the ionic dynamics of a system; - the number of adjustable parameters in the modeling process; - consideration is given to the internal dynamics of the memristive devices, including ion and electron transport processes within the filament; - the mathematical model is experimentally validated using electrical circuits and physical memristive devices; - availability of statistical data on the current-voltage characteristics of physical memristive devices with different compliance current values; - the number of memristive devices used in the model or circuit; - the operating voltage range of the memristive device; Table 1 Qualitative and quantitative comparison of the existed neuronal ionic dynamics models. \n N \n \n Items \n \n Our model \n \n The model 1 \n \n The model 2 \n \n 1 \n Insulator structure \n ZrO \n 2 \n \n NbO \n \n x \n \n \n VO \n 2 \n \n 2 \n Neuron model mFHN SRM neuron LIF \n 3 \n Number of model parameters 16 8 11 \n 4 \n Internal dynamics of the memristive devices Yes No No \n 5 \n Experimental validation Yes No No \n 6 \n Statistical data Yes No No \n 7 \n Number of memristive devices 2 1 1 \n 8 \n Operating voltage range, V 4 - 7 4 2 As shown in Table 1 , the proposed model offers several advantages. It was more flexible possessing 16 independent parameters that offered better opportunity for fine tuning the desired dynamical mode. Next, it permitted an experimental validation using physical memristive devices supported by statistical data. We also employed two memristive devices imitating different channels making the model more attractive in terms of its biological plausibility. Technically, our model demonstrated wider voltage range for the memristive devices what was also important for the model tuning. Finally, we believe that our model could be an appropriate candidate in the development of large-scale non-linear oscillators using memristive devices. Further exploring the model in theoretical part, we will focus on research into chimera states and analysis of potential chaotic oscillation modes. In technical development, particular attention will be given to integrated implementation. This is expected to lead to reduced energy consumption and increased stability, primarily due to the miniaturization of microscale memristors and other circuit components."
} | 2,664 |
22950603 | PMC3532602 | pmc | 2,827 | {
"abstract": "Summary To find links between the biotic characteristics and abiotic process parameters in anaerobic digestion systems, the microbial communities of nine full‐scale biogas plants in South Tyrol (Italy) and Vorarlberg (Austria) were investigated using molecular techniques and the physical and chemical properties were monitored. DNA from sludge samples was subjected to microarray hybridization with the ANAEROCHIP microarray and results indicated that sludge samples grouped into two main clusters, dominated either by Methanosarcina or by Methanosaeta , both aceticlastic methanogens. Hydrogenotrophic methanogens were hardly detected or if detected, gave low hybridization signals. Results obtained using denaturing gradient gel electrophoresis (DGGE) supported the findings of microarray hybridization. Real‐time PCR targeting Methanosarcina and Methanosaeta was conducted to provide quantitative data on the dominating methanogens. Correlation analysis to determine any links between the microbial communities found by microarray analysis, and the physicochemical parameters investigated was conducted. It was shown that the sludge samples dominated by the genus Methanosarcina were positively correlated with higher concentrations of acetate, whereas sludge samples dominated by representatives of the genus Methanosaeta had lower acetate concentrations. No other correlations between biotic characteristics and abiotic parameters were found. Methanogenic communities in each reactor were highly stable and resilient over the whole year.",
"conclusion": "Conclusions Although the control of anaerobic digestion processes has received much attention in the past few years, science is still far from understanding all of the process interactions occurring between biotic and abiotic parameters in this complex system. In this study, acetotrophic archaea dominated all reactors and revealed significant correlations with acetate levels. Archaeal communities remained highly stable over a whole year of operation, showing that seasonal changes in input materials had little effect on the apparently resilient microbiota in all biogas plants. This study has focused only on the archaea, but considering the close and efficient syntrophisms known to exist between bacteria and archaea, further research will target both phylogenetic groups.",
"introduction": "Introduction In the face of climate change and global warming, the rapid depletion of fossil fuel reserves and the accumulation of waste in our ‘throw‐away’ society, the production of clean bioenergy has undergone a rebirth in recent years. The production of biogas from the anaerobic digestion of organic wastes is one example. In fact, biogas technology is considered to be an excellent tool to avoid negative influences on the environment and climate (Insam and Wett, 2008 ). A number of anaerobic digesters have been developed and installed in Austria during the last 20 years and the number is continuously rising. In 2002, there were 97 plants in Austria, while by the end of 2010, there were 360 plants in operation (E‐CONTROL, 2010 ). In South Tyrol (Italy), 30 plants are in operation and several others are in the process of planning or under construction (INBIMO, 2011 ) due to the incentives currently being offered for the operation of anaerobic digestion plants. The anaerobic digestion process itself requires specific environmental conditions and is dependent on the microorganisms involved to cooperate in a close and efficient syntrophism (Schink, 1997 ). Digestion occurs in four major stages (hydrolysis, acidogenesis, acetogenesis and methanogenesis) and complex polymers are degraded in a stepwise manner to yield CO 2 and CH 4 . In the first step, hydrolysing and fermenting microbes degrade the organic macromolecules, such as proteins, carbohydrates and fats, into amino acids, sugars and fatty acids. Acidogenic bacteria convert the sugars, amino acids and fatty acids to organic acids, alcohols and ketones, acetate, CO 2 and H 2 . Acetogenic bacteria convert the fatty acids and alcohols into acetate, H 2 and CO 2 , products used by methanogenic archaea to form methane (Ahring, 2003 ). Methanogens thus hold the key position in the anaerobic digestion process. Since anaerobic digestion is a very complex process involving biotic and abiotic factors, improvements in operation can be difficult to achieve. There are however different approaches to increasing the biogas potential of a particular reactor, such as optimizing the reactor configuration, increasing the digestibility of the input material, optimizing process control and stability and improving the microbial processes and their efficiency (Ahring, 2003 ). Improvements in anaerobic biotechnology and an increased/stable biogas production require a better understanding of reactor functioning and the microbial communities involved. The aim of this study was to reveal and link anaerobic digester process functioning and environmental parameters with the microbial communities present.",
"discussion": "Discussion Anaerobic digestion is a renewable energy technology that is gaining increasing interest and importance worldwide. In this study, nine different anaerobic digestion plants were monitored and investigated to determine any links between the abiotic and biotic factors involved. All nine reactors monitored were found to be operating under stable conditions with good biogas yields, even when some values exceeded the critical value for good operation conditions (BMVIT, 2007 ). Correlation analysis indicated links between pH, conductivity, NH 4 ‐N, NH 3 ‐N and H 2 . This was not unexpected, since the pH approximates the negative logarithm (base 10) of the molar concentration of dissolved hydronium ions and thus regulates the balance of NH 4 ‐N and NH 3 ‐N (Fachagentur für Nachwachsende Rohstoffe, 2006 ). The thermophilic reactor F was found to have the highest acetate level of all reactors investigated. This correlation between acetate and reactor temperature has also been reported in previous studies (Van Lier et al ., 1996 ; Ahring et al ., 2001 ). The archaeal community structure was investigated using a variety of molecular‐biological approaches. Initially, 16S rRNA gene‐based microarrays and DGGE were used to obtain an overview of the archaeal community composition present in the different sludge samples. The dominating genera were then analysed by RT‐PCR to obtain quantitative data regarding the genera in the reactor samples. Microarray analysis with the ANAEROCHIP microarray as well as fingerprinting patterns as revealed by DGGE showed the microbial communities of the sludge samples to cluster into two main groups, dominated either by Methanosarcina or by Methanosaeta (Fig. 2 ). Both genera are comprised of acetoclastic methanogens, which have been reported to be responsible for approximately 70% of the methane produced in biogas reactors (Jetten et al ., 1992 ; Ahring et al ., 1995 ). Microarray data can however only be interpreted semi‐quantitatively, because of the variation in binding efficiency of different oligonucleotide probes with the various regions on the 16S rRNA gene. RT‐PCR is thus a perfect tool to counteract this disadvantage and results obtained in this study supported those obtained by microarray analysis (Fig. 4 ), whereby Methanosaeta dominated in the reactors A–E while Methanosarcina dominated in the reactors F–I. De Vrieze and colleagues ( 2012 ) stated that the ratio of Methanosarcina to Methanosaeta in reactors appears to be even more important than total archaeal numbers in monitoring reactor operational stability. All in all, microorganisms assumed to be acetotrophic were more abundant than hydrogenotrophic methanogens in the reactors studied, although cattle manure as a main substrate (except in reactor D) served as a constant inoculum of mainly hydrogenotrophic methanogens. The most frequently observed hydrogenotroph was Methanobacterium , which was detected in all investigated reactors using the ANAEROCHIP microarray. The hydrogenotrophic genera Methanoculleus and Methanobrevibacter were also detected, but yielded lower SNR signals. Furthermore, the number of hydrogenotrophic methanogens appeared to be higher in the Methanosaeta dominated reactors than in the Methanosarcina dominated reactors. Possibly, Methanosarcina , which is able to use the acetoclastic and the hydrogenotrophic methanogenesis pathways by utilizing H 2 and CO 2 (Jetten et al ., 1992 ; Kendall and Boone, 2006 ), was able to outcompete the hydrogen utilizing archaea. In this study we were able to confirm that acetate, which is considered to be the most important precursor during mesophilic anaerobic digestion and accounts for 60–80% of the CH 4 produced (Jeris and McCarty, 1965 ; Smith and Mah, 1966 ; Van den Berg et al ., 1974 ; Mountfort and Asher, 1978 ), was primarily responsibly for the clustering of the dominating archaea in the anaerobic digester sludges. Under low acetate concentrations (reactors A to E), Methanosaeta holds a competitive advantage over Methanosarcina spp., because of its 5–10 times higher substrate affinity. This higher substrate affinity is based on a high energy input in the activation of acetate by an acetyl‐CoA synthetase (Zinder, 1990 ; Jetten et al ., 1992 ; Kendall and Boone, 2006 ). Methanosarcina spp. cannot, reportedly, successfully compete under such limiting conditions. Raskin and colleagues ( 1996 ) reported acetate values of 0.1–0.18 mmol l −1 to be a minimum threshold for Methanosarcina growth. Min and Zinder ( 1989 ) and Westermann and colleagues ( 1989 ) also reported concentrations of 0.4–1.2 mmol l −1 acetate to be a minimum threshold for Methanosarcina growth. These acetate concentrations perfectly support the findings in this study, whereby Methanosaeta was able to dominate at low acetate levels (≤ 0.8 mmol l −1 , with the exception of two samples, where the acetate concentration was 1.8 mmol l −1 and 4 mmol l −1 ). In contrast, Methanosarcina outcompeted Methanosaeta in reactors F to I, where the acetate levels were mostly higher. According to the literature (Liu et al ., 1985 ; Schmidt et al ., 2000 ; Conklin et al ., 2006 ; Shin et al ., 2011 ) high changeover rates, low generation times (i.e. doubling times of 1.0–1.2 days), tolerance to sudden changes in pH (around 0.8–1.0 units) and higher ammonia and VFA tolerance ensure the dominance of Methanosarcina . While Methanosaeta sp. has a low maximum specific growth rate (μmax) of 0.20 day −1 and a half saturation constant (Ks) of 10–50 mg chemical oxygen demand (COD) l −1 , Methanosarcina sp. is characterized by a high μmax of 0.60 day −1 and a Ks of 200–280 mg COD l −1 (Gujer and Zehnder, 1983 ; McMahon et al ., 2004 ; Conklin et al ., 2006 ; Yu et al ., 2006 ; Qu et al ., 2009 ). Furthermore, Methanosarcina cells have a coccoidal form and grow in irregular flocks, protecting the cell against many harmful chemical agents (Demirel and Scherer, 2008 ). Methanogens like Methanosaeta , which occur as large filaments or non‐motile rods (Patel and Sprott, 1990 ) appear to be more sensitive to high ammonia concentrations than Methanosarcina cells (Zhilina, 1976 ). According to Calli et al . ( 2005 ) and Goberna and colleagues ( 2010 ) this resistance of Methanosarcina is attributed to its ability to form cell clusters and flocs. The size and form of Methanosarcina thus corresponds to a higher volume‐to‐surface ratio (four to seven times higher than for Methanosaeta sp.) correlating to a lower ammonia diffusion rate per unit of cell mass, compared with filamentous methanogens. However, we found no significant correlations between ammonium ions (NH 4 + ) or free ammonia (NH 3 + ) and the archaeal community in this study, since toxic concentrations as postulated in the literature (> 2.7 g l −1 NH 4 + and > 0.15 g l −1 NH 3 + ; Fachagentur für Nachwachsende Rohstoffe, 2006 ) were not exceeded. All the physicochemical parameters measured in this study were found to correlate positively (although not significantly) with Methanosarcina (Table 2 ), indicating a higher stress tolerance than Methanosaeta , for which the correlations were all negative. Syntrophic acetate oxidation (SAO), a two step process where acetate is first oxidized to CO 2 and H 2 (often through Clostridium sp.) and subsequently converted to methane by hydrogenotrophic methanogens, is reportedly favoured under thermophilic conditions (Zinder and Koch, 1984 ; Petersen and Ahring, 1991 ), at very low levels of H 2 partial pressure (i.e. between 2.6 and 74 Pa according to Hattori, 2008 ) and in the presence of very high inhibitor concentrations, particularly ammonium and VFAs (Schnurer et al ., 1999 ; Schnurer and Nordberg, 2008 ). Under such conditions, especially in Methanosaeta dominated reactors, which are very sensitive to stress, an increase in the organic loading rate can cause a shift from direct acetate cleavage towards syntrophic acetate oxidation coupled with hydrogenotrophic methanogenesis (Sasaki et al ., 2011 ). According to Schnurer and Nordberg ( 2008 ), Nettmann and colleagues ( 2010 ) and Sasaki and colleagues ( 2011 ) such community changes were reported in ammonium ranges of 3000 mg total ammonia nitrogen (TAN) l −1 . Also high acetate concentrations can result in a shift to syntrophic acetate oxidation (Hao et al ., 2011 ). Methanosarcina species, on the other hand were found in over 90% of the methanogenic population analysed in biogas reactors with SAO as the main methanogenic pathway and seem to act as hydrogen‐utilizing methanogens in reactors with SAO (Karakashev et al ., 2006 ; Karlsson et al ., 2012 ). According to these results, an interaction between SAO and Methanosarcina in reactors F–I is possible, but not proven, since the focus of this study was limited to methanogenic 16S rDNA. The archaeal communities appeared to be highly stable and resilient in all biogas reactors. As already known (Wittebolle et al ., 2009 ) community evenness corresponds with functional stability, by having a higher capacity to use redundant functional pathways. Traversi and colleagues ( 2011 ) detected a positive correlation between biogas production efficiency and the genera Methanosarcina and Methanosaeta . It seems that these organisms, primarily Methanosarcina , can act as an indicator for reliable reactor functioning, and help to diagnose imbalances in the microbial community (De Vrieze et al ., 2012 )."
} | 3,653 |
34179647 | PMC8223434 | pmc | 2,828 | {
"abstract": "In this paper, a\nmethod for preparing a high-stability superhydrophobic\npaper with temperature-induced wettability transition is proposed.\nFirst, a temperature-responsive superhydrophobic triblock polymer\nPHFMA–PTSPM–PNIPAAm was prepared by one-step polymerization\nof TSPM, HFMA, and NIPAAm in a mass ratio of 0.3:0.3:0.3, then a superhydrophobic\npaper with a good temperature response was successfully prepared by\ngrafting amino-modified SiO 2 with the polymer to modify\nthe surface of the paper. A further study found that when the mass\nratio of amino-modified SiO 2 to polymer is 0.2, the coating\nhas good superhydrophobicity and transparency. What is more, the prepared\nmodified paper is in a superhydrophobic state when the temperature\nis higher than 32 °C, and is in a superhydrophilic state when\nit is lower than 32 °C, which can realize free conversion between\nsuperhydrophobic and superhydrophilic states. In addition, the superhydrophobic\npaper prepared by this method not only has high oil–water separation\nefficiency, and the superhydrophobic coating shows good stability\nand transparency, but also has low requirements of environmental conditions\nfor preparation, relatively simple preparation process, and strong\nrepeatability, and it has a very broad application prospect.",
"conclusion": "3 Conclusions In this study, a one-step method was used to prepare a temperature-responsive\nsuperhydrophobic triblock polymer PHFMA–PTSPM–PNIPAAm\nby polymerizing the monomers HFMA, TSPM, and NIPAAm in a mass ratio\nof 0.3:0.3:0.3, then the amino-modified SiO 2 was grafted\nwith the polymer to modify the surface of the paper, thus a superhydrophobic\npaper with a good temperature response was successfully prepared.\nWhen the mass ratio of amino-modified SiO 2 to polymer is\n0.2, the coating has good superhydrophobicity and transparency. The\nprepared modified paper is in a superhydrophobic state when the temperature\nis higher than 32 °C, and is in a superhydrophilic state when\nit is lower than 32 °C, which can realize free conversion between\nsuperhydrophobic properties and superhydrophilic properties. On the\nwhole, the superhydrophobic paper prepared by this method not only\nhas high oil–water separation efficiency, and that the superhydrophobic\ncoating shows good stability and transparency, but also has low requirements\nof environmental conditions for preparation, a relatively simple preparation\nprocess, and strong repeatability, and it has a very broad application\nprospect in the fields of oil–water separation in actual industrial\nproduction.",
"introduction": "1 Introduction Superhydrophobic\nmaterials with an intelligent response are materials\nthat control the wettability transformation of a material surface\nthrough physical change or chemical reaction under external stimulation, 1 − 4 and environmental stimulus-sensitive superhydrophobic surfaces have\nbecome a hot area of academic research. 5 Studies have found some common environmentally responsive superhydrophobic\nsurfaces, such as light response, temperature response, pH response,\nelectric field response, ionic liquid response, and magnetic response. 6 − 14 Nowadays, a lot of waste oil and water mixtures are produced in\nthe industrial production process, and superhydrophobic materials\ncan not only separate these mixtures through filtration or single-phase\nselective absorption, but also realize the free switching of oil–water\nseparation according to production needs in different external environments,\nwhich shows their strong industrial application value. Therefore,\nthe preparation of intelligent responsive materials has become an\nimportant issue of industrial production. 15 − 18 However, due to the high cost\nand complex preparation process, their application value in industrial\nproduction is limited to a certain extent. 19 , 20 After experimental research and practical verification, it is gradually\nfound that temperature-responsive superhydrophobic surfaces have the\nadvantages of a fast response, low requirement of preparation equipment,\nand simple preparation process, which make them show a greater production\nand application value and also set off an upsurge of academic research\non temperature-responsive superhydrophobic surfaces. 21 , 22 From the existing research, many temperature-responsive superhydrophobic\nsurface preparation methods have been proposed, but these methods\nstill have certain limitations. First, by drip-coating PCL solution\n(poly-ε-caprolactone) on a substrate with a certain array structure\nto form a film on the surface, a reversible temperature conversion\nsuperhydrophobic surface with a critical temperature of 60 °C\nis obtained. Or free radical polymerization was used to react capric\nacid with butyl phthalate, immersing the fiber in the reactant to\nobtain a low surface energy, and then silica was used for roughness\nmodification to obtain an intelligent superhydrophobic surface with\npH control. However, due to the lack of adhesion between the polymer\nprepared by this method and the rough surface, the stability of the\nsuperhydrophobic surface is not high. 23 , 24 Second, by\ngrafting temperature-sensitive organics onto a silica gel substrate\nwith an array of nanopillars, the distance and width of the protrusions\nare controlled to affect the temperature change, thereby realizing\nthe reversible conversion of superhydrophobicity and superhydrophilicity.\nHowever, in this method, due to the poor flexibility of the substrate,\nthe difficulty of processing, the lack of recyclability and degradability,\netc., to a certain extent, its application value in production is\nlimited. 25 Third, the rough surface was\nconstructed by layer by layer self-assembly of nano-SiO 2 and polyacrylamide salt on the substrate. Then, a single layer of\nfluorinated azobenzene is modified on the rough surface to obtain\na surface that can realize the reversible conversion of superhydrophobicity\nand superhydrophilicity under ultraviolet light. However, the transparency\nof the coating prepared by this method is not ideal, which affects\nthe color and light transmittance of the substrate. 26 Fourth, a strawberry-like TiO 2 film is prepared\nby the seed growth method, whose surface contact angle can reach 163°,\nwhich can realize the conversion of superhydrophobicity and hydrophilicity\nunder natural light. However, from the perspective of the preparation\nprocess of this method, the reaction conditions of this method are\nharsh and the process is complex, and its applicability for commercial\nproduction is poor. 27 In this study,\na one-step method was used to promote the temperature-responsive\nmonomer N -isopropylacrylamide (NIPAAm), the fluorine-containing\nmonomer hexafluorobutyl methacrylate (HFMA), and the silicon-containing\ncross-linking agent 3-trimethoxysilyl propyl methacrylate (TSPM) polymerized\nin tetrahydrofuran solution to prepare a low surface energy triblock\npolymer with temperature-induced wettability transition. In order\nto improve the roughness of the paper and enhance the adhesion between\nthe polymer and silica, APTES was used to aminate silica nanoparticles,\nand the prepared polymer and amino-modified SiO 2 were grafted\nin solution to obtain a temperature-responsive superhydrophobic coating,\nwhich was sprayed on the surface of the substrate to obtain a superhydrophobic\npaper. NIPAAm in the polymer can make the paper surface have temperature-responsive\nproperties, and HFMA provides the low surface energy required for\nsuperhydrophobic surfaces. In the experiment, we used the silicon-containing\ncross-linking agent TSPM to cross-link the fluorine-containing monomer\nHFMA and the temperature-responsive monomer PNIPAAm to prepare high\nmolecular polymers, and then used TSPM to graft amino-modified silica\nonto the polymer to improve the surface roughness of the substrate\nto enhance the stability of superhydrophobic properties. 28 , 29 This method has the advantages of simple preparation process, high\nefficiency, and a wide range of applications (suitable for various\ntypes of papers). The prepared temperature-responsive superhydrophobic\npaper not only has the advantages of light weight, easy to carry and\ntransport, good printability, low cost, strong recyclability, and\ngood biodegradability, but can also achieve the reversible conversion\nbetween superhydrophobic and superhydrophilic properties by changing\nthe temperature, and the effective separation of oil–water\nmixtures can be achieved in the superhydrophobic state. This kind\nof coating surface with high transparency and high stability shows\ngreat potential for industrial applications. 30",
"discussion": "2 Results and Discussion 2.1 Effect\nof the Monomer Ratio on Wettability\nand Responsiveness The influence of different proportions\nof HFMA, TSPM, and NIPAAm on the wettability and response performance\nof the paper was explored. In this study, the polymer was synthesized\nin three proportions and then grafted with amino-modified SiO 2 , and three different superhydrophobic papers were prepared\non the basis of sufficient roughness in the structure, through comparative\nanalysis to determine the monomer ratio with the best overall performance. The three prepared papers were pretreated at 60 °C and then\nthe contact angle test was performed. It can be seen from Table 1 that when the ratio\nof HFMA, TSPM, and NIPAAm is 0.1:0.3:0.3, the contact angle is 143\n± 2°. With the increase of the ratio of HFMA, when the ratio\nreaches 0.3:0.3:0.3, the contact angle is 158 ± 3°, reaching\nthe superhydrophobic condition; and when the ratio is 0.6:0.3:0.3,\nthe contact angle reaches 159 ± 1°. Table 1 Surface\nContact Angle and Responsiveness\nof the Modified Paper Prepared with Different Monomer Ratios HFMA TSPM PIPAAm water contact\nangle responsiveness sample 1 0.1 0.3 0.3 143 ± 2 √ sample 2 0.3 0.3 0.3 158 ± 3 √ sample 3 0.6 0.3 0.3 159 ± 1 × The\nwater contact angle of the three papers was measured under\ndifferent temperature conditions to explore the effect of different\nmonomer ratios on the response performance of the paper. When the\nmonomer ratios of HFMA:TSPM:NIPAAm were 0.1:0.3:0.3 and 0.3:0.3:0.3,\nthe modified paper could respond to different temperatures. When the\nratio was 0.3:0.3:0.3, the paper was placed in a 60 °C environment\nfor pretreatment, and then the contact angle test was performed at\na constant temperature, the contact angle at this time is 157 ±\n3°, the water droplets were spherical on the surface of the paper,\nthe paper has superhydrophobic properties, and it remains superhydrophobic\nafter 5 h ( Figure 1 a). However, after the paper was pretreated in an environment of\n10 °C, the liquid droplets on the surface quickly penetrate into\nthe paper, and the water droplets were gradually absorbed by the paper\nafter a few seconds and spread completely on the surface within 30\ns, and the paper changed from the superhydrophobic state to superhydrophilic\nstate (the process is shown in Figure 1 b). It can be seen from Figure 1 a,b that the superhydrophobic paper already\nshows good temperature-response performance. Further research found\nthat this transition between superhydrophilic and superhydrophobic\nstates was completely reversible within 10 cycles ( Figure 1 c). When the monomer ratio\nHFMA:TSPM:NIPAAm was 0.6:0.3:0.3, because the fluorine atoms with\nlow surface energy play a superhydrophobic role in the polymer, with\nthe increase of the HFMA content, the proportion of hydrophobic blocks\nalso increased, which makes the paper lose its responsiveness and\nmaintain the superhydrophobic state at different temperatures. Figure 1 When the monomer\nratio of HFMA:TSPM:NIPAAm is 0.3:0.3:0.3, the\ncontact angle change of the paper at different temperatures was measured:\n(a) T = 60 °C, (b) T = 10 °C,\nand (c) reversible conversion of superhydrophilic and superhydrophobic\nproperties of the paper surface at 10 °C and 60 °C. Based on the above research results, it can be\nfound that the proportion\nof fluorine-containing monomer is the main factor affecting the hydrophobicity\nof the coating. However, when the block ratio is too high, the responsiveness\nof the coating surface will decrease or even lose responsiveness.\nConsidering hydrophobic performance and response performance comprehensively,\nthe best ratio of HFMA:TSPM:NIPAAm was found to be 0.3:0.3:0.3 in\nthis experiment. 2.2 Accurate Control of Roughness In\ngeneral, the superhydrophobic modification of the paper will not only\nchange the wettability of the paper, but also change the color, and\nthe relevant studies have also found that the content of SiO 2 will directly affect the transparency and superhydrophobicity of\nthe coating. 31 Figure 2 a shows the direct relationship between the\nsuperhydrophobicity (test the water contact angle under superhydrophobic\nconditions, i.e., T = 60 °C) and color difference\n(color difference from the original paper) of several modified papers\nprepared under different mass ratios of amino-modified SiO 2 and temperature-responsive polymers. Figure 2 b shows the reflectance spectra of three\npapers prepared with amino-modified SiO 2 and polymer under\ndifferent mass ratios. While testing the reflectance spectra of the\nthree papers, the L , a , and b values of the three are also tested ( L , a , and b represent the chromaticity\nvalue of the object color, L represents brightness,\na represents red-green, and b represents yellow-blue), the L *, a *, and b * values,\nquantitatively characterizing the color of the samples, are reported\nin Table S1 . Figure 2 c shows the scanning electron microscopy\n(SEM) image of the fiber on the surface of the paper. It can be seen\nfrom the figure that the surface roughness of the paper increases\nwith the increase of the content of amino-modified silica. Specifically,\nin Figure 2 a, curve\nA1 shows the effect of the SiO 2 content on the contact\nangle, and curve A2 shows the color difference Δ E caused by the SiO 2 content on paper surface. It can be\nseen from Figure 2 a\nthat as the content of SiO 2 increases within a certain\nrange, the surface of the paper will become rougher. While the superhydrophobic\nproperties of the paper are significantly improved, it also affects\nthe light transmittance of the coating and the color of the paper\nsurface, so that the chromatic aberration Δ E of the paper also increases. When SiO 2 is not added,\nthe contact angle of the paper surface is about 90°, and the\nfiber surface is smooth and flat ( Figure 2 c(Ι)). When the ratio of the SiO 2 to polymer content is 0.1, the paper has poor hydrophobicity\ndue to the insufficient surface roughness, and the contact angle is\nlower than 150° (the surface of the paper fiber is shown in Figure 2 c(II)). When the\nratio of the SiO 2 to polymer content is 0.2, the paper\nsurface obtains sufficient roughness (the surface of the paper fiber\nis shown in Figure 2 c(III)), the contact angle at this time is greater than 150°,\nand the paper obtains superhydrophobic properties. The reflectance\nspectrum is shown as curve B2 in Figure 2 b, compared with the reflectance spectrum\ncurve B1 of the original paper, the treated paper has a similar spectral\nreflectance in the range of 400–600 nm, a slight deviation\nin the red spectral region of 600–700 nm, and the measured\ncolor difference is Δ E = 0.91. When the ratio\nof the SiO 2 to polymer content reaches 0.3, the SEM image\nof the paper fiber is shown in Figure 2 c(IV). Compared with a silica content ratio of 0.2,\nalthough the change in the contact angle of the paper surface is very\nsmall at this time, the color difference is Δ E = 1.58, which has exceeded the acceptable color difference range\nof the human eye (Δ E < 1). 32 At this time, the reflective recording curve is shown as\nB3 in Figure 2 b, and\nits deviation from the reflectance spectrum curve B1 of the original\npaper in the 400–600 nm red spectral region is greater. Therefore,\naccording to Figure 2 a,b, the best mass ratio of amino-modified SiO 2 to polymer\nwas found to be 0.2. Figure 2 Characterization of the superhydrophobic paper prepared\nwith amino-modified\nSiO 2 and polymer under different mass ratios: (a) contact\nangle and chromatic aberration, (b) reflectance spectrum, (c) SEM\nimages under different mass ratios: (I) 0, (II) 0.1, (III) 0.2, and\n(IV) 0.3. 2.3 Structure\nCharacterization Figure 3 shows the infrared\nspectra of TSPM, temperature-responsive polymer PHFMA–TSPM–NIPAAm,\namino-modified SiO 2 , temperature-responsive superhydrophobic\nPHFMA–TSPM–NIPAAm/SiO 2 –NH 2 coating. Curve a is the infrared spectrum of TSPM. As can be seen\nfrom the figure, 2945 cm –1 is the stretching vibration\nabsorption peak of the methyl group, 2841 cm –1 is\nthe characteristic stretching vibration absorption peak of the methylene\ngroup, 1720 cm –1 is the stretching vibration absorption\npeak of C=O, and 1638 cm –1 is the stretching\nvibration peak of C=C, the symmetrical stretching vibration\npeak of Si–C appears at 813 cm –1 , and the\nstretching vibration peak of Si-OC appears at 1078 cm –1 . Curve b is the infrared spectrum of the polymer PHFMA–PTSPM–NIPAAm.\nIn addition to the characteristic peaks of the curve a, the stretching\nvibration absorption peak of CF appears at 1160 cm –1 , the peaks appearing at 3298 and 1546 cm –1 are\ncaused by the stretching and bending vibrations of the N–H\npeak in the amide of NIPAAm, and the characteristic peak of C=C\ndisappears at 1638 cm –1 , the changes in the infrared\nspectrum confirm the synthesis of the polymer PHFMA–PTSPM–PNIPAAm.\nCurve c is the infrared spectrum of amino-modified silica. As can\nbe seen from the figure, 1051 cm –1 is the antisymmetric\nstretching vibration peak of Si–O–Si, the characteristic\nabsorption peaks of methyl and methylene stretching vibration appear\nat 2973 and 2880 cm –1 , respectively, at 1549 cm –1 is the flexural vibration peak of NH, and the broad\npeak near 3300 cm –1 is caused by the stretching\nvibration of N–H, the appearance of the above peaks proves\nthe modification of silica. Curve d is the infrared spectrum of the\nPHFMA–PTSPM–PNIPAAm/SiO 2 –NH 2 coating. The stretching vibration peak of Si–O–C can\nbe observed at 1078 cm –1 , the stretching vibration\npeak of C–F can be observed at 1160 cm –1 ,\nthe antisymmetric stretching vibration peak of Si–O–Si\nin amino-modified SiO 2 can be observed at 1051 cm –1 , and the bending vibration peak and stretching vibration peak of\nN–H can be observed at 1550 and 3290 cm –1 , respectively. The characteristic peaks of curves b and c seen in\nthe figure can basically be observed in the Fourier transform infrared\n(FTIR) spectrum of the response polymer of the composite PHFMA–PTSPM–PNIPAAm/SiO 2 –NH 2 , and the results of FTIR also indicate\nthe successful synthesis and the successful introduction of amino-modified\nSiO 2 into polymers. 33 Figure 3 Fourier infrared\nspectrum analysis chart: (a) TSPM, (b) temperature-responsive\npolymer PHFMA–PTSPM–PNIPAAm, (c) amino-modified SiO 2 , and (d) temperature-responsive superhydrophobic coating\nPHFMA–PTSPM–PNIPAAm/SiO 2 –NH 2 . The surface elements of the original\npaper, the modified paper\ncoated with the polymer (PHFMA–PTSPM–PNIPAAm), and the\nmodified paper coated with the composite coating (PHFMA–TSPM–NIPAAm/SiO 2 –NH 2 ) were analyzed by X-ray photoelectron\nspectroscopy (XPS), although XPS cannot fully describe the chemical\ncomposition of paper samples due to their rough surface, it does provide\nqualitative information about the chemical changes before and after\nmodification. Figure 4 a shows the XPS spectrum of the unmodified original paper. It can\nbe seen from the figure that the surface of the original paper is\nmainly composed of C and O elements, corresponding to the positions\nnear 283 and 532 eV, respectively. Four more peaks appeared at 101,\n151, 396, and 689 eV on the surface of the paper coated with the copolymer,\ncorresponding to the appearance of the Si 2p, Si 2s, N 1s, and F 1s\nsignals, indicating that the polymer was successfully modified to\nthe paper surface. It can be seen from Figure 4 c of the coated copolymer (PHFMA–PTSPM–PNIPAAm)\npaper that the strength of the Si 2p peak centered at 101 eV on the\npaper surface is significantly stronger than that of the original\npaper ( Figure 4 b),\nwhich is mainly due to the influence of the Si element on PTSPM, thus\nproving the successful introduction of PTSPM. It can be seen from Figure 4 d that the Si peak\nsignal on the surface of the composite PHFMA–TSPM–NIPAAm/SiO 2 –NH 2 modified paper is stronger than that\nof the first two papers, which is caused by the addition of SiO 2 , also indicating the successful combination of PHFMA–TSPM–NIPAAm/SiO 2 –NH 2 and the paper. 34 Figure 4 (a)\nXPS analysis chart; XPS Si 2p core level spectra of (b) original\npaper, (c) PHFMA–PTSPM–PNIPAAm-coated paper, and (d)\nPHFMA–TSPM–NIPAAm/SiO 2 –NH 2 -coated paper. Figure 5 shows the\nthermogravimetric analysis (TGA) curves of paper samples a, b, c,\nand d. After heating all paper samples to 700 °C, the remaining\nweight percentage of the unmodified original paper sample a is 0%;\nthe remaining weight percentage of the paper sample b after spraying\nordinary silica dispersion is 4.28%, this is mainly due to the residual\nSiO 2 in the sample; the remaining weight percentage of\npaper sample c sprayed with amino-modified SiO 2 increased\nto 7.9%, compared with the pure SiO 2 curve b, this substantial\nchange in the weight loss rate means that APTES reacts with the hydroxyl\ngroups on SiO 2 to form amino-modified silica. The remaining\nweight percentage of the paper sample d coated with the superhydrophobic\ncoating PHFMA–R-PTSPM–PIPAAm/SiO 2 –NH 2 was 6.97%, and the weight loss rate reached 93.03%. These\nphenomena indicate that the successful modification of modified SiO 2 and the composite of PHFMA–PTSPM–PIPAAm on\nthe surface of amino-modified SiO 2 effectively affect the\nthermal stability of the material. 35 Figure 5 TGA cures of\nsamples a, b, c, and d, (a) original paper, (b) paper\ncoated with SiO 2 , (c) paper coated with amino modified\nSiO 2 , and (d) temperature-responsive superhydrophobic paper.\nSamples were heated to 700 °C in an air atmosphere at a ramp\nrate of 10 °C min –1 . 2.4 Microscopic Morphology of Superhydrophobic\nPaper In order to more intuitively reveal the changes of\nthe superhydrophobic paper before and after modification, the micromorphology\nof the paper before and after modification was compared through SEM\nimages. In Figure 6 a,c, the surface of the unmodified paper is flat and smooth, and\nthere is no rough structure similar to protrusions. The fibers of\nthe paper are tightly interwoven, and the gaps between the fibers\nare clearly visible, this reflects many excellent properties of the\npaper, such as good air permeability. Figure 6 b,d is the modified paper fiber diagram.\nIt can be observed from Figure 6 b that the modified coating adheres uniformly, without a large\namount of agglomeration and stacking, and the overall structure and\nshape of the fiber have not changed significantly. This indicates\nthat the introduction of copolymers and nanoparticles not only did\nnot destroy the structure of the fiber, but also did not have a significant\nimpact on the paper’s good air permeability and other properties.\nFrom the comparison of Figure 6 c,d, it can be clearly observed that the fiber surface becomes\nrough due to the attachment of SiO 2 particles, which gives\npaper the basic conditions for superhydrophobic properties. 36 Figure 6 SEM images of the paper surface: (a) original paper and\n(b–d)\ntemperature-responsive superhydrophobic paper. 2.5 Determination of Critical Temperature The\ndifferential scanning calorimetry (DSC) curves of different monomers\nand temperature-responsive polymers in the experiment are shown in Figure 7 . By comparing and\nanalyzing different DSC curves, we can determine the phase transition\nlow critical solution temperature (LCST) of temperature-responsive\npolymers, which is the peak point of the heat flow change during heating. 37 It can be seen from the figure that the DSC\ncurve of HFMA has no inflection point between 25 and 45 °C, so\nthere is no glass transition temperature in this temperature range.\nThe peak point of NIPAAm and PHFMA–PTSPM–PNIPAAm appeared\nnear 33 and 32 °C, respectively, in the DSC curve and the DSC\ncurves of the polymer are similar to those of monomer NIPAAm. It can\nbe seen from Figure 7 that there is little influence on the critical temperature change\nbefore and after polymerization, and the temperature response block\nis the PNIPAAm block in the polymer, which indicates that the phase\ntransition temperature of the polymer is about 32 °C. Theoretically,\nwhen the temperature is above the LCST, the temperature-responsive\npolymer exhibits superhydrophobicity, and when the temperature is\nbelow the LCST, the temperature-responsive polymer exhibits hydrophilicity. 38 , 39 Figure 7 DSC\ncurve of HFMA, NIPAAm, and PHFMA–PTSPM–PNIPAAm. In the experiment, we also determined the critical\nresponse temperature\nof the modified paper by gradually narrowing the temperature range.\nIn the specific operation, we have prepared several oil–water\nmixtures at different temperatures, and observed the separation state\nof these oil–water mixtures by the paper. First, six identical\noil–water mixtures were prepared in the beaker (50 mL oil red-stained\nbromobenzene and 50 mL blue deionized water), and the temperature\nof the oil–water mixture was adjusted in steps of 10 °C\n(10, 20, 30, 40, 50, and 60 °C). It can be seen from Figure 8 a that when bromobenzene\ncomes into contact with superhydrophobic paper, it quickly penetrates\nthe surface, while water droplets can remain spherical on the surface\nof the paper and cannot wet the surface, the prepared paper has superhydrophobic\nand superlipophilic properties in this state. The device shown in Figure 8 b is used to separate\nand collect each oil–water mixture, the process is shown in Figure 8 b–e. As shown\nin Figure 8 b, the surface\nof the small beaker was covered with the modified paper and moved\nit to the large beaker, and slowly poured the mixture of water and\nbromobenzene on the surface of the paper. Due to the modified paper’s\nsuperhydrophobic and lipophilic properties at a specific temperature,\nthe organic matter in the mixture can quickly penetrate through the\nsurface of the paper into the small beaker, while the water in the\nmixture stays on the surface of the paper and flows into the large\nbeaker by gravity. When the temperature of the oil–water mixture\nis 10, 20, or 30 °C, the oil and water will leak into the small\nbeaker together. At this time, the paper does not have the function\nof separating oil and water, which means that the paper does not have\nsuperhydrophobicity when the temperature was below 30 °C. When\nthe temperature of the oil–water mixture is higher than 40\n°C, bromobenzene dyed with oil red is collected in the small\nbeaker, and blue deionized water is collected in the large beaker,\nthis indicates that the paper has superhydrophobic and superlipophilic\ncharacteristics when the temperature is higher than 40 °C. Figure 8 f shows bromobenzene\nand water collected at 40, 50, and 60 °C. Through the above experiments,\nit can be determined that the critical temperature range of the temperature-responsive\nsuperhydrophobic paper is 20–30 °C. In this temperature\nrange, the temperature was adjusted with a gradient of 1 °C and\nrepeated the above experiment to collect the oil–water separation\nmaterial, and finally determined the critical temperature value. According\nto the test results, the critical response temperature of the prepared\ntemperature-responsive superhydrophobic paper is 32 °C, which\nis consistent with the LCST value (32 °C) of the copolymer, which\nfurther shows that the modification in this study is successful. Figure 8 Oil–water\nseparation process: (a) different wettability\nof modified paper to water and bromobenzene, (b–e) specific\nprocess of oil–water separation, and (f) water and bromobenzene\ncollected after oil–water separation. 2.6 Mechanism Analysis As shown in Figure 9 , PNIPAAm is a temperature-responsive\nblock, the principle of temperature response is that the hydrogen\nbond between the amide group and the water molecule changes with different\ntemperatures. Specifically, when the temperature is lower than the\nmolecular LCST of PNIPAAm, the C=O and NH groups in PNIPAAm\nwill form hydrogen bonds between molecules with external water molecules,\nand the molecular chain is in a stretched state. Therefore, the polymer\nexhibits superhydrophilic properties. In addition, the role of hydrogen\nbonding can cause the hydration expansion of PNIPAAm, which provides\nenough power for the surface segment of PNIPAAm to cover other polymers,\nso that PNIPAAm dominates, which also greatly enhances its hydrophilic\nproperties. When the temperature is higher than the LCST of PNIPAAm,\nthe hydrogen bond between the molecules will gradually weaken as the\ntemperature increases, making the hydrophobic interaction between\nisopropyl groups more obvious. Especially when the temperature increases\nto a certain level, the hydrogen bonds between the C=O and N –H groups and water molecules will break, forming\nintramolecular hydrogen bonds. Compared with the state when it is\nunder LCST, PNIPAAm at this time will shrink and cause it to be dehydrated\nand collapsed, which will expose the hydrophobic groups and other\nhydrophobic polymers used in the polymerization reaction, such as\nHFMA, etc., so that the prepared polymer surface exhibits superhydrophobic\ncharacteristics. 40 − 42 Figure 9 Mechanism of the temperature response. 2.7 Stability Test In practical applications,\nsuperhydrophobic materials will not only be corroded by various solutions,\nbut will also suffer abrasion due to various pressures or mechanical\ndamage. Therefore, the corrosion resistance and abrasion resistance\nof the superhydrophobic paper must be considered. Figure 10 a,b respectively, shows the\ninfluence of acid, alkali and salt solution immersion, and friction\ntimes on the contact angle of paper at different destruction times.\nThe three curves in Figure 10 a, respectively, represent the influence of acidic solution,\nalkaline solution, and salt solution on the contact angle of paper\nat different destruction times. Specifically, as the immersion time\nof hydrogen chloride solution (HCL, pH = 1), sodium hydroxide solution\n(NaOH, pH = 14), and sodium sulfate solution (Na 2 SO 4 , pH = 7) increases, the contact angle of the paper has been\nreduced to varying degrees, but it can still maintain good hydrophobic\nproperties. After being soaked for 150 min, the contact angles of\nthe three soaked papers were 155 ± 1, 151 ± 1, and 156 ±\n2°, respectively. After a long period of time, the alkali solution\nhas a greater impact on the durability of the superhydrophobic properties\nof the paper. This is mainly because SiO 2 in the coating\nis dissolved by sodium hydroxide solution (NaOH), which leads to the\ndegradation of superhydrophobic property. In order to test the abrasion\nresistance of the paper, sandpaper was pasted on the bottom of the\nweight, and then the friction experiment was carried out on the paper\n( Figure 10 b). It is\ndefined as a cycle that the weight is pushed to the front end of the\npaper and then pulled back to the original position. After the amino-modified\nsilica is grafted with the polymer PHFMA–PTSPM–PNIPAAm\nand modified on the paper, a uniform and stable coating was formed\non the surface of the paper covered by the polymer, so the superhydrophobic\npaper showed strong friction resistance in the experiment. With the\nincrease of friction times, the paper contact angle decreases and\nchanges little. After 100 friction cycles, the paper can still maintain\na contact angle of 155 ± 2°. Therefore, the superhydrophobic\npaper has good friction resistance. Figure 10 Effect of different destruction conditions\non the contact angle\nof the superhydrophobic paper: (a) effect of soaking in HCL, NaOH,\nand Na 2 SO 4 solutions on the contact angle of\nthe paper and (b) effect of 100 friction cycles on the contact angle\nof the paper. 2.8 Oil–Water\nSeparation Performance Test The device shown in Figure 11 a was used to carry\nout the oil–water separation\nexperiment, the test was carried out at 60 °C, and the oil–water\nseparation performance was analyzed. In the study, 45 mL of organic\nmatter stained with oil red O was mixed with 45 mL of deionized water\nstained with methylene blue to prepare an oil–water mixture,\nand further analyze the separation performance of the modified superhydrophobic\npaper for different oil–water mixtures. The oil–water\nseparation efficiency and recycling use are used to characterize the\noil–water separation performance of the prepared superhydrophobic\npaper. The separation efficiency is calculated by the following formula V 1 is the volume\nof oil collected after an oil–water separation experiment is\ncompleted and V 0 is the initial volume\nof oil before the separation experiment. Figure 11 (a) Schematic diagram\nof the paper oil–water separation\nprocess, (b) separation efficiency of the paper for different oil–water\nmixtures, and (c) paper separation efficiency changes with the number\nof separations. Take 1,2-dichloroethane\nas an example, fix the modified paper in\nthe middle of two glassware, and pour the 60 °C mixture of 1,2-dichloroethane,\nbromobenzene, chloroform, and deionized water into the upper part\nof the container. Due to the hydrophobicity and lipophilicity of the\nmodified paper, 1,2-dichloroethane quickly permeated the paper and\nflowed into the containers below, while deionized water remained on\nthe modified paper, thus completing the separation of oil–water\nmixture ( Figure 11 a). The oil–water separation efficiency was calculated\nusing\nthe above formula. It can be seen from Figure 11 b that the separation efficiency of the\nmodified paper for various oil–water mixtures such as 1,2-dichloroethane,\nbromobenzene, and chloroform can reach more than 98%. In addition,\ndue to the oil–water separation process, the silica and low\nsurface energy polymers on the paper surface will be partially dissolved\nin the organic solution, and some organic impurities will block the\ngaps between the paper fibers during the oil–water separation\nprocess, and the separation efficiency of the paper also decreases\nas the number of oil–water separation cycles increases. However,\nwe are pleased that the oil–water separation efficiency of\nthe paper prepared in this study does not decrease significantly with\nthe increase of recycling times, and our experiments have found that\nthe modified paper can still maintain a separation efficiency of over\n96.4% after being recycled 40 times ( Figure 11 c)."
} | 8,708 |
37697024 | PMC10495451 | pmc | 2,829 | {
"abstract": "Analog hardware-based training provides a promising solution to developing state-of-the-art power-hungry artificial intelligence models. Non-volatile memory hardware such as resistive random access memory (RRAM) has the potential to provide a low power alternative. The training accuracy of analog hardware depends on RRAM switching properties including the number of discrete conductance states and conductance variability. Furthermore, the overall power consumption of the system inversely correlates with the RRAM devices conductance. To study material dependence of these properties, TaOx and HfOx RRAM devices in one-transistor one-RRAM configuration (1T1R) were fabricated using a custom 65 nm CMOS fabrication process. Analog switching performance was studied with a range of initial forming compliance current (200–500 µA) and analog switching tests with ultra-short pulse width (300 ps) was carried out. We report that by utilizing low current during electroforming and high compliance current during analog switching, a large number of RRAM conductance states can be achieved while maintaining low conductance state. While both TaOx and HfOx could be switched to more than 20 distinct states, TaOx devices exhibited 10× lower conductance, which reduces total power consumption for array-level operations. Furthermore, we adopted an analog, fully in-memory training algorithm for system-level training accuracy benchmarking and showed that implementing TaOx 1T1R cells could yield an accuracy of up to 96.4% compared to 97% for the floating-point arithmetic baseline, while implementing HfOx devices would yield a maximum accuracy of 90.5%. Our experimental work and benchmarking approach paves the path for future materials engineering in analog-AI hardware for a low-power environment training.",
"conclusion": "Conclusion HfO x and TaO x RRAM devices have great potential as analog switching devices for the implementation of customized AI hardware. We demonstrated that ultrafast (300 ps) switching can not only speed-up the conductance update duration but also benefit analog switching performance. We report on an approach to obtain a large number of conductance states while maintaining low off state conductance by using low compliance current during initial device electroforming, followed by high compliance current switching. We report that TaO x devices have a 10x lower conductance compared to HfO x devices when integrated into 1T1R cells using an identical processing approach. As a result, this makes TaO x devices more suitable for large resistive array-based accelerators. Additionally, we adopted a device-aware training approach with hyperparameter optimization for each switching condition for system-level benchmarking. From this benchmarking approach, we demonstrated that TaO x device performance metrics can yield a higher system-level accuracy of 96.4% as compared to 90.5% for HfO x devices, with accuracy approaching the floating point baseline of 97% accuracy.",
"introduction": "Introduction In recent years, neural networks have been applied to challenging problems such as image recognition and natural language processing, with the ability to surpass human-level accuracy 1 , 2 ; however, a tremendous amount of power is required to train these models. For example, ChatGPT (which is a version of the GPT-3 model) required 1,287 MWh of power for training 3 . The equivalent CO 2 emissions for training this model are 552 metric tons, which is around 110 years of an average person’s CO 2 emission 4 . The computational power requirements to train a neural network (NN) is the result of large amounts of data transfer and weight updates, as well as repeated matrix multiplication operations. The conventional von Neumann computing architecture physically separates memory and logic units and has limited parallelism. Leveraging parallelism and repeated multiplication operations, GPUs increase the training throughput significantly 5 . As data transfer between memory and logic units introduce significant power consumption and latency, in-memory computation holds further opportunities for power and efficiency improvements 6 (Fig. 1 a,b). Additional power and latency reduction can be achieved by analog computation of multiplication operations. Studies have predicted that approximately 100 to 1000 times more efficient neural network training is possible on an analog, in-memory processing unit, as compared to state-of-the-art GPU computing 7 . Figure 1 Analog computation of AI workloads is a promising solution to this power-intensive task. RRAM based AI harware accelerators leverage in-memory computation compared to state-of-the-art von Neumann architectures. ( a ) The von Neumann computing architecture consists of physically separated memory and logic units, which are bottlenecked by data transfer (shown by the blue arrow). ( b ) In-memory computation reduces power and latency by computing directly within the memory unit and also utilizes the inherent parallelism of the memory architecture. ( c ) A simple neural network is shown with input layer (blue), hidden layer (green), and output layer (red). ( d ) An example of implementing a RRAM-based memory array for hardware realization of a neural network where each RRAM stores synaptic weight values. Simultaneous input at each row as voltage can result in column current for multiply and accumulate (MAC) operations. It should be noted that MAC operations are a significant contributor towards overall neural network training workload. ( e ) A 1-transistor 1-RRAM (1T1R) unit cell is shown, highlighting a cross-section of the RRAM structure with oxygen vacancies in the switching layer depicted as grey dots. Resistive Random Access Memory (RRAM) based analog accelerators have shown promising results for low-power and high-speed neural network training 5 , 8 – 12 . RRAM-based memory is non-volatile in nature with analog data storage capabilities (Fig. 1 e). RRAM are typically configured as a metal-insulator-metal (MIM) structure consisting of a top and bottom electrode with a switching layer in between. These devices store data as measurable conductance states, controlled by the position and relocation of ions within a conductive filament (Fig. 2 ). In oxygen vacancy based switching devices, a conductive filament is formed within the switching layer, connecting the bottom and top electrode. The filament consists of an accumulated column of oxygen vacancies 13 . An increase of the vacancy concentration (typically near the bottom electrode) will increase the conductance of the filament and thus change the conductance state of the RRAM. Likewise, under reverse bias, the concentration of oxygen vacancies can be reduced, yielding lower conductance. The conductive filament can be modified gradually by consecutive short pulses, making RRAM devices a viable option for synapse-like weight storage 14 (Fig. 2 ). In addition, RRAM arrays can be used to perform analog vector matrix multiplications or multiply–accumulate (MAC) operations by coding the input vector into a range of voltages and the matrix weights into conductance states within the array. The resulting output current vector (y n ) contains the encoded result of the operation (Figure 1 (c,d)). Analog matrix multiplication results can be achieved within a single cycle and therefore have a computational complexity of O ( n ) compared to O ( n 3 ) with conventional logic. Figure 2 ( a , b ) Electrical pulses and polarities required for analog switching. Short-duration positive pulses increase the conductance (or weight) of the device whereas negative pulses reduce RRAM device conductance. The repeated alternation between positive and negative pulses results in the device achieving the so-called symmetry point, where successive positive and negative pulses do not significantly alter device conductance. Number of states and programming noise definitions are shown on the right. ( c ) Schematic of the physical changes occurring to the conductive filament of a RRAM device during positive and negative pulses for analog switching using hour-glass model 13 , 14 . Positive pulses across the device grow the filament towards the bottom electrode, hence reducing the oxide gap in between. This increases the overall device conductance, whereas the opposite behavior occurs during negative pulses across the device, decreasing the device conductance. The most important performance metrics of the RRAM-based hardware accelerators are training accuracy and system power consumption. Training accuracy depends on the device switching behavior, such as the linearity, asymmetry during the process of potentiation or depression (increasing or decreasing the synaptic weight) 15 , 16 . Overall training accuracy also depends on the number of conductance states (defined as the conductance range divided by mean conductance change at the symmetry point) and programming noise (standard deviation of the change of conductance divided by mean change of conductance at the symmetry point) (Fig. 2 b) 17 , 18 . Generic neural network algorithms (e.g Stochastic Gradient Decent (SGD)) depend on ideal device weight updates (linear, symmetric and large number of states), and deviation from linear weight update results in a decreased training accuracy 19 , 20 . Most reported analog device behaviors with RRAM devices, however, are non-ideal with a non-linear asymmetric weight update, low number of states, and a relatively large noise level 8 , 21 . Hence, there is a need for a training algorithm that can handle the imperfect behavior of the RRAM devices and still provide high training accuracy. In this work we adopted a analog fully in-memory training algorithm named Tiki-Taka v2 (TTv2) that embraces practical non-ideal device switching behavior without compromising performance 17 , 22 . TTv2 results in higher accuracy compared to the generic SGD algorithm with a comparatively lower number of states, high non-linearity, and variations 17 . This algorithm relies on three RRAM unit cells, one RRAM for analog conductance/weight update and the second RRAM to store the “symmetry point” as a reference point between positive and negative weights, one RRAM for gradient accumulation around symmetry point 23 . This RRAM conductance “symmetry point” can be achieved by juxtaposing positive and negative pulses 23 , 24 (Fig. 2 ). Other device properties like conductance values significantly influence overall system power consumption while reducing IR drop 10 . The device switching linearity/non-linearity and conductance of the device are significantly influenced by the material stack of the RRAM device 15 , 16 . As a result, a RRAM-based accelerator’s overall efficacy depends largely on its material selection. For high volume manufacturability, conventional complementary metal oxide semiconductor (CMOS) fabrication compatibility of the RRAM material stack is one of the most important factors 13 . Due to proven CMOS compatibility, available material deposition/process tools in the foundry, HfO x and TaO x -based RRAM devices have had the highest research interest compared to other stacks 25 . In this work, we focus on benchmarking analog performance such as the number of states, dynamic range, and programming noise of HfO x and TaO x RRAM devices, both fabricated on 65nm CMOS technology process node. Furthermore, we benchmarked system-level training accuracy from both HfO x and TaO x device switching behaviors leveraging the TTv2 algorithm. In this work, RRAM devices with different switching layers (HfO x and TaO x ) were fabricated using a standard CMOS process technology (65 nm) for benchmark comparison of their potential for analog performance and utility in neural network training and accelerator approaches. Analog switching experiments with pulse length as short as 300 ps demonstrated the ability to achieve 35 analog states with TaO x versus 29 states with HfO x , with TaO x having lower off-state conductance, which is amenable to low power operation. We report the impact of pulse amplitudes on symmetry point convergence was systematically studied for the first time. Finally, we adopted an analog fully in-memory training algorithm (TTv2) for system-level benchmarking of neural network training and accuracy, and report on the baseline RRAM device switching data, fitting, and hyperparameter optimization to yield up to 96.4% accuracy versus a floating point peak accuracy of 97%.",
"discussion": "Discussion In this work we report on an approach to achieve a high number of states in RRAM-based synaptic devices with corresponding low off state conductance, by first using a reduced compliance current during the forming process and then higher compliance current during subsequent switching events. Previous reports suggest that a minimum compliance current is required to enable RRAM device switching 27 . The authors hypothesized that RRAM devices require a minimum current to assist with oxygen vacancy rearrangement, as a key driver for adjusting device conductance. We extend this hypothesis to explain the low off-state conductance. After forming at a low compliance current, a higher current is required during switching to assist with oxygen vacancy movement. Higher vacancy movement can result in a reduced concentration of vacancies at the end of the filament during the negative pulse cycle. In turn, this can result in lower conductance. Similarly low compliance current would result in lower vacancy movement as a result, higher off-conductance. The change in conductance during negative pulse cycling is affected by the prior forming or set condition that was used, which determines filament size and composition. Previous reports using analytical simulations 23 , 30 assumed that symmetry points can be achieved at any location and with any number of states. Our experiments using fully CMOS-integrated devices show that a stable symmetry point convergence can not be achieved at any indiscriminate location. We observed that symmetry point stability is largely dependent on the amplitude of the positive and negative voltage pulses during the switching (Fig. 4 ). The symmetry point can shift upward or downward, due to an imbalance of positive and negative voltage pulses. A large positive pulse with a smaller corresponding negative pulse tends to shift the symmetry point upward towards higher conductance. On the other hand, when the negative voltage pulse amplitude is larger, as compared to the positive pulse, the symmetry point tends to shift downwards. For our RRAM devices, symmetry point separation tends to occur when positive and negative voltage pulses are larger e.g. at + 1.5 V − 1.5 V. As fabricated in this study, CMOS-integrated TaO x RRAM devices exhibited a significantly lower conductance compared to HfO x RRAM devices that were integrated with an identical configuration (1T1R). Conductance of RRAM devices can be attributed to the effect of the different metal-insulator interface, the effect of oxygen scavenging layer, vacancy concentration of the switching layer and bulk oxide conductance. It has been suggested in previous publications that TaO x and HfO x have different defect migration energy for same oxygen scavenging layer metal which can also lead to different conductance values 13 , 29 . Experimental results reported previously showed strong dependence on the choice of oxygen exchange layer (OEL) for the RRAM stack and on and off conductance of the RRAM device 28 . This suggests further hardware and modeling experiments are required for further understanding and optimization of each material RRAM device stack."
} | 3,920 |
26130086 | PMC4486974 | pmc | 2,830 | {
"abstract": "Ralstonia eutropha is a facultative chemolithoautotrophic bacterium that uses the Calvin–Benson–Bassham (CBB) cycle for CO 2 fixation. This study showed that R. eutropha strain H16G incorporated 13 CO 2 , emitted by the oxidative decarboxylation of [1- 13 C 1 ]-glucose, into key metabolites of the CBB cycle and finally into poly(3-hydroxybutyrate) [P(3HB)] with up to 5.6% 13 C abundance. The carbon yield of P(3HB) produced from glucose by the strain H16G was 1.2 times higher than that by the CBB cycle-inactivated mutants, in agreement with the possible fixation of CO 2 estimated from the balance of energy and reducing equivalents through sugar degradation integrated with the CBB cycle. The results proved that the ‘gratuitously’ functional CBB cycle in R. eutropha under aerobic heterotrophic conditions participated in the reutilization of CO 2 emitted during sugar degradation, leading to an advantage expressed as increased carbon yield of the storage compound. This is a new insight into the role of the CBB cycle, and may be applicable for more efficient utilization of biomass resources.",
"discussion": "Discussion The 13 C abundance in P(3HB) synthesized by H16G∆∆ cbbLS , in which the CBB cycle cannot function owing to the lack of Rubiscos, was slightly increased from the natural abundance of 1.1% to 1.9% after incubation with [1- 13 C 1 ]-glucose for 12 h. Considering that R. eutropha H16 does not possess phosphofructokinase and 6-phosphogluconate dehydrogenase in the Embden–Meyerhof (EM) and pentose phosphate (PP) pathways, respectively, the ED pathway is a unique sugar degradation pathway, so that [1- 13 C 1 ]-pyruvate and unlabeled glyceraldehyde 3-phosphate (GAP) were produced from [1- 13 C 1 ]-glucose. When [1- 13 C 1 ]-pyruvate was converted to acetyl-CoA by decarboxylation or anaplerosis and the successive TCA cycle, 13 CO 2 was emitted and unlabeled acetyl-CoA was produced, even in the case of fixation of H 13 CO 3 − by anaplerotic reactions. Thus, the 13 C atom derived from [1- 13 C 1 ]-glucose was not incorporated into P(3HB) through the first round of the ED pathway ( supplementary Fig. S1 ). Besides decarboxylation, [1- 13 C 1 ]-pyruvate could be converted via the EM pathway, resulting in distribution of the 13 C atom at the 1-position of triose phosphates, followed by formation of [3- 13 C 1 ]-, [4- 13 C 1 ]- or [3,4- 13 C 2 ]-fructose-6-phosphate (F6P). While, [1- 13 C 1 ]-F6P was formed by 6-phosphorylation and isomerization of [1- 13 C 1 ]-glucose. These 13 C atoms were distributed at the 1-, 2-, 3- and/or 4-positions of F6P through interconversion of C 4 -C 7 sugar phosphates by a non-oxidative branch of the PP pathway [ supplementary Fig. S5(A) and (C)] or RuBP regeneration steps in CBB cycle [ supplementary Fig. S5(B) and (D) ]. The 13 C-labelled F6P molecules were converted to acetyl-CoA containing 13 C at the 1 and/or 2 positions by the ED pathway and oxidative decarboxylation of pyruvate. Thus, the 13 C atoms could be incorporated into P(3HB) without Rubisco-mediated CO 2 fixation ( supplementary Fig. S2 ). This event explained the slight increase of 13 C abundance in P(3HB) to 1.9% in the H16G∆∆ cbbLS strain. When the CBB cycle in the H16G strain operated during P(3HB) biosynthesis from glucose, the CO 2 molecules emitted by oxidative decarboxylation of pyruvate had a chance to couple with RuBP by the Rubisco-mediated reaction. As shown in supplementary Fig. S3 , the complete CBB cycle likely established a new pathway for conversion of the 13 C-labelled RuBP derived from [1- 13 C 1 ]-glucose to 13 C-labelled acetyl-CoA, even when 12 CO 2 was fixed. It was expected that the fixation of 13 CO 2 , generated from [1- 13 C 1 ]-pyruvate, by Rubisco enriched 13 C atoms in the sugar phosphates via interconversion of sugar phosphates, leading to incorporation of more 13 C into P(3HB) via 13 C-enriched acetyl-CoA ( supplementary Fig. S4 ). Indeed, the 13 C abundance in P(3HB) synthesized from [1- 13 C 1 ]-glucose by H16G increased to 5.6%. This observation demonstrated the actual fixation of CO 2 emitted by decarboxylation during glucose degradation by the CBB cycle, and incorporation of the fixed carbon into P(3HB) in R. eutropha . Metabolomic analysis revealed the increase of the [ 13 C 1 ]-isotopomers of most sugar phosphates, including RuBP and 3PGA, in the H16G strain during incubation with [1- 13 C 1 ]-glucose ( Fig. 2 ). This observation provided strong evidence for the actual flux of the CBB cycle with fixation of the [1- 13 C 1 ]-glucose-derived 13 CO 2 . In contrast, the [ 13 C 1 ]-isotopomers decreased in H16G∆∆ cbbLS . The presence of intracellular RuBP without increase of 13 C was reasonable, because RuBP could be generated by phosphoribulokinase (CbbP), but not be converted owing to the lack of Rubiscos in this strain. The absence of RuBP in H16G∆ cbbR was consistent with the too low expression of cbbP and other cbb genes caused by the deletion of the transcriptional activator CbbR. However, the abundances of the [ 13 C 1 ]-isotopomers for many metabolites in H16G∆ cbbR tended to show changes between those in H16G and H16G∆∆ cbbLS . These results suggested that the subtly expressed cbb genes even in the absence of the activator CbbR, as shown by qRT-PCR, which resulted in the slightly higher 13 C-abundance in P(3HB) synthesized from [1- 13 C 1 ]-glucose by H16G∆ cbbR (2.3%) than by H16G∆∆ cbbLS (1.9%). By R. eutropha H16G, one molecule of glucose is converted to P(3HB) monomer [( R )-3HB-CoA] along with two molecules of CO 2 and surplus energy and reducing equivalents via the ED pathway, as shown as the left-hand equation in Fig. 2 . The conversion of CO 2 to P(3HB) monomer by the combination of the CBB cycle and P(3HB) biosynthesis is shown as the right-hand equation in Fig. 2 . On the assumption that NADH is equivalent to NADPH and alternatively acted as an electron donor for generation of 2.5 ATP through aerobic respiration (P/O ratio = 2.5 27 ), it was estimated that the energy and reducing equivalents released during conventional P(3HB) synthesis from glucose correspond to those essential for fixation of 0.93 CO 2 into P(3HB) ( Fig. 3 ). Thus, when the CBB cycle is functional during heterotrophic P(3HB) biosynthesis not associated with cell growth, 0.23 molecules of 3HB monomer are expected to be additionally obtained from the fixed 0.93 molecules of CO 2 (46.5% recovery from 2 molecules of CO 2 emitted from 1 molecule of glucose), leading to increase of the P(3HB) yield to 123%. This calculation agreed well with the finding that R. eutropha H16G produced P(3HB) in 117–122% yield relative to H16G∆ cbbR and H16G∆∆ cbbLS ( Table 1 ). Apparently, the active CBB cycle under heterotrophic conditions was an advantage in P(3HB) production for R. eutropha . However, the net increase of 13 C abundance in P(3HB) synthesized from [1- 13 C 1 ]-glucose by H16G was 4.5% of the natural abundance (1.1%), which was lower than the 7.8% simply estimated from the fixation of 0.93 CO 2 in the two CO 2 molecules derived from [1- 13 C 1 ]-glucose. We initially supposed that this discrepancy was due to stable carbon isotope discrimination by Rubisco, but this reason is unlikely because the ratio of reaction rate toward 12 CO 2 to that toward 13 CO 2 ( ε -values) for Rubisco from R. eutropha was only 1.9% by in vitro assay 28 . Considering that the yields of P(3HB) by the R. eutropha strains were lower than the theoretical maximum yield, some acetyl-CoA molecules were inferred to have been completely degraded to CO 2 by the TCA cycle to obtain energy under aerobic conditions for maintaining various cellular functions. This process would reduce the abundance of 13 CO 2 within cells, reflecting the lower-than-expected 13 C abundance in P(3HB). The turnover of the TCA cycle in the P(3HB) accumulation phase not associated with cell growth was supported by the increase of the 13 C-labelled isotopomers of intermediate acids in the TCA cycle in H16G ( Fig. 2 ). Recently, Guadalupe-Medina et al . have reported that functional expression of type II Rubisco and phosphoribulokinase in Saccharomyces cerevisiae established a bypass for glucose degradation not producing excess NADH and resulting in reduced glycerol formation during bioethanol production under anaerobic conditions 29 . In the present study, we found that the CBB cycle in R. eutropha plays a role in fixation of CO 2 emitted by oxidative decarboxylation during sugar degradation and reutilized the fixed CO 2 as a source of P(3HB). It should be further noted that this novel function of CBB cycle was expressed under aerobic heterotrophic conditions, differently from those of the CBB cycle in purple non-sulphur bacteria under anaerobic photoheterotrophic conditions and chemoautotrophic bacteria under mixotrophic conditions. There are expected to be three important factors for this: heterotrophic derepression of the CBB cycle by an unique intercellular PEP sensor CbbR 8 at a probable low intracellular concentration of PEP attributed to P(3HB) formation, a high ratio of carboxylase activity to oxygenase activity (τ value) of 75 for the red-type Rubisco from R. eutropha 30 and simple modulation of Rubisco activity by a AAA + protein CbbX specific for red-type Rubiscos 31 . The more accumulation of the storage compound may be beneficial for survival in natural habitats. Moreover, the present results suggested that the CBB cycle, under heterotrophic conditions, could raise base yields of useful compounds by reutilization of the carbon atom emitted from carbon sources, when the reducing equivalents obtained by heterotrophic metabolisms were greater than those required for biosynthesis of the end products. Although the generation and consumption of reducing equivalents were equally balanced in typical anaerobic fermentation such as production of ethanol and 1-butanol, surplus reducing equivalents are available in biosynthesis of some bioproducts; for example, P(3HB), optically active 3-hydroxybutyrate and 2-propanol. The functional integration of carbon-fixing enzyme/pathways into the metabolic networks of industrial microorganisms may be a useful strategy for avoiding the loss of biomass-derived carbons in such cases."
} | 2,584 |
24691165 | PMC4101014 | pmc | 2,832 | {
"abstract": "We recently evaluated the relationship between abiotic environmental stresses and lutein biosynthesis in the green microalga Dunaliella salina and suggested a rational design of stress-driven adaptive evolution experiments for carotenoids production in microalgae. Here, we summarize our recent findings regarding the biotechnological production of carotenoids from microalgae and outline emerging technology in this field. Carotenoid metabolic pathways are characterized in several representative algal species as they pave the way for biotechnology development. The adaptive evolution strategy is highlighted in connection with enhanced growth rate and carotenoid metabolism. In addition, available genetic modification tools are described, with emphasis on model species. A brief discussion on the role of lights as limiting factors in carotenoid production in microalgae is also included. Overall, our analysis suggests that light-driven metabolism and the photosynthetic efficiency of microalgae in photobioreactors are the main bottlenecks in enhancing biotechnological potential of carotenoid production from microalgae."
} | 282 |
38380934 | PMC10880515 | pmc | 2,835 | {
"abstract": "Abstract Acetogenic gas fermentation is increasingly studied as a promising technology to upcycle carbon‐rich waste gasses. Currently the product range is limited, and production yields, rates and titres for a number of interesting products do not allow for economically viable processes. By pairing process modelling and host‐agnostic metabolic modelling, we compare fermentation conditions and various products to optimise the processes. The models were then used in a simulation of an industrial‐scale bubble column reactor. We find that increased temperatures favour gas transfer rates, particularly for the valuable and limiting H 2 , while furthermore predicting an optimal feed composition of 9:1 mol H 2 to mol CO 2 . Metabolically, the increased non‐growth associated maintenance requirements of thermophiles favours the formation of catabolic products. To assess the expansion of the product portfolio beyond acetate, both a product volatility analysis and a metabolic pathway model were implemented. In‐situ recovery of volatile products is shown to be within range for acetone but challenging due to the extensive evaporation of water, while the direct production of more valuable compounds by acetogens is metabolically unfavourable compared to acetate and ethanol. We discuss alternative approaches to overcome these challenges to utilise acetogenic CO 2 fixation to produce a wider range of carbon negative chemicals.",
"introduction": "INTRODUCTION To meet ambitious climate goals set by governments and institutions across the globe, the current reduction of emitted greenhouse gases (GHG) is not deemed sufficient (Masson‐Delmotte et al., 2021 ). As a result, carbon capture technologies have gained interest and are increasingly being implemented. However, the predominant carbon capture technologies rely on costly storage of sequestered carbon dioxide (CO 2 ), in large underground reservoirs, where it remains unused (Lackner, 2003 ). To allow circular re‐use of carbon, released or sequestered CO 2 can be upgraded into more valuable biochemicals using Carbon Capture and Utilisation (CCU). CO 2 is a thermodynamically stable form of carbon, and as such, relies on energy input to be condensed into longer carbon chains, for example in CO 2 electroreduction to acetate (Zheng et al., 2022 ) or ethanol (Jouny et al., 2018 ). Naturally, microbial cells function as biological catalysts capable of performing chemical reactions with lower energy inputs through their reliance on enzymes. Photosynthesis, the most abundant biologic carbon fixation pathway, harnesses light to reduce CO 2 , an energy source that is however poorly scalable in industrial‐scale CCU (Chen et al., 2010 ). Although less abundant, various chemoautotrophic ways of life exist, which use chemicals as energy sources. Organisms harbouring such metabolism are more suitable for industrial‐scale CCU and can be used to utilise carbon from gasified biomass or industrial waste gas, for example, from the steel industry (Köpke et al., 2011 ). A main group of interest is organisms using the Wood‐Ljungdahl pathway (WLP) to reduce CO 2 or carbon monoxide (CO) to form acetyl‐CoA, with redox potential from the oxidation of CO or hydrogen (H 2 ) (Ragsdale & Pierce, 2008 ; Wieringa, 1936 ; Wood & Harris, 1951 ). The generated acetyl‐CoA can be further fermented into various products, including acetate or ethanol (Daniell et al., 2012 ). The class of bacteria that possess the WLP are acetogens, a polyphyletic group characterised by their main metabolic product: acetate. The synthetic biology era has facilitated the introduction of novel phenotypes to various microbial hosts. Recent efforts have engineered synthetic microbial cell factories to introduce CO 2 as a carbon source into model organisms such as Escherichia coli and Pichia pastoris , however, the rates that can be achieved are by far outcompeted by the natural WLP harbouring organisms (Bar‐Even et al., 2010 ; Gassler et al., 2020 ; Li et al., 2017 ; Scheffen et al., 2021 ). To harness the natural efficient CO 2 fixating capabilities, various acetogens have been engineered to produce a wider range of valuable bioproducts (Cheng et al., 2019 ; Jia et al., 2021 ; Liew et al., 2017 ). Recently, the carbon‐negative production of acetone and isopropanol by an engineered mesophilic Clostridium has been demonstrated up to pilot scale (Liew et al., 2022 ). Acetogens are a diverse group showing large phenotypic variation, including their optimal growth temperature where mesophilic organisms grow in ranges up to 45°C, and thermophilic hosts have optimum growth temperatures above this threshold. The diversity of commonly used acetogens has been reviewed in regards to their growth temperature optima, ranging from 20 to 66°C, but also other characteristics such as their prevalent metabolic substrates and products (Lee et al., 2022 ). To date, most engineered acetogens are mesophilic, where only a few examples of engineered thermophilic acetogens producing alternative biochemicals exist (Kato et al., 2021 ). Temperature is a factor that directly affects virtually all physicochemical aspects of a fermentation process, including the biology. Working under thermophilic conditions provide various advantages over mesophilic production conditions, including decreased contamination risks, lower cooling costs, higher mass transfer rates, a key factor considering the gaseous substrates (Liew et al., 2016 ), and improved growth and production rates (Krüger et al., 2018 ; Zeldes et al., 2015 ). Additionally, higher fermentation temperatures allow more efficient in situ product recovery through gas stripping, specifically for volatile products. This facilitates simple downstream processing, whilst keeping product buildup in the fermentation broth low, in turn preventing product toxicity and feedback inhibition of production pathways (Brennan et al., 2012 ; Verhoef et al., 2009 ). As a result of these advantages, previously the economic feasibility of producing a volatile organic compound (VOC), acetone, from a theoretical thermophilic production strain was shown (Redl, Diender, et al., 2017 ). In the present study, we investigate the effect of process temperature on industrial acetogenic CO 2 fixation by using biological and mass transfer models, focusing especially on the effects of temperature on metabolism, gas uptake rates and gas stripping effectivity. These models are used as inputs for a bioprocess design and included in a final computational simulation of the chosen production process, as illustrated in Figure 1 , thereby highlighting the importance of tailoring biological parameters to process and product requirements. FIGURE 1 Overview of the models used in this work. Starting with a choice of reactor (A), mass transfer for both the gas substrate uptake rates (B) and the volatile product recovery are modelled (C). For a chosen production organism (D), a model based on general microbial growth, overall stoichiometry and reaction thermodynamics is built (E), complemented with a more detailed stoichiometric pathway analysis (F). These models are used to run a process simulation (G) of which the learnings (H) are used to reevaluate the initial choices. In bioprocess engineering, the choice of the host organism is often the first decision to be made. However, this choice is often limited to a few model organisms, or even more restrictive, to the specific organism that researchers have predominant experience with. While engineering might increase the performance of the specific organism under the desired conditions, better characteristics may be achievable by choosing a starting organism that is innately more suited to the envisioned product and hence required process. To capture this ambiguous starting point, and let process requirements dictate ensuing host selection, a host‐agnostic bacterial model of microbial growth dynamics can be used, based only on universal overall reaction stoichiometry and general microbial growth properties (Pirt, 1965 ). As summarised in Figure 1 , we start with assessing the substrate solubilisation by modelling the gas–liquid mass transfer rates through two‐film mass transfer theory. We then investigate the effect of temperature on acetogenic growth using temperature dependent, host‐agnostic black box biological models of acetogenic yields. These two models are then used in a process simulation for acetate production in an industrial‐scale Bubble Column Reactor (BCR). To test the possibility of expanding the product portfolio beyond acetate, we model acetogenic metabolism in more detail using a stoichiometric model of the WLP, thereby comparing the energetic balance of different possible products. A comparison of vapour pressures for a range of VOCs tests the feasibility of in‐situ product recovery.",
"discussion": "DISCUSSION Microbial gas fermentation is an efficient platform that enables the conversion of CO 2 ‐rich waste gas streams into biomass and organic carbon compounds, which can be used as sustainable products. To harness this power, optimisation of both the production strain and process are critical to achieve sufficiently high efficiencies and productivities, to outcompete petrochemical production in the economic‐incentive‐driven society. Applying host‐agnostic mechanistic models allows for tailored selection of microbial hosts based on final process requirements. Particularly advantageous to using host‐agnostic black box modelling is the low dependency on large a priori experimental datasets, which are often not available when pushing the boundaries of novel production processes and microbial phenotypes. In this way, a more efficient allocation of resources can be achieved when process and metabolic modelling are used to narrow the solution space that must be experimentally probed. Requiring limited data input, this approach offers a modular platform to explore novel bio‐manufacturing possibilities, moving the field away from traditional production organisms and processes. While the presented findings are purely based on modelling results, all modelling in this work builds on well‐established concepts that have been meticulously documented in literature. Indeed, the gas transfer rates and vapour pressure calculations are basic concepts that are commonly used in process engineering, as referenced in the methods section. Similarly, the stoichiometric model of acetogenic metabolism is the product of meticulous research of acetogens and their Wood‐Ljungdahl pathway. While the temperature correction of the maximum growth rate and the NGAM has not been used in the field of process design, the underlying phenomenon is widely accepted in other fields including ecology and food preservation, where NGAM would be described as the death rate. Although such modelling approaches are advantageous to limit the experimental test space, they can fall short of capturing intricate processes, either because they are a simplification of complex systems or because certain factors are unknown. The sole purpose of this analysis is to qualitatively compare the effect of temperature on gas fermentations and investigate which products can be formed. Therefore, looking forward, critical biological and experimental aspects presented here should be considered and tested in future work, to allow for accurate simulations. From a process engineering perspective the effect of media components and products in the bulk on the gas transfer rates and evaporation rates must be determined. From a biological perspective, the actual biological rates of growth and substrate uptake need to be measured, as well as the maintenance energy requirements and the inhibition of the various compounds. Nonetheless, this modelling approach has been demonstrated here to be a valuable tool to narrow down the experimental test space and could be applied to query other unconventional types of metabolism, such as the many other types of anaerobic respiration (Lovley & Coates, 2000 ). We show that for catabolic products, product yields increase with temperature due to the higher demand for maintenance energy. Specifically, for acetogenesis, metabolic modelling performed here shows that the energetic feasibility of acetogenesis depends solely on the downstream metabolism of product formation, to compensate the energy‐demanding WLP, with acetate being the most energetically favourable product whilst also utilising the most carbon per electron donor. Ethanol is also a suitable product, although it requires more H 2 because it is more reduced. However, producing ethanol comes with major advantages compared to acetate, which make it an interesting product nonetheless. As demonstrated in the volatile product recovery calculations, ethanol can more easily be recovered from the broth. Additionally, acetate production continuously acidifies the broth, requiring pH adjustment leading to salt formation, which is not a problem in ethanol formation. When other products than acetate or ethanol are desired, lower yields would be achieved, as by‐product formation is required for energy generation. This is often what is observed in gas fermenting organisms (Bengelsdorf et al., 2016 ; Jia et al., 2021 ), resulting in more complex downstream processing and overall lower titres. While most claims for the benefits of thermophilic processes have been put to the test in the presented analyses, the contamination risk and lower cooling cost have not been addressed. Indeed the surrounding mesophilic environment of the reactor is a barrier for the simple reason that it contains few thermophilic contaminants, however, contaminants can also come from the feed, which, depending on the origin of the gas can also be a thermophilic environment. Moreover, contaminants become increasingly an issue as they outperform the production strain, one could argue in the case of acetogens that their main contaminant, methanogens, should outperform them at higher temperatures because of the Gibbs free energy of reaction becoming increasingly favourable with temperature (Figure S1A ). Additionally, for production hosts producing a compound that is not their favoured catabolic product (e.g. all products besides acetate and ethanol here), the “contaminant” could come from the inoculum itself, through cells that loose the production pathway under selective pressure. Since thermophiles have an increased energy requirement of NGAM, this selective pressure is increased. Therefore, although there is a significant benefit of operating at higher temperature, there is still a risk of contamination. Regarding the reactor cooling cost, the overall chemical reactions remain exogenic, meaning that the same amount of heat needs to be removed from the reactor. While there might be more evaporation from the broth cooling the reactor, the main advantage is that the temperature difference between a thermophilic fermentation and its surroundings is such that water at room temperature cooler can be used, while a mesophilic fermentation would require the cooling liquid to be cooled, thereby making the cooling significantly more costly. Using mass transfer models and process simulations, we have investigated improved operating conditions for a BCR. We demonstrate in this study that higher temperatures are beneficial for gas uptake rates, essential for substrate availability in acetogenesis, while to date predominant industrial work with gas fermentation is performed in the mesophilic temperature range (Vees et al., 2020 ). While BCRs are a good starting point for gas fermentations due to their simplicity, scalability and low maintenance (Noorman & Heijnen, 2017 ), more complex and costly reactor types can also be envisioned if their properties address the shortcomings of BCRs. To our knowledge, there is currently no quantitative comparison of reactor performance between setups for the specific purpose of acetogenic gas fermentation. Such a study could provide valuable insights and further improvements into gas fermentation‐mediated CO 2 upcycling. Mass transfer models based on experimental data of continuous stirred tank (CSTR) (Liu et al., 2019 ), packed bed (PBR) (Steger et al., 2022 ) or external loop airlift (ELAR) (Puiman et al., 2022 ) reactors have been published. Especially the latter two reactor types are worth considering because of their potential to improve gas transfers (Nikakhtari & Hill, 2005 ). The packing of PBRs holds up and breaks up gas bubbles increasing gas mass transfer rates through both an increased total gas phase volume and a larger surface area (Nikakhtari & Hill, 2005 ). The packing can also serve as a support for immobilised cells, allowing for higher cell concentration and retaining the cells in the reactor, which allows for medium exchange and eases downstream processing (Pörtner & Faschian, 2019 ). ELARs are designed to have two columns, a riser column that is pushed up by the gas feed, and a downcomer recirculating the broth. With this design, the gas flow is used to circulate broth through the reactor, with each column functioning as a plug‐flow reactor. The broth entering the riser, having the lowest dissolved gas concentration, is exposed to the highest partial gas pressures, making for increased transfer rates. While the BCR can only be optimised by changing the reactor size, height‐to‐width ratio, gas flow and bubble size, PBRs and ELARs have more parameters that can be optimised, such as the packing type and pore size, or the riser to downcomer size ratio. Additionally, aspects from each reactor can be combined, for example, an ELAR can be designed with a packed bed and stirring. While these comparisons are not included in the present work and future work could focus on quantifying these effects and optimising the process, the main findings of this paper can be generalised to gas fermentation in any of these reactor designs. Indeed, with solely gaseous substrates, gas transfer is expected to be limiting regardless of the reactor type. Considering that the main difference between the reactor setups from a mass transfer perspective is the value of k \n L \n a (Doran, 2013 ), the findings of this research can also be applied to other reactor types. Gas transfer rates are only limiting as long as the fermentation is substrate limiting. If, through product accumulation, growth becomes product inhibited, product removal becomes the bottleneck. Then, in situ product removal can greatly benefit specific productivities. However, our calculations predict that product removal is only within reach for acetone and comes with the challenge of predicted dilute concentrations in the recovered gas, due to the large amounts of water evaporating. Other studies have also highlighted this challenge for other types of fermentation processes such as ABE fermentation (Díez‐Antolínez et al., 2018 ; Li et al., 2016 ; Liao et al., 2014 ; Rochón et al., 2017 ; Schiel‐Bengelsdorf & Dürre, 2012 ; Wen et al., 2018 ; Xue et al., 2016 ). When comparing the thermophilic gas fermentation to these mainly mesophilic ABE fermentations, the former comes with the advantage of gas feed already serving as the stripping gas, while the ABE fermentation uses a dissolved substrate, meaning that a separate stripping gas needs to be used, adding to the process complexity and operation cost. Additional processing steps can help the purification of the volatile product. The simplest addition is a second distillation after condensation of the off‐gas, possibly using high osmolarity to salt out the product (Wen et al., 2018 ). This is however purely downstream and does not affect the product concentration in the broth. If experimental work would show that the concentration needs to be pushed even lower, an option would be to increase the gas flow by adding a stripping gas. Increasing the overall gas flow in a BCR can break the flow of the air bubbles and cause a relative decrease of k \n L \n a (Heijnen & Van't Riet, 1984 ), therefore the feed rate can be kept the same while changing the gas composition to include N 2 or more CO 2 . More elaborate in situ product removal methods that do not rely on evaporation can also be considered, such as the use of membranes (Gössi et al., 2020 ) or polymeric resins (Nielsen & Prather, 2009 ) that separate the product based on other physiochemical properties, such as polarity. These are much more complex and come with a whole new set of challenges, especially when used in the relatively dirty environment of a fermentation. Overall, a BCR enables high gas absorption rates with minimal operation complexity and costs. Increased operation temperature favour both gaseous substrate uptake and formation rates and yields of catabolic products. Energy generation in acetogenic gas fermentation relies on product formation, with acetate being by far the favoured product. However, acetate is a product of limited value that cannot be recovered in situ by gas stripping. To reconcile this with the found benefits of a thermophilic gas fermentation in a BCR, several strategies can be considered. A co‐substrate could be added to expand the product range. Other products could be formed by acetogenesis despite their yields being less suitable, most likely producing a mixture of compounds. Acetate could be used as an intermediate metabolite for further conversion into more valuable compounds, either by chemical catalysis or by a second fermentation step by a heterotrophic production host. Taken together, this work highlights how the interplay of process and metabolic modelling can be used to steer bioprocess design to harness novel phenotypes and to allow probing of process parameters such as temperature and feed composition. Starting from the final production aim, we work our way back to choose suitable reactor operation conditions and a fitting production host. Specifically, the temperature‐dependent, host‐agnostic growth model is a powerful novel tool to assess and compare possible production hosts when the reactor conditions are not yet known. Combined with mass transfer models and metabolic pathway analysis into a process simulation, this approach has allowed to quantitively compare fermentation conditions and narrow down the range of process possibilities. Here, the study is specifically applied to microbial CO 2 fixation, aiming to assess the technology and push it further with the aim of providing a new tool to battle one of the most pressing global issues. Overall, this pushes the paradigm to rely less on conventional production organisms and organic carbon substrates, highlighting the possibility of efficiently converting gaseous waste streams into valuable chemicals, needed to achieve the ambitious climate goals society has set. Nomenclature \na\n area [m 2 ] \nc\n concentration ( c *: saturation c , c \n P * limiting inhibitory c ) [mM] \nD\n diffusion coefficient [m 2 /s] \nF\n flow rate [m 3 /h] \ng\n acceleration of gravity [m/s 2 ] \nH\n Henry's law constant [mol/m 3 *Pa] \nh\n height [m] \nk\n coefficient or constant, defined per case \n m \n GAM \n GAM energy requirement [kJ/mol biomass ] \n m \n NGAM \n NGAM energy requirement [kJ/h/mol biomass ] \nP\n (vapour) pressure (P m : mean logarithmic, P b : bottom, P t : top, P*: partial) [Pa] \nq\n specific consumption or production rate [h −1 ] \nR\n gas constant [m 3 *Pa/K/mol] or R Transfer rate [mol/m 3 /s] \nr\n radius [m] \nT\n temperature [K] \nV\n volume [m 3 ] \n v \n c \n gs \n superficial gas velocity [m 3 /h] \nv\n stoichiometric factor [unitless] \nx\n mol fraction [mol/mol] \nY\n yield (PS: product on substrate, XS: biomass on substrate, …) [(C)mol/(C)mol] Δ H \n r/f/vap \n enthalpy of reaction/formation/vaporisation [kJ/mol] Δ G \n \n r / f \n \n Gibbs free energy of reaction/formation [kJ/mol] \nε\n gas holdup fraction [unitless] \nθ\n temperature correction factor of k L a [unitless] \nμ\n specific growth rate [/h] \nπ\n pi [unitless] \nρ\n density [kJ/m 3 ] \n Subscripts i: compound I (or inhibition for K i ) X: biomass S: substrate P: Product max: maximum ana: anabolic reaction cata: catabolic reaction G: in the gas phase L: in the liquid bulk in: incoming (from feed flow) T: at temperature T 0: at a specified reference condition"
} | 6,084 |
26025340 | null | s2 | 2,836 | {
"abstract": "Self-assembled monolayers (SAMs) modified gold anodes are used in single chamber microbial fuel cells for organic removal and electricity generation. Hydrophilic (N(CH3)3(+), OH, COOH) and hydrophobic (CH3) SAMs are examined for their effect on bacterial attachment, current and power output. The different substratum chemistry affects the community composition of the electrochemically active biofilm formed and thus the current and power output. Of the four SAM-modified anodes tested, N(CH3)3(+) results in the shortest start up time (15 days), highest current achieved (225 μA cm(-2)) and highest MFC power density (40 μW cm(-2)), followed by COOH (150 μA cm(-2) and 37 μW cm(-2)) and OH (83 μA cm(-2) and 27 μW cm(-2)) SAMs. Hydrophobic SAM decreases electrochemically active bacteria attachment and anode performance in comparison to hydrophilic SAMs (CH3 modified anodes 7 μA cm(-2) anodic current and 1.2 μW cm(-2) MFC's power density). A consortium of Clostridia and δ-Proteobacteria is found on all the anode surfaces, suggesting a synergistic cooperation under anodic conditions."
} | 272 |
33846480 | PMC8041852 | pmc | 2,838 | {
"abstract": "The microbial electrolysis cell assisted anaerobic digestion holds great promises over conventional anaerobic digestion. This article reports an experimental investigation of extracellular polymeric substances (EPS), reactive oxygen species (ROS), and the expression of genes associated with extracellular electron transfer (EET) in methanogenic biocathodes. The MEC-AD systems were examined using two cathode materials: carbon fibers and stainless-steel mesh. A higher abundance of hydrogenotrophic Methanobacterium sp. and homoacetogenic Acetobacterium sp. appeared to play a major role in superior methanogenesis from stainless steel biocathode than carbon fibers. Moreover, the higher secretion of EPS accompanied by the lower ROS level in stainless steel biocathode indicated that higher EPS perhaps protected cells from harsh metabolic conditions (possibly unfavorable local pH) induced by faster catalysis of hydrogen evolution reaction. In contrast, EET-associated gene expression patterns were comparable in both biocathodes. Thus, these results indicated hydrogenotrophic methanogenesis is the key mechanism, while cathodic EET has a trivial role in distinguishing performances between two cathode electrodes. These results provide new insights into the efficient methanogenic biocathode development.",
"introduction": "Introduction The concept of electro-methanogenesis by combining the microbial electrolysis cell (MEC) and anaerobic digestion (AD) has become a promising method for process intensification and improving the stability of digesters 1 – 5 . The integrated process is called microbial electrolysis cell assisted anaerobic digester (MEC-AD). In MEC-AD systems, methane can be produced via multiple pathways, such as (1) direct electron transfer from the cathode to electrotrophic methanogens coupled with CO 2 reduction to methane, and (2) hydrogenotrophic methanogenesis of H 2 produced via cathodic hydrogen evolution reaction (HER) 4 – 7 . Moreover, methane can also be produced via direct interspecies electron transfer (DIET) between electroactive bacteria (EAB) and electrotrophic methanogens in cathode and anode electrodes 7 – 9 . Nonetheless, a considerable portion of methane would still be generated via conventional acetoclastic and hydrogenotrophic methanogenesis pathways.\n The activity of anodic EAB was identified as one of the key factors for boosting the methanogenesis process in MEC-AD systems. EAB can outcompete acetoclastic methanogens due to faster growth kinetics 10 , and divert electrons from acetate to anode via extracellular electron transport (EET). The transferred electrons can be utilized for hydrogen production via a cathodic HER. Thus, fast-growing hydrogenotrophic archaea can be augmented on the biocathode. Several studies reported enrichment of known hydrogenotrophic methanogens, such as Methanobacterium and Methanobrevibacter in the biocathode 8 – 11 . Thus, MEC-AD can provide faster methanogenesis rates compared to conventional anaerobic digesters. Furthermore, MEC-AD systems could provide better process stability due to the faster utilization of volatile fatty acids (VFAs) by EAB 10 , 12 . The accumulation of VFAs has been widely reported as a critical factor influencing failure or process instability of digesters operated at high organic loading rates 13 – 15 . A few studies also suggested that MEC-AD systems could provide better resilience to inhibitory compounds (e.g., phenol, ammonia, etc.) and decline of digester performance at lower temperatures 16 , 17 . Thus, MEC-AD systems can provide numerous benefits over conventional AD. Despite significant research efforts towards developing MEC-AD systems, studies exploring the significance of extracellular polymeric substance (EPS) in biocathode are limited. Biofilm EPS can have many functions, including attachment of cells to solid surfaces, maturation of biofilm structures, and protection of cells from harsh environmental conditions 18 – 20 . A few recent studies validated the significance of EPS in EET within electroactive anode biofilms 20 – 23 . In general, EPSs are composed of proteins, extracellular DNA (eDNA), humic acids, polysaccharides, etc., that are secreted by microbes in pure and mixed cultures 19 , 20 . Notably, humic acids, eDNA, and heme-binding proteins showed redox properties, serving as immobilized electron carriers in electroactive biofilms 20 , 22 , 23 . Interestingly, EPS extracted from anaerobic digesters also exhibited redox properties and identified as an essential route for DIET between syntrophic bacteria and methanogens 24 , 25 . As direct electron transport from the cathode-to-methanogen and bacteria-to-methanogens can promote electro-methanogenesis in the biocathode, it can be assumed that biocathode EPS can potentially be linked with MEC-AD performance. However, to the best of the authors’ knowledge, reports on biocathode EPS characteristics and expressions of EET genes in MEC-AD systems are still scarce. The optimization of applied voltage/potential and inoculation method has been broadly investigated to enrich a syntrophically balanced microbiome for MEC-AD systems 2 , 26 . Previous studies also substantiated the importance of persuasive system design 27 – 29 . Particularly, cathode materials with low overpotential, large surface area, and good conductive properties were found to play a deterministic role in MEC-AD performance 3 , 30 – 32 . Carbon-based electrodes, such as carbon fiber, carbon cloth, and carbon brush, have been mostly employed in previous studies due to their high surface area and biocompatibility properties 3 , 30 , 32 . Furthermore, low-cost 3D porous carbon-based composite materials have been developed for the efficient growth of biofilms 33 – 35 . However, carbon-based electrodes provide slow catalysis for cathodic HER, which seems to be critical for enriching hydrogenotrophic methanogens 1 , 30 , 36 . Some previous studies employed metal catalysts (e.g., nickel, platinum, etc.) on carbon electrodes to accelerate HER 1 , 30 , 36 , while these catalysts are still expensive. In contrast, non-precious metal electrodes, such as stainless steel, have shown an excellent low-cost alternative 11 , 31 , 37 – 39 . However, to date, limited information is available on how carbon and metal-based electrodes shape biocathode structures in terms of EPS, expression of EET genes, and microbial communities. Considering the research gaps mentioned above, the main goal of this study was to provide fundamental insights into the EPS characteristics and EET genes in methanogenic biocathode. The novelty of this study is two folds. First, this study presents, for the first time, a comprehensive characterization and significance of EPS and expression of EET genes for methanogenic biocathode. Second, underlying mechanisms of methanogenesis performance with carbon and metal cathodes were evaluated with a multifaceted approach combining molecular biology, microscopic and electrochemical tools.",
"discussion": "Results and discussion MEC-AD performance The performance of the two configurations was compared based on volumetric current density and methane productivity. As shown in Fig. S3 , the maximum current density from the CF–SS reactor reached 34.1 ± 0.3 A/m 3 , which was significantly higher ( p = 0.01) than CF–CF (27.6 ± 0.2 A/m 3 ). Although the methane generation patterns were comparable in both reactors (Fig. 1 ), CF–SS showed higher ( p = 0.03) daily methane production than CF–CF throughout the batch cycle. The total cumulative methane production was substantially higher in CF–SS (179.5 ± 6.7 vs. 100.3 ± 7.9 mL CH 4 ; p = 0.01). Both reactors used carbon fiber as the anode electrode and were operated under identical operating conditions (e.g., mixing speed, substrate, inoculation, etc.). Hence, the differences in system performance could be closely tied to the difference in the cathode electrode. As discussed later, stainless steel mesh cathode in CF–SS facilitated denser biofilms formation with more methanogenic biomass. Figure 1 Methane production from CF–CF and CF–SS reactors. The error bars indicate the standard deviation of three replicates (n = 3). Anode electrodes providing high specific surface areas have been efficient for enhancing the performance of various bio-electrochemical systems 3 , 28 , 30 – 32 . Therefore, carbon-based electrodes, such as carbon brush, activated carbon, have also been widely used for various biocathode applications, including electro-methanogenesis 3 , 30 , 32 . Notably, the rough surface of carbon fiber was found to be efficient for developing EAB biofilms 58 . Although we cannot rule out that different textures (diameter of carbon fiber and stainless steel wire) could also lead to distinct colonization of biomass 59 , 60 , the specific surface area provided by electrodes is often considered a critical factor. This study shows that stainless steel cathode having a relatively lower specific surface area than carbon fibers (4.23 vs. 3609 m 2 /m 3 ) resulted in a superior methanogenic activity. It has been previously suggested that the agglomeration of fibers in the liquid phase could reduce the available specific surface area for biofilms formation 61 . Nonetheless, considering all carbon fiber filaments in a bundle as a single fiber, the specific surface area provided by the carbon fiber was still higher than stainless-steel (4.23 vs. 41 m 2 /m 3 ). In general, carbon-based electrodes are considered inferior catalysts for HER than metal and carbon–metal composite electrodes 11 , 31 , 37 – 39 . Previous MEC-AD studies substantiated the role of hydrogenotrophic methanogenesis. It is also reasonable that acetoclastic methanogens would likely be washed out at low residence time (< 7 days) used in this study 7 , 62 . EIS analysis also indicated that stainless-steel biocathode could reduce various intrinsic internal resistances in CF–SS compared to CF–CF (see Supporting Information). As shown in the Nyquist plot (see Fig. S4 ), the overall internal impedance of CF–SS (52.62 Ω) was lower than that of CF–CF (77.28 Ω). Thus, stainless-steel cathode largely influenced the internal resistances, which influenced the HER kinetics and, ultimately, growth and activities of hydrogenotrophic methanogens. Previous studies also suggested that lower ohmic resistance in MECs could provide faster HER kinetics 63 , 64 . Thus, the inferior methane recovery from the CF–CF reactor than the CF–SS reactor was likely due to the inferior HER on carbon fibers and subsequent hydrogenotrophic methanogenesis. Organics removal and VFAs profiles The effluent COD concentration from CF–SS (215 ± 2.8 mg/L) was considerably lower ( p = 0.001) than that of CF–CF (382 ± 3.0 mg/L) (Fig. S5 a). Correspondingly, COD removal efficiency in CF–SS (89.9 ± 0.5%) was significantly higher ( p = 0.012) than that of CF–CF (83 ± 1.7%). Fig. S5 b, c show the VFAs profiles during batch operation. For both reactors, the acetate concentrations were relatively higher than propionate and butyrate throughout the operational period. The CF–SS reactor showed the highest acetate concentration of 439 ± 2 mg COD/L, while propionate (94 ± 0.1 mg COD/L) and butyrate (61 ± 0.3 mg COD/L) concentrations were relatively lower. In contrast, CF–CF exhibited the highest acetate concentration of 320 ± 0.4 mg COD/L, which was lower than that observed in CF–SS. Propionate concentrations were relatively higher in CF–CF, with the highest concentration of 118 ± 0.4 mg COD/L. The highest butyrate concentration (64.8 ± 1.3 mg COD/L) in CF–CF was comparable to CF–SS (61 ± 0.3 mg COD/L). CF–SS also showed a lower accumulation of VFAs in the final effluent than CF–CF (52.6 ± 0.5 vs. 133.5 ± 1.0 mg COD/L; p = 0.005). Throughout the batch operation, propionate concentrations in CF–SS remained relatively lower than those observed in CF–CF, indicating faster conversion of propionate in the CF–SS. The fermentation of propionate to acetate is a vital process towards anodic respiration (by EAB) and acetoclastic methanogenesis. However, propionate fermentation to acetate is energetically unfavorable in terms of Gibbs free energy 65 . Thus, maintaining lower hydrogen partial pressure would be critical for propionate fermentation to acetate. Even though stainless-steel cathode would be expected to provide superior HER than carbon fibers 11 , 38 , no hydrogen was detected in biogas from both reactors. This might be due to the rapid consumption of hydrogen by hydrogenotrophic methanogens, as suggested in previous studies 11 , 66 . Moreover, enhanced homoacetogenic activity (H 2 + CO 2 → acetate) could assist in maintaining lower hydrogen partial pressure in biocathode 11 , 66 . Microbial community analysis also coincided with these notions (discussed later). Thus, the VFA profiles suggest that the microbiome in CF–SS more rapidly utilized hydrogen produced via fermentation and cathodic HER. EPS characteristics As shown in Fig. 2 a, the EPS composition of anode biofilms in both reactors was quite similar and was not affected by the different cathode materials used. Protein was found as the major EPS component in anode biofilms, consistent with recent reports on EPS composition in pure culture Geobacter biofilms 23 , 46 . Geobacter species were also abundant in anode biofilms in both reactors in this study (discussed later). The concentrations of major EPS components (carbohydrates, proteins, and hemes) in the cathode biofilms in CF–SS were higher than those of CF–CF. Notably, carbohydrates and proteins in cathodic EPS were markedly higher in CF–SS than CF–CF (carbohydrates: 52.2 ± 0.2 vs. 25.8 ± 0.5 mg/cm 2 ; proteins: 212.8 ± 3.4 vs. 170 ± 1.3 mg/cm 2 ). The heme-binding proteins, uronic acid, and eDNA also showed the same patterns. Overall, cathodic biofilms developed on the stainless-steel electrode exhibited markedly higher EPS levels ( p = 0.03). Figure 2 EPS levels in biofilms ( a ), EPS quantitative analysis using CLSM; biovolume ( b ), and fluorescence intensity ( c ), and reactive oxygen species (ROS) intensities ( d ) of CF–CF and CF–SS reactors. Note. The error bars indicate the standard deviation of three replicates (n = 3). Moreover, electrode surfaces were visualized with CLSM (Fig. 3 ). The CLSM images showed that EPS was more uniformly distributed on the stainless-steel biocathode in CF–SS than the carbon fiber electrodes in both reactors. The biovolume of cathode biofilms in CF–SS was estimated at 30.2 ± 4.2 µm 3 /µm 2 , which was two times higher than that estimated for cathode biofilms in CF–CF (13.5 ± 2.8 µm 3 /µm 2 ) (Fig. 2 b). The biovolumes estimated for anode biofilms in both reactors were comparable ( p = 0.007). The intensities of EPS and eDNA were also quantified (Fig. 2 c). Like estimated biovolume, EPS and eDNA intensities in stainless steel biocathode were higher than those estimated for carbon fiber biocathode ( p = 0.008). Simultaneously, EPS and eDNA intensities were comparable for anode biofilms in both reactors ( p = 0.20). Thus, CLSM imaging and COMSTAT (COMSTAT2, Version 2.1, Dk, http://www.comstat.dk/) 51 – 53 analysis further confirmed that the stainless-steel biocathode resulted in the highest EPS production. The SEM imaging of biofilms also corroborated these results (Fig. S6 ). The biofilms did not fully cover the surfaces of carbon fibers, while biofilms grown on stainless steel cathode in CF–SS were evenly denser than anode/cathode biofilms grown on carbon fiber electrodes. The anode/cathode biofilms grown on carbon fibers exhibited substantial heterogeneity. In contrast, a large secretion of EPS could accelerate the surface attachment of cells on the stainless-steel. Figure 3 Representative confocal microscopic images of EPS with 3 µm scale; anode (CF–CF) ( a ), cathode (CF–CF) ( b ), anode (CF–SS) ( c ), and cathode (CF–SS) ( d ). The green color represents eDNA and the red color indicates EPS (This figure has been analyzed using COMSTAT2, Version 2.1, Dk, http://www.comstat.dk/ ). Studies on the EPS in electroactive biofilms received less attention and primarily focused on understanding their role in anodic EET. A few reports revealed redox-active features of anodic EPS in model EAB biofilms (e.g., Geobacter sulfurreducens , Shewanella oneidensis , and Pseudomonas putida ) 20 , 23 , 46 . Notably, higher levels of proteins in anode biofilms were correlated with higher EET efficiency. In this study, despite differences in volumetric current densities, both EPS composition and concentrations were quite similar in anode biofilms in both reactors. Instead, the difference in cathodic EPS levels was likely linked to current densities and methane productivity. As mentioned earlier, EPS can serve as immobilized redox cofactors (i.e., electron carriers) for facilitating EET in anodic EAB biofilms 20 , 23 . EAB can also regulate EPS generation to balance EET and protect cells 48 . The existing literature provides limited information on the roles of EPS in methanogenic biocathode. However, a few reports suggested that EPS could play similar roles (EET and cell protection) in archaeal biofilms in conventional digesters in the presence of conductive additives 24 , 67 . Interestingly, a recent study demonstrated that the addition of iron-based conductive materials in conventional anaerobic bioreactors could enhance redox-active EPS contents in methanogenic biomass 24 , which was positively correlated with methanogenesis rates. Conductive materials promote the syntrophic DIET from bacteria to archaea and thereby enhance methanogenesis 68 . Therefore, the CV of biocathode EPS from two reactors was performed to identify their redox activity (Fig. S7 ) qualitatively. As shown in Fig. S7 , the voltammograms of cathodic EPS extracted from both reactors showed distinct redox peaks, indicating their redox capability. However, redox peaks were observed at different potentials, suggesting that redox properties would be different for EPS extracted from two biocathodes. The peak current from stainless steel biocathode EPS was considerably higher than the EPS extracted from carbon fiber biocathode. This difference could be associated with higher levels of redox-active EPS in stainless steel biocathode, as previously suggested in the literature for anodic EAB biofilms 20 , 46 . Despite higher EPS levels in stainless-steel biocathode and differences observed in CV patterns, the expressions of genes associated with EET were comparable in both biocathodes (discussed later). Thus, it can be inferred that redox activities of EPS did not play a decisive role in differentiating between the performances observed from the two systems. Instead, EPS variations might be more associated with the protection of cells from harsh metabolic environments. Nonetheless, future investigation is warranted to reach a more thorough understanding and quantitative characterization of redox properties of EPS. A recent study reported that the current from anode biofilms was positively associated with EPS protein content and negatively correlated to carbohydrates in EPS 48 . In this study, both carbohydrates and proteins in EPS were considerably higher in stainless steel biocathode than that of carbon fiber biocathode (see Fig. 2 ). The secretion of carbohydrates could be associated with harsh environmental conditions 19 , 25 to provide a protective layer and maintain the redox activity of proteins involved in EET 22 , 25 . It is possible that enhanced HER in stainless steel cathode could create highly alkaline conditions near the cathode 7 , 69 , 70 , which might induce more EPS secretion. Based on a recent report, hydrogenotrophic methanogenesis could be the dominant pathway under alkaline pH 71 . As discussed later, stainless steel biocathode also showed a higher abundance of hydrogenotrophic methanogens in this study. Thus, it appeared that higher enrichment of hydrogenotrophic methanogens promoted by faster HER kinetics on stainless steel cathode was possibly associated with higher EPS excretion. While further investigation is needed to get more insights into the function of EPS on electro-methanogenesis, these results suggested that different cathode materials could influence EPS secretion and methanogenic activity due to differences in HER kinetics. ROS levels The quantitative measurement of ROS demonstrated a significant difference between biofilms grown on stainless steel and carbon fibers (Fig. 2 d). The lowest ROS level was observed for cathode biofilms formed on stainless steel, while ROS levels were very similar in anode/cathode biofilms developed on carbon fibers. Recent studies reported ROS accumulation in anaerobic digesters 21 , 72 , while ROS is usually thought to be produced during aerobic metabolism. It has been suggested that unfavorable metabolic conditions (e.g., inhibition by toxicants, pH changes) could lead to ROS accumulation in digesters 21 , 72 . ROS accumulation may suppress metabolic activities, leading to the deterioration of digester performance. As we used synthetic glucose medium as a substrate, the potential unfavorable metabolic conditions induced by any toxic compounds can be ruled out. Thus, potential local pH changes by HER can be considered as an unfavorable metabolic condition. The HER in both biocathode can lead to alkaline pH due to protons reduction (2H + + 2e − → H 2 ), while effects will likely be more intense on stainless steel cathode 11 , 31 , 37 – 39 . Thus, the lowest ROS level in stainless steel biocathode suggests that higher EPS levels provided some degree of protection to the cathodic microbiome from potential environmental stress (e.g., local alkaline pH due to superior HER). However, potential mechanisms relating to EPS and ROS levels should be further explored. Microbial quantity and diversity Figure 4 shows the quantitative assessment of microbial communities performed with qPCR. The total microbial cell counts (16S) in anode biofilms in CF–SS were slightly higher than that of CF–CF (9 × 10 8 vs. 8 × 10 8 cells/cm 2 ) (Fig. 4 a). An almost similar pattern was observed for cathode biofilms; however, the difference was more prominent (1 × 10 11 vs. 6 × 10 8 cells/cm 2 ). The archaeal cell numbers also showed similar patterns, with the highest archaeal cell numbers for the stainless steel biocathode. Figure 4 Total cell number using 16 s and archaeal primers ( a ), and mcr A gene copies ( b ). The error bars indicate the standard deviation of three replicates (n = 3). Furthermore, mcr A gene copies were quantified (Fig. 4 b), considered a biomarker for hydrogenotrophic methanogenesis 73 . A few recent reports also confirmed the positive link between mcr A gene copies and methanogenesis rates in MEC-AD reactors 56 , 73 , 74 . The highest number of mcr A gene copies was observed for the stainless steel biocathode (4 × 10 6 cells/cm 2 ; 100 times higher than carbon fiber biocathode). The mcr A gene copies in anodic biofilms for both systems were comparable. Thus, the higher abundance of mcr A gene copies within the stainless steel biocathode corroborated with higher methane productivity in the CF–SS reactor. The alpha diversity of microbial communities was also estimated (Table S3 ). The higher values of Chao 1, phylogenetic distance, OTUs, Pielou's evenness, and Shannon index clearly showed that the richness and diversity indices were relatively higher in CF–SS than CF–CF. Notably, cathode biofilms in CF–SS showed more diversity with the Shannon index of 5.10, as compared to CF–CF (3.95). These results indicated that the stainless-steel electrode persuaded the richness and diversity of the microbial communities. Microbial community composition, and gene expression 16S rRNA sequencing Microbial communities in two reactors were analyzed with specific bacterial, archaeal, and mcr A primers. Proteobacteria was the most abundant phylum in anode biofilms in both reactors; however, its relative abundance was much higher in CF–SS (85%) than CF–CF (47%) (Fig. S8 ). The relative abundances of Bacteroidetes (26%) and Firmicutes (14%) in CF–CF were considerably higher than CF–SS (6% and 4%, respectively). Also, Synergistetes (6%) and Lentisophaerae (4%) were present at slightly higher abundances in CF–CF, while in CF–SS, they were 1% and 3%, respectively. Proteobacteria was also the most abundant in both cathode biofilms; their relative abundances (64–68%) were also similar. However, the abundance of Bacteroidetes was higher in CF–CF (17%) than CF–SS (6%). On the contrary, the phylum Firmicutes (20%) was the second most abundant in CF–SS, while its abundance in CF–CF was considerably lower (9%). At the genus level, Geobacter, belong to Proteobacteria, was the most abundant in anode biofilms (CF–CF: 22%; CF–SS: 59%) in both systems (Fig. 5 a). Geobacter is a highly efficient EAB with the capability to facilitate EET from simple organic acids like acetate 42 , 48 . In CF–CF, Bacteroides was the second most dominant genus (12%), followed by Enterobacteriaceae (10%) and Dysgonomonas (5%). In contrast, the second abundant genus in CF–SS was Enterobacteriaceae (23%), followed by Dysgonomonas (3%) and Victivallis (3%). Figure 5 Relative abundance of microbial communities analyzed with bacterial primer ( a ), archaeal primer ( b ), and mcr A primer ( c ) at the genus level. The cathode biofilms in both reactors were dominated by the genus Enterobacteriaceae (CF–CF: 42%; CF–SS: 60%). In CF–CF, Bacteroides (12%), Pleomorphomonas (9%), and Desulfovibrio (4%) were the other dominant genera. In contrast, Acetobacterium was the second abundant genus (16%), followed by Bacteroides (5%), Dysgonomonas (3%), and Desulfovibrio (2%) in CF–SS. Acetobacterium , known homoacetogenic bacteria, can utilize H 2 and CO 2 to produce acetate 31 , 75 . Then, acetate can be consumed by either acetoclastic archaea or EAB 7 , 10 , 76 . The enrichment of Acetobacterium on the stainless-steel biocathode indicates the occurrence of higher catalysis of HER. As mentioned earlier, H 2 gas has not been observed in the biogas samples. This might be due to the rapid utilization of the generated H 2 via hydrogenotrophic methanogens and homoacetogenic Acetobacterium , as suggested in the literature 11 , 66 . The presence of the highest acetate concentration (439 ± 2 mg COD/L) in CF–SS corroborated with a higher abundance of Acetobacterium . Moreover, Acetobacterium can maintain a lower hydrogen partial pressure to provide thermodynamically favorable conditions for propionate and butyrate fermentation to acetate. This notion is also supported in part by the lower propionate concentrations in CF–SS compared to the CF–CF. Archaeal and mcr A primer sequencing For the archaeal phylum, relative abundances of Euryarchaeota were 32% and 51% in CF–CF and CF–SS, respectively (Fig. S8 ). At the genus level, the abundance of Methanobacterium was almost similar (13–14%) in the anode biofilms in both systems (Fig. 5 b). However, the abundance of Methanobacterium in cathode biofilms of CF–SS was higher than CF–CF (51% vs. 32%). Previous studies also reported the enrichment of known hydrogenotrophic methanogens in methanogenic biocathode 8 – 11 . Moreover, mcr A gene sequencing was performed (Fig. 5 c) to understand the taxonomy of methanogens 56 , 73 , 74 . In the anodic biofilms, the abundances of Methanobacterium species were almost similar, including formicicum (CF–CF: 67%; CF–SS: 71%) and subterraneum (CF–CF: 34%; CF–SS: 29%). In the cathodic biofilms, CF–SS showed more diverse species of Methanobacterium ; formicicum (54%), subterraneum (24.4%), and palustre (22%), as compared to CF–CF; formicicum (81%), and subterraneum (19%). Thus, the higher abundance and diverse species of hydrogenotrophic Methanobacterium on stainless steel cathode might have contributed to the faster methanogenesis via hydrogen utilization. Principal component analysis The PCA analysis of biocathode bacterial and archaeal communities was performed to evaluate the relation between genera and PCs (Fig. S9 ). Based on 16S rRNA bacterial sequencing of biocathode, the superior performance of CF–SS was related to the enrichment of homoacetogenic Acetobacterium (Fig. S9 a). However, the other genera might have an indirect relation to the superior performance of CF–SS. Based on archaeal sequencing of biocathode, hydrogen-consuming Methanobacterium and Acetobacterium primarily contributed to the superior performance of stainless steel biocathode (Fig. S9 b). Expression of EET genes The gene expression for pil A and c‐type cytochromes (Fig. 6 ) shows trivial differences in their expression levels in anode/cathode biofilms between both reactors. Moreover, compared to anode biofilms, the EET-associated genes were less expressed in cathode biofilms in both reactors. Based on the authors’ knowledge, this study first reports the expression of EET genes for methanogenic biocathode. The EET from EAB to the anode has been demonstrated to be facilitated via c-type cytochromes and conductive nanowire or pili 77 , while the significance of EET in methanogenic biocathode is still ambiguous. However, previous reports postulated that conductive pili and c-type cytochromes could play an important role in DIET from EAB to methanogens 78 , 79 . Notably, some bacteria (e.g., Enterobacteriaceae , Desulfovibrio , etc.) found in biocathode in this study could express different cytochromes and/or conductive pili 77 , 80 . Furthermore, a recent study suggested that Methanobacterium species could produce methane via DIET 81 . Nonetheless, the expressions of these EET genes were quite comparable in both systems, indicating higher current density and methane productivity from the CF–SS reactor was not attributed to the overexpression of EET genes. Figure 6 Expression of genes known to regulate extracellular electron transfer in biofilms. The error bars indicate the standard deviation of three replicates (n = 3). Implications This study provides new insights into the characteristics and significance of EPS and expressions of EET genes in methanogenic biocathode. As compared to the carbon fiber, significantly higher EPS levels were observed in the stainless steel biocathode. Protons reduction to H 2 during HER can create local alkaline pH on the cathode. Thus, it could be posited that the highest EPS secretion in stainless steel biocathode could be linked with faster HER. One important finding of this current study is that EET may not play a decisive role in differentiating performances in MEC-AD systems using different electrode materials. Instead, the effective catalysis of HER, lower internal resistance, and higher abundances of H 2 -utilizing methanogens and homoacetogens on stainless steel cathode appeared to be the primary reason behind the higher methanogenic activity. Nonetheless, based on EET gene expression patterns and redox activity of biocathode-derived EPS, EET would still be involved in cathodic electro-methanogenesis. Regarding the engineering significance of the results, carbon-based cathode electrodes have been mostly used in MEC-AD systems due to their excellent biocompatibility and higher surface area over metal-based electrodes 3 , 30 – 32 . While carbon fibers provided a higher specific surface area, stainless steel mesh outperformed carbon fibers under similar operating conditions (e.g., anode electrode, inoculum, mixing, etc.). Given that most of the single-chamber MEC-AD studies used carbon-based biocathode 5 , 7 , the results of this study are significant for selecting efficient cathode materials to realize improved performance. However, it should be noted that the results presented here are from specific operating conditions with two selected electrode materials. Hence, further research is warranted with more carbon and metal electrodes with similar textures and surface areas."
} | 8,023 |
25676143 | null | s2 | 2,839 | {
"abstract": "Inspired by how geckos abduct, rotate, and adduct their setal foot toes to adhere to different surfaces, we have developed an artificial muscle material called ion-exchange polymer-metal composite (IPMC), which, as a synthetic adhesive, is capable of changing its adhesion properties. The synthetic adhesive was cast from a Si template through a sticky colloid precursor of poly(methylvinylsiloxane) (PMVS). The PMVS array of setal micropillars had a high density of pillars (3.8 × 10(3) pillars/mm(2)) with a mean diameter of 3 μm and a pore thickness of 10 μm. A graphene oxide monolayer containing Ag globular nanoparticles (GO/Ag NPs) with diameters of 5-30 nm was fabricated and doped in an ion-exchanging Nafion membrane to improve its carrier transfer, water-saving, and ion-exchange capabilities, which thus enhanced the electromechanical response of IPMC. After being attached to PMVS micropillars, IPMC was actuated by square wave inputs at 1.0, 1.5, or 2.0 V to bend back and forth, driving the micropillars to actively grip or release the surface. To determine the adhesion of the micropillars, the normal adsorption and desorption forces were measured as the IPMC drives the setal micropillars to grip and release, respectively. Adhesion results demonstrated that the normal adsorption forces were 5.54-, 14.20-, and 23.13-fold higher than the normal desorption forces under 1.0, 1.5, or 2.0 V, respectively. In addition, shear adhesion or friction increased by 98, 219, and 245%, respectively. Our new technique provides advanced design strategies for reversible gecko-inspired synthetic adhesives, which might be used for spiderman-like wall-climbing devices with unprecedented performance."
} | 426 |
22509174 | PMC3325762 | pmc | 2,840 | {
"abstract": "Biofuels are anticipated to enable a shift from fossil fuels for renewable transportation and manufacturing fuels, with biohydrogen considered attractive since it could offer the largest reduction of global carbon budgets. Currently, lignocellulosic biohydrogen production remains inefficient with pretreatments that are heavily fossil fuel-dependent. However, bacteria using alkali-treated biomass could streamline biofuel production while reducing costs and fossil fuel needs. An alkaliphilic bacterium, Halanaerobium \n hydrogeniformans , is described that is capable of biohydrogen production at levels rivaling neutrophilic strains, but at pH 11 and hypersaline conditions. H. hydrogeniformans ferments a variety of 5- and 6-carbon sugars derived from hemicellulose and cellulose including cellobiose, and forms the end products hydrogen, acetate, and formate. Further, it can also produce biohydrogen from switchgrass and straw pretreated at temperatures far lower than any previously reported and in solutions compatible with growth. Hence, this bacterium can potentially increase the efficiency and efficacy of biohydrogen production from renewable biomass resources.",
"introduction": "Introduction As the price of fossil fuels increases and reserves diminish, biofuel production is seen as a viable contribution to current as well as future energy demands. Hydrogen (H 2 ), alcohol, and hydrocarbon generation by microbial fermentation of lignocellulosic plant materials holds promise as alternatives to petroleum based fuels. H 2 has an advantage in that combustion only results in water vapor, without the generation of carbon dioxide. Although the potential exists for microbially generated fuels from biomass to be economically attractive, several issues are still to be resolved. Of particular note is the large fossil fuel-dependent energy input of the typical steam blasting pretreatment of biomass as well as detoxification of the resulting compounds that are inhibitory to fermentation, and overall low hydrogen yields. While many technical issues are still to be resolved pertaining to using H 2 as a transportation fuel, particularly devising safe storage methods, the demand for sustainable forms of H 2 is great with regard to industry. Hydrogen gas is currently produced through the steam reforming of natural gas and coal gasification, and is used in chemical production, oil refining, and steel manufacturing. The vast majority of global H 2 production is by industrial consumers, with individual H 2 production facilities meeting the demands of their manufacturing processes (Mueller-Langer et al., 2007 ). As the price of natural gas continues to increase, demand is likely to increase for an economically and environmentally sustainable means of H 2 production that can fit into the current on-site production model. The further development of biological H 2 generation derived from plant biomass could be used to meet the demands of these industries while decreasing their reliance on fossil fuels. Fibrous plant material is made up of lignocellulose and principally contains cellulose, hemicellulose, and lignin. Cellulose is comprised of linear chains of the sugar glucose and is mostly crystalline; whereas, hemicellulose is more amorphous, being made up of both pentoses and hexoses, and is more readily hydrolyzed (Updegraff, 1969 ). Lignin has a poorly defined, heterologous structure providing strength and support to plant structure (Lebo et al., 2001 ) but is recalcitrant to bacterial degradation. It is due to lignins recalcitrance that pretreatments are required to facilitate its removal. Currently, the most common pretreatment method is steam blasting that causes the solid cellulose to become separated from the aqueous hemicellulose and lignin. However, this treatment requires electricity for steam generation and is typically a natural gas- or coal-dependent step (Datar et al., 2007 ; Liu et al., 2002 ). Hence, considerable amounts of CO 2 are emitted and the production of biofuels remains fossil fuel-dependent. Under these conditions, a host of compounds that are inhibitory to fermentation are generated from lignin degradation and include weak acids, furan derivatives, and phenolic compounds (Palmqvist and Hahn-Hägerdal, 2000b ). Before microbial fermentation of hemicellulose can occur, these inhibitors must be separated and removed from the liquid fraction. Current detoxification methods include enzyme treatment, fractionation of volatile compounds, and chemical treatment. However, alkali treatment is the most effective method of precipitating toxins and destabilizing inhibitors (Palmqvist and Hahn-Hägerdal, 2000a ). Overall, the current processes of pretreatment, detoxification, and subsequent fermentation are relatively inefficient and remain fossil fuel-dependent, thus counterproductive to the goal of reducing greenhouse gas emissions. An effective alternative is alkali treatment of the lignocellulose. This pretreatment separates cellulose from the hemicellulose and lignin, reduces the crystallinity of the cellulose making it more accessible for degradation and fermentation, and limits the production of inhibitory compounds (Gáspár et al., 2007 ; Jackson, 1977 ; Spencer and Akin, 1980 ). However, the resulting substrates are highly alkaline and potentially have a high concentration of salts. Previously, alkali treatment was evaluated against several others, including steam blasting, for bioethanol production from a number of biomass substrates. The results indicated that alkali treatment was the most efficient process, producing at least 3.5 times higher ethanol yields compared to steam-blasted or acid-treated biomass (Klinke et al., 2004 ). Hence, a fermentative haloalkaliphilic bacterium with naturally high H 2 yields may decrease the number of production steps for biofuel production. The application of such an organism would eliminate the need for steam blasting of biomass. The need for removal of inhibitory compounds would also be eliminated in addition to the pH neutralization step prior to sugar fermentation. Hence, a fermentative haloalkaliphile offers a potentially more efficient and cost-effective adaptation for biohydrogen production. The current study describes the isolation and characterization of such an organism, Halanaerobium hydrogeniformans , from Soap Lake, WA along with the development of a modified, alkaline pretreatment of lignocellulosic material that foregoes many of the existing pretreatment issues. H. hydrogeniformans was tested for the capacity to grow and produce biohydrogen on several 5- and 6-carbon sugars as well as with alkaline pretreated straw and switchgrass in batch and fed-batch bioreactors to assess the feasibility of haloalkaliphilic biohydrogen production.",
"discussion": "Results and Discussion There is a wide diversity of alkaliphilic bacteria spanning several phylogenic groups, most notably from soda lakes (Duckworth et al., 1996 ; Jones et al., 1998 ; Rees et al., 2004 ); however, their application for H 2 production has not been investigated. Soap Lake, Washington, a highly haloalkaline lake, possesses a pH value of 10, 15–140 g/L NaCl, and anaerobic lake bottom sediments with extraordinarily high sulfide concentrations (Anderson, 1958 ; Mormile et al., 1999 ; Pinkart et al., 2006 ; Sorokin et al., 2007 ; Dimitriu et al., 2008 ) making it an ideal location to obtain haloalkaliphilic bacteria. Lake bottom sediments were enriched using a culturing medium (Mormile et al., 1999 ) that mirrored the geochemistry of Soap Lake. The culturing medium was modified to include either 30 mM glucose or 15 mM cellobiose as a carbon source. Subculturing the initial enrichment into the modified media resulted in the accumulation of hydrogen gas. Serial dilution plating of the hydrogen producing mixed culture led to the isolation of a pure strain of bacteria, designated as strain SL-HP. Strain SL-HP was grown in the modified media to confirm the ability to produce hydrogen from the fermentation of both glucose and cellobiose as well as characterize its basic physiology. Description of the new species Halanaerobium hydrogeniformans Strain SL-HP was found to be an obligately anaerobic, gram negative, non-motile, non-sporulating, elongated rod that is ∼18 × 1 μm (Figure 1 ). Growth was observed between 20 and 37°C, pH 7–12, and NaCl concentrations of 2.5–15% (w/v) (Table A1 in Appendix). Optimum growth conditions were 30°C, pH 11, and 7.5% NaCl. In addition, increased growth was observed when SL-HP was grown in the presence sodium sulfide, even when other sources of reductant were present. The optimum sodium sulfide concentration was found to be 0.75 g/L (data not shown). Figure 1 (A,B) Scanning electron microscopy images of H. hydrogenoformans showing the long, curved nature of the isolate. Cells pictured here were grown on 15 mM cellobiose in liquid culture under non-agitated conditions at 30°C, and pH 11 with 7% (v/v) NaCl. Cells were fixed with 2.5% (v/v) anaerobic glutaraldehyde, subjected to ethanol dehydration, critical-point drying, and coated with palladium. In the interest of pursuing this bacterium as a potential strain for industrial biohydrogen production, the ability to ferment a variety of sugars derived from cellulosic biomass was observed. SL-HP was able to ferment and produce hydrogen from glucose, cellobiose, xylose, arabinose, mannose, and galactose. Observed products of fermentation include acetate, formate, and hydrogen (Table 1 ). Due to the high concentration of carbonate and the alkaline pH of the growth medium, accumulation of CO 2 was not detected in the headspace, but is most likely a product of fermentation. Glucose, ribose, and cellobiose fermentations resulted in the largest accumulation of biomass, while cellobiose, glucose, and mannose resulted in the highest levels of hydrogen production. In addition to these pentose and hexose sugars, growth was also observed on glycerol. Growth above control levels was not observed when acetate, lactate, or ethanol was the added carbon source. Table 1 Growth, metabolism and hydrogen production during non-shaking, uncontrolled batch cultivation . Initial substrate Final Whole Cell Protein (ug/ml) a Final optical density (600 nm) a % H 2 in headspace a H 2 molar yield a Acetate (mM) a Formate (mM) a 0.1% yeast extract control 12 ± 6 0.06 ± 0.01 0.4 ± 0.5 - 4.3 ± 0.1 0.0 15 mM cellobiose 83 ± 5 0.29 ± 0.03 38.0 ± 2.0 2.3 ± 0.2 31.3 ± 0.6 7.9 ± 0.4 30 mM glucose 91 ± 2 0.33 ± 0.03 31.0 ± 2.0 1.9 ± 0.1 25.7 ± 0.6 0.0 30 mM ribose 91 ± 4 0.26 ± 0.01 13.0 ± 1.0 0.7 ± 0.1 24.0 ± 2.0 0.0 30 mM xylose 73 ± 6 0.17 ± 0.01 8.2 ± 0.5 0.5 ± 0.1 17.0 ± 1.0 0.0 30 mM arabinose 59 ± 1 0.15 ± 0.01 7.0 ± 2.0 0.4 ± 0.1 14.0 ± 3.0 0.0 30 mM galactose ND ND 16.0 ± 1.0 0.9 ± 0.1 14.7 ± 0.6 0.0 30 mM mannose ND ND 29.0 ± 5.0 1.7 ± 0.3 26.0 ± 3.0 0.0 ND: not determined, although growth occurred . a All measurements were determined at the exhaustion of the carbon and electron source. All cultivations shown used a 5 mL culture with 23 mL headspace of 80% N 2 /20% C0 2 (v/v), under non-agitated conditions with 7% (v/v) NaCl and pH 11.0 at 30°C . Phylogenetic analysis clearly placed SL-HP in the Halanaerobium genus (Figure 2 ). The strain’s 16S rRNA gene sequence was most closely related to H. saccharolyticum subsp. saccharolyticum (97.4% sequence similarity), H. kushneri (97.4% sequence similarity), H. acetethylicum (97.3% sequence similarity), H saccharolyticum subsp. senegalense (97.2% sequence similarity), and H. praevalens (97.0% sequence similarity). The G + C content of SL-HP was determined to be 33 mol% through the sequenced genome (Brown et al., 2011 ), which is comparable with other members of the Halanaerobium genus (Supplemental Table 1 ). Though SL-HP possesses high 16S rRNA gene sequence similarities and G + C content with the other members of the genus Halanaerobium , it possesses a pH tolerance up to 12 and a growth optimum at pH 11. This is the first report of a true alkaliphilic strain of Halanaerobium . As such, we have proposed that this strain be recognized as a new species. We proposed the species name of H. hydrogeniformans : hy.dro.ge.ni.for’mans. N.L. n. hydrogenum (from Gr. n. hudôr, water; and Gr. v. gennaô, to produce), hydrogen; L. part. adj. formans, forming; N.L. part. adj. hydrogeniformans, hydrogen-forming. This name was used to describe the recently sequenced genome of this organism (GenBank deposition GQ215697; Brown et al., 2011 ). Figure 2 Evolutionary relationships of 16 taxa . The evolutionary history was inferred using the maximum parsimony method (Eck and Dayhoff, 1966 ). The bootstrap consensus tree inferred from 1000 replicates is taken to represent the evolutionary history of the taxa analyzed (Felsenstein, 1985 ). Branches corresponding to partitions reproduced in less than 50% bootstrap replicates are collapsed. The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test (1000 replicates) are shown next to the branches (Felsenstein, 1985 ). The MP tree was obtained using the Close-Neighbor-Interchange algorithm (Nei and Kumar, 2000 ), with search level 3 (Felsenstein, 1985 ; Nei and Kumar, 2000 ) in which the initial trees were obtained with the random addition of sequences (10 replicates). The tree is drawn to scale, with branch lengths calculated using the average pathway method (Nei and Kumar, 2000 ), and are in the units of the number of changes over the whole sequence. All alignment gaps were treated as missing data. There were a total of 1380 positions in the final dataset, out of which 301 were parsimony informative. Phylogenetic analyses were conducted in MEGA4 (Tamura et al., 2007 ). Hydrogen production by H. hydrogeniformans When grown in serum bottles with 15 mM cellobiose and without agitation at 33°C, increase in optical density ceased after 125 h while H 2 production continued at a constant linear rate until approximately 450 h (Figure 3 A). This was consistent with acetate and formate production along with decreasing cellobiose concentrations over the same time period (Figure 3 B). Also of note, the pH decreased from 10.5 to 10.0 during the first 100 h of growth, perhaps as a result of acetate production (Figure 3 B), but was maintained over the course of the cultivation. These data suggest a constant rate of sugar metabolism until substrate levels were exhausted with sugar metabolism not being obligately tied to growth. Figure 3 Quantification of growth and metabolic by-products over time in H. hydrogenoformans growing on cellobiose under non-agitating conditions at 30°C and pH 11 with 7% (v/v) NaCl . (A) Logarithmic growth (♦) occurred for the first 125 h of cultivation at which point the cells entered stationary phase, but H 2 (□) production continued beyond 600 h as did (B) acetate (⋄) and formate (O) production until cellobiose (▲) was exhausted. (B) The pH (X) of the culture initially decreased from 10.5 but never fell below 10.0. Under non-agitated conditions in serum bottles, H. hydrogeniformans grew as an opaque mass that adhered to the bottom of the glass bottle. The addition of agitation resulted in a dramatic increase in the growth rate and H 2 evolution as well as a visible decrease in the amount of mass adhering to the glass surfaces. Culture agitation decreased the generation time to ∼12 h and increased the biomass yield by three-fold while also increasing the rate of H 2 production five-fold and the overall H 2 yield by ∼30% (Figure 4 ).Assessment of fermentative process efficiency is routinely referred to as the “Hydrogen Molar Yield (HMY),” and is determined by measuring the moles of H 2 produced per mole of a hexose or pentose sugar oxidized. The theoretical HMY maximum for glucose via anaerobic fermentation is 4 (Hawkes et al., 2002 ; Shung and Wen-Hsing, 2008 ). Facultative anaerobes, such as enteric bacteria, have a HMY maximum of 2. With H. hydrogeniformans in serum cultures, 5-carbon sugar HMY ranged from 0.4 to 0.7; whereas 6-carbon sugars, including cellobiose, yielded a HMY of 0.9–2.3 (Table 1 ). While this value is low compared to the theoretical maximum, the highest published HMY from glucose by an obligate anaerobic wild type strain, Clostridium beijerinckii RZF-1108, is 1.97 (Zhao et al., 2011 ). In comparison, strains considered for fermentation of cellulosic sugars, such as Clostridium thermocellum , have been reported to produce only 1.5 mol H 2 /mol hexose (Levin et al., 2006 ) and engineered Clostridium paraputrificum , over-expressing a hydrogenase, produced a maximal 2.4 mol H 2 /mol hexose (Morimoto et al., 2005 ). E. coli has been engineered to achieve a HMY of 1.8, which is near the theoretical maximum for a facultative anaerobe (Akhtar and Jones, 2009 ).These results are comparable to the HMY of H. hydrogeniformans growing on cellobiose. Clearly though, there is room for improvement in biological H 2 production toward the theoretical maximum. Identification of new strains with higher HMY or metabolic engineering of current strains, including H. hydrogeniformans , are possible routes toward achieving this goal. Figure 4 Agitation of H. hydrogenoformans cultures growing on cellobiose resulted in increased growth yield and rate (♦) as well as greater H2 yield and production rate (⋄) as compared to growth ( ) and H2 production (□) in non-agitated cultures . A concomitant increase in the rate of cellobiose degradation and acetate production were observed with agitation (data not shown). One of the first steps toward determining industrial feasibility is to move from sealed bottles to small bioreactors, preferably in fed-batch mode. This was performed using H. hydrogeniformans on several carbon sources. Bioreactor grown cells consumed 30 mM glucose within 67 h and yielded stoichiometric amounts of acetate and formate in a molar ratio of 3:1 (Figure 5 A) with 75% of the available carbon from glucose being accounted for under the conditions tested (Table 2 ). Examination of the genome of H. hydrogeniforms resulted in the identification of several pathways for acetate production from pyruvate (Figure 6 ). If one assumes that all acetate produced from pathways other than through pyruvate formate lyase results in the production of CO 2 , 90% of the carbon can be accounted for. The exit gas H 2 productivity profile peak occurred at 50 h and then decreased sharply (Figure 5 B). Numerical integration of the exit gas H 2 concentration over time yielded the total molar H 2 production. Table 2 Growth, metabolism and hydrogen production during fed-batch controlled cultivation . Substrate Generation time (h) A 600 Glucose consumed (mmol) A Acetate produced (mmol) Formate produced (mmol) % carbon recovered Cumulative H 2 produced (mmol) H 2 production rate (mmol/L/h) D HMY 30 mM glucose 25.3 0.10 54.9 + 0.5 (99%) B 103.7 + 1.0 34.1 + 0.7 75 (90) c 151.0 1.14 2.42 15 mM cellobiose 26.0 0.11 52.5 + 0.3 (99%) 98.2 + 0.8 26.1 + 0.3 69 (92) 127.5 0.71 2.39 150 mM cellobiose 31.3 0.78 542.3 + 0.1 (99%) 864.2 ± 0.4 378.1 ± 17.5 69 (90) 1,257.0 0.70 2.34 Switchgrass hydrolysate 21.2 0.16 ND ND ND ND 108.8 2.93 ND A Amount of cellobiose consumed was multiplied by 2 to show glucose equivalents . B percent of carbon source consumed . C number in parentheses is the percent of carbon recovery assuming 15% of the carbon source was used for cellular biosynthesis . D value is the maximal H 2 production rate, observed during mid-exponential growth for glucose and at end of experiment for cellobiose and switchgrass . Figure 5 Fed-batch bioreactor growth and H2 production from H. hydrogenoformans . (A) Growth ( ) increased with time as 30 mM glucose (Δ) decreased until exhaustion with the concomitant increase in acetate (■) and formate (♦). (B) The [H 2 ] (○) increased until glucose exhaustion while the peak H 2 productivity (●) peaked at a slightly earlier time. (C,D) Fed-batch bioreactor growth and H 2 production with 15 mM cellobiose using the same symbols. Figure 6 Pathways for the metabolism of pyruvate by H. hydrogeniformans based the genome annotation . (1) Lactate dehydrogenase (Halsa_1287), (2) Pyruvate formate lyase (Halsa_0723), (3) Pyruvate dehydrogenase (Halsa_0919, Halsa_2297, Halsa_0164), (4) Pyruvate ferrodoxin/flavodoxin oxidoreductase (Halsa_2334), (5) Phosphotransacetylase (Halsa_1556), (6) Acetate kinase (Halsa_1555). (1) TotMolH 2 t = 10 3 F ρ MW ∫ 0 t w θ 10 6 - w θ d θ where TotMolH2 is the molar production of H 2 (mmol), F is the volumetric flow rate of the N 2 stream (mL/min), ρ is the density of the nitrogen stream (g/mL), MW is the H 2 molecular weight (2.016 g/mol), and w is the H 2 concentration in the exit gas sample (ppm). Dividing Eq. 1 by the liquid volume and the reaction time gives the overall H 2 productivity (2) H2prod t = TotMolH 2 t ⋅ 60 Vol ⋅ t where H2prod is the volumetric productivity of H 2 production (micromoles/L/h), Vol is the liquid volume (L), and t is the reaction time (min). The maximum H 2 productivity was 1.14 mmol/L/h (Table 2 ) which was ∼50% of the 2.35 mmol/L/h reported for serum cultures using glucose. The HMY was calculated as (3) HMY t = TotMolH2 t C 0 - C t V o l where HMY is (mmol/mmol), and C is the substrate concentration, (mM). The HMY at maximum productivity was 2.42 mmol H 2 /mmol glucose (Table 2 ). This is less than the theoretical yield of 4.0 when producing acetate, which suggests that some of the available NADH from glycolysis is not being oxidized to H 2 by the hydrogenase. Bioreactor grown cells consumed 15 mM cellobiose within ∼100 h, producing acetate and formate with a molar ratio of 3.8:1 (Figure 5 C) with ∼70% of the available carbon being accounted for under the conditions tested. If one accounts for assumed CO 2 production based on identified pathways, one can account for 92% of carbon from cellobiose (Table 2 ). The exit gas H 2 concentration profile peaked at 73 h, with a maximum hydrogen productivity of 0.71 mmol/L/h, occurring at 94 h (Figure 5 D). This agreed well with serum bottle culture value of 0.7 mmol/L/h. The HMY at maximum productivity was 2.39 mmol H 2 /mmol glucose equivalent and was similar to the glucose cultivation above. Bioreactor cultivations with 150 mM cellobiose exhausted the cellobiose supply within 100 h and yielded an acetate to formate molar ratio of 4:1 with a H 2 productivity of 0.7 mmol/L/h with ∼70% of the available carbon being accounted for; 90% again assuming CO 2 production., and an HMY of 2.34 (Table 2 ). H. hydrogeniformans produces hydrogen from alkaline treated biomass While this microorganism can use several pure sugars for the production of hydrogen, to be practically applicable, it would also need to produce H 2 from lignocellulosic material. Further, to do so at a lower temperature and pressure would also reduce the carbon footprint and energy input of the process, streamlining it from current methodologies as detailed above. To this end and to test feasibility with H. hydrogeniformans in serum bottle cultures, shredded straw and switchgrass were dried, passed through a 1 mm screen and then subjected to a Na 2 CO 3 alkaline pretreatment modified from that described by Bjerre et al. ( 1996 ) by the use of 40 g Na 2 CO 3 /L instead of 10 g/L at several temperatures. Using the lower Na 2 CO 3 concentration and high pressure with wheat straw, Aspergillus niger produced the highest ethanol concentration with a pretreatment of 170°C for 10 min after neutralization of the hemicellulosic liquor (Bjerre et al., 1996 ). When crystalline cellulose was tested by these researchers, the same pretreatment conditions resulted in the highest conversion to the more bioavailable amorphous cellulose (Bjerre et al., 1996 ). In the present work, digestion of the non-neutralized hemicellulosic liquor produced from switchgrass or straw was tested for hydrogen production with duplicate cultures using various pretreatment temperatures at ambient pressure. Pretreatments included room temperature (RT; overnight), 55°C (2 h) 75°C (2 h), 90°C (1 h), 130°C (15 min), 150°C (15 min), or 170°C (15 min) with treatments at 55°C yielding the best hydrogen production rate and extent (Figure 7 A,B), followed by RT and 75°C. With each lignocellulosic material, protein concentrations more than tripled by the end of the incubation for the RT and 55°C pretreatments (Figure 7 C,D), suggesting cellular growth from the pretreatment liquor while the higher pretreatment temperatures yielded far less protein. This was concomitant with the greatest production of acetate in both the straw (6 mM at RT; 2.5 mM at 55°C) and switchgrass (9 mM at RT; 14 mM at 55°C) as measured by HPLC analysis (data not shown). Individual sugars could not be identified in the liquors when compared with pure sugar standards (data not shown). From pretreated straw, hydrogen production was best at 55°C, with RT and 75°C yielding similar results but less then 55°C. Use of the RT pretreatment would require further exploration to determine if the smaller yield of biohydrogen could be offset by the further reduction of the carbon footprint for biohydrogen production from straw. With switchgrass, however, the difference was far more distinct and the overall H 2 yield was more than double that of straw. In both cases, the pH did not fall below 10.2 during the incubations. Other differences included the lack of H 2 production with switchgrass at 90 and 130°C but some production at 150 and 170°C (Figure 7 A), suggesting that inhibitory compounds from lignin degradation may have been produced with these temperatures and higher Na 2 CO 3 concentrations (Klinke et al., 2004 ). The rates of biohydrogen production were 0.37 and 0.88 μmol H 2 /h/mL for straw and switchgrass, respectively, as compared to a range of 0.59 μmol H 2 /h/mL with xylose or arabinose, and up to 2.35 μmol H 2 /h/mL with glucose. The rate of hydrogen production with switchgrass was highly similar to that found recently using A. thermophilum DSM 6725 at 70°C (Yang et al., 2009 ). The values of 0.37 and 0.88 μmol H 2 /h/mL for straw and switchgrass, respectively, are ∼12% of those generated by a fully grown microbial consortium producing H 2 from steam explosion-treated corn stover (Datar et al., 2007 ) or generated in various neutrophilic bioreactor configurations with glucose as the feedstock (Gavala et al., 2006 ). However, a 10% (v/v) inoculum was used in the present case in order to assess both H 2 production as well as the capacity to support growth. Hence, it would be expected that if the feedstock was pumped into a bioreactor with an already fully grown culture, this strain would likely produce H 2 at an increased rate. Figure 7 Biohydrogen production from the hemicellulosic liquors of (A) switchgrass and (B) straw in serum grown H. hydrogenoformans . In each case a pretreatment of 55°C (■) yielded higher H 2 concentrations than those of room temperature (♦), 75°C (▲), 90°C (×), 130°C (*), 150°(●), or 170°C (+). (C) switchgrass and (D) straw protein concentrations from incubations with the respective hydrolysate showing that the initial (white bar) protein values increased by the end of the incubation (black bar), with the highest ratios of protein increase ( ) at lower temperatures. Fed-batch bioreactor culture of H. hydrogeniformans \n (E) degraded switchgrass hydrolysate (Δ) with the concomitant increase in optical density ( ), acetate (■), and formate (♦). (F) The (H 2) (□) increased until switchgrass hydrolysate exhaustion while the peak H 2 productivity (●) peaked at a slightly earlier time (Zhao et al., 2011 ). For bioreactor studies with switchgrass, after it was milled and exposed to an alkaline pretreatment, the resulting supernatant possessed ∼8.5 mM cellobiose. Chromatographic peaks for monomers were blurred in the analysis, so it was impossible to determine the concentration of hemicellulose degradation products and/or glucose in the sample. H. hydrogeniformans consumed the initial cellobiose charge within ∼200 h, producing an acetate:formate molar ratio of 3.5 (Figure 7 E). The exit gas (H 2 ) peaked at 190 h and then decreased sharply with a maximum H 2 productivity of 2.93 mM/L/h (Figure 7 F, Table 2 ). This is approximately four times higher productivity than with pure cellobiose and twice that of pure glucose, but the time to achieve this productivity was almost three times that of the pure carbon sources. This suggested that while the degradation of the switchgrass liquor required a longer residence time, the eventual biohydrogen productivity may cancel out the greater time required. Further optimization of both the fermentation process and the pretreatment will likely decrease the residence time required for the fermentation process, thus resulting in equal or higher H 2 productivity with a faster turnover time. While growth and biohydrogen production were supported with the liquor, H. hydrogeniformans also used the solid residue remaining from the pretreatment, although to a diminished extent and only producing between 23 and 32 μmoles of H 2 with either feedstock (data not shown). The solid residue may primarily be insoluble, amorphous cellulose. The inability of H. hydrogeniformans to more efficiently metabolize the insoluble cellulosic material can be tracked to the genome which contains no genes annotated as cellulases. In order to achieve more complete conversion of plant biomass, additional treatments steps would be needed such as incubation with high pH tolerance cellulose degrading enzymes. An alternative would be genetic engineering of H. hydrogeniformans for the production cellulases to allow more complete conversion of plant biomass. Either option would serve to degrade the amorphous and crystalline cellulose to cellobiose and glucose, which could then be metabolized by this bacterium to produce H 2 . Genetic manipulation is likely the more cost efficient method in the longer-term and will require the development of a facile genetic system. Additionally, a system for genetic manipulation would allow pathways to be knocked out to increase flux through H 2 producing pathways. Given the large number of existing protocols, this is likely achievable in the shorter-term."
