pmid stringlengths 8 8 | pmcid stringlengths 8 11 ⌀ | source stringclasses 2
values | rank int64 1 9.78k | sections unknown | tokens int64 3 46.7k |
|---|---|---|---|---|---|
34260583 | PMC8312983 | pmc | 7,021 | {
"abstract": "Arbuscular mycorrhiza (AM) are mutualistic interactions formed between soil fungi and plant roots. AM symbiosis is a fundamental and widespread trait in plants with the potential to sustainably enhance future crop yields. However, improving AM fungal association in crop species requires a fundamental understanding of host colonisation dynamics across varying agronomic and ecological contexts. To this end, we demonstrate the use of betalain pigments as in vivo visual markers for the occurrence and distribution of AM fungal colonisation by Rhizophagus irregularis in Medicago truncatula and Nicotiana benthamiana roots. Using established and novel AM-responsive promoters, we assembled multigene reporter constructs that enable the AM-controlled expression of the core betalain synthesis genes. We show that betalain colouration is specifically induced in root tissues and cells where fungal colonisation has occurred. In a rhizotron setup, we also demonstrate that betalain staining allows for the noninvasive tracing of fungal colonisation along the root system over time. We present MycoRed, a useful innovative method that will expand and complement currently used fungal visualisation techniques to advance knowledge in the field of AM symbiosis.",
"introduction": "Introduction Arbuscular mycorrhiza (AM) fungi of the subphylum Glomeromycotina are soil fungi that engage in symbiosis with land plants [ 1 ]. Symbiotic associations with AM fungi date back to over 400 million years ago and can be formed by 70% to 72% of extant land plant species [ 2 – 4 ]. AM fungi are obligate biotrophs that receive all their carbon intake from the plant, which is estimated at up to 20% of the plant’s photosynthate [ 5 ]. In exchange, the fungus assists the plant with the acquisition of mineral nutrients, mainly phosphorus, whose availability in soils is often a limiting factor for plant growth [ 6 ]. Phosphorus contribution through the mycorrhizal pathway can be very high, and, in some instances, can account for the entire phosphorus consumption of a plant [ 7 ]. During AM symbiosis, fungal hyphae form dichotomously branched structures, named arbuscules, within root cortex cells. Hyphal extension and arbuscule formation are accompanied by the de novo extension of a specialised plant cell membrane that separates the fungal hyphae from the plant cytoplasm [ 8 ]. In order to accommodate arbuscule formation, plant cells undergo a series of changes in gene expression to aid the establishment of symbiosis [ 9 ]. Examples of such AM-induced genes include MtPT4 and MtBCP1 in the model legume Medicago truncatula . MtPT4 encodes a phosphate transporter, belonging to the phosphate transporter 1 (PHT1) subfamily, which is exclusively expressed in arbuscule-containing cells [ 10 ]. Mt PT4 localises to the plant cell membrane surrounding arbuscules and participates in the acquisition of phosphate released by the fungus during symbiosis. MtBCP1 encodes a member of the blue copper protein (BCP) family and is also specifically expressed in regions of the root hosting arbuscule development during AM symbiosis [ 11 ]. MtBCP1 expression is strongest in arbuscule-containing cells but can additionally be observed in adjacent cortical cells [ 11 ]. The study of AM symbiotic processes involves the detection, visualisation, and quantification of fungal colonisation. Current techniques rely on the specific staining of fungal cell walls through fast, simple, and cost-effective procedures. Commonly employed methods include the use of trypan blue [ 12 ], cotton blue and Sudan IV [ 13 ], acid fuchsin [ 14 ], ink–vinegar [ 15 ], or fluorescein-labelled wheat germ agglutinin (WGA) [ 16 ]. All of these methods are destructive, requiring the excision and chemical treatment of roots that are typically visualised with light or fluorescent microscopy. On the other hand, nondestructive methods for detection and visualisation of AM symbiosis offer a number of significant research opportunities but often rely on specialised equipment or are only available for certain species. For example, the foliar accumulation of blumenol-derived metabolites can function as a quantitative proxy for AM colonisation in a number of crop and model plants, with potential applications in field-based quantitative trait locus (QTL) mapping of AM fungi-related genes [ 17 ]. However, blumenol accumulation is not visible, and its detection requires specialised extraction and quantification steps. Furthermore, some cereal crops and species of the Liliaceae and Fabaceae naturally produce apocarotenoid yellow pigments in roots upon mycorrhizal colonisation [ 18 , 19 ], which has been useful, for example, for the identification of maize mutants affected in symbiotic interaction [ 20 ]. To date, however, the use of natural pigments as visual markers of AM symbiosis is limited to select species within these families and has not been implemented in other crops and model plants. Betalains are naturally occurring tyrosine-derived water-soluble pigments, comprising yellow to orange betaxanthins and red to purple betacyanins. Betalains have a number of bio-industrial applications, as natural food colourants and antioxidants [ 21 ], and as biosensors [ 22 – 26 ]. Betalains were first discovered in plants, where they are unique to the flowering plant order Caryophyllales [ 27 ], but have also been reported in the proteobacterium Gluconacetobacter diazotrophicus [ 28 ] and the fungi Amanita [ 29 ] and Hygrocybe [ 30 ]. The betalain biosynthesis pathway in Caryophyllales consists only of 3 main enzymatic steps with additional glycosylation and spontaneous chemical reactions ( Fig 1 ). Initially, tyrosine undergoes an hydroxylation reaction to 3,4-dihydroxy-ʟ-phenylalanine (ʟ-DOPA) catalysed by members of the CYP76AD family of cytochrome P450 enzymes (CYP76AD1/5/6 in Beta vulgaris ) [ 31 , 32 ]. ʟ-DOPA is then cleaved by the action of the ʟ-DOPA-4,5-dioxygenase (DODA) enzyme to form a 4,5-secodopa intermediate that spontaenously cyclises to betalamic acid [ 33 ]. Betalamic acid is the central betalain chromophore and can spontaneously condense with amino groups to produce yellow betaxanthins [ 34 ]. Alternatively, betalamic acid can spontaneously condense with cyclo -dihydroxyphenylalanine ( cyclo -DOPA) to produce the red betacyanin—betanidin. Cyclo -DOPA is obtained via oxidation of ʟ-DOPA, catalysed by CYP76AD1 [ 35 ]. Glycosylation of betacyanins is common and can occur at either the betanidin [ 36 ] or cyclo -DOPA stages [ 37 ], with the latter catalysed by the enzyme cyclo -DOPA 5- O -glucosyltransferase (cDOPA5GT). Betanin is the glycosylated form of betanidin. 10.1371/journal.pbio.3001326.g001 Fig 1 The betalain biosynthetic pathway. A simplified schematic representation of the main enzymatic and spontaneous reactions leading to the formation of red/purple betacyanins and yellow betaxanthins. Enzymatic steps: (1) tyrosine hydroxylation to ʟ-DOPA catalysed by CYP76AD cytochrome P450 enzymes; (2) cleavage of ʟ-DOPA to form betalamic acid by DODA; (3) ʟ-DOPA oxidation to cyclo -DOPA by CYP76AD1; and (4) cyclo -DOPA glycosylation to cyclo-DOPA 5- O -glucoside by the enzyme cDOPA5GT. S*, spontaneous reaction. Betacyanins are represented as a molecule of betanin. cDOPA5GT, cyclo -DOPA 5- O -glucosyltransferase; cyclo -DOPA, cyclo -dihydroxyphenylalanine; DODA, ʟ-DOPA-4,5-dioxygenase; ʟ-DOPA, 3,4-dihydroxy-ʟ-phenylalanine. Elucidation of these core enzymatic steps has enabled the engineering of the betalain biosynthetic pathway in a wide range of heterologous hosts, including microbes such as Saccharomyces cerevisiae [ 38 ]; plant model organisms like Arabidopsis thaliana , Nicotiana tabacum , and Petunia hybrida [ 31 , 39 , 40 ]; and a diversity of crops such as Oryza sativa (rice), Solanum lycopersicum (tomato), Solanum tuberosum (potato), and Solanum melongena (aubergine) [ 40 , 41 ]. Betalains have been used as biosensors in a number of heterologous contexts to report increased production of metabolites, such as tyrosine, dopamine, and ʟ-DOPA in Escherichia coli and Nicotiana benthamiana [ 23 , 24 , 26 ], to measure metabolic flux between competing pathways in the synthesis of benzylisoquinoline alkaloids (BIAs) in S . cerevisiae [ 22 ] and for the detection of copper by heavy metal–resistant bacteria in bioremediation processes [ 25 ]. In plants, specific promoters have been successfully used to target betalain production in specific tissues such as fruits and seed endosperm [ 40 – 42 ]. Expressing the betalain biosynthetic genes under the control of the AtYUC4 promoter in A . thaliana resulted in pigment production in tissues likely to present auxin biosynthesis activity [ 42 ]. Similarly, the use of the DR5 synthetic auxin-responsive promoter in O . sativa calli allowed for easier selection of transformed calli in in vitro transformation protocols [ 42 ]. The use of betalains as in vivo reporters offer a number of advantages: (1) the relative simplicity of betalain biosynthesis; (2) the potential for heterologous betalain expression in phylogenetically diverse hosts; and (3) the ease of betalain visualisation and quantification. Here, we present MycoRed, a betalain-based in vivo and noninvasive reporter system for the occurrence and progression of AM symbiosis in roots of the two model species M . truncatula and N . benthamiana . We leveraged known AM-responsive genes from M . truncatula to identify orthologous N . benthamiana promoters that are similarly responsive to AM fungal colonisation. Heterologous expression of betalain biosynthesis genes specifically driven by AM-responsive promoters effectively tracked AM colonisation dynamics in both species. Collectively, our work demonstrates the efficacy of betalain pigments as reliable in vivo visual markers for the previously inaccessible dynamic tracing of AM symbiosis within root systems, thereby providing a valuable resource for the plant–microbe research community.",
"discussion": "Discussion AM symbiosis is a fundamental and widespread trait in plants that greatly expands the root surface area for nutrient uptake. AM symbiosis is consequently a key agronomic trait in the drive to enhance future crop yields through environmentally sustainable mechanisms. However, enhancing AM fungal association in crop species requires the tools to understand its fundamental dynamics across varying agronomic and ecological contexts. To this end, we demonstrate the use of betalain pigments as in vivo visual markers for the occurrence and distribution of AM fungal colonisation in the roots of M . truncatula and N . benthamiana . We have generated multigene vectors in which AM colonisation specific plant promoters control the expression of core betalain synthesis enzymes in the production of betalain pigments. We show that AM-specific promoter-controlled betalain pigmentation is a powerful macroscopic tool to report and trace fungal colonisation in vivo along the root ( Fig 9 ). 10.1371/journal.pbio.3001326.g009 Fig 9 Betalains can be used as markers for AM colonisation in plant roots. Left: Red pigmentation is easily observable in whole plant root systems. Right: Red pigmentation is most prominent in colonised tissues as well as in adjacent tissue layers. Macroscopically, betalain colouration was specifically limited to regions of the root colonised by R . irregularis in both M . truncatula and N . benthamiana , as we detected few to no false positives. Furthermore, no betalain pigmentation was detected in mutant lines of M . truncatula impaired in AM symbiosis. Quantification of fungal structures revealed that pigmented root fragments were extensively colonised, showing arbuscule-containing cells over the majority of the root length for all tested promoters. We also observed some fungal structures in a low percentage of nonpigmented root fragments. In these cases, arbuscule colonisation was restricted over the length of the analysed root fragments, with the exception of NbBCP1b expressing N . benthamiana roots that exhibited a greater degree of unpigmented yet colonised root fragments. Thus, reporter ability to document the totality of colonisation events depends on the promoter chosen for vector construction. It appears that the NbBCP1b promoter-containing constructs are not equally activated in all colonisation events, or perhaps pigment accumulation becomes evident only at certain stages of colonisation. Promoter sequences of PT4 homologues such as NbPT5b were able to report the majority of colonisation events in the root system and are therefore preferred for reporting total root colonisation. A future solution to document total AM colonisation could involve the establishment of systems whereby the betalain biosynthesis genes are activated by transactivators, which could be then driven by promoters that are active at early, main, and late AM fungal colonisation stages [ 46 – 48 ]. Overall, betalain pigmentation effectively reports fungal colonisation with very little to no error and allows for simple selection of colonised root areas. Microscopy of pigmented tissues revealed that betalain accumulation extended beyond arbusculated cells and into adjacent cell layers in both M . truncatula and N . benthamiana . In M . truncatula , betalain pigmentation was most prominent in the endodermis and pericycle cells adjoining arbusculated inner cortical cells ( Fig 3 ). In N . benthamiana , we also observed betalain pigmentation in the endodermis and pericycle cells, yet pigmentation was more strongly retained in cortical cells ( Fig 6 ). However, colouration in N . benthamiana was sometimes present in non-arbusculated cells adjacent to arbusculated cortical cells, consistent with previously observed GUS staining patterns. These extended patterns occur even when using constructs where all three biosynthesis enzymes are driven by AM-specific promoters and cannot therefore be attributed to constitutive gene expression artefacts. Such pigment accumulation in non-arbusculated cells may be the result of a former colonisation or of moderate degree of pigment migration. Betalains are water-soluble pigments and, in native betalain-pigmented species, are produced in the cytoplasm and then stored in vacuoles [ 49 ], although the mechanisms responsible for the intracellular transport of betalains are unknown. As small water-soluble compounds, betalains may have the potential to move symplastically through root plasmodesmata, although this remains unproven. Transient expression of betalains in leaves of N . benthamiana leads to macroscopically well delimited pigmented areas [ 23 , 31 , 43 ], but cellular migration across the boundary tissue has not been studied. Unpigmented cells harbouring arbuscules could also be a result of mechanical disruption during the sectioning of root tissues for microscopy, where cells that previously contained betalains lose pigmentation after being sliced open. Nevertheless, the betalain reporter system remains a highly effective marker of AM colonisation, especially at the macroscopic level. The betalain biosynthetic pathway has been previously constitutively expressed in several plant species, including plants of the Solanaceae, with no report of observable developmental defects [ 31 , 40 ]. Yet, in our study, expression of betalain producing constructs where only CYP76AD1 was under the control of AM-specific promoters led to developmental defects in N . benthamiana lines. Affected plants displayed dwarf phenotypes, altered leaf, and flower morphology and were unable to set seed ( S6 Fig ). Our findings could be explained by a mechanism whereby constitutive expression of either, or both, DODA and cDOPA5GT in absence of CYP76AD1 expression cause developmental abnormality. Three observations are consistent with this hypothesis: (1) expression of constructs containing all three biosynthesis enzymes under symbiosis specific promoters substantially decreased the number of affected plants; (2) DODA expression appeared enhanced in leaves of severely affected plants; and (3) expression of CYP76AD1 was detected in leaves of plants with mild defects and seemed to decrease with phenotype severity. We speculate that, in absence of ʟ-DOPA, the DODA enzyme could be promiscuously acting on alternative N . benthamiana substrates, with negative developmental consequences. This effect is alleviated by the presence of CYP76AD1, which provides the DODA enzyme with an abundance of its native substrate, ʟ-DOPA, resulting in the production of inert betalain pigmentation. Constitutive DODA expression may therefore create a shoot developmental conflict that can be partially mitigated by compensatory expression of CYP76AD1 . Our selection process could therefore have been biased towards T0 plants with a degree of escaped CYP76AD1 expression, which would also explain the presence of betanin in a number of T1 plants descending from CYP76AD1 shoot-expressing lines ( S9 Fig ). Further experimentation is required to support this hypothesis, but in any case, developmental defects and vegetative betanin expression can be avoided when all three biosynthesis enzymes are driven by AM symbiosis–specific promoters. The most powerful application of our betalain-based AM reporter lies in its ability to noninvasively document colonisation in fully developed root systems over time (Figs 8 and S15 ). This will facilitate answering important questions in the field of AM symbiosis and its potential for agronomic improvement. These include understanding the dynamics of root system colonisation, including the time difference between lateral root emergence and its colonisation, or the differential colonisation susceptibility of different root orders [ 50 ]. Colonisation can be assessed in scenarios where plants compete with each other [ 51 ] or with shoot pests [ 52 ], different nutrient regimes, CO2 concentration, temperatures, soil structures [ 53 ], and other abiotic factors. A second important application is in the survey of induced plant genetic variation or fungal natural genetic variation that impacts on root system colonisation. Here, its use is only limited by the transformability of the plant species, and an important future step will be to test our approach in monocot crops, some of which have already been engineered to express betalain pigments in grains [ 41 ]. Finally, a third application derives from the ability to bulk collect betalain-pigmented root sections produced under AM fungal colonisation, especially when there is very little overall colonisation as is common in some symbiosis signalling pathway mutants or when colonisation is restricted to specific roots. Here, stage-specific early or late promoters will unlock targeted transcriptomic, proteomic, or metabolomic analysis of colonised roots without the dilution effect of non-colonised tissues. Such enrichment strategies may, for example, help in understanding communication mechanisms between AM fungi and plants. Targeted sampling of pigmented roots will also simplify low-throughput microscopy such as cryo-electron microscopy and drastically improve the signal-to-noise ratio inherent in current sampling processes. Selective sampling at red-to-white transition zones may additionally aid time-lapse microscopy of expanding fungal colonisation arrays within roots. In summary, betalains are plant pigments with a strong potential as visual markers for the study of physiological and developmental processes in plants and microorganisms. Here, we have expanded the application of these versatile pigments to add to an ever increasing range of betalain-based technologies as reporters of AM fungi colonisation in plant roots. MycoRed complements currently used fungal visualisation techniques and constitutes a powerful tool that will be of great value for the plant–microbe research community to advance knowledge in the field of AM symbiosis."
} | 5,021 |
36389645 | null | s2 | 7,022 | {
"abstract": "Plant reproduction in metalliferous habitats is challenged by elevated concentrations of metal trace elements in soil. As part of their survival strategy, metal-tolerant plants have adjusted reproductive traits, including seed morphology, dormancy, and germination rate. These traits are particularly relevant, yet poorly understood, in metal hyperaccumulators that are promising candidates for phytoremediation. We assessed seed shape characteristics, dormancy, and germination rate in the hyperaccumulating model species Seeds from non-metallicolous populations were on average 18% bigger than those from metal-contaminated post-mining sites, which contrasts the general expectation about reproductive parts in metallicolous plants. Irrespective of their origin, surface-sterilized seeds had up to ~ 20% higher germination rates and germinated earlier than non-sterilized seeds, hinting at a negative effect of seed-associated microbial communities. Surface sterilization also facilitated the emergence of an endophytic fungus ( Despite species-wide metal tolerance in "
} | 268 |
33420292 | PMC7794327 | pmc | 7,024 | {
"abstract": "In the present study we investigate the microbial community inhabiting As Burgas geothermal spring, located in Ourense (Galicia, Spain). The approximately 23 Gbp of Illumina sequences generated for each replicate revealed a complex microbial community dominated by Bacteria in which Proteobacteria and Aquificae were the two prevalent phyla. An association between the two most prevalent genera, Thermus and Hydrogenobacter , was suggested by the relationship of their metabolism. The high relative abundance of sequences involved in the Calvin–Benson cycle and the reductive TCA cycle unveils the dominance of an autotrophic population. Important pathways from the nitrogen and sulfur cycle are potentially taking place in As Burgas hot spring. In the assembled reads, two complete ORFs matching GH2 beta-galactosidases were found. To assess their functional characterization, the two ORFs were cloned and overexpressed in E. coli . The pTsbg enzyme had activity towards o-Nitrophenyl-β- d -galactopyranoside (ONPG) and p-Nitrophenyl-β- d -fucopyranoside, with high thermal stability and showing maximal activity at 85 °C and pH 6, nevertheless the enzyme failed to hydrolyze lactose. The other enzyme, Tsbg, was unable to hydrolyze even ONPG or lactose. This finding highlights the challenge of finding novel active enzymes based only on their sequence.",
"conclusion": "Conclusions The taxonomical analysis of As Burgas hot spring metagenome reveals a microbial community dominated by Bacteria in which Proteobacteria (68.25 ± 3.59%) and Aquificae (11.24 ± 1.15%) are the most abundant phyla. The prevalence of the genera Thermus (15.77%) and Hydrogenobacter (8.56%) and the relation of their metabolism suggests an association between these two genera. Moreover, the high relative abundance of sequences involved in the Calvin–Benson cycle and sequences annotated as key for the reductive TCA cycle unveils the dominance of an autotrophic population. Important pathways from the nitrogen and sulfur cycle such as DNRA, nitrification, or sulfur oxidation are potentially taking place in As Burgas hot spring, as was determined by the functional annotation of the metagenomic reads and in accordance with the microbial composition of the ecosystem. After assembling the metagenomic reads two complete ORFs annotated as β-galactosidases were found. Both of them showed 100% homology with T. scotoductus SA-01 and were cloned and overexpressed in E. coli . The enzyme Tsbg lacked β-galactosidase activity using ONPG and lactose as substrates. On the contrary, pTsbg showed β-galactosidase activity towards ONPG but was not able to hydrolyze lactose; it showed β-fucosidase activity on the substrate p-Nitrophenyl-β- d -fucopyranoside, which suggests a priori unexpected biotechnological application. Once more this result reveals that the presence of a gene in a metagenome does not mean that it is active in the way predicted from the sequence, and highlights the importance of combining both functional and sequence metagenomics to find novel enzymes from metagenomes. Our culture-independent study has provided an insight into the diversity of the microorganisms that inhabit As Burgas thermal environment, in an attempt to find novel β-galactosidases. Future research should be directed to characterize new environments, which will lead to better understanding of their ecological differences, and to find new enzymes of interest.",
"introduction": "Introduction Thermophiles, growing optimally at temperatures over 55 °C, are found in hot environments such as fumaroles, hydrothermal vents, hot springs, or deserts 1 – 4 . Apart from high temperatures, these habitats usually show other harsh conditions like extreme pH or high salt concentration. Therefore, the study of microorganisms inhabiting hot environments and their enzymes has drawn considerable interest from a biotechnological point of view, as these extremophiles have features suitable for industrial processes, in which high stability and activity at elevated temperatures, as well as high tolerance toward various reagents and solvents, are required.\n The potential of thermal water as a source of novel thermostable biocatalysts has been demonstrated since a considerable number of thermozymes such as lipases 5 , polymerases 6 , or cellulases 7 , among others, have been isolated from hot springs. In recent years, metagenomics has become a powerful tool to explore the microbiological community composition and activity of extreme environments, like hot springs, whose conditions are difficult to reproduce in a lab-bench. The metagenomic approach is based on the study of the whole environmental microbial DNA (metagenome) that is directly sequenced, in what is called sequence metagenomics, or ligated into a vector and transformed to generate a metagenomic library, in what is known as functional metagenomics. Sequence metagenomics has enabled the study of a large number of hot springs extended all over the world like Tuwa, Lasundra and Unkeshwar hot springs in India 8 – 10 , a hot spring in Kamchatka, Russia 11 , Sungai Klah hot spring in Malaysia 12 or several hot springs in Yellowstone National Park USA 13 , 14 . β-galactosidases catalyze the hydrolysis of lactose to glucose and galactose, and they have drawn considerable interest from the biotechnological industry for the production of low-lactose milk and the revalorization of whey. Furthermore, some β-galactosidases can transfer the galactosyl residue of lactose carrying transgalactosylations reactions, which are frequently used for the synthesis of galacto-oligosaccharides (GOS), with prebiotic effects 15 , and to synthesize other galactosylated products 16 . Metagenomics has contributed to the exploration of heated habitats such as hot springs, either for ecological study or for bioprospection of novel enzymes. Some thermal β-galactosidases have been isolated from hot springs using functional metagenomics 17 , 18 but there is only one reported study of thermostable β-galactosidases found in hot springs through sequence metagenomics 19 . In the province of Ourense (Spain), there are at least 13 geothermal springs widespread across the region. Because of its accessibility and its historical importance, in this study, we have focused on As Burgas hot spring. Although some authors have previously investigated its water composition 20 , or its culturable microorganisms 21 , the present is the first reported metagenomic study of this hot spring. From the unassembled reads obtained through shotgun metagenomic DNA sequencing, we have assessed taxonomical and functional characteristics of As Burgas water population. Then, metagenomic sequences were assembled and annotated, finding two potential β-galactosidases that have been cloned, purified, and characterized.",
"discussion": "Results and discussion Taxonomic and functional assignment of metagenomic sequences The BW1 and BW2 metagenomes consisted of 747,684 and 761,635 high quality reads, respectively (Table 1 ). There was no significant differences between the two samples (data not shown), thus the relative abundances of assigned reads to each taxon or function were expressed as an average. The taxonomical community analysis revealed a predominance of Bacteria (93.11 ± 1.86%), followed by Archaea (6.18 ± 1.84%), Eukaryota (0.67 ± 0.009%), and Viruses (0.02 ± 0.03%) (Fig. 1 A). From the 27 bacterial phyla detected, the most abundant were Proteobacteria (68.25 ± 3.59%), Aquificae (11.24 ± 1.15%), Deinococcus-Thermus (5.26 ± 1.01%), Firmicutes (4.29 ± 0.53%) and Bacteroidetes (1.95 ± 0.19%) (Fig. 1 B). More detailed information on the community structure is provided in supplementary material (Supplementary Tables S1 , S2 ). Table 1 Characteristics of the paired-end raw sequences obtained after Illumina MiSeq sequencing of As Burgas water before and after quality control (QC) with PRINSEQ. BW1 BW2 Read 1 Read 2 Read 1 Read 2 Before PRINSEQ QC Number sequences 867,096 867,096 873,846 873,846 Total bases 227,341,174 232,496,706 230,953,710 235,685,441 Seq. length (bp) 262.19 ± 46.53 268.13 ± 47.31 264.30 ± 44.08 269.71 ± 45.04 Mean GC content (%) 54.95 ± 11.23 55.32 ± 11.61 54.09 ± 11.76 54.46 ± 12.20 Number of pairs 867 096 (100% sequences) 873 846 (100% sequences) After PRINSEQ QC Number sequences 747,684 747,684 761,635 761,635 Total bases 193,410,210 192,903,192 199,007,260 198,412,539 Seq. length (bp) 258.68 ± 46.62 258.00 ± 45.55 261.29 ± 43.84 260.51 ± 42.86 Mean GC content (%) 54.22 ± 11.41 54.51 ± 11.63 53.31 ± 11.87 53.60 ± 12.11 Number of pairs 747,684 (100% sequences) 761,635 (100.00% sequences) Read 1 and Read 2 correspond to the paired reads. Figure 1 ( A ) Taxonomic assignment of the reads at domain level. The chart represents the percentage of reads assigned to each domain (relative abundance expressed as a percentage from the total assigned reads). ( B ) Taxonomic assignment of sequences within Bacteria domain. Percentage of reads annotated at phylum level is represented. Others include those phyla with less than 0.7% sequences assigned (Candidatus Poribacteria, Chlamydiae, Chlorobi, Chrysiogenetes, Deferribacteres, Dictyoglomi, Elusimicrobia, Fibrobacteres, Fusobacteria, Gemmatimonadetes, Lentisphaerae, Spirochaetes, Synergistetes, Tenericutes, Thermotogae, unclassified (derived from Bacteria) and Verrucomicrobia). Graphs were created with Excel for Windows version 14.0.0. The predominance of Bacteria followed by Archaea was also found in the soil and the water of the Lobios hot spring, located in the same Galician region 22 , 23 . Nevertheless, in contrast with the significant relative abundance of Proteobacteria found in As Burgas water, Acidobacteria was the major phylum in the Lobios sediment while Deinococcus-Thermus dominated the Lobios water. These differences might be due to the influence of physicochemical parameters, such as pH and temperature, on the microbial community composition. In fact, As Burgas water has a lower temperature (66.3 °C) and pH (7.56) 20 than Lobios water (76 °C, pH 8.2) 23 . It is also important to consider that taxonomical assignment in the study of Lobios water was done using assembled reads rather than the unassembled reads and thus, real phyla abundance might be lost 24 . Temperature has been reported as a key factor in the prevalence of Proteobacteria. Dominance of this phylum has been found in geographically distant but moderate-temperature (29–65 °C) geothermal springs like Deulajhari and Tattapani in India 25 , 26 , Aguas Calientes in the Amazon rainforest of Perú 27 , Chiraleu, Ciocaia, and Mihai Bravu in Romania 28 or El Coquito in the Colombian Andes 29 . Moreover, Power et al. 30 found that the phyla Proteobacteria and Aquificae dominated in 925 geothermal springs in New Zealand (65.2% total average relative abundance across all springs), especially in hot springs with temperatures below 50 °C where Proteobacteria were the most abundant phylum. Similar results were found by Najar et al. 31 that studied the microbial diversity of Polok (75–77 °C) and Borong (50–52 °C) hot springs in India finding that the dominance of the Phylum Proteobacteria was more pronounced in Borong hot spring, which had a lower temperature. Another distinctive aspect of Proteobacteria is that they are known to tolerate a higher concentration of sulfur and use reduced compounds of this element as an electron donor during their physiological processes 31 . Aquificae is the second most abundant phylum in the As Burgas ecosystem consisting of 11.24 ± 1.15% of the metagenome. This phylum encompasses strictly thermophilic bacteria with an optimum growth temperature above 65 °C 32 . The high relative abundance of Aquificae occurs in other hot springs with a broad range of pH and temperatures, including six geothermal springs in the Philippines (60–92 °C, pH 3.72–6.58) 33 , the Mihai Bravu in Romania 28 and the Ganzi Prefecture hot springs in China 34 . Members of this phylum dominate in environments with limited biomass and low ions concentrations, such as the King-Yu, Nono-Yu Koya, Yamanojo, and Jinata Onsen hot springs in Japan 35 , 36 , among others. Most Aquificae representatives are hydrogen-oxidizing bacteria that use hydrogen as electron donor, carbon dioxide as carbon source, and oxygen as the final electron acceptor. Alternatively, some species can oxidize thiosulfate or sulfur as energy sources 32 . Compared with other geothermal springs worldwide, the community structure of As Burgas is very similar to the Mihai-Bravu in Romania, which has similar temperature and pH (65 °C, pH 7.91) 28 , as both of the springs were dominated by phyla Proteobacteria, Aquificae, and Deinococcus-Thermus. This result suggests that chemolitotrophy by oxidation of H 2 and reduced sulfur compounds are important metabolic processes in these springs and that the members of phylum Aquificae play a main role in primary productivity in this community. Focusing on the genus level, the three most abundant genera in As Burgas water were Thermus (21,221 sequences (15.77%)), Hydrogenobacter (11,517 sequences (8.56%)) and Thiobacillus (5659 sequences (4.20%)). Thermus spp. has been traditionally described as heterotrophic thermophilic Gram-negative aerobic bacteria; although most are facultative anaerobes in the absence of oxygen and presence of nitrate 37 but some species from the genera have shown the ability to grow mixotrophically 38 , 39 . The dominance of Thermus in As Burgas water is consistent with this genus’ optimal growth temperature (62–75 °C) 37 , in fact, members of this genus are commonly found in other thermal springs with temperatures above 60 °C. For example, in the hot springs of Heart Lake Geyser Basin in Yellowstone National Park, a shift in the microbial population was detected from several cyanobacterial genera at 44 °C to the observation of Thermus members at 63 °C and finally a predominance of this genus in the 75 °C geysers 40 . Thermus genus was also dominant in the 65 °C Mihai-Bravu spring in Romania 28 and the Rupi Basin geothermal spring in Bulgaria 41 . This genus also dominates the water of the geographically close Lobios hot spring in Ourense 23 . Hydrogenobacter was the second most abundant genus in As Burgas. These extremely thermophilic representatives of phylum Aquificae are obligate chemolithotrophic organisms with anaerobic anabolism, but aerobic catabolism 42 . High relative abundance and co-existence of Hydrogenobacter with Thermus genera was found in Lobios (Ourense) 23 , Rupi Basin (Bulgaria) 41 , Elegedi (Eritrea) 43 and in Niujie (China) 44 thermal springs. The association between hydrogen-oxidizing Hydrogenobacter with hydrogen-producing Thermus in these hot springs suggests hydrogen metabolism as an essential component of these ecosystems. In addition to the community analysis, functional analysis was performed with MG-RAST. The sequences that passed MG-RAST quality control produced 347,814 and 368,188 predicted protein-coding features for BW1 and BW2, respectively. From these, 52.1% (181,371) for BW1 and 52.8% (194,410) for sample BW2, were assigned annotation by MG-RAST to SEED functional categories (Subsystems) (Table 2 ). Among the functional categories at Level 1 identified by the SEED subsystems annotation, the four most dominant were the clustering-based subsystems (functional coupling evidence but unknown function; 13.44 ± 0.55%), protein metabolism (10.77 ± 0.17%), carbohydrates (9.55 ± 0.11%) and miscellaneous (6.42 ± 0.24%), based in the relative abundance of assigned reads (Fig. 2 ). More detailed information is provided in supplementary material (Supplementary Table S3 ). Similar results were found in Lobios hot spring water where the clustering-based subsystems were found as the largest category followed by miscellaneous, carbohydrates, and protein metabolism 23 . The predominance of the clustering-based subsystems in both metagenomes shows how limited our knowledge is regarding the functional annotation of the microbial proteome, as the precise functions of most proteins in metabolic pathways are yet to be revealed. Thus, the strategy of discovering new activities by a functional-driven metagenomic approach rises as a valid alternative to overcome such challenges. Table 2 MG-RAST resume of the two replicates of As Burgas water metagenome (BW1 and BW2 samples MG-RAST Ids mgm4709017.3 and mgm4709018.3 respectively). BW1 BW2 Processed Predicted protein features 347,814 368,188 Predicted rRNA features 45,681 47,293 Alignment Identified protein features 181,371 194,410 Identified rRNA features 452 519 Annotation Identified functional categories 152,744 163,890 Figure 2 Functional profile of As Burgas hot spring at SEED subsystems level 1. Of 347,814 and 368,188 protein-coding regions predicted from BW1 and BW2 reads by MG-RAST, 52.1% (181,371) and 52.8% (194,410) were assigned by MG-RAST to SEED functional categories (Subsystems). The percentage of reads assigned to each function is represented. Others include those functions with less than 2.11% reads assigned (Cell Division and Cell Cycle; Dormancy and Sporulation; Fatty Acids, Lipids, and Isoprenoids; Iron acquisition and metabolism; Metabolism of Aromatic Compounds; Phages, Prophages, Transposable elements, Plasmids; Phosphorus Metabolism; Photosynthesis; Potassium metabolism; Regulation and Cell signaling; Secondary Metabolism; Sulfur Metabolism; Motility and Chemotaxis). Graph was created with Excel for Windows version 14.0.0. Since O 2 concentration is reduced in hot springs due to lower oxygen solubility in heated water, other electron acceptors are important, such as nitrate, elemental S, sulfate, or CO 2 . Thus, an overrepresentation of sequences related to nitrogen and sulfur metabolism could be expected in these kinds of habitats. Consequently, in this study, we specially review those pathways involved in nitrogen and sulfur metabolism. Analysis of the nitrogen metabolism at subsystem level 3 revealed a high abundance of sequences involved in nitrate and nitrite ammonification, also known as dissimilatory nitrate reduction to ammonium (DNRA) (Table 3 ). DNRA is the result of anaerobic respiration by chemoorganoheterotrophic microorganisms using nitrate (NO 3 − ) as a final electron acceptor, producing ammonia (NH 4 + ). This metabolic pathway results in nitrogen (N) conservation in the ecosystems and is favored in habitats where NO 3 − is limiting in relation to organic carbon 45 . Therefore, the low NO 3 - content found in As Burgas water in comparison to other proximal geothermal springs such as Outariz, Tinteiro and Chavasqueira 20 might be promoting the prevalence of DNRA bacteria like Proteobacteria 46 , 47 . This result is in accordance with the dominance of phylum Proteobacteria found in the taxonomical analysis of As Burgas metagenomic sequences. Nevertheless, it is important to remark that the presence or relative abundance of a gene in a metagenome does not mean that it is active. Metatranscriptomic studies are necessary to determine if DNRA is an important pathway in this ecosystem. In this aspect, other studies have reported the occurrence of an active DNRA pathway in some hot springs 48 – 50 . Table 3 Analysis of Subsystems at level 3. From the 28 subsystems at level 3 registered by MG-RAST, only those subsystems with more than 2000 reads assigned were collected in the table. SEED subsystems No. of reads Level 1 Level 2 Level 3 BW1 BW2 Amino acids and derivatives Branched-chain amino acids branched-chain_amino_acid_biosynthesis 2799 2930 Lysine, threonine, methionine, and cysteine methionine_biosynthesis 3119 3134 Carbohydrates CO 2 fixation Calvin-benson_cycle 2204 2351 One-carbon Metabolism Serine-glyoxylate_cycle 3311 3219 Cell Wall and capsule NULL Peptidoglycan_biosynthesis 2039 2106 Clustering-based subsystems NULL Bacterial_cell_division 2293 2119 Cofactors, vitamins, prosthetic groups, pigments Tetrapyrroles Heme_and_siroheme_biosynthesis 2244 2205 Folate and pterines YgfZ 2527 2741 DNA metabolism DNA replication DNA-replication 2446 2332 DNA repair DNA_repair,_UvrABC_system 2038 2019 DNA repair DNA_repair,_bacterial 2532 2661 Fatty acids, lipids, and isoprenoids Fatty acids Fatty_Acid_biosynthesis_FASII 2365 2459 Membrane transport ABC transporters ABC_transporter_branched-chain_amino_acid_(TC_3.A.1.4.1) 2598 2380 Motility and chemotaxis Flagellar motility in prokaryota Flagellum 3061 2874 Nitrogen metabolism NULL Ammonia_assimilation 2028 2279 NULL Nitrate_and_nitrite_ammonification 4965 4169 Nucleosides and nucleotides Purines De_novo_purine_biosynthesis 2868 3365 Purines Purine_conversions 2695 2672 Phosphorus metabolism NULL Phosphate_metabolism 3204 3396 Protein metabolism Protein folding Protein_chaperones 2584 2551 Protein degradation Proteolysis_in_bacteria,_ATP-dependent 2054 1933 Protein biosynthesis Ribosome_LSU_bacterial 4896 4258 Protein biosynthesis Ribosome_SSU_bacterial 3084 2958 Protein biosynthesis Universal_GTPases 2110 2012 Respiration Electron donating reactions Respiratory_complex_I 5524 5578 Electron accepting reactions Terminal_cytochrome_C_oxidases 3054 2826 RNA metabolism Transcription RNA_polymerase_bacterial 3475 3402 RNA processing and modification tRNA_modification_archaea 1801 2035 Sulfur metabolism NULL Sulfur_oxidation 3112 2665 A high number of reads with similarity to ammonia assimilation were found in As Burgas water metagenome (Table 3 ). The abundance of sequences annotated as glutamine synthetase and glutamate synthase, key enzymes in this metabolic pathway, were already expected as they are widely distributed among microorganisms, playing an important role in nitrogen metabolism 51 . Reads annotated as Nitrogenase ( Nif ) genes for nitrogen fixation were also abundant in the metagenome. Although the distribution of these genes seems to be widespread in nature, as they have been described in different environments 52 including hot springs 53 – 55 , active nitrogen fixation has been reported in several thermophilic organisms 56 , 57 . Nitrogen fixation could be important in As Burgas as this ecosystem harbors phyla with known diazotrophic representatives such as Proteobacteria and the phylum Aquificae in which some members of Hydrogenobacter were recently described as nitrogen-fixing bacteria 58 . Furthermore, nitrogen fixation has been demonstrated in other geothermal springs such as several hot springs from Yellowstone National Park 59 , 60 and Nakabusa hot springs in Japan 61 , among others. Nitrification might also take place in As Burgas ecosystem, as sequences matching the ammonia monooxigenase (AMO) enzyme were detected in the two metagenomes. This enzyme catalyzes the oxidation of ammonia to hydroxylamine and it is essential for chemolithotrophic ammonia-oxidizing bacteria. The oxidation of ammonia to nitrite in As Burgas hot spring water could be associated with the abundant Proteobacteria, as several members of this phylum have been described as autotrophic nitrifiers 62 , 63 . Another important component in the nitrogen cycle is denitrification, which competes with DNRA, due to the dependence of both metabolic pathways on NO 3 − . Members of the genus Thermus are important denitrifiers in heated ecosystems, as they can perform facultative anaerobic respiration using NO 3 − as the final electron acceptor, producing N 2 or nitrous oxide (N 2 O) 37 . In addition, representatives from another abundant genus in As Burgas, Thiobacillus , also perform denitrification processes 64 , 65 . Unexpectedly, not many sequences related to denitrification were annotated in the metagenome (771 sequences in BW1 and 692 in BW2), even though these potential denitrifiers were two of the most abundant genera found in As Burgas. At the function level, sequences related to denitrification such as nitrite reductase ( nir ), nitric-oxide reductase ( nor ) and nitrous-oxide reductase ( nos ), were present in both metagenomes, but not in high abundance. Functions involved in sulfur oxidation were also abundant in As Burgas water (Table 3 ). The high abundance of these sequences can be attributed to the prevalence of Proteobacteria in the microbial community, since this is an important sulfur-oxidizing phyla 66 , 67 . Numerous members of the abundant phylum Aquificae and Deinococcus-Thermus can oxidize thiosulphate or sulfur as an energy source and thus harbor sox genes 38 , 39 , 68 . Moreover, some sulfur-oxidizing bacterial species of the genus Thermus and Thiobacillus are also nitrate-reducing bacteria that accept electrons from the oxidation of reduced inorganic sulfur compounds and have been frequently identified in a diverse range geothermal springs 38 , 39 , 64 . Therefore, sulfur oxidation coupled with denitrification could be an important source of energy for carbon fixation in this hot spring, as was previously described for other hot springs 69 and diverse heated habitats like hydrothermal vents 70 .\n In relation to carbon-fixation metabolism, a high abundance of sequences associated with the reductive pentose phosphate cycle (Calvin–Benson cycle) (Table 3 ) was found. This cycle has been described as the principal pathway of carbon fixation in Cyanobacteria and Proteobacteria 71 and some studies have reported the presence of genes related to this cycle in several Thermus strains 72 . The number of sequences affiliated to the tricarboxylic acid (TCA) cycle was also representative (1742 for BW1 and 1775 for BW2), but slightly lower than those for the Calvin-Benson cycle. Most enzymes involved in the TCA cycle function in an oxidative way (releasing stored energy through the oxidation of acetyl-CoA into ATP and CO 2 ), but they can be used by some microorganisms in a reductive TCA cycle that is essentially the oxidative TCA cycle running in reverse, leading to the fixation of two molecules of CO 2 and the production of one molecule of acetyl-CoA 73 . Reverse TCA is suggested to be a more ancient pathway for carbon fixation 74 and the main route for primary production at high temperatures (above 70 °C) 75 . The ability to perform the reverse TCA cycle is typical of bacteria from the phylum Aquificae such as Hydrogenobacter 75 , 76 and was confirmed in a variety of anaerobic and microaerobic bacteria, including several proteobacteria 73 . Moreover, reads annotated as pyruvate: ferredoxin oxidoreductases (POR) were found in the two metagenomes. POR enzyme decarboxylates pyruvate to form acetyl-CoA and is crucial for the reverse TCA cycle, as it is able to act as pyruvate synthase catalyzing the reverse reaction 77 , 78 . The high abundance of sequences involved in the Calvin–Benson and reverse TCA cycles reveals that autotrophy is an important source of energy of the ecosystem, as was expected, in accordance with the low organic content of this kind of thermal habitats. A high relative abundance of reads associated with one-carbon metabolism such as YgfZ, a folate-binding regulatory protein 79 and sequences related to the serine-glyoxylate cycle (Table 3 ) were identified. Serine-glyoxylate cycle is a carbon assimilation pathway found in aerobic methanotrophs belonging to the classes Alpha-, Gammaproteobacteria, and the phylum Verrucomicrobia 80 . Sequences annotated as crucial enzymes for methanotrophic metabolism such as methane monooxygenase, methanol dehydrogenase or hydroxypyruvate reductase 81 , 82 were present in the two replicates of As Burgas metagenome. A similar result was previously reported for the nearby Lobios hot spring, in which a high abundance of sequences associated with YgfZ and the serine-glyoxylate cycle was also detected. However, Lobios metagenome lacks the methane monooxygenase and methanol dehydrogenase encoding genes 23 . The methanogenic microorganisms frequently found in hot springs microbial mats 83 would be the methane producers for methanotrophs in As Burgas. In fact, sequences annotated to the methanogenic orders Methanobacteriales, Methanocellales, Methanomicrobiales, Methanosarcinales, and Methanopyrales were found among the archaeal reads in the taxonomical analysis of As Burgas. Moreover, sequences matching several proteins involved in methanogenesis such as heterodisulfite reductase, formate dehydrogenase, and carbon monoxide dehydrogenase were found in the metagenome. Nevertheless, the presence of methyl-coenzyme M reductase gene, a key enzyme in methanogenesis 84 , was not detected in the metagenome. Sequence assembly and screening for sequences annotated as β-galactosidase From the 873,846 quality paired-end BW2 raw reads, a total of 28,296 contigs with a maximum length of 263,962 bp and an average length of 932 bp (26,379,150 bp) were obtained using SPADes. From these, 26,417 sequences (93.36%) were annotated to the functional level with the MG-RAST. A search for β-galactosidase sequences with this tool resulted in only 2 sequences that harbor complete coding ORFs that were chosen for further study. Both selected ORFs belong to Themus scotoductus SA-01, as their nucleotidic sequence had 100% alignment with the T. scotoductus SA-01 complete genome, deposited in the GenBank by Gounder et al. 85 under the accession number CP001962.1. This result is consistent with the dominance of Thermus genera reported in the taxonomical analysis. The deduced protein sequence of Tsbg and pTsbg consisted of 574 and 690 residues, respectively, and showed 100% homology with two different β-galactosidases from T.scotoductus with GeneBank accession number WP_015717803.1 and WP_015717801.1 for Tsbg and pTsbg respectively. The two proteins have been registered in GeneBank as part of a whole shotgun genome sequencing and annotation, but their cloning and expression have never been reported, therefore we selected both ORFs for further study and characterization. Both protein sequences contain a Glycosyl hydrolases family 2 (GH2) TIM barrel Domain (PF02836) according to Pfam protein database 86 . Therefore they are within the GH2 superfamily, in agreement with other thermostable microbial β-galactosidases like those from Thermotoga maritima 87 or Streptococcus thermophilus 88 . Cloning, expression, and purification of T. scotoductus β-galactosidases Both sequences were efficiently amplified, cloned in pDEST-527 vector and overexpressed in T7 Express E. coli . As no activity towards ONPG or lactose was detected for Tsbg, the gene was cloned in pDEST-527 without the histidine tag, in an attempt to discard the possibility of an incorrect folding or blocking of the active site due to the tag. Nevertheless, purified Tsbg protein without tag did not show activity using both lactose and ONPG as substrates. The lack of β-galactosidase activity in Tsbg is similar to the results obtained for T. scotoductus DSM 8553, as no β-galactosidase activity was detected in this strain 89 , 90 , which suggests that the cause is the protein itself rather than the expression host . Therefore, the successive characterization steps were only performed with the pTsbg. Effect of pH and temperature on activity and stability of recombinant pTsbg pTsbg showed maximal activity at pH 6.0 in Britton-Robinson buffer using ONPG as substrate (Fig. 3 A). This result is slightly lower than the optimum pH reported for other bacteria from Thermus genera like T. thermophilus HB8 91 , T. thermophilus HB27 92 and it is comparable to the optimal pH reported for other thermostable β-galactosidases such as those from Bacillus licheniformis 93 , Caldicellulosiruptor saccharolyticus 94 , Marinomonas sp. BSi20414 95 and much lower than the pH 7.8 reported for T. oshimai DSM 12092 β-galactosidase 96 . Figure 3 Effect of pH ( A ) and temperature ( B ) on the activity of pTsbg in Z Buffer using ONPG (4 mg mL −1 ) as substrate. Graphic was created using GraphPad Prism 6 for Windows (GraphPad Software, San Diego, California USA, www.graphpad.com ). As shown in Fig. 3 B, maximal pTsbg β-galactosidase activity towards ONPG was found at 85 °C. This optimal temperature is higher than described using the same substrate for other counterparts of the genus Thermus such as T.thermophilus HB8 91 , T.thermophilus HB27 92 , T. aquaticus YT‐1 97 , T. oshimai DSM 12092 96 and is the same reported as optimal to T.thermophilus KNOUC114 β-galactosidase 98 . When compared to other genera of thermophilic bacteria β-galactosidases, pTsbg showed higher optimal temperature than documented for the extremely thermophilic C. saccharolyticus and Marinomonas sp. BSi20414, which showed the optimum temperature at 80 °C and 60 °C respectively 94 , 95 . Nevertheless, the optimal temperature described for Thermotoga naphthophila RUK-10 β-galactosidase is higher 99 . In relation to the thermal stability, pTsbg was able to retain up to 60% of its maximal activity towards ONPG after 24 h of incubation at 75 °C (Fig. 4 ). Figure 4 Effect of temperature on the stability of purified pTsbg. Graphic was created using GraphPad Prism 6 for Windows (GraphPad Software, San Diego, California USA, www.graphpad.com ). Determination of substrate specificity of pTsbg Although substrate specificity of the enzyme was studied using the eight chromogenic substrates described in the “ Methods ” section, pTsbg was only active towards ONPG and p-Nitrophenyl-β- d -fucopyranoside. Moreover, the enzyme was unable to hydrolyze lactose and no transgalactosylation was observed in the presence of this substrate, as was determined by HPLC after carrying the reaction with 40% lactose at 70 °C and using a mix of galactose, glucose, lactose, raffinose and stachyose as standard (data not shown). The preference for β-linked galactosidic substrates such as ONPG or p-Nitrophenyl-β- d -fucopyranoside over lactose has been frequently described in the characterization of β-galactosidases 99 , 100 . Similar to our results with pTsbg, other studies have reported β-galactosidases with activity towards ONPG but unable to hydrolyze their natural substrate lactose in vitro such as YesZ β-galactosidase from Bacillus subtilis 101 or the β-Gal II from Bifidobacterium adolescentis DSM 20083 102 . The lack of β-galactosidase activity towards lactose reduces considerably the biotechnological potential of pTsbg, as it could not be applied to produce GOS from lactose and to generate lactose-free dairy products. Nevertheless, more studies focused on the fucosidase activity should be conducted, since pTsbg showed high activity with p-Nitrophenyl-β- d -fucopyranoside and may harbor fucosyltransferase activity that could be used for the synthesis of fucosylated oligosaccharides (FUCOS) with biological interest 103 such as those from human milk."
} | 8,709 |
31135958 | null | s2 | 7,025 | {
"abstract": "Plastids evolved from a cyanobacterium that was engulfed by a heterotrophic eukaryotic host and became a stable organelle. Some of the resulting eukaryotic algae entered into a number of secondary endosymbioses with diverse eukaryotic hosts. These events had major consequences on the evolution and diversification of life on Earth. Although almost all plastid diversity derives from a single endosymbiotic event, the analysis of nuclear genomes of plastid-bearing lineages has revealed a mosaic origin of plastid-related genes. In addition to cyanobacterial genes, plastids recruited for their functioning eukaryotic proteins encoded by the host nucleus and also bacterial proteins of noncyanobacterial origin. Therefore, plastid proteins and plastid-localised metabolic pathways evolved by tinkering and using gene toolkits from different sources. This mixed heritage seems especially complex in secondary algae containing green plastids, the acquisition of which appears to have been facilitated by many previous acquisitions of red algal genes (the 'red carpet hypothesis')."
} | 269 |
34790178 | PMC8591293 | pmc | 7,028 | {
"abstract": "Light is a ubiquitous source of both energy and information in surface environments, and regulates gene expression not only in photosynthetic microorganisms, but in a broad range of photoheterotrophic and heterotrophic microbes as well. Actinobacteria are keystone species in surface freshwater environments, where the ability to sense light could allow them to coordinate periods of nutrient uptake and metabolic activity with primary production. The model freshwater Actinobacteria Rhodoluna ( R. ) lacicola strain MWH-Ta8 and Aurantimicrobium ( A. ) photophilum strain MWH-Mo1 grow faster in the light than in the dark, but do not use light energy to support growth. Here, we characterize transcription throughout a light-dark cycle in R. lacicola and A. photophilum . In both species, some genes encoding carbohydrate metabolism and storage are upregulated in the light. However, expression of genes of the TCA cycle is only coordinated with light availability in R. lacicola. In fact, the majority of genes that respond to light and darkness in these two species are different, even though their light-responsive phenotypes are similar. The ability to respond to light and darkness may be widespread in freshwater Actinobacteria, but the genetic networks controlled by these two stimuli may vary significantly.",
"introduction": "Introduction Light is a ubiquitous resource in surface environments, and widely used by microbes. In fact, light-sensing proteins that control gene expression are common in photosynthetic microbes, photoheterotrophs, and non-phototrophic heterotrophs ( Eelderink-Chen et al., 2021 ). The organisms that do not use light energy for carbon fixation can use it for supplementary energy ( Calisto et al., 2021 ), phototaxis ( Wilde and Mullineaux, 2017 ), or to entrain circadian rhythms ( Sartor et al., 2019 ). In non-phototrophic bacteria, light regulates multiple biological processes, including motility, pigment production, and stress responses ( Burchard and Dworkin, 1966 ; Bhaya, 2004 ; Ziegelhoffer and Donohue, 2009 ). In photoheterotrophs, light often regulates expression of the photosystems, but may also regulate expression of the biosynthetic pathways of pigments or other photoactive cofactors, electron transport pathways, and the metabolic pathways that intersect with those ( Frühwirth et al., 2012 ; Kumka et al., 2017 ; Navid et al., 2019 ). These pathways can also be regulated by oxygen tension, nutrient availability, and other environmental factors, resulting in complex regulatory networks in photoheterotrophs. In illuminated freshwater environments, Actinobacteria in the Microbacteriaceae family are ubiquitous and abundant keystone species which mediate fluxes of organic carbon and nitrogen, reduced sulfur, and vitamins ( Eiler et al., 2012 ; Garcia et al., 2018 ; Linz et al., 2018 ). The freshwater clades have in common small genomes (<2 Mbp) with low GC content compared to other Actinobacteria (∼50% GC), and a variety of auxotrophies ( Neuenschwander et al., 2018 ). Both environmental metagenomic analyses and laboratory studies suggest that many members of these clades may be photoheterotrophs: actinorhodopsins and heliorhodopsins are common in their genomes, and both rhodopsin types can act as light-activated proton pumps ( Ghai et al., 2014 ; Keffer et al., 2015 ; Dwulit-Smith et al., 2018 ; Pushkarev et al., 2018 ; Maresca et al., 2019 ). Our previous work demonstrated that two species of these freshwater Actinobacteria, Rhodoluna ( R. ) lacicola strain MWH-Ta8 and Aurantimicrobium (A.) photophilum strain MWH-Mo1, grow faster in blue light than in the dark, even though neither has a functional rhodopsin under laboratory conditions ( Maresca et al., 2019 ). Both species are free-living heterotrophs Hahn et al. (2021 , 2014) Their growth rate phenotype strongly implies either that the cells have different activities in these two conditions, or that metabolic rates increase in the light. RNA-seq analysis of gene expression in stationary-phase cells grown in constant light or darkness further indicated that cells of both strains grown in constant light had higher expression of carbohydrate transport and metabolism pathways, while cells grown in constant darkness expressed more genes related to protein production and oxidative stress ( Maresca et al., 2019 ). These transcriptional differences suggest that different metabolic pathways are active in light and darkness, and that they are transcriptionally regulated in response to light. To begin to characterize the genetic and regulatory networks underlying the light-enhanced growth phenotype, we grew R. lacicola and A. photophilum in a 12-h light/12-h dark cycle and sequenced RNA from samples collected throughout the cycle. R. lacicola is representative of the Luna-1 clade of freshwater Actinobacteria, and A. photophilum is representative of the Luna-2 clade ( Newton et al., 2011 ). The R. lacicola genome is smaller (∼1.4 Mbp as compared to ∼1.8 Mbp), and the two genomes share 879 genes, representing 65% of the R. lacicola genome and 50% of the A. photophilum genome ( Supplementary Figure 1 ; Maresca et al., 2019 ). We initially predicted that the transcriptional responses of the two strains to light and darkness would be similar, because they belong to the same family (Microbacteriaceae), their genomes are so similar, and their light-enhanced growth phenotypes are similar. Here, we show that light and darkness alter transcription of distinct suites of genes in R. lacicola and A. photophilum , that approximately half of the genes in both are regulated in response to light or darkness, and that although the growth rate phenotypes of these strains are similar, their transcriptional responses to light and darkness are quite different.",
"discussion": "Discussion We hypothesized that R. lacicola and A. photophilum would have similar responses to light, because they belong to the same family (Microbacteriaceae) and their genomes are broadly similar. We had also expected that light would stimulate or repress expression of specific genes, and that the effect would decay over time. Instead, we found that different pathways respond to light in the two strains, and that light and darkness alter transcription differently in both strains. An important difference between the two strains is in energy-conserving pathways: the TCA cycle and the F 1 F 0 ATP synthase. In R. lacicola , expression of the genes of the TCA cycle and the ATP synthase clearly increases in the light and decreases in response to darkness. Further, the expression patterns of carbohydrate metabolism genes suggest that R. lacicola may prioritize early steps of carbohydrate metabolism (glycolysis) in the dark and the TCA cycle as well as synthesis of carbohydrates and storage molecules such as starch in the light. In A. photophilum , there are three different expression patterns among the TCA cycle genes, but only one ( AURMO_01262 , encoding a class II fumarate hydratase) was included in an expression group ( Supplementary Table 4 ), and transcription of genes encoding the ATP synthase appears to increase in response to both light and darkness ( Figure 7 ). Greater expression of the TCA cycle genes in the light would likely contribute to the faster growth observed in the light in R. lacicola ; the fact that A. photophilum has the same phenotype without the same gene expression patterns suggests that a different mechanism might underlie its increased growth rate in the light. An intriguing difference between the two strains is the expression of genes encoding a putative tad/flp pilus and other putative competence genes. Both strains have two putative operons encoding components of the pilus, as well as the com system for competence. In A. photophilum , these genes are not coordinately expressed; in R. lacicola , expression of all of them clearly increases in response to darkness ( Figure 8 ). The ways that genetic repair and exchange by these two species are regulated seem to be so different that they may have very different mechanisms for environmental adaptation and diversification. Rapid microdiversification in related freshwater Actinobacteria has been hypothesized previously ( Mehrshad et al., 2018 ), and consistent, daily uptake and incorporation of exogenous DNA could provide a mechanism for these rapid changes in genetic makeup. Some interesting similarities between the two strains occur in expression of potential light-sensitive proteins and the genes with similar patterns. Expression of the rhodopsin in both strains increases greatly in the dark, and the heliorhodopsin ( heR ) in A. photophilum has a similar expression pattern to the carotenoid biosynthetic genes. In R. lacicola , actR expression is similar to expression of about half of the carotenoid biosynthetic genes; the different expression patterns observed in this pathway suggests that different carotenoid products may be produced at different times. The protein that senses light and signals the cells to change their activity has not yet been identified, though based on the growth rate data, it would likely be a blue-light sensing protein ( Maresca et al., 2019 ). A new type of blue light sensor was recently identified in Leptospirillum ( Xu et al., 2021 ), but no homologs to this gene were found in the genomes of R. lacicola or A. photophilum . We previously hypothesized that the putative CryB-type cryptochrome both strains have could be the light sensor ( Maresca et al., 2019 ). Here, we observed that in R. lacicola , transcription of cryB increases during the light period, increases transiently when the light turns off, then decreases during the rest of the dark period. In contrast, transcription of cryB in A. photophilum appears to decrease through the light period, then increase greatly at the beginning of the dark period. If the CryB homologs in R. lacicola and A. photophilum regulate the light response, this difference in expression could explain why the transcriptional responses in the two strains have such different dynamics. The transcription dynamics observed here raise questions about circadian rhythms. We previously observed that both R. lacicola and A. photophilum lack homologs of the core clock proteins (KaiA, KaiB, and KaiC) that control circadian rhythms in cyanobacteria, and that the only predicted proteins with homology to light-sensing domains such as PAS, GAF, and BLUF domains are the photolyases and cryptochromes ( Maresca et al., 2019 ). The genomes likewise lack homologs to YtvA, a putative blue-light sensor, and the histidine kinase KinC , both of which have PAS domains and are expressed with circadian patterns in Bacillus subtilis ( Eelderink-Chen et al., 2021 ). Although it is tempting to speculate about circadian rhythms in freshwater bacteria that inhabit surface waters and would therefore consistently be exposed to sunlight, evidence of circadian rhythms would have to come from observing an entrained pattern of activity in the absence of the light signal ( Sartor et al., 2019 ). The oxidation state of peroxiredoxin activity has been suggested to be an indicator of circadian cycles in eukaryotes, bacteria, and archaea ( Edgar et al., 2012 ). Both strains studied here have two peroxiredoxins in their genomes. Transcription of both peroxiredoxins in A. photophilum increases from 1 h after the light turns on until the light turns off, then drops sharply. In contrast, in R. lacicola , the peroxiredoxins have inverse transcription patterns, with one increasing early in the light period and the other decreasing, then reversal of those trends. Although we did not quantify peroxiredoxin oxidation state, peroxiredoxin transcription in both strains appears to vary in coordination with light availability. This may reflect changes in oxidative stress in the light – other genes in this category have similar transcriptional profiles – but is intriguing, given the consistent association of peroxiredoxin activity with circadian rhythms ( Edgar et al., 2012 ). We observed that in R. lacicola , light stimulates or represses expression of most light-responsive genes, and that effect decays over time, leading to curved gene expression profiles with different maxima or minima. The relatively small number of dark-responsive genes, however, suggest that to this strain, darkness is not just an absence of light, but a different stimulus altogether. We see this even more strongly in A. photophilum , where both light and darkness induce transient changes in gene expression. It is possible here, as in Rhodobacter sphaeroides , a light sensitive protein is a master regulator, and any change in light availability disrupts expression of a variety of genes ( Frühwirth et al., 2012 ). Regardless, it suggests that R. lacicola and A. photophilum may have different signal transduction pathways and/or regulatory proteins that respond to light and darkness. Since they clearly have different networks of light- and dark-responsive genes, this is not surprising. These data lay the groundwork for experiments testing the effects of light on physiological and biochemical properties of freshwater Actinobacteria, and identifying the cellular activities that correspond to light-induced transcriptional changes in these organisms."
} | 3,354 |
17849010 | PMC1964534 | pmc | 7,031 | {
"abstract": "Background Several types of networks, such as transcriptional, metabolic or protein-protein interaction networks of various organisms have been constructed, that have provided a variety of insights into metabolism and regulation. Here, we seek to exploit the reaction-based networks of three organisms for comparative genomics. We use concepts from spectral graph theory to systematically determine how differences in basic metabolism of organisms are reflected at the systems level and in the overall topological structures of their metabolic networks. Methodology/Principal Findings Metabolome-based reaction networks of Mycobacterium tuberculosis , Mycobacterium leprae and Escherichia coli have been constructed based on the KEGG LIGAND database, followed by graph spectral analysis of the network to identify hubs as well as the sub-clustering of reactions. The shortest and alternate paths in the reaction networks have also been examined. Sub-cluster profiling demonstrates that reactions of the mycolic acid pathway in mycobacteria form a tightly connected sub-cluster. Identification of hubs reveals reactions involving glutamate to be central to mycobacterial metabolism, and pyruvate to be at the centre of the E. coli metabolome. The analysis of shortest paths between reactions has revealed several paths that are shorter than well established pathways. Conclusions We conclude that severe downsizing of the leprae genome has not significantly altered the global structure of its reaction network but has reduced the total number of alternate paths between its reactions while keeping the shortest paths between them intact. The hubs in the mycobacterial networks that are absent in the human metabolome can be explored as potential drug targets. This work demonstrates the usefulness of constructing metabolome based networks of organisms and the feasibility of their analyses through graph spectral methods. The insights obtained from such studies provide a broad overview of the similarities and differences between organisms, taking comparative genomics studies to a higher dimension.",
"conclusion": "Conclusions The results presented in this work show that the reaction networks of M. tuberculosis , M. leprae and E. coli are scale-free, small-world networks that differ significantly from random networks. The networks of the three organisms have similar network properties despite a difference in the overall sizes of their metabolomes. Graph spectral theory serves as a tool useful for analysing the topological structure and organisation of large complex networks. This technique yields information about the sub-clustering of nodes in the network and identifies the cluster centres by a single numeric computation. Analysis of sub-clusters of the mycobacterial reaction networks detected by this method suggests that modularity of metabolic networks is possibly less well-defined at the level of biochemical reactions; clusters have been discerned well from metabolite networks [21] . It was observed that the top 50 GS hubs of M. tuberculosis and M. leprae exclusively comprised reactions involving L-glutamate while the top GS hubs in E. coli only consisted of reactions involving pyruvate. We showed that the most highly connected biochemical reactions in the reaction network of an organism are not necessarily the reactions most central to the metabolism of that organism. Moreover, the reactions and metabolites forming the centre of the metabolic networks are not common across all organisms, but are specific to the metabolism of each organism. A systematic comparison of the topological properties of the mycobacterial metabolic networks reveals that massive gene decay in M. leprae does not significantly affect the global structure of its metabolic network. We showed that the most highly connected reactions central to the metabolism of both organisms as well as the finer grouping of reactions within the giant component of their networks are essentially conserved and differ from that of E. coli . Additionally, we have determined that metabolic streamlining in the leprae bacillus has led to the preservation of the shortest paths between reactions in its core metabolome while reducing the total number of alternate paths that exist between them. The results obtained in this work are also useful likely to find applications in rational drug design and metabolic engineering.",
"introduction": "Introduction Recent advances in high throughput technologies and network theory have made it possible to reconstruct and analyse large genome-scale networks of organisms in silico. Several types of networks reflecting different aspects of metabolism and regulation in organisms have been reconstructed. The transcriptional networks based on microarray data, protein-protein interaction networks based on high-throughput yeast two-hybrid type of experiments and metabolic networks based on reaction annotation of the individual proteins coded by the genome are some examples. Several of these studies have focused on elucidating the general principles underlying the structure and organisation of metabolic networks of a large number of organisms. For example, a protein–protein interaction network of Saccharomyces cerevisiae constructed based on systematic two-hybrid analyses [1] indicates that highly connected proteins with a central role in the network's architecture are three times more likely to be essential than proteins with only a small number of links to other proteins [2] . Similarly, a transcriptional regulatory network of Escherichia coli , has been reconstructed [3] based on the RegulonDB database as well as published literature, and has been used to identify important structural network motifs and their role in network function. The dynamics of the Saccharomyces cerevisiae biological network has been investigated computationally, with the integration of transcriptional regulatory information and gene-expression data for multiple conditions [4] . Metabolic networks have also been constructed for a number of genomes such as E. coli \n [5] and Staphylococcus aureus \n [6] which have been used to study the metabolic capabilities of organisms and gene essentiality through flux balance analyses. Protein-protein interaction networks have been previously used for comparative genomics [7] , [8] , [9] , [10] , [11] . Here, we seek to exploit the reaction-based networks of three organisms for comparative genomics. We use concepts from spectral graph theory to systematically determine how differences in the basic metabolism of various organisms are reflected at the systems level. In the present study, we have constructed and characterised the metabolic networks of two closely related organisms: Mycobacterium tuberculosis and Mycobacterium leprae , which are obligate intracellular pathogens [12] . A broad comparison with the network of E. coli has also been presented. Both the mycobacteria are important pathogens and hence of interest. Additionally, their comparison is of particular interest given that the genome sequencing of M. leprae has revealed massive gene decay as compared to other mycobacteria. Despite having genomes of comparable sizes the leprae genome codes for only 1406 proteins in comparison to the 3989 proteins in M. tuberculosis , leading to the consideration of the leprae genome as a ‘minimal genome’. The elimination of many important metabolic activities in this organism is thought to result in severe metabolic streamlining [12] . Various concepts from graph theory [13] , [14] , [15] , [16] , [17] have previously been used to construct and analyse metabolic networks for several fully sequenced organisms. Representing metabolic networks as graphs makes them amenable to various analyses, such as the detection of shortest and alternate paths. Such analyses have also resulted in the identification of the highly connected giant strong components of the networks, as well as metabolites central to the network. Graph spectral analysis can be carried out to obtain information on central hubs as well as the sub-clustering and organisation of the metabolic network. We represent the metabolic networks of these organisms in the form of a reaction based graph with the biochemical reactions as the nodes and an edge existing between the nodes if they share at least one metabolite. Such a representation is essential to ascertain the importance of different biochemical reactions in the metabolic networks of these organisms. Alternative representations have been previously used for representing metabolite networks. One is a substrate graph, wherein all substrates are represented by nodes, with edges between them indicating their participation in the same reaction [18] . In another representation, the metabolic network is built up of nodes, the substrates, which are connected to one another through links, which are the actual metabolic reactions. The physical entity of the link is the temporary educt–educt complex itself, in which enzymes provide the catalytic scaffolds for the reactions yielding products, which in turn can become educts for subsequent reactions [13] . Metabolic networks may also be represented as a directed bipartite graph, with two types of nodes indicating reactions and metabolites separately, with edges from metabolites directed towards the reaction they are substrates of and edges from reactions directed towards their products [19] . Further, in our representation of the network, we have chosen to leave out the links generated by currency metabolites. These metabolites are ubiquitously present in metabolic networks and although these substrates are necessary for a given reaction to take place, they cannot be considered as valid intermediates for path finding or establishing biologically meaningful network connections. Currency metabolites have been excluded from metabolic networks in previous studies such as [17] . Besides characterising the constructed networks for their various graph properties, we have systematically determined if the differences observed between these organisms at the genomic level reflect on the overall topology and characteristics of their metabolic networks. To compare features of the metabolic networks of mycobacteria with those of a more standard and well studied organism, we have also constructed and characterised the metabolic network of E. coli . Metabolic networks have properties similar to other real world networks, such as social networks and the World Wide Web. Particularly, they have been shown to be scale-free, with a non-random power-law distribution of node connectivity (number of interactions of each metabolite) and distinguished by the presence of ‘hubs’, a few highly connected nodes that are essential to the integrity and robustness of the network [13] . These networks are also small-world networks, characterised by a low average path length between nodes [13] , [17] , [18] , [20] . Recent studies have revealed metabolic networks to be modular in nature, comprising several small, functional modules that combine together in a hierarchical manner to form larger, less cohesive units [21] , [22] . In this work, we have studied the topological organisation of the constructed metabolic networks by using concepts from spectral graph theory. The graph spectral method has been applied earlier by our group in the identification of side chain and backbone clusters in proteins and to identify amino acid residues important for protein structure, folding, stability, function and dynamics [23] , [24] . It has also been used to successfully identify domains in multi-domain proteins [25] and to detect clusters of structurally similar proteins in protein chain universe graphs [26] . Bu and co-workers [27] have applied spectral analysis to study the topological structure of the protein interaction network in yeast. Jernigan and co-workers [28] have used spectral graph theory for analysing the functional clustering of the yeast protein–protein interaction network. Here, we explore the applications of the spectral method in analysing the topological organisation of large reaction networks. We show that this method is useful for identifying sub-clusters of reactions in the networks by a simple one step numerical computation.",
"discussion": "Results and Discussion Several pathway databases have recently become available, with curated information on biochemical reactions. Most databases are incomplete, with missing information on various biochemical reactions. A study undertaken recently by Kettner [29] illustrates the poor quality in well-known databases such as BRENDA even for pathways such as glycolysis. They conclude from their study that the difficulty in curation and enhancing quality in databases was largely due to both incomplete descriptions of material and methods in the papers and difficulties considering method-dependent results, since extraction of kinetic data from literature is necessarily carried out manually. They suggest that it might be useful to establish a deposition system to which authors can submit their data to ensure maximal accuracy and accessibility and to replace possibly the traditional retrospective process of manual data extraction. The situation is a bit better when it comes to reaction annotation without the quantitative data, for obvious reasons. At this point of time, there does not appear to be an automated method for detecting and correcting errors. Although highly desirable, availability of a comprehensive accurate database, hence may be quite a while away. We have used KEGG as our primary source of data, since it was the largest available curated database, particularly for mycobacteria. A major problem we observed was in describing the reversibility of the reactions, which we have corrected manually to the extent possible. Although there are omissions and errors in detail in the database, on the whole, most of the reactions and proteins are annotated correctly, from the manual checks we performed. Hence, the KEGG serves as a good starting point for systems biology studies. KEGG has also been extensively used by several groups previously for systems analyses [16] , [17] , [30] , [31] and even genome-scale metabolic reconstructions [32] . Felix and Valiente have performed an exhaustive validation of a substantial portion of the KEGG LIGAND database [33] , concluding that over 90% of the reactions in the KEGG are consistent. The reaction networks of M. tuberculosis H37Rv, M. leprae and E. coli K-12 MG1655 were reconstructed from a dataset primarily obtained from the KEGG LIGAND database [34] . We used reaction files containing compound IDs, so that there was no discrepancy in the reactions, based on the usage of synonymous compound names. Details of the total numbers of reactions, metabolites and enzymes comprising the networks of these organisms have been summarised in Table 1 . The total size of the networks in the three organisms is roughly of the same order, making their comparison quite meaningful. The networks essentially capture the core metabolic processes, such as the metabolism of carbohydrates, nucleic acids, amino acids and lipids, in the three organisms. Sequence analyses of the genomes have indicated that they share similarity among a number of individual genes/proteins, leading to several similar reactions in the three networks. However, the precise connections with which the reactions are associated have not been analysed previously in a comparative context. Graphs presented here are amenable to such comparative analysis, providing a handle to understand the similarities and the differences between a given pair of organisms from a systems perspective. Analysis of the various network properties, clustering patterns, shortest and alternate paths, that aid in this process are elaborated below. 10.1371/journal.pone.0000881.t001 Table 1 Network properties of the reconstructed metabolic networks of M. tuberculosis , M. leprae and E. coli and the corresponding random networks. Property \n M. tuberculosis \n \n M. leprae \n \n E. coli \n \n DETAILS OF THE RECONSTRUCTED NETWORK \n Total no. of reactions (nodes) 1906 1325 2080 Total no. of edges 14,100 8,508 20,316 Reversible reactions 209 152 274 Irreversible reactions 1488 1021 1532 No. of metabolites participating in the reactions 1649 1139 1633 Total no. of proteins catalyzing the reactions 1097 469 1062 No. of currency metabolites eliminated 102 84 107 \n NETWORK PARAMETERS \n Percentage of nodes belonging to the largest cluster 73.40 76.15 73.85 Percentage of ‘orphan’ nodes 17.00 16.30 16.39 Highest degree of connections 72 50 96 Average path length 5.58 5.48 4.94 Clustering co-efficient 0.01669 0.01600 0.01614 Degree exponent of the power law degree distribution 1.3952 1.4423 1.4093 \n PROPERTIES OF RANDOM NETWORKS \n * \n Percentage of nodes belonging to the largest cluster 99.94 99.85 99.99 Percentage of ‘orphan’ nodes 0.0588 0.1487 0.0058 Highest degree of connections 22 21 27 Average path length 4.00±0.02 4.07±0.03 3.61±0.01 Clustering co-efficient (3.78±0.56) * 10 −3 \n (4.84±0.89) * 10 −3 \n (4.67±0.45) * 10 −3 \n Characteristic Scale 7.40±0.0122 6.42±0.0137 9.76±0.0134 * Random networks have been generated using the Erdős-Rényi model, with the same total number of vertices and edge probability as the corresponding reaction networks. Analysis of network parameters Analysis of various network parameters such as the degree distribution, clustering coefficient, average path length and size of the largest cluster of the reaction networks of all three organisms revealed them to have a similar overall topology with comparable graph properties ( Table 1 ). The degree distributions of the nodes in the reaction networks of all three organisms exhibit a power law behaviour as shown in Figures 1A−C . Hence, the reaction networks of these organisms are scale-free in nature, consisting of a few ‘hubs’ that are highly connected and hold together numerous nodes having a small degree. The log-log plots of the degree distributions ( Figures 1D–F ) for the three networks also are characteristic of a power law behaviour, with the degree exponent γ∼1.4. This result differs from the results of Wagner and Fell [18] who did not obtain a clear power-law degree distribution for a reaction based graph of the metabolic network of E. coli. This was perhaps due to the reduced dataset used in their study. The clustering coefficients of the reaction networks of the three organisms are of the same order of 10 −2 , indicating that these networks are quite sparse and have approximately the same density of connections between nodes despite a difference in their overall sizes. Further, it implies that the observed connections may have evolved to suit a particular metabolic requirement and are far from random. The average path length of the three networks are also comparable, with the networks of M. tuberculosis and M. leprae having average path lengths of 5.58 and 5.48 respectively, while the reaction network of E. coli has a slightly smaller average path length of 4.94. The small values of the average path length in these organisms imply a small-world character, where the distance between any two reactions in the network is smaller than what is expected from traditional biochemical pathways, as will be shown later. 10.1371/journal.pone.0000881.g001 Figure 1 A–C: Plots of the degree distributions of nodes in the reaction networks of M. tuberculosis , M. leprae and E. coli. \n D–F: Log-log plots of the degree distribution function P(k) versus the degree k . P(k) defines the probability of a given node making exactly k connections in the network. The fit to the curve shows a power law behaviour and has an exponent of ∼1.4 for all three networks. G–I: Representative degree distributions of nodes in random networks generated with the same total number of nodes and edges as the reaction networks of M. tuberculosis , M. leprae and E. coli. \n Clustering patterns The largest clusters (giant components) in the networks of M. tuberculosis , M. leprae and E. coli were obtained by the depth-first-search method and comprised 73–76% of the total number of biochemical reactions in their metabolome; while 8–10% of the total reactions form other clusters in their respective networks. It is therefore interesting to note that the size of the giant component in the reaction networks of these organisms is conserved and is unaffected by the metabolic streamlining of the leprae bacillus. It is also unaffected by differences that exist between the metabolism of E. coli and the mycobacterial organisms. Figures 2A–C are representations of the clusters obtained in the reaction networks of the three organisms. Graphviz [35] was used for the generation of the cluster diagrams. Clustering analysis of the three networks by the depth-first-search method also revealed several ‘orphan’ reactions that were completely unconnected in their respective networks. Some of these were reactions that involved only currency metabolites and hence were unconnected with the rest of the network due to our elimination of interactions mediated by the currency metabolites. Other orphans represented reactions which were either inaccurately curated in the database and did not occur in the metabolism of these organisms or whose links with the rest of the metabolome have not been determined to date. A few examples of orphan reactions are shown below: \n R00004 (EC 3.6.1.1.) Pyrophosphate+H2O→2 Orthophosphate \n \n R00281 (EC 1.6.99.3) Acceptor+NADH→Reduced-acceptor+NAD+ \n As an extreme example, we even have: \n R02501 (EC 1.14.14.1) Testosterone+H++Oxygen+NADPH→19-Hydroxytestosterone+NADP + \n 10.1371/journal.pone.0000881.g002 Figure 2 Clusters in the reaction networks of M. tuberculosis (A), M. leprae (B) and E. coli (C). The giant component in the networks comprises 73–76% of the total number of nodes, while 8–10% of the total nodes form other clusters in the networks. The orphan nodes have been eliminated for better visualisation. Such information on orphan reactions can be used to improve the curation of the database and the annotation of the genome itself. We compared the network properties of the three reaction networks with those of random graphs comprising the same total number of vertices (V) and connecting edges (E) as the corresponding reaction network, using the Erdős-Rényi model (see Table 1 ). It was observed that for the given probability of connections, the nodes of the random network formed a single connected cluster comprising almost 100% of the nodes. Further, the degree distributions of the nodes of the random networks having the same total number of (V, E) as that of the M. tuberculosis , M. leprae and E. coli reaction networks followed a Poisson distribution, with an average scale of 7.40, 6.42 and 9.76 connections per node respectively. Moreover, the highest degree of connections in these networks was significantly lower with values of 22, 21 and 27 as compared to the highest degree in the reaction networks of M. tuberculosis (72), M. leprae (50) and E. coli (96) respectively. Analyses of sub-clusters in the giant component Metabolic networks are highly integrated and complex in nature. Hence, a rational reduction of these networks to their basic structural and functional units is essential to gain a deeper understanding of their design principles and functioning. Several studies have been carried out to detect modularity in metabolic networks [21] , [22] , [36] , [37] . Guimera & Nunes Amaral [38] have shown that metabolites in the cell group together to form functional modules with typically 80% of the nodes in the network only connected to other nodes within their respective modules. To determine if this modularity of metabolic networks is also reflected at the level of the constituent biochemical reactions, we carried out analyses to detect sub-clusters of reactions in the giant component by graph spectral analysis. As mentioned in the methods section, the 2evc plots of the Laplacian matrix of the graph provide the sub-cluster information. The 2evc plots constructed for the reaction networks of M. tuberculosis and M. leprae comprised numerous plateau regions which represented sub-clusters of reactions in the giant component of these networks ( Figures 3A,B ). A few examples of the sub-clusters obtained in the giant component of M. tuberculosis have been shown in Table 2 . 10.1371/journal.pone.0000881.g003 Figure 3 2evc plots for the giant components of the reaction networks of M. tuberculosis (A), M. leprae (B) and E. coli (C). Plateaus represent sub-clusters of reactions. The giant component in the reaction network of E. coli does not resolve into sub-clusters as indicated by the single plateau in plot C. Arrows indicate plateaus representing the sub-cluster of mycolic acid pathway reactions in the mycobacterial networks. 10.1371/journal.pone.0000881.t002 Table 2 Examples of sub-clusters in the giant component of the metabolic network of M. tuberculosis . 2evc Node no. RID * \n Pathway-ID Pathway CLUSTER 1 (SIZE: 12) 0.00902 552 R01700 map00020 Citrate cycle (TCA cycle) 0.00902 1129 R04231 map00770 Pantothenate and CoA biosynthesis 0.00902 1130 R04233 rn00770 Pantothenate and CoA biosynthesis 0.00902 181 R00439B rn00230 Purine metabolism 0.00902 168 R00429 rn00230 Purine metabolism 0.00902 278 R00722B rn00230 Purine metabolism 0.00902 173 R00434 rn00230 Purine metabolism 0.00902 118 R00332B rn00230 Purine metabolism 0.00902 164 R00416B map00530 Aminosugars metabolism 0.00902 331 R00896 rn00272 Cysteine metabolism 0.00902 200 R00480 rn00260 Gly, Ser and Thr metabolism 0.00902 443 R01214F rn00280 Val, Leu, Ile degradation CLUSTER 2 (SIZE: 11) 0.00921 483 R01343F map00220 Urea cycle and metabolism of amino groups 0.00921 453 R01248F rn00220 Urea cycle and metabolism of amino groups 0.00921 455 R01251F rn00220 Urea cycle and metabolism of amino groups 0.00921 999 R03646B rn00330 Arg and Pro metabolism 0.00921 457 R01253 rn00330 Arg and Pro metabolism 0.00921 834 R02788 rn00300 Lys biosynthesis 0.00921 135 R00376F rn00230 Purine metabolism 0.00921 612 R01857B rn00230 Purine metabolism 0.00921 701 R02235F rn00790 Folate biosynthesis 0.00921 703 R02236F rn00670 One carbon pool by folate 0.00921 1542 R05688 rn00100 Biosynthesis of steroids CLUSTER 3 (SIZE: 10) 0.00675 1204 R04536B rn00061 Fatty acid biosynthesis 0.00675 1206 R04543B rn00061 Fatty acid biosynthesis 0.00675 1343 R04958B rn00061 Fatty acid biosynthesis 0.00675 1346 R04961B rn00061 Fatty acid biosynthesis 0.00675 1349 R04964B rn00061 Fatty acid biosynthesis 0.00675 1351 R04966B rn00061 Fatty acid biosynthesis 0.00675 1340 R04955B rn00061 Fatty acid biosynthesis 0.00675 1180 R04429B rn00061 Fatty acid biosynthesis 0.00675 1202 R04534B rn00061 Fatty acid biosynthesis 0.00675 1338 R04953B rn00061 Fatty acid biosynthesis CLUSTER 4 (SIZE:24) −0.07038 1810 MAP123 fasII Mycolic acid pathway −0.07038 1812 MAP125 fasII Mycolic acid pathway −0.07038 1813 MAP126 fasII Mycolic acid pathway −0.07038 1814 MAP127 fasII Mycolic acid pathway −0.07038 1816 MAP129 fasII Mycolic acid pathway −0.07038 1794 MAP107 fasII Mycolic acid pathway −0.07038 1818 MAP131 fasII Mycolic acid pathway −0.07038 1819 MAP132 fasII Mycolic acid pathway −0.07038 1820 MAP133 fasII Mycolic acid pathway −0.07038 1822 MAP135 fasII Mycolic acid pathway −0.07038 1824 MAP137 fasII Mycolic acid pathway . . . −0.07038 1825 MAP138 fasII Mycolic acid pathway −0.07038 1826 MAP139 fasII Mycolic acid pathway −0.07038 1795 MAP108 fasII Mycolic acid pathway * In the RID, F indicates the reaction proceeding in the forward direction, and B indicates the reaction proceeding in the backward direction. It is observed that reactions belonging to fatty acid biosynthesis and the FAS-II cycle of the mycolic acid pathway in M. tuberculosis form distinct, tightly connected sub-clusters. This may be due to the iterative nature of these cycles where the metabolite passes through several cycles of the same reactions consecutively in order to obtain a product of the requisite carbon chain length. Hence, the reactions within these pathways are more tightly connected with each other than to the other reactions in the metabolome. The mycolic acid pathway [39] , [40] , [41] is a critical pathway in M. tuberculosis and is important for its survival and pathogenicity. The other sub-clusters obtained are less specific and contain reactions belonging to different pathways identified in the KEGG. For example, cluster 1 in Table 2 comprises reactions involved in purine metabolism as well in pantothenate and coenzyme-A biosynthesis. Further, reactions occurring in a particular pathway are not contained in a single cluster but different reactions from the same pathway are observed in different sub-clusters. For example, reactions involved in purine metabolism occur in both clusters 1 and 2. The reaction network of M. leprae comprised sub-clusters that were fewer in number and smaller in size than those obtained in M. tuberculosis. This is because the metabolome of M. leprae comprises fewer reactions than that of M. tuberculosis. Analysis of these sub-clusters revealed them to be similar in nature to those of M. tuberculosis , with reactions of the FAS-II cycle of the mycolic acid pathway forming closely related sub-clusters while the other sub-clusters were less specific in nature. Therefore, the overall topological structure and nature of the giant component is conserved between the two mycobacteria, indicating that the large scale downsizing of the M. leprae metabolome has not significantly altered the global structure of its core reaction network. Interestingly, unlike the mycobacterial networks, the giant component in the reaction network of E. coli did not resolve into constituent sub-clusters ( Figure 3C ), implying that reactions occurring in the metabolome of E. coli are more strongly interconnected across various biochemical pathways. Thus, differences can be seen to exist in the finer grouping of biochemical reactions within the giant components of the reaction networks of the mycobacteria and E. coli . Lastly, from the analyses of sub-clusters in the giant components of the three reaction networks, it can be seen that functional modules are less well defined at the level of biochemical reactions, with the reactions forming a single, large, connected cluster ensuring free flow of metabolites between them. However, it is important to note that similar to other studies, our model also is limited by the assumption of constitutive expression of all enzymes catalysing these reactions, i.e. the temporal expression of the enzymes on account of regulation is ignored. This may affect the clustering of biochemical reactions. Identification of hubs in the reaction networks Metabolic networks are scale-free networks characterised by the presence of hubs that are highly connected nodes serving to hold together numerous smaller nodes having a lower degree [13] . With a large number of links, hubs in metabolic networks integrate all substrates in the cell into a single, complex web of biochemical interactions. Hubs are essential to the integrity and robustness of the network against random attacks [42] , [43] . They are also responsible for the small-world behaviour of networks as any two nodes in the network can be reached by a relatively short distance by traversing a hub [43] . Furthermore, in biological networks, the hubs are thought to be functionally important and phylogenetically oldest [18] , [20] , [21] , [44] . To identify highly connected reactions essential and central to the metabolism of the three organisms under study, we elucidated the hubs in their reaction networks by graph spectral analysis as well as by degree analysis. As mentioned in the methods section, the largest vector component of the highest eigenvalue of the Laplacian matrix of the graph corresponds to the node with high degree as well as low eccentricity. Two parameters, degree and eccentricity, are involved in the identification of graph spectral (GS) hubs. In a graph representing a scale-free network, the highest vector component would therefore correspond to a hub with high degree and also closest to the geometric centre of the network. Alternatively, hubs can be ranked based on their connectivity alone (degree hubs). Thus, hubs obtained from graph spectral analysis may differ from the degree hubs. However, a comparison of the ranks of GS hubs and degree hubs did not show any significant difference (Supplementary Table S1 ). This is perhaps due to the topology of the network. However, when the reactions corresponding to the hubs were examined in detail, we find discrimination between the two sets of hubs. It was observed that the top 50 degree hubs in the reaction networks of the three organisms comprised reactions involving the metabolite L-glutamate as well as reactions involving pyruvate. However, the top 50 GS hubs of M. tuberculosis and M. leprae exclusively comprised reactions involving L-glutamate while the top GS hubs in E. coli only consisted of reactions involving pyruvate (see Supplementary Table S2 ). The difference in the degree and GS hubs suggests that the most highly connected reactions are not necessarily the most central reactions in the metabolome of the organism (by centrality, we mean that node which has the least eccentricity; eccentricity is the distance between a node and the node farthest from it). Furthermore, reactions (and metabolites) forming the centres of reaction networks are not common across all organisms but are specific to the metabolism of each organism. Previous studies by Ma & Zeng [17] on a substrate based metabolic network of E. coli have shown pyruvate to be central to the metabolism of this organism; our results corroborate their observation. By constructing a reaction based graph of the metabolic network, we have elucidated specific reactions involving pyruvate that form the centre of the E. coli metabolome. Moreover, our method of identifying hubs central to the network by graph spectral analysis is simpler and more advantageous as it uses a single numerical computation that takes into account both the degree and eccentricity of the hub in the overall network. The quality of the initial database used to reconstruct the network would affect the results obtained in this analysis. An incomplete database may lead to several false positives in the identification of hubs in the network. To determine the reliability of the identified hubs, we randomly knocked out half the nodes in the original reaction networks of the three organisms to generate highly reduced networks that simulate an incompletely curated database. This exercise was repeated 30 times for each organism and it was observed that reactions involving L-glutamate comprised a majority of the top 20 GS hubs in the reduced M. tuberculosis and M. leprae networks 83% and 67% of the time respectively. Similarly reactions involving pyruvate formed the majority of the top 20 GS hubs in the reduced E. coli network 67% of the time. Hence, we believe that the results of our analysis presented here are fairly reliable. Identification of mycobacterial hubs absent in the human metabolome Hubs are essential to the integrity of the network. They make the network vulnerable to targeted attacks on them, because once the highly connected hubs are attacked, the network starts disintegrating [42] , [43] . This property makes the hubs ideal drug targets; by targeting a drug against a suitable hub or group of hubs, it is possible to break down the cellular network of an organism completely, which would result in the death of the organism. Further, targeting hubs that are unique to the pathogen and absent in humans would minimise side-effects of the drug in the host. We analysed the GS hubs in M. tuberculosis and M. leprae to identify highly connected reactions that are central to the metabolism of these organisms but absent in humans (as detailed in the Methods section). The enzymes catalysing these reactions can be explored further as potential drug targets in the field of knowledge-based, rational drug design. The top fifteen ‘unique’ hubs obtained in M. tuberculosis ( Table 3 ) mainly comprise reactions involved in nitrogen metabolism and in the biosynthesis of essential amino acids that are not synthesised in humans. It is particularly interesting to note that the first reaction of the FAS-I cycle of the mycolic acid pathway unique to mycobacteria ranks as the 38 th unique hub in mycobacterium. Thus, the enzyme acyl carrier protein-fatty acid synthase (ACP-FAS) involved in this reaction can be explored as a potential target for drugs against mycobacteria. Independent studies by Sassetti and co-workers [45] and Raman and co-workers [41] have also identified the FAS enzyme as one of the putative anti-tubercular drug targets. Sassetti and co-workers performed a high throughput Transposon Site Hybridisation Mutagenesis study to identify essential genes in M. tuberculosis while Raman and co-workers performed a flux balance analysis of the mycolic acid pathway in M. tuberculosis , followed by in silico gene deletions to identify essential genes/putative drug targets. The list of the top hubs unique to M. leprae is similar and has not been tabulated separately. 10.1371/journal.pone.0000881.t003 Table 3 Top fifteen hubs in M. tuberculosis that are absent in the metabolome of humans. Rank as a unique hub Rank as a GS hub Node no. Hub RID E.C. No. of catalyzing enzyme Name of Enzyme Pathway(s) for which the enzyme is required 1 3 368 R00986 4.1.3.27 Anthranilate synthase Phenylalanine, tyrosine and tryptophan biosynthesis 2 4 482 R01339F 2.6.1.- Transferases, transaminases Nitrogen metabolism 3 8 1197 R04475B 2.6.1.17 Succinyldiaminopimelate aminotransferase Lysine biosynthesis 4 12 197 R00457B 2.6.1.36 Transferases, Transaminases Lysine biosynthesis 5 14 717 R02283B 2.6.1.11 Acetylornithine transaminase Urea cycle and metabolism of amino groups 6 28 160 R00411 3.5.1.- Hydrolases acting on carbon-nitrogen bonds, other than peptide bonds in linear amides Arginine and proline metabolism 7 35 714 R02282f 2.3.1.35 Glutamate N-acetyltransferase Urea cycle and metabolism of amino groups 8 48 19 R00114 1.4.1.13 L-glutamate synthase Glutamate metabolism, Nitrogen metabolism 9 49 562 R01716 6.3.5.8 Aminodeoxychorismate synthase Folate biosynthesis 10 54 87 R00260F 5.1.1.3 Glutamate racemase Glutamate metabolism, D-Glutamine and D-glutamate metabolism 11 55 1440 R05225 6.3.5.10 Cobyric acid synthase Porphyrin and chlorophyll metabolism 12 64 1439 R05224 6.3.5.9 Hydrogenobyrinic acid a,c-diamide synthase (glutamine-hydrolysing) Porphyrin and chlorophyll metabolism 13 83 1707 R06860 2.5.1.64 2-succinyl-6-hydroxy-2,4-cyclohexadiene Ubiquinone biosynthesis 14 85 1458 R05320 1.14.11.17 Taurine dioxygenase Taurine and hypotaurine metabolism 15 86 442 R01197 1.2.7.3 2-oxoglutarate synthase Citrate cycle . . 38 * \n 172 1692 MAP005 - Acyl carrier protein-fatty acid synthase I (ACP-FAS) Mycolic acid pathway * ACP-FAS enzyme involved in the fifth reaction of the mycolic acid pathway ranks as the 38 th unique hub in M. tuberculosis. \n Analysis of shortest paths Metabolic networks are small-world networks characterised by a small average path length between the nodes. Thus, in the metabolic network of most organisms, on an average, any metabolite can be converted to any other metabolite by rather small number of biochemical reactions [13] . Besides analysing the reaction networks of the three organisms under study to determine their average path lengths as described in a previous section, we have also elucidated the actual steps in the shortest path between any two nodes in their networks. We compared the shortest paths obtained in these organisms across six different metabolic pathways: (a) glycolysis, (b) citric acid cycle, (c) pentose phosphate pathway, (d) phenylalanine, tyrosine and tryptophan biosynthesis, (e) valine, leucine and isoleucine biosynthesis and (f) folate biosynthesis and observed that a majority of the paths obtained between reactions from these pathways were the same in all three organisms. Hence, the streamlining of the M. leprae metabolome has not significantly altered the shortest routes that exist between reactions in its core metabolism. Further, the shortest paths between reactions of the above pathways are also unaffected by the significant differences in the metabolism of mycobacteria and E. coli. \n As predicted from the low values of the average path length for these reaction networks, the paths obtained in this analysis are shorter than the traditionally annotated biochemical pathways. For example, the conversion of pyruvate to oxalosuccinate by the experimentally determined citric acid cycle requires a minimum of four steps (pyruvate→oxaloacetate→citrate→isocitrate→oxalosuccinate) while the shortest path obtained between them by our analysis comprised only three consecutive steps: Steps in terms of RIDs: \n R00344F→R00355B→R00268b \n \n R00344 (EC 6.4.1.1) ATP+Pyruvate+HCO3−→ADP+Orthophosphate+Oxaloacetate \n \n R00355 (EC 2.6.1.1) Oxaloacetate+L-Glutamate→L-Aspartate+2-Oxoglutarate \n \n R00268 (EC 1.1.1.42) 2-Oxoglutarate+CO2→Oxalosuccinate \n Hence, it can be inferred that the traditionally annotated pathways are not the shortest possible routes for the conversion of one metabolite into another. They are therefore a result of several constraints imposed within the cell, such as intracellular compartmentalisation and thermodynamic constraints. Whether the shortest pathways predicted here are really feasible pathways or not should be evaluated in the context of the constraints in the cell. Interestingly, our analysis also revealed shortest paths in which one or more reactions in the path produce metabolites that are utilised by reactions other than those that would be involved in the commonly accepted notion of the steps of a given pathway. Though such paths do not represent an ideal shortest path comprising consecutive steps between a pair of reactions (as in the above example), it provides important information regarding the interaction of reactions from different biochemical pathways. For example, the set of reactions obtained as the shortest path between reaction R00959 (the 1 st reaction) and R00658 (the 8 th reaction) in the glycolytic pathway are indicated in Table 4 . 10.1371/journal.pone.0000881.t004 Table 4 Shortest path between steps in glycolytic pathway. Steps in terms of RIDs: R00959F→R02740F→R01830F→R01826F→R00658F R00959 (EC 5.4.2.2) D-Glucose-1-phosphate→alpha-D-Glucose-6-phosphate R02740 (EC 5.3.1.9) alpha-D-Glucose-6-phosphate→beta-D-Fructose-6-phosphate R01830 (EC 2.2.1.1) beta-D-Fructose-6-phosphate+(2R)-2-Hydroxy-3-(phosphonooxy)-propanal→D-Erythrose-4-phosphate+D-Xylulose-5-phosphate R01826 (EC 2.5.1.54) Phosphoenolpyruvate+D-Erythrose-4-phosphate+H2O→2-Dehydro-3-deoxy-D-arabino-heptonate-7-phosphate+Orthophosphate R00658 (EC 4.2.1.11) 2-Phospho-D-glycerate→Phosphoenolpyruvate+H2O The path comprises reactions from different metabolic pathways – reaction R02740 occurs in the glycolysis and pentose phosphate pathways, reaction R01830 occurs in the pentose phosphate pathway and reaction R01826 occurs in the phenylalanine, tyrosine and tryptophan biosynthetic pathway. Further, in reaction R01826 phosphoenolpyruvate produced in the glycolytic pathway reacts with D-erythrose-4-phosphate produced in the pentose phosphate pathway to yield 2-dehydro-3-deoxy-D-arabino-heptonate-7-phosphate, a precursor for tryptophan biosynthesis. It is clear that the flux through each of these reactions will influence the direction of metabolites into specific biochemical pathways. For example, the phosphoenolpyruvate produced in reaction R00658 can either be converted into pyruvate by the glycolytic route or be diverted into tryptophan biosynthesis. The results of this analysis are also likely to be useful for metabolic engineering. A systematic study of shortest paths between all reaction pairs in the metabolome of an organism can reveal paths that are shorter than traditionally annotated biochemical pathways. Understanding of such paths can have implications in the production of industrially important secondary metabolites and for the manipulation of particular reaction fluxes to direct intermediates into specific metabolic pathways for obtaining larger quantities of the desired product. Analysis of alternate paths We have extensively analysed the number of alternate paths, that exist between reaction pairs in the metabolomes of the three organisms (for this, all 312 reactions common to the three organisms were considered). It was observed that E. coli contained the most number of alternate paths between a given pair of reactions, followed by M. tuberculosis , while the leprae bacillus had the least number of alternate paths of the three organisms. This is due not only to the differing sizes of the networks of the three organisms, but also to the difference in the degree distributions. Though this result is not surprising in itself, it is interesting to note that the reductive evolution of the M. leprae metabolome has led to a loss of multiple paths between reactions pairs in the core metabolic pathways, while keeping the shortest possible route between them intact. This is possibly because only the core of the metabolome has been conserved; the alternate routes (in M. tuberculosis ) add to the redundancy. Similarly, mycobacteria and E. coli have evolved to have differing number of alternate paths between reactions while keeping the shortest paths between the reactions conserved. Conclusions The results presented in this work show that the reaction networks of M. tuberculosis , M. leprae and E. coli are scale-free, small-world networks that differ significantly from random networks. The networks of the three organisms have similar network properties despite a difference in the overall sizes of their metabolomes. Graph spectral theory serves as a tool useful for analysing the topological structure and organisation of large complex networks. This technique yields information about the sub-clustering of nodes in the network and identifies the cluster centres by a single numeric computation. Analysis of sub-clusters of the mycobacterial reaction networks detected by this method suggests that modularity of metabolic networks is possibly less well-defined at the level of biochemical reactions; clusters have been discerned well from metabolite networks [21] . It was observed that the top 50 GS hubs of M. tuberculosis and M. leprae exclusively comprised reactions involving L-glutamate while the top GS hubs in E. coli only consisted of reactions involving pyruvate. We showed that the most highly connected biochemical reactions in the reaction network of an organism are not necessarily the reactions most central to the metabolism of that organism. Moreover, the reactions and metabolites forming the centre of the metabolic networks are not common across all organisms, but are specific to the metabolism of each organism. A systematic comparison of the topological properties of the mycobacterial metabolic networks reveals that massive gene decay in M. leprae does not significantly affect the global structure of its metabolic network. We showed that the most highly connected reactions central to the metabolism of both organisms as well as the finer grouping of reactions within the giant component of their networks are essentially conserved and differ from that of E. coli . Additionally, we have determined that metabolic streamlining in the leprae bacillus has led to the preservation of the shortest paths between reactions in its core metabolome while reducing the total number of alternate paths that exist between them. The results obtained in this work are also useful likely to find applications in rational drug design and metabolic engineering."
} | 12,103 |
29670508 | PMC5893757 | pmc | 7,032 | {
"abstract": "Artificial neural networks can harness stochasticity in multiple ways to enable a vast class of computationally powerful models. Boltzmann machines and other stochastic neural networks have been shown to outperform their deterministic counterparts by allowing dynamical systems to escape local energy minima. Electronic implementation of such stochastic networks is currently limited to addition of algorithmic noise to digital machines which is inherently inefficient; albeit recent efforts to harness physical noise in devices for stochasticity have shown promise. To succeed in fabricating electronic neuromorphic networks we need experimental evidence of devices with measurable and controllable stochasticity which is complemented with the development of reliable statistical models of such observed stochasticity. Current research literature has sparse evidence of the former and a complete lack of the latter. This motivates the current article where we demonstrate a stochastic neuron using an insulator-metal-transition (IMT) device, based on electrically induced phase-transition, in series with a tunable resistance. We show that an IMT neuron has dynamics similar to a piecewise linear FitzHugh-Nagumo (FHN) neuron and incorporates all characteristics of a spiking neuron in the device phenomena. We experimentally demonstrate spontaneous stochastic spiking along with electrically controllable firing probabilities using Vanadium Dioxide (VO 2 ) based IMT neurons which show a sigmoid-like transfer function. The stochastic spiking is explained by two noise sources - thermal noise and threshold fluctuations, which act as precursors of bifurcation. As such, the IMT neuron is modeled as an Ornstein-Uhlenbeck (OU) process with a fluctuating boundary resulting in transfer curves that closely match experiments. The moments of interspike intervals are calculated analytically by extending the first-passage-time (FPT) models for Ornstein-Uhlenbeck (OU) process to include a fluctuating boundary. We find that the coefficient of variation of interspike intervals depend on the relative proportion of thermal and threshold noise, where threshold noise is the dominant source in the current experimental demonstrations. As one of the first comprehensive studies of a stochastic neuron hardware and its statistical properties, this article would enable efficient implementation of a large class of neuro-mimetic networks and algorithms.",
"introduction": "1. Introduction A growing need for efficient machine-learning in autonomous systems coupled with an interest in solving computationally hard optimization problems has led to active research in stochastic models of computing. Optimization techniques (Haykin, 2009 ) including Stochastic Sampling Machines (SSM), Simulated Annealing, Stochastic Gradients etc., are examples of such models. All these algorithms are currently implemented using digital hardware which first creates a mathematically accurate platform for computing, and later adds digital noise at the algorithm level. Hence, it is enticing to construct hardware primitives that can harness the already existing physical sources of noise to create a stochastic computing platform. The principal challenge with such efforts is the lack of stable or reproducible distributions, or functions of distributions, of physical noise. One basic stochastic unit which enables a systematic construction of stochastic hardware has long been known—the stochastic neuron (Gerstner and Kistler, 2002 )—which is also believed to be the unit of computation in the human brain. Moreover, recent studies (Buesing et al., 2011 ) have demonstrated practical applications like sampling using networks of such stochastic spiking neurons. There have been some attempts for building neuron hardware (Indiveri et al., 2006 ; Pickett et al., 2013 ; Mehonic and Kenyon, 2016 ; Sengupta et al., 2016 ; Tuma et al., 2016 ), but building a neuron with self-sustained spikes, or oscillations, which are stochastic in nature and where the probability of firing is controllable using a signal has been challenging. Here, we demonstrate and analytically study a true stochastic neuron (Jerry et al., 2017a ) which is fabricated using oscillators (Shukla et al., 2014a , b ; Parihar et al., 2015 ) based on insulator-metal transition (IMT) materials, e.g., Vanadium Dioxide (VO 2 ), wherein the inherent physical noise in the dynamics is used to implement stochasticity. The firing probability, and not just the deterministic frequency of oscillations or spikes, is controllable using an electrical signal. We also show that such an IMT neuron has similar dynamics as a piecewise linear FitzHugh-Nagumo (FHN) neuron with thermal noise along with threshold fluctuations as precursors of bifurcation resulting in a sigmoid-like transfer function for the neural firing rates. By analyzing the variance of interspike interval, we determine that for the range of thermal noise present in our experimental demonstrations, threshold fluctuations are responsible for most of the stochasticity compared to thermal noise.",
"discussion": "4. Discussion In this paper, we demonstrate and analyse an IMT based stochastic neuron hardware which relies on both threshold fluctuations and thermal noise as precursors to bifurcation. The IMT neuron emulates the functionality of theoretical neuron models completely by incorporating all neuron characteristics into device phenomena. Unlike other similar efforts, it does not need peripheral circuits alongside the core device circuit (an IMT device and a transistor) to emulate any sub-component of the spiking neuron model like thresholding, reset etc. Moreover, the neuron construction not only utilizes inherent physical noise sources for stochasticity, but also enables control of firing probability using an analog electrical signal—the gate voltage of series transistor. This is different from previous works which control only the deterministic aspect of firing rate like the charging rate. A comparison of spiking neuron hardware characteristics in different works is shown in Table 1 . Table 1 Comparison of this work (experimental details from Jerry et al., 2017a ) with other spiking neuron hardware works based on different characteristics of spiking neurons. Tuma et al., 2016 Pickett et al., 2013 Sengupta et al., 2016 Indiveri et al., 2006 This work (VO 2 ) Neuron type Integrate & Fire Hodgkin Huxley Integrate & Fire Integrate & Fire Piecewise Linear FHN Material/Platform Chalcogenide Mott insulator NbO 2 MTJ 0.35 μm CMOS Vanadium Dioxide (VO 2 ) Material phenomenon Phase Change IMT Spin transfer torque (STT) – IMT Spontaneous spiking using only device No Yes No – Yes Peripherals needed for spiking Yes, for spike generation and reset No Yes, for spike generation and reset – No Integration mechanism (I&F) Heat accumulation – Magnetization accumulation Capacitor charging Capacitor charging Threshold mechanism (I&F) External reset by measuring conductance Spontaneous IMT External reset by detecting magnet flip Reset using comparator Spontaneous IMT Stochastic Yes – Yes No Yes Kind of stochasticity (I&F) Reset potential – Differential – Threshold and differential Source of stochasticity / noise Melt-quench process – Thermal noise – IMT threshold fluctuations & Thermal noise Control of stochastic firing rate Only integration rate – Only integration rate Only integration rate Yes Status of experiments Constant stochasticity, variable integration rate Deterministic spiking None Deterministic spiking Sigmoidal variation of stochastic firing rates Peak current 750–800 μA – 200 μA Power or Energy/spike 120 μW – 900 pJ / spike 196 pJ / spike Voltage 5.5 V 1.75 V – 3.3 V 0.7 V Maximum firing rates 35–40 KHz 30 KHz – 200 Hz 30 KHz We also show that the neuron dynamics follow a linear “carricature” of the FitzHugh-Nagumo model with intrinsic stochasticity. The analytical models developed in this paper can also faithfully reproduce the experimentally observed transfer curve which is a stochastic property. Such analytical verification of stochastic neuron experiments is one of the first in this work. It is an important result as it indicates reproducibility of stochastic characteristics and helps in creating the pathway toward perfecting these devices. With a growing concensus that stochasticity will play a key role in solving hard computing tasks, we need efficient ways for controlled amplification and conversion of physical noise into a readable and computable form. In this regard, the IMT based neuron represents a promising solution for a stochastic computational element. Such stochastic neurons have the potential to realize bio-mimetic computational kernels that can be employed to solve a large class of optimization and machine-learning problems."
} | 2,199 |
30966150 | PMC6415098 | pmc | 7,034 | {
"abstract": "Self-healing materials have attracted much attention because that they possess the ability to increase the lifetime of materials and reduce the total cost of systems during the process of long-term use; incorporation of functional material enlarges their applications. Graphene, as a promising additive, has received great attention due to its large specific surface area, ultrahigh conductivity, strong antioxidant characteristics, thermal stability, high thermal conductivity, and good mechanical properties. In this brief review, graphene-containing polymer composites with self-healing properties are summarized including their preparations, self-healing conditions, properties, and applications. In addition, future perspectives of graphene/polymer composites are briefly discussed.",
"conclusion": "6. Conclusions In this review, we have summarized the recent progress of graphene-containing polymer composites with self-healing capability. The preparation methods, self-healing conditions, and properties and applications of the graphene/polymer composites have been briefly discussed. Finally, the further perspectives of the composites have been proposed. Intelligent materials and self-healing materials, specifically for graphene-containing composites, are still in the initial stage. There will be a great research space in the field. The progress will guide further development of the self-healing graphene/polymer composites.",
"introduction": "1. Introduction Regenerative abilities allow creatures to repair damaged functions to prolong their life span. Researchers are inspired to design and prepare self-healing materials to increase the lifetime of materials and reduce the total cost of systems during the process of long-term use. Recently, great progress has been made in self-healing composite materials that possess the ability to restore their structure and functionality after damage. Early self-healing materials were focused on microcapsule or microtubule by release of healing agents to achieve repairing. However, the self-healing times of these methods are dependent on the amounts of healing agents in the microcapsule or microtubules [ 1 ]. To address these limitations, dynamic chemistry involving dynamic covalent chemistry (e.g., imine bonds [ 2 , 3 ], disulfide bonds [ 4 , 5 , 6 ], acylhydrazone bonds [ 7 , 8 , 9 ], and boronate ester bonds [ 7 ]) and non-covalent interactions, such as hydrogen bonds [ 10 , 11 ], π−π stacking [ 12 ], hydrophobic interactions [ 13 , 14 ], host-guest interactions [ 15 ], ionic interactions [ 16 ], electrostatic interactions [ 17 ], and metal-coordination interactions [ 18 , 19 , 20 ], has been recently introduced to construct self-healing materials with multiple reversible healing ability. Graphene, as a new type of two-dimensional planar monolayer of sp 2 carbon atoms, has attracted widespread attention in all kinds of research areas due to its large specific surface area, excellent electrical conductivity, thermal conductivity, and unique mechanical properties [ 21 , 22 , 23 , 24 , 25 ]. Recently, graphene or graphene derivatives have been widely introduced into polymer matrices. The excellent performance of graphene or graphene derivatives, combined with the advantages of the polymer matrix, makes graphene/polymer composites suitable for application in conductive devices, coating, and biological and pharmaceutical field [ 26 , 27 , 28 , 29 , 30 ]. Although graphene-based composites have been well established [ 31 , 32 , 33 , 34 ], graphene-containing composites with self-healing capacity have not been summarized up to now. Introduction of self-healing capability into graphene/polymer composites will endow them with the ability of repairing themselves after damage and enlarge their service life. A lot of studies have been reported on the self-healing of the graphene/polymer composites due to their wide applications ( Figure 1 ). Therefore, it is necessary to review self-healing graphene/polymer composites, which combine the outstanding properties of graphene with advantages of the polymeric matrix and can be used in the field of mechanics, thermology, photology and electricity. In this review, the current advances in self-healing graphene/polymer composites have been summarized, including their preparation methods, self-healing conditions, properties and applications. Finally, the future prospects of the self-healing graphene/polymer composites are discussed."
} | 1,107 |
11580860 | PMC56995 | pmc | 7,035 | {
"abstract": "Background Genomic sequence analyses have shown that horizontal gene transfer occurred during the origin of eukaryotes as a consequence of symbiosis. However, details of the timing and number of symbiotic events are unclear. A timescale for the early evolution of eukaryotes would help to better understand the relationship between these biological events and changes in Earth's environment, such as the rise in oxygen. We used refined methods of sequence alignment, site selection, and time estimation to address these questions with protein sequences from complete genomes of prokaryotes and eukaryotes. Results Eukaryotes were found to evolve faster than prokaryotes, with those eukaryotes derived from eubacteria evolving faster than those derived from archaebacteria. We found an early time of divergence (~4 billion years ago, Ga) for archaebacteria and the archaebacterial genes in eukaryotes. Our analyses support at least two horizontal gene transfer events in the origin of eukaryotes, at 2.7 Ga and 1.8 Ga. Time estimates for the origin of cyanobacteria (2.6 Ga) and the divergence of an early-branching eukaryote that lacks mitochondria ( Giardia ) (2.2 Ga) fall between those two events. Conclusions We find support for two symbiotic events in the origin of eukaryotes: one premitochondrial and a later mitochondrial event. The appearance of cyanobacteria immediately prior to the earliest undisputed evidence for the presence of oxygen (2.4–2.2 Ga) suggests that the innovation of oxygenic photosynthesis had a relatively rapid impact on the environment as it set the stage for further evolution of the eukaryotic cell.",
"conclusion": "Conclusions Our analyses of prokaryotic and eukaryotic genomic sequence data support two symbiotic events in the origin of eukaryotes: one premitochondrial (2.7 billion years ago, Ga) and a later mitochondrial event (1.8 Ga). Our time estimate for the divergence of an early-branching eukaryote ( Giardia ) that lacks mitochondria, 2.2 Ga, suggests that it is a primary and not secondary amitochondriate organism. Our time estimate for the origin of cyanobacteria (2.6 Ga) is more recent than expected and suggests that earlier fossils claimed to be of cyanobacteria are of other organisms (or artifacts). Moreover, the appearance of cyanobacteria immediately prior to the earliest undisputed evidence for the presence of oxygen (2.4–2.2 Ga) suggests that the innovation of oxygenic photosynthesis had a relatively rapid impact on the environment as it set the stage for further evolution of the eukaryotic cell.",
"discussion": "Discussion The purpose of this study was to examine the temporal relationship between the origin of eukaryotes and events in Earth history. However, some unexpected results required refinement in methodology. These included finding greater among-site rate variation in the calibration group and different rates of sequence change between prokaryotes and eukaryotes, and between eukaryotes derived from different groups of prokaryotes. By taking into account these variables, the resulting time estimates are more robust and have fewer assumptions. For example, the time estimate for the origin of eukaryotes (BK-o) is not based on a general assumption of rate constancy between prokaryotes (or even eubacteria) and eukaryotes because rates are adjusted for each protein and each comparison. Also, the calibration used for BK-o is not a general eukaryotic calibration but one based exclusively on eukaryote sequences derived from eubacteria. A tradeoff in these improved methods was a reduction in the number of proteins that could be used, which increased the variance of the time estimates. Nonetheless, the phylogenies and time estimates obtained in this study have a bearing on current models for the evolution of eukaryotes. Until about five years ago, it was generally accepted that there was a prior period (before mitochondria) in the history of eukaryotes [ 2 , 26 ]. The basal position of eukaryotes lacking mitochondria (amitochondriate) in phylogenetic trees [ 27 ] was consistent with this supposition as was evidence from sequence signatures [ 6 ]. However, molecular phylogenetic studies of several proteins in recent years have suggested that some or all amitochondriate eukaryotes once possessed mitochondria in the past [ 9 ]. Based on this new evidence, most current models for the origin of eukaryotes assume only a single symbiotic or fusion event between an archaebacterium and an α-proteobacterium [ 8 , 28 , 29 ]. Under the single-symbiosis model, eukaryotes should cluster exclusively with an α-proteobacterium (e.g., Rickettsia ), among eubacteria. However, our phylogenetic analyses (Fig. 4 ) instead indicate, significantly, that many eukaryotic proteins originated from one (or more) eubacterial lineages other than α-proteobacteria. The reduced genome of Rickettsia [ 25 ] would not explain this result because Rickettsia possesses all of the proteins used in the combined analysis (Fig. 4B ). Protein function and location also are consistent with a premitochondrial origin. Only one of the 32 BK-o proteins is restricted to the mitochondrion whereas eight of the nine BK-m proteins are restricted to that organelle. Also, all six of the proteins involved in cellular respiration are in the BK-m group. Based on the serial endosymbiosis theory, the first symbiotic event involved a spirochete [ 3 ]. On the other hand, sequence signatures of the heat shock molecular chaperone protein HSP-70 and other evidence have indicated that the first symbiotic event involved a gram-negative eubacterium [ 6 ]. Our data are unable to distinguish between these two alternatives but agree with both in implicating an earlier, premitochondrial event. Predation by prokaryotes on early eukaryotes also may have led to HGT. If two or more symbiotic events were involved, this does not necessarily confirm that any of the living lineages of amitochondriate eukaryotes arose prior to the second (mitochondrial) event. All may have once possessed mitochondria. However, because Giardia arose at an early time (Table 1 ) and branches near the base of the eukaryote phylogeny, the simplest explanation is that it never possessed mitochondria and is a primary (not secondary) amitochondriate. Although the position of Giardia in some protein phylogenies [ 30 ] has been proposed as evidence that it is a secondary amitochondriate, others have urged caution until additional, more conclusive, data become available [ 6 ]. The number of symbiotic events was important for our primary concern of estimating a timescale for the early evolution of eukaryotes. We find that the divergence between archaebacteria and the lineage leading to eukaryotes (K A ) was quite early (~4 Ga), which is about the time of the earliest biomarker evidence of life (3.9–3.8 Ga) [ 31 ]. We interpret that divergence to be a speciation event between two lineages of archaebacteria, with K A not becoming \"eukaryotic\" until the first symbiotic event at 2.7 Ga. The remaining time estimates cluster around the mid-life of Earth (1.8–2.7 Ga). The order of those events falls in a logical sequence: BK-o, BC, and BK-m. For example, the origin of mitochondria appears as the second (not first) symbiotic event, and the origin of cyanobacteria comes before the oxygen-utilizing organelles, mitochondria. Moreover, the timing of these biological events is consistent with the timing of events in geologic and atmospheric history (Fig. 5 ). Cyanobacteria appear before the major (undisputed) evidence of the rise in oxygen (2.4–2.2 Ga) and mitochondria appear after the rise in oxygen. Also, the estimates for the origin of cyanobacteria and eukaryotes are consistent (within one SE) with the earliest biomarker evidence for those two groups (~2.7 Ga.) [ 11 , 15 ]. Phylogenetic analyses of photosynthetic genes and sequence signatures also support a relatively late order of appearance of cyanobacteria among photosynthetic prokaryotes [ 32 , 33 ]. Figure 5 Summary diagram showing relationship between timing of evolutionary events (Table 2) and that of Earth and atmospheric histories. Time estimates are shown with ± 1 standard error (thick line) and 95% confidence interval (narrow line). The phylogenetic tree illustrates the radiation of extant eubacterial lineages (blue), and dashed lines with arrows indicate the origin of eukaryotes (BK-o) and origin of mitochondria (BK-m). The earliest divergence (last common ancestor) was not estimated but is placed (arbitrarily) just prior to the AK divergence. The increasing thickness of the eukaryote lineage represents eubacterial genes added to the eukaryote genome through two major episodes of horizontal gene transfer. The rise in oxygen represents a change from <1% to >15% present atmospheric level [34,52], although the time of the transition period and levels have been disputed [19,53]. Extensive glaciations occurred in the Paleoproterozoic (~2.4 Ga), and may have been global in extent [ 34 ]. It has been proposed that a major rise in oxygen at this time lowered global temperatures and may have triggered the glaciations [ 35 ]. If this is true, and given the time estimates here, the evolutionary innovation of oxygenic photosynthesis may have had a relatively rapid impact on the environment. Moreover, this innovation may have caused a mass extinction of prokaryotes at that time, as a result of the toxic effects of oxygen, as suggested by the virtual absence of lineages prior to ~2.5 Ga and subsequent rapid radiation of lineages (Figs. 4 , 5 )."
} | 2,388 |
40263605 | PMC12015412 | pmc | 7,036 | {
"abstract": "Reinforcement learning algorithms that handle continuous action spaces have the problem of slow convergence and local optimality. Hence, we propose a deep deterministic policy gradient algorithm based on the dung beetle optimization algorithm (DBOP–DDPG) and priority experience replay mechanism. This method first adopts the simultaneous search policy of multiple populations by introducing the dung beetle optimizer (DBO), which can effectively keep the algorithm from falling into the local optimum solution and improve global optimization capability. Then, we design a criterion for determining the priority of sample data. The experience replay mechanism sampling is improved, and sample data in the experience replay mechanism are stored in three replay mechanisms based on importance for subsequent sampling training to then improve the algorithm’s convergence speed. Finally, tests were conducted in three classic control environments of OpenAI Gym. The results showed that the improved method improved the convergence speed by at least 10% compared with the comparison algorithm, and the cumulative reward value was increased by up to 150.",
"conclusion": "Conclusion This paper proposes an improved DDPG algorithm that combines the DDPG and DBO algorithms. To improve the convergence speed of the algorithm, we design a sample priority function. Taking both TD-error and immediate reward values into account, we perform data normalization and select maximum value as the basis for setting sample importance. During the sampling process, three experience playback bodies are designed to sample excellent experiences, general experiences, and failure experiences. To improve the global optimization ability of the algorithm, we introduce the DBO algorithm with its strong optimization ability. Through experiments in three continuous action space reinforcement learning experimental environments, we find that our proposed algorithm has improved convergence speed and global optimization compared with existing algorithms. However, although this paper proves that the proposed algorithm shows better performance, it still has many shortcomings. For example, it performs poorly when dealing with certain environmental problems and sometimes converges slowly when dealing with complex environments.",
"introduction": "Introduction In recent years, with the continuous development of reinforcement learning (RL), we have seen promising results in processing continuous action RL tasks 1 – 5 . In dealing with some continuous action RL tasks, RL is currently most commonly used with the deep deterministic policy gradient (DDPG) algorithm as a baseline. This algorithm has been continuously improved upon, and multiple variants have achieved remarkable results in processing RL tasks 6 – 9 . Although some achievements have been made in solving continuous action space tasks using the DDPG algorithm, there is still room for improvement in terms of the algorithm’s convergence speed and the problem of falling into local optimality. This is the goal of the present article. First, we explain the principles behind the DDPG reinforcement learning algorithm and analyze its flaws. Second, the study aims to solve the problem of poor optimization ability of the algorithm, we introduce the dung beetle optimization algorithm and adopt a strategy of simultaneous search of multiple populations to avoid falling into the local optimum. Furthermore, in view of the problem of slow convergence speed, we recognize that the experience replay (ER) adopts random sampling of experience samples during the training process without considering the difference in importance of the experience samples. Therefore, we design a criterion for determining the importance priority of sample data. We also construct an ER mechanism that contains excellent ER bodies, general ER bodies, and failure ER bodies. Moreover, we set three methods of playback volume sample number. Finally, we examine the performance of the improved algorithm and four existing algorithms on three continuous action space tasks. The remainder of this article is structured as follows: section “DDPG algorithm” introduces the DDPG algorithm; section “Improved Algorithm” presents our approach to improving the algorithm; section “Experiments” describes our experiments and gives an analysis of the results; and section “Conclusion and prospects” provides conclusions and future prospects.\n\nIntroducing the DBO algorithm to improve optimization ability When solving the value function, the DDPG algorithm uses a neural network for approximate representation. This approach leads to a series of problems such as overestimation of value and suboptimal strategies. This article introduces the DBO algorithm to combat these issues."