} | 7,633 |
34265706 | null | s2 | 2,841 | {
"abstract": "The hyperaccumulation trait allows some plant species to allocate remarkable amounts of trace metal elements (TME) to their foliage without suffering from toxicity. Utilizing hyperaccumulating plants to remediate TME contaminated sites could provide a sustainable alternative to industrial approaches. A major hurdle that currently hampers this approach is the complexity of the plant-soil relationship. To better anticipate the outcome of future phytoremediation efforts, we evaluated the potential for soil metal-bioavailability to predict TME accumulation in two non-metallicolous and two metallicolous populations of the Zn/Cd hyperaccumulator Arabidopsis halleri. We also examined the relationship between a population's habitat and its phytoextraction efficiency. Total Zn and Cd concentrations were quantified in soil and plant material, and bioavailable fractions in soil were quantified via Diffusive Gradients in Thin-films (DGT). We found that shoot TME accumulation varied independent from both total and bioavailable soil TME concentrations in metallicolous individuals. In fact, hyperaccumulation patterns appear more plant- and less soil-driven: one non-metallicolous population proved to be as efficient in accumulating Zn on non-polluted soil as the metallicolous populations in their highly contaminated environment. Our findings demonstrate that in-situ information on plant phytoextraction efficiency is indispensable to optimize site-specific phytoremediation measures. If successful, hyperaccumulating plant biomass may provide valuable source material for application in the emerging field of green chemistry."
} | 408 |
23339487 | PMC3556108 | pmc | 2,842 | {
"abstract": "Background The modern society primarily relies on petroleum and natural gas for the production of fuels and chemicals. One of the major commodity chemicals 1,2-propanediol (1,2-PDO), which has an annual production of more than 0.5 million tons in the United States, is currently produced by chemical processes from petroleum derived propylene oxide, which is energy intensive and not sustainable. In this study, we sought to achieve photosynthetic production of 1,2-PDO from CO 2 using a genetically engineered cyanobacterium Synechococcus elongatus PCC 7942. Compared to the previously reported biological 1,2-PDO production processes which used sugar or glycerol as the substrates, direct chemical production from CO 2 in photosynthetic organisms recycles the atmospheric CO 2 and will not compete with food crops for arable land. Results In this study, we reported photosynthetic production of 1,2-PDO from CO 2 using a genetically engineered cyanobacterium Synechococcus elongatus PCC 7942. Introduction of the genes encoding methylglyoxal synthase ( mgsA ), glycerol dehydrogenase ( gldA ), and aldehyde reductase ( yqhD ) resulted in the production of ~22mg/L 1,2-PDO from CO 2 . However, a comparable amount of the pathway intermediate acetol was also produced, especially during the stationary phase. The production of 1,2-PDO requires a robust input of reducing equivalents from cellular metabolism. To take advantage of cyanobacteria’s NADPH pool, the synthetic pathway of 1,2-PDO was engineered to be NADPH-dependent by exploiting the NADPH-specific secondary alcohol dehydrogenases which have not been reported for 1,2-PDO production previously. This optimization strategy resulted in the production of ~150mg/L 1,2-PDO and minimized the accumulation of the incomplete reduction product, acetol. Conclusion This work demonstrated that cyanobacteria can be engineered as a catalyst for the photosynthetic conversion of CO 2 to 1,2-PDO. This work also characterized two NADPH-dependent sADHs for their catalytic capacity in 1,2-PDO formation, and suggested that they may be useful tools for renewable production of reduced chemicals in photosynthetic organisms.",
"conclusion": "Conclusion In this work, we demonstrated the 1,2-PDO production from CO2 for the first time by the engineered cyanobacterium S. elongatus PCC 7942 . By exploiting sADHs which have not been reported for 1,2-PDO production previously, a completely NADPH dependent pathway was built to channel the CBB cycle intermediate DHAP for 1,2-PDO production without accumulating the pathway intermediate, acetol. The best strain LH22, which harbors mgsA and yqhD both from E. coli and the adh from C. beijerinckii , produced ~150mg/L 1,2-PDO. This work revealed the great potential of the vast NADPH pool in photosynthetic cyanobacteria as a robust driving force for the production of chemicals. Among the chemicals that have been produced biologically in industrial scale, a significant number of them are synthesized by NADPH consuming pathways. For example, in amino acid production, studies have shown that increasing the NADPH pool can improve the production performance [ 26 , 27 ]. However, in most of the heterotrophic microorganisms, NADPH is mainly generated through the pentose phosphate pathway and TCA cycle and its pool size is relatively small compared to that of the NADH. On the other hand, photosynthetic organisms maintain high intracellular NADPH level. The unique metabolic feature of photosynthetic organisms provides great opportunities for the production of chemicals through NADPH dependent pathways.",
"introduction": "Introduction of the 1,2-PDO biosynthesis genes As described above, the genes mgsA , yqhD , and gldA from E. coli are needed to construct the 1,2-PDO biosynthesis pathway in S. elongatus from the CBB cycle intermediate DHAP (Figure 1A , B). These genes were cloned into an artificial operon driven by the P trc promoter under the control of lacO (Figure 2A ). The operon was inserted into the S. elongatus chromosome by homologous recombination at the Neutral Site I (NSI). A lacI gene and a spectinomycin resistant gene were also inserted together with the operon to achieve inducible gene expression (Additional file 1 : Figure S2) and facilitate antibiotics selection, respectively. The resulting strain was named LH21. Figure 2 Construction of the 1,2-propanediol pathways. ( A ) Illustration of artifical operons inserted into the Synechococcus elongatus PCC 7942 Neutral Site I (NSI) to build strains LH21, LH22, and LH23. gldA , yqhD and mgsA are from E.coli. Secondary alcohol dehydrogenases ( sADHs ) are from C. beijerinckii and T. brockii . ( B ) Test of the total RNA quality and the RT-PCR system using housekeeping gene rnpB . ( C ) Verification of the heterologous gene expression in engineered cyanobacteria strains by RT-PCR. To check if all the genes were successfully introduced and transcribed, reverse transcription polymerase chain reaction (RT-PCR) was performed. After induction with Isopropyl β-D-1-thiogalactopyranoside (IPTG), total RNA was extracted from LH21 and wildtype S. elongatus PCC 7942 . RT-PCR of the house keeping gene rnpB , whose transcription product is the RNA component of RNase P, was performed to verify the RT-PCR system. Using the rnpB specific primers, PCR products were obtained using cDNA synthesized from both wildtype and LH21 total RNA (Figure 2B ). On the other hand, the no-reverse-transcriptase (NRT) controls did not yield any products, suggesting that the genomic DNA contamination during the total RNA extraction was minimal and that the positive signals of RT-PCR are representative for the transcription of the target genes. Using the verified system, mgsA , yqhD , and gldA genes from E. coli were tested and showed to have expression in LH21 under inductive condition (Figure 2C ). Activity assays using cell lysate further suggested that these heterologous enzymes were functional in cyanobacterial cells (Figure 3A , B, Additional file 1 : Figure S1). Figure 3 Optimization of the 1,2-Propanediol production by exploiting different acetol-reducing enzymes. LH22 and LH23 were constructed which harbored C. beijerinckii and T. brockii . sADH. A ) methylglyoxal synthase activities were measured using cell lysate of LH21, LH22, and LH23, showing that mgsA was functionally overexpressed in all three strains. To confirm the functional expression of yqhD , NADPH-dependent methylglyoxal reduction activities were also measured (See Additional file 1 : Figure S1 and text in supporting material), although assay method with higher specificity needs to be developed. ( B ), ( C ) NADPH- and NADH-dependent acetol reduction activities measured using cell lysates of the wildtype Synechococcus elongatus PCC 7942 and engineered strains that overexpress gldA from E.coli (LH21), adh from C. beijerinckiiadh (LH22), and adh from T. brockiadh (LH23). (D) Acetol and 1,2-Propanediol accumulated by different strains after 10 days of production.",
"discussion": "Results and discussion Designing of the 1,2-PDO production pathway In light conditions, cyanobacteria fix CO 2 via the Calvin-Benson-Bassham (CBB) cycle which is powered by ATP and NADPH generated by the photosystems (Figure 1A ). Two CBB cycle intermediates, fructose-6-phosphate (F6P) and glyceraldehydes-3-phosphate (GAP), serve as the branch points of carbon leaving the CBB cycle to the central metabolism for glycogen synthesis and glycolysis, respectively. While glycogen synthesis is the major carbon and energy storage pathway, glycolysis and TCA cycle produce building blocks for cell growth. The synthesis of 1,2-PDO, on the other hand, starts from another CBB cycle intermediate, dihydroxyacetonephosphate (DHAP). The introduction of one extra branch point can potentially increase the flux of output carbon from the CBB cycle, which has been suggested to be beneficial for increasing photosynthesis efficiency in higher plants [ 18 , 19 ], but may also disrupt the normal flux distribution in the cell. To synthesize 1,2-PDO, DHAP is first converted to methylglyoxal (Figure 1A ) by methylglyoxal synthase (encoded by mgsA in E. coli ). Methyglyoxal is very toxic to the cells [ 11 ] and needs to be efficiently utilized by downstream enzymes. Two different metabolic routes have been shown to synthesize 1,2-PDO from methyglyoxal (Figure 1B ) [ 11 ]. The first involves reduction of methyglyoxal by the glycerol dehydrogenase (encoded by gldA in E. coli ) to lactaldehyde, which is further reduced by the 1,2-propanediol reductase (encoded by fucO in E.coli ) to yield the final product. The second route includes an alcohol dehydrogenase (such as the broad-substrate range aldehyde reductase encoded by yqhD in E. coli ) to produce acetol as the intermediate, which is then converted to 1,2-PDO by gldA. The latter route was chosen to introduce into S. elongatus because yqhD gene has been previously overexpressed in this organism for biofuel production and showed relatively good performance, possibly due to its NADPH-specific cofactor preference. Introduction of the 1,2-PDO biosynthesis genes As described above, the genes mgsA , yqhD , and gldA from E. coli are needed to construct the 1,2-PDO biosynthesis pathway in S. elongatus from the CBB cycle intermediate DHAP (Figure 1A , B). These genes were cloned into an artificial operon driven by the P trc promoter under the control of lacO (Figure 2A ). The operon was inserted into the S. elongatus chromosome by homologous recombination at the Neutral Site I (NSI). A lacI gene and a spectinomycin resistant gene were also inserted together with the operon to achieve inducible gene expression (Additional file 1 : Figure S2) and facilitate antibiotics selection, respectively. The resulting strain was named LH21. Figure 2 Construction of the 1,2-propanediol pathways. ( A ) Illustration of artifical operons inserted into the Synechococcus elongatus PCC 7942 Neutral Site I (NSI) to build strains LH21, LH22, and LH23. gldA , yqhD and mgsA are from E.coli. Secondary alcohol dehydrogenases ( sADHs ) are from C. beijerinckii and T. brockii . ( B ) Test of the total RNA quality and the RT-PCR system using housekeeping gene rnpB . ( C ) Verification of the heterologous gene expression in engineered cyanobacteria strains by RT-PCR. To check if all the genes were successfully introduced and transcribed, reverse transcription polymerase chain reaction (RT-PCR) was performed. After induction with Isopropyl β-D-1-thiogalactopyranoside (IPTG), total RNA was extracted from LH21 and wildtype S. elongatus PCC 7942 . RT-PCR of the house keeping gene rnpB , whose transcription product is the RNA component of RNase P, was performed to verify the RT-PCR system. Using the rnpB specific primers, PCR products were obtained using cDNA synthesized from both wildtype and LH21 total RNA (Figure 2B ). On the other hand, the no-reverse-transcriptase (NRT) controls did not yield any products, suggesting that the genomic DNA contamination during the total RNA extraction was minimal and that the positive signals of RT-PCR are representative for the transcription of the target genes. Using the verified system, mgsA , yqhD , and gldA genes from E. coli were tested and showed to have expression in LH21 under inductive condition (Figure 2C ). Activity assays using cell lysate further suggested that these heterologous enzymes were functional in cyanobacterial cells (Figure 3A , B, Additional file 1 : Figure S1). Figure 3 Optimization of the 1,2-Propanediol production by exploiting different acetol-reducing enzymes. LH22 and LH23 were constructed which harbored C. beijerinckii and T. brockii . sADH. A ) methylglyoxal synthase activities were measured using cell lysate of LH21, LH22, and LH23, showing that mgsA was functionally overexpressed in all three strains. To confirm the functional expression of yqhD , NADPH-dependent methylglyoxal reduction activities were also measured (See Additional file 1 : Figure S1 and text in supporting material), although assay method with higher specificity needs to be developed. ( B ), ( C ) NADPH- and NADH-dependent acetol reduction activities measured using cell lysates of the wildtype Synechococcus elongatus PCC 7942 and engineered strains that overexpress gldA from E.coli (LH21), adh from C. beijerinckiiadh (LH22), and adh from T. brockiadh (LH23). (D) Acetol and 1,2-Propanediol accumulated by different strains after 10 days of production. Production of 1,2-PDO 1,2-PDO production by LH21 was performed under high light condition (100 μE/s/m 2 ) with 50mM bicarbonate supplementation in the medium. After induction, LH21 produced around 16mg/L 1,2-PDO in 4 days, with the highest production rate of 7mg/L/day. However, although no defect in cell growth was seen, the production rate decreased rapidly and the total titer was only around 22mg/L after 10 days (Figure 4A , B). The drastically decreased and eventually ceased production by LH21 led to one hypothesis: key substrate(s) might become limited at certain stage of cell growth which decreased the flux of the synthetic 1,2-PDO production pathway. Three substrates are needed to produce 1,2-PDO in LH21: NADPH, NADH, and DHAP (Figure 1A , B). Among these substrates, NADPH and DHAP are made through photosynthesis light and dark reactions (Figure 1A ), respectively, and could be continuously generated under light condition. However, NADH may mainly be generated by the NADH-dependent glyceraldehyde-3-phosphate dehydrogenase in the glycolysis [ 20 ]. And the reaction catalyzed by the putative NADH-dependent malate dehydrogenase in TCA cycle may also contribute to the cellular NADH level. It has been suggested that the main function of glycolysis and TCA cycle in cyanobacteria is to generate essential metabolites for biomass synthesis under light conditions, rather than to produce reducing equivalents and energy [ 21 ]. As such, when the growth rate of LH21 cells slowed down in the stationary phase, the activities of these NADH generating pathways may also decrease. Figure 4 Cell growth and 1,2-Propanediol production by engineered Cyanobacteria strains. If NADH is really the limiting factor in our production scenario, the partially reduced intermediate acetol may accumulate. In fact, at the end of the production, around 16mg/L acetol was accumulated, which was comparable to the level of 1,2-PDO (~22mg/L) (Figure 3D ). In addition, acetol was only detected after 4 days and kept accumulating during the late stage of production (data not shown). These results are consistent with the above-mentioned hypothesis and suggest that the NADH-dependent reduction of acetol catalyzed by gldA might be the limiting step in the 1,2-PDO production pathway. Improving 1,2-PDO production using NADPH-dependent secondary alcohol dehydrogenases To overcome the bottleneck of the 1,2-PDO production in LH21, one possible strategy is to overexpress the soluble transdehydrogenase (STH) which produces NADH at the expense of NADPH. However, genes encoding this enzyme have not been found in S. elongatus genome. Heterologous overexpression of the Pseudomonas aeruginosasth gene in cyanobacteria has been shown to be instable and caused growth defect [ 22 ]. Alternatively, it could be beneficial to find the NADPH-dependent counterpart of gldA. To convert acetol to 1,2-PDO, a hydroxyl group on the secondary carbon has to be made, which can be catalyzed by the secondary alcohol dehydrogenase (sADH) family of enzymes. Several sADHs have been characterized previously that are NADPH-dependent [ 4 ], including the sADH encoded by the adh gene in Thermoanaero bacterbrockii [ 23 ] and Clostridium beijerinckii [ 24 ]. To test their catalytic activity for the substrate acetol, these two sADHs as well as the E. coli gldA were purified and their kinetics parameters were measured (Table 1 ). The kinetics studies showed that the C. beijerinckii sADH has the highest K cat among all three enzymes tested. However, the large K m value of C. beijerinckii sADH suggested that the enzyme has relatively low affinity to the substrate acetol. On the other hand, the T. bacterbrockii sADH has the highest acetol affinity and a two-fold higher K cat than that of the gldA enzyme, which is the most commonly used enzyme for this reaction step in previous studies. In summary, these two sADH have distinct kinetic features and both showed activity towards the substrate acetol, which suggested that they may be used in cyanobacterial cells for 1,2-propanediol production in vivo . Table 1 Kinetics parameters of gldA and secondary alcohol dehydrogenases (sADH) for acetol Enzyme Cofactor K m (mM) K cat (S -1 ) K cat /K m (mM -1 S -1 ) E.coli gldA NADH 1.64 0.59 0.36 C.beijerinckii sADH NADPH 7.98 4.95 0.62 T.brockii sADH NADPH 0.23 1.26 5.48 C. beijerinckii and T. brockii adh were cloned and introduced into the cyanobacterial genome to replace gldA . The resulting strains are named LH22 and LH23, respectively (Figure 2A ). RT-PCR was also performed to verify the expression of these genes (Figure 2B , C). Enzyme assays with crude cell extract of LH22 and LH23 further verified that both C. beijerinckii and T. brockii sADH were functionally overexpressed and showed higher activities of NAD(P)H-dependent acetol reduction compared to that in LH21. Especially, the C. beijerinckii sADH overexpression in LH22 delivered the highest activity (Figure 3B , C). Production using strains LH22 and LH23 yielded significantly higher 1,2-PDO titer (~150 and 80mg/L, respectively) compared to that of LH21 (Figure 4B ). Notably, the high production rate was maintained through the 10 days of production. In consistent with the hypothesis mentioned in the previous section, the high level of NADPH-dependent acetol reduction activity in LH22 and LH23 also significantly reduced the accumulation of the intermediate acetol (Figure 3D ). Despite its great significance to metabolic engineers, the information on intracellular NAD(P)H level during different growth phases and growth conditions in cyanobacteria is very limited. Although it is believed that NADPH is more abundant than NADH in cyanobacteria [ 25 ], only a few studies discussed its role in biofuel/biochemical synthesis from CO 2 [ 6 , 22 ]. The science behind efficient conversion of CO 2 to chemicals and fuels is still in its infancy and the NADPH driving force theory still needs to be extensively tested, which requires the accumulation of empirical evidence in more production scenario, as well as fundamental studies on NAD(P)H levels and their regulation. In our case, other factors may also contribute to the difference between the production levels of the NADH and NADPH-dependent pathways. For example, the NADPH-dependent enzymes may be better folded and more active when expressed in cyanobacteria. And different level of physiological fitness may be caused by overespression of different enzymes, although all production strains showed the same growth phenotype as the wildtype."
} | 4,805 |
33468210 | PMC7816431 | pmc | 2,843 | {
"abstract": "Background Renewable chemicals have attracted attention due to increasing interest in environmental concerns and resource utilization. Biobased production of industrial compounds from nonfood biomass has become increasingly important as a sustainable replacement for traditional petroleum-based production processes depending on fossil resources. Therefore, we engineered an Enterobacter cloacae budC and ldhA double-deletion strain (namely, EC∆budC∆ldhA) to redirect carbon fluxes and optimized the culture conditions to co-produce succinic acid and acetoin. Results In this work, E. cloacae was metabolically engineered to enhance its combined succinic acid and acetoin production during fermentation. Strain EC∆budC∆ldhA was constructed by deleting 2,3-butanediol dehydrogenase ( budC ), which is involved in 2,3-butanediol production, and lactate dehydrogenase ( ldhA ), which is involved in lactic acid production, from the E. cloacae genome. After redirecting and fine-tuning the E. cloacae metabolic flux, succinic acid and acetoin production was enhanced, and the combined production titers of acetoin and succinic acid from glucose were 17.75 and 2.75 g L −1 , respectively. Moreover, to further improve acetoin and succinic acid production, glucose and NaHCO 3 modes and times of feeding were optimized during fermentation of the EC∆budC∆ldhA strain. The maximum titers of acetoin and succinic acid were 39.5 and 20.3 g L −1 at 72 h, respectively. Conclusions The engineered strain EC∆budC∆ldhA is useful for the co-production of acetoin and succinic acid and for reducing microbial fermentation costs by combining processes into a single step.",
"conclusion": "Conclusions In this study, we engineered an E. cloacae budC and ldhA double-deletion strain (namely, EC∆budC∆ldhA) to produce succinic acid and acetoin. The highest acetoin and succinic acid titers achieved by this engineered strain were 39.5 and 20.3 g L −1 , respectively, during optimization of fed-batch fermentation conditions. Our findings demonstrated that the EC∆budC∆ldhA strain would be useful for the simultaneous production of commercial products (acetoin and succinic acid) and the prevention of by-product formation, thus reducing the cost of microbial fermentation in a single step.",
"discussion": "Discussion Several studies have also shown that the inactivation of budC significantly improves the production of acetoin. Indeed, previous reports have shown that the deletion of the budC gene could decrease 2,3-butanediol. Three butanediol stereoisomers, namely, (2R,3R)-2,3-butanediol, (2S,3S)-2,3-butanediol, and meso-2,3-butanediol, are found in many bacterial species, such as Enterobacter cloacae [ 10 , 22 ], Klebsiella pneumoniae [ 23 ], and Bacillus licheniformis [ 24 ], and meso-2,3-butanediol and (2S,3S)-2,3-butanediol are the major forms that accumulate in E. cloacae [ 25 ]. However, when the budC gene was deleted, a small amount of 2,3-butanediol could still be detected [ 3 , 22 , 23 , 26 , 27 ]. In this study, the budC gene was knocked out, and we observed that the production of meso-2,3-butanediol and (2S,3S)-2,3-butanediol decreased (data not shown). A previous study characterized a budC and glycerol dehydrogenase (encoded by gldA and dhaD )-deficient Klebsiella pneumoniae strain , which removes 2,3-butanediol under conditions wherein glycerol is used as a carbon source. These findings suggested that dhaD and gldA may be involved in 2,3-butanediol formation [ 22 ]. Another study reported diacetyl production by inactivating budA , budC , and diacetyl reductases (also known as glycerol dehydrogenase, encoded by gdh ) in E. cloacae SDM. When the gdh and budC genes were both inactivated in the strain E. cloacae SDM (∆budA), (2R,3R) 2,3-butanediol could be slightly detected [ 10 ]; these results show that there is a third enzyme responsible for 2,3-butanediol production in the E. cloacae strain. In the present work, disruption of the budC gene remarkably decreased the production of 2,3-butanediol by almost 2.7-fold compared to that of the wild type and EC∆ldhA strains (Table 1 ). However, small amounts of 2,3-butanediol were still detected in a few of the EC∆budC and EC∆budC∆ldhA strains, indicating the presence of other genes encoding enzymes that convert acetoin to 2,3-butanediol in E. cloacae . Theoretically, the formation of 1 mol succinic acid from glucose requires 1 mol of CO 2 [ 20 , 28 ]. Therefore, CO 2 is indispensable for succinic acid biosynthesis, and many studies have demonstrated that succinic acid production can be increased by adding HCO 3 − to the fermentation medium [ 12 , 28 ]. Cheng et al. [ 28 ] increased succinic acid production in K. pneumoniae by adding HCO 3 − to the fermentation medium. In another study, Wu et al. [ 20 ] reported yields of 40.67 g L −1 2,3-butanediol and 21.79 g L −1 succinic acid by adding NaHCO 3 to E. cloacae . In this study, supplying NaHCO 3 during batch fermentation may enhance succinic acid production by improving the quantity of dissolved CO 2 and by increasing the carbon flux to succinic acid. When grown in fermentation medium without NaHCO 3 , the final acetoin production (16.45 g L −1 ) was slightly higher; however, the final amount of succinic acid produced was only 1.15 g L −1 . When grown in fermentation medium supplemented with NaHCO 3 , succinic acid production was 34.8% higher than the amount produced during batch fermentation without NaHCO 3 (Table 2 ). In general, the production of succinic acid was higher under anaerobic conditions, and bacterial producers of succinic acid can be found among facultative and strictly anaerobic rumen bacteria such as Mannheimia succiniciproducens [ 29 ], Actinobacillus succinogenes [ 30 ], and Anaerobiospirillum succiniciproducens [ 31 ]. E. cloacae is a facultative anaerobe, and when it is cultured under anaerobic conditions, the glucose consumption rate of the ΔbudCΔldhA strain is slower, resulting in lower production concentration of acetoin. In addition, when cultured under anaerobic conditions, the ΔbudCΔldhA strain was found to produce lactic acid (Additional file 1 : Fig. S2). Although we only knocked out d -lactate dehydrogenase, this may activate other lactate dehydrogenases under anaerobic conditions, such as l -lactate dehydrogenase, leading to the production of lactic acid. A previous study showed that reducing the carbon flux to lactate, ethanol, and acetate by-products can be performed by deleting the ldhA , adhE , and pta genes in K. pneumoniae [ 32 ]. In this study, by blocking lactic acid synthesis pathways to redirect more carbon sources to succinic acid synthesis in wild type E. cloacae , the engineered EC∆budC significantly increased succinic acid yield. This engineering approach may represent a practical strategy involving the deletion of ldhA and budC genes to reduce carbon flux towards the formation of by-products."
} | 1,744 |
37409033 | PMC10318857 | pmc | 2,844 | {
"abstract": "Over the past two decades, nanofillers have attracted significant interest due to their proven chemical, mechanical, and tribological performances. However, despite the significant progress realized in the application of nanofiller-reinforced coatings in various prominent fields, such as aerospace, automobiles and biomedicine, the fundamental effects of nanofillers on the tribological properties of coatings and their underlying mechanisms have rarely been explored by subdividing them into different sizes ranging from zero-dimensional (0D) to three-dimensional (3D) architectures. Herein, we present a systematic review of the latest advances on multi-dimensional nanofillers for enhancing the friction reduction and wear resistance of metal/ceramic/polymer matrix composite coatings. Finally, we conclude with an outlook for future investigations on multi-dimensional nanofillers in tribology, providing possible solutions for the key challenges in their commercial applications.",
"conclusion": "6. Conclusions and outlooks The energy consumption of mechanical systems caused by friction and wear is directly related to their service life and application accuracy. Accordingly, achieving low friction and wear on moving surfaces has become a focus of attention. Multi-dimensional nanofillers have attracted significant attention because of their proven tribological potential in coatings. Nowadays, nano-coatings containing a large amount of various nanofillers have demonstrated excellent friction and wear behaviors, implying an excellent route to reduce the energy consumption of the ceramic, metal, and polymer-based components. 6.1. Conclusions The well-distributed dispersibility, ideal nanostructure, and interfacial bonding strength of multi-dimensional nanofillers have been increasingly recognized as crucial factors that determine tribological performances of coatings. Preliminary guidelines were provided for selecting appropriate nanofillers to achieve the expected lubrication behaviors of various coatings. The unique spherical structures and high chemical stability endow 0D-nanofillers with good friction and wear behaviors. The incorporation of 0D-nanofillers has confirmed to be effective for improving the hardness of MBCs and CBCs, which is considered to be the leading factor contributing to the improvement in tribological coatings. Furthermore, hard nanofillers at the frictional interfaces serve as distance holders and ball bearings, thus enhancing the load-bearing capacity and optimizing the anti-wear performance of polymer coatings. 1D-nanofillers with good dispersion improve the fracture toughness of coatings by preventing the propagation and coalescence of short cracks, thus reducing the wear degree of composite coatings. In addition, surface functionalization and chemical structure are two critical characteristics of 1D-nanofillers, which account for their tribological behaviors and wear applications. According to numerous recent reports, the excellent mechanical strength, thermal stability, and low surface energy of 2D-nanofillers make them suitable for a wide range of applications under friction and wear of interfacial coatings. In the case of polymer coatings, the transfer film plays an important role in the anti-wear enhancement and the friction reduction; therefore, 2D-nanofillers provide a feasible approach to improve the adhesion strength and load-bearing capacity of the transfer film. Furthermore, the application of nanofillers in coatings has accelerated the exploration of transparent conduction coatings with excellent anti-wear behavior, suggesting possible substitutes for the expensive indium tin oxide films in many optoelectronic devices. 3D-nanohybrids inherit the characteristics of their multiple components, which are directly responsible for their excellent tribological performance. Well-designed 3D-nanofillers certify their multidimensional collaboration effects on the friction and wear behaviors of composites, enabling nanohybrids to exhibit superior functions with friction reduction and wear resistance of tribological coatings. In addition, recent attention on 3D-nanofillers has broadened their extensive applications. For example, HIF-MoS 2 /RGO and Cu/PDA/MoS 2 have been applied in ionic grease and sunflower oil for the purpose of improving their lubricant performances. Table 5 presents a summary of the recent developments and specific applications of multi-dimensional nanofillers with metal, ceramic, and polymer coatings in the tribological field. It can be seen in Table 5 that metal, ceramic, and polymer coatings have their respective applications. Firstly, in the case of metal-based coatings with nano-additives, the addition of nanofillers significantly increases the microhardness of the coating. Due to their high wear resistance, the application of metal-based coatings is confirmed to be the most extensive among the three types of aforementioned coatings. Furthermore, with an increase in the content of nanofillers, ceramic-based coatings exhibit a clear increase in toughness, and these coatings are mainly suitable for parts operating at high temperature. Finally, the synergistic effect of nanofillers and polymer-based coatings not only results in excellent performances in the fields of friction and wear, but endows polymer-based coatings with potential application in the field of anti-corrosion. The latest progress and specific application of nanofillers with different dimensions in the field of tribology Matrix Nanofillers Recent developments Specific applications Metal 0D Ni–B/Al 2 O 3 improving wear resistance; 73 Ni–P–1.5TiN reduced wear loss (64.3 wt%). 75 Petrochemical, automotive, electronics, nuclear, and piston ring-cylinder liners 1D Graphite/copper–zinc promoting load-bearing capability; 121 Ni–CeO 2 NRs improving the microhardness (∼3 times). 128 Contact brushes, bearing materials, and automobiles 2D WC and WC/Graphene increasing wear resistance (∼18%); 179 the hardness increasing (∼4.4 times). 190 Electrical contacts, micro devices, bio-implants and mould materials 3D Ni–B–CeO 2 improving the microhardness (∼61%); 230 nano TiN particles grafted onto GO solving the dispersion problem. 232 Oil refinery, shipping, and machinery manufacturing Ceramic 0D Silver or copper resulting film structure densification; 87 YSZ ceramic matrix coatings possessing excellent thermal stability. 91 Sections of low- and high-pressure compressors, and aero-engines 1D The fracture toughness and bonding strength increasing (63.18% and 9.9 times); 136 fracture toughness increasing (∼38%). 141 Brakes, high-temperature heat exchangers, and nose cones piston 2D Hardness, elastic modulus and fracture toughness increasing (19%, 18% and 300%); 203 the friction coefficient and wear rate decreased (∼13% and ∼19%). 211 Power generation, plunger pumps, bearings, and turbine blades 3D Thick CrN/AlN coatings exhibiting a maximum micro hardness up to 3800 Hv; 241 the wear resistance increasing (5.57 times). 243 Valves in oil and gas field, piston ring and cylinder liners Polymer 0D The precipitated type of nano-silica replacing the expensive fumed nano-silica; 50 MD simulation to design different types of materials. 101 Cars manufacturing, industrial production, and astronautics 1D SWCNTs enhancing the hardness (∼66%) and elastic modulus (∼58%); 158 modification with HDI improving the dispersion of TiNTs. 170 Anticorrosion coating, bearings self-cleaning coatings, and gears 2D Transfer film determined polymer tribological performance; 212 PI/3 GP exhibiting low friction (19–29%) and higher wear-resistant (35–78%). 219 Electrical submersible pumps, air-conditioning, and cable coating 3D Wear rate of HBN-TiO 2 /EP composite coating decreased (65.8%); 249 the GO-CNTs-PI three-phase structure improving the mechanical properties of the film. 252 Corrosion protection, ocean, aviation, and microelectronics 6.2. Challenges and future recommendations Although the design of new nanofiller coatings provides broads application prospects for controlling friction and wear, there are still many challenges, as follows. (1) It is a great significance to study new nanofiller coatings under extreme friction conditions. With the rapid development of advanced industrial equipment, the working environments of machinery need to withstand extreme conditions such as high pressure, speed and temperature. Nanofiller coatings can work under some extremely harsh conditions, but their working times are too short to fully exert their anti-wear and friction reducing effects. Therefore, exploring new nanofillers with high tribological properties will be the focus of future basic research and application, especially ideal working hours under extreme conditions. Moreover, the application of nanofillers in composite materials has promoted the exploration of the high wear resistance of thermoelectric coatings, providing a good opportunity for the industrial application of intelligent thermoelectric devices in the future. (2) The impact of nanofillers on the tribological behavior of coatings is complex, given that the unavoidable defects and contamination of nanofillers reduce the tribological behaviors and reproducibility. Therefore, the complex wear behavior of nanocomposites should be evaluated to eliminate potential hazards related to their tribological applications. (3) Although well-designed friction and wear coatings have broad application prospects, there are still difficulties in improving the dispersion stability of nanofillers. Physically treated nanoparticles are prone to re-aggregation. Chemical modification may mask the natural characteristics of nanoparticles, but chemical modifiers may degraded easily during wear. Therefore, further research is needed to improve the dispersion method. The self-dispersion method does not require any modifiers and has good dispersion stability, and thus it is a development trend. (4) Although multi-dimensional hybrid nanofillers have also been produced in the laboratory, their production on an industrial scale is limited due to their complex synthesis procedures. Therefore, further exploration should focus on the production processes, repeatability and practicality of synthesis methods. (5) Finally, many studies on the coating mechanisms have shown that the formation of high-performance friction coatings largely depends on nanofillers. In this case, carefully designed functional nanofillers make them the best candidate materials for regulating the nanostructure and self-lubricating properties of coatings. Among the numerous nanofiller coating systems, the wear-resistant friction coatings formed using MXene systems possess excellent performance. However, the complex mechanical behavior of MXenes seriously hinders their tribological application. Thus, to achieve efficient friction and wear systems, more fundamental research is necessary to understand the potential friction and wear mechanisms. The tribology of nano-fillers in coatings has not been fully reviewed. In particular, insight into the critical role of multi-dimensional nanomaterials in different coating substrates has not been reported. Herein, initially, nanofillers were divided into four aspects based on their dimensions, including 0D, 1D, 2D, and 3D. The applications of nanofillers in metal, ceramic, and polymer coatings were described in each case. Furthermore, the tribological mechanism of nanofillers in the coating was expounded. These findings are beneficial in understanding the interface-related tribological behaviors of the coatings, thus widening their applications in various industries. It is expected that the present work will inspire new possibilities for next-generation coating applications, together with superior tribological properties, and stimulate further developments in tribological coating nanotechnology.\n\n6.1. Conclusions The well-distributed dispersibility, ideal nanostructure, and interfacial bonding strength of multi-dimensional nanofillers have been increasingly recognized as crucial factors that determine tribological performances of coatings. Preliminary guidelines were provided for selecting appropriate nanofillers to achieve the expected lubrication behaviors of various coatings. The unique spherical structures and high chemical stability endow 0D-nanofillers with good friction and wear behaviors. The incorporation of 0D-nanofillers has confirmed to be effective for improving the hardness of MBCs and CBCs, which is considered to be the leading factor contributing to the improvement in tribological coatings. Furthermore, hard nanofillers at the frictional interfaces serve as distance holders and ball bearings, thus enhancing the load-bearing capacity and optimizing the anti-wear performance of polymer coatings. 1D-nanofillers with good dispersion improve the fracture toughness of coatings by preventing the propagation and coalescence of short cracks, thus reducing the wear degree of composite coatings. In addition, surface functionalization and chemical structure are two critical characteristics of 1D-nanofillers, which account for their tribological behaviors and wear applications. According to numerous recent reports, the excellent mechanical strength, thermal stability, and low surface energy of 2D-nanofillers make them suitable for a wide range of applications under friction and wear of interfacial coatings. In the case of polymer coatings, the transfer film plays an important role in the anti-wear enhancement and the friction reduction; therefore, 2D-nanofillers provide a feasible approach to improve the adhesion strength and load-bearing capacity of the transfer film. Furthermore, the application of nanofillers in coatings has accelerated the exploration of transparent conduction coatings with excellent anti-wear behavior, suggesting possible substitutes for the expensive indium tin oxide films in many optoelectronic devices. 3D-nanohybrids inherit the characteristics of their multiple components, which are directly responsible for their excellent tribological performance. Well-designed 3D-nanofillers certify their multidimensional collaboration effects on the friction and wear behaviors of composites, enabling nanohybrids to exhibit superior functions with friction reduction and wear resistance of tribological coatings. In addition, recent attention on 3D-nanofillers has broadened their extensive applications. For example, HIF-MoS 2 /RGO and Cu/PDA/MoS 2 have been applied in ionic grease and sunflower oil for the purpose of improving their lubricant performances. Table 5 presents a summary of the recent developments and specific applications of multi-dimensional nanofillers with metal, ceramic, and polymer coatings in the tribological field. It can be seen in Table 5 that metal, ceramic, and polymer coatings have their respective applications. Firstly, in the case of metal-based coatings with nano-additives, the addition of nanofillers significantly increases the microhardness of the coating. Due to their high wear resistance, the application of metal-based coatings is confirmed to be the most extensive among the three types of aforementioned coatings. Furthermore, with an increase in the content of nanofillers, ceramic-based coatings exhibit a clear increase in toughness, and these coatings are mainly suitable for parts operating at high temperature. Finally, the synergistic effect of nanofillers and polymer-based coatings not only results in excellent performances in the fields of friction and wear, but endows polymer-based coatings with potential application in the field of anti-corrosion. The latest progress and specific application of nanofillers with different dimensions in the field of tribology Matrix Nanofillers Recent developments Specific applications Metal 0D Ni–B/Al 2 O 3 improving wear resistance; 73 Ni–P–1.5TiN reduced wear loss (64.3 wt%). 75 Petrochemical, automotive, electronics, nuclear, and piston ring-cylinder liners 1D Graphite/copper–zinc promoting load-bearing capability; 121 Ni–CeO 2 NRs improving the microhardness (∼3 times). 128 Contact brushes, bearing materials, and automobiles 2D WC and WC/Graphene increasing wear resistance (∼18%); 179 the hardness increasing (∼4.4 times). 190 Electrical contacts, micro devices, bio-implants and mould materials 3D Ni–B–CeO 2 improving the microhardness (∼61%); 230 nano TiN particles grafted onto GO solving the dispersion problem. 232 Oil refinery, shipping, and machinery manufacturing Ceramic 0D Silver or copper resulting film structure densification; 87 YSZ ceramic matrix coatings possessing excellent thermal stability. 91 Sections of low- and high-pressure compressors, and aero-engines 1D The fracture toughness and bonding strength increasing (63.18% and 9.9 times); 136 fracture toughness increasing (∼38%). 141 Brakes, high-temperature heat exchangers, and nose cones piston 2D Hardness, elastic modulus and fracture toughness increasing (19%, 18% and 300%); 203 the friction coefficient and wear rate decreased (∼13% and ∼19%). 211 Power generation, plunger pumps, bearings, and turbine blades 3D Thick CrN/AlN coatings exhibiting a maximum micro hardness up to 3800 Hv; 241 the wear resistance increasing (5.57 times). 243 Valves in oil and gas field, piston ring and cylinder liners Polymer 0D The precipitated type of nano-silica replacing the expensive fumed nano-silica; 50 MD simulation to design different types of materials. 101 Cars manufacturing, industrial production, and astronautics 1D SWCNTs enhancing the hardness (∼66%) and elastic modulus (∼58%); 158 modification with HDI improving the dispersion of TiNTs. 170 Anticorrosion coating, bearings self-cleaning coatings, and gears 2D Transfer film determined polymer tribological performance; 212 PI/3 GP exhibiting low friction (19–29%) and higher wear-resistant (35–78%). 219 Electrical submersible pumps, air-conditioning, and cable coating 3D Wear rate of HBN-TiO 2 /EP composite coating decreased (65.8%); 249 the GO-CNTs-PI three-phase structure improving the mechanical properties of the film. 252 Corrosion protection, ocean, aviation, and microelectronics",
"introduction": "1. Introduction Presently, the significant increase in energy consumption in modern industry has attracted increasing attention; however, its reduction is still a considerable challenge. 1–3 According to numerous reports, friction is responsible for approximately one-third of the energy consumed in automobiles. 4,5 It is believed that even just a 20% reduction in wear substantially lowers the economic costs in terms of energy usage and environmental impact. 6–8 Hence, controlling friction and wear has attracted great scientific interest and is technically valuable for reducing the consumption of energy and promoting industrial development. The application of surface coatings has been increasingly regarded as an effective method to optimize the tribological performance of materials. 9–11 In this case, the two fundamental prerequisites for the design of surface coatings are low friction and high wear resistance. 12,13 However, under increasingly severe working conditions, traditional coatings fail to meet the ever-increasing requirements for their tribology performance. 14,15 In this regard, an improvement in the friction and wear of coatings can be achieved by incorporating functional additives or other components. Among the various additives, nanofillers have attracted particular interest due to their attractive tribological potential based on their excellent corrosion resistance, unique mechanical behaviors, and high thermal conductivities. 16–18 According to their macro/microscale structures, nanofillers can be classified into zero-dimensional (0D), one-dimensional (1D), two-dimensional (2D) and three-dimensional (3D) architectures. 19,20 As shown in Fig. 1 , 0D-nanofillers mainly include nanoparticles and nanospheres, where all their dimensions are maintained at the nanoscale. 21 1D-nanofillers refer to nanofibers, nanorods and nanotubes, which have nanoscale diameters but high aspect ratios. 22,23 The family of 2D-nanofillers is comprised of various nanosheets and their corresponding multilayer structures, where their ultrathin sheet-like structures have a thickness ranging from a few to dozens of nanometers. 24,25 Alternatively, 3D-nanofillers include pure 3D nanomaterials or hybrids mainly composed of one or more 0D, 1D, and 2D basic structural units. 26,27 For example, when nanodiamonds are ∼5 nm in size, they are classified as 0D; 28 in contrast, their pure 3D structures have a size of ∼12 nm and above. 29 In addition, it has been shown that three-dimensional nanostructures composed of graphene oxide nanolayers and C60 have excellent lubricity at a wide strain level of 0–62%. 30,31 Fig. 1 Typical nanofillers for coating applications such as 0D nanofillers (nano-ZnS and nano-W 2 C particles 52,53 ), 1D nanofillers (carbon nanofiber (CNF) 54,55 ), 2D nanofillers (MoS 2 (ref. 56 ) and graphene nanoplates (GNPs) 57 ), 3D nanofillers (CNF/MoS 2 hybrid 55 and Ti/ND power 58 ). As a nanomaterial, 0D-fillers have been demonstrated to be suitable for a wide range of tribological applications in ceramic, metal and polymer coatings owing to their nearly spherical nanostructure, high thermal conductivity, and low thermal expansion coefficient. 32–34 However, attention has gradually shifted from 0D nanofillers to candidates in other dimensions with the introduction of multi-dimensional nanomaterials. Similar to 0D-nanofillers, 1D- and 2D-fillers have stimulated extraordinary interest in tribology. 1D-nanofillers with high aspect ratios and unique layered 2D-nanofillers are considered ideal materials for achieving excellent lubricity. 2,35–37 The enhancement of the tribological behaviors of nanofiller-reinforced coatings mainly depends on their inherent characteristics, such as their micro/nanoscale structure, dispersion state, and bonding strength of the resulting composite. 38 The nano-rolling effect of 1D-nanofillers has been proven to be an effective strategy for reducing the friction and wear of coatings. Multilayer 2D-nanomaterials are combined through weak van der Waals forces, which can result in a relatively low shear strength, 39–41 endowing 2D-filler nanosystems with excellent self-lubricating performances. In contrast to 0D, 1D and 2D nanofillers, 3D-nanofillers are defined as pure 3D nanomaterials or hybrid mainly composed of one or more 0D, 1D, and 2D structural units. 42,43 The use of well-designed 3D structures presents a feasible strategy for enhancing the tribological behaviors of surface coatings, enabling the applications of their composites in many important fields, such as electricity, medical science, and superlubricity. As rapidly developing nanotechnology globally, nanofillers have attracting interest in the design of solid lubrication coatings due to their unique structures and excellent physical/chemical properties. Compared with micro-scale additives, nano-additives have the advantages of smaller particle size and larger specific surface area. Furthermore, they are not only capable of improving the interface compatibility between additives and solid matrices, but also easily enter the tribo-interface during the friction process, providing excellent lubrication effects. 44 Recently, the widespread use of nanofillers in coatings has promoted the design of friction interface optimization strategies, playing an important role in improving the friction performance and reducing the energy consumption. Numerous studies have shown that multi-dimensional nanofillers can serve as the solid lubrication components of the coatings 45–48 to form friction films during the friction process. This helps to reduce friction and wear. To date, the studies on coatings have mainly focused on their synthesis and potential applications, 49–51 whereas the tribological functions of nanofillers in coatings have not been reviewed fully. In particular, insight into the crucial roles of multi-dimensional nanomaterials in different coating substrates has not been provided thus far. In this review, we focus on the tribological performance of coatings in three key areas. It is divided into the following five parts. After a brief introduction on the macro/microstructures and properties of nanofillers, the tribological behaviors and wear mechanisms of 0D-nanofillers as effective additives for ceramic, metal, and polymer coatings are described in Section 2. Additionally, the recent progress in the development of 1D and 2D nanofillers for coating tribology is elaborated in Section 3 and 4, respectively. In section 5, we explore the increasing efforts on 3D-nanofillers to realize superior anti-wear and friction-reduction behaviors. Finally, the important conclusions and outlooks of nanofillers in crucial tribological fields are discussed in Section 6."
} | 6,256 |
37609449 | PMC10440698 | pmc | 2,845 | {
"abstract": "The potential low-energy feature of the spiking neural network (SNN) engages the attention of the AI community. Only CPU-involved SNN processing inevitably results in an inherently long temporal span in the cases of large models and massive datasets. This study introduces the MAC array, a parallel architecture on each processing element (PE) of SpiNNaker 2, into the computational process of SNN inference. Based on the work of single-core optimization algorithms, we investigate the parallel acceleration algorithms for collaborating with multi-core MAC arrays. The proposed Echelon Reorder model information densification algorithm, along with the adapted multi-core two-stage splitting and authorization deployment strategies, achieves efficient spatio-temporal load balancing and optimization performance. We evaluate the performance by benchmarking a wide range of constructed SNN models to research on the influence degree of different factors. We also benchmark with two actual SNN models (the gesture recognition model of the real-world application and balanced random cortex-like network from neuroscience) on the neuromorphic multi-core hardware SpiNNaker 2. The echelon optimization algorithm with mixed processors realizes 74.28% and 85.78% memory footprint of the original MAC calculation on these two models, respectively. The execution time of echelon algorithms using only MAC or mixed processors accounts for ≤ 24.56% of the serial ARM baseline. Accelerating SNN inference with algorithms in this study is essentially the general sparse matrix-matrix multiplication (SpGEMM) problem. This article explicitly expands the application field of the SpGEMM issue to SNN, developing novel SpGEMM optimization algorithms fitting the SNN feature and MAC array.",
"introduction": "1. Introduction Coupling spatial and temporal information, the SNN shows promise in simulating biologically related models more comprehensively and efficiently. The CPU-based system is widely used for simulating these brain-inspired neural networks by taking advantage of flexibility. However, the efficient input spike encoding way is still in the exploration stage, and a gap still exists between the current encoding efficiency and that of the human brain, which reduces the expected sparsity of the input signal and extends the CPU running time. Moreover, to accommodate the serial operation mechanism, the model needs to introduce additional information when deployed to the hardware, such as the storage address of neurons, extra memory occupation owing to non-equivalent connections, and intermediate state storage buffers. This not only burdens the memory space but also inevitably requires more time to execute the corresponding pre- and post-neuron matching algorithm for information transfer and neural update, which is detrimental to the operation of real-time SNN inference. To address these issues caused by pure CPU systems, we introduced the parallel computing concept into SNN inference. The feasibility of parallel architecture processing SNN lies in the neurons of SNN being typically governed by the same type of equations (Yavuz et al., 2016 ). As a result, the single-instruction-multiple-data (SIMD) architecture of MAC fits SNN calculation. This study targets a wholly parallel calculation based on the more efficient matrix parallelism. We emulate the SNN inference on SpiNNaker 2 (Mayr et al., 2019 ), which integrates the MAC array in each processing element (PE). The parallelism of this integrated hardware component has the potential of speeding up SNN inference in a sufficiently parallel manner. Nevertheless, there are two challenges to tackle: Memory alignment : The memory alignment for catering to the MAC array architecture triggers an issue of data volume surges, blocking the possibility of deploying more neurons and synapses on limited hardware resources. Multi-core distribution : The unconsidered multi-core distribution of the large-scale model can differentiate the spatial-temporal overhead among activated PEs, wasting the resources in space and time and affecting the performance of applications with strict requirements. This study addresses these two challenges by lossless densifying the memory-aligned model information and splitting matrix multiplication operands into multiple PEs in a spatial-temporal load-balancing way. Essentially, accelerating SNN inference with our algorithms is the SpGEMM problem, as explained in Section 2.2. SpGEMM is very popular in high-performance computing, mainly used in algebra and graph analysis (Gao et al., 2020 ). The vast majority of the relevant studies, such as Davis ( 2018 ), Zhang et al. ( 2020 ), and An and Çatalyürek ( 2021 ), are based on the “row-wise” algorithm proposed by Gustavson ( 1978 ), also known as compressed sparse row format (CSR) or Yale sparse matrix format. This traditional algorithm is unsuitable for using MAC array accelerating SNN, so we propose a brand-new optimization algorithm set, which can accelerate the SNN processing when alleviating the ineffective memory footprint. This algorithm set, consisting of four algorithms up to now, provides an alternative to the traditional method for solving the SpGEMM problem. To the best of our best knowledge, our work is the first to build a bridge between the concept of SpGEMM and SNN, expand the application field of SpGEMM to SNN, and tackle the SNN inference using the MAC array with new SpGEMM algorithms. As a follow-up to our previous study that states three algorithms of information densification (Huang et al., 2023 ), this study proposes Echelon Reorder, filling in the unoptimized aspects of that work, completing the optimization algorithm set to fully resist the data sparsity caused by the SNN characteristics and fixed MAC array hardware structure. The corresponding splitting and deployment strategies proposed in this study extend the application range of the whole optimization algorithm set from single PE to multi-core, enabling accelerating the larger model on SpiNNaker 2 effectively. Furthermore, the compact splitting strategy fully uses each PE's memory resource, paving the way for the subsequent high-performance multiple tasks deployment on this multi-core neuromorphic platform. This study briefly introduces the hardware and software cornerstones in Section 2. Then, based on them, we elaborate on the Echelon Reorder algorithm for weight and input pure and mixture processor splitting strategies and also multi-core role-based SNN model deployment in Section 3. Next, Section 4 evaluates the performance of this proposed processing chain. Finally, we conclude this article in Section 5.",
"discussion": "5. Conclusion and discussion 5.1. Summary This study describes the processing chain for accelerating SNN inference with multi-core MAC arrays, including Echelon Reorder information densification algorithm, Multi-core two-stage splitting and multi-core authorization deployment strategies. These algorithms and strategies alleviate the intrinsic memory issue of excessive usage originating from memory alignment and data sparsity. They also realize the multi-core spatial-temporal load balancing for the large SNN layer. We benchmark with constructed and actual SNN models. The former explores how model feature and algorithm selection affect the spatial-temporal optimization performance, and the latter demonstrates two actual SNN models (the radar gesture recognition SNN model and balanced random cortex-like network) on SpiNNaker 2 hardware. They prove the feasibility of the whole optimization process and achieves performance increase. Based on the theoretical analysis and the experiment result, we found those as follows: The proposed algorithms and strategies are applicable to various densities of matrix multiplication operands, and the memory optimization degree increases with the weight operand getting sparse. In addition to the weight sparsity, the memory optimization rate is also positively correlated with delay range. The number of post-neurons periodically affects the memory optimization rate, and the overall trend is downward. The number of pre-neuron is generally independent of the memory optimization performance but has intense correlation with running time. The echelon mixed processor algorithm behaves better regarding memory but has less temporal efficiency than the echelon pure MAC solution. The proposed algorithms not only provide a concrete solution for accelerating SNN on the multi-core MAC arrays of SpiNNaker 2 but also has a referential value for hardware systems embedded with multi-core MAC arrays that intend to solve the SpGEMM issue. 5.2. Related work Some researchers have introduced the parallel computing concept into SNN inference to tackle the problems caused by CPU-based parallel processing. One of the representative works of parallel processors accelerating SNN computation is GeNN (Yavuz et al., 2016 ), a code generation framework speeding up the SNN simulation process using the graphics processing unit (GPU). Specifically, it speeds up the synaptic processing by utilizing a serially executed atomic add operation to add weight to the delay ring-buffer after reading the index of post-neuron and weight and delay by each thread of the GPU (Yavuz et al., 2016 ; Knight and Nowotny, 2018 ). However, this mixture of thread/vector parallel processing and serial add operation does not take full advantage of the parallelism of the processor. Another related study regarding parallel processing SNN inference is SpiNeMap (Balaji et al., 2020 ) and its follow-up (Balaji et al., 2021 ). SpiNeMap is a design methodology to partition and deploy SNNs to the crossbar-based neuromorphic hardware DYNAP-SE (Moradi et al., 2018 ). The proposed unrolling techniques decompose a neuron function with many presynaptic connections into a sequence of computation nodes with two presynaptic connections. This approach improves the crossbar utilization but introduces spike latency, i.e., distorts interspike intervals (ISIs) for global synapses. This issue can be relieved by reducing the number of spikes on global synapses as reported in Balaji et al. ( 2020 ), probably realized by modifying the model parameters or decreasing the input spike rate, both of which can negatively impact the accuracy of the original SNN. Our study targets no ISIs, no accuracy loss, and a wholly parallel calculation based on the more efficient matrix parallelism rather than vector-based computing. 5.3. Macro significance The proposed algorithms act on synaptic processing, which is part of the SNN inference. The impact of algorithms on the optimization of the whole network is a question worth discussing. To review the motivation of our study, it is discussed in Section 1 that the serial ARM baseline suffers from a large time-consuming issue and the naive parallel solution original MAC is unfriendly to limited memory space, it is necessary to explore the portion of synaptic processing of serial ARM in time and original MAC in memory in order to have a macro view of the degree of optimization of our proposed algorithms in the whole SNN inference. SNN inference consists of synaptic processing and neural update. We estimated the time consumption rate of these two steps for the serial ARM baseline with the following equation referring to ARM Cortex-M4 Technical Reference Manual (ARM, 2023 ): \n (4) \n t s y n a p t n e u r a l = n p r e × n p o s t × d × M L A n p o s t × ( M L A + C O M P ) ≊ 0.67 × n p r e × d . \n This equation shows that the time consumption rate depends on the number of pre-neuron and delay range. A meaningful model is supposed to have more than one pre-neurons, so this rate is always greater than 1. For the two benchmark actual models in Section 4.2, the synaptic processing time is 5,488.64 and 2,144.0 times larger than neural update. As for the memory comparison, suppose that the number of pre- and post-neuron is memory aligned, the decay and threshold values of the neural update of all neurons are identical, and only count one output matrix for the multi-core scenario, we have the following equations: \n (5) \n m e m s y n a p m e m n e u r a l = 4 × n p r e + n p r e × n p o s t × d + 4 × n p o s t + 4 × n p o s t 4 × n p o s t + 0.125 × n p o s t ≊ 1.94 + 0.24 × n p r e × ( 4 n p o s t + d ) . \n This equation provides the rate of memory cost of synaptic processing to neural update for the original MAC baseline. The rate always greater than 1.94 means synaptic processing dominates memory cost. For the two benchmark actual models, this rate indicates the memory cost of synaptic processing is 2,066.32 and 773.78 times larger than neural update, respectively. The above search confirms that synaptic processing has the lion's share in SNN inference, both in terms of running time and memory storage. Therefore, the optimization of synaptic processing is crucial and largely determines the optimization performance of the whole SNN inference. 5.4. Limitation and future work By analyzing formula 2 and comparing two actual models regarding the temporal performance of synaptic processing, we found that the larger number of pre-neuron hinders the running time optimization degree that is primarily governed by ARM calculation parts of the Echelon Reorder and matrix-multiplication of the echelon mixed processor approach. Future optimization for time can be placed on finding a more efficient algorithm for the Echelon Reorder and further compressing the data of the serial operation range in the echelon mixed processor approach. The spatial performance of our proposed optimization algorithms is limited by the smaller delay range, denser weight connection, and the larger number of post-neuron. How to further optimize the memory optimization rate r opt and r align _ c is defined in Section 4.1 and Section 4.2.3 and increasing the area of the white area of Figure 11 are the next things that are worth investigating. This study merely elaborates on the optimization mechanism of the Echelon Reorder that is based on the data transformation of the Operand Stack. In fact, the other two optimization algorithms (Zero Elimination and Proportion Merger) in Table 1 proposed in our previous study (Huang et al., 2023 ) of one PE accelerating SNN inference also fit multiply PEs. The memory optimization will benefit from their synergy. Figure 11 Summary of the memory cost differences of delayed weight of the four approaches (two baselines and two proposed approaches) with three “rate.” The memory alignment rate of row and column r align _ rc equals the memory aligned area in the weight matrices of delay x from Figure 2 , and other two “rate” represent the proportions of the pointed white area in the whole area of weight-delay map after using the Echelon Reorder. Although Section 3.3 provides the solution of authorizing PEs that contain the split echelon matrix and input buffer, an efficient multi-core deployment strategy and routing algorithm are not included in this study. Randomly deploying the six cores from Figure 8 on SpiNNaker 2 may cause a relatively low-efficient communication between the “Dominant PE” and the “Subordinate PE.” This issue will be of greater concern when we extend the deployment of one SNN layer to an entire network. At that time, a reasonable multi-layer deployment topology and global routing algorithm can avoid the potential traffic congestion and reduce the communication latency by fully utilizing the bandwidth resources of PE to PE and PE to DRAM. Thus, improving the spatio-temporal efficiency of the entire SNN even multiple SNNs on SpiNNaker 2 will be one of our future research priorities. The traditional serial processing for SNN inference, as the current mainstream method, is constantly being optimized and iterated, and the performance has been improving. It has a good performance in the condition of very sparse input spike train and weight-delay operand, which is what the approaches proposed in this study is yet to be improved. If we can find the sweet spot of SNN model structure factors including input and weight connection density between the traditional approach and our algorithms, it will help the neuromorphic community to have a deeper understanding of the serial and parallel processing methods and contribute to the mechanism of the hybrid processors jointly processing SNN inference in a more efficient way."
} | 4,113 |
33850944 | PMC8035421 | pmc | 2,846 | {
"abstract": "Background: Lignin is the largest natural aromatic polymer in nature and is also a unique aromatic-based biopolymer, accounting\nfor nearly 30% of the earth’s organic carbon. Generally, lignin is regarded as waste and is mainly used as a low- value\nfuel that is burned to generate heat and energy to solve the problem of biomass waste; for this obstacle of lignin,\nhighly efficient biodegradation plays a critical role in developing an environmentally friendly technique for lignin biotransformation. Objectives: This study intends to isolate and purify several microbial strains from nature. It also explores how their lignin degradation\nis able to enhance the biodegradation and recycling of biomass and the reclamation of lignin in wastewater from pulp and paper mills. Materials and Methods: Lignin-degrading microbial strains were isolated from soil using medium containing sodium lignosulphonate as the sole carbon\nsource. They were then screened by aniline blue and guaiacol plate, and then the best strain was chosen and identified.\nThe conventional one-factor method was used to optimize various parameters that affect lignin’s degradation ability. Results: The strain possessing the highest lignin biodegradation ability was identified and denominated as Aspergillus Flavus \nF-1. After optimization, the maximum degradation rate of lignin, 44.6% within 3 days, was obtained at pH 7.0, 30 ℃, 2.5 g·L -1 ammonium\nsulfate, 2 g·L -1 lignin and 0.5 g·L -1 glucose. The results show the LiP and Lac secreted from Aspergillus Flavus \nF-1 played the main role in the degradation of lignin. Conclusion: One microbial strain, Aspergillus Flavus F-1, was successfully isolated with a lignin-degrading ability\nthat can cut the lignin into fragments. This provides a promising candidate for the transformation and utilization of crop waste biomass for various industrial purposes.",
"conclusion": "6. Conclusion The strain of Aspergillus flavus F-1 from soil presented ligninolytic enzyme activities and could degrade 44.6% of the lignin into fragmentation at pH 7.0 and 30 °C in 3 days. Aspergillus flavus F-1 mainly produced lignin peroxidase and laccase. It could be Lip that played a major role in lignin degradation at first and then the cooperation of two enzymes that especially worked in lignin degradation after three days. The degradation characteristics of the strain showed that Aspergillus flavus strain F-1 could be a nitrogen-deficient strain, and at lower nitrogen source concentration it was slightly better than that at higher nitrogen source concentration. Aspergillus flavus F-1 proved a good candidate for the transformation and utilization of crop waste resources. At the same time, this work also provided a feasible way for the utilization of lignin in papermaking black liquor.",
"discussion": "5. Discussion The F-1 strain grows well in the PDA medium, and the color of the colony on the surface ranges from white,\nlight pink and gray to blue-green. Microscopic observations of appearance ( Fig. 1B ) shown that the mycelium\nis transparent, branched and separated, with smooth walls and a width of 3-12 μm. As seen in Figure 1C ,\nthe inverted microscope (10×40) showed that the mycelium had a septum, was branched and had no false root;\nthe spore pedicle was erect; the terminal sporangium had a rough surface of spherical conidial spore\nformed on the terminal sporangium; and the sporangium was flask or nearly globose. The sequence has been registered in the NCBI GenBank database, and the accession number is KC146411.\nThe BLAST analysis shows that the partial ITS rDNA of F-1 had 100% identity with those of Aspergillus flavus \nisolate F4 (JF951750.1). Based on the phylogenetic analysis, this strain was identified as Aspergillus flavus \nand named Aspergillus flavus strain F-1. Compared with Figure 3A and Figure 3B , the degradation rate reached 30.7% in three days and then increased slowly.\nThe maximum biodegradation of lignin (CL) was up to 36.4% at 12 days of incubation. The degradation of lignin\nincreased with the growth of mycelium during the early and middle growth cycles; meanwhile, in the late growth\ncycles, the degradation degree still increased when Aspergillus flavus stopped growing.\nThis phenomenon might have caused a series of free radical chain reactions to triggered the oxidation of\nthe substrate and the degradation of lignin by ligninolytic enzymes of strain F-1 while the nutrition was exhausted. Previous reports showed that white-rot fungi were the most effective microorganisms in nature to degrade lignin,\nand the degradation rate was 30%~40% when the longest culture period is 30 days (19). In contrast, the lignin\ndegradation rate of Aspergillus flavus F-1 reached 36.4% in only 3 days, which greatly reduced\nthe culture cycle. Moreover, the growth rate of Aspergillus flavus F-1 was fast, and it was\neasy to culture. As a result, Aspergillus flavus F-1 has the potential to be a highly effective lignin-degrading strain. It has been shown that fungi that produce laccase and Lip or only laccase can efficiently degrade lignin.\nLignin is degraded by side-chain oxidation, demethylation/ demethoxylation and aromatic ring breaking under the action of this degradation enzyme system. As shown in Figure 3C , MnP activity was not detected in the strain F-1, and the activity of Lip increased quickly\nand reached its peak at 3 days, while the activity of Lac increased extremely slowly in the early and middle growth\ncycles and only occurred actively in the late growth cycle. It was clear that Lip played a major role in lignin degradation\nduring the first three days. Three days later, although the activity of Lip decreased, the activity of Lac increased slowly\nand steadily. Therefore, the cooperation of two enzymes could have caused lignin degradation after three days, which made\nthe rate increase, as shown in Figure 3 (b) . It can be concluded that Lip and Lac were crucial factors for lignin degradation\nof Aspergillus flavus F-1. Experiments show that inorganic nitrogen sources have higher degradation\nefficiency for lignin. Because inorganic nitrogen sources are available nitrogen sources, they are more conducive\nto the utilization of strains. It was also possible that the strain mainly used the organic nitrogen source as a carbon\nsource at the early stage of growth and only began to decompose and utilize the sole carbon source, sodium lignosulfonate,\nafter nitrogen source consumption, which resulted in the existence of competitive inhibition of substrate. It can be seen from Figure 4B that there was no significant difference\nin lignin degradation of Aspergillus flavus F-1 under different nitrogen source concentrations,\nbut the overall trend showed that the degradation effect of Aspergillus flavus F-1 at lower nitrogen\nsource concentrations was slightly better than that at higher nitrogen source concentrations. The suitable substrate concentration was more favorable to the degradation of lignin. Because of its surfactant\nproperties, sodium lignosulfonate adsorbed on its surface and was surrounded by complex reticular structure,\nwhich was not favorable for the absorption of a small amount of glucose to promote growth, resulting in\nslow growth. Figure 4B shows that the lower the substrate content, the higher the degradation rate,\nreaching about 30%. However, while the substrate concentration was higher, the degradation system became\nviscous because of the colloidal property of sodium lignosulfonate swelling in the medium, which affected the dissolution and transfer of oxygen. The degradation rate of lignin showed a trend of first increasing and then decreasing with the increase in temperature.\nIt can be seen from Figure 4C that the degradation rate of lignin was the highest at 30 °C. When the temperature was\nlower than its optimal temperature, the metabolism of microorganisms became active with the increase in temperature,\nand when the temperature was higher beyond its optimal temperature, the microorganisms would die, leading to the lower degradation rate of lignin. Neutral pH greatly affects the growth of microorganisms and the membrane transport of nutrients. Studies\non optimizing the pH of the medium show that the maximum lignin degradation rate is reached at pH 7.0, which may\nbe due to the extreme pH directly affecting the pH of the microbial cell cytoplasm, which in turn affects the growth\nand enzyme productivity of the microbe. As shown in Figure 4D , the strain F-1 could degrade lignin in a wide range\nof acid–base culture conditions. When the pH of the culture system was low, the growth of the strain was seriously\ninhibited. It may be the case that the environment of strong alkali or strong acid destroyed cell membrane permeability\nand intracellular enzyme activity, thus affecting the metabolism of cells and adversely affecting the utilization\nof lignin by the strain. Therefore, pH 7.0 was selected as the degradation medium’s initial pH, and under the best conditions,\nit could cause 44.6% of the lignin to fragment within 3 days."
} | 2,269 |
28947739 | PMC5705400 | pmc | 2,847 | {
"discussion": "Discussion The work described herein identified an operon that was essential for assimilating LA into the β-oxidation pathway of P. putida . Through an integrated genetic and in vitro biochemistry study, we demonstrated that the genes lvaABCDE were upregulated in the presence of LA and were sufficient for the conversion of LA to 3HV-CoA, an intermediate of native β-oxidation. Removing any enzyme from the reaction mixture abolished 3HV-CoA production, indicating all five enzymes were necessary for this pathway. The biochemical assays confirmed the presence of 4PV-CoA, an intermediate previously observed in the metabolism of LA in rat livers. In sum, the pathway consumed at least 2 ATP and one reducing equivalent to produce 3HV-CoA ( Figure 1e ). β-oxidation of 3HV-CoA to acetyl-CoA and propionyl-CoA would recover the reducing equivalent. Given the energy demands of the pathway, growth on LA must be performed aerobically or in the presence of an alternative electron acceptor to enable ATP synthesis via respiration. Like many catabolic pathways, expression of the lva operon is regulated by the presence of the pathway substrates. Using a transcriptional reporter assay, we demonstrated that the lva operon is upregulated by a transcriptional activator encoded by the divergent lvaR gene. Additionally, we suspect that the lva operon is also regulated by Crc, a global carbon catabolite repressor. Crc is an mRNA binding protein that prevents protein translation when bound to a specific mRNA sequence in P. putida , AAnAAnAA 37 , 38 . This sequence pattern is found immediately upstream of lvaE ( Supplementary Figure 1d ), which encodes an acyl-CoA synthetase that initiates the pathway. The presence of the Crc target sequence suggests that the operon is also subject to P. putida’s carbon catabolite repression system which may explain the diauxic growth curves observed for mixtures of glucose and LA. The lva operon is highly conserved among the various Pseudomonas species ( Supplementary Table 5 ). Gene clusters comprised of the main enzymatic proteins can also be found in a variety of alpha-, beta- and gamma-proteobacteria, graphically represented in Figure 4 . The alpha-proteobacteria species ( Azospirillum, Bradyrhizobium, Rhodopseudomonas, Sphingobium ) are primarily isolated from soil environments, similar to Pseudomonas putida . The beta-proteobacteria species ( Azoarcus, Limnobacter ) and the gamma-proteobacteria species ( Acinetobacter, Marinobacter ) are isolated from both soil and ocean environments. Supplementary Table 5 lists all species that were found to contain individual homologs to LvaABCD positioned throughout the genome. Supplementary Table 6 lists all species that contain LvaACD homologs. Further investigation into the utilization of LA by these species could help determine whether spatial relationship of the lva operon genes is important. While LvaB was shown to be essential for LA catabolism, its exact role remains unclear. LvaB is a small protein (~100 aa) that is unlikely to contain enzymatic activity by-itself. Furthermore, LvaB co-purifies with LvaA, is essential for the phosphorylation of 4HV-CoA, and its orthologs are consistently found adjacent to orthologs of lvaA in the genomes of other organisms. Similar examples where small proteins provide critical support to enzymatic function have been indentified 39 , 40 . For example, nonribosomal peptide synthetase gene clusters often contain a small protein that belongs to the MbtH-like protein family, a family of proteins that are known to bind adenylation domains and enable catalytic activity. MbtH-like proteins form the necessary complexes required for domain activation but are not predicted to interact directly with the catalytic site 41 , 42 . Although LvaB does not share significant sequence homology with known MbtH-like proteins, we speculate that it could be playing a similar role with LvaA, where the presence of LvaB is required to form an active LvaAB complex. Without a crystal structure, the specific interaction between LvaA and LvaB and its role in catalysis will be difficult to unlock. Interestingly, the isomerization of 4HV-CoA to 3HV-CoA in P. putida proceeds through a phosphorylated intermediate, 4PV-CoA, a compound also observed in a study of LA metabolism in rat livers 20 . This study suggested the 3HV-CoA was generated via a pathway comprised of complex phosphorylated intermediates. We did not detect MS peaks corresponding to any of these compounds in our in vitro reaction mixtures. Instead, based on changes we observed in total ion abundance over time, we propose that 4PV-CoA is dephosphorylated to an enoyl-CoA and subsequently rehydrated to 3HV-CoA. We suspect that the phosphorylation of 4HV-CoA by LvaAB generates a better leaving group and makes the subsequent dehydration more thermodynamically favorable. However, the mechanism for these last steps remains unclear. Previous groups studying the nonmevalonate pathway have identified phosphate elimination steps for the formation of a double bond that is reminiscent of the intermediates we observed 43 , 44 , but these reactions do not include a rehydration step. The timecourse measurements collected for the full reaction indicate that the formation of the pentenoyl-CoA happens quickly, but the transition from the pentenoyl-CoA to the 3HV-CoA is a much slower reaction ( Figure 2f ). Our tests indicate that LvaC is capable of converting 4PV-CoA to 3HV-CoA, but those reactions still contain a higher abundance of pentenoyl-CoA compared to 3HV-CoA. A more detailed mechanistic study of the final steps may clarify the specific role of LvaC. We selected the E. coli strain LS5218 for our complementation studies because it constitutively expresses enzymes involved in beta-oxidation and catabolism of organic acids. Interestingly, this choice of host led to the requirement of additional mutations in fadE and atoC to permit robust growth on LA. These deletions likely prevent detrimental side reactions catalyzed by FadE and AtoDA (activated by AtoC(con) 45 , 46 ) that would compete with the desired catabolic flux to central metabolism. FadE is an acyl-CoA dehydrogenase that catalyzes the formation of a trans-2-enoyl-CoA from an acyl-CoA 47 . We hypothesize that FadE, which is upregulated in FadR-deletions, may act on LA-CoA, adding a double bond between the 2- and 3-positions of the γ-ketovaleryl-CoA species and sequestering the molecule from further metabolism. Unfortunately, FadE is an inner membrane protein 48 , 49 that has not been purified or characterized in vitro . While a deletion of atoC was not a necessary mutation, it did confer a growth benefit. We suspect that this mutation was isolated during through the directed evolution process because we were screening for mutants with rapid initial growth. Constitutive activation of the ato regulon by the atoC (Con) mutation in LS5218 causes overexpression of an acetoacetyl-CoA transferase (encoded by atoDA ), an acetyl-CoA acetyltransferase (encoded by atoB ) and a short chain fatty acid transporter (encoded by atoE ) 46 , 50 . We suspect that 3-ketovaleryl-CoA, the product of FadB acting on 3HV-CoA, was diverted away from the desired FadA reaction by increased AtoDA activity that released 3-ketovalerate. This sequestration of LA as 3-ketovalerate would reduce overall carbon flow to central metabolites and stunt growth until cells adapt to consume 3-ketovalerate. Reducing expression of AtoDA through the deletion of atoC would prevent the shunt pathway and allow direct flux of LA to central metabolites. Further investigation of the competing metabolic pathways will be critical to developing LA-based bioconversions."
} | 1,945 |
30151429 | PMC6108567 | pmc | 2,848 | {
"abstract": "First springtail-inspired omniphobic surface by hierarchical structure to repel liquids even with high pressure of droplets.",
"introduction": "INTRODUCTION Springtails are small soil-dwelling arthropods that have remarkable cuticles with intrinsically omniphobic surfaces displaying both static repellency and pressure resistance to impacts of drops such as raindrops. The geometry of their cuticles has been evolutionarily adapted to avoid complete wetting by water and organic liquids, features critical for their survival because they respire through the skin. For springtails, static repellency induces self-cleaning, which maintains dry cuticles, while pressure resistance plays a significant role in providing resistance to dripping liquids. These unique characteristics result from the hierarchical structures of the cuticle, composed of mushroom-shaped nanostructures as primary granules and microscale grooves as secondary granules ( Fig. 1A ) ( 1 – 4 ). It has been recently reported that the nanoscale doubly reentrant primary granules play a critical role in omniphobicity because they can provide energetically stable pinning points with liquids having low surface tensions ( 1 , 3 ). Although reentrant structures have been applied on the microscale to achieve static omniphobicity, a hierarchical omniphobic surface inspired by the springtail with both static repellency and high robustness to liquid pressure has yet to be realized. Previously reported omniphobic surfaces have not simultaneously achieved both features because of a general trade-off between these characteristics ( 5 – 8 ). Here, we took a significant step toward overcoming the limitations of previous biomimetic omniphobic surfaces. Fig. 1 Rational design of a hierarchical system inspired by the springtail cuticle. ( A ) Photograph (courtesy of B. Valentine) of a springtail displaying liquid repellency and resistance to high-pressure raindrops in a flooded habitat (left). SEM images showing the hierarchical system in a springtail cuticle composed of primary and secondary granules (middle and right panels). ( B ) Schematic of the steps used to fabricate serif-T–shaped nanostructures. Nanoimprinting and the SSL method were used here. The serif-T–shaped nanostructures were made with ~400-nm-diameter dots and ~ 400-nm spacing between the dots, both dimensions similar to those for the primary granules of springtail cuticles. ( C ) Scheme to fabricate microscale wrinkles via heat-induced shrinkage after nanostructure fabrication.",
"discussion": "DISCUSSION The hierarchical system we designed showed outstanding performance in terms of both static omniphobicity and pressure resistance even with low-surface-tension liquids. These results were attributed to the presence of sufficient air pockets both between the serif-T–shaped nanostructures and in the microscale grooves that together can provide a highly stable Cassie-Baxter regime. In addition, the serif-T geometry with doubly reentrant features was highly effective at providing a robust Cassie-Baxter regime, much more so than the other geometries despite their having the same air pocket dimensions. The microstructures resulting from the wrinkles effectively reduced the solid-liquid contact fraction of the entire surface, which was also highly important for achieving robustness. In addition, because the fragmentation of the liquids presumably originated from air layers at the solid-liquid interface, we concluded that the valley-like microstructures contained additional air pockets based on our observation of earlier commencement of fragmentation for the highly wrinkled surface as We increased ( Fig. 4 , A and C) ( 24 , 25 ). Finally, the presence of additional air pockets in the grooves of the highly wrinkled surfaces helped to reduce energy loss during spreading and retraction on the surface after droplet impact, leading to the drops bouncing off the surface to a greater a height than from a flat surface. As a result, the artificial springtail surface, specifically the highly wrinkled surface studded with serif-T–shaped nanostructures, provided extreme pressure resistance with the liquids we tested. In conclusion, we were inspired by the structures and properties of the cuticles of springtails to design and fabricate an artificial hierarchical system and verified its outstanding omniphobic characteristics. We fabricated this system by using a combination of lithographic tools and a wrinkling method. This fabrication for the most part yielded both the nanoscale and microscale features and the dimensions of these features as found in the springtail cuticle. In most existing omniphobic surfaces showing only features on the microscale, there has been an inevitable trade-off between static repellency and wetting resistance to pressure, with few investigations showing a stable Cassie-Baxter regime to both static repellency (apparent contact angle, low contact angle hysteresis) and pressure resistance (robustness upon being impacted with droplets at high velocity) ( 32 ). Our novel design showed high omniphobicity resulting from the doubly reentrant features of serif-T–shaped nanostructures and the hierarchical system. Although the role of the primary nanostructures of the springtail cuticle in omniphobicity is clear, the influence of microgrooves has yet to be properly elucidated. Hence, the inclusion of these microstructures in our hierarchical system may help to shed light on their role in the springtail system. Among the hierarchical systems we fabricated with different nanostructures, including disk-, overhang-, and serif-T–shaped structures, we demonstrated the serif-T–shaped nanostructure and a highly wrinkled surface to be the most advantageous. We attributed these results to these geometries being most effective at reducing the fraction of the surface in contact with the liquid and providing for a sufficient amount of air pockets. We believe this fabricated system to be one of the most optimally designed springtail-inspired systems for creating a superomniphobic surface with extreme pressure resistance."
} | 1,521 |
36426084 | PMC9679665 | pmc | 2,849 | {
"abstract": "Psychrophilic methanotrophic bacteria are abundant and play an important role in methane removal in cold methanogenic environments, such as boreal and arctic terrestrial and aquatic ecosystems. They could be also applied in the bioconversion of biogas and natural gas into value-added products (e.g., chemicals and single-cell protein) in cold regions. Hence, isolation and genome sequencing of psychrophilic methanotrophic bacteria are needed to provide important data on their functional capabilities. However, psychrophilic methanotroph isolates and consequently their genome sequences are rare. Fortunately, Leibniz Institute, DSMZ-German Collection of Microorganisms and Cell Cultures GmbH was able to revive the long-extinct pure culture of a psychrophilic methanotrophic tundra soil isolate, Methylobacter psychrophilus Z-0021 (DSM 9914), from their stocks during 2022. Here, we describe the de novo assembled genome sequence of Methylobacter psychrophilus Z-0021 comprising a total of 4691082 bp in 156 contigs with a G+C content of 43.1% and 4074 coding sequences. The preliminary genome annotation analysis of Z-0021 identified genes encoding oxidation of methane, methanol and formaldehyde, assimilation of carbon and nitrate, and N 2 fixation. In pairwise genome-to-genome comparisons with closely related methanotrophic strains, the strain Z-0021 had an average nucleotide identity (ANI) of 92.9% and 78.2% and a digital DNA-DNA hybridization (dDDH) value of 50.6% and 22% with a recently described psychrophilic, lake isolate, Methylobacter sp. S3L5C and a psychrotrophic, arctic wetland soil isolate, Methylobacter tundripaludum SV96, respectively. In addition, the respective similarities between genomes of the strains S3L5C and SV96 were 78.1% ANI and 21.8% dDDH. Comparison to widely used ANI and dDDH thresholds to delineate unique species (<95% ANI and <70% dDDH) suggests that Methylobacter psychrophilus Z-0021, Methylobacter tundripaludum SV96 and Methylobacter sp. S3L5C are different species. The draft genome of Z-0021 has been deposited at GenBank under the accession JAOEGU000000000."
} | 532 |
29988523 | PMC6026345 | pmc | 2,850 | {
"abstract": "In the wake of the uprising global energy crisis, microalgae have emerged as an alternate feedstock for biofuel production. In addition, microalgae bear immense potential as bio-cell factories in terms of producing key chemicals, recombinant proteins, enzymes, lipid, hydrogen and alcohol. Abstraction of such high-value products (algal biorefinery approach) facilitates to make microalgae-based renewable energy an economically viable option. Synthetic biology is an emerging field that harmoniously blends science and engineering to help design and construct novel biological systems, with an aim to achieve rationally formulated objectives. However, resources and tools used for such nuclear manipulation, construction of synthetic gene network and genome-scale reconstruction of microalgae are limited. Herein, we present recent developments in the upcoming field of microalgae employed as a model system for synthetic biology applications and highlight the importance of genome-scale reconstruction models and kinetic models, to maximize the metabolic output by understanding the intricacies of algal growth. This review also examines the role played by microalgae as biorefineries, microalgal culture conditions and various operating parameters that need to be optimized to yield biofuel that can be economically competitive with fossil fuels.",
"conclusion": "Conclusions Microalgae bear immense potential as bio-cell factories in terms of producing key chemicals, recombinant proteins, enzymes, lipid, hydrogen, alcohol etc. Abstraction of such high-value products (algal biorefinery approach) facilitates to make microalgae-based renewable energy as an economically viable option. Synthetic biology is an emerging field that harmoniously blends science and engineering to help design and construct novel biological systems, with an aim to achieve rationally formulated objectives. The microbial genetic information, which is easily amenable to modification via metabolic engineering, systems biology and pathway reconstruction coupled with synthetic biology, allows researchers to produce required biomolecules. However, resources and tools used for such nuclear manipulation, construction of synthetic gene network and genome-scale reconstruction of microalgae are limited. The use of synthetic biology in algal biofuel production is still in its infancy, wherein challenges in the development of more advanced genetic tools, high biomass and improved CO 2 fixation capacity need to be resolved. In addition to the aforementioned, novel consolidated bioprocess wherein a single microbe can generate renewable biofuel as an alternative to depleting fossil fuels is the need of the hour. This review also examines the role played by microalgae as biorefineries, microalgal culture conditions and various operating parameters that need to be optimized to yield biofuel that can be economically competitive with fossil fuels. To summarize, algal biorefinery in the present state strategically produces multiple products—bulk and specialized co-product to increase the total revenue from cultivation and make the bulk production economically feasible. In practice, multiproduct large-scale biorefinery still poses several problems, which need to be soon addressed. Some economically important value-added products such as astaxanthin and oil; lutein and oil; EPA and oil are being produced in lab scale. Attempts should be made to facilitate their large-scale production through proper pathway prediction and growth and metabolic modeling, wherein the domains of systems biology and process optimization will play a crucial role. Process optimization can lead us to optimized combination of growth conditions such as light, temperature, mass transfer, reactor configuration, media formulation and supplementation, which produces improved growth kinetics. Systems biology, on the other hand, can aid in optimal use of carbon and energy balance so that every possible carbon atom is put to use and every ATP is spent judiciously taking into account the metabolism, physiology and induced stress response. This can fructify only when channeling metabolic flux and partitioning are efficient, both of which are predicted by reconstructed genome-scale metabolic models. These in silico models are based on systems biology data that stem from omics and labeling analysis."
} | 1,088 |
35479645 | PMC9037143 | pmc | 2,851 | {
"abstract": "The application of bacterial inoculums for improving plant growth and production is an important component of sustainable agriculture. However, the efficiency of perennial crop inoculums depends on the ability of the introduced endophytes to exert an impact on the host-plant over an extended period of time. This impact might be evaluated by the response of plant-associated microbiome to the inoculation. In this study, we monitored the effect of a single bacterial strain inoculation on the diversity, structure, and cooperation in plant-associated microbiome over 1-year period. An endophyte (RF67) isolated from Vaccinium angustifolium (wild blueberry) roots and annotated as Rhizobium was used for the inoculation of 1-year-old Lonicera caerulea (Haskap) plants. A significant level of bacterial community perturbation was detected in plant roots after 3 months post-inoculation. About 23% of root-associated community variation was correlated with an application of the inoculant, which was accompanied by increased cooperation between taxa belonging to Proteobacteria and Actinobacteriota phyla and decreased cooperation between Firmicutes in plant roots. Additionally, a decrease in bacterial Shannon diversity and an increase in the relative abundances of Rhizobiaceae and Enterobacteriaceae were detected in the roots of inoculated plants relative to the non-inoculated control. A strong effect of the inoculation on the bacterial cooperation was also detected after 1 year of plant field growth, whereas no differences in bacterial community composition and also alpha and beta diversities were detected between bacterial communities from inoculated and non-inoculated roots. These findings suggest that while exogenous endophytes might have a short-term effect on the root microbiome structure and composition, they can boost cooperation between plant-growth-promoting endophytes, which can exist for the extended period of time providing the host-plant with long-lasting beneficial effects.",
"conclusion": "Conclusion Plant growth-promoting endophytes can be used as BCA to reduce the use of fertilizers and pesticides in agricultural systems including production of perennial crops. The efficiency of these BCA depends not only on the ability of the microorganism to promote plant growth but also on their ability to establish symbiosis with the plant and the stability of the introduced microbes in host-plant tissue over the extended growth periods. In this study, we used a single bacterial inoculation to monitor its effect on the soil and plant-associated microorganisms over a 1-year period. We determined that while bacterial inoculations might have a short-term effect on the composition and structure of soils and root-associated microbiomes, they can boost cooperation between plant growth-promoting endophytes inside the plant roots. We demonstrated that this cooperation could exist for an extended period of time. Therefore, the application of BCA might promote the establishment of symbiosis between naturally occurring plant growth-promoting microorganisms and perennial crops and might provide additional benefits for plant health and production.",
"introduction": "Introduction Endophytes can influence plant production by improving plant growth and resistance to biotic and abiotic stresses ( Bulgarelli et al., 2013 ; Busby et al., 2016 ). Plant microbial endophylism is a widespread relationship that often provides mutual benefits for both micro- and macro-symbionts by directly supplying microbial metabolites to the host-plants or stimulating specific plant responses, which leads to increased enzymatic catalysis and defense responses, and also enhancing nutrients and water uptake ( Brader et al., 2014 ; Chaudhry et al., 2021 ). Additionally, some endophytic microorganisms can outcompete phytopathogens by occupying the same ecological niche and preventing or decreasing disease occurrence in plants. This type of endophytes can be used as biofertilizers or biocontrol agents (BCAs) to reduce the use of fertilizers and pesticides in agricultural systems that include the production of perennial crops ( Compant et al., 2013 ; Carvalho et al., 2016 ; Mercado-Blanco et al., 2018 ; Santos et al., 2019 ). For example, several plant-associated bacteria and fungi were found to have a mitigation effect on tree diseases, such as canker ( TF223, 2020 ; Shuttleworth, 2021 ), apple scab ( Köhl et al., 2015 ), and replant diseases ( Duan et al., 2021 ; Wang H.W. et al., 2021 ). Moreover, the application of exogenous synthetic communities composed of naturally occurring, highly abundant plant-bacteria, becomes a new approach to increased biomass and enhanced root system development in agricultural crops ( de Souza et al., 2016 ; Armanhi et al., 2017 , 2021 ). However, achieving the full benefit of the application of BCAs, it is important to understand the extent of ecological effect of the introduction of exogenous microorganisms into environments and its impact on the host-plant over an extended period of time. How plant and soil microbiomes respond to an introduction of exogenous microorganisms is a fundamental question in microbial ecology. It is especially important in light of the increasing use of microbe-based supplements in modern agriculture ( Santos et al., 2019 ). There are a few reports addressing this question. For example, it was shown that inoculation with exogenous microbes can improve leguminous crop nodulation and nitrogen fixation ( Marek-Kozaczuk and Skorupska, 2001 ; Vessey and Buss, 2002 ; Masciarelli et al., 2014 ; Korir et al., 2017 ; Puente et al., 2019 ; Xie et al., 2019 ). This improved nodulation was correlated with an increased abundance of nitrogen-fixing bacteria in rhizosphere soil, which can be partially explained by an alteration of plants and metabolism reflected in root exudates addition ( Xie et al., 2019 ; Wang H. et al., 2021 ). Furthermore, it was reported that exogenous arbuscular mycorrhiza affected microbial community diversity and structure ( Smith and Read, 2010 ; Kohler et al., 2016 ; Wu et al., 2021 ) and the inoculation of Mimosa pudica with Paraburkholderia phymatum induced significant alterations in the root-associated microbiome ( Welmillage et al., 2021 ). In this study, we used a single bacterial endophyte inoculation to address the questions: (1) how the inoculation affects plant-associated microbiome, and (2) how long these effects can be persistent in the plant environment. To address the first question, we evaluated the differences in composition, diversity, and cooperation between the microbiomes associated with inoculated and non-inoculated plants after 3 months post-inoculation. Since soil microorganisms provide a foundation for the plant microbiome ( Zarraonaindia et al., 2015 ), we also looked at the response of soil bacteria to the inoculation. To address the second question, we extended our analysis of the root microbiome for another year to detect long-lasting effect of the inoculation.",
"discussion": "Discussion This study examined the response of plant-associated microbiome to the colonization by exogenous bacterial endophyte in perennial crops. However, we started our analysis with evaluation of the transformation of non-inoculated bacterial community over the first 3 months of plant growth, which can provide useful information regarding the dynamic of soil and plant-associated microbiomes during an initial adaptation of the plant to the new environments. Our results indicated significant variations in diversity and structure of bulk and rhizosphere soil and root microbiomes, although these variations did not follow the same pattern in all niches. For example, we detected an increase in bacterial alpha diversity in bulk soils (S2) and rhizosphere (RS2) after 3 months of plant growth, while this parameter was unchanged in root microbiome (R2) in 3 months post-planting and was decreased in the roots of the plants after field growth. This agrees with the previous reports that indicated a decrease in bacterial alpha-diversity along the soil-endosphere continuum ( Trivedi et al., 2020 ). On the other hand, based on the variation in sample groupings and the number of differentially represented taxa between time points, rhizosphere microbiome exhibited more stability over time compared to bulk soil and root microbiomes. This differential dynamic of the microbiomes might be a result of an increased complexity of interactions within microbiomes and plant holobiont ( Yurgel et al., 2018 ). In the non-inoculated roots, the composition of bacterial community changed over time. While several Proteobacteria taxa significantly decreased, Actinobacteria taxa and Bacillales were increased. It was shown that the plant microbiome is affected by both soil and plant ( Bulgarelli et al., 2013 ; Trivedi et al., 2020 ). In our study, the changes detected in root microbiome might reflect the changes in soil microbiome during the shift from nursery propagation to growth in the pots and in the fields—as well as the physiological status of the plants—maturation over the 3-month and 1-year period. For example, the profile of soil microbiome also underwent significant changes, including decrease in the relative abundance of Chitinophagales, Sphingomonadales, Xanthomonadales, and unclassified Gammaproteobacteria, the pattern similar to the changes in root microbiome over time. On the other hand, the relative abundances of several Actinobacteria taxa were increased in plant roots over time but not in soils, confirming that the previous studies show that the plant exerts control over its microbiota ( Tkacz and Poole, 2015 ; Trivedi et al., 2020 ). The strain RF67 used in our experiments was isolated from roots of perennial crop Vaccinium angustifolium and annotated as Allorhizobium–Neorhizobium–Pararhizobium–Rhizobium group. There is a substantial scientific evidence showing that rhizobial species can successfully infect and colonize cereal crops ( Rosenblueth et al., 2018 ). To ensure endophytic properties of RF67, we used GFP reporter gene to verify that the strain was able to establish infection and colonization of the host-plant root. Fluorescent bacteria were detected on the root surface and also in the apoplast of both root apical meristem regions where the cell walls are soft and also in the root differentiation zone where the cell walls are harder. Bacterial cells were normally detected two to three cell layers below the root surface. These facts demonstrate the ability of RF67 to penetrate an epidermis cell layer in different root zone and spread through the entire root apoplast. However, bacterial cells were not observed in the cotyledons even after 48 h of co-cultivation. This fact suggests that RF67 can only colonize root tissues. We also showed, that even though RF67 was isolated from perennial plant preferential to high acidic soils, the strain grew well on the plates with pH 7 typical for Haskap agricultural soils ( Iheshiulo et al., 2018 ) and colonize roots on medium with pH 5.7. Additionally, the initial isolation of the strain RF67 was done on MMNH4 medium with pH 7. This suggests that the strain can survive and grow in the soils used for Haskap cultivation. Despite the ability of RF67 to establish symbiosis with host-plant and its adaptation to less acidic environment, the ASV corresponding to RF67 was barely detected in inoculated soil, rhizosphere, and root microbiomes. Although it was significantly overrepresented in inoculated roots in 3 months post-inoculation (R1), it was only represented on average by 1.45 reads per sample. We also detected the sporadic presence of RF67 ASV in inoculated rhizosphere and soils after 3 months post-inoculation (RS1), as well as in bare (R0) and non-inoculated roots (R2). To account for potential MiSeq bleed-through between runs, ASV which accounted for less than 0.1% of the total sequences was removed during data processing. However, it is possible that an error in the steps of PCR amplificon during library preparation and sequencing might introduce a single-nucleotide polymorphism, resulting in the false detection of RF67 ASV in the non-inoculated microbiomes. It was recently estimated that the overall observed error rate for samples from the MiSeq platform is 0.473% with standard deviation 0.938 ( Stoler and Nekrutenko, 2021 ). Additionally, a small fragment, such as 16S rRNA V6-V8 region, does not present a determined taxonomic validity ( Flores-Felix et al., 2019 ; Young et al., 2021 ). This also could explain the fact that these sequences can be found in non-inoculated samples, where other naturally occurring rhizobial spices could be considered. While we did not detect a high presence of R67 on inoculated microbiomes, the inoculation with RF67 significantly affected in the root and soil microbiomes 3 months post-inoculation. The inoculation induced a significant shift in overall bacterial community structure in the roots and soils and increased bacterial alpha-diversity in the roots. The inoculation also resulted in the increase in the relative abundances of families Enterobacteriaceae and Rhizobiaceae in the inoculated roots compared to non-inoculated ones. Both these families contain a number of taxa with plant growth promotion capabilities. Interestingly, the inoculation had an opposite effect on the relative abundance of Rhizobiaceae in soils. In general, the relative composition of soil microbiome was much stronger affected by inoculation and resulted in the decrease in the relative abundances of a number of bacteria belonging to Alphaproteobacteria and Gammaproteobacteria classes, as well as Chthoniobacteraceae , Gaiellales, Nitrospiraceae , and Chloroflexi taxa, compared to non-inoculated soil. However, no effect of inculcation was detected in the rhizosphere microbiome’s diversity, composition, and structure. This is consistent with the previous findings which show that root-associated microbial communities were more affected by inoculation compared to the rhizosphere microbiome ( Welmillage et al., 2021 ). Based on the overall structure of the co-occurrence network, after 3 months post-inoculation (R1), the introduction of RF67 affected the interaction pattern within the root-associated community. The microbiome of the inoculated roots exhibited a stronger cooperation compared to the non-inoculated roots (R2). This was reflected in the 2-fold increase in the number of interactions, compared to non-inoculated roots, and in the formation of a large cluster of strongly cooperating Proteobacteria, Bacteroidota, Gemmatimonadota, and Actinobacteriota in the inoculated roots. In the non-inoculated roots, Firmicutes were the major taxa with strong cooperation among them. When the synthetic community derived from root endophytes was used for inoculation, an increase in the relative abundances of potential plant growth promotion microorganisms was detected ( Armanhi et al., 2021 ). We did not detect a large number of bacterial taxa differentially represented between inoculated and non-inoculated roots, but a number of taxa harboring plant beneficial microbes were among the most connected in the co-occurrence network of inoculated root microbiome. These taxa included Caulobacteraceae ( Pepe et al., 2013 ), Xanthobacteraceae ( Lee et al., 2008 ), Gaiellaceae ( Lazcano et al., 2021 ), Sphingomonadaceae ( Asaf et al., 2020 ), and Chitinophagaceae ( Madhaiyan et al., 2015 ). These results suggested that the introduction of RF67 boosted cooperation between plant growth-promoting endophytes. After a year of plant growth, the changes in bacterial community diversity and structure linked to RF67 inoculation were undetectable. The communities from inoculated (R4) and non-inoculated roots (R3) had similar Shannon diversity, as well as no significant effect of inoculation was detected on the significance of sample grouping. However, the co-occurrence network detected strong cooperation between bacteria in the inoculated roots, which was not detected in the non-inoculated roots. Similar to the co-occurrence network from R1, the R4 network had nearly 2-fold increase in the number of interactions, compared to non-inoculated roots. Nevertheless, overtime, the composition of tightly cooperated bacterial had changed. While Proteobacteria, Bacteroidota, Gemmatimonadota, and Actinobacteriota were among the highly connected taxa, a number of Firmicutes also became a part of this group. Interestingly, all the most connected taxa (with the degree at least 17) in R4 co-occurrence network were also found in the co-occurrence networks of R1 and/or R2. This might indicate that after initial strong perturbations in the cooperation between bacteria caused by RF67 inoculation, naturally occurring cooperation (found in R2 but not in R1) began to form in microbiome over time."
} | 4,230 |
27147218 | PMC4857112 | pmc | 2,852 | {
"abstract": "Microalgae biosynthesize high amount of lipids and show high potential for renewable biodiesel production. However, the production cost of microalgae-derived biodiesel hampers large-scale biodiesel commercialization and new strategies for increasing lipid production efficiency from algae are urgently needed. Here we submitted the marine algae Phaeodactylum tricornutum to a 4-day dark stress, a condition increasing by 2.3-fold the total lipid cell quotas, and studied the cellular mechanisms leading to lipid accumulation using a combination of physiological, proteomic (iTRAQ) and genomic (qRT-PCR) approaches. Our results show that the expression of proteins in the biochemical pathways of glycolysis and the synthesis of fatty acids were induced in the dark, potentially using excess carbon and nitrogen produced from protein breakdown. Treatment of algae in the dark, which increased algal lipid cell quotas at low cost, combined with optimal growth treatment could help optimizing biodiesel production.",
"discussion": "Discussion The present study shows that total lipid cell quotas in P. tricornutum , one of the best strains for biodiesel production 5 , can strongly increase after a darkness treatment. When photosynthesis was abolished in the dark and synthesis of the proteins involved in the biosynthesis of the photosynthetic machinery was decreased, we found that cellular C, N and energy were redirected toward lipid biosynthesis ( Fig. 4 ). Re-allocation of cellular energy toward lipid biosynthesis was also previously observed under N limitation and optimal light 14 21 . After a prolonged darkness period, the transcription and expression of key proteins involved in the production of the photosynthetic machinery (see Supplementary Table S6 ) as well as the biosynthesis of photosynthetic pigments (see Supplementary Table S5 ) only decreased by less than a factor of three even though C fixation was previously abolished for 4 d during the prolonged darkness period, a finding consistent with the results of Nymark et al. 22 . This suggests that a significant part of the photosynthetic apparatus, which contains a major fraction of cell N in algae, remains functional in the dark explaining why diatoms typically start to grow rapidly when the light suddenly becomes available after a dark stress 22 23 . Under such an intense stress on the photosynthetic apparatus, we indeed observed a decrease in the proportion of cell C allocated to carbohydrates and proteins, but carbohydrate oxidation through the glycolysis increased suggesting that P. tricornutum could either use an external source of sugars (such as sugars exuded by the algae in the culture medium during the growing period preceding the experiment in the dark) in the glycolysis pathway or mobilized polysaccharide reserves. The acetyl-CoA produced through the glycolysis was increasingly used for lipid biosynthesis. Our combined analyses of the whole proteome, key gene transcripts, organic acids, and enzymatic activities thoroughly describe the intricate cellular mechanisms at the root of the dark-enhanced lipid production in P. tricornutum . The long-term dark stress inhibited the synthesis of the photosynthetic machinery but stimulated carbohydrate oxidation As previously observed by Qian and Borowitzka 24 in N-limited P. tricornutum cells, the long-term dark stress we used in the present study decreased the cell carbohydrate content and affected the synthesis of the photosynthetic machinery, but stimulated lipid biosynthesis. The (nearly) generalized induction in the expression of glycolytic proteins (B7G5Q1, B7G6K6, B7GA05, and B7G585) and in the transcription of genes coding for glycolytic enzymes upon a prolonged dark stress clearly indicates that glucose conversion into acetyl-CoA was up-regulated. The expression of one enzyme (B7FXN2) of the pyruvate dehydrogenase complex (containing three enzymes: B7FXN2, B7GB47 and B7G3I7) was increased in response to the prolonged darkness period, but the expression of the two others were down-regulated. B7G317 (dihydrolipoamide acetyltransferase or E2) expression could decrease because it is an enzyme involved in the translation of psb A mRNA into the D1 protein in photosystem II (PSII) 25 , which was inhibited under dark stress. The cell content in pyruvate and acetyl-CoA effectively increased after the prolonged darkness period. However, the glucose cell content did not increase significantly in response to the dark treatment indicating that glucose was oxidized rapidly. The acetyl-CoA increasingly produced in the dark was then utilized for boosting fatty acid synthesis (see the section below). Furthermore, the dark-induced decrease in the expression of the enzyme UDP-glucose-6-dehydrogenase (B5Y5J6) suggested that the biosynthesis of polysaccharide reserves was inhibited, which could increase TAG biosynthesis 26 . Down regulation of nitrogen assimilation under long-term darkness The decrease of two key enzymes of N assimilation in darkness, nitrate reductase and glutamine synthetase, suggested that N assimilation into amino acids and proteins decreased under the prolonged dark stress. A decrease in the protein cell content was indeed observed after the 4-d dark stress. Furthermore, the decrease in the expression of five ribosomal proteins and two translation elongation factors in the dark is also consistent with a dark-induced inhibition of protein biosynthesis ( Fig. 1B ). An inhibition of the N assimilatory pathway in response to the prolonged darkness period contrasted with the up-regulation in the expression of the enzymes involved in N assimilation observed under N limitation in P. tricornutum 14 . This suggests that after a prolonged period of time in the dark, P. tricornutum , decrease energy investment in the N assimilation biochemical machinery by contrast to the observed up-regulation in the expression of the enzymes involved in N assimilation under N limitation in P. tricornutum 14 . However, both N limitation/starvation and a prolonged dark stress inhibit net protein biosynthesis. For instance, the transcripts level of two ribosomal proteins (ProtIDs 21235 and 15259) and of a translation factor (ProtID 269148) decreased in T. pseudonana grown under N starvation 27 . We found that the dark stress not only decreased protein biosynthesis, but it also increased protein breakdown by inducing the expression of an ubiquitin-activating enzyme E1 (B7FTU2), which catalyzed the first step in protein degradation process 28 29 . P. tricornutum cells redirect the acetyl-CoA produced from glycolysis toward lipid production, while concurrently, inhibiting the allocation of carbon skeleton to the nitrogen assimilation machinery ( Fig. 4 ). Oxidative stress increased in prolonged darkness Imbalance in C and N assimilation in algal cells can lead to oxidative stress 30 . In the present study, the observed increase in the expression of two superoxide dismutase isozymes (B7FPQ3 and B7G0L6) and heat-shock proteins (B7FXQ8 and B5Y472) after the 4-d dark stress suggests that the production of reactive oxygen species (ROS) in dark-treated cells increased and that reparation of proteins damaged by ROS was increasingly needed. Moreover, the increase in fatty acid synthesis measured in the dark-stressed cells could serve not only as an energy store, but also as antioxidant as suggested by Li et al. 31 . The later authors argued that TAG incorporation in thylakoids membranes could alleviate ROS accumulation by blocking redundant electrons passing through the photosynthetic electron transport chain. Stimulation of fatty acid synthesis under a long-term dark stress The first step in fatty acid synthesis is the conversion of acetyl-CoA to malonyl-CoA, catalyzed by acetyl-CoA carboxylase (ACCase). Even though the dark treatment did not increase the expression of the ACCase of P. tricornutum , it did increase the expression of the enzyme enoyl-acp reductase (FabI), which is the key enzyme to elongate fatty acid carbon chains, hence facilitating the synthesis of long-chain fatty acids. Real-time quantitative PCR also demonstrated the significant increase in the transcription of a number of genes involved in fatty acid biosynthesis (i.e. FABI, FABFa, SCD and PTD9) under prolonged darkness. Furthermore, the expression and gene transcription of a number of enzymes involved in the β-oxidation of fatty acids was down-regulated in response to the 4-d dark stress suggesting that fatty acid degradation decreased in the dark relative to the control. The results suggest that oxidation of lipids was not a major energy and C source for P. tricornutum in the dark. Hence, the observed decrease in the expression of enzymes involved in fatty acid β-oxidation and the demonstrated increase in the expression of several enzymes of fatty acid biosynthesis in the dark-treated cells act together to increase the TAG cell content in the dark-treated cells. By contrast, it has been shown that the increase in TAG production by N-limited P. tricornutum cells is coupled to an increased allocation of C and reductant to lipid biosynthesis rather than to an increase in the activity of the enzymes involved in fatty acid biosynthesis 14 . In the dark-treated cells, we observed that hexadecanoic acid (C16:0) and octadecanoic acid (C18:0) production increased to high level; both fatty acids can be esterified to acyl carrier protein (ACP), 16:0-ACP and 18:3-ACP, and these two ACPs fatty acids may contribute to the biosynthesis of glycerolipids in the chloroplast and thylakoid membranes 32 . Chloroplast membranes regulate metabolites in and out of cells, and thylakoid membranes are the site of the primary process of photosynthesis. Interestingly, the dark treatment also increased the production of long chains unsaturated fatty acids (e.g. linolenic acid and cis-5,8,11,14,17-eicosapentaenoic acid or EPA) by inhibiting fatty acid β-oxidation and increasing the expression of the enzyme enoyl-acp reductase (FabI). These long chains unsaturated fatty acids are of high nutritional values for farmed animals as well as humans 33 . A dark treatment of P. tricornutum thus needs to be considered in future development of algal cultivation technique in aquaculture and health food industries. Fluxes in carbon and nitrogen assimilatory pathways and regulation of lipid biosynthesis Our comprehensive physiological, gene transcripts, and proteomic analyses show that a long-term dark stress stimulated glycolysis, protein degradation, and fatty acid synthesis explaining the observed increase in lipid cell content in the dark ( Fig. 4 ). Under long-term darkness, up-regulation of the glycolysis pathway and the associated increase in the production of NADPH, pyruvate, and acetyl-CoA all contribute to increase lipid biosynthesis whereas, under N limitation, redundant photosynthetically fixed C and NADPH enhanced fatty acid and TAG biosynthesis 31 34 . The flux of C toward protein synthesis in the marine alga P. tricornutum tended to dominate in optimal growth conditions (balance C:N ratio) ( Fig. 4A ), while under a long-term dark stress, cell C and N from protein breakdown tended to be redirected toward lipid biosynthesis to minimize the imbalance in C and N assimilation ( Fig. 4A ). We demonstrated that a long-term dark stress in the marine microalga P. tricornutum enhanced neutral lipid cell content by 3.7-fold without strongly decreasing cell biomass by redirecting C and N flux into lipid synthesis (see Supplementary Table S1 ). This increase in lipid cell content after 4 day in darkness is higher than the increase in lipid cell quotas measured in several microalgae species growing in conditions of low N concentrations 3 35 . This storage of high-energy lipids could be of paramount importance for the survival of P. tricornutum in conditions of prolonged darkness such as in the deep ocean. Our study brings new insights into the fundamental mechanisms leading to modulation of lipid production in the dark in P. tricornutum , a species with high potential for biofuel production. Dark treatment could strongly stimulate lipid cell quotas, and represented a new cheap way to enhance algal lipid cell quotas. If combined with optimal culture conditions, a dark stress could potentially increase lipid productivity for biofuel production."
} | 3,097 |
30233626 | PMC6129962 | pmc | 2,853 | {
"abstract": "Food web theory predicts that current global declines in marine predators could generate unwanted consequences for many marine ecosystems. In coastal plant communities (kelp, seagrass, mangroves, and salt marsh), several studies have documented the far-reaching effects of changing predator populations. Across coastal ecosystems, the loss of marine predators appears to negatively affect coastal plant communities and the ecosystem services they provide. Here, we discuss some of the documented and suspected effects of predators on coastal protection, carbon sequestration, and the stability and resilience of coastal plant communities. In addition, we present a meta-analysis to assess the strength and direction of trophic cascades in kelp forests, seagrasses, salt marshes, and mangroves. We demonstrate that the strength and direction of trophic cascades varied across ecosystem types, with predators having a large positive effect on plants in salt marshes, a moderate positive effect on plants in kelp and mangroves, and no effect on plants in seagrasses. Our analysis also identified that there is a paucity of literature on trophic cascades for all four coastal plant systems, but especially seagrass and mangroves. Our results demonstrate the crucial role of predators in maintaining coastal ecosystem services, but also highlights the need for further research before large-scale generalizations about the prevalence, direction, and strength of trophic cascade in coastal plant communities can be made.",
"conclusion": "Conclusion Above we have discussed how predators can help protect the ecosystem services provided by coastal plant communities. However, our meta-analysis highlighted that the availability of studies in all four coastal plant systems is far below the volume needed to make broad generalizations about trophic cascades in these systems. Thus, the prevalence of top-down control in marine ecosystems is still debatable, especially for seagrass and mangroves. Furthermore, even if one concedes that top-down control is common in coastal plant communities, arguments about whether predators predominantly have positive or negative effects on plant communities remains an open question. Although marine predator conservation is important for multiple reasons (e.g., biodiversity), the argument that through top-down control they promote the persistence of coastal plant communities needs further study. However, this brings up a troubling question, if the science is not currently sufficient to make broad generalizations, can it get there in time to make a real contribution to our conversations about predator conservation in coastal communities? Continuing declines in marine predator populations suggest that we are running out of time to quantitatively “prove” to the world and ourselves that marine predators are important to the persistence of coastal plant communities and the ecosystem services they provide. This leaves coastal scientists with a conundrum. We can choose to wait, methodically building the evidence for or against predator effects on coastal plant communities, with a specific focus on increasing studies in seagrass and mangroves, underrepresented regions, and large-bodied predators and herbivores. However, conservation decisions must be made while there is still an opportunity to do so, otherwise the results of such studies will become obsolete ( Martin et al., 2012 ). Our other option is to be more aggressive in our recommendations about predator conservation, despite our less than perfect knowledge about their effects in coastal plant communities.",
"introduction": "Introduction The green world hypothesis predicts that the loss of predator control on herbivores could result in runaway consumption that would eventually denude a landscape or seascape of vegetation ( Hairston et al., 1960 ). Since the inception of the green world hypothesis, ecologists have tried to understand the prevalence of indirect and alternating effects of predators on lower trophic levels (i.e., “trophic cascade;” Figure 1 ), and their overall impact on ecosystems ( Estes et al., 2011 ). Multiple lines of evidence now suggest that top predators are key to the persistence of some ecosystems ( Estes et al., 2011 ). FIGURE 1 Predicted effects of predators, or lack thereof, on ecosystem services (carbon sequestration, coastal protection, and ecosystem stability) in coastal plant communities. It is predicted that predators, through direct and indirect interactions with lower trophic levels, support increased carbon uptake in plants and soils, protect coasts from storm surges and flooding, and support stability and resistance. With an estimated habitat loss of >50%, coastal plant communities are among the world’s most endangered ecosystems ( Zedler and Kercher, 2005 ; Waycott et al., 2009 ; Duarte et al., 2013 ). As bleak as this number is, the predators that patrol coastal systems have fared far worse. Several predatory taxa including species of marine mammals, elasmobranchs, and seabirds have declined by 90–100% compared to historical populations ( Lotze et al., 2006 ; Paleczny et al., 2015 ). Interestingly, predator declines pre-date habitat declines ( Lotze et al., 2006 ), suggesting that alterations to predator populations may be a major driver of change for coastal systems ( Jackson, 2001 ; Jackson et al., 2012 ). There is little doubt that collapsing marine predator populations results from overharvesting by humans. Localized declines and extinctions of coastal predators by humans began over 40,000 years ago with subsistence harvesting ( McCauley et al., 2015 ). However, for most large bodied, marine predators (toothed whales, large pelagic fish, sea birds, pinnipeds, and otters) the beginning of their sharp global declines occurred over the last century, coinciding with the expansion of coastal human populations and advances in industrial fishing ( Lotze et al., 2006 ; Lotze and Worm, 2009 ). Following global declines in marine predators, evidence of trophic cascades in coastal ecosystems started to emerge ( Estes and Palmisano, 1974 ; Sala and Zabala, 1996 ; Myers et al., 2007 ; Heithaus et al., 2014 ), with the disturbing realization that they affected more than just populations of lower trophic levels ( Estes et al., 2011 ). Understanding the importance of predators in coastal plant communities has been bolstered by their documented ability to influence ecosystem services ( Figure 1 ). Multiple examples have shown that changes to the strength or direction of predator effects on lower trophic levels can influence coastal erosion ( Coverdale et al., 2014 ), carbon sequestration ( Wilmers et al., 2012 ; Atwood et al., 2015 ), and ecosystem resilience ( Hughes et al., 2016 ). The idea that the extirpation of predators can have far-reaching effects on the persistence of coastal plants and their ecosystem services has become a major motivation for their conservation in coastal systems ( Estes et al., 2011 ; Atwood et al., 2015 ). Here, we discuss some of the effects of predators on coastal plant communities and the ecosystem services they provide. Although these examples provide evidence that the loss of predators has negative consequences for important ecosystem services, they do not give a sense of prevalence of trophic cascades in coastal plant communities. Furthermore, our examples highlight cases where predators had positive effects on the plant community, which in turn had a positive effect on ecosystem services. To determine whether our examples represent the rule for predator effects on ecosystem services in coastal plant communities or the exception, we conducted a meta-analysis to assess the strength and direction of trophic cascades in kelp forests, seagrasses, salt marshes, and mangroves. Coastal Protection Coastal flooding and erosion are major threats to coastal areas, with the frequency and magnitude of such events expected to increase with climate change ( Arkema et al., 2013 ). Protection by coastal plant communities has been identified as a relatively cheap and effective tool for mitigating the effects of coastal erosion and flooding ( Arkema et al., 2013 ; Duarte et al., 2013 ; Möller et al., 2014 ). The aboveground structure of coastal plants attenuates wave energy, dissipating and reflecting it away from the shore ( Coops et al., 1996 ; Koch et al., 2009 ; Möller et al., 2014 ; Rupprecht et al., 2017 ). In addition, the complex root structures of seagrass, salt marshes, and mangroves help trap and stabilize sediments, allowing shorelines to accrete, further attenuating wave energy and reducing coastal erosion. Herbivores and omnivores can negatively impact the structure of coastal plant communities ( Coverdale et al., 2014 ). Many types of marine herbivores consume the leaves, seeds/propagules, or roots of coastal plants with negative consequences for plant density and canopy height ( Cargill and Jefferies, 1984 ; Heithaus et al., 2014 ). As the extent of vegetation is correlated with the degree of wave attenuation ( Koch et al., 2009 ), such changes in the structure of coastal plants could significantly reduce their ability to attenuate wave energy. Predator declines have been implicated at least in part, in the collapse of several coastal plant communities, reducing coastal protection ( Silliman et al., 2005 ; Coverdale et al., 2012 ). For example, recreational fishing of predatory marine crabs and fish along the east coast of the USA has relaxed predation pressure on sesarmid crabs. In response to lower predator abundance, sesarmid crab densities have increased six-fold and their burrows can cover up to 90% of the surface area of some marshes ( Coverdale et al., 2012 ). These burrows undermined the structural integrity of the marsh, causing >10 cm of horizontal erosion annually, effectively removing >150 years of coastal accretion <30 years ( Coverdale et al., 2014 ). These results suggest that reinstating top-down control in degraded coastal plant communities could help alleviate coastal flooding and erosion. However, further studies linking trophic cascades to changes in coastal protection are needed, especially in vulnerable areas like Florida, California, and New York where coastal habitats provide the greatest risk reduction from coastal hazards ( Arkema et al., 2013 ). Enhance Carbon Sequestration Kelp forests, seagrass, salt marsh, and mangroves are among the world’s most productive ecosystems, with global net primary production rates of 0.01–0.64 Pg carbon yr -1 ( Duarte et al., 2013 ). In addition to storing carbon in plant biomass, seagrasses, salt marshes, and mangroves also store a significant amount of carbon in their soils ( McLeod et al., 2011 ; Fourqurean et al., 2012 ; Duarte et al., 2013 ; Atwood et al., 2017 ; Macreadie et al., 2017b ). With carbon turnover rates that are an order of magnitude slower than terrestrial soils, coastal wetlands represent the ultimate sink for otherwise rapidly cycled carbon ( McLeod et al., 2011 ). Although kelp does not accumulate large soil carbon deposits, kelp forests are carbon donors, exporting carbon to shelf and deep sea sediments ( Krause-Jensen and Duarte, 2016 ; Filbee-Dexter et al., 2018 ). Predators can shape the structure of coastal plant communities through consumptive (lethal) and non-consumptive (risk) effects on herbivorous prey, altering the storage of carbon in plant biomass ( Griffin et al., 2011 ; Christianen et al., 2014 ). The return of sea otters to the North American west coast revived overgrazed kelp forests, increasing carbon captured by kelp by upward of 8.7 Tg ( Wilmers et al., 2012 ). Conversely, declines in other marine predators along the California coast allowed epiphytes to smoother the leaf surface of seagrass, reducing photosynthetic rates and dropping seagrass production by 50% ( Lewis and Anderson, 2012 ). Although in both the above cases the negative effects of predation on herbivory had positive effects on plant growth, herbivory is an important process in coastal plant communities. Under natural or low levels, grazing can stimulate primary production by encouraging new growth ( Cargill and Jefferies, 1984 ). This suggests that a delicate balance in herbivory, which can be accomplished through predation, is needed to ensure maximum productivity. Not only do predators protect carbon sequestration in plant biomass, they also increase carbon sequestration in coastal soils ( Atwood et al., 2015 , 2018 ; Macreadie et al., 2017a ). For example, a 50% reduction in the density or canopy height of macrophytes has been shown to increase sediment resuspension 10-fold ( Gacia et al., 1999 ). Predators can moderate the effects of herbivores on coastal plant communities by significantly reducing their consumption of plant biomass and reducing the spatial extent of herbivory ( Griffin et al., 2011 ; Atwood et al., 2015 ). Such cascading effects of predators has been shown to increase carbon accumulation rates and belowground carbon stocks in seagrass, salt marshes, macroalgal systems, and to a lesser extent mangroves ( Atwood et al., 2015 , 2018 ). Promote Ecosystem Stability and Resilience Over the past 50 years we have lost 20–50% of global seagrass, salt marsh, and mangrove ecosystems ( Zedler and Kercher, 2005 ; Duarte et al., 2013 ). For kelp the story is more variable, with region-specific changes that reflect both increases and decreases in kelp forest abundance ( Krumhansl et al., 2016 ). Declines to coastal plant communities can be attributed to a multitude of interacting stressors (e.g., rising water temperatures, sea level rise, and eutrophication), many of which are anthropogenically driven. In the face of so many disturbances, stability and resilience may be the key to ensuring ecosystem persistence. A few natural experiments have provided evidence that predators serve an important role in the stability of coastal plant communities. In Elkhorn Slough, an estuary in California, nutrient loading from agriculture led to a sharp decline in eelgrass in the area between 1965 and 1984 ( Hughes et al., 2013 ). The recovery of eelgrass in the estuary in 1985 and again in 2005 coincided with the return of sea otters. Sea otters generated a four-tiered trophic cascade that ultimately resulted in the reduction of epiphytes, which reduce seagrass growth through shading ( Hughes et al., 2013 , 2016 ). Although rare, experiments investigating the ecosystem-level effects of predator recovery represent one of the few ways that we can examine their influence on ecosystem stability/resilience. This is because in many cases marine predator populations are already severely depleted ( Dulvy et al., 2014 ; Paleczny et al., 2015 ), and their role in ecosystems altered. A re-occurring issue with quantifying how predators impact ecosystems is that measures of stability and resilience are inherently multifaceted ( Donohue et al., 2016 ). One of the goals of understanding stability and resilience is to aid in the recovery of ecosystems, which in itself is a multifaceted problem. Managers tasked with restoring ecosystems by promoting predators should establish baseline data, and set clear measurable targets (e.g. “restore plant biomass to 75% and sediment retention to 50% of historical levels”). Only through quantifying recovery targets can the impacts of disturbances associated with the loss of predators be quantified, and mitigated ( Donohue et al., 2016 )."
} | 3,875 |
34781330 | null | s2 | 2,854 | {
"abstract": "We report the design of 'slippery' nanoemulsion-infused porous surfaces (SNIPS). These materials are strongly anti-fouling to a broad range of substances, including microorganisms. Infusion with water-in-oil nanoemulsions also endows these slippery coatings with the ability to host and control or sustain the release of water-soluble agents, including polymers, peptides, and nucleic acids, opening the door to new applications of liquid-infused materials."
} | 114 |
34255962 | PMC8320526 | pmc | 2,860 | {
"abstract": "DNA origami has emerged as a powerful molecular breadboard with nanometer resolution\nthat can integrate the world of bottom-up (bio)chemistry with large-scale, macroscopic\ndevices created by top-down lithography. Substituting the top-down patterning with\nself-assembled colloidal nanoparticles now takes the manufacturing complexity of\ntop-down lithography out of the equation. As a result, the deterministic positioning of\nsingle molecules or nanoscale objects on macroscopic arrays is benchtop ready and easily\naccessible."