} | 1,186 |
35263134 | PMC8906750 | pmc | 7,037 | {
"abstract": "The factors controlling lignin composition remain unclear. Catechyl (C)–lignin is a homopolymer of caffeyl alcohol with unique properties as a biomaterial and precursor of industrial chemicals. The lignin synthesized in the seed coat of Cleome hassleriana switches from guaiacyl (G)– to C-lignin at around 12 to 14 days after pollination (DAP), associated with a rerouting of the monolignol pathway. Lack of synthesis of caffeyl alcohol limits C-lignin formation before around 12 DAP, but coniferyl alcohol is still synthesized and highly accumulated after 14 DAP. We propose a model in which, during C-lignin biosynthesis, caffeyl alcohol noncompetitively inhibits oxidation of coniferyl alcohol by cell wall laccases, a process that might limit movement of coniferyl alcohol to the apoplast. Developmental changes in both substrate availability and laccase specificity together account for the metabolic fates of G- and C-monolignols in the Cleome seed coat.",
"introduction": "INTRODUCTION Lignin is the second most abundant biopolymer on Earth. Its monomeric composition determines not only its properties as a biomaterial but also its impact on biomass processing and forage digestibility ( 1 – 6 ). Lignin composition is more flexible than previously assumed, with a number of noncanonical “monolignol” building blocks found in recent years ( 7 ). Monolignols are polymerized via free radical reactions initiated/catalyzed by laccases and peroxidases ( 8 ), both of which are encoded by large gene families in plants ( 9 , 10 ). Recent studies have suggested that some laccases may show preference for specific monolignols ( 11 – 13 ), although it is unclear whether the final composition of lignin is determined by monomer availability, laccase/peroxidase specificity, or both. Catechyl (C)–lignin is a recently found homopolymer of caffeyl alcohol that occurs naturally in the seed coats of phylogenetically diverse plant species including orchids, cacti, several oil seed species including Jatropha and castor bean, and the ornamental plant Cleome hassleriana ( 14 – 16 ). In contrast, classical lignin is a complex heteropolymer of two to three different monolignol units found in the cell walls of secondarily thickened plant tissues such as xylem and interfascicular fibers in angiosperms and a guaiacyl (G)–rich polymer in secondary cell walls of gymnosperms ( 17 ). Heterogeneity and cross-linking of classical lignin limit its utility as a biomaterial ( 2 ). In contrast, C-lignin has been identified as an ideal subject for lignin valorization in the biorefinery. This is because its homogeneity (caffeyl alcohol units linked exclusively through benzodioxane structures) favors its selective reductive catalytic fractionation to simple products for subsequent biological funneling ( 18 – 20 ), and its linearity, with lack of cross-linking either internally or to other cell wall polymers, makes it a good material for production of carbon fibers ( 21 ). Furthermore, C-lignin is stable to pretreatment conditions used during processing of lignocellulosic biomass for fermentation to liquid biofuels ( 18 ), in contrast to classical lignin that can dissolve and reprecipitate during pretreatment. However, other than limited evidence for the presence of C-lignin in root cultures of genetically modified pine ( 22 ), the polymer currently appears to be limited to seed coats. Lignin biosynthesis in the C. hassleriana seed coat switches markedly from production of G-lignin to production of C-lignin at around 12 to 14 days after pollination (DAP) ( 16 ). This makes C. hassleriana an excellent model system to interrogate the molecular mechanisms underlying C-lignin biosynthesis. Synthesis of caffeyl alcohol requires suppression of the O -methylation reactions involved in the biosynthesis of the major monolignol coniferyl alcohol (the G unit of lignin) ( 16 ), and C-lignin accumulation in the Cleome seed coat is associated with a rapid decline in the gene expression and enzymatic activities of caffeoyl coenzyme A (CoA) 3- O -methyltransferase ( CCoAOMT ) and caffeic acid/5-hydroxyconiferaldehyde 3/5- O -methyltransferase ( COMT ) ( 23 ). Although COMT has been ascribed a role primarily in syringyl (S)–lignin biosynthesis ( 24 ), it is likely associated with the biosynthesis of G-lignin in the C. hassleriana seed coat, as S-lignin is absent ( 14 ). Although a dehydrogenation polymer with identical composition and linkage type to natural C-lignin can be generated in vitro by peroxidase-mediated coupling of caffeyl alcohol ( 14 ), the biosynthesis of caffeyl alcohol and its precursor caffealdehyde has yet to be demonstrated in plant tissues producing C-lignin. Here, we use a combination of metabolite profiling, isotopic labeling analyses, metabolic flux analysis (MFA), and enzyme specificity studies to determine the origin, levels, and fates of the precursor pools for lignin biosynthesis during the development of the C. hassleriana seed coat. Unexpectedly, coniferyl alcohol is still made during the period of C-lignin biosynthesis but not incorporated into lignin. We present a model in which caffeyl alcohol production directed to C-lignin and C-lignans accompanies the maintenance of a pool of coniferyl alcohol, presumably for other cellular functions. This is achieved, in part, through caffeyl alcohol inhibiting the oxidation of coniferyl alcohol by cell wall laccases and peroxidases, potentially preventing its movement to the apoplast.",
"discussion": "DISCUSSION Monolignol precursor pools do not predict lignin composition in the Cleome seed coat LC-MS–based targeted metabolite profiling of both seed coats and whole seeds revealed that the level of the G-lignin precursor coniferyl alcohol was, paradoxically, approximately fourfold higher at 16 DAP, during the period of C-lignin biosynthesis, than at 8 DAP when G-lignin is being made. The C-lignin precursor caffeyl alcohol was not detected at 8 DAP and was present at approximately 300 nmol/g DW (less than half the level of coniferyl alcohol) at 16 DAP. Levels of the H-lignin precursors coumaraldehyde and p -coumaryl alcohol were very low at 8 DAP but substantially increased at 16 DAP, although no H-lignin is produced from phenylalanine at this time. This suggests the possibility of a pathway from p -coumaryl alcohol to caffeyl alcohol in the seed coats. However, such a conversion is not catalyzed by the bifunctional ascorbate peroxidase/C3H ( 35 ), and, furthermore, MFA did not support its operation. The metabolic role/fate of p -coumaryl alcohol at the later times of seed coat development therefore remains unclear. MFA reveals nonclassical routes to monolignol pathway intermediates during Cleome seed coat development The MFA analysis was based on a model of the monolignol pathway in Brachypodium distachyon that postulates equilibration of pools of metabolites between soluble and endoplasmic reticulum (ER)–associated compartments ( 33 , 34 ). To obtain minimum SSR values, it was necessary to include an additional flux into unlabeled coumarate. Studies in B. distachyon have shown the presence of more than one pool of coumarate with different metabolic fates ( 46 ), associated with the presence of both phenylalanine and tyrosine ammonia lyases in this species ( 46 ). To the best of our knowledge, cinnamate, tyrosine, and various coumarate esters are the only direct precursors of coumarate. Our labeling strategy using half seeds was optimal for label incorporation, but it is possible that the large unlabeled pool of coumarate external to the seed coat could provide substantial flux into monolignol biosynthesis. The 3-hydroxylation of the monolignol aromatic ring can occur either at the level of the shikimate ester through the cytochrome P450 C3′H ( 47 ) or at the level of the free acid (coumarate) through C3H ( 35 ). The genes encoding the two key enzymes of the classical shikimate esters pathway, HCT and C3′H, are expressed at similar levels throughout seed coat development ( 23 ), and, after initially dropping, levels of both caffeoyl shikimate and caffeic acid increase between 12 and 16 DAP. MFA suggested that the esters pathway is the major route for G-lignin biosynthesis at 8 to 10 DAP. However, the acids route via C3H then becomes dominant at the time of the switch to C-lignin formation, with both pathways likely operating at later times. Accumulation of lignans and CTs is temporally separated in the Cleome seed coat The Cleome seed coat contains a wide range of lignans derived from caffeyl alcohol. Some of these accumulate to very high levels, but only during the phase of C-lignin biosynthesis. One specific benzodioxane-linked C-lignan, trans -isoamericanol, was the most abundant metabolite detected in the seed coats after sucrose and glucose. Two aspects of these results are unexpected. The first is that no lignans derived from coniferyl alcohol, other than dehydrodiconiferyl alcohol and traces of pinoresinol, or from p -coumaryl alcohol, accumulated to substantial levels in the seed coats at any stage of development. These molecules are commonly found in lignifying tissues ( 48 , 49 ). It is also unexpected that no mixed G-C lignans were observed, as down-regulation of CCoAOMT results in formation of isomers of benzodioxane-linked coniferyl alcohol-8- O -4-caffeic acid dimer in Arabidopsis vacuoles ( 48 ). Second, the diversity of C-lignans, with multiple linkage types, is in notable contrast to the structure of C-lignin, in which all the units are joined through β- O -4 linkages, with subsequent internal cyclization to create benzodioxane units ( 14 ). Most commonly, lignans are linked via so-called β-β′ bonds between the central atoms of the respective side chains (position 8); 3-3′, 8- O -4′, or 8-3′ bonds are observed less frequently, and, in these cases, the dimers are called neolignans. The exact localization of the lignans in the Cleome seed coat remains to be determined. Lignans are thought to be synthesized in the apoplast by the combined activities of a laccase and a dirigent protein ( 50 ). However, the presence of at least three different linkage types in the Cleome seed coat, along with different stereochemistries, suggests that C-lignan formation may only be under chemical control and not involve dirigent proteins to control the stereospecificity of the coupling ( 50 ). The presence of a wide range of glycosylated monolignol dimers and higher oligomers in the vacuoles of Arabidopsis has led to the hypothesis that lignans may also be made in the cytosol ( 48 ). The G-lignan dehydrodiconiferyl alcohol has been shown to exist as its glucoside in Arabidopsis vacuoles ( 48 ). It is notable that no C-lignan glycosides were detected in the present work, suggesting that the C-lignans may be present in the Cleome seed coat cell wall, where they might act as defensive compounds. The fact that ChLAC8, with an extracellular transit peptide, can form in vitro the multiple C-monolignol dimer types (resinol and benzodioxane stereoisomers) extracted from seed coat tissues is consistent with these lignans being made and accumulated in the apoplast. CTs accumulate early during Cleome seed coat development ( 12 ) and are shown here to be of the less common propelargonidin type derived from the flavan 3-ol monomer afzelechin. Recent studies indicate that CT polymerization is nonenzymatic, initiated by attack of a carbocationic extension unit at the nucleophilic C8 position of a flavan-3-ol starter unit ( 51 , 52 ). The laccase TT10 has been ascribed roles in the oxidation of CTs and also in lignin polymerization in the Arabidopsis seed coat, but a role in tannin oligomerization seems unlikely ( 44 , 53 ). ChLAC15 , an ortholog of Arabidopsis TT10 , although expressed most highly at 10 DAP, is still strongly expressed during the period of C-lignin biosynthesis. ChLAC15 could oxidize caffeyl alcohol but did not appear to oxidize monomeric afzelechin, and the flavan 3-ol did not appear to affect laccase-mediated monolignol polymerization in vitro. These results suggest that lignin/lignan and CT oligomerization are nonoverlapping processes in the Cleome seed coat. Different metabolic fates of coniferyl and caffeyl alcohols during the period of C-lignin biosynthesis The fact that monolignols are made in the cytosol but polymerized in the apoplast is important for interpreting the present labeling experiments. Monolignol synthesized from an upstream precursor such as L-Phe or trans -cinnamic acid will have to transfer across the plasma membrane, whereas monolignols applied to a cut surface with vacuum infiltration could also find their way directly to the apoplast or be involved in ectopic lignification at the cut (wounded) surface. This distinction might explain why there was incorporation of applied coniferyl alcohol into G-lignin at 16 to 18 or 22 to 24 DAP, whereas 13 C-Phe did not label G-lignin at these times. The incorporation of 13 C-caffeyl alcohol into G-lignin at 8 to 10 DAP, but not at 12 to 14 DAP, is explained by the presence of COMT activity during the earlier labeling period; caffeyl alcohol is an excellent substrate for COMT ( 54 ). Bearing these points in mind, the labeling data show that the substrate pool available for polymerization in the apoplast is not simply a reflection of the overall cellular plus extracellular monolignol pool sizes. The higher level of coniferyl alcohol during the period of C-lignin biosynthesis than during G-lignin biosynthesis might simply reflect the cessation of polymerization of coniferyl alcohol after around 12 DAP, consistent with the inhibition of coniferyl alcohol oxidation by caffeyl alcohol, discussed further below. If this were the only explanation, then the coniferyl alcohol accumulating at late time points would have been synthesized during the phase of G-lignin biosynthesis. However, this model is inconsistent with the continued de novo synthesis of coniferyl alcohol from L-Phe during the phase of C-lignin biosynthesis. The retained commitment for coniferyl alcohol biosynthesis raises two major questions: Why is coniferyl alcohol made when it is not being incorporated into lignin, and how is it made when the methylation machinery for its biosynthesis has been turned off to enable C-lignin biosynthesis ( 23 )? In addition to lignin, coniferyl alcohol is also a precursor of lignans such as the dehydrodiconiferyl alcohol glucosides that have been proposed to function as factors controlling cell division ( 55 , 56 ). Coniferyl alcohol can also act as a signal molecule in plant-microbe interactions ( 57 ). Thus, it may be critical for more than one reason for plants to maintain a pool of coniferyl alcohol, even under conditions in which the nonmethylated caffeyl alcohol must be made in large amounts to support C-lignin biosynthesis. The most extreme negative growth phenotypes in monolignol pathway mutants in Arabidopsis and Medicago truncatula result from simultaneous loss of function of COMT and CCoAOMT; the resulting dwarf seedlings die well before reaching maturity ( 49 , 58 ). Although both classical COMT and CCoAOMT genes are strongly down-regulated at the onset of C-lignin biosynthesis in Cleome , seed coat development continues and coniferyl alcohol is still made. This suggests that the Cleome seed coat has a pathway to coniferyl alcohol that is independent of the activities of classical COMT and CCoAOMT, and this is supported by the relatively high incorporation of caffeyl alcohol into coniferyl alcohol between 16 and 22 DAP. Studies are underway to identify the O -methyltransferase(s) involved. Caffeyl alcohol is an inhibitor of coniferyl alcohol oxidation by cleome seed coat laccases In A. thaliana , loss of function of AtLAC4 , AtLAC11 , and AtLAC17 results in plants in which lignin can only be detected in the Casparian strip in the root ( 40 ), a tissue in which laccases are not necessary for lignin polymerization ( 40 , 59 ). The fact that plants of the lac4 , lac11 , and lac17 triple mutant express a normal complement of peroxidase genes suggests that laccase is essential for initiating lignification in tissues other than the Casparian strip ( 40 ). Several laccases are expressed in the seed coat of Cleome ; on the basis of transcript profiles, ChLAC4 and ChLAC17 are expressed mainly during the phase of G-lignin biosynthesis, as is ChLAC15 , the ortholog of the TT10 gene of Arabidopsis ( 43 ). ChLAC8 expression is more closely correlated with C-lignin accumulation ( 12 ), and this enzyme, when expressed in a prokaryotic system that does not allow for glycan substitution of the laccase protein, demonstrates a strong preference for oxidation of caffeyl alcohol as compared to coniferyl alcohol ( 12 ). This enzyme is shown here to maintain this substrate preference when expressed transiently as a glycosylated protein in N. benthamiana . Although the complement of laccases in the Cleome seed coat changes significantly during the switch from G- to C-lignin accumulation, the inhibitory effect of caffeyl alcohol on oxidation of coniferyl alcohol appears similar at all developmental stages. Of the recombinant seed coat-expressed Cleome laccases, only ChLAC4 could effectively oxidize coniferyl alcohol and then only in the presence of a yet-to-be identified water-soluble factor that is unlikely to be a protein/enzyme based on its extraction in methanol, followed by air drying and water/ethyl acetate partitioning. This reaction was totally inhibited by an equimolar concentration of caffeyl alcohol and by much lower concentrations when experiments were conducted with crude cell wall laccase preparations. The demonstration of noncompetitive inhibition of cell wall laccases by caffeyl alcohol with an apparent Ki of only 1.9 μM suggests that levels of caffeyl alcohol are sufficient to block oxidation of coniferyl alcohol in the apoplast and that increasing concentrations of coniferyl alcohol will not easily overcome this. Furthermore, the inhibition of peroxidase-mediated oxidation of coniferyl alcohol by caffeyl alcohol provides a further mechanism to ensure C-homopolymer biosynthesis at late stages of seed coat development. A model for multilevel regulation of C-lignin biosynthesis Phylogenetic studies suggest that C-lignin has evolved recently and often in certain plant lineages ( 15 ). The underlying mechanism of C-lignin biosynthesis should therefore not require a number of independently controlled events but rather should feature a simple switch that turns on a cascade of downstream events. The defining event is likely the transcriptional switch that turns on the down-regulation of COMT and CCoAOMT ( 23 ), resulting in the formation of caffeyl alcohol, a powerful noncompetitive inhibitor of coniferyl alcohol oxidation. A comparison of the labeling of lignin from coniferyl alcohol generated in vivo or supplied exogenously appears to support control of the pathway at the level of monolignol transport across the apoplast. There is no incorporation of coniferyl alcohol formed in vivo from 13 C– l -Phe between 16 and 22 DAP, whereas 13 C-coniferyl alcohol supplied exogenously is incorporated at a level similar to that observed when it is fed at 8 to 14 DAP, possibly as a result of ectopic formation of G-lignin at cut surfaces and/or following vacuum infiltration. Figure 9A presents a model to explain the changes of lignin composition during development of the Cleome seed coat based on the present data. Before around 12 to 13 DAP, lignin biosynthesis proceeds by the classical pathway involving the shikimate shunt. No S-lignin accumulates because F5H is not expressed ( 23 ). At around 12 DAP, a switch (likely involving transcription factor expression) leads to suppression of the two monolignol O -methyltransferases, COMT and CCoAOMT , resulting in the biosynthesis of caffeyl alcohol, assisted by the expression of CAD5 , a form of CAD with preference for caffealdehyde ( 23 ). The mechanism behind the shift in the monolignol pathway from the shikimate shunt to the “acids” pathway during the onset of C-lignin biosynthesis requires further study. Fig. 9. A model for the control of lignin composition in the Cleome seed coat based on substrate availability, transport, and polymerization. ( A ) Between 8 and 10 DAP, coniferyl alcohol is synthesized by the classical shikimate esters pathway via caffeoyl shikimate. Feruloyl CoA is converted to coniferaldehyde and thence to coniferyl alcohol by a proposed complex of ChCCR1 and ChCAD4. Formation of coniferyl alcohol close to the plasma membrane may facilitate its passive diffusion to the apoplast, where it is oxidized by ChLAC4 and ChLAC17 to oligomers; further polymerization may involve cell wall peroxidases. To date, there is no evidence for a specific coniferyl alcohol transporter and if the diffusion is passive, it may be facilitated by the “pull” from the reduced apoplastic concentration. After approximately 13 DAP, down-regulation of COMT and CCoAOMT ( 23 ) results in formation of caffeoyl CoA but now with a major contribution from the acids pathway involving C3H ( 35 ), which accounts for essentially all the flux to caffeyl alcohol at 12 to 14 DAP. Caffeoyl CoA is reduced via ChCCR1 and ChCAD5 ( 23 ) to caffeyl alcohol, which may again reach the apoplast by passive diffusion, possibly assisted by active transport. Caffeyl alcohol inhibits the polymerization of coniferyl alcohol, thereby also reducing its potential diffusion to the apoplast, and is itself polymerized to C-lignin via ChLAC8 ( 12 ). DDCA, dihydrodiconiferyl alcohol. ( B ) Spatiotemporal deposition of lignin in the Cleome seed coat. At 8 DAP, G-lignin is being deposited in an outer lignified sublayer (OLS) of cells interior to the extratestal sublayer (ETC). A distinct layer of cells interior to the OLS is visible microscopically but does not start to accumulate lignin (C-lignin) to become the interior lignified sublayer (ILS) until around 14 DAP. By 20 to 24 DAP, the two cell layers have merged (merged lignified layer, MLL) and contain distinct C- and G-lignin homopolymers ( 16 ). The concentrations of coniferyl alcohol (black circles) and caffeyl alcohol (green circles) in the seed coat and rest of the seed (internal to the seed coat) are indicated (from fig. S2). Despite a previous demonstration of adenosine 5′-triphosphate–binding cassette transporters active with monolignol aglycones in plasma membrane vesicles ( 60 ), no G-monolignol (or S-monolignol) transporter has yet been identified at the molecular level ( 61 ). We therefore suggest that coniferyl alcohol likely arrives in the apoplast via passive diffusion ( 62 ). Work is in progress to determine whether there are specific transporters for caffeyl alcohol, although it is predicted to diffuse across a model plasma membrane more readily than coniferyl alcohol ( 62 ). Because of the low Ki value for caffeyl alcohol as an inhibitor of coniferyl alcohol oxidation, transporters may not be necessary to generate a sufficiently high concentration of free caffeyl alcohol to maintain inhibition of coniferyl alcohol oxidation, despite the higher concentrations of coniferyl alcohol than caffeyl alcohol in the seed coat ( Fig. 9B ). Furthermore, the passive diffusion model is suitably parsimonious in removing the need for additional factors for C-lignin accumulation, consistent with evolutionary considerations ( 15 ). Although not critical for the model, it is possible that both cinnamyl alcohol dehydrogenase 4 (CAD4) and CAD5 are located near the plasma membrane through association with cinnamoyl-CoA reductase (CCR), an enzyme that has been shown to interact with a RAC–guanosine triphosphatase that is part of a complex spanning the plasma membrane and orchestrating the localization of an enzyme complex for generation of apoplastic oxidizing equivalents that can support monolignol polymerization [reviewed in ( 32 )] ( Fig. 9 ). Such a localization could help facilitate diffusion of monolignols across the plasma membrane, driven by a negative concentration gradient as the monolignol is polymerized within the apoplast. Within the apoplast, caffeyl alcohol is preferentially oxidized by LAC8 ( 12 ). Inhibition of coniferyl alcohol oxidation by caffeyl alcohol might disrupt the concentration gradient and impede the passage of coniferyl alcohol to the apoplast. This feature of the model remains speculative absent a reliable technique to measure the relative concentrations of monolignols in the cytosol and apoplast but is not essential to explain the present data. Furthermore, the model supports either laccase- or peroxidase-mediated lignin polymerization. Compartmentation of G- and C-lignin biosynthesis to different cell types might obviate the need for controls of the type outlined above. Figure 9B summarizes our knowledge of the sites of lignin biosynthesis in the seed coat. Although C-lignin is initiated in a separate cell layer from G-lignin, this layer is interior to the G-lignin layer and therefore likely to have high levels of coniferyl alcohol represented by that in the “rest of the seed.” Ultimately, the G- and C-lignin layers fuse but retain individual G- and C-homopolymers. In summary, our results explain how C-lignin biosynthesis proceeds when coniferyl alcohol is still being formed and show that monolignol supply is not itself sufficient to determine lignin composition. The enzyme(s) that support COMT and CCoAOMT-independent formation of coniferyl alcohol remain to be identified, as does the mechanism that allows polymerization of caffeyl alcohol to C-lignin simultaneously with its oligomerization to a range of lignans with different linkage types."
} | 6,461 |
37426991 | PMC10325728 | pmc | 7,038 | {
"abstract": "Legumes are well-known for establishing a symbiotic relationship with rhizobia in root nodules to fix nitrogen from the atmosphere. The nodulation signaling pathway 2 ( NSP2 ) gene plays a critical role in the symbiotic signaling pathway. In cultivated peanut, an allotetraploid (2n = 4x = 40, AABB) legume crop, natural polymorphisms in a pair of NSP2 homoeologs ( N a \n and N b \n ) located on chromosomes A08 and B07, respectively, can cause loss of nodulation. Interestingly, some heterozygous ( N B n b \n ) progeny produced nodules, while some others do not, suggesting non-Mendelian inheritance in the segregating population at the N b locus. In this study, we investigated the non-Mendelian inheritance at the N B \n locus. Selfing populations were developed to validate the genotypical and phenotypical segregating ratios. Allelic expression was detected in roots, ovaries, and pollens of heterozygous plants. Bisulfite PCR and sequencing of the N b \n gene in gametic tissue were performed to detect the DNA methylation variations of this gene in different gametic tissues. The results showed that only one allele at the Nb locus expressed in peanut roots during symbiosis. In the heterozygous ( N b n b \n ) plants, if dominant allele expressed, the plants produced nodules, if recessive allele expressed, then no nodules were produced. qRT-PCR experiments revealed that the expression of N b \n gene in the ovary was extremely low, about seven times lower than that in pollen, regardless of genotypes or phenotypes of the plants at this locus. The results indicated that N b \n gene expression in peanut depends on the parent of origin and is imprinted in female gametes. However, no significant differences of DNA methylation level were detected between these two gametic tissues by bisulfite PCR and sequencing. The results suggested that the remarkable low expression of N b \n in female gametes may not be caused by DNA methylation. This study provided a unique genetic basis of a key gene involved in peanut symbiosis, which could facilitate understanding the regulation of gene expression in symbiosis in polyploid legumes.",
"introduction": "Introduction Nitrogen is an essential element for all living organisms, particularly for legume crops producing seeds of high protein content. In nature, legumes are mostly able to self-supply nitrogen due to the establishment of symbiosis with rhizobia in root nodules for biological nitrogen fixation (BNF), which contributes to sustainable agriculture ( Peoples et al., 1995 ). During the symbiosis, nitrogen from the atmosphere is converted into ammonia by rhizobia in root nodules, as nutrients to the plants ( Wilson and Burris, 1947 ). In return, rhizobia take carbon from the host plants for their energy. During the interaction between legumes and rhizobia, rhizobia first enter host plant roots either intracellularly via root hair or intercellularly via cracks on root surface ( Madsen et al., 2010 ). Nodules are then initiated in the root cortex beneath the rhizobial infection site as the infection of rhizobia proceeds ( Oldroyd, 2013 ). Rhizobia infect plant cells in the nodule primordia, where they differentiate into bacteroid, a form of rhizobia at nitrogen-fixing state. Up to date nearly two hundred genes required for the symbiosis process have been identified and characterized, mainly in two model legumes, Lotus japonicus and Medicago truncatula ( Roy et al., 2020 ). The molecular signaling processes initiate when the legumes release flavonoids to attract rhizobia. In response, rhizobia secrete nodulation (Nod) factors (NFs), lipochitooligosaccharide (LCOs) signaling molecules, which can be recognized by NF receptor 1 (NFR1) and NFR5 based on the studies in L. japonicus ( Limpens, 2003 ; E. B. Madsen et al., 2003 ). The interaction between NFR1/5 and NFs stimulates symbiotic calcium oscillations, which are further decoded by Ca 2+ /calmodulin-dependent protein kinase (CCaMK) and CYCLOPS ( Levy et al., 2004 ; Yano et al., 2008 ). The output of the symbiotic signal is further transmitted by the GRAS transcriptional factors (TFs), nodulation signaling pathway 1 (NSP1) and NSP2 ( Oldroyd and Long, 2003 ; Kalo, 2005 ; Arrighi et al., 2006 ) as an NSP1- NSP2 complex to induce Nodule Inception ( NIN) and required for Nodulation 1 ( ERN1) genes for nodule organogenesis process ( Stracke et al., 2002 ; Marsh et al., 2007 ). Cultivated peanut ( Arachis hypogaea L.) is an important legume crop grown worldwide. It is an allotetraploid (2n= 4x = 40, AABB) with a genome size of ~ 2.7 Gb ( Bertioli et al., 2019 ). As a legume species, cultivated peanut plants can establish symbiosis with Bradyrhizobia , a genus of soil-born slow-growing bacteria, and produce nodules regularly. Mutants of non-nodulating (Nod–) peanuts were first reported in progeny derived from the cross between two normally nodulating (Nod+) lines, PI 262090 and UF 487A-4-1-2 ( Gorbet and Burton, 1979 ). A recent effort of a forward genetics approach uncovered that natural nucleotide polymorphisms at a pair of NSP2 homoeologs caused the Nod– mutants in cultivated peanuts ( Peng et al., 2021 ). This pair of homoeologous NSP2 genes are located on chromosomes 08 and 17 of two subgenomes A and B, and thus were named AhNSP2-A08 ( N a \n ) and AhNSP2-B07 ( N b \n ), respectively. The natural mutant allele n a \n is a single nucleotide polymorphism (SNP) from cytosine to thymidine at the 673th nucleotide in the coding region, which leads to a premature stop codon that reduces the polypeptide size from 513 to 224 amino acids ( Peng et al., 2021 ). The natural mutant allele n b \n is a single nucleotide deletion of the cytosine at the 119 th nucleotide in the coding region, which causes a reading frame shift and leads to a premature stop codon, reducing the polypeptide size from 516 to 113 amino acids ( Peng et al., 2021 ). Predictably, the protein encoding sequences of both mutated alleles n a \n and n b \n miss all or most of the functional domains of the GRAS TF. In our previous research ( Peng et al., 2021 ), crossing of the Nod+ PI 262090 and UF 487A-4-1-2 with the confirmed genotypes of N a N a n b n b \n and n a n a N b N b \n produce a Nod+ F 1 ( N a n a N b n b \n ), the F 2 selfing population derived from which segregates the Nod– phenotype with the homozygous recessive alleles at both loci ( n a n a n b n b \n ), confirming the roles of both N a \n and N b \n in nodule production. Our further analysis revealed that the segregation of the nodulation phenotype follows the Mendelian inheritance with a 3:1 ratio of Nod+: Nod– at the N a \n locus in a selfing population of a heterozygous plant with the genotype of N a n a n b n b \n . The plants with the genotype N a _n b n b \n are Nod+, and the plants with the n a n a n b n b \n genotype are Nod–. By contrast, the phenotypic segregation ratio of Nod+ to Nod– varied from 5:3 to 1:1 at the N b \n gene locus in the selfing population of a heterozygous plant with the genotype of n a n a N b n b \n , violating a Mendelian 3:1 ratio. We further confirmed that the phenotype of the heterozygous lines with the genotype of n a n a N b n b \n can be either Nod+ or Nod– ( Peng et al., 2021 ), suggesting that some other mechanisms are underlying in regulation of the nodule phenotype at the N b \n locus. In this study, to understand the genetic mechanisms of the AhNSP2-B07 gene in controlling peanut nodulation, we analyzed the segregating ratios of peanut nodulation at the N b \n locus in the field, allelic expression of N b \n in peanut roots and gametic tissues, and DNA methylation at the N b \n locus. Our results revealed that the expression of AhNSP2-B07 is paternally expressed or imprinted in the peanut genome showing parent-of-origin effects. Although the bisulfite sequencing of the promoter and coding regions of AhNSP2-B07 revealed no significant differences in the methylation level between ovary and pollen, it is suspected that either the tissues we sampled didn’t resolve to detect the gametic methylation variations or the imprinting is not due to DNA methylation. To our knowledge, this is the first functional embryonic imprinted gene identified in crop species.",
"discussion": "Discussion Expression bias of homoeologs in polyploids Polyploidy, or whole genome duplication, creates novel phenotypes that enable the polyploids to better adapt to environmental changes, which is critical for crop evolution and crop domestication ( Chen et al., 2007 ; Yoo et al., 2014 ). As a result of polyploidization, duplicated genes from two subgenomes known as homoeologs, may share the same functions and expression patterns or have different functions. Several potential fates could occur for homoeologs after whole genome duplication. For example, both genes keep the original function, one copy gets silenced and becomes pseudogene, or the genes diversify with different functions or expression patterns ( Lynch and Force, 2000 ). Both natural polyploids and synthetic polyploids have been investigated for homoeolog expression bias at the whole genome level. Allotetraploid cotton ( Gossypium hirsutum ) (AADD) is one of the most widely cultivated crops from hybridization between G. arboreum (A2) and G. raimondii (D5), with 15.90–37.96% of genes showed different expression biases towards A or D subgenomes in fibers among different cultivated cotton cultivars ( Mei et al., 2021 ). In another study, Raphanobrassica (RRCC), which was artificially synthesized from Raphanus sativus (RR) and Brassica oleracea (CC) showed genome-wide unbalanced biased expression bias towards B. oleracea ( Zhang et al., 2021 ). In Arabidopsis suecica , more sequence deletions were detected in the less-expressed subgenome ( Chang et al., 2010 ; Garsmeur et al., 2014 ), which might lead to biased expression of the homoeologs. While most of the species in the genus Arachis are diploid, A. hypogaea is an allotetraploid species, which is most likely derived from the hybridization between two wild diploid Arachis ancestors, followed by polyploidization ( Husted, 1936 ; Burris and Roberts, 1993 ; Kochert et al., 1996 ). Overall, the total number of biased expressed homoeologs towards the A subgenome was similar to the number towards the B subgenome ( Bertioli et al., 2019 ). However, this difference is significant in some specific tissues, such as pericarp, perianth, root, and peg, which have significantly higher number of highly expressed genes in the B subgenome than in the A subgenome ( Bertioli et al., 2019 ). For the NSP2 homoeologs in peanuts, the segregation of AhNSP2-A08 fits the Mendelian segregation ratio, while the segregation of AhNSP2-B07 does not. Based on the survey of the mutation rate of the two NSP2 homoeologs in a US mini core collection, AhNSP2-A08 has a much lower loss of function mutation rate than AhNSP2-B07 ( Peng et al., 2021 ), which may suggest that AhNSP2-A08 has undergone stronger purifying selection during evolution while AhNSP2-B07 is gradually losing its function with a higher frequency of mutation rates. Moreover, it has been revealed that AhNSP2-A08 was highly expressed only in roots, whereas AhNSP2-B07 was highly expressed in both roots and flowers ( Peng et al., 2021 ). This indicated that AhNSP2-B07 may be related to the reproductive process. Parental effect controlling AhNSP2-B07 \n In a previous study ( Gallo-Meagher et al., 2001 ), reciprocal crosses were made between Nod– M4-2 ( n a n a n b n b \n ) and Nod+ PI 262090 ( n a n a N b N b \n ), the parental lines of E4 and E5 used in this study. All 30 F 1 progeny were Nod+ with uncountable nodules only when the wild-type allele N b \n was from the male parent ( \n Table 3 \n ). However, when the wild-type allele N b \n was inherited from the female parent, most of the F 1 progeny (32 out of 33) were Nod– showing non-nodulation or only few unusual big nodules ( \n Table 3 \n ). The dramatic difference in F 1 phenotypes between reciprocal crosses suggested a parent-of-origin effect ( Gallo-Meagher et al., 2001 ) at N b \n . For the reciprocal crosses between Nod+ UF487A ( N a N a n b n b \n ) and Nod- M4-2 ( n a n a n b n b \n ), all F 1 progeny (32 and 26), except one, were Nod+, suggesting that N a \n had no imprinting effect. Therefore, we hypothesized that the allele from female gametes in peanuts at the N b \n locus is inhibited or imprinted with little or no expression, whereas the allele from male gametes at N b \n can express normally in the offspring ( \n Table 4 \n ), which was further validated by qRT-PCR on gametic tissue ( \n Figure 4 \n ). According to this hypothesis, an equal number of Nod+ and Nod– N b n b \n offspring should be produced from the selfing population of N b n b \n plants, and thus a 1 Nod+:1 Nod– segregating ratio was expected to be observed in the field. In reality, the segregating ratio of Nod+: Nod– fell within the range of 5:3 to 1:1 ( Peng et al., 2021 ). Considering the Nod– plants have a relatively lower survival rate in the field due to nitrogen depletion, as starter nitrogen fertilizer was only applied during planting time. The practice of limiting nitrogen fertilizer maximized the phenotypic differences between Nod+ and Nod– peanut plants in the field, which also reduced the survival rate of low vigor Nod– plants inevitably. Therefore, the segregating ratio observed in the field during harvest is very likely shifted from the 1:1 ratio to some extent. Table 3 Phenotypes of F1 progeny of reciprocal crosses between different genotypes of nodulating and non-nodulating plants (Adapted from Gallo-Meagher et al., 2001 ). Cross Genotypes Phenotypes of F 1 \n Female Male Nod+ Few nodules Nod– UF 487A M4-2 \n N a N a n b n b \n × n a n a n b n b \n \n 32 0 1 M4-2 UF 487A \n n a n a n b n b \n × N a N a n b n b \n \n 26 0 0 PI262090 M4-2 \n n a n a N b N b \n × n a n a n b n b \n \n 1 24 8 M4-2 PI 262090 \n n a n a n b n b \n × n a n a N b N b \n \n 30 0 0 Table 4 Punnett square showing offspring genotypes of selfing N b n b \n peanut plants under genomic imprinting. Male gametes \n N b \n \n \n n b \n \n Female gametes \n N b * \n \n N b \n * N b \n (Nod+) \n N b \n * n b \n (Nod–) \n n b * \n \n N b n b \n * (Nod+) \n n b \n * n b \n (Nod–) * Indicates inhibition of the expression of AhNSP2-B07 in the ovary and the marked allele is not expressed. “+” means Nod+ phenotype; “–” means Nod– phenotype. Based on the qRT-PCR analysis, the expression of AhNSP2-B07 showed a significantly higher expression in pollen than that in ovary tissues. This result further supports that allele of AhNSP2-B07 in female gametes is inhibited, which is likely due to genomic imprinting. In very rare cases, the dominant N b \n allele was not completely inhibited in the heterozygous N b n b \n plants, resulting in few unusual big nodules being produced. The heterozygous plants with few big nodules were not sampled and focused as a separate group mainly due to the extreme low occurrence. The inhibited allele derived from female parents remains low or no expression in the offspring and only the allele from male gametes is expressed, which was validated in rhizobial infected roots, where any allele of AhNSP2-B07 gene inherited from male parents is actively expressing. Bisulfite sequencing reveals no significant difference in methylation between ovaries and pollens Gene imprinting can be caused by methylation and the process starts in gametes, where the allele is imprinted with low or no expression and subsequently remains inactive in the embryo. We compared the methylation levels of the N b \n gene spanning a 1690bp promoter region and the full CDS region between the gametic tissues, ovary and pollen. However, no significant differences were detected at all methylation contexts. This result could be due to the technical difficulties in obtaining pure male and female gametic cells or the detected regulatory region of 1690bp is not sufficient. In this study, the ovary and pollen tissues were used, which still contained a significant number of somatic cells, thus adding huge background noise to precisely detect the DNA methylation level difference between the two different gametic cells as a very small proportion of the total cells in the tissues. Therefore, a single-cell epigenomic technology to precisely reveal the methylation variation at the single-cell level might be critical to further investigate the methylation variations between gametic cells ( Luo et al., 2020 ). For example, single-microspore sequencing of maize has been performed to explore methylation reprogramming during the different developmental stages in plant sexual reproduction ( Kawashima and Berger, 2014 ). However, cultivated peanuts have much smaller volume of gametic tissues with a more complicated genome. Thus, the tissue dissection and sequencing approaches need to be specifically optimized to overcome the challenges in peanut research. Besides DNA methylation, histone methylation such as trimethylation of histone H3 lysine 27 (H3K27me3) catalyzed by Polycomb-Repressive Complex 2 (PRC2) is also associated with imprinting in plants ( Köhler and Makarevich, 2006 ; Wolff et al., 2011 ). It was reported that the imprinted paternal allele in Arabidopsis was mediated by the PcG complex consisting of histone methyltransferase, while no parental DNA methylation asymmetry was detected in the promoter of the MEDEA ( MEA ) gene ( Wöhrmann et al., 2012 ). As a future step, the histone modification of the AhNSP2-B07 gene needs to be investigated. In summary, this study revealed that the non-Mendelian inheritance of nodulation segregating at AhNSP2-B07 is due to the inhibition of its expression in female gametes, resulting in further genomic imprinting in the embryo of offspring. This study provided a rare example of a vital symbiosis gene that has parent of origin or maternal imprinting effect in polyploid crops. The monoallelic expression of N b \n in root samples explained the phenotypical distortion of Nod+: Nod– from the 3:1 ratio. Combining the reciprocal cross and qRT-PCR results using ovary and pollen, it is confirmed that N b \n was maternally imprinted. However, no significant difference of DNA methylation level of the promoter and CDS region of AhNSP2-B07 was detected between ovary and pollen, and the mechanism underlying the maternal imprinting of this gene remains unclear. Our findings facilitated the understanding of gene regulation in polyploid species, which could be helpful to unveil the evolutionary process after hybridization and polyploidization in cultivated peanuts."
} | 4,680 |
31467977 | PMC6707778 | pmc | 7,039 | {
"abstract": "Click chemistry can create freeze/heat-resistant, nonflammable, and highly robust ionic liquid–based click-ionogels.",
"conclusion": "CONCLUSION A click-reaction process that introduces click-ionogels prepared under ambient conditions in a single step has been demonstrated. Because thiol-ene click chemistry is highly efficient under mild conditions, the click-ionogels can be readily prepared by simply blending two solutions, without additional oxygen, humidity, or heating requirements. This simplicity will enable shape-design and large-scale production. The prepared click-ionogels continue to exhibit excellent mechanical properties and resilience after 10,000 fatigue cycles. Moreover, due to several unique properties of ILs and the formation of a hydrogen bond between IL-BF 4 and polymer network, the click-ionogels show high electrical conductivity, broad thermal compatibility (−75° to 340°C), good optical transparency, and nonflammability. Because of these properties, these click-ionogels have promising applications in various electrical devices, including flexible sensors, energy storage, and electronic skin.",
"introduction": "INTRODUCTION Stretchable conductive materials can be used as stretchable components in resistive circuits or capacitive elements and have attracted extensive attention ( 1 , 2 ) because of their potential uses in flexible energy storage devices, actuators, and sensors ( 3 , 4 ). Extant stretchable conductive materials include conducting polymers ( 5 ), inorganic/organic compounds ( 6 ), and quasi-solid conductive gels ( 7 ). Among these materials, conductive hydrogels offer notable potential because of their tunable mechanical properties, relatively high electrical conductivity, and low interfacial resistance ( 8 – 10 ). Hydrogels with outstanding strength and broad thermal compatibility are desirable for practical applications, and great progress in preparing tough hydrogels using a variety of strategies has occurred in the past decade ( 11 – 14 ). Among them, double-network (DN) hydrogels, which comprise a brittle initial sacrificial network and a tough and covalently cross-linked second network, show extraordinarily high elasticity and toughness ( 15 , 16 ). Li et al. reported the preparation of alginate-polyacrylamide–based DN hydrogels, consisting of both ionic and covalently cross-linked polymeric networks. The prepared DN gel exhibited high elastic modulus and fracture energy ( 17 ). However, covalently cross-linked networks are typically obtained using free radical polymerization of vinyl monomers, which requires heating or ultraviolet (UV) light and anaerobic conditions ( 18 ). Moreover, Kamio et al . reported that most DN gels are prepared using a two-step process, which forms inhomogeneous spatial network clusters within their respective parent-gels, limiting the improvement of mechanical properties ( 19 ). Li et al . further pointed out that it is conducive to form a more homogeneous sacrificial network one-step process via controllable gelation, which improves the resilience and fracture energy of DN gels ( 20 ). Therefore, it is desirable to develop a one-step, controllable polymerization method to prepare DN gels under ambient conditions. In addition, most previously reported hydrogels freeze at subzero temperatures and dehydrate at high temperatures, and each of these behaviors severely limits their practical applications ( 21 , 22 ). The use of an organic solvent system has enabled the fabrication of anti-freezing and nondrying hydrogels for long-term applications ( 23 ). However, the introduction of flammable organic solvent will bring potential fire and explosion hazards, especially at high temperatures. Therefore, freeze- and heat-resistant safe gels are desirable for practical applications. Ionic liquids (ILs) are composed of cation-anion pairings of an organic ion and inorganic counterion ( 24 – 26 ). They have attracted a great deal of interest in academic research because of their unique physicochemical properties, such as high ionic conductivity, nonflammability, and thermal, electrochemical, and chemical stabilities ( 27 , 28 ). Moreover, their freezing points can be tuned by adjusting the respective structures of these paired cations and anions. Poly(ionic liquid)s (PILs) are polymers whose backbones comprise IL monomers (repeat units) ( 19 , 29 ). PILs combine polymer chain durability with small-molecule IL functionality and have received a considerable amount of attention in polymer and materials sciences. Here, based on the concept of DN, we prepared IL click-ionogels under mild conditions. Ionic interactions between poly(1-butyl-3-vinyl imidazolium fluoborate) (PIL-BF 4 ) and benzene tetracarboxylic acid (BTCA) were selected to form a sacrificial network. Due to high reactivity, mild reaction conditions, and good selectivity, we demonstrate click chemistry to form a covalent network. Because of different gelation mechanisms, our synthesis of an ionic cross-linked network and a thiol-ene click network can be conducted simultaneously in a one-pot reaction scheme. These click-ionogels exhibit excellent mechanical properties and resilience even after 10,000 fatigue cycles. Moreover, due to several unique properties of ILs, these click-ionogels show excellent performance over a wide temperature range (−75° to 340°C), high ionic conductivity, transparency, and nonflammability. These attractive features suggest that click-ionogels will make good candidates for safe stretchable conductive materials.",
"discussion": "RESULTS AND DISCUSSION Preparation of click-ionogels Figure 1 shows our preferred preparation method of click-ionogels based on a thiol-ene click reaction. Because many compounds are poorly soluble in ILs, methanol was used as the first solvent in this process. First, poly(ethylene glycol) diacrylate (PEGDA), pentaerythritol tetraacrylate (PETA; cross-linker for covalent network), and anionic BTCA (cross-linker for the ionic bond network) were dissolved in methanol to obtain a homogeneous and transparent solution (solution A). Similarly, a mixture containing PIL-BF 4 , triethylamine (TEA; a catalyst for the thiol-ene click reaction), and 1,2-ethanedithiol (ED) in methanol (solution B) was also prepared. When solutions A and B were mixed at room temperature, a gel formed, which was preconditioned by soaking in a room temperature IL at 80°C under vacuum for 24 hours. Here, a room temperature IL, 1-propyl-3-methylimidazolium fluoborate [IL-BF 4 ; melting point (m.p.) = −79°C, boiling point (b.p.) > 300°C], was selected for the preparation of ionogels (table S1). The methanol (b.p. = 65.4°C) in the gel was removed (evaporated) and replaced by IL-BF 4 to form a tough and transparent click-ionogel (fig. S1). Varying the catalyst amount allowed the speed of the thiol-ene reaction to be adjusted from 240 min to 40 s, as shown in table S2. A very fast reaction speed inhibits any formation of a uniformly distributed ion-paired network and ultimately results in deteriorated mechanical properties. Hence, minimal amounts of catalysts ( m TEA : m ED = 0.1) were used to control the click reaction rate and covalent network gelation time. Fig. 1 Preparation of click-ionogels. ( A ) Composition of solutions A and B. ( B ) Preparation process and structural characterization of the click-ionogels. ( C ) Photographs showing that the click-ionogels can be molded in various shapes, including a circle (undyed), butterfly and leaves (dyed with methyl orange), and fish (dyed with methylene blue). Photo credit: Yongyuan Ren, Soochow University. The TEA-catalyzed thiol-ene Michael addition reaction was monitored via Fourier transform infrared (FT-IR) spectroscopy and solid-state nuclear magnetic resonance (SSNMR). As shown in fig. S7, the peak at 1630 cm −1 corresponding to the C═C bonds of PEGDA nearly disappeared after the click reaction ( 30 ). Similar results were observed in the SSNMR spectra. The peak of the C═C bonds disappeared after gelation, while the S─CH 2 bond at around 2.5 parts per million (ppm) was observed (fig. S8). These results indicate that these click-ionogels had successfully formed. Considering that both FT-IR and SSNMR spectra are qualitative tests, a more accurate reaction conversion is further calculated on the basis of Wijs titration, following the Association of Official Analytical Chemists (AOAC) methods. As shown in fig. S9, iodine chloride was used for C═C bond saturation analysis and the consumed iodine was measured by titration with 0.1 M standard sodium thiosulfate solution. On the basis of the titration results and formulas, the conversion and cross-linking density (Mc) can be determined to be 99.3% and 6940 g mol −1 , respectively. These mild click gelation conditions give our click-ionogels outstanding processability and wide shape designability. Figure 1C shows that the click-ionogels can be readily designed to form different shapes, such as circles, butterfly, fish, flowers, and leaves, illustrating this latitude, and simple processing consisting of blending solutions A and B at room temperature. Mechanical properties of click-ionogels To meet the higher requirements of flexible devices, it is not only important but also challenging to accommodate large and reversible deformations of conductive gels. Superior mechanical properties of DN hydrogels have been attributed to both ionic bonds and covalent bond cross-linking that form such hybrid networks. To determine the role of each component in click-ionogels, ionic cross-linked gels (IC gels) and covalently cross-linked gels (CC gels) were prepared for comparison. As shown in Fig. 2 (B and C) , the CC gels display high tensile stress but poor compressive strain behavior, with a failure tensile strain of 83% and a failure compressive strain of 28%. This is because there was no sacrificial network present to dissipate energy during deformation. In contrast, the IC gels, which lacked covalent cross-linking, showed soft and unrecoverable deformation with a failure tensile stress of 0.2 MPa and a failure compressive stress of 0.08 MPa. Click-ionogels combine the advantages of both IC and CC gels and exhibit enhanced mechanical properties with a failure tensile stress of 2.28 MPa at a strain of 1390% and a failure compressive stress of 23.7 MPa at a strain of 92%. Fig. 2 Mechanical properties of click-ionogels. ( A ) Photographs demonstrating stretching and compression of click-ionogels. Mechanical properties of click-ionogel, IC gel, and CC gel; ( B ) tensile stress strain; ( C ) compressive stress-strain; and ( D and E ) cyclic tensile stress-strain fatigue tests. Photo credit: Yongyuan Ren, Soochow University. It should be noted that an ionic cross-linking network can be formed even if IL-BF 4 is used as a solvent. Although the storage modulus (G′) and loss modulus (G″) of IC gels decreased slightly after replacing methanol with IL-BF 4 , G′ is dominant over G″ from 0.1 to 10 Hz, suggesting that the gelled state is maintained (fig. S10). Moreover, the interaction energy between polyacid-based BTCA and PIL-BF 4 (or IL-BF 4 ) was calculated via density functional theory (DFT), which showed that the interaction energy between BTCA and PIL-BF 4 is higher than that between BTCA and IL-BF 4 (fig. S11). The higher BTCA/PIL-BF 4 interaction prevents IL-BF 4 from destroying the ionic cross-linking network via robust quadruple ionic cross-linking. In addition, the optimal mechanical properties of ionogels were attained by tuning the contents of the IC network and the covalent cross-linking density of the CC network. Because of the existence of sacrificial bonds in the IC network, the stretchability and toughness of ionogels increase with the increase of the IC network content. However, an excessive amount of BTCA cannot be fully dissolved in methanol, which seriously diminished the mechanical properties and transparency of ionogels. Moreover, the covalent cross-linking density of the CC network, the molar ratio of PETA (cross-linker) to PEGDA (monomer), also significantly affected the mechanical properties of ionogels. As shown in table S3, smaller or higher cross-linking density is not conducive to improving the mechanical properties of the ionogels. When the molar ratio of PETA to PEGDA is lowered to 0.05, the obtained ionogels are soft and show poor fracture stress, while the ionogels become brittle at a molar ratio exceeding 0.2. The ionogels with optimal mechanical properties can be obtained at the molar ratio of 0.1. The IC network content is 78 mg/ml. The resilience and cycling stability of these gels highly determine the durability of flexible devices. Here, the cyclic compressive tests of the click-ionogels showed that these gels are compressed with 70% strain without fracture and then recover nearly their original state after the force is removed ( Fig. 2, A and D ). Moreover, a small residual strain (approximately 3%) is observed, indicating the occurrence of an internal rupture of the IC network in ionogels. The IC network plays an indispensable role in improving resilience and durability of ionogels. When the ionogels undergo deformation, the ionic bonds between PIL-BF 4 and BTCA dissipate energy and quickly reform during unloading, which avoids the damage to the ionogel caused by external forces. Because of this coupled energy dissipation system, these click-ionogels also show excellent resilience even after being subjected to 10,000 compressive fatigue cycles ( Fig. 2E ). Anti-freezing properties of click-ionogels Considering that all solvents have a freezing point, the elasticity of the gels will inevitably be influenced by temperature. Hydrogels tend to lose their elasticity at subzero temperatures since water freezes, which greatly limits their practical application at subzero temperatures. In this work, gels were formulated with low–freezing point IL-BF 4 as an antifreeze solution, and our click-ionogels showed an expected stable mechanical strength and elasticity at subzero temperatures. This behavior is illustrated in Fig. 3A and movie S1, where the click-ionogels can be stretched over the surface of liquid nitrogen (about −50°C). Figure 3 (B and C) shows that when temperature decreased below −40°C, the click-ionogels still had more than 1000% stretching elasticity and 85% compressive strain. Figure 3D shows the storage moduli (G′) and loss moduli (G″) of these click-ionogels and hydrogels. When hydrogel temperature is decreased to the range of 0° to −30°C, its storage modulus and loss modulus abruptly show an approximately 1000-fold increase, which indicates that the hydrogel froze. In contrast, the click-ionogels exhibited stable storage and loss moduli from −70° to 200°C because IL-BF 4 has a low freezing point. Fig. 3 Mechanical properties of anti-freezing click-ionogels at low temperatures. ( A ) Photographs of a click-ionogel stretched above liquid nitrogen (about −50°C). ( B ) Tensile stress-strain and ( C ) compressive stress-strain curves for the click-ionogels from −40° to 0°C. ( D ) Storage moduli (G′) and loss moduli (G″) of hydrogels from −30°C to 100° and the click-ionogels from −80° to 180°C. ( E ) Dynamic scanning calorimetry (DSC) results of neat IL-BF 4 , dry polymer (drying the click-ionogel when the solvent is methanol), click-ionogel, and hydrogel between −100° and 20°C. The inset is the spectra enlargement of the dry polymer and click-ionogels. ( F ) Schematic hydrogen bonds between IL-BF 4 and PEGDA segments. ( G ) 1 H NMR spectra of IL-BF 4 , PEGDA, and IL-BF 4 /PEGDA. Photo credit: Yongyuan Ren, Soochow University. To more accurately determine thermal limits of low temperature freezing in these click-ionogels, differential scanning calorimetry (DSC) was done from −100° to −20°C. For (water-based) hydrogels, a sharp peak was observed at 0°C, the melting point of bulk water. In contrast, these DSC results show that neat IL-BF 4 has a freezing point of −79°C, which is nearly consistent with that reported previously ( 31 ), and a glass transition temperature ( T g ) of the dry (un-solvated) polymer occurs at −62°C. The thermogram of IL-BF 4 in the click-ionogels does not display a freezing peak and only showed a peak similar to a glass transition at −75°C. This disappearance of a freezing point is consistent with IL-BF 4 being an effective solvent for similarly structured units in the PIL-BTCA network, which prevents any formation of a crystalline IL-BF 4 phase. In addition, a confinement effect on the IL in polymer networks has been observed in IL/poly(propylene oxide) composites and explained by a Gibbs-Thomson model and hydrogen-bonding interactions ( 32 , 33 ). Because of their high electropositivity, the critical “active” hydrogens in the imidazolium cation (e.g., sites C-2, C-4, and C-5) can serve as hydrogen bond donors, while the electronegative oxygens in the PEGDA segments act as acceptors ( Fig. 3F ). As shown in Fig. 3G , the intensity of active hydrogens on imidazole decreases with the addition of PEGDA, and a new C─O ... H bond peak is observed at about 3.1 ppm, indicating the formation of hydrogen bonds between IL-BF 4 and PEGDA segments. Because of the interaction, IL-BF 4 tends to form composite aggregates within a polymer network, which causes a transition temperature of the click-ionogel to lie between the freezing point of IL-BF 4 and the T g of the polymer network. Therefore, our click-ionogels have not only good elasticity and high toughness but also outstanding anti-freezing properties down to −75°C. Thermal stability of click-ionogels In addition to their anti-freezing properties, another important property of gels is their long-term thermal stability. Because the water in typical water-based hydrogels continuously evaporates, even at room temperature ( 22 ), their solvation under ambient conditions is a limiting design feature. The very low partial vapor pressure of IL-BF 4 allows these click-ionogels to exhibit long-term stability at temperatures above 100°C. With a further increase in temperature, the elastic modulus of these click-ionogels shows a slight decrease because the polymer chains have more mobility. However, these click-ionogels still have notable toughness and strength, with more than 1.5-MPa tensile strength and 15-MPa compressive strength, as shown in Fig. 4 (A and B) at 120°C. Figure 4C shows a long-term thermal stability of a hydrogel and a click-ionogels at 50° and 250°C, respectively. This hydrogel lost about 75 weight % (wt %) within half an hour at 50°C. In contrast, even after 5 days at 250°C, the weight of this click-ionogel remained constant, which indicates that click-ionogels have excellent high-temperature stability. The excellent mechanical properties of ionogels at high temperature come not only from the high–boiling point IL-BF 4 but also from a very strong interaction force between ILs and polymer chains. High temperature stimulates and activates random thermal motions of solvent molecules in the gel. In this case, the solvent will leach out of the polymer network and ultimately affect the mechanical properties of the gel, if there is a lack of interaction between the solvent molecule and the polymer chain. In addition, IL-BF 4 used in this work can form strong multiple hydrogen bonds with the polymer chains ( Fig. 3G ), providing the high-temperature thermal stability of the ionogels. Thermogravimetric analysis was used to further study the thermal stability of the click-ionogels. As shown in Fig. 4D , these click-ionogels have very high decomposition temperatures, up to 343°C in air, which indicates that these click-ionogels could be used in service temperature applications greater than 250°C. On the basis of the results discussed above, it can be concluded that the prepared click-ionogels exhibit excellent performance over a wide temperature range (from −75° to 340°C). Compared with the reported freeze/heat-resistant gels (listed in table S4), it is clear that our click-ionogel is a highly competitive candidate for freeze/heat-resistant conductive material. Fig. 4 The mechanical performances of thermally stable click-ionogels. ( A ) Tensile stress and ( B ) compressive stress-strain curves for click-ionogels from 30° to 120°C. ( C ) Stability of the hydrogel and click-ionogels under long-term exposure to air at 50° and 250°C, respectively. ( D ) Thermal decomposition curves of the hydrogel and click-ionogels from 30° to 800°C. In addition to this high-temperature stability, the nonflammability of IL/PIL further ensures thermal safety features of click-ionogels in practical applications. Flammable organogels [such as propylene carbonate (PC)– and ethylene glycol (EG)–based gels] are often used as electrolytes in electronic devices, which brings potential safety hazards such as fire and explosion. As shown in fig. S12 and movie S2, the PC- and EG-based organogels were ignited in only a few seconds, followed by a slight burst and a lot of smoke. However, even after being exposed to flames for 30 s, these click-ionogels could not be ignited, and only some charring was observed, likely due to the decomposition of the click-ionogels at high temperatures. Therefore, these heat-resistant and nonflammable click-ionogels show both excellent mechanical performance and flame retardancy over a wide temperature range. Applications in triboelectric nanogenerator devices In addition to having excellent mechanical properties at both low and high temperatures, it is important in some applications for these click-ionogels to have high electrical conductivities. Here, the conductivity [σ = L /( SR ); L , S , and R correspond to the thickness, effective overlap area, and resistance of click-ionogels, respectively] of these click-ionogels was investigated between −60° and 200°C. As shown in fig. S13, abundant pores were formed after drying-out treatment, while the pores disappeared completely after soaking in IL-BF 4 . A porous structure facilitates the filling of more IL-BF 4 in polymer networks, and the smooth surface of ionogel indicates that IL-BF 4 has a good affinity with polymer networks. Both of them are conducive for improving the conductivity of ionogels. As shown in Fig. 5A , the conductivity of the click-ionogels designed in this study is about 0.83 S m −1 at room temperature. This conductivity increases as temperature increases because the viscosity of ILs decreases at higher temperatures (163 cP at 10°C and 24 cP at 70°C), which leads to more rapid ion transport and higher conductivity. Fig. 5 The electrochemical performances of click-ionogels. ( A ) Conductivity of click-ionogels from −60 to 200°C. ( B ) Decomposition voltage of the hydrogel and click-ionogels. ( C ) Scheme of TENG mechanism. The pictures and output currents of TENG under extreme conditions, such as ( D and F ) strain, ( E and G ) bent, and ( H ) low temperature and high temperature. Photo credit: Yongyuan Ren, Soochow University. In addition to electrical conductivity, it is also important for gels to have a stable and wide electrochemical window. For example, aqueous supercapacitors and zinc-ion batteries have relatively small electrochemical windows (usually <1.0 V), which severely limits their output voltage and energy density. To measure a decomposition voltage of these click-ionogels, linear sweep voltammetry was performed at a scan rate of 0.5 mV/s from 0 to 5.0 V. Figure 5B shows that the measured hydrogel current increased sharply when an applied voltage exceeds ~1.5 V, accompanied by bubble generation on the interfaces of the electrodes. Such bubble formation indicates electrolysis of water in the hydrogel. In contrast, the click-ionogels exhibit a much higher decomposition voltage at about 3.5 V due to well-known high electrochemical stability of IL-BF 4 , which meets the voltage requirements for most electronic devices (e.g., lithium-sulfur batteries, zinc-ion batteries, and supercapacitors). Recently developed triboelectric nanogenerators (TENGs), which convert mechanical energy to electricity by exploiting a coupling effect of contact electrification and electrostatic induction ( Fig. 5C ), have advantages of simple structures, low cost, and many material design choices ( 34 – 37 ). To further explore an application of click-ionogels in electronic devices, a flexible TENG was prepared by using an elastomer (Very High Bond, VHB) and click-ionogels as an electrification layer and electrode, respectively ( Fig. 5C ). Because of the excellent stretchability and flexibility of click-ionogels and VHB, the prepared TENG exhibited outstanding mechanical properties when stretched and bent ( Fig. 5, D and E ) and corresponding electrical outputs were recorded when polyamides experienced contact-separation motion relative to the VHB. Energy harvesting and stability of a resulting TENG were evaluated under extreme conditions, such as large deformations, low temperature, and high temperature. An instantaneous peak value of an ac current was about 0.1 μA/cm 2 at room temperature, which is close to the value reported in a previous study that used a hydrogel as an electrode ( 38 ). Compared with the original state without strain, the output current of the TENG ( Fig. 5F ) is greatly improved from ~0.05 to ~0.2 μA after being stretched to 500%, which was attributed to an increase in the contact area for the electrification as the surface area of the device increased under tension. In contrast, because the effective contact area between polyamides and VHB did not change, the device exhibited a stable output current even when it was bent and twisted, as shown in Fig. 5G . In addition, operating temperature ranges of previously reported TENGs that use hydrogels, such as polyvinyl alcohol and alginate gels as the electrodes, are seriously limited by water freezing and evaporation. Although hydrogels are encapsulated in VHB tape to reduce water evaporation, water vapor bubbles inside the TENG have been observed at 80°C ( 2 ). Thus, temperature effects on this TENG device were evaluated. It should be noted that because the glass transition temperature and melting point can influence properties, if the temperature is either too low or too high, the VHB will harden and soften, respectively. As shown in Fig. 5H , there was no observed variation in the current between −60° and 200°C, which indicates that this device had good stability at both low and high temperatures. This can be explained by noting that a TENG is primarily a capacitive sensor that is based on electrostatic charge, and a temperature-induced resistance change of click-ionogels will not cause the current to vary much. Hence, the TENG that was prepared using click-ionogels as the electrode displayed high mechanical stability even upon being stretched, bent, and twisted, as well as a wide operating temperature range."
} | 6,743 |
31222170 | null | s2 | 7,040 | {
"abstract": "Eukaryotes evolved from a symbiosis involving alphaproteobacteria and archaea phylogenetically nested within the Asgard clade. Two recent studies explore the metabolic capabilities of Asgard lineages, supporting refined symbiotic metabolic interactions that might have operated at the dawn of eukaryogenesis."