} | 131 |
36297934 | PMC9612328 | pmc | 2,861 | {
"abstract": "A superhydrophobic composite coating consisting of polytetrafluoroethylene (PTFE) and poly(acrylic acid)+ β-cyclodextrin (PAA + β-CD) was prepared on an aluminum alloy AA 6061T6 substrate by a three-step process of electrospinnig, spin coating, and electrospraying. The electrospinning technique is used for the fabrication of a polymeric binder layer synthesized from PAA + β-CD. The superhydrophilic characteristic of the electrospun PAA + β-CD layer makes it suitable for the absorption of an aqueous suspension with PTFE particles in a spin-coating process, obtaining a hydrophobic behavior. Then, the electrospraying of a modified PTFE dispersion forms a layer of distributed PTFE particles, in which a strong bonding of the particles with each other and with the PTFE particles fixed in the PAA + β-CD fiber matrix results in a remarkable improvement of the particles adhesion to the substrate by different heat treatments. The experimental results corroborate the important role of obtaining hierarchical micro/nano multilevel structures for the optimization of superhydrophobic surfaces, leading to water contact angles above 170°, very low contact angle of hysteresis (CAH = 2°) and roll-off angle ( α roll − off < 5°). In addition, a superior corrosion resistance is obtained, generating a barrier to retain the electrolyte infiltration. This study may provide useful insights for a wide range of applications.",
"conclusion": "4. Conclusions In the present study, we have developed a composite PAA + β-CD + PTFE coating through the combination of ES, SC and SP techniques and successive heat treatments. The addition of PTFE nanoparticles in the PAA + β-CD electrospun mat turn the surfaces from hydrophilic to hydrophobic (WCA = 142°) because of the fluorine groups (CF 2 ). In order to produce superhydrophobic surfaces and with excellent water mobility, the SP of modified PTFE dispersion leads to water contact angles above 170° and very low contact angle hysteresis and roll-off angle (CAH = 2°, <5°). In this way, the FE-SEM images corroborate the important role of obtaining hierarchical nano/microparticle composite and the creation of micro/nano multilevel structures for the optimization of superhydrophobic surfaces. In addition, the superhydrophobic surface exhibits a significant improvement of the mechanical properties with a scratch resistance in the range of Al/epoxy coatings. This increase is mainly obtained by a strong bonding of the electrosprayed PTFE nanoparticles with each other and with the PTFE particles of lower layers fixed in the PAA + β-CD fiber matrix, where there are a high density of carboxylic groups and negative charges. Therefore, the different heat treatments (HTx) play a fundamental role not only for wetting agent removal, but for the improvement of particles bonding with each other and their adhesion to the substrate. Finally, the corrosion resistance shows an excellent performance, where a remarkable protection is achieved, blocking the corrosion current by water repulsion, which creates a barrier to retain the electrolyte infiltration. Thus, the surface remains dry with the electrolyte in the SHS surface and chemical inertness by the PTFE.",
"introduction": "1. Introduction In recent years, nanocomposites have become very attractive materials with which to design different functional properties for a broad range of applications. The addition of nanoparticles to the polymeric material may lead to changes in their properties that could be interesting for their applications. Among them, superhydrophobicity (SH) has been attracting attention due to properties that make them suitable to use in numerous fields [ 1 , 2 ]. One of the most representative examples of these materials are polymeric coatings, to which the addition of PTFE nanoparticles significantly improves their hydrophobicity, low sliding (roll-off) angles, low hysteresis, immersion stability, corrosion resistance, low ice adhesion, high thermal conductivity, dielectric and mechanical properties, and corrosion resistance [ 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 ]. As a result, there is a wide range of interesting applications, such as self-cleaning, liquid transport efficiency, naval coatings, seaplanes, off-shore wind power and oil platform applications, among others [ 11 , 12 , 13 , 14 , 15 , 16 ]. PTFE is a thermoplastic semicrystalline fluoropolymer with excellent chemical resistance, high thermal stability, low-friction coefficient, and superhydrophobicity characteristics as a consequence of the strong bonds between the fluorine and carbon atoms. However, the high crystalline melting point (337 °C) and its insolubility in common solvents make PTFE difficult to process [ 17 , 18 , 19 ] and adhere to different surfaces [ 20 , 21 ]. These difficulties, coupled with the need to manufacture materials that combine high surface roughness with the PTFE low surface energy [ 22 , 23 ], have led to a number of different production techniques. Among them, we can include roughening with sandpaper, microetching processes, laser-plasma X ray, laser/femtosecond laser, 3D printing, laser ablation, hot embossing process, and spray coating among others [ 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 ]. Recently, these procedures have been joined by the electrodynamic techniques, electrospinning [ 32 , 33 , 34 ] and electrospraying [ 20 , 35 ], which are facile, cost effective, and flexible methods that utilize an electrically charged jet of polymer solution for different scales of fiber production [ 36 ]. In addition, another method compatible with PTFE coatings is spin coating. This technique uses the centrifugal forces created by a spinning substrate to spread a coating solution on a surface and obtain films with a given and homogeneous thickness [ 37 , 38 ]. In this work, a superhydrophobic PTFE coating with excellent water mobility and superior corrosion protection is proposed. In the preparation of this kind of coating, the adhesion of the PTFE particles to the coated surface becomes a key point that still needs to be optimized [ 20 , 21 ]. In order to maximize the hydrophobic properties and improve the coating adhesion, a three-step process combining electrospinning, electrospraying, and spin-coating techniques has been used. In the first one, a PAA + β-CD electrosprayed fiber membrane is deposited onto the substrates to take advantage of its adhesive properties [ 39 , 40 , 41 ]. In the second one, a PTFE spin-coated layer is deposited, thus obtaining a uniform particle distribution embedded in the porous PAA + β-CD network that leads to WCA values similar to those reported in literature for PTFE fibers [ 42 , 43 , 44 , 45 ]. Finally, an extra layer of electrosprayed PTFE particles leads to the formation of a high roughness, increasing the hydrophobicity with WCA values at the level of the best ones found in the literature [ 46 , 47 ]. In each step, samples were subjected to different heating treatments to improve PTFE adhesion to the coated surface leading also to a high corrosion resistance, making this material a good candidate for multiple outside applications.",
"discussion": "3. Results and Discussion 3.1. Sample Thickness and Surface Morphology Sample thickness was measured by using confocal image microscopy (S1, S2, and S3), where a step function of the cross-section vertical profile at a different location [ 51 ] was employed obtaining the average values. On the S4, a contact profiler was used as it was not possible to obtain an image with the confocal microscope. A step profile was obtained with the contact profiler at several locations, resulting in the mean values. In addition, samples’ fiber diameter and their structure were analyzed with FE-SEM. The surface roughness was studied by interferometric profilometry (S1, S2, and S3) and contact profilometry (S4) to understand the effect of the different procedures during their preparation. On S4, the contact profiler was applied, because it was not possible to obtain an image with the interferometric profilometer. The obtained values for the thickness, fiber diameter and surface roughness are summarized in Table 3 . The fiber diameters in the Figure 2 a histogram are presented with diameter size for samples S1 (a) and S2 (b) having a log normal distribution shape, typically observed in nanoparticles’ size distribution that leads to the average values presented in Table 3 . The observed fiber diameter decreases from S1 to S2 as is clear in the FE-SEM images shown on Figure 3 a,b. As can be seen in Figure 3 a there is a network of fibers with irregular sizes and shapes, where partial crosslinking occurs between the β-CD and PAA for the S1 sample. The first thermal treatment HT 1 leads to the excedent solvent evaporation (ethanol and water) and an increase of the crosslinking agent (β-CD) activation with PAA. The rest of unreacted β-CD is detached from the fibers, decreasing their diameter; it then agglomerates and crystallizes in cubic shapes as can be seen in Figure 3 b. The β-CD crystallization process has been extensively studied [ 52 , 53 , 54 , 55 ]; theform depends on the chemical compounds surrounding the polymer and the solvent used. The presence of the β-CD cubic crystals within the fiber network increases both the roughness and the sample thickness. As for the rest of the samples, roughness has a huge decrease from S2 to S3, resulting in a smooth surface due to the SC 1 + HT 2 procedure. The FE-SEM images of Figure 3 b,c show how the PTFE-1 covers uniformly the fibers of S2 and, as a result, a fairly uniform coating of PTFE nanoparticles is achieved for S3. On the contrary, the SC 2 procedure of PTFE-1 with the SP of PTFE-2 dispersion increases surface roughness from S3 to S4. The PTFE particles of S4 (see Figure 3 d) are randomly deposited and form strong bond with each other and with the PTFE particles of lower layers obtained from S3 + SC 2 process during HT 3. This results in a random network of spheres with a micro/nano multilevel structure through the combination of different size particles leading to a rough surface structure [ 20 ]. 3.2. Chemical Composition In order to identify the chemical composition of the samples surface, FTIR and XPS studies were performed. In Figure 4 , FTIR spectra of the electrospun samples S1 and S2 show no significant differences between them. They have absorption bands characteristics of ester bonds, produced by the crosslinking agent β-CD in the PAA solution by the esterification process. More specifically, a strong absorption band in the 1750–1700 cm −1 region is attributed to the ester carbonyls and to the carboxylic carbonyl bonds, present in both the free carboxylic acid groups of PAA and in the β-CD-PAA ester bonds. The bands in the 1300–1100 cm −1 region are attributed to the C-O stretching vibrations of carboxylic and ester functional groups and, the bands in the 1050–900 cm −1 region could be associated to the OH deformation mode of the alcohol groups of β-CD, which have not crosslinked with PAA fibers, giving evidence of a partial crosslinking reaction between the β-CD and PAA. Moreover, the O-H absorption peak shows a decrease for S2 as a result of an increase of the crosslinking agent (β-CD) activation with PAA after HT 1 [ 39 ]. In the case of S3, the FTIR spectrum shows strong absorption peaks at 1210 cm −1 and at 1151 cm −1 , which correspond to the C-F bond and C-C bond of the PTFE-1, respectively. The peak at 640 cm −1 could be attributed to the absorption of CF 2 wagging [ 56 ]. As can be observed S3 presents the same contributions as the PTFE reference sample as expected because of the SC 1 process. Finally, in the S4 sample, the FTIR spectrum is completely flat, reflecting light in the infrared region and avoiding the obtention of any information. To solve this and obtain complementary information on the samples surface, XPS studies were carried out. The overall sample composition was determined from survey spectrum which showed intense signals from C, F, and O. C, F, and O atomic concentrations were determined by measuring the integral areas of C 1s, F 1s, and O 1s spectra, taken at a pass energy of 20 eV, after background subtraction and normalization by using the sensitivity factors proportional to the Scofield cross-section provided by the electron energy analyzer manufacturer. S1 and S2 data are in agreement with those obtained by FTIR showing PAA and β-CD presence. However, some extra information was extracted from S3 and S4 in comparison with a reference PTFE sample. The elemental content of C and F of S3 and S4 correspond to the nominal atomic of PTFE (70% F; 30% C). From the quantitative analysis of XPS peaks upon sample processing under electron beam irradiation, it is observed that the F vs. C atomic content ratio is reduced from (initial 70–30% F, final 50–50%for PTFE). This fact is in line with the damage of CF 2 bonds and the decrease of both the associated CF 2 peak in the C 1s ( Figure 5 a) and the continuous loss of F 1s core level emission. In addition, this is also related to the increase signal of C 1s in the energy region of the photoelectrons from C-C, and the emission of C-OH and C=O groups, whose intensity also follows the trend of the O1s signal. Regarding our samples, XPS spectra of S3 and S4 exhibit the characteristic photoelectron emission of PTFE comprised of the C 1s and F 1s intense peaks centered at the binding energy of ca. 292 to 690 eV, respectively. Detailed peak shape analysis of C 1s emission were performed by the deconvolution of the C 1s spectrum with several Gaussian/Lorentzian symmetric components (ratio of 70/30) by using a least-squares fitting routine [ 48 ]. The energy position of the peaks and their relative heights were determined to account for the emission ascribed to the different chemical environment of carbon atoms) according to the values reported in previous works [ 57 , 58 ]. The result of the fit provides symmetric peaks from CF 2 , CF, CF x representative of the PTFE nature of the samples and several components C=O, C-OH, C-C attributed to the presence of various oxygen functional groups that were present on the irradiated PTFE sample. Upon comparison with the line shapes, no significant differences are observed other than the intensity of the peaks for the oxygen functional groups as can be seen in Figure 5 b. However, the O 1s emission signal is quite different for the PTFE, S3 and S4 samples as can be seen in Figure 5 c. The PTFE sample shows a clear presence of O which, as we can see in Figure 5 c, decreases for S3 and is nil for S4. This presence explains the differences discussed above and may be indicative of the hydrophobic qualities of the samples being S4 the better one. Regarding the bonding stability, PTFE reference sample shows a decrease for the CF 2 signal giving way to the appearance of the secondary components mentioned as the irradiation time goes on. This phenomenon is smaller for S3 sample and non-existent for S4 suggesting that chemical bonds reduce their quality but are more stable for higher-temperature treatments. 3.3. Coating Adhesion Study by Scratch Test The adhesion of the samples was evaluated by the scratch test to compare the adhesive failure and strength of the different coatings on the substrate. The S1 sample had poor adhesion, not good enough to be tested in the scratch test machine. The S2 sample did not show any scratch resistance, with gross delamination and adhesion failure from the early stages of the test (see Table 4 ). The S3 sample exhibits an improvement of scratch resistance compared to S2 due to the SC 1 of PTFE-1 dispersion and HT 2 . This higher value is associated with the weak polyelectrolyte nature of the polymeric precursor of PAA, which can display a high density of carboxylic groups and negative charges, making possible a complex balance of interactions (mostly electrostatic and Van der Waals interactions) that can also promote a better adhesion onto the substrate [ 59 ]. In this way, the PTFE nanoparticles are fixed in the fiber matrix, resulting in a composite, wherein a higher cross-linking leads to higher scratch values [ 60 ]. Finally, the scratch resistance is increased in S4 compared to S3. This increase is mainly obtained by a strong bonding of the electrosprayed PTFE nanoparticles with each other and with the PTFE particles of lower layers during HT 3 , where the particles are melted and progressively cooled. The HT 3 not only leads to particles bonding, but also improves the adhesion to the substrate. In addition, the highest adhesion strength obtained in S4 (see Table 4 ) is comparable to scratch critical loads of Al/epoxy coatings (from 6 N to 12 N) [ 61 ]. According to previous work on electrosprayed PTFE coatings, one of the main problems is the adhesion of PTFE nanoparticles to the substrate [ 20 , 21 ]. The optical images of the representative scratches are presented in Figure 6 . As can be seen, the surface trace of the samples exhibits a cleaner trace in those with lower critical loads, unlike those with higher critical loads, which show more tearing. Therefore, this work implies a remarkable improvement in the bonding of PTFE particles to each other and their adhesion to the substrate. 3.4. Wetting Properties The PAA + β-CD coatings (S1 and S2) have shown a superhydrophilic behavior having different surface roughness (see Table 3 ). The S1 sample was not considered in this study because the water insolubility of the fiber mat was achieved by a thermal treatment (HT 1 ), inducing a thermal crosslinking reaction between the PAA and the β-CD molecules. Thus, the fibers of S2 remain almost unaltered, showing high chemical and mechanical stability absorbing the water droplets [ 39 , 62 ]. In both cases, S1 and S2 show a superhydrophilic behavior with static WCA < 5°. However, in sample S3, the coating shows a complete change of behavior, exhibiting a clear hydrophobicity with a WCA = 142° ± 7° in spite of being the smoothest of all the samples (see Figure 7 ). The water repulsion of the fluoride functional groups previously mentioned fully explains this behavior [ 56 , 63 ]. On the other hand, the geometrical structure of the surface plays an important role in the static wettability properties. This effect can be clearly seen between samples S3 and S4, where the hydrophobicity of S3 is mainly caused by the chemistry and the SH of S4 (static WCA = 170° ± 4.2°) by the combined effect (chemistry and the geometrical structure), especially as a result of the remarkable increase in roughness due to the SP process of PTFE nanoparticles (see Figure 7 ). In addition, the roll-off water angle ( α roll − off ) and the contact angle of hysteresis (CAH) of the S3 and S4 samples (see Table 5 ) were measured to check the dynamic wetting properties. The S4 sample obtained a higher performance through lower CAH and α roll − off values. This implies a low water interaction generated by the electrosprayed PTFE nanoparticles. The significant improvement of the dynamic wettability properties obtained in S4 is mainly caused by the combination of the PTFE chemical repulsion and the multilevel roughness of the surface. This is induced by the formation of micro/nano-sized particles, resulting in a Cassie–Baxter state [ 64 , 65 ]. The presence of air entrapped inside the gaps particles, makes the penetration of the liquid into the roughness or texture of the surface, difficult due to capillary forces, leading to a superhydrophobic surface [ 66 , 67 ]. 3.5. Anticorrosion Performance In order to check the anticorrosion behavior, Tafel polarization tests were performed obtaining the polarization curves presented in Figure 8 . First, the aluminum substrate (AA6061T6) was tested as a reference ( Figure 8 ) to be later compared with the samples (S3 and S4). In this case, the sample S1 and S2 were not studied because of their superhydrophilic behavior, where there are almost no differences to the substrate. In polarization curves, the excellent corrosion resistance has a lower corrosion rate (CR), which is related to a lower corrosion current density (Jcorr) and a higher corrosion potential (Ecorr) [ 68 , 69 , 70 ]. The results show that all the samples minimize the corrosion rate and current density of the aluminum substrate (see Table 6 ). Furthermore, Table 6 indicates that an increase of superhydrophobicity and a low water adhesion strength are related to a considerable decrease in Jcorr and CR, leading to an improvement in the substrate protection efficiency (η). This phenomenon could be due to a better blocking of the corrosion current because of water repulsion, which creates a barrier to retain the electrolyte infiltration to a larger extent. This effect is more intense in S4, where the tested surface was completely dry after removal of the electrolyte. However, in S3, the surface was wet as a consequence of the electrolyte infiltration. Thus, S4 and S3 minimize the corrosion rate of the aluminum substrate by four orders and one order of magnitude, respectively. Finally, these results are in concordance with reported literature [ 21 , 71 , 72 ], where PTFE coatings show a superior corrosion resistance, showing a great potential to be used in harsh acidic industrial environments. This effect is associated with the presence of a robust air layer entrapped on the superhydrophobic surface, which considerably reduced the contact area between coating and liquids, acting as an efficient barrier to prevent the penetration of aggressive media to the underlying metal substrate."
} | 5,411 |
31565323 | PMC6607131 | pmc | 2,862 | {
"abstract": "Abstract Transparent, durable coating materials that show excellent liquid repellency, both water and oil, have multiple applications in science and technology. In this perspective, herein, a simple aqueous chemical formulation is developed that provides a transparent slippery coating without any lubricating fluids, on various substrates extended over large areas. The coatings repel liquids having a range of polarity (solvents) as well as viscosity (oils and emulsions) and withstand mechanical strains. Exceptional optical transparency of 99% in the range of 350–900 nm along with high stability even after cyclic temperature, frost, exposure to sunlight, and corrosive liquids like aqua regia treatments, makes this material unique and widens its applicability in different fields. Besides, being a liquid, it can be coated on an array of substrates independent of their underlying topography, by various easily available techniques. Aside from these interesting properties, the coating is demonstrated as a potential solution contributing to the remediation of one of the biggest global issues of tomorrow: affordable drinking water. The coated surface can capture 5 L of water per day per m 2 at 27 °C when exposed to an atmosphere of 63% relative humidity.",
"conclusion": "4 Conclusion In conclusion, a waterborne, easy to synthesize, robust coating material has been formulated that shows high liquid repellency, without the use of any lubricating fluids. The coating showed excellent stability toward mechanical strains with uncompromised optical transparency. Transparency as well as liquid‐repellent properties of the coating was maintained even after extreme thermochemical treatments. Being a water‐based liquid material, it enables the creation of large surface area slippery surfaces with a simple coating procedure and decreases environmental concerns and risk of organic solvents at the same time. While extreme repellency toward a wide variety of liquids can widen its industrial use by minimizing transportation cost of fluids through pipelines, transparency in extreme conditions along with other properties can provide easy solutions for display and automobile industries. Beside these, application of this surface toward solving one of the biggest global issues, namely affordable clean/drinking water, is demonstrated as a proof of concept.",
"introduction": "1 Introduction Materials capable of imparting amphiphobic (hydrophobic and oleophobic) coatings are highly desirable for today's varied applications such as touch screen displays to glasses used in buildings, automobiles, etc. and are being intensely researched upon. 1 Although robustness toward chemical and mechanical stresses is one of the most needed/desired criteria of such coatings, high optical transparency has also drawn much attention of both industries and academia.[[qv: 1h,2]] Liquids usually have a high contact angle (CA) and a low contact angle hysteresis (CAH = advancing angle [θ A ] − receding angle [θ H ]) on these surfaces. In some cases, the surface energy of these coatings is so low that it makes liquids to bounce or roll over the surface and sit as a sphere, an effect known as superamphiphobicity.[[qv: 1b,e,3]] While lotus leaf effect is the inspiration behind these developments, 4 the trapped air in the microstructured surface of lotus leaf is ineffective against liquids with low surface tension. 5 To address this, recent research has come up with a “re‐entrant surface curvature technology,” a hierarchically developed surface structure with sufficient recess between the surface structure and its base causing liquids to sag below, without coming in contact with its sides or the base. 3 , 6 These materials have ultralow CAH and consequently high repellency for oils. Modulating the aspect ratio of these microstructures allowed the construction of surfaces with a static CA close to 160° for hexadecane, although the flat surface was oleophilic. 7 However, the creation of these microstructures compromises the transparency[[qv: 6d]] of the material due to increased refractive index.[[qv: 1b,7,8]] Moreover, intricate microfabrication technology needed to create such surfaces makes them rather expensive and in addition offer limited surface compatibility.[[qv: 3,6a]] On the other hand, stability or longevity of such surfaces are questionable, although a few reports on robust liquid repelling coatings exist.[[qv: 1h,9]] Furthermore, high CA leading to lower contact area between the liquid drops and these surfaces makes it difficult to accomplish various industrially significant features including heat transfer, condensation, and many others. 10 In this context, slippery liquid‐infused porous surfaces (SLIPS), with equally efficient liquid repellent property, known to possess low CA and low CAH while allowing high contact area, is an alternative. 5 , 11 \n Nepenthes pitcher ‐plant inspired surfaces of this kind were developed by infusing low surface tension liquids, such as perflurinated oils, inside the nano/microstructured porous matrices.[[qv: 2b,5,12]] Recently, Chen et al. have reported a cellulose‐based transparent slippery surface that repels both liquids and ice.[[qv: 2b]] Kang et al. have developed transparent hydrophobic electrodes in the context of outdoor solar cell devices where nonwetting property keep the surface clean and allows efficient/effective absorption of sunlight. 13 Application of these bioinspired surfaces is also known in different fields of science and technology.[[qv: 1d,5,14]] Furthermore, such water repelling surfaces having large contact area of water droplets are more efficient for condensation‐based technologies like water/humidity harvesting that can help to solve the water scarcity. In this context, Kim et al. have developed a graphene‐based hydrophobic surface. 15 However, large scale production of such surfaces can be an issue and expensive as it was obtained at very high temperatures (≈1000 °C) through in situ chemical‐vapor‐deposition. Therefore, designing a simple coating material to develop an affordable and scalable liquid repelling surface (both for water and for other liquids) devoid of lubricating fluids with durability is important. Incorporation of transparency can also explore the applications of such surfaces toward different global issues including energy crisis. Several methods have been introduced to create amphiphobic surfaces.[[qv: 1h,16]] Among these, designing coating materials through the sol–gel process is one of the increasingly developing methods and is intensely researched upon because of its reduced complexity (in production) and diverse substrate compatibility.[[qv: 1e,f,17]] However, most of the time, organic solvents, such as ethanol, acetone, hexadecane, dimethylformamide, and tetrahydrofuran, are heavily used that increase the concern related to the production cost and associated environmental problems.[[qv: 6a,16a,18]] Similar problem exists for slippery coating material as well.[[qv: 2b]] A few reports on aqueous coating materials for superhydrophobic surfaces exist. 19 However, such materials for superamphiphobic or slippry surfaces are not explored much. 20 Recently Lin and co‐workers have demonstrated a robust superamphiphobic surface by spraying a stable aqueous fluorinated nanoparticle dispersion.[[qv: 1e]] Such waterborne coatings yielding high transparency are desired for various applications. In this perspective, use of polydimethylsiloxane, a widely used hydrophobic coating material, also gets limited attention because of its solvent (organic) and substrate (except glass) compatibility. It also possesses inherent limitation in transparency when exposed to different temperatures. This suggests the necessity to develop a coating material in water that can provide a robust liquid repellent slippery coating over various substrates, irrespective of their shape, size, and surface morphology. In this work, we present a novel waterborne material which is a liquid at room temperature and shows excellent liquid repellent property (without any lubricating fluids) upon curing over the surface. The material can be painted or coated as a thin film on various substrates, such as metal, glass, hard plastic, and paper, etc., despite their varying surface morphology. Being a liquid, large area coating by processes such as spray coating, spin coating and doctor blading are possible which widen its applicability. Coated substrates show excellent liquid repellency with 99% transparency when compared to clean room treated glass slides. Interestingly, coating withstands various thermo‐mechanochemical damages without any adhesives and retains its properties intact. We believe, a combination of reduced surface energy along with rigid nanoscale structures, which form during the rapid polymerization process, helps these coatings to repel a wide variety of liquids irrespective of their polarity and viscosity. Beside these multiple effective properties, applicability of this coating for efficient water condensation is demonstrated as a proof of concept for atmospheric water capture that can resolve one of the biggest global issues namely, the water crisis.",
"discussion": "3 Results and Discussion \n Figure \n \n 1 \n A demonstrates that a wide variety of liquids from water to toluene and even corrosive acids such as aqua regia sit over the coated surface without spreading. This indicates low surface free energy of the coating which was investigated further. Inset shows the static CA of those respective fluids over the surface. Exceptionally high optical transparency of the coated glass was observed using UV–vis spectrometry (Figure 1 B). About 99% transmission in comparison to clean room‐treated glass is shown in Figure 1 B, Insets 1 and 2, and Video S1 in the Supporting Information. No observable differences in the visibility were found even when the coated surfaces were tested in front of an electronic display, ≈8–10 cm away from the surface (Figure 1 B, Inset 2). Initially, the wetting property of the coated surface was tested with the movement of the water drop on the surface when it was tilted manually for a few degrees (details are in the Experimental Section and the Supporting Information). The velocity of the slipped water drop was measured as 5.4 cm s −1 (Figure 1 C and Video S2, Supporting Information). Such a property of the coating resembles the slippery surfaces and was further studied in detail later on. This phenomenon can lead to a range of applications in energy reduction, in the context of liquid transport. Figure 1 A) Photographs of the coated surface with different liquids; 1‐water, 2‐toluene, 3‐aqua regia. Inset: The static contact angle of the respective liquids. B) Percentage transmission of coated glass in comparison to a normal glass. The coated surface showed 99% transmission when compared to the normal uncoated surface. 1. Coated surface just before the written letters. 2. Electronic letters through the coated surface (distance: 8–9 cm approximately). C) Image shows that at low tilting angle (5°) of the coated surface water rolled off at a speed of 5.4 cm s −1 . Low surface tension liquids (such as oils and emulsions) usually wet the surface and spread easily. Here, the wetting behaviors of the coated surface toward various liquids having different surface tensions were assessed by both static and dynamic CA measurements where dynamic CA is represented in terms of CAH. For instance, water droplet placed on the coated surface formed a static contact angle of 134° ± 2°; whereas, for toluene and silicone oil, it was 86° ± 2° and 52° ± 2°, respectively ( Figure \n \n 2 \n A). It correlates with the surface tensions of the respective liquids. However, low CAH in the range of 10° ± 2°, 4° ± 2°, and 3° ± 2° for water, toluene, and silicone oil, respectively, reveals the extent of the slippery nature of the coating toward different liquids despite having low static contact angle on the coated surface (Figure 2 A). Inset shows a picture of advancing and receding angle of the respective fluids. Surface structure and the chemical composition, being the underlying reasons of the wetting property, the coated surface was characterized in detail through various spectroscopic and microscopic techniques. X‐ray photoelectron spectroscopy (XPS) spectra reveal that the coating is largely composed of silica and fluorocarbons (Figure 2 B). Peak at in the region of 103.5 eV corresponds to the deconvoluted Si 3p peak of Si 4+ which matches exactly with that of silica (SiO 2 ). The presence of silicon, nitrogen, and other elements along with fluorine was proved from the energy dispersive analysis of X‐rays (EDAX) spectrum and mapping as well (Figure 2 C and Figure S1, Supporting Information). Wide area powder X‐ray diffraction of the film showed an amorphous background, similar to the glass substrate (Figure S2, Supporting Information). This was also observed in the qualitative elemental distribution of silicon and oxygen (1:2) in the EDAX mapping of the surface (Figure S1, Supporting Information). This detailed chemical characterization concludes that the backbone of the coating is made of silica network. This was also reflected in the robustness of the coating toward thermo‐mechanochemical perturbation, explained later on. Though scanning electron microscopy (SEM) reveals the absence of micrometer scale structure (Figure 2 D), atomic force microscopy (AFM) imaging confirms that the roughness of the coatings is very low, less than 1 nm (Figure 2 E). Both of these studies (spectroscopic and microscopic) suggest that the nanostructuring and the presence of appropriate functionalization are the reasons of this observed wetting property. To demonstrate the extent of water repellency of the coating material a glass slide was coated in the shape of ‘’ consisting of two uncoated patches in the middle. The hydrophobic coating in the periphery of the glass slides acted as an invisible barrier to contain ≈4 mL of water in the uncoated hydrophilic patches having a surface area of 7.8 cm 2 . As a result of this confinement, the water could reach a height of 0.4 cm in each of the patches (Figure S3 and Videos S3 and S4, Supporting Information). This concept of the invisible barrier has been used by others too and can be used in trapping, directing as well as retaining water and can have immediate applications in biological and environmental areas. 22 \n Figure 2 A) The advancing angle (θ A ), receding angle (θ R ), and contact angle hysteresis (θ H ) of water, toluene, and silicone oil, respectively. Inset shows the corresponding images. Physicochemical properties of the coated substrate. B) The XPS spectra of the coated surface showing fluorine 1s and silicon 2p region. C) EDAX spectrum shows the elemental composition of the coating. D) SEM image reveals the absence of micrometer scale morphology. E) AFM image and the average roughness of the coating, which is in nm scale. For applications in touch screens, goggles, and windscreens, the coating needs to be resistant to acute temperature fluctuations and chemical disruptions without compromising transparency of the surface. The stability and wettability of the coating against thermal and chemical damages were evaluated with extreme temperatures (high and low) and aqua regia (corrosive acid mixture). For all the cases, water, toluene, and silicone oil were used to study the wettability of the treated surfaces in terms of static CA and CAH. Glass substrates were assessed primarily to monitor the transparency and integrity of the coating. For high temperature treatment, surface was annealed at 200 °C for 4 h and it was observed to retain its surface free energy intact compared to the control sample (coated slides at room temperature). This was reflected in the CA and CAH of the liquids over the treated surface ( Figure \n \n 3 \n A and Video S5, Supporting Information). The stability of the coating at low temperatures was tested by incubating the surface at −80 °C for 8 h. In this case, a similar liquid repellent property was observed for the treated surface (Figure 3 A and Video S6, Supporting Information). Chemical robustness of the material was tested by incubating the coated glass in aqua regia for 10 min. Interestingly, the wettability of the coating remained unaltered and the surface functioned properly (Figure 3 A and Video S7, Supporting Information). Optical transparency of all the treated surfaces was also found to remain unchanged from that of the control (Figure 3 B). Inset of the figure pictorially represents the treated surfaces. These seem to be highly advantageous for places where frost formation on windshields is a serious concern. To be used as a nonwettable coating material for day to day use, mechanical stability is a mandatory compliance. This was assessed by knife scratch, peeling off, and abrasion tests. These tests were done using scissor, scotch tape, and a sand paper (keeping a load of 50 g on the sand paper) (Figure 3 C, 1–3). For all the cases, even after 20 complete cycles, the coatings remained pristine with uncompromised wetting behavior toward different liquids (Figure S4, Supporting Information). Transparency of the treated surfaces also remained intact although there was some sign of knife scratches on the particular surface (Figure S5, Supporting Information). Reusability as well as stability of the coating was further evaluated by write and erase tests (Figure 3 C, 4) where pencil streaks were easily erasable without damaging the unique properties of the coating. In this case also treated surface was also checked for wettability and transparency (Figures S4 and S5, Supporting Information). Figure 3 A) Stability of the coating upon high temperature (200 °C), low temperature (−80 °C), and aqua regia treatments. CA and CAH of the liquids (water, toluene, and silicone oil) on the treated surfaces compared to the control. B) Corresponding percentage transmission of the treated substrates shows no change in comparison to the control. Inset: Pictorial representation of treated surfaces. C) Test for mechanical robustness. 1) Peeling‐off experiment, 2) sand paper abrasion, 3) knife scratch, and 4) reusability measurements using write and erase experiments. D) Durability test in cyclic fashion. Sets 1 and 2: Treating at high temperature (200 °C) and low temperature (−80 °C). Set 3: Effect of chemicals (surface was dipped inside different organic solvents, oil and emulsion). Set 4: Direct exposure to sunlight. Figure 4 Properties of the coated surface. A). Response of the surface to aqua regia treatment. The metal surface (stainless steel) was dipped into aqua regia. B) There was an evolution of hydrogen gas from the uncoated part and C) it eventually corroded, while the coated region remained intact. D) Schematic representation of atmospheric water capture. E) Real time experimental setup with condensed water drops over the coated surface where surface temperature was cooled down to 8 °C by a peltier cooling system. The environmental temperature was 27 °C with 63% relative humidity. Longstanding durability of the coating to different cyclic perturbations was studied by subjecting the sample to consecutive cycles of different sets of conditions such as high temperature, frost, chemical treatment (organic solvents, oil and emulsion), and exposure to sunlight. Change in the wettability was measured (CA and CAH) after each and every cycle of different sets (details in the Experimental Section and the Supporting Information), which shows a constant value of 134° ± 2° and 86° ± 2° for CA and 10° ± 2° and 4° ± 2° for CAH on an average for water and toluene, respectively. These values remained unaltered even after four different sets of experiments (total of 20 cycles) (Figure 3 D). Application of such materials is not confined to glass substrates alone. To explore the universality or compatibility of the material with different substrates, this material was applied to a wide variety of substrates starting from metal to wood and plastic (Figure S6, Supporting Information). Consistency in the physical appearance of the original substrates even after the coating provides an added advantage (Figure S7, Supporting Information). All the substrates (coated glass and wood surface shown here) showed excellent resistance to wetting by nonpolar fluids such as oil and oil‐water emulsion (Figure S8 and Video S8, Supporting Information). Stability of the coatings on metal substrates was evaluated by treating them with the corrosive acid, aqua regia. The metal surface coated with the newly synthesized material remained unaffected ( Figure \n \n 4 \n A–C and Video S9, Supporting Information) while uncoated surface changed its color immediately with the evolution of hydrogen gas. Figure 4 D schematically demonstrates effective application of such slippery coatings in real life. Affordable drinking water being a global issue to concern, atmospheric water capture has become a hot topic of research. This needs efficient condensation of humidity and transportation of the droplet formed on the surface. In this context, low hysteresis and low contact angle (high contact area) surfaces having excellent durability can be a good solution as water condenses over such surfaces easily. Figure 4 E demonstrates a proof of concept experiment. Humidity and temperature are the governing parameters for this phenomenon. At 63% humidity and 27 °C, our coated surface enables condensation of 5 L of water per m 2 in a day. Here, a peltier cooling system was used to cool the surface down to 8 °C. We believe that the efficiency of such water collection can be maximized by patterning the surface. We note that overall collection efficiency is not only an issue of efficient condensation but also transport which is mostly controlled by the wettability of the surface as well as CAH. High liquid repellent nature of the coating originates from the presence of low surface energy molecules as well as the nanostructures formed in situ. While efficient adhesion property of silanes on the surface of various substrates makes this coating universal, the formation of amorphous silicate structure upon curing, which is inert toward a range of chemicals including strong acids, makes this material robust toward various mechanical and chemical perturbations. Stability and high transparency of the coating at varying temperatures also can be explained easily from the physical and the chemical structure of the material, which are similar to silicate glass."
} | 5,660 |
33137653 | null | s2 | 2,863 | {
"abstract": "A major question remaining in the field of evolutionary biology is how prokaryotic organisms made the leap to complex eukaryotic life. The prevailing theory depicts the origin of eukaryotic cell complexity as emerging from the symbiosis between an α-proteobacterium, the ancestor of present-day mitochondria, and an archaeal host (endosymbiont theory). A primary contribution of mitochondria to eukaryogenesis has been attributed to the mitochondrial genome, which enabled the successful internalisation of bioenergetic membranes and facilitated remarkable genome expansion. It has also been postulated that a key contribution of the archaeal host during eukaryogenesis was in providing 'archaeal histones' that would enable compaction and regulation of an expanded genome. Yet, how the communication between the host and the symbiont evolved is unclear. Here, we propose an evolutionary concept in which mitochondrial TCA cycle signalling was also a crucial player during eukaryogenesis enabling the dynamic control of an expanded genome via regulation of DNA and histone modifications. Furthermore, we discuss how TCA cycle remodelling is a common evolutionary strategy invoked by eukaryotic organisms to coordinate stress responses and gene expression programmes, with a particular focus on the TCA cycle-derived metabolite itaconate."
} | 334 |
33137653 | null | s2 | 2,864 | {
"abstract": "A major question remaining in the field of evolutionary biology is how prokaryotic organisms made the leap to complex eukaryotic life. The prevailing theory depicts the origin of eukaryotic cell complexity as emerging from the symbiosis between an α-proteobacterium, the ancestor of present-day mitochondria, and an archaeal host (endosymbiont theory). A primary contribution of mitochondria to eukaryogenesis has been attributed to the mitochondrial genome, which enabled the successful internalisation of bioenergetic membranes and facilitated remarkable genome expansion. It has also been postulated that a key contribution of the archaeal host during eukaryogenesis was in providing 'archaeal histones' that would enable compaction and regulation of an expanded genome. Yet, how the communication between the host and the symbiont evolved is unclear. Here, we propose an evolutionary concept in which mitochondrial TCA cycle signalling was also a crucial player during eukaryogenesis enabling the dynamic control of an expanded genome via regulation of DNA and histone modifications. Furthermore, we discuss how TCA cycle remodelling is a common evolutionary strategy invoked by eukaryotic organisms to coordinate stress responses and gene expression programmes, with a particular focus on the TCA cycle-derived metabolite itaconate."
} | 334 |
35173573 | PMC8842996 | pmc | 2,865 | {
"abstract": "We present an efficient and scalable partitioning method for mapping large-scale neural network models with locally dense and globally sparse connectivity onto reconfigurable neuromorphic hardware. Scalability in computational efficiency, i.e., amount of time spent in actual computation, remains a huge challenge in very large networks. Most partitioning algorithms also struggle to address the scalability in network workloads in finding a globally optimal partition and efficiently mapping onto hardware. As communication is regarded as the most energy and time-consuming part of such distributed processing, the partitioning framework is optimized for compute-balanced, memory-efficient parallel processing targeting low-latency execution and dense synaptic storage, with minimal routing across various compute cores. We demonstrate highly scalable and efficient partitioning for connectivity-aware and hierarchical address-event routing resource-optimized mapping, significantly reducing the total communication volume recursively when compared to random balanced assignment. We showcase our results working on synthetic networks with varying degrees of sparsity factor and fan-out, small-world networks, feed-forward networks, and a hemibrain connectome reconstruction of the fruit-fly brain. The combination of our method and practical results suggest a promising path toward extending to very large-scale networks and scalable hardware-aware partitioning.",
"introduction": "1. Introduction There has been a growing interest in the scientific community to attain a comprehensive understanding of the brain (Markram et al., 2011 ; Kandel et al., 2013 ) using actual in-vivo brain recordings or simulation models using spiking neural networks (SNNs). However, simulating such large brain-size networks (Ananthanarayanan and Modha, 2007 ) with massive size and complexity of neurons and interconnections between them is extremely challenging to realize using the computational capability of today's digital multiprocessors. Thus, extreme-scale distributed computing is being explored as an alternative route to the physical limitations in traditional computing methods. Computing systems with high-bandwidth interconnects between individual compute elements are crucial for such enormously distributed processing, and to demonstrate performance efficiency at brain scale. Such processing architectures are also energy-efficient edge ML acceleration tasks such as audio, image, and video processing. Data movement through a Network-On-Chip (NoC) becomes the most challenging part in the synchronization and event exchange of many-core spiking processors. This communication becomes the limiting factor in the processing, while the computation scales linearly with the number of cores (Musoles et al., 2019 ). To minimize the inter-core communication, we require both hardware optimization as well as an efficient compiler to generate an optimal network partitioning and mapping to the available computational resources. Distributed computing for spiking networks is the most efficient when performed with a combination of load-balancing, with which computation at the lowest latency is realized for any given network; and inter-core connectivity minimization, which ensures the most optimum reduction in traffic volume over the network. We use an extended version of the hierarchical address-event routing (HiAER) architecture (Park et al., 2017 ) for scalable communication of neural and synaptic spike events between different cores. HiAER implements a tree-based interconnect architecture of fractal structure in the connectivity hierarchy, where the communication bandwidth at each node is relatively constant at each level in the hierarchy due to decreasing fan-out at increasing levels. The notation L i corresponds to the i 'th level of communication hierarchy. The lowest level ( L 0 ) in this hierarchy represents the intra-core communication, which is just governed by a synaptic routing table (postsynaptic neuron destinations stored in the local memory allocated to the same core), which incurs no routing cost. As we organize these cores within the larger system, we can create further hierarchy, i.e., L 1 (communication within a cluster of cores) and L 2 (communication between clusters). For example, in order to arrange 32 cores onto two layers of hierarchy, we can arrange them into eight L 2 clusters of four L 1 cores, 16 L 2 clusters of two L 1 cores, or any other arrangement of two factors with a product of 32. This type of organization allows us to easily map our cores and routing nodes to hardware, where there are specific routing requirements and network connectivity while obviating the need for all-to-all connections through an extremely high bandwidth interconnect. We can extend our hierarchically structured communication network further by adding additional levels of hierarchy. Peak efficiency in near-memory or in-memory compute architectures requires all compute cores to get efficiently utilized, assuming a network structure that is locally dense and globally sparse. The previously discussed HiAER protocol for routing of neural spike events allows network connectivity that scales to networks of virtually unlimited size, due to decreasing connectivity density with an increasing distance that permits near-constant bandwidth requirements in event routing across the spatial hierarchy. An ideal neuromorphic computing system attains both the computational efficiency of intra-core dense local connectivity with near-memory compute cores, and the functional flexibility of HiAER sparse long-range connectivity. An open problem in the practical realization of this system is to efficiently map a given network with arbitrary topology onto the implemented hierarchy of cores with rigid dimensions. This requires automated means to partition the network in such a way to maximally align its connectivity with maximal fill density of the connectivity matrices within cores, and minimal communication of neural events across cores. The optimal partitioning method to align large, arbitrarily structured networks onto a scalable HiAER topology must satisfy several conditions. First, the partitioning scheme should be fast and scalable to different levels of the HiAER communication scheme. A network should be able to be partitioned over K cores. Neurons in these cores will use different levels of off-core communication depending on the location of the destination core in the communication hierarchy. We assume that each neuron has equal processing time in the cores. The total size of the network can be written as n 0 * K , with n 0 being the number of neurons in each of the K cores. Depending on hardware constraints and space that the user provides, the partitioning method should be able to partition over different possible values of n 0 , K , and L i . Figure 1 shows different potential configurations of the network around K cores. The partitioning method should be able to find an optimal partition and core configuration for each case shown. Figure 1 Different potential tree-based HiAER network configurations for 8 cores with varying L i and a fixed core count K . Additionally, the network of N neurons distributed across K cores must produce balanced partitions where each core contains roughly N / K neurons. Cores with too many neurons will be bottlenecks in network performance due to longer processing time, while cores with too few neurons will be underutilized. Distributing the neurons in a balanced fashion allows for balanced processing time among the cores and maximizes the overall network speed. Finally, the partitioning method should result in savings in both memory and communication across cores. This means that the neurons should be arranged in a way such that cross-core communication is minimized, which involves both grouping neurons with similar incoming connections inside the same core, as well as minimizing cross-core communication. These savings must be applicable to partitions that are at single or multiple levels of hierarchy. In most partitioning methods, minimum edge-cut is used as the metric for judging the quality of the partition and can be defined as the number of edges whose incident vertices belong to different partitions. However, minimum edge-cut does not suffice to describe the network communication when using AER or mask bits due to the fact that multiple connections can be encoded in the same communication packet. For our evaluation, we do not use the minimum edge-cut in order to evaluate the quality of the partition, and instead, we use our own set of routing rules defined for the HiAER network routing. For a given input network, it is difficult to partition based on the activity of each neuron in the network. Depending on the neuron model, inputs, and synaptic strengths, networks of the same underlying structure can behave unpredictably. Because of this, algorithms like SNEAP (Li et al., 2020 ) use spike traces from simulations of the network in order to better partition based on the network activity. In this work, we chose to develop a partitioning strategy that only depends on the connectivity of the input graph. For this reason, it makes sense that a partitioning method for reconfigurable hardware should be performed on a purely topological basis. Various types of SNN toplogies exist, including liquid state machines (Maass, 2011 ), deep spiking networks (Sengupta et al., 2019 ), and large-scale brain models (Potjans and Diesmann, 2012 ). This is due to the fact that it is not feasible to simulate large-scale models for activity data every time the model needs to be partitioned. Splitting an input network into a balanced set of partitions is known as the NP-hard balanced graph partitioning problem . Several balanced graph partitioning approximation methods exist, including METIS (Karypis and Kumar, 1999 ), a multilevel partitioning scheme, which is commonly used for its speed, flexibility, and performance. Solutions like Spinner (Martella et al., 2017 ), which runs on Giraph (a large-scale graph analytics platform) easily scale to massive graphs and a large number of compute cores. Streaming graph algorithms such as FENNEL (Tsourakakis et al., 2014 ) also offer very fast balanced partitioning solutions, where vertices are partitioned one-by-one, minimizing the computation required. These algorithms run on both weighted and unweighted undirected graphs, and have typically been used to partition very large graphs of social media networks or in very large-scale integration (VLSI) circuit partitioning (Alpert et al., 1996 ). There are some previous works on SNN mapping methods to neuromorphic platforms. Some of these methods include PACMAN (Galluppi et al., 2012 ), SCO (Lee et al., 2019 ), and SpiNeMap (Balaji et al., 2020 ). PACMAN, which is the partitioning and configuration framework for SpiNNaker (Painkras et al., 2013 ), partitions on a population level, and then sequentially maps the result to a huge number of ARM processors emulating SNN cores. This works well for the torus interconnect in SpiNNaker but clearly doesn't suit our hierarchical tree-based interconnects. The hierarchical tree-based interconnects are more scalable due to the fact that new branches can be added to the tree that maintains constant local bandwidth, in contrast to linear network-on-chip where congestion can arise from a large number of connections (Park et al., 2017 ). SCO minimizes the hardware resources for network execution but doesn't have any performance gains in reducing global communication traffic. SpiNeMap reduces the power consumption and latency for crossbar-based neuromorphic cores where the communication fabric is a single shared time-multiplexed interconnect. Previously, METIS has been used to partition spiking networks in Barchi et al. ( 2018b ) and Li et al. ( 2020 ). However, in the former, the cortical microcircuit network used for analysis is quite small, scaled down to only roughly 4,000 neurons and seven hundred thousand synapses. In the latter, spike trace information is used as weights during the METIS partitioning, the mapping strategy used does not take into account a HiAER network structure, and the networks used are relatively small. Additionally, there is no experimentation on different toplogies of spiking neural networks. Other works, such as Barchi et al. ( 2018a ), use several other graph partitioning methods such as spectral analysis, and simulated annealing in order to partition their input SNN, but do not extend its methods to different layers of network hierarchy. These previous works are not fully optimized for a HiAER network structure and to our knowledge, there is no existing work that has optimized the partitioning and mapping algorithm to be implemented using a HiAER framework. As HiAER offers the maximum flexibility in network connectivity as well as provides the highest scalability, our hardware-aware partitioning algorithm has the potential to scale to networks at the scale of the human brain.",
"discussion": "4. Discussion The goal of this work was to create and evaluate a flexible method of hierarchical partitioning for use in large-scale neuromorphic systems. We designed and developed an algorithm for performing this partitioning on various networks including synthetically generated networks, small-world networks, deep feedforward networks, and the fly hemibrain. In all cases, we observed a significant improvement of the hierarchical partitioning method over randomly balanced neuron placement and even flat METIS partitioning. The method is very flexible and scales to any number of layers of hierarchy, allowing fast reconfigurability and improved performance for neuromorphic systems. The hierarchical partitioning algorithm mainly takes advantage of the structure of an input network. In many cases, graphs that have a large number of random connections will not result in a reduction in message traffic. This was observed both in the synthetic networks at high spread factor and in the small-world graphs with large fan-out. However, as demonstrated in the fly hemibrain connectome, real networks can be partitioned and simulated with significant benefits. Further analysis must be done in order to validate the advantages of this partitioning method on data from more complex organisms, such as mice and humans. We also evaluated the cost/benefit of unicast and multicast communication protocols over the network. While unicast message reduction often benefits significantly more from hierarchical partitioning than from multicast, overall multicast communication allows for a drastic reduction in the overall number of messages and total network traffic. Multicast communication also benefits from requiring a fixed mask in each message, while unicast communication requires the full destination address. With the optimal placement generated from the hierarchical partitioning algorithm, multicast communication typically can take better advantage of the structure to reduce the number of relay messages. Many neuromorphic systems such as Loihi (Davies et al., 2018 ) and TrueNorth (Akopyan et al., 2015 ) use a mesh NoC on a single chip. These topologies are popular because of their low complexity and planar 2D layout properties. It is easier to generate the most optimal placement of the neurons and cores in this type of mesh. However, large planar systems may suffer from excessive hop counts when communicating end-to-end. The hierarchical structure we discussed has the advantage of improving bandwidth for long-distance connections. For each level of hierarchy, the communication sparsity and address-space both increase in each level, multiplying to a constant bandwidth at each level of hierarchy. Assuming constant average fan-in, fan-out, and event-rate for each neuron, the hierarchical connectivity that we presented will scale linearly with the number of neurons in the network. If additional cores are needed for larger networks, the partitioning method has proven to be able to find an optimal arrangement for a particular hierarchical structure of cores. However, this structure may be significantly worse than the most optimal hierarchical structure for the network. Finding the optimal hierarchical structure may require prior information about the spiking network connectivity. One area of further study is the idea of an optimal hierarchical structure for the input network. If N neurons can be placed inside of a single core, then placing all N neurons in a single core eliminates network traffic but takes longer simulation time. On the other hand, distributing the neurons over all available cores will improve the speed of simulation to the extent that the router can handle all of the messages. In order to further validate the partitioning method, we plan to test this on reconfigurable neuromorphic digital hardware such as a field-programmable gate array (FPGA) such as in Pedroni et al. ( 2020 ), and analyze the total speedup and communication volume over the router, as well as try to identify quantitative guidelines to determine the optimal partition. Additionally, further evaluation can be done on more complex network topologies, such as spiking Convolutional Neural Networks (Lee et al., 2018 ), and larger real biological networks. This opens the door to fast and efficient neural simulations with neuromorphic hardware on a very large scale."
} | 4,387 |
37111876 | PMC10141480 | pmc | 2,866 | {
"abstract": "The pollution of soil by trace elements is a global problem. Conventional methods of soil remediation are often inapplicable, so it is necessary to search intensively for innovative and environment-friendly techniques for cleaning up ecosystems, such as phytoremediation. Basic research methods, their strengths and weaknesses, and the effects of microorganisms on metallophytes and plant endophytes resistant to trace elements (TEs) were summarised and described in this manuscript. Prospectively, bio-combined phytoremediation with microorganisms appears to be an ideal, economically viable and environmentally sound solution. The novelty of the work is the description of the potential of “green roofs” to contribute to the capture and accumulation of many metal-bearing and suspended dust and other toxic compounds resulting from anthropopressure. Attention was drawn to the great potential of using phytoremediation on less contaminated soils located along traffic routes and urban parks and green spaces. It also focused on the supportive treatments for phytoremediation using genetic engineering, sorbents, phytohormones, microbiota, microalgae or nanoparticles and highlighted the important role of energy crops in phytoremediation. Perceptions of phytoremediation on different continents are also presented, and new international perspectives are presented. Further development of phytoremediation requires much more funding and increased interdisciplinary research in this direction.",
"conclusion": "7. Conclusions Soil pollution of TEs is a serious problem in the modern world. Unlike air or water pollution, soil-polluting TEs remain there much longer than other elements of the biosphere. All TEs in high concentrations are toxic to humans, animals, plants and microorganisms. Conventional soil remediation methods are often inapplicable, so it is necessary to intensively search for innovative and environmentally friendly techniques for ecosystem clean up using phytoremediation. Phytoremediation, referred to as green technology, is widely used to remediate soils contaminated with TEs and is used to treat sediments, groundwater and surface water. Like any method and this procedure has both advantages and disadvantages. In recent years, great progress has been seen in improving the efficiency and quality of the phytoremediation process. This method, combined with burning the resulting biomass to produce heat and electricity, may prove to be one of the key techniques for environmental clean-up [ 302 ]. At this stage, it seems essential to create effective transgenic plants that are good phytoremediators. Thus, a huge challenge is to obtain genetically modified plants that will result in the ability to accumulate pollution in their large biomass. In the case of TEs, the preference is focused on aboveground parts that can then be easily harvested. Maintaining translocation from the root to the shoot, followed by sequestration in vacuoles and/or other parts of the cells of the plant’s aboveground organs, are the most commonly used strategies for genetic modification [ 9 ]. Genetically modified plants should also exhibit high viability and be more resistant to environmental stress, which will make them better competitors among native plant varieties. In addition, it is important for scientists to understand the mechanisms of natural phytoremediation, which is still not fully understood. Until these undiscovered mechanisms are clarified, the trial-and-error method seems to be the only reasonable tool [ 303 ]. Purification of soils on an industrial scale will most likely be possible in the future through the use of genetically modified organisms. It is estimated that over the next 25 years, the European Union will allocate about 100 trillion euros to clean up degraded areas [ 222 ]. It is, therefore, necessary to intensify the research being carried out in this direction in order to create a plant that can remove and accumulate these pollutants sparsely and in large quantities as soon as possible. Today’s engineering bioremediation offers quite a few effective solutions in the form of the use of various organic substances (e.g., sewage sludge, sorbents, enzymatic and microbial preparations or nanoparticles). However, it is extremely important that the preparations or sorbents used do not adversely affect the environment and are easily and quickly biodegradable. This is because ignorance and unawareness of the far-reaching effects of their use can be a danger. The technique of assisting bioremediation with genetic engineering still arouses much controversy. There are a number of restrictions on its use. This is due to strict regulations and safety considerations. It should be remembered that there is always a significant risk of gene transfer from transgenic plants or microorganisms to the environment. Another huge drawback is that genetic research on microbiota and plants capable of efficient phytoremediation is usually conducted in specialized laboratories, which unfortunately does not reflect natural conditions. Great hope has been placed in international projects. One such project currently underway in Poland is the international ‘GOLD’ project called “Bridging the gap between phytoremediation solutions on growing energy crops on contaminated lands and clean biofuel production,” which has received sizable funding from Horizon in 2020. This project is very important because naturally polluted areas are being studied. This is because the above practices are translated into reality (two municipalities in Upper Silesia). Based on this research, universal strategies will be developed that can be applied to other potentially contaminated sites and used in various countries in the European Union and Asia. Thus, phytoremediation is becoming one of the elements of both an integrated and sustainable approach to the revitalization of polluted areas and the protection and shaping of the space in which we live. The future of phytoremediation development must therefore involve the development of technologies for the utilitarian use of the biomass obtained. Remediation of polluted soil is time-consuming and, in hyperaccumulating plants, takes 2–60 years, while in non-hyperaccumulating plants, it takes 25–2800 years [ 230 ]. Phytoremediation may be a viable option for the removal of TEs contamination from environments, as the biomass created in the process could be economically used in the form of bioenergy [ 304 ]. A holistic approach is therefore needed to assess the effectiveness of phytoremediation, requiring the joint efforts of engineers, agronomists, plant biologists and microbiologists to work together with policy makers, regulators and industry representatives. Key tasks for phytoremediation are the valorization of phytoremediation biomass to offset remediation costs. In addition, it is clear that all stakeholders expect the creation of phytoremediators that will ensure that all risks are minimized while maximizing both economic, ecological and social benefits.",
"introduction": "1. Introduction As a result of the ongoing industrialization of the world, which undoubtedly brings considerable economic benefits, the pollution of the natural environment has increased significantly. Soil is the largest reservoir of chemical pollutants, including trace elements, and it is a key element in the soil-plant-animal-human trophic chain [ 1 ]. Therefore, the pollution of soils with trace elements (TEs) and metalloids poses a threat to the normal function of the pedosphere. TEs are metallic elements with a density of more than 4.5 g·cm −3 . They are characterized by relatively high atomic weights and atomic numbers [ 2 ]. They have adverse effects on living organisms and, when in excess, block basic life processes [ 3 , 4 , 5 , 6 ]. TEs do not decompose in biological and physical processes; therefore, they persist in the soil and present a long-term (thousands of years) environmental threat [ 7 , 8 , 9 , 10 , 11 , 12 , 13 ]. For example, lead (Pb) can persist in soil for more than 150–5000 years and remains at high concentrations for up to 150 years after sludge application to the soil [ 14 , 15 ], whereas the biological half-life of cadmium (Cd) is approximately 10–30 years [ 15 , 16 ]. Therefore, TEs are an important factor limiting the abundance, activity and biodiversity of microorganisms and plants [ 8 , 17 ]. Their sources can be divided into natural and anthropogenic [ 2 , 18 ]. TEs can come from two sources—natural (products of bedrock weathering, volcanic eruptions, ocean evaporation, forest fires) and anthropogenic (mining, metallurgy, municipal and household waste, sewage discharges, industrial and commercial activities, oil industry, warfare, nuclear power plants, use of agrochemicals, active and inactive military zones—weapons testing, bomb disposal, shooting exercises) [ 9 , 11 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 ]. TEs are persistent inorganic chemicals. They have cytotoxic, genotoxic and mutagenic effects on plants [ 34 ]. We can divide these elements into: essential micronutrients for plants (Cu, Fe, Mn, Mo, Ni and Zn), non-essential elements or toxic, even in small amounts, elements for plants (As, Cd, Co, Cr, Hg, Pb, Sb, Cr) [ 35 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 , 49 , 50 , 51 ]. TEs can limit important processes such as enzymatic activity and photosynthesis [ 52 ]. These negative impacts occur because metals disrupt regular metabolic pathways in plants [ 53 ]. Micronutrients are usually components of enzymes and other proteins crucial to metabolic processes. When the concentration exceeds the threshold value, these TEs become toxic for plants. For example, excess arsenic (As) causes photosynthesis inhibition and decreases biomass and yield. Cadmium (Cd) is a highly toxic TE due to its fast mobility and persistency. A very small concentration of Cd is lethal to plants [ 37 ]. Cadmium toxicity causes chlorosis, reduced water and nutrient uptake, browning of root tips, and ultimate death. Chromium (Cr) and lead (Pb) stress cause reduced nutrient uptake and disturbance in metabolic pathways, respectively. Mercury (Hg) and zinc (Zn) toxicity cause reduced photosynthesis due to the inhibition of photosystems I and II. Furthermore, excess nickel (Ni) causes retarded seed germination, reduced plant height, reduced root length, and also reduced chlorophyll content [ 37 ]. Plants, to defend themselves from the negative effects of TEs, use their defense systems [ 54 ]. At the very beginning, plants use an avoidance strategy, which involves limiting the uptake of TEs or blocking their access to the root. This can involve sequestration, immobilization or complexation of metals through root exudates [ 37 , 55 ]. If the previously mentioned defense systems are not sufficient, plants activate TEs tolerance mechanisms, such as metal ion trafficking, metal binding, metal chelation, accumulation of osmolytes and osmoprotectants or intracellular complexation [ 56 ]. However, the presence of significant amounts of TEs in the soil inhibits the development and activity of microorganisms, which leads to the disruption of processes related to the decomposition and transformation of organic matter [ 57 , 58 ]. The deficit of soil microbes and humic compounds contributes to ionic imbalance and increases the pool of bioavailable forms of TEs in the soil sorption complex [ 59 ]. There are various methods of cleaning up an environment contaminated with TEs. Conventional physicochemical methods require significant financial resources and most often involve the complete replacement of the contaminated soil layer [ 60 , 61 ]. These methods are also energy-intensive and produce large amounts of toxic waste [ 62 , 63 ]. The cost of conventional methods is estimated at $10–1000 per m 3 of soil. Methods that involve the treatment of soils with plants are much cheaper (about $0.05 per m 3 of soil) and more effective [ 64 ]. The degree of bioaccumulation (the accumulation of harmful substances in the plant) depends on various factors, e.g., the TEs content in the soil, the organic matter content, the soil type and structure, soil moisture, soil pH and the plant species [ 65 ]. There is a group of plants that have developed a number of mechanisms (e.g., polypeptides called phytochelatins) that allow TEs to accumulate in their tissues [ 64 ]. However, the synthesis of phytochelatins depends on the plant organ, the duration of exposure and the concentration of metal in the medium. It is also worth mentioning that this process is associated with slower plant growth [ 66 , 67 , 68 , 69 ]. Therefore, it is important to look for innovative solutions to clean up endangered ecosystems. Biological methods are becoming increasingly important. Numerous scientific studies have shown that certain plant species, thanks to their specific characteristics, have both the ability to take up and degrade xenobiotics polluting the environment. The effectiveness of phytoremediation for the treatment of heavily contaminated soils is generally low, as the plants used take up a hundred percent small amounts of TEs. This would require their use for hundreds of years, with the reduction or complete removal of emission sources. On the other hand, they can be effective for the reclamation of less polluted soils located along traffic routes or parks, squares and urban greenery, i.e., places of frequent residence of various age groups. The idea of using plants to reduce and level pollution in the environment has been known for a long time. In addition to aesthetic value, protection from noise, providing oxygen, plant species with high phytoremediation abilities planted in urban areas (maple leaf plane, Japanese larch, poplar, ash, field maple, white and sessile dogwood, wrinkled rose, common yew), have the opportunity to play a health-promoting role. This is because they contribute to a significant improvement in the urban environment in which we live. The ability of plants to take up TEs and accumulate PAHs and particulate matter (products of traffic pollution) in the wax overhang makes phytoremediation a very attractive technology dedicated to urban areas. Plants with phytoremediation capabilities act as a “green liver” in the urban environment. From the literature collected for this review article, it appears that research to improve and refine phytoremediation methods is practiced and actualized, but further steps in this direction are still needed. In practice, the use of only one method or treatment for effective phytoremediation will not be sufficient or satisfactory. Plant-microbiome interactions are proving to be an extremely effective approach for TE uptake and translocation in plants. Our work holds high hopes for further exploration of new metabolites and pathways for the efficient degradation of contaminants through the plant-microbiota system. With modern bioengineering techniques, it is possible to modify plants with desirable traits, as well as to isolate microorganisms and then introduce them into the soil to improve phytoremediation using appropriate plant species. In-depth and interdisciplinary research in this direction with significantly increased funding is needed in order to obtain, through these modifications, both plants and microorganisms that are effective in the remediation of contaminated land and, in addition, resistant to difficult and often changing environmental conditions. For this literature review, papers from 2000–2023 were used. Older papers were used only for the clarification of terms. Data were searched in Scopus, PubMed, Web of Science, ScienceDirect, Public Library of Science and AGRO databases. Search engines such as Google Scholar, MDPI Search and ResearchGate were also used. Searches were mainly conducted by using key-words, synonyms, combining terms and database search limits, e.g., source type and topic."
} | 3,999 |
29396559 | PMC5797222 | pmc | 2,867 | {
"abstract": "Ostreobium sp. (Bryopsidales, Ulvophyceae) is a major microboring alga involved in tropical reef dissolution, with a proposed symbiotic lifestyle in living corals. However, its diversity and colonization dynamics in host’s early life stages remained unknown. Here, we mapped microborer distribution and abundance in skeletons of the branching coral Pocillopora damicornis from the onset of calcification in primary polyps (7 days) to budding juvenile colonies (1 and 3 months) growing on carbonate and non-carbonate substrates pre-colonized by natural biofilms, and compared them to adult colonies (in aquarium settings). Primary polyps were surprisingly already colonized by microboring filaments and their level of invasion depended on the nature of settlement substrate and the extent of its pre-colonization by microborers. Growth of early coral recruits was unaffected even when microborers were in close vicinity to the polyp tissue. In addition to morphotype observations, chloroplast-encoded rbc L gene sequence analyses revealed nine new Ostreobium clades (OTU99%) in Pocillopora coral. Recruits and adults shared one dominant rbc L clade, undetected in larvae, but also present in aquarium seawater, carbonate and non-carbonate settlement substrates, and in corals from reef settings. Our results show a substratum-dependent colonization by Ostreobium clades, and indicate horizontal transmission of Ostreobium -coral associations.",
"introduction": "Introduction Reef-building scleractinian corals are associated with complex microbial communities, distributed in the mucus, tissue and skeletal compartments 1 . Microborer communities that actively penetrate carbonate skeletons by chemical means 2 are among the relatively less studied partners of coral holobionts. Microborers (euendoliths) include filamentous cyanobacteria, fungi and algae (Chlorophytes and Rhodophytes) 2 , 3 . In living corals, the low-light environment inside the skeleton (often <1% of the photosynthetically active radiation) 4 , 5 selects a few adapted euendolithic species. These microboring communities are usually dominated by the photosynthetic Ulvophyceae Ostreobium spp. (Chlorophyte) 6 , 7 , which was until recently called: Ostreobium quekettii Bornet et Flahault 8 . Dense populations of this euendolithic siphonous alga form visible green bands in the skeleton of adult colonies of massive, slow-growing corals of the genus Porites sp. 7 , 9 . In fast-growing branching corals such as Stylophora pistillata , euendolithic filaments gradually decrease in density upwards and the colored bands are absent 10 . Molecular tools have during the past 10 years revealed a huge genetic diversity of Ulvophyceae that penetrate living scleractinian corals, with special attention to the siphonous euendolithic Ostreobium . The Ulvophyceae class contains multiple orders, including Ulvales (encompassing for instance families Ulvaceae, Phaeophilaceae), and Bryopsidales with 3 suborders (Ostreobidineae, Halimedineae and Bryopsidineae) 11 , each encompassing multiple families. Seven phylotypes of the RuBisCo large subunit of the chloroplast-encoded gene ( rbc L) were recorded in Ostreobium colonizing 2 species of massive corals from the Red Sea along a depth gradient 12 . A recent study revealed four families within the Ostreobidineae suborder, using the tuf A plastid gene coding for the protein elongation factor EF- Tu as a metabarcode marker 13 while exploring the diversity of Ostreobium spp. and associated endolithic green algae in limestone substrates. An environmental genome survey (plastid 16S rDNA and rbc L; nuclear18S rDNA) 14 focused on Ostreobium phylogeny and ecology, and suggested a possible coevolution between Ostreobium and the main coral endosymbionts, the dinoflagellates Symbiodinium sp. In an independent study of the skeletal microbiome of massive slow-growing coral genera from various Pacific habitats, Marcelino and Verbruggen (2016) 3 determined that the tuf A Ostreobium clade includes more than 80 taxonomic units at the near-species level. They combined tuf A with ribosomal RNA gene markers (nuclear 18S rDNA, and plastid 16S and 23S rDNA) and suggested that Ostreobidineae form a complex that has evolved over the last 500 million years. None of these studies, however, investigated the possible relationship between the developmental stages of the coral host and the Ostreobium clade diversity, and the colonization dynamics of microborers in corals remained unclear. Early life stages of corals offer the unique opportunity to investigate the transmission of Ostreobium -coral associations and the mechanisms of holobiont assembly. The development of corals starts with a critical recruitment step, when a planktonic larva (planula) swimming in seawater settles to become the benthic primary polyp. Substrate settlement triggers larval metamorphosis and the onset of skeletal deposition, forming elements of the calcareous basal plate of the initial primary polyp within 24 h after settlement 15 , 16 . Secondary polyps then develop by clonal budding at the periphery of the primary polyp, forming the juvenile colony, which further grows into the adult colony, building the framework of reef ecosystems. In coral recruits, the polyp tissue layers cover tightly the growing carbonate skeleton, which may prevent colonization by microborers from the surrounding seawater. Thus, it is hypothesized that colonization occurs most likely through the substrate of larval settlement. However, this has not been experimentally tested. In reefs, coral larvae generally settle on biogenic carbonates such as dead coral rubble covered by coralline crustose algae 17 , substrates which are natural reservoirs for microborers 18 , 19 . However, coral larvae can sometimes settle on artificial, non-carbonate substrates 16 covered by epilithic biofilm-forming microorganisms a priori free of microborers. Here we aim to study the microborer colonization process of coral recruits and to address the following questions: (i) At what developmental stage does the colonization of coral skeleton by microborers take place, and how fast does it spread? (ii) What are the sources and reservoirs of colonizing microborers? (iii) Does the dominant Ostreobium clade depend on the site origin of the host and does it change in the course of the coral development? We present a pilot study of colonization dynamics by microborers of early life stages of the coral Pocillopora damicornis type beta 20 , 21 . This species is a pioneer, fast-growing branching coral in tropical Indo-Pacific reefs 22 , with a life cycle that can be completed in captivity. Larval settlement experiments and long-term coral cultures in closed-circuit at ATPD-aquarium (Aquarium Tropical, Palais de la Porte Dorée, Paris, Fr) provide controlled conditions to study and map microborer abundance and distribution at unprecedented high temporal resolution in coral skeletons of three early life stages (7 days, one month, and three months post-metamorphosis), and in adult fully grown colonies. Two carbonate (dead Porites skeleton, calcite spar) vs two non-carbonate (plastic lumox ® -Sarstedt - and underwater paper) settlement substrates, previously pre-colonized (during 2–3 or 7 months) by microborers vs epilithic biofilms, were compared for their influence on microborer colonization of coral recruits. In addition to morphological criteria to detect Ostreobium filaments, the rbc L gene was amplified and Sanger sequenced to highlight the dominant Ostreobium clades in early coral life stages, their corresponding environmental seawater/settlement substrates, and adult colonies of Pocillopora corals from 3 French aquaria and 2 reef sites.",
"discussion": "Discussion At the reef scale, Ostreobium sp. is the main agent of microbioerosion 25 , 26 , i.e. the principal carrier of biogenic dissolution of carbonates. Despite host skeletal erosion, a few authors have proposed that this phototrophic euendolith may improve the survival of thermally stressed corals during bleaching events (e.g. Oculina patagonica in the eastern Mediterranean Sea) via a transfer of alternative photoassimilates to the host 27 , 28 . To clarify their role in the coral holobiont, it is important to understand the sequence of microborer colonization of the skeletons of living corals. Evaluating the timing of the establishment of coral-microborer association, the potential colonization of specific Ostreobium clades during that process, and its effect on early coral life stages is especially relevant in the current context of anthropogenic global changes. Carbonate dissolution of reef substrates by microborers is indeed enhanced by eutrophication and rising aqueous pCO 2 29 , 30 so that the combination of several environmental factors including rising sea surface temperatures threaten the coral recruitment and growth 31 , 32 . Here, we provide the first evidence on the entry of the euendolithic alga Ostreobium during the early ontogeny of a coral host. We show that colonization occurs within a few days (between 2 and 7 days) following the larval settlement on pre-colonized substrates, at the onset of skeletal deposition and basal plate formation in the primary polyp. Microboring filaments penetrate from the settlement substrate into the skeletons of coral recruit in a similar way as previously observed in live thalli of the coralline alga Hydrolithon onkodes 18 . The observed phototrophic (and phototropic) euendolith, Ostreobium must keep up with the vertical extension of its also largely phototrophic host to survive, taking in consideration that coral’s tissue-shaded skeletons constitute an extreme low light environment 4 , 5 . This adaptation was first suggested by Le Campion-Alsumard et al . 7 while explaining the pattern of green bands in adult colonies of the massive coral Porites , which correlated with periods of slower coral growth, permitting dense growth of multiply ramified Ostreobium filaments. Here, in fast accreting adult colonies of the branched Pocillopora damicornis , no colored bands underneath coral tissues were observed, instead, the euendolith abundance decreased by a factor 6 towards the tissue-covered branch tips. This pattern is similar to that found in the branching species Stylophora pistillata 10 , where fast growth leaves the euendoliths behind, resulting in an upward-progressive filament density decrease. However, no such virtual filament ‘dilution’ effect could be observed during the early deposition and expansion of carbonate in juvenile coral colonies. The rapidly growing coral recruits developing on heavily pre-colonized dead Porites were invaded by euendolithic microorganisms, which expanded through the polyp skeletal basal plate up to the areas in close vicinity to the coral tissue. We show for the first time that this early colonization did not slow the host extension rates, indicating that the fitness of recruits was not altered by early assembly of the coral-microborer association. Pre-colonized dead Porites skeleton fragments with highest concentration of endolithic Ostreobium proved also to be the best source for euendolith recruitment to coral juveniles that settled on them, providing a faster and more extensive microborer colonization compared to calcite and non-carbonate substrates. This pattern, together with similar succession of microboring communities in dead Porites skeletons, becoming mature i.e. dominated by Ostreobium after 7 months of exposure in our controlled aquarium settings (ATPD), just as shown in natural coral reefs 26 , indicates the representativeness of our ex-situ model to in situ processes. A really unexpected result of the present study was the early colonization by microborers of coral juveniles settled on non-carbonate substrates (plastic and paper). We checked that the soft tissues of these recruits covered the entire skeleton and were not damaged. We then hypothesized that some microborers were present on non-carbonate substrates as epilithic forms. Indeed, Golubic et al . 33 showed that Ostreobium filaments can exit coral pores and become temporary crypto-endolithic organisms. Kobluk and Risk 34 also suggested that Ostreobium filaments can exit their carbonate substrate to become epilithic. We prove the correctness of this hypothesis by observing filaments typical of Ostreobium within epilithic biofilm developed at ATPD, and by amplifying the Ostreobium rbc L gene on non-carbonate substrates covered by natural biofilms (plastic and paper) and in seawater. Ostreobium filaments have been reported to display sporangial bags, allowing the production of quadriflagellate zoospores in the environment 35 . However, the life cycle of Ostreobium remains poorly known, especially regarding the way spores are expelled from the presumed sporangial bags located inside the substrate into the seawater. We propose that spores or detached fragments of this siphonous alga became trapped in the epilithic biofilms and started to develop into filaments while inside the primary corallite of the coral recruits (skeleton of the primary polyps). The molecular aspects of our study confirm that microboring communities in living corals are reservoirs of new Ostreobium diversity. Indeed, 9 new Ostreobium rbc L clades (OTU99%) were detected in branching Pocillopora damicornis (type beta 21 ) from long-term aquarium cultures, P. verrucosa from Eilat reef (Haplotype A and E 36 ) and Pocillopora sp. from a New Caledonian reef, of which only clade K (OTU99%) was previously known from a shallow water massive Porites coral from the Red Sea. These results provide new information on the diversification of the genus Ostreobium , adding to the growing datasets of operational taxonomical units, which have recently been reported from the skeleton of mostly massive scleractinian corals, using the rbc L gene as well as complementary gene markers 3 , 11 , 13 , 14 . Here, the association of Pocillopora corals with Ostreobium clades seems quite variable and substratum-dependent. It should be noted that the Ostreobium clades present in settlement substrates may be affected by the geographic distribution 14 of these clades and their depth distribution 12 . Further investigations will require further broad scale sampling using for example metabarcoding. The strongly supported Ostreobium rbc L clade P1 (OTU99%, close to clade K) dominated coral grown in ATPD aquarium, both in adults and juveniles, and was absent from larval stages but present in environmental reservoirs such as dead Porites carbonate substrates as well as non-carbonate substrates and seawater. Altogether these data support horizontal intergenerational transmission of the Ostreobium -coral association. Besides, the P1 clade was confirmed in adults from another aquarium settings (Océanopolis) and in situ in Red Sea shallow reef settings in a parent Pocillopora verrucosa species, suggesting widespread occurrence in Pocillopora . Future studies are needed to further investigate this molecular diversity and the multiple factors and communication mechanisms potentially driving the selection of Ostreobium clades at various stages of the host development. More generally, the functional interactions between microborers of specific Ostreobium clades, and their living coral host need to be studied to understand the capacity of corals to adapt to changing environmental conditions. In conclusion, this study provides novel information on the timing of microborer colonization of early coral recruits, which occurs as early as 7 days post-metamorphosis in the primary polyp, before budding into a juvenile colony. Recruits settled on non-carbonate substrates are also colonized by Ostreobium clades, indicating the existence of life stages (propagules) in seawater and among epilithic biofilms which are able to penetrate newly deposited carbonate skeletons. Occurrence analyses indicate substratum-dependent Ostreobium -coral associations, with a widespread rbc L clade P1 of Ostreobium , detected both in environmental reservoirs and across Pocillopora sp. benthic life stages. Its dominance in aquarium microcosms suggests that it may be an ecologically dominant strain in Pocillopora corals but this needs further exploration in natural reef settings. Combined together this data show horizontal transmission of the Ostreobium -coral associations. Interestingly, the presence and abundance of microborers did not affect the coral skeletal extension rates, and deeply penetrating borings were observed in close proximity to coral tissues. These findings reveal the early incorporation of microborers into the coral holobiont, and have implications for a potential role of these microbial associates on coral host health and development."
} | 4,232 |
32050664 | PMC7077729 | pmc | 2,868 | {
"abstract": "Fluorinated (F6) and zwitterionic, as well as phosphorylcholine (MPC) and sulfobetaine (MSA), copolymers containing a low amount (1 and 5 mol%) of 3-(trimethoxysilyl)propyl methacrylate (PTMSi) were prepared and covalently grafted to glass slides by using the trimethoxysilyl groups as anchorage points. Glass-surface functionalization and polymer-film stability upon immersion in water were proven by contact angle and angle-resolved X-ray photoelectron spectroscopy (AR-XPS) measurements. Antifouling performance of the grafted films was assayed against the yeast Candida albicans , the most common Candida species, which causes over 80% of candidiasis. Results revealed that the F6 fluorinated, hydrophobic copolymers performed much better in reducing the adhesion of C. albicans , with respect to both corresponding zwitterionic, hydrophilic MPC and MSA counterparts, and were similar to the glass negative control, which is well-known to inhibit the adhesion of C. albicans . A composition-dependent activity was also found, with the films of copolymer with 99 mol% F6 fluorinated co-units performing best.",
"conclusion": "4. Conclusions Fluorinated (F6) and zwitterionic (MSA and MPC) copolymers containing low amounts of trimethoxysilyl groups were synthesized, and films therefrom were firmly anchored onto glass slides, for surface functionalization and chemical modification via a “grafting-to” reaction. AR-XPS measurements proved that the grafted polymer films were stable upon contact with water at least up to seven days of immersion, including the water-soluble copolymers p(MPC- co -PTMSix). It was also found that the film surface of fluorinated copolymers was largely populated by fluorinated moieties, both before and after immersion in water, consistent with their hydrophobic, low-surface-energy character. However, the film surface of zwitterionic copolymers was enriched in silicon, resulting in a more hydrophobic character than expected of their hydrophilic, high-surface-energy nature. Biological assays against the most aggressive fungal pathogen C. albicans clearly pointed out that fluorinated copolymers were much more able to reduce the adhesion of C. albicans than the corresponding zwitterionic copolymers. The amount of F6 counits in the copolymer played a role in decreasing cell attachment, with the copolymer p(F6- co -PTMSi1) richer in fluorinated component being the best performer. Therefore, this work helps to overcome some of the discrepancies present in the literature about the role of zwitterionic and fluorinated polymers in preventing the adhesion of C. albicans and can provide initial guidelines for the synthesis of polymers with a tailored hydrophilic/hydrophobic balance to combat biofouling. The question of the contrasting “preferences” of biofouling agents to adhere and settle on surfaces is highly relevant in diverse fields of advanced applications. For one example, such fluorinated copolymers might be considered as surface modifiers of biomedical silicones, to be potentially employed in other-than-typical dental materials, to enhance resistance to C. albicans colonization.",
"introduction": "1. Introduction Biofouling is generally an undesirable phenomenon that involves the organic matter deposition and/or organism colonization of surfaces upon contact with biological fluids, freshwater or seawater [ 1 , 2 ]. Thus, biofouling affects medical, industrial and marine fields, resulting in detriments to health and environment and increased operational costs [ 3 , 4 , 5 ]. In the medical field, biofilm-associated microorganisms are responsible for up to 80% of all microbial infections in humans [ 6 , 7 ]. Biofilms are linked to recurrent invasive infections that are difficult to eradicate because of their intrinsic resistance to antimicrobial treatments and host defense mechanisms and their excellent ability to adhere to biomaterials [ 8 ]. The increasing use of biomaterials and medical devices, such as catheters, stents, prostheses, contact lenses and implants, has led to a concomitant increase in the incidence of device-related infections, with the most common fungal infection due to Candida albicans [ 9 , 10 ]. Management of biofilm-associated Candida infections can be challenging due to the intrinsic drug-resistant phenotype of sessile fungal cells, and removal of the infected device is often required [ 9 ]. However, removal of the contaminated implant can be accompanied by complications, negatively affecting the patient’s condition and the economic burden. Therefore, prevention of biofilm-associated infections currently represents a major challenge. The surface properties, notably the surface chemical composition, of materials susceptible to fouling are known to significantly affect the biofilm formation on medical devices [ 11 ], as well as the micro- and macro-fouler colonization of ship hulls, maritime equipment and industrial implants [ 12 , 13 , 14 , 15 ]. One of the most promising strategies to overcome this problem is to develop anti(bio)fouling surfaces, which prevent microorganism adhesion. The first interactions occurring among C. albicans cells and materials surfaces are usually hydrophobic interactions and electrostatic forces, taking place within the first 12 h; subsequent stages involve a stronger adhesion, displayed by cell-wall glycoproteins (e.g., Als or Hwp1 proteins) [ 16 ]. This leads to formation of microcolonies (3–4 h) and biofilm aggregates organized in a bilayer composed of yeast and hyphal cells embedded in a self-produced extracellular polymeric matrix (11–30 h). C. albicans mature biofilm is then consolidated up to 38–72 h [ 17 ]. Among the different parameters that are reported to possibly affect Candida adhesion, including surface preconditioning with different biological fluids [ 12 ], surface roughness [ 8 , 9 , 10 , 11 ] and surface charge [ 9 ], one of the most relevant is surface wettability. It is widely acknowledged that C. albicans attaches more rapidly to hydrophobic, nonpolar surfaces, such as Teflon and other plastics, than to hydrophilic surfaces, such as glass and metals [ 18 ]. However, results of these studies have at times been contradictory, due to the absence of standardized methods for determining surface hydrophobicity, and the wide variability observed between different Candida species, as they may show differences in the cell-wall composition [ 19 ]. A considerable extent of variability in adhesive properties has also been described even among C. albicans isolates [ 20 ]. For instance, Kang et al. evaluated the effect of surface properties of four different types of denture-lining material (tissue conditioners, acrylic and silicone soft liners and hard reline materials) on the adhesion of C. albicans . Surface-energy parameters of the different materials were evaluated, and the data obtained indicated that acrylic soft liners were more hydrophilic than other materials and, together with tissue conditioners, displayed greater Candida adhesion than did silicone soft liners and hard reline materials [ 21 ]. Since hydrophobic interactions appear to play a key role in the adhesion of C. albicans to prosthetic materials [ 10 ], hydrophilic surface modification of biological devices has been regarded as an effective strategy to inhibit C. albicans colonization. Attempts have been performed to modify denture-base materials to make them more hydrophilic. In particular, Lazarin et al. coated denture-base acrylic resins with photopolymerized coatings containing zwitterionic or hydrophilic monomers and found out that those based on sulfobetaine methacrylate and 3-hydroxypropyl methacrylate significantly reduced the adhesion of C. albicans with respect to the untreated acrylic [ 22 , 23 ]. Superhydrophilic sulfobetaine-methacrylamide-based copolymers on denture surfaces were shown to enhance the hydrophilicity of the denture-base acrylic resin and reduce the initial adhesion of C. albicans [ 24 ]. Inorganic silica was also used for hydrophilic modification of acrylic denture resins, to reduce the colonization of C. albicans [ 25 ]. Acrylic resin surfaces were also modified by different plasma treatments, including argon, argon/oxygen and argon/sulfur hexafluoride atmospheres, in order to obtain surfaces with different wettability. Interestingly, it was found that both the hydrophobic Ar/SF 6 and the hydrophilic Ar/O 2 treated resins were able to significantly reduce the adhesion of C. albicans , regardless of the different initial contact angle values, the presence or absence of saliva and the surface roughness [ 26 ]. Silicone rubber is also used for oral cavity applications, including voice prostheses and denture soft-liners, and its surface modification with fluorinated trialkoxysilane has been proposed to enhance the resistance of silicone medical devices to C. albicans adhesion [ 27 ]. The present work aimed to better understand the effect of surface phobicity/philicity against the adhesion of the yeast pathogen C. albicans on hydrophobic, fluorinated polymers, as opposed to hydrophilic, zwitterionic polymers. In particular, we synthesized three sets of copolymers based on a hydrophobic fluorinated (F6) monomer, a hydrophilic zwitterionic, phosphorylcholine (MPC) or sulfobetaine (MSA), monomer with 3-(trimethoxysilyl)propyl methacrylate (PTMSi). The latter was incorporated to a minimal amount (1 and 5 mol%), to ensure covalent anchoring of the polymer film to a glass substrate. The copolymers were then used to produce films whose surfaces were investigated by contact angle and angle-resolved X-ray photoelectron spectroscopy measurements, both before and after immersion in water, to characterize the chemical structure of the polymer surface. Adhesion of C. albicans to the chemically different polymer films was assessed and related to the surface phobic/philic chemical nature. To the best of our knowledge, this is the first paper where the antifouling properties of zwitterionic and fluorinated copolymers against the yeast pathogen C. albicans are directly compared.",
"discussion": "3. Results and Discussion 3.1. Synthesis of Copolymers Zwitterionic copolymers p(MSA- co -PTMSix) and p(MPC- co -PTMSix) based on hydrophilic MSA and MPC were synthesized by free-radical polymerization at 50 °C, by using anhydrous methanol as solvent (10 mL g –1 monomers) and V50 (1 wt%) as initiator ( Figure 1 and Figure 2 ). V50 was chosen in place of more common thermal initiators, such as AIBN, owing to its higher solubility in methanol and its lower activation temperature ( t 1/2 = 13 h at 50 °C [ 32 ]). Fluorinated copolymers p(F6- co -PTMSix) based on hydrophobic (and lipophobic) F6 counterpart were synthesized in anhydrous toluene solution at 70 °C, by using AIBN (1 wt%) as free-radical initiator ( Figure 3 ). For any set of copolymers, an intentionally low amount of PTMSi counits (x = 1 and 5 mol%) was incorporated in the copolymer to warrant covalent anchorage of the copolymer to the glass surfaces through a “grafting-to” reaction during the polymer film preparation. The MPC-based copolymers were soluble in methanol, ethanol and water and poorly soluble in less polar solvents, such as THF and acetone. Therefore, they were purified by repeated precipitations from methanol solutions into THF. Differently, MSA-based copolymers were insoluble in ethanol, methanol and water and thus were purified by repeated precipitations from the very polar TFE) into methanol. The poor solubility of the MSA-based copolymers in water at room temperature was consistent with the known UCST behavior of the corresponding p(MSA) homopolymer. The fluorinated copolymers were purified by repeated precipitations from HFB solutions into hexane. The actual formation of copolymers was confirmed by 1 H NMR and 19 F NMR (p(F6- co -PTMSix)). The composition was calculated from the integral intensities of the signals of PTMSi at 3.6 ppm (OCH 3 ) or 0.7 ppm (SiCH 2 ) and the typical signals of the corresponding comonomers (4.47 ppm (COOCH 2 , F6), 3.27 ppm (N + (CH 3 ) 3 ), MPC) and 2.35 (C H 2 CH 2 SO 3 – , MSA)). The copolymerization runs were carried out up to almost complete conversion of monomers (95%–99%), which made the subsequent work-up steps much easier. The monomer conversion p was followed by 1 H NMR spectroscopy (see Figure 4 for one illustration example) and calculated according to Equation (1): p = 1 − I (6.05 t=x) / I (6.05 t=0) (1) \nwhere I (6.05 t= x ) and I (6..05 t=0) are the integrals of the signals of vinyl proton at 6.05 ppm at different reaction times and at the initial time, respectively. At such high final comonomer conversions, the copolymer composition was equal to that of the respective monomer feed. At the end of the polymerization, a suitable solvent (TFE, methanol and TFT for p(MSA- co -PTMSix), p(MPC- co -PTMSix) and p(F6- co -PTMSix), respectively) was added to dilute the copolymer solutions that were then stored under nitrogen at −20 °C. The molar masses and molar mass distributions of the copolymers could not be determined, owing to their poor solubility in the common solvents normally used for GPC analyses. Moreover, it was observed that the copolymers were difficult to handle because of the hydrolysis and sol-gel condensation of the Si(OCH 3 ) 3 groups of PTMSi counits. 3.2. Preparation of Polymer Films Polymer films of the three different classes of copolymers were prepared via a “grafting-to” approach to covalently anchor the films to the glass surface, in order to prevent their delamination during biological assays, especially of those containing hydrophilic zwitterionic MSA and MPC. Before functionalization, glass slides were activated by immersion in hot piranha solution, thus cleaning the glass surface and forming additional silanol (Si–OH) groups at the outermost surface layers (~20 Å). The formation of such Si–OH groups was also inferred by the decreased water contact angle from 28 ± 1° to 20 ± 3° in going from the non-activated to the activated glass. After cleaning, the glass slides were stored in deionized water. The Si–OH groups were then exploited as anchorage points to bind the polymer films through a condensation sol-gel reaction with the silanol functionalities derived from the hydrolysis of the pendant PTMSi trimethoxysilyl groups promoted by ambient humidity ( Figure 5 ). A 5 wt% copolymer solution was deposited by dip-coating on activated glass slides and thermally annealed at 130 °C in an oven, under vacuum, in order to remove the solvent and drive the cure reaction to completeness. The solvent of choice varied on the basis of the copolymer, being TFE, methanol and HFB for p(MSA- co -PTMSix), p(MPC- co -PTMSix) and p(F6- co -PTMSix), respectively. The polymer films were then copiously rinsed with and sonicated in water and TFT for zwitterionic and fluorinated copolymers, respectively, to remove the physically adsorbed polymer and residual traces of unreacted monomers. Films were finally annealed at 130 °C for 1 h. 3.3. Water Contact Angle and Surface Chemistry Static contact angles with water ( θ w ) were measured on polymer films covalently grafted to the glass slide and compared with those of the bare glass and the respective homopolymers. Data are collected in Table 1 . Copolymers p(F6- co -PTMSix) showed values of θ w higher than 90° similar to that of the respective homopolymer p(F6) and consistent with the hydrophobic nature of fluorinated (co)polymers [ 33 , 34 , 35 ]. On the other hand, both MSA and MPC copolymers displayed lower θ , even though significantly higher than those of the corresponding homopolymers p(MSA) and p(MPC). This result was unexpected on account of the high content of zwitterionic counits (95 mol%) in p(MPC- co -PTMSi5) and p(MSA- co -PTMSi5). A moderate hydrophilic nature was previously observed for poly(MPC- co - n -butyl methacrylate) and poly(MPC- co −2-methacryloyloxy-4-azidobenzoate) films with θ w values of approximately 60° and 50°, respectively [ 36 ]. However, most of the studies in literature showed that zwitterionic copolymers are markedly hydrophilic materials ( θ w = 15°–30°) [ 37 ]. The comparatively high hydrophobicity of the zwitterionic copolymers of this work is possibly due to the surface segregation of low surface energy siloxane groups derived from the condensation between trialkoxysilyl moieties during the covalent grafting to the glass substrate. To gain a better comprehension of the film surface composition on account of the findings from biological assays with C. albicans (see below, Section 3.4 ), angle-resolved X-ray photoelectron spectroscopy (AR-XPS) measurements were performed at photoemission angles ϕ of 70° and 20° on the films, before and after water immersion for 24 h on the films of copolymers p(MPC- co -PTMSi5) ( Figure 6 ) and p(F6- co -PTMSix) ( Figure 7 ). The former was also taken as illustration examples of the zwitterionic polymer films for determination of their surface charge by ζ potential measurements at different pH values. Quantitative XPS data are reported in Table 2 . In p(MPC- co -PTMSi5) film, the presence of P (P(2p) and P(2s)) and N (N(1s)) moieties at the surface suggests that the glass slides were effectively covalently functionalized with the copolymer ( Figure 6 ). Their atomic percentages were slightly lower than the theoretical ones and did not change significantly by varying the photoemission angle, indicating that the concentration of the phosphorylcholine side chain was essentially constant along the investigated sampling depth. On the other hand, Si (Si(2p) and Si(2s)) was detected with a content that, although relatively low in absolute value, was significantly higher than the theoretical one and decreased by increasing the sampling depth. These results indicate that siloxane groups were concentrated at the outermost surface layers, being the lowest surface energy components in the copolymer system. These findings are consistent with the contact angle analysis and suggest the hypothesis of siloxane bond formation as a result of the condensation of methoxysilyl/silanol groups of PTMSi. Upon immersion in water for up to one week, the surface of p(MPC- co -PTMSi5) film still presented Si moieties at the surface and overall the chemical composition after one week of immersion did not differ significantly from that before immersion. This is in contrast to what is normally expected of amphiphilic surfaces in which reconstruction of the outer surface occurs to respond to the changed aqueous outer environment [ 38 ]. We speculate that the interchain siloxane groups once formed were stable and did not rearrange upon contact with water. AR-XPS analysis of p(F6- co -PTMSix) films revealed that their surfaces were highly enriched in F (F(1s)) moieties, with a percentage similar for both p(F6- co -PTMSi1) and p(F6- co -PTMSi5) and higher than the theoretical one, especially at ϕ = 70° ( Table 2 ). Moreover, the F percentage decreased with increasing the sampling depth, while C and O percentages followed the opposite trend. Thus, the fluorinated chains were selectively located at the outermost surface layers. For these samples, the amount of Si at the surface was very low, consistent with the theoretical amount in the copolymer. Accordingly, the siloxane groups were segregated in the bulk of the film. Similar remarks are also valid for the films after immersion in water, which presented a slightly increase in Si and a decrease in F concentrations. The F moieties, however, remained the predominant components ( Figure 7 a). The presence of the resolved peak at ~292 eV in the C(1s) spectrum ( Figure 7 b) due to the CF 2 and CF 3 groups of F6 confirmed that the fluorinated side chains populated the film surface after immersion in water. As a result, no variation in water contact angle after 24 h of immersion was detected ( θ w = 115° ± 2 and θ w = 116° ± 1 for p(F6- co -PTMSi5) and p(F6- co -PTMSi1), respectively). The values of electrokinetic ζ potential of coated glass slides with copolymer p(MPC- co -PTMSi5) exhibited an ascending trend with decreasing pH and comprised a broad negative plateau (–55 ± 5 mV, pH > 6.5), an isoelectric point (pH = 3.7) and a narrow positive plateau (~20 ± 1 mV, pH < 3). Bare glass slides were negatively charged over the whole pH investigated range (plateau value of –75 ± 2 mV, pH > 4.5) and an isoelectric point at pH = 2.6. Thus, functionalization of the glass surface by MPC was further confirmed, including accumulation of phosphate anion groups at the polymer–water interface. Such values of plateau of ζ potential are consistent with those reported for different phosphorylcholine-modified [ 39 , 40 ] and sulfobetaine-modified [ 41 , 42 ] surfaces. The coated surfaces appeared not to readjust under the adopted experimental conditions, as inferred by the reproducibility on repeated measurements, nor was detachment of polymer film from the glass surface observed in any case. These findings could ensure stability of the tested films during the subsequent biological assays. 3.4. Biological Assays with C. albicans Results on the ability of C. albicans to adhere to, and form biofilm on, the films of zwitterionic and fluorinated copolymers are summarized in Figure 8 , which shows the numbers of sessile cells adhered (CFU cm –2 , surface area 1.8 cm × 1.8 cm) to each slide, following incubation for 24 h to a cell concentration of approximately 2 × 10 7 cells mL –1 for 24 h. Glass surface was also tested as a negative standard, which is known to inhibit the adhesion of C. albicans . The results demonstrate a marked reduction in C. albicans adhesion on fluorinated surfaces with respect to zwitterionic ones, with the fluorinated p(F6- co -PTMSi1) copolymer, showing comparable antifouling performance to that of the negative standard. Inverted microscope analyses of different surfaces revealed the presence of the filamentous forms associated with biofilm formation, with a higher prevalence of hyphal forms on zwitterionic polymers ( Figure 9 ). Since the high variability observed between zwitterionic and fluorinated polymers prevented to assess potential differences in fungal adhesion on fluorinated surfaces with different chemical composition, a second set of experiments of colonization on p(F6- co -PTMSix) films was performed to specifically address this point ( Figure 10 ). Oscillations were noticed in the CFU cm –2 values, as was expected of the variability of adhesion tests carried out with C. albicans at different times. The results confirmed that adhesion of cells on p(F6- co -PTMSi1) was significantly lower than on p(F6- co -PTMSi5) (1.0 × 10 3 CFU cm –2 vs. 2.3 × 10 4 CFU cm –2 ), and consistently a higher number of fluorinated units in the copolymer better reduced the adhesion of Candida ."
} | 5,735 |
38236040 | PMC10868220 | pmc | 2,869 | {
"abstract": "ABSTRACT Here we report the complete genome sequence of two moderately thermophilic methanotrophs isolated from a landfill methane biofilter, Methylococcus capsulatus (Norfolk) and Methylocaldum szegediense (Norfolk)."
} | 55 |
39585927 | PMC11649087 | pmc | 2,870 | {
"abstract": "Despite almost a century of research on energetics in biological systems, we still cannot explain energy regulation in social groups, like ant colonies. How do individuals regulate their collective activity without a centralized control system? What is the role of social interactions in distributing the workload amongst group members? And how does the group save energy by avoiding being constantly active? We offer new insight into these questions by studying an intuitive compartmental model, calibrated with and compared to data on ant colonies. The model describes a previously unexplored balance between positive and negative social feedback driven by individual activity: when activity levels are low, the presence of active individuals stimulates inactive individuals to start working; when activity levels are high, however, active individuals inhibit each other, effectively capping the proportion of active individuals at any one time. Through the analysis of the system’s stability, we demonstrate that this balance results in energetic spending at the group level growing proportionally slower than the group size. Our finding is reminiscent of Kleiber’s law of metabolic scaling in unitary organisms and highlights the critical role of social interactions in driving the collective energetic efficiency of group-living organisms.",
"introduction": "Introduction As a complex social system—such as an ant colony or a human organization—increases in size, so does the need to better regulate the activities of its members [ 1 ]. Without regulation, individuals would randomly distribute their effort across time, wasting energy if too many individuals were active when the global workload does not require it, or wasting opportunities if too few individuals were to respond to increasing needs in the population. The question of activity—and hence energy—regulation is not unique to complex social systems; it is also increasingly studied in the context of artificial distributed systems such as fleets of robotic devices [ 2 , 3 ] and Internet-of-Things networks [ 4 ], where redundancies are frequent and energy-wasting. Both in natural and artificial distributed systems, the control of activities is often fully decentralized, making it impossible to consider global solutions for optimizing energy consumption. Studies on social insect colonies have shown that living systems can be surprisingly efficient in managing their collective activities [ 5 ]. In particular, their energy use per unit of mass appears to scale hypometrically with their colony size [ 6 , 7 ]. In other words, larger colonies are more energy-efficient relative to their size than smaller ones, a phenomenon akin to Kleiber’s law [ 8 ] that states that the rate of energy use by a biological system scales hypometrically with its size. Understanding how ant colonies—a fully decentralized system—regulate their activities to achieve energy efficiency could, therefore, have important repercussions for the design of human organizations and artificial distributed systems. In the literature, activity regulation in social insects is typically divided between activation mechanisms that boost the number of individuals engaged in work, and inactivation mechanisms that reduce it [ 9 , 10 ]. On the activation side, two types of mechanisms are usually invoked. First, individuals may sense workload-associated stimuli and, if their intensity exceeds a certain value (called the “response threshold”), start performing work to reduce this stimulation [ 11 ]. Such response thresholds have been found in several social insect species, for instance, for triggering a defense reaction to threats [ 12 ] or a foraging response to the presence of nutrients [ 13 , 14 ]. Second, individuals already engaged in work can stimulate others to work, for instance, when the workload has outgrown their capacity [ 15 ]. This is frequently observed in social insects in the context of foraging. For example, ant workers that have found a resource will use a combination of chemical and tactile signals to stimulate other workers to join them in exploiting it [ 16 ]. Likewise, honeybee scouts that have found a new food patch execute a stereotypical “waggle dance” back at their colony that stimulates other bees to leave the nest and encodes the direction and distance to the resource [ 17 ]. Such recruitment processes are akin to a form of “social contagion”, whereby the state of being active spreads in the group through local social interactions. As a result of this positive feedback loop, social organisms can mobilize a large portion of the available workforce quickly, facilitating rapid monopolization of resources [ 18 ] or overwhelming attackers by swiftly assembling defense forces around them [ 19 ]. While activation mechanisms are well studied and supported by documented examples, inactivation mechanisms are overlooked—especially those mediated by social interactions [ 20 ]. In existing models of activity regulation, inactivation is often treated as an intrinsic property of the individual rather than a socially driven one. For instance, inactivation has been modeled as a workload-associated threshold [ 21 ], a limit on the time an individual can be active [ 22 , 23 ], or a constant probability per unit time [ 9 , 11 ]. In each of these cases, the social environment does not influence the duration of the activity of an individual, or, at best, does it very indirectly (for example, through the impact other individuals have on the quantity of work remaining to be done). Yet, socially driven inactivation mechanisms have been shown to play an important role in counterbalancing activation mechanisms in social insects. For instance, honeybees use inhibitory signals to slow down the recruitment of foragers when food storing at the nest cannot match the influx of new nectar and pollen [ 24 ] and to delay the maturation of hive workers into foragers when the forager population is already high [ 25 ]. In ants, crowding along a foraging trail can inhibit the deposition of trail pheromone, reducing the risk of traffic jam [ 26 ]. Repellent pheromone can also be used to discourage foragers from visiting unrewarding routes [ 27 ], for instance because the resource they lead to is now depleted or overcrowded. In a recent work [ 28 ], we investigated the impact of a socially driven inactivation mechanism on the scaling of energy use in eusocial organisms. We proposed an explanation for Kleiber’s law [ 8 ] in the context of colonies of harvester ants ( Pogonomyrmex californicus ) using data from Waters et al. [ 29 ]. Our explanation was based on scaling arguments adapted from urban science [ 30 ] and relied on a key biological phenomenon that we called “reverse social contagion”—a mechanism by which an individual engaged in a given behavior becomes more likely to interrupt this behavior as it interacts with more neighbors also engaged in the same behavior [ 31 ], see Fig 1a . Reverse social contagion can be observed, for instance, in the form of stop-signaling mechanisms [ 32 ] and blocking interactions [ 33 ], and has been proposed as a factor regulating collective decision-making [ 34 , 35 ] and energy spending in social systems [ 28 , 36 ]. 10.1371/journal.pcbi.1012623.g001 Fig 1 (a) Illustration of the concepts of social contagion (top) and reverse social contagion (bottom). (top) An inactive ant interacts with an ant engaged in a foraging task: through social contagion (for example, caused by active recruitment), it also begins foraging. (bottom) Two ants engaged in foraging interact: through reverse social contagion (for example, caused by steric exclusion), one of them ceases their activity to become inactive. Image courtesy of Isabella Muratore, reprinted from Porfiri et al. [ 28 ]. (b) State transition diagram for three classes of individuals (A, active, I, inactive, and R, refractory). The diagram describes two competing social feedback mechanisms (social contagion and reverse social contagion), along with a spontaneous transition from refractory to inactive state (that is, the completion of rest). Social contagion and completion of rest are well established in the literature since the seminal work by Goss and Deneubourg [ 9 ] (therein, inactive ants are called “activable inactives” and refractory ants “inactives” to specify that only the former ones can be activated). Reverse social contagion has been, instead, overlooked so far, thereby hindering our understanding of energy regulation in social groups. In that study, we focused on reverse social contagion in the context of movement, where an individual would cease to move (inactivate) in response to a social environment where many neighbors are also moving (being active). Each colony was composed of N individuals, of which A were active. The individuals in the colony interacted through a network with E ∝ N 3/2 links, as estimated from the data. To explain activity regulation in the form of a reduction in the fraction of active individuals as the colony grows, we proposed a balance between reverse social contagion and spontaneous social activation. The former should scale with A 2 / N 1/2 , given that the number of interactions of active individuals with other active individuals is estimated as the product between the total count of interactions in the colony, E , and the probability that two individuals are simultaneously active, ( A / N ) 2 . The latter, instead, should scale with N , being an inherent property of the individuals. Balancing reverse social contagion and spontaneous social activation, one predicts a hypometric scaling of the colony’s activity with respect to its size, A ∝ N 3/4 , akin to Kleiber’s law [ 8 ]. In this article, we draw inspiration from this simple scaling argument to formulate a dynamic model for activity regulation in eusocial systems. The state-of-the-art modeling of the dynamics of eusocial systems has, so far, largely focused on socially driven positive feedback mechanisms to predict rhythmic activation patterns [ 22 , 37 , 38 ], ignoring the impact of social information on inactivation patterns. Here, instead, we explicitly include socially driven negative feedback through reverse social contagion. We present a detailed study of the model, in the form of a system of coupled differential equations and stochastic Monte Carlo simulations.",
"discussion": "Discussion We showed that efficient activity regulation can emerge in highly integrated social groups from a balance between two socially driven feedback processes. First, an autocatalytic process of social contagion results in the activation of inactive individuals, after stimulation by already active individuals. This positive feedback is then counterbalanced by an “autoinhibitory” process that we call reverse social contagion where active individuals are increasingly likely to stop working as they interact with more active individuals. As a consequence, the number of active individuals in the group will be naturally capped at a proportion of the group size, whose value depends on the balance between the two aforementioned processes. This is compatible with observations of “lazy” individuals with very low levels of activity and commonly observed in large colonies of eusocial insects [ 54 ]. We also demonstrated that, when the number of interactions between the members of the group grows faster than the size of the group, the proportion of active individuals grows hypometrically with the group size. Hence, the group becomes energetically more efficient as its size increases. Both network and energy scalings are commonly observed in group-living organisms [ 6 , 7 , 41 , 47 ], supporting the hypothesis that social interactions are critical in explaining the energetic properties of social groups (including human beings). Notably, this scaling does not occur when the refractory state is removed from the model, suggesting that the refractory state serving as a reservoir is critically important to recovering the relationship between activation and colony size observed in experimental data. Finally, the reverse social contagion mechanism that we study here is fundamentally different from the inactivation mechanisms proposed in existing models of activity dynamics. Indeed, the latter are typically not socially driven, and they rely on constant duration of activity [ 22 ], constant rate of inactivation [ 9 , 11 ], or the ability of an individual to estimate the amount of remaining work [ 21 ]. In the first two cases, activity regulation is entirely dependent on the activation process and the hypometric energy scaling disappears. In the last case, the inactivation process can depend on the social environment as in our model, albeit indirectly via stigmergic communication since the amount of remaining work is impacted by the actions of the active individuals. This scenario is already included in our model since it does not require specifying the mode of interaction (direct or indirect) between active individuals but simply its rate. In conclusion, our study provides a generalizable explanation for activity regulation in social groups. Its predictions are in agreement with qualitative and quantitative observations of general activity patterns and energetic scaling found in highly integrated social species, such as ant colonies. Our next step is to incorporate multiple activity types and mobility in the model, towards a framework for the study of more complex and specific patterns of emergent division of labor and their impact on energy use in animal and human groups."
} | 3,409 |
27782372 | PMC5132096 | pmc | 2,871 | {
"abstract": "Abstract Microalgae have long been considered as one of most promising feedstocks with better characteristics for biofuels production over conventional energy crops. There have been a wide range of estimations on the feasibility of microalgal biofuels based on various productivity assumptions and data from different scales. The theoretical maximum algal biofuel productivity, however, can be calculated by the amount of solar irradiance and photosynthetic efficiency (PE), assuming other conditions are within the optimal range. Using the actual surface solar irradiance data around the world and PE of algal culture systems, maximum algal biomass and biofuel productivities were calculated, and feasibility of algal biofuel were assessed with the estimation. The results revealed that biofuel production would not easily meet the economic break‐even point and may not be sustainable at a large‐scale with the current algal biotechnology. Substantial reductions in the production cost, improvements in lipid productivity, recycling of resources, and utilization of non‐conventional resources will be necessary for feasible mass production of algal biofuel. Among the emerging technologies, cultivation of microalgae in the ocean shows great potentials to meet the resource requirements and economic feasibility in algal biofuel production by utilizing various marine resources.",
"conclusion": "4 Concluding remarks The global maximum microalgal biofuel productivity (when other substrates are sufficient and conditions are within the optimal growth range) is estimated at 3.2–14.8 TOE ha −1 y −1 with a global average of 8.4 TOE ha −1 y −1 when open ponds are used as the culture system and the lipid content of microalgae is 25%. Enhancements in the PE (from 3.1% PAR to 10.3% PAR) and lipid content (from 25 to 50%) could lead to increases of 233% and 33% in maximum biofuel productivity, respectively. Economic assessment showed that CAPEX and OPEX of an algal biofuel production facility should be considerably reduced while maintaining or increasing biofuel productivity. Production of co‐products and convert/recycle every molecule in the biomass can also significantly improve the economy of algal biofuels, but since they would be produced in vast quantities, large markets need to be secured beforehand. For example, the bioplastic market is rapidly growing and estimated to reach 10 billion USD by 2020, while growth of the global fish feed market is projected to 123 billion USD by 2019. Demand for protein feeds for livestock is also estimated to be over 200 Mt y −1 . Offshore algal culture systems that do not require extensive land construction, are built with low‐cost materials, use nutrients in wastewater or seawater, and utilize ocean waves for culture mixing can also be an option to substantially reduce CAPEX and OPEX in algal biofuel production. Furthermore, cultivation of microalgae in areas with high annual solar irradiance (e.g. Southeast Pacific Ocean) would help ensure high microalgal biofuel productivity. Global‐scale algal cultivation for biofuel production was not sustainable using conventional algal culture technologies. Indeed, enormous quantities of freshwater, nitrogen, phosphorus, and potassium comparable to the current global consumption would be needed to produce enough algal biofuel to meet 30% of the transportation fuel demand. Algal production will likely conflict with food production for resources. Cultivation of marine microalgae in seawater could easily alleviate the freshwater demand. Use of non‐fertilizer nutrients, such as dissolved inorganic nutrients in seawater and wastewater, and reclaiming nutrients in algal biomass by a process like HTL would be necessary for sustainable production of algal biofuel. Microalgae hold great potential for biofuel production, but mass production of algal fuel will require enormous amounts of resources. We showed that harnessing abundant marine resources could be a way to sustainably meet resource requirements. As offshore microalgal cultivation is relatively a new technology, there are challenges and potential problems that need to be investigated and solved, including fouling, transportation of products, detailed financial analysis, ecological impacts, political issues, etc. Nevertheless, with extensive research and development (R&D) efforts, ocean‐based algal culture systems would provide another option for sustainable and economic production of microalgae in the future.",
"introduction": "1 Introduction Various renewable energy sources have been harnessed for energy production, but many of these produce electricity, which cannot be efficiently stored. Moreover, the volumetric and gravimetric energy density of the most‐advanced batteries are much lower than those of liquid fuels 1 . Biofuels have long been considered potent, renewable alternatives to petroleum‐based fuels because they are carbon neutral and can be utilized by existing infrastructure and internal combustion engines with little or no modifications 1 , 2 . Microalgae have recently received a great deal of attention because of their superior characteristics as feedstocks for biofuel production. Specifically, they have high photosynthetic efficiency (PE), ability to accumulate energy storage biochemicals, broad range of growth conditions, can be harvested frequently (e.g. every two weeks), do not require arable land, and are not major food sources 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 . Despite the excellent characteristics of microalgae as feedstock for biofuels, there are still many obstacles to large‐scale commercial production of microalgal biofuel. Although a wide range of prices for algal biofuel have been reported (from 1.3–2.4 USD L −1 \n 11 , 12 , 13 , 14 to 20 USD L −1 \n 13 ), it is clear that algal biofuel is not cost competitive with petroleum‐based fuels or other biofuels, yet. Many studies have estimated the maximum productivity of algal biofuels since microalgae have gained attention. Microalgal biofuel productivity ranges from 5.6 TOE ha −1 y −1 to 106.8 TOE ha −1 y −1 \n 4 , 9 , 13 , 15 , 16 , 17 , 18 , 19 , 20 , 21 . In some cases, the data from lab‐scale experiments are used to estimate the productivity of large‐scale production, and arbitrary values for biomass productivity and lipid content are assumed in some cases. Moreover, the diversity of microalgae contributes to the highly variable estimations of algal biofuel productivity 4 , 6 , 9 , 15 , 16 . In most photoautotrophic algal cultures, the amount of light energy provided is one of the major factors determining microalgal biomass productivity. Some algal cultures use artificial lights for constant illumination or supplying light during the night to enable higher productivity 22 . Use of artificial lighting may be feasible for production of highly‐valued compounds, such as pharmaceuticals and antioxidants. However, LEDs, which are one of the most energy‐efficient lights, have up to 48% wall plug efficiency 23 , while the theoretical maximum value of PE is 25.9% of photosynthetically active radiation (PAR) 24 . Therefore, consuming electricity to cultivate microalgae for biofuels has a very low energy conversion efficiency and is economically infeasible 25 . Consequently, sunlight should be used as the only energy source for algal cultures for biofuel production. In such cases, the maximum productivity of algal biofuels is ultimately limited by the amount of solar irradiance and PE of the culture system. Several reports estimated microalgal biofuel productivity using solar irradiance information 9 , 19 , 26 , but they only analyzed a few locations or considered only land area when determining biofuel potentials. Land‐based open raceway ponds (ORPs) and closed photobioreactors (PBRs) have been the major technologies employed for algal cultivation, but development of water‐based algal culture systems has been reported recently, claiming various advantages over conventional land‐based systems 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 . Thus, it would be worth including the vast ocean area in analysis of maximum microalgal biofuel productivity. In this study, the maximum microalgal biofuel productivity was calculated based on surface solar radiation, limits for capital and operation costs for economic feasibility of algal biofuel production were determined, and sustainability of global scale microalgal biofuel production was assessed. Various methods to improve economic feasibility and sustainability were then discussed.",
"discussion": "3 Results and discussion 3.1 Maximum microalgal biofuel productivity based on surface solar irradiation The maximum biofuel productivity ranged from 3.2 TOE ha −1 y −1 to 14.8 TOE ha −1 y −1 with an average of 8.4 TOE ha −1 y −1 for case 1 (Table 1 ). The corresponding maximum biomass productivities were 14.9–68.8 t ha −1 y −1 with 38.9 t ha −1 y −1 as the global average (Table 1 ). The world maximum biofuel productivity map indicates that regions at lower latitude generally have higher maximum biofuel productivity, (Fig. 1 ). The tendency is more apparent when in the results based on the four cases are plotted against the latitude (Fig. 2 ). However, as mentioned above, the equatorial region did not have the highest biofuel productivity because of the Intertropical Convergence Zone, in which the weather is frequently cloudy and rainy all over the year. Nevertheless, the tropical zone had the highest overall biofuel productivity because of the substantially higher solar irradiance in the region (Fig. 2 ). Similar effects of climate conditions on the productivity were found at the sub polar zones, near 70°; however, unlike the tropical zone, biofuel productivity was lower in these areas than in the neighboring locations. The productivities for case 2, 3, and 4 were 133, 333, and 444% of those of case 1, respectively (Table 1 ). The lipid content increased from 25% (cases 1 and 3) to 50% (cases 2 and 4), resulting in a 33% increase in maximum biofuel productivity (compare cases 1 vs. 2 and 3 vs. 4) and higher PE values in cases 3 and 4 led to 3.3‐fold higher productivity compared to the lower PE values in cases 1 and 2 (Table 1 ). Table 1 also shows that the improvement in the productivity in response to the enhanced PE and lipid content (344%) was comparable to that observed in response to the maximum difference by location (362%). These results indicate that algal biofuel productivity can be substantially enhanced by improving algal culture technology for higher PE and lipid content. Figure 2 Maximum microalgal biofuel productivities by latitude based on 22‐year average of surface solar irradiance. WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim The maximum algal biofuel productivities from each case were compared with the values from the literature (Fig. 3 ). The estimations and results of other studies were within the range of predictions from this study, except for one case. An exceptionally high algal biofuel yield prediction of up to 106 TOE ha −1 y −1 was reported by Chisti 10 . A very high lipid content, 70%, was assumed, which is difficult to achieve in outdoor cultivation 40 , 41 , 42 , and temporal and spatial conditions were not taken into account in the estimation. Case 4 is based on the maximum PE and lipid content, reflecting the upper limit of algal biofuel productivity using sunlight. Therefore, the high productivities suggested would be difficult to achieve without breakthroughs in genetic modifications to enhance the PE of microalgae or PBR engineering. Indeed, the data from actual outdoor cultivation by Rodolfi et al. and Feng et al. were close to the estimations from case 3 16 , 20 , showing that the current maximum productivity has not yet been achieved. Figure 3 Comparison of microalgal biofuel productivities. The bars indicate maximum values if both average and maximum values were available. WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim The global map of maximum algal biofuel productivities for case 1 also shows variations by longitude (Fig. 1 ). For example, the west coast of the North Americas has higher maximum algal biofuel productivity than the east coast because of climate differences. Interestingly, a large area of China shows lower biofuel potentials than the adjacent locations, appearing as a blue island. Many Chinese metropolitan cities suffer from extensive atmospheric pollution, which considerably reduces the amount of solar radiation reaching the surface 48 . Although the highest biofuel productivity was obtained from the mountains of Chile, the Pacific, Atlantic, and Indian Oceans offer very large areas with high maximum biofuel productivities (Fig. 1 ). Cultivation of microalgae in such regions using water‐based algal culture systems could be an attractive alternative to land‐based culture systems. Particularly, for countries without large areas of land with high solar radiation (e.g. Korea and the United Kingdom), deploying algal cultures in their exclusive economic zone (EEZ) or creating a joint offshore algae farm in international waters with high solar irradiance would be an attractive option for algal biofuel production. While other regions can be affected by tropical storms such as typhoons, hurricanes, and cyclones, the Southeast Pacific Ocean and South Atlantic Ocean are free from such storms and would serve as great locations for microalgal biofuel production. 3.2 Economic feasibility of microalgal biofuels by culture systems and productivity The limits for CAPEX and OPEX were estimated based on the global average maximum algal biofuel productivities using Eq. (3) (Fig. 4 ). The periods for return of investment (ROI) were assumed to be five or 20 years. As the total production cost cap was determined by the set price of biofuel and productivity, CAPEX and OPEX were inversely correlated to each other, refer to Eq. (3) . While increases in productivity elevated the limits for both CAPEX and OPEX, ROI period only affected CAPEX. These results indicate that CAPEX and OPEX must be below one million USD ha −1 and 50 000 USD ha −1 y −1 , respectively, to achieve an algal biofuel price of 1 USD L −1 under the maximum productivity scenario (case 4) (Fig. 4 B). For ORPs, approximately 300 000 USD ha −1 and 15 000 USD ha −1 y −1 were the maximum CAPEX and OPEX, respectively (Fig. 4 B). Three predicted values of capital and operation costs for ORP and a hybrid of ORPs and PBRs are plotted in Fig. 4 \n 12 , 17 , 21 . For a 400‐ha algal biofuel plant using an open raceway pond, CAPEX and OPEX were estimated to be at least 250 000 USD ha −1 and 20 000 USD ha −1 y −1 (square in Fig. 4 ) 12 . For a hybrid system, CAPEX and OPEX were predicted to be 272 482 USD ha −1 and 15 270 USD ha −1 y −1 (triangle in Fig. 4 ) 17 , while for another hybrid system they were 228 000 USD ha −1 and 19 900 USD ha −1 y −1 (circle in Fig. 4 ), respectively 21 . With a five‐year ROI, none of the culture systems were economically feasible (Fig. 4 A), but when the ROI was increased to 20 years, they could be profitable with adequate algal biofuel productivity, roughly 23 TOE ha −1 y −1 (Fig. 4 B). Moreover, the capital cost of PBRs was estimated to be substantially greater, at 940 000 USD ha −1 \n 17 \n . Another study also reported 906 255 USD ha −1 as the capital cost for PBRs 14 . In such cases, an exceptionally high productivity (case 4) and low annual cost would be required to generate profit by producing only biofuel, but the estimated OPEX for PBRs was 216 232 USD ha −1 y −1 \n 14 , which was beyond the limit for operation cost of 50 000 USD ha −1 y −1 , even with the productivity of case 4. Without generating extra revenue by selling other products, the facility will not reach the break‐even point. In contrast to the estimated values, an actual facility of Cyanotech in Hawaii required about 460 000 USD ha −1 for site preparation for of the raceway pond alone 49 . The characteristics of the site for Cyanotech's facility, which was covered by volcanic rocks, added an extra 34 398 USD ha −1 for land clearing; nevertheless, it is still notable how expensive land construction for an algal culture systems can be. Figure 4 CAPEX and OPEX limits to produce algal biofuel at 1 USD L ‐1 based on microalgal biofuel productivity. A: ROI = 5 years, B: ROI = 20 years. WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim As the results indicate, substantial reductions in CAPEX and OPEX while maintaining or improving biofuel productivity are needed to deliver economic feasibility. The algal culture systems account for 53% to 83% of the capital cost 8 , 12 , 21 , and as seen in the case of Cyanotech, the cost for land construction also has a significant impact on the CAPEX. Thus, low‐cost algal culture systems that do not require extensive construction would be needed. Labor, electricity, and nutrient supply are generally the major contributors to OPEX 3 , 6 , 8 , 12 . In particular, decreasing the cost for nutrient supply can contribute to considerable reductions (>50%) in total cost in algal biomass and biofuel production 3 , 6 . Therefore, recovering and reusing nutrients in algal biomass or utilizing non‐fertilizer nutrients would be essential to producing microalgal biofuel at a competitive price. Use of wastewater in algal cultivation has been very popular recently as freshwater and nutrients can be supplied at the same time and credits for wastewater treatment could be granted 4 , 5 , 6 , 12 , 15 , 18 , 29 , 30 , 31 . In addition, production of by‐products, such as protein for animal feed, char for biofertilizers, and carbohydrates for fermentation, could help improve the overall economy of the algal biofuel production 50 . Revenues made by other co‐products allow increased target oil price. Doubling the target algal oil price also doubles the limits for CAPEX and OPEX, refer Eq. (3) . Therefore, even algal biofuel production facilities with high operating cost can achieve economic feasibility if additional revenue can be generated by other means. Culturing microalgae in the ocean could be a way to alleviate the CAPEX and OPEX in algal biofuel production. In contrast to land‐based culture systems, offshore algal culture systems do not require extensive land constructions, purchase of land, or expensive durable materials for construction because seawater supports the system. Moreover, seawater can be supplied on‐site, eliminating the need for drilling water wells or installing long pipelines. Wastewater and flue gas can also be used in ocean‐based algal culture systems especially when they are located near the coast for supply of CO 2 and other nutrients while removing pollutants as well 29 , 30 , 31 , 51 . CO 2 can also be supplied in the form of sodium bicarbonate salt or concentrated solution 52 , 53 . For offshore microalgal cultivation far away from the coast, nutrients dissolved in seawater can be utilized by using technology such as semi‐permeable membrane PBRs 27 , 35 . In such case of relying on dissolved nutrients, obtaining high nitrogen and phosphorus supply rate will be important as CO 2 is relatively abundant in comparison to the others 27 . Culture mixing by harnessing the energy from ocean waves instead of the paddle wheels traditionally used in pond systems is also a potential advantage of use of ocean‐based algal culture that leads to decreased power consumption, thereby lowering OPEX 28 . If the aforementioned advantages of ocean‐based culture systems are effectively delivered, significant cost reductions would be possible. Offshore cultivation of microalgae also brings challenges not present in land‐based algal cultures. For instance, fouling is a prevailing phenomenon in marine environment that could negatively affect algal biomass productivity and needs to be dealt with 33 . In case of culturing marine microalgae in seawater, salts in the biomass could increase CAPEX and OPEX in the downstream processes. While salts in marine microalgae did not affect transesterification reaction 54 , they might cause solid deposition and corrosion in the reactors 55 . On the contrary, using marine strain can bring advantages in the process as introduction of salt in hydrothermal liquefaction (HTL) yielded higher composition of hydrocarbon in bio‐oil 56 , and hydrothermal microwave processing showed better performance for marine microalgae than freshwater strains 57 . Costs for harvesting and transportation of produced biomass are also of concerns. Development of in situ harvest methods, such as floatation, for concentration and dewatering of algal biomass could reduce the cost for transportation 58 . At a full‐scale, construction of an offshore platform adjacent to an offshore algal culture facility would be a more economic option, so that final products, algal biofuel and other byproducts, could be transported as how fossil fuels from offshore platforms are extracted and transported. As land‐based open ponds and PBRs have been thoroughly studied and developed for decades, offshore culture systems would need to be extensively tested and carefully developed to become a viable option in large‐scale algal biofuel production. 3.3 Potentials of utilizing marine resources to improve sustainability of microalgal biofuels Resource requirements for global‐scale microalgal biofuel production were assessed for each productivity case (Table 2 ). For the basic scenario (case 1), 0.12 ha, 291 m 3 of freshwater, 2.4 t of carbon, 0.31 t of nitrogen, 0.07 t of phosphorus, and 0.04 t of potassium are needed for 1 TOE of microalgal biofuel. The area needed to replace 30% of annual global liquid transportation fuel (gasoline, diesel, and jet fuel) consumption ranged from 23 Mha to 100 Mha, accounting for 0.20–0.87% of the total non‐arable land area, depending on the areal productivity. Because of the inherent nature of the culture systems, PBRs require much less freshwater (37 km 3 y −1 ) than ORPs (244 km 3 y −1 ). These values correspond to 3.2 and 21% of total non‐agricultural freshwater consumption. Nutrient demands were lower when the lipid content was higher. When the lipid content was 50%, 997 Mt of carbon, 130 Mt of nitrogen, 29 Mt of phosphorus, and 18 Mt of potassium would be needed. The amounts of nutrients required for algal cultivation are doubled if the lipid content is 25%. When compared to the global consumption, 74–324% more nitrogen, phosphorus, and potassium must be produced to meet 30% of the global fuel demand. When the demand for carbon was compared to the world CO 2 emissions, 2.9–5.8% of CO 2 would be consumed for algal cultivation. Table 2 Resource requirements for production of microalgal biofuel to replace 30% of global transportation fuel consumption. Case 1 Case 2 Case 3 Case 4 Area (Mha) 100 75 30 23 % of non‐arable land 0.87 0.65 0.26 0.20 % of ocean area 0.27 0.21 0.08 0.06 Freshwater demand a) (km 3 · y ‐1 ) 244 244 37 37 % of non‐agricultural fresh water consumption 21 21 3.2 3.2 Freshwater demand b) (km 3 · y ‐1 ) 58 58 – – % of non‐agricultural fresh water consumption 5.0 5.0 – – Carbon demand (Mt · y ‐1 ) 1995 997 1995 997 % of CO 2 emissions 5.8 2.9 5.8 2.9 Nitrogen demand (Mt · y ‐1 ) 259 130 259 130 % of N consumption 235 118 235 118 % of ocean inventory 0.04 0.02 0.04 0.02 Phosphorus demand (Mt · y ‐1 ) 57 29 57 29 % of P consumption 324 162 324 162 % of ocean inventory 0.06 0.03 0.06 0.03 Potassium demand (Mt · y‐1) 37 18 37 18 % of K consumption 148 74 148 74 a) When freshwater is used for cultivation b) When seawater is used for cultivation WILEY‐VCH Verlag GmbH & Co. KGaA, Weinheim On the global scale, the area for algal cultivation is not a great concern. An area roughly equal to the land area of Egypt would suffice for cultivation of microalgae. However, evaluation of the freshwater and nutrient requirements indicates that algal biofuel may not be sustainable and would compete with agriculture for resources on a global scale. The main reason for the high water consumption in algal culture is the need to compensate for evaporation losses, especially in open culture systems 59 . Thus, the freshwater demand was substantially lower for cases with closed PBRs (fourth row in Table 2 ). Unlike traditional energy crops, many microalgae thrive in seawater. Therefore, if marine microalgae are cultivated using seawater for large‐scale open cultures, the evaporation loss can be replenished continuously from the sea, which would reduce the need for freshwater by 88% (fourth and sixth rows in Table 2 ) 59 . Nitrogen fertilizer can be produced from atmospheric nitrogen gas by chemical reactions. Therefore, if necessary, nitrogen fertilizer production can be expanded to meet the extra demand for algal biofuel production. However, the process consumes power and results in GHG emissions 60 ; therefore, synthetic nitrogen should be the last choice when selecting the source of nitrogen. Other fertilizers pose a much severe problem. Phosphorus and potassium are finite underground resources like crude oil, and phosphorus reserves are being rapidly depleted 61 . As shown in Table 2 , enormous amounts of phosphorus and potassium are required for biofuel production from microalgae. Use of these fertilizers for biofuels production will likely lead to food vs. fuel conflicts, which is contrary to the idea of using microalgae instead of conventional energy crops for biofuel production. Therefore, non‐fertilizer nutrients must be utilized for algal cultures and nutrients should be reclaimed. Linking municipal or livestock wastewaters to algal cultivation is an excellent option as discussed above. HTL has recently been receiving a great deal of attention because of its potential for higher energy balance and nutrients recycling. One of the products of the HTL process is an aqueous phase (AP) containing the nutrients assimilated into the algal biomass, which could be used for growing microalgae 57 , 62 , 63 . However, growth inhibitors may also be present in the AP, requiring heavy dilution prior to use as a nutritional supplement to the culture medium 57 . In other studies, 50–75% of nutrients could be supplied using the AP 62 , 63 , but further improvements are required to close the loop. Another approach to supply nutrition could be utilization of nutrients dissolved in seawater. The oceanic inventories of inorganic nitrogen and phosphorus in the ocean are approximately 660 and 93 billion tons, respectively 64 , 65 . Potassium is far more abundant than other nutrients in seawater, and carbon is continuously replenished by dissolution of atmospheric CO 2 . Even if the nutrients currently available in seawater are consumed for algal culture without replenishment or recycling, microalgal biofuel can be produced for several thousands of years. Using these vast amount of dissolved nutritional salts could alleviate the demand for extra fertilizer supply. However, the concentrations of nitrogen and phosphorus are insufficient to use seawater directly as a culture medium; therefore, methods to concentrate nutrients or microalgae, such as semi‐permeable membranes, would be required to achieve significant algal biomass productivity 27 , 35 . No biofuels can be sustainable if any parts of the biomass are to be wasted or any of the nutrients are to be continuously supplied. The reasons why we call the current economy as the fossil‐based economy or the petroleum‐based economy are (i) many commodities other than the energy are supplied from fossil resources and (ii) every single molecule in the crude oil is consumed. A sustainable bio‐based economy won't be realized until we find a way to close the loop by converting and/or recycling every atom in the biomass."
} | 6,975 |
35586069 | PMC9108730 | pmc | 2,872 | {
"abstract": "Summary Many naturally occurring bacteria lead a lifestyle of metabolic dependency for crucial resources. We do not understand what factors drive bacteria toward this lifestyle and how. Here, we systematically show the crucial role of horizontal gene transfer (HGT) in dependency evolution in bacteria. Across 835 bacterial species, we map gene gain-loss dynamics on a deep evolutionary tree and assess the impact of HGT and gene loss on metabolic networks. Our analyses suggest that HGT-enabled gene gains can affect which genes are later lost. HGT typically adds new catabolic routes to bacterial metabolic networks, leading to new metabolic interactions between bacteria. We also find that gaining new routes can promote the loss of ancestral routes (”coupled gains and losses”, CGLs). Phylogenetic patterns indicate that both dependencies—mediated by CGLs and those purely by gene loss—are equally likely. Our results highlight HGT as an important driver of metabolic dependency evolution in bacteria.",
"introduction": "Introduction Naturally occurring bacteria lead one of two metabolic lifestyles: autonomy or dependency ( Ochman and Moran, 2001 ; McCutcheon and Moran, 2007 , 2012 ; Luo et al., 2013 ). Although autonomy reflects complete self-sufficiency in converting nutrients to biomass, bacteria with dependencies often require crucial metabolites from others ( D’Onofrio et al., 2010 ; Davis, 1921 ; Morris et al., 2008 ; Suzuki et al., 1988 ; Watsuji et al., 2007 ). These metabolites are secreted by neighboring community members. Because such dependencies are common in bacterial communities, it is instructive to ask: what processes and factors affect their evolution; in other words, what drives the switch from metabolic autonomy to dependency? To answer this question, recent studies have put forth the Black Queen hypothesis, which states that dependencies evolve through adaptive gene loss ( Morris et al., 2012 ; Mas et al., 2016 ; Fullmer et al., 2015 ). Individuals lose costly, dispensable genes in leaky environments, trading autonomy for better growth (or fitness). As both experiments and models show, this is feasible—administering the loss of even a few specific biosynthetic genes in bacteria repeatedly leads to strong metabolic dependencies ( Pande and Kost, 2017 ; Pande et al., 2014 ; D’Souza and Kost, 2016 ; Goyal, 2018 ). This also explains how endosymbionts undergo severe genome reduction ( McCutcheon and Moran, 2012 ). These bacteria lack many biosynthetic pathways and instead depend on their hosts for the required biomass components. However, many extant free-living bacterial species are also metabolically dependent, despite the “free-living” label ( D’Souza et al., 2014 ; Monk et al., 2013 ; Goyal and Maslov, 2018 ; Wang et al., 2019 ; Enke et al., 2019 ). These species do not have merely reduced genomes, i.e., they do not differ from their ancestors only by gene losses, as expected under the Black Queen hypothesis ( Zelezniak et al., 2015 ; Garcia et al., 2013 ; Giovannoni et al., 2014 ). Over time, they have also gained many genes, primarily by horizontal gene transfer (HGT) and at times by de novo gene birth ( Pál et al., 2005 ; Vos et al., 2015 ; Szappanos et al., 2016 ; Press et al., 2016 ; Maslov et al., 2009 ). For these often-dependent bacteria, we ask: could gene gains also have contributed to which dependencies we observe today? Specifically, during dependency evolution, which genes will be gained, and do those gains affect and which genes will later be lost? Here we explore the role of horizontal gene transfer in the evolution of metabolic dependencies in bacteria. Specifically, we measure the potential of HGT to drive dependency evolution by affecting the likelihood of subsequent gene loss events. To do this, we reconstructed the evolutionary history of 835 phylogenetically diverse, nonendosymbiont bacterial species. By inferring ancestral gene content, we mapped gene gains and losses along a large, deep-branching phylogeny and assessed their impact on bacterial metabolic capabilities. Our analyses suggest that horizontally transferred genes can indeed affect which genes are later lost and which dependencies emerge as a result. We have two lines of evidence to support this. First, we find that gene gains add new catabolic routes to bacterial metabolic networks. These gained catabolic routes increase the chance of new metabolic interactions between bacteria, a prerequisite for dependency evolution. Next, we show how these new routes can promote the loss of preexisting ancestral routes (a process we call “coupled gains and losses,” CGLs). We find that phylogenetic patterns indicate that both processes—CGLs and pure gene loss—are equally likely to lead to dependencies. Collectively, these results highlight horizontal gene transfer as an important driver of metabolic dependency evolution in bacteria.",
"discussion": "Discussion To summarize, here we showed that horizontal gene transfer (HGT) can play a significant role in metabolic dependency evolution in bacteria. Specifically, if an alternate metabolic pathway (or “route”) is gained by HGT, it can promote the loss a preexisting, otherwise indispensable route. Such alternate routes often catabolize metabolic byproducts from coexisting bacteria, thus making bacteria dependent on them. Overall, this is a new mechanism for dependency evolution: coupled gains and losses (CGLs). Phylogenetic evidence suggests that CGLs have occurred much more frequently across bacterial evolutionary history than expected by chance ( Figure 4 A). Further, phylogenetic evidence also suggests that CGLs can often be adaptive, as gained pathways are often shorter and more energy-efficient when compared with preexisting pathways ( Figures 2 B and 2C). As a mechanism for metabolic dependency evolution, CGLs are contrasted with pure gene loss, also called the Black Queen hypothesis. We found that although in communities with low diversity, pure gene loss is the more likely cause of dependencies, in communities with high diversity, CGLs are more likely ( Figure 4 B). Our results thus enrich and supplement the Black Queen hypothesis, by explaining the role of prior gene gains on eventual gene loss. Note that both mechanisms assume that all metabolic intermediates in pathways can in principle be secreted by cells, and in turn all secreted byproducts may be imported by cells. Previous work has shown this to be a reasonable assumption, which can qualitatively predict metabolic dependencies (our focus), but has shown that it fails to quantitatively predict resulting growth rates, which is why we refrained from making such predictions ( Goyal and Maslov, 2018 ; Zelezniak et al., 2015 ; Dal Bello et al., 2021 ). The reason that this assumption works qualitatively, but not quantitatively, is that there always exist generic transporters (such as several ABC transporters) that can enable the intake and outflow of most metabolites from cells (exceptions include phosphates), but the specific rates at which they enable flow for different metabolites remains unknown. Contrasted with gene loss, which may only lead to dependencies, gene gain via HGT can result in a variety of outcomes, only one of which is evolving dependencies. Other outcomes include gaining a pathway (or part of a pathway) without losing an alternative pathway, resulting in metabolic redundancies (64% of HGT events), and even losing dependencies and becoming increasingly independent, from only being able to use byproducts to being able to use at least one nutrient for a biomass component (21% of HGT events). These two alternate outcomes are not mutually exclusive of each other but are exclusive of dependency evolution. Thus, metabolic network evolution by HGT takes genomes along richer and more varied evolutionary trajectories compared with gene loss. We believe that our approach can also aid in a more accurate classification of bacterial lifestyles. Conventionally, bacteria are classified as either free-living or symbiotic in biological databases. Although this classification suggests that free-living bacteria would often be independent (and symbiotic ones, dependent), these labels can be misleading. For instance, free-living bacteria are often metabolically dependent ( D’Souza et al., 2014 ). In our analyses, we wanted to avoid relying on such a binary classification. We acknowledged that the degree of dependency of a bacterial genome lies along a spectrum and measured it by inferring which key biomass components each genome could synthesize in various nutrient environments. In this way, our approach is more precise and ecologically relevant. The mechanism we proposed, CGLs, also makes the following prediction about experimental evolution: when co-evolved in a diverse community, bacteria are more likely to lose biosynthetic pathways that they have alternate pathways for; this is less likely when they are evolved alone. As a corollary, adding alternate pathways to bacteria will promote the loss of preexisting pathways. Crucially, pathways do not have to be completely lost. Our work suggests—but we did not quantitatively analyze—cases where only a part of a pathway may be lost when an externally available byproduct happens to be an intermediate in that pathway. Both predictions can be tested via laboratory evolution in a community context. Limitations of the study The framework we used here, combining phylogenetic analyses with metabolic network analyses, can also help quantify the relative contributions of drift and selection to the reduction of bacterial genomes. Progressive genome reduction is often termed “genome streamlining,” and a key question in bacterial genome evolution asks how parallel, or repeatable, streamlining events are. The logic is that more parallel events reflect selection being dominant in genome reduction. We can systematically study these questions within our framework. For instance, we can measure how often we detect the same dependencies evolve along a phylogenetic branch and quantify how similar the corresponding gene loss events are. Similar, or repeatable, gene loss events would be consistent, with selection playing a major role in streamlining: perhaps “weeding out” genes no longer required in certain environments. Dissimilar gene loss events, on the other hand, would suggest that drift dominates. Such analyses are outside the scope of this study and the subject of future work. Finally, our analyses focused on changes in metabolic network architecture, but dependency evolution can also occur via changes to gene regulatory networks. In experiments, we observe that both metabolic and regulatory changes are responsible for evolved dependencies ( Lercher and Pál, 2007 ; D’Souza and Kost, 2016 ; Shitut et al., 2019 ). However, we do not understand how to incorporate the effect of regulatory changes on bacterial phenotypes as well as we do the effect of metabolic changes. Future work in this direction can help better understand the role of regulation on metabolic dependency evolution."
} | 2,773 |
25719969 | null | s2 | 2,873 | {
"abstract": "Microorganisms can form biofilms, which are multicellular communities surrounded by a hydrated extracellular matrix of polymers. Central properties of the biofilm are governed by this extracellular matrix, which provides mechanical stability to the 3D biofilm structure, regulates the ability of the biofilm to adhere to surfaces, and determines the ability of the biofilm to adsorb gases, solutes, and foreign cells. Despite their critical relevance for understanding and eliminating of biofilms, the materials properties of the extracellular matrix are understudied. Here, we offer the reader a guide to current technologies that can be utilized to specifically assess the permeability and mechanical properties of the biofilm matrix and its interacting components. In particular, we highlight technological advances in instrumentation and interactions between multiple disciplines that have broadened the spectrum of methods available to conduct these studies. We review pioneering work that furthers our understanding of the material properties of biofilms."
} | 265 |
22710417 | null | s2 | 2,874 | {
"abstract": "Biofilms, or surface-attached communities of cells encapsulated in an extracellular matrix, represent a common lifestyle for many bacteria. Within a biofilm, bacterial cells often exhibit altered physiology, including enhanced resistance to antibiotics and other environmental stresses. Additionally, biofilms can play important roles in host-microbe interactions. Biofilms develop when bacteria transition from individual, planktonic cells to form complex, multi-cellular communities. In the laboratory, biofilms are studied by assessing the development of specific biofilm phenotypes. A common biofilm phenotype involves the formation of wrinkled or rugose bacterial colonies on solid agar media. Wrinkled colony formation provides a particularly simple and useful means to identify and characterize bacterial strains exhibiting altered biofilm phenotypes, and to investigate environmental conditions that impact biofilm formation. Wrinkled colony formation serves as an indicator of biofilm formation in a variety of bacteria, including both Gram-positive bacteria, such as Bacillus subtilis, and Gram-negative bacteria, such as Vibrio cholerae, Vibrio parahaemolyticus, Pseudomonas aeruginosa, and Vibrio fischeri. The marine bacterium V. fischeri has become a model for biofilm formation due to the critical role of biofilms during host colonization: biofilms produced by V. fischeri promote its colonization of the Hawaiian bobtail squid Euprymna scolopes. Importantly, biofilm phenotypes observed in vitro correlate with the ability of V. fischeri cells to effectively colonize host animals: strains impaired for biofilm formation in vitro possess a colonization defect, while strains exhibiting increased biofilm phenotypes are enhanced for colonization. V. fischeri therefore provides a simple model system to assess the mechanisms by which bacteria regulate biofilm formation and how biofilms impact host colonization. In this report, we describe a semi-quantitative method to assess biofilm formation using V. fischeri as a model system. This method involves the careful spotting of bacterial cultures at defined concentrations and volumes onto solid agar media; a spotted culture is synonymous to a single bacterial colony. This 'spotted culture' technique can be utilized to compare gross biofilm phenotypes at single, specified time-points (end-point assays), or to identify and characterize subtle biofilm phenotypes through time-course assays of biofilm development and measurements of the colony diameter, which is influenced by biofilm formation. Thus, this technique provides a semi-quantitative analysis of biofilm formation, permitting evaluation of the timing and patterning of wrinkled colony development and the relative size of the developing structure, characteristics that extend beyond the simple overall morphology."
} | 711 |
35407790 | PMC8999688 | pmc | 2,878 | {
"abstract": "Hydrophilic or superhydrophilic materials in some cases are considered to be potentially icephobic due to a low ice-adhesion strength to such materials. Here, the evolution of the properties of a superhydrophilic aluminum alloy with hierarchical roughness, fabricated by laser processing, was studied in contact with water during prolonged cyclic variation in temperature. It was shown that the chemical interaction of rough alumina with water molecules caused the substitution of the surface oxide by polymorphic crystalline gibbsite or bayerite phases while preserving hierarchical roughness. Due to such substitution, mechanical durability was notably compromised. Thus, in contrast to the superhydrophobic laser-processed samples, the superhydrophilic samples targeted on the exploitation in an open atmosphere as a material with anti-icing properties cannot be considered as the industrially attractive way to combat icing.",
"conclusion": "5. Conclusions Our studies showed that the superhydrophilic samples of aluminum alloy with the hierarchical alumina texture are not stable when in contact with water and cyclic variations from room temperature to low negative values. Chemical interaction of the hierarchically rough alumina with water molecules caused the oxide to be replaced by polymorph crystalline phases (gibbsite or bayerite). Although the preservation of the hierarchical roughness during such substitution allowed the retention of superhydrophilic properties, it was accompanied by a notable compromise in the mechanical resistance. In particular, the poor mechanical properties of aluminum-layered hydroxides led to weaker abrasive resistance. Thus, in contrast to the superhydrophobic laser-processed samples, which demonstrated very high durability in various atmospheric conditions [ 28 ], the superhydrophilic samples showed chemical transformation and mechanical degradation when in contact with water under cyclic temperature variation. Therefore, despite their potential anti-icing properties, superhydrophilic aluminum surfaces cannot be considered as an attractive way for industry to combat icing in open atmospheric conditions.",
"introduction": "1. Introduction Hydrophilic or superhydrophilic materials in some cases are considered potentially icephobic due to low ice-adhesion strength to hydrophilic surfaces at negative ambient temperatures [ 1 ]. The main mechanism for this low adhesion is related to the presence of nonfreezing water, which serves as a lubricating interfacial layer. Two mechanisms related to the formation of such a nonfrozen water layer are discussed in the literature. The first one considers the formation of a hydrated water layer on top of hydrophilic polymers, polymer gels, or polyzwitterion brushes, which have a high affinity to water molecules [ 2 , 3 , 4 , 5 ]. This hydrated water layer does not crystallize even at a temperature below −50 °C [ 3 ]. The second one can be induced by hydrophilic surfaces with nanopores when the surface forces inside the pore cause an alteration in the properties of water by shifting freezing temperatures inside the pore toward low negative values [ 1 , 6 , 7 ]. Although the idea to use superhydrophilic surfaces––for example as an anti-icing material––in open atmospheric conditions looks attractive, the critical point in development is related to scarce data on the evolution of their properties during cyclic variations in temperature from low negative to high positive. At the same time, the stability of the properties and the nanoporous structure of such material is a prerequisite for preserving the lubricating bound water layer and hence for their use in continuous contact with liquid and solid aqueous atmospheric precipitation. Herein we present the first study of the evolution of the properties of a superhydrophilic aluminum alloy with hierarchical roughness in contact with water during prolonged cyclic temperature variation and discuss the possibility of applying such material in cold environmental conditions in the presence of atmospheric water.",
"discussion": "4. Discussion It has been discussed in the literature that aluminum oxide surfaces are stable in an aqueous environment with a wide range of pH from 3 to 9 [ 17 ]. However, experimental studies of η-Al 2 O 3 and γ- Al 2 O 3 dispersed in aqueous media having different pH indicate oxide chemical activity and transformation to hydroxide phases [ 18 , 19 , 20 , 21 , 22 ]. This transition takes place in both the liquid aqueous phase and in a humid atmosphere [ 19 , 23 ]. The literature shows that, in the presence of water, hydroxides demonstrate a higher thermodynamic stability than alumina, supported by both experiment and thermodynamic calculations [ 18 , 19 , 22 ]. The reaction of alumina particles with water might result in different products such as bayerite β-Al(OH) 3 , gibbsite α-Al(OH) 3 , or boehmite AlOOH, depending on temperature, pH of the aqueous phase, and type of oxide [ 18 , 19 , 20 , 21 , 22 , 23 , 24 ]. For example, at room temperature and low pH the gibbsite phase prevailed, while a high pH facilitated bayerite formation [ 18 , 19 , 20 , 21 , 22 ]. Increased temperatures promoted the formation of boehmite or pseudo-boehmite [ 25 , 26 ] when in contact with water on a textured aluminum surface. In our experiments, hierarchical surface morphology was constituted by nanoparticle aggregates having a highly developed surface, which came in contact with cold/room temperature water in a homogeneous (Wenzel) wetting regime. EDX, SEM, Fourier-IR reflectance spectroscopy and XRD data as presented above showed the formation of either gibbsite or bayerite. The complex surface morphology made it difficult to differentiate between these two hydroxides with the same composition but different structures: different relative stackings of the same unit layers. For a better understanding of the phase type for the hydroxide that formed upon temperature cycling, we measured the pH variation of the aqueous phase in contact with our samples during 5 thermocycles. Since we changed the water inside the polypropylene containers with the sample every 5 cycles, as described in the experimental section, the duration of surface contact with the same water bath exceeded 900 min. Such a protocol was selected to mimic the contact of a superhydrophilic surface with atmospheric precipitation under outdoor conditions. It was found that the pH of the aqueous medium for the freshly textured sample quickly deviated from the value of pH 5.6 ± 0.1 characteristic of water at the beginning of this experiment. After one cycle, the alkalinity of the medium increased to pH 6.6±0.2, while after 5 cycles it reached pH 7.2 ± 0.2. On the one hand, this increase in the alkalinity of the aqueous phase was related to the oxidation of the sample surface, where nanoparticles of alumina decorated the substrate, containing patches of nonoxidized aluminum. For the nonoxidized aluminum patches, the following reaction took place: 2Al + 6H 2 O = 2Al(OH) 3 + 3H 2 ↑. For the superhydrophilic sample in contact with the aqueous phase for tens of cycles, the rate of pH variation was somewhat lower, from pH 5.6 ± 0.1 to 6.8 ± 0.2. The chemical interaction of alumina with water, characterized by the negative value of the free energies of hydration [ 19 ], is described by the reaction: Al 2 O 3 (s) + 3H 2 O(l) = 2Al(OH) 3 (s) Notations (s) and (l) denote the solid or liquid state of the reaction components. In the literature, three different transformation mechanisms from the oxide to hydroxide phase were considered [ 18 , 27 ]. The first one was associated with alumina surface hydration through hydrolysis of Al–O-Al surface bonds and the formation of Al-OH. It was expected that after this mechanism the transformation of the texturing elements composed of alumina should result in the preservation of the macro- and microsurface texture. The second mechanism was based on the dissolution of oxide, followed by the precipitation of the hydroxide from the saturated or supersaturated dispersions of hydroxide nuclei [ 18 ]. This mechanism resulted in the formation of a new poorly ordered type of texture atop our hierarchically rough samples. Finally, according to [ 27 ], the third mechanism related the conversion of alumina to aluminum hydroxide through the occupation of the vacancies left due to inward oxygen diffusion by OH − ions. Following this mechanism, the preservation of initial micro and macrotexture was highly likely. From the SEM pictures of our samples after prolonged thermocycling ( Figure 1 b,d) it can be concluded that the morphology of the initial sample’s surface was partially inherited, while microaggregates of nanoparticles were replaced by a layered texture formed by ordered plates. It looks like the hydroxide was formed by the joint action of all mentioned mechanisms of formation. The substitution of the oxide texture by the hydroxide caused a weakening of the mechanical resistance. In our recent studies [ 8 , 20 ], grain refinement, hardening, and enrichment of the surface layer with nano-inclusions of aluminum oxynitride during laser processing provided a significant increase in the resistance of the surface layer to abrasion. Thus, the laser-processed samples demonstrated good mechanical stability. For the hydroxide, the crystalline structure of gibbsite and bayerite differed only in the packing arrangement of the Al–O–Al layers [ 3 ], which were bound to each other by a network of hydrogen bonds. The weakness of the H bonds between the layers caused easy destruction of the layer network for both polymorphs. It should be noted that in this study we chose the AMG2 aluminum–magnesium alloy because it is widely used in the industry. This alloy has quite a low bulk Mg concentration of 2.9%, however, the surface concentration of magnesium after laser processing was even lower. The absence of a peak corresponding to Mg on the EDX spectra ( Figure 3 ) led to the conclusion that the Mg concentration in surface layers was lower than 0.2–0.3%. This result, known in the literature, was not surprising and can be explained by the high volatility (in comparison with Al) of magnesium during laser-induced vaporization. Thus, the effect of additional alloy compounds on the processes considered in this manuscript is negligible."
} | 2,581 |
38259349 | PMC10802671 | pmc | 2,882 | {
"abstract": "Microbial ecosystems are commonly modeled by fixed interactions between species in steady exponential growth states. However, microbes often modify their environments so strongly that they are forced out of the exponential state into stressed or non-growing states. Such dynamics are typical of ecological succession in nature and serial-dilution cycles in the laboratory. Here, we introduce a phenomenological model, the Community State model, to gain insight into the dynamic coexistence of microbes due to changes in their physiological states. Our model bypasses specific interactions (e.g., nutrient starvation, stress, aggregation) that lead to different combinations of physiological states, referred to collectively as “community states”, and modeled by specifying the growth preference of each species along a global ecological coordinate, taken here to be the total community biomass density. We identify three key features of such dynamical communities that contrast starkly with steady-state communities: increased tolerance of community diversity to fast growth rates of species dominating different community states, enhanced community stability through staggered dominance of different species in different community states, and increased requirement on growth dominance for the inclusion of late-growing species. These features, derived explicitly for simplified models, are proposed here to be principles aiding the understanding of complex dynamical communities. Our model shifts the focus of ecosystem dynamics from bottom-up studies based on idealized inter-species interaction to top-down studies based on accessible macroscopic observables such as growth rates and total biomass density, enabling quantitative examination of community-wide characteristics.",
"introduction": "Introduction Microbial communities in natural environments are often highly dynamic [ 1 – 7 ]. For example, many environments feature periodic replenishment of resources (e.g., the gut microbiome [ 8 ], the ocean [ 9 ]), or resetting of other environmental factors with periods of growth between these perturbations [ 10 , 11 ]. Lab-scale experiments [ 12 – 14 ] on microbial ecosystems frequently adopt serial dilution cycles with dynamic environments. Recent studies have found that stable microbial communities do not settle simply into a fixed state, but are instead driven through dynamic phases involving complex changes in the environment such as depletion of oxygen and build-up of toxic waste [ 15 , 16 ]. These changes, in turn, alter the physiological states of the microbes in the community, slowing down or even halting their growth. Changing physiological states also often change metabolic secretion and uptake profiles, and induce more complex interactions such as aggregation, motility, toxin secretion, and even contact-dependent killing [ 17 – 20 ]. These observations have been interpreted using models from theoretical ecology that typically explain ecosystem assembly and stability [ 21 , 22 ] in terms of resource competition [ 12 , 13 , 23 ], niche differentiation [ 24 ] and competitive exclusion [ 25 ]. However, these models typically assume that communities and the organisms in them are at steady state [ 21 , 26 – 29 ]. This difference between empirical observations and theoretical models raises questions about the role of dynamic physiological state changes in forming complex communities. One possibility is that physiological state changes are merely details, not essential for understanding factors that enable community assembly. In this perspective, nothing is lost by coarse-graining over dynamics and modeling communities as if they are at steady state. Microbial communities would be expected to show similar complexity and structures if microbes stay in fixed states (e.g., exponential growth) and independent of whether interactions through metabolic secretion and uptake occur in a temporally staged manner. Another possibility is that physiological state changes create dynamic niches that support complex communities. Since microbes have a plethora of non-growing states, this scenario could significantly expand the ways of generating niches beyond well-studied cases such as distinct metabolites [ 28 ], space [ 30 ], and externally dictated temporal epochs (e.g., diel or annual cycles) [ 31 ]. Further, the nature of such self-generated dynamic niches, if they exist, might have signatures that are predicted to be observed in microbial communities. We cannot easily address this question about the role of state changes using the current bottom-up theoretical frameworks (e.g., Lotka-Volterra or Consumer-Resource models) since these models typically characterize organisms and their interactions with fixed parameters. In these models, community dynamics only involves changes in species abundances and nutrient concentrations and is justified by assuming organisms are in a fixed physiological state, (e.g., Monod growth for exponentially-growing cells [ 32 ]). If one is to adopt a model of interactions between each species and its environment (as in Consumer-Resource Models) or other species (as in Lotka-Volterra Models), then each physiological state would minimally involve a different set of uptake and excretion parameters; a given species would effectively be modeled as multiple species over time. Thus, bottom-up models of dynamic communities require extensive characterization and unconstrained assumptions on specific details about what different cells do in different conditions. As a first step towards quantitatively modeling communities of species that undergo physiological changes, we introduce a minimal top-down phenomenological framework, the Community State Model. Our model is phenomenological at the level of species density; the physiological state and thus the growth rate of each species in a community is assumed to depend only on the community biomass at any given time, and as a result, community states are defined by regions of biomass density. Such a model can be solved explicitly (numerically and in simple cases analytically) to yield the temporal organization of community dynamics at a quantitative level. Analysis of the Community State model points to sequential dynamics as a strategy to form a stable community involving a large number of species [ 33 ]. In this simple model, each species grows rapidly in one (or a few) community states that persist over specific intervals of biomass accumulation, with slower or even no growth in other parts of the inter-dilution period (hereby referred to as the growth period). This strategy is a distinct alternative to the co-growth strategy based on steady-state models with fixed physiological states where species grow on resource niches simultaneously. For this sequential coexistence strategy, our model allows us to uncover a number of key features of community dynamics. We find (a) tolerance of community diversity to fast-growing species if such growth is limited to specific community states, (b) enhanced community stability through staggered dominance of different species in different community states, and (c) a requirement of increased growth dominance for late-growing species. These features counteract the dominant notions regarding species competition derived from analysis of steady-state systems, and serve as principles to guide the understanding of complex dynamical ecosystems",
"discussion": "Discussion In this work, we have investigated models of growth and survival of microbial species in communities subjected to cyclic environmental fluctuations, focusing on the case of prolonged periods between nutrient replenishment as seen often in the wild [ 34 , 44 ]; under these conditions, exponential steady-state growth cannot be sustained. In the lab, non-steady-state growth can occur during serial-dilution cycles where the cycle length is long enough for nutrient depletion or build up for toxic waste that limits growth [ 12 , 36 , 45 ]. Accurate bottom-up models are not feasible given our limited understanding of microbial behaviors outside of exponential growth [ 44 , 46 ]. Inspired by a model experimental system, we investigated the creation of dynamical niches by a combination of physiological states taken by members of a community in response to self-generated environmental changes. In our model, each species X is assigned a growth rate r X ( S ) in each state S of the community. A community state S was taken to last for an interval of total community biomass ρ tot accumulated during the growth of the community, and the corresponding growth rate of each species in that community state S reflects the physiological state each species is in as well as its environmental context (which includes the physiological states of all other species). Using total community biomass ρ tot as a driver of community state transitions gives a simple model that allows us to derive quantitative self-consistency conditions on temporal dynamics during serial-dilution cycles using experimentally measurable quantities: Suppose a set of species in repeated serial-dilution cycles are observed to grow at growth rates { r α } at time t where the community has total biomass ρ tot ( t ), to what extent can the set of data {r α , ρ tot } recapitulate the existence of species and the dynamics of their abundances during the cycle? And how robust is the observed dynamics to perturbations in environmental factors and community composition? Most of the results derived in this study are centered around these questions. One major finding is that community states cannot be taken for granted as “niches” - even when species “take turns” dominating growth in different community states, many species can go extinct. Instead, we find quantitative constraints on how fast or for how wide an interval the dominant species in each community state niche can grow. These constraints can be summarized as: (1) Tolerance to growth dominance: increasing the growth rate of an individual species in its favored community state beyond a point does not impact coexistence. This effect contrasts starkly with steady-state coexistence, where the increased growth rate of one species can drive other species extinct. (2) Community stability through staggered growth dominance: Stability of a diverse community requires minimizing simultaneous fast growth of multiple species; that is, stability requires that species stagger their growth dominance across distinct community states. (3) Increased growth dominance for late species: species growing in late community states must grow faster or for larger biomass intervals than species in early states. By showing that such niches can arise for a distinct mechanistic reason – transitions between physiological states – the Community State model makes distinct predictions about the relationship between physiology and ecology. Unlike in many other models, niches here are not created by a balance between microbes in exponential growth but originate through an interplay of switches between multiple growing and non-growing states. Our mutual invasibility criteria Eq. 3 offer a quantitative and intuitive understanding of the nature of these niches, factors that widen them, and the nature of competition between species in overlapping niches. The explicit dependence on the ratio of the growth rate of the invading species to that of the resident species in I A,B and I B,A make them a fitness-like measure for the current context (cyclic environments with multiple physiological states) where other fitness measures (e.g., growth rate difference) are not applicable. These effects persist in extensions of the model to many-species communities provided that the number of community states does not greatly exceed the growth dominance p , where p is the ratio of the growth rate of the dominant species in a community state to the basal growth rate of subordinate species in that state. Trajectories of species abundances ( Fig. 5H ) show that the dynamics in such systems no longer follow orderly succession dynamics but instead, show a seemingly-disordered array of growth curves that are nevertheless cyclic and hence maintain coexistence. To our knowledge, such disorderly, yet cyclic growth characteristics represent a new class of non-steady-state dynamics that has not been described in the ecology literature. The tendency in ecology, with an emphasis on steady states, has been to “coarse-grain” or ignore dynamics seen in real systems. Indeed, dynamics in some communities are merely complications (e.g., periodic perturbations about a stable steady state) that can be coarse-grained without any loss of understanding. In fact, in 1973, Stewart and Levin noted mathematically that two species could survive on a single “seasonal resource” [ 47 ]. Their work has often been dismissed (including in their own paper [ 47 ]) as a mathematical observation relevant only for an assumed growth-affinity trade-off, narrow resource competition, and other idealizations. Our work argues that their simple mechanism of dynamic coexistence – also explored in recent works [ 41 , 48 – 52 ] – is more relevant, not less, given the observed complex physiology and nonlinear growth dependencies in real microbial ecosystems. If physiological state changes turn out to be dominant drivers of dynamical niches, as seen in [ 12 , 16 , 53 ], dynamics cannot be “averaged” over but become the essential link between physiology and ecology. We regard an attractive feature of the top-down Community State model to be its direct quantitative connection to experimentally-accessible variables, as well as its avoidance of often inaccessible interaction parameters. Instantaneous growth rates of individual species can be obtained from transient changes in species abundances (via e.g., 16S sequence as proxy), and the total community biomass can be obtained by measuring total protein or total RNA as proxies, or simply by the optical density if the culture does not aggregate. The model studied here, therefore, provides a roadmap for the quantitative analysis of community-wide data to learn about community dynamics, going beyond taxonomic characterization, without invoking fitting parameters. This contrasts starkly with dynamical analysis based on commonly used bottom-up models which invariably involve a large number of unconstrained interaction parameters (e.g., the species interaction matrix in generalized Lotka-Volterra models, or the nutrient consumption matrix in Consumer-Resource models). Additionally, it emphasizes intra-cycle dynamics which has been largely neglected except for a few recent studies [ 16 , 50 , 51 , 53 ], and gives concrete predictions, e.g., on the growth rate and duration of early vs late species, that can be tested directly by data. In this sense, the Community State model is a phenomenological model that can be updated directly from data. Our approach shares common elements with other top-down approaches like the Stochastic Logistic Model [ 54 , 55 ] and recent data-driven models [ 56 , 57 ] without explicit interspecies interactions. While these other models attribute growth rate fluctuations to external factors, our model focuses on endogenously-driven environmental change. Our model can be extended to incorporate external fluctuations that randomly perturb growth niches, either across hosts or across cycles, predicting various abundance distributions as in [ 54 ]. However, a key distinction is that our approach imposes closure conditions on growth rate variations in repeated cycles needed for stable but dynamic coexistence. The key idea in our work is the existence of global community states that can be sensed by microbes in that community. Our results suggest that it would be advantageous for organisms to use this information to adjust their behavior and grow in specific community states since such regulation would maximize their chance of survival in the community. For example, organisms occupying early phases of the cycle may benefit from limiting their own growth so as not to eliminate other species active later in the cycle, as late species could be important for the survival of all species in later phases of the cycle – as is the case for acid-induced stress relief [ 16 ], the early blooming acid-producing sugar eater is rescued from death by the late-blooming acid consumer which removes the excreted acid and restores the environment. A more speculative aspect of the Community State model is that the sequence of community states can be parameterized by a one-dimensional eco-coordinate (as opposed to environmental factors or abundances of individual species). A further assumption that allowed for deriving quantitative coexistence criteria and relating them to empirical data is that community biomass can serve as this coordinate parameterizing the sequence of community states. We believe this hypothesis is biologically plausible: First, a number of key physiological parameters, e.g., pH, oxygen content, waste products, and iron availability, change with the accumulation of community biomass [ 58 ], and the values of these parameters to cause transitions in the physiological states of individual organisms are known. Other physiological effects such as lag time and cell death might introduce limitations for our framework that require further study [ 50 , 51 , 59 ]. Second, it is known that several autoinducers are produced and sensed by a wide range of both gram-positive and gram-negative bacteria [ 60 – 62 ]. In fact, AI-2 has been proposed to serve as a “universal signal” for inter-species communication [ 63 – 65 ]. Third, it is common for microbes to develop sensors to detect important features of their environment [ 66 – 69 ]; as total community biomass is clearly an important dynamical variable that can be used to forecast the fate of the community (e.g., how close to the carrying capacity), it would not be surprising if organisms have evolved various proxy schemes to sense the total biomass. As bacteria feature multiple sensors and regulatory processes, they may detect various (and possibly distinct from other species) aspects of the global state of the community and integrate the available information through diverse regulatory mechanisms. Thus, community biomass may be viewed as a simplified description to summarize the effects of the different sensors. The defining feature of the Community State model is the ability of organisms in a community to sense common features of the community and their ability to modulate their own physiology in response to such community-wide signals. Indeed, the existence of a group of organisms that can sense and respond to common features in the environment may be taken as a key characteristic that defines a “community”."
} | 4,720 |
38260536 | PMC10802591 | pmc | 2,883 | {
"abstract": "Microbial ecosystems are commonly modeled by fixed interactions between species in steady exponential growth states. However, microbes often modify their environments so strongly that they are forced out of the exponential state into stressed or non-growing states. Such dynamics are typical of ecological succession in nature and serial-dilution cycles in the laboratory. Here, we introduce a phenomenological model, the Community State model, to gain insight into the dynamic coexistence of microbes due to changes in their physiological states. Our model bypasses specific interactions (e.g., nutrient starvation, stress, aggregation) that lead to different combinations of physiological states, referred to collectively as “community states”, and modeled by specifying the growth preference of each species along a global ecological coordinate, taken here to be the total community biomass density. We identify three key features of such dynamical communities that contrast starkly with steady-state communities: increased tolerance of community diversity to fast growth rates of species dominating different community states, enhanced community stability through staggered dominance of different species in different community states, and increased requirement on growth dominance for the inclusion of late-growing species. These features, derived explicitly for simplified models, are proposed here to be principles aiding the understanding of complex dynamical communities. Our model shifts the focus of ecosystem dynamics from bottom-up studies based on idealized inter-species interaction to top-down studies based on accessible macroscopic observables such as growth rates and total biomass density, enabling quantitative examination of community-wide characteristics.",
"introduction": "Introduction Microbial communities in natural environments are often highly dynamic [ 1 - 7 ]. For example, many environments feature periodic replenishment of resources (e.g., the gut microbiome [ 8 ], the ocean [ 9 ]), or resetting of other environmental factors with periods of growth between these perturbations [ 10 , 11 ]. Lab-scale experiments [ 12 - 14 ] on microbial ecosystems frequently adopt serial dilution cycles with dynamic environments. Recent studies have found that stable microbial communities do not settle simply into a fixed state, but are instead driven through dynamic phases involving complex changes in the environment such as depletion of oxygen and build-up of toxic waste [ 15 , 16 ]. These changes, in turn, alter the physiological states of the microbes in the community, slowing down or even halting their growth. Changing physiological states also often change metabolic secretion and uptake profiles, and induce more complex interactions such as aggregation, motility, toxin secretion, and even contact-dependent killing [ 17 - 20 ]. These observations have been interpreted using models from theoretical ecology that typically explain ecosystem assembly and stability [ 21 , 22 ] in terms of resource competition [ 12 , 13 , 23 ], niche differentiation [ 24 ] and competitive exclusion [ 25 ]. However, these models typically assume that communities and the organisms in them are at steady state [ 21 , 26 - 29 ]. This difference between empirical observations and theoretical models raises questions about the role of dynamic physiological state changes in forming complex communities. One possibility is that physiological state changes are merely details, not essential for understanding factors that enable community assembly. In this perspective, nothing is lost by coarse-graining over dynamics and modeling communities as if they are at steady state. Microbial communities would be expected to show similar complexity and structures if microbes stay in fixed states (e.g., exponential growth) and independent of whether interactions through metabolic secretion and uptake occur in a temporally staged manner. Another possibility is that physiological state changes create dynamic niches that support complex communities. Since microbes have a plethora of non-growing states, this scenario could significantly expand the ways of generating niches beyond well-studied cases such as distinct metabolites [ 28 ], space [ 30 ], and externally dictated temporal epochs (e.g., diel or annual cycles) [ 31 ]. Further, the nature of such self-generated dynamic niches, if they exist, might have signatures that are predicted to be observed in microbial communities. We cannot easily address this question about the role of state changes using the current bottom-up theoretical frameworks (e.g., Lotka-Volterra or Consumer-Resource models) since these models typically characterize organisms and their interactions with fixed parameters. In these models, community dynamics only involves changes in species abundances and nutrient concentrations and is justified by assuming organisms are in a fixed physiological state, (e.g., Monod growth for exponentially-growing cells [ 32 ]). If one is to adopt a model of interactions between each species and its environment (as in Consumer-Resource Models) or other species (as in Lotka-Volterra Models), then each physiological state would minimally involve a different set of uptake and excretion parameters; a given species would effectively be modeled as multiple species over time. Thus, bottom-up models of dynamic communities require extensive characterization and unconstrained assumptions on specific details about what different cells do in different conditions. As a first step towards quantitatively modeling communities of species that undergo physiological changes, we introduce a minimal top-down phenomenological framework, the Community State Model. Our model is phenomenological at the level of species density; the physiological state and thus the growth rate of each species in a community is assumed to depend only on the community biomass at any given time, and as a result, community states are defined by regions of biomass density. Such a model can be solved explicitly (numerically and in simple cases analytically) to yield the temporal organization of community dynamics at a quantitative level. Analysis of the Community State model points to sequential dynamics as a strategy to form a stable community involving a large number of species [ 33 ]. In this simple model, each species grows rapidly in one (or a few) community states that persist over specific intervals of biomass accumulation, with slower or even no growth in other parts of the inter-dilution period (hereby referred to as the growth period). This strategy is a distinct alternative to the co-growth strategy based on steady-state models with fixed physiological states where species grow on resource niches simultaneously. For this sequential coexistence strategy, our model allows us to uncover a number of key features of community dynamics. We find (a) tolerance of community diversity to fast-growing species if such growth is limited to specific community states, (b) enhanced community stability through staggered dominance of different species in different community states, and (c) a requirement of increased growth dominance for late-growing species. These features counteract the dominant notions regarding species competition derived from analysis of steady-state systems, and serve as principles to guide the understanding of complex dynamical ecosystems",
"discussion": "Discussion In this work, we have investigated models of growth and survival of microbial species in communities subjected to cyclic environmental fluctuations, focusing on the case of prolonged periods between nutrient replenishment as seen often in the wild [ 34 , 44 ]; under these conditions, exponential steady-state growth cannot be sustained. In the lab, non-steady-state growth can occur during serial-dilution cycles where the cycle length is long enough for nutrient depletion or build up for toxic waste that limits growth [ 12 , 36 , 45 ]. Accurate bottom-up models are not feasible given our limited understanding of microbial behaviors outside of exponential growth [ 44 , 46 ]. Inspired by a model experimental system, we investigated the creation of dynamical niches by a combination of physiological states taken by members of a community in response to self-generated environmental changes. In our model, each species X is assigned a growth rate r X ( S ) in each state S of the community. A community state S was taken to last for an interval of total community biomass ρ t o t accumulated during the growth of the community, and the corresponding growth rate of each species in that community state S reflects the physiological state each species is in as well as its environmental context (which includes the physiological states of all other species). Using total community biomass ρ t o t as a driver of community state transitions gives a simple model that allows us to derive quantitative self-consistency conditions on temporal dynamics during serial-dilution cycles using experimentally measurable quantities: Suppose a set of species in repeated serial-dilution cycles are observed to grow at growth rates { r α } at time t where the community has total biomass ρ tot ( t ) , to what extent can the set of data { r α , ρ tot } recapitulate the existence of species and the dynamics of their abundances during the cycle? And how robust is the observed dynamics to perturbations in environmental factors and community composition? Most of the results derived in this study are centered around these questions. One major finding is that community states cannot be taken for granted as “niches” - even when species “take turns” dominating growth in different community states, many species can go extinct. Instead, we find quantitative constraints on how fast or for how wide an interval the dominant species in each community state niche can grow. These constraints can be summarized as: (1) Tolerance to growth dominance: increasing the growth rate of an individual species in its favored community state beyond a point does not impact coexistence. This effect contrasts starkly with steady-state coexistence, where the increased growth rate of one species can drive other species extinct. (2) Community stability through staggered growth dominance: Stability of a diverse community requires minimizing simultaneous fast growth of multiple species; that is, stability requires that species stagger their growth dominance across distinct community states. (3) Increased growth dominance for late species: species growing in late community states must grow faster or for larger biomass intervals than species in early states. By showing that such niches can arise for a distinct mechanistic reason – transitions between physiological states – the Community State model makes distinct predictions about the relationship between physiology and ecology. Unlike in many other models, niches here are not created by a balance between microbes in exponential growth but originate through an interplay of switches between multiple growing and non-growing states. Our mutual invasibility criteria Eq. 3 offer a quantitative and intuitive understanding of the nature of these niches, factors that widen them, and the nature of competition between species in overlapping niches. The explicit dependence on the ratio of the growth rate of the invading species to that of the resident species in I A , B and I B , A make them a fitness-like measure for the current context (cyclic environments with multiple physiological states) where other fitness measures (e.g., growth rate difference) are not applicable. These effects persist in extensions of the model to many-species communities provided that the number of community states does not greatly exceed the growth dominance p , where p is the ratio of the growth rate of the dominant species in a community state to the basal growth rate of subordinate species in that state. Trajectories of species abundances ( Fig. 5H ) show that the dynamics in such systems no longer follow orderly succession dynamics but instead, show a seemingly-disordered array of growth curves that are nevertheless cyclic and hence maintain coexistence. To our knowledge, such disorderly, yet cyclic growth characteristics represent a new class of non-steady-state dynamics that has not been described in the ecology literature. The tendency in ecology, with an emphasis on steady states, has been to “coarse-grain” or ignore dynamics seen in real systems. Indeed, dynamics in some communities are merely complications (e.g., periodic perturbations about a stable steady state) that can be coarse-grained without any loss of understanding. In fact, in 1973, Stewart and Levin noted mathematically that two species could survive on a single “seasonal resource” [ 47 ]. Their work has often been dismissed (including in their own paper [ 47 ]) as a mathematical observation relevant only for an assumed growth-affinity trade-off, narrow resource competition, and other idealizations. Our work argues that their simple mechanism of dynamic coexistence – also explored in recent works [ 41 , 48 - 52 ] – is more relevant, not less, given the observed complex physiology and nonlinear growth dependencies in real microbial ecosystems. If physiological state changes turn out to be dominant drivers of dynamical niches, as seen in [ 12 , 16 , 53 ], dynamics cannot be “averaged” over but become the essential link between physiology and ecology. We regard an attractive feature of the top-down Community State model to be its direct quantitative connection to experimentally-accessible variables, as well as its avoidance of often inaccessible interaction parameters. Instantaneous growth rates of individual species can be obtained from transient changes in species abundances (via e.g., 16S sequence as proxy), and the total community biomass can be obtained by measuring total protein or total RNA as proxies, or simply by the optical density if the culture does not aggregate. The model studied here, therefore, provides a roadmap for the quantitative analysis of community-wide data to learn about community dynamics, going beyond taxonomic characterization, without invoking fitting parameters. This contrasts starkly with dynamical analysis based on commonly used bottom-up models which invariably involve a large number of unconstrained interaction parameters (e.g., the species interaction matrix in generalized Lotka-Volterra models, or the nutrient consumption matrix in Consumer-Resource models). Additionally, it emphasizes intra-cycle dynamics which has been largely neglected except for a few recent studies [ 16 , 50 , 51 , 53 ], and gives concrete predictions, e.g., on the growth rate and duration of early vs late species, that can be tested directly by data. In this sense, the Community State model is a phenomenological model that can be updated directly from data. Our approach shares common elements with other top-down approaches like the Stochastic Logistic Model [ 54 , 55 ] and recent data-driven models [ 56 , 57 ] without explicit interspecies interactions. While these other models attribute growth rate fluctuations to external factors, our model focuses on endogenously-driven environmental change. Our model can be extended to incorporate external fluctuations that randomly perturb growth niches, either across hosts or across cycles, predicting various abundance distributions as in [ 54 ]. However, a key distinction is that our approach imposes closure conditions on growth rate variations in repeated cycles needed for stable but dynamic coexistence. The key idea in our work is the existence of global community states that can be sensed by microbes in that community. Our results suggest that it would be advantageous for organisms to use this information to adjust their behavior and grow in specific community states since such regulation would maximize their chance of survival in the community. For example, organisms occupying early phases of the cycle may benefit from limiting their own growth so as not to eliminate other species active later in the cycle, as late species could be important for the survival of all species in later phases of the cycle – as is the case for acid-induced stress relief [ 16 ], the early blooming acid-producing sugar eater is rescued from death by the late-blooming acid consumer which removes the excreted acid and restores the environment. A more speculative aspect of the Community State model is that the sequence of community states can be parameterized by a one-dimensional eco-coordinate (as opposed to environmental factors or abundances of individual species). A further assumption that allowed for deriving quantitative coexistence criteria and relating them to empirical data is that community biomass can serve as this coordinate parameterizing the sequence of community states. We believe this hypothesis is biologically plausible: First, a number of key physiological parameters, e.g., pH, oxygen content, waste products, and iron availability, change with the accumulation of community biomass [ 58 ], and the values of these parameters to cause transitions in the physiological states of individual organisms are known. Other physiological effects such as lag time and cell death might introduce limitations for our framework that require further study [ 50 , 51 , 59 ]. Second, it is known that several autoinducers are produced and sensed by a wide range of both gram-positive and gram-negative bacteria [ 60 - 62 ]. In fact, AI-2 has been proposed to serve as a “universal signal” for inter-species communication [ 63 - 65 ]. Third, it is common for microbes to develop sensors to detect important features of their environment [ 66 - 69 ]; as total community biomass is clearly an important dynamical variable that can be used to forecast the fate of the community (e.g., how close to the carrying capacity), it would not be surprising if organisms have evolved various proxy schemes to sense the total biomass. As bacteria feature multiple sensors and regulatory processes, they may detect various (and possibly distinct from other species) aspects of the global state of the community and integrate the available information through diverse regulatory mechanisms. Thus, community biomass may be viewed as a simplified description to summarize the effects of the different sensors. The defining feature of the Community State model is the ability of organisms in a community to sense common features of the community and their ability to modulate their own physiology in response to such community-wide signals. Indeed, the existence of a group of organisms that can sense and respond to common features in the environment may be taken as a key characteristic that defines a “community”."
} | 4,723 |
23940737 | PMC3737133 | pmc | 2,884 | {
"abstract": "Coral harbor diverse and specific bacteria play significant roles in coral holobiont function. Bacteria associated with three of the common and phylogenetically divergent reef-building corals in the South China Sea, Porites lutea , Galaxea fascicularis and Acropora millepora , were investigated using 454 barcoded-pyrosequencing. Three colonies of each species were sampled, and 16S rRNA gene libraries were constructed individually. Analysis of pyrosequencing libraries showed that bacterial communities associated with the three coral species were more diverse than previous estimates based on corals from the Caribbean Sea, Indo-Pacific reefs and the Red Sea. Three candidate phyla, including BRC1, OD1 and SR1, were found for the first time in corals. Bacterial communities were separated into three groups: P . lutea and G . fascicular , A. millepora and seawater. P . lutea and G . fascicular displayed more similar bacterial communities, and bacterial communities associated with A. millepora differed from the other two coral species. The three coral species shared only 22 OTUs, which were distributed in Alphaproteobacteria , Deltaproteobacteria , Gammaproteobacteria , Chloroflexi , Actinobacteria , Acidobacteria and an unclassified bacterial group. The composition of bacterial communities within each colony of each coral species also showed variation. The relatively small common and large specific bacterial communities in these corals implies that bacterial associations may be structured by multiple factors at different scales and that corals may associate with microbes in terms of similar function, rather than identical species.",
"conclusion": "Conclusions In this study, bacterial communities associated with corals from the South China Sea were investigated in detail for the first time. The results showed that coral-associated bacteria are highly diverse and are divergent from the seawater bacterial community. Furthermore, the bacterial community associated with A. millepora was distinct from P. lutea and G. fascicular . In comparison with previous studies, bacterial communities associated with A. millepora and P. lutea in the South China Sea were distinct from those located in the Great Barrier Reef and in Indo-Pacific reefs. It was observed that different coral species share a small common bacterial community, and the composition of the bacterial communities within each colony of each coral species also showed variation. The coexistence of specificity and uniformity reflects the complexity of coral-associated bacterial community and suggests that corals combine the functional bacterial associates in a subtle and sophisticated manner. This study provides novel insights into the complex structure of coral bacterial associates.",
"introduction": "Introduction The abundance of bacteria has been shown to be an important part of the coral holobiont [1] . Coral-associated bacteria are ubiquitous in the coral holobiont temporally and spatially. Planulae older than 79 h harbor internalized bacteria cells [2] . Subsequently, abundant and various bacterial communities were associated with adult corals, for example, Stylophora pistillata and Pocillopora damicornis \n [3] , [4] . Evidence has also accumulated suggesting that coral-associated bacterial communities respond to dynamic environmental conditions at different scales [5] – [7] . In divergent compartments of corals, such as mucus, tissues and the calcium carbonate skeleton, dissimilar bacteria communities have been detected [8] , [9] . Although coral-associated bacterial communities are diverse, they are distinct from ambient seawater bacterial communities [1] , [10] , [11] . Diverse and dynamic coral-associated bacteria assemblages potentially have functions related to nitrogen, carbon and sulfur metabolism, coral disease resistance and abiotic stress tolerance [12] . Our understanding of the specificity of coral-associated microorganisms is changing because the information on coral-derived microbial sequences is increasing at a staggering rate. The bacterial communities associated with the corals Montastraea franksi , Diploria strigosa and Porites astreoides from Panama and Bermuda [1] support the argument that coral-associated bacterial assemblages are most likely species-specific. In contrast, Littman et al. [13] reported that the bacterial communities in three species of Acroporid corals on the Great Barrier Reef were more crucially shaped by location than by the host coral species. Meanwhile, the argument that coral bacterial communities may be both site and species specific has been recently reported [14] . Although all of these studies support the conclusion that corals possess specific microbiota, the inconsistency of the findings on specificity across studies should not be overlooked. These results were mainly obtained using conventional cloning and sequencing or DGGE methods. Therefore, the major limitation of these studies is that the characterization of the microbial communities is not comprehensive. More recently, pyrosequencing has been employed to investigate the bacterial community associated with corals [11] , [15] – [19] . These studies have further supported the conclusion that the bacterial communities appear to be regulated by the host coral species. In addition to the coral species, the significant influence derived from environmental factors has also been emphasized. Therefore, the specificity of coral bacterial communities is more complex than initially thought and still obscure. To better understand the nature of the specificity of coral-associated microorganisms, more comprehensive surveys about more corals from different environments at different scales are required. Coral reefs are widely distributed across the South China Sea, with a total reef area of approximately 7974 km 2 , matching the Great Barrier Reef in size, latitudinal range and biodiversity [20] . However, the microbial consortium has been rarely documented in South China Sea corals. The Luhuitou fringing reef located in Sanya, southern Hainan Island, is approximately 3500 m long and 250–500 m wide and consists of approximately 70% of the coral species so far reported for Hainan Island and its surrounding islands [20] . Luhuitou is a popular tourist location; therefore, investigating of the bacteria associated with local coral colonies is crucial for us to estimate anthropogenic impacts on the coral reef. The aim of this study is to comprehensively investigate the diversity and structure of bacterial communities associated with the three dominant coral species Porites lutea , Galaxea fascicularis and Acropora millepora from the South China Sea. Furthermore, we compared the bacterial communities among coral species and individual colonies to define the common and specific bacteria communities in these corals. Such information will provide a further understanding of the specificity of coral-associated bacteria.",
"discussion": "Discussion Coral Bacterial Community Analysis Presently, the high-throughput pyrosequencing technique combined with barcoded PCR primers has been used for the survey of coral-associated bacterial communities by several researchers [10] , [11] , [15] – [19] . The number of OTUs detected in a single species of coral in the present study is similar to results obtained from Isopora palifera collected from Tan-Tzei Bay [11] . In comparison with the results shown by previous studies, this study revealed a higher bacterial diversity in corals from Sanya Bay than those from the Caribbean Sea [10] , [17] , [18] , Indo-Pacific reefs [18] and the Red Sea [19] . This difference may be due to technical factors, including PCR primer selection and sequencing depth, and may still reflect the essential distinction among different coral species in different environments. Sunagawa et al [15] found that mounding corals ( Montastraea faveolata , M. franksi , D. strigosa and P. astreoides ) had higher estimated diversities than branch-forming acroporid corals and, therefore, speculated that coral morphology plays a role in determining the diversity of coral bacteria. In this study, although the estimated diversities of coral-associated bacteria were not significantly different, they were grouped into A. millepora or P. lutea and G. fascicularis categories, which supports the previous hypothesis to a certain extent. Similar to previous reports, Alphaproteobacteria , Gammaproteobacteria , Firmicutes , Bacteroidetes and Actinobacteria were ubiquitous major groups detected in three coral species. Although Cyanobacteria was predominant in the Red Sea corals [19] , in Acropora formosa and P. lutea from Indo-Pacific reefs [18] and in M. faveolata from the Caribbean Sea [17] , Cyanobacteria were rare in the three coral species studied in this work as well as in the results presented by Chen et al [11] . Members of five candidate phyla, including BRC1, OD1, SR1, TM7 and WS3, with the exception of WS3 and TM7 [19] , [28] , were not previously known to inhabit corals. In previous studies, the most abundant bacteria of A. millepora in the Great Barrier Reef were Gammaproteobacteria , Alphaproteobacteria and Betaproteobacteria or Deltaproteobacteria \n [13] , while in this study, Gammaproteobacteria and Firmicutes and Deinococcus-Thermus were dominant in the A. millepora affiliated bacterial community. The divergence of habitat between this and previous studies may contribute to the different bacterial communities. Moreover, Deinococcus-Thermus has also been observed in stony coral Pocillopora verrucosa , Astreopora myriophthalma and S. pistillata and soft coral Sarcophyton sp. from the Red Sea [19] . In contrast to P. lutea in Indo-Pacific reefs, Actinobacteria , Bacteroidetes and Planctomycetes were more abundant and Cyanobacteria and Gammaproteobacteria were less abundant in the P. lutea -associated bacterial community in Sanya Bay [18] . Additionally, Chlorobi was a major group in the P. lutea -associated bacterial community from Sanya Bay, but it was absent in bacterial communities associated with Indo-Pacific P. lutea \n [18] . All of the observations mentioned above suggest that these differences are most likely due to geographical separation and distinct environmental conditions. Potential Functional Groups As coral reefs often reside in nutrient limited waters, nitrogen-fixing microbes are important for compensating the nitrogen deficit in coral holobionts [10] . Several bacteria potentially involved in nitrogen-fixing have been detected in this study, including Chlorobia , Chloroflexi , Clostridia and Cyanobacteria . Scleractinian corals are significant contributors to the production of dimethylsufoniopropionate (DMSP) and dimethysulfide (DMS), which are key compounds in the global sulfur cycle [29] . Diverse coral-associated bacteria take part in the degradation of DMSP and DMS. In this study, bacterial groups capable of metabolizing DMSP/DMS, such as Ruegeria , Pseudomonas , Acinetobacter , Desulfovibrio , Flavobacterium , Cytophaga , Oceanicola and Comamonas \n [29] were observed in three coral libraries. These diverse and metabolic potential bacterial groups play a crucial role in the biogeochemical cycle. Actinobacteria were observed in abundance in coral samples but were rare in seawater. This group may generate a diverse array of antibacterial compounds that protect the coral from pathogens [30] . The coexistence of various potential functional groups should be essential to the coral holobiont. Therefore, the detailed ecological functions of the bacterial groups identified in this study warrant further research. \n Lachnospiraceae was previously suggested to be a bacterial group for fecal source tracking [31] ; however, Newton et al [32] further suggested that the single phylotype Lachno2, which is closely related to the genus Blautia , would be a candidate for a host-associated fecal indicator. A high proportion of Lachnospiraceae was detected in G. fascicularis colony 3, and most of them belonged to an unclassified group. Whether they are related to human fecal bacteria still need further investigated. Additionally, Escherichia , which are assumed to be animal-associated bacteria [33] , appeared prominently in all A. millepora colonies. Because Sanya Bay is a popular tourist spot, the presence of Lachnospiraceae and Escherichia indicates that we should pay attention to the pollution sources in Sanya Bay, and the real reason for the appearance of these bacteria needs more study. Common and Specific Bacterial Communities Associated with Corals As in sponges and the human gut, the common bacterial community in corals was rather small [34] , [35] . The representative sequences of these 22 OTUs shared by three coral species, except 2 OTUs that also observed in seawater, showed ≦98% similarity to sequences in GenBank, most of which were previously found in sponge- or coral-associated microbial communities ( Table 2 ). It appears that these 20 OTUs might be coral-specific bacteria adapted to the coral reef niche. The species-specific community was large in contrast to the common bacterial community. Although bacterial communities associated with corals were grouped into A. millepora or P. lutea and G. fascicularis categories, bacterial composition in each colony varied at the 97% OTU level. This variation has also been detected in the I . palifera bacterial community [11] . The extensive specificity of coral-associated bacteria might result from varied coral development stages or exterior environments [1] , [28] . Previous studies have indicated that the specific bacterial lineages present in individual sponge and human gut microbiomes vary [34] , [35] . Turnbaugh et al [35] further proposed that different bacterial species assemblages shared genes among sampled individuals, comprising a “core microbiome” at the genomic level rather than the bacterial lineage level. Different sets of microbial species observed in coral individuals sampled in this study allow for us to speculate that these diverse combinations of species may fulfill the same functional roles required by corals through functional-overlap. Whether this pattern exists in coral-associated bacterial assemblages still needs further global investigation and more direct evidence. Conclusions In this study, bacterial communities associated with corals from the South China Sea were investigated in detail for the first time. The results showed that coral-associated bacteria are highly diverse and are divergent from the seawater bacterial community. Furthermore, the bacterial community associated with A. millepora was distinct from P. lutea and G. fascicular . In comparison with previous studies, bacterial communities associated with A. millepora and P. lutea in the South China Sea were distinct from those located in the Great Barrier Reef and in Indo-Pacific reefs. It was observed that different coral species share a small common bacterial community, and the composition of the bacterial communities within each colony of each coral species also showed variation. The coexistence of specificity and uniformity reflects the complexity of coral-associated bacterial community and suggests that corals combine the functional bacterial associates in a subtle and sophisticated manner. This study provides novel insights into the complex structure of coral bacterial associates."
} | 3,903 |
34258158 | PMC8261503 | pmc | 2,887 | {
"abstract": "Abstract High‐performance biodegradable electronic devices are being investigated to address the global electronic waste problem. In this work, a fully biodegradable ferroelectric nanogenerator‐driven skin sensor with ultrasensitive bimodal sensing capability based on edible porcine skin gelatine is demonstrated. The microstructure and molecular engineering of gelatine induces polarization confinement that gives rise the ferroelectric properties, resulting in a piezoelectric coefficient ( d \n 33 ) of ≈24 pC N −1 and pyroelectric coefficient of ≈13 µC m −2 K −1 , which are 6 and 11.8 times higher, respectively, than those of the conventional planar gelatine. The ferroelectric gelatine skin sensor has exceptionally high pressure sensitivity (≈41 mV Pa −1 ) and the lowest detection limit of pressure (≈0.005 Pa) and temperature (≈0.04 K) ever reported for ferroelectric sensors. In proof‐of‐concept tests, this device is able to sense the spatially resolved pressure, temperature, and surface texture of an unknown object, demonstrating potential for robotic skins and wearable electronics with zero waste footprint.",
"conclusion": "3 Conclusion With the aim of producing sustainable and eco‐friendly electronic devices without any e‐waste footprint, we demonstrated a fully biodegradable and ferroelectric gelatine e‐skin nanogenerator with the ability to simultaneously sense temperature, pressure, and surface texture variations. We showed that the physically confined gelatine within the interlocked microdome structure significantly enhanced the polarization and ferroelectric properties. Our gelatine e‐skin nanogenerator displayed excellent sensing capabilities to simultaneously detect and discriminate the temperature and pressure variations with the lowest detection limit ever reported. Proof‐of‐concept demonstrations of our gelatine e‐skin nanogenerator toward healthcare monitoring, simultaneous temperature–pressure mapping, and texture perception verified that it has great potential for applications in environmentally benign wearable sensors, robotic skins, and prosthetic limbs. Furthermore, our gelatine e‐skin nanogenerator, composed of active gelatine extracted from porcine skin and Mg electrodes, was biodegradable and fully decomposed in a body fluid environment within a month, demonstrating an ideal recycling process of decomposable transient electronics with zero waste footprint. Our devices have numerous advantages, such as excellent ferroelectric properties, low‐cost fabrication, edibility, and biodegradability, and are based on a simple yet powerful design strategy. Hence, we expect that these devices could be useful components in current electronic devices (such as sensors, actuators, transistors, memory, and capacitors) and the future internet‐of‐everything devices including wearable, implantable, and decomposable devices that do not generate any pollution or e‐waste.",
"introduction": "1 Introduction Materials with ferroelectricity, a subset of piezo/pyroelectricity, enable various applications including but not limited to non‐volatile memory devices, temperature and tactile sensors, transducers, actuators, and energy harvesters. [ \n \n 1 \n , \n 2 \n \n ] However, traditional inorganic and organic ferroelectric materials rarely meet the requirements of next‐generation electronics considering the biocompatibility, biodegradability, or at least recyclability. Hence, the development of eco‐friendly and sustainable technologies is required. [ \n \n 1 \n , \n 2 \n , \n 3 \n \n ] Most existing electronic devices are non‐decomposable, leading to the generation of large volumes of electronic waste (e‐waste) (53.6 million tonnes in 2019 and an expected 74.7 million tonnes by 2030 worldwide), which is threatening the health of ecological systems. [ \n \n 4 \n \n ] Since e‐waste contains hazardous components, even the recycling of these materials contaminates ecosystems with toxins that can enter living bodies by inhalation, skin exposure, and oral intake of contaminated foods. [ \n \n 5 \n \n ] Hence, transient and ingestible electronics made from completely biodegradable and edible materials are particularly attractive. Such devices can be completely degraded by decomposition and digestion processes after a certain period of use, thus reducing the amount of e‐waste produced. [ \n \n 6 \n \n ] Over the last decade, a range of biodegradable and bioresorbable devices were demonstrated as in vivo sensors and stimulators based on triboelectric effect which shown a new direction toward transient implantable technology. [ \n \n 7 \n , \n 8 \n , \n 9 \n \n ] One potential use of transient electronics is in flexible electronic skin (e‐skin) sensors for wearable devices and robotic skin applications, where ultrasensitive temperature, pressure, and texture perception capabilities are required to monitor the body conditions and surrounding environment. [ \n \n 10 \n \n ] In the pursuit of this goal, the combination of pyro/piezoelectricity is required for highly sensitive, multimodal, and self‐powered functionalities. [ \n \n 11 \n \n ] However, most previously developed multimodal e‐skins are based on non‐decomposable organic/inorganic materials, which continue to contribute to the global e‐waste problems. The development of a biodegradable and edible e‐skin with highly sensitive and multimodal sensing capabilities remains challenging. [ \n \n 12 \n , \n 13 \n \n ] \n Gelatine is a natural biological material with intrinsic biodegradability, low temperature processability, low cost (compared with other biodegradable polymers; Table S1 , Supporting Information), and flexibility. Hence, it is considered as the most promising candidate for the development of biodegradable and eco‐friendly transient electronics systems. [ \n \n 14 \n , \n 15 \n \n ] Gelatine has been extensively used as an edible material in drug delivery systems [ \n \n 16 \n \n ] and recently in electronics and soft robotics. [ \n \n 17 \n \n ] The main challenge of gelatine for e‐skin applications lies in its low pyro/piezoelectric coefficients and instability in humid environments. [ \n \n 18 \n \n ] In this scenario, nature‐inspired engineering design can enhance the functionality of gelatine for high‐performance e‐skins. Of particular interest, an important feature of biological skin is the interlocked micro‐ridge structure of epidermal‐dermal layers. This interlocked structure plays a key role in the precise perception of pressure and temperature through the localization and amplification of stress and temperature within the intermediate ridge tips and their effective transmission to the underlying mechanoreceptors in the dermis layer. [ \n \n 19 \n \n ] Imitating the structure and functionality of human skin, herein, a fully biodegradable gelatine‐based material with high ferroelectric properties is demonstrated using the physical confinement of molecular structures of gelatine within interlocked microstructures. The ferroelectric gelatine e‐skin is self‐powered and highly sensitive to both temperature and pressure stimuli due to its large pyro/piezoelectric coefficients, respectively, and finally biodegraded with zero waste footprint."
} | 1,784 |
38187274 | PMC10770442 | pmc | 2,888 | {
"abstract": "The Internet of Things (IoT) connects devices, enabling real-time data acquisition, automation, and collaboration. Wireless sensor networks are one of the important components of the Internet of Things, consisting of many wireless sensor nodes distributed in space. These nodes can perceive environmental information and transmit it to other nodes through wireless communication. In wireless sensor network routing optimization methods, improved ant colony algorithm can be used to find the optimal routing scheme. Ant colony algorithm simulates the behavior of ants in the process of searching for food, and optimizes factors such as transfer probability and pheromone concentration to enable ants to find the shortest path. In wireless sensor networks, node positions can be used as reference nodes and anchor nodes, combined with the objective function of wireless sensor network routing optimization, and improved ant colony algorithm can be used to solve the optimal path, thus obtaining the optimal wireless sensor network routing optimization scheme. Through experimental results, it can be found that the proposed method performs well in terms of energy consumption, transmission delay, number of dead nodes, and network throughput. These optimization results have positive implications for the sustainable development and practical application of the Internet of Things, which can improve the development of the digital economy and enhance the construction of smart cities.",
"conclusion": "4 Conclusion Wireless sensor network routing optimization is one of the key issues in the Internet of Things, which is of great significance for improving network performance, extending network lifespan, providing reliable data transmission, and supporting real-time applications. The current routing optimization methods have problems such as high energy consumption, high transmission delay, high number of dead nodes, and low network throughput. To address these issues, this article proposes a wireless sensor network routing optimization method based on improved ant colony algorithm in the Internet of Things. Through experimental testing, it has been proven that the proposed method has absolute advantages over the current method in terms of energy consumption, transmission delay, number of dead nodes, and network throughput. This research achievement can not only improve the energy efficiency of wireless sensor networks, but also improve the quality of data transmission and provide better support for real-time applications. At the same time, this method can also optimize network capacity and meet the network performance requirements of different application scenarios. These contributions provide important support for the reliability, intelligence, and efficiency of wireless sensor networks, and promote the further development of wireless sensor network technology in various application fields. However, there are still some challenges in the research of routing optimization in wireless sensor networks. For example, when optimizing routing in a dynamic environment, it is necessary to consider issues such as changes in network topology and the balance between energy consumption and data transmission quality. These challenges require further in-depth research to propose more comprehensive and adaptable routing optimization methods for wireless sensor networks, and to promote the development and application of animal networking technology.",
"introduction": "1 Introduction At present, wireless sensor networks have been widely applied in various fields such as environmental monitoring, agriculture, and the Internet of Things. In these applications [ 1 , 2 ], wireless communication between nodes in wireless sensor networks plays a crucial role. The Internet of Things has characteristics such as connectivity, intelligence, and real-time. It connects devices through wireless communication technology and endows them with perception, computing, and decision-making capabilities. This enables the Internet of Things to obtain and transmit data in real-time, automate collaboration and data sharing, improve production and operational efficiency, and reduce costs. At the same time, the Internet of Things utilizes big data analysis and predictive maintenance to provide insight into market trends and user needs, helping enterprises make more accurate decisions. In addition, the application of the Internet of Things in fields such as smart homes and health monitoring is improving people's quality of life and promoting the process of sustainable development. Therefore, the Internet of Things has significant advantages in improving efficiency, reducing costs, enhancing insight, improving quality of life, and promoting sustainable development. The routing optimization of wireless sensor networks has become an important and challenging issue due to the limited energy of sensor nodes, wide and random distribution of nodes, and other characteristics. The goal of wireless sensor network routing optimization is to improve network performance and energy utilization efficiency by selecting appropriate paths, minimizing energy consumption, maintaining low latency, and maximizing network lifespan [ 3 ]. It involves techniques and mechanisms such as path selection and transmission scheduling to ensure efficient and accurate data transmission between nodes. The research on routing optimization in wireless sensor networks is of great significance for improving network performance, extending network lifespan, providing reliable data transmission, and supporting real-time applications. By adopting effective routing optimization algorithms and mechanisms, more reliable, intelligent, and efficient wireless sensor network systems can be achieved. Wireless sensor network routing optimization plays an important role in the Internet of Things, as it can improve network performance and energy utilization efficiency, extend network lifespan, provide reliable data transmission and support real-time applications for various application fields. Therefore, routing optimization in wireless sensor networks is of great significance in promoting the development and application of animal networking [ 4 , 5 ]. Many research work has been carried out on the optimization of wireless sensor network routing. For example, reference [ 6 ] proposed a routing optimization method for wireless sensor networks based on grey prediction. Firstly, the characteristics of wireless sensor networks and the background of routing optimization problems were introduced, and the principle and application of grey prediction method were elaborated in detail. Then, a routing optimization method for wireless sensor networks based on grey prediction was proposed, which mainly utilizes the established grey prediction model to predict the status and communication load of nodes in the network, and makes routing decisions and optimizations based on the prediction results. Select nodes with lower load and energy consumption on the communication path as relay nodes to achieve network load balancing and energy conservation, in order to achieve the goal of optimizing network routing. However, after the application of this method, the network transmission delay increases. Reference [ 7 ] proposed a routing optimization method for wireless sensor networks based on energy and path constraints. Firstly, the energy constraints between nodes are considered, and through the use of energy balance strategies, the energy resources of nodes are reasonably allocated and managed to avoid situations where energy depletion leads to node failure or inability to complete tasks. Secondly, this method combines path constraints and considers factors such as distance between nodes and communication consumption in the optimization routing process to select the optimal transmission path. Energy and distance weights are added to ensure a uniform and reasonable energy distribution between network nodes, thus achieving wireless sensor network routing optimization. However, during testing, it was found that there were a large number of node deaths in the application of this method, which still lags behind the ideal application goal. Reference [ 8 ] proposed a wireless sensor network routing optimization method based on an improved grey wolf optimization algorithm. Initialize a group of virtual grey wolf representative nodes in a wireless sensor network and give them random position, speed, and fitness values. By simulating the search behavior of grey wolves, the optimization algorithm updates the speed and position of each grey wolf based on its position in the current solution space, and calculates its fitness value. Define a fitness function based on the routing optimization objective to evaluate the quality of each grey wolf's solution. In the process of updating the location of the grey wolf, combining competition strategy and cooperation mechanism, the grey wolf can conduct local and global searches based on the information of neighboring nodes to obtain better optimization solutions. Applying this method to practical applications, it was found that this method has the problem of low network throughput, and the actual application effect is not good. Wireless sensor network routing optimization faces problems such as energy constraints, network topology dynamics, bandwidth and capacity limitations. Improving ant colony algorithm can help solve these problems [ 9 , 10 ]. In wireless sensor network routing optimization, improving ant colony algorithm can improve the performance and effectiveness of wireless sensor network routing optimization. However, unlike the improved approach of introducing the alternating search strategy of regional ants and public ants in Ref. [ 10 ] and limiting the number of candidate nodes for public ant search, in this study, the ant colony algorithm was improved by optimizing the transfer probability and pheromone concentration. Optimizing the transition probability can effectively guide ants to better choose paths during the search process. By adjusting the concentration of pheromones, the importance of different paths can be dynamically adjusted during the search process. The method of optimizing pheromone concentration can make ants more intelligent in updating pheromones and guide other ants to pay more attention to passing through high-quality paths in subsequent searches. In this way, the optimized ant colony algorithm can more accurately find better solutions and reduce the exploration of invalid paths. This algorithm can fully consider energy constraints, dynamic network topology, and resource constraints, and can achieve self-adaptability and distributed properties during the search for the optimal path. Therefore, a new routing optimization method for wireless sensor networks based on improved ant colony algorithm is proposed. The innovative technology route is as follows. (1) Edge computing and intelligent perception: push computing power to IoT nodes to enable them to have intelligent perception and decision-making capabilities. This allows for real-time data processing and analysis on the device side, reducing the pressure of data transmission and cloud computing, and improving response speed and energy efficiency. (2) Artificial intelligence and machine learning: Apply artificial intelligence and machine learning to IoT systems, enabling them to have autonomous learning and adaptability. By extracting patterns from massive data, discovering patterns, and automatically adjusting and optimizing the system, intelligent decision-making, automated management, and predictive maintenance functions are achieved. (3) Security and privacy protection: With the increasing number of IoT devices and data, security and privacy have become important considerations. The innovative technological path should include strengthening the security of IoT devices and communication networks, adopting data collection and processing methods that comply with privacy protection standards, and promoting technological innovation in identity verification, encryption, and vulnerability repair. (4) Cloud edge collaboration and resource optimization: By optimizing the collaborative work between cloud and edge devices, efficient utilization of resources and distributed computing can be achieved. This can reduce network transmission latency and provide a better user experience and flexibility. (5) Integrated application and scenario innovation: Promote the integration of animal networking with other fields, such as smart cities, healthcare, industrial manufacturing, etc. By innovating applications and scenarios, the Internet of Things can be widely applied in different fields, further improving efficiency, reducing costs, and contributing to sustainable development."
} | 3,207 |
35336196 | PMC8953468 | pmc | 2,890 | {
"abstract": "Modern temperate alley-cropping systems combine rows of trees with rows of crops (agroforestry), which allows for diverse interspecific interactions such as the complementary and competitive use of resources. The complementary use of resources between trees and crops is considered the main advantage of these multifunctional land use systems over cropland monocultures. Moreover, several studies demonstrated that agroforestry systems are environmentally more sustainable than cropland monocultures. Over two decades of research on soil microorganisms in temperate alley-cropping systems are characterized by a variety of different methodological approaches and study designs to investigate the impact of agroforestry on the soil microbiome. Here, we review the available literature on the abundance, diversity, and functionality of soil microorganisms in temperate alley-cropping systems. Further, we identify current knowledge gaps as well as important experimental factors to consider in future studies. Overall, we found that temperate alley-cropping systems increase soil microbial abundance, diversity, and functions as compared to cropland monocultures, which is expected to contribute to enhanced biological soil fertility in these systems.",
"conclusion": "8. Conclusions Our review revealed that tree rows in temperate alley-cropping agroforestry systems increase soil microbial population sizes, and that this beneficial effect can extend gradually into the crop rows. Additionally, fungi may benefit more than bacteria, as several studies indicated an increase in the fungi:bacteria ratio. Agroforestry systems increase soil microbial diversity mainly through the establishment of a tree–row association microbiome that is compositionally distinct from the microbiome of the crop rows (i.e., increased beta diversity within agroforestry systems) rather than increased alpha diversity. Furthermore, soil microbial populations in the tree rows of agroforestry system have greater catabolic potential and substrate-use efficiency and are functionally more diverse than those in the neighboring crop rows or in cropland monoculture systems. Microbial communities in soil of agroforestry systems are likely shaped by multiple factors (e.g., soil management, plant species, and cultivation cycles) that act simultaneously but to varying degrees across space and time. To disentangle the effects of these factors on soil microbial communities, future studies require study designs adapted to the complexity of these systems. Overall, temperate alley-cropping agroforestry systems increase soil microbial abundance, diversity, and functionality as compared to cropland monocultures, which is expected to contribute to enhanced biological soil fertility in these systems.",
"introduction": "1. Introduction Agroforestry systems are agricultural systems that combine trees or other woody plants with crops or livestock. Several different systems fall under the umbrella of agroforestry. Temperate agroforestry systems include, among others, alley-cropping systems, shelterbelts, orchard meadows, and forest pastures. In the temperate zone, alley-cropping systems are gaining popularity as they can maintain or even increase production while being environmentally more sustainable than cropland monocultures [ 1 , 2 ]. Modern alley-cropping systems in the temperate zone combine fast-growing trees (e.g., poplar, willow) with annual crops (e.g., wheat, maize, and oilseed rape). The tree and crop components of these systems are often arranged in alternating alleys with North–South orientation to minimize shading. The spatial proximity of the trees and crops allows for a variety of interspecific interactions in these systems such as competition for resources as well as complementary resource use [ 3 ]. Over 25 years ago, Cannell et al. [ 4 ] argued that yield benefits of agroforestry systems only occur if resources are used in a complementary manner by the trees and crops—a hypothesis dubbed the “central hypothesis of agroforestry”. Twenty-five years later, a large body of literature has shown that the benefits of complementary resource use can outweigh the disadvantages of resource competition in agroforestry systems. The most prominent example of complementary resource use is the altered soil–nutrient (re)cycling in agroforestry systems. In 2004, Allen and coworkers [ 5 ] found evidence that trees in temperate agroforestry systems can reduce nutrient leaching by taking up leachable nutrients that were not utilized by the crops. These otherwise lost nutrients are incorporated into the biomass of the trees and will partly reenter the soil through tree litter. This recycling of nutrients has been described as ‘nutrient pumping’ [ 6 ]. The supply of nutrients through tree litter is likely to account for increased soil fertility close to the trees [ 7 ]. In addition to their function as nutrient pumps, trees effectively reduce wind speed [ 8 ] and, thus, can contribute to reduced soil erosion. Furthermore, trees can also decrease arthropod pests, increase the abundance of natural enemies, e.g., [ 9 ], and enhance pollination services, e.g., [ 10 ]. Unsurprisingly, the integration of trees through agroforestry also increases biodiversity, e.g., [ 11 ], and C sequestration of agroecosystems, e.g., [ 12 ]. In addition to environmental benefits, the woody biomass of the trees serves as an additional product for farmers [ 13 ]. This economic diversification may help to compensate for fluctuations in crop yields and market prices and, thus, lowers risks in crop production [ 14 ]. Furthermore, temperate agroforestry can be more profitable than cropland monocultures [ 15 ], especially when negative externalities are considered [ 16 ]. Although farmers recognize and value the benefits of agroforestry, many are deterred by the increased labor and work complexity [ 17 , 18 ]. In this context, it was suggested that agroforestry practice would become more readily adopted by farmers if promoted and supported appropriately [ 17 ]. Apart from all the benefits, crop yields have been shown to decrease close to the trees as a result of resource competition [ 19 , 20 , 21 , 22 ]. While there is initial evidence that long-term yields in alley-cropping systems may be comparable to those in cropland monocultures [ 22 ], yield depressions in close proximity to the trees are inevitable and increase with tree height. Other than yield, crops cultivated in agroforestry systems are as healthy as crops cultivated in cropland monocultures [ 23 ]. Plant health and productivity as well as nutrient cycling are strongly driven by soil organisms, especially microorganisms, e.g., [ 24 , 25 ]. In recent decades, several soil organisms have been investigated in temperate agroforestry systems, ranging from soil macrofauna, e.g., [ 26 ], to microorganisms, e.g., [ 27 ]. While the influence of agroforestry systems on soil fauna and their functions has extensively been reviewed recently by Marsden et al. [ 28 ], the last review on soil microorganisms and their functions in agroforestry systems was published over one decade ago with little focus on temperate systems [ 29 ]. Since then, many new studies from temperate systems have been published, especially due to the rapid development of new molecular tools. In 2016, Banerjee and colleagues [ 30 ] published the first study that investigated soil bacterial communities in temperate agroforestry systems using 454 pyrosequencing. From 2019 onwards, several studies that used Illumina sequencing to profile microbial communities were published [ 31 , 32 , 33 , 34 ]. The work of Banerjee and colleagues [ 30 ] was also the first to quantify soil microbial groups in temperate agroforestry systems by using real-time PCR. As for amplicon sequencing, several studies that employed real-time PCR have been published since 2019, e.g., [ 35 , 36 , 37 , 38 ]. Additionally, various studies that used other non-molecular methods have been published in the last ten years, e.g., [ 27 , 39 , 40 ]. All these studies greatly enhanced our understanding of soil microbial communities and their functions in temperate agroforestry systems. In this work, we review the available literature on the abundance, diversity, and function of soil microorganisms in alley-cropping agroforestry systems of the temperate zone. Based on our literature review, we provide recommendations for future studies of soil microorganisms in these systems."
} | 2,115 |
39957275 | PMC11831097 | pmc | 2,893 | {
"abstract": "Abstract Marine foundation species are critical for the structure and functioning of ecosystems and constitute the pillar of trophic chains while also providing a variety of ecosystem services. In recent decades, many foundation species have declined in abundance, sometimes threatening their current geographical distribution. Kelps (Laminariales) are the primary foundation species in temperate coastal systems worldwide. Kelp ecosystems are notoriously variable, challenging the identification of key factors controlling their dynamics. Identification of these drivers is key to predicting the fate of kelp ecosystems under climatic change and to informing management and conservation decisions such as restoration. Here, we used in situ data from long‐term monitoring programs across 1350 km of coast spanning multiple biogeographic regions in the state of California (USA) to identify the major regional drivers of density of two dominant canopy‐forming kelp species and to elucidate the spatial and temporal scales over which they operate. We used generalized additive mixed models to identify the key drivers of density of two dominant kelp species (Nereocystis luetkeana and Macrocystis pyrifera ) across four ecological regions of the state of California (north, central, southwest, and southeast) and for the past two decades (2004–2021). The dominant drivers of kelp density varied among regions and species but always included some combination of nitrate availability, wave energy and exposure, density of purple sea urchins, and temperature as the most important predictors. These variables explained 63% of the variability of bull kelp in the northern and central regions, and 45% and 51.4% of the variability in giant kelp for the central/southwest and southeast regions, respectively. These large‐scale analyses infer that a combination of lower nutrient availability, changes in wave energy and exposure, and increases in temperature and purple sea urchin counts have contributed to the decline of kelp observed in the last decade. Understanding the drivers of kelp dynamics can be used to identify regional patterns of historical stability and periods of significant change, ultimately informing resource management and conservation decisions such as site selection for kelp protection and restoration.",
"conclusion": "Conclusion Understanding how biotic and abiotic factors contribute to the temporal variation in the abundance of foundation species is a major focus of ecology and biogeography and is key to understanding biodiversity in a changing climate. Our results demonstrate the benefits of combining long‐term in situ monitoring data that provide information about species interactions (not yet obtainable through remote‐sensing) with remote‐sensing datasets to interpret population modeling results. Using both in situ and remote‐sensing data to understand the dynamics of surface canopy kelps allowed for the validation of the robustness of the model outputs. The maps produced from these robust models provide valuable information for managers and stakeholders about the locations that are more likely to support healthy kelp ecosystems and the functional relationships identified, form a basis for future studies focused on the spatiotemporal dynamics of kelp forests under a changing climate (Giraldo‐Ospina, Bell, et al., 2023 ).",
"introduction": "INTRODUCTION Of the many species whose distributions and dynamics are exhibiting dramatic changes in response to a changing global climate, few are as important and concerning as foundation species ( sensu Dayton, 1972 ; Ellison et al., 2005 ). These species support ecosystem productivity, create structural habitat, act as ecological engineers ( sensu Jones et al., 1994 ), and underpin a multitude of ecosystem services. In coastal marine ecosystems, foundation species such as kelps (Graham et al., 2007 ; Steneck et al., 2002 ), corals (De'ath et al., 2012 ; Hoegh‐Guldberg et al., 2007 ; Hughes et al., 2003 ), seagrasses (Koch et al., 2013 ; Serrano et al., 2021 ; Short & Neckles, 1999 ), mangroves (Alongi, 2015 ; Ward et al., 2016 ), and oysters (Beck et al., 2011 ), among others, are experiencing marked changes in distribution and dynamics in response to gradual and episodic changes in water temperature (e.g., marine heatwaves, MHW) and other anthropogenic stressors. These spatial and temporal responses in such population attributes as abundance, productivity, and demographic and genetic structure are complex because of the interactions among multiple simultaneously changing environmental and ecological drivers (e.g., Crain et al., 2008 ; Hewitt et al., 2016 ). This complexity is compounded by the many spatial (local, regional, global) and temporal (seasonal, interannual, decadal) scales over which environmental conditions vary and interactions among scale‐specific sources of variation (de Amorim et al., 2023 ; Gunderson et al., 2016 ). Consequently, one of the most challenging goals in ecology and conservation biology is to elucidate the spatial and temporal relationships between species abundance and variable environmental and ecological conditions. Such relationships are central to explaining and predicting variability in the distribution and dynamics of foundation species and the ecosystems they create. In light of the accelerating effects of climate change, there is an urgent need to identify the key drivers of geographical and temporal variation in foundation species in order to make effective management decisions that protect them and the ecosystem services they provide. Globally, kelps (Laminariales) are among the most important foundation species in temperate coastal oceans, inhabiting shallow temperate rocky reefs throughout the world (Assis et al., 2020 ; Bolton, 2010 ; Eger et al., 2023 ; Graham et al., 2007 ; Steneck et al., 2002 ). Like many other foundation species, their primary production both fuels food webs and creates physical habitat structure for highly species‐rich ecosystems. These productive ecosystems support a multitude of culturally and economically important fisheries species, among other provisioning, regulating, supporting, and cultural services (Eger et al., 2023 ; Vásquez et al., 2014 ). With declines of kelp forests in many parts of the world associated with gradual and episodic (MHW) increases in water temperature and other anthropogenic stressors (Arafeh‐Dalmau, Cavanaugh, et al., 2021 ; Beas‐Luna et al., 2020 ; Krumhansl et al., 2016 ; Wernberg et al., 2016 ), there is global interest in conservation and restoration of kelps and their associated ecosystems (Eger et al., 2022 ; Morris et al., 2020 ). From Baja California, Mexico, to southwest Alaska, USA, the dominant canopy‐forming kelps are the bull kelp ( Nereocystis luetkeana ) and the giant kelp ( Macrocystis pyrifera ), both of which create large, floating canopies (Carr & Reed, 2016 ; Graham et al., 2007 ; Schiel & Foster, 2015 ). Over the past decade, both bull and giant kelp forests on the West coast of North America experienced two major disturbances: the 2013 sea star wasting disease that led to local extinction of a key sea urchin predator, the sunflower star ( Pycnopodia helianthoides ), and the Northeast Pacific MHW of 2014–2016 (NE Pacific MHW) (Di Lorenzo & Mantua, 2016 ; Michaud et al., 2022 ). These events were rapidly followed by extensive loss (>90%) of bull kelp along the coast of northern California (McPherson et al., 2021 ; Rogers‐Bennett & Catton, 2019 ) and substantial losses of giant kelp in central California (Smith et al., 2021 , 2024 ) and Baja California, Mexico (Arafeh‐dalmau et al., 2019 ; Beas‐Luna et al., 2020 ). The loss of kelp in northern California drove closures of important recreational and commercial fisheries such as red abalone and red sea urchins, and critically, almost a decade later, these lost forests have yet to recover (McPherson et al., 2021 ; Rogers‐Bennett & Catton, 2019 ). The detrimental consequences of the widespread loss of kelp have given rise to an urgent need to better understand the drivers of kelp dynamics in order to optimize decisions about conservation and restoration of this important marine habitat. In California, the geographic distributions of these two kelp species extend across two well‐recognized biogeographic regions distinguished by persistent differences in oceanographic conditions (e.g., water temperature, wave energy, and coastal upwelling) (Blanchette et al., 2008 ; Briggs, 1974 ; Horn et al., 2006 ; Reed et al., 2011 ). This environmental variability generates geographic differences in the structure and dynamics of kelp forest communities (Carr & Reed, 2016 ), making California an ideal place to evaluate the drivers of climate‐induced losses, gains, and distributional shifts. To further advance our understanding of the environmental and ecological variables that explain the distribution and dynamics of canopy‐forming kelps, including observed changes before and after the NE Pacific MHW, we ask the following questions: (1) What environmental and biological variables best explain and predict the density dynamics of bull kelp and giant kelp along the coast of California? (2) How do the relative and combined impact of these variables on the spatiotemporal dynamics of kelp density vary across the bioregions of the California coast? and (3) How well can these variables explain and predict regional dynamics of kelp (canopy) abundance detected by satellite imagery? To answer these questions, we created species distribution models (SDMs) of kelp densities from long‐term data, which allowed us to construct a coast‐wide time series of interannual kelp density dynamics. We used these results to inform resource managers about which sites are more likely to support dense forests in the face of future MHWs, and this knowledge can be used to prioritize sites for protection and restoration while considering the broad range of responses across a large geographical area (Giraldo‐Ospina, et al., 2023 ).",
"discussion": "DISCUSSION Kelp forests are dynamic ecosystems that naturally experience great variability across space and time, yet, globally and in California, they have become increasingly threatened by multiple stressors that are exacerbated by climate change (Arafeh‐Dalmau, Brito‐Morales, et al., 2021 ; Krumhansl et al., 2016 ; Wernberg et al., 2016 ). Indeed, many regions have lost all or much of their kelp forests and are struggling with management and restoration decisions to stem further losses and/or rebuild populations (Butler et al., 2020 ; Hynes et al., 2021 ; Miller et al., 2023 ; Rogers‐Bennett & Catton, 2019 ). The first step to developing management actions to ensure the persistence of kelp forest ecosystem functioning and services under a changing climate is to understand the key environmental and ecological factors that shape both temporal and spatial patterns of abundance. Kelps are unlike many terrestrial forest systems, being highly variable in time and space. Here we have identified the key environmental and biological drivers of two dominant surface canopy‐forming kelp species (bull kelp and giant kelp) across the 1350‐km coast of California. The results from our models are robust as the drivers identified in this approach were useful in reconstructing regional patterns of historical kelp density from remotely sensed canopy cover. However, we also identified important mismatches in the predicted dynamics resulting from the two data sources, that arise from ecological processes (e.g., grazing by sea urchins) which are not captured by remote‐sensing yet are critical to kelp population dynamics. The results from this study highlight the importance of long‐term in situ monitoring surveys of complex ecosystems such as kelp forests (Hughes et al., 2017 ; Magurran et al., 2010 ), as well as the role for remote‐sensing to fully understand population dynamics (Cavanaugh et al., 2021 ) which are key for effective management of these species and the ecosystems they sustain. Drivers of bull kelp distribution and dynamics The best model we identified for bull kelp covered both the north and central coasts combined, indicating that the drivers of bull kelp density were the same for the entire extent of its distribution in the state of California. Increasing wave height had the largest effect on bull kelp density, as expected for a species that dominates exposed coastlines such as found in northern California (Carr & Reed, 2016 ; Springer et al., 2010 ). In central California, bull kelp commonly occurs in mixed beds with giant kelp where it thrives in the shallower areas that experience breaking waves as evidenced by its higher density in areas with increased wave heights compared with giant kelp (Springer et al., 2010 ). Higher water motion, brought on by waves may aid in the uptake of nutrients and carbon (Hurd, 2000 ; Koehl & Alberte, 1988 ), increase irradiance by pushing the fronds in different directions (Koehl & Alberte, 1988 ), and enhance blade production (Breitkreutz et al., 2022 ). However, an upper threshold of water motion for bull kelp appeared to be reached at maximum orbital velocities higher than ~5 m/s, beyond which density was reduced, likely the result of the ripping of kelp stipes from increased drag in high flow speeds (Johnson & Koehl, 1994 ). Bull kelp inhabits areas of significant coastal upwelling, which delivers cold and nutrient‐rich water to coastal areas (Springer et al., 2010 ). Temperature has been found to be an important factor affecting growth and performance of bull kelp on the Pacific coast (Fales et al., 2023 ; Korabik et al., 2023 ; Weigel et al., 2023 ). The optimal growing temperature for adult and early life stages of bull kelp in Canada and Washington state has shown to be ~10–16°C with an upper thermal limit between 18 and 20°C (Supratya et al., 2020 ; Weigel et al., 2023 \n ) . For California we found, similarly, that bull kelp density increased at mean upwelling season temperatures lower than 12°C. During the upwelling season SST is negatively associated with nutrient availability, indicating that nutrient limitation during this time is usually what drives the interannual dynamics of bull kelp (García‐Reyes et al., 2022 ; McPherson et al., 2021 ). Further evidence for the importance of nutrient limitation to bull kelp dynamics was the positive relationship between bull kelp density and minimum annual nitrate which indicated that the higher the nitrate concentration from the physiological thresholds of bull kelp, the better it does in general. Increasing SST and lower nutrient levels combined with an increase in herbivory levels have been shown to be important drivers in the collapse of bull kelp populations after the NE Pacific MHW (McPherson et al., 2021 ). Densities of purple sea urchins spiked after the NE Pacific MHW, potentially due to the combination of a lack of top‐down control of sunflower sea stars on herbivorous sea urchins (McPherson et al., 2021 ) and anomalously high settlement of sea urchins around the Fort Bragg region between 2013 and 2015 resulting in high recruitment (Okamoto et al., 2020 ). Densities of bull kelp did not show a response at low urchin densities but declined dramatically at urchin densities greater than ~4.5 log urchins/60‐m 2 transect (equivalent to about 1.5 urchins/m 2 ). This parallels previous findings that show urchin densities of about 0.1 to 1.7 urchins/m 2 before 2013, which increased up to 60‐folds during the NE Pacific MHW and persisted through the collapse of bull kelp coverage in northern California (Rogers‐Bennett & Catton, 2019 ). Importantly, after the NE Pacific MHW, our model predicted a recovery of bull kelp in 2018 and 2021, when our models predicted a decline in urchin populations and favorable abiotic conditions for kelp had returned. However, this predicted recovery of bull kelp abundances was not evident in satellite imagery or in situ diver surveys in the region (Cavanaugh et al., 2023 ). This result may be evidence of hysteresis in the system driven by sea urchin overgrazing (Filbee‐Dexter & Scheibling, 2014 ). Net primary productivity was another factor associated with bull kelp density dynamics. High values of NPP may indicate that kelp is competing for nutrients and light with phytoplankton in the water column (Kavanaugh et al., 2009 ), explaining why NPP values higher than 2500 mg C m −2 day −1 resulted in a decline in bull kelp. Evidence of competition for nutrients at large spatial scales has been shown in other kelps, especially during high‐nutrient conditions driven by interannual climatic oscillations (Dayton et al., 1999 ). Densities of bull kelp declined linearly with depth likely due to light attenuation (Goldberg & Kendrick, 2004 ). In bull kelp, light is the most important factor in the development of gametophytes and sporophytes and allows sporophytes to reach sexual maturity (Vadas, 1972 ). We also found that grazers like sea urchins were less abundant in shallower areas, potentially due to higher wave energy in these areas (Duggins et al., 2001 ). The combination of higher light levels and lower grazer density potentially creates an optimum environment in the shallows for bull kelp to thrive. Drivers of giant kelp distribution and dynamics Unlike bull kelp, two models were required to best describe the distribution of giant kelp in California and were generally consistent with different oceanographic conditions between the central and southern coasts. These domains are not defined by the typical biogeographic limit of Point Conception (Blanchette et al., 2008 ; Claisse et al., 2018 ; Hamilton et al., 2010 ), but rather by an approximate midpoint in Santa Barbara Channel, which has been shown to be the limit between a system with more nutrient availability and lower SSTs driven by upwelling conditions to the west of the Santa Barbara Channel, including Santa Rosa and San Miguel islands (Broitman et al., 2005 ; Claisse et al., 2018 ; Gosnellet et al., 2014 ; Hamilton et al., 2010 ). Large waves are one of the most important causes of giant kelp mortality in California (Dayton et al., 1984 ; Reed & Foster, 1984 ) and wave height was an important driver of kelp density in our models, but the effect was different in the two regions. In the central/southwest coast, which naturally experiences larger waves than the southeast coast (Reed et al., 2011 ), kelp density was resilient to mean annual wave heights up to approximately 1 m, while in the southeast coast, kelp density was the highest at lower wave heights of approximately 0.75 m. This regional difference in wave disturbance has been shown to determine giant kelp net primary productivity in California (Castorani et al., 2022 ; Reed et al., 2011 ). Stronger and more frequent wave events in the central/southwest coast may explain why orbital velocity was a significant factor negatively correlated with kelp density in that region but not in the calmer, southeast coast. In shallow waters, the horizontal motion of a wave may be up to five times greater than the wave height, creating drag forces which, when strong enough, break stipes or remove holdfasts (Seymour et al., 1989 ). Previous studies in central California have found wave orbital velocity to be the most frequent disturbance responsible for tearing out plants (Graham, 1997 ) and giant kelp in the region can occur across a range of wave orbital velocities but is most abundant in a moderate wave environment (~ 0.86 m/s) (Young et al., 2016 ). The nutrient environment varies widely along the distribution of giant kelp in California, which covers approximately 10° of latitude. Central California generally has a more stable and consistent nutrient supply through upwelling due to the exposed coastline and narrow continental shelf than the south coast (Huyer, 1983 ; Zimmerman & Kremer, 1984 ). Here, we add empirical evidence suggesting that kelp populations are adapted to local nutrient conditions, a process which has been observed in a number of marine species (Bennett et al., 2015 ; Howells et al., 2011 ; Sanford & Kelly, 2011 ). Interestingly, in central/southwest California, kelp density was positively related to the number of days above 4 μmol L −1 while in the south coast, the maximum nitrate anomaly during the summer season was the more important predictor, and was negatively associated with kelp growth. Positive growth of giant kelp in southern populations has been shown to occur at extremely low nitrate concentrations (less than 1 μmol L −1 ), while in central California a positive effect on giant kelp biomass is only visible past the 4 μmol L −1 (Kopczak et al., 1991 ). During the summer, declines in giant kelp are usually related to the reduction in nutrient availability (Gerard, 1982 ; Jackson, 1977 ; North & Zimmerman, 1984 ), with the principal nutrient source during summer and fall through internal wave propagation (Zimmerman & Kremer, 1984 ). The negative relationship between maximum nitrate anomaly in the summer months and kelp density that we found in the south coast is hence unexpected. This may be explained by low rates of frond production during summer, which may not be able to keep up with natural frond loss (Zimmerman & Robertson, 1985 ). Previous work has found that frond dynamics are better explained by intrinsic biological processes, such as frond age, rather than external environmental conditions (Rodriguez et al., 2013 ), indicating that kelp senescence, which, in southern California occurs in summer, can overpower external environmental conditions (Bell & Siegel, 2022 ). As with nutrient availability, there is also evidence of adaptation to thermal stress in the microscopic reproductive stages of giant kelp (Hollarsmith et al., 2020 ). A strong inverse relationship between SST and nitrate availability is known to exist in California (Zimmerman & Robertson, 1985 ) with warm sea temperatures typically associated with reduced upwelling and nutrient limited conditions. Although the number of days above 21°C was selected as an important driver for kelp density in central and southern California, we found it had a small effect in both regions. Previous studies have also found SST metrics not to be significantly related to the resilience of giant kelp to warming events (Cavanaugh et al., 2019 ), indicating that nutrients and other local processes are more determinant at driving kelp populations than temperature, both in central and southern California. We also found NPP to be an important factor explaining kelp density, with a small negative effect on kelp density overall. Net primary productivity (NPP) can indicate the levels of plankton abundance, and extensive blooms of phytoplankton can reduce light intensity (Kavanaugh et al., 2009 ) especially during upwelling periods (Strub & Powell, 1987 ) when kelp is uptaking nutrients and growing. This could result in competition for both light and nutrients between phytoplankton and the kelp recruits, explaining the negative effect of increasing NPP. Competition for nutrients between giant kelp and other species has been shown to be more noticeable in shallow depths and during large‐scale low‐frequency events that drive nutrient‐rich conditions (such as La Niña), as these drive surface nutrient availability and have long‐term influence on the surface canopy of giant kelp (Dayton et al., 1999 ). Both the density and foraging behavior of grazers have been shown to be an important factor determining the distribution and biomass density of kelp beds (Bell et al., 2015 ; Filbee‐Dexter & Scheibling, 2014 ; Johnson et al., 2011 ; Young et al., 2023 ). Similar to studies in other regions (Balemi & Shears, 2023 ; Filbee‐Dexter & Scheibling, 2014 ; Ling & Keane, 2024 ), we found that giant kelp declined in both central/southwest and southeast regions of California when densities of sea urchins exceeded approximately 2.5 urchins/m 2 (5 log urchins/60‐m 2 transect). However, a previous study of giant kelp abundance in the central coast of California identified some sites with a positive relationship between urchins and kelp (Bell et al., 2015 ). Indeed, under some circumstances, high density of giant kelp can coexist with high densities of grazers (Karatayev et al., 2021 ; Randell et al., 2022 ; Rennick et al., 2022 ). However, this was not the case for giant kelp in central and northern California following the NE Pacific MHW, where urchins increased dramatically to previously undocumented levels (Rogers‐Bennett & Catton, 2019 ). This increase in grazing combined with increased physiological stress in the kelp caused by high SST and low nutrient levels may explain the post‐MHW decline in giant kelp, which although not as dramatic as for bull kelp in the north coast, was more evident in the central/southwest, than the southeast region. Demographic connectivity is key to the persistence of kelp metapopulations; however, it is highly variable due to spore supply and dispersal (Castorani et al., 2015 ; Hanski, 1998 ). Previous findings in southern California found that temporal variation in fecundity (i.e., spore supply) had a larger effect on the persistence and recovery of giant kelp beds than variation in the physical transport of spores (Castorani et al., 2017 ). Here, we used a spore supply metric that accounted for annual variation based on the maximum kelp biomass from the previous year (a proxy for spore production) and found that it was one of the most important drivers of giant kelp density in both regions. While this might also be the case for bull kelp, we currently do not have a relationship between adult biomass and spore production, or dispersal information for that species. Depth is known to be one of the most important factors affecting kelp growth and recruitment as it is correlated with light availability (Gerard, 1984 ). Our models found that giant kelp densities declined below depths of 15 m in both regions, with a stronger negative effect in the southeast coast. In addition to depth, light attenuation beneath the kelp canopy can be reduced up to 99% at 20 m of depth (Dean, 1985 ). Such low irradiances may reduce kelp density by limiting recruitment, as microscopic stages of kelp have high light requirements and are vulnerable to intraspecific competition through shading (Stewart et al., 2009 ). Additionally, the growth and recruitment of kelps are dependent on sufficiently cold, nutrient‐rich seawater provided by upwelling (Deysher & Dean, 1986 ). During the NE Pacific MHW, there was a significant warming and reduction in nitrate concentration within the mixed layer and an overall deepening of the nitracline across the coast of California (Landry et al., 2024 ). These processes likely decreased the available nutrients to kelp across all reef depths. Comparison with Landsat observations Our comparison of the regional model predictions of in situ kelp density with canopy cover estimated from Landsat indicated that dynamics from the two sources match quite well for both kelp species. Thus, the kelp dynamics model performed well in predicting the dynamics of forest density of both canopy‐forming species and provided confidence in spatially projecting forest stability of both kelps throughout their California ranges. But the comparison also revealed a critically important limitation of the model. The bull kelp density dynamics model predicted the recovery of forests in the years following the NE Pacific MHW whereas Landsat did not detect a recovery. This mismatch is likely due to a hysteresis effect that the modeling approach did not capture, where abundances of kelp did not track abundances of urchins after the MHW as they did before the MHW. Persistent overgrazing of macroalgae by urchins has been known to shift algal ecosystems into an alternative stable state from which it is challenging to recover because although high urchin densities are required to overgraze kelp, much lower densities are required to maintain the deforested, urchin barren state (Filbee‐Dexter & Scheibling, 2014 ; Ling et al., 2019 ). One study found that the biomass of urchins required to tip a kelp forest into a barren state is one order of magnitude higher than that required to maintain the urchin barren state (Ling et al., 2015 ). As such, the ecosystem shift into this alternative state can be difficult to reverse, even when abiotic conditions are conducive for forest recovery, and such change in ecological relationships is not adequately captured by the model. In the case of northern California, further analyses of urchin densities from in situ surveys are required to confirm such change in the kelp‐urchin relationship, once again reinforcing the necessity to couple in situ and remote‐sensing monitoring data to inform the interpretation of ecological processes in modeling results. Implications for understanding future disturbances and management of kelp forests Identifying the biotic and abiotic factors that drive spatial and temporal variation in abundance and condition of foundation species is key for successful management of biodiversity under a changing climate. Our kelp density models quantified the functional relationships between various environmental and ecological drivers and kelp. These models explained and predicted spatial and temporal variation in the two kelp species quite well across markedly different geographic regions. However, the bull kelp model failed to predict the persistence of a deforested state as abiotic conditions were favorable for kelp. In this case, ecological interactions, such as overgrazing by sea urchins, superseded the ability of kelps to naturally recover. This result reinforces the importance of coupling multiple sources of data, including in situ and remotely sensed, to predict future dynamics. It also highlights the need to develop models that incorporate shifts in ecological interactions, such as that between urchins and kelp (e.g., Arroyo‐Esquivel et al., 2023 ; Karatayev et al., 2021 ). While we expect that the environmental drivers identified in this study will vary in strength and spatial distribution with future climate change, we also assume that the functional relationships between kelp growth or loss and those drivers will remain constant in the near future (years to decades). If the assumption holds, then the models created here, when coupled with knowledge or predictions of urchin densities and projections of environmental variables, should accurately reproduce kelp dynamics across the state for the upcoming years and can be used to predict the dynamics of kelp into the future. However, if the fundamental nature of the relationships changes with climate change, then those relationships will need to be explored with experiments, and new models will need to be developed. How might the relationship between the drivers of kelp dynamics as we know them now, change in the future? Nutrients and temperature are important to primary producers, all of which possess specific photosynthetic temperature response curves that define an optimal temperature for photosynthesis and critical temperature thresholds. These are known for California kelps (Bell et al., 2015 ; Cavanaugh et al., 2019 ; Zimmerman & Kremer, 1984 ); but as sea temperature increases, local adaptation could change the response curves and the thresholds. Changes in the spatial distribution of various ecotypes of kelp through movement (passive or assisted) could also change these relationships. Similarly, interventions to assist the recovery of kelp forests, may accelerate local adaptation and alter the response to environmental variables found here, so as to contribute toward more resilient or resistant populations that can withstand contemporary and future environmental conditions. As we and others have demonstrated (Bell et al., 2015 ; Cavanaugh et al., 2019 ; McPherson et al., 2021 ), kelp forests are dynamic systems that can respond rapidly to fluctuating environmental drivers. Much of our understanding of the dynamic relationships between these oscillations and kelp dynamics is due to the availability of multidecadal time series of kelp canopy from remotely sensed imagery (Bell, Cavanaugh, Saccomanno, et al., 2023 ) and increasingly from in situ long‐term monitoring of kelp forests. The in situ diver surveys from long‐term monitoring programs in California provided key data that enabled the model construction of kelp and urchin densities, and identification of ecological processes (i.e., grazing) that contribute to their temporal and spatial patterns of abundance in the study region. The colocated and simultaneous sampling of kelp and urchin densities was key to capture the covariation of urchin and kelp densities and to incorporate that covariance into the SDMs (GAMs). These models revealed the important inverse relationships between purple urchin density with either kelp species. Furthermore, our analyses allowed for an independent assessment of the drivers of kelp dynamics, and the important drivers were similar to those identified using Landsat data from previous studies (Bell et al., 2015 ), providing added confidence in both methods. Conclusion Understanding how biotic and abiotic factors contribute to the temporal variation in the abundance of foundation species is a major focus of ecology and biogeography and is key to understanding biodiversity in a changing climate. Our results demonstrate the benefits of combining long‐term in situ monitoring data that provide information about species interactions (not yet obtainable through remote‐sensing) with remote‐sensing datasets to interpret population modeling results. Using both in situ and remote‐sensing data to understand the dynamics of surface canopy kelps allowed for the validation of the robustness of the model outputs. The maps produced from these robust models provide valuable information for managers and stakeholders about the locations that are more likely to support healthy kelp ecosystems and the functional relationships identified, form a basis for future studies focused on the spatiotemporal dynamics of kelp forests under a changing climate (Giraldo‐Ospina, Bell, et al., 2023 )."
} | 8,615 |
30952088 | null | s2 | 2,895 | {
"abstract": "Microbial communities can perform a variety of behaviors that are useful in both therapeutic and industrial settings. Engineered communities that differ in composition from naturally occurring communities offer a unique opportunity for improving upon existing community functions and expanding the range of microbial community applications. This has prompted recent advances in various community design approaches including artificial selection procedures, reduction from existing communities, combinatorial evaluation of potential microbial combinations, and model-based in silico community optimization. Computational methods in particular offer a likely avenue toward improved synthetic community development going forward. This review introduces each class of design approach and surveys their recent applications and notable innovations, closing with a discussion of existing design challenges and potential opportunities for advancement."
} | 235 |
30559729 | PMC6286981 | pmc | 2,896 | {
"abstract": "Sulfate is the predominant electron acceptor for anaerobic oxidation of methane (AOM) in marine sediments. This process is carried out by a syntrophic consortium of anaerobic methanotrophic archaea (ANME) and sulfate reducing bacteria (SRB) through an energy conservation mechanism that is still poorly understood. It was previously hypothesized that ANME alone could couple methane oxidation to dissimilatory sulfate reduction, but a genetic and biochemical basis for this proposal has not been identified. Using comparative genomic and phylogenetic analyses, we found the genetic capacity in ANME and related methanogenic archaea for sulfate reduction, including sulfate adenylyltransferase, APS kinase, APS/PAPS reductase and two different sulfite reductases. Based on characterized homologs and the lack of associated energy conserving complexes, the sulfate reduction pathways in ANME are likely used for assimilation but not dissimilation of sulfate. Environmental metaproteomic analysis confirmed the expression of 6 proteins in the sulfate assimilation pathway of ANME. The highest expressed proteins related to sulfate assimilation were two sulfite reductases, namely assimilatory-type low-molecular-weight sulfite reductase (alSir) and a divergent group of coenzyme F 420 -dependent sulfite reductase (Group II Fsr). In methane seep sediment microcosm experiments, however, sulfite and zero-valent sulfur amendments were inhibitory to ANME-2a/2c while growth in their syntrophic SRB partner was not observed. Combined with our genomic and metaproteomic results, the passage of sulfur species by ANME as metabolic intermediates for their SRB partners is unlikely. Instead, our findings point to a possible niche for ANME to assimilate inorganic sulfur compounds more oxidized than sulfide in anoxic marine environments.",
"introduction": "Introduction The anaerobic oxidation of methane (AOM) is an important biogeochemical process in the global carbon cycle, and is the primary sink for methane in anoxic ocean sediments ( Reeburgh, 2007 ). The diffusion of seawater sulfate into sediments serves as the major electron acceptor for this process, fueling a syntrophic association between uncultured anaerobic methanotrophic archaea (ANME) and sulfate-reducing bacteria (SRB) in regions where methane seepage occurs. Since the discovery of the AOM syntrophy ( Hinrichs et al., 1999 ; Boetius et al., 2000 ; Orphan et al., 2001 ), a number of hypotheses have been proposed on how ANME and SRB function together ( Knittel and Boetius, 2009 ), but they have not been fully resolved. Diffusible intermediates such as hydrogen, formate, or acetate could be exchanged between ANME and SRB to allow energy metabolism of AOM coupled to sulfate reduction ( Valentine and Reeburgh, 2000 ; Moran et al., 2008 ; Alperin and Hoehler, 2009 ). However, these hypotheses are inconsistent with results from incubation experiments ( Nauhaus et al., 2002 , 2005 ; Meulepas et al., 2009 ; Wegener et al., 2016 ). More recent work has suggested that ANME could be syntrophically coupled to SRB via direct interspecies electron transfer ( Meyerdierks et al., 2010 ; McGlynn et al., 2015 ; Wegener et al., 2015 ; Scheller et al., 2016 ; Skennerton et al., 2017 ). Alternatively, ANME (in particular ANME-2a and ANME-2c lineages) have been hypothesized to couple methane oxidation to sulfate reduction, releasing zero-valent sulfur which is subsequently disproportionated by SRB ( Milucka et al., 2012 ). Recent attempts to culture the syntrophic SRB partners of ANME using zero-valent sulfur were unsuccessful ( Wegener et al., 2016 ). Furthermore, a genetic and biochemical basis for dissimilatory sulfate reduction by ANME is currently lacking. Aside from members of the distantly related Archaeoglobales ( Pereira et al., 2011 ), no other euryarchaeotal group has been shown to have the genetic capability for energy conservation through dissimilatory sulfate reduction. Components of the assimilatory sulfate reduction pathway were found previously in ANME-1 and ANME-2c lineages, suggesting the genomic potential for biochemical transformation of oxidized forms of sulfur ( Meyerdierks et al., 2010 ; Krukenberg et al., 2018 ). On the other hand, marker genes or proteins for canonical dissimilatory sulfate reduction have not been detected in ANME ( Meyerdierks et al., 2010 ; Milucka et al., 2013 ; Wang et al., 2014 ; Krukenberg et al., 2018 ). All cultured methanogens to date can use sulfide for biosynthesis ( Liu et al., 2012 ). Given that ANME live in highly sulfidic environments, it stands to reason that they too would preferentially assimilate sulfide rather than invest energy in sulfate assimilation. However, the genomic capacity for sulfur metabolism has not been fully explored in different ANME lineages. An important step in sulfate reduction is the six electron reduction of sulfite to sulfide by assimilatory or dissimilatory sulfite reductases. Sulfite reductases can be classified into different phylogenetic groups and are found in the genomes of methanogens ( Dhillon et al., 2005 ; Loy et al., 2008 ; Susanti and Mukhopadhyay, 2012 ). The assimilatory-type low-molecular-weight sulfite reductase (alSir, also called Group I Dsr-LP) have been biochemically characterized and shown to reduce sulfite ( Moura et al., 1982 ). While alSir is not involved in dissimilatory sulfur metabolism in the bacteria Desulfuromonas acetoxidans , its physiological role remains unclear ( Moura et al., 1986 ; Moura and Lino, 1994 ). Another sulfite reductase, coenzyme F 420 -dependent sulfite reductase (Fsr), was more recently characterized in Methanocaldococcus jannaschii ( Johnson and Mukhopadhyay, 2005 ). Fsr is a fusion protein consisting of the beta subunit of the F 420 H 2 dehydrogenase at the N-terminus and a sulfite reductase at the C-terminus, together couple F 420 H 2 oxidation to sulfite reduction ( Johnson and Mukhopadhyay, 2005 ). When Fsr from M. jannaschii was heterologously expressed in sulfite-sensitive Methanococcus maripaludis , M. maripaludis was able to tolerate and assimilate sulfite as the sole sulfur source ( Johnson and Mukhopadhyay, 2008 ). Both alSir and Fsr were found in ANME-1 and ANME-2c genomes and expressed in the metatranscriptome ( Hallam et al., 2004 ; Susanti and Mukhopadhyay, 2012 ; Krukenberg et al., 2018 ), but their physiological roles remain unknown. Here we focus on identifying potential sulfur pathway genes in ANME, building from a collection of newly sequenced genomes to cover different lineages. Our genome observations were then combined with metaproteomics and microcosm experiments to gain further insight into the role of sulfur on ANME and their partner SRB. The capacity for sulfur usage by different ANME lineages is an important aspect to understanding energy conservation and syntrophy in AOM.",
"discussion": "Results and Discussion Survey of Sulfur Metabolism in ANME and Methanogen Genomes Sulfate can be reduced to sulfide for anabolism or catabolism, and distinct assimilatory or dissimilatory pathways have been characterized previously ( Verschueren and Wilkinson, 2001 ; Rabus et al., 2015 ). Analysis of the genomes from diverse ANME lineages revealed multiple candidate genes for assimilatory but not dissimilatory sulfate reduction (Figure 1 ). Nitrate-reducing Ca. Methanoperedens (formerly known as ANME-2d) recovered from freshwater environments showed a more expanded genetic capacity to reduce sulfate compared to the marine ANME lineages (ANME-1b, ANME-2a, ANME-2b, and ANME-2c) that perform AOM coupled sulfate reduction with deltaproteobacterial partners (Figure 1 ). This study focuses on the genetic potential of sulfate reduction to sulfide in the marine ANME lineages. The Supplementary Information includes sulfate reduction pathways separated by ANME lineage and a more detailed discussion on Ca. Methanoperedens. Putative sulfate transporters were identified in all ANME lineages, but given the substrate promiscuity of these transport systems for different oxyanions ( Marietou et al., 2018 ), the specificity and enzyme activity for sulfate is uncertain. Once sulfate is transported into the cell, the first step in sulfate reduction is the activation of sulfate (sulfur oxidation state +6) using ATP that can be catalyzed by two non-homologous ATP sulfurylase enzymes (Sat or CysDN). The heterodimeric sulfate adenylyltransferase (CysDN) used for sulfate assimilation is composed of a regulatory GTPase subunit CysN and a catalytic subunit CysD, and was previously reported in ANME-1 ( Meyerdierks et al., 2010 ). Our ANME-2b genome also contained a CysDN homolog (Figure 1 ). CysN and elongation factor 1-alpha (EF-1α) are homologous ( Mougous et al., 2006 ). Phylogenetic analysis confirmed that the CysN in ANME-1 and ANME-2b clustered together with characterized CysN as opposed ot EF-1α (Figure 2A ). In addition, the ANME CysN homolog were found next to CysD, which showed a similar evolutionary pattern (Figure 2B and Supplementary Table 1 ). The CysDN found in ANME would operate at a high energetic cost, requiring one GTP and one ATP per sulfate activated ( Liu et al., 1994 ) and therefore unlikely involved in dissimilatory sulfate reduction. In comparison, only three known methanogens ( Methanoregula formicica , Methanococcoides methylutens , Methanolobus tindarius ) contained CysDN, which were not monophyletic with the ANME proteins, suggesting that these methanogens may have acquired cysDN separately through horizontal gene transfer (Figures 2A,B ). The alternative protein for sulfate activation, the homo-oligomeric ATP sulfurylase (Sat), was found in ten methanogens as well as Ca. Methanoperedens, but not marine ANME lineages with partner SRB (Supplementary Table 1 ). Sat is involved in both assimilatory and dissimilatory sulfate reduction and uses one ATP per reaction ( Sperling et al., 2001 ; Ullrich et al., 2001 ). It is interesting to find CysDN and Sat in a few methanogens and Ca. Methanoperedens (see the Supplemental Information for details on sulfur pathway genes in methanogens). Future genetic studies of CysDN and Sat will be needed to confirm their roles in sulfate activation and assimilation in ANME and methanogens. FIGURE 2 Phylogeny of heterodimeric ATP sulfurylase subunits (CysDN). (A) Bayesian phylogeny of 416 amino acid residues of sulfate adenylyltransferase subunit 1 (CysN) and elongation factor 1 alpha (EF-1A) or elongation factor thermo unstable (EF-Tu) proteins. CysN, in green, formed a separate phylogenetic cluster from the homologous EF-1A and EF-Tu in blue. ANME proteins are bolded in red. The phylogenetic analysis distinguished CysN from their homologous elongation factor in ANME. (B) Bayesian phylogeny of 270 amino acid residues of sulfate adenylyltransferase subunit 2 (CysD) in green. They are found in ANME genomes next to CysN confirming that they are the heterodimeric ATP sulfurylase subunits. Asterisks ( ∗ ) indicate proteins that have been studied biochemically or structurally ( Liu et al., 1994 ; Andersen et al., 2000 ; Vitagliano et al., 2001 ; Mougous et al., 2006 ; Schmeing et al., 2009 ; Kobayashi et al., 2010 ; Thirup et al., 2015 ). Protein accession numbers from the NCBI database or gene IDs from the IMG database are shown in parentheses. Black dots on the branches represent Bayesian posterior probability values greater than 90%, and scale bar indicates the number of amino acid substitutions per site. Activated sulfate in the form of adenosine-5′-phosphosulfate (APS, sulfur oxidation state +6) can be reduced to sulfite (sulfur oxidation state +4) directly through APS reductase, or indirectly via 3′-phosphoadenosine-5′-phosphosulfate (PAPS, sulfur oxidation state +6) that uses APS kinase (CysC) followed by PAPS reductase ( Verschueren and Wilkinson, 2001 ). Genes for dissimilatory APS reductase (AprAB) and the essential membrane complex QmoABC in sulfate reducing bacteria and archaea ( Pereira et al., 2011 ) were not identified in any ANME genomes as reported in previous studies ( Meyerdierks et al., 2010 ; Wang et al., 2014 ; Krukenberg et al., 2018 ). We identified APS kinase ( cysC ) in our ANME-1b and ANME-2c genomes (Figure 1 ), which is in line with previous observations ( Meyerdierks et al., 2010 ; Krukenberg et al., 2018 ). Previous studies also mentioned the presence of assimilatory APS/PAPS reductase homolog in ANME-1, which we have also identified in ANME-2a and ANME-2b genomes (Figure 1 and Supplementary Table 1 ). Assimilatory APS reductase and PAPS reductase are homologous and use the same catalytic mechanism ( Carroll et al., 2005 ). These APS/PAPS reductase homologs are also widespread in methanogen genomes (Supplementary Table 1 ). We further investigated their phylogenetic relationship with characterized homologs, and found a separation between assimilatory APS/PAPS reductases in archaea and those commonly found in bacteria and eukarya (Figure 3 ). Based on their phylogenetic clustering with biochemically characterized homologs from Methanocaldococcus jannaschii ( Lee et al., 2011 ; Cho, 2013 ), we propose that one cluster is involved in APS reduction while the other cluster is involved in PAPS reduction (Figure 3 ). Assimilatory APS reductase of M. jannaschii is a small protein containing a 4Fe-4S domain ( Lee et al., 2011 ), while the assimilatory PAPS reductase of M. jannaschii contains an extra iron-sulfur binding domain at the N-terminus ( Cho, 2013 ). In comparison, homologs from ANME and other methanogen genomes contained additional domains including extra iron-sulfur cluster binding domains at the N- or C-terminus, or a cysteine desulfurylase domain at the C-terminus (Figure 3 ). Given these sequence differences, we refer to these homologous proteins as putative APS/PAPS reductases. It is possible that the homologs’ enzyme substrate specificity is the same as those in M. jannaschii , while the added iron-sulfur clusters could be facilitating electron transfer. The source of APS or PAPS is unclear, as many of the ANME and methanogen genomes lack the genes involved in activating sulfate and phosphorylating APS (Supplementary Table 1 ). FIGURE 3 Bayesian phylogeny of assimilatory adenyl-sulfate (APS) reductases and phosphoadenylyl-sulfate (PAPS) reductases. APS reductases and putative APS reductases are in green, PAPS reductases and putative PAPS reductases are in blue, bifunctional APS and PAPS reductase of Bacillus subtilis is in teal, and ANME proteins are bolded and in red. Archaeal and Bacterial/Eukaryal sequences formed separate clusters. Asterisks ( ∗ ) indicate proteins that have been studied biochemically from Archaea ( Lee et al., 2011 ; Cho, 2013 ), or Bacteria/Eukaryotes ( Gutierrez-Marcos et al., 1996 ; Savage et al., 1997 ; Suter et al., 2000 ; Berndt et al., 2004 ; Kim et al., 2004 ; Yu et al., 2008 ). Length of the proteins ranged from 239 to 896 amino acids with the addition of protein domains. The protein domains, if found, are shown with filled symbols. Only 172 amino acid residues of the central shared region were used for phylogenetics. Given that two copies of APS/PAPS reductases were found in each ANME-2 lineage and clustered separately, it is likely one is for APS and the other is for PAPS reduction similar to M. jannaschii ( Lee et al., 2011 ; Cho, 2013 ). Ca . Methanoperedens and four other methanogens in Methanosarcinales also contained a second putative assimilatory APS reductase more closely related to the bacterial/eukaryotic homologs, while ANME-1b contained a gene that does not cluster with assimilatory APS or PAPS reductases of known substrate. Protein accession numbers from the NCBI database or gene IDs from the IMG database are shown in parentheses. Black dots on the branches represent Bayesian posterior probability values greater than 90%, and scale bar indicates the number of amino acid substitutions per site. The final step in sulfate reduction involves a reduction of sulfite to sulfide (sulfur oxidation state −2). There are at least seven groups of homologous sulfite reductases that have a proposed assimilatory (aSir, alSir and Fsr) or dissimilatory (DsrA, DsrB, AsrC) function, in addition to a biochemically uncharacterized group Group III Dsr-LP (Dsr-Like Protein) ( Dhillon et al., 2005 ; Loy et al., 2008 ; Susanti and Mukhopadhyay, 2012 ). All known dissimilatory sulfite reductases encoding genes were absent from ANME and methanogen genomes (DsrA, DsrB and AsrC, Supplementary Figure 1 and Supplementary Table 1 ). In addition, genes for the essential membrane complex for dissimilatory sulfate reduction, DsrMK ( Pereira et al., 2011 ), found in all known sulfate-reducing bacteria and archaea were also absent in the ANME genomes investigated. However, all marine ANME lineages with SRB partner contained alSir and Fsr in their genomes (Figures 1 , 4 ), in line with previous ANME genomes ( Hallam et al., 2004 ; Meyerdierks et al., 2010 ; Wang et al., 2014 ; Krukenberg et al., 2018 ). Furthermore, in our phylogenetic analysis of sulfite reductases, it was observed that the previously studied coenzyme F 420 -dependent sulfite reductase (Fsr) from M. jannaschii ( Johnson and Mukhopadhyay, 2005 , 2008 ) clusters with Fsr genes from other non-cytochrome containing methanogens, here referred to as Group I Fsr. The Fsr homologs in ANME (with the exception of Ca. Methanoperedens) and other Methanosarcinales genomes formed a distinct well-supported clade, referred to here as Group II Fsr (Figure 4 ). FIGURE 4 Bayesian phylogeny of sulfite reductases. Two well-supported groups of Fsr were identified in exclusion of alSir and other sulfite reductases. ANME proteins are bolded in red. The phylogenetic tree was constructed based on 224 amino acid residues of the shared catalytic and siroheme-binding region. Asterisks ( ∗ ) indicate proteins that have been studied biochemically ( Huynh et al., 1984 ; Moura et al., 1986 ; Johnson and Mukhopadhyay, 2005 ). Protein accession numbers from the NCBI database or gene IDs from the IMG database are shown in parentheses. Black dots on the branches represent Bayesian posterior probability values greater than 90%, and scale bar indicates the number of amino acid substitutions per site. The fully expanded tree can be found be found in Supplementary Figure 1 . To show that Group II Fsr could be found in different methane seep sediments, we designed sets of specific and degenerate PCR primers based on alignments of ANME fsr sequences and used them to screen 4 different samples from Hydrate Ridge, United States (Supplementary Table 3 ). Positive amplicons were recovered from all four samples and the resulting fsr sequences clustered with fsrs recovered from ANME-2a/2b/2c genomes (Supplementary Figure 5 ). The ANME-2a reconstructed genome ( Wang et al., 2014 ) has two copies of Group II Fsr, but a primer set designed to specifically target one of the variants (IMG gene ID 2566126432) failed to amplify from our samples. All Group II Fsr sequences were then analyzed together with alSir and well-characterized DsrA to assess conservation of key amino acid residues. Sulfite reductases in general have conserved amino acid residues involved in the binding of siroheme and sulfite independent of their different physiological roles ( Crane et al., 1995 ; Dhillon et al., 2005 ; Schiffer et al., 2008 ). Alignments of both Fsr and alSir showed strong conservation of siroheme-[FeS] binding cysteines also present in DsrA (Supplementary Figure 3 ). However, the key residues that bind sulfite were changed in the Group II Fsr. Two arginine residues in the sulfite binding site ( Crane et al., 1995 ; Schiffer et al., 2008 ) were replaced with lysine and glycine in all Group II Fsr sequences (Supplementary Figure 3 ). This variation was also evident in models of protein homology which showed conservation in the overall structure and 3D positioning of siroheme-[FeS] binding cysteines (Supplementary Figure 4A ), but predicted an altered active site pocket due to the replacement of Arg with amino acids Lys or Gly smaller in size (Supplementary Figure 4B ). The amino acid changes may suggest a different substrate specificity of Group II Fsr compared to biochemically characterized Group I Fsr. Metaproteomic Expression of ANME Assimilatory Sulfur Metabolism Genes Environmental metaproteomic analysis of methane seep sediments confirmed the active expression of Group II Fsr and other sulfur metabolism genes from ANME (summarized in Table 1 , and manual validation of spectra corresponding to these peptides is provided in Supplementary Data Sheet 1 ). Peptides assigned to CysN, APS kinase and a putative APS/PAPS reductase homolog associated ANME-1 were detected (Table 1 ), suggesting that ANME-1 may be actively assimilating sulfate in the environment. Assimilation of sulfate would be particularly beneficial for ANME-1 at the base of or below the sulfate-methane transition zone where sulfate levels are low ( Beulig et al., 2018 ). In contrast, the only detected proteins closely affiliated with ANME-2a and ANME-2b were two sulfite reductases, alSir and Group II Fsr, and a putative sulfate transporter (Table 1 ). Table 1 Specific search for sulfur pathway proteins of marine ANME lineages in methane seep metaproteomes. Protein accession Description Organism Averaged normalized spectral counts (nSpc) in methane seep metaproteomes Hydrate ridge Santa monica 0–4 cm Santa monica 8–12 cm Eel river 0–10 cm Eel river 10–20 cm RCV62684 Unknown APS/PAPS reductase ANME-1b b.d. b.d. b.d. b.d. 71.5 RCV63267 CysN ANME-1b b.d. 484.4 b.d. b.d. b.d. RCV64987 SulP family inorganic anion permease ANME-1b b.d. b.d. b.d. 17.7 b.d. CBH38748 APS Kinase ANME-1b 1562.6 b.d. b.d. b.d. b.d. 2566123967 DASS family sodium-coupled anion symporter ANME-2a b.d. b.d. b.d. 103.6 b.d. PPA79744 Group II Fsr ANME-2b 2490.6 b.d. b.d. b.d. b.d. PPA80122 alSir ANME-2b 4762.7 b.d. b.d. b.d. b.d. AAU83232 Group II Fsr ANME-2c 3964.9 b.d. 1243.2 b.d. 19 AAU83223 alSir ANME-2c 5775.2 1684 b.d. b.d. b.d. MH823235 Group II Fsr Unknown ANME 6395.9 b.d. 1824.2 b.d. 38 MH823238 Group II Fsr Unknown ANME 1215.6 b.d. b.d. b.d. b.d. The search used a streamlined database containing only sulfur proteins of interest, and peptide fragmentation spectra of proteins found to be expressed were also manually validated in Supplementary Data Sheet 1 . b.d., below detection limits of mass spectrometry. Of all the ANME sulfur pathway proteins recovered, alSir and Fsr had the highest relative expression levels (Table 1 ). However, expression was at least 10-fold below the relative expression of methane oxidation genes and the dissimilatory sulfate reduction genes present in the syntrophic SRB partner (Supplementary Table 2 ). This result is similar to findings in a recent metatranscriptomic study of AOM enrichments ( Krukenberg et al., 2018 ), and appears inconsistent with a role in energy generating, dissimilatory functions, such as sulfate reduction to zero-valent sulfur ( Milucka et al., 2012 ). In our genomic survey of ANME and methanogens, alSir was more widespread than fsr and most of the alSir -containing species did not have the full assimilatory sulfate reduction pathway (Supplementary Table 1 ). The physiological role of alSir could be sulfite assimilation, but a source for in situ sulfite production remains unclear. Another possible role of alSir could be intracellular production of the essential sulfite for coenzyme M biosynthesis ( Graham et al., 2009 ) by the reverse reaction (oxidizing sulfide to sulfite) as previously proposed ( Moura et al., 1982 ). Given the high levels of in situ protein expression of Group II Fsr by ANME-2 (Table 1 ) and change in their active site residues (Supplementary Figure 3 ), further biochemical investigation are needed to confirm the enzyme substrate and reaction. Metabolic Response of ANME and Methanococcoides burtonii to Sulfite and Zero-Valent Sulfur To explore the potential roles of these sulfite reductases in ANME, we conducted microcosm experiments using a methane seep sediment (sediment ID 7142, dominated by ANME-2a/2c) amended with sulfite. Given Group I Fsr’s potential sulfite detoxification role in M. jannaschii ( Johnson and Mukhopadhyay, 2008 ), we hypothesize that Group II Fsr may also function in sulfite detoxification. Addition of sulfite at concentration of 1.0 mM was found to be inhibitory to ANME, leading to an immediate decrease in the rate of AOM (Figure 5A ). Methanococcoides burtonii , a close relative of ANME-2 within the Methanosarcinales , also contains alSir and Group II Fsr (Figure 4 ). Similar to ANME experiments, sulfite was also found to be inhibitory to the growth of M. burtonii , as observed by optical density measurements of the cultures (Supplementary Figure 2 ). These results contrast previous publications showing the effect of Group I Fsr on sulfite tolerance, where heterologous expression of Group I Fsr of M. jannaschii resulted in growth of Methanococcus maripaludis with 2 mM sulfite ( Johnson and Mukhopadhyay, 2008 ). Although these experiments were conducted with different methanogens, there seems to be a difference in sulfite tolerance or maybe function between Group I and Group II Fsr. FIGURE 5 Metabolic response of ANME to (A) sulfite and (B) zero-valent sulfur additions, as measured by 13 C-labeled dissolved inorganic carbon (DIC) production from 13 CH 4 . Zero-valent sulfur was added in the form of polysulfide and polythionate. Arrows indicate time of sulfur compound additions. Sulfite (1 mM) and zero-valent sulfur (>0.25 mM polysulfide or 1.0 mM polythionate) additions showed an inhibitory effect on methane oxidation in contrast to control or other sulfur compounds. Methane seep sediment ID 7142, dominated by ANME-2a and ANME-2c lineages, was used in these experiments. Zero-valent sulfur has been proposed as a metabolic intermediate in the AOM symbiosis ( Milucka et al., 2012 ). We used microcosm experiments to investigate the effect of zero-valent sulfur on ANME activity. An inhibitory effect of zero-valent sulfur in the forms of polythionate and polysulfide at concentrations of 1.0 and 0.25 mM, respectively, was observed on methane oxidation (Figure 5B ). Following thermodynamic predictions by Milucka et al. (2012) , product inhibition on methane oxidation by zero-valent sulfur should only occur at much higher concentrations (ΔG’ = 0 when [HS 2 - ] = 6193 M), assuming ANME directly coupled methane oxidation to dissimilatory sulfate reduction producing zero-valent sulfur in the form of disulfide. The effect of zero-valent sulfur on AOM measured in our experiments is therefore unlikely due to product inhibition but an alternative toxic mechanism unknown at the moment. M. burtonii , a closely related methanogenic archaeon to ANME-2, also stopped growing upon addition of 1 mM polysulfide (Supplementary Figure 2 ), supporting that zero-valent sulfur is toxic to this phylogenetic group rather than specifically to ANME. Furthermore, we could not enrich for the partner SRB in methane seep microcosms amended with polythionate or polysulfide (Supplementary Figure 6 ). This similar finding has been reported previously ( Wegener et al., 2016 ). Combined, these results indicate that zero-valent sulfur is unlikely a metabolic intermediate in the AOM symbiosis. Ecological Relevance of Assimilatory Sulfate Reduction Genes in ANME By recovering new ANME genomes and surveying their sulfur pathways, our results revealed the genomic potential for several ANME lineages to assimilate sulfur species more oxidized than sulfide. There are predicted differences between major ANME lineages in both sulfate activation by heterodimeric ATP sulfurylases (CysDN) found in ANME-1/2a and Ca. Methanoperedens, and the formation of sulfite using assimilatory APS/PAPS reductases found in ANME-2a/2b and Ca. Methanoperedens (Figure 1 ). Two sulfite reductases, alSir and Group II Fsr, were found to be the highest expressed proteins in methane seep sediment related to sulfur cycling in ANME (Table 1 ). However, their expression levels were still much lower than that of primary metabolisms, i.e., methane oxidation in ANME and dissimilatory sulfate reduction in SRB. Together with information on their characterized homologs associated with assimilatory but not dissimilatory sulfate reduction, our results suggest that ANME are unlikely to perform dissimilatory sulfate reduction as proposed previously ( Milucka et al., 2012 ). Additional experiments are needed to determine the enzyme function of two sulfite reductases that are common to all marine ANME lineages, as well as the divergent homologs of ATP sulfurylase and assimilatory APS/PAPS reductases that were found in some ANME lineages. These genes may be important for the synthesis of essential organo-sulfur molecules, in particular coenzyme M that has a sulfonate group at +4 oxidation state. The differences in sulfur assimilatory genes between ANME lineages, representing novel order to genus-level diversity, underscore the phylogenetic as well as physiological differences between them (see Supplementary Information for a more detailed discussion). It is intriguing to find potential genes for assimilation of sulfate or other sulfur species more oxidized than sulfide in ANME genomes, especially ANME-1b/2a/2b lineages that live in syntrophy with SRB partners and high levels of sulfide. In marine sediments with active sulfur cycling, such as sulfate-methane transition zones where ANME thrive, sulfate and sulfide may not be the only sulfur species present. Sulfite and thiosulfate have previously been measured at low micromolar concentrations in different marine sediments including methane seep sediment ( Zopfi et al., 2004 ; Smith et al., 2017 ). Under these conditions, the ability to scavenge additional sulfur species for anabolism could be beneficial. In addition, ANME-1b and ANME-2a/2b/2c lineages have been found together with microorganisms other than deltaproteobacterial sulfate reducers that hints alternative syntrophic lifestyles ( Hatzenpichler et al., 2016 ), and ANME-2a/2c remained anabolically and catabolically active in laboratory incubations devoid of sulfate using electron acceptors including AQDS, humic acids and Fe (III) ( Scheller et al., 2016 ). In these scenarios, the ability to assimilate multiple sulfur sources using Group II Fsr or other enzymes in the assimilatory sulfate reduction pathway may provide ANME, or methane-cycling archaea in general, a broader environmental niche and the ability to survive in environments with different anabolic sources of sulfur."
} | 7,719 |
20305795 | null | s2 | 2,897 | {
"abstract": "Polymer networks possessing reversible covalent crosslinks constitute a novel material class with the capacity for adapting to an externally applied stimulus. These covalent adaptable networks (CANs) represent a trend in polymer network fabrication towards the rational design of structural materials possessing dynamic characteristics for specialty applications. Herein, we discuss the unique attributes of CANs that must be considered when designing, fabricating, and characterizing these smart materials that respond to either thermal or photochemical stimuli. While there are many reversible reactions which to consider as possible crosslink candidates in CANs, there are very few that are readily and repeatedly reversible. Furthermore, characterization of the mechanical properties of CANs requires special consideration owing to their unique attributes. Ultimately, these attributes are what lead to the advantageous properties displayed by CANs, such as recyclability, healability, tunability, shape changes, and low polymerization stress. Throughout this perspective, we identify several trends and future directions in the emerging field of CANs that demonstrate the progress to date as well as the essential elements that are needed for further advancement."