} | 77 |
36895758 | PMC9988672 | pmc | 7,041 | {
"abstract": "The data output from microbiome research is growing at an accelerating rate, yet mining the data quickly and efficiently remains difficult. There is still a lack of an effective data structure to represent and manage data, as well as flexible and composable analysis methods. In response to these two issues, we designed and developed the MicrobiotaProcess package. It provides a comprehensive data structure, MPSE , to better integrate the primary and intermediate data, which improves the integration and exploration of the downstream data. Around this data structure, the downstream analysis tasks are decomposed and a set of functions are designed under a tidy framework. These functions independently perform simple tasks and can be combined to perform complex tasks. This gives users the ability to explore data, conduct personalized analyses, and develop analysis workflows. Moreover, MicrobiotaProcess can interoperate with other packages in the R community, which further expands its analytical capabilities. This article demonstrates the MicrobiotaProcess for analyzing microbiome data as well as other ecological data through several examples. It connects upstream data, provides flexible downstream analysis components, and provides visualization methods to assist in presenting and interpreting results.",
"introduction": "Introduction A wide array of important roles of the microbiota in diverse environments have been investigated and explored substantially, 1 , 2 largely because of the development of high-throughput sequencing technologies and bioinformatics. During the last decades, many bioinformatics algorithms and tools for the exploration and analysis of microbiome data have been built in the scientific community, such as qiime2 , 3 \n dada2 , 4 \n usearch , 5 \n mothur , 6 \n MetaPhlAn . 7 , 8 These pipelines or tools conduct initial bioinformatics analysis of microbiome data, but the relevant data to a microbiome experiment became heterogeneous consisting of feature-oriented (operational taxonomic unit [OTU] or amplicon sequence variants [ASV]) data frames, sample-oriented data frames, the representative sequences, and the phylogenetic tree of the sequences after the initial analysis, which brings new challenges for the downstream data analysis and reproducibility. There are often many important links (such as the features of the feature table and the nodes of the phylogenetic tree, feature table, metadata, etc.) among the diverse data that need to be preserved throughout an analysis. If a researcher failed to integrate these data comprehensively, the information might be lost and thus generate errors. In addition, many intermediate steps that may confuse may be repeatedly performed in the downstream statistical analysis. For instance, the dissimilarity indices (such as Bray-Curtis , (Un)Weighted Unifrac , Jaccard , etc.) for the communities of microbes might need to be calculated and reused in hierarchical cluster analysis, principal coordinate analysis ( PCoA ), and permutational multivariate analysis of variance ( PERMANOVA ), and so on. If the intermediate data can be effectively integrated and stored, it will improve the efficiency of analysis, enhance reproducibility and avoid errors. 9 The R programming language has become one of the most popular tools for biomedical data analysis. 10 Some efforts have been made to build common representations and infrastructures for complex, highly interdependent datasets in R. 10 , 11 For example, SummarizedExperiment 12 class is widely used to integrate matrix-like objects of feature abundance, the normalized feature abundance, a sample- and feature-oriented metadata data frames as a standardized data structure across many Bioconductor 13 packages. To integrate the phylogenetic tree structure and heterogeneous associated data, we defined the treedata 14 , 15 class, which has also been widely used in several packages, such as tidytree , 14 \n treeio , 15 and ggtree . 16 However, these data structures did not cover all the heterogeneous data of a microbiome experiment, and the existing tools 17 , 18 , 19 , 20 , 21 , 22 for the downstream statistical analysis of the microbiome also did not well integrate the primary heterogeneous data and the useful intermediate data. For instance, the phyloseq 17 class is used in the phyloseq , 17 \n microViz , 19 and MicrobiomeAnalyst 22 packages, but the data structure can only store the primary input datasets; it cannot integrate the normalized data and the intermediate data such as the alpha diversity, dis-similarity indices, the result of differential analysis, and so on. The animalcules package was developed using the MultiAssayExperiment 23 class, and it cannot integrate intermediate data and the phylogenetic tree, which is often needed in the calculation of specified dissimilarity (( Un ) Weighed UniFrac ). These data structures for the microbiome cannot take into account the diversified needs of downstream analysis, which makes it more convenient for some specific needs, while other needs may be troublesome. Moreover, the differences in these data structures are too large, which restricts downstream integrated analysis. Aside from the data structure defects, downstream statistical, visual, and functional analyses of microbiome data are complex because the appropriate analysis workflow often needs to be explored and adjusted according to the research design (such as different sequencing methods: 16S, metagenome, or metatranscriptome) and the different statistical methods, which often are developed in different platforms or packages across various programming syntax and environments. There remains a lack of flexible and comprehensive packages that can streamline the personalized analysis of microbiome data with a unified and user-friendly syntax. Recently, the tidy concept was proposed and has gained popularity in the R data analysis community with tidy data tools, 9 , 10 such as dplyr , 24 \n tidyr , 24 and ggplot2 , 25 to allow users to freely and easily explore data and focus more on the special problems. This tidy concept has been applied to different disciplines, including genomic, 26 , 27 transcriptomic, 28 functional enrichment analysis, 29 and phylogenetic data analyses. 16 , 30 Through the human-readable data structure and analysis grammar, these tidy data tools make it easier for the community to develop modular manipulation, visualization, and analysis methods, decrease the learning curve for users, and facilitate reproducibility for the related studies. 9 , 24 , 31 However, this principle of tidiness has not been implemented in the existing tools 17 , 18 , 19 , 20 , 21 , 22 for microbiome data analysis. This factor greatly hinders the flexibility and ease of use of downstream data analysis in this discipline and also limits the possibilities for researchers to explore data and develop personalized analysis pipelines. To fill these gaps, we developed the MicrobiotaProcess package. We defined the MPSE class for storing the microbiome experiment or related ecological dataset and the related intermediate data for downstream analysis. The MPSE class defined in MicrobiotaProcess inherits both the SummarizedExperimen t 12 and the treedata 14 , 15 classes, which are popular and widely used in the Bioconductor 13 ecosystem. It takes merit from Bioconductor packages that are based on these two data structures. The importance of data structure is that it is the foundation for downstream analysis. A good data structure helps to unify the downstream analysis. When operating on certain data, the associated data in the data structure can be updated synchronously, which can decrease many errors caused by improper operations. To make downstream data analysis modular and easy to use, we introduce a user-friendly grammar (i.e., tidy interface) to process, analyze, and visualize microbiome data stored in the MPSE data structure. Additionally, we developed a differential abundance analysis method for finding prospective biomarkers with a better false-positive rate based on analysis results of real and simulated datasets (Supplemental file B). All the functions are developed under the tidy framework, which allows users to build human-readable and flexible analysis workflows. We believe it can remove a major obstacle for scientists to explore and analyze the microbiome (16S, metagenome, and metatranscriptome) and other ecological data.",
"discussion": "Discussion and conclusion Microbiomics technologies have become increasingly popular methods for exploring the relationship between microbial communities and the host or the environment (e.g., intestine, skin, soil, and the ocean). 1 , 50 , 60 , 61 , 62 , 63 , 64 Data analysis is still one of the bottlenecks in this field, especially in downstream analysis, the integration of heterogeneous data and the need for personalized analysis have brought new challenges. MicrobiotaProcess provides a comprehensive data structure, the MPSE class, to store heterogeneous microbiome data, including feature (e.g., species, OTUs) abundance, sample data (e.g., clinical information), and feature data (e.g., taxonomic relationships, functional profiles). Moreover, simultaneously with the analysis, intermediate results can be stored in an MPSE object. For example, the normalized or rarefied data, the diverse dissimilarity metrics, and the results of the differential analysis can be integrated with the microbiome data into an MPSE object. These intermediate results can be further processed, visualized, and reused. This can prevent repeated calculation, facilitate data sharing, and enhance analytic reproducibility. MicrobiotaProcess implements a set of functions within the tidy principles to unveil the characteristics and biological function of microbial communities in a diverse environment. The returned values of these functions are predictable and consistent. Each function is designed to accomplish a simple task and these functions can be combined to accomplish complex tasks. In this way, it has better flexibility, and the development of workflow through function series can meet most of the personalized analysis needs. Considering the interoperation between MicriobiotaProcess and other tools, some functions developed by other software can be applied to MPSE objects, making the downstream analysis function more comprehensive and having wider application scenarios. For example, differential analysis using tidybulk and functional analysis using clusterProfiler can be integrated into the analysis of the microbiome and other ecological data through MPSE objects. MicrobiotaProcess provides several functions to parse outputs obtained from upstream tools and allows converting commonly used objects to MPSE objects ( Figure 2 A). This enables the functionalities provided by MicrobiotaProcess to be well connected to upstream analysis. The downstream analysis results need to be interpreted through visualization. MicrobiotaProcess provides several functions for visualization. More important, the tidy interface extracts relevant data and connects to the visualization functions developed by the R community, allowing users to have more options for visual exploration and presentation of data. For example, differential analysis results can be extracted by mp_extract_tree , followed by a visual exploration of the result via the ggtree package suit ( Figures 4 A and 6 ). Although MicrobiotaProcess was mainly designed for analyzing microbiome data, some of its methods, such as alpha, beta diversity, (partial) constrained correspondence analysis ( pCCA or CCA ), and redundancy analysis ( RDA ), 65 are also applicable to other ecological data ( Figures 5 C and 5D). In addition, MicrobiotaProcess implements several methods from scratch, including the mp_cal_pd_metric for calculating several phylogenetic diversity metrics by combining the phylogenetic tree and the species abundance of a community ( Figure 3 B), the mp_diff_analysis for identifying potential biomarkers with better control of the false-positive rate ( Figures SB.3–SB.6 ), and a phylogenetic transform function ( mp_balance_clade ) for identifying differential clades of related bacteria at different resolutions on the phylogenetic tree ( Figures 4 A and 4C). Using differential clades can improve the performance of the supervised classification model ( Figures 4 D and SB.1 ), which can be used for disease diagnosis. Methods of microbiome data analysis are rapidly evolving. Some new methods not in the MicrobiotaProcess package will be developed by other software, such as the generalized Lotka-Volterra model for the analysis of microbial interactions, 66 \n LinDA , and ZicoSeq for the differential abundance analysis of microbiome data. Through the left_join method, users can integrate the analysis results of these new methods into an MPSE object, and the results can be explored, visualized, and further analyzed using the functions implemented in MicrobiotaProcess ( Figures SA.20–SA.24 ). Compared with other related tools, MicrobiotaProcess has many unique advantages ( Figure SA.33 ), and we will develop more functions as needed in the future. In summary, we believe that MicrobiotaProcess will be a valuable resource for analyzing microbiomes and other ecological data."
} | 3,356 |
39353557 | PMC11444771 | pmc | 7,042 | {
"abstract": "Mutualistic relationships with photosynthetic organisms are common in cnidarians, which form an intracellular symbiosis with dinoflagellates in the family Symbiodiniaceae. The establishment and maintenance of these symbionts are associated with the suppression of key host immune factors. Because of this, there are potential trade-offs between the nutrition that cnidarian hosts gain from their symbionts and their ability to successfully defend themselves from pathogens. To investigate these potential trade-offs, we utilized the facultatively symbiotic polyps of the upside-down jellyfish Cassiopea xamachana and exposed aposymbiotic and symbiotic polyps to the pathogen Serratia marcescens . Symbiotic polyps had a lower probability of survival following S. marcescens exposure. Gene expression analyses 24 hours following pathogen exposure indicate that symbiotic animals mounted a more damaging immune response, with higher levels of inflammation and oxidative stress likely resulting in more severe disruptions to cellular homeostasis. Underlying this more damaging immune response may be differences in constitutive and pathogen-induced expression of immune transcription factors between aposymbiotic and symbiotic polyps rather than broadscale immune suppression during symbiosis. Our findings indicate that in facultatively symbiotic polyps, hosting symbionts limits C. xamachana’ s ability to survive pathogen exposure, indicating a trade-off between symbiosis and immunity that has potential implications for coral disease research.",
"conclusion": "5 . \n Conclusions Together, our data demonstrate that there is a trade-off between photosymbiosis and immunocompetence in the facultatively symbiotic polyps of C. xamachana . Underlying this trade-off may be the differential regulation of immune transcription factors both constitutively and in response to S. marcescens exposure rather than broadscale immune suppression in symbiotic animals. Symbiotic C. xamachana mount a stronger and more damaging immune response following pathogen exposure, resulting in higher levels of oxidative stress, greater disruptions to cellular homeostasis and ultimately decreased survival rates. The trade-off between symbiosis and immunity seems to be shared across independent evolutions of facultative cnidarian–algal symbiosis, as E. diaphana are also more susceptible to S. marcescens in a symbiotic state [ 25 ]. There likely are more complexities to the nutritional aspect of this trade-off, as starvation has been shown to influence cnidarian immune gene expression [ 25 , 90 ]. With the expanding threat of disease to coral reef ecosystems, this trade-off could be a major factor in coral disease susceptibility and dysbiosis via bleaching or nutrient pollution. The cost of hosting symbionts should be investigated further and potentially incorporated into the paradigms of coral disease research [ 21 , 91 ].",
"introduction": "1 . \n Introduction Throughout the metazoan phylogeny, several taxa have evolved photosymbiosis or mutualistic relationships with photosynthetic organisms [ 1 ]. In these mutualisms, the symbionts provide their hosts with photosynthates, which often account for the bulk of the hosts’ nutrition, in exchange for nutrients such as nitrogen and the protection of being housed within the host [ 2 , 3 ]. Photosymbiosis is common throughout the cnidarian phylogeny, with the vast majority of symbiotic cnidarians forming an intracellular symbiosis with dinoflagellates in the family Symbiodiniaceae [ 2 , 4 ]. This symbiosis is best known for its vital role in coral reef ecosystems, as the nutrition provided by the symbionts allows the cnidarian hosts to live in oligotrophic environments that would otherwise be uninhabitable for them if they relied upon heterotrophy alone [ 5 , 6 ]. The extent to which the cnidarian host is reliant upon their symbionts for nutrition varies. This symbiosis can be obligate, as seen in tropical reef-building corals, facultative, as seen in some anemones and soft corals, or both depending on the life stage, as seen in the scyphozoan genus Cassiopea [ 5 , 7 – 9 ]. While all algal symbionts are housed intracellularly in a specialized acidic organelle called the symbiosome, the cell type in which the symbionts reside is variable [ 2 , 7 , 8 ]. This is likely owing to the complex evolutionary history of the cnidarian–algal symbiosis, which has independently evolved several times [ 4 , 9 , 10 ]. Members of the classes Hexacorallia and Octocorallia, which account for the vast majority of symbiotic cnidarians, house their symbionts in the gastrodermis [ 2 , 7 ]. This is in contrast to the more distantly related scyphozoans whose symbionts are housed in mobile cells called amoebocytes within the mesoglea [ 8 ]. The complete mechanisms of symbiosis establishment and maintenance within cnidarians are still unknown and may vary across the independently evolved symbioses [ 7 , 11 ]. Recognition of the algal symbionts likely occurs via pattern recognition receptors (PRRs), though many different classes of these receptors have been implicated [ 11 ]. However, there is evidence in soft and hard corals that lectins opsonize the symbionts prior to phagocytosis [ 7 , 12 – 14 ]. Following phagocytosis, non-compatible symbionts are expelled from the symbiont-hosting cells via vomocytosis, while compatible symbionts are retained to establish their intracellular niche [ 15 ]. The establishment of compatible symbionts is strongly associated with the suppression of the cnidarian hosts’ innate immune system [ 11 , 16 – 18 ]. Studies indicate that this immune suppression likely occurs via the suppression of the master immune regulator and transcription factor nuclear-factor kappa B (NFκB) or through the suppression of pathways upstream of NFκB [ 15 , 19 , 20 ]. This immune suppression persists in the symbiont-hosting cells in order to retain their symbionts [ 17 – 19 ]. Because symbiotic cnidarians suppress their immune systems while maintaining populations of intracellular symbionts, there is a potential trade-off between the nutrition these animals get from their symbionts and their ability to respond to pathogens. With recent increases in the frequency and severity of coral disease outbreaks, it is pertinent to understand how hosting symbionts influences cnidarians’ ability to defend themselves against pathogens [ 21 – 24 ]. Aposymbiotic Exaiptasia diaphana have been shown to be less susceptible to Serratia marcescens infection relative to their symbiotic counterparts [ 25 ]. This species has also shown marked differences in gene expression between symbiotic and aposymbiotic animals in response to Vibrio coralliilyticus [ 26 ]. However, it has not been established whether immune suppression and the subsequent trade-off between symbiosis and immunity are shared across independent evolutions of the cnidarian–algal symbiosis, given the differences in the symbiont housing cell type. Therefore, we tested the infection outcomes and responses of aposymbiotic and symbiotic Cassiopea xamachana to the known cnidarian pathogen S. marcescens. C. xamachana are benthic jellyfish that are facultatively symbiotic in their polyp life stage [ 27 ]. This facultative symbiosis can be leveraged to disentangle the role of symbiosis in pathogen-induced stress. We found that symbiotic C. xamachana , similar to E. diaphana , are more susceptible to bacterial infection relative to their aposymbiotic counterparts. To further investigate the mechanisms of this trade-off between symbiosis and immunity, we measured their acidic organelle activity and gene expression following pathogen exposure. These data give more insights into the trade-offs between symbiosis and immunity in cnidarians by identifying core shared responses and phenomena across the independently evolved symbioses.",
"discussion": "4 . \n Discussion The lower survival of symbiotic polyps following S. marcescens exposure indicates the presence of a trade-off between the nutritional benefit of hosting symbionts and immunocompetence in facultatively symbiotic C. xamachana . We found that hosting Symbiodiniaceae alters both constitutive and pathogen-induced C. xamachana gene expression. Symbiosis in C. xamachana altered the constitutive expression of metabolism, ion transport and innate immunity. When exposed to S. marcescens , symbiotic C. xamachana upregulated a stronger ROS and immune effector response, likely disrupting protein homeostasis in the endomembrane system and leading to low survival rates. We hypothesize that differences in constitutive and pathogen-induced expression of immune transcription factors drive symbiotic polyps’ greater susceptibility to bacterial pathogens. In comparing the two control groups of our study, we found several notable differences in the constitutive expression of symbiotic and aposymbiotic C. xamachana . Several genes involved in transport and metabolism were differentially expressed in symbiotic animals relative to their aposymbiotic counterparts. Consistent with findings in the symbiotic sea anemone E. diaphana , our symbiotic controls downregulated lipid catabolism relative to their aposymbiotic counterparts. Previous studies attribute these expression changes in lipid metabolism to be indicative of increased energy stores and thus characteristic of stable symbiosis [ 49 , 50 ]. Interestingly, we found no evidence of constitutive shifts in symbiotic polyps’ expression of nitrogen metabolism in any of our analyses. This is notable because shifts in nitrogen metabolism, specifically ammonium recycling via the glutamine synthetase/glutamate synthase system, are a well-established occurrence in symbiotic anthozoans and thought to be a mechanism of symbiont population control [ 9 , 12 , 17 , 51 – 54 ]. Isotopic studies have found that while C. xamachana’ s symbionts have limited access to nitrate in hospite , they retain access to ammonium derived from heterotrophic feeding [ 55 , 56 ]. Our results support this finding, as GLUL, a gene responsible for the removal of ammonium by converting it to glutamine, is constitutively downregulated in symbiotic polyps. This same gene is upregulated in symbiotic E. diaphana and overexpressed in the symbiont-hosting cells of Stylophora pistillata and Xenia sp. [ 9 , 12 , 51 , 57 ]. The lack of transcriptomic signatures of constitutive changes to nitrogen metabolism in symbiotic C. xamachana polyps could be owing to their different symbiont-hosting cell type and their potential use of their bacterial microbiome to limit symbiont access to dissolved inorganic nitrogen [ 55 , 56 ]. We found extensive evidence for the constitutive differential regulation of ion transport while in a symbiotic state, with both the ranked GO enrichment analysis between control groups and the W GCNA module positively correlated to symbiosis having enrichments involving the nervous system and transmembrane ion transport. Similar expression patterns have been observed in E. diaphana , with many of the upregulated genes associated with ion transport (KCNA2, GAABRR2, CNTNAP4 in our study) functioning to decrease membrane excitability [ 51 , 58 – 60 ]. Several intracellular protozoan parasites cause similar changes to their host’s ion transport, often to prevent the host cell from producing nitric oxide [ 61 – 63 ]. Symbiodiniaceae may employ similar strategies while residing intracellularly, as high levels of host-derived nitric oxide have been shown to lead to symbiosis breakdown [ 64 , 65 ]. However, nitric oxide synthase is upregulated in symbiotic controls relative to aposymbiotic controls, so the changes in expression of these ion transporters may serve a different function in the maintenance of symbionts. Our data do not suggest that symbiotic animals have large-scale downregulation of immunity, as there were no GO terms related to immunity that were significantly differentially expressed between the two control groups. This, along with the PCA showing symbiotic controls grouping with the aposymbiotic pathogen-exposed replicates across PC1, which is enriched for genes with innate immune response GO annotations, indicates that broadscale immune suppression owing to hosting symbionts is unlikely to be driving differences in survival rates between symbiotic states as has been hypothesized in other symbiotic cnidarians [ 11 , 16 ]. However, we did find several immune transcription factors with significantly different constitutive expression between the symbiotic states. Two of these transcription factors, NFκB and IRF4 paralog 2, have significantly lower expression in symbiotic controls relative to aposymbiotic controls. This NFκB expression pattern has been found in symbiotic anemones [ 19 , 20 ]. Additionally, there is evidence that stony corals negatively regulate NFκB pathways while hosting symbionts, indicating that NFκB downregulation is a common characteristic of the cnidarian–algal symbiosis across independent evolutions of the trait [ 17 , 18 ]. The other two immune transcription factors differentially regulated between the control groups are IRF1 and IRF4 paralog 1, which have significantly higher expression in symbiotic controls relative to aposymbiotic controls. Interferon regulatory factors have yet to be implicated in cnidarian symbiosis, but similar to NFκB these immune transcription factors are capable of mounting pro-inflammatory immune responses [ 66 , 67 ]. IRF expression can be both constitutive, providing basal levels of defence against microbes, and inducible, following danger-associated molecular pattern recognition [ 67 – 69 ]. Our data indicate that S. marcescens exposure can induce IRF1 and IRF4 paralog 1 expression, but not IRF4 paralog 2. As such, aposymbiotic animals are only constitutively upregulating a single immune transcription factor with S. marcescens -induced expression (NFκB), whereas symbiotic animals are constitutively upregulating two (IRF1 and IRF4 paralog 1). The differential regulation of these transcription factors may underlie why symbiotic polyps are less likely to survive S. marcescens exposure. The pathogen response of symbiotic C. xamachana shared some similarities with aposymbiotic C. xamachana . Regardless of symbiotic state, C. xamachana polyps upregulated GO terms indicative of the secretion of immune effector proteins. One such immune effector was MPEG1, a bactericidal protein [ 70 , 71 ]. However, many of these same GO terms are significantly upregulated in symbiotic pathogen-exposed polyps relative to aposymbiotic pathogen-exposed polyps, indicating that symbiotic animals are likely mounting a stronger immune effector response than their aposymbiotic counterparts. This could be owing to the utilization of different pro-inflammatory immune pathways in the aposymbiotic and symbiotic pathogen responses. Both aposymbiotic and symbiotic polyps are upregulating ‘response to interferon gamma’ when exposed to S. marcescens . As cnidarians lack interferons, this GO enrichment likely suggests the upregulation of IRFs and other genes under transcriptional control of IRFs [ 66 , 72 ]. This is supported by both symbiotic states upregulating IRF1 in response to the pathogen. However, as symbiotic polyps have higher baseline expression of IRF1 and IRF4 paralog 1, symbiotic pathogen-exposed polyps have significantly higher expression of these transcription factors relative to aposymbiotic pathogen-exposed polyps. This is further supported by both these transcription factors belonging to the turquoise module, which is significantly positively correlated to symbiosis and S. marcescens exposure. Additionally, symbiotic polyps are upregulating the GO term ‘regulation of IκBK/NFκB signaling’ in response to S. marcescens as well as NFκB and one of its activators, NFκB essential modulator (NEMO). However, symbiotic polyps are also upregulating IκBKB, the inhibitor of NFκB, in response to S. marcescens . As NFκB is often a component of the initial immune signalling following the introduction of a stressor, this could suggest that at 24 h following S. marcescens exposure, symbiotic polyps are transitioning to different immune stress response pathways [ 73 , 74 ]. Stronger pro-inflammatory immune signalling in symbiotic polyps relative to aposymbiotic polyps could explain why gene expression signatures of oxidative stress are higher in pathogen-exposed symbiotic polyps relative to their aposymbiotic counterparts [ 75 ]. While the pathogen response of both symbiotic states included the upregulation of the GO term ‘regulation of ROS biosynthetic process’, only the symbiotic pathogen response included the upregulation of GO terms associated with responding to ROS. Additionally, the turquoise module, which is positively correlated to both pathogen exposure and symbiosis, is enriched for oxidant production. Higher levels of oxidants in pathogen-exposed symbiotic polyps are further supported by their significantly higher expression of the antioxidant SOD1 and the oxidative stress transcription factor NRF1 relative to pathogen-exposed aposymbiotic polyps [ 76 , 77 ]. Pro-inflammatory factors can induce ROS production directly and indirectly [ 78 , 79 ]. Higher demands for immune effector secretion can result in the upregulation of oxidative phosphorylation during immune responses [ 80 ]. The upregulation of oxidative phosphorylation in symbiotic pathogen-exposed polyps relative to their aposymbiotic counterparts may be another contributor to the higher transcriptional signatures of oxidative stress in the symbiotic pathogen response. Together, the stronger upregulation of immune effector responses, oxidative stress responses and oxidative phosphorylation in symbiotic polyps indicates that their immune response to S. marcescens likely results in more severe disruptions to cellular homeostasis relative to aposymbiotic polyps [ 78 – 80 ]. Our gene expression data indicate that symbiotic polyps experience disruptions to endomembrane system homeostasis and the protein folding environment within the cell. Oxidative stress, high secretory demand and immune effector proteins are all capable of disrupting the protein folding environment within the ER and are more highly upregulated in symbiotic pathogen-exposed replicates relative to aposymbiotic pathogen-exposed replicates [ 81 – 83 ]. Symbiotic polyps upregulated both the endoplasmic-reticulum-associated protein degradation (ERAD) pathway and an unfolded protein response (UPR). These pathways are both indicative of disruptions to protein homeostasis and ER stress [ 84 ]. The ERAD pathway is responsible for removing misfolded or unfolded proteins from the ER. If these misfolded and/or unfolded proteins accumulate, they will trigger the UPR [ 84 , 85 ]. In response to the accumulation of misfolded and/or unfolded proteins, the UPR reduces protein synthesis and upregulates chaperonins to attempt to refold or repair the misfolded proteins [ 81 , 84 ]. Any proteins unable to be folded are degraded via either autophagy or the ERAD pathway [ 81 , 82 , 84 ]. If the cell is unable to correctly repair or refold its proteins or the misfolded proteins accumulate and are not able to be degraded, the UPR transitions into a cell death pathway [ 82 , 86 ]. Given that symbiotic polyps have considerably more disruptions to protein homeostasis at 24 h following S. marcescens exposure, they likely are transitioning to cell death pathways sooner than aposymbiotic polyps, resulting in lower survival. Differential expression of genes associated with autophagy is a common component of cnidarian immune responses, particularly at relatively early timepoints following pathogen exposure or in disease-resistant coral species that are not as severely impacted by a given pathogen [ 87 , 88 ]. In the context of innate immunity, autophagy is cytoprotective and anti-inflammatory, counteracting the damage that secreted inflammatory factors can cause to mitochondria and the endomembrane system [ 89 ]. Both symbiotic and aposymbiotic polyps significantly changed their regulation of autophagy and had increased acidic organelle activity following exposure to S. marcescens . However, given the signatures of higher oxidative stress and ER stress in symbiotic polyps, it is likely that upregulation of autophagy is insufficient to counteract the damage caused by their immune response, resulting in the negative regulation of autophagy and transition to cell death pathways [ 86 , 89 ]."
} | 5,143 |
21564335 | null | s2 | 7,043 | {
"abstract": "Since their inception 20 years ago, the biennial blast (Bacterial Locomotion and Signal Transduction) meetings instantly became the place to be for exchanging and sharing the latest developments in the field of bacterial motility and signalling. At the 11th edition, held last January in New Orleans, LA, researchers reported on the myriad of mechanisms involved in bacterial movement, sensing and adaptation, ranging from the molecular level to multicellular behaviour. New insights into bacterial signalling phenomena were gained, revealing previously unsuspected layers of complexity, particularly in mechanisms ensuring signal transduction fidelity and novel links to metabolic processes."
} | 173 |
38612017 | PMC11012369 | pmc | 7,044 | {
"abstract": "The microbial hybrid system modified by magnetic nanomaterials can enhance the interfacial electron transfer and energy conversion under the stimulation of a magnetic field. However, the bioelectrocatalytic performance of a hybrid system still needs to be improved, and the mechanism of magnetic field-induced bioelectrocatalytic enhancements is still unclear. In this work, γ-Fe 2 O 3 magnetic nanoparticles were coated on a Shewanella putrefaciens CN32 cell surface and followed by placing in an electromagnetic field. The results showed that the electromagnetic field can greatly boost the extracellular electron transfer, and the oxidation peak current of CN32@γ-Fe 2 O 3 increased to 2.24 times under an electromagnetic field. The enhancement mechanism is mainly due to the fact that the surface modified microorganism provides an elevated contact area for the high microbial catalytic activity of the outer cell membrane’s cytochrome, while the magnetic nanoparticles provide a networked interface between the cytoplasm and the outer membrane for boosting the fast multidimensional electron transport path in the magnetic field. This work sheds fresh scientific light on the rational design of magnetic-field-coupled electroactive microorganisms and the fundamentals of an optimal interfacial structure for a fast electron transfer process toward an efficient bioenergy conversion.",
"conclusion": "4. Conclusions In summary, a microbe hybrid system was successfully synthesized by coating γ-Fe 2 O 3 MNPs on bacterial surface, and the bioelectrocatalytic performance of the hybrid system improved under an electromagnetic field. The γ-Fe 2 O 3 MNPs improved the bacterial conductivity and served as an electron transport pathway for long-distance electron transfer, enhancing the efficiency of EET and the power generation performance of the MFC. In addition, the oxidation peak current of CN32@γ-Fe 2 O 3 increased to 2.24 times under an electromagnetic field, which, due to the electromagnetic field, can greatly boost extracellular electron transfer. The enhancement mechanism is mainly due to the fact that the surface modified microorganism offers an elevated contact area for the high microbial catalytic activity of the outer cell membrane’s cytochrome, while the magnetic nanoparticles provide a networked interface between the cytoplasm and the outer membrane for boosting a fast multidimensional electron transport path under a magnetic field. This work successfully constructed a hybrid-coupled bioelectrochemical system that synergistically promotes an efficient electron transfer of MFCs, which is of great significance for increasing clean energy production using electromagnetic fields.",
"introduction": "1. Introduction Microbial fuel cells (MFCs) are bioelectrochemical systems that convert chemical energy into electrical energy using electroactive microbes (EAMs) as catalysts [ 1 ], representing a clean energy technology [ 2 ]. Moreover, they can be used to treat waste/wastewater and have been widely investigated due to their dual efficacy [ 3 , 4 ]. EAMs can utilize an electrode as the terminal electron acceptor for reduction, with the electrode serving as either the electron donor or acceptor depending on whether it functions as the anode or cathode [ 5 , 6 ]. They possess several extracellular electron transfer (EET) strategies for anaerobic respiration, including a direct electron transfer (DET) mediated by outer membrane C-type cytochromes (OM c-Cyts) [ 7 ] and nanoconductors [ 8 , 9 ], and an indirect electron transfer (IET) mediated by exogenous or endogenous electron mediators. However, limited by the inefficient EET process and the slow transmembrane process, the power density of MFC is far from reaching the levels of industrial application [ 10 , 11 ]. Hence, there lies a pressing need to construct a simple and highly efficient approach that expedites the EET process. With the burgeoning advancement in the field of nanoscience, the use of nanomaterial for MFC modification has received widespread attention from researchers and has achieved significant results [ 12 ]. A lot of studies have proven that the implementation of functional nanomaterials can significantly reduce charge transfer resistance, leading to a notable enhancement in microbial colonization and biofilm growth [ 13 , 14 , 15 ]. Additionally, this innovative approach has demonstrated a considerable potential for improving the efficiency of electron transfer to extracellular receptors [ 16 ]. Furthermore, highly conductive nanomaterials can serve as electron shuttle channels, which can greatly improve the efficiency of an EET [ 17 ]. In these contexts, researchers have applied various advanced nano-functional materials, including metal oxides, carbon-based nanomaterials, metal-based nanomaterials, conductive polymers, and their composites, to the study of MFC [ 18 ]. Initially, researchers used a single or composite nanomaterial to modify the anode surface to increase the surface area for bacterial growth and optimize the electrode surface properties [ 19 ]. Nonetheless, a noteworthy observation has emerged, revealing that within the natural biofilm formed on the anode surface, a majority of bacteria are located at a considerable distance from the functional nanomaterial. Consequently, only the bacteria inhabiting the innermost layer of the biofilm maintain direct contact with the nanomaterials positioned on the electrode surface [ 20 ]. The efficiency of EET is improved only to a limited extent by relying on slow electron jumps in redox centers for electron transfer between bacteria [ 21 ]. To address these issues, researchers then proposed the strategy of using nanomaterials hybridized with biofilms [ 22 ]. Through a disordered mixed contact of functional nanomaterials with bacteria inside biofilms, the nanomaterials can help facilitate long-distance electron transfer, further improving the efficiency of EET [ 23 ]. However, the establishment of a tightly coupled and efficient pathway for electron transfer between EAMs and conductive non-biological surfaces continues to be elusive. To exploit the active sites on the bacterial outer membrane, researchers have proposed the use of nanomaterials to modify the surface or interior of bacteria [ 24 , 25 , 26 , 27 ]. The efficient construction of the microbial–nanomaterial interface can optimize the EET efficiency and enhance the power generation performance of MFC. As an example, the strategic utilization of the carbon particle point modification on bacterial surfaces has demonstrated its ability to enhance bacterial adhesion and facilitate the formation of biofilms. Owing to the presence of surface carbon particle points, the maximum current and power output are increased by 7.34 times each, and the EET efficiency is improved [ 28 ]. Silver nanoparticles successfully introduced into the transmembrane and outer membrane dramatically enhance the charge extraction rate of MFC, achieving a maximum current density of 5 mA/cm 2 and a power density of 0.66 mW/cm 2 [ 29 ]. The concept of a single-cell electron collector has been proposed. The team utilized dopamine in situ polymerization on the surface of S. oneidensis MR-1 cells to form a primary electron collector, followed by the further assembly of a more efficient electron collector through FeS NPs biomineralization. The remarkable electron transfer rate and electron recovery efficiency have led to a record-breaking bioelectricity generation in S. oneidensis MR-1, achieving an impressive power output of 3.21 W/m 2 [ 30 ]. Magnetic nanoparticles (MNPs), regarded as a pivotal class of functional nanomaterials, have garnered extensive attention owing to their remarkable nanoscale characteristics and distinctive magnetic properties, thus triggering substantial research endeavors. Among the MNPs, magnetic hematite (γ-Fe 2 O 3 ) is considered one of the most ideal materials for various applications due to its inherent biocompatibility, oxidative stability, high surface area, and good magnetism [ 31 ]. Also, Shewanella belongs to the group of dissimilatory metal-reducing bacteria with a unique EET behavior, using iron oxides as terminal electron acceptors to complete metabolic and electron transfer processes. Moreover, the trivalent iron ions in magnetic hematite have reducibility, which is an important electron carrier property for OM c-Cyts and iron oxide proteins. However, there are currently a few reports on the application of γ-Fe 2 O 3 for the cell surface modification of MFC power generation bacteria. At the same time, recent investigations have demonstrated that the application of a magnetic field (MF) can facilitate the rapid proliferation of biofilms, enhance the electrochemical activity of power-producing microorganisms, shorten startup time, improve open-circuit voltage, reduce reactor resistance, and enrich power-producing bacteria [ 32 , 33 , 34 ]. Therefore, the application of an MF in MFCs has been extensively studied. A reduced startup time and an enhanced biofilm electrocatalytic activity were observed with the application of a 100 mT magnetic field to a single-chamber microbial fuel cell using mixed wastewater [ 35 ]. A range of MFCs, including single-chamber, double-chamber, and three-electrode cell designs, were fabricated utilizing pure cultured Shewanella . Remarkably, under the influence of MF, high voltage outputs were obtained consistently across all configurations. The study revealed that the application of an MF stimulation resulted in an enhanced secretion of mediators and improved catalytic activity. As a consequence, the electron exchange efficiency was markedly improved [ 36 ]. In addition, research has indicated that an appropriate magnetic field intensity can increase power generation and reduce internal resistance, while a stronger magnetic field can suppress MFCs performance [ 37 ]. In addition to the static magnetic field mentioned above, researchers have also introduced an electromagnetic field (EMF). An EMF can serve as a driving force for controlling the metabolic kinetics of EAMs, and a constant high-intensity magnetic field can inhibit microbial metabolism and normal growth [ 38 ]. Therefore, the short-term intermittent application of a magnetic field can have a good regulatory effect on bioelectrocatalysis. Pulse electromagnetic fields (PEMFs) can enhance the enrichment of exoelectrogenic bacteria and accelerate extracellular electron transfer, thereby improving the power generation efficiency. A PEMF causes changes in microbial community and uniformity, leading to a decrease in the microbial diversity of the biofilm [ 39 ]. Applying a 2 mT solenoid magnetic field (SOMF) in an osmotic microbial fuel cell (OMFC) increased the coulomb efficiency in a study by 20–30%, producing a current density of 26.58 ± 12 mW m −2 , power density of 266.29 mA m −2 , and shortening startup time by 1–2 days. However, performance was reduced when the electromagnetic flux of the coil was increased to 3 mT [ 40 ]. In particular, the synergistic application of MNPs and a MF exhibits tremendous potential in enhancing bioelectrochemical electricity generation, facilitating the production of high-value byproducts and efficiently removing pollutants from wastewater sludge [ 41 , 42 , 43 ]. However, there are still a few reports on the bioelectrochemical system coupling controllable electromagnetic fields with magnetic nanoparticles. In this study, the magnetic nanomaterials γ-Fe 2 O 3 and S. putrefaciens CN32 were self-assembled to form hybrid bacteria CN32@γ-Fe 2 O 3 ( Figure 1 ). The magnetic nanoparticles in the hybrid bacteria can serve as electron shuttle media and be coupled with electromagnetic fields to construct an MNPs hybrid bacteria-coupled electromagnetic field system to accelerate the bioelectrocatalytic process at the interface, opening up new channels for electron transfer and to improve the EET transfer efficiency and power generation performance. The experimental findings unequivocally demonstrated that the magnetic nanomaterials on the surface of the bacteria enhanced the DET mediated by OM c-Cyts and significantly improved the efficiency of the bacterial and interface EET processes. Under magnetic field stimulation conditions, electrochemical results showed that the MFC system constructed by CN32@γ-Fe 2 O 3 had a smaller peak spacing, a more negative anodic peak potential, a larger oxidation–reduction area, and a lower charge transfer resistance compared to controls. This experiment provides an uncomplicated, efficient, and low-cost technique for modifying S. putrefaciens CN32 using magnetic nanomaterials, which can promote the EET process and has potential value for BES.",
"discussion": "3. Results and Discussion 3.1. Assembly of CN32@γ-Fe 2 O 3 To demonstrate the feasibility of the hybridization of the magnetic nanomaterial γ-Fe 2 O 3 with microorganisms, a morphology analysis of CN32 and CN32@γ-Fe 2 O 3 was performed using SEM and TEM. Figure S2 presents the X-ray diffraction pattern of the modified material γ-Fe 2 O 3 nanoparticles. The structure reveals characteristic diffraction peaks at 2θ = 30.2, 35.5, 43.2, 53.7, 57.3, and 62.8 corresponding to the (220), (311), (400), (422), (511), and (440) crystal planes. These planes align with the standard JCPDS No. 00-039-1346 crystal structure features [ 44 ], confirming the material’s identity as γ-Fe 2 O 3 . By comparing ( Figure 2 a,b), it can be seen that the unmodified S. putrefaciens CN32 was rod-shaped and had a smooth surface. After γ-Fe 2 O 3 MNPs were added for hybrid cultivation, the bacteria surface became rough due to the particle encapsulation, indicating the successful hybridization of γ-Fe 2 O 3 MNPs on the surface of the bacteria. As shown in Figure 2 c, the elemental analysis results clearly indicate that elements C, P, Fe, and O were uniformly distributed on the surface of the bacteria, further confirming the successful preparation of CN32@γ-Fe 2 O 3 . Corresponding data on the elemental composition of the surfaces are presented in Figure S3 . The surface elemental analysis data reveal the presence of carbon (C), oxygen (O), iron (Fe), and phosphorus (P), with C and P attributed to bacterial surface constituents and Fe and O indicative of γ-Fe 2 O 3 modifications on the surface. 3.2. Exploration of the Optimal Coating Amount of γ-Fe 2 O 3 MNPs To further investigate the effect of the surface modification of γ-Fe 2 O 3 MNPs on bacterial bioelectrocatalytic and to identify the optimal coating amount, we assembled half-cells with different CN32@γ-Fe 2 O 3 bioanodes, further conducted CV, DPV, EIS, and IT electrochemical tests. Figure 3 a reveals the lack of a discernible redox peak in the CV curve of the naturally occurring S. putrefaciens CN32 strain, and only a weak redox peak at approximately −0.45 V (vs. SCE) is observed, which can be attributed to the presence of endogenous electron mediators. It is worth noting that an obvious reversible redox peak pair appears at around −0.4 V and 0 V (vs. SCE) of heterozygous bacteria at all concentrations, which correspond to the endogenous electron mediators and outer membrane C-type cytochrome protein response, respectively. Interestingly, from the CV analysis, it is evident that the oxidation peak current value exhibits a notable upward trend with an increase in the coating amount. This trend reaches its zenith at a coating amount of 7.5 mM γ-Fe 2 O 3 MNPs, after which it gradually diminishes. At the same time, it can be observed that the cathodic and anodic peak separation is minimal for CN32@γ-Fe 2 O 3 + 7.5 mM, indicating a faster electrochemical reaction. γ-Fe 2 O 3 MNPs may serve as efficient electron conduits, facilitating bridging the gap for electron transfer both intra- and inter-cellularly, thus overcoming the limitations associated with long-distance electron transmission and enhancing extracellular electron transfer efficiency [ 45 ]. However, high concentrations of MNPs will decrease the electrocatalytic activity of electricity-producing bacteria, inhibiting their growth and metabolic activity. In addition, the DPV of CN32@γ-Fe 2 O 3 had a clear peak around −0.1 V ( Figure 3 b), which can be attributed to the OM c-Cyts, and the peak current density was higher than that of the undecorated bacteria, which was consistent with the trend of the CV curve. The hybrid bacteria with 7.5 mM γ-Fe 2 O 3 MNPs had the highest electrocatalytic response. However, there was not much change in the oxidation peak attributed to self-secreted electron mediators around −0.45 V, which meant that the magnetic nanomaterials main enhanced DET mediated by OM c-Cyts, probably by improving the conductivity through the encapsulation of γ-Fe 2 O 3 MNPs on bacterial surfaces, constructing a long-distance electron transport channel, and creating an efficient conductive network together inside and outside the biofilm. As expected, the fact that 7.5 mM was the optimal coating concentration was confirmed in the experiment. From Figure 3 c, it can be seen that CN32@γ-Fe 2 O 3 7.5 mM had the highest stable current, which was due to the magnetic nanoparticles promoting the growth and enrichment of EAMs, expediting the growth of a biofilm on the electrode surface, and improving the output current. Through the study of EIS, which was used to evaluate the conductivity of bioanodes [ 46 ], it was found that all the electrodes had similar impedance spectra, composed of a distinct semicircle and a straight line in Figure 3 d. At the electrode–electrolyte interface, the diameter of the semicircle in the impedance spectrum represented the electron transfer resistance ( R ct ). A smaller R ct value indicated a greater efficiency in electron transfer, corresponding to a faster rate of electron transfer at the interface. The wild-type S. putrefaciens CN32 had the highest charge-transfer resistance, indicating the poor electrical conductivity of the native bacterial cells. After modification with γ-Fe 2 O 3 MNPs, R ct was significantly reduced because of the improved conductivity of EAMs, which was beneficial for enhancing the power generation performance of MFCs. Therefore, we chose 7.5 mM γ-Fe 2 O 3 MNPs as the optimal coating concentration to enhance the CN32 electrocatalytic activity and improve the EET efficiency. The following experiments used CN32@γ-Fe 2 O 3 prepared by 7.5 mM γ-Fe 2 O 3 MNPs hybridization for research. 3.3. Effect of EMF on Functionalized CN32@γ-Fe 2 O 3 Further, to explore the impact of EMF on bioelectrocatalysis, we applied the same field strength EMF to CN32@γ-Fe 2 O 3 doped with different concentrations and carried out CV, EIS, and IT electrochemical test analyses. As seen in Figure 4 a, the CN32@γ-Fe 2 O 3 + MF redox peak currents were both substantially increased compared to the natural S. putrefaciens CN32 and showed reversible redox curves. CN32 CC@γ-Fe 2 O 3 7.5 mM + MF electrocatalytic activity was the highest. The electrocatalytic activity of the electroproducing bacteria may have been enhanced due to the stimulation of expression of OM c-Cyts and changes in oxidoreductase activity, which could be attributed to the presence of γ-Fe 2 O 3 /EMF. Figure 4 b demonstrates that EMF has the capability to decrease the charge transfer resistance, thus enhancing the extracellular electron transfer capacity of electroactive bacteria. Notably, the current density exhibited a progressive increase over time, which corresponds to the process of bacteria enrichment on the electrode surface leading to the formation of a biofilm ( Figure 4 c). Meanwhile, CN32 CC@γ-Fe 2 O 3 7.5 mM + MF has the largest amount of power production, indicating that EMF can increase the specific enrichment of electroproducing bacteria at the anode. Interestingly, we also found that the effect of the applied EMF on the MFC was transient and reversible, with the current rising immediately when EMF was switched on, gradually decreasing when EMF was switched off, and gradually increasing with respect to the previous cycle during the three EMF stimulation cycles. This suggests that the EMF stimulation has a superimposed effect on MFCs, and the effect on the interior of the electroproducing bacteria is sustained, which ultimately promotes electron transfer and current generation, which is of great significance in revealing the mechanism of EET. The observed trend in the aforementioned test results aligns closely with the findings depicted in Figure 2 , demonstrating an overall consistency. 3.4. The Mechanism of MNPs and EMF Synergistically Enhance the DET Process Further investigation was conducted to explore the mechanism of hybrid bacteria coupled with EMF to synergistically promote EET. As shown in Figure 5 a, CN32@γ-Fe 2 O 3 exhibited a significant improvement in the oxidation–reduction peak current compared to CN32 or CN32 CC@γ-Fe 2 O 3 . The main reason was that the γ-Fe 2 O 3 MNPs modified on the bacterial surface had a strong interaction with the electrogenic bacteria, thereby enhancing microbial activity, in contrast to anode electrode modification [ 47 ]. It is worth noting that the current density of the CN32 and CN32 CC@γ-Fe 2 O 3 coupled electromagnetic field was only slightly increased. However, the oxidation peak current of CN32@γ-Fe 2 O 3 + MF was 2.24 times higher than that of CN32@γ-Fe 2 O 3 , at 0.451 mA cm −2 and 0.201 mA cm −2 , respectively. This not only indicated that the surface modification of the bacteria was beneficial to the improvement of the electrocatalysis but also suggested that the EMF played a synergistic role in promoting extracellular electron transfer mediated by γ-Fe 2 O 3 MNPs coated on the bacterial surface. On the other hand, the oxidation–reduction peak potential of CN32@γ-Fe 2 O 3 was −0.421 V and 0.053 V (vs. SCE), while that of CN32@γ-Fe 2 O 3 + MF was −0.394 V and −0.07 V. The peak-to-peak distance of CN32@γ-Fe 2 O 3 + MF was reduced by 0.15 V (0.324 V vs. 0.474 V). The reduction in peak-to-peak distance also demonstrated that MNPs and EMF synergistically accelerated EET. To further explore the interfacial redox kinetics, we tested the CV curves of different bioanodes at various scan rates, as shown in Figure S4 . In the results of CN32@γ-Fe 2 O 3 and CN32@γ-Fe 2 O 3 + MF bioanodes, we found that the oxidation–reduction peak current and the square root of the scan rate showed a great linear relationship (5–100 mV s −1 ), indicating that diffusion control dominated the reaction process. As shown in Figure 5 b, the DPV curves of CN32@γ-Fe 2 O 3 and CN32@γ-Fe 2 O 3 + MF exhibit two peaks in the range of −0.8 V to 0.6 V. The peak at around −0.45 V is attributed to flavins, while the peak at around −0.1 V is related to outer membrane cytochrome protein-mediated DET. It is noteworthy that the anodic peak current density of CN32@γ-Fe 2 O 3 reaches 0.06 mA cm −2 , while the curve of CN32@γ-Fe 2 O 3 + MF shows a peak current of 0.146 mA cm −2 , which is 2.43 times higher than without the magnetic field. Correspondingly, the peak potential of CN32@γ-Fe 2 O 3 + MF exhibits a negative shift (−0.136 V vs. −0.12 V). Moreover, it is observed that the peak of a direct electron transfer for CN32 CC@γ-Fe 2 O 3 and CN32 CC@γ-Fe 2 O 3 + MF is not enhanced. This may be attributed to the fact that the heterogeneous coupled electromagnetic field dual system enhances the direct electron transfer in two ways: on the one hand, γ-Fe 2 O 3 MNPs on the bacterial surface act as electron transport channels to improve EET efficiency; on the other hand, the magnetic field promotes the specific enrichment of electrogenic bacteria, which produce additional magnetic electrons (electrons produced by magnetic-field-stimulated microorganisms) and transfer electrons through the newly formed magnetic channel, significantly enhancing the kinetics of electrochemical reactions and the efficiency of a direct electron transfer. As depicted in Figure S5 , using EIS to study interfacial charge transfer behavior, CN32 modified with magnetic nanoparticles exhibited a lower R ct compared to the wild-type CN32. Further application of the EMF resulted in smaller charge transfer resistance, indicating that MNPs and EMF enhanced the conductivity of the bioanode, which is more conducive to extracellular electron transfer and the construction of a well-established biological and non-biological EET interface. It was shown that MtrC and UndA are important OM c-Cyts in the EET process of S. putrefaciens CN32 [ 48 ]. In order to further investigate the mechanism of electroactive bacterial EET, the electrocatalytic properties of MNPs and EMFs for the deletion of bacterium ∆MtrC/UndA CN32 were tested. No significant oxidative reduction current was observed for ∆MtrC/UndA CN32 and ∆MtrC/UndA CN32 + MF. It may be because the deletion of the major external receptor protein of S. putrefaciens CN32 hinders the original EET, and a large number of electrons are unable to carry out the normal EET process. Notably, after the MNPs encapsulated the deletion bacteria, the cyclic voltammetry results showed that ∆MtrC/UndA CN32@γ-Fe 2 O 3 and ∆MtrC/UndA CN32@γ-Fe 2 O 3 + MF still exhibited reversible redox peaks, but the peak currents and integral areas were significantly weaker than those of CN32@γ-Fe 2 O 3 + MF. This may be the result of electron transfer mediated by OM c-Cyts other than MtrC/UndA. The demonstrated reversible extracellular electron transfer proves that OM c-Cyts MtrC/UndA is the major mediating protein because γ-Fe 2 O 3 MNPs can act as terminal electron acceptors to promote EET, but the absence of OM c-Cyts (UndA and MtrC) cuts off the major respiratory chain of electron transfer. Meanwhile, ∆MtrC/UndA CN32@γ-Fe 2 O 3 + MF has greater electrocatalytic activity compared to ∆MtrC/UndA CN32@γ-Fe 2 O 3 . This result illustrates that, firstly, the binding of MNPs to outer membrane pigment proteins promotes electron transfer, and, subsequently, EMF-driven electroproducing bacteria enhance electron transfer, corroborating the synergistic promotion of EET by γ-Fe 2 O 3 MNPs and EMF ( Figure 5 c). Differential pulse voltammetry test results showed that the oxidation peak positions attributed to outer membrane pigment proteins by deletion bacteria at −0.1 V (SCE) were all positively shifted, predicting a slower electrocatalytic reaction compared to CN32@γ-Fe 2 O 3 + MF, and the oxidation peak currents were all significantly reduced because direct electron transfer was limited by the absence of mediator proteins OM c-Cyts MtrC/UndA, which suggests a critical role for the synergistic facilitation of direct electron transfer by the γ-Fe 2 O 3 MNPs and EMF in the synergistic promotion of DET ( Figure 5 d). The hypothetical mechanism of MNPs and EMF synergistically promoted EET, as shown in Figure 6 . The effective interfacial electron transfer depended on close contact between the electron conduit and the receptor interface. Only the electrogenic bacteria at the outermost layer of the biofilm could perform interfacial electron transfer through direct contact. The electron transfer process at the distal end was extremely slow and may not have been fully utilized. However, when γ-Fe 2 O 3 was coated on the surface of bacteria, acting as an electron conduit, it enhanced the conductivity of the bacteria, allowing for a direct connection between bacteria through the electron conduit. This expanded the transfer distance of electrons, resulting in the formation of an inside-out electron transfer pathway in the biological membrane. Moreover, under the influence of magnetic fluid effects caused by the applied electromagnetic field, the catalytic activity of the outer membrane cytochrome protein and metabolic reductase was improved. It opened up new magnetic channels, connecting the cytoplasm and the outer membrane for the transfer of more electrons, thus accelerating the electron transfer process. The growth metabolism of microorganisms was stimulated by the electromagnetic field, inducing cells to secrete additional magnetic electrons, thereby improving the efficiency of interfacial electron transfer in the bioelectrochemical interface."
} | 7,083 |
32049993 | PMC7015315 | pmc | 7,045 | {
"abstract": "Honey bee ( Apis mellifera ) colonies are valued for the pollination services that they provide. However, colony mortality has increased to unsustainable levels in some countries, including the United States. Landscape conversion to monocrop agriculture likely plays a role in this increased mortality by decreasing the food sources available to honey bees. Many land owners and organizations in the Upper Midwest region of the United States would like to restore/reconstruct native prairie habitats. With increasing public awareness of high bee mortality, many landowners and beekeepers have wondered whether these restored prairies could significantly improve honey bee colony nutrition. Conveniently, honey bees have a unique communication signal called a waggle dance, which indicates the locations of the flower patches that foragers perceive as highly profitable food sources. We used these communication signals to answer two main questions: First, is there any part of the season in which the foraging force of a honey bee colony will devote a large proportion of its recruitment efforts (waggle dances) to flower patches within prairies? Second, will honey bee foragers advertise specific taxa of native prairie flowers as profitable pollen sources? We decoded 1528 waggle dances in colonies located near two large, reconstructed prairies. We also collected pollen loads from a subset of waggle-dancing bees, which we then analyzed to determine the flower taxon advertised. Most dances advertised flower patches outside of reconstructed prairies, but the proportion of dances advertising nectar sources within prairies increased significantly in the late summer/fall at one site. Honey bees advertised seven native prairie taxa as profitable pollen sources, although the three most commonly advertised pollen taxa were non-native. Our results suggest that including certain native prairie flower taxa in reconstructed prairies may increase the chances that colonies will use those prairies as major food sources during the period of greatest colony growth and honey production.",
"conclusion": "5 Conclusion While we did not find evidence that reconstructed prairies provide a highly attractive resource for honey bees in May, June, and July, we did find evidence that reconstructed prairies can become very attractive in the later season, potentially leading to significant health benefits to honey bee colonies. In addition, we found that honey bee foragers perceived seven native prairie taxa found in reconstructed prairies as worth advertising to their nestmates. Our results suggest that including these taxa, especially Dalea purpurea , Dalea candida , and Agastache sp. at high densities may make prairies more attractive to honey bees in July, which is useful information for both land managers who want to provide food for honey bees and those that want to avoid potential competition between honey bees and native bees on their land. In cases where landowners are concerned about competition [ 87 ], they may either want to avoid planting these species at high densities, or, if their goal is to conserve specialist bees that rely on those species for pollen (ex. Colletes specialists on Dalea [ 88 ]), they may want to limit the number of honey bee colonies with access to the prairie planting. In addition, our results highlight the current importance of several species of non-native pollen sources, including species in the genus Trifolium , Melilotus officinalis , and Lotus corniculatus to honey bee colonies. We are currently examining the full diets of honey bee colonies located near reconstructed prairies to provide more information about the most attractive prairie species for honey-bee friendly plantings. Future experiments involving planting carefully-controlled patches of flowers and looking at honey bee recruitment behavior could help to determine how important both patch area and planting density are in attracting honey bee foragers.",
"introduction": "1 Introduction Many insect populations around the world have faced rapid declines in recent decades due to human-induced landscape changes, including massive increases in the area of land devoted to monocrop agriculture [ 1 , 2 , 3 , 4 ]. Declines in bee populations in particular have raised alarm because of the essential pollination services that bees provide [ 5 , 6 ], contributing to the global economy and improving human nutrition by making diverse fruits and vegetables cheaper to grow [ 5 ]. The most widely managed crop pollinator species, Apis mellifera L., the European honey bee, contributes an estimated $14 billion yearly in pollination services in the United States alone [ 7 ]. In recent years, beekeepers have seen increased honey bee colony mortality in several regions, including the United States [ 8 , 9 ]. While exposure to pathogens, parasites, and pesticides undoubtedly contribute to high colony mortality, poor nutrition likely plays a key role [ 6 , 9 , 10 , 11 , 12 , 13 , 14 ]. Like all bee species, honey bees require both pollen, their primary source of proteins and lipids, and nectar, their primary source of carbohydrates [ 15 , 16 , 17 ]. In temperate regions, colonies may need to collect an estimated 25 kg of pollen [ 17 , 18 ] and potentially over 300 kg of nectar [ 19 ] to function during the summer and survive the cold winters. In addition to quantity, diet quality is also very important. Diverse pollen sources help honey bees combat pathogens and parasites [ 20 , 21 , 22 , 23 ] and increase their ability to detoxify pesticides [ 24 ]. Colonies living in the temperate zone must respond to frequent changes in the species of blooming flowers from spring to fall [ 25 , 26 ] and may experience periods of dearth where temperatures remain high but few rewarding flowers bloom [ 27 , 28 ]. Many groups around the world are interested in helping maintain healthy bee populations by planting flowers for bees [ 29 , 30 , 31 ]. Simultaneously, many organizations are more broadly interested in restoring native habitats that had been converted to agriculture or other human uses [ 32 ]. Before European colonization, the Upper Midwest of the United States mainly consisted of prairie lands, defined as temperate grasslands with a moderate rainfall and deep-rooted perennial forbs [ 33 , 34 ]. Today less than 2% of the original prairies remain [ 34 ], and there is great interest in restoring native prairie habitats [ 35 ]. Unfortunately, governments and organizations interested in helping bees often have limited information about how different land management schemes [ 36 ] or seed mixes [ 29 ] will affect bee foraging success. Prairie restoration projects are very likely to benefit native bee species, especially bees that specialize on prairie flowers [ 37 , 38 , 39 ]. It is less clear to what extent non-native honey bees will be attracted to and use patches of flowers in reconstructed prairies. Honey bees’ unique life cycle and foraging strategy may affect their use of flowering resources in prairies. Honey bees are generalist foragers that have a very wide foraging range, with most foraging trips occurring within 4 km of the nest [ 40 , 41 , 42 , 43 ] but some trips as far as 14 km away [ 44 , 45 ]. Honey bee colonies contain thousands of foragers that can communicate with each other about the locations of the most rewarding patches of flowers using a signal called a waggle dance, a behavior unique to bees in the genus Apis [ 26 , 44 ]. This signal involves repeated figure-eight runs in which the dancer waggles her abdomen back and forth during the straight middle portion of the figure-eight, called the waggle run. The waggle run provides dance follower bees with a vector containing both the direction of the flower patch relative to the azimuth of the sun and the distance to the patch [ 44 ]. Foragers will only dance to advertise the locations of flower patches that they perceive as profitable (having a favorable ratio of nutrients gained to energy expended) [ 46 ]. These characteristics of honey bee foraging behavior may lead colonies to focus on the densest patches of flowers and ignore sparser flowers within non-rewarding grasses, as is common in reconstructed prairie habitats. However, the density and size of flower patches in prairies change across the season as different species of flowers bloom. Even if honey bee foragers only perceive flowers in prairies as profitable resources during part of the foraging season, access to prairies may boost colony health by supplying nectar or pollen when there is a dearth of non-prairie food sources [ 47 ]. The honey bee waggle dance provides us with a window into how honey bee foragers perceive the resources that they encounter in the landscape around their hives [ 36 ]. At the level of the colony, the proportion of dances advertising flower patches in a given habitat serves as a measure of the decision-making process that allocates foragers among habitat types based on their relative profitability [ 36 ]. At the level of the individual, if a forager brings back food from a particular species of flowers and dances to advertise the site that she visited, it indicates that she perceived those flowers as sufficiently profitable to recruit nestmates. It is currently feasible to determine which flower taxon was advertised in a dance from the pollen that the dancer carried but not from nectar carried by dancers [ 48 , 49 ]. In addition, characteristics of dances advertising nectar sources can give more nuanced information about the perceived profitability of the resource. The total number of waggle runs that a nectar dancer performs in a dance is correlated with her assessment of the profitability of the resource advertised [ 46 ]. This relationship has been demonstrated multiple times with artificial sugar-water feeders [ 50 , 51 , 52 ], but so far has not been demonstrated with pollen or pollen substitutes ([ 53 ] but see [ 54 ]). Multiple studies have also shown that colonies tend to advertise sites at greater distances in order to find profitable nectar sources during dearth periods [ 27 , 40 , 42 , 55 ]. Therefore, we took advantage of the information in honey bee waggle dances to answer two main questions about how honey bee colonies perceive flowers in prairie habitats: first, is there any part of the season in which the foraging force of a honey bee colony will devote a large proportion of its recruitment efforts (waggle dances) to pollen or nectar sources within prairies? To better understand seasonal changes in the proportion of dances advertising prairies, we asked two additional questions: 1) During times of year when a large proportion of nectar dances advertise sources within prairies, do dances for prairie sites include more waggle runs and thus indicate a higher perceived profitability than dances for sites outside of prairies, and 2) Is there a seasonal change in the average distance of advertised nectar sources? For our second main question, we asked whether honey bee foragers will advertise patches of native prairie flowers as high-quality pollen sources, and, if so, which taxa will they advertise? To answer these questions, we placed honey bee colonies in glass-walled observation hives with access to two large, reconstructed prairies. The glass-walled hives allowed us to record the dances performed by members of these colonies throughout the summer and early fall. From these recordings we decoded the direction and distance information within the dances, mapped them as probability density distributions using a Bayesian modeling approach [ 56 ], and determined what percentage of the dances advertised sites within prairies during different parts of the season. At sites with a seasonal change in the proportion of dances for food sources in prairies, we also quantified waggle runs per dance to determine if the bees perceived the prairie flowers as more profitable compared to flowers outside of prairies. We also explored whether there were seasonal changes in the average distance of advertised nectar sources. Finally, to both map and identify the taxa that the foragers considered to be profitable pollen sources, we captured a subset of dancing bees who were carrying pollen loads and identified the pollen source using microscopy and DNA barcoding.",
"discussion": "4 Discussion The results of our study provided answers to two main questions about how honey bee foragers behave when given access to large reconstructed prairies. First, although most dances advertised non-prairie food sources, at Belwin Conservancy, we saw a significant increase in the proportion of dances for nectar sources within prairies at the end of the foraging season. Second, we determined that honey bee foragers do perceive seven taxa of native prairie flowers as profitable pollen sources and advertise them with dances: Solidago spp., Dalea purpurea , Agastache sp., Dalea candida , Ambrosia spp., Chamaecrista fasciculata , and a member of the tribe Heliantheae (probably a Rudbeckia species based on the top BLAST hits). To better understand the seasonal changes in the proportion of nectar dances advertising prairie sites at Belwin Conservancy, we determined whether foragers advertising nectar sources in prairies performed more or fewer waggle runs than foragers advertising nectar sources outside of prairies across the foraging season. We found no significant relationship between the probability that a dance advertised a prairie site and the number of waggle runs performed. A significantly greater number of waggle runs for nectar sources within prairies in August/September would have supported the idea that foragers perceived late season prairie flowers as more profitable sources of nectar than late season non-prairie flowers. The lack of difference suggests that the increased proportion of dances in August/September may instead reflect changes in the abundance of rewarding flowers inside of prairies relative to other areas. However, we did see a significant effect of season, with July having the longest dances. The greater number of waggle runs indicates that foragers invested more energy into recruiting other foragers during this period, which could result from foragers perceiving July nectar sources as more profitable or perceiving a greater colony need for or ability to store nectar during July [ 72 ]. Recruits that follow more waggle runs are significantly more likely to find a food source [ 73 ] so an average increase of 7–9 waggle runs in July could potentially lead to an average higher recruitment success per dance. Because each recruit will likely also dance, this per-dance difference in waggle runs would likely increase recruitment to that site exponentially as more foragers advertise it [ 74 ]. We also looked at the average distances of advertised nectar sources across the season at Belwin Conservancy and found a significant seasonal change with shorter distances advertised in the later season. The average distances that we found are within the range of previous studies [ 36 , 43 , 71 , 75 ]. However, the decrease in distances advertised is surprising because multiple studies have indicated that honey bee colonies in the Upper Midwest and surrounding states face a dearth period when there is a lower diversity of blooming flowers at the end of the summer and into the fall (North Dakota: [ 76 ]; Ohio: [ 75 ]; Michigan: [ 77 ]; Iowa: [ 78 ]). The period from the beginning of July to early August is generally the time of year when honey bee colonies in Minnesota gain the most weight and produce the most honey [ 79 , 80 ]. A nectar dearth period generally results in longer distances advertised [ 27 , 40 , 41 , 42 ]. Our results suggest that being close to a reconstructed prairie in the later season may provide a significant fitness benefit to colonies by allowing them to forage closer to home and expend less energy per trip at a time of year when they might otherwise have needed to travel much farther. A recent study found that colonies in landscapes dominated by corn and soybeans lose weight and their nurse bees lose fat stores starting in August but moving colonies to a large prairie in August rescued them from these effects [ 78 ]. The lower flight distances that we saw in August/September could potentially contribute to this rescue effect. There are several possible explanations for why foragers in our colonies showed a seasonal change in the proportion of waggle dances for nectar sources in prairie sites at Belwin Conservancy but not Carleton College. Differences in the weather between the two sites may have affected nectar production of native prairie and non-prairie species differently. Both temperature and rainfall can have significant effects on the flowering time and nectar production of many species of forbs [ 81 , 82 , 83 ]. The average distance between our colonies and the reconstructed prairies at Belwin Conservancy was smaller than the average distance between our colonies and the reconstructed prairies at Carleton College ( Fig 1 ), which would make them easier to fly to and, thus, likely more attractive [ 71 ]. We also recorded a greater diversity of flowering plants at Belwin Conservancy, including Solidago rigida , blooming in the later season, which may have been particularly attractive sources of both nectar and pollen at that time of year ( S1 Fig ). Unfortunately, our pollen identification methods cannot differentiate between species in the genus Solidago , but they did reveal the genus as one of the most commonly advertised native prairie pollen sources ( S1 Table ). Solidago species tend to grow in large, dense clusters due to their ability to produce colonies of clones using rhizomes [ 84 ]. This growth pattern may make patches of Solidago particularly attractive resources for honey bees given that many foragers can be recruited to the same patch of flowers without depleting it of food. It is possible that the Solidago plants around Belwin Conservancy were more concentrated within the reconstructed prairies while a larger proportion of the Solidago plants blooming near our colonies at Carleton College were growing on roadsides or in other non-prairie habitats. Given that we were only able to survey flowers within prairies, we cannot be certain. In addition to the seven prairie taxa, foragers in our study advertised a diverse set of non-native and native pollens from non-prairie habitats. The three most commonly advertised pollen taxa were Trifolium repens/hybridum , Melilotus officinalis , and Lotus corniculatus ( S1 Table ). These taxa are non-native forbs from Europe brought to North America as forage crops for livestock, and they have since become established as common weeds [ 85 ]. All three have a long bloom period centered around July ( S1 Fig ), which may help to explain the large number of patches advertised. The fact that honey bees were also brought to North America from Europe raises the question of whether honey bees may have evolved strong preferences for cues from flowers in their native range. On the other hand, the success of European honey bees on six continents and the fact that honey bees are the most frequently recorded floral visitors in natural habitats across the globe [ 86 ] suggests that they have very plastic foraging preferences. Dances advertised a number of other non-native weedy taxa, including Trifolium pratense/incarnatum , Securigera varia , Arctium minus , and Brassica spp. ( S1 Table ). These non-native taxa match taxa found in pollen collected by honey bee colonies in a separate nearby study, except for Arctium minus and Lotus corniculatus [ 13 , 76 ]. Maps of dances advertising them indicate that most, but not all, dances for those taxa were for sites outside of prairies ( Fig 5 ). Flower surveys confirmed that a number of non-native species bloomed in prairies at both sites ( S1 Fig ). In addition, dances in May, June, and July advertised a number of native non-prairie taxa, including woody species in genera such as Gleditsia , Rhus , Cornus , Tilia , and Parthenocissus that provide honey bee foragers with very concentrated patches of flowers ( S1 Table )."