} | 317 |
27386542 | PMC4928936 | pmc | 2,898 | {
"abstract": "Researchers demonstrate the replication of 3D natural gyroid nanostructures with superior optical performance and properties.",
"introduction": "INTRODUCTION Gyroid structures within butterfly wings are chiral periodic structures with a cubic symmetry ( 1 – 4 ). They are the subject of rapidly increasing interest in photonics, with applications from photonic crystals (PCs) ( 5 , 6 ) and metamaterials ( 7 ) to optical materials with topological complexity ( 8 ), owing to their unique geometrical properties. The strong chirality phenomenon of gyroid structures results in the ability to manipulate optical circular dichroism ( 6 ) and has even been used as a new miniature chiral beam splitter ( 9 ). Gyroid structures have also been predicted to exhibit frequency-isolated Weyl points with gapless surface dispersions and line nodes ( 10 ), similar to Dirac points in two-dimensional periodic systems. In addition, gyroid metamaterials made by metals or coated with metals rather than dielectric materials are demonstrated to have a wide variety of tunable nonlinear optical properties with ultrafast response ( 11 ). The experimental application of gyroid structures requires appropriate ways to fabricate these three-dimensional (3D) complex nanostructures ( 3 , 6 , 12 ). Natural gyroid structures from butterfly wings are found to have several drawbacks to their optical performance, including a different crystallite orientation, the presence of both left- and right-handed single gyroid enantiomers, and uncontrolled structure disorders ( 1 , 13 ). Because of these, natural butterfly wings contain gyroid structures lacking significant circular dichroism ( 1 ). Different fabrication techniques have been used to mimic the said natural structures with the same geometrical configuration but with a much larger size scale than that found in nature ( 14 – 17 ). This is attributable to the lack of 3D fabrication techniques with resolutions comparable to biological nanoscales of less than 100 nm. Thus, the application of gyroid structures in optics, photonics, and biomimetics at the short-light wavelength region, including the visible and ultraviolet ranges, is greatly limited. For example, the predicted Weyl points and line nodes can only be experimentally demonstrated at microwave wavelengths ( 18 ). Gyroid metamaterials work at the terahertz frequency but not in the optical range ( 11 ). To fully replicate and exceed these 3D nanostructures, we need a 3D fabrication technique with a feature resolution of 100 nm and a feature separation of 300 nm. High-resolution lithography techniques, such as electron beam lithography, can give a resolution below 100 nm ( 19 , 20 ); however, it does not have intrinsic 3D capability. Multiphoton lithography is an ultimate approach to 3D nanofabrication, but lacks resolution below 100 nm ( 9 , 12 ). Here, using optical two-beam super-resolution lithography ( 21 , 22 ), we demonstrate that this recently developed technique can fabricate biomimetic photonic structures with superior resolution, uniformity, and controllability.",
"discussion": "DISCUSSION We have demonstrated the replication of 3D biomimetic gyroid nanostructures at size scales smaller than their natural counterparts. The optical two-beam super-resolution lithography technique enables smaller unit cell sizes than nature. It offers more flexibility in structure design than nature, including unit cell size, filling fraction, and control of orientation and termination. It allows designable fabrication to demonstrate the applicable functionality. This work opens the door to the investigation of more gyroid-related applications in optics and photonics in the visible or near-ultraviolet wavelength region. It can also further enhance our understanding of these biological nanostructures and their functionality in nature."
} | 956 |
37061714 | PMC10105947 | pmc | 2,899 | {
"abstract": "Background Amino acid production features of Corynebacterium glutamicum were extensively studied in the last two decades. Many metabolic pathways, regulatory and transport principles are known, but purely rational approaches often provide only limited progress in production optimization. We recently generated stable synthetic co-cultures, termed Communities of Niche-optimized Strains (CoNoS), that rely on cross-feeding of amino acids for growth. This setup has the potential to evolve strains with improved production by selection of faster growing communities. Results Here we performed adaptive laboratory evolution (ALE) with a CoNoS to identify mutations that are relevant for amino acid production both in mono- and co-cultures. During ALE with the CoNoS composed of strains auxotrophic for either l -leucine or l -arginine, we obtained a 23% growth rate increase. Via whole-genome sequencing and reverse engineering, we identified several mutations involved in amino acid transport that are beneficial for CoNoS growth. The l -leucine auxotrophic strain carried an expression-promoting mutation in the promoter region of brnQ (cg2537), encoding a branched-chain amino acid transporter in combination with mutations in the genes for the Na + /H + -antiporter Mrp1 (cg0326-cg0321). This suggested an unexpected link of Mrp1 to l -leucine transport. The l -arginine auxotrophic partner evolved expression-promoting mutations near the transcriptional start site of the yet uncharacterized operon argTUV (cg1504-02). By mutation studies and ITC, we characterized ArgTUV as the only l -arginine uptake system of C. glutamicum with an affinity of K D = 30 nM. Finally, deletion of argTUV in an l -arginine producer strain resulted in a faster and 24% higher l -arginine production in comparison to the parental strain. Conclusion Our work demonstrates the power of the CoNoS-approach for evolution-guided identification of non-obvious production traits, which can also advance amino acid production in monocultures. Further rounds of evolution with import-optimized strains can potentially reveal beneficial mutations also in metabolic pathway enzymes. The approach can easily be extended to all kinds of metabolite cross-feeding pairings of different organisms or different strains of the same organism, thereby enabling the identification of relevant transport systems and other favorable mutations. Supplementary Information The online version contains supplementary material available at 10.1186/s12934-023-02078-2.",
"conclusion": "Conclusions Even after decades of research, the genome annotations of C. glutamicum and other biotechnologically relevant production organisms still contain many uncharacterized segments, harboring potential for not only increasing metabolic understanding but also for increasing metabolite production. The co-culture evolution-guided metabolic engineering approach presented in this study represents one additional tool for exploiting this potential through putting higher selective pressure on communities to grow faster comparable to monoculture approaches. This enabled the identification of amino acid transport systems not identifiable with other evolution approaches so far. Especially the deletion of the identified argTUV in existing high-yield l -arginine-producers [ 55 ] could therefore be worthwhile. Co-culture evolution-guided metabolic engineering could also easily be extended not only to other CoNoS published before [ 9 ] but also to a number of different metabolite cross-feeding pairings, enabling the identification of more transport systems. Additionally, further rounds of evolution with already import-optimized strains with lower levels of production could result in mutations occurring also in metabolic pathway enzymes, since also other co-culture pairings suggested that increasing production via community evolution is possible [ 6 ]. Employing the new best CoNoS for future work will enable further progress in improving small molecule production with highly efficient microbial communities [ 3 ].",
"discussion": "Discussion Native microbial communities have usually evolved over thousands of years toward an extremely efficient use of the available resources, thereby heavily relying on cooperation and cross-feeding among the community members [ 1 , 3 , 5 ]. Recent advances in evolving synthetic co-cultures of strains of the same species [ 6 , 40 ] or mixed species communities [ 41 , 42 ] underlined the potential of increasing product cross-feeding, which could be of high biotechnological value. In this work, we successfully evolved a synthetic community composed of amino acid auxotrophic strains [ 9 ], identified the relevant mutations and used these to increase l -arginine production also in monoculture. Automated ALE using repetitive batch cultures, which has so far almost exclusively been demonstrated for monocultures [ 15 , 43 ], proved to be easily applicable for CoNoS to select for faster-growing communities. During the ALE, both the ΔARG and the ΔLEU strains accumulated mutations beneficial for community growth, which was also observed for synthetic co-cultures consisting of l -leucine and l -lysine auxotrophic E. coli strains [ 44 ]. Several mutations we found increased growth in a co-culture setting, but did not result in better growth of a monoculture supplemented with the required amino acid. This is also in agreement with the E. coli approaches, where single evolved community members showed decreased growth in supplemented monocultures [ 44 ]. Let’s have a closer look on the mutations we found in the evolved CoNoS. These were i) a mutation in the cystathionine β-lyase MetC, ii) mutations in or upstream of amino acid uptake systems, iii) mutations in the multi-subunit Na + /H + antiporter Mrp1, and iv) mutations in uncharacterized genes. Based on our MSA and AlphaFold2 data, we proposed the MetC S322F mutation to alter the dynamics of the C-terminal domain, thus modulating substrate binding and/or catalysis. In the context of the ∆LEU ARG + strain, the resulting changes to the steady-state levels of homocysteine, cystathionine, and methionine may indirectly influence either l -leucine utilization or l -arginine export. Conflicting results regarding the applicability of AlphaFold2 for predicting the impact of single point mutations in protein structures have been reported [ 45 – 48 ]. In the case of MetC and Cg2850, despite the high similarity of the structures predicted for the WT and the mutated version, small structural changes can be detected in the vicinity of the mutation site. This was not observed for Cg1874. Notably, even if an amino acid exchange does not significantly alter the mean structure, it may nevertheless affect other properties such as protein dynamics, stability, enzymatic activity, or protein–protein and protein–ligand interactions, the investigation of which is beyond the scope of this study. In our CoNoS with auxotrophic strains, efficient amino acid export and uptake appears to be one of the key factors for community growth. In total, we found four different mutations that presumably increased the amount of available transporters in the cell (Table 1 ). In this context, we identified and characterized ArgTUV as an l -arginine and l -citrulline importer. Despite ArgT showing significant homology to other secreted substrate-binding proteins such as ArtJ of G. stearothermophilus and hypothetical ancient precursors binding also l -histidine, l -lysine, l -cysteine or l -glutamine [ 38 , 39 ], ArgT bound exclusively to l -arginine and its molecular precursor l -citrulline. When synthetic communities of two E. coli strains auxotrophic for histidine or one other metabolite were evolved, several mutations appeared in promoter and regulatory regions that increased e.g. l -histidine and 2-oxoglutarate uptake [ 40 ]. Only very few mutations were found in the coding region of transporters, which might alter transporter activity or codon usage or translation by influencing mRNA structure [ 40 ]. In a further study, two E. coli strains auxotrophic for either l -tryptophan or l -tyrosine were evolved together and the resulting strains produced more of the amino acid required by the partner strain [ 6 ]. The evolved strains were not sequenced, thus it is unknown whether also other factors, such as amino acid import, was affected [ 6 ]. In another study with a co-culture consisting of two E. coli strains auxotrophic for either l -tryptophan or l -tyrosine, mutations were identified in a porin and in the global transcriptional regulator Lrp [ 49 ]. The evolution of a lactic acid bacterium, which is naturally auxotrophic for amino acids, together with a Saccharomyces cerevisiae, auxotrophic for riboflavin or folate, also revealed several mutations that regulate transcription or are associated with amino acid uptake [ 41 ]. Interestingly, most of the mutations influenced transcription or translation of the transporter protein, and only a few the activity of the protein itself, and mutations were almost exclusively associated with uptake systems but not with exporters. The Mrp1 mutations resulted in severe impairments of the function of the multi-subunit Na + /H + antiporter Mrp1, analogous to gene deletion or other mutations identified in Mrp1 subunits before [ 25 , 50 ]. The fact that mutations of Mrp1 and the MetC/P brnQ * mutation evolved together twice in independent experiments suggested a functional link between these two proteins. In most organisms, the l -leucine-import via BrnQ depends on the proton motive force (reviewed in [ 51 ]), coupling l -leucine and Na + -symport across an energy gradient [ 52 ]. Mrp1 is the main Na + /H + antiporter in C. glutamicum and required to establish the gradient for Na + -coupled uptake [ 20 , 50 ]. A defect in Mrp1 presumably leads to a decreased Na + gradient and thus a reduced l -leucine import. Therefore, it is not obvious how the Mrp1 mutations are beneficial for the ΔLEU strain. The fact that the mutations in the uncharacterized proteins Cg1874 and Cg2850 evolved several times independently from each other is a strong hint that they may be somehow beneficial for CoNoS growth. Their specific role is still unclear, as their reconstruction had no obvious effect in monoculture and in a CoNoS setting (Additional file 1 : Fig. S12). However, the reconstructed strains were only tested in a CoNoS with the parental ΔLEU ARG + strain, so maybe the beneficial effect is only apparent with a partner carrying mutations MetC S322F /P brnQ * and/or mutations in Mrp1. In summary, these results suggested that the metabolite uptake is often the major bottleneck under the very low metabolite concentrations in a CoNoS as observed before [ 9 ]. The identified mutations support the view that transport may be mostly limited by the availability of transporter proteins, because all mutations presumably led to an increased transporter availability. Metabolite production and export appears to be less critical in our setup because we did not find any mutation obviously related to these processes. Nevertheless, rationally increased amino acid production also increased the community growth, suggesting that sufficient amino acid production is still one major bottleneck, leaving room for improvement. For the rational design of synthetic communities, this means that one should concentrate both on metabolite production and on metabolite import to obtain optimal community growths. At the end of this study, we would like to discuss what kind of mutations we expected to find by evolution of a CoNoS and how this differs from the ALE of monocultures. When selecting for faster growing strains or cultures, the selection pressure is highest on the bottleneck that is limiting growth most strongly. In our case, this was most likely amino acid import, because we found mutations in promoters leading to an increase in l -leucine and l -arginine import. Elevated uptake can not only result from promoter mutations that promote RNA polymerase binding upstream of the importer gene, but also be caused by mutations of regulators, mutations of the transporter start codon to a more favorable one, mutation of the RBS, mutation of the transporter itself increasing binding affinity or transport speed, mutations that lead to more favorable codons and several other mechanisms reviewed elsewhere [ 53 , 54 ]. Thus, there are numerous potential targets which can mutate to increase import. If export is the limiting factor, the transporter and the corresponding regulatory mechanisms can mutate in a similar way. If transport is no longer limiting, we would also expect mutations in the amino acid biosynthesis pathways themselves. Here, again, regulatory processes can be affected, or the biosynthetic enzymes mutate to release e.g. feedback inhibition or increase reaction speed. Thus, to find mutations in the biosynthetic pathways using ALE, it is necessary to generate a CoNoS that is no longer limited in amino acid import and export."
} | 3,279 |
35235034 | PMC9033746 | pmc | 2,900 | {
"abstract": "Productive biofilms are gaining growing interest in research due to their potential of producing valuable compounds and bioactive substances such as antibiotics. This is supported by recent developments in biofilm photobioreactors that established the controlled phototrophic cultivation of algae and cyanobacteria. Cultivation of biofilms can be challenging due to the need of surfaces for biofilm adhesion. The total production of biomass, and thus production of e.g. bioactive substances, within the bioreactor volume highly depends on the available cultivation surface. To achieve an enlargement of surface area for biofilm photobioreactors, biocarriers can be implemented in the cultivation. Thereby, material properties and design of the biocarriers are important for initial biofilm formation and growth of cyanobacteria. In this study, special biocarriers were designed and additively manufactured to investigate different polymeric materials and surface designs regarding biofilm adhesion of the terrestrial cyanobacterium Nostoc flagelliforme (CCAP 1453/33). Properties of 3D-printed materials were characterized by determination of wettability, surface roughness, and density. To evaluate the influence of wettability on biofilm formation, material properties were specifically modified by gas-phase fluorination and biofilm formation was analyzed on biocarriers with basic and optimized geometry in shaking flask cultivation. We found that different polymeric materials revealed no significant differences in wettability and with identical surface design no significant effect on biomass adhesion was observed. However, materials treated with fluorination as well as optimized biocarrier design showed improved wettability and an increase in biomass adhesion per biocarrier surface.",
"introduction": "Introduction Phototrophic biofilms composed of terrestrial cyanobacteria are embedded in a matrix of extracellular polymeric substances (EPS) [ 1 ]. The EPS contain various polysaccharides, lipids, and extracellular proteins that lead to a stable production environment, enhanced mechanical stability, and increased surface adhesion [ 2 , 3 ]. Due to the discovery of several valuable compounds with antibacterial and antifungal activities [ 4 , 5 ] in biomass and EPS [ 6 – 8 ], cyanobacteria are beneficial to the development of drugs such as antibiotics and offer a great potential for pharmaceutical applications [ 9 , 10 ]. To use this potential and to enable an industrial grade of utilization, the optimization of production technologies for cyanobacteria is necessary. For the controlled phototrophic cultivation of cyanobacteria and production of bioactive substances, special reactors, called photobioreactors, are used. Over the past few decades, their operation and technology have been progressively improved and aspects such as light distribution [ 11 , 12 ] or temperature control [ 13 , 14 ] have been examined. Most cultivation systems are designed for aquatic cyanobacteria and therefore a submerged cultivation with liquid media as suspension. For terrestrial cyanobacteria, originated from emersed air-exposed habitats, however, the cultivation conditions in submerged fermentations are not optimal and their productivity is limited [ 15 , 16 ]. To address these limitations, a special biofilm cultivation system, the emersed photobioreactor (ePBR), was developed by Kuhne et al. [ 17 ] and steadily refined over the past few years [ 18 – 20 ]. The ePBR imitates the natural habitat of terrestrial cyanobacteria by an air exposed cultivation and nutrient supply through an aerosol. The aerosol-based cultivation results in reduced water consumption and allows better process control, for example, the aimed desiccation to induce the production of EPS [ 21 ]. Surface attached cultivation of cyanobacteria in biofilm photobioreactors led to higher biomass productivity and more efficient harvesting processes compared to suspended cultivations [ 18 , 22 , 23 ]. When cultivated as biofilms, the biomass production depends on the available cultivation surface. To maximize this cultivation surface and to keep the volumetric requirements low, regular shaped biocarriers can be used as a substrate for surface attached biofilm growth. Biocarriers and packings are designed to increase the reactive surface area for biochemical, chemical, or thermal processes. Therefore, they are already implemented in several industrial applications. In thermal process engineering, for example, packings are used as packed columns for heat and mass transfer processes in rectification, absorption, and extraction [ 24 ]. Heterotrophic biofilm reactors in wastewater treatment use biocarriers to provide cultivation systems with decreased volume requirement, consistent production, and enhanced mass transfer [ 25 , 26 ]. Related to the variety of different application fields, biocarriers and packings are available in various sizes, geometries, and materials. Biocarriers must fulfill specific requirements for the application in photobioreactors. A high specific surface is required to increase the surface–volume ratio and optimize volumetric productivity. Due to the phototrophic cultivation, light supply is an important factor, which is affected by geometry, porosity, and material transparency. Flow characteristics are also crucial as they influence the mixing efficiency in submerged and the aerosol distribution in emerged fermentations. Furthermore, the choice of biocarrier material regarding performance (e.g. toxicity, flexibility), the material and media composition (e.g. pH value) and surface properties such as roughness and wettability are essential for initial biofilm adhesion [ 27 ], growth, and harvesting. However, not only the biocarrier material, but also the manufacturing process plays a key role. Additive manufacturing (AM), commonly known as 3D-printing, is a rapid evolving manufacturing technique. Due to the variety of inexpensive print materials and 3D-printers, fused filament fabrication (FFF) is one of the most common AM processes. A polymeric filament is used in FFF to generate a solid model, layer by layer, on a build platform. Single extruded lines are combined to a slice of the model in each layer. Thereby, material is extruded through a heated nozzle with a cylindrical outlet, resulting in lines with circular cross-section. These lines are placed close to each other and fused together, but their circularity remains noticeable on the surface of the model and especially on the top layer. Orientation of the top layer lines can be influenced by model placement on the printer and presets in the slicing software. Due to the circular extrusion, the FFF process leaves small grooves between each extruded line vertically and horizontally. Another AM process is digital light processing (DLP), which manufactures solid models out of liquid photopolymers. In DLP 3D-printing, an image of one layer of the model is projected to a build platform and the photopolymer is cured in designated zones using photopolymerization. During printing, photopolymer is supplied in a resin tank and models are built layer by layer to the build platform. There are no grooves on the surface and overall finish is smooth because the projected image consists of square pixels. AM has already been used to investigate biocarrier geometry [ 28 ] and adhesion of aquatic microalgae and macroalgae to different surface topographies [ 29 ]. However, there is no literature available regarding biofilm adhesion of terrestrial cyanobacteria to additively manufactured biocarriers. This study focused on the impact of biocarriers’ surface and material characteristics for biofilm adhesion of terrestrial cyanobacteria to optimize emerged and submerged fermentations. AM was used to manufacture polymeric biocarriers and material samples for contact angle and surface roughness measurements. To specify the influence of wettability for biofilm adhesion, surface characteristics of 3D-printed materials were altered by gas-phase fluorination. In addition, terrestrial cyanobacteria were cultivated on suspended biocarriers by shaking flask cultivation to investigate biofilm formation and attachment on different polymeric biocarriers.",
"discussion": "Discussion Surface properties of 3D-printed materials Wettability directly corresponds to contact angle. Generally, lower contact angles indicate a better wetting and more hydrophilic surfaces, while higher contact angles represent poor wetting and more hydrophobic surfaces. Measurement of contact angles showed that the grooves on the surface of FFF printed material samples influenced drop formation and led to an elliptical deformation in direction of the line pattern as illustrated in Fig. 7 for PLA. The deformation effect occurred on all tested FFF printed surfaces and was even enhanced through fluorination. As the line patterns were oriented differently in relation to the camera view on the samples for contact angle measurements (Fig. 3 ), the deformation effect resulted in variation of drop width and height and thus caused varying contact angles for the different top layer orientations. Complex geometries like biocarriers resulted in a 3D-printed object with surface patterns that consisted of a mixture of the analyzed line orientations. Due to the different line orientations, varying contact angles on the active surface of the biocarriers are to be expected. Therefore, no general contact angle of the biocarrier could be given. To characterize the surface properties, the average contact angles were determined at four different line orientations. Fig. 7 Top view of the drop formation of 300 µL BG 11 drops on smooth glass, without deformation, and on the surface of untreated and fluorinated PLA samples, with deformation in direction of the top layer lines For PLA, the average contact angle of BG 11 medium was 82.3° for untreated and 62.4° for fluorinated samples. This is close to the water contact angle of 81.1° for untreated PLA and 61.0° for fluorinated PLA observed in a similar fluorination process [ 31 ]. Surface roughness of the analyzed samples was comparable to each other and depended on the line orientation of the top layer. The number of grooves on the surface, caused by FFF 3D-printing, related to the differentiation in surface roughness. Fluorination resulted in either reduction or increase of surface roughness. The highest alteration of arithmetical mean deviation by fluorination was approximately 10 µm. Since individual samples were used for untreated and fluorinated measurements, their roughness difference could also be caused by the fabrication process as the alteration lies well within the expected accuracy of FFF printing. Therefore, it is not clear that differences in surface roughness between untreated and treated samples originate from the fluorination process. However, it is assumed that fluorination does not alter surface roughness, which corresponds to the observations by Schroepfer et al. [ 31 ]. Influence of wettability and surface roughness on biofilm adhesion The influence of wettability on biofilm adhesion is a highly discussed topic. Apart from various biofilms investigated, in most studies several materials have been used to achieve different wettability for the investigation of biofilm adhesion. Material composition also influences biofilm adhesion [ 27 ], which makes it difficult to isolate the effect of wettability. In this study gas-phase fluorination showed a significant reduction of contact angle and thus caused an alteration of the wettability for all materials. Although the chemical composition of the materials was also modified through fluorination, this resulted in different wettabilities for the same material and simplified the comparison and isolation of the influence of wettability. The correlation between attached biomass and the average contact angle of the biocarrier materials is shown in Fig. 8 . Based on the investigation of the surface properties between the untreated FFF printed biocarriers (ABS, PETG, PLA, and PP), there was no considerable difference found in biofilm adhesion, which corresponded to the similar wettability observed for the tested materials. Fluorination however resulted in a significant reduction of the average contact angle of approximately 20° for PLA and 28° for PP. For all investigated biocarriers, fluorination increased wettability and hydrophilicity while the surface roughness remained almost unaltered. The trend of higher biomass attachment on fluorinated biocarriers of the same material, compared to untreated carriers, indicated that biofilm attachment by N. flagelliforme (CCAP 1453/33) is preferred on hydrophilic surfaces with high wettability. Several other studies reported enhanced biofilm adhesion to hydrophobic surfaces with low wettability [ 27 , 32 – 34 ], no influence of wettability to biofilm adhesion [ 35 ] as well as increased adhesion to hydrophilic surfaces with high wettability [ 36 ]. The different findings relate to different species such as algae and cyanobacteria [ 27 , 32 , 34 , 35 ], macroalgae [ 33 ] and active sludge in wastewater treatment [ 36 ]. The diversity in these findings indicates that the preferred surface characteristic (hydrophobic or hydrophilic) highly depends on the used microorganism and that the results are not directly transferable to other cyanobacteria species. Furthermore, there are several reports for the same microalgae species ( Chlorella vulgaris) describing enhanced adhesion to hydrophobic surfaces [ 27 , 32 ] and no influence of hydrophobicity on biofilm adhesion at all [ 35 ]. This indicates, although wettability has a major influence on the initial biofilm formation, that other factors such as adaptation of the microorganisms [ 35 ], media composition [ 32 ], material compatibility [ 27 ], or cultivation conditions are also crucial and make it difficult to generalize the prediction of biofilm formation. Fig. 8 Biomass dry weight (BDW) per biocarrier surface in relation to average contact angle for untreated and fluorinated FFF biocarriers. Given values represent the mean of 10 biocarriers and the average material contact angle for BG 11 media with standard deviation In this study, the surface roughness was measured primarily to show that the fluorination process does not have a significant impact on surface roughness. The observed changes in roughness were rather small, occurred randomly, and most likely due to inaccuracies in the manufacturing process. The influence of surface structure and roughness on biofilm adhesion has been previously studied by other researchers [ 28 , 29 , 36 – 39 ]. It is reported that surface features of similar size to the individual cells in the biofilm are beneficial for surface adhesion, while smaller features decrease biofilm adhesion [ 37 ]. Furthermore, the optimal surface roughness highly depends on the used microorganisms [ 29 ]. However, in terms of biocarriers, several findings report increased biomass adhesion to rougher surfaces [ 28 , 36 , 39 ] due to protection of cells from hydrodynamic forces [ 38 ] and an increased effective surface due to microscopic surface features [ 28 ]. Apart from surface roughness, biocarrier design and dimensions are also important for the initial biofilm formation. Influence of biocarrier design on biofilm adhesion Increased biofilm attachment per biocarrier surface on optimized PAR biocarriers showed the importance of biocarrier design for biofilm formation and growth. The immobilization of biomass on biocarriers depends not only on surface properties like wettability and surface roughness, but also on hydrophobicity of the microorganisms, electrophoretic mobility and steric effects [ 40 , 41 ]. Hydrodynamic forces also play a key role for the initial biofilm formation on the biocarriers. For the determination of biofilm attachment in this study, a static cultivation was performed to reduce the risk of biomass detachment from the carriers. In static cultivation, sedimentation reduces the amount of biomass in suspension and can lead to poor biofilm formation because less biomass is available for attachment. This is supported by the low amount of biomass on the biocarriers observed in our experiments. However, mixing during cultivation results in increased shear forces that could lead to detachment of biomass from the biocarrier surfaces. To prevent biofilm detachment by erosion due to fluid shear forces, abrasion due to collision with other particles, or sloughing [ 42 ], biocarriers should have recessed areas to protect the biomass from shear forces. Furthermore, optimized biocarriers should have a highly effective contact area to facilitate biomass adhesion and a high loading capacity to ease the diffusion of nutrients [ 25 , 42 ]. A biocarrier density similar to the medium density reduces hydrostatic forces and is expected to ensure optimal flow and homogeneous biofilm attachment around the biocarrier in submerged cultivations. Because inoculation for emerged cultivations is usually carried out in suspension, a neutrally buoyant biocarrier could also ensure uniform biofilm inoculation before emerged cultivation. Regarding different biocarrier structures, the highest growth of heterotrophic and phototrophic biofilms has often been observed in the protected internal structures and cavities when used in fluidized beds [ 36 , 43 , 44 ]. As the cyanobacteria tended to adhere to the grooves of the lateral area on the cultivated biocarriers (Fig. 4 ), it is assumed that cyanobacteria also prefer protected areas for initial biofilm formation in static cultivation. Although the internal areas provide good protection against shear forces, light transmitting into these areas can be affected by several factors. Biomass on the outer shell of the biocarrier, adjacent biocarriers and suspended cells surrounding the biocarriers [ 45 ] reduce light penetration and thereby affect photosynthesis and biomass production in the protected internal structures. However, cyanobacteria and especially immobilized terrestrial cyanobacteria are highly adaptable to extreme and fluctuating light conditions, which is shown by their natural occurrence in harsh environments like the Antarctica [ 46 ], tropical regions [ 47 ], or deserts [ 48 ]. Light acclimation and adaptation enables cyanobacteria to produce biomass under extremely low light intensities [ 49 ] and even increases production of certain valuable pigments like chlorophyll, phycobilisomes or specific carotenoids [ 50 ]. The biomass in the internal structures of the biocarriers is expected to continue to grow due to light adaptability of cyanobacteria, even if most of the light is blocked by biofilms on the outer shell or other biocarriers. Apart from the biocarrier design, material transparency and color are also expected to show an effect on light availability and biofilm growth [ 51 ]. However, further investigations are needed to clarify the impact of the light limitation on biofilm production in the internal structures. In conclusion, no significant differences of biomass adhesion were observed on the tested untreated biocarrier materials. In contrast, fluorination, particularly by PP and PLA biocarriers, showed significant improvement in biofilm formation because of increased wettability. The higher biofilm attachment per carrier surface to optimized PAR biocarriers revealed great potential for further optimization of biocarrier geometry and thus biotechnological production of phototrophic biofilms."
} | 4,907 |
40269528 | PMC12018788 | pmc | 2,902 | {
"abstract": "Abstract Restoration of soil microbial communities, and microbial mutualists in particular, is increasingly recognized as critical for the successful restoration of grassland plant communities. Although the positive effects of restoring arbuscular mycorrhizal fungi during the restoration of these systems have been well documented, less is known about the potential importance of nitrogen‐fixing rhizobium bacteria, which associate with legume plant species that comprise an essential part of grassland plant communities, to restoration outcomes. In a series of greenhouse and field experiments, we examined the effects of disturbance on rhizobium communities, how plant interactions with these mutualists changed with disturbance, and whether rhizobia can be used to enhance the establishment of desirable native legume species in degraded grasslands. We found that agricultural disturbance alters rhizobium communities in ways that affect the growth and survival of legume species. Native legume species derived more benefit from interacting with rhizobia than did non‐native species, regardless of rhizobia disturbance history. Additionally, slow‐growing, long‐lived legume species received more benefits from associating with rhizobia from undisturbed native grasslands than from associating with rhizobia from more disturbed sites. Together, this suggests that native rhizobia may be key to enhancing the restoration success of legumes in disturbed habitats.",
"introduction": "INTRODUCTION Legumes (plants in the family Fabaceae) are integral parts of grassland communities that enhance biodiversity, provide resources for pollinators, provide high‐quality forage, and can improve soil quality through their interactions with nitrogen‐fixing bacteria (Koricheva et al., 2000 ; Potts et al., 2009 ; Tilman et al., 2001 ). Despite their importance, legumes (particularly long‐lived, slow‐growing late‐successional species) are underrepresented in grasslands restored via seed broadcasting relative to undisturbed remnants (Kindscher & Tieszen, 1998 ; Urban, 2020 ). Given that microbial mutualists play strong roles in plant establishment (Delavaux et al., 2021 , 2022 ) and can influence restoration success (Koziol et al., 2018 ), it is possible that native late‐successional legume establishment is limited in restorations by microbial mutualists. That is, communities of microbial mutualists may be degraded by anthropogenic disturbance of soils, which then affect the establishment success of these species. Strong evidence has accumulated that the establishment of high‐quality late‐successional prairie plant species can be limited by native arbuscular mycorrhizal (AM) fungi. AM fungi associate with most prairie plants, including most prairie legumes, and provide plants with many benefits, including increased access to phosphorus (Smith & Read, 2008 ). Late‐successional native prairie species generally benefit more from AM fungi (Bauer et al., 2018 ; Bryant & Bever, 2024 ; Koziol & Bever, 2015 ) and are more sensitive to AM fungal composition (Cheeke et al., 2019 ; Koziol & Bever, 2016a ) than early‐successional native or non‐native species. AM fungal composition has been shown to be degraded by disruption of prairie soil such as tillage (House & Bever, 2018 ; Jansa et al., 2002 ; Kabir, 2005 ), and late‐successional native prairie species are particularly responsive to AM fungi from undisturbed habitats (Koziol et al., 2022 ). Inoculation with AM fungi from remnant prairies enhances the establishment and growth of late‐successional plant species, including legumes, in disturbed sites (Koziol et al., 2022 ; Koziol & Bever, 2016b ; Middleton et al., 2015 ). Overall, this indicates that inoculation with native AM fungi can be used to enhance the restoration success of late‐successional, difficult‐to‐establish legume species. Less is known about the potential importance of nitrogen‐fixing rhizobium bacteria, a second group of microbial mutualists associated with legumes. Rhizobia can greatly enhance plant fitness by providing plants with fixed atmospheric nitrogen in exchange for carbon, which suggests that rhizobia have the potential to influence the establishment of legumes in restorations. There is some evidence that, like for AM fungi, land‐use change may degrade rhizobium communities. Long‐term nitrogen addition can lead rhizobia to evolve to provide fewer benefits to their hosts (Weese et al., 2015 ), which can lead to degradation of rhizobium communities in former agricultural fields. Grman et al. ( 2020 ) found that plants inoculated with microbial communities (including, but not limited to, rhizobia) from remnant prairies produce more root nodules (structures to house rhizobia) than plants inoculated with soil from site disturbed by agriculture, suggesting that disturbance may potentially decrease rhizobium quantity/quality. Similarly, inoculation with certain rhizobium strains improved legume establishment in a restored prairie (Beyhaut et al., 2014 ), indicating that high‐quality strains may be missing from disturbed sites. Whether potential degradation of the rhizobium community differentially inhibits late‐successional legumes compared to early‐successional species is unclear. In Grman et al.'s ( 2020 ) study, inoculation increased plant growth regardless of plant successional status. By contrast, Herzberger et al. ( 2015 ) found that growth of a late‐successional legume was greater when inoculated with microbial communities from recently restored (i.e., recently anthropogenically disturbed) sites than those from remnant prairies. Better understanding of legume interactions with rhizobium communities in degraded grasslands may be essential for improving legume establishment in these systems. In this study, we examine the responses of grassland legume species varying in life history (late‐successional native, early‐successional native, and non‐native species) to microbial and rhizobium communities from grasslands varying in land‐use history (remnant prairies, post‐agricultural grasslands, and agricultural fields) to better determine when and where rhizobia may be most useful in restoration efforts. Specifically, in a series of greenhouse and field experiments, we test the following hypotheses: (1) growth of late‐successional legume species will respond more strongly to rhizobia than that of early‐successional or invasive species; (2) legumes (particularly late‐successional legumes) will benefit most from rhizobia from undisturbed, remnant prairies; and (3) inoculation with rhizobia from remnant prairies will increase legume establishment in a post‐agricultural grassland restoration. By identifying patterns in legume responses to rhizobia, we can determine how rhizobia may best be used to enhance the establishment of legumes in degraded grassland systems.",
"discussion": "DISCUSSION Rhizobia can have large effects on legume fitness, with consequences for plant productivity and distributions. Here we find that land use, specifically agricultural practices, affects native rhizobium communities by altering nitrogen‐fixer relative abundance and community composition. These effects remain even after agricultural practices are abandoned, and affect legume survival and growth. These results suggest that native rhizobium communities do not recover on their own following disturbance and add to the growing body of evidence that rhizobium distribution may be limited by dispersal or co‐limited by the difficulty of simultaneous colonization of compatible host plants. Analyses of global patterns of legume distribution show that legumes that associate with rhizobia are less likely to establish as invaders in novel habitats than legumes that do not associate with rhizobia (Delavaux et al., 2022 ; Simonsen et al., 2017 ), likely due to the absence of compatible rhizobia outside their native ranges. Native rhizobia may similarly limit the establishment of native legumes in restoration, and inoculation with native legumes may be critical for enhancing the restoration success of legume species in disturbed habitats. Native rhizobia may be especially important for restoration given our findings that native legumes are more dependent on rhizobia than non‐native invasive legumes. In our greenhouse experiment, both late‐ and early‐successional native legume species that are commonly included in restoration seed mixes generally benefited more from association with rhizobia than did invasive species that are commonly found in post‐agricultural grasslands and restored prairies. This mirrors broad patterns in plant responses to AM fungi, where native plants generally respond more strongly than invasives (Koziol et al., 2022 , 2023 ), but we did not find response differences between late‐ and early‐successional native species, which have been shown for AM fungi (Bauer et al., 2018 ; Bryant & Bever, 2024 ; Cheeke et al., 2019 ; Koziol & Bever, 2015 , 2016a , 2016b ). However, we also found that late‐successional native legumes, which are often prioritized in grassland restorations, are more sensitive to rhizobia origin than early‐successional or invasive species. In the greenhouse, late‐successional natives received fewer growth benefits and made fewer root nodules when associating with whole microbial communities from agricultural sites, and they tended to also grow less with rhizobia‐only inoculation from these sites, suggesting that rhizobia that are particularly beneficial to late‐successional native legumes may be less abundant in these very recently disturbed sites. We found similar sensitivity differences in the field, where inoculation with native rhizobia from undisturbed sites, but not rhizobia from post‐agricultural sites, increased late‐successional legume survival but had no effects on the other legume species. Overall, this suggests that native rhizobia can provide benefits to late‐successional native legumes that do not extend to weedier, easy‐to‐establish legumes. Although our greenhouse and field experiments consistently show that native late‐successional legumes benefit from interacting with rhizobia, we found greater plant sensitivity to native rhizobia than post‐agricultural rhizobia in the field but not in the greenhouse. This could potentially be due to the short length of our greenhouse experiments (4 months), which may not have been enough time to observe differential effects on slow‐growing late‐successional species like we observed in the field, where significant effects of native rhizobia inoculation were not observed until the second year of the experiment. However, differential effects of remnant and post‐agricultural microbial communities were found in another greenhouse study of similar length with late‐ and mid‐successional legumes (Grman et al., 2020 ), indicating that growth responses can be observed in this timeframe. Alternatively, the lack of difference between the effects of microbial communities from remnant and post‐agricultural sites in our greenhouse studies may stem from the locations where we collected those communities. At each of the two locations we sampled (each with a remnant, post‐agricultural, and agricultural site), the remnant and post‐agricultural sites were in close proximity to each other (<50 m apart), and agricultural practices had been abandoned >60 years prior to sampling. Although the plant communities in the post‐agricultural sites differed from those in the remnant sites (higher cover of non‐native weedy species and fewer native species, S. M. Magnoli, personal observation), their microbial communities could be similar if microbes dispersed over the short distance from the remnant sites over time. By contrast, the remnant and post‐agricultural sites where we collected the rhizobia used in our field experiment came from locations where these sites were separated by larger distances (>150 m) and agricultural practices had ceased more recently in the post‐agricultural sites (ranging from <10–25 years ago). If microbial dispersal/recovery was less likely to occur in these post‐agricultural sites due to distance and time, it may explain why we observed differential effects of rhizobia from these sites in our field experiment. This suggests that when degraded sites are not directly adjacent to undisturbed remnant sites, inoculation with native rhizobia will likely be beneficial for native legume establishment. Differences between the results of our field and greenhouse experiments could also stem from the simplified greenhouse environment, where plants interact with mutualists in the absence of other biotic interactions such as competition and herbivory that are present in the field. This underscores the importance of considering context dependence when evaluating plant–mutualist interactions. Although our study focuses mainly on legume interactions with rhizobia, the fact that legumes are simultaneously interacting with other soil microbes can alter the importance of rhizobia. We observed, for example, differences in the magnitude of plant response to whole soil versus rhizobia only inoculation, which is consistent with rhizobia effects being at least partially influenced by the background soil microbial community. AM fungi are an obvious component of the soil microbial community that could alter rhizobium impacts on plant growth, as legumes, particularly perennial legumes, can receive synergistic benefits from associating with both mutualists at once (i.e., plants grow much larger than expected with both mutualists based on growth with individual mutualists) (Magnoli & Bever, 2023 ; Primieri et al., 2022 ). Inoculation with both native rhizobia and AM fungi could potentially have synergistic benefits to late‐successional legumes, thereby enhancing restoration success. In our field experiment, which had both rhizobia and AM fungi inoculation treatments, we observed strong growth responses of all legumes to native AM fungal inoculation, whereas rhizobia only benefitted late‐successional species survival and even had some negative effects on invasive species growth. These differences in mutualist effects could be due to environmental context dependence (e.g., soil fertility or water availability). Our observation of positive growth responses to native AM fungi of C. fasciculata in particular is consistent with previous tests (Reynolds et al., 2020 ). The high dependence of native and non‐native legumes on AM fungi is also consistent with previous field inoculation results in which both native and non‐native legumes benefit from these symbionts (Koziol et al., 2022 ). However, we found no evidence of synergistic effects of dual‐inoculation with both AM fungi and rhizobia. This was somewhat surprising, given that the late‐successional species in our experiment ( A. canescens ) had been shown to experience strong synergism when associating with both mutualists in a previous greenhouse experiment (Larimer et al., 2014 ). Annual legumes are less likely to experience synergism than perennial legumes (Primieri et al., 2022 ), which could explain why the early‐successional species in our field experiment tended to have lower growth when inoculated with both mutualists. Exploring under what conditions and for which types of legumes synergism may occur in grassland systems would improve our understanding of how to use both these important mutualists to increase restoration success. Restoration is essential to improving and enhancing grassland biodiversity and the ecosystem functions that grasslands provide. Legumes are integral components of grassland floras, and here we show patterns in how they interact with mutualistic rhizobia and demonstrate that rhizobia inoculations can be used to enhance the restoration success of a desirable legume species in a degraded habitat. While native AM fungi inoculants are increasingly used by restoration practitioners to enhance the establishment success of native plants in prairie restorations, our work demonstrates that including native rhizobia inoculants is also beneficial, particularly when restoring native late‐successional legumes. Future work should focus on long‐term field experiments that include more legume species in order to better generalize these findings and determine the best ways to use rhizobium inoculations as a tool to improve restoration success."
} | 4,103 |
31611703 | PMC6858561 | pmc | 2,903 | {
"abstract": "Synthetic microbial consortia have an advantage over isogenic synthetic microbes because they can apportion biochemical and regulatory tasks among the strains. However, it is difficult to coordinate gene expression in spatially extended consortia because the range of signaling molecules is limited by diffusion. Here, we show that spatiotemporal coordination of gene expression can be achieved even when the spatial extent of the consortium is much greater than the diffusion distance of the signaling molecules. To do this, we examined the dynamics of a two-strain synthetic microbial consortium that generates coherent oscillations in small colonies. In large colonies, we find that temporally coordinated oscillations across the population depend on the presence of an intrinsic positive feedback loop that amplifies and propagates intercellular signals. These results demonstrate that synthetic multi-cellular systems can be engineered to exhibit coordinated gene expression using only transient, short-range coupling among constituent cells.",
"introduction": "Introduction Synthetic biologists are now adept at engineering transcriptional gene circuits to create novel phenotypes within microbes. To date, a large variety of synthetic genetic devices have been developed, including toggle switches 1 , 2 , oscillators 3 – 5 , and logic gates 6 , 7 . In each of these, transcription factors and their cognate promoters are rearranged in order to regulate gene transcription within single cells. Intercellular signaling pathways have also been engineered to regulate gene expression in populations of cells 8 . To do this, synthetic biologists have generally used quorum sensing systems taken from Gram-negative bacteria that utilize N-acyl homoserine lactones (HSLs) 9 . Synthetic versions of rewired intercellular pathways have allowed synthetic biologists to generate multicellular systems that mediate intercellular communication 10 , mimic predator-prey systems 11 , display population-level oscillations 12 , 13 , and create spatial patterns 14 , 15 . Such coordination of gene expression across time and space through intercellular signaling pathways will be important if we are to use synthetic multicellular systems in complex environments such as soils or the gut microbiome, or interface with materials and bioelectronics. Existing multicellular synthetic microbial systems have been constructed to operate in either well-mixed 13 , 16 , or resource-limited environments 15 , 17 . Intercellular signaling is simplified in these environments, as either coupling between cells is uniform (in well-mixed environments), or cells quickly go to stationary phase (in resource-limited environments). However, as the size of synthetic multicellular systems increases it becomes important to consider environments that are not well mixed, i.e. those in which intercellular signaling via small molecules has a limited range within the population. Such considerations are important because large multicellular synthetic systems will need to coordinate their behaviors across both space and time. Some efforts have been made to coordinate gene expression in large synthetic colonies. For instance, Prindle et al. showed that oscillations between colonies of synthetic bacteria could be synchronized through engineered cell-cell communication mediated by hydrogen peroxide gas exchange 18 . However, gas exchange is not always the best option for cell-cell communication, as non-vaporous chemical means (especially HSLs) are more commonly accessible, and do not trigger native redox signaling pathways. Therefore, we need to better understand how to temporally coordinate gene expression in spatially extended bacterial communities using intercellular chemical signals. Here, we show that gene expression within a spatially extended synthetic bacterial consortium can be temporally coordinated through chemically mediated intercellular communication. We examine the dynamics of a two-strain synthetic bacterial consortium in which the two strains emit two orthogonal quorum sensing molecules to generate a regulatory network with linked positive and negative feedback. When co-cultured in a small (~100μm) microfluidic device, the two strains exhibit emergent transcriptional oscillations of genes within the synthetic network. We find that when these two strains are co-cultured in a spatially extended microfluidic trap (~2mm), synchronization of the entire population is possible even though the diffusion of the signaling molecules provides only short-range interactions among the cells. Through a combination of experimental perturbations and computational simulations, we find that the temporal coordination of gene expression in our system depends on the regulatory structure ( i.e. the presence or absence of various transcriptional feedback loops) of the network controlling intercellular signaling. In particular, a key positive feedback loop allows cells in an oscillating consortium to amplify signals locally. This amplification can reduce the phase difference between neighboring bacterial subpopulations, and is thus critical for synchronization across a spatially extended system. This is in contrast to smaller, well-mixed populations, which can exhibit synchronous oscillations for a variety of regulatory structures, including those without intrinsic positive feedback. We thus show how molecular interactions within individual cells are crucial for synchronization of spatially extended populations. Our findings represent a major step towards the creation of large, synthetic, multicellular systems. To be useful, such systems need to quickly coordinate their activity through a distributed network of local interactions. The mechanisms we describe show how to design communities of synthetic microbes to achieve this goal, and suggest how their counterparts in the wild have evolved to do so.",
"discussion": "Discussion We demonstrated that local coupling in synthetic bacterial communities can synchronize oscillations in a spatially extended population. Interestingly, in our system positive feedback within an activator strain of the population was necessary to achieve this. Without such feedback, the intercellular signals that couple the constituent cells are not amplified, and the limited spatial range of diffusion hinders the emergence of globally synchronous oscillations. Positive feedback plays a critical role in generating robust oscillations in synthetic oscillators constructed in single, isogenic cellular populations 4 , 28 . However, in a previous study we found that, in a small chamber, the addition of a negative feedback loop was needed to maintain robust oscillations in the face of fluctuating population ratios between activator and repressor strains of a consortium 13 . We have designed the spatially extended chamber to minimize such fluctuations in population ratios, and observed robust oscillations even in the absence of the negative feedback ( Figs. 1 and 2 ). Therefore, different features of the gene circuit architecture may be responsible for supporting robust local dynamics, and globally coherent dynamics on the population level. Various natural systems also achieve spatially coherent oscillations using only local coupling 29 . Interestingly, many of them appear to use intracellular positive feedback loops to achieve such synchrony, in accord with our findings. For instance, after starvation, a population of Dictyostelium is driven to aggregate via synchronous oscillations in cAMP signaling 30 . The spatial scales of the chemical gradients of cAMP are small compared to the size of the population, and thus distant cells communicate only indirectly. Dictyostelium has been shown to use a local positive feedback loop to amplify local signals: the extracellular cAMP inhibits the degradation of intracellular cAMP, which triggers local excitation ( i.e. amplification) 30 – 32 allowing for the signal to propagate more effectively. A second example is provided by the interlinked positive and double-negative feedback loops among Cdk1/Wee1A/Cdc25C that lead to the spatially coordinated mitosis in the large, fertilized X. laevis egg 33 . Finally, the master circadian clock located in the suprachiasmatic nucleus consists of ~20,000 individual oscillators coupled via various excitatory and inhibitory neurotransmitters. The circadian clock generates synchronized oscillations and functions as a timekeeper of our body 34 – 36 . Interestingly, each individual circadian oscillator has an intracellular transcriptional positive feedback loop mediated by Rors 37 , although its role in achieving globally coherent rhythms has not been investigated. These examples suggest that signal amplification through local positive feedback is used by a variety of biological systems to drive emergent behaviors in spatially extended systems. We have shown that a similar mechanism can be used to engineer spatially extended synthetic microbes that exhibit collective behavior even when they interact only locally."
} | 2,262 |
34746565 | PMC8567373 | pmc | 2,904 | {
"abstract": "Energy harvesting\nfrom natural resources has gained much attention\ndue to the huge increase in the demand for portable electronic devices\nand the shortage of conventional energy resources in general. In the\npresent work, the fabrication and realistic applications of a piezoelectric\nnanogenerator (PENG) using polydimethylsiloxane (PDMS) and the abundantly\navailable, environment-friendly natural fiber Sonchus\nasper (SA) have been discussed. The biocompatible,\nlow-cost SA fibers were flexible enough and showed high piezoelectric\nproperties as active materials in the study. The SA pappus based piezoelectric\nnanogenerator demonstrated its ability to convert the harvested biomechanical\nenergy into electrical energy from the various mechanical energy sources\navailable in our environment. The SA pappus/PDMS thin film based piezoelectric\nnanogenerator (SPENG) fabricated in the laboratory showed colossal\noutput performances (open circuit output voltage, V OC ∼81.2 V; short circuit current, I SC ∼1.0 μA) by continuous finger impartation.\nUniform output performance was also obtained by the application of\nuniform force on the devices (e.g., ∼42 V for 5 N force at\n10 Hz frequency). The SPENG was capable to charge a 2.2 μF capacitor\nto 3.2 V within a short time span (16 s) under continuous finger impartation\nand illuminate 39 commercial high-power blue LEDs that were connected\nin series. Thus, the fabricated SPENG can be used as a green and portable\nenergy source to power up portable electronic devices. Apart from\nthis, the SPENG may also be used as a self-powered energy supply for\npacemakers or different types of health care units if properly improvised.",
"conclusion": "Conclusions The observations of the studies indicated that the naturally and\namply available fiber of Sonchus asper pappus, being piezoelectric in nature, might be successfully employed\nto develop a bio-inspired nanogenerator (SPENG). The fabricated SPENG\nperformed quite efficiently to harvest waste energy from various mechanical\nand biomechanical movements. Electrical energy was produced by harnessing\nthe mechanical energy from various living and nonliving sources. This\nbiocompatible equipment was capable of generating a high output voltage\n( V OC ) of ∼81.2 V and a short circuit\ncurrent ( I SC ) of ∼1.0 μA\nunder periodic finger imparting and releasing (with an axial force\nof 13.5 N) with an average frequency of 6 Hz. The calculated instant\npower density of the device was 182.06 μW cm –3 . The fabricated SPENG showed remarkable response in exploiting the\nvibration energy of the vortex mixture (∼6 V) and low force\nimpact of the 9.8 g ball falling from 10 cm height on the surface\nof SPENG (∼15 V). The device was sensitive enough to respond\nto mere hand movement and finger touching on it. The realistic application\nof SPENG was verified by successfully lighting up to 39 blue LEDs\nconnected in series via finger impartation only and also by charging\na capacitor (2.2 μF) to 3.2 V within a very short span time\n(16 s). All the results were reproducible, and SPENG showed an exceptionally\ngood mechanical stability with brilliant performances for a year that\nindicated the excellent sustainability of the experimental device. The bio-piezoelectric material, SA pappus, collected from natural\nwaste material is a low-cost and easily and amply available green\nenergy material, which can be successfully utilized in several renewable\nenergy applications. The flexible, biocompatible, and biodegradable\nSPENG, made of SA pappus, has the prospect to be used in different\nhandy gadgets as an energy harvester, which can convert mechanical\nenergy from different sources easily available in the environment.\nAfter proper improvisation, the device may certainly be used in many\nin vivo applications as an energy harvester, which has been proven\nto be proficient enough to convert biomechanical energy, even when\nnegligibly small, into a usable form of electrical energy.",
"introduction": "Introduction To reduce the use of\nconventional power sources such as fossils\nfuels, firewood, etc., energy harvesting from living natural systems\nhas attracted more interest in the development of self-powered biocompatible\nand portable electronic gadgets such as mobile phones, implantable\nmedical appliances, roll-up displays, actuators, sensors, and wearable\nelectronic devices. 1 − 7 For low-energy devices, bio-based materials are considered to be\nsuitable alternatives over regular rechargeable batteries. 8 , 9 Nowadays, piezoelectric nanogenerators (PENGs) and triboelectric\nnanogenerators (TENGs) have been identified as promising green energy\nharvesting devices. These PENGs and TENGs can harvest electrical energy\nfrom various natural and artificial mechanical energy sources (such\nas human body movement, water flow, airflow, vibrations, and rotations)\nvia piezoelectric and triboelectric effects. 5 , 10 − 23 The triboelectric nanogenerator (TENG) has emerged as a newly\nevolved\nenergy harvesting device for transforming mechanical energy into electrical\npower due to the couple effect of contact electrification and electrostatic\ninduction. Due to its high efficiency, low cost, easy availability,\nand eco-friendliness, TENG is applicable in harnessing small mechanical\nenergy and large-scale energy as well. 24 , 25 Recently,\nPENGs have attracted additional interest of scientists\ndue to their flexibility, mechanical durability, high performance,\nsensitivity, etc. 22 , 23 , 26 To upgrade PENGs, various piezoelectric ceramic and metal oxide\nmaterials such as ZnO, PMN–PT, (Na, K)-NbO 3 , BaTiO 3 , and PZT 27 − 30 have been used. Apart from the piezoelectric ceramic and metal oxide\nnanomaterials, certain polymers such as polyvinylidene fluoride (PVDF), 31 its copolymers with trifluoroethylene [P(VDF-TrFE)], 32 and poly(vinlyacetate) (PVAc) 33 also show high piezoelectric properties. The announced\nmaterials have some limitations because of their toxicity, high cost,\nnonbiodegradability, and breakability, along with the lengthy and\ndifficult synthesis process. 5 To address these problems,\nit is necessary to design devices using biodegradable, low-cost, and\nflexible materials for upcoming days. Therefore, amply available natural\nand nontoxic bio-piezoelectric materials have attracted more interest\nin the development of eco-friendly smart devices. 34 − 40 Many bio-based PENGs, using a virus, 41 fish scale, 42 fish bladder, 43 cellulose, 44 paper-BaTiO 3 -bacterial cellulose, 45 cellulose-BaTiO 3, 46 biowaste crab shell, 6 and biowaste onion skin 5 as piezoelectric materials, have been studied for harnessing green\nenergy. Recently, biowaste material based PENGs have been reported\nwith\ncomparatively low output performances. Maiti et al. 5 described\ntheir study on an onion skin based PENG that showed an output voltage\nof ∼18 V, short circuit of ∼166 nA, and instantaneous\npower density of ∼1.7 μW cm –2 . A fish\nscale based PENG with an output voltage of 10 V and a short circuit\ncurrent of 51 nA was reported by Ghosh and Mandal. 42 Due to their cost-effectiveness and recyclability, biowaste\nmaterials have attracted many researchers for harvesting electrical\nenergy. Thus, we have designed a simple and cost-effective biowaste\nmaterial based piezoelectric nanogenerator with high output performance,\nflexibility, high durability, and nontoxicity. The environment-friendly Sonchus asper (SA) is the strongest natural polymer\nfiber. 47 SA is known to possess diuretic,\nrefrigerant, sedative,\nand antiseptic properties and is mainly used in medical treatment. 48 Various chemical studies exposed the presence\nof ascorbic acid, carotenoids, and fatty acids in SA. 49 Also, the highly crystalline SA pappus forms stress-induced\npolarization and generates piezoelectricity. In our study, we\nfabricated a PDMS/SA microfiber thin film based\npiezoelectric nanogenerator (SPENG) for harvesting energy from a small\namount of mechanical work such as walking, talking, bending, stretching,\npulse vibration, and finger imparting. SA pappus was chosen as the\nkey component of SPENG as it is not only an abundantly available biowaste\nmaterial but biocompatible, biodegradable, nontoxic, robust, sustainable,\nand renewable also. The PDMS/SA pappus film based PENG, named SPENG,\nshowed a maximum output voltage ( V OC )\nof 81.2 V and a short circuit current ( I SC ) of 1.0 μA, with an instantaneous power density of 182.06\nμW cm –3 during continuous finger impartation.\nMoreover, the fabricated PENG was capable of producing output voltage\neven in the case of low pressure of heart beats and pulses. The study\nindicated the prospect of the nanogenerator SPENG as a green and portable\nenergy harvester to power up portable electronic devices. Furthermore,\nafter proper improvisation, the fabricated nanogenerator might be\nused as an energy supply for pacemakers and other types of health\ncare devices.",
"discussion": "Results and Discussions Output Performance The open circuit output voltage\n( V OC ) and short circuit current ( I SC ) generation through human finger imparting\nand releasing of an axial force of ∼13.5 N with ∼5 to\n6 Hz frequency is depicted in Figure 1 a,b. Different output voltages were generated by the\nSPENG depending on the application of various forces and frequencies.\nWe studied the voltage output performance of the SPENG by applying\na uniform force of 5 N under varied frequencies ranging from 2 to\n10 Hz by using a self-designed instrument (from Ocean College, Zhejiang\nUniversity, PR China). Figure 1 (a) Open circuit voltage ( V OC ) and\n(b) short circuit current ( I SC ) generated\nby human finger impartation at a frequency of ∼5 to 6 Hz. (c)\nFrequency-dependent V OC . (d) Magnified\nview of V OC for two successive cycles\ncorresponding to 10 Hz frequency under 5 N force. Voltage generation\nby using the vibration energy of an (e) air drier and (f) vortex machine\nat different speeds. With increasing frequency\nof the applied forces, the output performance\n( V OC ) increased gradually from ∼7.5\nV at 2 Hz to ∼36 V at 10 Hz ( Figure 1 c). The output voltage of the SPENG depended\non the impedance and applied strain frequency. Even at low-pressure\nfinger impartation, the device showed high sensitivity. The maximum\noutput voltage (observed by Keysight, Oscilloscope DSO-X 3012A) and\nshort circuit current (recorded by Keysight, Electrometer B2985) of\nthe SPENG were V OC ∼81.2 V and I SC ∼1.0 μA, respectively, as shown\nin Figure 2 a,b. It\nis evident from the figures that the values of both V OC and I SC were non-uniform,\nwhich differed slightly for periodic pressing and releasing by finger. Figure 2 (a–e)\nOpen circuit voltage generated by dropping balls of\n108.793, 35.214, 9.18, 59.34, and 15.01 g on the device from a height\nof ∼10 cm. (f and g) Output voltage generated by forefoot and\nheel pressing. (h) Wrist bending signal. (i) Light finger touching\non SPENG. Such variation was due to the\nmanual inconsistency of the applied\nstrain on the energy scavenging devices during finger impartation.\nThe enlarged view and working process of the nature of growth and\ndecay of V OC for two successive cycles,\ngenerated by instrument impartation at 10 Hz frequency, are shown\nin the Figure 1 d. The\nhighest positive peak was obtained when the strain was applied. After\nwithdrawal of the strain, the device recovered its original state,\nand consequently, the negative peak appeared. The output performance\nof SPENG is greater than that of any other previously reported biomaterial-based\ndevices, given in table form in the Supporting Information ( Table S1 ). The fabricated device was also tested\nby bending and twisting and the output voltage performance was recorded,\nbut no remarkable voltage generation was found, proposing that there\nwas negligible triboelectric effect. Also, switching polarity data\nhave been taken and represented in Figure S2 in the Supporting Information. Recently, Lee et al. 41 presented a virus-based PENG with an output voltage of\n∼0.4 V and current of ∼4 nA. Ghosh and Mandal 43 reported a piezoelectric energy harvester derived\nfrom a fish swim bladder with an output voltage of ∼10 V. Maiti\net al. also reported an output voltage ( V OC ) of 18 V and a short circuit current of 166 nA in the case of an\nonion skin based PENG. Hoque et al. 6 also\nfabricated a biowaste crab shell chitin nanofiber-based PENG with\nan output voltage ∼49 V and a short circuit current ∼1.9\nμA. Thus, the output performance of the SA pappus based piezoelectric\nenergy harvester, SPENG, was definitely better than that of previously\nreported similar bio-based devices. The SPENG also showed a high power\ndensity of ∼182.06 μW cm –3 . The details\nof power density and external force calculation of the fabricated\nSPENG are in the Supporting Information section. Energy harvesting from different vibration energies\nof an air drier\nand vortex machine in vibrating condition was also tested. Placing\nthe SPENG under the air drier and operating the air drier at three\ndifferent settings of low, medium, and high speed, SPENG produced V OC ∼0.2, ∼0.42, and ∼0.6\nV, respectively, as shown in Figure 1 e.When attached to a vortex machine body, the SPENG\nshowed remarkable voltage output of ∼1.3, ∼2.4, ∼3.25,\nand ∼4.1 V depending on the vortex rotating speed of low, medium,\nmedium high, and high speed, respectively, as presented in Figure 1 f. All the above-mentioned\nexperimental findings show that the SPENG can harvest electrical energy\nquite efficiently from any kind of mechanical energy generated by\nvibration. Ultrasensitivity of the SPENG under Different\nBody Movements The fabricated SPENG was also capable of harvesting\nmechanical\nenergy from the various body movements. Furthermore, the sensitivity\nof the fabricated device was tested by dropping balls of different\nmasses from a height of 10 cm on the device surface, shown in Figure 2 a–e. The output\nvoltages were ∼6, ∼10, ∼12, ∼15, and ∼17\nV for dropping balls of masses 108.793, 35.214, 9.18, 59.34, and 15.01\ng, respectively. When balls were dropped on the device, different\noutput voltages were produced depending on the masses of balls and\nthe contact surface area. The ball with the maximum mass showed a\nlow output voltage compared to the other balls due to the harder (lower\nelastic properties than other balls) surface of the bigger-size ball.\nFor this reason, different pressures were generated on the nanogenerator,\nwhich resulted in different output voltages. Figure 3 f–i indicates that the\nself-powered flexible SPENG device was sensitive enough to utilize\nbiomechanical energy from various body movements like heel and forefoot\npressing, wrist up–down movement, and just finger touch. High\noutput voltages were obtained during forefoot pressing (∼36\nV) and heel pressing (∼15 V) by attaching the device to the\nfoot and heel, depicted in Figure 2 f–g. The generated output voltage during forefoot\npressing was greater than that during heel pressing. A possible reason\nbehind this result may be the greater contact surface area for the\nforefoot with device than the heel with device surface. The maximum\noutput voltage during wrist movement is ∼1.9 V, as represented\nin Figure 2 h, which\nalso confirms the capability of the SPENG device in harvesting green\nenergy from body movement. The sensitivity was further checked by\njust touching the device with a finger, where V OC was ∼0.65 V ( Figure 2 i). The results indicate that the fabricated SPENG\nis competent enough to detect even a minute amount of pressure and\nconvert the mechanical energy, caused by the application of pressure,\ninto electrical energy. The output voltage was recorded each month\nover the course of a year to check the stability of our device, and\nresults suggested that the experimental device possessed almost the\nsame performances throughout the year. Thus, the fabricated SPENG\nis cost-effective, biocompatible, flexible, and sustainable, which,\nafter proper improvisation, can charge biomedical devices by utilizing\nthe biological pressure from various body movements, muscle contraction/relaxation,\nand blood circulation. Figure 3 (a) Charging performance of the 2.2 μF capacitor;\ninset:\ncircuit diagram for the charging capacitor and LEDs light up and (b)\nsnapshot of LEDs in glowing condition. Moreover, the experimental SPENG can be used to harness energy\nfrom a wide range of green biomechanical energy resources by attaching\nthe SPENG to running machines like treadmills, dancing floors, interior\nof shoes, wheels of cars, etc. The electrical energy produced can\nalso be used to switch on LEDs, LCD screens, and other different types\nof battery charging equipment. Working Principle of the\nFabricated SPENG The performances\nof SPENG can be explained with the help of the synergetic effect of\nthe electro activity. Most of the biomaterials such as cellulose,\ngelatin, chitin, and collagen exhibit shear piezoelectricity due to\ntheir fibrous configuration. Such piezoelectric effect is due to the\ninternal rotation of polar atomic groups, which are interrelated with\nthe asymmetric carbon atom. 50 , 51 SA pappus is composed\nof an amide linkage, −OH groups, and carbonyl groups arranged\nin α-helical and β-sheet structures. In the SA pappus,\nα-helical and β-sheet elastic structures of amino acids\nare connected strongly through intra- and intermolecular H-bonding\namong themselves. 30 Due to the high aspect\nratio of SA pappus fibers in conjunction with the strong hydrogen\nbonding energy and high binding energy, these fibers can be used undoubtedly\nas bio-piezo materials. The presence of amide and hydroxyl groups\nassists SA pappus to produce electric dipoles inside the fiber. These\ndipoles show a high piezoelectric response under applied external\nstrain. Under mechanical compression, highly crystalline SA pappus\nforms stress-induced polarization and generates piezoelectricity.\nThus, SA pappus is a promising piezoelectric material that has the\ncapability to generate output voltage under applied mechanical compression. The working principle of the experimental device (SPENG) depends\nprimarily on the changes in the molecular structure of SA pappus under\nmechanical stress. Application of vertical pressure on the PENG causes\nthe movement of total polarization charge toward the electrodes, which\nacts as the key for the generation of piezoelectric potential across\nthe two electrodes. 52 Thus, the applied\nvertical force eventually initiates the accumulation of positive and\nnegative charges separately on two electrodes and thus generates a\npositive voltage signal. The polarization charge increases with increasing\nmagnitude of mechanical force and frequency of vibration. When the\nvertical strain is released, the collected charges on the electrodes\nmove toward the reverse direction, and hence, a negative signal is\nproduced. Due to the coincidence of the negative signal and weak damped\npiezoelectric potential, we observe the first negative potential peak.\nAgain, a small positive signal, indicated by the second positive peak,\nis observed due to the assemblage of free charges at the electrodes.\nAfter returning to the original state, the accumulated electrons fall\nback again and give a second negative peak. The second positive and\nnegative signal peaks occur due to the damping effect of devices.\nThus, a continuous pulse or the application of an external force is\nthe main source to obtain the output performances of PENG. For\nthe piezoelectric effect, σ p ( z )\n= piezoelectric polarization charge density of PENG and σ( x ) = free electron charge density in the electrode. Considering the strain introduced by the applied compression, σ( x ) is a function of the thickness of the device ( z ). Now, for the mechanical force, the piezoelectric equation\nand constituter are given by: 1 2 3 where ( e ) ijk denotes the piezoelectric third-order tensor\nand S is the mechanical strain. C E and T are the elasticity tensor and\nstress tensor,\nand ε is the dielectric tensor. Due to media polarization,\nthe displacement current can be expressed\nas follows: 4 Equation 4 illustrates\nthat the changing rate of the\napplied strain is proportional to the output current density of the\nPENG. The frequency dependence of output performance of the device\n(shown in the Figure 1 c) matches well with eq 4 . Realistic Application of SPENG The ability of the fabricated\nSPENG to charge a 2.2 μF capacitor was investigated by connecting\nthe device with the capacitor through a simple bridge rectifier circuit\n( Figure 3 a inset) and\nswitching on the SPENG by periodic finger impartation at a frequency\n∼5–6 Hz. The exponential nature of the time–voltage\ncurve during the charging ( Figure 3 a) of the capacitor indicated a high energy storage\ncapability. Using SPENG, the capacitor (2.2 μF) was charged\nup to 3.2 V within 16 s under periodic finger impartation. So, we\ncan easily state that the charging efficiency of SPENG is higher than\nthat of any other previously reported biowaste-based PENGs. Since\nthe charging ability of SPENG is very high, it can be used as an alternative\nenergy source for portable or medical devices via capacitor charging. Figure 3 b shows the competence\nof the SPENG to light up LEDs under continuous finger impartation.\nThe power produced from SPENG can illuminate 39 blue LEDs connected\nthrough a full wave bridge rectifier. The video is in the Supporting\nInformation ( Video S1 )."
} | 5,363 |
36440170 | PMC9685601 | pmc | 2,905 | {
"abstract": "A novel contact–separation triboelectric generator\nconcept\nis proposed in this study, which is composed of a double-sided tape\nwith acrylic adhesive material and a metalized polyester (PET/Al)\nfilm (an aluminum layer coating on one side). The proposed concept\nis very cost-effective and easy to fabricate compared to existing\ntriboelectric nanogenerators (TENGs), which require special equipment\nand sophisticated procedure to build. The strong bonding nature of\nacrylic adhesive on the tape induces a significant charge when contacting.\nThe peak power generation depends on the induced pressure at the impact.\nDuring the separation phase, the air breakdown between triboelectric\nlayers allows most existing electrons to flow back from the ground\ndue to rapid charge removal at the interface. A higher voltage can\nbe generated when the PET is interfaced with the double-sided tape\ncompared to the Al-acrylic configuration because of the effect of\ntriboelectric series and a Schottky barrier formation for electrons\nat the tape–Al interface during contact. A double-electrode\nconfiguration with an assembly of Al/PET–tape–PET/Al\nsignificantly improved the performance, in which a 21.2 mW peak power\nis achieved compared to 7.6 mW in the single-electrode design with\ntape–PET/Al assembly when excited at 20 Hz in a shaker test.\nThis double-electrode triboelectric generator can power 476 LEDs with\nan active area of 38 mm × 25 mm. Moreover, a direct power of\na 650 nm laser diode was demonstrated. In summary, the proposed triboelectric\ngenerator concept using tacky materials shows the potential for higher-energy\nharvesting via triboelectrification and advances the state of the\nart by offering low cost and easy fabrication options. It is expected\nthat such newly proposed triboelectric generators are able to meet\npower requirements in many engineering applications.",
"conclusion": "Conclusions In this paper, a simple triboelectric generator\nconcept was proposed,\nevaluated, and demonstrated, which is composed of a double-sided tape\nand a PET/Al sheet. Power generation is comparable with the current\nstate of the art of TENG devices. Key conclusions are summarized below: This is the first attempt to use tacky materials in\nthe triboelectric generator. This novel\ndouble-sided tape triboelectric generator\nshows the potential to improve the performance of energy harvesting\nvia triboelectrification. Such a simple\ndesign leads to easy fabrication and integration,\nwhich only requires a craft-level skill compared to the nanotechnology-based\nsophisticated fabrication scheme used in current TENGs. In polymer-based adhesives, the bonding strength depends\non the electrical double layer at the interface with positive and\nnegative charges on each side. The strong\nbonding nature of acrylic adhesive in a double-sided\ntape induces a significant charge when contacting the PET layer. Air breakdown between triboelectric layers\noccurs during\nthe separation phase to enable rapid charge removal, which actually\nhelps electrons to flow much easier. The first attempt to power a 650 nm laser diode using\nour TENG, which is crucial for optoelectronic devices and sensor applications.",
"introduction": "Introduction Various adhesives have been developed\nfor thousands of years using\nanimal parts, natural rubbers, and recently synthetic materials to\nbond different objects. From the late 19th and the early 20th centuries,\nadhesive tapes, which contain adhesive-coated backing materials, have\nbeen introduced. Today, adhesive tapes are one of the essential items\nfor everyday life. In the 21st century, these tapes, specifically\npressure-sensitive tapes, which can bond when pressure is applied,\nhave been used in various applications. Nobel Prizes in physics were\ngiven to Andre Geim and Konstantin Novoselov for “groundbreaking\nexperiments regarding the two-dimensional material graphene.” 1 Common pressure-sensitive tapes were used for\nseparating a graphene layer from bulk graphite in their research.\nResearchers from UCLA were able to generate X-rays by unrolling pressure-sensitive\ntapes in vacuo , which is called “triboluminescence.” 2 Accelerated electrons generate Bremsstrahlung\nX-rays when they strike the positive side of the tape during tape\nunrolling. The intensity of the X-ray was enough to take a radiogram\nof a human finger. In this paper, we will show that pressure-sensitive\ntapes can also be utilized to generate electricity using the triboelectric\nphenomenon as small-scale energy harvesters. Electrical power generation\nby a small-scale energy harvesting device is needed to power low-power\nprofile electronic systems in many different applications. 3 − 12 These include structural health monitoring, wearable sensors, and\nenvironmental monitoring systems. To minimize the environmental impact,\nthese should be built as stand-alone systems with the capability of\nharvesting ambient mechanical energy and converting it into electrical\npower. Triboelectricity, defined as generating power by electrostatic\ncharges through a friction of two surfaces with different materials,\nhas been introduced for being used as an energy harvesting device.\nSince its invention in 2012, the triboelectric nanogenerator (TENG)\nis one of the most promising candidates for small-scale energy harvesters. 8 , 9 , 13 , 14 This process allows converting mechanical energy into electrical\nenergy using the triboelectric effect and electrostatic charges. Triboelectric\nlayers can be selected from the various combination of materials from\nthe triboelectric series to maximize the power generation. For example,\nCu and poly(tetrafluoroethylene) (PTFE: Teflon) combination/poly(methyl\nmethacrylate) (PMMA) and polyimide combination can be used for positive\nand negative triboelectric layers. 15 , 16 Those combinations\ncan generate electrical currents by touching each other like contact–separation\nand sliding. Previously, researchers have developed TENGs with many\ndifferent modifications such as circuitry designs, different triboelectric\nmaterials, different methods of contacting, and different atmospheres.\nThe improvement of TENG performance has been conducted in this way\nto increase the charge density to achieve higher power density. For\nexample, instead of a normal direction contact–separation,\nthe sliding of triboelectric layers is also commonly used to generate\nelectricity. 17 Even though there have been\nmany significant improvements in the performance of TENGs, to design\nand build a TENG is complicated. Therefore, it is very costly to integrate\nTENG in real applications due to the sophisticated structure layout,\ncircuitry optimization, and fabrication scheme. 8 , 9 , 13 One challenge is mostly avoiding the air\nbreakdown problem between two triboelectric layers. For example, researchers\nfabricated a main TENG device with a charge excitation TENG. 16 Moreover, voltage multiplying circuits (VMCs)\nwith Zenor diode were used to boost and stabilize the output voltage. 18 With this design, the maximum current density\nof 1.25 mC/m 2 and the maximum power density of 38.2 W/m 2 were obtained at a load resistance of 4 MΩ and a frequency\nof 4 Hz. The charge density of 2.20 mC/m 2 and the power\ndensity of 40 W/m 2 are achieved using a ferroelectric P(VDF-TrFE)\nlayer and a VMC circuit. 18 Even though\nthe excitation TENG was no longer used, the ferroelectric layer needs\nto be integrated into the design with all of the different layers.\nVMC is still required for the TENG system. Recently, the highest power\ndensity of a TENG at 115.6 W/m 2 with an external charge\nexcitation module to avoid the air breakdown between triboelectric\nlayers was established with a carbon/silicone gel electrode in the\ncontact–separation TENG. 19 Therefore,\nthe design of contact–separation-based TENGs shows a high complexity\nfor harvesting electrical power from environmental sources. Moreover,\nthe fabrication process for a complete TENG system is preferred to\nnot give a large environmental impact by avoiding the chemical and\nevaporation processes. To advance the development of TENG devices\nin various engineering\napplications, we need to address two major challenges, 20 which include the enhancement of power density\nin triboelectric series and the simplification of fabrication. Detailed\nquantification on the triboelectric properties of a wide range of\nmaterials was conducted to serve as a standard for implementing the\napplication of triboelectrification for energy harvesting. 21 , 22 One could build a TENG device by exploring those material candidates.\nHowever, the fabrication difficulties involved in creating microstructures\nor special patterns on the surface of selected triboelectric materials,\nthe selection of special equipment, and carefully designed fabrication\nscheme as observed in the typical TENG device development make this\ndifficult for practical uses. In this paper, a novel triboelectric\ngenerator concept is proposed,\nin which a simple design was adopted by using a conventional double-sided\ntape (with acrylic adhesive layer) and a metalized polyester film\n(i.e., an aluminum layer coating on one side) (PET/Al) to serve as\ntriboelectric layers. It is worth noting that no special fabrication\nscheme is required in the current design. Power generation is achieved\nvia a contact–separation motion, as shown in a similar TENG\ndevice. During the contact phase, the proposed triboelectric generator\nbehaves similarly as observed in a typical contact–separation\nTENG device, in which charges are developed on the interface between\na double-sided tape and a PET/Al layer. However, the strong bonding\nnature of acrylic adhesive on the tape induces a significant charge\ncompared to that of a non-tacky triboelectric layer. During the separation\nphase, the air breakdown via an electric spark occurs at the interface.\nSuch rapid charge removal in triboelectric layers allows existing\nelectrons from the ground flow back to the aluminum layer to realize\na different mechanism compared to the capacitance behavior, as observed\nin a typical TENG device. The tackiness in the adhesive layer contributes\nto the improved energy generation during the contact and separation\nphase in the current design. Output power also depends on the induced\npressure during the impact in a contact–separation motion.\nIt has better efficiency for contacting PET to a tape surface compared\nto contacting Al to a tape surface because of the effect of triboelectric\nseries as well as a Schottky barrier formation at the aluminum interface.\nBoth single-electrode and double-electrode designs were investigated.\nA single-electrode design involves a double-sided tape and a PET/Al\nlayer, in which the aluminum side is served as one electrode along\nwith a ground connection. A double-electrode design leads to a three-layered\nconfiguration in which a double-sided tape layer is sandwiched between\ntwo PET/Al sheets and both aluminum sides serve as electrode connections.\nComprehensive shaker tests from 15 to 25 Hz were conducted to evaluate\nthe performance of the proposed triboelectric generators with double-sided\ntape materials. We are able to obtain a 21.2 mW peak power in a double-electrode\ndesign and 7.6 mW in a single-electrode design, respectively, in which\nthe excitation frequency is at 20 Hz. Using this double-electrode\nsystem, a number of different manually\noperated generator prototypes were designed and developed, including\na spring load prototype, which produced 169.6 W/m 2 while\noperating approximately at 2 Hz, and a V-shaped prototype, which was\ncapable of powering 476 LEDs with an active area of 38 mm × 25\nmm. Moreover, we are able to directly power a 650 nm laser diode using\nthe double-electrode triboelectric generator driven by a shaker at\n20 Hz. In summary, the first contact–separation triboelectric\ngenerator\nwas proposed, prototyped, and evaluated by exploring tacky materials\nas a triboelectric layer. Comparable performance is achieved in our\nsimple design using double-sided tape material.",
"discussion": "Results and Discussion Figure 1 c,d shows\nthe voltage and the current time history under 20 Hz excitation during\na shaker test, in which a 4 MΩ load resistor is used because\nit shows the optimum power generation, as described in the later part\nof the paper. Interestingly, the voltage signal shows a large negative\nvalue of −166.17 V during the contact phase. Different polarity\nof voltage amplitude is observed in the separation phase, and its\npeak amplitude is 122.67 V. Compared to the existing contact/separation\nTENGs in the literature, the opposite polarity of voltage occurs instantaneously\nat the initiation of separation. In those studies, it was shown that\nsuch positive voltage occurs at a later stage with almost a half-cycle\ndelay. 16 , 19 However, as shown in Figure S2 , our generator shows amplitude inversion at the\nmoment between the contact and separation. This strongly indicates\nthat rather than the field-induced charge movement as observed in\ntriboelectric layers, charges in our system instantly interact during\nthe operation. 24 It will be explained in\nthe later part of this paper. The amount of power, which is a product\nof voltage and current, was plotted as a function of time, as shown\nin Figure 1 e. As aforementioned,\nour power calculation was obtained by multiplying the measured voltage\nfrom a 100 Ω resistor, and the peak value is 7.6 mW. The corresponding\npower density is 6.2 W/m 2 by assuming a 35 mm × 35\nmm contact area. Impressively, our current triboelectric design with\na double-sided tape and a PET/Al sheet achieves a comparable power\ndensity to existing TENGs in the literature. Note that the charge\ninteraction plays a crucial role in our simple triboelectric generator\ndesign. Figure 2 a shows\nthe charge development during the contact phase. Similar to existing\nTENGs, a formation of an electrical double layer occurs at the interface\nbetween two triboelectric layers. Accordingly, the double-sided tape\ngenerates positive charges, while PET generates negative charges after\nthe contact due to the electrification in triboelectric series. 20 , 25 , 26 Additionally, based on the electrostatic\ntheory of adhesive, the contact between polymer adhesive materials\nand adherend generates a strong attractive force. 27 Since the adhesive bonding strength depends on the attractive\nforce in the double layer, a higher charge generation can be achieved\ncompared to the configuration using a non-tacky triboelectric layer\nin the existing TENGs. Since the PET layer is negatively charged,\nelectrons in the Al layer on the PET will experience a repulsive force\nso that those electrons will flow to the ground. Therefore, a negative\nvoltage occurs during the contact process, as shown in Figure 1 c. In the separation process,\nthe discharging behavior is quite different compared to the case in\nexisting TENGs. Instead of the field-induced phenomenon as experienced\nin existing TENGs, an air breakdown induces a sudden electron removal\nin the PET layer. Such air breakdown causes neutralization of the\nPET layer so that electrons can flow back from the ground to the Al\nlayer with positive charges, as shown in Figure 2 b. 19 , 28 − 30 As a result, a positive voltage appears during the separation. From\nthe electrostatic theory of adhesive, it is expected that the separation\nof adhesive will lose charges by field emission and gas ionization. 27 The actual electric spark generation was observed\nand will be discussed in the later part of this paper. Figure 2 Schematics of a triboelectric\ngenerator. (a) In contact, the charge\ngeneration occurs at the interface between the double-sided tape and\nPET layer. (b) In separation, because the double-sided tape and PET\nwill give an electric spark, almost all negative charges are gone\nin the PET layer. The remaining positive charges in Al will be compensated\nby electrons from the ground. As shown in Figure 3 a, the power is a function of the resistance load and\ndriven frequency\nin the shaker test. It is very important to determine the frequency\nand resistance load effects on the performance of our triboelectric\ngenerator. A 4 MΩ resistance load yields the peak power under\nall frequencies, which corresponds to the impedance matching condition. 16 , 31 Also, the power generation shows frequency-dependent characteristics.\nA higher power value was observed around 20–22 Hz. As aforementioned,\na load transducer was used to record the force during the impact.\nAs shown in Figure 3 b, the pressure was calculated using the measured force data under\ndifferent driven frequencies. A higher pressure (force/unit area)\nvalue by the resonance frequency at 20–22 Hz corresponds to\nhigher power generation, which indicates that a higher force generates\ncharges more rapidly at a given time in the double layer during the\ncontact phase. As a result, different frequencies with various pushing/pulling\nforces during a contact–separation process determine how fast\nelectrical charges are generated through the system. Namely, if the\ncontact and the separation processes occur with a faster impact, the\nelectrical charge generation and dissipation occur abruptly, which\nresults in having higher amplitudes. Figure 3 (a) Power generation with different load\nresistances at different\nfrequencies from 15 to 25 Hz. Since the natural frequency of the system\nis between 20 and 22 Hz, higher peak power generation occurs at this\nfrequency range. (b) Comparison of peak power generation and applied\npressure for triboelectric generator as a function of frequency. In\nthe natural frequency range of system, the applied pressure increases\nas resonance occurs so that the generation of power increases significantly. To further understand the effect of triboelectric\nseries on power\ngeneration, two configurations were evaluated, in which a double-sided\ntape is interfaced with either PET or Al side. Figure 4 c shows the time history of collected voltage\ndata, as plotted in black for the case with the interface between\na double-sided tape and PET, and in red for the case with the interface\nbetween a double-sided tape and Al. The amount of power generation\nfor both configurations is shown in Figure 4 d. Higher peak power is observed in the case\nof the interface between a double-sided tape and PET due to the triboelectric\nseries and Schottky effects. The triboelectric difference between\na double-sided tape (i.e., acrylic adhesive layer) and a PET layer\nis higher than the case between a double-sided tape and an Al layer.\nWhen a double-sided tape layer is interfaced with an Al metal layer,\nelectrons are created on the Al side during the contact phase. However,\nsuch electrons must overcome the Schottky barrier back and forth in\na way of thermionic emission at the interface. 27 , 32 It is well known that the thermionic emission of metal will decrease\nthe efficiency of triboelectric generators. 32 Since electrons are generated, an n-type Schottky barrier needs\nto be considered as it is defined as a product of the metal work function\nsubtracted by the electron affinity of material at the interface.\nThis strongly indicates that the electrostatic bonding nature between\nacrylic adhesive and Al is significantly less strong than the case\nwith the interface between a double-sided tape and a PET sheet due\nto Schottky barrier formation at the metal interface. 27 It was also discussed that the negatively charged dielectric\nlayer with higher work function (acrylic work function: 5.5 eV) than\nmetal (Al work function: 4.2 eV) gives electrons to a metal after\nforming a contact and aligning the Fermi level. 21 , 33 , 34 Figure 4 (a) Shaker operating triboelectric generator\nwith Al/PET/tape–PET/Al\ncombinations. During its operation, an electric spark occurs between\nthe double-sided tape and PET/Al. This strongly indicates that the\ndischarge between triboelectric layers occurs. (b) Within 33 ms after\ndischarge (1 frame), the Al layer brings electrons from the ground\nbecause it still has the remaining positive charges. Therefore, lighting\nup of 296 LEDs is confirmed. (c) Comparison of voltage amplitude between\ndouble-sided tape–PET /Al(black) and double-sided tape–Al/PET\nconfigurations (red). Due to the Schottky barrier height for tape/Al\ninterface, less electrons will overcome the barrier thermionically.\n(d) Consequently, power generation is significantly lower for double-sided\ntape–Al/PET combinations (red) compared to that for double-sided\ntape–PET/Al(black). Knowing that PET made a better contact surface\nfor the double-sided\ntape than aluminum, a double-electrode design was proposed to improve\nthe performance of our triboelectric generator, as shown in Figure 5 a, in which an additional PET/Al sheet is used to form a double-electrode\nconfiguration. Similar to the single-electrode design as shown in Figure 2 a, we expect to utilize\nthe electric charges on both sides of the double-sided tape during\nthe contact and separation. Two separate electrical circuits were\nbuilt to collect the voltage data separately. During the phase, the\nassembly on the right-hand side will generate positive charges in\nthe tape layer and negative charges in the PET layer, as shown in Figure 5 b. The Al electrode\nlayer will flow electrons to the ground, as previously described in Figure 2 b. No charge will\nbe generated in the assembly on the left-hand side since the PET/Al\nlayer was initially attached to the tape layer. Instead, the Al electrode\nlayer on the left will bring electrons from the ground because the\ntape was positively charged due to the contact. During the separation\nphase, electrons will flow from the ground in the Al electrode layer\non the right since the negatively charged PET layer becomes neutral\nby sparking after the separation from the tape. The Al electrode layer\non the left will bring electrons back to the ground since the charges\non the tape were dissipated by sparking, as described in Figure 2 b. Therefore, both\nAl electrode layers will have an opposite sign of voltage amplitude,\nas explained during the contact and separation motion. Figure 5 d,e shows the voltage amplitude\nof the left Al layer (denoted as an impactor in red)/the right Al\nlayer (denoted as a stator in black) and the combined power as a function\nof time, respectively, at 20 Hz. As expected, the voltage amplitude\nsigns are exactly opposite on both sides. The combined power shows\na peak value of 21.2 mW. The associated power density is calculated\nat 17.3 W/m 2 , which is comparable to the value in existing\nTENGs. Furthermore, the efficiency of the system can be estimated\nby calculating both the input kinetic energy and output electrical\nenergy. 35 The kinetic energy (KE) is given\nby Here, m is the mass of Al\nimpactor. An average velocity ( v ) is assumed. Figure 5 (a) Schematic\nof triboelectric generator with Al/PET/tape and PET/Al\nconfigurations. Separate electrical circuits are built for measuring\nvoltage, current, and power on each side. (b) In the contact process,\nthe electric double layer occurs at the interface between the double-sided\ntape and PET layer on the right side. Since PET on the left side was\nnot used as a triboelectric layer, it was not charged up. Due to the\npositive charges in the double-sided tape, electrons will flow from\nthe ground to Al layer. On the right side, with negatively charged\nPET, electrons will be pushed out from the Al layer to the ground.\nTherefore, the voltage amplitude signs on each side are opposite to\neach other. (c) In the separation process, the electrons are discharged\nin PET by the spark generation between the double-sided tape and PET\non the right side. On the left Al layer, since the double-sided tape\ndoes not have any charges to attract electrons, electrons in the Al\nlayer flow to the ground. On the right side, positive charges in Al\nwill attract electrons from the ground. It shows similar behavior\nwith contact process but with opposite signs of voltage amplitudes.\n(d) Amplitudes as a function of time for the left side Al layer (impactor—red)\nand right side Al layer (stator—black). As expected, amplitude\nsigns are opposite for stator and impactor. (e) Corresponding power\nfrom combining the left and right sides. Peak power at 21.2 mW is\nobtained, which is significantly higher than that of a one-electrode\nAl layer configuration. The output electrical energy is 27.4 μJ by\nintegrating the\npower over time. Finally, the efficiency can be determined by taking\nthe ratio between the input and output power, which is 8.6%. A bridge rectifier circuit was used to convert AC voltage to DC\nvoltage to demonstrate the direct power of 296 LEDs, which are connected\nin series as shown in Figure 6 a during a shaker test with a driven frequency of 20 Hz. The\nvoltage amplitude across 296 LEDs is up to 700 V, as shown in Figure 6 b. Compared to the\nvoltage data, a narrow-banded feature was observed in the power data,\nas shown in Figure 6 c. The peak power has a value of 25 mW, i.e., 20.4 W/m 2 power density, which is able to power LEDs with high brightness,\nas shown in Figure 6 d. The corresponding video footage is shown in S4 . Figure 6 (a) Schematic of the triboelectric generator with Al/PET/tape–PET/Al\ncombinations for DC pulse generation. A bridge rectifier was connected\nwith the triboelectric generator and 296 LEDs in series. (b) DC voltage\namplitude as a function of time for 296 LEDs. Negative amplitude generation\nbecomes positive after passing the rectifier. (c) Corresponding power\nfor 296 LEDs. A maximum power at 25 mW was generated during the process.\n(d) 296 LEDs light up during the triboelectric generator operation\nat 20 Hz. Several manually operated triboelectric energy\nharvester prototypes\nwere developed to further demonstrate the operation under low-frequency\n(i.e., < 5Hz) motion, in which a double-electrode configuration\nis adopted with an active area of 25 mm × 25 mm. As shown in Figure 7 a, the pressing (contact)\nwill generate a very strong restoring force by loaded springs during\nthe separation process. Both soft spring and hard spring cases were\ninvestigated, which correspond to the spring constants at 0.91 N/mm\nand 1.75 N/mm, respectively, with four springs at each corner with\na 10 mm gap between plates. A 4 MΩ load resistor and a 100 Ω\nresistor were introduced to collect voltage and current data during\nthe manual press/release operation approximately at 1 Hz. As shown\nin Figure 7 c–e,\nthe voltage, current, and power responses show higher values when\nthe spring constant is high. The highest peak power is 106 mW, and\nthe associated power density (defined by power/unit area) is 169.6\nW/m 2 , which exceeds 47% to the highest value observed in\nexisting TENGs. 19 Figure 7 (a) Cross-sectional view\nof a manual triboelectric generator. Triboelectric\nlayers are separated by four loaded springs at each corner. Once pressure\nis applied, triboelectric layers touch each other and are released\nquickly due to compressed springs. (b) 3D printed manual harvester\nwith two electrical wires connecting both PET/Al layers to the circuit.\n(c) Voltage amplitude and (d) current data of the manual harvester\nwith 0.91 and 1.75 N/mm spring constants, respectively. (e) Peak power\ngeneration for the manual harvester with 0.91 and 1.75 N/mm spring\nconstants, respectively. A second V-shaped triboelectric generator prototype\nwas developed\nas well for a “hand castanets” operation, in which a\nfolded polypropylene plate with a triboelectric assembly on each inner\nsurface is employed. This manual triboelectric generator lights up\n476 LEDs operated approximately at 2 Hz, which has a 38 mm by 25 mm\nactive area. Associated videos are shown in S5 and S6 . Additional 20 more LEDs were\nlit compared to that in a typical TENG device. 19 A 650 nm laser diode was powered using our double-electrode\ntriboelectric\ngeneration under a shaker test, as shown in Figure 8 a. Videos of power LEDs and a laser diode\nin a two-in-one flashlight are shown in S7 and S8 , respectively. Figure 8 b shows a photodetector amplitude\nsignal of the laser diode as a function of time. Clearly, the overcoming\nof the lasing threshold is observed from the laser diode with an indication\nof the stimulus emission for the first time in the triboelectric generator.\nIn addition, the laser operation shows a 50 ms interval, which corresponds\nto the driven frequency of 20 Hz in the shaker test. Figure 8 (a) Schematic of triboelectric\ngenerator with Al/PET/double-sided\ntape and PET/Al configuration for a DC pulse generation used for a\nlasing with a stimulus emission at 20 Hz of shaker frequency. (b)\nA stimulus emission of a 650 nm diode laser was confirmed with a 50\nms duration from a Si photodetector. A V-shaped manual triboelectric generator with\na 38 mm × 38\nmm active area assembled as the Al/PET/double-sided tape and PET/Al\nwas inserted at the bottom of a shoe, as shown in Figure 9 a, to demonstrate energy harvesting\nvia walking. A bridge circuit and 51 LEDs are connected to convert\nthe AC electric signals into DC for lighting up LEDs. The amount of\npeak power generation was confirmed, as shown in Figure 9 b–d. It shows that the\npeak power is over 10 mW, which indicates that the double-sided tape\ntriboelectric generator can be easily adapted to real-world applications\nwithout any issues at low-frequency ranges. A video for this demonstration\nis shown in S9 . Figure 9 (a) Triboelectric generator\n(Al/PET/double-sided tape and PET/Al\nconfiguration) with a 38 mm × 38 mm area installed on the bottom\nof a shoe. A bridge rectifier circuit is used to convert AC to DC\npower to demonstrate the lighting up of 51 LEDs; (b) voltage; (c)\ncurrent; and (d) power under a 4 MΩ electrical load. Figure 10 shows\nthe endurance test results for a triboelectric generator with a double-electrode\nconfiguration, in which an electrical load of a 4 MΩ resistor\nis connected. Both voltage and temperature data were recorded up to\n100,000 contact–separation cycles. Since the temperature increases\ngradually, the voltage amplitude increases accordingly. Clearly, the\ntemperature has an effect on the performance of power generation because\nmore charges will be transmitted between insulating triboelectric\nlayers. 32 After 40,000 contact–separation\nevents, the temperature decreases gradually so that the performance\ndegrades slightly. Comparable power generation is achieved up to 100,000\ncycles, which varies from 19 to 24 mW. Figure 10 (a) Voltage amplitude\nfor a triboelectric generator with a double-electrode\nconfiguration in a shaker test to reach 100,000 contact–separation\ncycles. Temperature data in red were collected at the Al framing (shown\nin Figure 1 ). (b) Corresponding\npeak power under a 4 MΩ electrical load Figure 11 shows\nSEM images of both tape and PET surfaces before and after the test.\nBefore the test, the PET shows many particles on its surface due to\nthe electrostatic attraction. After the test, less particles are observed.\nThere is no major morphological change in the double-sided tape. This\nindicates that there is no substantial degradation in adhesiveness. Figure 11 SEM\nimages of PET (a) before and (b) after endurance test and double-sided\ntape (c) before and (d) after endurance test at 20,000× magnification."
} | 7,825 |
35012155 | PMC8747654 | pmc | 2,906 | {
"abstract": "The paper presents the viscoelastic properties of new hybrid hydrogels containing poly(vinyl alcohol) (PVA), hydroxypropylcellulose (HPC), bovine serum albumin (BSA) and reduced glutathione (GSH). After heating the mixture at 55 °C, in the presence of GSH, a weak network is formed due to partial BSA unfolding. By applying three successive freezing/thawing cycles, a stable porous network structure with elastic properties is designed, as evidenced by SEM and rheology. The hydrogels exhibit self-healing properties when the samples are cut into two pieces; the intermolecular interactions are reestablished in time and therefore the fragments repair themselves. The effects of the BSA content, loaded deformation and temperature on the self-healing ability of hydrogels are presented and discussed through rheological data. Due to their versatile viscoelastic behavior, the properties of PVA/HPC/BSA hydrogels can be tuned during their preparation in order to achieve suitable biomaterials for targeted applications.",
"conclusion": "4. Conclusions The paper presents the viscoelastic properties of self-healing hybrid hydrogels prepared by physical interactions between poly(vinyl alcohol) (PVA), hydroxypropylcellulose (HPC), bovine serum albumin (BSA) and reduced glutathione (GSH). After heating the mixture at 55 °C, a weak network is formed due to partial BSA unfolding and HPC hydrophobic interactions. By applying three successive freezing/thawing cycles, a porous structure is developed, as evidenced by SEM. The viscoelastic behavior was investigated in order to evidence the self-healing ability as a function of hydrogel composition, applied strain or temperature. The viscoelastic properties of the healed polymer/protein hydrogels were recovered to almost 100% of the original values if the strain and temperature are below a critical value. The thixotropic behavior is influenced by the BSA content, loaded deformation and temperature, and it is attributed to multiple interactions developed in the system: hydrogen bonding, hydrophobic interactions, S-S bonds, PVA/PVA interactions in the crystalline zones. Due to their versatile viscoelastic behavior, the properties of PVA/HPC/BSA hydrogels can be tuned during their preparation in order to achieve suitable biomaterials for targeted applications. Hydrogels with moderate and low BSA content are appropriate for wound dressings and tissue engineering applications. From a rheological point of view, high BSA content formulations are suitable as injectable or targeted drug delivery biomaterials; also, they can be used for bio-printing or electrospinning at temperatures lower than 50 °C.",
"introduction": "1. Introduction In nature, the damage or failure of different living tissues can be spontaneously healed without the action of some external stimuli. Inspired from biological systems, this extraordinary ability was translated to smart polymer networks or other systems, and it is of high interest for various engineering applications [ 1 , 2 , 3 ]. Thus, during the recent years, many efforts were devoted to the design and investigation of materials with self-healing behavior in order to improve the required performances in using conditions or to extend their service lifetime in various applications. These smart materials are able to recover spontaneously their initial state or to repair themselves after damage or degradation. Thus, they recover the functionality and structural integrity by assuring a high rate of healing as compared with the rate of damage. Analogous to the structures present in biological systems at different length scales, a remarkable progress was registered in developing new functional materials for regenerative medicine, pharmaceutics or electronics and advanced technologies [ 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 ]. Due to high elasticity, superabsorbence and excellent mechanical properties, synthetic hydrogels with porous structure are widely used in biomedical applications for controlled drug release, tissue engineering, implantable devices, wound dressings, contact lenses, biosensors or actuators, etc. [ 10 , 11 , 12 , 13 , 14 ]. In particular, various biomaterials can be easily obtained by environmental-friendly chemical or physical procedures applied to poly(vinyl alcohol) (PVA) [ 15 , 16 ], or to various combinations of PVA with other macromolecules or small molecules [ 17 , 18 , 19 , 20 , 21 , 22 ]. The high interest in the PVA-based hydrogels is due to the exceptional mechanical properties [ 4 , 23 ] and their performances in multifunctional biomaterials [ 19 , 20 , 21 , 22 , 23 , 24 ], but also to their self-healing properties [ 5 , 9 , 11 ] and high elasticity proved in creep and recovery tests [ 9 , 11 , 19 ]. Certain other specific functions that belong to natural tissue can be conferred by introducing natural molecules (proteins, peptides, polysaccharides, clays, etc.) in the hydrogel structure: biocompatibility, bioadhesivity, biodegradability, cell proliferation ability, antimicrobial activity, low cytotoxicity, etc. [ 25 , 26 , 27 , 28 , 29 ]. However, the hydrogels composed only from natural biomolecules [ 26 , 27 ] often exhibit a very long gelation time and poor mechanical properties, limiting their applications. Many new strategies to form double networks have emerged in order to improve the mechanical properties and biological compatibility of materials, such as dual physically/chemically crosslinking hydrogels [ 12 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 ]. Multifunctional hydrogels based on synthetic or natural polymers combined with proteins, peptides or active enzymes represent in most cases a synergistic approach that allows the design of materials with targeted properties in various applications. These hybrid materials present improved mechanical and structural stability, functionality and biocompatibility, and they are able to integrate and regenerate the tissue [ 26 , 27 , 29 , 37 ]. Serum albumins have the ability to bind various biological molecules, either hydrophobic or hydrophilic, cationic or anionic [ 38 , 39 ]. Bovine serum albumin (BSA) is a globular protein with a structure homologous with human serum albumin, often used as a model bioactive protein [ 40 ] or for enzyme separation and purification by specific complexation [ 41 ], but also to prepare materials for biomedical applications [ 42 , 43 ]. Reduced glutathione (GSH) is a tripeptide (γ-L-glutamyl-L-cysteinylglycine) well known as a natural antioxidant due to its redox properties. The majority of diseases are associated with a decrease in GSH content in the living cells, combined with various oxidative stress states. Different approaches are considered as emerging tools to increase GSH bioavailability and the potential uses in nanotechnologies, as for example the design of stimuli responsive systems for targeted drug delivery [ 44 ]. According to in vitro and in vivo investigations [ 45 ], GSH conjugated BSA nanoparticles represent a promising non-toxic brain drug delivery system. In presence of GSH, the tris-citric acid extender improves the freezability, post-thaw quality of buffalo bull spermatozoa and in vivo fertility [ 46 ]. The addition of 1 mM GSH to the freezing and thawing extenders increases the motility of the human sperm by improving the level of sulfhydryl groups on membrane proteins [ 47 ]. Recently, it was shown that GSH inhibits the dimerization and wild type SARS-Cov-2 Mpro activity [ 48 ]. PVA, in combination with BSA and hydroxypropylcellulose (HPC), in the presence of GSH, was recently used to prepare porous hydrogels by the freezing/thawing method. It was found that the obtained physical networks have high hydrophilicity and swelling capacity in solution [ 29 ]. In this paper, double protein/polymer networks are prepared. The physical hydrogels are formed by S-S interactions of unfolded BSA in the presence of GSH, and then PVA junction points resulted by applying successive freezing/thawing cycles. The entire range of the polymer/BSA composition was explored, and the self-healing ability was investigated in different conditions of deformation and temperature."
} | 2,034 |
35627790 | PMC9141142 | pmc | 2,907 | {
"abstract": "Microbial fuel cells (MFCs) could achieve the removal of antibiotics and generate power in the meantime, a process in which the bacterial community structure played a key role. Previous work has mainly focused on microbes in the anode, while their role in the cathode was seldomly mentioned. Thus, this study explored the bacterial community of both electrodes in MFCs under sulfadiazine (SDZ) pressure. The results showed that the addition of SDZ had a limited effect on the electrochemical performance, and the maximum output voltage was kept at 0.55 V. As the most abundant phylum, Proteobacteria played an important role in both the anode and cathode. Among them, Geobacter (40.30%) worked for power generation, while Xanthobacter (11.11%), Bradyrhizobium (9.04%), and Achromobacter (7.30%) functioned in SDZ removal. Actinobacteria mainly clustered in the cathode, in which Microbacterium (9.85%) was responsible for SDZ removal. Bacteroidetes, associated with the degradation of SDZ, showed no significant difference between the anode and cathode. Cathodic and part of anodic bacteria could remove SDZ efficiently in MFCs through synergistic interactions and produce metabolites for exoelectrogenic bacteria. The potential hosts of antibiotic resistance genes (ARGs) presented mainly at the anode, while cathodic bacteria might be responsible for ARGs reduction. This work elucidated the role of microorganisms and their synergistic interaction in MFCs and provided a reference to generate power and remove antibiotics using MFCs.",
"conclusion": "4. Conclusions In summary, the addition of SDZ has a limited effect on the electrochemical performance of air-cathode MFCs, with the maximum output voltage kept at 0.55 V. Anodic bacteria were mainly responsible for power generation, and part of them could remove contaminants. Cathodic bacteria were responsible for pollutants removal. The nature (aerobic/anaerobic) and function of bacteria decided their distribution in the anode and cathode. The electrons provided by exoelectrogenic bacteria could increase the metabolic reaction of degradation-related bacteria, and degradation-related bacteria might contribute to the power generation by producing secondary metabolites. They could remove SDZ and generate power through synergistic interactions. The potential hosts of ARGs mainly presented in anodic bacteria, while cathodic bacteria possibly played a role in ARG reduction. Regardless, the spread risk of antibiotic resistance should be seriously concerned for both electrodes.",
"introduction": "1. Introduction The misuse and overuse of antibiotics has not only caused environmental pollution, but it has also stimulated the selection of antibiotic resistance genes (ARGs) [ 1 ]. Hence, exploring economical and effective treatments to eliminate these pollutants has become a hot topic. A microbial fuel cell (MFC) is regarded as a promising alternative treatment to realize waste resource utilization as well as the removal of antibiotics [ 2 , 3 ]. It includes anodic reactions with complex organic compounds as electron donors and cathodic reactions with O 2 , nitrate and nitrite as electron acceptors [ 2 , 3 ]. The mechanism of MFC to remove antibiotics is the combined effect of anaerobic biodegradation and electrical stimulation. Persistent electrical stimulation stimulates the microbial metabolism by transmitting electrons to bacterial cells, and stimulated microorganisms rapidly metabolize antibiotics by secreting enzymes [ 4 ]. Antibiotics, including sulfadiazine (SDZ), sulfamethoxazole (SMX), Tetracycline (TC), Oxytetracycline (OTC), and Chloramphenicol (CAP), could be removed in two dual chamber MFCs [ 4 , 5 , 6 ], and the voltage output was virtually not affected. Different from two dual chamber MFCs, single-chamber MFCs (mainly air-cathode MFCs) could directly use O 2 in the air as an electron acceptor without the cathode chamber, so they could increase the mass transfer effective and reduced cost, and O 2 diffusion from air to the cathode would help the removal of nitrogen without aeration. Therefore, air-cathode MFCs also showed excellent performance in removing antibiotics [ 4 , 7 , 8 ]. Bacterial community structure played a key role during the power output and contaminant removal in MFCs. Therefore, it is necessary to understand the microbial status in the MFCs to clarify the relationship between power generation and pollutant removal. Previous research revealed that the dominant phylum in MFCs were Proteobacteria, Bacteroidetes, Firmicutes, and Actinobacteria. The bacteria affiliated with these phyla were classified into two groups (degradation related and exoelectrogenic) [ 9 ]. The mechanisms of electroactive microorganisms have been summarized in several recent reviews [ 10 , 11 , 12 ]. These cases of research mainly focused on anodic bacteria, but the growth of bacteria in the cathode is inevitable for air-cathode MFCs, which was an important factor reducing the electrochemical performance [ 13 , 14 ]. Conversely, cathode-biofilm also could prevent the diffusion of O 2 in the cathode side to the anode chamber to increase the power generation of MFCs [ 15 ]. Therefore, the role of cathodic bacteria is controversial. There is limited literature on cathodic bacteria of air-cathode MFCs. To gain insight into anodic and cathodic communities, Daghio et al. firstly operated a single chamber MFC to investigate microbial communities. It was found that degradation-related bacteria were enriched in the cathode, but electricigens or closely related microbes were clustered in the anode to attribute chlorinated herbicide removal or power generation in soil MFCs [ 16 , 17 ]. Yuan et al. found that the COD/N of wastewater would affect the bacteria, both in the anode and cathode of MFCs [ 18 ]. The distribution trend of nitrifiers and denitrifiers in cathodes varied with the cathode-biofilm depth. These studies demonstrated that the cathode played an important role in power generation and contaminants removal in air-cathode MFCs; however, the research on the degradation of antibiotics in wastewater by MFCs mostly focused on anode biofilm, and the function of cathodic bacteria was seldom mentioned. Therefore, it is necessary to clarify the roles of both the anode and cathode in power generation, antibiotic degradation, and ARG propagation. This work investigated the efficacy of air-cathode MFCs in wastewater treatment, with SDZ as the representative antibiotic. SDZ is reported to be one of the most common sulfonamides and is used frequently in veterinary medicine [ 19 ]. The electrochemical and physicochemical performance of MFCs under SDZ pressure was studied. Especially, the structures and interactions between anodic and cathodic microorganisms were mainly analyzed. Finally, the occurrence of ARGs and integrons in the anode and the cathode are discussed. The study will elucidate the role of microorganisms in both electrodes and provide reference for the further application of air-cathode MFCs in pollution control and power generation.",
"discussion": "3. Results and Discussion 3.1. Performance of Air-Cathode MFCs under Sulfadiazine Pressure The removal ratio of COD, NH 4 + -N, and TDN was 86.55%, 45.15%, and 45.64% in the MFCs, respectively, and it was 83.46%, 9.56%, and 11.27% in the open circuit treatment, respectively ( Figure 1 a). The degradation of SDZ was gradually accelerated with the process ( Figure 1 b). Twelve cycles later, over 50% of SDZ could be removed within a cycle, which suggests that the microbes in the MFCs gradually acquired the ability for SDZ degradation. Compared with the open circuit, MFCs reduced the increase of pH and conductivity ( Figure S2 ) to keep a suitable environment for microorganisms to degrade pollutants, even under SDZ pressure. In air-cathode MFCs, oxygen diffused from the air into the biofilm from the cathode, which played a role in nitrification and mono-oxygenation reactions for the removal of SDZ [ 21 , 22 ]. Compared with other research (13.39–80%) [ 6 , 23 ], our experiment was conducted in the dark, which might have caused a relatively low removal ratio of SDZ (58.72%). The maximum output voltage was 0.55 V ( Figure 1 c), and the addition of SDZ had no drastic effect on the output voltage of MFCs. Previous studies have demonstrated that the output of MFCs would be enhanced rather than be inhibited under antibiotic pressure [ 7 , 8 , 24 ]. An obvious peak was at −0.4 V vs. Ag/AgCl in CV curves ( Figure 1 d), which was close to the formal potential of outer membrane cytochromes of G. sulfurreducens (−0.398 V vs. Ag/AgCl) [ 25 ]. It was consistent with a previous study that concluded that some antibiotics could increase the permeability of exoelectrogens membranes and then facilitate the direct electron transfer [ 7 ]. In addition, the maximum output power ( Figure S3a ) and Nyquist plots ( Figure S3b ) demonstrate that the addition of SDZ showed little influence on the electrochemical activity of MFCs. 3.2. Bacterial Community Shift in the Anode and Cathode under Sulfadiazine Pressure The goods coverage of all tested samples was 99%, which indicated that the sizes of all libraries were enough to cover the bacterial communities. As shown in Figure 2 , the bacterial community showed a lower Chao1 index and a higher Shannon index in the cathode compared with that of the anode. This indicated that although the bacteria community in the cathode was not rich, it had a higher diversity. The principal coordinate analysis (PCoA) based on UniFrac distances showed that the sample had a good reproducibility ( Figure S4a ). The distance between samples further validated the difference of the bacterial community structures between the anode and cathode. The unique species also sharply reduced in the anode and cathode ( Figure S4b ), which will be discussed in detail below. These indicated that the bacteria in the inocula were selectively enriched in the anode and cathode. Firmicutes, Ignavibacteriae, Chloroflexi, Lentisphaerae, and Euryarchaeota were mainly enriched in the anode, while Actinobacteria mainly clustered in the cathode; Proteobacteria and Bacteroidetes existed both in the anode and cathode ( Figure 3 ). Among these phyla, Proteobacteria, Bacteroidetes, and Actinobacteria were three dominant phyla with the relative abundance of 88.6% (anode) and 97.1% (cathode), respectively. The function of these bacteria determined their selective enrichment in the anode and cathode, which will be discussed in more detail at follow-up. Proteobacteria was most abundant in the anode (70.7%) and cathode (62.6%). The difference was further explicated at the class-level ( Figure 4 a). The most abundant class in anodic community was Deltaproteobacteria (40.91%), for which almost all the detected sequences belonged to the genus Geobacter ( Figure 4 b). Geobacter was the dominant genus in the MFCs anode, with the relative abundance of 40.3% as Gram-negative and electroactive bacteria [ 26 , 27 ]. Geobacter is strictly anaerobic and could adapt to the low surface potentials of the anode [ 26 , 27 ]; therefore, they mainly colonized in the anode and were responsible for power generation by using acetate. Xanthobacter (11.11%, p < 0.01), Bradyrhizobium (9.04%, p < 0.01), and Mesorhizobium (2.68%, p < 0.001) affiliated with Alphaproteobacteria were enriched in the cathode ( Figure 4 b). Xanthobacter was dominant in the microbial electrolysis cells (MECs) but was scarcely reported in MFCs. They accumulated in the cathodic community in this experiment, possibly due to the fact that Xanthobacter thrives in micro-oxygen environments [ 18 ], as well as that they could degrade the aromatic structure of SDZ in the cathode [ 28 ]. Similarly, Bradyrhizobium could degrade antibiotics via co-metabolism with acetate [ 29 ]. Mesorhizobium is an aerobic bacterium using oxygen as the terminal electron acceptor to respire. The possible role of Mesorhizobium was to remove acetate and SDZ degradation products because it could metabolize amino salts, nitrates, and various amino acids using various carbohydrates and organic acid as the carbon source [ 30 ]. Azospirillum (1.88%, p < 0.01) affiliated with Alphaproteobacteria was enriched in the anode, which have been reported to be resistant to antibiotics using a wide range of carbon sources [ 31 ]. It was reported that Bradyrhizobium , Mesorhizobium , and Azospirillum could produce extracellular polysaccharides, which supported their metabolism [ 30 , 31 ]. Achromobacter (7.30%, p < 0.001) and Castellaniella (2.24%, p < 0.01) affiliated with Betaproteobacteria were enriched in the cathode ( Figure 4 b). Achromobacter was able to degrade sulfonamides [ 32 ], and Castellaniella could degrade pyrene under denitrifying conditions [ 33 ], so they clustered in the cathode to degrade the aromatic structure of SDZ. Azospira (5.45%, p < 0.01) affiliated with Betaproteobacteria was enriched in the anode as denitrification organisms, which could denitrify under anaerobic conditions and could use the electrode as the electron acceptor [ 34 ]. They colonized in the anode for denitrification and power generation. Hydrogenophaga , an autotrophic H 2 -oxidizing bacteria, could utilize hydrogen as the energy source [ 35 ]. Comamonas as facultative anaerobic microorganisms have been isolated from MFCs, which contributed to the power generation with acetate as the electron donor [ 36 ]. They belong to Betaproteobacteria, with lower abundance in the anode and the cathode ( Hydrogenophaga (0.49% vs. 2.28%, p = 0.114) and Comamonas (0.78% vs. 0.05%, p = 0.403)). Pseudomonas (4.58%, p < 0.01), which belong to Gammaproteobacteria, were enriched in the anode ( Figure 4 b). As electroactive bacteria, Pseudomonas were also able to degrade sulfonamides, which were the dominant bacterial genus in the anodic chamber of MFCs [ 37 ], so they accumulated in the anode for SDZ removal and power generation. Stenotrophomonas (1.59% vs. 2.68%, p = 0.186) and Dokdonella (1.20% vs. 2.18%, p < 0.01) affiliated with Gammaproteobacteria were enriched in both the anode and cathode. Stenotrophomonas is able to degrade many xenobiotic aromatic compounds [ 38 ]. Dokdonella is usually found in the aerobic biological systems to remove nitrogen and degrade aromatic hydrocarbons simultaneously [ 39 ], so they colonized both the anode and cathode, possibly for the degradation of SDZ. Actinobacteria were widely used in water treatment field, as they can use glucose, starch, and cellulose as carbon sources [ 40 ]. The abundance of Actinobacteria in the cathode (19.68%) was higher than that in the anode (7.65%, p < 0.01) ( Figure 4 a), due to the anaerobic condition that inhibited Actinobacteria [ 41 ]. Microbacterium (9.85%, p < 0.05) and Pseudoclavibacter (2.78%, p < 0.001), which belong to Actinobacteria, clustered in the cathode ( Figure 4 c). Microbacterium is an important degradation-related microorganism, which showed a higher degradation rate for sulfonamides, especially for SDZ [ 42 ]; therefore, it was most likely responsible for the degradation of SDZ in the cathode. Pseudoclavibacter is a Gram-positive and aerobic genus; although this genus has not been reported in the wastewater treatment field, it was reported to be responsible for the biodesulfurization of organic sulfur [ 43 ]; therefore, they might cluster in the cathode, helping the removal of sulfur in SDZ. Rhodococcus (2.69% vs. 1.99%), Mycobacterium (1.03% vs. 0.43%), and Gordonia (1.56% vs. 3.69%, p < 0.01) affiliated with Actinobacteria existed both in the anode and the cathode ( Figure 4 c). Rhodococcus degraded SDZ and regulated biofilm thickness in both the anode and cathode, for they could degrade various aromatic compounds via a ring-cleavage pathway under anaerobic conditions [ 44 ]. They could also improve MFC performance via controlling the biofilm thickness on the anode surface [ 45 ]. Mycobacterium and Gordonia were beneficial to degrade SDZ, for they could degrade polycyclic aromatic hydrocarbons (PAHs) into phthalate, CO 2 , and other chemicals by decarboxylation, dioxygenation, hydrolysis, and ring-cleavage reactions [ 6 , 46 ]. Gordonia is an aerobic bacterial genus, but it was found in the anode in previous studies [ 46 , 47 ], which is possibly due to less attention being paid to microorganisms in the cathode of MFCs. Bacteroidetes were enriched both in the anode (10.28%) and cathode (14.86%) without a significant difference between each other. Proteiniphilum (2.02%, p < 0.05) and Petrimonas (1.83%, p < 0.05) affiliated with Bacteroidetes mainly accumulated in the anode ( Figure 4 c), possibly because they are typical fermenters. Proteiniphilum and Petrimonas could degrade complex substrates, such as PAHs, to simple organic compounds by syntrophic metabolism [ 48 , 49 ], which indicates that they could degrade the product of SDZ in anode. There was no significant difference for the relative abundance of Chryseobacterium between the anode (1.27%) and the cathode (1.17%) ( Figure 4 c). Considering their ability to degrade aromatic compounds [ 50 ], they might degrade SDZ in both electrodes. Generally, Proteobacteria played an important role both in the anode and cathode, without a significant difference as the predominant phylum: Deltaproteobacteria mainly accumulated in the anode, represented by the genus of Geobacter ; it was responsible for power generation. Alphaproteobacteria, Betaproteobacteria, and Gammaproteobacteria colonized both the anode and cathode and were associated with the degradation of SDZ; Actinobacteria mainly clustered in the cathode, which were responsible for the removal of SDZ; Bacteroidetes showed no significant difference between the anode and cathode, and they were associated with the degradation of SDZ. The nature (aerobic/anaerobic) and function of bacterial genera decided their distribution in the anode and cathode. 3.3. Potential Bacterial Roles in SDZ Degradation The microorganisms in air-cathode MFCs can be functionally categorized into the following two groups: exoelectrogenic bacteria ( Geobacter , Pseudomonas , Azospira , and Comamonas ) and degradation-related bacteria ( Xanthobacter , Bradyrhizobium , Achromobacter , Azospirillum , Microbacterium , Pseudoclavibacter , Mycobacterium , Castellaniella , Dokdonella , Rhodococcus , Mycobacterium , Gordonia , Proteiniphilum , Petrimonas , and Chryseobacterium ). Based on the variation of bacterial communities in the biofilms, the ecological model of SDZ biodegradation in air-cathode MFCs was proposed ( Figure 5 a). Substrates (acetate and SDZ) diffused into the biofilm both in the anode and cathode, while oxygen diffused from the cathode side. In the anodic biofilms, Geobacter was mainly responsible for power generation by the metabolizing acetate, and SDZ degradation products were further oxidized by fermentation bacteria (e.g., Proteiniphilun and Petrimonas ) or Gordonia to low molecular organics. Nitrification bacteria used the O 2 from cathode. The oxygen-utilizing of cathodic bacteria can create anoxic zones, offering a favorable zone for denitrifying bacteria and exoelectrogenic bacteria. Almost all cathodic bacteria and part of anodic bacteria were involved in nitrification and denitrification. Based on published literature, the possible degradation pathway of SDZ is shown in Figure 5 b. Furthermore, the role of microorganisms in the degradation of SDZ and their relationship in air-cathode MFCs was explained according to specific functions. It is noteworthy that the initial electrophilic attack by oxygenases of aerobic bacteria is often a rate-limiting step and the first of a chain of reactions which is responsible for the biodegradation of many organic compounds [ 52 ]. This might be the main reason why degradation bacteria were mostly distributed in the cathode. In the pathway I, SDZ obtained electrons in the presence of O 2 to form dihydroxyl SDZ (I-①). As mentioned in Section 3.2 , Mycobacterium might work in the process by deoxygenation. Moreover, Microbacterium was able to degrade SDZ into 2-aminopyrimidine, which was initiated by NADH-dependent ipso-hydroxylation [ 53 ], corresponding to the S-N bond hydrolysis, (I-②) or (II-①). The p -anilinesulfonic acid was further hydrolyzed to generate aniline and sulfate (I-③), or N atoms were bio-consumed to form benzenesulfinic acid (II-②), then leading to the formation of benzene and sulfate. During this process, Pseudoclavibacter was responsible for biodesulfurization. The benzene and catechol can be decomposed into acetate and pyruvate by Proteiniphilun , Petrimonas , Rhodococcus , and Castellaniella (I-⑥, II-④). 2-amino-4,6-dihydro-pyrimidine contains more electron-donating functional groups, which render the molecules more prone to electrophilic attack by oxygenases of aerobic bacteria (mainly cathodic bacteria); then, it was decomposed into N 2 , formic acid, and acetate by nitrifiers/denitrifiers (I-⑦, II-⑥) [ 52 ]. Finally, formic acid, acetate, and pyruvate can be used by a Geobacter for power generation. The degradation of SDZ was the result of the synergistic reaction of anodic and cathode bacteria in air-cathode MFCs. O 2 was harmful to power generation in MFCs, but it was conducive to remove contaminants. Therefore, the role of O 2 in air-cathode MFCs should be reconsidered. All the reactions in air-cathode MFCs can be summarized as follows: (1) Exoelectrogenic bacteria were responsible for power generation in the anode of air-cathode MFCs; (2) The electrons provided by exoelectrogenic bacteria could increase the metabolic reaction of degradation-related bacteria located in the anode and cathode; (3) Degradation-related bacteria might contribute to the power generation by producing metabolites, which could be used as electron donors by the electroactive bacteria. These functional bacteria could effectively remove pollutants and generate power via complex synergistic interactions. 3.4. ARGs in the Anode and Cathode under Sulfadiazine Pressure As shown in Figure 6 a, 16S rRNA gene, intI 1, intI 2, sul 1, and sul 2 were detected in the anode and cathode. This is consistent with previous work that showed that sul 3 and sul A were not detected in any samples [ 54 ]. The integrons, intI 1 and intI 2, are natural mobile genetic elements that can capture, integrate, and express resistance gene cassettes with the help of integrase genes. Integrase genes, being important players in ARG transfer, were the driving force for bacterial evolution. Integrons have been reported to be involved in the occurrence of new resistant and pathogenic species [ 8 ]. The resistance of bacteria to sulfonamides are mutated of the enzyme dihydropteroate synthase (DHPS) or acquired alternative DHPS gene, among which sul1 and sul 2 belong to the latter [ 55 ]. The absolute abundance of 16S rRNA gene, intI 1, intI 2, sul 1, and sul 2 in the anode was higher than that of the cathode ( Figure 6 a), and a significant difference was observed between the anode and cathode. Moreover, the relative abundance of these ARGs in the anode was also higher than that in the cathode, except for sul 2 ( Figure 6 b). The relative abundance of sul 1 and sul 2 ranged from 6.59 × 10 −4 to 3.43 × 10 −2 , and sul 2 > sul 1 in the cathode. Similar to other studies, the abundance of sul 1 was usually lower than sul 2 [ 56 ]. A network analysis was conducted in order to further explore the relationship of the bacterial community, integrons, and ARGs in the air-cathode MFCs ( Figure 6 c). intI 1 showed a positive correlation with Geobacter and presented a negative correlation with Bradyrhizobium , Mesorhizobium , Achromobacter , and Hydrogenophaga . It indicated that Geobacter might be the host bacteria of intI 1, while the proliferation of others might be effective to reduce intI 1. Geobacter mainly clustered in the anode, and others were enriched in the cathode, which explains why the abundance of intI 1 in the anode was higher than that in the cathode. The sul 1 presented a positive correlation with Proteiniphilun , Petrimonas , and Azospirillum and showed a negative correlation with Microbacterium , Dokdonella , Stenotrophomonas , and Castellaniella . Proteiniphilun , Petrimonas , and Azospirillum were mainly enriched in the anode, and Microbacterium and Castellaniella mainly clustered in the cathode. Though Stenotrophomonas and Dokdonella were both enriched in anode and cathode, the abundance of these bacteria in the cathode was higher than that in the anode. All of these were consistent with the abundance of sul 1 in the anode being higher in the cathode. The sul 2 showed a positive correlation with Hydrogenophaga and a negative correlation with Geobacter . In addition, the relative abundance of Geobacter in the anode was substantially higher than the abundance of Hydrogenophaga in the cathode. This was consistent with the relative abundance of sul 2 being lower in the anode than that of the cathode. The results showed that many anodic bacteria were potential hosts of the tested ARGs, while the cathodic bacteria might play a role in the reduction of these ARGs. The findings were consistent with previous research that aerobic conditions showed better removal capacities of ARGs than anaerobic conditions [ 57 ]."
} | 6,395 |
29556049 | null | s2 | 2,910 | {
"abstract": "Post-translational modification of proteins is a strategy widely used in biological systems. It expands the diversity of the proteome and allows for tailoring of both the function and localization of proteins within cells as well as the material properties of structural proteins and matrices. Despite their ubiquity in biology, with a few exceptions, the potential of post-translational modifications in biomaterials synthesis has remained largely untapped. As a proof of concept to demonstrate the feasibility of creating a genetically encoded biohybrid material through post-translational modification, we report here the generation of a family of three stimulus-responsive hybrid materials-fatty-acid-modified elastin-like polypeptides-using a one-pot recombinant expression and post-translational lipidation methodology. These hybrid biomaterials contain an amphiphilic domain, composed of a β-sheet-forming peptide that is post-translationally functionalized with a C"
} | 243 |
39469026 | PMC11514727 | pmc | 2,911 | {
"abstract": "Self-cleaning glass surfaces have emerged as a focal point in the field of materials science due to their potential to reduce the accumulation of pollutants, enhance transparency, and improve durability. In recent years, significant advancements have been made in self-cleaning technologies based on photocatalysis and wettability regulation, particularly in the development of superhydrophobic and superhydrophilic surfaces. This article provides a systematic review of the research progress in self-cleaning technologies for glass surfaces. It analyzes and summarizes the applicability of self-cleaning effects induced by special properties such as photocatalysis, superhydrophobicity, superhydrophilicity, and omniphobicity on glass surfaces. Subsequently, the article delves into a discussion of the durability of these surface treatment technologies in practical applications, especially their stability and long-term performance under harsh environmental conditions. Furthermore, the paper explores the current status of applications for self-cleaning glass surfaces across various fields and proposes potential solutions and future research directions to address the challenges encountered in the practical application of self-cleaning glass surfaces.",
"conclusion": "5 Conclusions and outlook This review summarizes the various applications of self-cleaning glass surface technology and its durability issues, and discusses the applicability and limitations of different types of self-cleaning surfaces. By analyzing the strengths and weaknesses of these surfaces in practical applications, we can see the significant progress in self-cleaning surface technology, as well as the limitations encountered in real-world use. Superhydrophobic surfaces are often vulnerable to mechanical damage and chemical corrosion in long-term use due to the complexity of their surface micro–nano structures, which limits their durability to a certain extent. In contrast, the photocatalytic surface has good chemical stability and can resist the erosion of the external environment to a certain extent, but its dependence on light intensity and the weakening of photocatalytic activity by pollutant accumulation are also a major challenge. Other types of self-cleaning coatings, such as super-hydrophilic coatings, perform well in preventing liquid contamination, but their ability to resist contamination is often limited in high humidity environments. In addition, the existing coating preparation process is complex and costly, which limits the possibility of large-scale industrial application. To address these issues, future research needs to further tackle the following challenges: (1) Current superhydrophobic and superhydrophilic surfaces are easily affected by the external environment in practical applications, leading to a decline in their self-cleaning performance. Therefore, future research should focus on developing more wear-resistant and stable materials, while optimizing the surface micro–nano structures to enhance their impact resistance and corrosion resistance. (2) A major obstacle to the existing self-cleaning surface technology is its high preparation cost, which hinders widespread application. Future research should prioritize the development of efficient, low-cost preparation processes to achieve the feasibility of large-scale application. (3) With the continuous development of self-cleaning surface technology, how to endow these surfaces with additional functions (such as antibacterial, corrosion resistance, anti-icing, etc. ) will become an important direction for future research. Multifunctional coatings will further expand their application scenarios, especially in harsh environments. (4) Currently, there is a lack of uniform assessment standards for self-cleaning surfaces, and there is a significant variation in evaluation results between different laboratories. Therefore, future research needs to focus on establishing standardized testing methods to accurately assess the long-term performance of self-cleaning surfaces and their feasibility in practical applications. Overall, self-cleaning surface technology has shown tremendous potential for application, but it still faces many challenges in practical use. Future research should continue to strive to improve the durability of materials, reduce production costs, and expand their functional applications to promote the widespread application of this technology across various industrial fields.",
"introduction": "1 Introduction As an important industrial material, glass is widely used in various fields such as construction, automobiles, photovoltaics, electronics and daily utensils. However, glass surfaces are susceptible to contamination, scratching, and corrosion, which can diminish their optical transparency, service life, and aesthetics. With technological advancements, scientists and engineers have been endowing glass surfaces with special properties like self-cleaning and durability through functional surface treatments, thereby expanding the application fields of glass and enhancing its value. 1,2 Among these surface treatment technologies, self-cleaning glass has garnered widespread attention and research due to its effectiveness in resisting the accumulation of pollutants and maintaining long-term cleanliness. Research on self-cleaning surfaces dates back to the natural phenomena known as the “Lotus effect” and the “Photocatalytic effect”. The Lotus effect is characterized by superhydrophobicity, enabling water droplets to form a high contact angle on the surface, and roll off while carrying away dust and other contaminants. In contrast, the Photocatalytic effect facilitates the degradation of pollutants on the surface through the catalytic action of ultraviolet radiation, leading to their automatic decomposition. 3,4 Inspired by these phenomena, researchers have developed various self-cleaning glass surface treatment technologies with biomimetic design and the application of nanomaterials, such as superhydrophobic coatings and photocatalytic films. These technologies enable glass to maintain cleanliness automatically under the action of rainwater or sunlight, reducing reliance on traditional cleaning methods. Currently, research on self-cleaning glass surfaces primarily focuses on two aspects: one is the self-cleaning surfaces based on the photocatalytic principle, and the other is the self-cleaning surfaces based on the regulation of surface wettability. 4,5 Photocatalytic self-cleaning surfaces utilize photocatalysts to produce strong oxidizing substances under light exposure, breaking down surface contaminants into carbon dioxide and water, thus achieving self-cleaning. Wettability-based self-cleaning surfaces, on the other hand, regulate the surface's wettability to render it superhydrophilic or superhydrophobic, using the spread or roll of water droplets to remove surface contaminants. However, merely possessing self-cleaning properties is insufficient to meet the diverse demands of practical applications. The durability of glass surfaces is also a critical issue, especially under harsh conditions such as outdoor exposure, acid rain erosion, high temperatures, and mechanical wear, where the functional surface of the glass may be significantly compromised. 6,7 Therefore, enhancing the long-term stability and durability of self-cleaning glass has become a focal point of research. Moreover, different application scenarios impose varying functional requirements on glass; for instance, photovoltaic glass demands high light transmittance and self-cleaning surfaces to improve the efficiency of solar energy conversion, while automotive glass focuses more on multiple properties such as water resistance, anti-fog, and scratch resistance. 8,9 Consequently, the development of self-cleaning surface materials and treatment processes tailored to specific application needs has also become a hot topic of research. This review aims to systematically recount the advancements in the research of self-cleaning glass surfaces, with a particular focus on the principles of self-cleaning technology, surface modification methods, durability issues, and their functional performance in practical applications. Through the analysis and summarization of existing literature, this article will outline the current status and challenges of self-cleaning glass surface applications across various fields and look forward to future development directions. Self-cleaning glass surface technology has not only brought convenience to modern industry and daily life but also provided significant technological support for environmental protection and sustainable development. Thus, in-depth research and optimization of this technology hold broad application prospects and economic benefits."
} | 2,202 |
37427092 | PMC10327653 | pmc | 2,913 | {
"abstract": "Swarming is a collective bacterial behavior in which a dense population of bacterial cells moves over a porous surface, resulting in the expansion of the population. This collective behavior can guide bacteria away from potential stressors such as antibiotics and bacterial viruses. However, the mechanisms responsible for the organization of swarms are not understood. Here, we briefly review models that are based on bacterial sensing and fluid mechanics that are proposed to guide swarming in the pathogenic bacterium Pseudomonas aeruginosa . To provide further insight into the role of fluid mechanics in P. aeruginosa swarms, we track the movement of tendrils and the flow of surfactant using a novel technique that we have developed, Imaging of Reflected Illuminated Structures (IRIS). Our measurements show that tendrils and surfactants form distinct layers that grow in lockstep with each other. The results raise new questions about existing swarming models and the possibility that the flow of surfactants impacts tendril development. These findings emphasize that swarm organization involves an interplay between biological processes and fluid mechanics."
} | 292 |
28878822 | PMC5584521 | pmc | 2,914 | {
"abstract": "Background Over three-fifths of the world’s known crude oil cannot be recovered using state-of-the-art techniques, but microbial conversion of petroleum hydrocarbons trapped in oil reservoirs to methane is one promising path to increase the recovery of fossil fuels. The process requires cooperation between syntrophic bacteria and methanogenic archaea, which can be affected by sulfate-reducing prokaryotes (SRPs). However, the effects of sulfate on hydrocarbon degradation and methane production remain elusive, and the microbial communities involved are not well understood. Results In this study, a methanogenic hexadecane-degrading enrichment culture was treated with six different concentrations of sulfate ranging from 0.5 to 25 mM. Methane production and maximum specific methane production rate gradually decreased to 44 and 56% with sulfate concentrations up to 25 mM, respectively. There was a significant positive linear correlation between the sulfate reduction/methane production ratio and initial sulfate concentration, which remained constant during the methane production phase. The apparent methanogenesis fractionation factor ( α \n app ) gradually increased during the methane production phase in each treatment, the α \n app for the treatments with lower sulfate (0.5–4 mM) eventually plateaued at ~1.047, but that for the treatment with 10–25 mM sulfate only reached ~1.029. The relative abundance levels of Smithella and Methanoculleus increased almost in parallel with the increasing sulfate concentrations. Furthermore, the predominant sulfate reducer communities shifted from Desulfobacteraceae in the low-sulfate cultures to Desulfomonile in the high-sulfate cultures. Conclusion The distribution of hexadecane carbon between methane-producing and sulfate-reducing populations is dependent on the initial sulfate added, and not affected during the methane production period. There was a relative increase in hydrogenotrophic methanogenesis activity over time for all sulfate treatments, whereas the total activity was inhibited by sulfate addition. Both Smithella and Methanoculleus , the key alkane degraders and methane producers, can adapt to sulfate stress. Specifically, different SRP populations were stimulated at various sulfate concentrations. These results could help to evaluate interactions between sulfate-reducing and methanogenic populations during anaerobic hydrocarbon degradation in oil reservoirs. Electronic supplementary material The online version of this article (doi:10.1186/s13068-017-0895-9) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions This study not only demonstrated the competition and coexistence of sulfate-reducing and methanogenic populations during anaerobic hexadecane degradation processes, but also revealed a linear positive relationship between the ratio of sulfate reduction/methane production and initial sulfate concentration. The removal efficiency of hexadecane could be enhanced by increased sulfate addition, although the amount of methane accumulation decreased. CO 2 reduction was the dominant methanogenic pathway, but the activity was inhibited by increasing sulfate addition. Both Smithella and Methanoculleus , the key alkane degraders and methane producers, could be adapted to the sulfate stress, and different SRP populations were stimulated at various sulfate concentrations. Taken together, our results would help to understand the interactions between methanogenic populations and SRPs during alkane degradation under mixed electron acceptors conditions.",
"discussion": "Discussion Impacts of sulfate on hexadecane degradation and methane production The results of this study join those of a previous body of reports demonstrating that methanogenic cultures can degrade alkanes anaerobically in the presence of sulfate [ 11 , 15 , 31 ]. Methane production was enhanced under low-sulfate concentrations (<2–5 mM) [ 15 , 31 ], whereas significant inhibition was observed at sulfate concentrations greater than 5 mM [ 31 ]. Gieg et al. [ 11 ] reported that the extent of methanogenic degradation of crude oil alkanes was not affected by sulfate in the 10 mM range, but the methane production rate slightly decreased. The results of this study demonstrated that the rate and extent of methane production were gradually inhibited by increasing sulfate concentration (Fig. 1 ). However, the presence of excessive sulfate at 25 mM, which is higher than the 21 mM needed for complete conversion of hexadecane to CO 2 through sulfate reduction, did not completely inhibit methane production (Eq. 2 ). The concurrent methane production and sulfate reduction in the excessive presence of sulfate may be attributed to the adequate supply of noncompetitive substrates or an abundance of competitive substrates for the syntrophs and methanogens [ 43 ]. However, the possibility of cooperation between the incomplete-oxidizing sulfate reducers and methanogens cannot be ignored if incomplete oxidation of organic intermediates into H 2 or acetate through sulfate reduction presented in the cultures [ 44 ]. This study also demonstrated that sulfate reduction increased hydrocarbon degradation efficiency, although the enhancement may not be directly used to address questions of methanogenic and sulfidogenic competition [ 45 , 46 ]. The increased activity of sulfate reduction after the treatment with increased sulfate concentration probably enhanced H 2 conversion rate and caused a much lower hydrogen concentration below the threshold value for hydrogenotrophic methanogens [ 44 , 47 , 48 ], which may facilitate the syntrophic alkane oxidation process from a thermodynamics standpoint [ 18 ]. The concept of “electron flow” was introduced by Isa et al. [ 49 ] to quantify the competition between methanogenesis and sulfidogenesis. The most common explanation for the electron distribution between sulfate reduction and methane production during organic substrate degradation is the effect of the COD/SO 4 \n 2− ratio [ 49 – 51 ], which also includes other factors, such as H 2 S, substrate concentration, and inoculum volume [ 49 ]. As expected, electron flux toward sulfidogenesis increased with increasing sulfate concentrations in this study (Fig. 2 a). However, unlike most previous work showing how electron flow can be affected by the COD/SO 4 \n 2− ratio [ 50 – 53 ], the relative distribution of electrons between the SRPs and the methanogenic populations depended on the initial sulfate addition and remained constant during methane production (Fig. 2 b), which has not been reported before. The results suggested that the concentration of sulfate could be an effective indicator of hexadecane metabolism pathway under mixed electron acceptors conditions. Giving the complex composition of crude oil and diverse electron acceptors present in oil reservoirs, it is also necessary in the future to evaluate the in situ effects of sulfate on the carbon and electron flow during methanogenic crude oil degradation, which could aid in predicting the biomethane potential of crude oil in field trials. Impact of sulfate on the methanogenic pathway The α \n app is a generally accepted index to coarsely estimate the dominant methane production pathway [ 42 ], with values above 1.065 and below 1.025 proposed to represent hydrogenotrophic and aceticlastic methanogenesis, respectively, according to recent reports [ 54 , 55 ]. However, the α \n app value apparently changes according to species and growth conditions, and the values measured in various cultures of hydrogenotrophic methanogens range between 1.031 and 1.077 [ 56 , 57 ]. Two methanogen genera, Methanosarcina and Methanothrix , are capable of performing aceticlastic reactions [ 58 ], whereas only Methanothrix species were observed in this study (Fig. 4 a), and the fractionation factor of this genus during aceticlastic methanogenesis is less than 1.01 [ 56 , 59 ]. Moreover, the acetate concentration was always lower than the threshold value (100 μM) during incubation, indicating that carbon isotope fractionation did not occur during the conversion of acetate to CH 4 by aceticlastic methanogens [ 60 ]. Therefore, the increase in α \n app observed during the incubation of all sulfate addition cultures likely revealed that CO 2 reduction was the predominant methanogenic pathway, although acetate disproportionation would also contribute to hexadecane degradation. Meanwhile, α \n app decreased gradually with increasing sulfate concentration (Fig. 3 c), indicating that sulfate reduction often outcompetes hydrogenotrophic methanogenesis for H 2 utilization from thermodynamic and kinetic points of view [ 48 , 49 ]. Microbial community response to sulfate stress The aceticlastic Methanothrix was the dominant archaea, not only in the hexadecane- and sulfate-containing cultures but also in the control culture in which hexadecane and sulfate were not added (Fig. 4 a). The dominant Methanothrix did not show a positive relationship with either methane production or sulfate reduction (Table 3 ). These observations indicated that Methanothrix may play a minor role in methanogenic hexadecane degradation, which was not only supported by the aforementioned changes in methanogenic communities but also by increasing α \n app values over time in all sulfate treatments. Regardless, Methanothrix may still play a role by leaking H 2 from acetate degradation to SRPs for sulfate reduction [ 61 ]. Both Methanoculleus and Methanolinea species, which use H 2 /CO 2 for methanogenesis [ 62 , 63 ], were positively associated with methane production (Fig. 4 a; Table 3 ). The Methanoculleus populations steadily increased with increasing initial sulfate concentrations and became the second most abundant after Methanothrix in the 2–25 mM sulfate cultures. This increase could be explained by the fact that Methanoculleus may have a greater affinity than other hydrogenotrophic methanogens for H 2 [ 64 ] or the fact that Methanoculleus is involved in the syntrophic alkane oxidation process, together with SRPs [ 34 ]. \n Smithella is the predominant bacterial phylotype responsible for alkane degradation under methanogenic conditions [ 33 , 34 ]. Detection of the ass A gene and methyl pentadecyl succinic acid in the methanogenic hexadecane-degrading culture has demonstrated that Smithella can initiate alkane degradation via fumarate addition [ 39 ]. However, no dsr -like gene was detected in the binned genome of strain M82_1 [ 39 ]. The observation of a positive correlation between Smithella abundance and sulfate addition (Fig. 4 b; Table 3 ) may reflect cooperation rather than competition between syntrophic alkane degraders and SRPs, as supported by reported increased hexadecane degradation efficiency with increasing sulfate consumption (Additional file 1 : Figure S1). The core bacterial phylotypes affiliated with Spirochaetaceae , Chloroflexi , Candidatus Cloacamonas, and Thermotogaceae were proposed to play roles as either secondary degraders during the degradation of hexadecane [ 12 ] and terephthalate [ 65 ] or contributors to scavenging anabolic products (protein and lipids), presumably derived from detrital microbial biomass [ 65 ]. Most of these bacteria were positively correlated with methane production, but their relationships with sulfate reduction were either negative or irrelevant (Table 3 ), which may reflect the competition of these phylotypes with SRPs. SRPs in the methanogenic cultures have the capacity to cooperate with methanogens and syntrophic bacteria under sulfate-free conditions but also possess robustness to compete with these organisms through sulfate reduction when sulfate becomes available [ 66 ]. The abundance levels of unclassified Desulfobacteraceae (OTU 69) and Desulfomonile (OTU 54) increased, and these genera became the predominant SRPs in the low- and high-sulfate addition cultures, respectively (Fig. 5 ). Most members of Desulfobacteraceae can utilize sulfate and acetate as electron acceptors and carbon sources, respectively, and some can oxidize organic substrates completely to CO 2 , whereas others perform an incomplete oxidation of organic substrates to acetate [ 67 ]. Desulfomonile spp., known as dehalogenators, is also capable of H 2 scavenging, and H 2 uptake can increase by more than fourfold when sulfate is used as an electron acceptor [ 47 ]. These findings suggest that the response of the sulfate reduction bacterial community to sulfate varies according to sulfate concentrations, although their ecophysiological roles in methanogenic hexadecane-degrading cultures in the presence of sulfate require further examination."
} | 3,196 |
31851400 | null | s2 | 2,916 | {
"abstract": "The extracellular matrix (ECM) has force-responsive (i.e., mechanochemical) properties that enable adaptation to mechanical loading through changes in fibrous network structure and interfiber bonding. Imparting such properties into synthetic fibrous materials will allow reinforcement under mechanical load, the potential for material self-adhesion, and the general mimicking of ECM. Multifiber hydrogel networks are developed through the electrospinning of multiple fibrous hydrogel populations, where fibers contain complementary chemical moieties (e.g., aldehyde and hydrazide groups) that form covalent bonds within minutes when brought into contact under mechanical load. These fiber interactions lead to microscale anisotropy, as well as increased material stiffness and plastic deformation. Macroscale structures (e.g., tubes and layered scaffolds) are fabricated from these materials through interfiber bonding and adhesion when placed into contact while maintaining a microscale fibrous architecture. The design principles for engineering plasticity described can be applied to numerous material systems to introduce unique properties, from textiles to biomedical applications."
} | 296 |
34111183 | PMC8191904 | pmc | 2,920 | {
"abstract": "Liquid manure (slurry) from livestock releases methane (CH 4 ) that contributes significantly to global warming. Existing models for slurry CH 4 production—used for mitigation and inventories—include effects of organic matter loading, temperature, and retention time but cannot predict important effects of management, or adequately capture essential temperature-driven dynamics. Here we present a new model that includes multiple methanogenic groups whose relative abundance shifts in response to changes in temperature or other environmental conditions. By default, the temperature responses of five groups correspond to those of four methanogenic species and one uncultured methanogen, although any number of groups could be defined. We argue that this simple mechanistic approach is able to describe both short- and long-term responses to temperature where other existing approaches fall short. The model is available in the open-source R package ABM ( https://github.com/sashahafner/ABM ) as a single flexible function that can include effects of slurry management (e.g., removal frequency and treatment methods) and changes in environmental conditions over time. Model simulations suggest that the reduction of CH 4 emission by frequent emptying of slurry pits is due to washout of active methanogens. Application of the model to represent a full-scale slurry storage tank showed it can reproduce important trends, including a delayed response to temperature changes. However, the magnitude of predicted emission is uncertain, primarily as a result of sensitivity to the hydrolysis rate constant, due to a wide range in reported values. Results indicated that with additional work—particularly on the magnitude of hydrolysis rate—the model could be a tool for estimation of CH 4 emissions for inventories.",
"conclusion": "Conclusions With multiple groups (populations) of methanogens, a mechanistic model of methane production from animal manure or similar wastes can reproduce complex observed responses to temperature, in particular, the distinctly different short- and long-term responses to temperature change. The new model described here, implemented in the ABM R package, is a flexible tool that can facilitate research on CH 4 emission and its control. Accounting for temporary inactivation of methanogens, methane oxidation, and possibly other processes may be necessary for the most accurate predictions, and model extension is possible. An application of the model to field data showed that detailed measurements of slurry organic matter composition will be needed for model extension and future application at all scales.",
"introduction": "Introduction Methane (CH 4 ) emissions from livestock production make a significant contribution to global warming, and manure management on farms contributes about 6.5% of global anthropogenic CH 4 emissions [ 1 , 2 ]. Current emissions estimates in national inventories are based on guidelines from the IPCC [ 3 ], which offer a simple “Tier 1” approach with default emission factors for livestock categories and average annual temperature, and a more detailed “Tier 2” approach considering effects of organic matter (as volatile solids, VS) loading, retention time, and temperature, i.e., properties that vary with farming practices and location. Tier 2 estimates are currently based on a modification of the model presented by Mangino et al. [ 4 ], in which the fraction of VS converted to CH 4 within each month is calculated from a van ’t Hoff-Arrhenius equation with an empirical estimate of activation energy and a reference point corresponding to 100% degradable VS conversion at 30 or 35°C. Although this provides a more site-specific estimate of CH 4 emissions than fixed emission factors, the method has been found to poorly describe both temporal dynamics and total CH 4 emissions in farm- and pilot-scale experiments [ 5 – 7 ]. Thus, a more accurate approach is needed to describe and quantify CH 4 production in manure environments. Models with a dynamic description of microbial decomposition of organic matter in anaerobic digesters already exist [ 8 – 11 ]. The ADM1 model was originally developed for anaerobic digestion almost two decades ago [ 10 ], and it remains a useful and popular tool for research and possibly even plant management [ 12 , 13 ]. Despite its complexity (at least 26 differential equations), ADM1 and similar models were not developed to predict responses to temperature change known to affect CH 4 production in stored slurry. The distinction between short- and long-term responses to environmental changes is also not included in these or most other models, which therefore cannot be used to assess slurry management practices such as cooling as a means to reduce CH 4 emissions, or even for accurate estimation of seasonal variations. For example, an empirical model that accounted for daily temperature and VS degradation still failed to capture the observed dynamics of CH 4 emissions, and it was concluded that the description of methanogenic activity under variable slurry storage conditions was inadequate [ 5 ]. In storage experiments with both fresh and aged slurry, a period of days to months with low CH 4 emission rates has often been observed [ 5 , 14 – 18 ]. Such a lag phase may reflect the time required for substrates of methanogenesis to reach a threshold concentration supporting growth, or the time required for development of an adapted methanogenic community. Some studies have highlighted the importance of residual aged manure in a storage acting as an inoculum, which suggests that community development is central for the temporal dynamics of CH 4 emissions [ 19 – 21 ]. Thus, short-term changes due to temperature variation may reflect the activity of an existing methanogenic community, while long-term changes include the effects of successional changes of the community. Recent measurements of CH 4 production rates in manure and digestate at temperatures between 5 and 52°C [ 22 ] highlight the difference between short- and long-term responses. These measurements show that the short-term (hours to days) response to a change in temperature is generally a shift away from the optimum of the active methanogenic community, but over time the activity at the new temperature will increase ( Fig 1 ). Although some of the long-term differences in CH 4 production were undoubtedly due to changes in substrate availability, considering them would tend to magnify the differences between short- and long-term responses. The general trend shown in Fig 1 is typical in studies of temperature change during anaerobic digestion [ 23 – 26 ]. 10.1371/journal.pone.0252881.g001 Fig 1 Example of short- and long-term responses of methane production to temperature change. Differences between short- and long-term response to temperature change measured by Elsgaard et al. [ 22 ]. Labels identify source: C = cattle manure (from barn), D = fresh digestate (directly from anaerobic digester), S = stored (> 1 month) digestate. Red arrows show short-term effects (differences between samples from the same source when incubated for 17 hours), and blue arrows apparent long-term effects (differences for samples stored for short and long time (weeks or months) at the same temperature)). The distinction between short- and long-term temperature responses was recently quantitatively addressed through development of an anaerobic digester model that included gradual changes in the temperature optimum of kinetic parameters for a single population of methanogens [ 27 ]. However, an empirical approach was used that does not explicitly represent the underlying mechanism. Gene sequencing has revealed that temperature changes shift the relative abundance of taxonomic groups of methanogens, indicating that changes in CH 4 production rates are due to selective growth of adapted methanogenic populations, rather than adaptation of an already established consortium [ 24 , 28 , 29 ]. Presumably the response to environmental stresses other than temperature also varies among methanogenic populations [ 30 – 32 ], and accordingly models need to include multiple groups of methanogens with different responses to temperature, and perhaps other stressors, in order to accurately predict both lag phases and the difference between short- and long-term response to changes in temperature and the chemical environment. This discussion also highlights an important challenge for prediction of CH 4 emission: measurements of short-term temperature responses in the laboratory, however careful, may not reflect long-term seasonal or geographic responses that are important for total CH 4 emissions. Besides methanogens, there is evidence that sulfate reducing bacteria can affect CH 4 emission by competing for substrate (acetate or hydrogen) [ 33 , 34 ], or by the production of inhibitory hydrogen sulfide (H 2 S) [ 35 , 36 ]. This is particularly important for acidification of liquid manure with sulfuric acid, where prolonged suppression of CH 4 emission has been observed, even when pH returned to or remained at near-neutral [ 15 , 37 ]. In the absence of suitable electron acceptors, processes other than methanogenesis are not expected to play a major role beyond fermentation in this anaerobic environment [ 38 ]. Although sulfate may be used to oxidize ammonia [ 39 ], it is unlikely that this autotrophic process is important in organic-rich slurry. At the slurry-air interface, there is a potential for production of nitrous oxide (N 2 O) through nitrification and denitrification [ 40 ], as well as for bacterial methane oxidation [ 41 , 42 ], but in both cases this depends on the development of a partly dry surface crust. While the model described below includes oxidation of organic matter in the surface layer, a crust represents a different environment that is outside the scope of the presented model. Still, these processes are part of a complete assessment of greenhouse gas emission from livestock operations. Existing mechanistic models of organic matter degradation in anaerobic digesters describe biochemical pathways in detail, but are difficult to apply to highly variable manure environments due to lack of data. For example reported hydrolysis rate constants for slurry vary between 0.004 and 0.13 d -1 [ 43 – 45 ], possibly due to effects of feeding practice, manure management, or manure age [ 43 , 46 ]. The main substrates for methanogenesis in slurry are volatile fatty acids (VFAs) and hydrogen [ 47 , 48 ], but hydrogen is mainly derived from VFA oxidation [ 49 , 50 ], and its regulatory role may be exaggerated [ 51 ]. Consequently, some models and simplified versions of detailed models merge hydrogen and VFA consumption kinetics to more simply represent organic matter degradation pathways [ 8 , 11 ]. The considerations presented above strongly support the need for a new approach to predict CH 4 emission from stored manure that describes both short- and long-term responses to management and storage conditions—including temperature and substrate availability—more accurately than current models. We propose that these responses can be accurately described using a simple dynamic model that includes multiple methanogenic populations with different temperature responses. Both inhibition and competition can easily be incorporated into this framework. The objective of the study was to develop and implement this approach as a new mechanistic model that can be used to better understand and predict of CH 4 emission from slurry storage environments, including pits or channels inside barns and outside storage facilities.",
"discussion": "Results and discussion Model behavior In this section effects of residual slurry and methanogen enrichment, temperature changes, and pH on predicted CH 4 production are presented in order to show the behavior of the model. Residual fraction and methanogen enrichment The residual fraction of slurry after export and the enrichment of active methanogens in the residue are expected to enhance methane production from fresh excreta. Fig 5 shows predicted effects of varying the residual fraction of slurry between 1 and 50% with or without enrichment of methanogens in the residue. Total methanogen biomass ( Fig 5a ) and CH 4 production ( Fig 5b ) were correlated, and both quantities were substantially reduced when the slurry channel was emptied to 0.5% ( f resid = 0.005) as compared to 10% or 50% ( f resid = 0.1 or 0.5). The differences could be partially explained by the average amount of slurry in the tank, but was primarily a result of the smaller methanogen population being retained when the residual slurry fraction was low. Changing the microbial enrichment factor ( a enrich ) significantly impacted methanogen biomass only for the lowest residual fraction f resid = 0.005. At f resid = 0.5, and f enrich = 5, methanogen retention was effectively close to 100%, resulting in complete consumption of available VFAs, and hydrolysis limiting CH 4 production, which explains why CH 4 production was almost identical for f enrich = 5 and 0. Altogether, the results in Fig 5 demonstrate that a substantial CH 4 reduction can be expected by near-complete emptying of the slurry channel. 10.1371/journal.pone.0252881.g005 Fig 5 Predicted effects of residual slurry fraction on methanogens and methane production. (a) Total methanogen biomass and (b) Cumulative CH 4 emission as affected by the residual fraction ( f resid ) of slurry after slurry removal assuming a high degree of microbial enrichment ( a enrich ), or no enrichment. Temperature effects The implementation of temperature sensitive methanogen groups allowed for simulating effects of gradual (seasonal) temperature changes, as well as transient and prolonged effects of rapid temperature change (i.e. slurry export from a barn to an outside storage, or from the animal tract to a slurry channel or pit). Fig 6a shows methanogen groups responding to a seasonal change in slurry temperature (temperatures from Kariyapperuma et al. [ 6 ]). It was predicted that at high residual slurry fraction ( f resid = 0.95) the methanogen biomass ( Fig 6a ) and CH 4 emission ( Fig 6c ) will peak when the temperature peaks, but the response to the temperature change was not immediate neither with increasing nor decreasing temperature. The delay reflects the time required for a change in microbial biomass. On the other hand, significant hysteresis was observed when the residual fraction was low ( f resid = 0.1), with methanogen biomass and CH 4 emission peaking 1–2 months after the temperature. A similar delay was seen in Kariyapperuma et al. [ 6 ], where relatively large portions of manure were added and removed, similar to the use of a small residual fraction. In Fig 6c , a large CH 4 emission spike was predicted in response to increasing temperature, which reflects the rapid consumption of VFAs that had accumulated during the preceding cold period. This result reflects a low sensitivity of hydrolysis and fermentation to decreasing temperature in the model, compared to methanogenesis, although in reality this dynamic may be more complex. However, the accumulation of VFAs after rapid temperature changes has been observed in multiple studies [ 24 , 27 , 85 ]. In Fig 6b and 6d the effects of an instantaneous temperature decrease lasting 10 or 300 days are shown. Methanogen numbers declined only slightly during a 10-day temperature decrease ( Fig 6b ), resulting in a significant CH 4 spike once the temperature was raised again ( Fig 6d ). In contrast, during a 900-day temperature decrease, m3 completely disappeared, resulting in a longer period of low CH 4 production when the temperature later increased. Concomitantly, while a surplus of VFA substrate was transiently available after the temperature increase ( Fig 6d ), m4 grew alongside m3, but as VFAs were depleted, m4 was outcompeted again. In reality both m3 and m4 may need to acclimatize to the environment before consuming VFAs, which in the model could be accounted for by implementing a lag phase where CH 4 production from a single methanogen population is temporarily halted in order to adjust cellular metabolism to new conditions, before resuming CH 4 production. This mechanism, however, seems very difficult to describe with equations relating to actual biochemical processes due to complexity and lack of knowledge about the lag phase [ 86 ]. 10.1371/journal.pone.0252881.g006 Fig 6 Predicted temperature effects on methanogens and methane production. (a) Methanogen biomass and (c) CH 4 emission during gradual slurry temperature changes as predicted with a high and low residual slurry fraction in the storage (f resid ). (b) Methanogen biomass and (d) CH 4 emission during short- and long-term temperature changes using a large residual fraction of slurry ( f resid ) of 0.95. The new model effectively assumes a linear temperature dependency of long-term CH 4 production rate (under non-limiting conditions), an assumption also made in other models [ 27 , 51 , 87 ]. However, for short-term dynamics of CH 4 emissions there is solid evidence for an Arrhenius-like temperature dependency of hydrolysis and methanogenesis [ 22 , 45 , 58 ], and the long-term link between temperature and CH 4 production rate remains to be studied systematically, particularly in the psychrophilic temperature range where slurry is often stored. Acidification Acidification to suppress ammonia volatilization has been shown to reduce CH 4 emissions from cattle and pig slurry by 70–90% [ 15 , 31 , 37 ]. Fig 7 shows the predicted response to an instantaneous drop in pH, as in acidification in a storage tank. The immediate reduction in pH resulted in net microbial decay ( Fig 7a ) while immediately reducing CH 4 emissions ( Fig 7b ). The apparent dominance of the m3 group during low pH was a consequence of its naturally higher abundance in the fresh slurry that was added each day. Once pH increased, the methanogens recovered and CH 4 emissions rose again. However, in practical slurry management systems, the pH drop is typically achieved by sulfuric acid treatment, which inevitably raises the SO 4 2- concentration to a level where sulfate reducing bacteria gain a thermodynamic advantage over methanogens. Therefore the model includes an optional sulfate reducer group (sr1) as demonstrated in S5 Appendix . Inclusion of sr1 decreased the magnitude and delayed by several months the CH 4 peak after the pH was raised again. This response resulted primarily from increased competition between methanogen groups and sr1 for VFA substrate when SO 4 2- was abundant, reflecting the known competition between the two groups [ 88 ]. A simulation with sr1 is probably more realistic for modelling acidification of slurry with sulfuric acid, but has the disadvantage of introducing additional parameters with associated uncertainty. 10.1371/journal.pone.0252881.g007 Fig 7 Predicted pH effects on methanogens and methane production. (a) Methanogen biomass and (b) CH 4 emission responses to pH changes. The residual fraction of slurry ( f resid ) was set to 0.95. Similar to the effect of pH, the inhibiting effects of total ammoniacal nitrogen (TAN) and H 2 S were modelled by factoring a term directly onto the substrate utilization rate. An example of TAN inhibition is presented in S6 Appendix . Sensitivity analysis Cumulative CH 4 production was most sensitive to the hydrolysis rate constant parameter, α opt , and the temperature input variable, T ( Fig 8a and 8b ). The model response is relatively insensitive to increases in the Monod parameters, but sensitive to decreases in the yield ( Y i ) and maximum substrate utilization rate ( q max , opt ). It is important to state that non-default parameters may significantly change model sensitivity to other parameters. Hydrolysis remains the least well-defined step in anaerobic digestion [ 43 ], and it can be considered the most critical input for the model performance. Significant effort should therefore be made to determine α opt . 10.1371/journal.pone.0252881.g008 Fig 8 Model sensitivity to parameters and input variables. Model output sensitivities to (a) parameters and (b) input variables. Initial microbial biomass refers to changes in both the initial concentration of methanogens in the slurry inoculum and the fresh influent slurry. For parameter values, see S3 Appendix . Model application In Fig 9 the model was applied to the case study of Kariyapperuma et al. [ 6 ] to show the qualitative responses of the model against real measurements. The simulation was run with different hydrolysis rates ( Fig 9a ), which resulted in CH 4 emission peaks of different magnitude, with α opt = 0.02 matching best with measurements in terms of peak height. However, emissions were predicted to occur much too soon. The substrate utilization rate ( q max , opt ) was reduced, which postponed the CH 4 emission peaks to match better with data ( Fig 9b ). As suggested in Fig 4b , methanogenesis is rate-limiting at low temperatures, explaining the observed effect of reducing q max , opt . There is a need for model validation in controlled experiments, and with the necessary input data. The total concentration and relative fractions of degradable and slowly degradable particulate material can be measured [ 82 ], but this is rarely done. Furthermore, emission studies on full-scale storages often report only the composition of slurry in the storage and not information about the influent slurry composition [ 6 , 89 ]. The new model presented here relies on characterization of the influent slurry composition to determine hydrolysis rate and CH 4 potential, and we stress therefore the importance of accurately measuring degradable particulate material and VFAs for accurate prediction of CH 4 emission. 10.1371/journal.pone.0252881.g009 Fig 9 Model application to case study. Model results versus measured [ 6 ] CH 4 emission from a full scale slurry tank with periodic slurry introduction applying different (a) hydrolysis rates (α opt ) and (b) maximum substrate utilization rates at optimum temperature ( q max , opt )."
} | 5,574 |
35702426 | PMC9115875 | pmc | 2,922 | {
"abstract": "Although microbial fuel cells (MFCs) have been widely studied as wastewater treatment technologies that convert organic matter to electricity, there are few reports of large-scale MFCs that treat both organic matter and nitrogen compounds. In this study, a 226 L reactor equipped with 27 MFC units was partially aerated at 10% of its total volume. The MFC unit consists of a cylindrical air core covered with a carbon-based air cathode, an anion exchange membrane, and a graphite non-woven fabric anode. The air-cathode MFC with 13 L min −1 aeration rate produced a current density of 0.0012–0.15 A m −2 with 40 to >93% biological oxygen demand (BOD) removal to have an effluent BOD of <5–36 mg L −1 at a hydraulic retention time (HRT) of 12–47 h. Meanwhile, 55 ± 17% of the total nitrogen (TN) was removed, resulting in 9.7 ± 3.8 mg L −1 TN in the effluent, although the TN removal was limited at ≥20 °C. The mono-exponential regression for BOD and TN (≥20 °C) estimated that an HRT of 21 h could meet the Japanese effluent quality standards of BOD and TN. Calculation of the total energy recovered via current generation and energy consumed by aeration suggested an energy consumption of 0.22 kW h m −3 . Decreasing the aeration rate and HRT in the reactor would further reduce energy consumption and increase energy production.",
"conclusion": "4 Conclusions The present study revealed that 10% partial aeration in air-cathode MFCs facilitates the meeting of the discharge standards for BOD and TN with an energy consumption of 0.22 kW h m −3 . However, it decreased the current with a reduction in current recovery and dominated the total energy consumption. For practical application as a less energy consuming treatment, decreasing the aeration rate and shortening the HRT by increasing the biomass are required.",
"introduction": "1. Introduction Municipal wastewater treatment systems have received much attention as infrastructure that recycles biomass energy and nutrients 1 and maintains water quality. Energy self-sufficient wastewater treatments can be realized by recovering energy from the wastewater treatment process and improving energy consumption efficiency. 2 However, practical energy capture is limited to the use of sludge for biogas fermentation and solid fuel, while dissolved organic matter in wastewater rather consumes energy to meet discharge standards and thus is not recovered as an energy source. Chemical oxygen demand (COD) removal consumes approximately 0.6 kW h kg COD −1 of energy via aeration, 3 and total nitrogen (TN) removal 4 consumes 6.08 kW h kg TN −1 , which accounts for approximately half of the total energy consumption. 1 In terms of the volumetric energy density, conventional activated sludge consumes 0.27–1.89 kW h m −3 energy for organic matter removal. 2,3 Microbial fuel cells (MFCs) have received much attention as a promising technology for simultaneously recovering energy and treating wastewater without aeration; 5 however, there are still challenges in power production with a low concentration of organic substrate. For instance, an insufficient supplement of organic fuel to anodic microbes raises the anodic potential with low external resistance and leads to a higher anode resistance. 6 Increasing the specific surface area of anode 7–9 or flow velocity 10 successfully increases the current recovery and COD removal efficiency. Further improvement of the anode results in a comparatively higher cathode resistance and separator membrane resistance. 11 Although MFCs have progressed, the primary goal of MFC wastewater treatment is to provide good quality effluent to meet discharge standards. MFCs are capable of successfully treating domestic wastewater without aeration, even at a > 100 L scale. 6,12–19 However, the COD of the effluent from anaerobic MFC was higher than 81 mg L −1 with less than 43 h of HRT and rarely met discharge standards. 6,12–19 However, there have been few exceptions; dual-chambered MFC with an aerated cathode chamber yielded an effluent with a COD 25 mg L −1 . 20 Thus, MFCs generally require post or partial treatments such as aeration, 21 anaerobic membrane filtration, 22 and activated carbon filtration. 19 Furthermore, partial or post-aeration conferred external oxidizing power for nitrogen removal. The aerated cathode chamber of dual-chambered MFCs 20 successfully removed nitrogen from municipal wastewater. 20,23–25 Nitrogen removal has also been achieved without aeration in an air-cathode MFC, especially with a gas diffusion layer (GDL) 26 and at a relatively smaller scale owing to the high GDL area/wastewater volume ratio for sufficient oxygen supplement via GDL. An anaerobic MFC using persulfate as oxidant in the cathode chamber successfully demonstrates an anammox reaction. 27 An air-cathode MFC separated with an ion exchange membrane cannot achieve nitrogen removal and requires additional air supply via effluent sprinkling 18 or partial aeration. Thus, nitrogen removal from municipal wastewater has been optimized for various types of MFCs and is still at the stage of feasibility, along with MFC itself. A one-meter deep air-cathode MFC with an anion exchange membrane (AEM) was demonstrated for the first time in our previous study, 28 and it successfully recovered electric power from sewage wastewater over a year. 6 Previous studies have revealed that a comparison of electric power production by MFCs using AEMs and other separators shows the advantage of AEMs in mitigating pH imbalances that are often observed in MFCs with cation exchange membrane or GDL. 29–31 However, the MFC with AEM could not remove TN and was estimated to require 90 W h m −3 of external energy for aeration to decrease TN from 30 to 15 mg L −1 . In this study, an air-cathode MFC with an AEM was operated in a 226 L reactor with partial aeration, and the current production, biological oxygen demand (BOD) removal, nitrogen removal, and energy efficiency were evaluated.",
"discussion": "3. Results and discussion 3.1 Current production \n Fig. 2 shows the current densities produced in each of the 27 MFC units during operation with continuous primary sedimentation tank effluent inflow. The operation time represents the time elapsed since the start of aeration in the reactor, which originally ran for more than 500 days under anaerobic conditions. 6 The HRTs varied in the range 12–47 h throughout the operation owing to the accidental clogging of the tubing pump ( Fig. 2A ). The current production was almost zero at 0–19 days with an aeration rate of 34 L min −1 and then gradually increased to 0.15 A m −2 at day 20 after the aeration rate decreased to 13 L min −1 ( Fig. 2B ), indicating that aeration inhibits the current production. The current density drastically decreased and approached zero with the increase in HRT to 23–36 h (days 34–63). The decrease in HRT to 17–20 h again increased the current production, resulting in an average current density of 0.015 A m −2 (days 64–93). The lack of current production at days 94–110 decreased the current due to the clogged inflow tube during the long vacation. Thereafter, the MFCs maintained the current production response to HRTs until the end of the operation on day 184 ( Fig. 2B ). The lower current at longer HRT reflects the lower BOD in the reactor that resulted from higher BOD removal. Fig. 2 Current production in 27 microbial fuel cell (MFC) units in the continuous flow reactor. Panels (A) and (B) indicate the hydraulic retention times (HRTs) and current densities, respectively. The partial aeration at 10% volume significantly decreased current production using MFCs even in the presence of partition; 34 L min −1 of aeration almost eliminated the current production using MFC, and 13 L min −1 of aeration reduced the current production using MFC without aeration by 59–71% at an HRT of 12–24 h. 6 Inhibition of current production by aeration in the anodic chamber has been reported in other MFCs. 32 The oxygen of the alternative oxidant increases the anode potential by eliminating the charge on the anode, resulting in lower current. Because the current recovered immediately after decreasing the aeration rate, temporary exposure of the biofilm to oxygen did not inhibit current production irreversibly. Another reason for the decrease in current is degradation of BOD by aerobic bacteria, resulting in a lower CE. 33,34 The overall trend observed was that a lower aeration rate and a shorter HRT produced more current, as observed for all MFCs, although the current varied in each MFCs due to position and unit variances. The highest current was obtained in G9, followed by G8 and G7. The higher current in the MFCs in the vicinity of aeration is possibly attributed to the enhancement of substrate supplement to anode by turbulence in wastewater due to aeration. 10 3.2 BOD removal \n Fig. 3 shows BOD removal by MFCs with partial aeration in the continuous flow reactor. The influent BOD (BOD IN ) was maintained at 64 ± 15 mg L −1 throughout the study. Partial aeration (10%) at 34 L min −1 (day 0–19) yielded effluent BOD (BOD EF ) ranging from <5 mg L −1 to 12 mg L −1 at an HRT of 18–24 h. The decrease in partial aeration rate to 13 L min −1 did not remarkably change BOD EF , which ranged from <5 mg L −1 to 21 mg L −1 at corresponding HRTs of 18–24 h. HRT affected BOD-removal more than aeration rate; HRT < 19 h (12–18 h) reduced BOD by 40–88%, with a BOD EF of 5.9–36 mg L −1 , whereas HRT of >19 h (19–47 h) resulted in 69% to >93% of BOD removal efficiency to yield BOD EF of <5 mg L −1 to 20 mg L −1 . Considering the permitted discharge water quality standards of the activated sludge process, which is 15 mg BOD L −1 , as per the Sewerage Act in Japan, 19 h of HRT with partial aeration (10% volume) is the optimum condition. Fig. 3 Biological oxygen demand (BOD) removal by the continuous flow reactor with microbial fuel cells (MFCs) and partial aeration. Panel (A) and (B) indicate the hydraulic retention times (HRTs) and BODs, respectively. Red-line closed plots of effluent BOD (BOD EF ) and BOD removal efficiency (BOD-RE) represent BOD removal efficiency values at the detection limit (<5 mg L −1 ). Limited aeration in wastewater treatment is a rapidly growing strategy to save energy and achieve the simultaneous removal of organic matter and TN. An HRT of 19 h with partial aeration at 13 L min −1 in a 200 L reactor corresponded to 74 100 L HRT −1 m −3 and achieved better removal of organic matter and TN better than the reactors with limited aeration. A microbial electrochemical system (MES) with 0.5/5.5 min intermittent aeration at an air-flow rate of 400 L min −1 achieved 92% COD-reduction to 22 mg L −1 within 7 h of HRT; the accumulated airflow was calculated to be approximately 20% of that in the minimum aeration performed in this study. 20 Other studies with limited aeration, such as vertical flow constructed wetlands 35 and up-flow partially aerated biological filters, also achieved sufficient COD reduction with aeration, corresponding to 20% and 85% of that obtained in this study, respectively. These results suggested that a maximum 80% of aeration can be reduced (from 13 to 2.6 L min −1 ) to achieve sufficient effluent quality. The airflow affected current rather than BOD removal, and an 80% reduction in airflow can produce more current with lower energy consumption. 3.3 Nitrogen removal in the chemostat reactor The influent TN (TN IN ) was 21 ± 4.7 mg L −1 throughout the operation ( Fig. 4 ) and was successfully reduced in the reactors at ≥20 °C ( Fig. 4B ). The TN removal efficiency (TN-RE) achieved was 55 ± 16%, regardless of aeration rate and yielded an effluent TN (TN EF ) of 9.4 ± 3.5 mg L −1 ( Fig. 4C ). The data at day 17 had a significantly low TN-RE despite a longer HRT, which was probably attributed to the lower TN IN . In contrast, following day 92, TN-RE declined drastically to 13 ± 14% at <20 °C with TN EF of 19 ± 4.2 mg L −1 at an HRT of 12–47 h ( Fig. 4B, C ). Thus, the TN EF and TN-RE were more sensitive to water temperature than aeration rate and HRT, and a temperature of ≥20 °C was preferable for TN reduction. Fig. 4 Nitrogenous compound removal in the continuous flow reactor with microbial fuel cells (MFCs) and partial aeration. Panels (A) and (B) demonstrate the hydraulic retention times (HRTs) and water temperature, respectively. Panels (C) and (D) show the total nitrogen (TN) removal and the constituted nitrogenous compounds, respectively. \n Fig. 4D shows the breakdown of TN, i.e. , NH 4 + , NO 2 − , and NO 3 − . TN IN mostly comprised NH 4 + , which was oxidized to NO 2 − and NO 3 − by partial aeration. NH 4 + removal by nitrification was supported by the absence of nitrification ability in the air-cathode MFC operated anaerobically in a previous study. 6 This is typical for air-cathode MFCs using IEM instead of GDL. Therefore, the NH 4 + removal is attributed to nitrification in the aeration tank rather than the MFC itself, although only the total removal was available with no breakdown of how the anaerobic and aerobic parts contributed. The imbalance between NH 4 + and the oxidation products and the decrease in NH 4 + suggested further denitrification. During day 0–19 with an aeration rate of 34 L min −1 , 19–86% of the TN EF (0.58 ± 0.12 mM) remained as NO 3 − (0.13–0.53 mM), indicating insufficient denitrification due to excessive aeration. During day 19–84 with a temperature ≥ 20 °C and an aeration rate of 13 L min −1 , 81 ± 25% of the TN EF (0.69 ± 0.27 mM) comprised NH 4 + (0.60 ± 0.34 mM). NO 3 − concentration was 0.18 ± 0.11 mM during day 34–62. Additionally, after day 92 with <20 °C, the TN EF remained as NH 4 + at 1.3 ± 0.31 mM. TN removal generally involves aerobic nitrification (conversion of NH 4 + to NO 3 − ) and subsequent anoxic denitrification (conversion of NO 3 − to N 2 ). Hence, BOD and NH 4 + are competitive electron donors for aerobic bacteria, while electrode and NO 3 − are competitive electron acceptors for anaerobic bacteria. In addition, NH 4 + has been reported to be the electron donor for current production by using sulphate peroxide as the oxidant in MFCs. However, there was no result suggesting that either of these competitions or the ammonia-driven current production occurred in this study. The most sensitive factor affecting TN removal was the water temperature in the MFC reactor, rather than the aeration rate, and 20 °C seemed to be the turning temperature for TN reduction. This is well agreed by the Arrhenius relationship for oxidation of ammonia, which suggests activation energies of 87.1 and 38.6 kJ mol −1 in the temperature ranges 10–20 °C and 20–30 °C, respectively. 36 In contrast, nitrite oxidation has an activation energy of 34.2 kJ mol −1 in the range of 10–30 °C. These calculations suggest that the reduction in TN removal can be attributed to the requirement of twice the activation energy for ammonia oxidation. Dissolved oxygen (DO) has been recognized as an important factor for nitrification; 0.5 mg L −1 of DO is required for oxidation of both nitrite and ammonia. 37 The DO in the aeration tank was approximately 1.0 mg L −1 (data not shown), with the minimum aeration rate in this study. This suggests that the aeration rate could be further decreased. 3.4 Calculation of BOD and TN degradation and current recovery BOD degradation was determined using mono-exponential regression with BOD IN and BOD EF at different HRTs, resulting in the value C = 63e −0.069 t ( Fig. 5A ). Based on this calculation, an HRT of 21 h was required to meet the discharge standard of 15 mg L −1 . The degradation rate constant (0.069) in partial aeration MFCs was higher than that of the MFCs without aeration (0.046) (ref. 6 ) (ESI Fig. S1 † ), and the HRT was reduced by approximately half to meet the permitted standard discharge water quality. The TN removal was calculated as C = 22e −0.037 t and C = 19e −0.0019 t at ≥20 °C and <20 °C, respectively. Thus, the MFC operation at ≥20 °C of water temperature and 10 h of HRT is necessary to meet the effluent quality standards of TN, i.e. , 15 mg L −1 for advanced wastewater treatment by the Sewerage Act, Japan. Fig. 5 Calculation of effluent BOD (BOD EF ) and effluent TN (TN EF ) based on the mono-exponential regression. Panels (A) and (B) show the HRT-dependent plots of BOD and TN, respectively. Red and blue plots are the data at ≥20 °C and <20 °C, respectively. The measured CEs ranged from 0.078 to 21% (average CE: 4.9 ± 5.3%) and varied widely even under similar HRT and BOD (equivalent to BOD EF ). There was no trend between the CE and HRT, BOD IN , and BOD EF (ESI Fig. S2 † ). Therefore, the CE is assumed to be a constant value independent of the HRT and BOD concentrations and is determined using the measured current and BOD EF ( C = 63e −0.069 t ) to obtain the best fit for eqn (2) . The determined CE was 2.5%, which was approximately half of the measured average, although the calculated current seemed to well demonstrate the observed current ( Fig. 6A ) and the electric power ( Fig. 6B ). The electrical energy and EGE varied and did not reproduce the experimental data well ( Fig. 6C and D ). Fig. 6 Comparison of measured and calculated electric power production. Panels (A), (B), (C), (D) indicate current density (A), electric power density (B), electric energy density (C), and energy generation efficiency (EGE) (D), respectively. The CE obtained in this study drastically decreased from 24 ± 13% to 4.9 ± 5.3% due to aeration (HRT: 12–42 h) (ESI Fig. S3 † ). 6 The CE with partial aeration was still higher than that in the biocathode-MES using intermittent-aerated cathode portion in half of the total wastewater (calculated as 0.64%), 20 although the aeration rate was much higher in the MFC operated in this study. In our study, the CE of the MFC operated with aeration was comparable to that of two anaerobic MFCs: an air-cathode MFC with a GDL 12,14 and an air-cathode MFC with a cation exchange membrane. 18 This contradiction suggests that CE was determined by multiple factors rather than just aeration, which included the concentration of organic matter, electrode or separator, GDL specific surface ratio, and cathode reaction rate. 3.5 Towards practical application of MFC TE, the sum of the total energy generated by the MFC and the energy consumed by partial aeration, was calculated for the treatment of 1 m 3 of sewage at an HRT of 21 h, which enabled both BOD and TN to meet the discharge water qualities. Energy production was negligible (0.013 W h m −3 ), and energy consumption by partial aeration (220 W h m −3 ) dominated the TE. The TE (0.22 kW h m −3 ) corresponds to 73–11% of conventional activated sludge (0.30–1.89 kW h) and 50–11% of that in oxidation ditch plants (0.44–2.07 kW h m −3 ) in Japan. 38 The presented total energy could reduce energy consumption but is less effective in comparison to other advanced technologies. For example, the biocathode-MES achieved 12% energy consumption of the conventional activated sludge 20 with an HRT of 5 h. The superiority of MES over MFC is attributed to its shorter HRT requirement. The system combining anaerobic digestion and nitrification–anammox achieved 96% COD and 81% TN removal, with a net energy consumption of 0.09 kW h m −3 . 39 The high energy-consumption efficiency is attributed to the lack of aeration in the system. Thus, decreasing the aeration rate in the present reactor or lack of aeration and shortening the HRT by increasing biomass with more anode specific surface area 40 are required to reduce energy consumption and increase energy production."
} | 4,940 |
29255486 | PMC5729428 | pmc | 2,923 | {
"abstract": "Background Recent studies have suggested that addition of electrically conductive biochar particles is an effective strategy to improve the methanogenic conversion of waste organic substrates, by promoting syntrophic associations between acetogenic and methanogenic organisms based on interspecies electron transfer processes. However, the underlying fundamentals of the process are still largely speculative and, therefore, a priori identification, screening, and even design of suitable biochar materials for a given biotechnological process are not yet possible. Results Here, three charcoal-like products (i.e., biochars) obtained from the pyrolysis of different lignocellulosic materials, (i.e., wheat bran pellets, coppiced woodlands, and orchard pruning) were tested for their capacity to enhance methane production from a food waste fermentate. In all biochar-supplemented (25 g/L) batch experiments, the complete methanogenic conversion of fermentate volatile fatty acids proceeded at a rate that was up to 5 times higher than that observed in the unamended (or sand-supplemented) controls. Fluorescent in situ hybridization analysis coupled with confocal laser scanning microscopy revealed an intimate association between archaea and bacteria around the biochar particles and provided a clear indication that biochar also shaped the composition of the microbial consortium. Based on the application of a suite of physico-chemical and electrochemical characterization techniques, we demonstrated that the positive effect of biochar is directly related to the electron-donating capacity (EDC) of the material, but is independent of its bulk electrical conductivity and specific surface area. The latter properties were all previously hypothesized to play a major role in the biochar-mediated interspecies electron transfer process in methanogenic consortia. Conclusions Collectively, these results of this study suggest that for biochar addition in anaerobic digester operation, the screening and identification of the most suitable biochar material should be based on EDC determination, via simple electrochemical tests. Electronic supplementary material The online version of this article (10.1186/s13068-017-0994-7) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusions This study confirms the capacity of different biochar materials of enhancing the anaerobic methanogenic conversion of food waste fermentate. This is achieved most likely by accelerating rate-limiting interspecies electron transfer (IET) processes (e.g., between acetogens and methanogens), critically involved in the syntrophic conversion of organic substrates, such as volatile fatty acids. Compared with other conductive materials such as iron oxide particles or activated carbon, biochar holds several advantages such as the lower cost (particularly with reference to the activated carbon) and the unique possibility to be sustainably and safely disposed of along with the digestate, thus eliminating the need for further separation and treatment costs. Interestingly, all the herein tested biochars were found to enhance comparably the methanogenic conversion process relative to unamended (or sand-amended) controls, despite being characterized by remarkably different physico-chemical properties (e.g., electron-donating, electron-accepting capacities, surface area, bulk electrical conductivity), with these latter being primarily dependent on the biochar production conditions (e.g., pyrolysis temperature), and the starting lignocellulosic material. Importantly, for the first time, we could demonstrate that the positive effect of biochar is directly related to the electron-donating capacity (EDC) of the material, yet virtually independent of its bulk electrical conductivity, specific surface area, and electron accepting capacity, properties which were all previously hypothesized to play a major role in the DIET process. Hence, although these results will certainly have to be consolidated through the analysis of the impact of biochars on the anaerobic digestion under a broader range of conditions, they clearly suggest that estimation of EDC via mediated electrochemical tests or hydrodynamic electrochemical techniques with rotating disc electrodes (RDE) [ 61 ] may be a straightforward strategy to individuate and select the most effective biochar material and devise possible application strategies.",
"discussion": "Results and discussion Influence of biochar particles on the methanogenic degradation of FWF Figure 1 a–d shows the time course of VFAs concentration in the different treatments, during the 1st feeding cycle. Interestingly, a statistically significant ( p < 0.05) effect of biochar was observed only on propionate degradation, which commenced earlier (on day 55 vs. day 76) and proceeded at a higher initial rate in bottles supplemented with conductive particles, relative to the unamended controls (Fig. 1 c). By contrast, no statistically relevant differences were observed on acetate and butyrate profiles. Fig. 1 Time course of total VFAs ( a ), n -butyrate ( b ), propionate ( c ), and acetate ( d ) concentration in bottles supplemented with conductive particles and in unamended controls, during the 1st feeding cycle. Error bars represent the standard deviation of two replicate bottles \n As for methane production, no statistically relevant differences were observed among treatments (Fig. 2 ), although in all biochar-supplemented bottles methane production (as determined from chromatographic analysis carried out on gas-phase samples) was slightly delayed with respect to the unamended controls, consistently with the reported capacity of biochar to adsorb methane gas [ 58 ]. Fig. 2 Time course of methane formation yield (%) from fermentate VFAs, in bottles supplemented with conductive particles and in unamended controls, during the 1st feeding cycle. Error bars represent the standard deviation of two replicate bottles \n At the end of the 1st feeding cycle (i.e., on day 99), in all bottles methane production accounted for 92–110% of initial VFA (on a COD basis), regardless of the presence of conductive particles (Fig. 2 ). This finding suggested that methane almost exclusively derived from VFA, while the non-VFA fraction of the COD of the fermentate (approx. 15%) was somewhat recalcitrant to the anaerobic degradation. On day 126, all bottles were flushed to remove the produced methane and were re-spiked with a 2nd dose of freshly prepared FWF, which could have a slightly different initial composition and concentration with respect to the FWF used during the first feeding cycle. As expected, during this 2nd feeding cycle, in all treatments the overall FWF transformation proceeded at a substantially higher rate compared to the 1st cycle (Fig. 3 a–d), most likely due to biomass growth and acclimation on the supplied substrates. Indeed, the complete (91–106%) methanogenic conversion of fermentate VFAs, spiked at an initial concentration of approximately 1500 mgCOD/L, was achieved in less than half of the time needed in the 1st feeding cycle (43 days vs. 99 days). Fig. 3 Time course of total VFAs ( a ), n -butyrate and i -butyrate ( b ), propionate ( c ) and acetate ( d ) concentration (mgCOD/L) in bottles supplemented with conductive particles, with non-conductive sand, and in unamended controls, during the 2nd feeding cycle. Error bars represent the standard deviation of two replicate bottles \n Notably, in all biochar-amended bottles the lag phase prior to the onset of VFAs degradation was almost eliminated compared to the unamended controls bottles, whereby it typically exceeded 10 days as in the case of butyrate (Fig. 3 b). Importantly, VFAs concentration profiles in control bottles supplemented with non-conductive silica sand (Fig. 3 a–d) were almost completely indistinguishable from those of unamended controls, hence providing a further line of evidence that the observed stimulatory effect on methanogenic degradation was ultimately linked to the electrical conductivity of particles, as previously suggested in the literature [ 9 ]. Differently from the 1st feeding cycle, isobutyrate (insert of Fig. 3 b) also transiently accumulated during FWF degradation, reaching a peak concentration of 36–50 mgCOD/L on day 136 in biochar-amended bottles, and on day 146 (of approx. 40 mgCOD/L) in unamended controls and in sand-amended controls (insert in Fig. 3 b). In agreement with the observed VFAs time profiles, methane production in biochar-amended treatments proceeded more rapidly to completion with respect to the unamended (or amended with silica sand) controls (Fig. 4 ). Specifically, the fold enhancement of the initial (from day 129 to day 140) methane production rate in biochar-supplemented bottles relative to the unamended controls was 3.9 ± 0.2 for wheat bran biochar, 4.6 ± 0.2 for the wood biochar, and 5.0 ± 0.1 for the orchard biochar. Fig. 4 Time course of methane formation yield (%) from fermentate VFAs, in microcosms supplemented with conductive particles, with non-conductive sand, and in unamended controls, during the 2nd feeding cycle. Error bars represent the standard deviation of two replicate microcosms \n Influence of biochar supplementation on pH and ammonia concentration Throughout the experimental period, in all treatments the reaction pH remained stable in the range of pH 7.0–7.5, irrespective of the presence of biochar or of non-conductive sand. This finding ruled out the possibility that the observed differences in the methanogenic conversion process could be due to differences in pH values among treatments. During to the 2nd feeding cycle, the concentration of ammonia nitrogen (NH 3 -N) was also monitored over time (Fig. 5 ), in order to evaluate whether the observed improvement in the methanogenic conversion process could be due to the biochar particles alleviating ammonia inhibition via physical adsorption [ 48 ]. After an increase during the first 10 days of the 2nd feeding cycle, which was most likely due to the hydrolysis of proteins and/or other ammonia-bearing substrates contained in the FWF, ammonium concentration remained, in all treatments, nearly constant at around 200–250 mg NH 3 -N/L and, hence, values that are substantially lower than those reported to exert inhibitory effects on methanogenic biomass (i.e., > 2 g NH 3 -N/L) [ 59 ]. It is also worth mentioning that, throughout the whole experimental period, the observed differences among treatments were not statistically significant, hence providing further indication that the effect of biochar on methanogenic activity was not related to its direct interaction with ammonia. Fig. 5 Time course of ammonia nitrogen (mg NH 3 -N/L) concentration in microcosms supplemented with conductive particles, with non-conductive sand, and in unamended controls, during the 2nd feeding cycle. Error bars represent the standard deviation of two replicate microcosms \n Physical and electrochemical characterization of biochars The three biochars differed markedly for their lumped physical and electrochemical properties (Table 1 ), which in turn determine their ultimate capacity to interact and/or exchange electrons with soluble (e.g., organic and inorganic substrates) and or insoluble (e.g., microorganisms) components occurring in the surrounding environment. More specifically, the biochars from wheat bran and wood exhibited a BET-specific surface area (55 ± 1 and 61 ± 1 m 2 /g, respectively) that was substantially higher (i.e., > fourfold) than the orchard biochar (13.7 ± 0.5 m 2 /g). Consistently, wheat bran and wood biochars also displayed total pore volumes that were threefold larger (0.0445 and 0.0483 cm 3 /g, respectively) compared to the orchard biochar (0.0165 cm 3 /g). In all biochar samples, porosity was mostly (40–55%) associated with the presence of micropores having an internal diameter of less than 2 nm, hence unlikely accessible to microorganisms. Biochar can donate, accept, and/or transfer electrons to/from soluble and insoluble components occurring in its surrounding environment, either via abiotic or biotic pathways [ 35 , 43 ]. Typically, electron transfer by biochars is reported to involve at least two types of redox-active structures, namely surface-bound quinone/hydroquinone moieties, which are responsible for the electron accepting and donating capacity (i.e., EAC and EDC) of the material, and conjugated π -electron systems associated with condensed aromatic sub-structures of the biochars, which are mainly responsible for its bulk electrical conductivity [ 43 ]. Here, the EAC and EDC of the different biochars were assessed by means of mediated electrochemical experiments, conducted as reported elsewhere [ 43 ]. Further details on the results of mediated electrochemical experiments are included in Additional file 1 : Figures S1–S3. As far as the EDC is concerned, the orchard biochar exhibited the highest value (0.30 ± 0.02 meq/g), followed by the wood biochar (0.20 ± 0.02 meq/g), and the wheat bran biochar (0.05 ± 0.01 meq/g). A somewhat similar trend was observed also for the EAC, with the only exception for the wheat bran biochar, which displayed an unexpectedly high EDC value (0.43 ± 0.05 meq/g) and a substantial difference between the EAC and EDC. In spite of that, however, all EAC and EDC values herein determined fall within the range of those reported in the literature for biochars produced under comparable conditions [ 43 ]. Interestingly, the trend observed for EDC (with wheat bran exhibiting substantially lower values compared to the wood and orchard biochars) is in agreement with the reported effect of pyrolysis temperature on the chemical composition of the produced biochar. Indeed, the wheat bran biochar was obtained at a pyrolysis temperature of 800 °C, whereas the wood and orchard biochars were at 500 °C. Along this line, it has been reported that EDC (and also EAC) typically increases with pyrolysis temperature up to about 400–500 °C before decreasing at higher temperature values as a consequence of the degradation of previously formed quinoid structures. This latter process is frequently coupled with the onset of aromatization of the biochars that typically starts at around 450–550 °C [ 35 ]. The bulk electrical conductivity, a parameter that determines the capability of the material to function as an electron conduit, also differed markedly among the tested biochars, as reported in Table 1 . In particular, the measured electrical conductivities of wood and orchard biochars were similar (i.e., 1.6 and 0.5 S/m, respectively), and nearly one order of magnitude lower than that of wheat bran biochar (49.9 S/m), in agreement with the higher pyrolysis temperature at which this latter material was produced (800 °C) relative to the others (500 °C), which possibly resulted in a higher abundance of condensed aromatic and/or graphitic structures. Correlating methanogenic activity to the physical and electrochemical properties of biochar Despite the increasing number of studies proving the capability of biochar and other conductive materials to stimulate the anaerobic digestion process, little efforts have been made, so far, to correlate the extent of the observed stimulatory effect on methanogenic activity to the specific physical and electrochemical properties of the added materials. As a consequence of this lack of knowledge, the choice of the most appropriate materials to be employed remains based on purely empirical considerations, as its impact on the overall process performance can hardly be predicted. In order to contribute filling this scientific gap, an attempt was made here to correlate the observed performance of the digestion process to the measured biochars properties. To this aim, the fold of increase of the initial methane production rate of the 2nd feeding cycle, relative to the unamended control, was plotted as a function of the EDC, EAC, specific surface area, and electrical conductivity of the different biochars. Interestingly, under the herein applied experimental conditions, a positive linear correlation ( R \n 2 = 0.9967) was observed exclusively with the EDC (Fig. 6 ). It should be emphasized, however, that a statistically relevant difference (i.e., p < 0.05) among biochars was observed only between the orchard biochar and the wheat bran biochar, whereas no relevant differences were apparent between the orchard and wood biochars and between the wheat bran and wood biochars (Fig. 6 a). Fig. 6 Correlation between the initial methane formation rate (mgCOD/L d) and the electron donating capacity (EDC) (meq/g) ( a ); the electron accepting capacity (EAC) (meq/g) ( b ); the electrical conductivity (S/m) ( c ), and the specific surface area (m 2 /g) ( d ) of the different biochars. Legend: (1) wheat bran (this study); (2) wood (this study); (3) orchard (this study); (4) pine [ 29 ]; (5–6) rice straw [ 62 ]; (7–9) rice straw [ 63 ]; (10) corn stover [ 64 ]; (11) pine [ 64 ] \n To strengthen this analysis, data retrieved from the scientific literature, specifically dealing with the impact of biochar on the methanogenic digestion process, are also included in Fig. 6 . Surprisingly, a lack of correlation was observed between the electrical conductivity and the stimulatory effect on methanogenic activity (Fig. 6 c), suggesting that this biochar property was probably over-considered in the scientific literature. Clearly, this finding does not necessarily imply that conductivity is not involved in DIET-driven methanogenic processes [ 60 ], rather it suggests that it is not often rate-limiting process performance under most conditions. The fact that, in previous studies, biochar was found to promote DIET with similar rates and stoichiometry as those observed with granular activated carbon (GAC), despite having a nearly 1000-fold lower electrical conductivity with respect to GAC, seems to support this latter hypothesis [ 31 ]. As far as the specific surface area is concerned, unexpectedly, data reported in Fig. 6 d apparently suggest an inverse correlation between this parameter and the methanogenic activity, with very high surface area biochars being outperformed by low surface area materials. Most probably, this is due to fact that high values of surface area correspond to extremely small pore diameters, often in the range of nanometers, which in turn are too small to be accessible to microorganisms. It should be noted that according to the best of our knowledge, no other literature studies have examined the impact of the biochar EDC and EAC on the methanogenic degradation process, both suggesting that the herein obtained results will necessarily have to be confirmed in future studies and that EDC and EAC measurements probably deserve greater consideration than previously thought. FISH–CLSM analysis At the end of the 2nd feeding cycle, suspensions (containing planktonic cells and small biochar and sand particles) from the different treatments were sampled and analyzed by FISH–CLSM in order to visualize the spatial distribution of Bacteria and Archaea in biochar-supplemented bottles, in unamended controls, and in sand supplemented controls. As expected, irrespective of the treatment, images (Fig. 7 ) revealed the presence of large microbial aggregates, with archaea (in red) laying in close proximity to bacteria (in green), consistent with the fact that a syntrophic association between these metabolically distinct groups of microorganisms is anyhow necessary during the methanogenic conversion of volatile fatty acids mixtures. Silica sand and biochar particles, visualized by their reflection signal, appear gray (Fig. 7 ). Fig. 7 CLSM combined images showing the spatial distribution ( X – Y and Y – Z planes) of Archaea (red) and Bacteria (green) cells identified by FISH in aggregates from the unamended control ( a ), silica sand-supplemented control ( b ), wheat bran biochar ( c ), wood biochar ( d ), and orchard biochar ( e ). Biochar particles, visualized by their reflection signal in the same microscopic field, appear gray. Each image is composed by 32–40 optical sections of the aggregate thickness every 0.4–0.5 μm \n The 3D image reconstruction ( x – z plane) clearly showed an intimate association among prokaryotic cells and particles. Interestingly, the aggregate thickness, ranging between 12 and 20 μm, showed the highest values in unamended control ( A ), thus, suggesting that the presence of particles of both conductive and not conductive materials could stimulate bulk biomass compactness. Confocal microscope analyses of aggregates, in combination with FISH, revealed that the presence of particles also shaped aggregate microarchitecture beyond bulk biomass. Control bottles (Fig. 7 a, b) were found to contain a large number of filamentous archaea, resembling the distinct morphology of Methanosaeta species, along with irregular spheroid bodies occurring alone or typically in aggregates of cells, resembling the distinct morphology of Methanosarcina species. By contrast, in bottles supplemented with conductive materials (Fig. 7 c–e), Methanosaeta -like filaments were almost completely absent, with Methanosarcina -like cells and aggregates accounting for the greatest share of Archaea. Interestingly, previous co-culture investigations clearly pointed out that Methanosaeta and Methanosarcina species were both capable to participate in DIET, either via direct cell-to-cell electron exchange or via conductive carbon-based materials [ 9 , 13 , 14 ]. Collectively, the results of this study, however, suggest that herein used biochar materials specifically favored the growth of M ethanosarcina -like Archaea over Methanosaeta -like archaea. This finding is fully in agreement with the results of a previous study showing that Methanosarcina preferentially enriched, with respect to Methanosaeta , over the surface of coarse biochar particles (2–5 mm) during the methanogenic conversion of glucose [ 47 ]. Further microbiological investigations are, however, required to shed light onto this interesting finding."
} | 5,549 |
36683768 | PMC9827592 | pmc | 2,924 | {
"abstract": "Polyvinylidene fluoride (PVDF) is a favorite polymer with excellent piezoelectric properties due to its mechanical and thermal stability. This article provides an overview of recent developments in the modification of PVDF fibrous structures and prospects for its application with a major focus on energy harvesting devices, sensors and actuator materials, and other types of biomedical engineering and devices. Many sources of energy harvesting are available in the environment, including waste-heated mechanical, wind, and solar energy. While each of these sources can be impactively used to power remote sensors, the structural and biological communities have emphasized scavenging mechanical energy by functional materials, which exhibit piezoelectricity. Piezoelectric materials have received a lot of attention in past decades. Piezoelectric nanogenerators can effectively convert mechanical energy into electrical energy suitable for low-powered electronic devices. Among piezoelectric materials, PVDF and its copolymers have been extensively studied in a diverse range of applications dealing with recent improvements in flexibility, long-term stability, ease of processing, biocompatibility, and piezoelectric generators based on PVDF polymers. This article reviews recent developments in the field of piezoelectricity in PVDF structure, fabrication, and applications, and presents the current state of power harvesting to create completely self-powered devices. In particular, we focus on original approaches and engineering tools to design construction parameters and fabrication techniques in electro-mechanical applications of PVDF.",
"conclusion": "Conclusions Considering that the crystallization of PVDF can be controlled, PVDF is a significant processible inorganic material whose final properties depend on its fabrication method, which has been shown to have excellent ferroelectric properties. Its remanent polarization is 0.05–0.1 C m −2 , leading to large piezoelectricity in which the piezoelectric constant is 24–34 pC/N. The coercive field and Curie temperature of PVDF are larger than those of typical inorganic ferroelectrics. The phase transition in polymorphs of PVDF to achieve high piezoelectric properties, and the mechanism by which the orientation of the dipole moment of a unit cell is changed, are quite interesting. The electrospinning technique demonstrated potent efficacy with high β-phase fraction, as an effective way to develop piezoelectric properties.",
"introduction": "Introduction Piezoelectricity is the ability of certain crystalline materials to develop an electric charge proportional to a mechanical stress. Soon it was realized that materials showing this phenomenon must also show the converse: a geometric strain (deformation) proportional to an applied voltage. 1 Both natural and artificial materials exhibit a range of piezoelectric effects. 2 Although piezoceramics are one of the most efficient piezoelectric materials, a polymer exhibiting transducer characteristics has remarkable advantages over ceramics because of the less brittle nature and greater flexibility of polymers. 3 Various ceramics, such as PZT, 4 ZnO, BaTiO 3 , Na/KNbO 3 , and polymers like (PVDF) and its copolymers with hexafluoropropylene [P(VDF-HFP)] and tetrafluoroethylene [P(VDF-TrFE)], and poly(vinyl acetate) (PVAc) are used as piezoelectric materials. The drawbacks of these materials are brittleness, toxicity, non-biodegradability/non-biocompatibility and complicated fabrication processes. However, polymers show advantages, such as toughness, non-toxicity, and biocompatibility, making them suitable materials for applications requiring high bending and twisting and for biomedical devices. 5 Piezoelectric polymers can be classified as amorphous, crystalline, or semicrystalline structures. Semicrystalline piezopolymers include polyvinylidene fluoride (PVDF). 6 Crystallinity is based on the polymer's molecular structure and is determined by the ratio of semicrystalline to amorphous regions, which changes with fabrication methods and thermal history. Molecular dipoles are primarily responsible for the piezoelectric effect. Randomly oriented dipoles get arranged with thermal annealing and high voltage, and stretching the polymer speeds up the process. The electric voltage in semicrystalline polymers enhances the piezoelectricity, while in amorphous materials, poling near T g is needed to align the dipoles and freeze them. The piezoelectric coefficient of amorphous polymers increases 4–5 times by poling. 7 After PVDF-TrFE, in comparison with other piezopolymers, PVDF has demonstrated the highest dielectric constant and a high piezoelectric coefficient, as shown in Table 1 . 8 Piezoelectric natural and synthetic polymers Polymer Dielectric constant (1 kHz; 25 °c) Piezoelectric coefficient (pC/N) PLA 3.0–4.0 9.82 Polyhydroxy butyrate PHB 2.0–3.5 1.6–2.0 PVDF 6–12 24–34 Poly(vinylidenefluoride-trifluoroethylene) (PVDF-TrFE) 18 38 Polamide-11 5 4 As mentioned above, PVDF has a piezoelectrically active part that can be in the form of nanostructures, wires, fibers, ribbons, tubes, etc. , 9 and has excellent mechanical strength, flexibility, chemical resistance, ruggedness, high hydrophobicity, biocompatibility, and thermal stability, compared to other commercialized polymeric materials. PVDF is used in energy harvesters, battery separators, sensors, and biomedical applications. 10,11 There are no engineering fields in which piezoelectric materials have not found applications. 12 Engineers are investigating enhancing the polarity of PVDF materials and further improving their mechanical–electrical conversion efficiency to increase the piezoelectric coefficient. 1 However, improving the polarization characteristics of flexible piezoelectric materials has been challenging. In this respect, it has faced unsolved problems concerning lead perovskite based piezoelectrics, and a lot of attention has been paid to obtaining a self-powered generator and supercapacitor by the development of flexible lead-free perovskite devices. 13 Bhavya et al. 14 published a novel strategy to develop a self-charging supercapacitor using lead-free perovskite as a piezoelectric and polyvinyl alcohol–potassium (PVA–KOH) film. The fabricated supercapacitor can quickly self-charge to 200 mV by simple thumb impact, 155 mV by bending it to 150°, 160 mV by twisting it, and a high value of 266 mV by applying a compression of 20 N. The piezoelectric coefficient of PVDF can be impressively amended by doping with inorganic piezoelectric materials, such as piezoelectric ceramics like PZT. In addition, Ag, BTO, nanoclay, graphene, and other nanomaterials can change the crystal structure of PVDF and increase the β-phase percentage. Researchers have utilized different ways to access the optimizes fraction of β-phase. In addition, some new fabrication methods can also amend the polarity of flexible piezoelectric materials. Techniques undertaken to achieve a self-polarized β-phase in PVDF include spin coating, phase-inversion techniques, and the Langmuir–Schaefer (LS) method, none of which need electrical poling. 9 The trend towards smaller flexible electromechanical devices based on PVDF is predicted for the future, such as using the help of piezoelectric properties of biocrystals, to improve biocompatibility for the development of bionic devices. 1"
} | 1,846 |
20576425 | null | s2 | 2,925 | {
"abstract": "The ability to recognize and react to specific environmental cues allows bacteria to localize to environments favorable to their survival and growth. Synthetic biologists have begun to exploit the chemosensory pathways that control cell motility to reprogram how bacteria move in response to novel signals. Reprograming is often accomplished by designing novel protein or RNA parts that respond to specific small molecules not normally recognized by the natural chemosensory pathways. Additionally, cell motility and localization can be coupled to bacterial quorum sensing, potentially allowing consortia of cells to perform complex tasks."
} | 159 |
40011535 | PMC11865525 | pmc | 2,927 | {
"abstract": "Ectomycorrhizal fungi (EMF) play pivotal roles in determining temperate forest ecosystem processes. We tracked root EMF community succession across saplings, juveniles, and adults of three temperate broadleaf trees ( Acer mono , Betula platyphylla , and Quercus mongolica ) in Northeast China. Adult stages showed higher alpha diversity but lower community dissimilarity compared to earlier stages. In particular, the EMF alpha diversity of Quercus mongolica marginally increased along with host developmental stages and ranked as sapling < juvenile < adult. Unlike those of Acer mono and Quercus mongolica , the EMF community composition of Betula platyphylla showed greater variation between the sapling and juvenile stages than between the sapling and adult stages. Cooccurrence networks revealed increasing interconnectivity with host maturity, dominated by positive correlations (> 99%). LEfSe was employed to identify stage- and/or host-specific EMF indicators. This study highlighted the assembly of EMF community during the development of broadleaf trees in temperate forests, thereby advancing understanding of the succession and coevolution of symbiotic relationships. Supplementary Information The online version contains supplementary material available at 10.1038/s41598-025-91411-3.",
"conclusion": "Conclusion In summary, we revealed root EMF dynamics and succession across the various development stages of broadleaf trees in Changbai Mountain. The transition from early-stage heterogeneity to late-stage homogeneity illustrates adaptive strategies of symbiontic relationships as host trees mature. This suggests a natural selection process driving the convergence of symbiotic networks, which is crucial for the survival and success of heterospecifics in temperate forests. This adaptive mechanism helps select specific EMF partners as keystone taxa to the hosts at each stage. The EMF–host relationship is not strictly species-specific but exhibits overlap as hosts mature. This highlights the succession of EMF and contributes to understanding how these interactions evolve within forest ecosystems. These insights lay a framework for further investigations on the mechanisms underlying EMF dynamics and implications for the management and protection of forests.",
"introduction": "Introduction Ectomycorrhizal fungi (EMF) form mutualistic relationships with plant roots, where their extramatrical mycelia extend many centimeters into soil, facilitating root nutrient (e.g., nitrogen) uptake and, in turn, receiving photosynthesized carbon from host plant 1 , 2 . This symbiosis is crucial for the survival, health, and growth of temperate forests 3 , 4 . For example, EMF hyphae enable plants to access a larger soil bank 5 , secrete hydrolytic exoenzymes to mobilize nutrients from soil mineral–organic matter 6 , and trigger plant defense mechanisms against pathogens and abiotic stressors 7 . Broadleaf trees in temperate forests, particularly those of the Betulaceae, Fagaceae and Sapindaceae families, rely heavily on EMF symbiosis for establishment and growth 8 . Investigating the assembly of the root EMF communities during host development is essential for understanding the survival strategies of these obligately ectomycorrhizal trees in temperate ecosystems 8 . Previous studies document that EMF diversity increases as hosts mature. For instance, EMF richness of Quercus liaotungensis notably increases across sapling (1–3 years), juvenile (20–30 years) and mature stage (50–70 years) 9 . Similarly, the Shannon index of Fagus sylvatica EMF rises as hosts mature from early (~ 20 years) to later stages (~ 100 years) 10 . These shifts are attributed to the expansion of well-established root systems in adult trees, which allows for greater soil exploration and facilitates the recruitment of new fungal species via ectomycorrhizal networks (EMNs) shared by neighboring trees 8 . Albeit typically increasing with host maturity, EMF richness and evenness may decrease at very late host stages due to the competitive pressures of conspecific neighbors, which can elevate mortality rates via species-specific pathogens 11 – 14 . In contrast, a study on Betula papyrifera suggests minimal changes in EMF diversity between the host stages of 5 and 65 years, probably because paper birch regenerates through stump sprouts and thereby maintains the host-specific EMF 15 . In addition, Corylus avellana exhibits little variation in EMF community composition between the ages of 50–100 years and 100–200 years, with younger (< 50 years) stages differing more significantly 16 . Furthermore, the EMF community compositions of young (20–30 years) and mature (50–70 years) Quercus liaotungensis are more similar to each other than to saplings (1–3 years) 9 . These suggest that adult host trees serve as major key repositories of forest biomass and pivotal hubs in EMNs, facilitating local coexistence of EMF 17 , 18 . Despite extensive research on the EMF of broadleaf trees, few studies have examined the effects of host development across multiple coexisting tree species 19 – 22 . The zonal top broadleaf Korean pine mixed forests in Changbai Mountain are renowned for their high biodiversity and complex stand structure 23 . For instance, the second poplar–birch forests, which developed following disturbances in the original broadleaf Korean pine forests, mainly comprise of Betula platyphylla and Populus davidiana . These forests serve as vital repositories of various EMF symbionts and are relatively short-lived, typically declining after 50–60 years 24 . This study aims to investigate the succession patterns of root EMF communities across the sapling, juvenile, and adult stages of three typical broadleaf trees— Acer mono (Sapindaceae), Betula platyphylla (Betulaceae) and Quercus mongolica (Fagaceae). By utilizing Illumina MiSeq sequencing, we examine root EMF communities and identify specific EMF taxa associated with each host species and developmental stage. The correlations among the root EMF ASVs across different developmental stages were statistically analyzed.",
"discussion": "Discussion To explore the dynamics of EMF communities in coexistent broadleaf trees at a regional scale, we analyzed 77 root tips from a chronosequence encompassing sapling, juvenile, and adult stages in temperate forests of Northeast China. EMF alpha diversity was significantly higher in adult trees than their younger counterparts. Furthermore, EMF community composition diverged to a greater extent during early host development, with mature trees hosting more converged EMF community composition. Cooccurrence network analysis confirmed that, as trees mature, EMF taxa became increasing integrated, displaying higher centrality, connectivity, and modularity. This suggests that mature trees foster more robust mutualistic networks, enhancing symbiotic interactions. Mature broadleaf trees tended to host EMF communities with lower alpha diversity but higher structural dissimilarity compared to saplings and juveniles, consistent with previous studies 15 , 19 . This can be attributed to the limited contact with EMF spores in undeveloped root systems of early-stage hosts 25 . These young pioneers also face higher mortality rates due to environmental stressors such as nutrient depletion, moisture stress, herbivore damage, and wind disturbance 26 . However, mature trees benefit from fully extended root systems that connect them to a broader range of EMF species, leading to more stable nutrient and water acquisition and increased EMF diversity 27 – 29 . These long-term obligate symbioses may streamline EMF partners and optimize nutrient-use-efficiency, facilitating adaptations to similar biotic and abiotic stressors over the host’s life span 30 . Across premature host stages, the stochastic EMF recruitment 25 , compounded by environmental stressors, disrupt the continuity of the mycorrhizal networks, increasing the divergence of the EMF communities 26 . While we did not directly examine whether the EMF community of younger trees is nested within that of mature individuals, this remains a plausible hypothesis for future investigation. Network analysis demonstrated increased EMF integration in mature hosts, with higher modularity and centrality suggesting stabilized and interconnected mutualistic networks. This pattern aligns with prior findings on the stability of adult-hosted EMF communities 29 , 31 , 32 as long-lived trees selectively recruit EMF to synchronize growth and facilitate community convergence. Even though stage-specific EMF did not always directly interact, they were consistently linked to common EMF species, especially in adult hosts. This suggests that the evolution of stage-specific EMF communities heavily rely on common species that mediate the development of mutualistic networks during host maturation 33 . We found that positive associations among EMF species predominated (over 99%), which likely stems from niche overlap, ecological compatibility, and resource exchange. This promotes plant-soil feedbacks, supports plant maturation, and contributes to forest succession 34 . The minimal negative correlations between stage-specific and common EMF species suggest low competition, facilitating the equilibrium and coexistence of EMF species 35 . Several EMF taxa demonstrated broad host spectra across multiple developmental stages. For instance, Suillus is known as an obligate symbiont of Pinaceae and can significantly alters the morphology and hormone content in host seedlings 36 , and it also benefits the growth of Quercus 37 . The mixed forest composition at the current sites supports the likelihood of multihost colonization, highlighting a trend towards symbiotic promiscuity rather than specificity. Furthermore, the host preference of Suillus underscores the evolutionary constraints of EMF species 38 , as its persistence in mature hosts reflects legacy effects of past host community dynamics 39 . Moreover, the adult-stage indicator Plioderma , found in Acer mono and Quercus mongolica , is adapted to low nitrogen availability and typically associated with late-stage hosts 40 – 42 . This indicates that trees at very late developmental stages acquire resources via specific EMF species, thereby supporting their persistence across different succession stages 43 . EMF species adopt different life strategies at various stages of host development. For example, Pyronemataceae, adapted to nutrient-poor environments 44 , are suitable partners for Acer mono saplings that thrive in highly heterogonous niches 45 . Additionally, Inocybe and Amphinema , preferring nutrient-rich and disturbed environments, aid the establishment of Betula platyphylla in post-disturbance sites 46 . In contrast, Russula prefers mature forests, probably due to challenges with spore dispersal and germination at early developmental stages of hosts 15 , 47 . In addition, we also observed that Acer mono adults were predominated by Phaehelotium and Hebeloma , and Quercus mongolica adults by Cenococcum and Thelephora . All these families are optimized for short-distance exploration 48 , maximizing hydrophilic hyphal extension to efficiently absorb nutrients and enabling energy-conserving late-stage growth."
} | 2,819 |
30715050 | PMC6267296 | pmc | 2,928 | {
"abstract": "Triboelectric nanogenerators (TENG), which utilize contact electrification of two different material surfaces accompanied by electrical induction has been proposed and is considered as a promising energy harvester. Researchers have attempted to form desired structures on TENG surfaces and successfully demonstrated the advantageous effect of surface topography on its electrical output performance. In this study, we first propose the structured Al (SA)-assisted TENG (SA-TENG), where one of the contact layers of the TENG is composed of a structured metal surface formed by a metal-to-metal (M2M) imprinting process. The fabricated SA-TENG generates more than 200 V of open-circuit voltage and 60 µA of short-circuit current through a simple finger tapping motion. Given that the utilization of the M2M imprinting process allows for the rapid, versatile and easily accessible structuring of various metal surfaces, which can be directly used as a contact layer of the TENG to substantially enhance its electrical output performance, the present study may considerably broaden the applicability of the TENG in terms of its fabrication standpoint.",
"introduction": "1. Introduction The development of portable electronic devices has received considerable attention due to their unique advantages such as convenience and excellent functionality, which has enhanced the lives of people in various aspects. These devices require power without using wires from an external power source. As a result, energy-harvesting technologies, which convert the available sustainable energies into electricity has also attracted attention worldwide [ 1 ]. The eminent concept of the triboelectric nanogenerator (TENG), which utilizes contact electrification of two different material surfaces accompanied by electrical induction, has been proposed [ 2 ]. Since its first proposal in 2012, TENG has been actively studied, hence, it is considered to be a promising energy harvesting technology to operate portable electronic devices without spatio-temporal limitations on the basis of its advantages, such as material selection diversity, high efficiency and high shape adaptability [ 3 , 4 , 5 , 6 , 7 , 8 ]. The most actively conducted research topic in the field of the TENG is the enhancement of its electrical output performance, which is similar to those of other energy-harvesting technologies [ 9 ]. Given that the fundamental operating mechanism of the TENG is based on the contact and separation between two surfaces, the simplest and most widely used strategy to enhance the electrical output performance is the introduction of micro- and nanoscale structures onto the surfaces where friction occurs [ 3 , 5 , 10 , 11 , 12 ]. The formation of the surface structures significantly increases local contact pressure resulting in further increases in the contact area between two contact surfaces, thereby generating a high amount of electrical charge on the contact surface [ 13 ]. Given that the most widely utilized material is based on polymer, many researchers have attempted to form desired structures onto polymer surfaces through various subtractive and additive fabrication methodologies. These researchers have successfully demonstrated the benefits of surface topography on the electrical output performance of the TENG. Although the electrical output performance of the TENG has drastically increased due to these efforts, the introduction of structures on the polymer surface accelerates the mechanical wear of the surface of the friction due to the increased local contact pressure. Considering that the operation of a TENG is based on the friction between two surfaces, one of the major issues to be resolved is the problem of mechanical wear [ 5 , 14 , 15 ]. However, the introduction of structures onto the polymer has relatively weak mechanical characteristics compared with other engineering materials, which is definitely unfavorable in terms of mechanical wear. A potential strategy to reduce mechanical wear in the TENG can be the formation of the structures on metal surfaces instead of polymer surfaces. Metal is a widely utilized engineering material with better mechanical characteristics, thereby resulting in higher abrasion resistance than those of the polymer. Metal is also widely utilized in the TENG causing friction with the polymer surface. The common approach to the fabrication of small-scale structures on metal is first to fabricate a prepatterned polymer and use it as a sacrificial layer. Then, subsequent metal deposition on its surface followed by lift-off or an etching process, which are considered to be complex and labor intensive multistep processes due to their requirement for difficult processing conditions [ 16 , 17 ]. Consequently, most TENGs, which utilize a polymer-metal friction to generate electrical output performance, only have a structure on the polymer surface because it takes considerable effort to form structures on the metal surface than on the polymer surface [ 3 ]. Hence, the proposal of facile and highly accessible strategies to form structures onto the metal surface will open another horizon to enhance the electrical output performance of the TENG. In this study, we propose a structured Al (SA)-assisted TENG (SA-TENG), where one of the contact surfaces of the TENG is composed of the structured metal surface formed by a metal-to-metal (M2M) imprinting process. The imprinting process is a low-cost and rapid process that involves transcribing structures onto the substrate of interest by using a stamp. The present study utilizes a precise femtosecond (fs) laser to fabricate steel stamps, which have optically induced micro- and nanoscale patterns and these patterns are transcribed on Al substrates by applying heat and pressure. The M2M imprinting process is optimized by applying a systematic approach. As a result, conical microstructures and line nanostructures are successfully transcribed onto the Al substrates. The structures on the Al substrates considerably enhance the electrical output performance of the present SA-TENG. Given that the utilization of the M2M imprinting process allows the rapid, versatile and easily accessible structuring of the metal substrate, which can be directly used as a contact layer on the TENG to significantly enhance its electrical output performance, the present study may significantly broaden the applicability of the TENG in terms of its fabrication standpoint.",
"discussion": "4. Discussion In this study, the steel stamps, which have nanometer or micrometer scale surface structures are successfully fabricated by using an fs laser. The fabricated steel stamps were then directly utilized in the M2M imprinting process to pattern the Al substrate, which can be directly utilized as one of the contact layers of the SA-TENG. The transcribed structures on the Al substrate were shown to play a role in enhancing the electrical output performance of the TENG through corresponding increased local pressure. By using the fabricated SA-TENG, the 200 V of V OC and 60 µA of I SC were generated through single finger press motion. Given that the utilization of the M2M imprinting process enables rapid, versatile and easily accessible structuring of the metal substrate, which can be directly used as a contact layer on the TENG to significantly enhance its electrical output performance, the present study might significantly contribute to broadening the applicability of the TENG from a fabrication standpoint."
} | 1,872 |
31333404 | PMC6621912 | pmc | 2,929 | {
"abstract": "Real-world applications such as first-person video activity recognition require intelligent edge devices. However, size, weight, and power constraints of the embedded platforms cannot support resource intensive state-of-the-art algorithms. Machine learning lite algorithms, such as reservoir computing, with shallow 3-layer networks are computationally frugal as only the output layer is trained. By reducing network depth and plasticity, reservoir computing minimizes computational power and complexity, making the algorithms optimal for edge devices. However, as a trade-off for their frugal nature, reservoir computing sacrifices computational power compared to state-of-the-art methods. A good compromise between reservoir computing and fully supervised networks are the proposed deep-LSM networks. The deep-LSM is a deep spiking neural network which captures dynamic information over multiple time-scales with a combination of randomly connected layers and unsupervised layers. The deep-LSM processes the captured dynamic information through an attention modulated readout layer to perform classification. We demonstrate that the deep-LSM achieves an average of 84.78% accuracy on the DogCentric video activity recognition task, beating state-of-the-art. The deep-LSM also shows up to 91.13% memory savings and up to 91.55% reduction in synaptic operations when compared to similar recurrent neural network models. Based on these results we claim that the deep-LSM is capable of overcoming limitations of traditional reservoir computing, while maintaining the low computational cost associated with reservoir computing.",
"conclusion": "5. Conclusions We proposed a new approach for performing spatio-temporal tasks on a budget. The proposed deep-LSM has promising results in video activity recognition achieving 84.78% on a representative dataset and surpasses state-of-the-art algorithms in accuracy. More importantly, the deep-LSM consumes significantly lower synaptic memory storage and computational resources. Edge devices naturally benefit from this computationally light algorithm and the following benefits ensue. Edge intelligence framework: Suitable for real-time on-device learning and inference. Local unsupervised plasticity mechanisms: Enable fine-grained tuning to trade-off compute complexity vs. accuracy. Broaden applicability of RC approaches to complex temporal problems that require integration of information over multiple time-scales. An overall reduction in energy consumption and memory requirements compared to current recurrent networks.",
"introduction": "1. Introduction Enabling intelligence on the edge minimizes the round trip delay in decision-making, lowers communication costs, load-balances for the end user, and enhances security with caching or local algorithms to pre-process the data. An emerging input source for edge devices is streaming visual data from first person cameras, such as in smart vehicles, or wearable devices. Being able to accurately process streaming video is crucial for edge devices to understand and react to their environment in a wide range of applications ( eg: path planning, action selection, or surveillance ). A popular application for demonstrating understanding of first-person video data in machine learning and computer vision is video activity recognition. However, majority of state-of-the-art methods for video activity recognition do not target low-end embedded platforms. Complex networks are not amenable for on-device intelligence due to their compute and memory intensive operations (networks with 10–60 million synapses require 0.32–2 GB to store synaptic weights Alom et al., 2018 ) and long training times (in the order of hours to days with GPUs Fu and Carter, 2016 ). In the early 2000s, a computationally light algorithm known as reservoir computing (RC) was proposed by two research groups independently. The two algorithms are otherwise known as the Echo State Network (ESN) (Jaeger, 2001 ) and the Liquid State Machine (LSM) (Maass et al., 2002 ). The main difference between the two is that the LSM is a biologically inspired spiking neural network (SNN), whereas the ESN is a rate-based approximation. In this work we focus on the LSM, a neurally inspired algorithm, with innate characteristics for edge devices that bring in size, weight, and power constraints. In particular SNNs can store the neuronal activation's in a single bit (all or nothing signal), can consume as low as ≈20 pJ per spike (Neftci et al., 2017 ), and shown to be computationally at least as powerful as sigmoid and threshold neurons (Maass, 1997 ). The LSM is a three-layer neural network which consists of an input layer, a liquid layer, and a readout layer. The recurrent connections in the liquid layer allow it to capture dynamic information, where information fades out over time. The advantage of the LSM is that all the synaptic connections, except for those which connect to the readout layer, are randomly initialized and remain fixed. Unique inputs will produce distinct perturbations in the state of the high-dimensional liquid layer from which information can be extracted. By using fixed connections, the LSM can circumvent the need for expensive learning rules and the problem of vanishing gradients which can impede learning with gradient descent approaches in recurrent neural networks. In Soures et al. ( 2017 ), it was shown that these networks are robust to internal noise, making them a natural choice for embedded systems, particularly analog implementations which are prone to device noise. However, the conventional LSM model has shown limited applicability in complex real-world problems owing to the single dynamical layer driven by an input signal (Hermans and Schrauwen, 2013 ; Ma et al., 2017 ). The single layer constricts the temporal dynamics of the LSM resulting in very large reservoir networks to solve trivial tasks. Another drawback with LSM is its dependence on the initialization of random synaptic connections. Recent literature highlights the gaps in conventional LSM, RC networks in general, and the need to extend the capabilities of these networks (Jaeger, 2007 ; Triefenbach et al., 2010 , 2013 ; Gallicchio and Micheli, 2016 ; Wang and Li, 2016 ; Ma et al., 2017 ; Bellec et al., 2018 ). Motivated by these observations, we propose a novel framework that drastically reduces the overall computational resources without sacrificing the overall performance in complex spatiotemporal task. Specific contributions of this work are Deep-LSM, a semi-trained deep spiking recurrent neural network with LSM as a core building block, capable of capturing information over multiple time-scales. Demonstrate that a modular/deep architecture significantly reduces the memory requirements for storing synaptic weights. Use local, unsupervised plasticity mechanisms to partially train the network yields state-of-the-art performance while minimizing the cost of training. Design an attention modulated readout layer to selectively process information in the deep-LSM with limited computational resources. Analyze the model performance on first-person video activity recognition with DogCentric dataset (Iwashita et al., 2014 ) and demonstrate state-of-the-art performance. Observe ≈ 90% memory savings and reduction in number of operations compared to a LSTM and ≈ 25% reduction of memory consumption in comparison to a standard LSM and 16% decrease in number of operations."
} | 1,864 |
31387595 | PMC6683508 | pmc | 2,931 | {
"abstract": "Background Acetoin (AC) and 2,3-butanediol (2,3-BD) as highly promising bio-based platform chemicals have received more attentions due to their wide range of applications. However, the non-efficient substrate conversion and mutually transition between AC and 2,3-BD in their natural producing strains not only led to a low selectivity but also increase the difficulty of downstream purification. Therefore, synthetic engineering of more suitable strains should be a reliable strategy to selectively produce AC and 2,3-BD, respectively. Results In this study, the respective AC ( alsS and alsD ) and 2,3-BD biosynthesis pathway genes ( alsS , alsD , and bdhA ) derived from Bacillus subtilis 168 were successfully expressed in non-natural AC and 2,3-BD producing Corynebacterium crenatum , and generated recombinant strains, C. crenatum SD and C. crenatum SDA, were proved to produce 9.86 g L −1 of AC and 17.08 g L −1 of 2,3-BD, respectively. To further increase AC and 2,3-BD selectivity, the AC reducing gene ( butA ) and lactic acid dehydrogenase gene ( ldh ) in C. crenatum were then deleted. Finally, C. crenatum Δ butA Δ ldh SD produced 76.93 g L −1 AC in one-step biocatalysis with the yield of 0.67 mol mol −1 . Meanwhile, after eliminating the lactic acid production and enhancing 2,3-butanediol dehydrogenase activity, C. crenatum Δ ldh SDA synthesized 88.83 g L −1 of 2,3-BD with the yield of 0.80 mol mol −1 . Conclusions The synthetically engineered C. crenatum Δ butA Δ ldh SD and C. crenatum Δ ldh SDA in this study were proved as an efficient microbial cell factory for selective AC and 2,3-BD production. Based on the insights from this study, further synthetic engineering of C. crenatum for AC and 2,3-BD production is suggested. Electronic supplementary material The online version of this article (10.1186/s12934-019-1183-0) contains supplementary material, which is available to authorized users.",
"conclusion": "Conclusion In this study, we successfully engineered C. crenatum SYPA5-5 to overexpress alsS , alsD and/or bdhA from B. subtilis 168 for selective AC and 2,3-BD biosynthesis from glucose. After depressing competition pathways, recombinant C. crenatum Δ butA Δ ldh SD and C. crenatum Δ ldh SDA further increased AC and 2,3-BD production to 76.93 g L −1 and 88.83 g L −1 , respectively. Overall, respective non-natural AC and 2,3-BD biosynthesis pathway constructed in this study efficiently reduced the by-products accumulation and improved AC and 2,3-BD selectivity in C. crenatum . The optimal selection of AC and 2,3-BD biosynthesis strategies with further metabolic engineering should lead to the development of a promising microbial cell factory for AC and 2,3-BD production.",
"discussion": "Results and discussion Overexpression of ALS, ALDC, and AR/BDH in C. crenatum Here, an artificial operon including alsS , alsD , and bdhA from B. subtilis was selected to overexpress in C. crenatum to investigate its effect on AC and 2,3-BD production. The confirmed recombinants were named C. crenatum /pXMJ19- alsS ( C. crenatum S), C. crenatum /pXMJ19- alsD ( C. crenatum D), C. crenatum /pXMJ19- bdhA ( C. crenatum A), C. crenatum /pXMJ19- alsSD ( C. crenatum SD), and C. crenatum /pXMJ19- alsSD - bdhA ( C. crenatum SDA), respectively. The SDS-PAGE was used to confirm the expression of ALS (AHAS), ALDC, and AR/BDH in C. crenatum recombinants (Additional file 1 : Figure S1). As expected, ALS (AHAS), ALDC, and AR/BDH on gel demonstrated their respective molecular mass about 60, 29, and 38 kDa, respectively. Enzyme activity assays also confirmed the functional expression of ALS (AHAS), ALDC, and AR/BDH in C. crenatum transformants. As shown in Table 1 , the wild type strain showed negligible acetohydroxyacid synthase (AHAS) activity, which might be inhibited by the anabolism of l -valine and l -leucine in C. crenatum [ 25 , 26 ]. Meanwhile, no ALDC activity was detected in C. crenatum WT. After enhanced expression of ALS and ALDC, the enzyme activities in C. crenatum SD and C. crenatum SDA was increased to 2.22/2.32 and 8.08/5.68 U mg −1 , respectively. However, it should be noted that the lower increased activities than that of only ALS or ALDC expression probably due to lower efficiency in the tandem co-expression driven by one promoter for two or three genes. On the other hand, compared to C. crenatum WT, C. crenatum SDA showed a high AR and BDH activities (0.32 and 0.10 U mg −1 ) comparable to C. crenatum A (0.38 and 0.19 U mg −1 ). In addition, although AR/BDH was a bidirectional reversible enzyme [ 24 , 27 ], the heterogenous AR/BDH from B. subtilis present a higher AR activity than BDH activity in C. crenatum A and C. crenatum SDA. With ALS, ALDC, and (or) AR/BDH, the recombinant C. crenatum can catalyze the pyruvate to either AC or DA, thus allowing the respective conversion of glucose to AC and 2,3-BD. Table 1 The specific enzyme activities of ALS (AHAS), ALDC, and AR/BDH were determined in recombinant C. crenatum cell extracts Strains Specific enzyme activity (U mg −1 ) ALS (AHAS) ALDC AR/BDH C. crenatum WT 0.01 ± 0.01 0 0.05 ± 0.01/0.01 ± 0.01 C. crenatum S 4.86 ± 0.85 nd nd C. crenatum D nd 13.59 ± 1.22 nd C. crenatum A nd nd 0.38 ± 0.03/0.19 ± 0.01 C. crenatum SD 2.22 ± 0.43 8.08 ± 0.94 nd C. crenatum SDA 2.32 ± 0.31 5.68 ± 0.87 0.32 ± 0.02/0.10 ± 0.01 Using C. crenatum WT as positive control, 1% recombinant strains were transferred to 10 mL LBG medium with 10 g mL −1 chloramphenicol. After 12 h of activation, 1% bacterial solution was transferred to 50 mL LBG liquid medium. After incubation for 3–4 h at 30 °C, 180 r min −1 , the expression of heterologous enzyme in the recombinants was induced by adding 1 mM IPTG and incubated for 12 h. Specific enzymatic activities of ALS (AHAS), ALDC, and AR/BDH in the crude enzyme solution were assayed after disrupting the recombinant cells with sonicator. The specific enzyme activity results are the mean ± standard of three replicates nd not detected \n Production of AC and 2,3-BD by shake flask fermentation of recombinant C. crenatum Batch fermentations with 120 g L −1 glucose as the sole carbon source were first performed in shack flask to investigate the metabolic impact of ALS, ALDC, and BDH expression. Since C. crenatum is a facultative anaerobe, cell growth under different dissolved oxygen condition often leads to generate different products [ 28 , 29 ]. Previous study has demonstrated that C. crenatum shows a fast growth rate of biomass when using glucose as the substrate, and is diffusely used in the yielding of sundry amino acids such as l -glutamic acid and l -arginine under sufficient oxygen supply conditions [ 28 ]. Therefore, the by-product l -arginine was also determined at the end of fermentation in this study. The results are shown in Table 2 . Table 2 Fermentation kinetics by recombinant C. crenatum in shack flask Strain Time (h) Glucose consumption (g L −1 ) Glucose uptake rate (g L −1 h −1 ) Diacetyl (g L −1 ) Acetoin (g L −1 ) 2,3-Butanediol (g L −1 ) l -Arginine (g L −1 ) C. crenatum WT 120 110 ± 2 0.91 ± 0.02 nd nd nd 23.59 ± 1.7 C. crenatum S 120 36 ± 4 0.30 ± 0.03 2.17 ± 0.13 0.76 ± 0.12 0.45 ± 0.13 7.09 ± 0.70 C. crenatum D 120 78 ± 3 0.65 ± 0.02 nd nd nd 15.66 ± 1.31 C. crenatum A 120 73 ± 4 0.61 ± 0.03 nd nd nd 14.27 ± 1.45 C. crenatum SD 120 70 ± 3 0.58 ± 0.02 nd 13.59 ± 1.25 10.61 ± 1.33 8.74 ± 1.12 C. crenatum SDA 120 70 ± 6 0.58 ± 0.05 nd 9.43 ± 1.50 15.16 ± 1.31 7.83 ± 0.84 Using C. crenatum WT as positive control, 10% recombinant C. crenatum was transferred to the fermentation medium containing 120 g L −1 glucose for shake flask fermentation (120 g L −1 glucose was used as substrate). The titer of DA, AC, 2,3-BD and l -arginine were determined after 120 h of fermentation. The results of the fermentation parameters are the mean ± standard of three biological replicates nd not detected \n As expected, C. crenatum WT yield 23.6 g L −1 \n l -arginine, and no AC and 2,3-BD were detected in the fermentation broth. Theoretically, overexpression of alsS , alsD , and bdhA gene, respectively, cannot produce AC and 2,3-BD in C. crenatum . However, except for little AC and 2,3-BD producing, C. crenatum S also produced 2.17 g L −1 of unexpected DA, which could not be accumulated in B. subtilis [ 30 ]. This might be due to the spontaneous non-enzymatic decarboxylation reaction of α -acetolactate under natural oxidative aerobic conditions [ 31 ]. In addition, our previous work has also proved that the BDH from C. crenatum could catalyze DA to AC and then convert to 2,3-BD [ 24 ]. This explains why the chain reaction was got through in C. crenatum S without expression of ALDC. Moreover, it has been reported that DA is a bacteriostatic food additive that can inhibit bacteria growth [ 3 ]. Thus, C. crenatum S consumed only about 36 g L −1 glucose and then suspended the fermentation while accumulated comparable amount of DA, leading to a low efficiency of the multiple-step reaction from pyruvate to AC and 2,3-BD. On the contrary, after introducing the alsSD operon with one promoter into C. crenatum , co-expression of ALS and ALDC effectively converted pyruvate to 13.59 g L −1 AC and 10.61 g L −1 2,3-BD, respectively. To enhance 2,3-BD production, bdhA was further expressed in C. crenatum SD and the generated C. crenatum SDA dramatically increased 2,3-BD titer by 43%. These results indicated that ALS and ALDC are both important for AC and 2,3-BD biosynthesis in C. crenatum . In addition, enhanced heterogeneous BDH activity was confirmed the positive effect on 2,3-BD biosynthesis. However, C. crenatum SD and C. crenatum SDA still produced comparable amount of l -arginine as by-product, resulting in a low yield of AC and 2,3-BD. Therefore, l -arginine biosynthesis must be decreased. Bioconversion of glucose to AC and 2,3-BD by recombinants C. crenatum resting cells l -Arginine as growth factor is necessary for C. crenatum growth, which cannot be directly depressed by gene knock-out. l -arginine production by C. crenatum needs high level of dissolved oxygen, while there is no strict demand on dissolved oxygen for the acid-butanediol fermentation. Therefore, we attempted to produce AC and 2,3-BD using resting cell bioconversion with glucose as substrate by C. crenatum SD and C. crenatum SDA, respectively. As demonstrated in Fig. 2 , no l -arginine was detected during the resting cell bioconversion process (data not shown). C. crenatum SD converted 100 g L −1 glucose to 9.86 g L −1 AC and 12.76 g L −1 2,3-BD, respectively. Similarly, C. crenatum SDA further accumulated 2,3-BD to 17.08 g L −1 with a high 2,3-BD/AC ratio. However, the content of organic acids, especially lactic acid, were also detected in both C. crenatum SD and C. crenatum SDA, which was consistent with previous study that C. crenatum can produce various organic acids such as succinic acid, acetic acid and lactic acid in the case of limited dissolved oxygen [ 29 ]. As the accumulation of organic acids increases the consumption of carbon sources [ 32 , 33 ], the synthetic pathways of organic acids may competitively inhibit the accumulation of AC and 2,3-BD. Fig. 2 Resting cell bioconversion analysis of C. crenatum and the recombinant strains in shack flask. Recombinant C. crenatum (100 µL) was transferred to 10 mL LBG medium and incubated for approximately 12 h at 30 °C, 180 r min −1 . The cultured bacterial solution (3 mL) was transferred into 30 mL of fermentation medium at the same culture condition for 24 h. The cultured cells were harvested by centrifugation at 8000 r min −1 for 10 min and then resuspended in resting cell bioconversion medium (containing 100 g L −1 glucose) for resting cell bioconversion. Formed metabolites and organic acid during the bioconversion of resting cells are shown. The results of resting cell bioconversion kinetics are shown as the mean ± standard of three replicates. +: overexpression, Δ: knockout \n Construction of ldh and butA blocked recombinant C. crenatum Although resting cell bioconversion reduced l -arginine formation, some pyruvate was still converted to lactic acid. In addition, unexpected high yield of 2,3-BD was still produced by C. crenatum SD. To further decrease by-products and enhance the AC and 2,3-BD selectivity, blocking the competitive synthesis pathways in C. crenatum was considered to be necessity. The homologous lactate dehydrogenase ( ldh ) and butanol dehydrogenase ( butA ) genes in C. crenatum were knocked out to increase respective AC and 2,3-BD selectivity. The suicide vectors pK18-Δ ldh (Additional file 2 : Figure S2) and pK18-Δ butA (Additional file 3 : Figure S3), respectively, was constructed and transformed into C. crenatum . The mutant strains C. crenatum Δ ldh and C. crenatum Δ butA were finally obtained after two rounds of homologous recombination. Then, pK18-Δ ldh was further introduced into C. crenatum Δ butA and generated C. crenatum Δ butA Δ ldh . All of mutant strains were verified by PCR using the upstream and downstream primers of ldh and butA (Additional file 4 : Figure S4). LDH and AR/BDH activities of C. crenatum mutants were determined, and the results were shown in Table 3 . LDH activity was reduced by 83–88% after blocking of ldh gene. Meanwhile, AR and BDH activities were both reduced by 62–82% and 83–87%, while butA gene was knocked out. The results indicated that the AC reduction and lactic acid biosynthesis pathway were depressed by knocking out butA and ldh genes. By comparing the dry cell weights of C. crenatum WT and the knockout strains (Additional file 5 : Table S1), it was found that the knockout of ldh and butA genes had little effect on the growth of the bacteria. Table 3 Specific enzyme activity assays of LDH and AR/BDH in crude cell extracts of recombinant C. crenatum strains Strains LDH (U mg −1 ) AR/BDH (U mg −1 ) C. crenatum WT 1.04 ± 0.03 0.055 ± 0.003/0.003 ± 0.001 C. crenatum Δ ldh 0.18 ± 0.01 nd C. crenatum Δ butA nd 0.021 ± 0.003/0.0005 ± 0.001 C. crenatum Δ butA Δ ldh 0.13 ± 0.01 0.010 ± 0.002/0.0004 ± 0.001 Using C. crenatum WT as positive control, 1% of recombinant C. crenatum was transferred to 50 mL LBG medium and incubated in shake flask at 180 r min −1 , 30 °C for 12 h. The supernatant of the disrupted cells was taken to LDH and AR/BDH activity assay. The results of the specific enzyme activity assays are the mean ± standard of three replicates nd not detected \n Then, the recombinant C. crenatum Δ butA , C. crenatum Δ ldh , and C. crenatum Δ butA Δ ldh were subjected to resting cell and the bioconversion results were shown in Fig. 2 . As expected, after knocking out ldh gene, the acetic and succinic acid production were slightly increased, while the lactic acid production dramatically decreased by about 85% in C. crenatum Δ ldh and C. crenatum Δ butA Δ ldh . Moreover, compared with C. crenatum WT and C. crenatum Δ ldh , there is no significant effect on biocatalysis after butA delection. Efficient one-step bioconversion of glucose to AC by recombinant C. crenatum Δ butA Δ ldh /pXMJ19- alsSD resting cell To construct the one-step bioconversion of glucose to AC, plasmid pXMJ19- alsSD was then transformed into ldh and butA deletion strain to construct C. crenatum Δ butA Δ ldh /pXMJ19- alsSD ( C. crenatum Δ butA Δ ldh SD). C. crenatum Δ butA Δ ldh SD consumed 94 g L −1 glucose after 60 h with an uptake rate of 1.57 g L −1 h −1 , which was significantly higher (2.18-fold and 2.34-fold, respectively) than that of C. crenatum WT and C. crenatum Δ butA Δ ldh , indicating that the expression of heterologous AC biosynthesis pathway strongly promoted glucose for AC production. Consequently, AC production dramatically increased from 9.86 to 23.56 g L −1 , about 2.39-fold than before. Meanwhile, only a small amount of 2,3-BD (1.28 g L −1 ) and lactic acid (1.83 g L −1 ) were detected, resulting in an AC molar yield of up to 0.51 mol mol −1 . Consistence with that of C. crenatum Δ butA Δ ldh , acetic acid and succinic acid has hardly changed (Fig. 2 ). Efficient one-step bioconversion of glucose to 2,3-BD by recombinant C. crenatum Δ ldh /pXMJ19- alsSD - bdhA resting cell To further extend the products chain for 2,3-BD production, pXMJ19- alsSD - bdhA was introduced into C. crenatum Δ ldh , resulting in C. crenatum Δ ldh /pXMJ19- alsSD - bdhA ( C. crenatum Δ ldh SDA) which co-expresses both homologous and heterologous AR/BDH. Generally, compared with C. crenatum SDA as the positive control, C. crenatum Δ ldh SDA resting cell converted more glucose to 2,3-BD. After 60 h, C. crenatum Δ ldh SDA consumed 95 g L −1 glucose with an uptake rate of 1.58 g L −1 h −1 . With overexpression of AR/BDH and depressing LDH, 2,3-BD production increased from 17.08 to 25.93 g L −1 , about 52% higher than the control. Meanwhile, lactic acid was decreased about 95%, leading to a high 2,3-BD molar yield of 0.55 mol mol −1 (Fig. 2 ). It should be noted that the bioconversion for 2,3-BD production was still accompanied by certain AC accumulation because of the reversible reaction of AR/BDH. Moreover, 2,3-BD and lactic acid are both NADH-dependent products. Therefore, depressing LDH also releases additional reducing equivalent, which further releases the constraint of co-enzyme poll on 2,3-BD synthesis. However, the yields of acetic acid (1.6 g L −1 ) and succinic acid (1.1 g L −1 ) were slightly higher than C. crenatum SDA. Repeated batch resting cell bioconversion for AC and 2,3-BD production in 5 L bioreactor To investigate the long-term stability and performance of recombinant C. crenatum , repeated batch resting cell bioconversion were performed in 5 L bioreactors, and the results are shown in Fig. 3 . Generally, the resting cell bioconversion was repeated for 3 cycles with 100 g L −1 of glucose as the substrate. In the first two batch conversions, AC production was stably increased by C. crenatum Δ butA Δ ldh SD. However, AC productivity of the third batch has gradually declined. Finally, after total 60 h of bioconversion, 76.93 g L −1 of AC with the yield of 0.67 mol mol −1 was produced by C. crenatum Δ butA Δ ldh SD (Fig. 3 a). Meanwhile, after eliminating the lactic acid production and enhancing 2,3-butanediol dehydrogenase activity, C. crenatum Δ ldh SDA also synthesized 88.83 g L −1 of 2,3-BD with the yield of 0.80 mol mol −1 (Fig. 3 b). It should be noted that both average AC and 2,3-BD productivity throughout the repeated batch conversion process reached 1.28–1.48 g L −1 h −1 , which increased ~ 3.5-fold compared to flask batch biocatalysis. However, significant decrease of enzyme activities was observed during the third process (data not shown). Therefore, further strategies should be developed to maintain the cell viability and stability during the whole bioconversion process. Fig. 3 Resting cell bioconversion analysis of recombinant C. crenatum in 5 L bioreactor. Recombinant C. crenatum (200 µL) was transferred to 20 mL LBG medium and incubated for 24 h at 180 r min −1 , 30 °C. After the 10% cultures were transferred and cultured in 200 mL seed medium for 18 h, the whole cultures were transferred to 5 L bioreactor containing 2 L of fermentation medium, and incubated for 24 h at 30 °C, 600 r min −1 . 4 L recombinant C. crenatum cultures were collected and resuspended in 2 L resting cell bioconversion medium containing 100 g L −1 glucose. Resting cell bioconversion was performed in batches at 30 °C, 250 r min −1 . Consumed glucose, formed metabolites, and organic acid during the bioconversion of resting cells are shown. The results of resting cell bioconversion kinetics are shown as the mean ± standard of three replicates \n Comparison with other study Currently, microbial fermentation is still the main method for producing AC and 2,3-BD from various substrates. However, its relatively long fermentation period and low productivity and substrate conversion rate are currently obstructed the industrial application [ 18 , 34 , 35 ]. Biocatalysis, as a highly efficient and environmentally friendly methods, can significantly improve substrate utilization rate and product yield. As shown in Table 4 , previous studies have demonstrated that engineered E. coli and B. subtilis can be used as host for biocatalytic synthesis of AC and 2,3-BD, in which the highest yield of AC and 2,3-BD can be reached to 0.98 mol mol −1 and 0.96 mol mol −1 . However, most of these biocatalysts cases were carried out using AC or 2,3-BD as a substrate, which are not suitable for industrial production at all. Although 2,3-BD and AC bioconversion from glucose were achieved by K. pneumoniae and B. subtilis , two step batch strategy still cannot selectively separate mixed (2 S ,3 S )-2,3-BD and (3 S )-AC [ 41 ]. In this study, we developed a new mono-bioconversion system using C. crenatum as the only host for AC and 2,3-BD production directly from glucose. After depressing competition pathways and overexpressing AC and 2,3-BD biosynthesis genes from B. subtilis , one-step biosynthesis of AC and 2,3-BD, respectively, were achieved in C. crenatum Δ butA Δ ldh SD and C. crenatum Δ ldh SDA without additional complex nutrients. Table 4 Comparison of titer and yield of AC and 2,3-BD produced by different strains through bioconversion in recent years Strains Bioconversion method Substrates Products Titer (g L −1 ) Yield (mol mol −1 ) References \n E. coli \n Fed batch DA AC 39.4 0.83 [ 36 ] E. coli BL21 (DE3) Batch meso -2,3-BD (3 R )-AC 86.7 0.94 [ 37 ] E. coli BL21 (DE3) Batch meso -2,3-BD (3 S )-AC 36.7 0.88 [ 21 ] E. coli BL21 (DE3) Batch (2 R ,3 R )-2,3-BD (3 R )-AC 41.8 0.98 [ 21 ] B. subtilis 168 Fed batch 2,3-BD AC 91.8 0.78 [ 22 ] \n E. coli \n Batch meso -2,3-BD (3 S )-AC 72.38 0.92 [ 38 ] \n E. coli \n Batch meso -2,3-BD (2 S ,3 S )-2,3-BD 38.41 0.96 [ 38 ] E. coli BL21 (DE3) Fed batch DA (2 S ,3 S )-2,3-BD 31.7 0.86 [ 39 ] B. subtilis 168 Repeated batches AC + formate 2,3-BD 115.4 0.96 [ 40 ] B. subtilis 168 Batch AC + glucose 2,3-BD 63.7 0.96 [ 40 ] K. pneumoniae and B. subtilis 168 Two-step sequencing batch Glucose (3 S )-AC 56.7 0.85 [ 41 ] C. crenatum Δ butA Δ ldh SD Repeated batches Glucose AC 76.9 0.67 This study C. crenatum Δ ldh SDA Repeated batches Glucose 2,3-BD 88.8 0.80 This study \n Compared to usual 100–120 g L −1 AC and 2,3-BD produced by microbial fermentation, further improvements of C. crenatum are necessary for industrial application. Although depressing ldh and butA and enhancing AC or 2,3-BD biosynthesis pathway activity, C. crenatum Δ butA Δ ldh SD and C. crenatum Δ ldh SDA showed a higher AC and 2,3-BD selectivity, respectively, some pyruvate still converted to acetic and succinic acid. Therefore, increased AC and 2,3-BD production can be realized by disruption of the acetate and succinate biosynthesis pathway, as demonstrated in K. oxytoca and B. subtilis [ 11 , 12 ]. Moreover, modification of key enzymes is critical for promoting cell metabolism. Previously, we have relieved the feedback inhibition of l -arginine by site-directed mutation of the key enzyme (NAGK) of C. creantum , and overexpressed the l -arginine operon, which effectively increased the yield of l -arginine by 41.7% [ 42 ] and 29% [ 43 ]. Therefore, the site-specific mutagenesis of butanol dehydrogenase ( butA ) gene might be further improved the AC and 2,3-BD yield and selectivity in C. creantum . In addition, although the repeated batch biocatalysis showed a high average productivity for AC and 2,3-BD production, the catalytic efficiency significantly deceased after only two batches, which can be further improved by cell immobilization to increase the cell viability and stability [ 44 , 45 ]. On the other hand, process engineering, including buffer optimization, substrate concentration optimization, multiple biocatalysis strategies, and high cell density et al., can further improve AC and 2,3-BD production for commercial development. These synthetic and process engineering strategies can be applied to together to develop an efficient microbial cell factory for selective AC and 2,3-BD production in C. creantum ."
} | 6,117 |
24907284 | PMC4158751 | pmc | 2,932 | {
"abstract": "Ruminant livestock represent the single largest anthropogenic source of the potent greenhouse gas methane, which is generated by methanogenic archaea residing in ruminant digestive tracts. While differences between individual animals of the same breed in the amount of methane produced have been observed, the basis for this variation remains to be elucidated. To explore the mechanistic basis of this methane production, we measured methane yields from 22 sheep, which revealed that methane yields are a reproducible, quantitative trait. Deep metagenomic and metatranscriptomic sequencing demonstrated a similar abundance of methanogens and methanogenesis pathway genes in high and low methane emitters. However, transcription of methanogenesis pathway genes was substantially increased in sheep with high methane yields. These results identify a discrete set of rumen methanogens whose methanogenesis pathway transcription profiles correlate with methane yields and provide new targets for CH 4 mitigation at the levels of microbiota composition and transcriptional regulation.",
"discussion": "Discussion CH 4 emitted from sheep is formed by methanogenic archaea in the rumen as an end product of microbial degradation of forage material. It is therefore likely that the ruminal microbiome contributes to the host CH 4 yield phenotype. The exact mechanism causing the high and low CH 4 yield phenotypes observed in sheep is still unclear, and our understanding of the microbial contribution to differences in CH 4 yields among sheep has been limited by the low throughput of previous cellular and molecular manipulations ( Warnecke et al. 2007 ; Pope et al. 2010 ; Hess et al. 2011 ). Also, previous studies have suggested that microbial-derived phenotypes, including CH 4 production levels, are primarily determined by microbial abundance profiles ( Kao-Kniffin et al. 2011 ; Fox 2012 ). In contrast, the deep metagenomic and metatranscriptomic sequencing of rumen content in the present study revealed that increases in CH 4 output are primarily associated with increases in the expression of methanogenesis pathway genes. A possible mechanism explaining CH 4 yield differences between animals is based on the amount of time that feed particles are retained in the rumen ( Benchaar et al. 2001 ), with longer particle retention times leading to higher CH 4 yields. Particle retention time in ruminants is known to be a heritable trait ( Orskov et al. 1988 ; Smuts et al. 1995 ) and may explain at least some of the CH 4 yield variation observed in sheep ( Pinares-Patiño et al. 2003 ). Recently, CH 4 yield in sheep in Australia has been directly correlated with the retention time of feed particles and liquid and with the total amount of feed particles and rumen volume ( Goopy et al. 2013 ), further supporting this view. Differential particle retention time may explain our findings of altered expression of methanogenesis pathway genes in sheep via a substrate-mediated effect. Differences in the passage rate of particles through the rumen is predicted to affect ruminal H 2 levels according to a model based on microbial growth kinetics and fermentation thermodynamics ( Janssen 2010 ). In this model, an increased particle passage rate is associated with higher rumen H 2 concentrations, a thermodynamic negative feedback of H 2 that results in less H 2 formation by the fermentative microbes and, hence, less CH 4 formation. Conversely, slower particle passage results in lower H 2 concentrations, enhanced H 2 formation during fermentation, and more CH 4 . These hypotheses are consistent with the finding that low CH 4 yield sheep have fewer H 2 -producing bacteria and high CH 4 yield sheep have more H 2 -producing bacteria in their rumens (Kittelmann et al., unpubl.). Under ruminal conditions of slower particle passage rate and lower H 2 concentrations, there will be a higher turnover rate of a smaller H 2 pool through the methanogenesis pathway to account for the elevated CH 4 formed. The lower ruminal H 2 concentration means that methanogens have to increase expression of methanogenesis genes to maintain the H 2 turnover rate. This is because enzyme concentrations as well as substrate concentrations can limit the flux through a pathway, and increasing enzyme expression partially overcomes the limitation of lower substrate concentrations ( Morgan et al. 1997 ; Enoki et al. 2011 ; Walker et al. 2012 ; Browne and Cadillo-Quiroz 2013 ) Conversely, a high particle passage rate and high H 2 conditions would require a lower level of expression of methanogenesis pathway genes to permit the same flux. While there have been few studies on characterizing rumen microbial populations associated with natural variation in ruminant CH 4 yields ( Kittelmann et al. 2013 ), there have been numerous investigations on feedlot cattle selected for efficiency of feed conversion (also known as residual feed intake [RFI]), for which some CH 4 yields' data are also available. Low-RFI animals are considered to be feed efficient and have lower CH 4 yields compared with high-RFI, or feed-inefficient, animals ( Nkrumah et al. 2006 ; Hegarty et al. 2007 ). Comparisons of ruminal microbiomes between low- and high-RFI animals using a variety of methods have shown differences in bacterial and archaeal community profiles correlated with RFI, although these associations are often influenced by the energy content of the diet ( Guan et al. 2008 ; Zhou et al. 2010 ; Carberry et al. 2012 ; Hernandez-Sanabria et al. 2012 ). Methanogen-related differences observed in these studies included a specific high-RFI–related PCR-DGGE band associated with Methanobrevibacter smithii PS ( Zhou et al. 2010 ), an elevated abundance of Methanosphaera stadtmanae , and Methanobrevibacter sp. strain AbM4-like sequences in high-RFI animals ( Zhou et al. 2009 ), and a higher abundance of M. smithii genotypes in high-RFI animals ( Carberry et al. 2014 ). Where measured, total methanogen densities in the rumen contents did not differ between the feed efficiency groups, indicating that the composition of the methanogenic community was the important difference. These observations are generally consistent with our findings of no changes in total methanogen numbers and an increase in the relative abundance of the M. gottschalkii group within the Methanobacteria. However, the elevated levels of Methanosphaera spp. in high-RFI cattle relative to low-RFI animals ( Zhou et al. 2009 ) differ from our observation of elevated Methanosphaera in the low-CH 4 -yielding sheep. This may be due to the large difference between the diets fed (high grain feedlot diet for cattle [ Zhou et al. 2009 ] vs. pelleted lucerne diet for sheep [this study]) or to innate differences between ruminant species (cattle vs. sheep). The main findings of this study indicate that there are strong correlations between the expression levels of the hydrogenotrophic methanogenesis pathways in rumen methanogens and CH 4 yields in sheep, in the absence of significant changes in methanogen community structure or relative abundance. This indicates a response of methanogenesis functions of the resident methanogens to the supply of their main substrate, H 2 . We predict that these gene expression changes are indirectly controlled by particle retention time or digesta passage rate in sheep. This is an avenue for future investigation within New Zealand’s sheep CH 4 screening program, with the long-term goal of selecting animals with lower CH 4 yields without compromising their productivity or reproductive ability. Furthermore, the identification of specific groups of methanogens that encode up-regulated methanogenesis genes correlated with high CH 4 yield in sheep confirms current gene targets under investigation and provides new microbial and pathway targets for CH 4 mitigation technologies in ruminants."
} | 1,983 |
35455118 | PMC9031894 | pmc | 2,933 | {
"abstract": "The spiking neural network (SNN) is regarded as a promising candidate to deal with the great challenges presented by current machine learning techniques, including the high energy consumption induced by deep neural networks. However, there is still a great gap between SNNs and the online meta-learning performance of artificial neural networks. Importantly, existing spike-based online meta-learning models do not target the robust learning based on spatio-temporal dynamics and superior machine learning theory. In this invited article, we propose a novel spike-based framework with minimum error entropy, called MeMEE, using the entropy theory to establish the gradient-based online meta-learning scheme in a recurrent SNN architecture. We examine the performance based on various types of tasks, including autonomous navigation and the working memory test. The experimental results show that the proposed MeMEE model can effectively improve the accuracy and the robustness of the spike-based meta-learning performance. More importantly, the proposed MeMEE model emphasizes the application of the modern information theoretic learning approach on the state-of-the-art spike-based learning algorithms. Therefore, in this invited paper, we provide new perspectives for further integration of advanced information theory in machine learning to improve the learning performance of SNNs, which could be of great merit to applied developments with spike-based neuromorphic systems.",
"conclusion": "5. Conclusions In this invited paper, we first presented an ITL-based scheme for robust spike-based continual meta-learning, which is improved by the RMEE criterion. A gradient descent learning principle is presented in a recurrent SNN architecture. Several tasks are realized to demonstrate the learning performance of the proposed MeMEE model, including autonomous navigation, robust working memory in the store–recall task and robust meta-learning capability for the sMNIST data set. In the first autonomous navigation task, the SNN model learns to find the correct destination by continual meta-learning from the task reward and punishment. This demonstrates that the MeMEE model based on the proposed RMEE criterion realizes the meta-learning capability for navigation and outperforms the conventional RSNN model. In the second task, the proposed MeMEE model improves the working memory performance by recalling the stored noisy patterns. In the third task, the proposed MeMEE model with RMEE criterion can enhance the robustness in the meta-learning task for noisy sMNIST images. This invited paper provides a novel insight into the improvement of the spike-based machine learning performance based on information theoretic learning strategy, which is critical for the further research of artificial general intelligence. In addition, it can be implemented by the low-power neuromorphic system, which can be applied in edge computing of internet of things (IoT) and unmanned systems.",
"introduction": "1. Introduction In recent years, deep learning has shown a superior performance that exceeds the human-level performance in various types of individual narrow tasks [ 1 ]. However, in comparison with human intelligence that can learn to learn continually in order to execute unlimited tasks, the current successful deep learning methods still have a lot of drawbacks and limitations. In fact, humans can learn to learn by accumulating knowledge across their life time, which is a great challenge for artificial neural networks (ANNs) [ 2 ]. From this point of view, continual meta-learning aims at realizing machine intelligence at a higher level by providing machines with the meta-learning capability of learning to learn continually [ 3 ]. The human brain can realize meta-learning continually and avoid the catastrophic forgetting problem based on a combination of neural mechanisms [ 4 ]. The catastrophic forgetting problem is the critical challenge for developing the capability of continual meta-learning [ 5 ]. The human brain has implemented an efficient and scalable mechanism for continual learning based on neuronal activity patterns that represent previous experiences [ 6 ]. Neurons communicate with each other and process the neural information by using neural spikes, which is one of the most critical fundamental mechanism in the brain. Based on this mechanism, the human brain can realize superior performance in different aspects, such as low power consumption and high spatio-temporal processing capability [ 7 ]. Therefore, implementing a brain-inspired continual meta-learning algorithm based on spike patterns and the brain’s mechanisms is a promising technique. The spiking neural network (SNN) uses the biologically plausible neuron model based on spiking dynamics, while the conventional ANN only uses the neurons based on a static rate [ 8 ]. SNNs are applied to reproduce the brain’s mechanisms and to deal with the cognitive tasks [ 9 ]. In addition, the neuromorphic hardware based on SNNs can realize high performance in artificial intelligence tasks, including low power consumption, high noise tolerance, and low computation latency [ 10 ]. Previous neuromorphic hardware researches have proven these advantages by using various types of tasks, such as Tianjic, Loihi, BiCoSS, CerebelluMorphic, and LaCSNN [ 11 , 12 , 13 , 14 , 15 ]. Researchers have proposed SNN models to realize the short-term memory capability in a spike-based framework [ 16 ]. However, the current SNN models still suffer from the continual meta-learning problem under the non-Gaussian noise, and no previous study has solved this problem. Therefore, this is the focus of this study. Information theoretic learning (ITL) has attracted increasing attention in the field of machine learning in recent years to improve the learning robustness and enhance the explainable capability [ 17 , 18 , 19 ]. Previously, Chen et al. proposed researches focusing on maximum correntropy theory and minimum error entropy criteria to improve the robustness of machine learning theory [ 20 , 21 , 22 ]. In addition, a series of entropy-based learning algorithms have been presented to deal with the robustness improvement of machine learning models, including guided complement entropy and fuzzy entropy [ 23 , 24 , 25 ]. Nevertheless, there is no application of the ITL-based approach in the spike-based continual meta-learning to improve its learning robustness. Therefore, in this invited article, we aim to propose a novel approach to deal with this challenging problem. A novel model is presented, which is called meta-learning with minimum error entropy (MeMEE). We test the meta-learning capability of the proposed SNN model. Then, we investigate the robust working memory capability in non-Gaussian noise. Finally, the robust transfer learning performance is explored under a non-Gaussian noisy condition. Experimental results strongly suggest the robust meta-learning capability of the SNN model with a working memory feature in a non-Gaussian noisy environment.",
"discussion": "4. Discussion This paper presents an information theoretic learning framework for robust spike-driven continual meta-learning. Different from the previous SNN learning research, we first introduce the RMEE criterion to develop and improve the spike-based learning framework, which is significantly general and can also provide a series of theoretic insights. Moreover, the information theoretic framework allows us to obtain a direct understanding and better interpretation of the robust learning solutions of SNN models, compared with some previous studies focusing on improving the learning robustness of SNNs [ 36 ]. As a first step in establishing a rigorous framework for SNN continual meta-learning with RMEE, the presented research can be extended in both theoretical and practical aspects. From the theoretical point of view, one extension is to use the information potential to train the presented SNN model. For example, as shown in [ 37 ], Chen et al. presented a survival information potential algorithm for adaptive system training. This does not require computing of the kernel function and has good robustness performance accordingly. The other extension is to apply the proposed framework in other spike-based learning paradigms, including few-shot learning, multitask learning, and unsupervised learning [ 38 ]. From a practical point of view, the model is expected to be implemented on neuromorphic platforms to realize low-power and real-time systems for various types of applications. The state-of-the-art digital neuromorphic systems include Loihi [ 12 ], Tianjic [ 11 ], BiCoSS [ 13 ], CerebelluMorphic [ 14 ], LaCSNN [ 15 ], TrueNorth [ 39 ], and SpiNNaker [ 40 ]. By implementing embedded neuromorphic systems, it can be applied in different fields such as edge computing devices, brain–machine integration systems, and intelligent systems [ 41 , 42 , 43 ]."
} | 2,226 |
33392171 | PMC7775722 | pmc | 2,935 | {
"abstract": "Due to the non-renewable nature of fossil fuels, microbial fermentation is considered a sustainable approach for chemical production using glucose, xylose, menthol, and other complex carbon sources represented by lignocellulosic biomass. Among these, xylose, methanol, arabinose, glycerol, and other alternative feedstocks have been identified as superior non-food sustainable carbon substrates that can be effectively developed for microbe-based bioproduction. Corynebacterium glutamicum is a model gram-positive bacterium that has been extensively engineered to produce amino acids and other chemicals. Recently, in order to reduce production costs and avoid competition for human food, C. glutamicum has also been engineered to broaden its substrate spectrum. Strengthening endogenous metabolic pathways or assembling heterologous ones enables C. glutamicum to rapidly catabolize a multitude of carbon sources. This review summarizes recent progress in metabolic engineering of C. glutamicum toward a broad substrate spectrum and diverse chemical production. In particularly, utilization of lignocellulosic biomass-derived complex hybrid carbon source represents the futural direction for non-food renewable feedstocks was discussed.",
"conclusion": "Conclusion and Perspectives In this review, non-food carbon sources for the fermentation cultivation of C. glutamicum and chemical production were summarized. These feedstocks, including xylose, methanol, arabinose, glycerol, mannitol, N-acetylglucosamine, cellobiose, and cellulose hydrolyzates, provide alternative and renewable substrates to produce biobased chemicals. Microbial fermentation using these substrates is expected to alleviate the problem of food competition between industrial fermentation and human nutrition. However, there are still many technical bottlenecks in the practical application of these carbon sources. First, the tolerance of C. glutamicum to toxic materials, such as methanol and cellulose hydrolyzate, requires additional improvement to meet the demands of industrial fermentation. Fermentation inhibitors, including furfurals and phenolic aldehydes generated in the dilute-acid hydrolysis process of cellulose, are crucial factors restricting cellulose conversion. In addition to fermentation inhibitors, some nutrients may themselves hinder biosynthesis and secretion of target chemicals. Second, the instability of allogenically assembled metabolic pathways frequently restricts the biosynthesis of chemicals. These metabolic pathways disrupt the original metabolic balance, resulting in low growth and relatively low substrate utilization rates. Dynamic regulation of allochthonous metabolic pathways and chromosome insertion of genes involved in substrate utilization provide an effective solution for unstable pathway engineering. Third, the abundance and outlook of the alternative feedstock are important factors to consider when evaluating the technological index. As sea levels rise, land desertifies, the total human population of the earth increases, and the area of arable land decreases; biomass from terrestrial plants will no longer be the ideal substrate source. Utilization of marine biomass in the form of mannitol, mannose, and trehalose is one alternative, owing to the heavy specific proportion of these substrates in oceans. Meanwhile, stimulated by the development of sea-rice ( Chen R. et al., 2017 ) and sand rice ( Genievskaya et al., 2017 ), the utilization of rice straw hydrolyzate will also play an important role in industrial fermentation in the future.",
"introduction": "Introduction Since isolation in 1957, Corynebacterium glutamicum has been intensively applied for biobased chemical production due to its ability to secrete amino acids, which are traditionally used as drugs or health products ( Kogure and Inui, 2018 ; Wu et al., 2019 ). These amino acids and their derived chemicals are worth billions of dollars per year. Development of engineered strains with higher production performance is an active field that has attracted numerous researchers ( Lee and Wendisch, 2017b ; Li et al., 2017 ; D’Este et al., 2018 ; Zhang et al., 2018a , b ). In particular, most amino acids, including L-glutamate, L-lysine, L-arginine, L-valine, and L-ornithine, have achieved industrial-scale production due to rapid development of gene-editing and fermentation-manipulation techniques. These amino acids are used in human nutrition, food additives, and drug preparation, applications that benefit from the use of biologically safe C. glutamicum ( Lee and Wendisch, 2017a ). In addition to amino acids, C. glutamicum has also been extensively modulated to produce a multitude of valuable products, including bulk chemicals, natural products, polymers, proteins, and biofuels ( Becker et al., 2018b ; Shanmugam et al., 2018 , 2019 ; Wendisch et al., 2018 ). This vigorous development has benefited from fossil fuel depletion and anthropogenic climate change caused by the emission of toxic gases generated from oil decomposition, which traditionally served as the major source of manufactured chemicals ( Sun et al., 2018 ). However, biobased production of metabolites using C. glutamicum consumes large amounts of glucose, obtained from the hydrolysis reaction of starch, creating competition for food with humans. Hence, it is critical to exploit alternative renewable carbon sources, such as agricultural wastes, industrial wastes, and others for the cultivation of industrial model strains. In the past few years, research efforts have shifted toward biobased production of metabolites from non-food renewable feedstocks. Here are summarized recent advances in the utilization of alternative C-resources, including xylose, arabinose, methanol, glycerol, and mannitol, to produce high-value chemicals using C. glutamicum ( Figure 1 ). FIGURE 1 Overview of non-food renewable carbon resource utilization and chemicals production in C. glutamicum . G6P, glucose-6-phosphate; F6P, fructose-6-phosphate; F-1,6-P, fructose-1,6-diphosphate; G3P, glyceraldehyde-3-phosphate; DA3P, dihydroxyacetone-3-phosphate; PEP, phosphoenolpyruvate; PYR, pyruvate; Gluc6P, gluconate-6-phosphate; Ribu5P, L-ribulose-5-phosphate; Ribo5P, ribose-5-phosphate; Xylu5P, xylulose-5-phosphate; Ery4P, erythritol-4-phosphate; Sde7P, sedoheptulose-7-phosphate; F1P, fructose-1-phosphate; Gly3P, glycerol-3-phosphate."