} | 5,063 |
35505868 | PMC9052801 | pmc | 7,046 | {
"abstract": "Liquid–liquid\nphase separation (LLPS) is an emerging and\nuniversal mechanism for intracellular organization, particularly,\nby forming membraneless organelles (MLOs) hosting intrinsically disordered\nproteins (IDPs) as scaffolds. Genetic engineering is generally applied\nto reconstruct IDPs harboring over 100 amino acid residues. Here,\nwe report the first design of synthetic hybrids consisting of short\noligopeptides of fewer than 10 residues as “stickers”\nand dextran as a “spacer” to recapitulate the characteristics\nof IDPs, as exemplified by the multivalent FUS protein. Hybrids undergo\nLLPS into micron-sized liquid droplets resembling LLPS in vitro and\nin living cells. Moreover, the droplets formed are capable of recruiting\nproteins and RNAs and providing a favorable environment for a biochemical\nreaction with highly enriched components, thereby mimicking the function\nof natural MLOs. This simple yet versatile model system can help elucidate\nthe molecular interactions implicated in MLOs and pave ways to a new\ntype of biomimetic materials.",
"conclusion": "Conclusion By describing an IDP FUS with a “stickers-and-spacers”\nmodel, we designed a minimalistic representation of the protein. The\nIPH, containing short peptides derived from RACs and arginine-rich\nsequences grafted onto a flexible polymer backbone, exhibited LLPS\nbehavior reminiscent of the formation of natural MLOs. Systematic\nvariation of DW, MW, the Y/R ratio further reviewed the molecular\ndeterminants of LLPS of IPHs, and agreement was found with FUS. The\ndroplets formed by IPH acted as artificial MLOs, enabling recruitment\nand enrichment of model RNAs and proteins and providing liquid compartments\nfor localizing and enhancing an enzymatic reaction. We believe that\nIPHs afford simple yet useful model systems for elucidating molecular\ninteractions for the assembly of MLOs. As a new type of biomaterials,\nIPHs create new possibilities for the dynamic delivery of proteins,\nnucleic acids, as well as in situ biochemical reactions.",
"introduction": "Introduction Cellular\nmetabolism requires precise spatiotemporal regulation\nof numerous biomolecules. Besides lipid bilayer membrane-delimited\ncompartmentalization, 1 membraneless organelles\n(MLOs) formed by liquid–liquid phase separation (LLPS) provide\nanother universal intracellular organization. MLOs have aroused intense\ninterests from multidisciplinary scientists owing to the ubiquitous\nbiological implications in cellular physiology and disease. 2 , 3 In contrast to the stable and static amyloid-like structures, 4 MLO structures are labile, dynamic, and reversible. 5 Most MLOs contain IDPs harboring low-complexity\ndomains, 6 which are responsible for driving\nLLPS via weak and multivalent interactions. Reported in natural living\ncells 7 − 10 and reconstituted systems in vitro, 11 − 15 large low-complexity domains 8 , 12 , 16 (usually over 100 residues) or engineered\nproteins harboring low-complexity domains 11 − 13 are the building\nblocks for LLPS. Notwithstanding, a chemically synthetic construct\nto recapitulate the essential features of IDP is yet to be reported. The intricate molecular interactions implicated in IDPs can be\ndepicted using a simplified “stickers-and-spacers” 17 framework derived from Flory–Huggins\ntheory, wherein the mean-field free energy was enhanced from “sticker”\ninteractions. 18 Modules driving molecular\nattractions are considered as “stickers”, while modules\nproviding a flexible linkage between “stickers” with\nno significant attractions are considered as “spacers”. 18 Taking a reductionist approach, we reason that\nthe “stickers-and-spacers” 18 interaction mode with prominent multivalency 12 , 19 , 20 can help extract the molecular determinants\nof LLPS of IDPs. Hence, we are inspired to employ a bottom-up and\nminimalist approach to design biomimetics of the scaffolding proteins\nof MLOs, which we term “IDP-mimicking polymer–oligopeptide\nhybrid” (IPH). We aim to create a simplified model system with\nconcise and well-defined interaction modules to help elucidate biological\nLLPS from the molecular level as well as to synthesize artificial\nMLOs and evaluate their properties. IPH was chemically synthesized\nby grafting hydrophilic, flexible polymer chains (as the “spacers”)\nwith weakly interacting, low-complexity-domain-derived, short oligopeptides\n(as the “stickers”). The key molecular characteristics,\nincluding molecular weights, patterning, and composition of structural\nmotifs were chosen to mimic natural IDPs. We employed turbidimetry\nand optical microscopy for characterizing micron-sized liquid droplet\nformation as well as fluorescence recovery after photobleaching (FRAP)\nfor characterizing molecular dynamics in a condensed phase. The LLPS\nbehavior was further investigated for IPH with modulated structural\nparameters, namely, the degree of modification 21 (DM) of the peptide, the molecular weight (MW) of the polymer,\nand the tyrosine/arginine ratio (Y/R). MLOs are hubs for numerous\nintricate biochemical reactions, owing\nto the coexistence of compartmentalized structures and high dynamics\nof molecules. 2 Generally hosting protein\nand RNA, MLOs, being ubiquitous in both cytoplasm 22 and nucleus, 23 are widely implicated\nin multifaceted RNA metabolism. 24 We thus\nevaluate the hypothesis that artificial MLOs can harbor certain functions\nof natural MLOs, including the preferential and reversible recruitment\nof cargo proteins, RNA molecules, and enhancement of the biochemical\nreactions.",
"discussion": "Results and Discussion FUS-Mimicking IPH Forms Droplets via LLPS\nIn Vitro As a general structural feature, IDPs contain binding\nelements with\nhigh valency and modest affinity, between which long and flexible\nlinkers are interspersed. 25 We hypothesize\nIDP-mimicking hybrid materials with multivalent weak molecular interactions\ncould undergo LLPS to form MLO-like compartments. We designed and\nsynthesized IPH via grafting CGGSYSGYS/CGGRGG dual peptides to vinylsulfone-modified\ndextran (Dex–VS), which aims to recapitulate the structural\nfeatures of FUS, a representative IDP ( Figure 1 a). We chose charge-neutral dextran as the\nbackbone material because of its hydrophilicity, good biocompatibility,\nand disordered and flexible random-coil-like nature of the polymeric\nchains, 26 which are essential molecular\nfeatures of “spacer” modules of IDPs. The hydroxyl groups\non the dextran backbone were functionalized with thiol-reactive vinylsulfone\n(VS) groups 21 to provide chemical anchorage\npoints for oligopeptides. 1 H NMR was applied to quantify\nthe DMs ( Figure S1b,c ). Figure 1 Chemical structure of\nIPH* and behavior indicative of liquid–liquid\nphase separation. (a) Schematic structure of IPH*. Vinylsulfone-modified\ndextran is conjugated to two types of thiolated oligopeptides, with\nsequences as shown. (b) Formation of micron-sized droplets and associated\nturbidity change (left, left cuvette: IPH* solution; right cuvette:\ncognate buffer for comparison) and images of IPH* droplets under light\nmicroscope (right). Scale bar, 20 μm. (c–e) Liquid-like\nnature of IPH* droplets indicated by wetting phenomenon (glass without\npassivation, scale bar, 20 μm, c), fusion event (induced by\noptical tweezer, scale bars, 5 μm, d), and FRAP (measuring both\ninternal and external molecular rearrangement, scale bars, 10 μm,\ne). Unless otherwise specified, in this paper, the concentration of\nIPH* used is 6 μM, and the solvent used is intracellular physiology-mimicking\nbuffer comprising 150 mM NaCl, 10 mM HEPES, and 10 wt % PEG 8000 at\npH 7.4. Weak and reversible molecular\ninteractions are prevalent in low-complexity\ndomains and pivotal for driving LLPS. 27 Recently, short (<10 residues) reversible amyloid cores (RACs),\nthe low-complexity motifs of RNA-binding IDPs, such as FUS 28 ( Figure S2a ), TDP-43, 29 and the hnRNP family, 30 were revealed to form labile and reversible fibrils reminiscent\nof structures formed by full-length IDPs, thereby indicating that\nRACs could be drivers of intra- and intermolecular interactions in\nMLOs. Moreover, RACs harbor aromatic residues that stabilize weak\nmolecular interactions and labile self-assemblies 27 that exhibit as kinked beta sheets. We thus hypothesized\nthat RACs could be exploited as minimalist structural motifs as the\n“sticker” modules for recapitulating the LLPS behavior\nof full-length IDPs. 31 − 33 Specifically, inspired by work reported on the FUS\nprotein, we designed and synthesized two peptides: cysteine-terminated\nlow-complexity-domain-like peptide (CGGSYSGYS) and cysteine-terminated\n(arginine–glycine–glycine)-containing peptide (CGGRGG)\n( Figure S2d,e ). CGGSYSGYS contains a flexible\nsegment CGG conjugated to an aromatic-rich RAC segment (SYSGYS). 37 SYSGYS 42 from FUS protein can form labile fibrils\nwith a physiologically relevant melting temperature, namely, between\n20 and 50 °C. 28 CGGRGG contains a\nflexible segment CGG, linked to repeats of RGG. RGG repeats are included\nfor three reasons: the abundance of the RGG segment in FUS protein, 8 the dominance of the cation−π-type\ninteraction from tyrosine–arginine (Y–R) pairs in IDPs, 8 , 20 , 31 , 34 , 35 and the presence of a positive charge for\nnucleic acid recruitment. 36 , 37 The thiolated ends\nof both peptides enable facile conjugation to Dex–VS via a\nclick reaction. During the design phase, another RAC with a similar\nsequence from FUS, 28 54 SYSSYG 59 , was considered but was not adopted because of its lower\nmelting temperature (between 4 and 20 °C). We reason that the\ninteraction is too weak to engender LLPS under physiological conditions.\nThe hybrids constructed using oligopeptides with higher stability\nwere shown to form irregular assemblies ( Figure S7g–i ), which will be discussed later in this paper.\nIn the design of IPH, we also considered three molecular features\nof FUS protein, namely, the size of macromolecules, the Y/R ratio,\nand the patterning of RACs ( Figure S2b ).\nOne construct, termed IPH*, was designed ( Figure S2a–c ) and synthesized ( Figure S3a ) to mimic the three aspects to the greatest extent. As the molecular\nmodule sufficient for driving LLPS, 8 the\nlow-complexity domains of FUS, harboring 214 residues, are mimicked\nby selecting a 40 kDa dextran backbone with 247 repeating units. As\nan intrinsic parameter of IDPs, governing LLPS propensity, 8 the Y/R ratio of FUS protein, equaling 0.973,\nis mimicked by using an IPH with a comparable Y/R ratio (Y/R = 1.03).\nThe DM of CGGSYSGYS (13.3%) was chosen to allow the spacing of oligopeptide\nstickers to mimic the patterning of RACs in the low-complexity domains\nof FUS protein, which is slightly higher than designer DM (9.35%, Figure S2b and Supporting Methods ) to ensure\nsufficient driving force of LLPS. IPH* underwent LLPS to form\nmicron-sized droplets under physiological\nconditions in vitro ( Figure 1 b), which is reminiscent of LLPS of parental full-length FUS\nprotein in living cells 5 , 38 and in vitro, 34 presumably driven by the collective contribution of promiscuous\ninteractions including cation−π, π–π,\nand cation–cation interactions. By contrast, the simple mixing\nof CGGSYSGYS and/or CGGRGG with dextran formed no phase separation\nin solution (data not shown). The liquid-like nature of droplets was\nconfirmed by the wetting phenomenon, the fusion event, and FRAP ( Figure 1 c–e and Video S1 ). The droplets formed by IPH* allowed\nrapid material rearrangement with apparent diffusivity D app = 0.0114 μm 2 /s, which is within the\nrange found for an in vitro LLPS system constructed by FUS protein 39 ( D app = 0.002–0.016\nμm 2 /s) and LAF-1 protein 20 ( D app = 0.025–0.01 μm 2 /s). The propensity of LLPS was evaluated by the density and\nsize of droplets formed under microscopy as well as turbidimetry.\nThe extent of LLPS of IPH* depended on the ionic strength and pH conditions\n( Figures 2 a–c\nand S4c ). Interestingly, LLPS exhibited\nthe highest propensity at physiological ionic strength ([NaCl] = 150\nmM), while LLPS was less sensitive to a pH change across the broad\nrange tested, as the arginine sites always remained predominantly\nprotonated ( Figure 2 b,c). 40 Nevertheless, LLPS demonstrated\nsome sensitivity to pH under low-ionic-strength conditions (50 mM\nNaCl or lower, Figures 2 a and S4c ), while the mechanisms are not\nstudied in this paper. IPH* phase-separated in the absence and presence\nof PEG (as a crowding agent) in a concentration-dependent fashion\n( Figures 2 d and S4a ). Low-dose incorporation of crowding agent\ncould enhance the propensity of phase separation in a dose-dependent\nmanner, while high-dose incorporation could undermine the propensity\nthereof ( Figures 2 e\nand S4b ). The upper critical solution temperature\n(UCST) behavior, i.e., the higher propensity to phase-separate under\nlower temperature, was characterized by a temperature-dependent turbidimetry\nassay ( Figure 2 f) and\noptical microscopy ( Figure S4e ) under physiologically\nrelevant concentrations, whereas LLPS gradually evolves into dispersed\nsolution upon heating across the 35 to 45 °C regime, reminiscent\nof the UCST behavior of FUS 41 and some\nother IDPs. 12 , 42 We employ a 1,6-hexanediol assay\nto test the metastability and reversibility of IPH droplets. Generally,\nassemblies of a liquid-like and labile nature can be disrupted by\n1,6-hexanediol, while strong assemblies such as amyloid plaques cannot. 43 − 46 The dose-dependent disruption of droplets and droplet recovery after\n1,6-hexanediol removal were confirmed by both a turbidimetry assay\n( Figure 2 g) and optical\nmicroscopy ( Figure S4d ), supporting the\nmetastability and reversibility of IPH droplets, respectively. Figure 2 Environmental\nresponsiveness of IPH*. (a–c) LLPS is dually\nresponsive to ionic strength and pH. (d,e) IPH phase-separates in\na concentration-dependent (d) and crowding condition-dependent (e)\nmanner. (f,g) LLPS is sensitive to temperature (f) and 1,6-hexanediol\n(g) stimuli. LLPS Behavior Is Dependent\non the Molecular Property and Structure\nof IPH Valency 12 , 47 and “sticker–sticker”\naffinity 8 , 47 of IDPs have been shown as the molecular\ndeterminants of multivalent interactions driving LLPS. We thus sought\nto investigate specific molecular determinants of LLPS for the IPH\nsystem. IPHs varying in DM, MW, and Y/R ratio were synthesized ( Figure S3b–d ). For a systematic study,\nonly one parameter was changed, while the other two remained the same\nas those for IPH* ( Figure S12 ). Based on\nthe “stickers-and-spacers” model, we hypothesize that\nDM and MW will affect the valency of interactions, while the Y/R ratio\nwill influence the strength of “sticker–sticker”\ninteractions. First, we investigated the effect of DM by oligopeptides\non phase behavior. Only IPH with DM higher than a threshold DM (22.3%\n< DM threshold < 37.9%) could exbibit LLPS under physiological\nconditions in vitro, and LLPS propensity increased with increasing\nDM (37.9, 50.2, and 91.7%) ( Figures 3 a and S5a,b ). This is consistent\nwith the strong dependence of the phase behavior on the valency of\nthe “stickers” of IDPs, that is, higher valency allows\nthe formation of LLPS at lower IDP concentration. 12 All high-DM IPHs were prone to phase-separate under moderate\nionic strength, and in particular, maximum turbidity was observed\nunder the physiological value (150 mM NaCl) ( Figures 3 b and S5c ). IPHs\nwith lower DMs were prone to phase-separate under alkaline conditions,\nand LLPS propensity at acidic pH increased drastically with increasing\nDM ( Figures 3 c and S5d ). All higher DM IPHs exhibited responsiveness\nto temperature in a UCST fashion and 1,6-hexanediol in a dose-dependent\nand recoverable manner. The critical temperature and the critical\n1,6-hexanediol concentration (for LLPS disruption) increased with\nincreasing DM ( Figures 3 d,e and S5e,f ). Figure 3 Molecular structural\nfeatures modulate LLPS of IPH. (a–e)\nModulation of DM contributes to modification of LLPS propensity (a),\nresponsiveness to ionic strength (b), pH (c), temperature (d), and\n1,6-hexanediol disruption (e). (f–j) Backbone MW of IPH affects\nLLPS propensity (f), responsiveness to ionic strength (g), pH (h),\ntemperature (i), and 1,6-hexanediol disruption (j). (k–o) Modulation\nof Y/R contributes to modification of LLPS propensity (k), responsiveness\nto ionic strength (l), pH (m), temperature (n), and 1,6-hexanediol\n(HDO) disruption (o). The x -axes of (b), (g), and\n(l) were plotted in logarithmic scale. Next, we investigated the effect of the MW of the dextran backbone\non LLPS at a fixed mass concentration, namely, a fixed concentration\nof “stickers”. The valency of IPH refers to the total\nnumber of oligopeptide “stickers” on a single macromolecule,\nwhich is affected by the MW of dextran backbone. The effect of branching 48 could be neglected, as the degrees of branching\n(DBs) of backbone dextran with different MWs were confirmed to be\ncomparable 49 ( Figure S1a ). The valencies of IPH Dex-6k , IPH Dex-40k , IPH Dex-550k , and IPH Dex-2000k were about 15, 94, 1300, and 4605, respectively.\nHigher MW IPHs could exhibit prominent LLPS under physiological conditions\n( Figures 3 f and S6a,b ), in contrast with the dispersed phase\nbehavior of IPH Dex-6k , thereby underscoring the\nsignificance of valency for initiating LLPS. This result is consistent\nwith an observation of LLPS of IDP in living cells that higher valency\nallows phase separation at a lower fractional saturation of “stickers”. 47 Enzymatic cleavage of the dextran backbone could\nundermine LLPS in a time/dose-dependent and highly efficacious fashion,\nfurther supporting the significance of multivalency for maintaining\nLLPS ( Figure S6g ). A plausible explanation\nis that LLPS is formed via a two-step nucleation–growth pathway.\nWe reason valency is essential for both the initiation (nucleation)\nof phase separation and the growth of liquid droplets. At the same\nmass concentration and same degree of modification, the number density\nof the peptide stickers is identical independent of the MW. In other\nwords, the components that provide the driving force (or enthalpic\ninteraction) are present in the same amount. IPHs of different MWs\nare marked by the difference in the distribution of the peptide stickers.\nThe valency is higher in IPH of a higher molecular weight, that is,\nthe number of peptide stickers on the polymer backbone is higher.\nWe reason that the higher valency favors intramolecular interaction\nof stickers due to their proximity. This self-nucleation process expediates\nliquid phase separation at lower mass concentration. However, a dominant\nintramolecular interaction can compete with intermolecular interaction\nand hinder the growth of droplet size. With a higher MW, IPH was more\nprone to phase-separate under higher ionic strength ( Figures 3 g and S6c ) and exhibited a robust LLPS under a wider pH range ( Figures 3 h and S6d ). All phase-separated IPHs exhibited responsiveness\nto temperature in a UCST fashion and 1,6-hexanediol in a dose-dependent\nmanner, with higher-MW IPHs showing lower sensitivity under the conditions\nexamined ( Figures 3 i,j and S6e,f ). Notably, IPH* (IPH Dex-40k ), designed through structural mimicking of FUS,\ncould undergo prominent LLPS under physiological conditions (in contrast\nwith the dispersed state of IPH Dex-6k ), withstanding\na larger range of ionic strength and showing more responsiveness to\ntemperature and 1,6-hexanediol (in comparison to MW IPH Dex-2000k ). IPHs of medium MW (IPH Dex-40k and IPH Dex-550k ) exhibited similar phase separation behavior, suggesting that under\nconditions where fractional saturation of stickers is fixed, an optimal\nwindow of valency exists. We further investigated the effect\nof the tyrosine/arginine (Y/R)\nratio on the LLPS behavior. IPHs with varying Y/R ratios were able\nto undergo LLPS, despite differing in the extent ( Figures 3 k and S7a,b ). As indicated by the turbidity measurement, IPHs phase-separated\nto a lesser extent at Y/R = 0, 0.268, and ∞ and more prominently\nat the intermediate Y/R values of 1.03 and 3.25 ( Figures 3 k and S7a,b ). For IPH with Y/R = ∞, only CGGSYSGYS oligopeptides\nare present. The result indicates that the RAC-harboring “stickers”\nCGGSYSGYS alone are sufficient to drive LLPS and recapitulate the\nformation of micron-sized liquid compartments by FUS, in contrast\nto the irregular solid assemblies formed by Aβ peptide-inspired\nconjugates ( Figures S2d,e, S3e, and S7g ). Two RAC peptides NFGAFS 29 and SGYDYS 30 with higher melting temperatures (stability),\nnamely, higher than 70 and 50 °C, respectively, were also used\nto construct hybrids with dextran. Both hybrid constructs form only\nirregular assemblies under physiological conditions ( Figure S7h,i ), thus suggesting weak molecular interactions\nare crucial for LLPS. The addition of CGGRGG, the charge-containing\noligopeptides, to IPH was found to increase the responsiveness of\nLLPS to changes in ionic strength ( Figures 3 l and S7c ). IPHs\nat all Y/R ratios showed UCST phase behavior ( Figures 3 n and S7e ) and\n1,6-hexanediol responsiveness ( Figures 3 o and 7f ), reflecting the\nlabile and dynamic nature of phase-separated liquid droplets. Note\nthat the extent of LLPS of IPH* (Y/R = 1.03) was most sensitive to\nchanges in ionic strength ( Figures 3 l and S7c ), temperature\n( Figures 3 n and S7e ), and 1,6-hexanediol ( Figures 3 o and S7f ) while\nremaining robust over a broader range of pH ( Figures 3 m and S7d ). The\nstimuli-responsive behavior of an IPH system containing dual peptides\nshould be beneficial for imparting functions, such as the reversible\nrecruitment and release of biomolecules. In addition, we tested the\nsimple mixing of dextran–CGGSYSGYS and dextran–CGGRGG\n(with fluorescence labeling), which can also form droplets similar\nto IPH*, with the homogeneous distribution of both hybrid macromolecules\nwithin droplets ( Figure S13 ). IPH* Droplets\nas Artificial MLOs The LLPS of IPH* implies\npotential of displaying functionalities of natural MLOs. MLOs act\nas subcellular condensates that enrich various biomolecules including\nRNA and proteins 50 , 51 and host biochemical reactions. 2 Thus, we investigated whether IPH* droplets could\nfunction as artificial MLOs, in terms of preferential recruitment\nof compositional macromolecules and compartmentalized reaction enhancement.\nModel RNA and protein molecules, namely, polyuridylic acid (polyU)\nand green fluorescent protein (GFP), were both recruited and highly\nenriched within artificial MLOs with 716 (±132)- and 102 (±29.8)-fold\nenrichment, respectively ( Figure 4 a,b and Table S1 ). Additionally,\nthe horseradish peroxidase (HRP) enzyme was enriched within artificial\nMLOs by 246 ± 65.5-fold ( Table S1 ).\nNotably, for polyU and HRP, the enhancement of fluorescence intensity\nis one order of magnitude lower compared with the calculated enrichment\nof cargoes ( Figure S14 ), presumably owing\nto the quenching effect of fluorophores. The charge–charge\nattractions between arginine residues from IPH* and phosphate groups\nand the aromatic interactions between tyrosine residues of IPH* and\nuracil of polyU could explain the recruitment of RNA. Similarly, the\nfavorable enrichment of GFP could be attributed to nonspecific charge–charge\nand aromatic interactions between IPH* and the protein. Moreover,\nartificial MLOs demonstrate reversible release and recruitment of\nGFP and polyU in response to temperature change in physiologically\nrelevant range ( Figure 4 a, b). Addition of polyU was found to promote the LLPS of IPH* when\nthe amount reached a stoichiometric ratio or above ( Figure S8b,c ), presumably by reinforcing the network interactions\nwithin droplets. Note that sphericity was maintained for all the RNA-containing\ncondensates, implying the preservation of the liquid-like property.\nOther model RNAs (polyA and tRNA) tested also showed a similar modulation\nof IPH*’s LLPS behavior ( Figure S8d,e ). Figure 4 Dynamic recruitment and release of cargoes from artificial MLOs.\n(a) Protein (GFP at 60 μg/mL loading). (b) RNA (polyU incorporated\nat N / P = 0.01, methylene blue added\nat 25 μM). IPH* droplets release protein/RNA upon heating to\n45 °C, concomitant with droplet dissolution, whereas recruiting\nprotein/RNA upon recooling to 25 °C is concomitant with droplet\nformation. Scale bars, 10 μm. Next, we demonstrated the possibility to carry out compartmentalized\ncatalysis using HRP-catalyzed decomposition of hydrogen peroxide as\na model reaction with the fluorescent resorufin as a reporting molecule\n( Figure 5 a). HRP was\nenriched in IPH* droplets by 246 (±65.5)-fold, thereby confining\nthe reaction inside the liquid compartment. This was confirmed by\nreal-time confocal imaging, which demonstrated a gradual change of\nred fluorescence in the droplet interior (0–480 s) ( Figure 5 b). Moreover, the\nincrease in fluorescence signal was much faster in the condensed phase\nthan in the dispersed phase ( Figure 5 c), supporting that the local reaction rate was about\n15 times faster in the droplet. Assuming the reaction follows Michaelis–Menten\nkinetics, we estimated that V max increased\nby 5.0-fold in the condensed phase in comparison to the dispersed\nphase. The turnover rate constant, k cat , was lower in the condensed phase, at 2.0% that of the dispersed\nphase ( Figure S10 and Supporting Information ). The reduced k cat can be explained\nby the higher viscosity of the condensed phase. Thus, the overall\nincrease in reaction rate in artificial MLOs is attributed to the\nenzyme enrichment. Combined reactions (in both condensed and dispersed\nphase) were quantified by a time-dependent absorbance change of resorufin\nin bulk solution. The bulk reaction rate in the presence of LLPS was\nalso higher than that in the absence of LLPS ( v 0 = 51 ± 2.2 nM·s –1 versus v 0 = 22 ± 6.2 nM·s –1 ). Note that the enhancement of bulk reaction rate is not as significant\nas the local reaction rate, owing to the limited volume fraction of\ncondensed phase (ϕ con = 0.32 ± 0.026%, Table S1 ). The enhancement of the bulk reaction\nrate can be improved by increasing the extent of LLPS, as has been\nachieved by increasing the concentration of IPH* and ϕ con ( Figure S11 ). These results show that\nartificial MLOs formed by IPH* have the potential not only to mimic\nbiophysical properties but functions of MLOs in providing a dynamic\nand hierarchical organization of biomolecules. Figure 5 Artificial MLOs formed\nby IPH* as compartments for biochemical\nreactions. (a) Diffusion of the hydrogen peroxide into the IPH* droplet\ncontaining HRP initiates oxidation of the Amplex Red and its conversion\ninto fluorescent resorufin. (b) Time-lapse fluorescent confocal microscopy\nshowing production of resorufin within the IPH* droplet. (c) Average\nchange of fluorescence intensity over time in the interior (condensed\nphase) and exterior (dispersed phase) of the IPH* droplet ( n = 6). Scale bars, 10 μm."
} | 6,751 |
29568284 | PMC5853052 | pmc | 7,047 | {
"abstract": "In this study, we designed a microbial electrochemical fluidized bed reactor (ME-FBR), with an electroconductive anodic bed made of activated carbon particles for treating a brewery wastewater. Under a batch operating mode, acetate and propionate consumption rates were 13-fold and 2.4-fold higher, respectively, when the fluidized anode was polarized (0.2 V) with respect to open circuit conditions. Operating in a continuous mode, this system could effectively treat the brewery effluent at organic loading rates (OLR) over 1.7 kg m -3 NRV d -1 and with removal efficiencies of 95 ± 1.4% (hydraulic retention time of 1 day and an influent of 1.7 g-COD L -1 ). The coulombic efficiency values highly depended upon the OLR applied, and varied from a 56 ± 15% to 10 ± 1%. Fluorescence in situ hybridization (FISH) analysis revealed a relative high abundance of Geobacter species ( ca . 20%), and clearly showed a natural microbial stratification. Interestingly, the Geobacter cluster was highly enriched in the innermost layers of the biofilm (thickness of 10 μm), which were in contact with the electroconductive particles of bed, whereas the rest of bacteria were located in the outermost layers. To our knowledge, this is the first time that such a clear microbial stratification has been observed on an anode-respiring biofilm. Our results revealed the relevant role of Geobacter in switching between the electrode and other microbial communities performing metabolic reactions in the outermost environment of the biofilm.",
"conclusion": "Conclusion In this study, we expand the classical use of static electrodes for performing microbial electrochemistry by demonstrating that a fluidized anode made of electrically conductive particles is a suitable configuration for treating a real industrial wastewater such as a brewery effluent. The design of our ME-FBR was able to remove COD from wastewater at a rate further than an OLR of 1.7 kg-COD m -3 NRV d -1 . Our results showed that the proportion of electrogenic metabolism highly depended on the OLR applied. Increasing the OLRs in the ME-FBR lead to a decrease in electron recovery on the fluidized anode. Thus, one should examine methods to stimulate the degradation of simple organic matter by microbial electrogenesis rather than by other metabolisms. Working with the fluidized anode serving as electron acceptor (polarization conditions) resulted in improved organic matterremoval rates. Furthermore, we observed that the nature of the fluidized bed (electrically or non-electrically conductive) highly influenced the treatment efficiency and the microbial colonization of the reactor. The microbial analysis revealed the development of a thin biofilm on the fluidized particles of the ME-FBR. A microbial stratification was observed in it, in which Geobacter was naturally located at the inner layers of the biofilm, in intimate contact with the fluidized anode. This scenario shows the existence of a microbial organization in space created by the presence of an extracellular electron acceptor like a fluidized anode. The fact that Geobacter species are stacked immediately adjacent to most internal layers shows the competitive advantage of this species over others for respiring at an anode, and the relevant role of these Geobacter species on interspecies electron transfer. Further studies regarding the factors that promote this microbial distribution should be addressed in order to better understand the role of Geobacter in natural environments where interspecies electron transfer is a key survival strategy for energy conservation.",
"introduction": "Introduction Water purification technologies based on biological processes require a suitable electron acceptor to consume the electrons generated in the oxidation of organic waste. In this context, microbial electrochemical technologies (METs) represent a promising field based on the effective redox coupling between microbial metabolism and electrically conductive materials ( Du et al., 2007 ). Although urban wastewater ( Min and Logan, 2004 ; Brown et al., 2015 ) has been the most common biodegradable fuel tested in METs, industrial organic matter sources such as food industry residues have been extensively tested in the last decade ( Cercado-Quezada et al., 2010 ; Kelly and He, 2014 ; Çetinkaya et al., 2015 ). From the very beginning of research in this field, brewery wastewater has received much attention since the organic components in brewery effluents (consisting of sugars, soluble starch, ethanol and volatile fatty acids) are highly biodegradable ( Feng et al., 2008 ; Dong et al., 2015 ). A potential advantage of bioelectrochemical systems is the higher resistance of electrogens to disturbances caused by shocks in the organic loading when compared to methanogens ( Kaur et al., 2014 ). Anaerobic digestion has been typically the technology used by brewery plants for eliminating the organic matter of its effluents. One of the problematic factors of anaerobic digesters is the slow growth and the high sensitivity of methanogens to a wide variety of inhibitory compounds ( Chen et al., 2008 ). This can lead to an accumulation of the volatile fatty acids (VFAs) (acetic and propionic acids principally) and a pH drop ( Franke-Whittle et al., 2014 ) if the feeding presents disturbances like organic load shocks. In this regard, METs have been shown to be systems in which VFAs can be rapidly consumed by electrogens under the presence of an anode acting as electron acceptor ( Asensio et al., 2016 ; Cerrillo et al., 2016 ). The capacity of biological treatment systems is determined by the biomass amount (concentration, volume, etc.) and its activity. One of the engineering designs that have allowed one to optimize the mix of these two variables is the fluidized bed reactor. This configuration has been merged with a MET by using an electrically conductive fluid-like bed, resulting in a microbial electrochemical fluidized bed reactor (ME-FBR). In this system, the fluidized bed functions as a fluid-like polarized three-dimensional electrode with large specific surface to stimulate the degradation of organic matter by microbial electrogenesis. Interestingly, direct extracellular electron transfer between Geobacter planktonic cells and a fluidized anode has been reported in a ME-FBR ( Tejedor-Sanz et al., 2017a ). This fact suggests a new paradigm in bioelectrochemical systems in which direct extracellular electron transfer (EET) does not proceed through biofilm formation. In contrast to this planktonic interaction, by using porous particles in motion and with a more hydrophilic and irregular surface than glassy carbon beads (i.e., activated carbon), a biofilm architecture of a mixed culture can also be achieved ( Kong et al., 2011 ; Deeke et al., 2015 ; Tejedor-Sanz et al., 2017b ). Electroactive bacteria commonly interact with electrodes directly by forming a biofilm. Techniques based on sequencing the 16S rRNA gene have allowed one to identify the main microbial communities enriched in anode-reducing biofilms. Actually, Geobacter species have been reported to dominate the microbial communities found in anodes composed of mixed populations ( Kan et al., 2011 ; Kiely et al., 2011 ; Yates et al., 2012 ). This genus has also been identified in the granules of an upflow anaerobic sludge blanket (UASB) reactor treating brewery waste ( Shrestha et al., 2014 ). In spite of the absence of an electrode, Geobacter was found to perform direct extracellular electron transfer (DEET) by exchanging the electrons with methanogenic communities, through direct interspecies electron transfer (DIET). Specifically, Methanosarcina barkeri has been shown to be capable of performing DIET in co-cultures with Geobacter species ( Rotaru et al., 2014 , 2015 ). DIET can also take place with a mineral as a mediator, a process in which different species use as conduits of electrons nano-mineral particles or conductive surfaces such as activated carbon granules or biochar ( Kato et al., 2012 ; Liu et al., 2012 ). This phenomena has also been described to stimulate methane production and Geobacter growth ( Cruz Viggi et al., 2014 ; Shrestha and Rotaru, 2014 ; Li et al., 2015 ). All these findings suggest that co-aggregation of Geobacter species and methanogens may be a common phenomenon in methanogenic environments and that might be relevant with respect to methane production in anaerobic digesters. Thus, besides the engineering aspects, the study of the microbial diversity and their interactions in these reactors can provide potential tools for optimizing its performance. In our work we aim to characterize a ME-FBR as a technology for removing the organic matter from a brewery wastewater. This effluent contains both fermentable and non-fermentable matter, and thus promotes the proliferation of a wide range of microbial communities that we aim to characterize. In addition to studying the performance of the system, we investigate the anodic biofilm structure developed on the fluidized electrode in order to provide insights into the interspecies interactions in these electroactive biofilms.",
"discussion": "Results and Discussion The general characteristics of the brewery wastewater are presented in Supplementary Table 1 . Since the wastewater came from a previous coagulation pretreatment, low amounts of suspended solids and nutrients (N and P) were found in this effluent ( ca . 25 and 6.8 mg L -1 , respectively). Nitrate and nitrite were below the detection limits, indicating that most of the nitrogen in the wastewater was in the form of ammonia and insoluble matter not removed during the pretreatment step. The COD load was highly variable from the time of collection (from 0.6 to 2.8 g-COD L -1 ). This allowed us to study the response of the ME-FBR to different organic loads when the biomass was already adapted without the need of dilution or addition of external an organic source to the real wastewater. Due to the inoculum in the ME-FBR and the broad range of energy substrates, diverse anaerobic-based metabolisms were likely to coexist. Therefore, the organic matter (mainly soluble) in the ME-FBR might be degraded throughout a serial of anaerobic biological reactions including fermentations (complex organic matter like sugars and proteins), acidogenesis (further complex matter degradation to short chain acids, H 2 , alcohols, ammonia…), acetogenesis, methanogenesis and electrogenesis ( Figure 1 ). These three last pathways may compete for the acids and H 2 (electron sources) coming from the acidogenesis step. The brewery effluent fed to the ME-FBR contained as possible main electron acceptors a fluidized anode and carbon dioxide. This represents a niche for methane-producing microorganisms, acetate-synthesizing microbes, and electroactive bacteria that are able to harvest energy from the reduction of those substrates. FIGURE 1 Schematic of the different anaerobic communities that might coexist within the ME-FBR and the possible competitive reactions among them for the electron donors. ME-FBR for Treating Organic Matter and VFAs Degradation Tests The ME-FBR could effectively degrade the organic matter from the brewery effluent over a wide range of influent organic loads (Supplementary Figure 2 ). Actually, as the ORL in the ME-FBR increased, the removal of organic matter followed the same trend, reaching values always over 74%. No VFA accumulation was observed during the experimental period, although the levels of these species increased with the increase of the OLR (Supplementary Figure 3 ). The levels of acids were always low (below 35 mg L -1 of total VFAs), revealing that the bio-electrochemical system efficiently oxidized most of the acids formed from the complex organic substrates. Actually, these acids were far below the concentrations reported to inhibit the methanogenic community in anaerobic digesters (can start from 900 mg L -1 of propionic acid at levels of 2400 mg L -1 of acetic acid) ( Wang et al., 2009 ). Acetic acid, the main contributor to methane generation, was indeed the most abundant acid in the effluents (up to 28 mg L -1 ). Butyric acid was detected as well but at lower concentrations. Although the brewery wastewater already contained propionic acid (Supplementary Table 1 ), it could not be detected in the effluent of the reactor, indicating that it was rapidly consumed or not produced as much as acetic or butyric. In order to further analyze the oxidation of the acids in the ME-FBR, we analyzed the individual consumption of acetate and propionate. These two acids have been reported as preferential acids by electroactive bacteria over others with higher chain lengths ( Freguia et al., 2010 ). The VFA degradation tests were assayed by providing either the fluidized anode as final electron acceptor (electrolysis mode), or the fluidized bed as a mere carrier at open circuit (no current flow). Interestingly, both acids were consumed notably faster ( Figure 2 ) under bio-electrochemical conditions (removal rates of ca. 35 μmol-acetate min -1 and 4.8 μmol-propionate min -1 ) than under OC conditions ( ca. 2.7 μmol-acetate min -1 and 2 μmol-propionate min -1 ). Under the electrolysis mode, the removal rates of acetate and propionate per working reactor volume were of 5.4 g-COD L -1 \n NRV d -1 and 1.3 g-COD L -1 \n NRV d -1 , respectively. However, at OC the rates were of 0.4 g-COD L -1 \n NRV and 0.54 g-COD L -1 \n NRV , respectively. In contrast with the high enhancement of acetate removal (13-fold), propionate removal rate was enhanced only by 2.4-fold when the fluidized bed acted as electron acceptor. The removal rates in the ME-FBR described for other METs with flat-electrode designs ( Freguia et al., 2010 ). This rapid VFA consumption in our fluidized anode could represent an attractive tool for stabilizing anaerobic digestion processes when inhibitory processes occur that lead to acid accumulation, as previously described ( Cerrillo et al., 2016 ). FIGURE 2 Consumption of acetic acid and propionic acid in the ME-FBR operated under batch mode, under both open circuit (fluidized anode can not act as electron acceptor) and polarized anode conditions (fluidized anode acting as electron acceptor, E = 0.2 V vs. Ag/AgCl). Propionate consumption led to acetic acid production, which is represented as well in the propionic acid consumption graph. The current response to acetate addition was inmediately detected, whereas for propionate pulse the increase in current production was not as sharp and rapid (Supplementary Figure 4 ). This suggests that acetate was a ready-to-use substrate for electroactive bacteria in the ME-FBR, but propionate had to be fermented first (to acetate or formate) in order to be bioelectrochemically oxidized, as previous studies have observed ( Hari et al., 2016 ). We observed that current was still produced at the fluidized anode even when no acids were detected in the medium of the ME-FBR. This could have been due to a possible adsorption of these acids within the biofilm and on the surface of the particles. The coulombic efficiencies (CE) values obtained were low (38% for acetate and 5% for propionate). These numbers suggest the presence of non-electrogenic microorganisms consuming those acids. These could be acetothrophic methanogens or syntrophic propionate-oxidizing bacteria leading to methane production ( Kouzuma et al., 2015 ). Actually, methane was detected in the headspace of the reactor after an acid pulse, but not quantified (data not shown). Influence of the OLR Over the Bioelectrochemical Performance Next, we performed a serial of assays for testing the treatment capacity of the reactors at different OLRs. Our results showed that the removal efficiency significantly rise with higher OLRs within the range from 0.62 to 1.26 kg m -3 NRV d -1 ( Figure 3A ) (statistical analysis shown in Supplementary Table 3 ). Meanwhile, the organic removal rate (ORR) increased linearly with the OLR over the ranges applied. By operating the ME-FBR at higher OLRs than 1.26 kg m -3 NRV d -1 , like 1.7 kg-COD m -3 NRV d -1 , we did not achieve better COD removal efficiencies. In contrast, it was possible to obtain higher values of removal rates. This indicates that our configuration could further operate at ORLs higher than 1.7 kg-COD m -3 NRV d -1 while maintaining performance levels of 95% of COD removal. FIGURE 3 Performance of the ME-FBR on the treatment of a brewery effluent. (A) Chemical oxygen demand (COD) removal of the system and organic removal rates (ORRs) for all the organic loading rates tested (OLRs). (B) Current density harvested in the fluidized anode and coulombic efficiencies for each OLR tested in the ME-FBR. The fluidized bed was polarized to 0.2 V (vs. Ag/AgCl). Regarding the current density values, our ME-FBR achieved values from 17 ± 4 to 27 ± 5 A m -3 NRV ( ca. 131 ± 33 to 206 ± 40 A m -3 bed ). As the OLR in the ME-FBR increased, lower values of CE were achieved. This actually occurred in spite of obtaining higher current density values at increasing OLRs for most of the conditions assayed ( Figure 3B ). The increase in current density was significant at lower OLRs (Supplementary Table 2 and statistical analysis shown in Supplementary Table 3 ). This effect has been extensively observed in bio-electrochemical systems with mixed cultures, and it seems to be due to the out-competition of other microbial metabolisms, as methanogenesis, at high organic loadings ( Sleutels et al., 2016 ). Another possibility is that the existence of DIET (either via electrode-mediation or by direct cell contact) might be another cause minimizing electron recovery at the fluidized anode ( Liu et al., 2012 ; Shrestha et al., 2013 ). When we ran the system under open circuit conditions at an OLR of 1.46 ± 0.2 kg COD m -3 NRV d -1 and then was operated back with the fluidized bed polarized, we saw an increase in COD removal of a 17% ( Figure 4 ). This indicates that the electrodes polarization (current production) had a real impact in the organic matter removal. FIGURE 4 Chemical oxygen demand removal when the ME-FBR was operated under open circuit conditions and under current-circulating conditions (fluidized bed polarized to 0.4 V vs. Ag/AgCl) (OLR = 1.46 ± 0.2 kg COD m -3 reactor d -1 ). The additional control, the biolite M-FBR (non-bioelectrochemical configuration) showed a different response to the variation of the OLR. This reactor showed much lower treatment efficiencies than the ME-FBR when operated under identical conditions (Supplementary Figure 5 ). The removal of COD in this reactor severally decreased when either the OLR or the flow rate were increased, which means that the system was being overloaded. This difference could be associated with the double role of the electrically conductive bed: (a) for attachment, (the chemical nature of the bed may influence bacteria attachment and biofilm development) and (b) as the electron accepting element (the bed may act as a respiratory substrate by accepting electrons from cell metabolism). Biolite has been reported as a highly biocompatible material that promotes high biomass adhesion, and rapid start-up periods during the treatment of industrial wastewaters ( Balaguer et al., 1997 ). For this reason, we hypothesize that the nature of the material was not limiting biofilm development and biomass acclimation. Probably, the presence of an unlimited electron acceptor (polarized fluidized anode) caused the enhancement in the microbial colonization of the anodic particles. After testing the influence of the OLR over the ME-FBR removal capacity, we analyzed the effect of eliminating the motion of the polarized particles. When the recirculating pump was switched off, the system becomes a fixed bed reactor. The performance of the reactor severally decreased from an average of 87 ± 5% of COD removal when the bed was fluidized to a minimum of 36% when the bed was fixed (Supplementary Figure 6 ). Under this last condition, fresh substrates, metabolites and end products are slowly transported across the biofilm and bed particles. Furthermore, at any extracelullar electron acceptor respiration, e.g., an anode, the reaction is typically consuming electrons but not protons, so this charge uncoupling may cause an acidification of the biofilm as a result of the bioelectrochemical oxidation of organic matter ( Torres et al., 2008 ). Eventually it would lead to a depletion of the electroactive activity of the community. However, the recirculation flow of the ME-FBR can play a key role in avoiding such a biofilm acidification. Indeed, this could be one of the reasons for which eliminating the fluidization state of the bed negatively affected the ME-FBR. In this study, we show the treatment of a brewery wastewater in a ME-FBR as a proof of concept. However, for developing a scalable prototype, each of the elements of design (the electrode material and active surface, distance between anode and cathode, recirculating flow, bed quantity, particle size or pH-control) should be properly studied and optimized. To compete on a commercial scale with high-rate anaerobic reactor configurations, such as UASB (Upflow Anaerobic Sludge Blanket), EGSB (Expanded Granular Sludge Bed), or fluidized bed configurations, which can already treat up to 40 kg-COD m -3 NRV d -1 ( van Lier et al., 2016 ), the OLRs in the ME-FBR should be considerably enhanced. However, under this scenario METs so far face the problem of low electron recovery efficiencies. This minimizes the revenues (recovery of hydrogen or other by-products from cathodic reactions) that can be subtracted from bio-electrochemical treatment. Thus, until METs are able to show higher reaction efficiencies, ME-FBRs and other bio-electrochemical-based configurations are confined to specific applications in which anaerobic reactors experience operational problems. Likewise, METs can find use as a complementary technology for further reducing the COD of the effluents coming from a first biological step. In fact, our ME-FBR has shown good bio-electrochemical behavior at low OLRs, achieving CE values up to 55 ± 15% treating a real effluent. Microscopic Biofilm Examination The biocatalysis in the ME-FBR is mainly performed by the biofilm colonizing the electroconductive bed. As such, a deep analysis of the surface of the bed particles by SEM and FISH provided further insights into the physical distribution of the microbial community. We observed microbial colonization of the activated carbon particles in big pores ( Figure 5 ). The biofilm developed a high content of EPS and matrix, which is typical from systems exposed to shearing stress in fluidized beds. It has been previously observed that the thickness, structure and even the metabolism of fluidized-bed biofilms are highly correlated to the detachment force ( Chang et al., 1991 ; Liu and Tay, 2001 ). Thus, the limited thickness of the biofilm on the polarized particles could be a consequence of high shear conditions in the ME-FBR. FIGURE 5 Scanning electron microscopy (SEM) images of the colonization on the particles of the ME-FBR after 4 months of operation. Regarding the colonization of the biolite particles of the M-FBR, we observed a poor microbial attachment after ca . 2.5 months of operation (Supplementary Figure 5 ). Rather than forming a biofilm, cells seemed to be attached forming aggregates. The difference between the biomass growths on the polarized fluidized anode and on the non-electrically conductive particles could be the reason for the variation in treatment efficiencies found between the two reactors. We further characterized the electroactive biofilm on the activated carbon particles of the ME-FBR by FISH analyses. The results with the probes and the DAPI staining showed a partial coverage with biomass of the surface of the particles ( Figure 6A ). The estimation from the DAPI staining showed that an average of ca. 40% of the surface (shown on the images with the bright field) was colonized with biofilm ( Figure 6C ). The existence of bare spaces without an attached biomass may actually favor the electrical connection among the fluidized particles and between the particles and current collector. FIGURE 6 Fluorescence in situ hybridization (FISH) results on the polarized particles using DAPI (blue signal, all nucleic acids) and the following probes: for (A) Eubacteria probe (red signal) and Geobacter cluster probe (green signal) (probes combination 4) and for (B) Eubacteria (green signal) and Archaea probe (red signal) (probes combination 5). (A,B) 3D Images projected from a sequence of images (sections were taken at 2-μm interval) of the biofilm formed over the surface of the particles (bright field). (C) (a) Biomass coverage of the fluidized particles (average of 2 independent projected sequences) and (b) relative abundance of Eubacteria and Geobacter cluster estimated from at least 2 sequences of images taken for each sample (average of 3 independent projected sequences). (D) Sections of the biofilm from image B in the 3 dimensions. The 3D images were built by the software from projected sequences of 18 images, each one with an interval of 2 μm. From the fluorescence images (Supplementary Figures 7 , 8 ) we could estimate the relative abundance of each hybridized probe. Almost 60 ± 17% of the total biomass was composed of Eubacteria domain. Interestingly, the images showed a relative high abundance of Geobacter species (green) in the biofilm of the polarized particles ( ca . 20 ± 10%) ( Figure 6C ). The presence of this genera characterized by performing efficient direct EET suggests a strong role of Geobacter for oxidizing VFAs and donating the resulting electrons to the fluidized anode ( Aklujkar et al., 2009 ). It also indicates that the polarization of the conductive particles highly determined the microbial diversity in the biofilm and its structure. Similar observations have been described in other studies with METs treating real wastewaters from the food industry, in which the anode presents a high proportion of Geobacter species ( Kiely et al., 2011 ; Blanchet et al., 2015 ). The analysis with the Archaea probe revealed the important presence of species of this domain ( Figures 6B,D ). An estimated average of ca. 50% of Archaea domain from the total biomass was obtained (see Supplementary Figure 8 ). This could be explained by the fact that the particle samples were collected when the ME-FBR was operated at the highest OLR (1.7 kg-COD m -3 NRV d -1 ). This condition probably stimulated the growth of methanogens. Microbial Stratification Within the Biofilm The images taken with the confocal microscope allowed us to visualize a biofilm thickness of ca. 10 μm ( Figures 6D , 7A ). This thickness is relatively low in comparison with either previously reported anodic biofilm ( Jana et al., 2014 ) or with non-electroactive fluidized bed biofilms ( Liu and Tay, 2001 ). Probably the fluidized particles in the ME-FBR were exposed to a high shearing stress that could be limiting the biofilm growth. FIGURE 7 (A) FISH image from the section of the biofilm developed on an activated carbon particle, using Eubacteria probe (red signal), and Geobacter cluster probe (green signal) (probes combination 3). The left side of the red signal corresponds to the outermost environment (ME-FBR medium), whereas the right side of the green signal corresponds to the surface of a fluidized activated carbon particle. This image is representative of one out of 30 sections taken from the biofilm developed on a surface of ca . 50 × 100 μm. (B) Proposed microbial electron transfer mechanisms among the different microbial communities colonizing the particles and toward the fluidized anode. Hybridization using both Archaea and Eubacteria probes did not show a clear organization of these two microbial communities ( Figure 6D ). Nevertheless, we observed a pronounced microbial stratification of the biofilm when the Geobacter cluster and Eubacteria probes were employed (probes combination 3). The most internal layers of the biofilm were mainly composed of bacteria from the Geobacter genus, whereas at the outermost zones of the film other kinds of bacteria were present ( Figure 7A ). This strongly suggests that members of the Geobacter genus could be responsible for the direct and ultimate transfer of electrons to the anode in the ME-FBR. These electrons may be generated from Geobacter metabolism, or from other microorganisms able to perform interspecies electron transfer ( Shrestha and Rotaru, 2014 ). It has already been reported through gene sequencing that the Geobacter genus dominates over other microbial communities at the most internal layers of current-producing mixed-species biofilms ( Malvankar et al., 2012 ). However, we show for the first time images of this microbial stratification on an anodic biofilm. This scenario was observed again when the Geobacter cluster probe was used in combination only with DAPI (probes combination 6, see Supplementary Figure 10A ). Indeed, a higher presence of bacteria domain at the internal layers of the biofilm was observed (probes combination 4, Supplementary Figure 10B ). This microbial stratification and Geobacter selection might have been favored by the initial enrichment strategy used for starting-up the ME-FBR (acetate was added to the ME-FBR). This might have stimulated an initial Geobacter colonization of the particles surface. The brewery wastewater initially contained VFAs such as acetate and propionate, which are also generated during the fermentation processes in the ME-FBR together with H 2 . Thus, it is likely that these conditions promoted as well an additional Geobacter enrichment on the particles (secondary biofilm development). While in the case of anode-reducing biofilms a clear microbial conformation ( Kiely et al., 2011 ) has not been identified, a defined stratification was found on electrode-oxidizing biofilms by using fluorescence in situ hybridization (FISH) ( Virdis et al., 2011 ). In this study, electroactive denitrifying microorganisms using a cathode as electron donor were located at the inner layers of the biofilm on the electrode, whereas ammonia-oxidizing species were found in the outermost layers of the film. This microbial stratification was due to the existence of an oxygen gradient within the biofilm. The scenario found in our anodic fluidized particles suggests the existence of several kinds of microbial interactions within the biofilm. We hypothesize two mechanisms that may act independently or coexist, through either direct or indirect interspecies electron transfer (DIET vs. IIET, see Figure 7B ). The first hypothesis is that members of the Geobacter genus could be responsible for taking up the electrons from the metabolism of other species, acting like a biological plug between outermost cells and the fluidized anode surface by performing DIET (mechanism shown as case A in Figure 7B ) ( Lovley, 2011 ; Shrestha et al., 2013 ). Thus, this direct electron transfer might be directional, from the outer to the innermost layers of the biofilm. This mechanism would allow to indirectly recovering of electrical current from complex substrates that are not substrates for Geobacter (such as sugars) . Previous studies have described the stimulation of DIET between bacteria and methanogens in the presence of granular activated carbon (GAC), without any polarization of this material ( Liu et al., 2012 ). In that study, the mere electrically conductive nature of GAC promoted the exchange of electrons without the need of cell aggregation as typically occurs when performing DIET. However, no stratification was observed within that biofilm, which suggests that the polarization value (e.g., 0.2 V versus Ag/AgCl) of our fluidized activated carbon particles may have determined preferential directions for the electron flux, from the outermost layers of the biofilm to the anode. The cell assemblage along the biofilm thickness might be a consequence of the resulting redox gradient. It remains unknown whether the operation of the ME-FBR at open circuit potential for a long period of time would have led to a different microbial conformation within the biofilm. Further investigation regarding this approach might provide interesting information concerning strategies, which might be employed for interspecies electron transfer. Our second hypothesis is that the assemblage of Geobacter cells at the inner layers of the polarized fluidized particles could be directly metabolizing intermediate metabolites that reach probably by diffusion the deeper layers of the biofilm ( Lovley, 2011 ) (this mechanisms is shown as case B in Figure 7B ). Under this scenario, the communities would be sharing metabolites as energy currency, and cooperating to degrade the organic substrates contained in the brewery effluent. Our ME-FBR reactor was fed by brewery wastewater rich in complex substrates that were used by fermenters and converted into smaller organic molecules like VFAs. Thus, synthrophic metabolic relations were likely to occur among the different microbial communities. Bacteria from Geobacter genus have been reported to be capable of oxidizing a wide variety of short-chain acids, like acetate ( Bond and Lovley, 2003 ), lactate ( Call and Logan, 2011 ; Speers and Reguera, 2012 ), formate ( Speers and Reguera, 2012 ) and directly transferring the resulting electrons to an anode. In addition, hydrogen, which is an end-product from the metabolism of acetogens and an electron shuttle in interspecies electron transfer ( McInerney et al., 2009 ; Stams and Plugge, 2009 ), can also be used as electron donor to Geobacter ( Bond and Lovley, 2003 ). This second scenario has been described as mediated interspecies electron transfer via soluble metabolites ( Rotaru et al., 2012 ; Shrestha and Rotaru, 2014 ). The outer- Geobacter cells within the biofilm might be provided by fresh and ready-to-use non-fermentable substrates. Those cells, not in intimate contact with the anode, may use long-range electron transfer strategies to reach the fluidized anode, like cell-to-cell or cell-to-electrode electron transfer via electrically conductive nanowires as has been described for insoluble electron acceptors ( Reguera et al., 2005 ; Strycharz-Glaven et al., 2011 ). In contrast, the first layer of cells might be respiring the anode by directly contacting the electrode surface via c -type cytochromes ( Busalmen et al., 2008 ). As previously suggested, both mechanism proposed in Figure 7B may coexist, prevailing one over the other depending upon the electron donor available, and the distance to the anode."