} | 1,599 |
27512389 | PMC4962555 | pmc | 2,936 | {
"abstract": "Microbial chemosynthesis within deep-sea hydrothermal vent plumes is a regionally important source of organic carbon to the deep ocean. Although chemolithoautotrophs within hydrothermal plumes have attracted much attention, a gap remains in understanding the fate of organic carbon produced via chemosynthesis. In the present study, we conducted shotgun metagenomic and metatranscriptomic sequencing on samples from deep-sea hydrothermal vent plumes and surrounding background seawaters at Guaymas Basin (GB) in the Gulf of California. De novo assembly of metagenomic reads and binning by tetranucleotide signatures using emergent self-organizing maps (ESOM) revealed 66 partial and nearly complete bacterial genomes. These bacterial genomes belong to 10 different phyla: Actinobacteria, Bacteroidetes, Chloroflexi, Deferribacteres, Firmicutes, Gemmatimonadetes, Nitrospirae, Planctomycetes, Proteobacteria, Verrucomicrobia. Although several major transcriptionally active bacterial groups (Methylococcaceae, Methylomicrobium, SUP05, and SAR324) displayed methanotrophic and chemolithoautotrophic metabolisms, most other bacterial groups contain genes encoding extracellular peptidases and carbohydrate metabolizing enzymes with significantly higher transcripts in the plume than in background, indicating they are involved in degrading organic carbon derived from hydrothermal chemosynthesis. Among the most abundant and active heterotrophic bacteria in deep-sea hydrothermal plumes are Planctomycetes, which accounted for seven genomes with distinct functional and transcriptional activities. The Gemmatimonadetes and Verrucomicrobia also had abundant transcripts involved in organic carbon utilization. These results extend our knowledge of heterotrophic metabolism of bacterial communities in deep-sea hydrothermal plumes.",
"conclusion": "Conclusion In summary, a total of 66 bacterial genomes from 10 different phyla in the GB hydrothermal plume were reconstructed through a combination with high-throughput sequencing, de novo genomic assembly and tetranucleotide binning. The widespread presence of extracellular peptidase and carbohydrate metabolizing enzyme genes in most of bacterial genomes suggests that the utilization of organic carbon is a common microbial metabolism in the GB hydrothermal plume. Furthermore, the higher relative abundance of transcripts of extracellular peptidases and carbohydrate metabolizing enzymes in the plume than in background indicate these bacteria might be involved in degradation of organic carbon derived from hydrothermal chemosynthesis. Hence, primary and secondary production in hydrothermal plumes may be tightly coupled. Although the mechanism of trophic transfer is unknown, the presence of abundant viruses that infect the autotrophs in GB and other hydrothermal plumes ( Anantharaman et al., 2014 ) suggests that viral lysis is one potential source of carbon for heterotrophs. Among the most abundant and active heterotrophic bacteria in deep-sea hydrothermal plumes are members of the Planctomycetes, Gemmatimonadetes, and Verrucomicrobia. These results extend our knowledge of heterotrophic metabolism of bacterial communities within deep-sea hydrothermal plumes, highlighting their critical role in organic carbon cycling in the deep-sea hydrothermal plumes.",
"introduction": "Introduction Deep-sea hydrothermal vents are typically distributed along the mid-ocean ridges throughout the world’s oceans, where hot and chemically reduced hydrothermal vent fluids mix with cold and oxidizing seawater, forming hydrothermal plumes that rise 100s of meters off the seafloor and disperse 100s of kilometers from their sources. The hydrothermal inputs, such as H 2 S, H 2 , CH 4 , NH 3 , Mn 2+ and Fe 2+ , serve as energy sources that support microbial chemosynthesis ( Winn et al., 1986 ; Deangelis et al., 1993 ). Evidence has suggested that this chemosynthesis is a significant source of organic carbon to the deep ocean ( McCollom, 2000 ; Lam et al., 2004 , 2008 ; Baker et al., 2012 ; Lesniewski et al., 2012 ). Although the global annual biomass production in hydrothermal plumes is a small fraction (~10 12 g) of total primary production in the oceans, plume chemosynthesis may contribute a substantial fraction of organic carbon in the deep sea ( McCollom, 2000 ). Therefore, understanding the cycling of organic carbon in deep-sea hydrothermal plumes is potentially important in terms of both deep-sea microbiology and microbial food web interactions. Studies have begun to elucidate the importance and role of microorganisms and their metabolisms that operate within hydrothermal plumes ( Lesniewski et al., 2012 ; Dick et al., 2013 ). Surveys of small subunit (SSU) ribosomal RNA (rRNA) genes using clone libraries ( Sunamura et al., 2004 ; Dick and Tebo, 2010 ) and tag pyrosequencing ( Sylvan et al., 2012 ) have revealed the composition of plume microbial communities. Metagenomic and metatranscriptomic results have provided insights into the roles of dominant microorganisms involved in oxidation of sulfur, hydrogen, methane, and ammonia in hydrothermal plumes, such as SUP05 ( Anantharaman et al., 2013 ), Methylococcaceae ( Lesniewski et al., 2012 ; Li et al., 2014a ), Marine Group I Thaumarchaea ( Baker et al., 2012 ) and SAR324 ( Sheik et al., 2014 ). Studies also show that rare members of the plume microbial community such as Alteromonadaceae ( Li et al., 2014b ) and Nitrospiraceae ( Baker et al., 2013 ) are potentially keystone species with roles in iron uptake and nitrite oxidation, respectively. These results have greatly enhanced our knowledge of deep-sea hydrothermal plume microbiology. However, previous studies focused on metabolisms related to autotrophy and inorganic electron donors, and little work has addressed the fate of organic carbon produced via chemosynthesis. Two recent studies presented metagenomic and metatranscriptomic evidence that widespread archaea ( Li et al., 2015 ) and Alteromonas bacteria ( Baker et al., 2013 ) play roles in scavenging a variety of organic compounds in the deep sea. Another recent study also inferred a microbial food web in which chemoautotrophy supports and heterotrophy in hydrothermal plumes at the Mid-Cayman Rise ( Bennett et al., 2013 ). However, the broader role of bacteria in processing organic carbon in deep-sea hydrothermal plumes, both in terms of specific groups and pathways, remains unclear ( Dick et al., 2013 ). Guaymas Basin (GB), a submarine depression located on the seabed in the central area of the Gulf of California, hosts an unusual deep-sea hydrothermal system because of its location in a semi-enclosed basin and it proximity to the coast ( Lonsdale and Becker, 1985 ). The ridge axis in GB has been blanketed by a 400 m layer of organic-rich sediment that chemically modifies hydrothermal fluids as they ascend toward the seafloor ( Vondamm et al., 1985 ). The deep-sea vents inject hydrothermal solutions into the deep waters of a semi-enclosed basin, resulting in a plume where concentrations of methane (30 μM), ammonia (3 μM), and Mn (250 nM) are highly enriched over ambient deep sea levels (<0.5 μM, 0.25 μM, and 5 nM, respectively) at GB ( Dick et al., 2009b ). Methane and ammonia are energy sources that fuel substantial and diverse chemoautotrophy that provides a significant source of organic carbon to the deep oceans ( Lam et al., 2004 ; Lesniewski et al., 2012 ). The objective of this study was to understand organic matter utilization by heterotrophic bacterial communities within GB hydrothermal plumes via shotgun metagenomic and metatranscriptomic sequencing. The results of this study shed light on the ecological and physiological properties of heterotrophic bacteria and highlight their critical role in oceanic carbon cycling.",
"discussion": "Discussion In the GB hydrothermal plume, the dominant active methanotrophs and chemolithotrophs produce labile organic matter that may support heterotrophic microorganisms ( Baker et al., 2012 ; Lesniewski et al., 2012 ; Anantharaman et al., 2013 ; Li et al., 2014b ). A recent study has reported the heterotrophic metabolisms of deep-sea archaea, indicating the critical roles of ubiquitous archaea in scavenging various organic matter ( Li et al., 2015 ). To understand the bacterial heterotrophic metabolisms, we have reconstructed a total of 66 bacterial genomes from the GB hydrothermal plume through the combined use of shotgun metagenomic sequencing, de novo genomic assembly, tetranucleotide signature ESOM binning, and manual curation ( Figure 1 ; Supplementary Table S2 ). The identified 66 bacterial genomes affiliate to 10 different bacterial phyla, indicating a diverse community structure and greatly expanding the genomic coverage of GB plume microorganisms. Furthermore, most of reconstructed bacterial genomes are nearly complete, including many uncharacterized deep-sea bacteria, such as members of Deferribacteres (Bin98-1, Bin98-2, and Bin57) and unknown Deltaproteobacteria (Bin52-1, Bin52-2, Bin55, Bin56, and Bin61; Supplementary Table S3 ). Thus, the data presented here provide substantial genomic and transcriptomic information to understand the function of bacteria in deep-sea hydrothermal plumes. We note that the metatranscriptomic data presented in this paper are in terms of relative abundance, thus we are unable to make inferences regarding levels of gene expression . Rather our results are in terms of relative abundance of transcripts, which provides a valuable measure of the relative contributions of genes and organisms to the pool of community transcripts. Although the microbial community structures in the GB hydrothermal plume are similar to these in surrounding background waters ( Dick and Tebo, 2010 ; Lesniewski et al., 2012 ), most of the bacteria in the plume had higher relative abundance of transcripts than in background, suggesting bacteria in the GB plume are stimulated by abundant substrates ( Figure 3 ). Consistent with previous results, both genomic and transcriptomic evidence show that the most abundant and transcriptionally active bacterial groups were members of SUP05, methanotrophs, and SAR324, which display evidence of chemolithoautotrophic metabolisms ( Baker et al., 2012 ; Lesniewski et al., 2012 ; Anantharaman et al., 2013 ; Li et al., 2014a ; Sheik et al., 2014 ). However, the presence of genes encoding extracellular peptidases and carbohydrate metabolizing enzymes in most of bacterial genomes in the GB plume suggests that the potential for organic matter utilization in the deep-sea bacteria is also widespread ( Figure 4 ). Taken together with the results of archaea in GB ( Li et al., 2015 ), the utilization of organic matter is one of most common microbial metabolisms in the GB hydrothermal plume. Furthermore, transcriptomic evidence indicated that 66 bacterial groups from 10 bacterial phyla contain transcripts for different families of extracellular peptidases and carbohydrate metabolizing enzymes, suggesting these bacteria from different phyla are transcriptionally active for different organic matter utilization in the GB hydrothermal plume. Given the predominant chemolithoautotrophic metabolisms in the GB hydrothermal plume, the much higher abundance of transcripts of extracellular peptidases and carbohydrate metabolizing enzymes in the plume than in the background support that these bacteria may be degrading organic carbon derived from hydrothermal chemosynthesis. The Planctomycetes are a bacterial phylum that is ubiquitous in natural environments. All currently described pure cultures of Planctomycetes are aerobic heterotrophs (the members of Brocadiaceae that conduct anaerobic ammonia oxidation are still not purified) capable of growth on several sugars and sugar alcohols, such as glucose, fructose, mannitol, xylose, ribose, and fucose ( Youssef and Elshahed, 2014 ). Many grow exceptionally well on specific substrates, such as N -acetylglucosamine (NAG) presented as a side chain on mucin and chondroitin sulfate in nature. Furthermore, several described Planctomycetes species can metabolize polysaccharides, including starch, gelatin, and carboxymethyl cellulose ( Schlesner et al., 2004 ; Kulichevskaya et al., 2009 ; Bondoso et al., 2011 ; Kulichevskaya et al., 2012 ). This preference for monomers and sulfated polymers has lead to suggestions that degradation of sulfated polymeric carbon, e.g., marine snow, is a natural role for Planctomycetes in various habitats ( Glöckner et al., 2003 ; Woebken et al., 2007 ). Consistent with previous results, seven genomes of Planctomycetes identified in the GB hydrothermal plume contain abundant genes and transcripts for extracellular peptidases and carbohydrate metabolizing enzymes, particularly genes related to sulfatases, suggesting they perform important roles in organic carbon cycling in the GB plume ( Figure 6 ). Bacteria belonging to phylum Gemmatimonadetes have been identified as one of the top nine phyla found in soils, comprising ~2% of soil bacterial communities ( Bossio et al., 1998 ; DeBruyn et al., 2011 ). Despite their frequency and persistent abundance in soils, information about this phylum in marine environments is quite limited. Currently, only few representatives from this phylum have been isolated and partially characterized ( Joseph et al., 2003 ; Zhang et al., 2003 ; Davis et al., 2005 ). Members of Gemmatimonadetes are aerobic heterotrophs capable of utilization of multiple substrates, including east extract, polypepton, succinate, acetate, gelatin, and benzoate ( Zhang et al., 2003 ). A recent study reported five genomes of Gemmatimonadetes from estuary sediments and found that members of Gemmatimonadetes contain a large number of genes of carbohydrate hydrolysis and protein degradation ( Baker et al., 2015 ). Here, we have reconstructed a nearly complete (85.8%) genome of Gemmatimonadetes from deep oceans, which has the most abundant extracellular peptidase genes (9.2/Mb). Together with results from transcriptomes, as well as carbohydrate metabolizing enzymes, the data presented here strongly support Gemmatimonadetes bacteria as key mediators of the organic carbon cycle in the GB hydrothermal plumes ( Figure 6 ). The third interesting bacteria group are members belonging to the phylum of Verrucomicrobia, which are widely detected in different freshwater and marine habits ( Wagner and Horn, 2006 ). Previous studies have confirmed that members of Verrucomicrobia are capable of heterotrophic, carbohydrate-degrading metabolisms ( Arnosti, 2011 ; Steen et al., 2012 ; Cardman et al., 2014 ). The genomic evidence from a metagenome in Baltic seawater indicated that the genome of Verrucomicrobia bacteria encodes a diversity of GHs that likely allow degradation of various complex carbohydrates, such as cellulose, mannan, xylan, chitin, and starch ( Herlemann et al., 2013 ). A recent report also proposed Verrucomicrobia are involved in polysaccharide hydrolysis in the water column and sediment ( Cardman et al., 2014 ). Here, the four genomes of Verrucomicrobia contain diverse GH and CE genes, which also have relative high transcripts in the GB hydrothermal plume ( Figure 6B ), implying important roles of these microorganisms in the deep-sea carbon cycle."
} | 3,828 |
37223699 | PMC10202376 | pmc | 2,937 | {
"abstract": "Repelling liquid drops from engineering surfaces has attracted great attention in a variety of applications. To achieve efficient liquid shedding, delicate surface textures are often introduced to sustain air pockets at the liquid–solid interface. However, those surfaces are prone to suffer from mechanical failure, which may bring reliability issues and thus limits their applications. Here, inspired by the aerodynamic Leidenfrost effect, we present that impacting drops are directionally repelled from smooth surfaces supplied with an exogenous air layer. Our theoretical analysis reveals that the synchronized nonwetting and oblique bouncing behavior is attributed to the aerodynamic force arising from the air layer. The versatility and practicability of our approach allow for drop repellency without the aid of any surface wettability treatment and also avoid the consideration of mechanical stability issues, which thereby provides a promising candidate for the applications that necessitate liquid shedding, e.g., resolve the problem of tiny raindrop adhesion on the automobile side window during driving.",
"conclusion": "Conclusion In summary, the repulsion and directional removal of impacting drops were realized by simply imposing an exogenous air layer on a solid substrate. By controlling the thickness and flow velocity of the air layer, in principle, the solid surface can be kept nonwetting under the continuous impact of a water drop or other liquids. This kind of rebound is based on the effect of dynamic pressure provided by the air layer. The vertical component of the aerodynamic force pushes the drop upward to rebound for one thing, but also, the horizontal component along the flow direction involves the long-distance movement of the drop on the surface. The dynamics of the vent can spend as a form of an aerodynamic Leidenfrost effect by controlling the air layer between the liquid drop and the solid surface. The effect is more productive for smaller drop sizes and more sloping surfaces. We finally applied this method to the surface of the car side windshield in the lab, showing a strong potential to solve the problem of driving safety caused by raindrop adhesion on the side windshield on rainy days. Such aerodynamic repellency of liquid drops has universal applicability, regardless of the physical properties of the solid surface and impacting drops.",
"introduction": "Introduction Traffic safety threatens the lives of humankind. Every minute, more than 2 people die because of traffic [ 1 ]. Notably, the traffic accident rate on rainy days is much higher than on sunny days [ 2 – 4 ], for which an important factor is the obstruction of vision caused by raindrop adhesion. Generally, raindrops on the front windshield or rearview mirror can be eliminated by wipers operating or heating, while raindrop adhesion on the side windshield, which may also cause undesirable consequences, is still an unsolved issue. The adhering raindrop on the side windshield acts like a convex lens, which scatters lights and thus distorts the visual scene image (Fig 1 A), as the scene of blurred vision we simulate in Fig. 1 B. This greatly worsens the drivers’ horizons and may lead to catastrophic car accidents [ 5 – 8 ]. Therefore, a strategy to efficiently remove the adhered drops and eliminate blurred vision is expected. Fig. 1. Blurred vision caused by drop adhesion. (A) A photo shows the blurred vision caused by raindrop adhesion. (B) Effect of surface raindrop adhesion on light scattering, reflection, and refraction. (C) Surface modification to resolve the problem of raindrop adhesion. (D) The poor mechanical stability of the micro/nanostructured superhydrophobic coating. To address the issue of blurry side windshield, surface modification might be considered at first thought. By rendering the surface superhydrophobicity or superhydrophilicity [ 9 – 18 ], the solid–liquid interaction could be alleviated or reinforced and thus facilitates rapid liquid drop shedding or forms liquid films to avoid the optical effects of drop curvature (Fig. 1 C). However, commercial superwetting materials often suffer from durability issues due to the fragility of their encompassing micro/nanostructures, especially in dynamic working conditions (e.g., reciprocating sliding) [ 19 – 25 ]. This is evidenced by Fig. 1 D and Fig. S1 , where the stability of a glass with a commercial superhydrophobic coating is tested with a glass scraper (mimicking the action of rolling up or down the side windshield). Specifically, the original superhydrophobic-treated area is out of function for water repellency after being scraped once. In addition, the introduction of micro/nanostructures inevitably leads to light scattering, which deteriorates light transmission through surfaces. Therefore, surface modification shows limited performance in dynamic working conditions due to its poor mechanical stability and transparency, although they may work well at initial use. Apart from providing air pockets by introducing a physical structure, an alternative strategy for obtaining the nonwetting surface property is to generate an air cushion at the solid–liquid interface, which is capable of preventing the incoming liquid drops from contacting the surface fundamentally [ 26 – 28 ]. The well-known Leidenfrost effect induced by overheating enables volatile liquids to levitate above a vapor cushion [ 29 – 32 ]. More intriguingly, the dynamical motion of surfaces (over a threshold velocity) can also engender an air layer to repel the incoming drop, known as the aerodynamic Leidenfrost effect [ 33 – 36 ]. Unfortunately, these methods necessitate external actuation, which makes the experimental setup complex and cumbersome and may bring reliability issues. Inspired by the aerodynamic Leidenfrost effect, in this work, we demonstrated that incoming drops are repelled and directionally removed by glass surfaces supplied with an exogenous air layer. The threshold speed of drop bouncing, in particular as a function of the drop impact velocity, is explored experimentally. Theoretical analysis discloses that the synergic effect of nonwetting and oblique bouncing is induced by the aerodynamic force arising from the air layer. Such a strategy can effectively realize drop repellency avoiding the surface inherent wettability and mechanical stability, providing a promising candidate for resolving the problem of raindrop adhesion on the automobile side windshield.\n\nIntroduction of an exogenous air layer An air compressor was used to introduce an air layer onto a glass surface through an air introduction device (Fig. 2 A and Fig. S2 ). The device consisted of a round air inlet and a rectangular air outlet with adjustable thickness. The air inlet was easily connected to the air compressor via a rubber pipe, while the air outlet’s length and width ( m and n , respectively) were optimized for even airflow distribution. The device's internal structure is shown in Fig. S2 , where the transitional design between the air inlet and outlet benefits the stabilization of the airflow. The airflow rate ( Q ) was measured by a digital display gas flowmeter (range from 0 to 200 l/min). The initial flow velocity of the air layer was determined by the size of m and n . Using a 3-dimensional (3D) printer, 4 different outlet thicknesses ( n = 0.6, 1, 2, and 3 mm) were printed with a fixed length m = 15 mm. Fig. 2. Aerodynamic repellency of impacting drop by introducing an air layer. (A) Design of air layer introduction device. The inset shows the side-view schematic diagram of a drop impacting the air layer surface. (B) Impacting drop bounced off an air layer glass surface. The impacting velocity v = 0.2 m/s, the airflow velocity u is 15 m/s, the thickness ( n ) of the air outlet is 3 mm, and the diameter of the impacting drop D is 2.5 mm. (C) Impacting drop wetted the glass surface at a velocity of 0.2 m/s without introducing an air layer. (D) The height ( h ) change curves of the lowest part of the impacting drop from the surface vs. time ( t ) during the whole impacting process.\n\nSurface treatment and air layer introduction device printing The glass slides (borosilicate glass material, 75 mm × 26 mm × 1 mm, MARIENFELD) were rinsed with ethanol (AR, ≥99.7%, HUSHI) and ultrapure water (18.2 MΩ/cm, from Synergy Water Purification System, Millipore SAS), respectively. The final glass surface exhibited a hydrophilic wetting state. The air layer introduction device was obtained by 3D printing technique (3D printer, 100-μm printing precision, i3 mega, ANYCUBIC) using polylactic acid as the printing material. The processing temperature was set as 200 °C."
} | 2,171 |
40128897 | PMC11934508 | pmc | 2,938 | {
"abstract": "Background Porous environments constitute ubiquitous microbial habitats across natural, engineered, and medical settings, offering extensive internal surfaces for biofilm development. While the physical structure of the porous environment is known to shape the spatial organization of biofilm inhabitants and their interspecific interactions, its influence on biofilm community structure and functional diversity remains largely unknown. This study employed microfluidic chips with varying micropillar diameters to create distinct pore spaces that impose different levels of spatial constraints on biofilm development. The impact of pore spaces on biofilm architecture, community assembly, and metabolic functions was investigated through in situ visualization and multi-omics technologies. Results Larger pore sizes were found to increase biofilm thickness and roughness while decreasing biofilm coverage over pore spaces. An increase in pore size resulted in reduced biofilm community evenness and increased phylogenetic diversity. Remarkably, biofilms in 300-μm pore spaces displayed the highest richness and the most complex and interconnected co-occurrence network pattern. The neutral model analysis demonstrated that biofilm assembly within different pore spaces was predominantly governed by stochastic processes, while deterministic processes became more influential as pore space increased. Exometabolomic analyses of effluents from the microfluidic chips further elucidated a significant correlation between the exometabolite profiles and biofilm community structure. The increased community richness in the 300-μm pore space was associated with the significantly higher exometabolome diversity. Conclusions Collectively, our results indicate that increased pore space, which alleviated spatial constraints on biofilm development, resulted in the formation of thicker biofilms with enhanced phylogenetic and functional diversity. \n Video Abstract Supplementary Information The online version contains supplementary material available at 10.1186/s40168-025-02075-0.",
"conclusion": "Conclusions Our study demonstrates that the physical structure of porous environments plays a critical role in shaping biofilm community structure, assembly processes, and functional diversity. By employing microfluidic chips with controlled pore sizes, we observed that larger pore spaces increased biofilm thickness and roughness, while decreasing biofilm coverage and alleviating spatial constraints. Larger pore size resulted in reduced biofilm community evenness and increased phylogenetic diversity, with deterministic processes exerting a greater influence on community assembly. Exometabolomic analyses of effluents from the microfluidic chips further elucidated a significant correlation between the exometabolite profiles and biofilm community structure. The increased community richness in large pore spaces was associated with the significantly higher exometabolome diversity. These findings highlight that mitigating spatial constraints in porous environments promotes the development of biofilms with greater phylogenetic and functional diversity, providing insights into managing microbial communities across environmental, industrial, and medical contexts.",
"discussion": "Discussion Porous environments serve as the predominant habitats for microbes [ 1 , 18 ]. The intricate pore structures within these environments modulate diverse microbial interactions and foster the development of complex microbial communities [ 19 ]. However, the role of pore space in shaping biofilm community structure and function remains largely unexplored. In this study, we employed laboratory-constructed microfluidic chips to investigate the assembly mechanisms of soil microbial biofilm communities in pore spaces under different levels of spatial constraints. The biofilm in smaller pore spaces exhibited reduced thickness and roughness while also demonstrating greater coverage by forming a streamer structure within these spatially confined pore spaces. Increasing the pore size led to a decrease in biofilm community evenness and a concomitant increase in phylogenetic diversity. The biofilm inhabiting the 300-μm pore space exhibited significantly higher richness compared to those in smaller pore spaces. The neutral model analysis revealed a diminishing contribution of stochastic processes in biofilm community assembly with increasing pore space. Exometabolomic analyses further revealed a significant correlation between exometabolite pools and biofilm community structure. A significantly higher exometabolome diversity was observed in the 300-μm pore space which associated with the increased biofilm community richness. Together, these results reveal that large pore spaces with reduced spatial constraints promote biofilm formation and foster the assembly of a more diverse microbial community, both in terms of phylogenetic composition and functional attributes. Soil biofilms typically range from 10 to 100 μm in thickness [ 1 , 20 ]. In this study, we fabricated microfluidic chips with pore sizes ranging from 20 to 300 μm to impose varying degrees of spatial constraints on soil biofilm development. According to soil pore size classification, pores between 10 and 50 μm were designated as small coarse pores, while those higher than 50 μm were defined as wide coarse pores [ 21 ]. These pore size ranges correspond to mechanical fissures, cracks, and biopores (from roots and earthworms), which are known hotspots for soil biofilm formation. Previous studies have reported divergent effects of pore space on microbial diversity [ 5 , 22 – 24 ]. These varying results were generally attributed to complex organic pools, different microbial communities, and distinct mineral compositions. To isolate the effect of pore space geometry, we investigated the biofilm assembly across different pore spaces with identical pore water chemistry, solid substratum, and inoculum community composition. Our findings revealed that the smaller pore size led to reduced community richness and phylogenetic diversity. Consistent with our findings, a previous study in bermudagrass ecosystems identified soil texture as the second most influential factor (after pH) shaping soil microbial communities, with clay content exhibiting negative and positive correlations with bacterial richness and evenness, respectively, particularly under wet conditions [ 24 ]. Our analyses further revealed that a stronger fit ( R 2 ) for the Sloan’s neutral community model within the smaller pore spaces indicates the colonization is predominantly governed by a stochastic, birth–death immigration process (Fig. 4 a, b, c, d) [ 25 ]. Deterministic processes, such as microbial interactions, appear to exert less influence on biofilm assembly in confined porous environments. The diminished importance of microbial interaction in shaping biofilm community can be attributed to biofilm clogging, which hindered nutrient diffusion and metabolite exchange among biofilm inhabitants. Consequently, the enhanced microbial diversity within confined pore spaces observed in previous studies may arise from factors beyond pore size. Meanwhile, a significant increase in community richness was observed at the 300-µm pore size. Neutral model analysis revealed a higher number of abundant taxa were selected against, while more intermediate taxa were actively selected at the pore size of 300 μm comparing to smaller pore spaces (Fig. 4 i, j, k, l). It indicates that abundant taxa were incapable of occupying all niches within the large pore space, thereby enabling the growth of intermediate taxa. This observation aligns with the reduced proportion of generalists and the increased proportion of opportunists as pore size expands (Supplementary Fig. S12). Metabolomic analysis of the 300-μm pore space unveiled greater chemical diversity within the exometabolome and the emergence of unique metabolic pathways not observed in other small pore spaces, indicating extensive metabolic interactions in the large pore space. These interactions were further supported by the highest level of network complexity and connectivity observed in co-occurrence analysis (Fig. 5 , Supplementary Table S1). It suggests that large pore spaces can provide more spatial niches to facilitate the colonization of non-abundant taxa and foster the coexistence of phylogenetically distant taxa. These findings have significant implications for the management, control, and engineering of biofilms in porous environments. In confined pore spaces, increased biofilm coverage reduces pore connectivity, inhibiting biofilm activity and leading to simpler, more homogeneous communities. Consequently, biofilms in such environments are more susceptible to removal by biocidal agents. In contrast, biofilms in larger pore spaces exhibit elevated EPS content, making EPS targeting crucial for effective biofilm eradication. Furthermore, sufficient pore space supports the development of thicker and more metabolically diverse biofilms, suggesting that increasing pore space may be an effective strategy for enhancing biofilm functionalities. Previous studies have highlighted the superior performance of large-pore membrane bioreactors and macroporous biofilm carriers in contaminant removal through biocatalysis [ 26 , 27 ]. This study has elucidated the influence of pore sizes on biofilm community assembly and function traits using a combined microfluidic and multi-omics approach. Future metagenomic studies could further characterize the genomic functional potential of biofilms formed in varying pore sizes and at different growth stages. These genomic data could corroborate the observed relationship between pore size and genomic functional potential, particularly the increased functional redundancy within larger pores that may enhance community resilience to environmental perturbations, such as fluctuations in pore water and nutrient availability, antimicrobial agent exposure, and microbial invasion. Further investigation is warranted into the spatial organization of different biofilm inhabitants within pore spaces and their mutual interactions. The spatial structure of complex biofilm communities can be visualized using techniques such as fluorescent gene tagging and high-phylogenetic-resolution microbiome mapping by fluorescence in situ hybridization (HiPR-FISH) [ 28 ]. Additionally, the effects of other key physicochemical factors on biofilm development within porous ecosystems can be investigated by leveraging microfluidic platforms. For example, substratum properties can be assessed by coating the internal surfaces of microfluidic devices with polymers or minerals through direct adsorption or light curing. Flow conditions can be precisely controlled to study hydrodynamic influences, including the impact of varying shear forces or unsaturated microenvironments."
} | 2,727 |
29556172 | PMC5844986 | pmc | 2,939 | {
"abstract": "Asynchronous event-based sensors, or “silicon retinae,” are a new class of vision sensors inspired by biological vision systems. The output of these sensors often contains a significant number of noise events along with the signal. Filtering these noise events is a common preprocessing step before using the data for tasks such as tracking and classification. This paper presents a novel spiking neural network-based approach to filtering noise events from data captured by an Asynchronous Time-based Image Sensor on a neuromorphic processor, the IBM TrueNorth Neurosynaptic System. The significant contribution of this work is that it demonstrates our proposed filtering algorithm outperforms the traditional nearest neighbor noise filter in achieving higher signal to noise ratio (~10 dB higher) and retaining the events related to signal (~3X more). In addition, for our envisioned application of object tracking and classification under some parameter settings, it can also generate some of the missing events in the spatial neighborhood of the signal for all classes of moving objects in the data which are unattainable using the nearest neighbor filter.",
"conclusion": "7. Conclusion and future work In this work, we presented a novel neural network-based noise filtering (NeuNN) approach and compared it with the typically used nearest neighbor (NNb) noise filter for event-based image sensors. We also described how we can map the NeuNN filter algorithm to TrueNorth architecture efficiently. Though the filter used is available on the TrueNorth platform for implementation, it can be adapted to any other neural network-based neuromorphic hardware for filtering noise from event-based sensors. We showed that the proposed NeuNN filter is capable of generating new events that can be associated with the signal while the output events of traditional filters are strictly a subset of input events. This approach results in much higher signal retention for NeuNN and also results in higher SNR for NeuNN-filtered images. We used manual annotation information for analysing metrics. In future, we will integrate a tracker and classifier after the filter and evaluate the performance of the filtering algorithm based on tracker and classifier performance. Finally, in the current study, the background noise characteristics change with respect to the time of the recording. When the noise in the environment changes, configuring the neuronal parameters for the task of filtering across varying noise conditions becomes more challenging. To improve the noise filtering performance, a learning mechanism for updating the parameters based on the noise characteristics could be introduced and the possibility of incorporating such a mechanism needs to be studied. The total number of cores used in this two-layer filtering approach is quite high and increases rapidly with the size of the filter and image. One way to reduce the number of cores is to combine both filtering operations into one layer. In this method, the neurons in the layer 2 filter can be reset to a low value below the threshold and can slowly recover to resting potential through a leak. The time taken by the neuron to recover to its resting potential acts like a refractory period. This method is similar to a relative refractory period (Gerstner and Kistler, 2002 ) because the neuron can still fire a spike in this period if it receives a large number of input spikes, whereas the method we propose is akin to absolute refractory period. However, this method has the problem that neurons in layer 2 become less excitable in general and do not generate output events even if inputs are coming from surrounding pixels within that refractory period. We found that the percentage of signal that remains after using this approach is less than our proposed solution. Another possibility is to use two layers, but instead of using three neurons per pixel in layer 1 for the refractory period, use one neuron per pixel using the relative refractory method described earlier. However, this still suffers from a non-constant refractory period due to the variable number of spike inputs the neuron receives during the refractory period.",
"introduction": "1. Introduction Inspired by the efficient operation of biological vision, research on neuromorphic event-based image sensors, or “silicon retinae,” took off a few decades back (Mahowald and Mead, 1991 ). Recently, the technology has matured to a point where the sensors are commercially available. Dynamic Vision Sensor (DVS) (Lichtsteiner et al., 2008 ), Asynchronous Time-based Image Sensor (ATIS) (Posch et al., 2011 ), the sensitive DVS (Leñero-Bardallo et al., 2011 ), and the Dynamic and Active pixel Vision Sensor (DAVIS) (Berner et al., 2013 ) are some of the popular Address Event Representation (AER) change detection sensors that can be employed for various applications. Unlike conventional image sensors that operate by sampling the scene at a fixed temporal rate (typically between 30 and 60 Hz), these sensors employ level crossing sampling pixels which asynchronously and independently signal an event if sufficient temporal contrast is detected (Posch et al., 2014 ). This results in a higher dynamic range, lower data rate and lower power consumption compared to frame based imagers. Several possible applications of these sensors have been investigated including traffic monitoring (Litzenberger et al., 2006 ), stereovision (Rogister et al., 2012 ), high speed sensory-motor loops (Delbruck and Lang, 2013 ), motion estimation (Orchard and Etienne-Cummings, 2015 ) and pose estimation (Valeiras et al., 2016 ). However, most works on processing the events generated by silicon retina are performed on FPGA or microcontrollers resulting in lower power efficiency of the whole system than that afforded by the sensors. Hence, interfacing these sensors with low-power neuromorphic processors would benefit the development of low power systems that could potentially be deployed in wearables or as sensors in the Internet of Things (IoT). Several low power neuromorphic processors have been proposed in recent years (Painkras et al., 2013 ; Benjamin et al., 2014 ; Chen et al., 2016 ) and some of them have been interfaced with spiking image sensors (Orchard et al., 2015 ). However, the power dissipation of these systems are still an order of magnitude more than that required for a wearable device. On the other hand, the TrueNorth processor developed by IBM (Merolla et al., 2014 ) has both the low power footprint and the large neuron count available on a single chip required to interface with image sensors. It processes spiking data in real-time with a fixed 1 kHz clock using distributed memory kept locally in each core instead of one central location. TrueNorth consumes very low power of approximately 50–70 mW for typical networks and results in a good combination of power efficiency and configurability. As a consequence, there have been recent reports of interfacing the TrueNorth with “silicon retinae” to create real-time low-power vision applications (Amir et al., 2017 ). In keeping with the above trend, we propose in this work a set of noise filtering primitives that may be used as a pre-processing block for event-based image processing applications on TrueNorth. The concepts presented in this paper can also be used to implement noise filtering for silicon retinae on other embedded hardware platforms. In this work, we focus on the case of a static camera observing a scene such as in a surveillance scenario. Noise filtering algorithms such as nearest neighbor for event-based imagers presented in the literature exploit the temporal correlation associated with activity across the neighborhood pixels (Dominguez-Morales, 2011 ; Ieng et al., 2014 ; Linares-Barranco et al., 2015 ; Liu et al., 2015 ; Czech and Orchard, 2016 ). Though these filters capture the activity of the fast moving objects, they typically filter out the activity by the small and slow-moving objects due to weak temporal support. The spiking neural network based noise filter proposed in this paper is shown to work better than other popular filters in this respect and preserve a large fraction of activity associated with the signal (and sometimes generate more events than input) while filtering out most of the noise events which is beneficial to object tracking and classification in our envisioned applications. This paper is structured as follows. In section 2, we describe the ATIS setup used in our experiments as well as the nature of data recorded. Section 3 provides an overview of TrueNorth, while in section 4, we introduce the noise filtering approaches. We present the implementation of the noise filter on TrueNorth in section 5 and the results of the experiments are presented in section 6. Finally, we conclude the paper with some discussions about future work in section 7.\n\n5.1. Layer 1: introducing refractory period using the neurons on TrueNorth cores As described in section 4.1, the refractory period layer incorporates a minimum time difference between the successive events at any pixel location before passing through the NeuNN filter. Each pixel needs three input axons and three output neurons available on the core of the TrueNorth to stream events for the refractory period function. The connectivity mapping and neuronal parameters specified in Figure 4A are used to create the refractory period operation on each input pixel. The output of the neuron labeled 2 produces the spike output with the desired refractory period by configuring the weight of neuron 0. The parameters for neurons 1 and 2 are identical thereby generating multiple copies of the neuronal output. The operation of this circuit may be described as follows: at the start when there is a spike input coming from axon 0, neurons 1 and 2 produce spikes, thereby allowing the first input spike to appear at the output. When there is no spike input, neuron 0 emits a spike every millisecond due to the leak value (+1), threshold value (1) and reset value (+1). However, when there is feedback input coming from the neuron 1 output and it has a negative weight of −( T ref − 1), it brings down the membrane potential and inhibits the spike, thus providing parameter control to set the desired refractory period. Because there are 256 input axon lines and 256 output neurons in each core, up to 85 pixel inputs can be mapped to each core taking a block of three axons and three neurons for each pixel. To map all the data coming from all pixels of ATIS at least a total of 859 (304 × 240/85) cores are required. Figure 4 (A) Connectivity mapping for refractory neuron function along with the neuronal parameters. (B) Statistics of the ISI for both signal and noise events show many noise events at ISI between 1 and 10 ms, whereas signal events mostly have ISI higher than 10 ms. This is used to choose a refractory period value in the range of 3–5 ms. To decide the appropriate setting for T ref , we have plotted the inter-spike interval ( ISI ) histogram (Figure 4B ) for both signal and noise events. To do this, signal events were extracted based on annotated tracks of valid objects in the scene while noise events correspond to those occurring outside these marked bounding boxes. It can be seen from Figure 4B that there is as many noise events with ISI < 10 ms, whereas most signal events are at ISI > 10 ms. To suppress these frequently occurring noise events, a refractory period between 3 and 5 ms seems an appropriate choice. Based on the above analysis, the refractory period T ref of 5 ms is used at the first layer of processing the data both on software and TrueNorth, before applying either of the second layer filters such as NNb and NeuNN on software and TrueNorth respectively. There are a total of 96,232,414 events in all the recordings combined and after passing the entire data set through the refractory layer, there is a loss of around 86.6 % of the original events and a total of 12,909,503 events (around 13.4 % of the original events) were retained for the second layer of filtering. The effect of the refractory period on the hardware adaptation of the refractory period layer on TrueNorth is identical to the software implementation with respect to the filtered events and discussed in detail in section 6.",
"discussion": "6. Results and discussion 6.1. Layer 2 filter parameter optimization 6.1.1. Software filter (NNb) For the NNb filter, we swept the value of T NNb in the range of 0.5–5 ms to observe the effect it has on the three metrics PSR f , PNR f and SNR f (in dB). The result of this exploration (on recording at 6:30 p.m.) is shown in Table 4 . From Table 4 , we can see that SNR f (in dB) is highest for the smallest value of T NNb due to the very small amount of remaining noise. However, at this setting, the amount of signal remaining PNR f is also very low, and in practice, we need a balance between SNR f (in dB) and PSR f . Hence, in practice, it is better to use values of T NNb in the range of 1–3 ms. In the rest of the simulations in this paper, we use T NNb = 1 ms. Table 4 Effect of varying parameter T NNb in the NNb filter on Filtering Metrics. T NNb (ms) PSR f PNR f SNR f 0.5 11.97 0.09 20.17 1 21.54 0.18 19.74 2 35.36 0.35 19.02 3 44.38 0.51 18.32 4 50.42 0.68 17.69 5 54.49 0.83 17.12 6.1.2. TrueNorth filter (NeuNN) The NeuNN network is configured with a single type of synapse with synaptic weight w for all the synapses and one neuron type with threshold α and leak λ for all the neurons. The design space exploration is carried out with ratio of threshold/weight, (α/ w ) and leak/weight, (λ/ w ) ranging from [1–3] and [0.150–0.7], respectively with a fixed synaptic weight. The filtering metrics presented in section 2.4 are used to study the effect of TrueNorth neuron parameters and the results are plotted in Figure 5 . Figure 5 Effect of TrueNorth neuron parameters on the metrics from section 2.4 (the above plot used 6:30 p.m. data). From Figure 5A , the signal remaining after filtering increases with the decrease of both (α/ w ) and (λ/ w ) and it can be higher than 100%. The reason for this is that the output neurons are receiving inputs from all the neighboring pixels within a particular distance and when (α/ w ) and (λ/ w ) is low, the filter spikes very often, resulting in more output events than input events for the signal. Therefore, instead of any signal loss that typically happens after using traditional filtering methods, we can observe a tremendous amount of signal gain for certain parameter ranges of the NeuNN filter. This behavior can be explained by thinking of NeuNN in this parameter range as a spatiotemporal version of the dilation operator used in traditional image processing (Gonzalez and Woods, 2008 ). From Figure 5B , we can see that the noise remaining after filtering is also affected similarly as signal remaining with both (α/ w ) and (λ/ w ). Because the objective of this filtering is to keep the noise as low as possible without compromising too much on the signal, we a have a window of parameter space, where (α/ w ) and (λ/ w ) can be configured. It can be seen from Figure 5C that the signal to noise ratio (SNR) increases with both increasing (α/ w ) and (λ/ w ). Choosing the parameters of the filter cannot depend on the high SNR (which is due to the reduction in the noise) alone; the signal remaining and noise remaining both need to be considered in conjunction with the requirements of the following processing steps (Tracking–prefers as low noise as possible and Classification–prefer high amount of signal). We chose two sets of parameters: (1) (α/ w ) = 2 and (λ/ w ) = 0.33, and (2) (α/ w ) = 2.5 and (λ/ w ) = 0.583 where the first set results in higher PSR f than NNb filter (due to the dilation like behavior described earlier) while the second set results in similar PSR f as NNb filter. The results in the paper are reported for parameter set 1 by default while we specifically mention when parameter set 2 is used. 6.2. Qualitative comparison of both filtering approaches From Figure 6 , the refractory layer reduces high-frequency temporally localized noise while layer 2 removes spatially spread noise. NeuNN also retains more signal events than the NNb filter. We can notice that the refractory layer significantly reduces the high-frequency noise pixel in yellow.We can also notice the missing activity due to the occlusion of objects and there is a small region of dead pixels (around 8–10 pixels) toward of the right of the frame and the activity is not possible to be removed by both the filters. Figure 6 Comparison of color-coded average spike frequency (events/s) at each pixel for all 304 × 240 pixels combining all the sample files for (A) original data; (B) after layer 1refractory period of 5 ms; (C) after layer 1 refractory filter and layer 2 NNb filter on software; and (D) after layer 1 refractory filter with layer 2 NeuNN filter on TrueNorth. Figure 7 that NeuNN retains many more signal events while keeping slightly more noise. The benefits of NeuNN are most evident for slow-moving objects like pedestrians, where the NNb filter removes almost all signal events. We can visually verify that there are more events in the tracker for all classes of objects in the scene from the NeuNN filter screenshots on the right compared with the NNb filter on the left. Slow-moving pedestrians are the worst performing class in terms of retaining the signal and they are barely visible on NNb filter and this makes them quite challenging to detect and classify. Their activity, however, is visible on the NeuNN filter. There are slightly more noise events present outside the tracker after filtering using the NeuNN filter in comparison with the NNb filter when there is very high back noise activity. Figure 7 Comparison of NNb filter on software (on the left) with the NeuNN filter on TrueNorth (on the right) using the same time instances and same files mentioned in Figure 1B . 6.3. Quantitative comparison of filtering approaches using the metrics Next, we quantitatively evaluate the benefits of the NeuNN filter over the NNb filter using the metrics described earlier. The results are plotted in Figure 8 . It can be seen from Figure 8A that the signal remaining after filtering using the NeuNN filter is significantly higher (~3X) compared with the NNb filtering. NeuNN is also capable of producing missing activity in the neighborhoodnone of the traditional filters are capable of doing this. Figure 8B shows that the remaining noise is slightly higher (~2.5X) when the background noise increases significantly at the later part of the night. Finally, Figure 8C shows the SNR before and after filtering. The SNR ratio before and after filtering is compared for the two filtering methods and the SNR (in dB) is significantly higher (by 14.75–29.3 dB) for filtering using NeuNN on TrueNorth due to the retention of the signal after using the neural network-based filter. Figure 8 Comparison of all the metrics of NNb filter and NeuNN filter with respect to each sample file at various recording times representing varying background noise characteristics. A refractory period of 5 ms is used for the first layer filter and the filtering performance for the NNb filter on software and NeuNN filter on TrueNorth are compared. Across all cases, NeuNN generates more than 5X events compared with the NNb filter for all classes of objects moving in the scene. We observe from Figure 9 that the signal loss for the NNb filter is significant for all objects (at around 80%). However, the NeuNN filter generates extra signal events in the case of fast-moving objects (all except pedestrians) compared with the NNb filter. This is a significant advantage of the NeuNN on TrueNorth over the NNb filter. Though slow-moving humans with least spike rate per pixel of 52 spikes/sec (averaged over all videos) are affected the worst in comparison with other objects using both approaches, the signal loss is only around 30% with the NeuNN filter compared with 83% with the NNb filter. This makes it easier to track and classify pedestrians using the NeuNN filter in the later stages of processing. Finally, we observe that the signal remaining is highest for bikes in both approaches. The reason for this is that bikes are compact and fast-moving objects with highest spike rate per pixel of 142 spikes/ sec (averaged across all videos). Other fast-moving objects, like buses, have regions of low contrast (e.g., glass windows) that do not generate events, making their spatial density of events low (approximate spike rate per pixel of 95 spikes/sec averaged across all videos). Figure 9 Average percentage of signal events remaining compared with the original data for each class after two-layer filtering using all the samples in Table 1 . 6.4. Discussion We have demonstrated a new noise filtering algorithm for event based imagers and shown its implementation on TrueNorth. There are some important points to note about our proposed method. First, it should be noted that since TrueNorth processes data at a millisecond clock tick, we lose the original microsecond resolution of events coming from the sensor. However, for our application of object tracking and classification in traffic monitoring or surveillance, this is acceptable and microsecond temporal resolution is unnecessary. Second, the response of a rectangular photodiode, such as the ones in ATIS, will be different for horizontally and vertically moving contours (Clady et al., 2017 ). However, for the traffic monitoring and surveillance applications we are considering, the orientation of the camera will be fixed and we presented results for this orientation. For changes of orientation, our filtering method will still work–however, the exact gains in SNR f might be different. Third, we expect the effect of adding more events in regions with signal events (spatio-temporal dilation) will be helpful in tracking under noisy conditions. However, the shape information of the object may be distorted by this filtering. We propose an architecture with two parallel processing paths in that case–the raw signal (or mildly filtered one) goes to a classifier in one path, while in the other path, we have the NeuNN filter followed by the tracker. The tracker informs the classifier about which spatial location to focus on for object classification. In this way, the classifier gets access to raw shape of the object as well as precise location. Also, for applications where the added events are undesirable, we can use other parameters such as parameter set 2 described earlier that has similar PSR f as NNb filter. Quantitatively, the average values of PSR f , PNR f , and SNR f for NNb filter are 21.54, 0.18, and ~20dB respectively while the same ones for NeuNN filter with parameter set 2 are 24.04, 0.09, and 29 dB respectively. This shows that even when NeuNN reduces signal similar to levels of NNb filter, it can produce better SNR due to stronger attenuation of noise. For reference, the average values of PSR f , PNR f , and SNR f for NeuNN with parameter set 1 are 92, 0.11, and 28 dB respectively. Though this setting can address the reduction of dilation effect it cannot ensure complete elimination of this phenomenon and in our envisioned applications it is not a significant issue. However, in applications such as event-based stereovision or event-based optical flow computation (which are principally based on small local spatiotemporal neighborhoods and precise timing) they could have a significant local effect.\n\n6.4. Discussion We have demonstrated a new noise filtering algorithm for event based imagers and shown its implementation on TrueNorth. There are some important points to note about our proposed method. First, it should be noted that since TrueNorth processes data at a millisecond clock tick, we lose the original microsecond resolution of events coming from the sensor. However, for our application of object tracking and classification in traffic monitoring or surveillance, this is acceptable and microsecond temporal resolution is unnecessary. Second, the response of a rectangular photodiode, such as the ones in ATIS, will be different for horizontally and vertically moving contours (Clady et al., 2017 ). However, for the traffic monitoring and surveillance applications we are considering, the orientation of the camera will be fixed and we presented results for this orientation. For changes of orientation, our filtering method will still work–however, the exact gains in SNR f might be different. Third, we expect the effect of adding more events in regions with signal events (spatio-temporal dilation) will be helpful in tracking under noisy conditions. However, the shape information of the object may be distorted by this filtering. We propose an architecture with two parallel processing paths in that case–the raw signal (or mildly filtered one) goes to a classifier in one path, while in the other path, we have the NeuNN filter followed by the tracker. The tracker informs the classifier about which spatial location to focus on for object classification. In this way, the classifier gets access to raw shape of the object as well as precise location. Also, for applications where the added events are undesirable, we can use other parameters such as parameter set 2 described earlier that has similar PSR f as NNb filter. Quantitatively, the average values of PSR f , PNR f , and SNR f for NNb filter are 21.54, 0.18, and ~20dB respectively while the same ones for NeuNN filter with parameter set 2 are 24.04, 0.09, and 29 dB respectively. This shows that even when NeuNN reduces signal similar to levels of NNb filter, it can produce better SNR due to stronger attenuation of noise. For reference, the average values of PSR f , PNR f , and SNR f for NeuNN with parameter set 1 are 92, 0.11, and 28 dB respectively. Though this setting can address the reduction of dilation effect it cannot ensure complete elimination of this phenomenon and in our envisioned applications it is not a significant issue. However, in applications such as event-based stereovision or event-based optical flow computation (which are principally based on small local spatiotemporal neighborhoods and precise timing) they could have a significant local effect."
} | 6,629 |
28582469 | PMC5476291 | pmc | 2,941 | {
"abstract": "Microalgae are promising microorganisms for the production of numerous molecules of interest, such as pigments, proteins or triglycerides that can be turned into biofuels. Heterotrophic or mixotrophic growth on fermentative wastes represents an interesting approach to achieving higher biomass concentrations, while reducing cost and improving the environmental footprint. Fermentative wastes generally consist of a blend of diverse molecules and it is thus crucial to understand microalgal metabolism in such conditions, where switching between substrates might occur. Metabolic modeling has proven to be an efficient tool for understanding metabolism and guiding the optimization of biomass or target molecule production. Here, we focused on the metabolism of Chlorella sorokiniana growing heterotrophically and mixotrophically on acetate and butyrate. The metabolism was represented by 172 metabolic reactions. The DRUM modeling framework with a mildly relaxed quasi-steady-state assumption was used to account for the switching between substrates and the presence of light. Nine experiments were used to calibrate the model and nine experiments for the validation. The model efficiently predicted the experimental data, including the transient behavior during heterotrophic, autotrophic, mixotrophic and diauxic growth. It shows that an accurate model of metabolism can now be constructed, even in dynamic conditions, with the presence of several carbon substrates. It also opens new perspectives for the heterotrophic and mixotrophic use of microalgae, especially for biofuel production from wastes.",
"conclusion": "Conclusions The dynamic metabolic model developed for the heterotrophic, mixotrophic and autotrophic growth of Chlorella sorokiniana on acetate and butyrate achieved a so far unequalled accuracy. The model efficiently fits the dynamic experimental data and correctly predicts the biomass yields for a broad range of experimental conditions. This new powerful simulation tool provides new insight into the mixotrophic microalgal process, and allows us to explore the different possibilities to overcome the inhibition induced by some of the substrates, in particular by adjusting the mixotrophic regimes. The model also highlights the dynamics of some internal compounds, especially under an auto- or mixotrophic regime, while light intensity is slowly affected by an increase in self-shading. As a consequence, the model shows that QSSA is not valid for mixotrophic growth as long as the light is variable in the culture medium. In the future, the model should be extended further in order to handle mixotrophic behavior under periodic light/dark cycles.",
"introduction": "Introduction Microalgae are unicellular eukaryote microorganisms that can grow autotrophically using light energy and CO 2 . Many species can also grow heterotrophically in darkness on various organic carbon sources, including glucose, or can combine heterotrophy and autotrophy for a mixotrophic growth [ 1 ]. Microalgae have been domesticated and used to synthesize many products with industrial applications, such as pharmaceutics or cosmetics (antioxidants, pigments, unsaturated long-chain fatty acids), agricultural products (food supplements, functional food, colorants) and animal feed (aquaculture, poultry or pig farming) [ 2 ]. They are also promising organisms for green chemistry (bioplastics), the environment (wastewater treatment, CO 2 mitigation), and even energy production (biodiesel, bioethanol, hydrogen) [ 2 ]. Autotrophic growth of microalgae is limited by light distribution to all the cells, constraining the cell concentration to below 10 g/l (for the thinnest and most concentrated cultivation systems). Heterotrophic growth does not have this limiting factor and higher biomass density can be achieved [ 1 ], drastically reducing the harvesting costs. In addition, heterotrophic growth is usually faster, reducing the cultivation time [ 3 ]. However, industrial production of heterotrophic microalgae is hampered by the high economic and environmental costs of glucose, commonly used as a substrate. One solution is to use the waste from other processes, such as glycerol, acetate (ACE) or butyrate (BUTYR), which represent low cost carbon substrates. For instance, dark anaerobic fermentation produces an effluent mainly composed of acetate and butyrate [ 4 ]. However, some substrates in waste, such as butyrate, can be inhibitory [ 5 ]. Moreover, the successive metabolic switches between different substrates are not well understood and are likely to significantly affect growth. Therefore, this bioprocess still needs to be mastered and optimized to produce microalgae and extract the targeted byproducts on an industrial scale and at a competitive price, with consistent quality and in a sustainable way. In this context, mathematical modeling of the metabolism has proven to be an efficient tool for optimizing growth and increasing the production of target molecules. To date, no models exist for heterotrophic microalgal metabolism dynamically switching between several substrates ( S1 Table ), including mixotrophic growth in light. So far, only static fluxes have been predicted under constant substrate consumption [ 6 – 9 ]. Representing the dynamic shifts for a blend of substrates typical of real wastewater is a major challenge, since some intracellular accumulation might occur, either during the transition between substrates, or due to the varying nature of the light. As a consequence, the quasi-steady state assumption (QSSA) required by most of the existing metabolic approaches may be an invalid hypothesis in this case [ 10 ]. The DRUM modeling framework recently proposed in [ 10 ] was used here to handle the non quasi-steady state (QSS). It allowed the development of a dynamic metabolic model for Chlorella sorokiniana grown on a single-substrate culture and a mixed-substrate (acetate and butyrate) culture, combined with various combinations of light. The model is thus designed to represent autotrophic, heterotrophic or mixotrophic modes under diauxic conditions. Our purpose is to propose a relatively generic model, instantiated and calibrated for C . sorokiniana . According to Baroukh et al. [ 11 ], such a generic model should be applicable to a wide range of microalga species.",
"discussion": "Results/Discussion Macroscopic scale simulation The model simulation accurately reproduces experimental data, even for the validation data sets that were not used for calibration (Figs 2 – 4 ). The diauxic growth is particularly well represented (Figs 3 and 4D ), and the transient behavior, together with the final biomass, is correctly predicted (Figs 2 – 4 ), showing that the biomass yields obtained from the metabolic network are accurate. Indeed, one of the advantages of metabolic modeling [ 20 ] is the prediction of biomass yields supported by the stoichiometry of the metabolic network. Here, the predicted conversion yield of acetate and butyrate to biomass is 0.514 grams of carbon biomass per gram of carbon in the incoming substrate. This yield contributes to correct prediction of the biomass for both acetate ( Fig 2B ) and butyrate ( Fig 2D ), thus validating the approach. Interestingly, the yields are identical between the two substrates. A possible explanation is the fact that more ATP is required for the transport of butyrate into the cell than for acetate, thus balancing the ATP created when converting butyrate and acetate to succinate. 10.1371/journal.pcbi.1005590.g004 Fig 4 Comparison between the model and experimental data for Chlorella sorokiniana mixotrophic and autotrophic growth. Simulations are represented by full lines (conditions used for calibration). Experimental results are represented by large dots, diamonds or triangles. Red: acetate; blue: butyrate; yellow: biomass. A. Autotrophic growth. B. Mixotrophic growth with 0.3 gC.L -1 acetate C. Mixotrophic growth with 0.3 gC.L -1 butyrate. D. Mixotrophic growth with 0.3 gC.L -1 acetate and 0.3 gC.L -1 butyrate. Only one of the experimental triplicates is represented here. The simulations for all triplicates are available in S7 Fig . The set of kinetic parameters matches both the single-substrate culture and the mixed-substrate culture. This implies that butyrate has no impact on acetate growth rate. However, the inverse is not true, since the acetate concentration at which butyrate consumption starts ( k D ) is very low (5.39*10 −10 M), illustrating the strong diauxic growth that occurs. Even the smallest amount of acetate inhibited butyrate uptake. The maximum acetate uptake rate was higher than the maximum butyrate uptake rate by nearly 15 fold, reflecting the preference of Chlorella sorokiniana for acetate. The non-inhibiting butyrate concentration ( Sopt MR 2 ) was very low (1.93*10 −5 M), which highlights the strong inhibition of butyrate in the medium on its uptake. It also explains why, in the butyrate-only experiments, no biomass growth was observed for butyrate concentrations above 0.1 g.L -1 ( Fig 2D ). In addition to substrates and biomass concentrations, light evolution inside the culture vessel was computed. During the first few days, the average light intensity decreases until equilibrium is reached around 16 μE.m -2 .s -1 ( S1 File section 4, S8 Fig ). It represents 11.7% of the incident light and is in agreement with the literature [ 21 ]. Interestingly, equilibrium is reached faster for mixotrophic growth, particularly on acetate, which supports fast heterotrophic growth ( S8 Fig ). In addition, the photosynthetic quotient for autotrophic growth varies between 1.0 and 1.16, matching the typical range of 1.0–1.8 for microalgae [ 6 ]. Intracellular scale simulation The predicted metabolic fluxes ( Fig 5 , S9 Fig ) are in accordance with previous studies [ 11 ]. Autotrophy ( S9C Fig ) is characterized by high fluxes in the photosynthetic pathways, which convert light and CO 2 to GAP. Beyond these pathways, fluxes drop considerably in terms of absolute magnitude. Upper glycolysis is in the gluconeogenic direction to produce the carbohydrate and sugar precursor metabolites (Glucose 6-phosphate (G6P), Ribose 5-phosphate (R5P), Erythrose 4-phosphate (E4P)) necessary for growth. In the heterotrophic mode, fluxes are more homogenous among reactions ( Fig 5B , S9A Fig ). Acetate and butyrate are converted to acetyl-CoA in the glyoxysome ( Fig 5B , S9D Fig ). Acetyl-CoA is then converted into succinate by the glyoxylate cycle and injected in the TCA cycle. Upper glycolysis also goes in the gluconeogenic direction to produce carbohydrate and sugar precursors. This can be achieved thanks to the anaplerotic reactions that convert oxaloacetate to phosphoenolpyruvate (PEP). Mixotrophy is a mixed combination of the autotrophic and heterotrophic modes ( Fig 5A , S9A Fig ). For mixotrophic growth on acetate, heterotrophic metabolism is dominant, whereas autotrophic metabolism is dominant for mixotrophic growth on butyrate. This is due to the fact that autotrophic growth is slower than growth on acetate but faster than growth on butyrate. 10.1371/journal.pcbi.1005590.g005 Fig 5 Flux maps for mixotrophic and heterotrophic growth of Chlorella sorokiniana on butyrate. Fluxes are normalized by unit of biomass. Dashed arrows indicate fluxes related to biomass formation. Metabolic fluxes vary greatly according to substrates and growth modes. The scale for converting metabolic fluxes into arrows width is presented for each case. A. Mixotrophic growth on 0.3 g.L -1 butyrate. Flux maps computed at time = 5.0 days. B. Heterotrophic growth on 0.1g.L -1 butyrate. Flux maps computed at time = 8.1 days. Flux maps on other substrates are available in S9 Fig . Avoiding the inhibitory effect of butyrate Interestingly, in agreement with the data, the model did not predict any growth on butyrate above 0.1 gC.L -1 , and at the same time successfully forecasted growth on 0.9 gC.L -1 butyrate in mixed substrate conditions ( Fig 3E ) and on 0.3 gC.L -1 butyrate in mixotrophic conditions. Indeed, in these conditions, the first-stage growth on acetate and/or light produces enough biomass to finally consume such an inhibiting quantity of butyrate. The substrate to biomass (S/X) ratio is known to be a key process parameter for overcoming the inhibitory effects of the substrate [ 22 ]. The model therefore represents a tool to compute and optimize the amount of co-substrate that must be added to overcome the inhibition and consume the butyrate. Different strategies could be tested to achieve a low S/X ratio and accelerate butyrate consumption. The simplest approach would involve adding a non-inhibiting substrate in order to reduce the amount of inhibitory substrate per unit biomass. For example, the addition of 0.5 gC.L -1 of acetate for a volume equal to half of the culture volume has been found to eventually lead to the consumption of 0.5 gC.L -1 of butyrate in 14 days ( Fig 6B ), which would not have been possible otherwise ( Fig 6A ). However, in general, such pure substrate is not available. We therefore simulated the addition of a mix of acetate and butyrate in proportions that are representative of fermentative digestate [ 4 ], for a volume equal to half of the culture volume. On the one hand, the acetate contained in the waste stimulated growth, but since it is associated with addition of butyrate, it also increased inhibition. Simulations show that the inhibition is overcome, but does not lead to the total consumption of butyrate within 15 days ( Fig 6D ). Furthermore, the mixotrophic potential can be exploited: autotrophic growth can be enhanced by illumination in order to ultimately dilute the inhibitory substrates. Illuminating the algae at an incident intensity of 136 μE.m -2 .s -1 leads to the consumption of the same quantity of butyrate in 13 days, and this delay can be reduced to 9 days using a light intensity of 272 μE.m -2 .s -1 ( Fig 6C ). Finally, if light is provided at the same time as the addition of fermentative digestate (for a volume equal to half of the culture volume), inhibition can be overcome after 10 days ( Fig 6D ). 10.1371/journal.pcbi.1005590.g006 Fig 6 Disinhibition of butyrate by addition of acetate, light and a mix of acetate/butyrate due to the biomass effect. A larger biomass implies a decrease in butyrate inhibition on growth. Biomass can be increased by addition of acetate (B), light (C) and/or a mix of acetate and butyrate (D). Red: acetate; blue: butyrate; yellow: biomass. A. Normal conditions, without any additions. B. Addition of acetate (0.5 gC.L -1 , volume half of culture volume) in the medium at day 5. C: Addition of light at day 5. Full lines: incident light intensity of 136 μE.m -2 .s -1 . Dashed line: incident light intensity of 272 μE.m -2 .s -1 . D. Addition of a mix of acetate (0.25 gC.L -1 ) and butyrate (0.45 gC.L -1 ) (volume half of culture volume) representative of a fermentative waste (4). Full lines: without light. Dashed lines: with light. Assessing the time constants of the metabolism The advantage of the DRUM approach is its ability to account for the accumulation of some intracellular metabolites and thus to characterize the time to reach steady state. It can also determine more quantitatively the time scales of flux variations in the cell than earlier frameworks. This analysis was applied to SUC and GAP, which are, in our model, the intermediate accumulating metabolites. Interestingly, SUC actually hardly accumulates in the simulations and rapidly achieves a QSS ( S10 Fig ) where its concentration evolves slowly compared to the other variables in the system (substrate consumption, biomass formation). We developed an algorithm to automatically detect the time needed to reach QSS ( t QSS ). In the experimental conditions of this study, approximately 3 minutes were necessary for succinate to achieve QSS ( S2 File ) thanks to a higher biomass synthesis rate (via a high k MR 4 ) compared to the substrate assimilation rate, implying that succinate is immediately consumed once it is synthesized from butyrate or acetate. A sensitivity analysis on the parameter k MR 4 revealed that the confidence interval of t QSS was [0.6; 34] minutes (model error less than 5% of the minimal error) ( S1 File section 8, S2 File ). After the brief transient succinate step, the QSSA for heterotrophic growth on butyrate and acetate is valid. Therefore, the macroscopic model can be reduced further, by merging reaction MR4 with reactions MR1 and MR2 ( S1 File section 6). The same kinetic parameters can be used for simulation, and the fit is nearly identical (increase of 0.6% of the error). As a consequence, results considering QSSA are very close to the ones based on DRUM. GAP, in contrast to SUC, does not reach a QSS rapidly ( S11 Fig ). First, GAP accumulates at high light intensities, reaching a maximum when average light intensity is approximately 60 μE.m -2 .s -1 . Then, it is consumed at low light intensities, reaching a QSS when average light intensity reaches a steady state at 16 μE.m -2 .s -1 ( S1 File section 8). This suggests that microalgal metabolism in autotrophic and mixotrophic modes only reaches a QSS when average light is constant in the culture media, meaning that growth has ceased. This behavior is similar to that of microalgae grown in day/night cycles [ 10 , 11 ], involving accumulation of carbon-reserve metabolites (carbohydrates, lipids) during the day, when the light is intense enough, and re-consumption during the night or at the beginning and end of the day, when light intensity is low. Here, the carbon reserve metabolite is GAP, because only GAP accumulated in the model. Nevertheless, it is probable that carbohydrates and/or lipids also accumulate. Further experiments are required to validate these results more extensively and to determine which carbon-only metabolite is stored inside the cell. To confirm these results, a Macroscopic Bioreaction Model of the system [ 23 ], relying on the QSSA assumption, was developed (see S1 File section 10 for details on the methodology). Without accumulation of SUC, the model error was almost unchanged (0.06% increase of the error). But without the possibility for GAP to accumulate, a 40% increase in the error is observed. This confirms our finding that GAP do accumulate inside the cell at high light intensities to be consumed later at lower light intensities. It is also interesting to note that the MBM approach is sufficient and produces accurate results, for applications in heterotrophy only cultures, without the need for accumulating metabolites. Conclusions The dynamic metabolic model developed for the heterotrophic, mixotrophic and autotrophic growth of Chlorella sorokiniana on acetate and butyrate achieved a so far unequalled accuracy. The model efficiently fits the dynamic experimental data and correctly predicts the biomass yields for a broad range of experimental conditions. This new powerful simulation tool provides new insight into the mixotrophic microalgal process, and allows us to explore the different possibilities to overcome the inhibition induced by some of the substrates, in particular by adjusting the mixotrophic regimes. The model also highlights the dynamics of some internal compounds, especially under an auto- or mixotrophic regime, while light intensity is slowly affected by an increase in self-shading. As a consequence, the model shows that QSSA is not valid for mixotrophic growth as long as the light is variable in the culture medium. In the future, the model should be extended further in order to handle mixotrophic behavior under periodic light/dark cycles."
} | 4,933 |
34708512 | PMC9298193 | pmc | 2,942 | {
"abstract": "Summary \n Terpios hoshinota is an aggressive, space‐competing sponge that kills various stony corals. Outbreaks of this species have led to intense damage to coral reefs in many locations. Here, the first large‐scale 16S rRNA gene survey across three oceans revealed that bacteria related to the taxa Prochloron , Endozoicomonas , SAR116, Ruegeria , and unclassified Proteobacteria were prevalent in T . hoshinota . A Prochloron ‐related bacterium was the most dominant and prevalent cyanobacterium in T . hoshinota . The complete genome of this uncultivated cyanobacterium and pigment analysis demonstrated that it has phycobiliproteins and lacks chlorophyll b , which is inconsistent with the definition of Prochloron . Furthermore, the cyanobacterium was phylogenetically distinct from Prochloron , strongly suggesting that it should be a sister taxon to Prochloron . Therefore, we proposed this symbiotic cyanobacterium as a novel species under the new genus Candidatus Paraprochloron terpiosi. Comparative genomic analyses revealed that ‘Paraprochloron’ and Prochloron exhibit distinct genomic features and DNA replication machinery. We also characterized the metabolic potentials of ‘Paraprochloron terpiosi’ in carbon and nitrogen cycling and propose a model for interactions between it and T. hoshinota . This study builds a foundation for the study of the T . hoshinota microbiome and paves the way for better understanding of ecosystems involving this coral‐killing sponge.",
"conclusion": "Conclusions This study makes several discoveries about bacterial communities associated with T. hoshinota . First, the study showed that although bacterial communities are governed by biogeography, four ASVs were prevalent in T. hoshinota and formed the core microbiome. The core microbiome of T. hoshinota includes a Prochloron‐ like bacterium, Endozoicomonas , SAR116, Ruegeria , and other unclassified Proteobacteria . Second, we found that the Prochloron‐ like bacterium was the predominant cyanobacterium in cyanobacteriosponge T. hoshinota and was present in all the samples and accounted for 30% relative read abundance on average, suggesting that this particular cyanobacterium is a potential obligate symbiont of the sponge. By genomic, phylogenetic, and pigment analyses, we found that LD05 and SP5CPC1 formed a sister clade adjacent to Prochloron , and they lack chlorophyll b and have phycobilins, which is inconsistent with the definition of Prochloron . Hence, we propose the new genus ‘Paraprochloron’ to accommodate this clade. Third, we demonstrated that ‘Paraprochloron’ has genomic features distinct from those of Prochloron . ‘Paraprochloron’ has much smaller genomes, higher coding gene densities, and uses dnaA‐independent DNA replication. In summary, our research will help future studies explore the detailed ecosystem inside the holobiont, and the complete genome reconstruction of Ca . Pp. terpiosi LD05 will help to extend our knowledge of cyanobacteria evolution and the functional diversity of symbiotic cyanobacteria.",
"introduction": "Introduction The coral‐killing sponge Terpios hoshinota has received attention since outbreaks were first detected in several coral reef regions in the western Pacific Ocean, South China Sea, and Indian Ocean (Rützler and Muzik, 1993 ; Liao et al ., 2007 ; Fujii et al ., 2011 ; de Voogd et al ., 2013 ; Hoeksema et al ., 2014 ; Montano et al ., 2015 ; Thinesh et al ., 2015 ). This sponge grows up to 23 mm −1 month −1 (Plucerrosario, 1987 ), and its fast‐growing and competitive nature enables it to kill scleractinian corals rapidly and at a high rate (30%–80% mortality) across biogeographic regions (de Voogd et al ., 2013 ; Hoeksema et al ., 2014 ; Montano et al ., 2015 ; Yomogida et al ., 2017 ). For instance, T . hoshinota overgrowth jeopardizes coral reefs in numerous regions of Taiwan, Indonesia, Malaysia, Japan, India, and the Great Barrier Reef (Fujii et al ., 2011 ; Shi et al ., 2012 ; Madduppa et al ., 2017 ; Yomogida et al ., 2017 ). The gradual spread of T. hoshinota poses a serious threat to coral reefs. However, little is known regarding the causes of such outbreaks. Sponges are commonly known to harbour complex microbial communities, and symbiotic microorganisms play vital roles in the development, health, and nutrient acquisition of their hosts (Hentschel et al ., 2012 ). The microbes and hosts form an ecological unit referred to as a holobiont. In T. hoshinota , the sponge is associated with a bacterial community of relatively low diversity that is dominated by cyanobacteria (Tang et al ., 2011 ). Ultrastructural observations have clearly shown that cyanobacteria are densely distributed in T. hoshinota , contributing to 50% of the total cellular volume (Rützler and Muzik, 1993 ; Hirose and Murakami, 2011 ; Tang et al ., 2011 ), and the blackish colour of T. hoshinota has been attributed to cyanobacteria (Tang et al ., 2011 ). Accordingly, the sponge has been called ‘cyanobacteriosponge' (Rützler and Muzik, 1993 ). Several studies have shown that cyanobacteria play important roles in the growth of T. hoshinota and in competition with corals. For instance, a high number of cyanobacteria can be observed in the larvae of T. hoshinota (Wang et al ., 2012a ; Hsu et al ., 2013 ), suggesting that they are transmitted vertically during embryogenesis in this particular sponge (Nozawa et al ., 2016 ). Second, in situ short‐term shading can cause a long‐term decrease in the biomass of symbiotic cyanobacteria and lead to irreversible damage to T. hoshinota , arresting its expansion (Soong et al ., 2009 ; Thinesh et al ., 2017 ). Third, when T. hoshinota encounters certain corals, the sponge forms a hairy tip structure packed with dense cyanobacteria to interact with corals (Wang et al ., 2012b ). These results indicate that T. hoshinota ‐associated cyanobacteria are vital for the overgrowth and rapid destruction of coral reef ecosystems. Although important, the identity and classification of dominant symbiotic cyanobacteria in T. hoshinota are still unclear. In 2015, Yu et al . isolated and cultivated Myxosarcina sp. GI1, a baeocytous cyanobacterium, from T. hoshinota at Green Island (Yu et al ., 2015 ). However, electron microscopy did not identify vegetative cell aggregates of baeocytes, a type of reproductive cell, in T. hoshinota (Hirose and Murakami, 2011 ; Tang et al ., 2011 ; Wang et al ., 2012a ). Moreover, our previous analyses of 16S rRNA gene sequences in T. hoshinota samples from Green Island revealed that the dominant cyanobacterium in T. hoshinota is closely related to Prochloron (Tang et al ., 2011 ). Prochloron , a genus comprised of a single species, is an obligate symbiont of certain ascidians; the hallmarks and definition of this genus are the presence of chlorophyll a and b and the lack of phycobilins, which are unusual in cyanobacteria (Whatley, 1977 ). However, pigment analysis identified that the cyanobacteria in T. hoshinota contain phycobilins (Hirose and Murakami, 2011 ). Hence, the characteristics and classification of the predominant cyanobacterium in T. hoshinota remain uncertain. Moreover, whether the predominant cyanobacterium is the same in all of the Indo‐Pacific regions requires validation. Finally, since cyanobacteria are attributed to the health and invasive capacity of T. hoshinota , ecological relationships and molecular interactions involving T. hoshinota and its cyanobacteria need to be determined. Other T. hoshinota ‐associated bacteria may also contribute to holobiont function. The microbial community in a sponge can be shaped by host‐related factors, such as the immune system and nutrient exchange (Pita et al ., 2013 ; Webster and Thomas, 2016 ; Pita et al ., 2018 ), or be determined by environmental factors, such as light availability, pH, and temperature (Webster and Thomas, 2016 ). Nevertheless, no study has investigated the microbiome of T. hoshinota from different biogeographical backgrounds. Hence, in this study, we conducted a 16S ribosomal RNA (rRNA) gene survey to investigate the microbial community structures and diversity of T. hoshinota samples from a wide geographical range across the western Pacific Ocean, South China Sea, and Indian Ocean. Knowing that the predominance of a cyanobacterium was ubiquitous, the complete genome of this uncultivated cyanobacterium was reconstructed by whole‐genome shotgun sequencing using Nanopore and Illumina platforms. Genomic and comparative genomic analyses elucidated the phylogenetic affiliation and taxonomy of the dominant symbiotic bacterium in T. hoshinota and identified putative symbiotic interactions between the bacterium and the host.",
"discussion": "Discussion \n T. hoshinota is one of the most important biological threats to corals in the Indo‐Pacific region. Its association with cyanobacteria and other symbiotic bacteria is essential for maintaining the function of T. hoshinota . Our study is the first to explore the composition and role of the T. hoshinota microbiome. An unprecedented large‐scale survey of the bacterial community of T . hoshinota based on 16S rRNA gene amplicon sequencing of samples from various regions across the western Pacific Ocean, Indian Ocean, and South China Sea enabled us to characterize the T . hoshinota microbiome. Biogeographical variation in the T . hoshinota microbiome and its core microbial members Certain sponges exhibit high microbiome stability due to strict selective pressures exerted by their hosts. For instance, Ircinia and Hexadella sponges exhibit host‐specificity and stability in their associated bacterial communities, despite large geographic distances between sampling sites (Pita et al ., 2013 ; Reveillaud et al ., 2014 ). In contrast, certain sponges, such as Petrosia ficiformis , harbour biogeography‐dependent bacterial compositions (Burgsdorf et al ., 2014 ). Our survey of T . hoshinota from three oceans enabled us to determine whether bacterial communities of T . hoshinota vary across different biogeographical backgrounds or experience strong selective pressure from their hosts. The dissimilarity analysis of the microbial community structures of T. hoshinota showed a correlation between sample sites and microbial community structure (Fig. 1 ). The differences in microbial community structures among T. hoshinota samples from different locations may be the result of local acclimatization. Samples from the same regions may represent holobionts with similar metabolic functions to cope with stress or increased fitness in certain environments. However, despite biogeographical variation, we still identified core bacterial members that were present in all samples (Table 1 and Fig. S2 ). This core bacterial community comprised only a few genera but accounted for around 80% of relative abundance across all samples, indicating that it is vital to the holobiont of T . hoshinota . Host‐related factors may help keep this core microbiome stable because its members carry out core functions of the holobiont (Pita et al ., 2018 ). Such core members may have the metabolic capability to utilize nutrients from the sponge host environment and play important roles in nutrient exchange, such as sulfur, carbon, and nitrogen cycling. This core group may also be responsible for defence against predators and for protecting sponge symbionts from toxins and pathogens (Pita et al ., 2018 ). The most dominant ASV in the core microbiome is closely related to Prochloron , a genus of symbiotic cyanobacteria found in various ascidians (Whatley, 1977 ; Kuhl et al ., 2012 ). The biogeographically independent prevalence and predominance of this Prochloron ‐related bacterium in T. hoshinota suggests that it plays a crucial role in the holobiont. On the other hand, symbiotic cyanobacteria in sponges can exhibit host specificity (Thacker and Starnes, 2003 ) or co‐exist with various sponges (Konstantinou et al ., 2018 ). The Prochloron‐ related ASV in T. hoshinota has not been observed in other organisms, implying that this Prochloron‐ related cyanobacterium may have high host specificity. \n Candidatus Paraprochloron terpiosi gen. nov., sp. nov., a Prochloron ‐related bacterium prevalent in T. hoshinota \n LD05 and SP5CPC1, two Prochloron ‐related bacteria in sponges, are distinct from Prochloron in terms of pigment content, phylogenetic divergence, and genomic features. Moreover, the pigment features of these bacteria are inconsistent with the definition of Prochloron . Hence, a new genus, ‘Paraprochloron’, is herein proposed to distinguish between these two groups. \n Candidatus Paraprochloron terpiosi LD05 was the dominant cyanobacterium in T . hoshinota specimens collected across three different oceans (Supporting Information Table S2 ). Why the dominant cyanobacterium species remain identical in T. hoshinota across different oceans is unclear. Recent studies suggest that T. hoshinota larvae, which carry vertically transmitted cyanobacteria, may have short dispersal distances because they are denser than the water, leading them to settle rapidly after leaving their mother sponge (Wang et al ., 2012a ; Hsu et al ., 2013 ; Nozawa et al ., 2016 ). Under these circumstances, the symbiotic Ca . Pp. terpiosi LD05 from various locations might accumulate genetic differences to adapt to local environments. The absence of evident speciation in our study may be the result of tight and stable symbiotic interactions between Ca . Pp. terpiosi LD05 and T. hoshinota , which restrict genetic changes in the bacterium. Another possibility is that the species becomes broadly dispersed through other mechanisms, such as ocean currents or transportation via vessels, which would enable T. hoshinota larvae to spread quickly across oceans with few genetic alterations. Evidence of recent T. hoshinota outbreaks supports the latter scenario (Liao et al ., 2007 ). Comparison between ‘Paraprochloron’ and Prochloron \n Comparative genomics involving ‘Paraprochloron’ and Prochloron enables us to infer their evolutionary histories and respective relationships with their hosts. Previous studies have shown that symbiotic bacteria usually have reduced genomes because certain genes erode as symbiosis develops (Gao et al ., 2014 ; Lo et al ., 2016 ). Furthermore, a model of symbiont evolution has proposed that during the evolutionary history of symbiosis, large‐scale pseudogenization can occur during transitional events, such as strict host association or vertical transmission, leading to a sudden decrease in coding density (Lo et al ., 2016 ). Eventually, the coding density gradually bounces back owing to deletion bias. ‘Paraprochloron’ had a comparatively smaller genome, while Prochloron had lower coding density compared to other phylogenetically closely related cyanobacteria (Supporting Information Table S3 ). These distinct genomic features between Prochloron and ‘Paraprochloron’ may indicate that the transition toward host‐restricted lifestyles occurred more recently in Prochloron than in ‘Paraprochloron’. This hypothesis may be supported by our observation that Prochloron harboured more transposase genes as the elevation of mobile genetic element quantities, such as transposons and insertion elements, which is thought to be associated with recent transition to a host‐restricted symbiotic lifestyle (Moran et al ., 2008 ). Another hallmark of Prochloron is its ability to produce patellamides and cytotoxic cyclic peptides. However, a gene cluster for the synthesis of patellamides was not found in Ca . Pp. terpiosi LD05 or SP5CPC1. In contrast, the LD05 and SP5CPC1 genomes have four terpene synthesis gene clusters, whereas Prochloron genomes harbour only two such gene clusters. Moreover, LD05 and SP5CPC1 genomes contain genes encoding terpene cyclases, enzymes that catalyze the cyclization of linear terpenes. This indicates that LD05 and SP5CPC1 have cyclic terpenes. Taken together, ‘Paraprochloron’ and Prochloron may produce distinct secondary metabolites to increase the fitness of themselves or of their entire holobiont. Certain cyanobacteria contain genes involved in sucrose metabolism. Although the role of sucrose in cyanobacteria remains understudied, several studies have shown that sucrose can be utilized as a compatible solute, serve as a signal molecule, or be employed for glycogen synthesis (Blumwald and Telor, 1982 ; Desplats et al ., 2005 ; Curatti et al ., 2008 ; Kolman et al ., 2015 ). Interestingly, our genomic analysis revealed that the genes involved in sucrose metabolism were present in Ca . Pp. terpiosi LD05 and SP5CPC1 but absent in Prochloron (Supporting Information Table S4 ). We hypothesize that ‘Paraprochloron’ can use sucrose as an osmolyte to cope with osmotic stress; this is supported by a previous finding that the genes related to osmotic stress are found in sponge‐associated bacterial genomes (Webster and Thomas, 2016 ). On the other hand, we also found that Ca . Pp. terpiosi LD05 and SP5CPC1 contained osmoprotectant transporter genes, which were not identified in Prochloron genomes (Supporting Information Table S4 ). These results indicate that the ‘Paraprochloron’ species may live in environments with higher osmotic stress, such as locations with high osmolarity or fluctuations in osmolarity, compared to Prochloron . Alternatively, ‘Paraprochloron’ and Prochloron may use different strategies to deal with osmotic stress. Another possible role of sucrose is that ‘Paraprochloron’ may provide cells of sponge hosts or other symbiotic bacteria with sucrose as an energy or carbon source to promote the growth of sponge holobionts. Another observation that drew our attention was the absence of the dnaA gene in Ca . Pp. terpiosi LD05 and SP5CPC1. DnaA is required for the initiation of DNA replication at oriC (Katayama et al ., 2010 ). Some bacterial symbionts do not possess the dnaA gene and have multiple copies of the same genome in a single cell (Akman et al ., 2002 ; Ran et al ., 2010 ; Ohbayashi et al ., 2016 ). The evolution of DnaA‐independent replication has been most extensively studied in cyanobacteria. A study showed that cyanobacteria lost DnaA dependency before becoming symbiotic bacteria, and such loss can drive free‐living bacteria to become symbiotic (Ohbayashi et al ., 2016 ; Ohbayashi et al ., 2020 ). However, in our analyses, Prochloron had dnaA , but ‘Paraprochloron’ did not, indicating the loss of dnaA occurred after the two bacteria separated from a common ancestor. This implies that symbiosis may also drive bacteria to lose dnaA . Symbiotic interactions between Ca . Pp. terpiosi and T. hoshinota \n The predominance of Ca . Pp. terpiosi LD05 highlights its role in T. hoshinota . Many sponges harbour photosynthetic symbionts that provide their host with nutrients (Erwin and Thacker, 2007 ; Thacker et al ., 2007 ). Some sponges acquire >50% of their energy demand from symbiotic cyanobacteria in the form of photosynthates (Wilkinson, 1983 ; Taylor et al ., 2007 ; Usher, 2008 ). Photosynthesis was observed in the T. hoshinota holobiont, and its efficiency increased when the coral‐killing holobiont came into contact with coral, which may help sponges overgrow corals (Wang et al ., 2015 ). As the dominant cyanobacteria, Ca . Pp. terpiosi LD05 may be an important carbon source for T. hoshinota and may facilitate competition with corals. Certain symbiotic cyanobacteria in sponges can fix nitrogen (Wilkinson and Fay, 1979 ). Ca . Pp. terpiosi LD05 does not harbour genes related to nitrogen fixation, but genes related to ammonium transporters and the GS‐GOGAT pathway were identified in Ca . Pp. terpiosi LD05, indicating that it could recycle ammonium from the cells of T . hoshinota in the mesohyl. On the other hand, Ca . Pp. terpiosi LD05 harbours urea transporter and urease genes. Therefore, urea can be used as an alternative nitrogen source. A previous study showed that levels of free amino acids in T. hoshinota ‐inhabiting cyanobacteria were elevated when the holobiont encountered a coral (Wang et al ., 2015 ). Our finding of an amino acid transporter in Ca . Pp. Terpiosi LD05 suggests that the bacterium may benefit from ‘coral killing’ by consuming amino acids or ammonium that are released as coral colonies disintegrate. Animals cannot synthesize essential vitamins, so symbiotic microorganisms are thought to be important sources of essential vitamins for sponges. Our analysis of the Ca . Pp. terpiosi LD05 genome identified biosynthetic pathways for vitamin B 1 , vitamin B 7 , and vitamin B 12 . Thus, Ca . Pp. terpiosi LD05 may help maintain T. hoshinota health by providing the holobiont with vitamins. One of the leading questions in the study of T. hoshinota is how it kills corals. Several mechanisms have been proposed to explain this. One argument is that the sponge produces cytotoxic allelochemicals that damage coral cells (Bryan, 1973 ). Another suggests that the sponge overgrows corals and competes with them for nutrients (Wang et al ., 2012b ). These hypotheses are not mutually exclusive. Several cytotoxic compounds, including nakiterpiosin, nakiterpiosinon, and terpiodiene, have been isolated from T. hoshinota holobionts (Teruya et al ., 2002 ; Teruya et al ., 2004 ). Nakiterpiosin and nakiterpiosinon are C‐nor‐D‐homosteroids. Previous studies have shown that cyanobacteria can produce sterols by cyclization of squalene, a triterpene (Fagundes et al ., 2019 ). In the Ca . Pp. terpiosi LD05 genome, biosynthetic pathways for squalene and squalene cyclases were identified. Hence, Ca . Pp. terpiosi LD05 may be responsible for toxin production. These toxins may facilitate the overgrowth of corals by damaging coral tissues directly or by weakening coral defences. In the future, the products of these gene clusters may be confirmed using molecular approaches."
} | 5,552 |
18416814 | PMC2375130 | pmc | 2,943 | {
"abstract": "Background Compared to other omics techniques, quantitative metabolomics is still at its infancy. Complex sample preparation and analytical procedures render exact quantification extremely difficult. Furthermore, not only the actual measurement but also the subsequent interpretation of quantitative metabolome data to obtain mechanistic insights is still lacking behind the current expectations. Recently, the method of network-embedded thermodynamic (NET) analysis was introduced to address some of these open issues. Building upon principles of thermodynamics, this method allows for a quality check of measured metabolite concentrations and enables to spot metabolic reactions where active regulation potentially controls metabolic flux. So far, however, widespread application of NET analysis in metabolomics labs was hindered by the absence of suitable software. Results We have developed in Matlab a generalized software called 'anNET' that affords a user-friendly implementation of the NET analysis algorithm. anNET supports the analysis of any metabolic network for which a stoichiometric model can be compiled. The model size can span from a single reaction to a complete genome-wide network reconstruction including compartments. anNET can (i) test quantitative data sets for thermodynamic consistency, (ii) predict metabolite concentrations beyond the actually measured data, (iii) identify putative sites of active regulation in the metabolic reaction network, and (iv) help in localizing errors in data sets that were found to be thermodynamically infeasible. We demonstrate the application of anNET with three published Escherichia coli metabolome data sets. Conclusion Our user-friendly and generalized implementation of the NET analysis method in the software anNET allows users to rapidly integrate quantitative metabolome data obtained from virtually any organism. We envision that use of anNET in labs working on quantitative metabolomics will provide the systems biology and metabolic engineering communities with a mean to proof the quality of metabolome data sets and with all further benefits of the NET analysis approach.",
"conclusion": "Conclusion anNET is the first tool publicly available for network-embedded thermodynamic analysis of metabolome data. The most immediate application of anNET is the consistency check of quantitative metabolome measurements [ 11 ]. As outlined in several recent papers [ 8 , 26 ], reliable quantification of intracellular metabolites is still extremely challenging. Thus, anNET can help here. In this context, however, it is important to note that thermodynamic feasibility approved by NET analysis is not a sufficient condition to certify that the measured concentrations reflect the true state of a cell. Nevertheless, despite the rather conservative quality filter that is given by NET analysis, a previous study showed that out of seven published metabolite datasets, three were thermodynamically not consistent [ 11 ]. A data set that fails to be thermodynamically consistent must be carefully checked before it is used for further analyses that rely on quantitative information. To this respect it is important to stress that in an unfeasible system not only the experimental data should be questioned, but also the respective input data (i.e. assumed reaction directions, thermodynamic data) as well as the inherently underlying assumptions (i.e. well-mixed compartments). The prerequisites for a consistency check by NET analysis is that (i) quantitative metabolomics data is available (although relative amounts in form of concentration ratios can also be integrated by anNET); and (ii) flux directions can be defined. Hence, this precludes the application of NET analysis to the consistency check of for example serum metabolome, or to cells grown in rich media were flux directions are uncertain. We hope that anNET will soon be used for quality check of quantitative metabolome data and thus, in consequence, the quality of published quantitative metabolite data sets will rise.",
"discussion": "Results and discussion Validation of the implementation To ascertain the correct NET analysis implementation in anNET, we analyzed the E. coli dataset published by Schaub et al. [ 20 ] with anNET using the iJR904 model [ 21 ] (see Additional file 1 ). The results obtained with anNET were compared to the published results that were obtained independently with the NET analysis implementation based on the non-generalized code [ 11 ]. The original model of 923 reactions and 762 metabolites was reduced to a core model with 166 reactions and 147 metabolites after the available thermodynamic information was propagated. The data set from Schaub et al. consisted of 6 metabolite concentrations and 4 sums of concentrations that resulted from not fully analytically resolved analytes. Further 3 ratios were added to the system to assess the feasible range of the adenylate energy charge (AEC) and the redox state of the cofactors NAD + /NADH and NADP + /NADPH. The input concentration ranges of these three ratios were chosen very wide to avoid that they become active constraints. Notably, the two analyses delivered equivalent results for all ranges of concentrations and of Δ r G (see Additional file 2 ). Minor variations are caused by the fact that in the previously published NET analysis an uncertainty for all Δ f G° of ± 0.5 kJ/mol was employed to account for possible errors in the thermodynamic input variables. Comparison between solvers Two different non-linear solvers can be used by anNET for the optimization, i.e. the LINDO API library, which relies on the CONOPT3 algorithm, or the fmincon function from the Matlab Optimization Toolbox. Independent NET analyses of the aforementioned E. coli dataset with the two solvers delivered identical results for metabolite concentration and Δ r G estimates (see Additional file 3 ), thus validating the robustness of the solution. However, it should be noted that fmincon occasionally failed to minimize/maximize the value of concentration ratios. For example, in the data set from Schaub et al. , the ranges for the summation constraints and the adenylate energy charge (i.e. resembling a ratio) were estimated in agreement with LINDO, while fmincon underestimated the feasible ratios between NADP/NADPH and NAD/NADH. Despite several modifications in the optimization settings, including the starting point and maximum duration, we were not able to find a universal configuration that lead to robust optimization of non-linear terms with fmincon . Furthermore, the LINDO solver consistently proved to complete the optimization 2–3 orders of magnitude faster than fmincon (Table 2 ). The speed of the solvers did not significantly improve when explicit functions to calculate the gradients of the non-linear terms or the objective function were provided. The computation time of the fmincon solver could be decreased by almost one order of magnitude by allowing less restrictive optimization tolerance criteria. Unfortunately, this resulted occasionally in premature termination and thus sub-optimal results. For reasons of robustness and speed, we opted to utilize the LINDO library for all following analyses. Table 2 Comparison of performance of fmincon and LINDO solver for estimation of feasible ranges. Solver Computation time for Ranges to estimate fmincon LINDO - parsing 20 ± 1 s 20 ± 1 s - feasibility check 1 25 ± 3 s 0.2 ± 0.1 s - ranges of concentrations 166 51 min 23 s - ranges of ΔrG 147 145 ± 20 min 30 s - non-linear constraints 7 n.d. a 1 s The time is given for at least duplicate analyses of the Schaub data set on a Pentium IV 3 GHz processor. Note: a , no runtime is provided because no robust optimization was possible (see text). Application of anNET to published metabolome data sets We tested the thermodynamic consistency of three recently published metabolome data by Schaub et al. [ 20 ], Hiller et al. [ 22 ], and Ishii et al. [ 23 ], all of which relate to wild-type E. coli glucose-limited continuous cultivations at a growth rate of 0.10–0.13 h -1 . For these conditions, fluxes in central carbon metabolism were measured experimentally by 13 C metabolic flux analysis [ 24 , 25 ]. We used this information to manually compile a list of 36 direction constraints in central carbon metabolism (which, in the following, we refer to as 'Set 1'). An independent second set of direction constraints was obtained in silico using our above mentioned tool for the prediction of the minimum set of essential flux directions. For growth on glucose and by using the biomass vector specified in the model iJR904 [ 21 ], we obtained a total of 131 reactions ('Set 2') that need to be active under the assumption that all reactions in the model are reversible. Notably, all these reactions are located in peripheral regions of the metabolism, where unique biosynthetic routes to the biomass precursors have to be active. A knockout in these genes is lethal unless the model topology or the biomass vector is ill-defined. Interestingly, by this approach no flux direction is predicted in central carbon metabolism, where multiple alternative pathways exist. Owing to the complementary nature of Set 1 and Set 2, we merged them to construct Set 3. We found that not all of the three data sets were thermodynamically feasible, even when we allowed a 10% error on all measured concentrations (Table 3 and Additional file 4 ). Consistent with the previous analysis [ 11 ], the Schaub data set was proven to be feasible with all sets of flux constraints. In contrast, both the Hiller and the Ishii data sets were not feasible when the set of flux constraints obtained from 13 C flux analysis was employed. Table 3 Consistency check of three recent E. coli metabolome datasets. Measured concentrations Constraints on flux directions Data set CCM Redox cofactors Energy carriers Others Set 1 Set 2 Set 3 (= Set 1 + Set 2) Schaub 8 0 2 0 F F F Hiller 8 3 3 1 NF F NF Ishii 14 5 3 71 NF F NF The flux directions sets are described in the main text. Abbreviations: F, feasible; NF, not feasible, CCM, central carbon metabolism. Details can be found in Additional file 4 Troubleshooting of non-feasible systems We used our troubleshooting routine to localize the conflicts that provoke the infeasibility in the above datasets. Despite the large number of measured metabolites in the dataset of Ishii et al. and the therewith involved increased risk for system infeasibility, only one apparent thermodynamic inconsistency was found to exist in the data set, which is the concentration range of ribulose-5-phosphate (ru5p-D) (see Additional file 5 ). Conflicts were found to exist with the concentration of ribose-5-phosphate (r5p) and the directions of three enzymes: ME2, ICDHyr, and RPI. Removal of the directions constraints for ME2 and ICDHyr did not relax the unfeasibility, thus locating the inconsistency around RPI, which catalyzes the isomerization between ru5p-D and r5p. In fact, removal of the measurement of ru5p-D or r5p, or of the RPI reaction direction constraint turned the system into a feasible system. Owing to the high confidence of the RPI flux direction estimate based on 13 C metabolic flux analysis, we conclude that the problem is likely due to an erroneous concentration. From thermodynamics, roughly equimolar concentrations are expected for the two intermediates ru5p-D and r5p, whereas a 4–5 fold higher amount was detected for ru5p-D. Interestingly, Ishii et al. reported additional wild-type metabolome data sets for different growth rates: four out of five wild-type data sets exhibited the same inconsistency. In the data set by Hiller et al. , our analysis identified two problematic concentration ranges: glucose-6-phosphate (g6p) and glyceraldehyde-3-phosphate (g3p) (see Additional file 6 ). In the first case, measured g6p concentrations are not compatible with the assumed direction of the phosphoglucoisomerase (pgi). In glucose-limited continuous cultures, the glycolytic flux through the pgi is directed from g6p to fructose-6-phosphate (f6p) [ 24 ]. Because of the resulting constraint on Δ r G'(pgi), the concentration of g6p has to be at least 3.1-fold larger than that of f6p, in contrast with the measured ratio of 2.3. The conflict is relieved when a relative error of at least 25% is allowed for both concentrations. In the second problem, g3p is incompatible with the concentrations of dihydroxyacetone-phosphate (dhap) and fructose-1,6-bisphosphate (fdp) and the connecting reactions catalyzed by the triosephosphate isomerase (tpi) and the fdp-aldolase (fba). The reaction directions imposed by the glycolytic flux dictate that the g3p concentration has to be in the range between 2–38 μM when the concentrations of fdp and dhap are assumed to be within 30% of the measured values. This range, however, is largely lower than the measured g3p value of 200 μM. Interestingly, no feasible system could be obtained when removing the experimental concentrations of either fdp or dhap from the dataset, because this resolved the infeasibility around either fba and tpi, respectively, but not both simultaneously. Overall, these examples demonstrate the usefulness of the troubleshooting function to identify the loci of thermodynamic infeasibility and to suggest potential error sources. In general terms, it is important to emphasize two aspects. Firstly, apparent inconsistencies in metabolite concentrations may be linked to bad measurements but also reflect faulty thermodynamic data or local differences in reactant activity. The troubleshooting routines can not distinguish between these causes, but diagnoses all of them simultaneously by the requisite to further relax concentration constraints around specific nodes. Secondly, the fact that modification or removal of one constraint (or more) in an unfeasible system lead to a feasible one proves neither that the modified constraints were wrong, nor that the others were correct. It is a mere indication that requires experimental verification."
} | 3,526 |
35653573 | PMC9191679 | pmc | 2,945 | {
"abstract": "Significance In this study, we ask how ant colonies integrate information about the external environment with internal state parameters to produce adaptive, system-level responses. First, we show that colonies collectively evacuate the nest when the ground temperature becomes too warm. The threshold temperature for this response is a function of colony size, with larger colonies evacuating the nest at higher temperatures. The underlying dynamics can thus be interpreted as a decision-making process that takes both temperature (external environment) and colony size (internal state) into account. Using mathematical modeling, we show that these dynamics can emerge from a balance between local excitatory and global inhibitory forces acting between the ants. Our findings in ants parallel other complex biological systems like neural circuits.",
"conclusion": "Conclusion Our results provide a simple, tractable example of a collective perception–action loop, where social dynamics are used to integrate the sensory perception of individual ants and to produce a coherent collective response. We show that under borderline conditions, individual ants suppress their own assessment or perception of sensory information about the external environment in favor of a collective decision. Moreover, the social dynamics enable the colony to integrate information not only about the external environment, but also about the state of the colony itself (its size in this case). The collective outcome is therefore more than a mere average of the “opinions” of the individual ants. Our modeling results also highlight the importance of heterogeneity in social groups. This group-level property is thought to contribute to a group’s ability to adapt to changing conditions in systems ranging from insect colonies to human societies ( 58 – 60 ). For example, in honey bee and bumblebee colonies, the variability of individual thresholds for fanning, a behavior that helps circulate the air, is hypothesized to contribute to the overall performance of collective thermoregulation ( 24 , 61 , 62 ). In our model, heterogeneity is expressed as the distribution of individual response thresholds in a colony. The collective threshold, which is an emergent property of the group, can then vary within the range of that distribution depending on the context, which in our case is the size of the colony. In other words, the variability between individuals is what enables the adaptation of the collective property. The collective sensory threshold seems to emerge from a balance between two opposing forces. As in other biological systems, these forces do not have to map to a single biological mechanism, but could rather represent a combination of various processes with a similar functional effect. For example, excitation and inhibition in a neural network arise from many different types of neurons and neurotransmitters whose effects differ in aspects such as timescale, spatial distribution, and plasticity. Likewise, transmembrane currents are the product of many types of ion channels, each modulating the membrane response in a different way. This mechanistic complexity underlies both the robustness of biological systems and their flexibility to adapt their responses to various conditions on multiple timescales ( 63 – 68 ). Similarly, the excitatory and inhibitory forces at play in an ant colony are likely composed of various chemical, physical, and possibly other types of interactions. The isolation of individual mechanisms and the understanding of their precise functional role in the collective dynamics will require further experiments. However, mesoscopic models such as the one employed here can provide a formal understanding of the principles of emergent collective computation even without detailed knowledge of the underlying mechanisms.",
"discussion": "Results and Discussion Ants Respond to Increasing Temperature by Evacuating the Nest. Controlling the sensory environment of a group of freely behaving animals is challenging. We decided to use thermosensation as the sensory modality, because temperature is a scalar property of the environment, and its sensation is minimally dependent on the position and specific behavior of the individual. To study how ants respond to a temperature increase, we developed a behavioral arena in which the ground temperature is controlled by an array of thermoelectric components ( SI Appendix , Fig. S1 and Materials and Methods ). A thin layer of plaster of Paris constitutes the surface of the arena. This setup allows us to rapidly change the arena’s ground temperature ( SI Appendix , Fig. S2 ), while maintaining ground moisture level constant ( SI Appendix , Fig. S3 ). In all of the following experiments, we placed O. biroi colonies of variable sizes in the arena. Unlike most other ants, O. biroi is queenless, and all experimental colonies were composed of workers and larvae in a 2:1 ratio. Adult ants were ∼1 mo old, and larvae were 6 to 7 d old. O. biroi reproduces asexually and clonally, providing precise experimental control over an individual’s genotype. We standardized genotypes by sourcing all individuals from the same clonal line and stock colony ( Materials and Methods ). At baseline, the set value of the ground temperature controller was 26 °C. When ants are placed in the arena, they quickly create a nest by settling around a brood pile ( Fig. 1 A , Inset and SI Appendix , Fig. S4 ). Typically, a few scouts explore the arena, while most ants remain inside the nest ( Fig. 1 A and B ). After a settling period of about 48 h, we studied how colonies respond to temperature changes by subjecting them to a sequence of perturbation events. In each perturbation, the set temperature was abruptly increased to a higher set value for a period of 15 min. Perturbations were spaced by intervals of 2 h to allow the ants to resettle ( SI Appendix , Fig. S5 ). When the perturbation temperature was relatively high, the colonies typically responded with a stereotypical coordinated evacuation of their nesting site. Fig. 1 B–E and Movie S1 show a representative example of a colony of 36 workers and 18 larvae responding to a 40 °C perturbation. Following the temperature increase, the ants gradually get excited and increase their activity levels ( Fig. 1 C ). After some delay, the colony initiates an ordered evacuation in which all ants leave the nest in a column ( Fig. 1 D and E and Movie S1 ). Because temperature is increased evenly across the entire arena, the ants remain in a high-activity “explorative” state following the initial response. Depending on the perturbation temperature, this state can be highly organized (for relatively low temperatures, Fig. 1 F ) or manifest as a more chaotic and disorganized collective pattern (for high temperatures, Fig. 1 G ). Once the temperature returns to baseline, the ants slowly relax and reform the nest cluster ( Fig. 1 H and I ). Fig. 1. The response of an ant colony to a step temperature perturbation. ( A ) A snapshot from a raw experimental video, showing a colony of 36 ants on a temperature-controlled plaster of Paris arena ( Materials and Methods and SI Appendix , Figs. S1 and S2 ). Each of the ants is marked with a unique combination of color tags to allow for individual behavioral tracking. The arena is confined by a black metal frame heated to 50 °C ( SI Appendix , Fig. S2 ). The ants form a nest (red square at the bottom and Inset at top right), with a few scout ants exploring the arena (top red square and Inset at top left). The light brown objects in the arena are food items. ( B–I ) Snapshots depicting the typical dynamics of the response of a colony to a strong temperature perturbation. Images are processed by removing the background for visual clarity. ( B ) Baseline state. Before the onset of the perturbation, most ants reside in the nest, with few scout ants exploring the arena. ( C ) Excitement. Following the onset of the perturbation, the ants first respond by increasing their activity level around the nest. ( D ) Evacuation onset. After a delay that lasts up to a few minutes, the ants suddenly begin to leave the nest in a well-defined direction. ( E ) Full evacuation. The colony forms a well-organized evacuation column. ( F ) Stable evacuation. ( G ) Disordered perturbed state. In some cases, especially under high-temperature perturbations, the organized evacuation column breaks, and the colony enters into a high-activity, swarm-like state, where the movements of the ants are only weakly correlated. ( H ) Relaxation. Following the return of the temperature to baseline, the ants slowly relax and begin to reform the nest, possibly in a different location. The relaxation process can take up to 1 h to complete. ( I ) New baseline. The colony has fully returned to its baseline relaxed state. The Colony Response to Temperature Perturbations Is Collective. Escape, or place-change behavior in response to changes in temperature or other environmental parameters, is a ubiquitous behavior that is often studied in the context of sensory decision making ( 1 , 39 – 41 ). Solitary animals make such decisions independently, and the correlations between individuals are generally low, both in the decision itself (whether to leave or not) and in the timing and direction of leaving. In contrast, the response of clonal raider ants to the temperature increase seemed highly coordinated, both in time and in space. To quantify the collectivity in this response, we performed an experiment to measure the coordination and correlation between the responses of individual ants. We perturbed three colonies of 36 ants and 18 larvae with a sequence of 24 temperature perturbations of 15 min duration, each with an amplitude of 33 °C, which does not produce a robust evacuation response. Ants in these colonies were marked with unique combinations of color tags and were individually tracked using a custom software ( 42 ). From the tracking results, the nest location before each perturbation was determined ( Materials and Methods ). To identify evacuating ants, we defined a circle with a 15-mm radius around the nest (around twice the typical radius of the nest blob, Fig. 2 A ). The binary response b i k of the i th ant to the k th perturbation was defined as 1 if she exited the nest circle at some point during the time of the perturbation and remained outside for a duration of at least 30 s and as 0 in the case that she did not. Ants that were outside the nest circle at the beginning of the perturbation were treated as missing values. For each of the three colonies, we found that the distribution of average responses across ants for each perturbation is bimodal ( Fig. 2 B ), suggesting the decisions to leave the nest are correlated between ants. To test this, we calculated the correlation between the response of each pair of ants in the colony and compared the distribution of these values to a null distribution, constructed by randomly shuffling the individual responses of the ants between perturbation events. These two distributions differ significantly (average pairwise correlation value of 0.424, P < 10 −5 ; Materials and Methods and Fig. 2 C ), showing that the decisions of ants in the colony are indeed highly correlated. Fig. 2. Ants respond collectively to temperature perturbations. ( A ) Measures of individual responses. We define a circle of radius R = 15 mm around the location of the nest. A schematic drawing of the trajectories of two ants is depicted in green and pink. For each ant, we record the binary response ( b ), the response direction ( α ), and the response latency ( τ ) as its first crossing of that circle for a duration longer than 30 s, as explained in the text and Materials and Methods . ( B ) Histograms of the average binary response across ants. Each histogram is constructed from one colony subjected to a sequence of 24 perturbations of 33 °C. ( C ) Pairwise correlations between the binary responses of ants in B , compared to correlations in shuffled responses. Shuffled responses are generated by shuffling the binary responses of each ant to all the perturbations independently of other ants in its colony, therefore eliminating any correlation. The real distribution is composed of 1,890 correlation values, produced from the responses of 108 ants from three colonies. The null distribution is composed of 189,000 correlation values, produced from 100 independent shuffles of the responses. ( D ) Scatter plot depicting the distributions of individual response latencies, from three colonies subjected to a sequence of 24 perturbations of 40 °C. Each column represents the responses of ants from a single colony to one perturbation. The events are sorted first by colony and then by the average individual response latency in each event. ( E ) Pairwise correlations between the response latencies of ants in D compared to a null distribution generated in the same way as in C . ( F and G ) Plots as in D and E , but for the individual response directions. Note that the response direction measure is cyclic, but because the per-event distribution (one column in F ) is narrowly distributed around the average colony direction, this does not have a significant effect on the analysis. To assess whether the ants are also correlated in the timing and direction of their response, we repeated the experiment with three additional colonies of the same size and composition, using the same protocol. However, this time we subjected the colonies to stronger perturbations of 40 °C, which generally produce a robust collective nest evacuation. We defined the individual response as before and also measured the response delay τ i k as the time elapsed between the onset of the k th perturbation and the time the i th ant crossed the circle and the response direction α i k as the angle between the point of crossing and the line connecting the nest center and the center of the arena ( Fig. 2 A ). Plotting the distributions of individual response latencies across perturbation events, we found that the variability between individual response latencies in the same event is lower than the variability between events ( Fig. 2 D ), suggesting that ants are coordinated in the timing of their response. We showed this formally by comparing the distribution of pairwise latency correlations to a null distribution in the same way as before (average pairwise correlation value of 0.558, P < 10 −5 ; Materials and Methods and Fig. 2 E ). We then repeated the same analysis for the response direction, showing that ants coordinate their response also spatially (average pairwise correlation value of 0.628, P < 10 −5 ; Materials and Methods and Fig. 2 F and G ). The Collective Response Is Characterized by a Size-Dependent Threshold. To better understand how the collective response depends on the amplitude of the perturbation, we performed an experiment with 10 colonies of 36 workers and 18 larvae each and subjected each colony to a sequence of perturbations of variable amplitude, ranging from 28 to 45 °C. We defined the collective state of the colony as the fraction of ants outside of the nest in any given frame ( Fig. 3 A ). We defined the binary collective response as 1 if a quorum of at least 90% of the colony was outside the nest for at least 30 consecutive seconds at some point during the time of the perturbation and 0 otherwise. For each perturbation temperature, we computed the probability of a collective response across all colonies, resulting in a sigmoidal psychometric-like response curve typical for systems with a noisy threshold response ( Fig. 3 B ). Using a logistic regression model ( Materials and Methods ), we estimated the threshold θ to be 34.12 °C for the colonies in this experiment, with a 95% CI of 33.3 to 34.8 °C. Fig. 3. The collective threshold depends on group size. ( A ) A trace from a 24-h-long perturbation protocol using a colony of 36 tagged ants. Perturbations are 15 min long and separated by intervals of 2 h. The set temperature is shown in black. The dispersion of the colony (defined as the fraction of ants outside the nest circle) is shown in brown. The interval allows the colony to relax back to baseline before the next perturbation. ( B ) The probability of a “full response” (defined as at least 90% of the ants being outside the nest at the same time for at least 30 s at some point during the perturbation; solid line) as a function of the perturbation temperature. The shaded band represents the 95% CI of probability (computed using asymptotic normal approximation for binary coefficient estimation). The dashed line represents a logistic regression fit of the response curve. Experimental colonies consisted of 36 tagged ants. ( C ) Fitted logistic regression curves as in B for different colony sizes, showing an upward shift in the response curve. ( D ) The collective threshold parameter θ c , estimated by logistic regression, as a function of colony size. The shaded band represents the 95% CI, estimated using the bootstrap method with 1,000 sample repetitions ( Materials and Methods ). To further investigate this collective threshold, we conducted an experiment with different colony sizes, ranging from 10 to 200 ants. The age, clonal line, and workers to larvae ratio were identical to the experiments described above ( Materials and Methods ). Each colony was subject to the same experimental protocol with varying temperature perturbations. The collective threshold was estimated for each colony size as above ( Fig. 3 C ). Plotting the threshold as a function of colony size ( Fig. 3 D and SI Appendix , Fig. S6 ), we found that larger colonies have a significantly higher collective threshold than smaller ones. This effect is robust to variation in the parameters defining the binary collective response (the quorum threshold and duration; SI Appendix , Fig. S7 ). Such a dependency of an emergent property on group size in a system characterized by many intricate interactions is not surprising from a complex systems science point of view, and an increase in social cohesion as a function of group size has indeed been observed experimentally ( 43 ). Nevertheless, it is unexpected in light of previously established paradigms for the study of collective sensing in social insects. These can roughly be divided into two classes that lead to two different predictions regarding group size effects. The first one is “wisdom of the crowd,” in which noisy independent individual estimates of an external signal are pooled to produce a more accurate collective estimate. Under this scenario, the variability in the response is a result of noisy estimation of the external environment by individual ants, and the threshold temperature is an objective quantity that is independent of the group. Accordingly, the prediction would be that larger groups should have higher accuracy (i.e., less variance) in estimating temperature, but the threshold temperature should not change with group size ( 44 – 46 ). A dependency of the threshold on group size would imply that group dynamics integrate information suboptimally and in a biased way ( 47 – 49 ). According to the second paradigm, collective sensing is used to increase the resolution, or sensitivity, to external events such as predator attacks ( 50 , 51 ). Perturbations of any strength should ideally elicit a response, and the existence of a response threshold is the result of a limited detection capacity. In this case, however, larger groups are predicted to have lower thresholds, because the probability of detecting a weak perturbation increases with the number of individuals in the group. In contrast to these previously documented dynamics, our finding that the collective temperature threshold increases as a function of group size suggests that the response threshold is not an objective quantity to be estimated, but rather a result of a decision-making process that integrates information about the environment with information about the internal state of the colony. It is in fact possible that the optimal collective response threshold differs for colonies of different sizes. For example, if nest evacuation is associated with a relatively higher cost in larger colonies, a threshold that increases with group size might be adaptive. The Response Dynamics Are Characterized by Distinct Timescales and Social Feedback. The strong correlation between the responses of individual ants and the existence of a collective response threshold implies that the collective response is coordinated using interactions between the ants, resulting in positive feedback dynamics. To visualize the dynamics underlying the emergence of the collective response, we plotted the average time course of the collective state (the number of active ants outside the nest; Materials and Methods ) over the perturbation events in the variable-amplitude experiment with colonies of 36 tagged ants ( Fig. 4 A ). We divided the perturbation events into three groups, according to the perturbation amplitude: weak perturbations, for temperatures in which the response probability is lower than 0.1; strong perturbations with a response probability greater than 0.9; and intermediate perturbations between those cutoffs. For intermediate perturbations, we separately plotted events in which the colony collective binary response was 1 and 0, respectively ( Movies S1 and S2 ). The time-course plot is indicative of two dominant processes with distinct timescales. The first one is a fast response, with a timescale of 1 to 3 min, in which some ants become excited. This fast response is slower than the timescale of the physical temperature increase ( SI Appendix , Fig. S2 E ), and the delay might correspond to the internal physiological and neural processing time, as well as the behavioral delay until an ant is considered to have “responded” by our definition. At this timescale, the number of ants responding to midrange temperature perturbations is widely distributed around half of the number of ants in the colony ( Fig. 4 B , green). This is what we would expect if the ants respond independently to perturbations around the typical individual threshold. It is therefore plausible that the colony response during this timescale is dominated by the ants’ intrinsic responses and less by interactions between the ants, which leads to a continuous distribution of the colony state variable ( Fig. 4 B ). Following this first stage of the response, a few minutes into the perturbation, a second dynamical process with a characteristic timescale of 5 to 10 min seems to kick in, in which the colony state converges on either a low or a high value ( Fig. 4 B , pink). This convergence indicates the dominance of interaction-driven social feedback during this stage. Fig. 4. The response dynamics are characterized by distinct timescales and social feedback. ( A ) The evolution in time of the colony state variable (the fraction of active ants) following a temperature change. Single-trial response curves from all experiments with 36 ants were averaged according to temperature and collective response condition. The dark brown curve represents the average response for high-temperature perturbations, for which the response probability was larger than 0.9. Events without a full response (b = 0) were excluded. The light brown curve represents the average response for intermediate-temperature perturbations (response probability between 0.1 and 0.9) in which the colony responded (b = 1). Responses in the same temperature range, but in which the colony did not respond (b = 0), are depicted by the light blue curve. Finally, the dark blue curve represents the average response for low-temperature perturbations [response probability smaller than 0.1; events with positive response (b = 1) were excluded]. ( B ) Histograms showing distributions of single-trial colony activity states for the intermediate-temperature range where the response probability is between 0.1 and 0.9, at two time points along the response curve. The green histogram shows the distribution for the interval between 3 and 4 min following perturbation onset, roughly corresponding to a time window in which the effect of the faster process has been exhausted, while the effect of the slower process is not yet apparent. Each datapoint in the histogram is the median value of a single perturbation event in that segment. The pink histogram shows the distribution for the interval between 14 and 15 min, when both transient dynamics of the response have run their course. A Binary Network Model Recapitulates the Emergence of a Collective Threshold. To better understand the interactions that might underlie a colony’s collective response, we implemented a simple spin-like binary network model. This type of model is commonly used for complex and collective systems ( 52 – 55 ). We represent each ant by a binary variable σ i , in which σ i = 1 represents the “perturbed” behavioral state and σ i = − 1 the “relaxed” behavioral state ( Fig. 5 A ). The behavioral state of each ant is decided by a logistic activation function ( Fig. 5 B ): [1] P ( σ i = 1 ) = 1 1 + e − β h i . Fig. 5. The emergence of the collective threshold can be modeled with two opposing forces. ( A ) An illustration of the model’s two-stage dynamics. In the first stage, the ants respond independently according to their individual response thresholds. As a result, a subset of ants becomes active. In the second stage, the interactions between the ants result in the colony being either fully active or fully inactive. ( B ) The logistic activation function of the individual ant. The ant is either active or inactive in a probabilistic manner depending on an integrated input parameter h i . The β parameter controls the width of the ambiguous response region. ( C ) The individual response thresholds θ i are sampled from a normal distribution of mean θ m and width θ S D . The collective threshold θ c is the temperature for which the cumulative probability equals m c . ( D ) Simulation of the collective threshold, showing the response probability as a function of temperature, averaged over 100 simulation runs. For each run, a new set of individual thresholds is sampled. See Materials and Methods for full details on the simulation parameters. ( E ) The collective threshold as a function of group size for the basic model (gray circles) and the asymmetric model (purple circles). The interaction parameters J p and J r were chosen to have the same collective threshold at N = 50 and to approximately replicate the range of thresholds observed in the experiment. ( F ) An illustration of possible ant interaction mechanisms. ( Top ) Global, pheromone-based interaction, in which ants in a given state contribute (black arrows) to the total concentration of pheromone in the environment (blue circle), which is then perceived by all ants (blue arrows). ( Bottom ) Local, contact-based interaction, in which ants in a given state affect only the behavior of nearby ants (red arrows). Here, P represents the steady-state probability of the ant to be in the perturbed state, h i is the integrated input of ant i , and β is a thermal noise-like parameter that measures the determinism of the individual response (with high values leading to a deterministic threshold-like response according to the value of h i ). In the general case, this integrated input is a combination of the external temperature perceived by the individual ant and the contribution of the social interactions. For simplicity, we take advantage of the apparent separation between the timescales of these two components and assume the dynamics to take place in separate stages ( Fig. 5 A ). In the first stage, the ants respond individually and independently to the external temperature, according to the equation: [2] h i = T − θ i . Here, T is the perturbation temperature and θ i is the individual threshold of ant i , randomly drawn from a normal distribution with mean θ m and SD θ S D ( Fig. 5 C ). By the end of this stage, some of the ants will be in the relaxed state and some in the perturbed state ( Fig. 5 A , Center ). The state of the colony is characterized by the order parameter m , defined as the fraction of perturbed ants in a colony with N ants: [3] m = 1 2 N ∑ i ( σ i + 1 ) . In the second stage, the states of the ants change over time according to the social dynamics: [4] h i ( t ) = ∑ j J i j ( t ) σ j ( t ) . Here, J i j ( t ) is the interaction strength between ant i and ant j at time t . In the most general case, the interaction between a pair of ants will depend on their relative position and on their behavioral states. We further simplify our model by ignoring space and assuming that all ants interact with all other ants all the time. This can be justified by the observation that the collective response dynamics of the ants are much slower than their movement speed in the nest during the “excitement” period that precedes the response ( Movies S1 and S2 ). Under this assumption, we can characterize the interaction by two parameters, representing the exciting effect of perturbed ants and the inhibiting effect of relaxed ants. The input to all the ants is then the same and can be written as: [5] h ( t ) = J p N p ( t ) − J r N r ( t ) = N J p m ( t ) − N J r ( 1 − m ( t ) ) , where J p and J r are the interaction parameters, and N p ( t ) and N r ( t ) are the numbers of perturbed and relaxed ants at time t , respectively. To reproduce the experimental results within the model, we assume β to be high enough for the model to be characterized by two stable fixed points at m = 0 and m = 1 , representing the collective relaxed and perturbed states ( SI Appendix , Fig. S8 A–D ), with a separatrix at: [6] m c = J r J p + J r . This implies that, on average, the value of m at time 0 will decide the final collective state of the colony. This m 0 is a result of the individual responses of the ants to the temperature increase during the first stage of the dynamics. Therefore, the threshold perturbation temperature θ c is the temperature at which the average fraction of perturbed ants will equal m c . In the case where the thermal noise is small compared to the individual threshold variability, this threshold satisfies the condition: [7] P ( θ i < θ c ) = m c . The exact value of this threshold will depend on the distribution of the individual thermal thresholds in the ant population, as well as the ratio between the excitatory and inhibitory interaction strengths ( Fig. 5 C and SI Appendix , Fig. S8 ). Group Size Dependency of the Collective Threshold Entails Asymmetric Interactions. While the model, as defined above, captures the emergence of the collective threshold from the interactions between the ants ( Fig. 5 D ), it does not include any group size dependency that can reproduce the experimental observations in Fig. 3 ( SI Appendix , Fig. S8 L and gray curve in Fig. 5 E ). Because ants in the experiments had a narrow age range, belonged to the same clonal line, and were sampled randomly from the same stock colony in each experiment ( Materials and Methods ), we can assume that their individual thermal thresholds were also sampled from the same distribution, regardless of colony size. This implies that, under the assumptions of the model, group size dependency can arise only if the excitatory and inhibitory components scale differently with group size N . For example, we can set the inhibitory interaction to scale linearly with N , but let the excitatory interaction be independent of N : [8] h ( t ) = J p m ( t ) − N J r ( 1 − m ( t ) ) [9] m c = N J r J p + N J r . Using these definitions, the threshold value will sublinearly increase with the size of the colony (purple curve in Fig. 5 E ). This choice is not arbitrary, because different types of interactions should produce different scaling with N . Clonal raider ants are blind and mostly communicate via pheromones and tactile interactions. For a volatile pheromone, the concentration in the air surrounding ants aggregated in close proximity in the nest should scale with the number of ants that emit the pheromone, which equals the fraction of pheromone-emitting ants multiplied by N . This pheromone concentration is then perceived by all ants in the colony ( Fig. 5 F , Top ). The exact form of the scaling will depend on the chemical properties of the pheromone, the physical properties of the environment, and the concentration/response curve of the ants themselves. On the other hand, physical interactions between pairs of ants will depend on the rate of encounters. Physical interaction is a common excitatory mechanism in ants, particularly in scenarios of recruitment ( 56 , 57 ). When the colony is dense, the encounter rate per ant saturates, meaning an ant is always in contact with other ants at the maximum capacity. This implies that the total excitatory force an ant feels is dependent on the fraction of active ants in the colony and does not scale with the size of the colony ( Fig. 5 F , Bottom ). Of course, the distinction between the two mechanisms does not have to be clear cut and can be quantified with a scaling parameter α : [10] h ( t ) = N α p J p m ( t ) − N α r J r ( 1 − m ( t ) ) . The value of α represents a gradual transition between local, nearest-neighbor interactions and global, all-to-all interactions. An increase of collective threshold with group size will then emerge for any α r > α p ( SI Appendix , Fig. S9 ). So far, we have considered interactions that are state dependent, that is, interactions in which ants signal their state to other ants. However, some interactions within the colony can be regarded as state independent or to vary slower than the typical timescale of the behavioral response. For example, weakly volatile “aggregation” pheromones could mark the nest site. The strength of this nest odor will then scale with the number of ants in the nest but will not change because of ants leaving the nest momentarily. The existence of such a pheromone is supported by the tendency of the ants to settle back at their original nest location following perturbations ( SI Appendix , Fig. S10 ). Such an aggregation signal would act as a constant pulling force that balances the excitation within the colony. In our model, we can account for such an interaction by removing the state dependency from the inhibitory term and write Eq. 10 as: [11] h ( t ) = N α p J p m ( t ) − N α r J r . Because the scaling of the inhibitory interaction strength is the same as in the initial version of the model, we again get an increase of the threshold with group size ( SI Appendix , Fig. S9 D and E ). However, the shape of the increase and the predictions of the model for larger group sizes differ ( SI Appendix , Fig. S9 F )."
} | 8,795 |
32936651 | null | s2 | 2,946 | {
"abstract": "Velocity of capillary flow in closed or open channels decreases as the flow proceeds down the length of the channel, varying as the inverse of the square root of time or as the inverse of travel distance. In order to increase the flow rate-and extend the duration of the flow-capillary pumps have been designed by mimicking the pumping principle of paper or cotton fibers. These designs provide a larger volume available for the wicking of the liquids. In microsystems for biotechnology, different designs have been developed based on experimental observation. In the present paper, the mechanisms at the basis of capillary pumping are investigated using a theoretical model for the flow in an open-channel \"capillary tree\" (i.e., an ensemble of channels with bifurcations mimicking the shape of a tree). The model is checked against experiments. Rules for obtaining better designs of capillary pumps are proposed; specifically, we find (1) when using a capillary tree with identical channel cross-sectional areas throughout, it is possible to maintain nearly constant flow rates throughout the channel network, (2) flow rate can be increased at each branch point of a capillary tree by slightly decreasing the areas of the channel cross section and decreasing the channel lengths at each level of ramification within the tree, and (3) higher order branching (trifurcations vs bifurcations) amplify the flow rate effect. This work lays the foundation for increasing the flow rate in open microfluidic channels driven by capillary flow; we expect this to have broad impact across open microfluidics for biological and chemical applications such as cell culture, sample preparation, separations, and on-chip reactions."