} | 8,745 |
31546779 | PMC6783927 | pmc | 7,048 | {
"abstract": "Polyhydroxyalkanoates (PHAs) are biodegradable plastic-like materials with versatile properties. Plant oils are excellent carbon sources for a cost-effective PHA production, due to their high carbon content, large availability, and comparatively low prices. Additionally, efficient process development and control is required for competitive PHA production, which can be facilitated by on-line or in-line monitoring devices. To this end, we have evaluated photon density wave (PDW) spectroscopy as a new process analytical technology for Ralstonia eutropha ( Cupriavidus necator ) H16 plant oil cultivations producing polyhydroxybutyrate (PHB) as an intracellular polymer. PDW spectroscopy was used for in-line recording of the reduced scattering coefficient µ s ’ and the absorption coefficient µ a at 638 nm. A correlation of µ s ’ with the cell dry weight (CDW) and µ a with the residual cell dry weight (RCDW) was observed during growth, PHB accumulation, and PHB degradation phases in batch and pulse feed cultivations. The correlation was used to predict CDW, RCDW, and PHB formation in a high-cell-density fed-batch cultivation with a productivity of 1.65 g PHB ·L −1 ·h −1 and a final biomass of 106 g·L −1 containing 73 wt% PHB. The new method applied in this study allows in-line monitoring of CDW, RCDW, and PHA formation.",
"conclusion": "5. Conclusions Here, we show that in-line PDW spectroscopy is a powerful PAT tool for monitoring R. eutropha -based PHB production. The reduced scattering coefficient µ s ’ and absorption coefficient µ a showed very reproducible signals during different biological cultivations. The new method described in this study allows in-line monitoring of CDW, RCDW, and PHB concentrations in R. eutropha cultivations up to a CDW of 84 g·L −1 . PDW spectroscopy could contribute to improving the scaling-up process and thus to performing PHA production processes in an economical efficient way with the ultimate goal to commercialize a green sustainable plastic.",
"introduction": "1. Introduction When the US Food and Drug Administration (FDA) announced their process analytical technology (PAT) directives, the investigation of PAT became a key research area in bioprocess development. The main objectives are designing, developing, and operating bioprocesses to guarantee a targeted final product quality [ 1 , 2 ]. The focus of this initiative was predominantly on biopharmaceutical processes, while novel PAT tools could be integrated into any bioprocess. Especially, the implementation of PAT for polyhydroxyalkanoate (PHA) production can provide significant benefits to facilitate a consistent and highly efficient production. Techniques such as FTIR, Raman spectroscopy, fluorescence staining associated with flow cytometry, and enzymatic approaches were reported as novel methods for a rapid characterization of PHA production [ 3 , 4 , 5 , 6 , 7 ]. A comprehensive overview of qualitative and quantitative methods for PHA analysis was published by Koller et al. [ 8 ]. However, the reported methods have not been applied for in-line or at-line measurements of the PHA production process so far. Photon density wave (PDW) spectroscopy is an in-line technique, which has been used as an analytical tool for measurements of various highly turbid chemical processes [ 9 , 10 , 11 , 12 ]. The method is based on the theory of photon migration in multiple light scattering material. If intensity-modulated light is introduced into a strongly light scattering but weakly light absorbing material, a PDW is generated. Absorption and scattering properties of the material influence the amplitude and phase of the PDW. By quantifying these shifts as a function of the emitter fiber and detector fiber distance and of the modulation frequency, the absorption coefficient µ a and the reduced scattering coefficient µ s ’ can be determined independently [ 9 , 13 , 14 ]. The mentioned features make PDW spectroscopy very attractive for the monitoring of high-cell-density bioprocesses. Currently, PHA production costs are not compatible with the low-priced production of conventional plastics. The main cost driving factors are the feedstocks for PHA accumulation and the recovery process. Thus, alternative low-cost substrates, e.g., biogenic waste streams, are of high interest to reduce the final production price. Other attempts concentrate on finding more sustainable and price efficient purification strategies [ 15 , 16 , 17 , 18 , 19 , 20 , 21 ]. Ralstonia eutropha (also known as Cupriavidus necator ) is one of the main species studied for polyhydroxybutyrate (PHB) accumulation and the model organism for PHA accumulation [ 22 ]. Growth of R. eutropha on oleaginous feedstocks is particularly attractive due to their high carbon contents, high conversion rates to PHA, and low culture dilution in fed-batch processes. Efficient growth on these feedstocks is facilitated by the expression of extracellular lipases, which emulsify the lipids [ 23 , 24 , 25 , 26 , 27 ]. A large biomass accumulation prior to PHA accumulation is very important for a high final product titer. In this context, it has been shown that urea is an inexpensive nitrogen source, which allows excellent growth [ 24 , 28 ]. Despite alternative substrates and downstream approaches, highly efficient bioprocesses are required for an economic feasible PHA production. Recently, high-cell-density cultivations with R. eutropha on various renewable feedstocks have been published presenting the production of over 100 g·L −1 PHA and space time yields from 1 to 2.5 g PHA ·L −1 ·h −1 [ 21 , 24 , 29 , 30 , 31 ]. However, none of the presented studies describe in-line PAT-based monitoring or control strategies for the enhancement of process results. This work aims to integrate PDW spectroscopy into high-cell-density bioprocesses, for the monitoring of the highly turbid and complex PHB production with R. eutropha in plant oil cultivations. As a result, total cell dry weight (CDW) and residual cell dry weight (RCDW, the difference of CDW and the PHB concentration) accumulation could be distinguished with the PDW spectroscopy probe as a new in-line tool for bioprocesses.",
"discussion": "4. Discussion The purpose of this study was to evaluate the potential of PDW spectroscopy for monitoring plant oil-based R. eutropha cultivations. The batch ( Figure 2 ) and pulse-based fed-batch ( Figure 3 ) cultivations showed that the reduced scattering coefficient µ s ’ correlates strongly with the CDW and the absorption coefficient µ a with the RCDW ( Figure 4 ). These results demonstrate that PDW spectroscopy is a valuable tool for in-line monitoring of the CDW, RCDW, and PHB accumulation. To the best of our knowledge, the results of this study are the first data showing in-line quantification of PHB. The lack of such in-line or on-line monitoring devices for an adaptive control of the production process was recently emphasized by Koller et al. [ 3 ]. Previously, Cruz et al. reported the possibility to use a NIR transflectance probe for an in-line quantification of PHB. However, the authors showed only at-line data quantifying the PHB and CDW concentration (up to 9.3 g·L −1 and 13.7 g·L −1 , respectively) during batch cultivations [ 34 ]. During the initial batch cultivations, PHB yields of 0.56–0.66 g PHB ·g Oil −1 were obtained ( Figure 2 ), which are similar to previously reported yields for R. eutropha H16 growth on palm oil [ 35 ]. During the high-cell-density fed-batch cultivation 106 g·L −1 CDW (72 wt% PHB) with a space time yield of 1.65 g PHB ·L −1 ·h −1 were reached, which is comparable to other published high-cell-density plant oil cultivations [ 21 , 24 , 25 , 29 , 36 , 37 ]. In the pulse experiment, a nitrogen bolus was added (32 h) after the PHB production phase to trigger PHB degradation, as described previously [ 38 ]. The PDW spectroscopy signal µ s ‘ decreased with the declining CDW while µ a increased with an increasing RCDW ( Figure 3 ). Currently, we do not understand why the strength of the signal was not proportional with the determined off-line value changes after that time point (32 h). For this reason, these measurement points were not used for the linear correlation. A potential hypothesis could be an unequal distribution of PHA granules during PHA mobilization, as it was reported for Pseudomonas putida [ 39 ], which might have an effect on scattering and absorption coefficients during PHA degradation. Atypical PHB formation before nitrogen depletion was detected during the high-cell-density cultivation ( Figure 5 ). This preliminary formation of PHB could explain the low yield coefficient of 0.43 g PHB ·g Oil −1 , which is significant lower than the typical yield of PHB in R. eutropha plant oil cultivations [ 37 ]. The formation of PHB without nutrient starvation could indicate a stress response triggered by the high urea levels. Stress responses typically involve the formation of the alarmone (p)ppGpp. For R. eutropha it is known that formation of this alarmone triggers PHB formation [ 38 , 40 ], but (p)ppGpp formation due to excess urea or ammonia availability was not studied so far. Additionally, it was reported that controlled induction of stress could also been used for an enhanced PHB formation [ 41 ]. Such stress responses should be thoroughly considered during the scale-up of a R. eutropha PHA production process, as zones of high or low substrate availability occur in large scale bioreactors. The high impact of such substrate gradients on a reduction of biomass and product yields were intensively studied for Escherichia coli [ 42 ]. Adapting the feeding strategy during the PHA production process by using in-line monitoring devices could be a potential scenario for avoiding such negative impacts on the process. A reduction of the µ s ’ signal after 35 h was observed during the high-cell-density fed-batch cultivation, whereas µ a stayed constant during that period ( Figure 5 ). The decrease of µ s ’ instead of a leveling off of the signal contradicts that a signal saturation effect was observed. Additionally, scattering coefficients in suspensions with particle contents of up to 40% (v·v −1 ) were measured successfully with this technology [ 9 , 10 ]. Heavy foaming, which occurred after 35 h, could be the reason for the observed signal reductions. The surplus of foam was constantly forced into the liquid phase, which increased the overall gas hold-up and total reaction volume in the system. This additional gas volume results in a dilution of the system, which could explain the µ s ’ decrease (35–45 h) even though the culture continued to accumulate PHB ( Figure 5 ). The foaming occurred after the end of the continuous rapeseed oil feed at 35 h. Before 35 h, the added oil functioned as a natural antifoam agent by decreasing the surface tension of the culture broth. During plant oil cultivations foaming occurs through the emulsification process. R. eutropha emulsifies plant oils before uptake, which is catalyzed by extracellular lipases. The lipases cleave the triacylglycerols in diacylglycerols, monoacylglycerols, glycerol, and free fatty acids (FFAs) [ 23 , 43 , 44 ], which causes heavy foaming during aerated bioreactor cultivations. Nevertheless, µ s ’ stayed constant after the CDW did not further increase at 45 h, which could indicate the perfect time point for harvesting in an industrial process. A reliable in-line quantification of CDW, RCDW, and subsequently PHA concentration was reached until a CDW of 84 g·L −1 . To increase the robustness of the method further at higher cell densities, the calculated gas-hold up in the bioreactor [ 45 ] could be integrated in the correlation of the PDW spectroscopy signals. In order to quantify the direct impact of the PHA concentration on the optical coefficients, further studies referencing cell counts and sizes by flow cytometry and microscopy need to be conducted. A wavelength of 638 nm was used to evaluate the PDW spectroscopy signals during this study, which did not show any correlations with the oil addition or emulsification ( Figure 3 ). The emulsification process is very important for an efficient growth. It was previously shown that an overexpression of lipases results in a reduced lag phase and subsequently a more efficient process [ 43 ]. In previous studies, it was shown that PDW spectroscopy was used to measure emulsions [ 46 , 47 ]. An in-line determination of the oil content or the emulsion formation would be a very valuable additional information for bioprocess development and process control. This information might result from integration of additional wavelengths into the PDW spectroscopy set-up. To summarize, PDW spectroscopy allows in-line estimation of the CDW, RCDW, and PHB content in real-time. In contrast, off-line analysis is typically carried out to determine the PHB content, which includes drying the cells and polymer derivatization before time-consuming HPLC or GC analysis. By using PDW spectroscopy, process development and scale-up could be accelerated. In addition, this technology could be used at a large scale for process monitoring and control of R. eutropha cultivations. Specifically, the real-time adjustment of feeding strategies according to the PHA production rates—determined by PDW spectroscopy signals—holds great potential."
} | 3,360 |
21219472 | PMC3285556 | pmc | 7,049 | {
"abstract": "Summary Quorum sensing is a process of bacterial cell–cell communication that enables populations of cells to carry out behaviours in unison. Quorum sensing involves detection of the density-dependent accumulation of extracellular signal molecules called autoinducers that elicit population-wide changes in gene expression. In Vibrio species, CqsS is a membrane-bound histidine kinase that acts as the receptor for the CAI-1 autoinducer which is produced by the CqsA synthase. In Vibrio cholerae , CAI-1 is ( S )-3-hydroxytridecan-4-one. The C170 residue of V. cholerae CqsS specifies a preference for a ligand with a 10-carbon tail length. However, a phenylalanine is present at this position in Vibrio harveyi CqsS and other homologues, suggesting that a shorter CAI-1-like molecule functions as the signal. To investigate this, we purified the V. harveyi CqsS ligand, and determined that it is ( Z )-3-aminoundec-2-en-4-one (Ea-C8-CAI-1) carrying an 8-carbon tail. The V. harveyi CqsA/CqsS system is exquisitely selective for production and detection of this ligand, while the V. cholerae CqsA/CqsS counterparts show relaxed specificity in both production and detection. We isolated CqsS mutants in each species that display reversed specificity for ligands. Our analysis provides insight into how fidelity is maintained in signal transduction systems.",
"introduction": "Introduction Many bacteria produce and release extracellular signalling molecules called autoinducers. By monitoring the accumulation of autoinducers, bacteria track changes in population density and species complexity in the vicinity. This process, known as quorum sensing, enables groups of bacteria to synchronize gene expression and carry out collective behaviours such as bioluminescence, biofilm formation, competence and virulence factor production, which presumably are not productive when performed by a single bacterium ( Fuqua and Greenberg, 2002 ; Pappas et al ., 2004 ; Novick and Geisinger, 2008 ; Ng and Bassler, 2009 ; Williams and Camara, 2009 ). Many Vibrio species possess two or more quorum-sensing systems that channel multiple autoinducer inputs into the output response ( Bassler et al ., 1994 ; Miller et al ., 2002 ; Henke and Bassler, 2004 ). For example, the human pathogen Vibrio cholerae activates quorum sensing in response to two different autoinducers, CAI-1 and AI-2 ( Chen et al ., 2002 ; Miller et al ., 2002 ; Higgins et al ., 2007 ) ( Fig. 1A ); whereas the marine bacterium Vibrio harveyi detects three distinct autoinducers, HAI-1, AI-2 and Vh-CAI-1 ( Cao and Meighen, 1989 ; Chen et al ., 2002 ; Henke and Bassler, 2004 ) ( Fig. 1A ). Presumably, bacteria extract unique information from each autoinducer. Consistent with this notion, HAI-1 ( Fig. 1A ), identified as 3-hydroxybutanoyl homoserine lactone, is produced by the LuxM synthase and detected by the LuxN receptor. The LuxM/LuxN system is present in V. harveyi and a few other very closely related Vibrio species, and HAI-1 is suggested to be used for intra-species communication ( Cao and Meighen, 1989 ; Bassler et al ., 1993 ; 1994 ; Henke and Bassler, 2004 ). AI-2, identified as (2 S , 4 S )-2-methyl-2,3,3,4-tetrahydroxytetrahydrofuran borate in Vibrios , is produced by the LuxS enzyme and detected by the LuxPQ receptor ( Bassler et al ., 1994 ; 1997 ; Schauder et al ., 2001 ; Chen et al ., 2002 ; Neiditch et al ., 2005 ). LuxS is present in many bacterial species; thus, AI-2 is considered a signal for inter-species communication ( Xavier and Bassler, 2003 ; Vendeville et al ., 2005 ; Federle, 2009 ). CAI-1, identified in V. cholerae as ( S )-3-hydroxytridecan-4-one, is produced by the CqsA synthase and detected by the CqsS receptor. The CqsA/CqsS system is conserved in many Vibrio species ( Miller et al ., 2002 ; Henke and Bassler, 2004 ; Higgins et al ., 2007 ), suggesting it may be used for communication between Vibrios . The structure of the CAI-1 signal produced by V. harveyi (Vh-CAI-1) has not been examined. Fig. 1 (A) Autoinducer molecules produced and detected by Vibrio species. Molecules in the left column are the CAI-1 type. HAI-1 and AI-2 are shown in the right column.B. The CqsA/CqsS quorum-sensing system of Vibrio species. Black arrows denote phosphate flow from the CqsS receptor to LuxO at low cell density. Under this condition, the Qrr sRNAs are transcribed and inhibit translation of luxR ( hapR ) transcripts. Thus, LuxR (HapR) proteins are not produced. At high cell density, in the presence of autoinducers, phosphate flow in the signal transduction pathway is reversed, the qrr genes are not transcribed, luxR ( hapR ) mRNA is translated, and LuxR (HapR) protein is produced, and it initiates the quorum-sensing response. The information contained in the three autoinducers in V. harveyi (two in V. cholerae ) is shuttled into a common phosphorelay signal transduction pathway ( Fig. 1B shows the CqsA/CqsS system as an example). Collectively these autoinducers activate production of the master quorum-sensing regulator LuxR (HapR in V. cholerae ). LuxR (HapR) acts as both a transcriptional activator and a transcriptional repressor. In V. harveyi , LuxR activates genes for bioluminescence and regulates at least 50 additional targets ( Showalter et al ., 1990 ; Waters and Bassler, 2006 ; Pompeani et al ., 2008 ). In V. cholerae , HapR activates competence genes and HapA protease production, and represses virulence and biofilm formation ( Kovacikova and Skorupski, 2002 ; Zhu et al ., 2002 ; Hammer and Bassler, 2003 ; 2009 ; Zhu and Mekalanos, 2003 ; Blokesch and Schoolnik, 2008 ; Tsou et al ., 2009 ). Specificity in ligand–receptor interactions presumably plays a role in preventing cross-talk between related signals and eliminating noise from molecules similar to autoinducers that exist in the environment. However, some quorum-sensing systems display low signal discrimination and multiple autoinducers can activate these circuits ( McClean et al ., 1997 ; Cha et al ., 1998 ). The CqsA/CqsS system belongs to the latter class. Specifically, spent culture fluids from a variety of Vibrio species trigger a quorum-sensing response in a V. cholerae CAI-1 reporter strain ( Henke and Bassler, 2004 ). This finding has been interpreted to mean that CAI-1 is used for inter- Vibrio cell–cell communication. Strikingly, however, examination of V. cholerae CqsS receptor mutants displaying altered responses to natural and synthetic CAI-1 analogues showed that the CqsS residue Cys 170 imparts a preference for a CAI-1 molecule that carries a 10-carbon hydrocarbon tail ( Ng et al ., 2010 ). Substituting residue 170 with bulky aromatic amino acids such as phenylalanine or tyrosine (C170F or C170Y) results in a receptor mutant that only recognizes a CAI-1 analogue carrying an 8-carbon tail (C8-CAI-1, Fig. 1A ). Examination of conservation among different CqsS homologues shows that residue 170 is substituted with a phenylalanine in V. harveyi CqsS and other homologues. A polymorphism at this particular position in CqsS suggests that a CAI-1 molecule with a 10-carbon tail is not likely to be the autoinducer detected by all CqsS receptors. In this study, through purification and chemical synthesis, we identify the V. harveyi CqsS ligand as ( Z )-3-aminoundec-2-en-4-one (Ea-C8-CAI-1, Fig. 1A ). Ea-C8-CAI-1 is also produced and detected by V. cholerae CqsA/CqsS. Although the CqsA/CqsS systems in V. harveyi and V. cholerae are highly conserved, the V. harveyi CqsA synthase is highly selective for its substrate octanoyl CoA, and the V. harveyi CqsS receptor likewise displays an exquisite specificity for its ligand Ea-C8-CAI-1. In contrast, V. cholerae CqsA has less substrate specificity, accepting both decanoyl CoA and octanoyl CoA. The V. cholerae CqsS receptor is similarly less stringent and does not discriminate well between ligands. We isolated V. cholerae and V. harveyi CqsS mutants displaying increased and decreased ligand specificity respectively. We propose that the CqsA/CqsS systems in these two organisms have diverged from a common ancestral origin, and that differences arose from selective forces that favoured decreased specificity in V. cholerae and/or increased specificity in V. harveyi.",
"discussion": "Discussion The CqsA/CqsS quorum-sensing system responds to molecules made by a variety of Vibrio species ( Henke and Bassler, 2004 ), yet the identities of the active signalling molecules have not been well characterized. The first CAI-1 signal examined, that of V. cholerae , is (S)-3-hydroxytridecan-4-one ( Higgins et al ., 2007 ) (CAI-1, Fig. 1A ). Here, we show that CAI-1 cannot be detected by the V. harveyi CqsS receptor due to an incompatibility between the 10-carbon tail in CAI-1 and the presence of a bulky F175 residue in the V. harveyi CqsS receptor. Using purification and total in vitro synthesis, we identified the ligand for the V. harveyi CqsS to be ( Z )-3-aminoundec-2-en-4-one (Ea-C8-CAI-1, Fig. 1A ). Ea-C8-CAI-1 fulfils the requirements for being an inter- Vibrio quorum-sensing signal because it is produced and detected by multiple Vibrio species ( Figs 3 and 4 , Table 1 ). Although we focused in this study on V. harveyi and V. cholerae as test cases, the CqsS and CqsA sequences from most other Vibrio species can be readily categorized into one of the two classes represented by these examples. The V. cholerae system has relaxed specificity in both substrate selection by CqsA and ligand detection by CqsS ( Figs 3 and 4 , Tables 1 and 2 ). V. cholerae produces and detects three different molecules: CAI-1, Ea-CAI-1 and Ea-C8-CAI-1. In addition, the absence of an enamino group in CAI-1 does not significantly hamper its agonist activity. In contrast, the V. harveyi system is stringent in both CqsA-dependent production and CqsS-directed detection of the ligand ( Figs 3 and 4 , Tables 1 and 2 ). V. harveyi produces C8-CAI-1 and Ea-C8-CAI-1, but only Ea-C8-CAI-1 is detected. The presence of the enamino group is critical for ligand activity in V. harveyi ( Figs 3 and 4 , Tables 1 and 2 ). Based on these results, we suggest that CqsA/CqsS systems similar to the V. harveyi system are stringent, in which the unique substrate for CqsA is C8-CoA and the unique signal for CqsS is Ea-C8-CAI-1. Consistent with this prediction, we found that only Ea-C8-CAI-1 and C8-CAI-1, but not CAI-1 and Ea-CAI-1, are present in spent culture fluids prepared from several of these species (e.g. Vibrio parahaemolyticus, Vibrio alginolyticus, Vibrio anguillarum and Vibrio furnissii , data not shown). In contrast, CqsA/CqsS systems similar to the V. cholerae system are relaxed (e.g. Vibrio mimicus and some sequenced natural Vibrio isolates). Both C8-CoA and C10-CoA can be used as substrates for CqsA, and multiple signals including CAI-1, Ea-CAI-1 and Ea-C8-CAI-1 can be detected by CqsS. Ea-C8-CAI-1 links the two types of systems as it is produced and detected by both. Ea-CAI-1 and Ea-C8-CAI-1 are both potent agonists for V. cholerae CqsS, and thus we wondered why we did not identify these enamino compounds previously ( Higgins et al ., 2007 ). We suspect that the enamino activities were destroyed during our previous sample preparation because our present results demonstrate that enamino molecules are labile following concentration or heat treatment. The procedures we used to initially isolate V. cholerae CAI-1 likely did not preserve the enamino molecules. However, because the V. cholerae CqsS receptor shows low stringency for the enamino group ( Fig. 4 , Table 1 ), we were able to identify CAI-1 as one of the V. cholerae CqsS ligands ( Higgins et al ., 2007 ). A detailed analysis of the mechanism by which Ea-CAI-1 is produced by CqsA and how Ea-CAI-1 is subsequently converted into CAI-1 is described elsewhere ( Y. Wei et al ., 2011 ). In terms of signal specificity, the residue conferring the difference in receptor specificity is located in the final transmembrane helix of CqsS ( Ng et al ., 2010 ). Receptor specificity can be manipulated by incorporation of a single substitution at the C170 or F175 position of the CqsS receptor of V. cholerae or V. harveyi respectively. This polymorphism is particularly important for chain length preference. The presence of a small amino acid residue such as cysteine decreases the receptor selectivity, while the presence of a bulky amino acid residue such as phenylalanine increases selectivity ( Fig. 4 , Table 1 ). The analogous mechanism underlying CqsA substrate preference is not understood. Based on the V. cholerae CqsA crystal structure, it is believed that the 10-carbon hydrocarbon chain sits in an enclosed hydrophobic pocket lined by H30, V32, F79, F257, I263, F264, C346, P348 and A349 ( Jahan et al ., 2009 ; Kelly et al ., 2009 ). All of these residues are conserved in V. harveyi CqsA so they cannot be responsible for specifying the C8 substrate tail length preference. Additional biochemical and genetic studies are required to understand CqsA substrate specificity in V. harveyi or lack thereof in V. cholerae . Although a few homologues of the CqsA/CqsS system are found in non- Vibrio species (see Tiaden et al ., 2010a ), to date, only one such system (i.e. the Legionella pneumophila LqsA/LqsS system) has been studied ( Tiaden et al ., 2007 ; 2008 ; 2010b ; Spirig et al ., 2008 ). An α-hydroxyketone molecule LAI-1 (3-hydroxypentadecan-4-one, that is, CAI-1 with a 12-carbon tail) is proposed to be the LqsS ligand. It is not known if the LqsA synthase uses a mechanism identical to CqsA to produce an enamino compound which is subsequently converted to LAI-1. The sequence homology is weak for the ligand–receptor interaction region in the sixth transmembrane helix of LqsS and CqsS; therefore, it is unclear how LqsS detects a ligand with a 12-carbon tail. In Gram-negative bacteria, LuxI/LuxR quorum-sensing systems are commonly used. LuxI type proteins produce acyl homoserine lactones, which are detected by cognate cytoplasmic LuxR type receptors ( Fuqua et al ., 2001 ; Pappas et al ., 2004 ). In Gram-positive bacteria, oligopeptides are usually detected by membrane-bound receptors for quorum sensing ( Havarstein et al ., 1995 ; Novick and Geisinger, 2008 ; Thoendel and Horswill, 2010 ). Examination of the phylogenetic relationship among all LuxI/LuxR systems leads to the conclusion that these quorum-sensing systems are ancient and thus cell–cell communication arose early in evolution ( Gray and Garey, 2001 ; Lerat and Moran, 2004 ). The majority of luxI/luxR genes are located contiguously on the chromosome and therefore presumably retain their pairwise functional relationships through co-evolution as single cassettes ( Gray and Garey, 2001 ; Lerat and Moran, 2004 ). The same genomic arrangement is observed for peptide-based quorum-sensing systems in Gram-positive bacteria ( Pestova et al ., 1996 ; Novick and Geisinger, 2008 ). That is, the gene encoding the autoinducer peptide is linked to the gene encoding the receptor. When we examine the gene organization of the different CqsA/CqsS homologues, we find that the cqsA and cqsS genes are usually adjacent in Vibrio species. In other species that possess homologous systems, such as L. pneumophila , Burkholderia xenovorans and Ralstonia eutropha , the genes that encode the synthase and the receptor are also contiguous (as in R. eutropha ) or at least in close proximity with an occasional insertion of an accessory gene (e.g. in L. pneumophila and B. xenovorans ). Finally, signal biosynthesis and signal detection profiles in V. cholerae and V. harveyi match closely ( Figs 3 and 4 ), further suggesting that the cqsA and cqsS genes in each species coevolved. We therefore suggest that an ancestral CqsA/CqsS system must have diverged to give rise to the V. cholerae and V. harveyi CqsA/CqsS systems. V. cholerae and V. harveyi reside in different environmental niches. V. cholerae is a human pathogen that cycles between periods in the aquatic environment and periods inside the host; while V. harveyi is a marine bacterium that is pathogenic to many marine vertebrates and invertebrates. Although we do not know the forces that drove these two species to evolve different signalling specificities, we suspect that stringent signalling in the case of V. harveyi and promiscuous signalling in the case of V. cholerae increases the fitness of each species in its respective environmental niche."
} | 4,176 |
34084321 | PMC8157749 | pmc | 7,051 | {
"abstract": "The perception of organic crystals being rigid static entities is quickly eroding, and molecular crystals are now matching a number of properties previously thought to be unique to soft materials. Here, we present crystals of a boronate ester that encompass many of the elastic and plastic mechanical properties of polymers such as bending, twisting, coiling and highly efficient self-healing of up to 67%, while they maintain their long-range structural order. The approach utilizes the concept of dynamic covalent chemistry and proves it can be applied towards ordered materials. This work expands our current understanding of the properties of crystalline molecular materials, and it could have implications towards the development of mechanically robust organic crystals that are capable of self-repair for durable all-organic electronics and soft robotics.",
"conclusion": "Conclusions Through our investigation of the mechanism of self-healing it is clear that both transesterification and metathesis occur in the solid state and both probably occur concurrently during the self-healing experiments. The two processes are both initiated by the same stimulus, mechanical force, and it is not possible to deconvolute the individual contributions of two mechanisms towards healing. In fact, both processes occurring simultaneously may be the reason the crystals can heal to such high efficiencies. From this, we believe that both transesterification and metathesis are responsible for the self-healing effect ( Fig. 5 ). If a portion of the restoration stemmed from physical adhesion by non-covalent interactions, we would have expected to have measured it during AFM adhesion experiments ( Fig. 6 , Tables S3 and S4 † ). On the other hand, the formation of the transesterification and metathesis products provides firm evidence that the bond shuffling between the boronic esters and catechol occurs in the solid state and appears to be the most likely reason for the self-healing. The reformation of the bonds between the layers is related to intrinsic self-healing capability. Within a broader context, we anticipate that self-healing by dynamic covalent chemistry is a more common phenomenon occurring in molecular crystals that can be deformed plastically, and can be explored within the realm of materials for electronics, robotics and pharmaceutical applications.",
"introduction": "Introduction Fatigue and wear are some of the most inevitable, yet undesirable fates of a material. A deviation from this common axiom was introduced by self-healing materials that are capable of autonomously recovering from damage. 1 The self-healing effect has been extensively explored in mesophasic materials and many self-healing polymers have been developed that use encapsulation, 2,3 metal–ligand interactions, 4,5 hydrogen bonding, 6,7 dynamic covalent chemistry, 8 pericyclic reactions, 9,10 and other covalent and supramolecular strategies. 11–14 Being inherently less flexible, the deformation of molecular crystalline solids has been given much less attention, however this view is evolving to appreciate the vibrant properties of crystalline solids. 15–19 More recently, the realm of atypical phenomena found in crystalline solids has been expanded by the discovery of methods to ‘weld’ crystals 20 and the first self-healing organic crystal, 21 although the material in the latter case showed a modest recovery of only less than 7%. We hypothesized that self-healing in molecular crystals is a more general property and it may be found by exploring other dynamic covalent chemistry motifs. 22,23 To that end, we focused on the reversible reactions between boronic acids and boronic esters that have been utilized to create self-healing polymers and hydrogels. 24–26 Here, we report crystals of a boronic ester that are capable of self-healing with an initial healing of 67% to 44% after five cycles. This material is unique in that unlike other organic crystals that can bend either elastically or plastically, 27,28 when exposed to local pressure it can be deformed elastically and plastically by bending, twisting and coiling while maintaining most of its crystallinity. By providing the first example of highly efficient repair of an organic crystal, this material bridges the gap between soft matter and ordered crystalline solids, and opens prospects for the discovery of other self-healing crystalline materials."
} | 1,099 |
27242863 | PMC4876123 | pmc | 7,052 | {
"abstract": "Shrubs have positive (facilitation) and negative (competition) effects on understory plants, the net interaction effect being modulated by abiotic conditions. Overall shrubs influence to great extent the structure of plant communities where they have significant presence. Interactions in a plant community are quite diverse but little is known about their variability and effects at community level. Here we checked the effects of co-occurring shrub species from different functional types on a focal understory species, determining mechanisms driving interaction outcome, and tested whether effects measured on the focal species were a proxy for effects measured at the community level. Growth, physiological, and reproductive traits of Euphorbia nicaeensis , our focal species, were recorded on individuals growing in association with four dominant shrub species and in adjacent open areas. We also recorded community composition and environmental conditions in each microhabitat. Shrubs provided environmental conditions for plant growth, which contrasted with open areas, including moister soil, greater N content, higher air temperatures, and lower radiation. Shrub-associated individuals showed lower reproductive effort and greater allocation to growth, while most physiological traits remained unaffected. Euphorbia individuals were bigger and had more leaf N under N-fixing than under non-fixing species. Soil moisture was also higher under N-fixing shrubs; therefore soil conditions in the understory may counter reduced light conditions. There was a significant effect of species identity and functional types in the outcome of plant interactions with consistent effects at individual and community levels. The contrasting allocation strategies to reproduction and growth in Euphorbia plants, either associated or not with shrubs, showed high phenotypic plasticity and evidence its ability to cope with contrasting environmental conditions.",
"introduction": "Introduction Plant interactions modulate the structure of plant communities and shape species distribution ( Callaway, 2007 ; Butterfield et al., 2010 ; Cavieres et al., 2014 ). Some plant species facilitate establishment and growth of other species through amelioration of physical stress ( Moro et al., 1997 ; le Roux and McGeoch, 2010 ) or resource supply ( Pugnaire et al., 1996 ; Prieto et al., 2012 ) while competition may counter facilitation effects determining net interaction outcomes ( Tielbörger and Kadmon, 2000 ; Pugnaire and Luque, 2001 ; Armas et al., 2011 ). Indeed, while positive interactions (i.e., facilitation) enhance plant growth, reproduction, and survival of understory species, eventually expanding their distribution range ( le Roux et al., 2012 ), negative interactions (i.e., competition, interference) limit growth and fitness of other species and may even completely exclude them from suitable habitats ( Choler et al., 2001 ; le Roux et al., 2013 ). Overall, there is now ample evidence of plant interactions enhancing coexistence in plant communities at local ( Pugnaire et al., 1996 ; Holzapfel and Mahall, 1999 ; Choler et al., 2001 ; Maestre et al., 2003 ; le Roux and McGeoch, 2010 ; Schöb et al., 2012 ; Tirado et al., 2015 ) and global scales ( Callaway et al., 2002 ; Kikvidze et al., 2005 ; Butterfield et al., 2013 ; He et al., 2013 ; Cavieres et al., 2014 ). Yet there is less evidence on the intensity of plant interactions within a plant community in a given environment ( Pugnaire et al., 2004 ; Callaway, 2007 ). Most studies on plant interactions focused on the effects of a particular species on one ( Callaway et al., 1996 ; Pugnaire et al., 1996 ; Sthultz et al., 2007 ) or several understory species ( Rodríguez-Echeverría and Pérez-Fernández, 2003 ; Gómez-Aparicio et al., 2004 ; Padilla and Pugnaire, 2009 ; Pistón et al., 2015 ). In plant communities, where usually several dominant species co-exist, there is little information on how interaction outcome may vary across species ( Pugnaire et al., 2004 ; but see Liancourt et al., 2005 ; Schöb et al., 2012 ; Pistón et al., 2016 ). In addition, there is a growing body of evidence suggesting that certain species are more likely to act as facilitators than others ( Callaway, 2007 ; Paterno et al., 2016 ), and that these differences may be result from e.g., specific functional traits related to plant phylogeny ( Garbin et al., 2014 ) or differential trait effect on surrounding abiotic conditions ( Schöb et al., 2013a ). Several mechanisms can generate species-specific and plant functional type specific facilitative relationships linked to the variety of ways facilitator species influence their environment and resources, e.g., increasing soil water or nutrient content ( Callaway, 1998 ). For instance, several studies have found that neighboring plants growing close to leguminous species can benefit from the additional N supply to the soil, resulting in greater N concentration in their leaves ( Pugnaire et al., 1996 ; Temperton et al., 2007 ). Thus, species belonging to different functional types (e.g., N-fixers and non-fixers) may differently affect performance of understory species, resulting in a range of interactions that could go from competition to facilitation within the same community ( Pugnaire et al., 2004 ). Moreover, whether such effects are consistent at the species and community levels remains unclear ( Soliveres et al., 2015 ). In stressful environments, such as dry mountains, shrubs can modify the environment under their canopy by improving microclimatic conditions ( Gómez-Aparicio et al., 2004 ; Schöb et al., 2013a ) or increase availability of soil resources ( Verdú and García-Fayos, 2003 ; Gazol and Camarero, 2012 ). There is also evidence that shrubs may improve conditions for the establishment and growth of other species under their canopy despite strong resource competition ( Holmgren et al., 1997 ; Moro et al., 1997 ; Schöb et al., 2014a , b ; Tirado et al., 2015 ). These mechanisms operate simultaneously ( Pugnaire and Luque, 2001 ), exerting both positive and negative effects (e.g., improving water availability while reducing light under the canopy) that would ultimately shape plant interaction outcome. However, we have less evidence about the factors that are more important in driving plant interactions in dry mountain systems. In the southern slopes of the Sierra Nevada Mountains in Spain, at elevations below 2500 m, precipitation is low and temperature and radiation high, leading to rather demanding conditions for plants ( Callaway et al., 2002 ; Sánchez-Marañón et al., 2002 ; Schöb et al., 2013a ). This system provides an opportunity to assess the contribution of different factors (e.g., water, nutrients, light or temperature) to plant interaction outcome. The co-occurrence of several shrub species with contrasting functional types (e.g., N fixers and non-fixers) makes this system suitable to address species-specific effects on understory species. Our aim was to assess the effects of shrubs of contrasting functional type on plant interaction outcome and how limitation of multiple resources (e.g., light, water) modulates plant interactions at the single-species and community levels. We selected four dominant shrub species belonging to two different functional types, two Fabaceae, Cytisus galianoi and Genista versicolor , and two non-N-fixing species, Bupleurum spinosum (Apiaceae) and Hormathophylla spinosa (Brassicaceae), and assessed shrub effects on performance of a focal understory species, Euphorbia nicaeensis , using a functional trait approach ( Violle et al., 2007 ). We additionally estimated shrub effects on the herbaceous plant community beneath shrubs, recording the number of species and of individuals per species. Hence we tested whether effects at the species level paralleled effects at community level, and whether these effects were consistent across different shrub species and functional types. We expected that shrubs from different functional type would differ in their effects on Euphorbia performance and the understory community, and that species-specific effects on Euphorbia would be scalable to the whole community.",
"discussion": "Discussion Our data show specific effects of each shrub species on its understory community, although functional type (N-fixing vs. non-N-fixing shrubs) was a reasonably good predictor of interaction outcome. The effects of each species were consistent at the species and community levels, and were a consequence of their functional type; although we did not measure individual shrub traits, the two non-legume shrubs exerted a net competitive effect on Euphorbia (species level) and on the number of individuals growing beneath them (community level) whereas the two legume shrubs presented neutral effects regardless of level. Finally, Euphorbia displayed a high phenotypic plasticity in response to the shrub presence. We did not find net facilitation effects but rather evidence of strong competition (under non-N-fixing shrubs) or a partial release of competition under N-fixing shrubs ( Holmgren et al., 1997 ; Maestre et al., 2003 ). At the species level, this was most evident regarding biomass, as RII was neutral under N-fixing shrubs ( Cytisus and Genista ) and negative under non-fixing shrubs ( Bupleurum and Hormathophylla ). At community level we recorded net competition (lower number of subordinate individuals) under Bupleurum and Hormathophylla , a neutral effect under Genista , and a facilitative effect under Cytisus . We should note that facilitative and competitive effects are species-specific and depend strongly on the species under study ( Pugnaire et al., 2004 ; Liancourt et al., 2005 ; Callaway, 2007 ; Padilla et al., 2009 ); however, the fact that these effects were similar at the species (e.g., for Euphorbia individuals) and community levels (number of individuals per m 2 ) suggest that these results are scalable, as suggested already by Schöb et al. (2013b) . This parallelism in the effects at the species and community levels suggests that using a single focal species provides reliable evidence on the processes shaping the plant community. The contrasting allocation patterns recorded in Euphorbia , along with strong correlations between PAR, SWC and reproductive traits, suggest different strategies of individuals growing under shrubs and in open areas. Plants under shrubs invested relatively more in growth at the expenses of reproduction, as reported for E. terracina by Riordan et al. (2008) . These differences in biomass allocation point also to the species considerable phenotypic plasticity and its ability to cope with stress meant by high light and low water availability. Since Euphorbia seeds can germinate under both light and dark conditions ( Narbona et al., 2006 ), seeds would germinate easily under shrubs and individuals would grow better thanks to the high water and nutrient availability. Individuals in open areas, however, must cope with drier soils, higher temperatures and lower soil nutrient content, investing relatively more into reproductive organs likely to compensate for higher juvenile or adult mortality ( Al Samman et al., 2001 ). Although reproductive allocation has a genetic basis, it can greatly vary within species depending on environmental conditions ( Karlsson and Méndez, 2005 ; Castellanos et al., 2014 ). Therefore, differences in allocation patterns are Euphorbia nicaeensis’ phenotypic responses to environmental variability under shrubs and in open areas. As the number of vegetative stems, a good index of plant age in Euphorbia nicaeensis ( Narbona, 2002 ), was similar among microhabitats, we assume a similar age for all selected individuals ruling out age driven differences on the reproductive outputs observed. The prevalence of competitive effects has been also reported for other Euphorbia species under dry conditions (e.g., Rinella and Sheley, 2005 ). Improvement of soil water availability and soil nutrients under perennial species is a major source of facilitation in semi-arid systems ( Pugnaire et al., 2011 ). Yet, given the interaction outcome in Euphorbia , regarding plant size (biomass) and plant fitness (reproductive traits), and the preference of the species for sunny places ( Benedí et al., 1997 ), it appears that light is the limiting factor for this species (cf. Grime, 1973 ). Our data support this idea since reproductive output (and reproductive-to-vegetative biomass ratio) was greater in open areas than under shrubs, despite lower soil water and nutrient content. Euphorbia individuals grew taller under shrubs than in open areas, which are a common response to light limitation ( Macek and Lepš, 2003 , 2008 ). Nevertheless, as height of Euphorbia individuals also differed between shrub microhabitats and did not correlate with PAR under shrubs ( P = 0.90), it suggests there was no etiolation but rather an increase in plant size. Although light is an important factor driving plant interactions ( Holmgren et al., 2012 ), facilitation through increased nutrient availability is also common in semi-arid environments, especially by N under legume species ( Pugnaire et al., 1996 ; Reynolds et al., 1999 ; Gómez-Aparicio et al., 2004 ; Armas et al., 2008 ). In our case, despite strong competition for light, Euphorbia benefited from improved N content and higher soil moisture under leguminous shrubs, as changes in leaf N positively correlated to plant height and size (i.e., taller plants and greater RII under N-fixing shrubs). Hence, the lower competition intensity observed under N-fixing shrubs was most probably a consequence of increased resource availability. This benefit may be, however, difficult to separate from improved soil water conditions, as plant N and water uptake are usually tightly linked ( Hu and Schmidhalter, 2005 ; Macek et al., 2009 ). Recent climate models predict an increase in aridity and temperature in Southern Europe ( IPCC, 2007 ), and particularly under Mediterranean climate, where drier conditions would be found at higher elevations as compared to current conditions ( Ackerly et al., 2010 ). Hence, plant association patterns could be altered by increased competition between shrubs and beneficiary species ( Schöb et al., 2014a ). An understanding of the mechanisms that drive these interactions could help forecast future changes in structure, function and assembly of plant communities under changing climate ( Callaway, 2007 ). We hypothesize that a competitive displacement of some shrub species and a change in shrub community structure with increasing environmental harshness may alter plant interaction patterns in dry mountain systems, leading to a community structure where the ratio between N-fixing and non-N-fixing shrubs will play an important role in the Sierra Nevada Mountains."
} | 3,740 |
39748701 | PMC11744930 | pmc | 7,055 | {
"abstract": "Methyl ketones, key\nbuilding blocks widely used in diverse\nindustrial\napplications, largely depend on oil-derived chemical methods for their\nproduction. Here, we investigated biobased production alternatives\nfor short-chain ketones, adapting the solvent-tolerant soil bacterium Pseudomonas putida as a host for ketone biosynthesis\neither by whole-cell biocatalysis or using engineered minicells, chromosome-free\nbacterial vesicles. Organic acids (acetate, propanoate and butanoate)\nwere selected as the main carbon substrate to drive the biosynthesis\nof acetone, butanone and 2-pentanone. Pathway optimization identified\nefficient enzyme variants from Clostridium acetobutylicum and Escherichia coli , tested with\nboth constitutive and inducible expression of the cognate genes.\nBy implementing these optimized pathways in P. putida minicells, which can be prepared through a simple three-step purification\nprotocol, the feedstock was converted into the target short-chain\nmethyl ketones. These results highlight the value of combining morphology\nand pathway engineering of noncanonical bacterial hosts to establish\nalternative bioprocesses for toxic chemicals that are difficult to\nproduce by conventional approaches.",
"conclusion": "Conclusion The biotechnological production of MKs and\nrelated molecules is\na sustainable alternative to oil-derived chemical production. Here, P. putida was adapted and optimized for short-chain\nketone production owing to its innate solvent tolerance and enhanced\nutilization of organic acids as a carbon source. Examining product\nand substrate toxicity in both E. coli and P. putida highlighted the value\nof Pseudomonas as a host for MK production. Next\nto this, whole-cell ketone production with butanoate as the main carbon\nsource was explored by implementing several pathway designs and synthetic\nexpression systems. This systematic analysis suggested an optimal\ncombination of Thl Ca , AtoDA Ec and Adc Ca elements,\nwhere the cognate genes are heterologously expressed under the transcriptional\ncontrol of the rhamnose-inducible RhaRS/ P rhaBAD system. This bioproduction setup is amenable\nto further optimization steps toward increasing product titers and\nyields, e.g., fed-batch cultivation 40 or\ncofeeding strategies. 94 Given the known\npromiscuity of the enzymes in this pathway, byproduct formation should\nbe carefully monitored during production process optimization, particularly\nat high MK titers. Engineered chromosome-free P. putida minicells were adopted to explore alternative\napproaches for short-chain\nMK bioproduction. E. coli minicells\nhave been exploited for sugar-dependent production of C 6 –C 10 alcohols and esters, 92 isobutanol and lycopene 95 —leading\nto titers that often exceeded those obtained by whole-cell biocatalysis.\nIn our study, all three MKs of interest (C 3 –C 5 ) could be produced by the engineered P. putida minicells through bioconversion of C n –1 organic acids. The catalytic lifespan of minicells\ngenerally surpassed that of their parental cells, providing a distinct\nadvantage for bioproduction. Yet, several aspects of the minicell\nbioconversion process can be subjected to optimization. For example,\nceftriaxone, while not essential for MK production, could potentially\nincrease costs and have potential environmental impacts. In this proof-of-concept\nstudy, the antibiotic was used to limit parental cell growth, ensuring\nthe purity of the minicell preparation and the accuracy of characterization\nexperiments. Additionally, alternative methods (e.g., CRISPR interference) 96 − 98 could be explored to induce minicell formation and ease strain construction,\nwhile potentially eliminating the need for purification steps. Together\nwith these optimization steps, a systematic analysis of resource allocation\n(e.g., ATP and reducing equivalents) would illuminate further engineering\nsteps to maximize minicell-based bioproduction of chemicals and proteins. Finally, the results in this study not only underscore an alternative\nbioproduction strategy for chemical commodities, 99 but they also suggest further applications for biosensing 100 and the delivery of other cargoes (e.g., proteins).\nIn this sense, bacterial minicells can be programmed to target cancer\ncells; 101 since some short-chain ketones\nhave been shown to impair carcinogenesis, 7 this feature could be leveraged for designing advanced therapies.",
"introduction": "Introduction Ketones, a class of\nstructurally diverse\nmolecules with the general\nformula R –C(=O)– R ’, where R and R ’\ncan be a variety of carbon-containing substituents, are currently\nproduced using traditional, oil-based chemistry. 1 Methyl ketones (MKs), a subset of this family of compounds\nwhere one ligand on the carbonyl carbon is a CH 3 group,\nplay an important commercial role. 2 Because\nof their strong solvent properties and generally high evaporation\nrates, ketones are used as building-blocks in the fragrance, flavor,\ntextile, pesticide and agrochemical industries, 3 − 5 as well as being\nkey precursors for the synthesis of pharma molecules. 6 2-Pentanone [CH 3 –(CH 2 ) 2 –C(=O)–CH 3 ], a prime\nexample of the broad MK family, has multiple applications\nin the fragrance sector and it has been recognized as a potent inhibitor\nof prostaglandin production, associated with colon carcinogenesis. 7 Greener alternatives for ketone production are\nneeded to fulfill the growing market demands, and the adoption of\nrobust microbial cell factories emerges as an attractive option to\nthis end. 8 − 11 Toxicity issues, imposed on the producer cells by either intermediates\nor products of solvent biosynthesis pathways, continue to be a major\nhurdle that impairs the development of stable bioprocesses. 12 , 13 This occurrence frequently leads to the loss of plasmid(s) encoding\ncomponents of the biosynthesis pathways or the accumulation of mutations\nin the genes thereof, resulting in nonproducing phenotypes. 14 − 16 Several strategies have been implemented toward extending\nthe lifespan\nof actively producing cell factories, yet growth-coupled bioprocesses\ninvolving harsh, reactive intermediates and products tend to display\nlimited yields due to the stability issues listed above. An alternative\napproach that can help overcoming such limitations is using anucleated,\nbacteria-derived vesicles as biocatalysts. 17 , 18 Chromosome-free minicells, for instance, naturally occur within\nbacterial populations albeit at very low frequencies. 19 Under normal growth conditions, the Z ring\nconstricts and recruits the peptidoglycan synthesis machinery together\nwith its associated proteins. 20 The dynamic\nassembly of FtsZ units in rod-shaped bacteria is controlled by the\nMin system, which encompasses the MinC, MinD and MinE structural proteins. 21 MinC and MinD inhibit FtsZ assembly, whereas\nMinE acts as a negative regulator of MinCD. Minicells are formed when\nthere is a limited abundance of MinC and MinD, whereupon FtsZ buildup\npromotes asymmetric cell division. 22 Unlike\ntheir parental counterparts, minicells do not contain chromosomal\nDNA and are unable to further divide; for this reason, they have historically\nbeen exploited as a model for studying protein synthesis. 23 In spite of their recognized potential as (mini)cell\nfactories, 24 these chromosome-free bacterial\nvesicles have not been extensively exploited for bioproduction. Enhanced\nprotein production, for instance, has been demonstrated in Escherichia coli minicells 25 —since plasmids are preferentially located\nat the poles of\nthe parental bacterium, 26 their enrichment\nin minicells is facilitated via either active partitioning\nor random distribution. Moreover, cellular resources in the minicells\n(e.g., reducing power and energy equivalents) can be reasonably assumed\nto be allocated to bioproduction instead of housekeeping processes,\ne.g., genome maintenance and duplication. Among the microbial chassis implemented for metabolic engineering thus far,\nminicell applications have been exploited mostly for E. coli and Bacillus subtilis . 23 No attempts have been reported on\nproducing these anucleated vesicles from Pseudomonas\nputida , an nonpathogenic soil bacterium extensively\nadopted for engineering applications. 27 − 30 Building on the well-known\nsolvent tolerance features of P. putida ( 31 − 34 ) and its ability of using a wide range of carbon sources, 35 − 39 we have engineered genome-reduced P. putida strains for biosynthesis of short-chain (C 3 –C 5 ) (methyl) ketones from organic acids, with a focus on 2-pentanone\nproduction. The capacity of Pseudomonas species to\nproduce long-chain MKs is illustrated by the model-guided engineering\nof P . taiwanensis VLB120, 40 which enabled the synthesis of C 11 –C 17 MKs from sugars. In our study, the tolerance\nof E. coli and P. putida to products and substrates was compared, and production conditions\nwere optimized for whole-cell 2-pentanone biosynthesis by using\nbutanoate as the main substrate. Different\nsynthetic pathway designs were likewise tested, analyzing the performance\nof constitutive and inducible expression systems across production\nconditions. Furthermore, we explored the potential of P. putida minicells for chemical production, adopting\nMKs as model compounds. Thus, we engineered P. putida -derived minicells to express selected pathway variants, demonstrating\nproduction of acetone, butanone and 2-pentanone from acetate, propanoate\nand butanoate, respectively. Programmable P. putida minicells had the ability to stably produce ketones over extended\ntimeframes (up to 4 months)—a first case example of an “off-the-shelf”\ncell factory for MK biosynthesis. This strategy underscores the potential\nof integrating synthetic morphology with pathway engineering as an\nalternative approach to sustainable ketone biosynthesis.",
"discussion": "Results and Discussion Exploring\nthe Potential of P. putida for Short-Chain\nKetone Biosynthesis from Organic Acids A major challenge\nfor efficient short-chain ketones production in\nengineered bacterial cell factories is the stress caused by the endogenously\nproduced chemicals, which (similarly to other solvents) 41 could inhibit growth or even cause cell death. 42 While the biosynthesis of acetone (2-propanone,\nC 3 ) and butanone (C 4 ) by engineered microorganisms\nhas been explored in the primary literature, 43 − 46 reports on biobased approaches\nfor 2-pentanone (C 5 ) production are scarce. A study by\nLan et al. 47 highlighted product toxicity\nas a key factor impairing 2-pentanone biosynthesis by engineered E. coli strains using glucose as a carbon source.\nThe growth of E. coli JCL299, one of\nsuch modified strains, 47 was reported to\nbe impaired by 50% in the presence of as little as 0.6 g L –1 2-pentanone, and bacterial growth\nwas fully arrested when the concentration of the ketone reached 5\ng L –1 . As P. putida is known to be a solvent-tolerant host, we explored its capacity\nfor short-chain MK production, with 2-pentanone as a proxy of this\nfamily of compounds. As a first step in testing the performance\nof P. putida as a production platform,\nwe examined the effect of adding 2-pentanone to cultures of P. putida SEM1.3.\nStrain SEM1.3 is a refactored derivative of P. putida EM42, 48 a reduced-genome version of the\nplatform strain KT2440 ( Table 1 ). 49 The modifications introduced\nin P. putida SEM1.3 comprise (i) deletion\nof the benABCD gene cluster, 50 to abolish oxidation of 3-methylbenzoate (3- m Bz),\nallowing the use of this molecule as a gratuitous (i.e., nonmetabolizable)\ninducer of the XylS/ Pm expression system without\ninterfering with measurements of the optical density at 600 nm (OD 600 ) typically caused by brown-colored catechols and other\nproducts of aromatic compound metabolism 51 and (ii) elimination of the native pha gene cluster, phaC1ZC2DFI , 52 to avoid any potential\nmetabolic cross-talk that could compete for acetyl-coenzyme A (CoA),\nkey precursor for short-chain ketone biosynthesis via acetyl-CoA–dependent chain elongation. P.\nputida SEM1.3 was grown either in rich lysogeny broth\n(LB) or de Bont minimal (DBM) medium supplemented with 1% (w/v) glucose\nas the main carbon source and spiked with 2-pentanone at different concentrations\n(i.e., 1, 2.5, 5, 10, 25 and 50 g L –1 ). Under all\nconditions tested, the impact of 2-pentanone on cell physiology was\nmore evident when\ncells grew in a mineral medium than in a rich broth. Bacterial growth,\nassessed as the OD 600 values after 24 h, started to be\naffected at ketone concentrations above 5 g L –1 ,\nbut bacterial growth was still observed with up to 25 or 50 g L –1 of 2-pentanone both\nin DBM medium or LB, respectively ( Figure 1 A). As an example, the final cell density\nwas only reduced to ∼40% in LB medium spiked with 2-pentanone\nat 50 g L –1 . In general, P. putida SEM1.3 fared better under these conditions than E.\ncoli JCL299, based on data reported in the literature. 47 These results illustrate the ability of P. putida to adapt to high solvent concentrations,\na phenotype mediated by modifications in the surface of the outer\nmembrane and other natural stress response mechanisms. 31 Next, we explored whether P.\nputida could utilize 2-pentanone as a carbon source\nby incubating strain SEM1.3 in DBM medium supplemented with 1 g L –1 of 2-pentanone as the sole carbon substrate. The\nMK should not mediate any toxicity at this relatively low concentration.\nControl experiments in DBM medium containing 1 g L –1 glucose were run in parallel to benchmark growth patterns, and these\ncultures reached full saturation within 24 h ( Figure 1 B). No growth was detected when strain SEM1.3\nwas incubated with the ketone over 100 h ( Figure 1 B), indicating that P. putida is unable to utilize 2-pentanone as a substrate and underscoring\nits potential for short-chain MK bioproduction. Table 1 Bacterial Strains and Plasmids Used\nin This Study Bacterial\nstrain Relevant\ncharacteristics a Reference\nor source Escherichia coli DH5α λ pir Cloning\nhost; F – λ – endA1\ngln X44 (AS) thiE1 recA1 relA1 spoT1 gyrA96 (Nal R ) rfbC1 deoR nupG Φ80( lacZ ΔM15)\nΔ( argF-lac ) U 169 hsdR17 ( r K – m K + ), λ pir lysogen Hanahan and Meselson 123 Pseudomonas putida KT2440 Wild-type strain; derivative\nof P. putida mt-2 cured of the catabolic\nTOL plasmid 124 Bagdasarian et al. 125 SEM1.3 Reduced-genome derivative\nof strain EM42; 48 Δ phaC1ZC2DFI (Δ PP_5003-PP_5008 ) Δ benABCD (Δ PP_3161-PP_3164 ) Kozaeva et al. 50 SEM1.3 Δ minD Minicell-forming\nderivative\nof strain SEM1.3, Δ minD (Δ PP_1733 ) This work KT2440 Δ minD Minicell-forming\nderivative\nof strain KT2440, Δ minD (Δ PP_1733 ) This work Plasmid Relevant\ncharacteristics Reference\nor source pSEVA4413 Standard vector for constitutive\ngene expression; P EM7 promoter; oriV (pRO1600/ColE1); Sm R /Sp R Silva-Rocha\net al. 63 pS4413· msfGFP Derivative of vector pSEVA4413\nfor constitutive expression of the monomeric superfolder GFP; P EM7 → msfGFP ; Sm R /Sp R Fernández-Cabezón\net al. 126 pSEVA2313 Standard vector for constitutive\ngene expression; P EM7 promoter; oriV (pBBR1); Km R Wirth et al. 127 pS2313·MKc Derivative of vector pSEVA2313\nharboring the genes encoding the canonical acetone biosynthesis pathway\nfrom Clostridium acetobutylicum ; P EM7 → thl Ca ctfAB Ca adc Ca ; Km R This work pS2313·MKs1 Derivative of vector pSEVA2313\nharboring the genes encoding a synthetic MK biosynthesis pathway; P EM7 → thl Ca atoDA Ec adc Ca ; Km R This work pS2313·MKs2 Derivative of vector pSEVA2313\nharboring the genes encoding a synthetic MK biosynthesis pathway; P EM7 → thl Ca pcaIJ Pp adc Ca ; Km R This work pS2313·MKs3 Derivative of vector pSEVA2313\nharboring the genes encoding a synthetic MK biosynthesis pathway; P EM7 → phaA Cn ctfAB Ca adc Ca ; Km R This work pSEVA438 Standard\nexpression vector\ncarrying a 3- m Bz–inducible expression system; oriV (pBBR1); xylS , Pm ;\nSm R /Sp R Silva-Rocha et al. 63 pS438·MKc Derivative of vector pSEVA2313\nharboring the genes encoding the canonical acetone biosynthesis pathway\nfrom C. acetobutylicum ; XylS/ Pm → thl Ca ctfAB Ca adc Ca ; Sm R /Sp R This work pS438·MKs1 Derivative of vector\npSEVA438\nharboring the genes encoding a synthetic MK biosynthesis pathway;\nXylS/ Pm → thl Ca atoDA Ec adc Ca ; Sm R /Sp R This work pSEVA4318 Standard expression vector\ncarrying a rhamnose-inducible expression system; oriV (pBBR1); rhaR , rhaS, P rhaBAD ; Sm R /Sp R Martínez-García\net al. 64 Kozaeva et al. 44 pS4318·MKc Derivative of vector pSEVA4318\nharboring the genes encoding the canonical acetone biosynthesis pathway\nfrom C. acetobutylicum ; RhaRS/ P rhaBAD → thl Ca ctfAB C adc Ca ; Sm R /Sp R Kozaeva et al. 44 pS4318·MKs1 Derivative of vector pSEVA4318\nharboring the genes encoding a synthetic MK biosynthesis pathway;\nRhaRS/ P rhaBAD → thl Ca atoDA Ec adc Ca ; Sm R /Sp R Kozaeva et al. 44 pS4318·MKc·Acs Derivative of vector\npSEVA4318\nharboring the genes encoding the canonical acetone biosynthesis pathway\nfrom C. acetobutylicum and an acetyl-CoA\nsynthase gene ( acs ) from Bacillus\nsubtilis ; RhaRS/ P rhaBAD → thl Ca ctfAB Ca adc Ca acs Bs ; Sm R /Sp R This work pS4318·MKs1·Acs Derivative of vector pSEVA4318\nharboring the genes encoding a synthetic MK biosynthesis pathway and\nan acetyl-CoA synthase gene ( acs ) from B. subtilis ; RhaRS/ P rhaBAD → thl Ca atoDA Ec adc Ca acs Bs ; Sm R /Sp R This work a Antibiotic markers and abbreviations: Km , kanamycin; Nal , nalidixic acid; Sm , streptomycin; Sp , spectinomycin; CoA , coenzyme A; MK , methyl ketone; and\n3- m Bz, 3-methylbenzoate. The source of relevant genes\nis indicated with a superscript as follows: Ca , Clostridium acetobutylicum ; Cn , Cupriavidus necator ; Ec , Escherichia coli ; Pp , Pseudomonas putida ; and Bs , Bacillus subtilis . Figure 1 Exploring the tolerance of P. putida and E. coli to 2-pentanone and short-chain organic\nacids. (A) Physiological response of P. putida SEM1.3 to increasing concentrations of 2-pentanone, added to either de Bont\nminimal (DBM)\nmedium or rich LB medium in microtiter plate cultures. Cell densities,\nestimated as the optical density at 600 nm (OD 600 ), were\nmeasured after 24 h of cultivation. (B) Growth profile of P. putida SEM1.3 cultivated on DBM medium containing\neither glucose or 2-pentanone as the sole carbon source. (C) Growth\nof E. coli MG1655 and P. putida SEM1.3 cultivated in DBM medium with different\ncarbon sources (i.e., glucose, acetate, butanoate, or a combination\nof the two short-chain organic acids). Cell densities in these shaken-flask\ncultures, estimated as the OD 600 , were measured after 24\nh of cultivation; specific growth rates (μ) were calculated\nduring exponential growth. In all cases, mean values ± standard\ndeviations were derived from independent biological triplicates. Significance\nlevels of the cell density values when compared to control conditions\n[0 g L –1 2-pentanone for panel (A) and 2 g L –1 glucose for panel (B)] are indicated as follows:\n* P -value <0.05, ** P -value <0.01\nand *** P -value <0.001. Another attractive metabolic feature of P. putida , i.e., efficient assimilation of organic\nacids as sole carbon source, 53 could be\nexploited for short-chain ketone biosynthesis.\nTo examine this possibility, we evaluated butanoate as a substrate\nfor 2-pentanone production. Butanoate, a C 4 carboxylic\nacid, can be processed by Pseudomonas through the\ncanonical β-oxidation to yield\nacetyl-CoA. 54 Even though this carboxylate\nhas not been actively investigated as a feedstock for bacterial fermentations,\nbutanoate is a promising building-block that can be obtained from\nrenewable resources 55 and it has been selected\nas a substrate for establishing high-cell-density P.\nputida cultures. 56 While\nthese properties position butanoate as an interesting biorefinery\nsubstrate, osmotic shock and acid stress may result in toxicity issues\neven at relatively low concentrations. 57 To test this scenario, the tolerance of P. putida SEM1.3 and E. coli MG1655 (adopted\nas a model Gram-negative bacterium, extensively used as a host in\ndiverse bioprocesses) 58 to increasing butanoate\nconcentrations was evaluated and compared with widely used sugar and\norganic acid substrates ( Figure 1 C). Glucose or acetate were selected as the primary\ncarbon feedstocks as representative examples of a glycolytic and a\ngluconeogenic substrate, respectively. P. putida SEM1.3 tolerated up to 7 g L –1 butanoate with\na reduction of ca. 45% in the final OD 600 values, while\nthe growth of E. coli MG1655 was virtually\nabolished at any butanoate concentration above 3 g L –1 . Neither bacterial species could grow on DBM medium containing 10\ng L –1 butanoate, which marks the practical upper\nconcentration to be used in production experiments. Interestingly,\nthe cultures of P. putida SEM1.3 reached\nsimilar final OD 600 values when butanoate was either used\nas the sole carbon substrate or added at 3 g L –1 in the presence of glucose or acetate. This observation indicates\nthat butanoate is not a preferred carbon source in the presence of\na cosubstrate. 59 The specific growth rate\n(μ) was not affected by butanoate at concentrations <5 g\nL –1 ( Figure 1 C), although we observed an increase (ca. 2 h) in the extension\nof the lag phase in these cultures. Building on these results, we\nadopted reduced-genome P. putida SEM1.3\nas the host to explore the butanoate-dependent biosynthesis of 2-pentanone. Implementing and Optimizing Pathways for 2-Pentanone Biosynthesis\nfrom Organic Acids The canonical short-chain ketone biosynthesis\nroute, key to the widely known fermentation process to produce acetone\nby Clostridium acetobutylicum ATCC\n824 (the Weizmann organism ), 60 was adopted to explore MK biosynthesis by engineered P. putida . We hypothesized that this pathway could\nbe coupled with CoA-dependent ketoacid chain elongation, 61 , 62 yielding butyryl-CoA (C 4 ) from acetyl-CoA (C 2 ) extender units. In this way, the core acetone biosynthesis pathway\nof C. acetobutylicum can be adjusted\nto produce a variety of short-chain ketones. For instance, biosynthesis\nof 2-pentanone (C 5 ) from\nbutanoate involves the condensation of acetyl-CoA and butyryl-CoA\n(forming 3-ketohexanoyl-CoA) followed by ester hydrolysis and decarboxylation\nto yield the final product ( Figure 2 A). The sequence starts with the condensation of acetyl-CoA\nand butyryl-CoA mediated by a thiolase (Thl; acetyl-CoA acetyltransferase,\nEC 2.3.1.9). Next, CtfAB, an acetoacetyl-CoA:acetate/butanoate CoA\ntransferase (EC 2.8.3.9) relocates a CoA moiety from 3-ketohexanoyl-CoA\nto acetate, forming 3-ketohexanoate. Finally, this molecule is reduced\nto 2-pentanone by an acetoacetate decarboxylase (Adc; EC 4.1.1.4),\nreleasing CO 2 ( Figure 2 A). The theoretical ketone yield from organic acids\nthrough the canonical biosynthesis pathway shown in Figure 2 A is Y P/S = 50% mol mol –1 . Figure 2 Pathway engineering and\noptimization for 2-pentanone biosynthesis\nfrom short-chain organic acids. (A) Biosynthetic pathway for 2-pentanone production from butanoate.\nMethyl ketone\nbiosynthesis relies on the enzymes of the canonical acetone pathway\nfrom C. acetobutylicum, comprising\nThl, thiolase (acetyl-CoA acetyltransferase, EC 2.3.1.9), CtfAB (acetoacetyl-CoA:acetate/butanoate\nCoA transferase, α and β subunits, EC 2.8.3.9), and Adc\n(acetoacetate decarboxylase, EC 4.1.1.4). Enzyme variants are likewise\nindicated whenever relevant. Abbreviations: CoA ,\ncoenzyme A; PPi , inorganic pyrophosphate. (B) Synthetic\noperons constructed for 2-pentanone biosynthesis. The elements in\nthis diagram are not drawn to scale. (C) Testing 2-pentanone biosynthesis\nfrom short-chain organic acids in engineered P. putida . Bacteria were individually transformed with plasmids pS2313·MK[c,s1-s3],\ncarrying the synthetic operons shown in panel (B) under control of\nthe constitutive P EM7 promoter ( Table 1 ), and incubated for 24 h in DBM medium with the different carbon\nsource combinations indicated; 2-pentanone titers were quantified\nin culture supernatants by GC-FID. (D) Plasmids for inducible expression\nof the synthetic operons for 2-pentanone biosynthesis. The synthetic\nXylS/ Pm and RhaRS/ P rhaBAD expression systems are induced by addition\nof 3-methylbenzoate and rhamnose, respectively; both plasmids carry\na streptomycin-resistance determinant (Str R ). (E) Exploring\nstrain performance by growing P. putida SEM1.3 with the selected plasmids in DBM medium containing 3 g L –1 butanoate for 24 h. 2-Pentanone titers in the culture\nsupernatant were quantified by GC-FID; methyl ketone titers in strains\ncarrying inducible expression systems are compared to the constitutive\nexpression mediated by the P EM7 promoter. Results shown in panel (C) and (E) correspond to\nmean values ± standard deviations from independent biological\ntriplicates. Significance levels of 2-pentanone titers when compared to\ncontrol conditions [pathway variant 1, spanning\nthe canonical set of enzymes, for panel (C), and constitutive gene\nexpression mediated by the P EM7 promoter for panel (E)] are indicated as follows: * P -value <0.05, ** P -value <0.01 and\n*** P -value <0.001. Since all enzymes within the canonical biosynthesis\nroute stem\nfrom a Gram-positive bacterium, we expanded the biochemical toolset\nfor short-chain ketone production by harnessing activities from species\nphylogenetically closer to our host. The PhaA thiolase from Cupriavidus necator and the AtoDA and PcaIJ CoA transferases\nfrom E. coli MG1655 and P. putida KT2440, respectively, were considered for\npathway design. The broad-host-range plasmids pS2313·MKc, MKs1,\nMKs2 and MKs3 ( Table 1 ) were constructed with different combinations of genes encoding\nall pathway enzymes under transcriptional control of the constitutive P EM7 promoter. A synthetic ribosome\nbinding site 63 (RBS) was incorporated upstream\neach coding sequence ( Figure 2 B), and all plasmids were formatted according to the Standard\nEuropean Vector Architecture (SEVA). 64 According\nto the plasmid nomenclature adopted in our study, constructs involving\ncanonical pathway components are identified with a c letter (e.g., plasmid pS2313·MKc), whereas those bearing synthetic\npathway variants are labeled with an s letter (e.g.,\nplasmid pS2313·MKs1, where the canonical ctfAB genes are replaced by atoDA from E. coli , Table 1 ). The resulting plasmids (pS2313·MKc, MKs1, MKs2\nand MKs3) were individually transformed into P. putida SEM1.3, and the ability of the engineered strains to produce 2-pentanone\nwas tested in shaken-flask cultures using different carbon substrates.\nCultures were grown for 24 h in DBM medium containing either 3 g L –1 of butanoate, 2 g L –1 of acetate\nor a combination of 3 g L –1 of butanoate and 2 g\nL –1 of acetate as the carbon substrate (s). By the\nend of the cultivation, 2-pentanone titers and the concentration of\nany residual carbon source(s) were determined in culture supernatants.\nFull consumption of the carbon substrate(s) was observed under all\nconditions. Acetate did not promote MK biosynthesis when used as a\nsole carbon substrate regardless of the biosynthetic pathway borne\nby strain SEM1.3 ( Figure 2 C). Constructs 1 (containing thl , ctfAB and adc ) and 3 (spanning thl , atoDA and adc ), in\ncontrast, mediated the highest 2-pentanone titers (3.2 mg L –1 and 4.5 mg L –1 , respectively) when butanoate was\nused as the feedstock. Cultures incubated in the presence of both\nacetate and butanoate grew to higher cell densities than those added\nwith the individual organic acids, but 2-pentanone titers were not\nsignificantly different—suggesting that, under these conditions,\nbutanoate was largely responsible for promoting ketone formation.\nThe higher 2-pentanone titers attained\nby P. putida SEM13 carrying plasmid\npS2313·MKs1 (construct 3) indicate that the AtoDA was the most\nefficient transferase variant. This feature is likely connected to\nthe substrate affinity of AtoDA, which is 22-fold higher than that\nof CtfAB ( Km for acetate = 53.1 mM and 1,200\nmM, respectively). 65 Considering\nthe trade-offs between growth and production, 66 − 68 which compete\nfor essential resources at the level of the acetyl-CoA\nnode, 69 we evaluated whether regulated\nexpression of the pathway genes could lead to higher ketone titers.\nThe performance of strain SEM1.3 carrying constructs 1 and 3 under\nthe control of different inducible expression systems was assayed\nfor this purpose. A number of expression systems are available for\nengineering P. putida ; including both\nnative Pseudomonas regulatory elements (e.g., XylS/ Pm , AlkS/ P alkB and NahR/ P sal ) 28 and heterologous modules (e.g., AraC/ P araB and RhaRS/ P rhaBAD ). 70 The XylS/ Pm and RhaRS/ P rhaBAD expression cassettes, inducible by 3-methylbenzoate\n(3- m Bz) and rhamnose, respectively, were selected\nowing to their reportedly tight inducer dependency 71 and the high levels of gene expression. 70 Hence, plasmids pS438·MKc and MKs1 (based on the XylS/ Pm system) and pS4318·MKc and MKs1 (based on the RhaRS/ P rhaBAD system) were constructed\nfor regulated expression of the best-performing pathway designs (constructs\n1 and 3, Table 1 );\nthe physical map of these plasmids is displayed in Figure 2 D. These four new plasmids\nwere individually introduced in P. putida SEM1.3, and the corresponding cultures were grown for 24 h in DBM\nmedium with 3 g L –1 of butanoate, assessing the\n2-pentanone titers and residual carbon source concentrations by the\nend of the incubation ( Figure 2 E). The two inducers, 3- m Bz and rhamnose,\nwere added at the onset of the cultivation (4 h) at 1 mM and 5 mM,\nrespectively. In general, inducible expression of the different production\npathways led to enhanced ketone titers when compared to constitutive\ngene expression. This positive effect was even more evident in cultures\nof P. putida SEM1.3 carrying construct\n3 (which includes the AtoDA transferase), with a 2-pentanone titer\nof 25.3 ± 0.9 mg L –1 paired to full substrate\nconsumption. We also investigated whether the activation of\nbutanoate into butyryl-CoA\ncould represent a metabolic bottleneck. 72 Previous experiments depended on the activity of the endogenous\nacetyl-CoA synthases (Acs) of P. putida (i.e., Acs-I and Acs-II), 73 necessary\nfor generating the CoA-ester derivatives of the short-chain organic\nacid substrates. To investigate the potential of a heterologous acetyl-CoA\nsynthetase in supporting butyryl-CoA formation, a well-characterized\nand highly efficient Acs from Bacillus subtilis ( 74 ) was integrated into pathway designs,\nresulting in plasmids pS4318·MK(c,s1)·Acs ( Table 1 ). Testing 2-pentanone biosynthesis\nby P. putida SEM1.3 harboring these\nconstructs, however, did not yield a substantial enhancement in product\ntiters from butanoate (data not shown). We concluded that the activity\nof the endogenous Acs-I and Acs-II is probably sufficient for sustaining\nketone biosynthesis, and the original pathway designs were retained\nfor further experiments. Establishing Asymmetrical Cell Division in P.\nputida to Produce Minicells Based on the\nMK production capabilities of P. putida , along with its inherent tolerance to toxic chemicals, we hypothesized\nthat minicells could be adopted as an alternative system for ketone\nbiosynthesis. Minicells are nanosized (100–400 nm in diameter)\nachromosomal bacteria, formed by abortive cell division in the mother\ncell poles. 23 Minicells cannot grow or\nduplicate, but they can support other vital cellular processes, e.g.,\nATP synthesis, replication and transcription of plasmid DNA and mRNA\ntranslation. 23 To the best of our knowledge,\nthe production and engineering of P. putida minicells have not been reported thus far. We started by exploring\nthe cell division mechanisms that could mediate abnormal bacterial\nsegregation in this species. The normal process of cell division in E. coli and other rod-shaped bacteria, triggered\nby the formation of the Z -ring, 22 yields two equally sized daughter cells 75 ( Figure 3 A). In Gram-negative bacteria, the dynamics of the Z -ring assembly and location are regulated by the Min system. 76 MinD binds to the polar membrane to form a polymer,\nand the MinCD-complex inhibits FtsZ polymerization of by recruiting\nMinE—interfering with the membrane assembly at an abnormal\nlocation. 24 Abnormal cell division ensues\nif the Z -ring is either formed at the cell pole or\nif the FtsZ division protein is overproduced, leading to minicell\nformation. 77 The Min system, thoroughly\ncharacterized in E. coli and B. subtilis , is widespread in bacterial species 19 including Pseudomonas . In strain\nKT2440, the minCDE genes form a cluster in the PP_1732-PP_1734 locus, with the gene encoding the septum\nsite-determining protein MinC as the last element in the sequence. 49 Previous experiments from our laboratory indicated\nthat transcriptional interference on minD ( PP_1733 , encoding the ATPase component of the Min system)\nwith a CRISPRi system tailored for Pseudomonas species 78 leads to the emergence of minicells. Figure 3 Production\nand characterization of P. putida minicells.\n(A) Key elements involved in cell division in Gram-negative\nbacteria. In normally shaped cells, the MinC–MinD–MinE\nsystem and the cell wall structural, actin-like MreB protein establish\na dynamic equilibrium with FtsZ, the division protein, to ensure proper\nseptation and segregation of daughter cells. Mutations in the Min\ncomponents, e.g., the Z -ring positioning MinD protein,\nlead to minicell segregation. (B) Overview of the protocol for P. putida minicells production, purification, storage,\nand downstream applications. The minicell suspension can be either\nused immediately or stored at −20 °C or −70 °C\nupon buffer exchange. Whenever needed, the minicell preparation is\nincubated in microtiter plates under specific conditions to support\nproduct formation, and the metabolites of interest are detected by\ndedicated analytical methods. (C) Representative scanning electron\ncryomicroscopy (CryoSEM) images of strains P. putida SEM1.3 and its minicell-producing derivative. CryoSEM microscopy\nwas performed on freeze-fractured samples from these cultures, imaged\nat 15 kV in an X-Max N 150 silicon drift detector (sensor-active\narea = 150 mm 2 ; Oxford Instruments NanoAnalysis) after\na Pt-coating treatment. Parental cells and minicells are identified\nwith green and blue arrowheads, respectively; the asymmetrical segregation\nof daughter cells leads to the formation of chromosome-free minicells\nthat retain plasmid(s) from the parental bacterium. (D) Representative\nphase-contrast (bright field) and fluorescence merged images of GFP-fluorescence\nand DAPI staining of a Δ minD derivative of P. putida KT2440 carrying plasmid pS4413· msfGFP . The arrow marks the position of a bacterial minicell.\n(E) Representative thin-sectioned negative-stain micrographs (transmission\nelectron microscopy, TEM) of the same strains (wild-type, WT, and\na Δ minD mutant) shown in panel (D). The MinD-deficient P. putida strain\nwas used to enrich the minicell population through a simple three-step\nprotocol, based on sequential centrifugation ( Figure 3 B). This procedure yielded ca. 0.25 g CDW (cell dry weight) L –1 minicells from\na 1.5 g CDW L –1 bacterial suspension.\nAs an additional, optional step in the purification process, ceftriaxone,\na broad-spectrum β-lactam, was added at 100 μg mL –1 to the samples to eliminate any parental cells that\nmight be present in the minicell suspension. At the last stage of\nthe protocol, glycerol was added to the minicell suspension at 15%\n(v/v) and the preparation can be kept at 4 °C or at −20\n°C or −70 °C for extended storage. To explore\nthe minicell-forming phenotype, P. putida Δ minD and a derivative expressing the monomeric\nsuperfolder GFP (msfGFP) gene ( Table 1 ) were grown in rich 2 × YT medium at 30 °C\nand harvested after 24 h. Aliquots of these cultures, along with the\nparental strain, were centrifuged, washed and prepared for scanning\nelectron cryomicroscopy (cryoSEM), transmission electron microscopy\n(TEM) and fluorescence microscopy as indicated in Methods . CryoSEM was performed on freeze-fractured samples derived from\nthese cultures, with imaging at 15 kV in an X-Max N 150\nsilicon drift detector. Image analysis identified the segregation\nof minicells at the poles of the Δ minD mutant\n( Figures 3 C and S1A ). The resulting anucleated vesicles were\nround-shaped, with a mean diameter in the 200–400 nm range,\nsimilar to the features reported for E. coli minicells. 79 In several instances, minicells\nwere observed budding from normally rod-shaped parental bacteria ( Figure S1B ). Under these conditions, we could\nnot observe any structure akin to outer membrane vesicles, 80 smaller than minicells and typically produced\nunder stressful conditions. 81 Fluorescence\nmicroscopy revealed that the Δ minD strain carrying\nplasmid pS4413· msfGFP ( Table 1 ) produced minicells loaded with the fluorescent\nprotein but lacking chromosomal DNA, as indicated by 4′,6-diamidino-2-phenylindole\n(DAPI) staining ( Figure 3 D). We also analyzed the phenotype of a Δ minD derivative of the wild-type strain KT2440 and observed that some\nminicells retained their flagella ( Figure 3 E), highlighting the value of adopting P. putida SEM1.3 Δ minD for\nbiocatalysis. This strain lacks the flagellar machinery, 50 thereby conserving energy that can be redirected\ninto bioproduction. Based on data from the literature, 82 − 84 a Δ minD bacterial cell typically undergoes\n3–7 normal divisions before malfunctioning; hence, 12–28\nminicells can be expected per mother cell. The features were exploited\nto test bioconversion of organic acids into value-added products as\nexplained in the next section. P. putida Minicells Mediate Bioconversion\nof Organic Acids into Short-Chain Ketones We tested the versatility\nof the short-chain MK biosynthesis pathway for testing the production\nof three different ketones by P. putida minicells. In particular, the enzymatic cascade composed by Thl\nand Adc from C. acetobutylicum , together\nwith AtoDA from E. coli , could be leveraged\nfor the synthesis of various (methyl) ketones by using different,\nreadily available organic acids as feedstocks (e.g., acetate, propanoate\nand butanoate; Figure 4 A–C). Sustainable methods for producing short-chain organic\nacids are attracting attention in biotechnology—a trend exemplified\nby the recent advances in C 1 assimilation processes for\nsynthesizing building-blocks (e.g., acetate). 85 − 88 Acetate, in turn, has been exploited\nas a feedstock to drive efficient acetone production, 89 showing the potential of this C 2 carboxylate\nas a sustainable carbon source for biorefinery. The core biosynthesis\npathway ( Figure 2 )\ncould be adapted to produce acetone, butanone or 2-pentanone by feeding acetate (C 2 , Figure 4 A),\npropanoate (C 3 , Figure 4 B) or butanoate (C 4 , Figure 4 C) as the\nmain substrate, respectively. In all cases, the theoretical (methyl)\nketone yield on the organic acid substrate is Y P/S = 50% mol mol –1 . At the same time, we\nassessed whether a genome-reduced background influenced ketone production\nby comparing production formation in both P. putida KT2440 and SEM1.3 ( Figure 4 D). No significant differences were found for butanone and\n2-pentanone, while increased substrate consumption and higher acetone\ntiters were observed in experiments with strain KT2440 ( Figure 4 D). Figure 4 Exploring (methyl) ketone\nbiosynthesis by P. putida . (A–C)\nSynthetic pathways tested for the bioconversion of\ndifferent organic acids into short-chain (methyl) ketones by P. putida minicells. The selected pathways mediate\nthe production of acetone from acetate (C 2 ) (A), butanone\nfrom propanoate (C 3 ) (B), and 2-pentanone from butanoate\n(C 4 ) (C). The theoretical yield for all (methyl) ketones\nshown in the diagram is Y P/S = 50% mol mol –1 . Abbreviations: CoA , coenzyme A; PPi , inorganic pyrophosphate.\n(D) Whole-cell biocatalysis experiments with P. putida SEM 1.3 and KT2440 were performed in the presence of acetate, propanoate\nor butanoate as the substrates for short-chain (methyl) ketone (MK)\nproduction. In all cases, P. putida strains were incubated with 50 mM of selected organic acid and 15%\n(v/v) glycerol. Both organic acid (OA) consumption and product (acetone,\nbutanone and 2-pentanone) titers were determined after 24 h. Results\ncorrespond to mean values ± standard deviations from three independent\nbiological triplicates. Based on these observations,\nminicells derived\nfrom P. putida SEM1.3 Δ minD transformed\nwith plasmid pS4318·MKs1 were used in bioconversion experiments\nby incubating the suspension with the three organic acids (i.e., acetate,\npropanoate and butanoate) at 50 mM. After 24 h, both ketone production\n( Figure 5 A) and substrate\nconsumption ( Figure 5 B) were assessed in the supernatants. Acetone titers reached 211\n± 18 mg L –1 , whereas the maximum butanone and 2-pentanone concentrations observed\nunder the same\nconditions were 30 ± 3 mg L –1 and 24 ±\n6 mg L –1 , respectively ( Figure 5 A). The profile of short-chain MK production\nby minicells followed a trend similar to that of carboxylate utilization\n( Figure 5 B). Under\nthese conditions, between 11% and 52% of the carboxylates were consumed\nby the P. putida minicells, with a\npositive correlation between chain length and substrate consumption\n( Figure 5 B). Figure 5 Short-chain\n(methyl) ketone production by P. putida minicells. (A) Bioconversion\nexperiments performed using acetate, propanoate or butanoate as the\nsubstrate for short-chain (methyl) ketone biosynthesis by minicells. P. putida minicells were incubated in the presence\nof the selected organic acid (50 mM) and 15% (v/v) glycerol. Both\nproduct (acetone, butanone and 2-pentanone) titers (A) and substrate\n(acetate, propanoate and butanoate) consumption (B) were determined\nafter 24 h. (C) The effect of glycerol on bioconversion of organic\nacids into (methyl) ketones was assessed after 18 h of incubation\nwith 15% (v/v) glycerol and/or 50 mM of the corresponding organic\nacid substrate (acetate, propanoate or butanoate). Ctrl ., control. Substrate consumption by purified P. putida minicells under the same experimental conditions is indicated in\npanel (D). Acetate, propanoate or butanoate could not be detected\nin samples that were not supplemented with the corresponding substrates;\ntherefore, these experiments are not included in the figure. (E) Acetone\ntiters were used as a proxy for evaluating production phenotypes in P. putida minicells after storage at −70 °C\nin the presence of 15% (v/v) glycerol. Engineered minicells, prepared\nas indicated in Figure 3 B, were tested for their ability to mediate the bioconversion of\nacetate (50 mM) into acetone. Acetone titers were compared to a control\n(Ctrl.) experiment conducted with freshly prepared P. putida minicells. Results shown in panels (A–E)\nrepresent mean values ± standard deviations from independent\nbiological triplicates. (F) Plasmid DNA content in P. putida SEM1.3 and its minicell-producing derivative.\nLyophilized samples (corresponding to 6.5 mg cell dry weight) were\nused to isolate plasmid DNA with a commercial kit; plasmid DNA content\nis expressed in ng DNA mg cell dry weight –1 . Results\ncorrespond to mean values ± standard deviations from three independent\nbiological triplicates; significance levels are indicated with ** P -value <0.01. Evaluation of Operating Conditions for Short-Chain (Methyl)\nKetone Biosynthesis by P. putida Minicells The minicell suspension, prepared as outlined in Figure 3 B, was used to demonstrate\nthe capacity of the anucleated bacterial vesicles for biotransformation\nof organic acids into methyl ketones. To investigate factors affecting\ncatalytic activity, we conducted a series of experiments focusing\non incubation conditions and metabolic stability of the minicells.\nIn this sense, the final step of the isolation protocol involved adding\nglycerol at 15% (v/v) as a cryoprotectant for extended storage. Since\nglycerol can also serve as an energy and carbon source for P. putida , 90 we tested\nthe potential contribution of this additive to product formation.\nGlycerol alone did not support acetone biosynthesis, and traces of\nbutanone and 2-pentanone could be detected in the absence of organic\nacid substrates ( Figure 5 C). In all cases, we observed an additive effect of glycerol and\nthe main substrate in promoting product formation. This observation\nwas mirrored in the consumption of organic acids during the bioconversion\nexperiments ( Figure 5 D). When glycerol was present in the incubation medium, substrate\nconsumption decreased (with butanoate being particularly affected).\nThese observations suggest a role for the polyol in supporting cellular\nprocesses, e.g., redox balancing, 91 beyond\nthe expected cryoprotectant effect. A relatively unexplored\nadvantage of minicells for bioproduction is their catalytic stability,\nas they do not require resource investment or energy expenditure for\ncell division. We systematically assessed the stability of P. putida minicells after enrichment and purification\n( Figure 3 B) by storing\naliquots of the minicell suspension at −70 °C for extended\nperiods (from 10 to 70 days). Periodically, individual minicell aliquots\nwere retrieved from the freezer and used in 24-h acetone bioconversion\nexperiments as described above. To establish a baseline, the ability\nof freshly purified minicells to convert acetate into acetone was\nanalyzed immediately after purification (control condition, Figure 5 E). Acetone titers\nshowed relatively limited variability across all samples, with >50%\nof the acetone concentration retained in experiments performed with\nminicells stored for up to 70 days ( Figure 5 E). Interestingly, storing the minicells\nat −20 °C during the same period did not result in any\nsignificant decrease in product titers (data not shown). The parental\ncells had a much lower catalytic activity (<10%) upon 10 days of\nstorage at −70 °C. Taken together, these results highlight\nthe metabolic stability and robustness of P. putida -derived minicells and demonstrate their potential to produce toxic\nchemicals even after prolonged storage. In connection with the\nstable production phenotypes observed in\nengineered minicells, we studied their plasmid DNA content. To this\nend, the pathway designed and optimized for 2-pentanone biosynthesis\n(borne by plasmid pS4318·MKs1, Figure 2 ) was transformed into P.\nputida SEM1.3 Δ minD . We reasoned\nthat the asymmetrical, abortive division in this mutant would lead\nto the formation of minicells carrying plasmids, but no chromosomal\nDNA ( Figure 3 C). We\ncompared the plasmid DNA content in minicells of P.\nputida SEM1.3 Δ minD , prepared\nas indicated in Figure 3 B, and the parental strain. The plasmid DNA content in the minicells\nwas 901 ± 19 ng mg CDW –1 , which was\n∼20% higher than that of the parental cells ( Figure 5 F). The stable maintenance\nof plasmids encoding the MK biosynthesis pathway helps to explain\nthe observed production phenotypes. During bioconversion experiments\nusing minicells, the acetone titers achieved corresponded to a Y P/S of ca. 28% of the theoretical maximum ( Figure 5 A), calculated based\non acetate consumption ( Figure 5 B). The butanone and 2-pentanone levels reached in these\nexperiments were higher than those attained\nby whole-cell biocatalysis ( Figures 2 E, 4 D and 5 A). This result is probably linked to increased stress tolerance,\nas previously suggested for E. coli minicells. 92 Taken together, these observations\nindicate that the minicells retain the biosynthetic capabilities of\ntheir nucleated parental cells. Furthermore, these P. putida nanoreactors were able to produce all three\ntarget compounds even without further optimization steps (e.g., enhancing\nprecursor channeling). 93"
} | 12,078 |
35848130 | PMC9796341 | pmc | 7,056 | {
"abstract": "Abstract Diversity of viruses infecting non‐extremophilic archaea has been grossly understudied. This is particularly the case for viruses infecting methanogenic archaea, key players in the global carbon biogeochemical cycle. Only a dozen of methanogenic archaeal viruses have been isolated so far. In the present study, we implemented an original coupling between stable isotope probing and complementary shotgun metagenomic analyses to identify viruses of methanogens involved in the bioconversion of formate, which was used as the sole carbon source in batch anaerobic digestion microcosms. Under our experimental conditions, the microcosms were dominated by methanogens belonging to the order Methanobacteriales ( Methanobacterium and Methanobrevibacter genera). Metagenomic analyses yielded several previously uncharacterized viral genomes, including a complete genome of a head‐tailed virus (class Caudoviricetes , proposed family Speroviridae , Methanobacterium host) and several near‐complete genomes of spindle‐shaped viruses. The two groups of viruses are predicted to infect methanogens of the Methanobacterium and Methanosarcina genera and represent two new virus families. The metagenomics results are in good agreement with the electron microscopy observations, which revealed the dominance of head‐tailed virus‐like particles and the presence of spindle‐shaped particles. The present study significantly expands the knowledge on the viral diversity of viruses of methanogens.",
"conclusion": "CONCLUDING REMARKS Predicting the hosts of viruses is a major bottleneck in microbial ecology. In recent years, besides the classic culture‐dependent approaches, several promising methods have been developed to identify the hosts for viruses within complex microbial communities, such as PhageFISH (Allers et al., 2013 ; Barrero‐Canosa & Moraru, 2019 ), Meta3C (Marbouty et al., 2014 ), viral tagging (Deng et al., 2014 ) or epicPCR (Sakowski et al., 2021 ; Spencer et al., 2016 ). These single cell level methods generated important new knowledge, but they also suffer from limitations, either due to the requirement of a pre‐existing knowledge on the viral genome sequence, of the capability to cultivate the host, or due to their complexity. Several purely bioinformatic approaches of host prediction have been also developed (Coclet & Roux, 2021 ; Edwards et al., 2016 ). However, the accuracy of in silico methods is limited, in most cases, by the lack of microbial genomes closely related to that of the true host. Viral host prediction, therefore, remains challenging for metagenomics studies. The original experimental approach developed in the present study is not strictly speaking a method for the identification of viral hosts but it still presents some advantages in this perspective. Relying on SIP enabled us to discriminate between the active microorganisms involved in formate metabolism and the other ones. Although this substrate was not consumed exclusively by methanogens, the use of formate strongly enriched the community in methanogens and their associated viruses, resulting in a simplified microbial community to study and, possibly, an improved MAG assembly. Coupling the results from different bioinformatic methods (taxonomic assignation of the contigs, alignment‐ and signature‐based host prediction), we were thus able to detect several previously uncharacterized genomes of viruses infecting methanogens, with high accuracy, and to determine whether they targeted methanogens actively involved in the formate bioconversion. It was especially possible thanks to the establishment of specific host databases, through the shotgun sequencing of cellular metagenomes. Interestingly, one of them likely corresponds to a new archaeal virus family within the Caudoviricetes (proposed name Speroviridae ), and two others to putative new families of spindle‐shaped archaeal viruses. These results significantly expand the knowledge on the diversity of viruses of methanogens, since no spindle‐shaped virus has been reported until now for the orders Methanobacteriales and Methanosarcinales. It illustrates the complementarity between SIP, metagenomics and specific bioinformatic tools for host‐virus analysis in complex microbial communities. Our approach can be viewed as complementary to the available experimental methods for host identification, and could likely be combined with most of them. In particular, PhageFISH or epicPCR would be nicely complementary as they require at least a partial knowledge of the viral sequence, and they would enable to fully confirm the host identity. To sum up, we proposed an original coupling between SIP, an experimental method, and metagenomic analyses with complementary bioinformatic tools to identify viruses of methanogenic archaea within anaerobic digestion microcosms. It enabled successful enrichment of the microbial community in methanogenic archaea, and their labelling with 13 C. The in silico approach we developed led to the identification of several dozens of contigs predicted to originate from viruses infecting methanogenic archaea. Their analyses through gene‐sharing network and comparative genomics highlighted the dominance of the Caudoviricetes class, with the discovery of a previously uncharacterized siphovirus infecting Methanobacteriales hosts and belonging to a new suggested viral family, Speroviridae . It also led to the discovery of two spindle‐shaped viruses representing two new putative families, not previously reported for the orders Methanobacteriales and Methanosarcinales. Our results significantly expand the knowledge on the diversity of viruses of methanogens and reinforce the notion of the wide environmental and phylogenetic distribution of spindle‐shaped viruses in Archaea (Krupovic et al., 2020 ). Our original experimental approach enables to identify viruses infecting key functional groups contributing to biogeochemical fluxes in communities of uncultured microbes. It can be invaluable for the study of viruses infecting metabolically active microorganisms in virtually any type of complex microbial community.",
"introduction": "INTRODUCTION Viruses infecting archaea represent one of the most unique parts of the global virosphere (Krupovic et al., 2018 ). Despite the limited number of archaeal viruses described so far compared to bacterial viruses, archaeal viruses show a great diversity of gene content and morphological properties. In particular, several morphotypes are specific to archaeal viruses, showing no similarity to viruses infecting bacteria and eukaryotes, such as bottle‐shaped ( Ampullaviridae ), coil‐shaped ( Spiraviridae ), or spindle‐shaped ( Bicaudaviridae , Fuselloviridae , Halspiviridae , Thaspiviridae ) ones. Archaeal viruses are currently classified into 33 families (Baquero et al., 2020 ; Liu et al., 2021 ), including cosmopolitan icosahedral viruses and archaea‐specific viruses. Methanogenic archaea play a major role in carbon cycling at the global scale, through methanogenesis. Known methanogenic archaea are currently grouped into eight different orders in the phyla Euryarchaeota and Halobacteriota, and a few candidatus taxa in the Euryarchaeota, Halobacteriota, and in the TACK group (Evans et al., 2019 ; Lyu et al., 2018 ) These were isolated from very diverse natural ecosystems such as wetlands, termite, human and livestock digestive tracts, rice fields, and deep‐sea hydrothermal vents, and also from anaerobic digesters (ADs) (Lyu et al., 2018 ). Indeed, their unique metabolic features are exploited in AD processes (Ahring, 2003 ) for valorization of organic waste and effluents into methane‐rich biogas, a renewable energy source. Among archaeal viruses, 10 have been reported to infect methanogenic archaea (methanogens) (Krupovič et al., 2010a ; Meile et al., 1989 ; Molnár et al., 2020 ; Nölling et al., 1993 ; Pfister et al., 1998 ; Thiroux et al., 2021 ; Weidenbach et al., 2017 ; Weidenbach et al., 2021 ; Wolf et al., 2019 ; Wood et al., 1989 ). In addition, several proviruses integrated in the genomes of diverse methanogens have been described (Krupovič et al., 2010 ; Krupovič & Bamford, 2008 ). Almost all of the viruses of methanogens described so far have been isolated from AD samples (Table S1 ). A total of five head‐tailed viruses or proviruses originate from thermophilic ADs, all infecting Methanobacteriales hosts: ΨM1 and ΨM2 (family Leisingerviridae ; siphovirus morphology) (Liu et al., 2021 ; Meile et al., 1989 ; Pfister et al., 1998 ) and related defective provirus ΨM100 (Luo et al., 2001 ) infect Methanothermobacter strains, whereas ΦF1 (unclassified) and ΦF3 (unclassified; siphovirus morphology) (Nölling et al., 1993 ) infect Methanobacterium species. Moreover, four viruses or virus‐like particles (VLPs) have been isolated from mesophilic ADs: Methanobacterium ‐infecting virus Drs3 (family Anaerodiviridae ; siphovirus morphology) (Liu et al., 2021 ; Wolf et al., 2019 ), Blf4 (unclassified; siphovirus morphology), infecting Methanoculleus strains (Methanomicrobiales) (Weidenbach et al., 2021 ), MetSV (unclassified) (Weidenbach et al., 2017 ), a spherical virus infecting Methanosarcina strains (Methanosarcinales), and finally the oblate or spindle‐shaped Methanococcus (Methanococcales)‐infecting A3 VLPs (Wood et al., 1989 ). MFTV1, an unclassified temperate head‐tailed virus (siphovirus morphology), has been induced from a Methanocaldococcus fervens AG86 strain (Methanococcales) isolated from deep‐sea hydrothermal vents (Thiroux et al., 2021 ); it is the first characterized virus infecting hyperthermophilic methanogens. In addition, MetMV (unclassified) (Molnár et al., 2020 ) has been suggested to infect mesophilic Methanosarcina strains, but this still needs to be confirmed. For the other four known orders of methanogens (Evans et al., 2019 ), no viruses have been isolated so far. Thus, the diversity of viruses infecting methanogenic archaea remains largely unexplored. Metagenomics can provide a less biased view on the diversity of viruses infecting methanogens, by circumventing the challenge of cultivating some of the methanogenic archaeal strains, as well as biases associated with virus isolation. In such context, AD ecosystems prove to be particularly well suited as they are relatively easy to access to, and to establish in laboratory reactors. Moreover, they encompass methanogenic archaea from several orders, such as Methanobacteriales, Methanomicrobiales and Methanosarcinales (Evans et al., 2019 ; Lin et al., 2016 ). Yet, identifying viruses infecting methanogens is challenging in AD metagenomes due to the complexity of the catalytic microbial communities, dominated by diverse bacteria, and due to usually low proportions of methanogenic archaea in AD processes. In the present study, an original experimental approach was applied with the aim of favouring the enrichment of AD microbial communities in methanogens, to help the discovery of their viruses in metagenomes. To this end, AD microcosms were fed with 13 C‐labelled formate, one of the known substrates for methanogenesis through the following equation (Sun et al., 2021 ): \n 4HCOO − + H + + H 2 O → CH 4 + 3HCO 3 − ∆ G 0 : − 170.2 kJ / mol \n \n In addition to favouring methanogens, such an experimental approach has the advantage of preserving the AD process and a certain level of microbial diversity. In particular, it can enable to reach higher proportions of methanogens, and it is also compatible with the presence of several methanogenic species or genera, offering the possibility of a relatively broad view on the diversity of viruses of methanogens. Another benefit of this method is the possibility to identify DNA viruses targeting microorganisms that actively assimilated the substrate. Indeed, within complex microbial communities, not all of the microorganisms are active or involved in the degradation of a specific substrate. Thus, identifying active microorganisms, and their associated viruses, are key issues in linking viral diversity with functional aspects. Hence, we coupled stable isotope probing (SIP) (Radajewski et al., 2000 ), applied to the total cellular DNA, to in silico host prediction of the viral contigs. Host prediction was applied by using the cellular metagenomes obtained from the heavy ( 13 C‐enriched) and light (not enriched in 13 C) DNA sequences, which were used to build host databases. Whenever the predicted host was identified in the heavy cellular DNA fraction, we considered it as evidence that the virus infected active microorganisms, assimilating the 13 C‐formate. This original coupling (SIP and bioinformatic analyses for host prediction) led to the discovery of several previously uncharacterized genomes of DNA viruses of methanogens. In particular, they included two contigs likely representing new families of spindle‐shaped viruses, thereby expanding the knowledge on the diversity of viruses infecting methanogens. One of these families was associated with the active hydrogenotrophic methanogenic archaea, while the other one seemed to target methanogens that were not assimilating the formate in the studied microcosms, which could only be evidenced thanks to the SIP analysis.",
"discussion": "RESULTS AND DISCUSSION A significant proportion of formate was converted to methane through hydrogenotrophic methanogenesis A total of six batch AD microcosms were established and monitored over time for a maximum of 17 days. They contained, as sole carbon source, either unlabelled formate (three replicates) or 13 C‐formate (three replicates). At each of days 8, 13 and 17, one pair of microcosms was sacrificed for metavirome analysis, one with unlabelled formate and one with 13 C‐formate. Chemical analyses (VFA, TIC, TOC) showed that both unlabelled and 13 C‐labelled formate were consumed, and more than 85% of the initial quantity was metabolized after 17 days of incubation (Supplementary Figure S1 ). The bioconversion resulted in a pH increase from ~7.50 to ~9.20 on average, consistent with the methanogenesis equation mentioned above. Biogas was produced after a lag phase of about 5 days (Supplementary Figure S1 ) to reach final cumulated productions of ~100 normo‐ml. Furthermore, reproducible dynamics were observed in microcosms, irrespective of the formate type. Methane (CH 4 ) was by far the dominant biogas component, followed by carbon dioxide (CO 2 ) as well as traces of hydrogen (H 2 ) and hydrogen sulfide (H 2 S). Collectively, these results support the successful establishment of the AD process in the microcosms. More precisely, the isotopic composition of the biogas demonstrated the dominance of 13 C in the methane (>95%) produced by the methanogens in the microcosms fed with 13 C‐formate and indicated the dominance of the hydrogenotrophic methanogenesis pathway (Whiticar et al., 1986 ) in the microcosms fed with 12 C‐formate. For unlabelled substrates, this method relies on the abundance of the stable isotope 13 C in nature (1.1%) and on the difference in reaction rate between 12 C‐ and 13 C‐containing substrate molecules. The results and detailed calculation of isotopic signatures are provided in the Supplementary Information (Section 2.1 ). The proportion of methanogenic archaea increased over time and they were dominated by Methanobacterium species To identify the methanogenic archaea and more broadly the microbial community composition, we applied 16S rRNA gene metabarcoding [Figure 1(A) ]. The relative abundance of the dominant archaea (Methanobacteriaceae, in green) increased from 1%–3% at day 0 to 26%–30% at day 17. Consistently, a functional analysis based on the cellular shotgun metagenomic data (Supplementary Figure S2 ) showed the dominance of methane metabolism, in relative abundance, at day 17. The KEGG category ‘methane metabolism’ includes methanogenesis, methane oxidation and metabolisms related to intermediate molecules of both these pathways. Among the microbial groups involved in methane metabolism [Figure 1(C) ], archaea were the main actors (Methanobacteriaceae in brown, Methanosarcinacea in green and Methanotrichaceae in cyan). Similar proportions of archaea were observed in the metagenomic dataset (Supplementary Figure S3 ). These results confirmed that the microbial communities were enriched in methanogenic archaea during the incubation. The detected archaea were mostly methanogens [Figure 1(B) ]. At day 0, Methanosarcina (Methanosarcinales order) was dominant. However, at the end of the incubation, at day 17, the genus Methanobacterium became dominant, followed by Methanobrevibacter (both from the order Methanobacteriales). It can be assumed that Methanobacteriales members were selected during the incubation since they are able to grow on formate through the hydrogenotrophic pathway, and since they likely outcompeted Methanosarcinales methanogens, due to their faster growth rate: the doubling time is generally lower than 1 h in Methanobacteriales, compared to more than 10 h in Methanosarcinales (Thauer et al., 2008 ). Concerning bacterial families present in these systems, a notable decrease in relative abundance was observed over time for Synergistaceae (phylum Synergistetes, from ~15% down to ~7%) and Syntrophomonadaceae (phylum Firmicutes, from ~24% down to ~7%) [Figure 1(A) ]. Synergistaceae members generally consume amino acids to generate short‐chain fatty acid (He et al., 2018 ), whereas Syntrophomonadaceae members are acetogens (Si et al., 2016 ). For both families, the decrease is understandable due to the lack of adequate substrate (proteins and short‐chain fatty acids). A notable increase in relative abundance was observed at day 17 for members of Anaerolineaceae (phylum Chloroflexi, from >1% up to ~4%), which are generally described as fermentatives or acetogens (Liang et al., 2015 ; Si et al., 2016 ), and some of them were previously shown to form syntrophic associations with methanogens (Lei et al., 2018 ). In the present experiment, Anaerolineaceae bacteria may degrade formate and/or play a role in electron transfer (Wang et al., 2021 ), in partnership with hydrogenotrophic methanogens. For the six microcosms, the profiles of cellular DNA concentrations in function of their mass density were established (Supplementary Figure S4 ). For microcosms fed with 13 C‐formate at day 17 [Figure 1(D) , red line], a second peak, of denser DNA, was visible in the profile, indicating that the labelled substrate was assimilated by some of the microorganisms. The separation of DNA in CsCl gradient depends not only on the mass of the isotope but also on the GC content of the DNA (Eason & Campbell, 1978 ). Hence, we pooled the DNA into three distinct fractions, according to their density. The first fraction (fr1, density <1.705) corresponded exclusively to non‐labelled DNA. The second fraction (fr2, 1.705 < density <1.738) possibly contained a mix of unlabelled and 13 C‐labelled DNA. Finally, the third fraction (fr3, density >1.738) contained exclusively 13 C‐labelled DNA. The 16S rRNA gene metabarcoding applied to these fractions showed that archaea reached their highest proportions (56%) in fraction fr3 at day 17 [Figure 1(B) ], whereas their proportions in fractions fr1 and fr2 were lower than 3%. Consistently, the highest abundance of genes involved in methane metabolism was in fraction fr3 at day 17, based on metabolic pathway analysis from the shotgun metagenomics data [Figure 1(C) ]. These observations confirmed that the 13 C‐formate incorporated into the microbial biomass was consumed mostly by methanogenic archaea of the genera Methanobacterium and Methanobrevibacter . \n TEM evidenced the presence of icosahedral and spindle‐shaped VLPs \n TEM observations revealed a great diversity of VLPs in the microcosms (Figure 2 ), especially after 13 and 17 days of incubation [Figure 2 (A,C)]. Besides cosmopolitan morphotypes common to the domains Bacteria and Archaea, uncommon and especially archaea‐specific morphotypes were also observed. FIGURE 2 Morphotypic diversity of VLPs in different microcosms, observed by TEM. One representative sample is shown for each incubation time point (A: day 8, B: day 13, C: day 17) Cosmopolitan morphotypes included VLPs with a head‐tailed morphology, typical of the class Caudoviricetes , as well as icosahedral tailless particles [Figure 2(G) ]. The latter could originate from tailless icosahedral viruses or from head‐tailed viruses with a broken tail. In both samples, these VLPs were the most abundant and presented a large diversity: myovirus‐like [i.e. icosahedral capsids with contractile tails; Figure 2(D) ], siphovirus‐like [i.e. icosahedral capsids with long non‐contractile tails; Figure 2(E) ] and also podovirus‐like [i.e. icosahedral capsids with short tails; Figure 2(F) ] morphotypes were observed, with capsid diameters ranging from 50 to 200 nm. Less common viral morphotypes were detected in the microcosms at days 13 and 17, such as rod‐shaped [Figure 2(H) ], spherical [Figure 2(I) ] and spindle‐shaped [Figure 2 (K–L)]. Spindle‐shaped morphotypes, which are specific to archaeal viruses, have been commonly observed in extreme geothermal and hypersaline environments (Krupovic et al., 2014 ), but have also been reported in moderate ones, such as freshwater and marine habitats (Borrel et al., 2012 ; Kim et al., 2019 ). Interestingly, the presence of spindle‐shaped VLPs has also been reported in AD plants (Calusinska et al., 2016 ), suggesting that the hosts of these viruses could possibly be involved in the AD process. Particles with unique morphotypes were also identified [Figure 2 (M–O)]. Some of them had a chain structure with multiple of two or three monomers and their size ranged from 100 to 400 nm. Moreover, other particles appeared as round‐shaped and less than 50 nm, with a hollow star‐like structure [Figure 2(M) ]. The nature of these particles is unclear. Overview of the shotgun metagenomics datasets and of the host prediction approaches Shotgun metagenomic sequencing was applied to seven selected DNA extracts. Four of the extracts were from the total cellular DNA and total viral DNA from samples collected at days 13 and 17 (D13cell, D13vir, D17cell, D17vir), selected due to their enrichment in methanogens observed in the metabarcoding analysis (Figure 1 ) and the presence of interesting VLP morphotypes (Figure 2 ). The three cellular DNA fractions obtained after density gradient centrifugation at day 17 were also sequenced (D17SIPfr1, D17SIPfr2, D17SIPfr3) [Figure 1(D) ]. The purpose was primarily to analyse the metaviromes, and to use the cellular metagenomes for building specific databases for host prediction. Based on classical metrics (Table 1 ), the sequencing data and the assemblies were of satisfactory quality. In the case of metaviromes, ViromeQC analysis indicated moderate contamination rates of 7.6% and 8.8% for D13vir and D17vir, respectively. Among 23,016 assembled contigs longer than 1 kb, 1927 were exclusively detected in D17vir, suggesting that most of the corresponding viruses were selected during the late stages of the incubation. Among the 87.59% of contigs that obtained a taxonomic annotation, 98.43% were affiliated to prokaryotes, which can be explained primarily by database biases: prokaryotes were more represented than their viruses in the NCBI nr database used for the taxonomic assignment, and many viral genes could therefore have their best match in prokaryotic genomes, in particular in proviruses. Such bias is a strong limitation for the identification of viruses and their taxonomic classification. Yet, it can bring information on the possible hosts of the viral contigs. TABLE 1 Overview of the shotgun metagenome datasets Cellular metagenomes (individual assemblies) Viral metagenomes (co‐assembly) D13cell D17cell D17SIPfr1 D17SIPfr2 D17SIPfr3 D13vir D17vir Number of raw reads (millions) 85.68 63.76 75.41 78.78 65.54 38.92 46.48 Reads obtained after trimming (%) 98.39 98.08 97.80 98.35 98.40 98.41 98.61 Number of contigs (≥1 kb) 77,032 68,382 58,308 33,123 21,159 23,016 Number of contigs (≥3 kb) 18,093 15,244 14,728 9432 5108 5571 Max contig length (b) 622,910 622,910 449,342 388,850 389,221 372,273 Contigs with taxonomic affiliation (%) 96.62 96.67 96.93 97.32 97.22 87.59 Reads mapped to contigs (%) 89.35 87.74 89.03 93.17 93.73 87.20 79.85 Number of MAGs 81 67 75 55 28 – Number of archaeal MAGs 8 10 8 7 5 – Number of selected MAGs (≥60% completeness and ≤5% contamination) 81 (including 17 from archaea) Viral genome detection was performed with CheckV, VIBRANT and VirSorter2. All these tools rely on protein similarity profiles and both latter detect protein profiles by machine learning. CheckV, VirSorter2 and VIBRANT identified 3898, 3968 and 3904 viral genomes (6484 distinct genomes in total), respectively, with most of the genomes being linear, double stranded and predicted to belong to virulent viruses. These numbers were low compared to the number of contigs longer than 1 kb, likely reflecting the fact that shortest contigs contain insufficient information for confident virus identification. For cellular metagenomes, a taxonomic affiliation was obtained for more than 96% of the contigs, suggesting that they would constitute a good‐quality database of host sequences. MAGs were reconstructed by binning of the cellular contigs: in total, 81 MAGs (completeness >60% and contamination <5%) were selected as reference hosts, including 17 assigned to the domain Archaea. For host prediction, two complementary methods were used. The first one relied on matching of CRISPR spacers to the viral contigs. Spacers are short sequences (26–72 bp) (Makarova et al., 2011 ) of plasmid or viral origin, which are integrated into the host genomes by CRISPR‐Cas (Nasko et al., 2019 ), adaptive immunity systems identified in 85% of Archaea and 40% of Bacteria (Makarova et al., 2020 ). The second approach relied on signatures, as most prokaryotic viruses have a genome k‐mer composition similar to the one of their hosts, due to their co‐evolution (Edwards et al., 2016 ). Similar genomic signatures were searched between viral contigs and the 81 microbial MAGs as potential hosts, using WIsH (Galiez et al., 2017 ). For contigs longer than 3 kb, the spacer‐based method revealed a total of 375 contigs with a predicted host, whereas the signature‐based method against our homemade MAG database predicted an acceptable host ( p ‐value ≤0.05) for 3239 contigs. Coupling of these methods resulted in a total of 3466 contigs with a predicted host, showing their complementarity to improve the capability and accuracy of in silico host prediction. The Caudoviricetes class dominated among the 39 contigs likely originating from viruses of methanogens As our aim was to identify contigs of viruses infecting methanogenic archaea, we applied successive stringent filters and manual curation, relying on the integration of results from complementary bioinformatic analyses (Figure 3 ). A first filter based on contig length was applied to eliminate contigs which would contain only limited information. The second filter aimed at removing an important fraction of the cellular contaminants. In the third step, contigs possibly originating from archaea or archaeal viruses were selected. Finally, the contigs obtained at this stage were analysed manually, to remove most remaining contaminant contigs (typically originating either from cells or from bacterial viruses). FIGURE 3 Strategy for the selection of contigs of interest, likely originating from viruses of methanogens A total of 39 contigs were thereby selected as contigs of interest for further analysis, as putatively originating from archaeal viruses. Due to the limited accuracy of some host prediction tools [e.g. of the order of 70%–80% at the phylum level (Galiez et al., 2017 )], this selection of contigs still possibly contained a few contaminants. The selected contigs ranged in lengths from 4.2 to 53 kb (median: 9.4 kb) and the reads per kilobase per million mapped reads (RPKM) values from 0.07 to 468.27 for D13vir, and from 0.20 to 229.40 for D17vir. Nine of these contigs could not be assigned to any known taxon (either cellular or viral), suggesting them to be previously uncharacterized. Two contigs, C1661 and C1697, had best‐predicted hosts as MAGs from the unlabelled metagenome D17SIPfr1, suggesting that they represent viruses infecting inactive and minor methanogens in our microcosms. Besides, 34 were predicted as virulent by VIBRANT and only one, C1485, was predicted as temperate (Supplementary Figure S5 , Supplementary Table S5 ). Genome scaffold quality results were obtained through three different bioinformatic tools: VIBRANT, CheckV and VirSorter2: 35 contigs had low or medium quality. However, given that all of these tools were developed based on the viral databases overwhelmingly dominated by bacterial Caudoviricetes , their accuracy on datasets including novel archaeal viruses, especially those with small or medium‐sized genomes, is not to be expected. All the results from the bioinformatics tools are available for this set of 39 contigs, in Supplementary Table S5 . A bipartite network was built (Figure 4 ) to evaluate the similarity between the contigs of interest and the archaeal (pro)viruses described in the literature. Such networks enable to represent complex systems comprising two distinct classes of components (nodes) (Iranzo et al., 2016b ). Here, the two classes of node correspond to viral contigs or genomes on the one hand, and to protein OGs on the other. Two viral nodes can be connected only indirectly, through shared OG nodes. Previously published archaeal pro(viral) genomes were labelled according to their taxonomic affiliation, when established. The top half of the network contained viral genomes from (pro)viruses with a cosmopolitan head‐tailed morphotype (siphovirus, myovirus, magrovirus, provirus). In the bottom half, the presence of various viral families specific to archaea was observed. Such a network topology is consistent with the spatial distribution in archaeal virus networks previously described in the literature (Iranzo et al., 2016a ). Moreover, head–tail viruses (class Caudoviricetes ) are known to be highly mosaic (Krupovič et al., 2010 ), hence generating tightly interconnected networks. FIGURE 4 Bipartite network of known archaeal viruses and of the 39 contigs of interest, and of protein orthologous groups (OGs). Most of the contigs of interest originate from archaeal viruses. The genomes and contigs are represented as circles coloured according to the legend shown in the figure, and OGs are denoted by the intersections of edges. Some shared OGs with a functional annotation are represented. TerL: Terminase large subunit, MCP: major capsid protein, mCP: minor capsid protein, Exo: exonuclease, TP: tail protein, PP: portal protein, Lig: ligase, DNApol, DNA polymerase, BH: baseplate hub protein, Heli: helicase, Trans: transposase, TR transcription regulator Out of 39 contigs of interest (green), 24 shared at least one OG with (pro)viruses belonging to the class Caudoviricetes , including virion morphogenesis proteins, such as terminase large subunit (TerL), various tail proteins, major/minor capsid proteins (MCP/mCP) and baseplate hub proteins (BH). To further explore the organization of the network and relationships among the 39 contigs of interest and archaeal (pro)viruses, they were clustered according to the presence/absence of shared genes (see Experimental procedures ). Consistent with the network analysis, the two main viral clusters (VC) were related to the class Caudoviricetes (Supplementary Figure S6 ). The first cluster, VC4, included four contigs, C1803, C174, C300 and C616, and was held together through OGs related to the capsid formation and packaging module, and/or DNA, RNA and nucleotide metabolism (Supplementary Figure S6 ). Moreover, in the longest contig of VC4, C174, a sheath protein was detected, also suggesting its affiliation to viruses with contractile tails (myovirus‐like morphology) (Fokine & Rossmann, 2014 ). For all the contigs in this cluster, only WIsH predicted an archaeal host, whereas the other tools employed showed either a bacterial host or no identified one. To ascertain the host assignment, the taxonomic and functional annotations of each gene from the cluster were examined: no single gene was assigned to Archaea or annotated as an archaeal protein, while many were related to Bacteria, suggesting that bacterial hosts were more probable. It highlights the importance of relying on different complementary tools for more accurate host prediction. The second cluster, VC10, included five contigs, C1666, C1908, C158, C673 and C1006. Several OGs shared among members of the cluster were related to various viral functional modules: virion morphogenesis (including capsid and tail formation, and genome packaging), DNA/RNA and nucleotide metabolism, as well as integration and excision. In particular, C158 seems to be closely related to C673 and C1908 as they share respectively six and two OGs. VC10 was located near the siphovirus nodes (cyan) in the network. Importantly, the taxonomic affiliation of these contigs and their host prediction by different tools were consistent with each other and confirmed that they originated from archaeal viruses, unlike contigs from VC4. Contigs located on the periphery of the network shared very few OGs with other known viruses (only one or two OGs per contig). Moreover, those shared OGs were often uncharacterized or hypothetical proteins. Several contigs shared annotated OGs belonging to viral families specific to archaea, and they could have originated from previously uncharacterized viruses of methanogens. For example, C775 shared an OG annotated as helicase with eight genomes belonging to the family Lipothrixviridae . However, only the presence of signature genes, that is characteristic of particular virus groups, such as those encoding major structural proteins, can provide a reliable taxonomic affiliation of contigs, as it is the case for VC4 and VC10 contigs. For contigs outside of these two VC, such signature genes were identified only for two contigs (C1359 and C1697). Indeed, a structural protein typical of archaeal spindle‐shaped viruses (Krupovic et al., 2014 ) was identified in these two contigs, suggesting that they represent new families of spindle‐shaped viruses, as described in more detail in the next section. Viruses infecting active methanogens during the incubation were identified Among the 39 viral contigs of interest, several had predicted hosts corresponding to the dominant and active methanogens in the studied microcosms, namely, those from the order Methanobacteriales ( Methanobacterium and Methanobrevibacter genera). In particular, contig C158 (belonging to VC10) was predicted to infect Methanobacterium species, according to its taxonomic affiliation and the signature‐based host prediction. Methanobacterium was the most abundant archaeal genus in the studied microcosms. C158 has a length of 42,490 bp and contains 49 predicted genes. Consistent with the abundance of the predicted host, C158 was the most abundant of the 39 contigs in metaviromes at days 13 and 17, with RPKM values of 468 and 229, respectively. Furthermore, C158 is a complete dsDNA circular genome, from a head‐tailed virus. All functional modules typical for the class Caudoviricetes were identified in this contig (Figure 5 ), such as those required for virion morphogenesis (HK97‐like MCP, mCP, tail proteins, terminase subunits, capsid maturation protease) and several enzymes necessary for the life cycle, such as an intracellular proteinase inhibitor and an integrase, this latter suggesting a temperate lifestyle. According to its proximity to siphoviruses in the network (Figure 4 ) and VIRFAM analysis (Lopes et al., 2014 ), C158 is likely to have a siphovirus‐like morphology, that is long non‐contractile tail. Considering the lack of very strong similarity with previously characterized viruses, it likely represents a new family of archaeal viruses within the Caudoviricetes . We suggest the name Speroviridae for this new viral family. FIGURE 5 Genome map for selected contigs of interest. (A) Contigs from VC10 affiliated to class Caudoviricetes , including C158 (proposed new family Speroviridae ). (B) Viral contigs not related to head‐tailed viruses, including predicted spindle‐shaped viruses. Genes were annotated with MMseq against PHROGs and with hhpred against four other databases (see Experimental procedures ). Genes are represented as arrows. The main functional modules are indicated by colours. Abbreviations of core viral proteins: Ad, adaptor protein; BP, baseplate protein; BH, baseplate hub protein; BS, baseplate spike protein; BW, baseplate wedge protein; DNA_pol, DNA polymerase; DNA_prim, DNA primase; DNA_prim/helic, DNA primase/helicase; Endo, endonuclease; HAPV1_prot2, similar to protein 2 from Halorubrum pleomorphic virus 1; His1_prot_16, similar to protein 16 from Haloarcula hispanica virus 1; His1_SP, His1‐like major capsid protein; Holin, holin; Int, integrase; Lys, endolysin; MCM, MCM helicase; mCP, minor coat protein; MCP, major capsid protein; MT, methyl transferase; MTP, major tail protein; NP, pre‐neck appendage protein; NT, tRNA nucleotidyltransferase; PP, portal protein; Prot, capsid maturation protease; Sc, scaffolding; Sh, tail sheath; SSP1_SP, similar to the main structural protein of Shigella phage SSP1; SSV1_SP, SSV1‐like major capsid protein; TCP, tail completion protein; TerL, terminase large subunit; TerS, terminase small subunit; TMP, tail tape measure protein; TP, tail protein; Tube, tail tube The second‐most abundant contig of interest was C1359, one of the two contigs predicted to have a spindle‐shaped morphology, with RPKM values at days 13 and 17 of 285 and 183, respectively. Surprisingly, this 9936 bp‐long contig (17 detected genes) was predicted to infect Methanoculleus sp. (order Methanomicrobiales), which was present at very low abundances over the incubation time (<0.5% of the archaea based on shotgun metagenomic data). This host prediction was based on spacer alignment (SpacePHARER) against a public spacer database. By contrast, WIsH and spacer alignment with blastn did not yield any high‐confidence predicted host. Based on the archaeal community composition (see section 3.1.2), this is likely a false prediction, despite a highly significant p ‐value (2.03 × 10 −7 ). Nevertheless, C1359 gene content confirmed the probable archaeal nature of its host, and its high abundance suggests that it infected a dominant methanogen (order Methanobacteriales). Indeed, it encoded a protein showing a significant similarity with the major capsid proteins of two spindle‐shaped archaeal viruses, the halophilic Haloarcula hispanica virus His1 (Bath & Dyall‐Smith, 1998 ) and the hyperthermophilic Sulfolobus shibatae virus SSV1 (Palm et al., 1991 ) (probability: 98.84% and 98.36% respectively). Moreover, spindle‐shaped viruses are specific to archaea (Krupovic et al., 2018 ; Pina et al., 2011 ; Prangishvili et al., 2017 ; Snyder et al., 2015 ), excluding the possibility that the virus infects bacteria. This predicted morphotype is consistent with the presence of spindle‐shaped VLPs observed by TEM [Figure 2 (K–L)] and reported in AD ecosystems (Calusinska et al., 2016 ). Due to its limited similarity with previously characterized viruses, C1359 likely represents a new viral family, and corresponds to the first spindle‐shaped virus reported in association with the order Methanobacteriales. Viruses infecting minor methanogens in the formate incubation microcosms were identified As mentioned above, methanogens of the orders Methanomicrobiales and Methanosarcinales were present in the studied microcosms at a low abundance (<5%). Nevertheless, previously uncharacterized viral genomes possibly infecting these methanogens were identified. Contig C889 has a length of 14,120 bp and contains 26 predicted genes. Its RPKM values at days 13 and 17 were 2.3 and 3.3, respectively. Based on its taxonomic affiliation, this contig seemed to originate from a virus infecting a Methanomicrobiales host. In this contig, a protein similar to replicative minichromosome maintenance helicases was detected. This is not a structural protein but it has been identified in several families of archaeal viruses and archaeal plasmids (Krupovič et al., 2010a ), strongly supporting an archaeal host. No structural protein could be identified in C889, suggesting either a partial genome or a new type of virus. Contig C1697 has a length of 8207 bp, contains 11 predicted genes, and had RPKM values of 4.1 at day 13 and 8.4 at day 17. Based on its taxonomic affiliation and on host prediction with WIsH, this contig originates from a Methanosarcina virus. It encodes one protein showing significant profile similarity with the major capsid protein of SSV1 ( Fuselloviridae ) (probability: 94.92%), suggesting a spindle‐shaped virion morphology. C1697 did not show pronounced similarity with C1359 (which also had a predicted spindle‐shaped morphotype) or with known spindle‐shaped viruses, suggesting that it could represent a second new family of spindle‐shaped archaeal viruses. Notably, however, similar to spindle‐shaped halspivirus His1 (Bath & Dyall‐Smith, 1998 ) and thaspivirus Nitrosopumilus spindle‐shaped virus 1 NSV1 (Kim et al., 2019 ), C1697 encodes a protein‐primed family B DNA polymerase, suggesting that the genome of this virus is linear."
} | 10,481 |
20575574 | null | s2 | 7,058 | {
"abstract": "DNA origami was used as a scaffold to arrange spherical virus capsids into one-dimensional arrays with precise nanoscale positioning. To do this, we first modified the interior surface of bacteriophage MS2 capsids with fluorescent dyes as a model cargo. An unnatural amino acid on the external surface was then coupled to DNA strands that were complementary to those extending from origami tiles. Two different geometries of DNA tiles (rectangular and triangular) were used. The capsids associated with tiles of both geometries with virtually 100% efficiency under mild annealing conditions, and the location of capsid immobilization on the tile could be controlled by the position of the probe strands. The rectangular tiles and capsids could then be arranged into one-dimensional arrays by adding DNA strands linking the corners of the tiles. The resulting structures consisted of multiple capsids with even spacing (approximately 100 nm). We also used a second set of tiles that had probe strands at both ends, resulting in a one-dimensional array of alternating capsids and tiles. This hierarchical self-assembly allows us to position the virus particles with unprecedented control and allows the future construction of integrated multicomponent systems from biological scaffolds using the power of rationally engineered DNA nanostructures."