} | 429 |
35071893 | PMC8771958 | pmc | 2,947 | {
"abstract": "Superwetting surfaces\nare widely used in many engineering fields\nfor reducing energy and resistance loss. A facile and efficient method\nusing laser etching has been used to fabricate and control superwettable\ndrag reduction surfaces. Inspired by the self-cleaning theory of lotus\nleaves, we propose controllable patterned bionic superhydrophobic\nsurfaces (BSSs) simulating the uneven micro/nanostructures of lotus\nleaves. The superhydrophobicity and drag reduction ratios at low velocities\nare highly improved using a laser ablation method on metal substrates.\nHowever, unstable air layers trapped on superhydrophobic surfaces\nare usually cut away by a high-velocity flow, which greatly reduces\nthe drag reduction performance. The fabricated bionic superhydrophobic/hydrophilic\nsurfaces (BSHSs) with alternated hydrophilic strips can build a large\nsurface energy barrier to bind the three-phase contact line. It maintains\nthe stable drag reduction by capturing the air bubbles attached to\nthe hydrophilic strips at a high velocity. Three-dimensional simulation\nanalysis and equipment to measure the weak friction of a self-assembled\nsolid–liquid interface are used to explain the drag reduction\nmechanism and measure the drag reduction ratios at different flow\nspeeds. BSSs achieve an improved drag reduction effect (maximum 52.76%)\nat a low velocity (maximum 1.5568 m/s). BSHSs maintain an improved\nand steady drag reduction effect at high speed. The drag reduction\nratios can be maintained at about 30% at high speed, with a maximum\nvalue of 4.448 m/s. This research has broad application prospects\nin energy saving, liquid directional transportation, and shipping\ndue to their robust superhydrophobic properties and stable drag reduction\neffect.",
"conclusion": "Conclusions BSSs and BSHSs are ablated\nby laser on Al–Mg alloys substrates.\nBionic lotus leaf micro/nanostructures can improve superhydrophobicity.\nThe nanoscale particulate melts on the surface after ablation, further\nincreasing the surface roughness. When the diameter of the micro/nanostructured\nunit is 40 μm, the biggest CA is 168°. The smallest SA\nfor 40 μm is 0.5°. The BSS with a diameter of 40 μm\nhas the best superhydrophobicity. The drag reduction ratio is above\n50% at low velocities ranging from 1.112 to 2.4468 m/s. The hydrophilic\nstrips pinning the three-phase contact line can form the wetting step.\nThe air bubbles sheared away can be captured to keep the stable drag\nreduction at a high velocity. Interfacial slippage formed by the hydrophilic\nstrips and micro/nanostructures can obviously improve the drag reduction\neffect by an ASHS simulation. The drag reduction ratios can be held\nnearby 30% at high velocities (2.4468–4.448 m/s). The BSHS\nwith a hydrophilic strip spacing of 1 mm has the optimal stable drag\nreduction effect. The bionic lotus leaf micro/nanostructures on the\nsuperhydrophobic surfaces and the hydrophilic strips are necessary\nfor the stable and high drag reduction properties. The experiments\nare in accordance with the simulation analysis. Effective stable drag\nreduction strategies for the BSS and ASHS at a low velocity and a\nhigh velocity, respectively, can reduce energy consumption, save cost,\nand realize the applications such as liquid directional transport,\nmarine vessels, microfluidic devices, and microchannels.",
"introduction": "Introduction Energy conservation\nand pollution reduction have had significant\nimplications in scientific research and engineering in the past. From\n1950 to 2001, fuel consumption of the shipping industry has increased\nfrom 65 to 280 million tons. The surface friction of the ship consumes\nmore than half of the energy in ocean exploitation and shipping engineering.\nShipping accounts for 10% of global carbon transport emissions. 1 , 2 The United States has built more than two million kilometers of\npipelines to carry oil and natural gas. 3 Pipe wall friction drag during liquid transportation can greatly\nreduce the efficiency of transportation. Constructing biomimetic shark-shaped\nmicro/nanostructures on the surfaces of aircraft can reduce surface\ndrag by 8% and reduce energy consumption by 1.5%. 4 , 5 Engineering\nefficiency, energy conservation, and emission reduction can be improved\nby drag reduction. 6 − 9 Therefore, achieving efficient drag reduction in many fields, such\nas aircraft, liquid pipeline transportation, power equipment, biomedicine,\nand so on, has become an important research topic. 10 − 14 Bionic materials with special wettability can\nachieve drag reduction,\nand there are many different synthesis methods that can be used to\nfabricate drag reduction surfaces. 15 − 20 Wang et al. prepared vertical and horizontal interlaced grids as\nmicro/nanostructures for superhydrophobic surfaces to maintain a drag\nreduction ratio of 13%. 21 Luo et al. used\nthe template method to prepare bionic shark skin with low viscous\nresistance surfaces at a drag reduction ratio of 12%. 22 Solomon et al. demonstrated the drag reduction mechanism\nby modeling the effect of lubricant viscosity ratio on fluid resistance\nin laminar flows. 23 However, most of the\nfabrication methods, such as emulsion and dispersion polymerization,\ntemplate forming, and spray, have application limitations. The superhydrophobic\nsurface morphology and wettability prepared by chemical methods are\nnot easy to be accurately controlled, and material failure is inevitable. 24 − 26 Nanosecond laser scanning ablation with simulation models can accurately\ndesign and control a variety of patterned micro/nanostructures with\nspecial wettability for optimal performance. Compared with the traditional\npreparation methods, laser ablation is a pollution-free and environmentally\nfriendly cleaning technology and almost does not cause significant\ndamage to the surface of the substrate. It also has the advantages\nof convenient processing, good stability, and high-degree automation. 27 − 31 The morphology, size, and distribution of micro/nanostructures can\nregulate and control superwettability. Bionic fish scales, horizontal\nstrips of bionic rice leaves, bionic butterfly wings, and other special\nmicro/nanostructures have anisotropy. The surfaces of these bionic\nstructures have been widely studied. The fabricated superhydrophobic\nsurfaces can achieve a certain drag reduction effect. 32 − 36 Wu et al. proved that the drag reduction effect of the bionic superhydrophobic\nsurface at a flow rate of 0.66 m/s was at the optimum level and that\nthe drag reduction rate was 2.805%. 37 The\nchanging trend of drag reduction rate decreases with an increase of\nthe flow velocity because the friction resistance increases gradually\nand the air bubble layer on the surface loses rapidly. The air\nlayer at the solid–liquid interface of the micro/nanostructures\non a superhydrophobic surface can reduce the area of liquid–solid\ndirect contact and convert it into part of gas–liquid contacts,\nwhich reduces the resistance of the fluid in the boundary layer and\nincreases the fluid velocity. This achieves a drag reduction effect. 38 However, the trapped air layer is destroyed\nby fluid shear stress, resulting in the instability of the superhydrophobic\nsurface. Under the conditions of bubbles disappearing, external force\ndisturbance, fault of construction, and phase transformation, the\ndrag reduction effect will be impaired when the surface changes from\nthe Cassie state to the Wenzel state. Some researchers have proposed\na continuous forced injection of air to maintain a layer of air on\na solid surface underwater for drag reduction. 39 , 40 With the increase of flow velocity, the instability of the fluid\nboundary layer caused by the loss of a large number of air bubbles\nbroke the drag reduction of the superhydrophobic surface. As a consequence,\nbuilding a stable air layer on the superhydrophobic surface can maintain\na good drag reduction effect at high velocities. Herein, we\npropose an efficient strategy to fabricate the controllable\npatterned bionic superhydrophobic surfaces (BSSs) and bionic superhydrophobic/hydrophilic\nsurfaces (BSHSs) inspired by the convex array micro/nanostructures\nof lotus leaves. A nanosecond laser ablation method was used to prepare\nthe patterned superhydrophobic surfaces with alternated hydrophilic\nstrips on an aluminum–magnesium alloy substrate. Enhanced superwettability\nand sliding of BSSs can be obtained by regulating the patterned surface\nparameters. The alternated hydrophilic strips of BSHSs can build a\nlarge energy barrier on the superhydrophobic surface as a wet phase\nstep and then strongly fix the air/water/solid three-phase contact\nline. Air bubbles attached to the hydrophilic zones are captured to\nmaintain a stable drag reduction effect at high velocities. We used\nCOMSOL to build three-dimensional simulation models to accurately\ncalculate and elucidate the stable and improved drag reduction of\nBSSs and BSHSs in laminar flows, and we verified the experimental\nresults. The friction resistance and drag reduction ratio were calculated\nby multiphase flow phase field and surface integral methods. The simulation\nanalysis chose the optimal solution of BSSs and BSHSs. A facile friction\nresistance measuring equipment is used to test the drag reduction\nratios at different flow velocities. This research will improve the\nsuperwetting and sliding properties of patterned bionic special wettability\nsurfaces and enhance the stability of drag reduction at high velocities.\nIt has application prospects in the field of reducing energy intensity,\ndrag losses, directional transport of liquids, and ship engineering.",
"discussion": "Results\nand Discussion Figure 2 a–c\nshows the photograph of a lotus leaf and scanning electron microscopy\n(SEM) images of the convex micro/nanostructures on the lotus leaf\nsurface at different magnifications. A BSS sample is fabricated on\nan aluminum–magnesium alloy substrate by laser ablation, as\nshown in Figure 2 d.\nEach bionic lotus leaf micro/nanostructured unit with a diameter of\n40 μm is constructed on the BSHS sample. The micro/nanostructure\nis similar to that of a real lotus leaf compared to those with other\ndiameters in Figure 2 e–g. We used the laser ablation method for each unit along\nthe edge of the bionic lotus leaf micro/nanostructure, and then the\nconvex micro/nanostructure began to take shape in the middle. The\nnanoscale particulate melts on the surface produced by laser ablation\nare shown in Figure 2 h. The surface roughness and superhydrophobicity can be enhanced\nby the micro/nanostructures of BSSs. The drag reduction effect will\nincrease with an increase in superhydrophobicity. Figure 2 (a) Photograph of a lotus\nleaf (Photograph courtesy of “Jinglin\nZhang.” Copyright 2018; and it is a free domain). (b, c) SEM\nimages of micro/nanostructures on the lotus leaf surface and enlarged\nSEM images (adapted with permission from ref ( 33 )). (d) As-prepared BSS\nsample. (e–h) Enlarged SEM images of the laser-treated BSS\nof the bionic lotus leaf micro/nanostructured unit with a diameter\nof 40 μm under various magnifications. The SEM images of the four as-prepared BSS samples with unit diameters\nof 20, 50, 60, and 70 μm are shown in Figure 3 a–h under various magnifications.\nThe surface roughness decreases with the increase of the diameter.\nWhen the unit diameter is 20 μm, the micro/nanostructured units\nare similar to the real convex surface of lotus leaves. But, the structural\nunits are irregular, and the size of the bulge units is smaller than\nthat in the structural unit with a diameter of 40 μm. When the\ndiameter of the structural unit is increased to 50 μm, the shape\nof each unit can still be seen clearly but the spacing between each\nunit is increased, and the height of the bulge is significantly decreased.\nAt this time, the surface roughness is decreased compared to that\nof the structural unit with a diameter of 40 μm. As the diameters\nare increased to 60 and 70 μm, regular and tight protruding\nstructures can no longer be formed. The spacing between units is larger\nand the protruding height is lower. The surface roughness will decrease\nwhen the diameter is greater than 40 μm. The diameter of 40\nμm affords the best values for the surface morphology and surface\nroughness. The superhydrophobicity can be improved by high roughness\nand it benefits the drag reduction effect of BSSs. Figure 3 (a, b) SEM images of\nthe BSS sample with a unit diameter of 20\nμm. (c, d) SEM images of a sample with a unit diameter of 50\nμm. (e, f) SEM images of a sample with a unit diameter of 60\nμm. (g, h) SEM images of a sample with a unit diameter of 70\nμm. The SEM images of the three as-prepared\nBSHS samples with a hydrophilic\nstrip spacing of 1, 2, and 3 mm are shown in Figure 4 . The bionic micro/nanostructured unit diameter\nis 40 μm, and each hydrophobic strip width is 50 μm. To\ncontrol variables, the surfaces prepared on each substrate have the\nsame laser-etched area as sample 1, sample 2, and sample 3. The best\nhydrophilic strip spacing and the improved stable drag reduction effect\nfor the BSHS are obtained by simulation analysis and drag reduction\nexperiments. Figure 4 SEM images of three BSHS samples with hydrophilic strip\nspacings\nof 1 mm (sample 1), 2 mm (sample 2), and 3 mm (sample 3). The EDS pattern of the BSS sample with a bionic lotus leaf\nmicro/nanostructured\nunit diameter of 40 μm is shown in Figure 5 a. The chemical compositions and contents\nof the BSS sample are as follows: Al (47.3%), O (27%), C (22.4%),\nF (0.8%), Mg (2.3%), and Si (0.1%). An Empyrean diffractometer was\nused to test the X-ray diffraction (XRD) pattern results. It analyzed\nthe composition of the BSS sample. The XRD pattern of the BSS sample\nis shown in Figure 5 b. Diffraction peaks of the BSS include Al [PDF: 04-003-2966], Al 0.963 Si 0.037 [PDF: 04-003-7126], Al 0.85 Si 0.15 [PDF: 04-003-7125], and Al 0.99 Si 0.01 [PDF: 04-001-2511]. The five strong peaks located at 38.583,\n44.787, 65.074, 78.139, and 82.480° are ascribed to the (111),\n(200), (220), (311), and (222) planes, respectively. Al and the solid\nsolution of Al and Si are the main compositions of the BSS sample.\nA 3D interference contour tester is used to measure the surface roughness\nof the BSS and BSHS samples. Figure 5 c,d shows the typical 3D morphologies of the BSS and\nBSHS samples. The prepared BSS and BSHS samples exhibit a homogeneous\ndistribution. The average mean square roughnesses ( R a ) of the BSS and BSHS samples are 3.3886 and 4.788 μm,\nrespectively. The maximum height fluctuation ( R z ) values of BSS and BSHS samples are 61.1984 and 73.7428 μm,\nrespectively. Many regular bionic lotus leaf micro/nanostructures\nand hydrophilic strips are present on the surface of the BSS and BSHS\nsamples. The local convex rough surface can improve superhydrophobicity. Figure 5 (a) EDS\npattern of the BSS sample with a micro/nanostructured unit\ndiameter of 40 μm. (b) XRD pattern of the BSS sample. (c) Three-dimensional\nmorphology of the BSS sample. (d) Three-dimensional morphology of\nthe BSHS sample. Figure 6 a shows\nthe diagram of the water contact angle (CA) as a function of the structure\nunit diameters of BSSs. When the diameter is 20 μm, the CA is\n160°, and when the diameter is 40 μm, the CA improves to\n168° at a maximum. The change in CAs may be attributed to the\ndifference in the surface morphology of BSSs with different diameters\nof the unit. After laser ablation, granular melt protrusions produce\nsmaller rough nanostructures. When the diameter is 40 μm, the\nCA is greater than those for units with diameters of 50–80\nμm shown in Figures 2 and 3 . Therefore, the CA for units\nwith 40 μm diameter is larger than those for 50–80 μm.\nDeep holes along the micro/nanostructure units of 20 μm are\nproduced by laser ablation, resulting in inhomogeneity of the surface\nmorphology. It indicates that the surface with a unit of too small\ndiameter will form a deeper hole by laser etching, and it is not the\nbest surface morphology and contact angle. The CAs decrease with the\nincrease of diameters. When the diameters increase to 50, 60, 70,\nand 80 μm, the CAs decrease to 158, 155, and 153°, respectively.\nWhen the diameters are large, the spacing between each microstructure\nunit increases, each convex structure becomes smaller and irregular,\nand the heights decrease compared with those of 20 and 40 μm.\nThe paths and quantities of laser decrease when they etch the same\nnumber of micro/nanostructure units. Fewer convex nanostructures of\nmolten materials are produced by the same amount of laser energy.\nThe lack of molten materials and smaller irregular units decrease\nthe superhydrophobicity of BSSs. The surface with a diameter of 40\nμm has an optimal size, shape, and roughness for the best superhydrophobicity.\nThe variation tendency of water sliding angle (SA) under various diameters\nis similar to that of water CA. The diameter is the key parameter\nto determine the sliding property of BSSs, as shown in Figure 6 b. The changing trend of SA\nis that it decreases first and then increases with the increase of\ndiameter, and then it reaches the maximum value at 40 μm. For\ndiameters of 20, 50, 60, 70, and 80 μm, the SAs are 1.5, 3,\n4.5, and 6.5°, respectively. The video frames of continuously\nsliding water droplets on the BSS sample with a diameter of 40 μm\nare shown in Figure 6 c. The smallest SA of 40 μm is 0.5°, which indicates that\nthe surface is more slippery when the micro/nanostructure unit is\nof the optimal size, shape, and roughness. Figure 6 d shows the water CA on a hydrophilic area\nof the BSS sample with a diameter of 40 μm. The water CA is\n67°, which indicates that the hydrophilic strips are hydrophilic.\nThe best roughness of micro/nanostructures will enhance the superhydrophobicity\nof the BSS and further promote the drag reduction effect. A BSS with\na diameter of 40 μm has the best superhydrophobicity as a test\nand analyzed the sample for the drag reduction property. The same\nparameters are used for the preparation of the BSHS samples. Figure 6 (a) CAs and\n(b) SAs of the BSS samples with various diameters.\n(c) Video frames of continuously sliding water droplets on the BSS\nsample with a diameter of 40 μm. (d) Water contact angle on\na hydrophilic area. The hydrophilic strips\nhave some effect on the three-phase contact\nline for the superhydrophobic/hydrophilic surface. The three-phase\ncontact line is bound at the boundary of the surface. A contact angle\nhysteresis phenomenon is shown in Figure 7 a. The contact line of the liquid does not\nchange as it reaches the superhydrophobic/hydrophilic interface regions.\nAs the contact angle of the droplet increases, the three-phase contact\nline always keeps the force in a balanced state. The force balance\nequation of the three-phase contact line at the hydrophilic and superhydrophobic\nboundary is 41 1 where θ′ is the apparent CA,\nγ slh and γ sgh are the solid–liquid\nand solid–gas surface tension on the hydrophilic surface, and\nγ sls and γ sgs are the solid–liquid\nand solid–gas surface tension on the superhydrophobic surface.\nThe CA hysteresis extent of superhydrophobic/hydrophilic surfaces\nis 41 2 where θ 1 and θ 2 are the CAs on the superhydrophobic/hydrophilic surface. Figure 7 (a) Schematic\nof the stress balance condition of the three-phase\ncontact line at the superhydrophobic/hydrophilic surface (adapted\nwith permission from ref ( 41 )). (b) Bubble blocking and interface slippage mechanism\nof a BSHS simulation model at the flow impact. It has a similar stress state of the air layer on the solid in\nair and underwater. Therefore, when bubbles are attached to BSHS underwater,\nthe theory can also be applied for the stress balance condition of\nthe three-phase contact line. BSHSs can bind to the three-phase contact\nline underwater. When the gas contact line moves outward from the\nsuperhydrophobic surface to the hydrophilic interface region, it is\nbound by a partial CA hysteresis. An air layer is attached on the\nsuperhydrophobic surface. 42 , 43 Interface slippage\nexists on the hydrophilic and superhydrophobic structures with a sealed\nstable air layer. Friction resistance will reduce by interface slippage.\nThe wetting step can be constructed by laser ablation of hydrophilic\nstrips on the BSHS by adjusting the spacing of hydrophilic strips\nreasonably. The bound three-phase contact line captures the air bubbles\nsheared away and locks them and attaches along the hydrophilic strips\nto maintain the stable drag reduction effect at a high flow velocity.\nBubble blocking and interface slippage mechanism of a BSHS simulation\nmodel at the flow impact are shown in Figure 5 b. The air bubble layer contains the air\nlayer existing near the hollow micro/nanostructures of the BSHS and\nthe sealing air bubbles around the hydrophilic strips. The presence\nof an air layer makes the liquid–gas interface partially replace\nthe original liquid–solid interface, resulting in interface\nslippage and reducing the original friction resistance of the solid–liquid\ninterface. Interface slippage is the main factor to improve the drag\nreduction effect. We use COMSOL by a three-dimensional finite\nelement method to establish\nthe BSS ( Figure 8 )\nand BSHS ( Figure 9 )\nsimulation models at a flow velocity. The models are based on the\nfluid–solid coupling physical field and the volume of fluid\nmethod. The left inlet and right outside of the model are velocity\nand pressure, respectively. The pressure is set to 0. The upper surface\nof the computational domain is set as the untreated surface for comparison.\nThe superhydrophobic surface with the same bionic micro/nanostructures\nis the lower surface as the prepared BSS sample. The other model simulates\nthe superhydrophobic/hydrophilic surface as the prepared BSHS sample.\nBased on the Reynolds equation, the laminar flow field under a stable\ncondition is simulated numerically. The characteristic height of the\ncalculated domain is 200 μm. In accordance with the best superhydrophobicity\nparameters, the height and diameter of the micro/nanostructured unit\nare 30 and 40 μm, respectively, while the hydrophilic strip\nspacings are 1, 2, and 3 mm, and the hydrophobic strip width is 50\nμm. Figure 8 Simulation analysis of the BSS. (a, b) Velocity contour and the\npartially enlarged photograph of the cross section in the y – z plane; (c, d) static pressure\ndistribution in the near-wall area and the cross-sectional photograph\nin the x – z plane; (e, f)\nliquid–gas volume fraction distribution and its partially enlarged\nphotograph on the cross section in the y – z plane. Figure 9 Simulation analysis of\nthe BSHS. (a, b) Velocity vector and the\npartially enlarged photograph of the cross section in the y – z plane; (c, d) static pressure\ndistribution in the near-wall area and the cross-sectional photograph\nin the x – z plane at the pressure\noutlet; (e, f) liquid–gas volume fraction distribution and\nthe partially enlarged photograph on the cross section in the y – z plane. The simulation models are described by the continuity equation\nof fluid motion with respect to solid as follows 3 The gravitational force is neglected\non the\nmicron scale, and so the momentum equations are 4 where u , v , and w are the fluid\nvelocities (m/s) in the x, y , and z directions, respectively, μ\nis the dynamic viscosity (Pa·s), ρ is the density (kg/m 3 ), and p is the pressure (Pa). 40 Figure 8 shows the\nsimulation analysis of the BSS at 2 m/s. The overall velocity contour\ndistribution and a partially enlarged photograph of velocity vector\ndistribution in the near-wall area on the cross section in the y – z plane are shown in Figure 8 a,b. As the water\nflows through, there is an obvious change in the velocity gradient\non the gas–liquid interface. As the friction resistance increases,\nthe fluid velocity near the upper wall surface and at the liquid–solid\ninterface near the lower wall gradually decreases, and the fluid velocity\nat the gas–liquid interface with micro/nanostructures is obviously\nhigher than that at the solid–liquid interface. The velocities\nof the bionic lotus leaf micro/nanostructures near the wall (the blue\nand green areas) are greater than those at the solid–liquid\ninterface. BSSs with the bionic lotus leaf micro/nanostructures can\ncapture an air layer to form interface slippage to improve the velocity.\nViscous resistance is produced by the velocity gradient near the wall.\nThe bubbles are sealed near the wall to form the liquid–gas\ninterface slippage. It can produce a drag reduction effect. Velocity\nin the x -direction increases by interface slippage,\nand then viscous resistance decreases. The total resistance of the\nBSS consists of viscous resistance and pressure resistance. Compared\nwith the viscous resistance, the pressure resistance is smaller in\nthe order of magnitude and can be ignored. Therefore, the friction\nresistance is mainly composed of viscous resistance. The BSS can reduce\nthe viscous resistance, improve the flow velocity near the wall, and\nreduce the resistance. The static pressure distribution near the wall\nand the cross-sectional photograph in the x – z plane are shown in Figure 8 c,d. Compared with the untreated surface, the BSS with\nmicro/nanostructures has a more stable static pressure field, as shown\nin Figure 8 c. The static\npressure field changes obviously only near the wall, as shown in Figure 8 d, and it leads to\npressure resistance. Figure 8 e,f shows the liquid–gas volume fraction distribution\nand partially enlarged photograph on the cross section in the y – z plane. Red and blue represent\nwater and gas, respectively. An air layer can be formed near the bionic\nlotus leaf micro/nanostructure to improve the drag reduction effect\nof the BSS. In Figure 9 , the\nsimulation analysis indicates that the hydrophilic strips of the BSHS\ncan capture the bubbles to attain a stable drag reduction. The velocity\nvector and the partially enlarged photograph on the cross section\nin the y – z plane are shown\nin Figure 9 a,b. When\nwater flows through, the velocity gradient at the gas–liquid\ninterface is influenced by the superhydrophobic/hydrophilic surface\nwith the micro/nanostructures. With the increase of friction resistance,\nthe fluid velocity near the gas–liquid interface is obviously\nhigher than that at the solid–liquid interface. Figure 9 c,d shows the static pressure\ndistribution near the wall and the cross-sectional photograph in the x – z plane at the pressure outlet.\nThe static pressure field is simultaneously influenced by the superhydrophobic/hydrophilic\nsurface to produce the pressure resistance. In the pressure outlet\nsection, the superhydrophobic/hydrophilic area near the wall causes\nthe change in the nearby flow field, and the overall stability is\nin the concentrated area of the velocity far from the wall. Figure 9 e,f shows the liquid–gas\nvolume fraction distribution. The liquid exists in the hydrophilic\nstrips. The bubbles are attached to the hydrophilic strips. The BSHS\nin synergy can achieve a stable drag reduction at high velocities. Figure 10 and Video S1 show a dynamic process of the stable\nexistence and interface slippage of the two-dimensional BSHS model\nat a flow velocity of 2 m/s. As the water flows across the BSHS, the\nair layer always exists and tends to be stably locked around the hydrophilic\nstrips after dynamic changes. The total air layer contains the air\nbubbles in the micro/nanostructures of the BSHS and the locked air\nbubbles around the hydrophilic strips. The diameters of the locked\nair bubbles around the hydrophilic strips trapped on the BSHS are\n30 μm in the model. According to the analysis of the final steady-state\nsolution, the air layer is stable. Its role in the slipping water\nis to create slippage around the hydrophilic strips for drag reduction. Figure 10 (a–f)\nDynamic process of air layer stable existence and\ninterface slippage of the two-dimensional BSHS model. The friction resistance and drag reduction ratios were calculated\nby the BSS and three BSHS with different hydrophilic strip spacing\nsimulation models at a flow velocity of 2 m/s in Figure 11 . The drag reduction ratio\nθ s is calculated as follows 5 where f gv , f lv , and f lp are\nthe total friction resistance values on the untreated surface, viscous\nresistance, and pressure resistance of the built BSS and BSHS simulation\nmodels, respectively. Figure 11 (a) Friction resistance of the untreated surface and the\nBSS model\nat different velocities. (b) Drag reduction ratios of the BSS model\nat different velocities. (c–e) Friction resistance of the untreated\nsurfaces and BSHS models with different hydrophilic strip spacings\nat different velocities. (f) Drag reduction ratios of the BSHS models\nwith different hydrophilic strip spacings at different velocities. The friction resistance of the untreated surface\nand the BSS model\nat different velocities (1.112–4.448 m/s) is shown in Figure 11 a. The friction\nresistance increases when the flow velocity increases, as shown in Figure 11 b. The maximum\nresistance is 10.5662 × 10 –5 N at the highest\nvelocity of 4.448 m/s. The drag reduction ratio decreases gradually\nwith the increase of the flow velocity. The maximum drag reduction\nratio is 56.43% at 1.112 m/s. The drag reduction ratio remains above\n50% in the range of 1.112–1.557 m/s. When the velocity is greater\nthan 3.7808 m/s, the drag reduction ratio decreases to less than 20%.\nIt indicates that when the velocity is small, the air layer existing\non the BSS can remain stable, and so the drag reduction ratio is large;\nhowever, when the flow velocity increases gradually, the trapped air\nbubble layer on the BSS is unstable and sheared away by the water\nflow, and thus the interface slippage is destroyed and the drag reduction\nratios drop rapidly. Therefore, the drag reduction property of BSSs\nis excellent at a low velocity by the simulation analysis. The\nfriction resistance of the untreated surfaces and the BSHS\nmodels with different hydrophilic strip spacings at various velocities\nis shown in Figure 11 c–e. The same computational area is used to control variables\nin the model. The friction resistance increases when the hydrophilic\nstrip spacings change from 1 to 3 mm. The drag reduction ratio fluctuates\nin stages in Figure 11 f. When the flow velocity is 1.112–2.4464 m/s, the drag reduction\nratio remains above 40%, and then it decreases and remains at about\n30% at 2.4464–4.448 m/s under the trend of fluctuation change.\nThe drag reduction ratio declines when the hydrophilic strip spacings\nincrease. There will be a significant reduction in the drag reduction\nwith the hydrophilic strip spacing increasing to a stable value. With\na hydrophilic strip spacing of 1 mm, the friction resistance of BSHS\nis 0.9319 × 10 –5 N at the minimum and the drag\nreduction ratio is 47.37% at the maximum at a velocity of 1.112 m/s.\nThe hydrophilic strips can seal the air bubbles sheared away, and\nso it can effectively exhibit a stable drag reduction at a high velocity\nby the simulation analysis. The results show that the BSHS with a\nhydrophilic strip spacing of 1 mm has the best drag reduction property\nunder the same condition. Figure 12 shows\nsnapshots of sealing air bubbles underwater in three BSHS samples\nwith hydrophilic strip spacings of 1, 2, and 3 mm in the experiments.\nThe experiment proves that the BSHS has the function of sealing the\nbubbles near the hydrophilic strips as the flow velocity increases.\nThe smaller the hydrophilic strip spacing, the greater and denser\nthe sealed air bubbles and the better the drag reduction effect. The\nexperimental results are identical to the theoretical simulation. Figure 12 (a–c)\nSnapshots of sealing air bubbles underwater in three\nBSHS samples with hydrophilic strip spacings of 1, 2, and 3 mm, respectively. The rising process from underwater to air of the\nBSHS is shown\nin Figure 13 and Video S2 . The BSHS sample blocks the bubbles\nas the acceleration of the velocity increases. The air layer generates\nand then remains stable during the movement for drag reduction. Therefore,\nwe can be sure of the presence of stable air bubbles in the experiments. Figure 13 (a–f)\nSnapshots of the rising process from underwater to\nair of the BSHS. A high-precision self-assembly\nfriction resistance measuring equipment\nwas built to measure the weak friction resistance and the drag reduction\nproperty of the samples, as shown in Figure 14 . It can also verify the results of simulation\nanalysis. Mechanical, drive, and signal processing modules constitute\nthe test device. Powered by a water pump, water flows through the\nnozzle and impacts the surface of the test sample and then backflows\ninto the water tank in the experiment. A friction resistance sensor\nis used to measure the friction resistance at the solid–liquid\ninterface on the sample surface. The output signal changes from the\nstrain gauge deformation are collected by the data collection. The\ncorresponding friction resistance value is obtained by conversion.\nThe experimental results are averaged by several measurements. Figure 14 Set of the\nfriction resistance measuring equipment. The BSS and BSHS samples are tested at different velocities. In\nthe experiments, the drag reduction ratio θ e is calculated\nas follows 6 where f ge and f le are the friction resistance values on the\nuntreated surface and the as-prepared samples, respectively. A bulk Al–Mg alloy serves as the untreated surface as a\ncontrol sample and the parameters are controlled in the same conditions. Figure 15 a shows that the\nfriction resistance of the untreated surface and the as-prepared BSS\nsample at different velocities in the drag reduction experiments.\nThe friction resistance increases with the increase of velocity, while\nthe drag reduction ratio decreases with the increase of flow velocity.\nThe drag reduction ratio is 52.76% at the maximum when the friction\nresistance is 1.7128 × 10 –3 N at a velocity\nof 1.112 m/s, and the minimum value is 10.15% when the friction resistance\nis 13.4368 × 10 –3 N at a velocity of 4.448\nm/s in Figure 15 b.\nThe overall changing trend of the drag reduction ratio is the same\nas that of simulation analysis, which shows the correctness of the\ntheory. At a low velocity, the drag reduction effect of the BSS is\nexcellent, while at a high velocity, the impact of water flow affects\nthe air bubble layer sealed by micro/nanostructures, resulting in\nincreasing the friction resistance and reducing the drag reduction\neffect. In Figure 15 c, the variation trend of friction resistance of BSHS samples at\ndifferent flow velocities is the same as that of BSS. The friction\nresistance increases when the hydrophilic strip spacing increases.\nThe smallest friction resistances of the BSHS with a hydrophilic strip\nspacing of 1 mm increase from 2.0432 × 10 –3 to 9.9793 × 10 –3 N at various velocities.\nThe drag reduction ratios of the BSHS samples with different hydrophilic\nstrip spacings at different velocities are shown in Figure 15 d. The drag reduction ratio\nof the BSHS fluctuates and remains at 40 ± 4% with a velocity\nless than 2.4464 m/s. As the velocity increases to 4.4468 m/s, the\noptimal drag reduction ratios can be held nearby 30%. When the flow\nvelocity is 1.112 m/s, the drag reduction ratio is 43.65%. The drag\nreduction ratios of BSHS with a hydrophilic strip spacing of 1 mm\nare superior to others at different flow velocities. The BSHS sample\nwith a hydrophilic strip spacing of 1 mm can achieve the best and\nstable drag reduction properties at high velocities of 2.4468–4.448\nm/s. The bionic lotus leaf micro/nanostructures and the hydrophilic\nstrips are significant for the stable drag reduction effect. The experiments\nverify the correctness of drag reduction simulation models. Figure 15 (a) Friction\nresistance of the untreated surface and the BSS sample\nat different velocities. (b) Drag reduction ratios of the BSS sample\nat different velocities. (c) Friction resistance of the untreated\nsurfaces and the BSHS samples with different hydrophilic strip spacings\nat different velocities. (d) Drag reduction ratios of the BSHS samples\nwith different hydrophilic strip spacings at different velocities."
} | 9,068 |
22761019 | null | s2 | 2,948 | {
"abstract": "This paper gives an overview of elastomeric valve- and droplet-based microfluidic systems designed to minimize the need of external pressure to control fluid flow. This Concept article introduces the working principle of representative components in these devices along with relevant biochemical applications. This is followed by providing a perspective on the roles of different microfluidic valves and systems through comparison of their similarities and differences with transistors (valves) and systems in microelectronics. Despite some physical limitation of drawing analogies from electronic circuits, automated microfluidic circuit design can gain insights from electronic circuits to minimize external control units, while implementing high-complexity and high-throughput analysis."
} | 197 |
33850714 | PMC8039717 | pmc | 2,949 | {
"abstract": "Increasing understanding of metabolic and regulatory networks underlying microbial physiology has enabled creation of progressively more complex synthetic biological systems for biochemical, biomedical, agricultural, and environmental applications. However, despite best efforts, confounding phenotypes still emerge from unforeseen interplay between biological parts, and the design of robust and modular biological systems remains elusive. Such interactions are difficult to predict when designing synthetic systems and may manifest during experimental testing as inefficiencies that need to be overcome. Transforming organisms such as Escherichia coli into microbial factories is achieved via several engineering strategies, used individually or in combination, with the goal of maximizing the production of chosen target compounds. One technique relies on suppressing or overexpressing selected genes; another involves introducing heterologous enzymes into a microbial host. These modifications steer mass flux towards the set of desired metabolites but may create unexpected interactions. In this work, we develop a computational method, termed M etabolic D isruption Work flow ( MDFlow ), for discovering interactions and network disruptions arising from enzyme promiscuity – the ability of enzymes to act on a wide range of molecules that are structurally similar to their native substrates. We apply MDFlow to two experimentally verified cases where strains with essential genes knocked out are rescued by interactions resulting from overexpression of one or more other genes. We demonstrate how enzyme promiscuity may aid cells in adapting to disruptions of essential metabolic functions. We then apply MDFlow to predict and evaluate a number of putative promiscuous reactions that can interfere with two heterologous pathways designed for 3-hydroxypropionic acid (3-HP) production. Using MDFlow , we can identify putative enzyme promiscuity and the subsequent formation of unintended and undesirable byproducts that are not only disruptive to the host metabolism but also to the intended end-objective of high biosynthetic productivity and yield. As we demonstrate, MDFlow provides an innovative workflow to systematically identify incompatibilities between the native metabolism of the host and its engineered modifications due to enzyme promiscuity.",
"conclusion": "5 Conclusion We presented MDFlow , a method for quantitatively evaluating the side effects of engineered modifications on the host metabolic network resulting from enzyme promiscuity. Without mitigation, such side effects may lead to either unexpected behaviors during experiments or failure to take advantage of potentially beneficial interactions. By combining PROXIMAL and flux analysis in a streamlined workflow, MDFlow is capable of both discovering new interactions and evaluating their effects without the need for costly, time-consuming studies of in vivo experiments. MDFlow can be used at all stages of the metabolic engineering process. Prior to any design work, the method can be applied to expose native background promiscuous activity, revealing potentially interfering enzymes and metabolites that may be present but not documented in the model, resulting in building Extended Metabolic Models (EMM), as we demonstrated in prior works ( Amin et al., 2019 ; Hassanpour et al., 2020 ). The workflow can then actively guide the pathway construction process by providing feedback on engineering decisions as they are being made, thus helping to identify modifications that create more robust strains. And finally, MDFlow may be leveraged to compare a fully designed pathway to other candidate pathways for the same target compound: when implemented, a pathway with less predicted metabolic disruption may have a higher yield. MDFlow is a first systematic automated analysis step towards the evaluation of underground metabolism and its interaction with engineered cellular machinery.",
"introduction": "1 Introduction Integrating heterologous synthesis pathways within microbial hosts has been instrumental in the biomanufacturing of industrial products such as biofuels, polymers, pharmaceuticals, therapeutics, flavors and chemical commodities ( Lee et al., 2008 ; Madison and Huisman, 1999 ; Nakamura and Whited, 2003 ; Trantas et al., 2015 ; George et al., 2015 ). One strategy to improve yield is to use well-established metabolic engineering techniques such as gene deletion, promoter engineering, media optimization, etc. ( Lee et al., 2009 ). Another strategy is to directly engineer enzymatic properties such as activity, selectivity, inhibition-resistance, and solubility ( Yoshikuni et al., 2008 ). Using one or more of these strategies has proven effective in the development of strains with desired target yields, productivity, and titers. Often, such metabolic engineering strategies yield unexpected enzyme-compound interactions. Some interactions can be beneficial for the survival of the host. For instance, Patrick et al. documented 41 rescue instances where the lethality of an essential protein deletion was suppressed by overexpression of one noncognate E. coli gene, attributing some of them to catalytic promiscuity and substrate ambiguity ( Patrick et al., 2007 ). In other cases, beneficial interactions can come at a cost. The overexpression or knockout of enzymes can result in interactions that are disruptive for growth and maintenance by siphoning off key metabolic intermediates like pyruvate, acetyl-CoA, and NADH. For example, while seeking to suppress lethality of inactivating the pyridoxal-5-phosphate (PLP) cofactor synthesis pathway, Kim et al. experimentally identified a four-step serendipitous pathway in E. coli that restored the strain’s ability to grow on glucose at the expense of consuming essential intermediates in the native serine biosynthetic pathway ( Kim et al., 2010 ), producing toxic byproducts ( Kim and Copley, 2012 ). The presence of high concentrations of heterologous enzymes and metabolites within microbial cells causes unexpected promiscuous interactions with host enzymes and metabolites. For example, the short-chain dehydrogenase YMR226C used to produce 3-hydroxypropionic acid (3-HP) in yeast is associated with 15 known substrates ( Fujisawa et al., 2003 ; Jessen et al., 2015 ). In E. coli strains featuring the malonyl-CoA pathway for 3-HP synthesis, significant quantities of lactate and acetate are produced, even after lactate dehydrogenase ( ldh ) and pta were knocked out ( Rathnasingh et al., 2012 ). Another instance of promiscuous activity can be observed for pathways intended for butanol production: the promiscuity of the bifunctional butyryl-CoA dehydrogenase (AdhE2) enzyme with substrate acetyl-CoA often results in concomitant synthesis of ethanol with butanol ( Inui et al., 2008 ; Atsumi et al., 2008a ; Nielsen et al., 2009 ). In yet another example, Liao and colleagues leveraged promiscuity of ketoacid decarboxylase (KVD) and alcohol dehydrogenase (ADH) enzymes to synthesize a spectrum of alcohols from branched-chain amino acid metabolic intermediates ( Atsumi et al., 2008b ). Yet, a caveat of this promiscuous activity is that no single alcohol can be made alone. That is, pyruvate itself is a ketoacid, which can be converted to ethanol by the same promiscuous activity of KVD and ADH. Thus, isobutanol synthesis is also coupled to ethanol synthesis due to enzyme promiscuity ( Trinh et al., 2011 ). While ubiquitous ( D’Ari and Casadesús, 1998 ; Nobeli et al., 2009 ; Khersonsky et al., 2006 ; Tawfik and D. S, 2010 ) and often observed, effects of enzyme promiscuity on the host metabolic network are often ignored during design and only identified during experimental studies. Predicting such interactions early in the design cycle could yield improved design outcomes and reduce experimental efforts. The prediction of enzymatic products due to substrate promiscuity has mainly relied on hand-curated rules that capture well-known enzymatic transformations. For example, a list of 50 reaction rules, each associated with one or more reaction, was previously defined to explore novel synthesis pathways ( Cho et al., 2010 ; Campodonico et al., 2014 ). Another set of rules was applied repetitively to generate novel synthesis ( Li et al., 2004 ) or degradation pathways ( Finley et al., 2009 ). Further use of such rules allowed the compilation of over 130,000 hypothetical enzymatic reactions that connect two or more KEGG metabolites ( Hadadi et al., 2016 ), and the compilation of predicted metabolic products into databases such as MINEs ( Jeffryes et al., 2015 ). MyCompoundID ( Li et al., 2013 ) utilizes a similar paradigm and generates products by the repeated application of addition or subtraction of common functional groups. BioTransformer ( Djoumbou-Feunang et al., 2019 ) predicts derivatives by utilizing five separate prediction modules in concert with machine learning and a rule-based knowledge base. PROXIMAL ( Yousofshahi et al., 2015 ) utilizes enzyme-specific reactant–product transformation patterns from the KEGG database ( Moriya et al., 2010 ) as a lookup table to predict products for query molecules. The PROXIMAL algorithm was utilized to create organism-specific E xtended M etabolic M odels (EMMs) that extend reference metabolic models catalogued in databases to include putative products due to promiscuous native enzymatic activities on native metabolites ( Amin et al., 2019 ; Hassanpour et al., 2020 ). Despite advances in predicting promiscuous products, however, these efforts have not been put forward in a systematic way to analyze metabolic network disruption in engineered microbial hosts due to enzyme promiscuity. We develop in this paper a computational method, M etabolic D isruption Work flow ( MDFlow ), to analyze the disruptive impact of enzyme promiscuity on engineered microbial hosts in a systematic manner. We define metabolic disruption as off-target changes in host metabolism that arise because of enzyme-substrate interactions upon gene or pathway overexpression, where such interactions neither exist in the wild-type chassis organism nor are they expected due to presence of recombinant enzyme(s). Accordingly, this definition encompasses all enzyme promiscuity that arises because of adding heterologous enzymes and their chemical products to a host microbe. Therefore, MDFlow is designed to consider two different disruption scenarios. In “Scenario 1”, promiscuous activity is predicted in the context of overexpressed enzymes, whether heterologous or native, acting promiscuously on native host metabolites. Meanwhile, in “Scenario 2,” predictions are made by assuming that native enzymes exhibit promiscuous interactions with synthesis pathway metabolites introduced with engineering changes. Of course, in a biological system, both scenarios would occur simultaneously to some degree, and even higher-order interactions would be possible (e.g., subsequent use of promiscuous reaction products as substrates for additional transformations). Using PROXIMAL ( Yousofshahi et al., 2015 ) and flux analysis ( Orth et al., 2010 ; Segrè et al., 2002 ), MDFlow models and evaluates such promiscuous scenarios. We demonstrate the use of MDFlow to evaluate how engineered microbial hosts are impacted by enzyme promiscuity under three engineering strategies. First, MDFlow is used to explain how essential gene deletion can be suppressed by overexpression of another native gene. Next, MDFlow is used to identify multi-step interactions that may compensate for essential gene knockouts. Third, MDFlow is used to evaluate the potential disruption when a heterologous pathway is added to a microbial host. The first two cases represent Scenario 1 and are evaluated against experimentally verified data. The last case represents a simultaneous application of both Scenarios 1 and 2 and demonstrates that the choice of synthesis pathway can impact metabolic disruption scenarios. This work is novel as it is the first to systematically investigate effects of heterologous and native enzyme promiscuity on host metabolism and its consequences on biocatalysts. Our method serves as the first computational tool that can assist metabolic engineers in (1) identifying sources of unexpected byproducts, (2) assessing the consequences of metabolic engineering on the host, and (3) quantifying pathway-host incompatibility using metabolic network disruption. Outcomes from this work will aid in future studies to design robust systems with more predictable behaviors and improved desired product yield.",
"discussion": "4 Discussion In metabolic engineering, a number of strategies are employed to produce a target metabolite of interest, including the introduction of heterologous enzymes and selective overexpression and deletion of certain genes. After such interventions or combinations thereof, it is not uncommon for unexpected and/or undesirable metabolite product profiles to arise from various interactions between the introduced and native machinery in the host. Using MDFlow , it is possible to predict such interactions in the form of single- and multi-step pathways enabled by promiscuous enzymatic activity. The method utilizes PROXIMAL to construct reactions arising from promiscuity and relies on FBA to assess their impact on yield or biomass growth rate in pre- and post-modification hosts. The results for the single-gene deletions that were rescued via single- and multi-step enzymatic pathways, which were experimentally validated in prior studies, provide evidence of the ability of MDFlow in predicting metabolic network disruption. The results for the 3-HP case illustrate how MDFlow can help identify promiscuous interactions early in the design cycle. Further, our sampling-based FBA analysis shows that promiscuity can cause unexpected byproducts and results in yield disruption. Importantly, MDFlow can be used to explain byproducts often observed but not well explained in the literature. Our Scenario 1 and 2 disruption classification has direct correspondence to the network inference classification proposed by Kim et al. ( Kim and Copley, 2012 ). Interference is classified into three groups: those due to (i) heterologous metabolites in new pathways interfering with native metabolism, (ii) native metabolites interfering with a heterologous pathway, and (iii) heterologous pathway intermediates being diverted by promiscuous activity of native enzymes. MDFlow identifies the same interactions as long as they’re caused by enzyme substrate promiscuity – with groups (i) and (iii) corresponding to Scenario 2 predictions and group (ii) represented by Scenario 1-type interactions. The computational methodology in MDFlow can be further enhanced. It currently uses PROXIMAL to predict promiscuous byproducts; it is possible to use alternatives for promiscuous product prediction. We selected PROXIMAL because we have established confidence in its capabilities in predicting organism-specific enzymatic transformations. Our earlier study of promiscuity using PROXIMAL on non-engineered E. coli allowed the discovery of 17 putative enzymatic reactions that explained metabolomics measurement ( Yousofshahi et al., 2015 ). Regardless of the tool, however, there always remains the issue of false positives. In our work with PROXIMAL, we discarded byproducts that were not documented prior in PubChem, KEGG, or i ML1515/1428. Using machine-learning tools that evaluate the likelihood of compound-enzyme interactions, such as EPP-HMCNF ( Visani et al., 2020 ), might provide further confidence in such predictions. Importantly, better prediction of enzymatic products and their likelihood can improve the workflow’s ability to uncover unexpected interactions and evaluate their impact on the engineered host. Metabolism simulation aspect of MDFlow also offers opportunities for improvement. In certain applications, MOMA can be an attractive alternative to FBA. While FBA utilizes linear programming to maximize an objective function, typically yield or biomass, MOMA approaches cell modeling from the perspective of minimizing redistribution of metabolic fluxes compared to the wildtype conditions. MOMA can provide improved correlation of predictions with experimental flux data over the steady-state modeling provided by FBA ( Segrè et al., 2002 ). Additionally, the model itself can be tailored to the specific circumstances of an experiment. For example, isozymes that are minimally expressed in glucose M9 media were removed from the i ML1515 model to create i ML1428. As a result of this reduction in degrees of freedom, the derivative context-specific model tends to offer more accurate lethality predictions than i ML1515 in gene knockout experiments ( Monk et al., 2017 ). Such adjustments can be informed by manual investigation of the host’s metabolic network, by leveraging detailed kinetic models ( Khodayari and Maranas, 2016 ), or via techniques such as 13 C Metabolic Flux Analysis ( Long and Antoniewicz, 2019 ). Of course, depending on the application, it may be helpful to consider using more complex techniques that can capture translation and regulation aspects of underground metabolism, though we believe – in the three cases we have considered – the overexpressed genes are the dominating factor. Knowing more about the biological sample, such as the concentrations of metabolites and enzymes can shed further light on the amount of disruption. Another area of improvement is the approach used to determine the direction and maximum flux limits of the predicted reactions. We rely on the preset 1% average mean coupling percentage to estimate the limits of all reactions, which may not be representative of the higher or lower actual mean coupling percentage under the conditions of a given experiment. Future studies may re-evaluate directions and flux limits in the context of each experiment individually. Despite these limitations, the presented results are promising and call for further design exploration of the impact of enzyme promiscuity on engineered microorganisms."
} | 4,547 |
29208745 | PMC5717390 | pmc | 2,952 | {
"abstract": "ABSTRACT The genome of the bacterium Burkholderia thailandensis encodes three complete LuxI/LuxR-type quorum sensing (QS) systems: BtaI1/BtaR1 (QS-1), BtaI2/BtaR2 (QS-2), and BtaI3/BtaR3 (QS-3). The LuxR-type transcriptional regulators BtaR1, BtaR2, and BtaR3 modulate the expression of target genes in association with various N -acyl- l -homoserine lactones (AHLs) as signaling molecules produced by the LuxI-type synthases BtaI1, BtaI2, and BtaI3. We have systematically dissected the complex QS circuitry of B. thailandensis strain E264. Direct quantification of N -octanoyl-homoserine lactone (C 8 -HSL), N -3-hydroxy-decanoyl-homoserine lactone (3OHC 10 -HSL), and N -3-hydroxy-octanoyl-homoserine lactone (3OHC 8 -HSL), the primary AHLs produced by this bacterium, was performed by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS) in the wild-type strain and in QS deletion mutants. This was compared to the transcription of btaI1 , btaI2 , and btaI3 using chromosomal mini-CTX- lux transcriptional reporters. Furthermore, the levels of expression of btaR1 , btaR2 , and btaR3 were monitored by quantitative reverse transcription-PCR (qRT-PCR). We observed that C 8 -HSL, 3OHC 10 -HSL, and 3OHC 8 -HSL are differentially produced over time during bacterial growth and correlate with the btaI1 , btaI2 , and btaI3 gene expression profiles, revealing a successive activation of the corresponding QS systems. Moreover, the transcription of the btaR1 , btaR2 , and btaR3 genes is modulated by cognate and noncognate AHLs, showing that their regulation depends on themselves and on other QS systems. We conclude that the three QS systems in B. thailandensis are interdependent, suggesting that they cooperate dynamically and function in a concerted manner in modulating the expression of QS target genes through a successive regulatory network.",
"conclusion": "Conclusion. The study described here provides for the first time an exhaustive portrait of the interplay between the QS-1, QS-2, and QS-3 systems in B. thailandensis E264 ( Fig. 8 ). We observed an interdependence between the QS-1 and QS-2 systems. While we confirmed that the QS-3 system is controlled by BtaR1, we also found that BtaR3 modulates the QS-1 system, which indicates that those two systems are linked. Interestingly, such an interaction between the QS-1 and QS-3 systems seems to be conserved in the closely related species of the Bptm group ( 14 , 17 , 20 ). Interestingly, the QS-2 and QS-3 systems that share common AHLs seem not to be transcriptionally linked, but instead they are temporally connected by their common AHLs. We also highlighted a surprising uncoupling of AHL production and expression of the corresponding synthase in the QS-1 system, which hints that QS regulation does not always follow a classic pattern. Collectively, the results of our study suggest that there are homeostatic regulatory loops provided by the various QS systems in B. thailandensis resulting from transcriptional and posttranscriptional interactions, allowing tightly controlled coordination of the expression of genes. Although we have found new connections and insights on the QS cascade, there are still many questions to be answered. Indeed, further work is needed to comprehend more about the mechanisms behind those links and regulation as well as the implications of recently characterized RsaM-like proteins. The temporal pattern of QS-controlled genes clearly shows that additional factors are involved ( 17 , 20 , 27 ).",
"introduction": "INTRODUCTION Quorum sensing (QS) is a global regulatory mechanism of gene expression depending on bacterial density ( 1 ). Gram-negative bacteria typically possess homologues of the LuxI/LuxR system initially characterized in the bioluminescent marine bacterium Vibrio fischeri ( 2 ). The signaling molecules N -acyl- l -homoserine lactones (AHLs) produced by the LuxI-type synthases accumulate in the environment throughout bacterial growth, providing information on cell density. These AHLs activate the LuxR-type transcriptional regulators that modulate the expression of QS target genes, which usually contain a lux box sequence in their promoter region. These genes include a luxI homologue encoding a LuxI-type synthase generally located in close vicinity of a luxR homologue that codes for a LuxR-type transcriptional regulator, resulting in a typical self-inducing loop of AHLs ( 3 ). Species belonging to the Burkholderia genus generally carry a unique AHL-based QS system referred as the CepI/CepR QS system ( 4 ). The CepI synthase is responsible for the production of N -octanoyl-homoserine lactone (C 8 -HSL), whereas the CepR transcriptional regulator modulates the expression of QS target genes in association with C 8 -HSL, including the cepI gene ( 4 ). Additionally, the cepR gene transcription can be autoregulated as well ( 5 , 6 ). Multiple QS circuitries were also reported for several Burkholderia spp., such as the members of the Bptm group that consists of the nonpathogenic soil saprophyte Burkholderia thailandensis and the closely related pathogens Burkholderia pseudomallei and Burkholderia mallei responsible for melioidosis and glanders, respectively ( 7 – 9 ). QS was reported to be involved in the regulation of several virulence factors in B. pseudomallei and to be essential to its pathogenicity ( 10 , 11 ). B. thailandensis , which is considered the avirulent version of B. pseudomallei ( 12 ), is commonly used as a surrogate model for the study of B. pseudomallei , which is considered a potential bioterrorism agent and whose manipulation is consequently restricted to biosafety level 3 (BSL3) laboratories. The members of the Bptm group contain homologous LuxI/LuxR QS systems that are involved in the biosynthesis of various AHLs ( 13 – 17 ). In B. thailandensis , the LuxI/LuxR QS systems are referred to as the BtaI1/BtaR1 (QS-1), BtaI2/BtaR2 (QS-2), and BtaI3/BtaR3 (QS-3) QS systems. The QS-1, QS-2, and QS-3 systems are also found in B. pseudomallei , whereas the QS-2 system is absent in B. mallei ( 18 ). These species also possess additional orphan luxR homologues, namely, btaR4 ( malR ) and btaR5 in B. thailandensis ( 7 – 9 , 19 ). The QS-1 system is composed of the btaI1 and btaR1 genes that code for the BtaI1 synthase and the BtaR1 transcriptional regulator, respectively. BtaI1 is responsible for the production of C 8 -HSL ( 13 ), and transcription of btaI1 is positively modulated by BtaR1 ( 20 ). The BtaI2 synthase and the BtaR2 transcriptional regulator encoded by the btaI2 and btaR2 genes, respectively, constitute the QS-2 system. BtaR2 directly activates expression of btaI2 involved in both N -3-hydroxy-decanoyl-homoserine lactone (3OHC 10 -HSL) and N -3-hydroxy-octanoyl-homoserine lactone (3OHC 8 -HSL) biosynthesis ( 16 ). The QS-3 system comprises the btaI3 gene encoding the BtaI3 synthase that also catalyzes the synthesis of 3OHC 8 -HSL ( 13 ), as well as the BtaR3 transcriptional regulator, the product of the btaR3 gene located next to btaI3 . The main goal of this study was to dissect the QS regulatory network of B. thailandensis E264 to reveal the interactions existing between the QS-1, QS-2, and QS-3 systems. Besides verifying previously proposed and established interactions, we uncovered several interconnections between the QS-1, QS-2, and QS-3 circuits, providing a comprehensive picture of the complex QS network in B. thailandensis E264. Ultimately, this study will contribute to a better appreciation of the QS regulatory mechanism of the expression of genes in B. thailandensis , and in particular those related to pathogenicity in B. pseudomallei .",
"discussion": "DISCUSSION Although the QS-1, QS-2, and QS-3 systems of B. thailandensis had been previously described ( 13 , 16 , 20 ), a detailed picture of the interactions between the elements composing this complex QS regulatory network was missing. Since the real impact of the BtaR transcriptional regulators on the biosynthesis of their cognate AHLs and expression of adjacent btaI genes was assumed in the literature but almost never confirmed experimentally, we investigated production of AHLs in all Δ btaR mutants and compared it with measurements of the levels of expression of btaI genes. As previously described for B. pseudomallei KHW ( 17 ), we observed variations in the biosynthesis of the main AHLs as well as in the transcription of the AHL synthase-coding genes btaI1 , btaI2 , and btaI3 throughout the growth phases in B. thailandensis E264 ( Fig. 1 ). These observations highlighted the timing of expression of the QS-1, QS-2, and QS-3 systems during the different stages of growth and consequently the existence of potential interactions between these QS circuits. While C 8 -HSL is generally considered the primary AHL produced by Burkholderia spp. ( 4 ) and is indeed predominately detected in stationary-phase cultures of B. pseudomallei K96243 and B. mallei ATCC 23344 ( 15 , 17 ), we confirmed that 3OHC 10 -HSL is actually the most abundant AHL found in B. thailandensis E264 cultures during the different stages of growth, revealing the importance of the QS-2 system in the QS circuitry of B. thailandensis E264 ( Fig. 8 ). FIG 8 Proposed interactions between the QS-1, QS-2, and QS-3 systems. While we confirmed that transcription of btaI2 and biosynthesis of 3OHC 10 -HSL are activated by BtaR2, a stronger activation by 3OHC 10 -HSL indicates that BtaR2 exhibits higher affinity for this AHL than for 3OHC 8 -HSL ( Fig. 6B ), which is also produced by the same synthase ( 16 ). Similarly, the bpsI2 gene that codes for the BpsI2 synthase was also shown to be substantially enhanced by 3OHC 10 -HSL in B. pseudomallei KHW ( 17 ). The fact remains that the levels of expression of btaI2 were similar in the wild-type E264 strain of B. thailandensis and in the non-3OHC 10 -HSL-producing Δ btaI2 mutant ( Fig. 3B ). Considering that 3OHC 8 -HSL is still produced in the absence of BtaI2 ( 16 ), we must conclude that both 3OHC 10 -HSL and 3OHC 8 -HSL can induce the transcription of btaI2 ( Fig. 8 ). Because we confirmed that BtaR2 does not function with C 8 -HSL ( Fig. S3 ), an alternative LuxR-type transcriptional regulator is likely involved in its effect on btaI2 expression, highlighting an interaction between the QS-1 and QS-2 systems. Although both BtaR1 and BtaR3 affect 3OHC 10 -HSL production ( Fig. 3A ), indicating that regulation of the biosynthesis of this AHL implies dynamic coordination between the B. thailandensis E264 QS-1, QS-2, and QS-3 circuits ( Fig. 8 ), neither one has an effect on btaI2 expression ( Fig. 3B ). Nevertheless, Majerczyk et al. ( 20 ) demonstrated that btaR2 expression is stimulated by 3OHC 8 -HSL, and we determined that the transcription of this gene is in fact affected by the absence of all AHLs found in B. thailandensis E264 ( Fig. 5B ). Thus, we hypothesize that BtaR1 and BtaR3 act indirectly through btaR2 control. We also do not exclude the possibility that additional transcriptional and/or posttranscriptional regulators are involved in the modulation of the QS-2 system. Interestingly, this system contains an additional gene between btaI2 and btaR2 that is conserved in the Burkholderia genus ( 21 ). It encodes a hypothetical protein that is 37% identical to the B. cenocepacia J2315 Bc RsaM ( 22 ), a homologue of the QS repressor RsaM originally identified in the plant pathogen Pseudomonas fuscovaginae ( 23 ), which we consequently renamed RsaM2 ( Fig. S5 ). Accordingly, we observed that C 8 -HSL, 3OHC 10 -HSL, and 3OHC 8 -HSL concentrations were all increased in an rsaM2 mutant compared to the wild-type strain ( 24 ), indicating that RsaM2 likely intervenes in the regulation of all QS systems of B. thailandensis E264. 10.1128/mBio.01861-17.5 FIG S5 Genetic organization of the QS regulatory genes in B. thailandensis E264. btaI1 and btaR1 are not located next to each other and are divergently transcribed in B. thailandensis E264. The promoter region of btaI1 contains a putative lux box sequence centered 73.5 bp upstream of the btaI1 translation start site (CCCTGTAAGGGTTAACAGTT). btaI2 and btaR 2 are also not located next to each other and are transcribed in the same direction on the genome of B. thailandensis E264. The promoter region of btaI2 contains a putative lux box sequence centered 65.0 bp upstream of the btaI2 translation start site (ACCTGTAGAAATCGTAGT). btaI3 and btaR3 are also transcribed in the same direction and are located next to each other in B. thailandensis E264. Download FIG S5, PDF file, 0.1 MB . Copyright © 2017 Le Guillouzer et al. 2017 Le Guillouzer et al. This content is distributed under the terms of the Creative Commons Attribution 4.0 International license . As described previously for the B. pseudomallei KHW BpsI and B. mallei ATCC 23344 BmaI1 synthases ( 11 , 15 ), Chandler et al. ( 13 ) demonstrated that BtaI1 is responsible for C 8 -HSL production. In agreement with the finding that the B. pseudomallei K96243 BpsR and B. mallei ATCC 23344 BmaR1 transcriptional regulators directly activate the BpsI- and BmaI1-encoding genes in response to C 8 -HSL, respectively ( 15 , 25 ), Majerczyk et al. ( 20 ) reported that btaI1 transcription is positively modulated by BtaR1. We observed a strong BtaR1-dependent induction of btaI1 through C 8 -HSL ( Fig. S1 ) and confirmed that the QS-1 system responds best toward its cognate AHL ( Fig. 6A ). While we demonstrated that BtaR1 constitutes the main regulator of btaI1 expression, we assume that BtaR1 represents the main regulator of C 8 -HSL biosynthesis as well. An uncoupling of AHL production and expression of the corresponding synthase was also reported in a Burkholderia RsaM-deficient strain ( 22 , 26 ). Bc RsaM from B. cenocepacia H111 was indeed described as an important repressor of C 8 -HSL biosynthesis and shown to activate the transcription of cepI and cepR encoding the LuxI-type synthase CepI and the LuxR-type transcriptional regulator CepR, respectively ( 22 , 26 ). Interestingly, a gene encoding a hypothetical protein sharing 63% identity with the B. cenocepacia J2315 Bc RsaM, hence called RsaM1, was also found between btaI1 and btaR1 ( Fig. S5 ). Investigating the effect of RsaM1 on the biosynthesis of AHLs in B. thailandensis E264 showed that C 8 -HSL is overproduced in an rsaM1 mutant compared to the wild-type strain ( 24 ), revealing a possible link between the QS-1 system and RsaM1. Additional experiments will be necessary to fully understand the mechanisms involved in the regulation of the QS-1 system as well as the implications of the RsaM-like proteins in B. thailandensis E264. We demonstrated that the biosynthesis of C 8 -HSL and transcription of btaI1 are both negatively controlled by BtaR2 ( Fig. 2 ). Because no overexpression of the btaI1 gene was observed in the Δ btaI2 mutant background, we assume that BtaR2 represses the QS-1 system in the absence of its ligands. This contrasts with the BtaR3-dependent regulation of btaI1 transcription in conjunction with 3OHC 8 -HSL, as well as with 3OHC 10 -HSL, albeit to a lesser extent ( Fig. 8 ). This is also further supported by the fact that BpsR3 was reported to directly activate bpsI in response to both 3OHC 8 -HSL and 3OHC 10 -HSL, with 3OHC 8 -HSL eliciting the strongest response from BpsR3 ( 17 ). Considering that bmaI1 was also shown to be directly controlled by BmaR3/3OHC 8 -HSL ( 14 ), we suppose that BtaR3 directly activates expression of the btaI1 gene as well. However, we believe the effect of BtaR3 on the QS-1 system is more complex. While the bpsR gene encoding BpsR was reported to be positively autoregulated ( 11 ), we determined that btaR1 expression is repressed by QS ( Fig. 5A ). Thus, negative regulation of C 8 -HSL biosynthesis by BtaR3 could be linked to btaR1 modulation. Altogether, these observations further highlight the existence of interactions between the QS-1, QS-2, and QS-3 circuits and reveal that the timing of expression of the QS-1 system is dependent on both the QS-2 and QS-3 systems ( Fig. 8 ). This might contribute to the successive activation of the B. thailandensis E264 QS circuits observed throughout bacterial growth. Similarly to the B. pseudomallei KWH BpsI3 and B. mallei ATCC 23344 BmaI3 synthases, BtaI3 was shown to produce 3OHC 8 -HSL ( 13 , 14 , 17 ). While the B. pseudomallei KHW BpsR3 and B. mallei ATCC 23344 BmaR3 transcriptional regulators specifically respond to 3OHC 8 -HSL, the bpsI3 and bmaI3 genes were not reported to be activated by BpsR3 and BmaR3, respectively, in conjunction with 3OHC 8 -HSL ( 14 , 17 ). Here, in B. thailandensis E264, we demonstrated that the transcription of btaI3 is positively controlled by BtaR3 and activated by 3OHC 8 -HSL ( Fig. S2 ). However, 3OHC 8 -HSL-dependent activation of btaI3 seems to be conditioned by the presence of other AHLs ( Fig. S4 ). The interaction between BtaR3 and 3OHC 8 -HSL, necessary to activate btaI3 expression, could be impeded by a competitive inhibition exerted by another AHL, as already proposed for B. pseudomallei KHW ( 17 ). In addition, we observed that btaI3 expression is activated by 3OHC 10 -HSL, albeit to a lesser extent ( Fig. 6C ). Indeed, the BtaR3-controlled genes identified in transcriptomic analyses were also generally affected by both 3OHC 8 -HSL and 3OHC 10 -HSL ( 20 ). This further supports the idea that BtaR3 functions with these two AHLs ( Fig. 8 ). Considering that BpsI3 and BmaI3 were both shown to produce 3OHC 10 -HSL in addition to 3OHC 8 -HSL ( 14 , 17 ), it is possible that BtaI3 intervenes in the biosynthesis of 3OHC 10 -HSL in B. thailandensis E264 as well. Remarkably, positive 3OHC 8 -HSL- and 3OHC 10 -HSL-dependent regulation of btaI3 occurred in the stationary growth phase ( Fig. 7 ), in agreement with the expression profile of this gene. Conversely, activation of btaI2 transcription by these AHLs was mainly observed during logarithmic growth. We thus hypothesize that the QS-3 system regulates the QS-2 system targets by producing 3OHC 8 -HSL in stationary phase, whereas production of this AHL by the QS-2 system occurs essentially during the exponential phase, implying a coordination between the QS-2 and QS-3 systems ( Fig. 8 ). Additionally, it seems that 3OHC 8 -HSL is produced by BtaI2 at the expense of 3OHC 10 -HSL. This would explain why there is an overlap between these QS circuits when it comes to genes modulated by 3OHC 8 -HSL and 3OHC 10 -HSL ( 20 ). Importantly, while sharing common AHLs, the QS-2 and QS-3 systems are apparently not transcriptionally linked. The BtaR1/C 8 -HSL-dependent control of btaI3 transcription, which starts in the exponential growth phase, is consistent with the idea that the QS-1 system is required for the expression of btaI3 ( 20 ), and might also account for the belated activation of the QS-3 circuit in comparison with the QS-1 and QS-2 systems. This again illustrates the successive expression of these QS circuits and points toward an interdependence between the QS-1 and QS-3 systems ( Fig. 8 ). Such an interconnection has already been observed among the members of the Bptm group, as bpsI3 transcription was reported to be stimulated by the BpsI/BpsR QS system ( 17 ). Nevertheless, the precise regulatory mechanism directing the QS-3 system through BtaR1 is currently unknown. While BtaR1 seems to act by activating btaI3 transcription, we propose that the negative impact of BtaR1 on 3OHC 8 -HSL biosynthesis does not result from a direct interaction with the btaI3 promoter but rather could imply the effect of BtaR1 on the level of btaR3 as previously suggested ( 20 ). Additional transcriptional and/or posttranscriptional regulators might also be involved in the BtaR1-dependent modulation of the QS-3 system. Conclusion. The study described here provides for the first time an exhaustive portrait of the interplay between the QS-1, QS-2, and QS-3 systems in B. thailandensis E264 ( Fig. 8 ). We observed an interdependence between the QS-1 and QS-2 systems. While we confirmed that the QS-3 system is controlled by BtaR1, we also found that BtaR3 modulates the QS-1 system, which indicates that those two systems are linked. Interestingly, such an interaction between the QS-1 and QS-3 systems seems to be conserved in the closely related species of the Bptm group ( 14 , 17 , 20 ). Interestingly, the QS-2 and QS-3 systems that share common AHLs seem not to be transcriptionally linked, but instead they are temporally connected by their common AHLs. We also highlighted a surprising uncoupling of AHL production and expression of the corresponding synthase in the QS-1 system, which hints that QS regulation does not always follow a classic pattern. Collectively, the results of our study suggest that there are homeostatic regulatory loops provided by the various QS systems in B. thailandensis resulting from transcriptional and posttranscriptional interactions, allowing tightly controlled coordination of the expression of genes. Although we have found new connections and insights on the QS cascade, there are still many questions to be answered. Indeed, further work is needed to comprehend more about the mechanisms behind those links and regulation as well as the implications of recently characterized RsaM-like proteins. The temporal pattern of QS-controlled genes clearly shows that additional factors are involved ( 17 , 20 , 27 )."
} | 5,439 |
29053149 | PMC5776459 | pmc | 2,954 | {
"abstract": "Rising sea surface temperature is the main cause of global coral reef decline. Abnormally high temperatures trigger the breakdown of the symbiotic association between corals and their photosynthetic symbionts in the genus Symbiodinium . Higher genetic variation resulting from shorter generation times has previously been proposed to provide increased adaptability to Symbiodinium compared to the host. Retrotransposition is a significant source of genetic variation in eukaryotes and some transposable elements are specifically expressed under adverse environmental conditions. We present transcriptomic and phylogenetic evidence for the existence of heat stress-activated Ty1- copia -type LTR retrotransposons in the coral symbiont Symbiodinium microadriaticum . Genome-wide analyses of emergence patterns of these elements further indicate recent expansion events in the genome of S. microadriaticum. Our findings suggest that acute temperature increases can activate specific retrotransposons in the Symbiodinium genome with potential impacts on the rate of retrotransposition and the generation of genetic variation under heat stress."
} | 286 |
37467178 | PMC10355445 | pmc | 2,957 | {
"abstract": "DNA origami purification is essential for many fields, including biophysics, molecular engineering, and therapeutics. The increasing interest in DNA origami has led to the development of rate-zonal centrifugation (RZC) as a scalable, high yield, and contamination-free method for purifying DNA origami nanostructures. RZC purification uses a linear density gradient of viscous media, such as glycerol or sucrose, to separate molecules according to their mass and shape. However, many methods for creating density gradients are time-consuming because they rely on slow passive diffusion. To expedite the preparation time, we used a LEGO gradient mixer to generate rotational motion and rapidly create a quasi-continuous density gradient with a minimal layering of the viscous media. Rotating two layers of differing concentrations at an angle decreases the time needed to form the density gradient from a few hours to minutes. In this study, the density gradients created by the LEGO gradient mixer were used to purify 3 DNA origami shapes that have different aspect ratios and numbers of components, with an aspect ratio ranging from 1:1 to 1:100 and the number of components up to 2. The gradient created by our LEGO gradient mixer is sufficient to purify folded DNA origami nanostructures from excess staple strands, regardless of their aspect ratios. Moreover, the gradient was able to separate DNA origami dimers from DNA origami monomers. In light of recent advances in large-scale DNA origami production, our method provides an alternative for purifying DNA origami nanostructures in large (gram) quantities in resource-limited settings.",
"introduction": "Introduction DNA origami, a method of self-assembling complex nanostructures from long, single-stranded DNA (scaffold strands) and large excess of shorter oligonucleotides (staple strands) [ 1 – 8 ], has proven to be a robust and efficient method for generating nanostructures with arbitrary shapes at ∼5 nm resolution. Furthermore, the exquisite positional control of DNA origami enables precise patterning of biomolecules and inorganic molecules [ 8 – 11 ]. The programmable DNA origami has been used in fields such as medicine [ 12 – 16 ], super-resolution microscopy [ 17 – 20 ] and electronics [ 21 – 23 ]. Although efforts have been made to optimize DNA origami assembly [ 24 , 25 ], the yield of well-folded origami for complex 3D structures is far less than 100% [ 26 ]. Most applications of DNA origami, however, require uncontaminated samples that are free of staple strands and misfolded structures [ 26 – 28 ]. Thus, for downstream applications, a purification step is added after the folding step to isolate the well-folded structures. Rate-Zonal Centrifugation (RZC) is a high-yield, contamination-free method for purifying DNA origami [ 27 ]. The technique subjects a linear density gradient to high centrifugal force in order to separate heterogeneous molecules by their distinct mass and shape [ 29 ]. Within the context of DNA origami, RZC can separate well-folded structures from other undesired species (misfolded structures, staple strands, and aggregates) as a purification step. RZC has several advantages compared to other purification methods, such as agarose gel electrophoresis (AGE). RZC keeps the samples in an aqueous solution throughout the process, is free from contaminants such as agarose gel residues, and is scalable to accommodate a large amount of sample [ 27 ]. In most cases, RZC purification requires fewer steps than multi-round spin filter purification or agarose gel extraction [ 30 ]. RZC purification starts with the traditional preparation of the density gradient, which can be time consuming [ 27 , 31 ]. There are several methods to prepare a density gradient of glycerol: (1) layering two solutions of different glycerol concentrations in a tube and resting the tube horizontally for 1–2 hours to allow the glycerol to passively diffuse, (2) layering several solutions of different glycerol concentrations and resting the tube upright so that the layers can passively diffuse, and (3) mixing the glycerol gradient using a commercially available gradient mixer. Since the first two methods employ passive diffusion, they do not require expensive equipment but come with a lengthy preparation period. Utilizing a gradient mixer reduces the preparation time, but it is costly and may require additional equipment or training. Thus, there is a need for a method for preparing a density gradient that is both fast and cost-effective. Here, we report a low-cost method for creating a linear glycerol density gradient that accelerates diffusion via a simple rotational motion. The prototype of the instrument was made using LEGO EV3 Mindstorms materials, and the diffusion process facilitated by the LEGO gradient mixer takes only one minute. To develop the gradient, a low-concentration solution of glycerol is gently layered on top of a high-concentration glycerol solution inside a centrifuge tube. The glycerol-filled tubes are then loaded into the LEGO gradient mixer and the program is initiated to tilt the glycerol-filled tubes horizontally and rotate them for 1 minute at low rpm (20 rpm). The LEGO gradient mixer then returns to its upright position, yielding the density gradients for the RZC purification. The DNA origami samples are loaded on top of the newly created glycerol gradients. The tubes are then transferred to an ultracentrifuge for RZC purification [ 27 ]. Fractions are collected from the RZC result to be analyzed using AGE, and the purified structures are verified using atomic force microscopy (AFM).",
"discussion": "Discussion DNA nanotechnology has the potential to advance research in the fields of microscopy [ 17 – 20 ], nanomedicine [ 12 – 16 ], and even molecular-scale electronics [ 21 – 23 ]. Its pace toward industry applications is paired with successful efforts to remove the key obstacle of scalability for both production [ 34 – 36 ] and purification [ 27 ] of DNA nanostructures. The LEGO gradient mixer presented here is an effective complement to the current state-of-the-art RZC method to purify DNA origami. The LEGO gradient mixer was able to produce a concentration gradient in a relatively short time compared to other gradient mixers [ 27 , 31 ]. In combination with the speed and ease of use of the LEGO gradient mixer, RZC purification could help pave the way for large-scale DNA nanostructure production that will potentially enable the separation of functional DNA nanostructures from the side products of bioconjugation reactions, such as excess streptavidins, quantum dots, gold nanoparticles, aptamers, and other moeties. The gradient produced by this method was able to purify 6-hb monomer from its staple strands and 6-hb dimer from its precursor monomer, demonstrating its ability to separate molecules with 1:2 size ratio. Although the purified dimer shows only 63±6% purity ( Fig 5J ), the observed monomer in the purified dimer likely results from the difficulty of extracting the dimer sample manually from the small gradient volume in which the experiment is conducted. Taking note that we used less than ideal equipment for the purification method, i . e . long neck gel loading tips, the yield could be significantly improved by upgrading or automating the fractionation technique. An improved and/or automated fractionation technique could produce more consistent and higher yield results and would therefore be integral to optimizing the effectiveness of RZC purification. Similar success was observed when the RZC purification and LEGO mixed gradient were applied to the purification of a DNA origami snubcube ( Fig 4H ). Though the Amicon spin column purification performed better in purifying the snubcube from its staple strands, the purification method using the LEGO mixed gradient offered comparable results. Translating the success of RZC from a more simplified 6-hb monomer and dimer to a more complex shape, like the snubcube, makes RZC a promising purification method for a wide range of DNA origami products. The strength of RZC is its ability to purify molecules that require finer precision, such as separating a 6-hb dimer from its monomer. We note that the reported method can be further improved to reduce the cost. The current cost of the EV3 LEGO Mindstorms pieces used to make this device is $349.99, while gradient mixers are typically quoted at around $500.00 to $650.00 for a Millipore Sigma gradient mixer (cat no. Z340391) and >$2K for a GradientMaster (Biocomp Instruments). The main components of the LEGO gradient mixer are a motor, a servo, and a custom holder for the tubes, all of which do not require a high degree of precision to be effective. Hence, the price to create such a LEGO gradient mixer can be reduced further by an order of magnitude (S2 Table in S1 File ). To lower the cost further, such machines can be powered manually using some gear and scaffolding materials. In the spirit of frugal science, our discovery could help density gradient centrifugation become a more accessible tool for scientific communities around the world. The myriad applications of density gradients to fractionate DNA origami, viral particles, and a variety of macromolecules [ 37 ] have been critical in the preparation of viruses for vaccines and other immunotherapy products. The methods introduced in this paper will contribute to science education and research in resource-limited settings."
} | 2,364 |
29259923 | PMC5723315 | pmc | 2,959 | {
"abstract": "Quorum Sensing (QS) mechanism, a bacterial density-dependent gene expression system, governs the Serratia marcescens pathogenesis through the production of virulence factors and biofilm formation. The present study demonstrates the anti-quorum sensing (anti-QS), antibiofilm potential and in vivo protective effect of phytol, a diterpene alcohol broadly utilized as food additive and in therapeutics fields. In vitro treatment of phytol (5 and 10 μg/ml) showed decreasing level of biofilm formation, lipase and hemolysin production in S. marcescens compared to their respective controls. More, microscopic analyses confirmed the antibiofilm potential of phytol. The biofilm related phenomenons such as swarming motility and exopolysccharide productions were also inhibited by phytol. Furthermore, the real-time analysis elucidated the molecular mechanism of phytol which showed downregulation of fimA, fimC, flhC, flhD, bsmB, pigP , and shlA gene expressions. On the other hand, the in vivo rescue effect of phytol was assessed against S. marcescens associated acute pyelonephritis in Wistar rat. Compared to the infected and vehicle controls, the phytol treated groups (100 and 200 mg/kg) showed decreased level of bacterial counts in kidney, bladder tissues and urine samples on the 5th post infection day. As well, the phytol treatment showed reduced level of virulence enzymes such as lipase and protease productions compared to the infected and vehicle controls. Further, the infected and vehicle controls showed increasing level of inflammatory markers such as malondialdehyde (MDA), nitric oxide (NO) and myeloperoxidase (MPO) productions. In contrast, the phytol treatment showed decreasing level of inflammatory markers. In histopathology, the uninfected animal showed normal kidney and bladder structure, wherein, the infected animals showed extensive infiltration of neutrophils in kidney and bladder tissues. In contrast, the phytol treatment showed normal kidney and bladder tissues. Additionally, the toxic effect of phytol (200 mg/kg) was assessed by single dose toxicity analysis. No changes were observed in hematological, biochemical profiles and histopathological analysis of vital organs in phytol treated animals compared to the untreated controls. Hence, this study suggested the potential use of phytol for its anti-QS, antibiofilm and anti-inflammatory properties against S. marcescens infections and their associated inflammation reactions.",
"introduction": "Introduction Urinary tract infection (UTI) is one among the utmost commonly detected infections in clinical settings (Hvidberg et al., 2000 ). In divergence to men, women are more vulnerable to UTI (Derbie et al., 2017 ). Almost 1 in 3 women will have had UTI by the age of 24 years. Nearly half of all female population will experience with UTI throughout their lifetime. The expenses have extended for the treatment of UTI infection is $2.4 billion a year by means of 4.5–6.8 million cases in worldwide (Foxman and Brown, 2003 ). The way of UTI is well-known, which starts from urethral to bladder and then moving up the ureters into the kidneys. Cystitis, a predominant UTI, takes place in the bladder of the lower urinary tract whereas the pyelonephritis, a severe kidney infection, that targets the upper urinary tract. Pyelonephritis is a potentially life threatening infection that often leads to renal damaging (Katsiari et al., 2012 ). Pyelonephritis occurs subsequently a sequence of events carrying the bacteria from outside of the human body, up to the bladder and finally settles down into the kidneys. If the pyelonephritis is not treated, this can lead to severe renal abscesses and sepsis along with renal failure (Ramakrishnan and Scheid, 2005 ). Most pyelonephritis infections are occurred by bacterial pathogens ascent through the urethra and urinary bladder. The etiologic agents of pyelonephritis are Escherichia coli, Proteus mirabilis, Pseudomonas aeruginosa , and Serratia marcescens (Ohno et al., 2003 ; Mittal et al., 2009 ; Chen et al., 2012 ; Kufel et al., 2016 ). Serratia marcescens , a Gram-negative bacterium, belongs to the family Enterobacteriaceae, frequently isolated from urinary and respiratory tracts and it can function as an opportunistic pathogen in immunocompromised patients (Kida et al., 2007 ). The infections caused by S. marcescens are hard to treat since it possesses inherent resistance to an extensive variety of antibiotics (Lee et al., 2005 ; Kim et al., 2013 ; Leclercq et al., 2013 ; Liou et al., 2014 ; González-Juarbe et al., 2015 ). Development of antibiotic resistance in S. marcescens demands the urgent need for the alternative treatment approaches. Host-pathogen interaction and the ability of pathogens to modify the host response is a crucial factor for establishing successful infections (Youn et al., 1992 ). This capability of a pathogen is typically attributed to their ability to secrete several of virulence factors and to alter host immune response (McMillen et al., 1996 ). The prominence of these responses has been exposed in several biological processes through an assortment of inflammatory mediators and cytokines, which include tissue inflammation, wound healing and immune defense (McMillen et al., 1996 ; Rumbaugh et al., 2004 ). Recently, several reports specified that the quorum sensing (QS) mediated virulence factors are important for successful establishment of bacterial infection in animal models (Kumar et al., 2009 ; Gupta et al., 2013b ). QS is a vital global gene regulatory machinery in bacteria that allows discrete bacteria to coordinate their virulence behavior in a cell density depended manner, which depends on self-produced signaling molecules called autoinducers (Rumbaugh et al., 1999 ). Serratia marcescens has a well described QS system (SmaI/SmaR) which utilizes different homoserine lactones (HSLs) such as C4-HSL, C6-HSL, and C8-HSL as signal molecules and governs the secretion of extensive range of virulence factors such as prodigiosin, lipase, protease, chitinase, nuclease, siderophore production, hemolysin production and most importantly biofilm formation (Hines et al., 1988 ; Eberl et al., 1996 ; Horng et al., 2002 ; Rice et al., 2005 ). Research targeting the bacterial QS system has paid a great deal of attention for the identification of effective anti-quorum sensing (anti-QS) and antibiofilm agents. These anti-QS agents aim the virulence factors production rather the growth of the bacterial pathogen, hence the emergence of selective pressure for the development of antibiotic resistance strain is nullified. Thus, it is foreseen that the inhibition of such QS mechanism would warrant as an effective approach to reduce the S. marcescens pathogenicity and infection (Labbate et al., 2007 ). Recently, numerous studies have been continuously reported the anti-QS and antibiofilm potential of several natural compounds from plant origin. Plant sources play a vital role in delivering the novel drugs candidates in medicinal field. Phytol, a diterpene alcohol compound majorly found in essential oils, extensively used as fragrant ingredient in shampoos, cosmetics, fragrances and other toiletries (Islam et al., 2015 ). As well, it is also used in the production of Vitamin K and E. In therapeutic field, phytol has shown antioxidant and antinociceptive activities as well as antiallergic, antimicrobial, antiradical, anti-cholinesterase, antiamyloidogenic, and anti-inflammatory properties along with adequate safety (Inoue et al., 2005 ; Lim et al., 2006 ; Ryu et al., 2011 ; Santos et al., 2013 ; Pejin et al., 2014 ; Lee et al., 2016 ; Sathya et al., 2017 ). Also, the phytol is a tremendous immuno stimulant, in respect of long term memory stimulation of both acquired and innate immunity. However, the report on anti-QS potential of phytol against bacterial pathogens is very much scarce (Pejin et al., 2015 ) and the protective effect of phytol on bacterial pathogens in animal model is nil. Based on these facts, this pioneering study primarily focused on assessing the in vitro anti-QS and antibiofilm potential of phytol against S. marcescens and the in vivo protective effect of phytol against S. marcescens associated acute pyelonephritis infection in rat model.",
"discussion": "Discussion UTI is an infection occurred wherever in the urinary system typically exposed to bacterial pathogens. Once bacterial pathogens reach the kidney through ascending infection, they are capable to adhere to the urothelium before raiding the renal tissue with subsequent pyelonephritis (Nickel et al. ( 1987 ). Such sort of infections are reported to be caused by Gram-negative bacteria like P. mirabilis, E. coli, Klebsiella pneumoniae, P. aeruginosa, S. marcescens , and Gram-positive bacteria such as Staphylococcus aureus and Enterococcus faecalis (Su et al., 2003 ; Behzadi et al., 2010 ; Kaur et al., 2014 ). Among which, S. marcescens is an important human opportunistic bacterial pathogen, causing numerous nosocomial infections such as respiratory tract infections, blood stream infections, ocular infections and most importantly urinary tract infections (Hejazi and Falkiner, 1997 ). It secretes array of virulence factors and forms biofilm via signal mediated QS mechanism. In our previous study, we assessed the anti-QS potential of phytol through primary assays such as prodigiosin production, protease inhibition assays and biofilm cells quantification by crystal violet assay (Srinivasan et al., 2016 ). Nevertheless, the present study further evaluated the potentials of phytol against S. marcescens by assessing various virulence assays such as biofilm cells quantification by XTT reduction assay, microscopic analyses of biofilm formation, swarming motility analysis, lipase, hemolysin and EPS quantification assays. In addition, the current study elucidated the molecular mechanism of phytol on QS system in S. marcescens through real-time expression analysis and confirmed its in vivo protective effect on acute pyelonephritis infection in rat model with satisfactory safety evaluated by single dose toxicity studies. Biofilms are the aggregation of microorganism, wherein the microbial cells stick to each other on biotic and abiotic surfaces and composed of extracellular DNA, polysaccharides and proteins (Abdel-Aziz and Aeron, 2014 ). Therefore, we tested the effect of phytol on biofilm formation and EPS production in S. marcescens by XTT reduction and EPS quantification assays. The obtained results showed decreasing level of metabolically active cells involved in biofilm formation and EPS production in phytol treatment compared to their respective controls (Figures 2A , 4A ). Further, the light and CLSM (2, 2.5, and 3 D) images confirmed the antibiofilm potential of phytol, in which, the 5 and 10 μg/ml of phytol treatment showed disintegration of biofilm formation. Divergently, the control slides showed thick coating of biofilm formation (Figures 3A,B ). Our results are going well with the findings of the previous researches, who have reported that the morin reduced the metabolically active cells involved in Listeria monocytogenes biofilm formation (Sivaranjani et al., 2016 ) and marine bacterial extract G-16 effectively inhibited the S. marcescens EPS production (Padmavathi et al., 2014 ). Several bacterial pathogens simultaneously grow and spread rapidly over a surface through the pattern of movement called swarming motility. This diminishes competition between bacterial cells for nutrients and speeding their growth (Kaiser, 2007 ). This typical virulent phenomenon in S. marcescens plays a vital role in catheter associated urinary tract infections. In this bacterial species the phenomenon of swimming and swarming motility is associated with QS. Hence, an attempt was made to examine the QSI potential of phytol in inhibiting the swarming movement. Results of the current study showed vigorous swarming motility in the untreated S. marcescens control plate, wherein the 5 and 10 μg/ml of phytol treatment showed concentration dependent swarming motility inhibition (Figure 3C ). Consistent with this result, Srinivasan et al. ( 2016 ) have reported that Piper betle extract effectively inhibited the S. marcescens swarming motility in a concentration dependent manner. Lipase is the secreted extracellular virulence enzyme in S. marcescens and their production is regulated by QS. Hemolysin production is accountable for the pathogenesis of various bacterial pathogens. Hemolysin produced by S. marcescens (ShlA), is a group of pore forming toxins, targets the cell membrane permeability (Shimuta et al., 2009 ). The result of lipase and hemolysin inhibition assays indicated a significant ( P ≤ 0.0005) decline in lipase and hemolysin production in S. marcescens upon treatment with 5 and 10 μg/ml of phytol (Figures 4A,B ). Previously, Anethum graveolens extract and farnesol were tested for their effects on lipase and hemolysin production in S. marcescens and P. aeruginosa respectively, and which showed promising lipase and hemolysin inhibitory properties (Hassan Abdel-Rhman et al., 2015 ; Salini and Pandian, 2015 ). Further to understand the anti-QS and antibiofilm potential of phytol at molecular level and to support the outcome of in vitro results, the real-time PCR analysis was performed. It is known that fimA and fimC are the major fimbrial subunits in S. marcescens . In 2007, a study done by Labbate et al. disclosed that the fimA disruption mutant unable to produce fimbriae and likewise they confirmed the absence of fimbrial structure in S. marcescens by electron microscopy. The products of the flhDC master operon, FlhD and FlhC are global gene regulators in S. marcescens , which expressed several inherent determinants such as cell differentiation, cell division, swimming and swarming motilities (Liu et al., 2000 ). Therefore, the impact of phytol on the fimA, fimC, flhC , and flhD gene expression levels were tested and the obtained real-time data showed a substantial downregulation of these fimbrial and motility genes expression in S. marcescens . The bsmB is a QS controlled virulence gene in S. marcescens . Labbate et al. ( 2007 ) reported that the bsmB mutant lacked biofilm formation, lipase, protease and S-layer protein productions. Phytol treatment decreases the expression level of bsmB gene up to 0.36-fold compare to the control. The RssA-RssB (RssA-sensor kinase and RssB -response regulator) is a two component system and it negatively regulates the S. marcescens swarming motility. RssB binds directly to the flhDC promoter and suppresses the flhDC transcription, leading to reduced production of hemolysin and flagellar mediated motilities (Lin et al., 2010 ). In Ang et al. ( 2001 ) stated that the overexpression of rsmA gene in S. marcescens inhibits the swarming motility and prodigiosin production. The pigP is the master transcriptional regulator and which controls the regulation of prodigiosin pigment production in S. marcescens under the QS mechanism (Gristwood et al., 2011 ). RssB binds directly to the promoter region of the pig operon, leading to negative regulation of prodigiosin production (Soo et al., 2014 ). The outcome of real-time data showed upregulation of rssB and rsmA genes expression and support the in vitro data of hemolysin, swarming motility and prodigiosin inhibition due to their binding on flhDC and pigP promoter regions. Likewise, phytol decreases the expression level of pigP gene upto 0.48 fold compare to the control. ShlA is a key virulence factor of S. marcescens , which has shown to wield cytotoxic effects on fibroblasts and epithelial cells (Di Venanzio et al., 2014 ) and shlBA mutant strains were extremely reduced in virulence in mice, Drosophila melanogaster and Caenorhabditis elegans models (Kurz et al., 2003 ). In S. marcescens , hemolysis and swarming motility are co-regulated (Shanks et al., 2013 ). In the current study the phytol inhibited the hemolysin production along with swarming motility inhibition. Similarly, the real-time data showed downregulation of shlA gene upon treatment with phytol (Figure 5 ). The recent reports stated that the QS mediated virulence factors are very important for establishment of successful UTI infection in animal models (Kumar et al., 2009 ; Gupta et al., 2013b , 2016 ; Saini et al., 2015 ). Only limited studies specified the pathogenesis of S. marcescens in animal models and also no reports are available on the protective effect of plant extracts or pure compounds against S. marcescens associated infection in animal models. To the best of our knowledge, the present study is the first of its kind has been made with a prime objective to establish the S. marcescens associated acute pyelonephritis in rat and assessing the protective effect of phytol against acute pyelonephritis induced rat. After successful establishment of acute pyelonephritis in rat model, the bacterial count in phytol treated and untreated rats were quantified by bacteriological assay. The infected control had 8.28 × 10 4 , 7.2 × 10 4 , and 3.72 × 10 4 CFU in kidney, bladder and urine samples, respectively compare to the 200 mg/kg body weight of phytol treated group in which 1.78 × 10 4 , 0.96 × 10 4 , and 0.5 × 10 4 CFU were observed in kidney, bladder and urine samples, respectively (Figures 6B–D ). This corresponds to nearly 4.6, 7.5 and 7.4 fold decrease in bacterial count in phytol (200 mg/kg body weight) treated kidney, bladder and urine samples respectively, compared to the infected control. These results correlate with the findings of Hvidberg et al. ( 2000 ), who have reported that the antibiotic gentamicin treatment significantly decreased the bacterial count in kidney, bladder and urine samples in UTI induced mice compare to the infection control. Colonization of bacterial pathogens on host tissue during the early stage of infection is an essential factor for the establishment of very infection. Virulence factors produced by the bacterial pathogens help in the host colonization and subsequent infection progress. The extracellular virulence enzyme protease plays a pivotal role in the pathogenesis of S. marcescens during infection and induces interleukin-6 and interleukin-8 mRNA expression through protease-activated receptor 2 (PAR-2) (Kida et al., 2007 ). A study made by Lyerly and Kreger ( 1983 ) state that the highly purified protease enzyme obtained from S. marcescens induced the acute pneumonia in mice and guinea pigs. A finding made by Ishii et al. ( 2014 ) revealed that the protease intricate in the pathogenesis of S. marcescens and leads to a huge loss of hemolymph in silkworm larvae. Like protease, the extracellular lipase enzyme also an extensive virulence factor and which involved in the pathogenesis of S. marcescens (Hejazi and Falkiner, 1997 ). Both of these virulence enzyme productions are controlled by the QS mechanism (Labbate et al., 2007 ). In support, the result stated by Elsheikh et al. ( 1987 ) indicated that the virulence enzyme protease enhances the pathogenesis of P. aeruginosa in experimental mouse burn infection. In Gupta et al. ( 2013a ) suggested that the QS mediated virulence enzymes such as protease and elastase are involved in the establishment and colonization of P. aeruginosa in mice during experimental UTI. Therefore, the inhibitory effect of phytol on virulence enzyme production in rat acute pyelonephritis model was evaluated. As expected the phytol treatment showed decreased level of protease and lipase enzymes production in both kidney and bladder tissues compared to the infected and vehicle controls (Figure 7 ). The extreme reduction in virulence enzyme productions of kidney and bladder tissues in phytol treated groups is go well with bacteriological assay. Hence, it is envisaged that the decreasing level of virulence enzymes in phytol treated groups might be due to the decreasing level of invading S. marcescens cells. MDA is an indicator of lipid peroxidation and which is a steady product of oxidative stress of reactive oxygen species on unsaturated fatty acid, a vital constituent of cell membrane. In the current study, the kidney and bladder tissues from infected and vehicle control groups showed a substantial increase in MDA level on 5th p.i.d, whereas the phytol treated groups showed decreasing level of MDA production in kidney and bladder tissues (Figure 8A ). Consistent with our results, synergistic combination of azithromycin and ciprofloxacin has been shown to decrease the MDA level in kidney tissue homogenates of P. aeruginosa infected mice on the 3rd and 5th p.i.d (Saini et al., 2015 ). MPO is an enzyme deposited in azurophilic granules of polymorphonuclear neutrophils and macrophages, which released during inflammatory process and oxidative stress into extracellular fluid. The MPO is a possible pathological marker for the confirmation of inflammation (Loria et al., 2008 ). In the present study, the MPO level was considerably low in case of infected rats treated with phytol compare to the infected and vehicle controls in both kidney and bladder tissues (Figure 8B ). The results of MPO assay go well with the findings of Vadekeetil et al. ( 2016 ), who have reported that the ajoene-ciprofloxacin combination effectively decreasing the MPO production in the mice infected from P. aeruginosa biofilm associated murine acute pyelonephritis. NO is produced by a different cell types by NO synthases, which are involved in the inflammatory processes. Stimulation of NO production during inflammatory progression signifies a protection mechanism against invading bacterial pathogens, however extreme formation of NO has also been involved in host tissue injury (Van Der Vliet et al., 1997 ). A significant ( P ≤ 0.0005) decline of nitrite in the levels of protein was observed in kidney and bladder tissues of phytol treatment groups compare to the infection and vehicle controls. Similar to the observed results, recently the combination therapy with ajoene and ciprofloxacin has been found to show decreasing level of NO production in mice infected with P. aeruginosa (Vadekeetil et al., 2016 ). To support the decreasing level of virulence enzymes and inflammatory markers in phytol treated groups, the histopathology analysis was done. Kidney sections of the normal uninfected rats looked histologically normal with no substantial pathological variations (Figure 9Aa ). The kidney sections of infection and vehicle control rats had extensive infiltration of neutrophils with destruction of renal tubules and shrinkage of glomeruli (Figures 9Ab,c ). In case of 100 mg/kg body weight of phytol treated group, a mild infiltration of neutrophils was noted and 200 mg/kg body weight of phytol treatment showed no considerable pathological changes (Figures 9Ad,e ). Recently, Balamurugan et al. ( 2015 ) found that the treatment of UTI QQ with gentamicin against rats infected with S. aureus showed minimal dilatation of renal tubules with no considerable pathological changes in kidney section. The bladder histology section of infection and vehicle controls showed extensive infiltration of neutrophils with severe abrasion in transitional epithelium (Figures 9Bb,c ). In contrast, the uninfected rat and infected rat treated with phytol showed no considerable pathological changes (Figures 9Ba,d,e ). Outcome of this bladder histology supports the results of Sabharwal et al. ( 2016 ), who have not observed any adverse pathological changes in divalent flagellin treated mice bladder tissue. The toxicological property of phytol has been tested in different animal models for different clinical applications (Hidiroglou and Jenkins, 1972 ; McGinty et al., 2010 ). The acute oral LD 50 of phytol in rats was described to be more than 5.0 g/kg body weight (McGinty et al., 2010 ). However, the rats were dosed for 28-day in sub chronic toxicity study showed the no-observed-adverse-effect-level (NOAEL) of phytol to be 500 mg/kg/day, based on organ weight changes. In contrast, the rats were dosed for a longer period of time (52–108 days) in a one-generation reproductive toxicity study, the lowest-observed-adverse-effect level (LOAEL) of phytol was to be 250 mg/kg/day, based on renal changes in male and female rats (Api et al., 2016 ). The overall mammalian toxicity of phytol is considered to be low only in least concentration. Hence, the protective effect of phytol was tested against S. marcescens associated acute pyelonephritis infection at the concentration of 100 and 200 mg/kg. On the other hand, we assessed the toxic effect of phytol (200 mg/kg) by single dose acute toxicity study. No significant differences were observed in the hematological profile of phytol treated group compared to the animal control (Table 3 ). The oral administration of phytol in rats did not show any significant changes in biochemical profile when compared to the animal control group (Table 3 ). However, an increase in ALP and SGPT serum blood levels were observed in the phytol treatment. ALP and SGPT are generally used as markers for liver function and indicators of liver toxicity. ALP and SGPT levels elevate in the blood when the hepatic cellular permeability is changed or cellular injury occurs in liver. The histopathological analysis of vital organs (Kidney, Liver, Heart, Lungs, and Spleen) in phytol treated group did not show any adverse pathological effects compared to the animal control, except liver section (Figure 10 ). The liver section of phytol treatment showed moderate degeneration of hepatocytes (Figure 10Bb ) and it was due to the increasing level of ALP and SGPT. The degeneration of hepatocytes and increasing level of liver enzymes support the outcome of Mackie et al. ( 2009 ), who have reported that the phytol induced the hepatotoxicity in mice. To the best of our knowledge, this is the pioneering study annex the anti-QS and antibiofilm capability of phytol in the counteractive action on S. marcescens infection through the serious of virulence inhibition assays. The real-time analysis disclosed the molecular mechanism of phytol on QS intervened virulence factors productions in S. marcescens . Further, the S. marcescens associated acute pyelonephritis infection in rat model unveiled the protective effect of phytol by reducing the bacterial counts, virulence enzymes and inflammatory markers productions with adequate safety. Therefore, the utilization of phytol is promising in the advancement of novel antipathogenic medications to control acute pyelonephritis infection caused by S. marcescens . However, further studies will be needed to reveal the mode of action of phytol against S. marcescens associated acute pyelonephritis infection."
} | 6,756 |
35847276 | PMC9280763 | pmc | 2,960 | {
"abstract": "The development of\nflexible materials with higher piezoelectric\nproperties and electrostrictive response is of great significance\nin many applications such as wearable functional devices, flexible\nsensors, and actuators. In this study, we report an efficient fabrication\nstrategy to construct a highly sensitive (0.72 kPa –1 ), red light-emitting flexible pressure sensor using electrospun\nEu 3+ -doped polyvinylidene fluoride–hexafluoropropylene/graphene\noxide composite nanofibers using a layer-by-layer technology. The\nhigh β-phase concentration (96.3%) was achieved from the Eu 3+ -doped P(VDF-HFP)/GO nanofibers, leading to a high piezoelectricity\nof the composite nanofibers. We observed that a pressure sensor is\nenabled to generate an output voltage of 4.5 V. Furthermore, Eu 3+ -doped P(VDF-HFP)/GO composite nanofiber-based pressure sensors\ncan also be used as an actuator as it has a good electrostrictive\neffect. At the same time, the nanofiber membrane has excellent ferroelectric\nproperties and good fluorescence properties. These results indicate\nthat this material has great application potential in the fields of\nphotoluminescent fabrics, flexible sensors, soft actuators, and energy\nstorage devices.",
"conclusion": "4 Conclusions In conclusion, the mechanical properties of the conductive filler\nsystem are obviously better than those of the inorganic filler system,\nand Young’s modulus is greatly reduced. Although the sandwich\nstructure conductive filler system can also improve the ferroelectric\nproperties of materials, the residual polarization and saturation\npolarization values are far less than those of the inorganic filler\nsystem. The synergism of the conductive filler and the fluorescence\ncomplex reduced the fiber diameter and promoted the transformation\nof α to β phase in P(VDF-HFP) nanofibers, and the β\nphase content increased by 96.3%. In terms of sensor performance,\nthe sandwich structure sensor is far superior to the monolayer, with\na sensitivity of up to 0.72 kPa –1 , an ultrahigh\noutput voltage of 4.5 V, and an electroforming variable of 7 mm. The\nstability, repeatability, durability, and applicability on practical\nfields of this multilayer nanofiber composites are also very important,\nwhich is the direction we need to further study in the future. The\nsandwich structure nanofiber membrane prepared has good hydrophobicity\nand certain fluorescence properties. The flexible sensor based on\nthe electrostatic spinning sandwich structure double-packing system\nnanocomposite fiber has ultrahigh and low voltage sensitivity, high\npiezoelectric output, good fluorescence function, excellent mechanical\nproperties, and so on, which has great application potential in multifunctional\nflexible sensors, actuators, and other fields.",
"introduction": "1 Introduction In\nrecent years, the application of piezoelectric materials has\nbeen expanding. Piezoelectric materials refer to the internal polarization\nphenomenon when deformation by external force, and the voltage between\nthe two ends of the crystal material, can transform mechanical energy\ninto electrical energy. 1 Electroactive\npolymers convert electrical energy into mechanical energy with their\nintrinsic ability, especially under the stimulus of the applied electric\nfield change size or shape (i.e., bending, contraction, or expansion). 2 Organic piezoelectric materials represented by\npolyvinylidene fluoride (PVDF) and its copolymers have good piezoelectric,\nferroelectric, and electrostrictive response properties, which have\nbeen widely used in sensors, actuators, energy storage applications,\nwaterproof coatings, and other fields. 3 − 7 PVDF and its copolymers mainly have α, β, and\nγ\ncrystal types. Among them, the β crystal structure with full\n“TTTT” anticonformation has the largest polarity and\nexhibits good piezoelectric and electrostrictive properties. 8 Therefore, choosing an appropriate method to\nincrease the content of the β phase is an important research\ndirection to improve the piezoelectric properties of PVDF-based materials,\nwhich has important practical application significance. At present,\nscientists mainly use methods such as high electric field, electrostatic\nspinning, and polarity induction of nanofillers to increase the content\nof the β phase. 9 − 11 The multilayer porous media film prepared using the\nelectrospinning technology has the advantages of easy control, small\npore size, and high specific surface area. Under the action of the\napplied electric field, the polarization effect and the high tensile\nratio of the high static electric field in the electrospinning process\nare similar to uniaxial mechanical drawing, which promotes the increase\nin the β phase content. 12 , 13 At the same time, based\non the excellent conductivity of conductive fillers, nanofiber membranes\nare endowed with excellent electrodeformation characteristics under\nthe application of electric field, which opens up the application\nof piezoelectric nanofiber modified materials for soft actuators. The doping of rare-earth ions often improves the piezoelectric\nproperties of piezoelectric materials. Eu 3+ ions were doped\nin the polyvinylidene fluoride–hexafluoropropylene [P(VDF-HFP)]\nmatrix as fillers to nucleate electrically active β. 14 During electrospinning, the uniform distribution\nof Eu 3+ in PVDF nanofibers greatly enhanced the interaction\nbetween single Eu 3+ and adjacent PVDF segments, thus improving\nthe thermal stability of photoluminescence. 15 Among conductive fillers, graphene oxide (GO) is a two-dimensional\ncarbon structure with sp 2 hybridization, with high specific\nsurface area, mechanical strength, thermal stability, and a variety\nof other properties noted. Its surface is bound with a variety of\noxygen functional groups, including −COOH and −OH, 16 , 17 which has a high affinity with surrounding molecules and can promote\nhydrophilicity and absorption capacity. 18 , 19 Adding very\nsmall amounts of GO with a weight percentage of less than 1% to polymer\nmembranes improves the physicochemical properties of the membranes\nand generates new functions. 20 The addition\nof modified particles in the electrospinning process can make the\ninteraction between the polymer and modified particles produce higher\ncharge density, promote the mutually reinforcing crystallization of\nβ, improve its crystallinity, and improve the piezoelectric\nproperties. 21 GO-based composites exhibit\nsignificant thermal and mechanical properties attributed to the chemical\ngroups of GO, improving interfacial interactions with other materials/substrates. 22 The mechanical properties of the composites\ncan be optimized by the synergistic effect of multiple fillers, and\nthe piezoelectric properties of the nanocomposite fibers can be enhanced\nby using multiple fillers. 23 Multiple\nconductive fillers were used to improve the electric field\ndistribution or conductive characteristics of the matrix so that the\ninternal charge of the material could be effectively exported to the\nmaterial surface. Local electric field can be used to make other fillers\nin the material fully express the effect of electrical coupling. 24 Layered, core-shell, and sandwich structures\nhave been extensively and deeply studied. 25 Sandwich structure polymer nanocomposites can overcome the contradiction\nbetween the dielectric constant and breakdown strength, and they have\nexcellent energy storage performance, 26 which contributes to improve the comprehensive performance of composite\nfilms. Ahmad et al. 27 found that GO filler\nhad good adsorption performance on the P(VDF-HFP) polymer matrix,\nand the addition of GO made P(VDF-HFP) PEM have good thermal stability,\nelectrolyte absorption, and morphology. Gao et al. 28 used porous thermoplastic polyurethane (TPU) as a flexible\nsubstrate and silver nanowires (AgNWs) as a conductive network system\nto easily achieve a “sandwich” composite conductive\nmaterial with high flexibility and low resistance. Rochel and Yalcinkaya 29 prepared PVDF nanofiber multilayers with good\nhydrophilicity and high mechanical properties through the lamination\nprocess, providing a new approach for the design and development of\nelectrospinning filter membranes. Therefore, the development of multifunctional\ncomposite piezoelectric materials modified by a foundation structure\ndesign is of great significance in many future applications. In previous studies, we studied the synergic effect of the BaTiO 3 system 30 to enhance the function\nof inorganic fillers and substrate material P(VDF-HFP) and prepared\nmultifunctional nanofiber films with fluorescence, piezoelectric,\nand ferroelectric properties, which can be used for flexible pressure\nsensors and energy storage devices. Two-dimensional conductive filler\nGO has its own unique and excellent performance, which can better\nimprove the mechanical properties and piezoelectric output of the\nmaterial. Herein, we use two-dimensional conductive filler GO and\nfluorescent complex Eu (TTA) 3 (TPPO) 2 to form\na double filler system and further explore the influence of double\nfiller components on polymer properties and applications. The β-phase\ncontent of the double packing Eu(TTA) 3 (TPPO) 2 /GO system is up to 96.3%, which is used as a sensor with a sensitivity\nof up to 0.72 kPa –1 , 4.5 V ultrahigh output voltage,\nand 7 mm electroforming variable, and it has good fluorescence characteristics.\nThe multifunctional piezoelectric polymer films developed are of great\nsignificance in many application fields such as intelligent wearable\ndevices, flexible sensors, and actuators.",
"discussion": "3 Results and Discussion 3.1 Membrane Morphology Nanofiber membranes\nwere prepared by electrospinning. The scanning electron microscopy\n(SEM) image and fiber diameter distribution are shown in Figure 1 a–c. The figures\nshow randomly oriented fibers without any lumps and an irregular fibrous\nmorphology. Compared with PFP fibers, the diameter of PFPC and PFPCGO\nnanocomposite fibers decreased from 1000∼2000 nm to 400∼700\nnm and 600∼800 nm because of the addition of C and C/GO. This\nis because the addition of the filler increases the electrical conductivity\nof electrostatic spinning solution, and the electrostatic repulsion\nand Coulomb force on Taylor’s cone increase during spinning,\nleading to the reduction of the fiber diameter. 31 There is no obvious granular bulge on the fiber surface,\nindicating that the doped filler has been well dispersed into the\nPFP matrix. As can be seen from the energy dispersive X-ray spectroscopy\n(EDS) in Figure 1 d–f,\nthe element composition is complete, and the element is evenly dispersed\nwithout aggregation. The uniform dispersion of C with fluorescent\nproperties and conductive GO lays a good foundation for the improvement\nof P(VDF-HFP) properties. Figure 1 SEM images and diameter distributions of (a)\nPFP fiber, (b) PFPC\nnanocomposite fibers, and (c) PFPCGO nanocomposite fibers. Diameter\ndistribution of (d) PFP fiber, (e) PFPC nanocomposite fibers, and\n(f) PFPCGO nanocomposite fibers. (g) PFPCGO nanofiber element distribution\ntotal spectrum. (h) Eu, (i) O element distribution total spectrum. 3.2 Structure and Phase Transformation The three common conformations of P(VDF-HFP) crystals are α\ncrystal, β crystal, and γ crystal, among which β\ncrystal exhibits piezoelectric properties because of its spontaneous\npolarization. 32 Different crystal types\nof P(VDF-HFP) correspond to different FT-IR absorption peaks. The\ncharacteristic absorption peaks of the α phase are located at\n530, 615, 763, 796, 976, and 1383 cm –1 , and the\ncharacteristic absorption peaks of the β phase are located at\n510, 840, and 1278 cm –1 . The peaks at 1234 and 841\ncm –1 are attributed to the γ-phase and the\nsuperimposed β- and γ-phases of P(VDF-HFP), respectively. 33 In order to determine the content of β-phases\nin the sample, 763 and 840 cm –1 were selected to\nrepresent the absorption peaks of α and β crystals, respectively.\nAs shown in Figure 2 a, it is obvious that the peak height decreases significantly after\nthe doping filler at 763 cm –1 because the addition\nof the filler promotes the transformation of P(VDF-HFP) from the thermally\nstable α-phase to the metastable β-phase, thus producing\nmore β crystal types. The β crystal content calculated\nusing the Lambert–Beer law is shown in Figure 2 b. The β crystal content of the PFP\nsample is only 80.8%, while the β crystal content of the three\nsamples doped with the filler is 94.1, 96.1, and 96.3%, respectively.\nIt can be seen that the codoping of complex C and GO double packing\ncan promote the increase of the β crystal content, which is\n15.5% higher than that of PFP. This result was also confirmed by XRD\n( Figure 2 c) for pure\nPFP films, and the peaks at 18.4° and 20.0° correspond to\nthe (020) and (110) crystal planes of the P(VDF-HFP) α phase.\nWith the addition of the filler, the characteristic diffraction peak\nof the α phase almost disappears, and the characteristic diffraction\npeak of the (100/200) crystal surface of the β phase at 20.4°\ngradually increases. A special interaction between P(VDF-HFP) and\nGO further improves the transfer from α to β phases. With\nthe addition of the filler, the intensity of the α phase diffraction\npeak gradually decreases or even disappears, while the intensity of\nthe β phase diffraction peak gradually increases. Figure 2 (a) FT-IR spectra\nfor polymer membranes. (b) β crystal content\ndiagram. (c) XRD pattern of the polymer membranes. Figure 3 shows\nthe\nmechanism by which P(VDF-HFP) increases the β phase content.\nFluorescence complexes Eu(TTA) 3 (TPPO) 2 and GO\nas nucleating agents provide a substrate for the nucleation of P(VDF-HFP)\ncrystals and induce β phase formation through strong interfacial\ninteractions. Because of the high voltage electrostatic field during\nelectrospinning and the high electronegativity of graphene, the H\natoms of P(VDF-HFP) tend to be close to the GO surface. The electric\nfield of electrospinning itself, the fluorescence complex, and the\nlocal amplified electric field of conductive GO can all generate induced\ncharges, resulting in a stronger Coulomb force, which attracts the\nP(VDF-HFP) chain to the GO surface to form hydrogen bonds and crystallize\ninto the β phase, promoting the increase of the β phase\nin P(VDF-HFP) composite nanofibers. Figure 3 Schematic diagram of increasing the β\nphase content. DSC thermographs are provided\nin Figure 4 a. The addition\nof the filler obviously affects\nthe thermal behavior of nanocomposites. For pure PFP, the exothermic\npeak is 143.6 °C and there is only one large melting peak. With\nthe addition of filler, it is obvious that the melting peak develops\ninto a double wide peak, and the maximum wave peak moves towards high\ntemperature. The melting peak displacement of PFPCGO samples doped\nwith double fillers is the largest, and the exothermic peak is 153.5\n°C. The higher melt peak displacement is attributed to the uniform\ndistribution and nucleation of nanoparticles. 34 The GO layer has a good affinity with the polar P(VDF-HFP) chain,\nresulting in substantial nucleation in the polymer matrix, thus promoting\nthe increase of β crystal content. Figure 4 b,c shows the thermogravimetric analysis\n(TGA) and differential thermal analysis (DTA) curves of the composite.\nAll curves show typical weightlessness at around 470 °C because\nof the degradation of the polymer chain. Although all samples showed\nsimilar degradation characteristics, the percentage of residual mass\nof the PFPCGO sample was higher than that of other nanocomposites,\nindicating that the thermal resistance of this particular sample was\nslightly higher. 35 PFPC and PFPCGO samples\nhave a weightlessness platform corresponding to the weightlessness\nof the fluorescence complex C at 300 °C, which also proves that\nthe doping of fillers is successful. Figure 4 (a) DSC results of polymer membranes.\n(b) TGA and (c) DTA curves\nfor polymer membranes. PFPCGO spinning solution\nwas scraped and coated with a thickness\nof about 5 μm. The AFM test was performed. Figure 5 shows the AFM amplitude, height,\nand phase images of the PFPCGO scratch-coated film. In these images,\nthe brightest areas display the highest point of the membrane surface,\nand the dark regions represent valley or membrane pores. It can be\ninferred from AFM images that there are fine particles on the surface\nof the film, which is due to the diffusion of the nanofillers into\nthe P(VDF-HFP) phase so that the synergistic effect of the double\nfillers promotes the increase in the β crystal content. Figure 5 AFM (a) amplitude,\n(b) height, and (c) phase images of the PFPCGO\ncomposite membrane. 3.3 Mechanical\nProperty Analysis The\nstress–stain curves of electrospun nanofibers are depicted\nin Figure 6 a,b, and\nthe sandwich structure of the PU/PFPCGO/PU nanofiber membrane has\na maximum strain of 200%, showing an excellent Young’s modulus\nof 1.0 MPa, but the strain is low, only 1.5 MPa. The maximum strain\nof the PU/PFPCGO/PU nanofiber membrane of the sandwich structure is\ndoubled than that of pure PU/PFP/PU, while the Young’s modulus\nis reduced by more than three times. It can be seen that doping GO\ncan significantly improve the elongation at break and significantly\nreduce the Young’s modulus of the material, which is far superior\nto doping complex C. The synergistic effect of the two fillers to\nenhance the mechanical properties is that the composite nanofibers\nobtain the maximum elongation at break and the lowest Young’s\nmodulus, which makes it more effective in the application of flexible\npressure sensors. Figure 6 (a) Stress–strain curves and mechanical properties\nof the\nnanofibers. (b) Contrast diagram of Young’s modulus. 3.4 Ferroelectric Performance\nAnalysis Figure 7 describes\nthe ferroelectric properties of nanofiber samples. It can be seen\nthat the addition of the filler and sandwich structure design obviously\nenhances the ferroelectric properties of the material. Saturation\npolarization ( P s ) was significantly enhanced,\nin which P s-PFP : 0.028 μC/cm 2 , P s-PU/PFP/PU : 0.033 μC/cm 2 , P s-PFPCGO : 0.039 μC/cm 2 , and P s-PU/PFPCGO/PU :\n0.055 μC/cm 2 . The remnant polarization ( P r ) of 4.0 × 10 –3 μC/cm 2 is obtained for PU/PFPCGO/PU, which is higher than that of\nPFPCGO (a negligible value of 1.7 × 10 –3 μC/cm 2 ), indicating that the sandwich structure can improve the\nferroelectric properties of the material efficiently. Although the\nsandwich structure conductive filler system can also improve the ferroelectric\nproperties of materials, the residual polarization and saturation\npolarization values are far less than those of the inorganic filler\nsystem [Eu(TTA) 3 (TPPO) 2 /BaTiO 3 ]. Figure 7 Polarization–electric\nfield (P–E) hysteresis of composite\nnanofibers. 3.5 Surface\nWCA of Electrospun Membranes The surface hydrophobicity of\ndifferent nanofibrous layers was evaluated\nusing dynamic WCA measurements [ Figure 8 a–c]. The WCA over 90° means that nanofibers\nsurface is hydrophobic in nature. PU has a hydrophobic surface with\na WCA of 117°. The WCA of PFPCGO nanofiber membrane was 122°.\nWe found that the sandwich structure of the PU/PFPCGO/PU nanofiber\nmembrane WCA was 123° close to the PFPCGO nanofiber membrane,\nwhich maintained the surface hydrophobicity of the material. Therefore,\nthe obtained PU/PFPCGO/PU film exhibited great hydrophobicity, making\nit a promising candidate to be fabricated into a flexible press sensor\nand some intelligent wearable devices. Figure 8 WCA images of (a) PFPCGO,\n(b) PU/PFPCGO/PU, and (c) PU composite\nnanofibers. 3.6 Fluorescence\nProperties of Electrospun Membranes Figure 9 shows the\nfluorescence emission spectrum of the PFPCGO nanofiber membrane. The\nemission peak is strong at 593 and 617 nm, corresponding to the 5 D 0 → 7 F 1 and 5 D 0 → 7 F 2 electron\ntransitions of Eu 3+ . The emission peak is the strongest\nat 617 nm, which is the characteristic peak of Eu 3+ . The\ncalculated peak points ( x = 0.65, y = 0.33) correspond to red emission in CIE color coordinates, which\nprovides a basis for the realization of flexible pressure sensors\nthat can be fluorescently labeled. Figure 9 Fluorescence spectra of PFPCGO. 3.7 Sensitivity of the Flexible\nPressure Sensor The flexible pressure sensor is prepared\nusing the method explained\nin Section 2.4 . Our\nflexible pressure sensor has a positive capacitance change and can\nshow a capacitance curve that increases with the application of pressure\nwhen applied. When the external pressure, the top and bottom when\nthe distance between the electrodes is reduced, the internal nanofibers\nin thickness can be reduced, nanofibers the increase of contact area,\nand separation to shorten the distance between electrode lead to pressure\nsensor capacitance increased significantly when compressed, so within\nthe scope of the low pressure with high stress sensitivity. The flexible\npressure sensor based on the piezoelectric sandwich nanofiber membrane\nacts as a capacitor. Therefore, when the piezoelectric voltage is\ngenerated under applied stress, the induced charge accumulates on\nthe electrode surface, which will cause the change of the capacitance\nof the device. The sensitivity of the device can be measured by the\nchange in the capacitance value with the change in pressure. As shown\nin Figure 10 a, when\nthe pressure is less than 1 kPa, it is obvious that the PU/PFPCGO/PU\nfilm has an excellent sensitivity of 0.72 kPa –1 ,\nabout 4.8 times that of the PFPCGO sample (0.15 kPa –1 ), and much higher than that reported in some literature. 36 − 39 [ Figure 10 b]. With\nincreasing pressure, sensitivity decreases and tends to equilibrium.\nTherefore, in the low-pressure range, the pressure sensitivity is\nmore excellent, making it useful as an electronic skin to detect small\nvibrations such as pulse and heart rate. 39 Figure 10 (a) Comparison of the sensitivity of PFPCGO and PU/PFPCGO/PU sensors.\n(b) Comparison of the sensitivity of sensors in this study and the\nreported literature. 3.8 Piezoelectric\nStudies of the Flexible Pressure\nSensor Figure 11 a shows the schematic diagram of the piezoelectric response\nmeasurement system. The computer control terminal can ensure that\nthe linear motor can provide a certain frequency and a certain value\nof pressure. Test results are displayed and recorded on the output\nterminal. We conducted a polarity test by reversing electrode connection,\nas shown in Figure 11 b,c, the output voltage is basically the same, but the opposite polarity,\nwhich confirmed the arrangement of dipole. It also shows that when\nthe sensor is subjected to external stress changes, piezoelectric\npotential will be generated due to the existing polarization changes,\nthat is, the measured output electrical signal is generated by the\npiezoelectric effect. Figure 11 (a) Schematic diagram of the piezoelectric response measurement\nsystem. Output voltage of (b) the forward connection and (c) the reverse\nconnection. It is observed that the piezoelectric\nproperty of polymer nanocomposites\nstrongly depends on the crystalline structure of the polymer, as well\nas on the electroactive polar phase formation in the nanocomposite. 35 In order to study the piezoelectric properties\nof the prepared flexible pressure sensor, the output voltage and current\nwere tested at the same pressure of 20 N and different frequencies\n( Figure 12 ). The output\nvoltage is not only dependent on the β crystal content but also\ndepends on the effect of the vibration frequency applied to the sample. 40 It can be seen from the figure that the output\nvoltage and current are proportional to the frequency, which means\nthat the sensor can have good response under different environmental\nchanges. With the addition of the filler, when the test condition\nis 3 Hz, the voltage output of PFP, PFPC, and PFPCGO is 0.8, 2.2,\nand 3.8 V, respectively. The voltage output of PU/PFP/PU, PU/PFPC/PU,\nand PU/PFPCGO/PU with the sandwich structure is 1.7, 3.2, and 4.5\nV, respectively. The output of the sandwich structure is better than\nthat of the single layer structure, and the addition of the filler\nis better than that of pure P(VDF-HFP). By providing more nucleating\nsites, GO increased the β phase proportion and therefore enhanced\nthe electrical properties of PVDF. 41 The\nstudy found that at 3 Hz, the piezoelectric output of the PU/PFPCGO/PU\ndouble-packing system with the sandwich structure was the highest,\nwith an output voltage of about 4.5 V and an output current of about\n35 nA, which were far better than those reported in the literature,\nas shown in Table 1 . In order to better demonstrate the piezoelectric properties of\nthe material, the piezoelectric charge coefficient ( d 33 ) of the PFPCGO spinning solution is 4 pC/N after scraping\nand coating the film, which proves that the material has certain piezoelectric\nproperties. There is still a certain gap with some reported data in Table 1 . The reason should\nbe that the electrode prepared by nanofiber film will always be broken\ndown, and the scraping process is adopted. Therefore, the device can\nbe used for pressure detection, signal monitoring, and electronic\nskin sensing. Figure 12 Piezoelectric output (a–c) voltage and (d–f)\ncurrent\nof the nanocomposites under different frequencies (1∼3 Hz). Table 1 Compared with Other Reported Piezoelectric\nOutputs materials area [cm 2 ] frequency\n[Hz] force [N] d 33 [pC/N] voltage [V] current [nA] ref. P(VDF-HFP)-Co-ZnO 50 2.5 2.8 ( 34 ) PVDF-GO-BTO 1 2 10 38 2.5 10.5 ( 42 ) P(VDF-HFP)/MWCNT 15 0.62 ( 43 ) PVDF-GO/graphene 35 24 2 600 ( 44 ) PVDF-GO 4 8 12 12.25 2.1 ( 45 ) PVDF-rGO 20 5 4.38 ( 46 ) PVDF-GO 35 2 0.49 0.65 0.08 70 ( 47 ) P(VDF-HFP)-GO-Eu 3+ 16 3 20 4 4.5 35 this work 3.9 Electrostrictive\nTest The electrostrictive\ntest device and schematic diagram is shown in Figure 13 . The sample was clamped with a clamp to\nexpose it for 20 mm, and the laser table was adjusted so that the\nlaser hits 5 mm upward at the bottom of the sample, that is, the displacement\n( D ) at 15 mm away from the fixed position. The positive\nand negative power supplies are connected to the conductive glass,\nand the electric field spacing is 40 mm. Without prepolarization,\nthe electrostrictive test was carried out. The electrostrictive test\nis to gradually increase the voltage from 400 to 6400 V and increase\nthe interval by 50 V/0.01 s each time. The electrostrictive test cycle\ntest is to momentarily apply a voltage of 4500 V, maintain for 20\ns, disconnect for 20 s, and cycle 10 times. The cycle test is to continue\nafter the electrostrictive test of the same sample. The nanofiber\nlayer was spun on the surface of aluminum foil using the layer spinning\nmethod. The difference is that electrospun 6 mL (expanded by 3 times)\nof PU/PFP/PU, PU/PFPC/PU, PU/PFPGO/PU, and PU/PFPCGO/PU precursors\nare electrically driven samples. The PU nanofiber layer (2 mL) was\nelectrospun on both sides of the surface, and the sample was cut into\n40 mm × 5 mm with a thickness of 0.2 ± 0.05 mm. The test\nresults are shown in Figure 13 . Figure 13 (a) Electrostrictive test device and (b) schematic diagram. It can be seen from Figure 14 and the Supporting Information Videos 1 and 2 that\nthe sample\nof the electrostatic spinning nanofiber film with a sandwich structure\ndeforms under the action of the applied electric field, and the deformation\nis restored after the electric field is removed. 48 In the absence of any applied electric field, the dipoles\nin the nanofibers are randomly oriented. When an electric field is\napplied, the dipoles are oriented according to the electric field,\nso the sample will shrink. 49 The maximum\ndeformation of sample PU/PFPCGO/PU is ∼7 mm, which is about\n4.4 times the deformation of sample PFPCGO (∼1.6 mm) and 8.8\ntimes that of PFP (∼0.8 mm). It can be seen from the curve\nslope that the response of the double filler system is also relatively\nfast, and the slope increases gradually with the increase in the electric\nfield, which indicates that the response and deformation increase\nsharply with the increase in the electric field. It can be seen that\nthe double filler system 50 and the sandwich\nstructure can more effectively promote the driving performance of\nthe film. Figure 14 b shows the bending displacement curve of the sample under the cyclic\nelectric field; it can be seen that the deformation increases rapidly\nat the beginning of applying the fixed electric field, and the deformation\nincreases slowly with the change in time. After removing the electric\nfield, the deformation decreases sharply, but it cannot return to\nthe initial position, where certain deflection is maintained, and\nthe deflection begins to increase with the increase in the number\nof cycles. The double filler system and the sandwich structure of\nthe nanofiber membrane have a more obvious electric deformation effect. Figure 14 Response\nperformance of samples (40 mm long, 5 mm wide, and 0.2\nmm thick) to high voltage electric field. Bending displacements ( D ) of (a) monolayer and (b) sandwich structures. Bending\ndisplacement curves of (c) monolayer and (d) sandwich structures under\ncyclic electric field."
} | 7,299 |
20161249 | null | s2 | 2,961 | {
"abstract": "The role of disturbance in community ecology has been studied extensively and is thought to free resources and reset successional sequences at the local scale and create heterogeneity at the regional scale. Most studies have investigated effects on either the disturbed patch or on the entire community, but have generally ignored any effect of or on the community surrounding disturbed patches. We used marine fouling communities to examine the effect of a surrounding community on species abundance within a disturbed patch and the effect of a disturbance on species abudance in the surrounding community. We varied both the magnitude and pattern of disturbance on experimental settlement plates. Settlement plates were dominated by a non-native bryozoan, which may have established because of the large amount of initial space available on plates. Percent cover of each species within the patch were affected by the surrounding community, confirming previous studies' predictions about edge effects from the surrounding community on dynamics within a patch. Disturbance resulted in lower percent cover in the surrounding community, but there were no differences between magnitudes or spatial patterns of disturbance. Disturbance lowered population growth rates in the surrounding community, potentially by altering the abiotic environment or species interactions. Following disturbance, the recovery of species within a patch may be affected by species in the surrounding community, but the effects of a disturbance can extend beyond the patch and alter abundances in the surrounding community. The dependence of patch dynamics on the surrounding community and the extended effects of disturbance on the surrounding community, suggest an important feedback of disturbance on patch dynamics indirectly via the surrounding community."
} | 458 |
36095129 | PMC9508829 | pmc | 2,962 | {
"abstract": "Abstract Natural methylotrophs are attractive methanol utilization hosts, but lack flexible expression tools. In this study, we developed yeast transcriptional device libraries for precise synthesis of value-added chemicals from methanol. We synthesized transcriptional devices by fusing bacterial DNA-binding proteins (DBPs) with yeast transactivation domains, and linking bacterial binding sequences (BSs) with the yeast core promoter. Three DBP–BS pairs showed good activity when working with transactivation domains and the core promoter of P AOX1 in the methylotrophic yeast, Pichia pastoris . Fine-tuning of the tandem BSs, spacers and differentiated input promoters further enabled a c onstitutive tr anscriptional d evice l ibrary (cTRDL) composed of 126 transcriptional devices with an expression strength of 16–520% and an i nducible TRDL (iTRDL) composed of 162 methanol-inducible transcriptional devices with an expression strength of 30–500%, compared with P AOX1 . Selected devices from iTRDL were adapted to the dihydromonacolin L biosynthetic pathway by orthogonal experimental design, reaching 5.5-fold the production from the P AOX1 -driven pathway. The full factorial design of the selected devices from the cTRDL was adapted to the downstream pathway of dihydromonacolin L to monacolin J. Monacolin J production from methanol reached 3.0-fold the production from the P AOX1 - driven pathway. Our engineered toolsets ensured multilevel pathway control of chemical synthesis in methylotrophic yeasts.",
"introduction": "INTRODUCTION In view of its quantitative, predictive and engineering characteristics, synthetic biology has pushed forward the conversion of life cognition to life design. Recently, the development of theories and methods in synthetic and molecular biology has allowed the sophisticated rewiring of non-natural life. Most successes have been achieved on model microbial hosts, such as Escherichia coli ( 1 , 2 ) and Saccharomyces cerevisiae ( 3 , 4 ), whereas other non-conventional hosts have not been explored. As different strains have different genetic backgrounds, they may have specific biological components that adapt to different application scenarios. Therefore, exploration of individualized set-ups for different strains is required. For instance, the bio-utilization of sustainable one-carbon (C1) substrates such as methanol, methane and CO 2 has attracted widespread attention in both academia and industry ( 5 ). Methanol is a major byproduct of the fossil industry and a promising product of methane oxidation or CO 2 reduction. It represents a probable C1 substrate for industrial bio-utilization, with its liquid state being compatible for transportation and fermentation control ( 6–8 ). Moreover, methanol has a 50% higher degree of reduction per mole of carbon than sugar substrates such as glucose, and thus provides more surplus electrons for compound synthesis ( 9 ). With mature genetic manipulation tools, some non-methylotrophic strains, such as E. coli ( 10–12 ), Corynebacterium glutamicum ( 13–15 ) and S. cerevisiae ( 16–18 ), have been engineered to utilize methanol by the reassembly of exogenous methanol assimilation pathways in cells. Nevertheless, their weak methanol utilization ability still lags far behind their production requirements ( 19 ). Instead, natural methylotrophs are known to efficiently utilize methanol, but lack sufficient genetic tools to enable multigene pathways ( 20–22 ). In the past two decades, the methylotrophic yeast, Pichia pastoris (syn. Komagataella phaffii ), has been shown to be an excellent workhorse for protein production ( 20 , 23–25 ) and a potential cell factory for chemical synthesis ( 26–29 ). This organism can efficiently utilize methanol with high-level expression of endogenous alcohol oxidase 1 (Aox1) and thus grows well on methanol as the sole carbon source ( 22 ). Most recently, it was further developed as an autotrophic strain capable of growth on CO 2 with methanol as a reducing power donor ( 30 , 31 ). The extraordinary characteristics of high cell density, strong expression ability and availability of post-translational modifications ( 20 ) support expression of enzymes which catalyse the synthesis of the desired products. Recently, Golden Gate, CRISPR–Cas9 and other genome editing strategies have facilitated pathway assembly in P. pastoris ( 32–40 ). Nevertheless, high-level biosynthesis of these products requires precise control of the multigene pathway, including an increase in precursor pools, up-regulation of positive pathways, down-regulation of competitive pathways and balance of pathway parts ( 28 , 29 , 41 ). In comparison with model expression hosts, P. pastoris lacks fine-tuned expression tools ( 21 ). Although promoters can be bioinformatically screened, it is difficult to obtain promoters with a broad range of expression levels and clear transcriptional regulatory mechanisms that can adapt to complicated pathway expression. Promoter engineering may provide variants of gradient strengths from either methanol-inducible ( 42 , 43 ) or other types ( 44–48 ) of promoters. However, these variants are obtained mainly through simple mutations or combinations of DNA sequences. They may share most DNA sequences or in situ transcription factors (TFs), which can easily cause uncontrollable cross-talk during transcription ( 49–51 ). Therefore, it is necessary to propose alternative expression toolboxes for non-conventional yeasts. In recent years, synthetic TFs with heterologous DNA-binding and transactivation domains have been designed to function with regulatory cis- elements. These engineered devices have demonstrated functional independence over native transcriptional regulation in various microbial species ( 52 , 53 ). Previously, we explored a t ranscriptional s ignal a mplification d evice (TSAD) in P. pastoris composed of a hybrid promoter lacO -cP AOX1 (core P AOX1 ) and a chimeric transactivator LacI–Mit1AD ( 29 ). An improved TSAD (iTSAD) was obtained using 18 combination groups of cis- and trans- acting elements in E. coli and P. pastoris . The reporter protein expressed by the iTSAD with glucose was 4.2-fold higher than that expressed by the strong methanol-inducible promoter P AOX1 with methanol ( 54 ). This represents a useful strategy that far exceeds the amplification of the P AOX1 variant library (up to 1.6-fold) ( 42 , 43 ). The functions of iTSAD inspired us to explore the synthetic transcriptional components of the full coverage intensity. This could be realized by combinatorial assembly of various cis- and trans- acting elements from bacteria and yeast, and by manipulation of the spacers between the binding and core sequences of the cis- acting elements. Constitutive control of these elements enabled a synthetic c onstitutive tr anscriptional d evice l ibrary (cTRDL) composed of 126 devices with an intensity range of 16–520% (P AOX1 as 100%). Methanol-inducible input promoters with gradient strength were used to construct a synthetic methanol-inducible TRDL (iTRDL) composed of 162 devices with an intensity range of 30–500% (P AOX1 as 100%). We subsequently tested the applicability of the iTRDL and cTRDL in multigene pathway expression and balance. We selected devices from the iTRDL to adapt to dihydromonacolin L (DML) biosynthesis from methanol using an orthogonal experimental design, which led to a 5.5-fold DML titre compared with that from the P AOX1 -driven pathway. Furthermore, the selected devices from the cTRDL were adapted to the downstream pathway of DML to monacolin J (MJ) using a full factorial design. It finally achieved a 3.0-fold MJ titre compared with that from the P AOX1 -driven pathway with methanol as the substrate. Our study provides an alternative expression toolbox and a strategy for methanol bio-utilization in natural methylotrophs.",
"discussion": "DISCUSSION Although remarkable progress has been made in reprogramming bacteria and baker's yeast into synthetic methylotrophs ( 10–18 ), breakthroughs in the methanol utilization capacity for industrial use remain a huge challenge ( 19 , 68 ). The production of chemicals by natural methylotrophs has provided an alternative path to methanol bio-utilization for industrial purposes ( 28 , 69–71 ). However, the lack of genetic tools severely prevents natural methylotrophs from synthesizing valuable chemicals via multigene pathways ( 20–22 ). Although the methylotrophic yeast P. pastoris has been used as an attractive protein expression host ( 20 , 23–25 ), its use for the production of chemicals via complicated pathways is still limited by insufficient fine-tuning expression tools ( 21 ). In this study, we demonstrated that transcriptional device libraries with broad expression levels and smooth changes can be achieved by rational design in addition to conventional promoter variants. We engineered sTFs by fusing endogenous TFADs with different bacterial DBPs, in which hybrid cis -elements were designed by linking DBP-paired BSs and TFAD-contributing CPs. Three workable bacterial DBP–BS pairs with minimal cross-talk, LacI– lacO , LexA –lexO and AraC– araI , were selected from the five tested groups ( Supplementary Figure S1 ). Their combinations with activation domains of three P. pastoris TFs, Mit1, Mxr1 and Prm1 ( 57 ), finally generated a cTRDL with 126 constitutive transcriptional devices when the TFs were constitutively expressed (Figure 1 ; Supplementary Figures S2 and S3 ). Subsequently, methanol-inducible promoters with gradually changing strengths were used to drive sTFs, thus generating an iTRDL with 162 methanol-inducible transcriptional devices (Figure 2 ; Supplementary Figure S3 ). With regard to the strength of the native P AOX1 (100%), our engineered cTRDL and iTRDL presented expression intensity ranges of 16–520% and 30–500%, respectively. They provide diverse and powerful expression tools compared with traditional promoter variants. In addition, the minimal cross-talk among various sTF–BS pairs may extend their application scenarios, which shows a unique advantage over the reported promoter libraries ( 42–45 , 48 ). The synthetic cTRDL and iTRDL were then tested to fine-tune the multigene biosynthetic pathway of the hypolipidaemic drug intermediate, MJ. The six essential genes of the MJ biosynthetic pathway were divided into two modules. Thus, we evaluated the regulatory functions of the selected cTRDL and iTRDL devices within each module and between the two modules. The statistical method of orthogonal experimental design was employed to arrange three expression levels of devices from the iTRDL, matching the four biosynthetic genes for the upstream module. The resulting nine single-copy recombinant strains produced various levels of the target compound DML, illustrating the effectiveness of the iTRDL devices (Figure 3C ). In particular, DML synthesis from methanol by the optimum strain A3 was superior to that of our previously constructed P AOX1 -based single-copy-producing strain D1 and multicopy-producing strain D9. Moreover, with the predicted optimal combination of expression levels matching pathway genes, we obtained the improved strain Opt (LovB with device intensity of ∼450%, LovC and NpgA with device intensity of ∼250%, and LovG with device intensity of ∼100%). It showed a noticeable advantage over the control strains D1 (5.5-fold for titre and 9.2-fold for productivity) and D9 (2.0-fold for titre and 4.0-fold for productivity). Generally, the DML biosynthetic pathway in Opt was stronger than in the control strains, as reflected by the relative transcription levels of biosynthetic genes ( Supplementary Figure S9A ). These results verify the strong and flexible expression modes of the engineered iTRDL. Notably, we used only four devices that shared the same DBP–BS pair, whereas 162 iTRDL devices with various DBP–BS pairs provided sufficient optional tools for multigene pathway control. Moreover, we proved the availability of the iTRDL with a statistical experimental design, which can reduce the number of experimental trials, especially for complicated pathways. With the success of the internal module control by the iTRDL, we next tested the iTRDL- and cTRDL-mediated pathway balancing between the two modules. A full factorial combination of three expression levels from iTRDL (or cTRDL), matching two biosynthetic genes, was designed for the downstream module (Figure 4 ). These downstream combinations were constructed in the Opt strain, generating nine strains that produced various levels of MJ and intermediates. The optimal expression combination, sLovA, using a device with an intensity of ∼250%, and CPR, using a device with an intensity of ∼100%, was observed for both the cTRDL and iTRDL. Moreover, the downstream module driven by cTRDL devices achieved an MJ titre which was 20.9% higher than that of iTRDL devices. This finding is in accordance with our previous study ( 29 ), which illustrated that constitutive accumulation of downstream enzymes can facilitate conversion of the generated intermediates to the final product after induction. The highest MJ production was obtained from strain OC4 by cTRDL, achieving 3.0- and 1.8-fold titres compared with the control strains J1 and J9, respectively (Figure 4 ). In addition, the MJ biosynthetic pathway in OC4 and OI4 was stronger than that in the control strains, as reflected by the relative transcription levels of the biosynthetic genes ( Supplementary Figure S9B ). Generally, it represents the highest reported level of DML and MJ fermented from methanol. Recently, we achieved an MJ production of 3.2 g/l by heterologous synthesis of this compound in P. pastoris using ethanol as a substrate ( 29 ). In addition, a high MJ production of 5.5 g/l from glucose achieved by rewiring metabolic pathways has been reported in the native fungus Aspergillus terreus ( 72 ). Therefore, there is still room for improvement in the synthesis of MJ using methanol as the sole carbon source. It is worth noting that the DBP–BS pair for devices used in the downstream module was independent of that used in the upstream module. Theoretically, this can avoid the cross-talk of transcriptional regulation control between the two modules. With the effectiveness of the statistically orthogonal experimental design and full factorial design, these synthetic TRDLs may be further applied to complicated pathway regulation using deeper algorithms, such as neural networks and machine learning, in future research. Overall, our TRDLs strengthened and balanced the pathway nodes inside and between the modules. It provides alternative yeast toolboxes for high-level biosynthesis of value-added chemicals from methanol. In addition, these TRDLs should be compatible with heterologous protein expression based on reporter protein levels. Non-model microorganisms, such as unconventional yeast and filamentous fungi, have been extensively used in many fields, including biopharmaceuticals, novel foods, biorefining, industrial enzymes, platform chemicals, biomaterials and environmental protection. However, these strains are difficult to explore in depth because of their unclear genetic background and insufficient molecular tools. Recently, genetic manipulation tools have become possible using CRISPR–Cas-derived genetic editing technologies. However, accurate expression control systems cannot be easily achieved. Although promoter mining is a common strategy for collecting expression tools, it is difficult to obtain promoters with a high intensity and tight regulation. Ideally, promoters for expression control should be strictly inducible to separate cell growth from protein expression and to avoid a cumulative metabolic burden ( 51 , 63 ). Promoters covering a wide range of intensities are also indispensable for fine-tuning control from tight down-regulation to high overexpression ( 51 ). Researchers have modified promoters to breed variants of different strengths. However, the high identity of promoter sequences may cause interpromoter recombination and transcription titration effects ( 49–51 ). Here, we propose a strategy for engineering diverse transcription machinery tools to fine-tune the expression of pathways. This represents a universal construction method of a combinatorial set-up of transactivators, CPs, DBPs and BSs (either native or heterologous), which can be reproduced in extensive methylotrophic hosts."
} | 4,131 |
20073090 | null | s2 | 2,963 | {
"abstract": "The development of renewable alternatives to diesel and jet fuels is highly desirable for the heavy transportation sector, and would offer benefits over the production and use of short-chain alcohols for personal transportation. Here, we report the development of a metabolically engineered strain of Escherichia coli that overproduces medium-chain length fatty acids via three basic modifications: elimination of beta-oxidation, overexpression of the four subunits of acetyl-CoA carboxylase, and expression of a plant acyl-acyl carrier protein (ACP) thioesterase from Umbellularia californica (BTE). The expression level of BTE was optimized by comparing fatty acid production from strains harboring BTE on plasmids with four different copy numbers. Expression of BTE from low copy number plasmids resulted in the highest fatty acid production. Up to a seven-fold increase in total fatty acid production was observed in engineered strains over a negative control strain (lacking beta-oxidation), with a composition dominated by C(12) and C(14) saturated and unsaturated fatty acids. Next, a strategy for producing undecane via a combination of biotechnology and heterogeneous catalysis is demonstrated. Fatty acids were extracted from a culture of an overproducing strain into an alkane phase and fed to a Pd/C plug flow reactor, where the extracted fatty acids were decarboxylated into saturated alkanes. The result is an enriched alkane stream that can be recycled for continuous extractions. Complete conversion of C(12) fatty acids extracted from culture to alkanes has been demonstrated yielding a concentration of 0.44 g L(-1) (culture volume) undecane."
} | 415 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.