} | 336 |
28895241 | null | s2 | 7,059 | {
"abstract": "The utility of thermoresponsive hydrogels, such as those based on poly(N-isopropylacrylamide) (PNIPAAm), is severely limited by their deficient mechanical properties. In particular, the simultaneous achievement of high strength and stiffness remains unreported. In this work, a thermoresponsive hydrogel is prepared having the unique combination of ultrahigh compressive strength (≈23 MPa) and excellent compressive modulus (≈1.5 MPa). This is accomplished by employing a double network (DN) design comprised of a tightly crosslinked, highly negatively charged 1st network based on poly(2-acrylamido-2-methylpropane sulfonic acid (PAMPS) and a loosely crosslinked, zwitterionic 2nd network based on a copolymer of thermoresponsive NIPAAm and zwitterionic 2-(methacryloyloxy)ethyl]dimethyl-(3-sulfopropyl)ammonium hydroxide (MEDSAH). Comparison to other DN designs reveals that this PAMPS/P(NIPAAm-co-MEDSAH) DN hydrogel's remarkable properties stem from the intra- and internetwork ionic interactions of the two networks. Finally, this mechanically robust hydrogel retains the desirable thermosensitivity of PNIPAAm hydrogels, exhibiting a volume phase transition temperature of ≈35 °C."
} | 296 |
32573899 | PMC7496599 | pmc | 7,063 | {
"abstract": "Summary Interspecies bacterial competition may occur via cell‐associated or secreted determinants and is key to successful niche colonization. We previously evolved Pseudomonas aeruginosa in the presence of Staphylococcus aureus and identified mutations in the Wsp surface‐sensing signalling system. Surprisingly, a Δ wspF mutant, characterized by increased c‐di‐GMP levels and biofilm formation capacity, showed potent killing activity towards S. aureus in its culture supernatant. Here, we used an unbiased metabolomic analysis of culture supernatants to identify rhamnolipids, alkyl quinoline N‐oxides and two siderophores as members of four chemical clusters, which were more abundant in the Δ wspF mutant supernatants. Killing activities were quorum‐sensing controlled but independent of c‐di‐GMP levels. Based on the metabolomic analysis, we formulated a synthetic cocktail of four compounds, showing broad‐spectrum anti‐bacterial killing, including both Gram‐positive and Gram‐negative bacteria. The combination of quorum‐sensing‐controlled killing and Wsp‐system mediated biofilm formation endows P. aeruginosa with capacities essential for niche establishment and host colonization.",
"introduction": "Introduction Whether in the environment or during host infection, bacteria have to compete with established microbial communities to gain access to nutrients and space. Typical examples for polymicrobial infections are burn wounds and cystic fibrosis (CF) lungs, where the dominant pathogenic species are Pseudomonas aeruginosa and Staphylococcus aureus . Although S. aureus is prevalent in young CF patients, P. aeruginosa occurs more frequently in adult respiratory samples (Folkesson et al ., 2012 ). This switch in colonization frequency might result from host immune responses, antibiotic treatments and/or from direct competition between the two microorganisms (Limoli and Hoffman, 2019 ). Focusing on the latter hypothesis, we have previously established an in vitro model in which we evolved P. aeruginosa for 150 generations in the presence and absence of S. aureus to assess how P. aeruginosa adapts to the presence of a pre‐established niche competitor (Tognon et al ., 2017 ). By comparing the genomes of ancestor and evolved populations, we observed emergence of mutations in the P. aeruginosa Wsp (wrinkly spreader phenotype) signal transduction system (D'Argenio et al ., 2002 ; Hickman et al ., 2005 ). The mutations occurred in the wspF gene, resulting in constitutive activation of the Wsp signalling cascade. Surprisingly, the wsp mutants showed increased killing activity against S. aureus in comparison to the ancestor strain PA14. The Wsp signal transduction system of P. aeruginosa is similar to the chemosensory system of Escherichia coli and uses methylation and demethylation of the WspA transmembrane sensor by WspE and WspF, respectively, to adapt to variations of external stimuli (D'Argenio et al ., 2002 ; Hickman et al ., 2005 ). The Wsp system in P. aeruginosa was reported to respond to surface attachment (Guvener and Harwood, 2007 ; O'Connor et al ., 2012 ; Song et al ., 2018 ) and to changes in membrane composition (Blanka et al ., 2015 ). These stimuli ultimately translate into the generation of the intracellular signalling molecule c‐di‐GMP, via the cognate di‐guanylate cyclase WspR, entailing increased polysaccharide production, cell aggregation and biofilm formation (Kulasakara et al ., 2006 ; Valentini and Filloux, 2016 ). Loss of function mutations in the methyl esterase WspF results in constitutive activation of the Wsp signalling cascade and hence increased c‐di‐GMP levels (Hickman et al ., 2005 ). Compared to the wild type, wspF mutants are less motile and display a typical wrinkly colony morphology and reduced colony size, a phenotype reminiscent of the wrinkly spreaders described initially in P. fluorescens (Spiers et al ., 2002 ). The secondary messenger c‐di‐GMP plays a major role in the switch between the two main lifestyles of bacteria: the motile, planktonic lifestyle, associated with low c‐di‐GMP levels, and the sessile biofilm mode of growth triggered by high intracellular levels of c‐di‐GMP (Valentini and Filloux, 2016 ). Wsp mutants emerge in vitro upon exposure of P. aeruginosa to sub‐lethal concentrations of hydrogen peroxide (Chua et al ., 2016 ), but also in vivo during chronic lung infection in CF patients and in burn wounds (Smith et al ., 2006 ; Starkey et al ., 2009 ; Marvig et al ., 2015 ; Gloag et al ., 2019 ). The observation that P. aeruginosa , programmed for the sessile lifestyle, also displays antibacterial activity, prompted us to analyse the supernatants of a wspF mutant, using metabolomics analysis. We show that metabolite clusters comprising rhamnolipids, alkyl quinoline N‐oxides and the siderophores pyochelin and pyoverdin are the most upregulated chemical species in the wspF supernatant compared to the ancestor PA14. We observed a coordinated antibacterial action of these compounds and were able to reconstitute an artificial cocktail showing a broad‐spectrum antibacterial activity against Gram‐positive and Gram‐negative species. Hence, switching‐on the Wsp signalling system during biofilm formation also protects against invading or established competitors.",
"discussion": "Discussion Our unbiased metabolomic comparison between a Δ wspF and PA14 wild type supernatant allowed us to identify specific P. aeruginosa metabolites, which individually showed no antibiotic activity, but when combined in a synthetic cocktail, resulted in a broad‐spectrum bactericidal activity. An essential component of this cocktail were rhamnolipids, produced by Pseudomonas and Burkholderia spp (Soberon‐Chavez et al ., 2005 ). Their main function seems to be the solubilization of hydrophobic compounds such as aliphatic C‐sources acquired from the environment (Noordman and Janssen, 2002 ) or the self–produced PQS‐signalling molecule (Calfee et al ., 2005 ). Rhamnolipids show intrinsic antimicrobial activity against both Gram‐positive and Gram‐negative bacteria (Haba et al ., 2003 ; Nitschke et al ., 2010 ; Samadi et al ., 2012 ), as well as amoeba (Cosson et al ., 2002 ) and fungi (Goswami et al ., 2015 ). These glycolipids also display surfactant activity (Abdel‐Mawgoud et al ., 2009 ), required for swarming motility (Köhler et al ., 2000 ), disrupt tight junctions in epithelial cells (Zulianello et al ., 2006 ) and lyse polymorphonuclear neutrophils (Jensen et al ., 2007 ). The amounts produced by the Δ wspF mutant ranged between 300 and 400 μg ml −1 , a concentration reported to have bacteriostatic activity against S. aureus , S. epidermidis and B. subtilis (Haba et al ., 2003 ; Nitschke et al ., 2010 ; Samadi et al ., 2012 ). However, incubation with a commercial mix of rhamnolipids at concentrations of up to 500 μg ml −1 caused only a modest 50‐fold reduction in S. aureus and S. carnosus viable counts, which did not result from cell lysis as we confirmed by microscopic analysis. These data support the notion that at the concentrations detected in the Δ wspF supernatant, rhamnolipids alone are not sufficient to explain a 5 to 6‐log decrease in S. aureus viable cell counts. We therefore conclude that rhamnolipids act mainly as a permeabilizing agent (Radlinski et al ., 2017 ). Rhamnolipids can form micelles, which might incorporate cargo molecules, like the hydrophobic AQNOs and pyochelin (Fig. 6 ). Along this line, the QS signals PQS and 3‐oxo‐C12‐HSL were shown to be integrated into outer membrane vesicles of P. aeruginosa , thereby promoting their dissemination within a bacterial community (Mashburn‐Warren et al ., 2008 ). Membrane vesicles of P. aeruginosa were reported to fuse to the membranes of other Gram‐negative bacteria and even to S. aureus membranes (Kadurugamuwa and Beveridge, 1996 ). Whether this is the case for rhamnolipid micelles remains to be determined. The critical micellar concentration for rhamnolipids is approximately 100 μg ml −1 (Klosowska‐Chomiczewska et al ., 2017 ), a concentration below the 300–400 μg ml −1 measured in Δ wspF supernatants. Hence, the majority of rhamnolipid molecules in the Δ wspF supernatant should be under the form of micelles. Indeed, preliminary experiments using Nile Red to stain rhamnolipid micelles are in agreement with this hypothesis (our own unpublished observations). In the case of S. carnosus , lysozyme could further enhance the access of outer membrane vesicles or rhamnolipid micelles to the cytoplasmic membrane. Fig 6 Proposed scheme for the action of compounds identified in the Δ wspF mutant supernatant. HQNO and likely other AQNOs inhibit binding of menadione to CydAB cytochrome oxydase. HCN blocks heme binding in CydAB and favors release of Fe from Fe‐S clusters and ferritins. Siderophores with higher Fe(III)‐affinity than pyochelin could capture Fe(III) outside the cell to generate apo‐pyochelin. Rhamnolipids (RLP) form micelles and deliver apo‐pyochelin and AQNOs to the cell membrane. Pyochelin increases intracellular ROS production by chelating Fe(III) generated during the Fenton reaction. This leads to DNA and membrane damage or DNA fragmentation in the presence of H 2 O 2 [Color figure can be viewed at wileyonlinelibrary.com ] AQNOs represent another family of specialized metabolites reported to have anti‐staphylococcal activity. In particular, HQNO and NQNO (Lightbown and Jackson, 1954 ; Machan et al ., 1992 ; Szamosvari and Bottcher, 2017 ) inhibit cytochrome oxidases of the electron transport chain (CydAB in S. aureus ) by blocking the access to respiratory quinones (menadione) (Voggu et al ., 2006 ). HQNO also selects small colony variants in S. aureus , which carry mutations in heme or menadione synthesis pathways (Hoffman et al ., 2006 ). AQNOs are produced in vitro by CF‐isolates in the range of 1–10 μM (Nguyen et al ., 2016 ), but were also detected in lung biopsies from CF‐patients (Garg et al ., 2017 ). Indeed, the Δ wspF Δ pqsL mutant deficient in AQNO production, as shown by the metabolome analysis, showed strongly diminished killing activity. PqsL is responsible for the hydroxylation of the AQs to generate AQNOs (Drees et al ., 2018 ). Addition of commercial HQNO alone had only a weak antimicrobial effect on S. aureus , suggesting that HQNO requires other components present in the Δ wspF supernatant. We cannot exclude that AQNOs with longer acyl chains (NQNO, DQNO, UQNO) or harbouring an unsaturated acyl chain (Szamosvari and Bottcher, 2017 ) detected in our extended metabolomics analysis, also participate in the killing activity. Interestingly, the activity of HQNO seems to increase under iron‐deficient conditions (Nguyen et al ., 2016 ), which pinpoints the potential role of siderophores in our cocktail. Surprisingly, we could demonstrate killing of S. carnosus , which in contrast to S. aureus expresses a cyanide and HQNO insensitive CydB enzyme and has a lysozyme‐susceptible peptidoglycan (Bera et al ., 2006 ; Voggu et al ., 2006 ). This would incriminate other compounds present in the Δ wspF supernatant and acting on a different S. carnosus target. Alternatively, CydB might be susceptible to long‐chain AQNOs, which do not gain access to its target in the absence of lysozyme treatment. Besides the QS‐controlled factors rhamnolipids and AQNOs, the Δ wspF mutant also overproduced the main siderophores pyochelin and pyoverdin. Pyochelin has been reported to have bacteriostatic effects on S. aureus and other bacterial species and was suggested to generate reactive oxygen species (Adler et al ., 2012 ; Ong et al ., 2017 ) in the presence of pyocyanin (Coffman et al ., 1990 ; Britigan et al ., 1992 ). We could show here that pyochelin participates in S. aureus killing because the pyochelin‐deficient pchAD mutant showed decreased killing activity and addition of pyochelin at 200 μM resulted in a 2‐log reduction in S. aureus viable counts, both in medium and in PA14 supernatant. In E. coli , addition of catechol siderophores (enterobactin) was shown to abrogate ROS induced damage generated by pyochelin (Adler et al ., 2012 ). However, in our study, we observed the opposite effect since (i) addition of pyoverdin (a cyclic peptide harbouring a dihydroxy quinoline chelating group) to the pyochelin containing cocktail increased S. aureus killing activity by 3‐logs and (ii) deletion of pyoverdin synthesis genes in the pyochelin deficient Δ wspF Δ pchAD strain decreased killing activity of the corresponding supernatants by 3‐logs. We hypothesize that pyoverdin, which has a higher affinity for iron than pyochelin, chelates Fe(III) outside the cell generating apo‐pyochelin (Fig. 6 ). In Gram‐negative bacteria, TonB‐dependent receptor proteins actively transport siderophores across the outer membrane, whereas Gram‐positive bacteria lack these transporters. We therefore suggest that the hydrophobic apo‐pyochelin is captured by membrane vesicles or rhamnolipid micelles, which eventually fuse with the membrane of target bacteria and deliver pyochelin into the cytosol. The other high‐affinity iron chelators tested (enterobactin, protochelin, 2,2′‐dipyridyl) also increased the killing activity supporting the role of iron chelation for bactericidal activity. The high‐affinity siderophores could remove Fe(III) generated during the Fenton reaction thereby increasing the production of hydroxyl radicals (Fig. 6 ). In summary, our data show that P. aeruginosa produces a set of metabolites, including respiratory chain inhibitors (AQNOs, HCN), siderophores (pyochelin, pyoverdin) involved in ROS production, and cell permeabilizers (rhamnolipids, LasA protease), which when combined synthetically, result in an efficient broad‐spectrum bactericidal cocktail. Hence, P. aeruginosa uses a probably unique combination of effector and membrane permeabilizing compounds, targeting a broad spectrum of Gram‐positive and Gram‐negative bacteria. Secretion of this lethal cocktail is likely beneficial to P. aeruginosa when establishing or defending niches in the environment or in the host."
} | 3,573 |
37476829 | PMC10354246 | pmc | 7,065 | {
"abstract": "To navigate in new environments, an animal must be able to keep track of its position while simultaneously creating and updating an internal map of features in the environment, a problem formulated as simultaneous localization and mapping (SLAM) in the field of robotics. This requires integrating information from different domains, including self-motion cues, sensory, and semantic information. Several specialized neuron classes have been identified in the mammalian brain as being involved in solving SLAM. While biology has inspired a whole class of SLAM algorithms, the use of semantic information has not been explored in such work. We present a novel, biologically plausible SLAM model called SSP-SLAM—a spiking neural network designed using tools for large scale cognitive modeling. Our model uses a vector representation of continuous spatial maps, which can be encoded via spiking neural activity and bound with other features (continuous and discrete) to create compressed structures containing semantic information from multiple domains (e.g., spatial, temporal, visual, conceptual). We demonstrate that the dynamics of these representations can be implemented with a hybrid oscillatory-interference and continuous attractor network of head direction cells. The estimated self-position from this network is used to learn an associative memory between semantically encoded landmarks and their positions, i.e., an environment map, which is used for loop closure. Our experiments demonstrate that environment maps can be learned accurately and their use greatly improves self-position estimation. Furthermore, grid cells, place cells, and object vector cells are observed by this model. We also run our path integrator network on the NengoLoihi neuromorphic emulator to demonstrate feasibility for a full neuromorphic implementation for energy efficient SLAM.",
"introduction": "1. Introduction Simultaneous localization and mapping (SLAM) is the computational process of keeping track of one's location while navigating an unknown environment (i.e., localization ) and, simultaneously, creating a map of the environment (i.e., mapping ). Accurate localization is required for building metric map from egocentric observations, but errors in localization accumulate when relying solely on internally generated signals or self-motion (i.e., path integration or dead reckoning). An allocentric environment map can be used to correct these errors, making localization and mapping interdependent processes. SLAM is a core problem in mobile robotics, particularly in applications where high-precision GPS data is not available, such as in autonomous underwater vehicles or planetary exploration (Kim and Eustice, 2013 ; Palomeras et al., 2019 ; Geromichalos et al., 2020 ). Biological systems have evolved to solve these problems. Animals are capable of navigating and creating maps of novel environments, deducing their current location, and retracing their steps. Considerable research has been conducted to investigate the neural mechanisms underlying spatial cognition in animals. It is known that many animals—including rodents (Mittelstaedt and Mittelstaedt, 1982 ; Etienne, 1987 ; Benhamou, 1997 ), bats (Aharon et al., 2017 ), and humans (Mittelstaedt and Mittelstaedt, 2001 ) – are capable of path integration. Tolman ( 1948 ) proposed that animals construct “cognitive maps”: internal mental constructs used to retain and retrieve information about the relative locations and features of an environment. Such maps are widely believed to be used to discover novel shortcuts and provide corrections to path integration, much like SLAM systems in robots. Indeed, animals have access to a plethora of external sensory information, such as visual landmarks and odor trails, which can be used to correct the errors that would accumulate when using path integration alone. The hippocampal formation is believed to be crucial for such computations, with place cells, head direction cells, and grid cells thought to play significant roles. In fact, Safron et al. ( 2022 ) have characterized the hippocampal-entorhinal system as “the most sophisticated of all biological SLAM mechanisms”. While SLAM is a well-studied problem, and modern mobile robots are capable of performing SLAM, animal navigational abilities are still superior; they are more robust, efficient and adaptive, making them more useful in challenging real-world environments. Animals can use information from multiple sensory modalities (e.g., visual, olfactory, auditory, magnetoreception, and idiothetic cues) for navigation. Additionally, animals are able to navigate and map their environment in real-time using power-constrained computational resources, which is something that robots are still not able to achieve—brains are far more energy efficient than the GPUs or CPUs used to execute typical SLAM algorithms. The brain consumes around 20 Watts of energy while a single modern graphics card requires around 350 Watts. By taking inspiration from biology, researchers are trying to develop SLAM algorithms that are optimized for online processing, and that can run on resource-constrained platforms. For instance, neuromorphic hardware—designed to mimic the functionality of biological neural networks—is particularly well-suited for resource-constrained computing because it is designed to be energy-efficient and can perform brain-like computations using minimal resources (Bersuker et al., 2018 ; Thakur et al., 2018 ; Rathi et al., 2021 ). Biology has influenced the development of a new category of SLAM models that includes RatSLAM (Milford et al., 2004 ), DolphinSLAM (Silveira et al., 2015 ), and NeuroSLAM (Yu et al., 2019 ), among others. Remarkably, some of these models have demonstrated performance comparable to contemporary state-of-the-art approaches. However, this is still an active area of inquiry, with questions remaining regarding scalability and biological plausibility of these approaches, as well as their deployability on neuromorphic hardware. While these types of SLAM algorithms have made notable progress, they have yet to fully explore the wealth of knowledge available from neuroscience and cognitive science. Animals extract and make use of higher-order semantic information about their environment and landmarks from raw sensory inputs while navigating. Recent advancements in robotics have successfully incorporated semantic information into SLAM models (Bowman et al., 2017 ; Zhang et al., 2018 ; Chen et al., 2019 ; Fan et al., 2022 ). Semantic SLAM models use deep neural networks to extract semantic information to build environment maps. By utilizing higher-level conceptualization of states grounded in cognitive meaning, these models can augment and improve upon purely metric SLAM. Consequently, the construction of maps containing semantic representations empowers such SLAM systems to interact with environments in sophisticated and intelligent ways. In the same way that biology can aid in the development of AI and robotics, computational modeling can also provide valuable insights into biological research questions. By creating computational models of SLAM that are constrained to be biologically plausible, we can gain a deeper understanding of the neural algorithms that may underlie spatial cognition in animals. For example, we can investigate hypotheses on how exactly cognitive maps may be learned, stored, and used to assist in navigation. Or how multi-modal sensory information is integrated during the construction of cognitive maps. Or how such maps may be accessed and queried to reason about space. In this work, we unite biologically inspired and semantic SLAM in our model SSP-SLAM, and consider how our computational model can explain neuroscientific observations. Specifically, we present a novel spiking neural network SLAM system, called SSP-SLAM. This model is built using the Neural Engineering Framework (NEF) (Eliasmith and Anderson, 2003 ) and the Semantic Pointer Architecture (SPA) (Eliasmith, 2013 ). The NEF provides a systematic method for embedding a state space model into a spiking neural network that can run on neuromorphic hardware. The SPA, which includes Spatial Semantic Pointers (SSPs), provides an approach for representing and processing symbol-like information in connectionist systems. The SPA provides an architecture and “semantic pointer” representations, for characterizing neural processing, including that of symbols, as manipulation of high-dimensional vectors. This enables the development of systems that can learn and reason about symbolic information in a scalable, differentiable, and compositional manner. These methods are used in SSP-SLAM to build environment maps. These maps are core to the functioning of SSP-SLAM, as they integrate semantic information, while being combined with an SSP-based path integrator. The resulting model provides the following contributions: We propose and implement a novel spiking neural network SLAM model. We constrain our model to only use quantities that are known to be represented in hippocampus, like spatial representations of head direction cells, object vector cells, place cells, and grid cells. Furthermore, biologically plausible, Hebbian-like rules are used to learn an environment map in the form of an associative memory. We explore compositional semantic map representations using the SPA and the principles of vector symbolic architectures more broadly. We demonstrate how such a map can be queried post-training to recall what landmarks were in particular areas, recall where landmarks of certain types or colors were located, and compute (online) the vector between self-position and landmarks in memory. We illustrate first steps toward a neuromorphic implementation of our model, showing that the path integration component of SSP-SLAM can be run on an emulator of Intel's Loihi neuromorphic chip.",
"discussion": "4. Discussion 4.1. Prior research The development and implementation of SLAM algorithms for mobile robots has garnered significant attention in academic and engineering communities. Approaches generally involve recursive Bayesian estimation—via various kinds of Kalman Filters (Smith et al., 1990 ; Brossard et al., 2018 ), Particle Filters (Montemerlo et al., 2002 ; Sim et al., 2005 ), or occupancy grid methods (Stachniss et al., 2004 )—or graph optimization (Thrun and Montemerlo, 2006 ; Sünderhauf and Protzel, 2012 ). In recent years, researchers have focused on incorporating semantic information into SLAM systems, using deep artificial neural networks, particularly convolutional or recurrent neural networks for object detection and semantic segmentation. The use of semantic information in SLAM has been found to improve performance and robustness of robot localization (Frost et al., 2016 ; Stenborg et al., 2018 ; Bowman, 2022 ). Furthermore, robots equipped with semantic SLAM hold the promise of performing higher-level tasks, such as planning paths based on human instructions that reference objects in the environment. Concurrently, an alternative approach to SLAM, drawing inspiration from the brain, has continued to develop novel algorithms with the goal of improving efficiency and robustness (Milford et al., 2004 , 2016 ; Silveira et al., 2015 ; Yu et al., 2019 ). In this line of research, models of neural path integration inspired by hippocampal cells are used for localization. Coupling such neural algorithms with recent developments in neuromorphic hardware, as we have done here, aims to both improve our understanding of how the brain accomplishes SLAM and to improve the power efficiency of engineered solutions. Neural localization models used in this alternative approach can generally be divided into two categories: Continuous Attractor Network (CAN) models (Samsonovich and McNaughton, 1997 ; Tsodyks, 1999 ; Conklin and Eliasmith, 2005 ) and Oscillator-Interference (OI) models (O'Keefe and Burgess, 2005 ; Burgess et al., 2007 ; Hasselmo et al., 2007 ; Welday et al., 2011 ). In CAN models, path integration is performed by a recurrently connected neural sheet, whose dynamics sustain a single Gaussian-like activity bump that represents the self-position estimate of an agent. In contrast, in OI models, the self-position estimate is encoded by the phase differences between Velocity-Controlled Oscillators (VCOs)—oscillators whose frequency is modulated by a velocity signal. The seminal application of neural-inspired methods to SLAM is RatSLAM, in which visual odometry is used to drive a CAN (Milford et al., 2004 , 2016 ). The CAN consists of “pose cells” (similar to the place and head direction cells found in the hippocampal formation) and maintains an estimate of self-position and orientation. Sensor data is processed outside the neural network to create a template array (for example, raw visual input is converted to an intensity profile vector). When a novel template is observed, a new “local view cell” (similar to the spatial view cells in the hippocampus) is added to the network. The population of these cells is sparsely connected to the CAN, with associations learned via Hebbian learning. Additionally, a graph is constructed and updated with a graph relaxation algorithm online to create a topological environment map. Its nodes store experiences in the form of activity of pose cells and local view cells along with robot pose estimates. In contrast, a hybrid OI-CAN model is used for path integration in SSP-SLAM and a graphical environment map is not learned—instead, the outgoing connection weights from the memory network implicitly store a map which can be retrieved by querying the network. The Voja rule, which is used to shift the associative memory population encoders toward observed input in SSP-SLAM, plays a similar role to the template novelty detection and addition of local view cells that occurs in RatSLAM. Furthermore, we have not implemented an external module for pre-processing of sensory data, and we use landmark semantic pointers and displacement SSPs in lieu of templates. Object detection and depth estimation algorithms would be required to obtain this input from visual data. Many models have since extended the original RatSLAM. CAN SLAM models with place cell-like activity were also used by BatSLAM (Steckel and Peremans, 2013 ), an extension to RatSLAM for handling environment information from sonar sensors, and DolphinSLAM (Silveira et al., 2015 ), developed for 3D SLAM in underwater environments. A CAN consisting of conjunctive grid cells was used in the SLAM model presented in Zeng and Si ( 2017 ). Three-dimensional SLAM in realistic environments with grid cells was also explored in NeuroSLAM (Yu et al., 2019 ). Unlike our work, none of these models use spiking neural networks. More recent research has focused on developing spiking networks for SLAM and testing them on neuromorphic hardware. Spiking 2D SLAM models were presented in Tang and Michmizos ( 2018 ), Tang et al. ( 2019 ), and Kreiser et al. ( 2020a , b ). In Kreiser et al. ( 2020a ), a SLAM system on the Loihi chip was used to estimate the head position of an iCub robot as it visually explored a wall with a dot pattern acting as the environment. Tang et al. ( 2019 ) made use of a depth camera and Bayesian updates on a posterior distribution represented by neural population. They found that their SLAM system, when run on Loihi, was more energy efficient by two orders of magnitude compared to a baseline method on a CPU. While the models discussed here use raw sensory input, it should be noted that non-spiking visual modules are used to process this information and obtain input for SLAM. For instance, intensity profile vectors or feature colors and distances from the observer are used. In contrast to SSP-SLAM, none of the models mentioned incorporate any elements of OI to perform path integration, or perform 3D SLAM. Furthermore, some of these models employ “localist”/discrete representations, such as using one neuron to represent each integer value for heading direction or discretized distance to features. This approach does not support generalization and does not scale well to higher dimensional representations, unlike SSPs. Taken together, and summarized in Table 2 , past work provides examples of spiking and non-spiking networks, using CANs for path integration. However, unlike SSP-SLAM, none of these approaches provides a methodology for incorporating semantic information or for online learning of semantic environmental maps. In addition, none of these employ SSPs, or the same combination of a OI-CAN network in a fully spiking model capable of functioning equally well in both 2D and 3D spatial environments, as demonstrated above. Table 2 Comparison of bio-inspired SLAM models. \n Model \n \n Sensors \n \n Input representation \n Dim . \n Localization \n \n Env. map \n \n Cells \n \n Experiment scale \n \n Spiking \n \n Neuromorhpic hardware \n SSP-SLAM None Displacement to features as an SSP and feature identities as SPs Any, tested on 2D & 3D OI-CAN hybrid Weights between landmark population to landmark locations HDC, GC, landmark cells, OVC Small Yes Partially RatSLAM (Milford et al., 2004 , 2016 ) Monocular camera Greyscale image intensity profile 2D CAN Topological map associating local views with position stored as a graph Pose cells, local view cells Large No No BatSLAM (Steckel and Peremans, 2013 ) Biomimetic sonar Intensity difference between left and right Echolocation Related Transfer Functions 2D CAN Topological map local views with position stored as a graph Pose cells, local view cells Small No No DolphinSLAM (Silveira et al., 2015 ) Sonar & visual One-hot representation obtained from FabMAP algorithm on top of a Bag of Words model 3D CAN Graph with nodes storing local view, place cell and position while edges store displacements 3D PC, local view cells Small No No NeuroSLAM (Yu et al., 2019 ) Panoramic camera Greyscale image intensity profile 3D CAN Topological map storing activities of local view cells, GCs, HDCs, and estimated pose 3D PC, conjuctive 3D GC and HDC, local view cells Large No No Kreiser et al. ( 2020a ) Event-based camera Detection of blinking LEDs at different frequencies 2D CAN Weights from landmark population to a HDC population HDC, landmark cells Small Yes Fully Tang et al. ( 2019 ) RGB-Depth camera Discretized distances to landmarks 2D CAN Weights from PC to a displacement-from-border population 2D PC, HDC, border cells, Bayesian cells Small Yes Fully 4.2. Performance We have presented the results of several experiments on SSP-SLAM to assess its performance and utility. The model demonstrates accurate localization capabilities on different paths, both two-dimensional and three-dimensional. To achieve this, a hybrid OI-CAN model is employed for path integration. Notably, this is the only SLAM model (to our knowledge) that uses OI techniques for localization. This approach has the advantage of easy generalization to higher dimensional spaces. Typically, CAN models describe a neural population as a 2D sheet or 3D array (often with periodic boundary conditions), where the geometry specifies the recurrent connectivity pattern required for localization. However, this only supports unimodal position estimates, and the connectivity pattern must be modified and made more complicated to move to higher dimensional path integration. 1 In contrast, in our approach the recurrent connectivity of the path integrator network remains the same regardless of spatial dimensionality. This allows the same model to switch seamlessly between SLAM in different spaces and domains. Furthermore, SSP-SLAM encodes environment maps in the outgoing connections of an associative memory network, which are learned online using biologically plausible learning rules. The map generated is a semi-metric, semantic map that uses symbol-like vector representations that have been leveraged in a variety of large-scale cognitive models (Eliasmith, 2013 ; Arora et al., 2018 ; Kajić et al., 2019 ; Kelly et al., 2020 ; Gosmann and Eliasmith, 2021 ). By working in the SSP and VSA paradigm, we are able to formulate the problem in such a way that unites metric and semantic SLAMs. This approach unites analytical models of vehicle motion and map construction with neural networks, resulting in a formulation that is compatible with modern ML approaches to robotics, while still maintaining the explainability of the system. These feature distinguishes SSP-SLAM from other bio-inspired SLAM models and makes it the first spiking semantic SLAM model to our knowledge. This inclusion of semantic information helps SSP-SLAM be more accurate. Specifically, SSP-SLAM performs loop closure via corrections to the PI network provided by the environment map, which leads to significant improvements in localization accuracy. After training, the map can be queried to obtain object locations given their symbol-like representation as a semantic pointer. Alternatively, item representations can be obtained by querying specific areas, or vectors between the agent and landmarks can be computed. These kinds of direct queries of semantic map knowledge cannot be easily made with past spiking network map representations. Finally, a key element of the model, the path integrator, was tested on a neuromorphic emulator. The results indicate that the model can maintain expected accuracy (given the absence of error correction mechanisms) on neuromorphic hardware. Notably, all additional operations used in the model have been implemented on neuromorphic hardware in other work (Knight et al., 2016 ; Mundy, 2017 ), so we believe this demonstration strongly suggests that a full neuromorphic implementation is achievable. Overall, this study presents a novel and promising approach to SLAM based on a fully spiking neural network. 4.3. Limitations This study presents a novel model that employs biologically-inspired mechanisms to solve SLAM. However, SSP-SLAM has several limitations. First, the full SSP-SLAM model has not been tested on a neuromorphic chip emulator nor has the model been deployed on an actual neuromorphic hardware platform. Second, the model was tested on a small scale and artificial environments, which restricts what conclusions we can draw as to its generalizability to more complex, real-world environments. To improve the model's utility, it is essential to test it on real-world input and integrate it with a network that can process raw sensory data. Such an approach would enhance the model's ability to handle more complex and diverse environmental conditions. Moreover, the current model's accuracy is inferior to that of non-biologically inspired SLAM methods, which limits its usefulness to mobile robotics. This accuracy drop and the use of small scale test environments is true of current spiking SLAM models more generally. Despite this, the use of neuromorphic computing and hardware has the potential to improve energy efficiency of SLAM systems, which is particularly useful in mobile robotics applications. This encourages further research into spiking SLAM systems. Reduced power demands permits the deployment of SLAMs in progressively more power-constrained environments, such as edge computing or operations in GPS-denied settings, like space or sub-sea exploration. An increasing number of algorithms have harnessed the advantages of spike-based computing to make gains in efficiency and speed (Yakopcic et al., 2020 ; Davies et al., 2021 ; Yan et al., 2021 ). Therefore, while the current model shows promise in enabling biologically-inspired SLAM, its limitations in terms of testing and accuracy should be addressed before considering its wider application in real-world scenarios. Further research could focus on testing the model on larger networks and more complex environments, as well as investigating ways to improve its accuracy. 4.4. Future work One clear direction for future work is ameliorating the limitations discussed in the previous section. Beyond this, there are several other directions that warrant further exploration—for example, explicit modeling of sensor uncertainties using SSPs, introducing coupling dynamics to increase localization accuracy, higher-dimensional SLAM, and integration with other cognitive models. Accurate of localization is vital and phase drift is one of the main factors contributing to SSP inaccuracy. As path integration progresses, errors can accumulate in the phases of the velocity-controlled oscillators (VCOs), resulting in inconsistencies that degrade the spatial information (e.g., see Figure 8 ). The loop-closure error corrections ( Figure 2 ) can shift the phases toward the true values, but the phase inconsistencies would still be present. However, one could take advantage of the redundancy in the SSP representation by adding coupling between the VCOs that enforce their proper phase relationships (Orchard et al., 2013 ). Additionally, higher-dimensional SLAM could be a promising area of investigation. The proposed model can be extended to localization and mapping in any dimension of space by modifying the input without changing the model or hyperparameters. Although SLAM is mainly applied to navigation and mapping in physical spaces, operating in dimensions equal to or less than three, it is possible that the same neural mechanisms underlying spatial navigation and mapping could be applicable to non-spatial domains, such as mapping in high-dimensional conceptual space. The idea that similar computations to those behind SLAM may be understood as core cognitive processes has been proposed in Safron et al. ( 2022 ). The application of SSP-SLAM to localization and mapping in various spaces (including non-spatial ones) via interactions with other cognitive systems is promising area for future research. By employing control mechanisms to manipulate the input to SSP-SLAM, it may be possible to model different cognitive functions. For instance, one could switch between motion input from sensory systems to perform localization and input from memory and cognitive maps to simulate path replay or planning. This could be realized by integrating SSP-SLAM with more complex memory, action selection, and reasoning systems. Since the proposed model was developed using the SPA, it fits naturally within the context of NEF and other SPA models, including Spaun (Stewart et al., 2012a ; Choo, 2018 ). Integration of the proposed SLAM model with other models constructed with these tools could be used to develop systems equipped with more sophisticated cognitive capabilities and able to tackle multiple tasks. Exploiting memory and reasoning capabilities in large spatial environments remains a challenge for models of biological cognition. 4.5. Summary In conclusion, we have proposed a novel spiking semantic SLAM model, SSP-SLAM, which is inspired by the hippocampal formation in the mammalian brain. The model is unique in its integration of a hybrid OI-CAN path integrator, online biologically-plausible learning of an environment map, and use of symbol-like object representations in a spiking network. This combination enables the model to perform SLAM accurately in small scale environments and learn representations that can be queried in powerful ways. For example, it can provide information about what is located in a particular area of the map, report vectors between landmarks, and identify the location of objects based on their properties, such as their color. Furthermore, these techniques advance the sophistication of biologically plausible SLAM networks, showing a wide variety of previously identified cell types while demonstrating functionality in 2D and 3D environments. Finally, we have tested a core component of the network on a neuromorphic hardware emulator, which represents an important step toward achieving a full system running on neuromorphic hardware. While significant work remains to achieve this goal, we believe that the methods and components employed in this study provide a foundation for future research in this area. With continued progress, this spiking semantic SLAM model could have important applications in a wide range of fields, including robotics, artificial intelligence, and neuroscience."
} | 7,094 |
25567946 | PMC3352509 | pmc | 7,066 | {
"abstract": "The root systems of most agronomic crops are colonized by diverse assemblages of arbuscular mycorrhizal fungi (AMF), varying in the functional benefits (e.g. nutrient transfer, pathogen protection, water uptake) provided to hosts. Little is known about the evolutionary processes that shape the composition of these fungal assemblages, nor is it known whether more diverse assemblages are beneficial to crop productivity. In this review we aim to identify the evolutionary selection pressures that shape AMF diversity in agricultural systems and explore whether promotion of AMF diversity can convincingly be linked to increases in agricultural productivity and/or sustainability. We then ask whether farmers can (and should) actively modify evolutionary selection pressures to increase AMF functioning. We focus on three agriculturally imposed selection regimes: tillage, fertilization, and continuous monoculture. We find that the uniform nature of these practices strongly selects for dominance of few AMF species. These species exhibit predictable, generally non-beneficial traits, namely heavy investment in reproduction at the expense of nutrient scavenging and transfer processes that are beneficial for hosts. A number of focus-points are given based on empirical and theoretical evidence that could be utilized to slow down negative selection pressures on AMF functioning, therein increasing crop benefit.",
"conclusion": "Conclusion The first two questions addressed in this review were: (i) What are the evolutionary selection pressures that promote or diminish microbial biodiversity in agricultural systems? And (ii) Can promotion of microbial biodiversity be convincingly linked to increases in agricultural productivity and/or sustainability? In regards to (i), studies on AMF have focused on documenting effects of various agricultural management schemes on AMF diversity. Although such studies are essential to understand the genetic hierarchy of AMF community response, the second question can only be answered through a shift of focus to ‘ functional diversity’, not just ‘diversity’ of AMF. In theory, there are numerous AMF attributes (pathogen and herbivory protection, alleviation of water stress, tolerance to salinity, pH, toxins, etc.) with the potential to increase agricultural productivity and sustainability. However, we are seeing that current management practices are more likely to favour AMF with attributes less beneficial for crop hosts, such as fast, abundant sporulation and increased carbon acquisition from hosts. A less beneficial AMF community fails to provide optimal functioning (nutrient acquisition and otherwise) and so agricultural practices (e.g. higher fertilization) are required to maintain crop productivity, with the result that these practices continue to degrade the AMF community. Calling for large-scale changes in management regimes is not practical, especially given that AMF functional attributes, at least at field scale, are still more theoretical than demonstrated. However, relatively small scale changes in agricultural practices may lead to a more functionally complex AMF community, potentially with the benefit of increasing productivity ( Fig. 3 ). Figure 3 Schematic representation of selective environment imposed by ‘sustainable’ management practices (left) and their influence on the AMF community (middle). Implementation of more sustainable management regimes is predicted to increase the benefit of AMF to agricultural systems by facilitating a re-establishment of AMF with functionally diverse attributes. The last question we aimed to answer with our review, is also the most urgent: what small-scale changes in management practices (e.g. particular crop rotations) will have large-scale benefits towards increasing the functioning of AMF communities? One route towards gathering more concrete data on the selection pressures that modify functional traits in AMF is through the use of microcosms (e.g. Boddington and Dodd 2000 ) inoculated with AMF communities from agricultural fields. Multigenerational experiments in which treatments mimic agricultural selection pressures such as tillage, fertilizer regimes, crop rotations and crop polyculturing may begin to capture how agronomic-like manipulations modify functional traits of AMF communities over time. The small size of the experiments would allow the tracking of genetic diversity (ideally both among and within AMF individuals), as well as functional diversity. Benefits of specific functional attributes could be measured and followed over multiple generations. We could then begin to determine the specific AMF strategies favourable in an agricultural context and ask what selection pressures facilitated their spread in the AMF community. Beneficial strains surviving over several generations of strong agronomic selection pressures could be isolated, propagated and potentially introduced with their host crop into the field. Two major pitfalls in this approach are (i), problems of scaling up from microcosms dynamics to agronomic fields ( Oehl et al. 2009 ), and (ii), the recently highlighted issues of introducing AMF inoculum strains into (albeit managed) ecosystems ( Schwartz et al. 2006 ; Mummey et al. 2009 ). From a purely evolutionary point of view, the incredibly high evolvability of AMF strains makes them an interesting model organism for investigations into rates of adaption. Ehinger et al. (2009) found that under laboratory conditions, AMF genetic composition could change within one propagation cycle upon nutrient or host selection pressures. Future research should focus on whether this change is random or directed. If random, local drift processes may occur but the population as a total may still harbour the same genetic information. If directed, however, this could mean rapid evolution of AMF strains with the likely loss (or gain) of valuable functions due to strong agronomic pressures. Ideally, molecular methods will be developed in the near future utilizing gene-expression as a way to approximate mutualistic benefit for specific functional traits ( Gamper et al. 2010 ). These types of tools could prove to be useful for the future management of agroecosystems, ultimately allowing farmers to maximize mutualistic benefit of soil microbes.",
"introduction": "Introduction Interest in the functional role of biodiversity has burgeoned in recent years ( Cardinale et al. 2007 ), including its potential to enhance the sustainability and output of agricultural systems ( Tilman et al. 1996 ; Moonen and Bàrberi 2008 ). This interest stems from the often assumed, and sometimes demonstrated ( Schlapfer and Schmid 1999 ), cause and effect relationship between particular components of biodiversity and long-term agricultural productivity. Biodiversity of organism groups such as decomposers, predators, pollinators, etc. are thought to increase the provision of specific agricultural services by providing an array of functionally complementary species. Such complementarity may raise agricultural productivity. The mechanisms are diverse, but in the broadest terms, biodiversity may act in an ‘insurance’ role, buffering systems against stresses or losses, while also increasing the multi-functionality of a system ( Hector and Bagchi 2007 ), known as the ‘niche differentiation effect’ ( Tilman 1999 ; Ptacnik et al. 2008 ; Marquard et al. 2009 ). Many of the problems encountered in high-input global agriculture (e.g. nutrient runoff, pests, weeds and erosion) are among those that a diverse ecosystem is predicted to counteract. In particular, the functional role of soil microbial diversity in agroecosystems has received much attention to date. Higher nutrient use efficiency, increased soil aggregate stability and respiration, improved organic matter formation and increased water regulation, are among the soil-related processes that microbially-diverse systems are hypothesized to promote ( Mäder et al. 2002 ; Brussaard et al. 2007 ). Despite numerous ecologically-focused studies on the role of microbial diversity in agroecosytem function ( Shennan 2008 ; Toljander et al. 2008 ), surprisingly little is known about: (i) the evolutionary selection pressures that promote or diminish microbial biodiversity in agricultural systems, (ii) whether promotion of microbial biodiversity can convincingly be linked to increases in agricultural productivity and/or sustainability (see e.g. Martini et al. 2004 ), and (iii) whether farmers can (and should) actively modify management practices to manipulate evolutionary selection pressures for increased service provisioning from soil microbes. Here, we examine these questions by focusing on one critical group of soil microbes, abundant in both agricultural and natural ecosystems: arbuscular mycorrhizal fungi (AMF). In this 450-million-year-old symbiosis, the mycorrhizal fungal partner forms an obligate symbiosis with its host plant, exchanging nutrients from the soil for carbon from the host. This interaction is arguably the world's most abundant symbiosis, responsible for massive amounts of nutrient transfer globally ( van der Heijden et al. 2008 ). In agricultural systems, arbuscular mycorrhizae form associations with almost all important crops (maize, wheat, soybean, but not: cabbage, mustard and beet), and are therefore an intricate component of the above- as well as the belowground ecosystem. Increasingly, the role of AMF in pathogen suppression ( Lendzemo et al. 2005 ), pollination enhancement ( Cahill et al. 2008 ), herbivore protection ( Gange and West 1994 ; Bennett et al. 2009 ) and improved water relations are being recognized ( Augé 2001 ; Wilson et al. 2009 )."
} | 2,423 |
38770267 | PMC11105003 | pmc | 7,067 | {
"abstract": "Transparency, flexibility, and high thermal conductivity\nare trade-offs.\nSpecifically, we have investigated a cross-linked acrylic liquid crystal\nelastomer (LCE) that exhibits both transparency and flexibility while\nmaintaining a high level of thermal conductivity. The transparent\nmonodomain LCE sheet was achieved through a process of stretching\nan initially opaque polydomain sheet to 80% elongation and subsequently\nsubjecting it to photocuring. The thermal conductivity in the stretching\ndirection ( x ) of the monodomain LCE sheet was found\nto be 1.8 times higher than that of the prestretched polydomain sheet,\nconsistent with findings from previous studies. However, in the orthogonal\ndirection ( y ) to the stretching ( x ) direction, the thermal conductivity exhibited an even higher value,\nbeing 1.7 times greater than in the x -direction,\nwith a value of 3.0 W/(m·K). This unique observation prompted\nus to conduct further investigation through higher-order structural\nanalysis of these LCE sheets using 2D wide-angle X-ray scattering\n(WAXS) analysis. In the transparent sheet, the LCE molecules were\naligned in the sheet in the stretching x -direction\n(monodomain structure) for the out-of-plane direction. However, in\nthe in-plane x -direction, the molecular plane spacing\nexhibited random orientation at a period of 0.45 nm. In contrast,\nwithin the y -direction of the inner layer, the molecular\nplane spacing exhibited a uniaxial horizontal orientation at the same\nperiod length as in the x -direction. The heat energy\nentering into the y -direction once spreads to the x -direction, but it was considered that the reason for the\nhigher thermal conductivity to the y -direction would\nbe forming covalent bonds that function as new heat transmission paths,\nin the direction intersecting to the x -direction\nduring photocuring. Therefore, we concluded that the synergistic effect\nof the high level of the ordered inner structure and covalent bonding\nstructure due to cross-linking in the y -direction\ncontributes to its higher thermal conductivity compared to that in\nthe x -direction, which exhibits a random in-plane\nstructure. Additionally, we have fabricated an LCE composite sheet\nfilled with 75 vol % of alumina particles using a polydomain-type\nLCE as the base material. The composite sheet exhibits remarkable\nthermal conductivity in the thickness direction, measuring at 9.8\nW/(m·K), while maintaining a flexibility characterized by an\nelastic modulus of 70 MPa. This thermal conductivity surpasses that\nof a nonmesogenic acrylic composite sheet with identical alumina particle\nfilling, which measured at 3.9 W/(m·K), more than twice as much.\nThe presence of the mesogen skeleton has been demonstrated to enhance\nheat transfer, even within soft composites, by facilitating the formation\nof an ordered structure.",
"conclusion": "4 Conclusions In this paper, higher ordered\nstructures were analyzed for the\ncross-linking acrylic liquid crystal elastomer (LCE), which exhibits\ntransparency and flexibility while possessing high thermal conductivity\nin the in-plane direction. This transparent monodomain LCE sheet was\nobtained by photocuring while stretching the polydomain LCE to 80%\nelongation. The thermal conductivity of the stretching direction ( x ) of this monodomain LCE sheet was 1.8 times higher, measuring\n1.79 W/(m·K), compared to that of the prestretched polydomain\nsheet. Furthermore, regarding the thermal conductivity of the orthogonal\ndirection ( y ) to the x -direction,\nwhich was measured at 3.0 W/(m·K), it was around 1.7 times higher\nthan that of the x -direction. From the XRD\nanalysis, for the out-of-plane direction, the molecules\nin the monodomain LCE sheet were aligned in the stretching x -direction. In contrast, for the in-plane direction, the\nmolecular plane spacing was randomly oriented to the x -direction (i.e., heat transfer direction). However, in contrast,\nthe molecular plane spacing was horizontally uniaxially oriented,\nexhibiting a highly ordered structure in the y -direction.\nTherefore, we conclude that the heat energy entering into the y -direction once spreads to the x -direction\ndue to the higher ordered structure; subsequently, the synergistic\neffect of the covalent bonding structure due to photo-cross-linking\nin the y -direction contributes to its higher thermal\nconductivity compared to the x -direction. Next,\nusing the polydomain LCE, composite sheets filled with 75\nvol % of alumina particles were fabricated and characterized. The\nalumina composite sheet exhibited high thermal conductivity along\nthe thickness direction, averaging 9.8 W/(m·K), coupled with\nflexibility characterized by an elastic modulus of around 70 MPa.\nThe thermal conductivity was over twice the value of 3.9 W/(m·K),\nwhich is measured in a nonmesogenic acrylic composite sheet with the\nsame alumina particle filling. The apparent thermal conductivity of\nthe base LCE resin in the composite was estimated to be 0.5 W/(m·K)\nusing Bruggeman’s composite’s thermal conductivity prediction\nequation. These flexible composite sheets are expected to be applicable\nas thermal interface materials (TIM) in semiconductor applications\nand other related fields.",
"introduction": "1 Introduction As electronic devices\nbecome more sophisticated and compact, the\nheat generation density in circuits is increasing. Consequently, there\nis an increasing demand for enhanced heat dissipation, especially\nfor electric insulating materials. 1 − 4 Although polymers are excellent electric\ninsulating materials, they are basically adiabatic and do not conduct\nheat effectively. This is because, unlike metals, which have free\nelectrons that facilitate heat conduction, in insulating polymers,\nphonons dominate thermal conduction. 5 Polymers\nare broadly classified into thermoplastic and thermosetting polymers,\nand it is known that the thermal conductivity of thermoplastic polymers\nincreases when they are stretched. 6 − 14 Recently, it has been reported that the thermal conductivity of\npolyethylene fiber, a thermoplastic polymer, was increased to 104\nW/(m·K), which is equivalent to that of metal, by stretching\nit to the utmost limit. 15 , 16 In contrast, thermosetting\npolymers that form a three-dimensional network structure, such as\nepoxy resin, which is an essential material for electronic circuits,\ncannot be stretched or otherwise oriented like thermoplastic polymers,\nmaking it difficult to significantly improve their thermal conductivity.\nTherefore, to reduce phonon scattering and achieve high thermal conductivity,\nit is effective to form a highly ordered higher-order structure inside\nthe polymer by self-alignment of the mesogen backbone, 4 , 17 − 38 and then to design molecules that increase the cross-linking density\nby selecting a curing agent. 2 Recently,\na research study utilized a mesogen epoxy resin film alone, without\nthe addition of a thermally conductive ceramic filler, and achieved\na thermal conductivity of 10 W/(m·K). 39 Although thermoplastic polymers exhibit a significant decrease in\nthe thermal conductivity above Tg, thermosetting polymers experience\na minor decrease in thermal conductivity. This is attributed to the\npreserved cross-linked orientation order in thermosetting polymers\neven when above Tg. 1 , 2 However, transparency, flexibility,\nand high thermal conductivity\nare in a trade-off relationship within polymers. Thus far, a polymer\nthat successfully embodies all three characteristics simultaneously\nhas not yet been developed. Several papers had reported transparent\npolymer films with high thermal conductivity, but not enough transparency\nproperties. 40 , 41 The introduction of a mesogen\nskeleton is effective in increasing thermal conductivity, but its\nhigh orientation and birefringence anisotropy result in a rigid and\nopaque structure. 2 , 18 , 35 , 36 , 39 Recently,\na new liquid crystalline acrylic elastomer (LCE) sheet with transparency,\nflexibility, and impact strength resistance has been published in\npapers. 42 − 44 In this paper, monodomain and polydomain liquid crystal\nelastomers (LCEs), as well as LCE composites, are fabricated, and\ntheir thermal conductivities are measured. The study aims to investigate\nthe relationship between higher-order structural analysis and thermal\nconductivity by utilizing XRD. The fabrication process of the monodomain\nLCE involves two stages. First, the LCE polydomain is achieved by\nemploying Michael addition of polyeher thiols and multifunctional\nthiols to an acrylic monomer containing mesogens (1st stage). Subsequently,\nthe material undergoes stretching and photo curing to transition into\nthe monodomain state (2nd stage). Additionally, a flexible composite\nsheet of polydomain LCE with a high alumina filler content was fabricated\nwithout the stretching process. Subsequently, its thermal conductivity\nwas evaluated.",
"discussion": "3 Results and Discussions 3.1 LCE Polydomain and Monodomain Sheets Figure 3 displays\nthe appearance of the fabricated polydomain and monodomain LCE sheets,\neach with a thickness of 300 μm. The monodomain sheet exhibits\ntransparency in the visible region. The transmittance spectrum of\neach LCE sheet was measured using a Hitachi UV–vis–NIR\nSpectrophotometer (U-4000), as depicted in Figure S2 . Figure 3 Appearance of the opaque polydomain and transparent monodomain\nLCE sheets (thickness = 300 μm). First, the planar orientation of the LCE polydomain\nand monodomain\nsheet birefringence was qualitatively confirmed by rotating the stage\nunder a crossed Nicols using a polarizing microscope (POM). The results\nare shown in Figure 4 . For polydomain sheets, the brightness of the interference image\ndid not change when the stage was rotated, whereas for monodomain\nsheets, the bright and dark fields repeated every 45° when the\nstage was rotated, indicating that the sheets were uniaxially oriented\nin the stretching direction within the plane. Figure 4 Observation of the orientation\nstate of monodomain and polydomain\nsheets using a polarizing optical microscope (POM). Measurements of the in-plane thermal conductivity\nof LCE sheets\nwere performed on 600 μm thick sheets. This choice was made\ndue to the sensor size of the TWA apparatus, which is 0.25 ×\n0.5 mm (as illustrated in Figure S1 ). Table 1 shows the measurement\nresults of thermal conductivities of each LCE sheets. The orientation\nof the thermal conductivity measurements is denoted by ∥ for\nin-plane and ⊥ for the out-of-plane (thickness) direction.\nThe thermal conductivity of 0.35 W/(m·K) in the out-of-plane\n(thickness) direction for the polydomain LCE sheet is approximately\nthree times higher than that of the monodomain LCE or reference acrylic\nelastomer sheet. It is hypothesized that the increase in the thermal\nconductivity of the polydomain LCE is attributed to the formation\nof an ordered structure facilitated by the presence of mesogens. In\ncontrast, for the in-plane direction, both sheets exhibited thermal\nconductivities more than three times higher than those observed in\nthe thickness direction. The reason for this phenomenon is assumed\nthat the in-plane orientation would be formed by the effect of the\nhydrophobic surface of the PET film and the surface tension on the\nsurface in contact with air. This is the typical characteristic of\nthermal conductivity anisotropy of polymer films; the longer molecular\nchains lie down randomly in the in-plane direction, as a result, facilitating\nheat transfer along the direction of the molecular chains rather than\nbetween molecules. 4 , 14 , 38 , 39 , 48 − 50 Table 1 Thermal Conductivities and Elastic\nModulus for LCE Monodomain, Polydomain, and Reference Nonmesogenic\nMultifunctional Acrylic Elastomer Sheets (Thickness = 600 μm) material monodomain LCE polydomain LCE reference direction ∥ (in plane) ⊥ (out of plane) ∥ (in plane) ⊥ (out of plane) ∥ (in plane) ⊥ (out of plane) thermal conductivity [W/(m·K)] (1) x -direction : 1.8 ± 0.15 (2) y -direction : 3.0 ± 0.17 0.13 ± 0.02 1.0 ± 0.13 0.35 ± 0.04 0.30 ± 0.03 0.10 ± 0.02 density [g/cm3] 1.25 ± 0.01 1.25 ± 0.01 1.06 ± 0.01 specific heat capacity [J/(g·K)] 1.20 ± 0.02 1.20 ± 0.02 1.93 ± 0.02 elastic\nmodulus [MPa] 0.15 ± 0.02 0.11 ± 0.01 0.05 ± 0.02 In particular, the in-plane directional thermal conductivities\nof LCEs were over 1 W/(m·K), surpassing the in-plane thermal\nconductivity of polyimide (0.7–2.1 W/(m·K)). 4 , 38 , 50 The in-plane thermal conductivity\nof the monodomain was 1.8 W/(m·K) in the x -direction\nand 3.0 W/(m·K) in the y -direction, demonstrating\nvery high values not seen in transparent and flexible polymers. In\ngeneral, the thermal conductivity of polymer films increases in the\ndirection of stretching and decreases tremendously in the thickness\ndirection, as has been reported for PE and acrylic resins in many\ncases. 7 − 17 For this LCE, the thermal conductivity increased in the stretch\ndirection as above and was even 1.7 times higher in the in-plane y -direction, which is orthogonal to the stretch direction.\nThis unique phenomenon prompted the study of higher-order structural\nanalysis for these LCE sheets by using X-ray diffraction (XRD) analysis. Figure 5 shows the\nresults of transmission 2D-WAXS measurements of LCE polydomain and\nmonodomain sheets to the out-of-plane (thickness) direction. In the\npolydomain sheet (a), the intermolecular periodic structure (about\n0.45 nm) around 2θ = 20° is isotropic and circularly scattered;\nalthough in the monodomain sheet (b), the intermolecular periodic\nstructure around 2θ = 20° is strongly scattered in parallel\nto the LCE sheet stretching direction. Figure 5 Higher-order structural\nanalysis for (a) polydomain LCE and (b)\nmonodomain LCE sheets, using transmission 2D-WAXS (out-of-plane direction)\n(thickness= 600 μm). Focusing on the small angle region, a 2θ\npeak was observed\nat 2.7° ( d = 3.25 nm) of polydomain sheet, but\nno peak existed for the monodomain sheet. Here, the molecular length\nof BABMB calculated from the linear distance between atoms in the\nMMFF94s force field using the structure optimization program Conflex\nis shown in Figure 6 . The 2θ = 2.7° ( d = 3.25 nm) value in Figure 5 is close to the\nmolecular length of the BABMB. Figure 6 Molecular lengths of BABMB, EDDET, and\nPETMP calculated from linear\ndistances between atoms in the MMFF94s force field using the “Conflex”\nprogram. Although the intensity of this peak is weak, it\ncould be attributed\nto the periodic structure of the BABMB molecule. As shown in the schematic\ndiagram in Figure 5 , it is hypothsized that several mesogens would be periodically aligned\nin all directions, and the alkyl chains (EDDET, PETMP) are connected\nto BABMB in a random structure. Next, the results of 2D-WAXS\ntransmission measurements on the inner\nlayer of LCE sheets were conducted and shown in Figure 7 and Figure S4 . No directional differences were observed in the polydomain sheets\nin Figure S4 ; the circular diffraction\nscatter image, which exhibits almost isotropic characteristics but\nwith a slight orientation toward the in-plane direction, was observed\nin the 2D-WAXS. However, significant differences in higher-order structure\nwere observed between the x - and y -directions in the monodomain LCE sheets, as depicted in Figure 7 b. Specifically,\nthe intermolecular periodic structure (about 0.45 nm) around 2θ\n= 20° in the inner layer of the orthogonal y -direction to the stretching direction was strongly in-plane oriented\nin parallel to the stretching direction of the sheet. The characteristic\nfour-point pattern to cybotactic nematics with stronger ordering 51 − 53 was not observed. In contrast, for the x -direction,\na circular diffraction scattering image was observed in the 2D-WAXS\nimage, suggesting that the intermolecular periodic structure is isotropic.\nNo peak was observed at 2θ = 2.7° ( d =\n3.25 nm) for all LCE sheets, indicating no periodic structure of the\nBABMB molecule would exist in the inner layer. Figure 7 Higher-order structural\nanalysis for (a) polydomain LCE and (b)\nmonodomain LCE sheets, using transmission 2D-WAXS (in-plane direction)\n(thickness= 600 μm). Focusing on the small angle region, a weak but\nsharp scattering\nwere observed in the vicinity of 2θ from 1.3° to 1.5°\nfor all the sheets. Although the d -spacing is around\n6 nm, which is nearly equivalent to two molecules of BABMB, the scattering\ndirection is in the thickness direction of the LCE sheet. Additionally,\nthe intensity is very weak compared to that of around 2θ = 20°;\ntherefore, it is considered that there is little contribution to the\nthermal conductivity properties. As shown in Figure 8 , the slight difference in\nthe periodic structure between the intermolecular\nplanes at the wide-angle region (around 2θ = 20°) had little\neffect on the thermal conductivity. The width and thickness of the\nLCE sheet after stretching are macroscopically reduced to around 70\nand 80%, respectively; however, the d -spacing is\nnot changed nanoscopically. Therefore, it is considered that the reason\nfor the higher thermal conductivity in the y -direction\nis due to the formation of covalent bonds that act as new heat transfer\npathways in the direction intersecting with the x -direction during photo curing. Figure 8 Relation between d -spacing\nvalues obtained from\nXRD analysis and their thermal conductivities (λ) measured by\nthe TWA method (thickness= 600 μm). Consequently, we concluded that the heat energy\nentering into the y -direction once spreads to the x -direction,\nbut the synergistic effect of the high level of ordered inner structure\nand the covalent bonding structure due to cross-linking in the y -direction contributes to its higher thermal conductivity\ncompared to that in the x -direction, which exhibits\na random in-plane structure. 3.2 LCE Alumina Composite Sheet Polydomain\nLCE, while soft and opaque, possesses high thermal conductivity in\nthe thickness direction, so it might be applied to thermal interface\nmaterials (TIM) if it is made into a composite. Therefore, we aimed\nto fabricate the LCE alumina composite sheet for this purpose. Figure 9 illustrates (a)\nappearance of the obtained cured LCE alumina composite sheet with\nan alumina filler content of 75 vol % and a thickness of 300 μm,\n(b) a photo of the composite sheet wrapped around and adhered to a\n6 mm diameter aluminum pipe demonstrating flexibility, and (c) confirmation\nof alumina filler dispersibility using POM (under crossed Nicols).\nAs depicted in Figure 9 b, the cured LCE alumina composite sheets have flexibility even with\na 75 vol % alumina filler. The elastic modulus is approximately 70\nMPa, as shown in Table 2 . The POM observation result indicates that different particle sizes\nof the alumina particles were evenly dispersed throughout the composite\nLCE material. Achieving good dispersibility is crucial as it greatly\ninfluences both thermal conductivity and insulation properties. 2 Table 2 shows the measurement results of thermal conductivities and\nthe elastic modulus, where the orientation of the thermal conductivity\nmeasurements is indicated by ∥ for in-plane and ⊥ for\nout-of-plane (thickness) directions. For comparison, the thermal conductivity\nof the acrylic elastomer composite without mesogen is also provided\nin Table 2 . Figure 9 Photographs\nof each: (a) appearance of the alumina composite sheet\nusing polydomain LCE, (b) appearance of wrapped and adhered to a 6\nmm diameter aluminum pipe, and (c) a POM image of the alumina composite\nsheet under crossed Nicols (thickness= 300 μm). Table 2 Thermal Conductivities and Elastic\nModulus for LCE Alumina Composite Sheets and Reference Nonmesogenic\nMultifunctional Acrylic Elastomer Composite Sheets (Thickness= 300\nμm) composite\nmaterial 75 vol % of alumina\nof LCE 75 vol % of alumina\nof Reference direction ∥ (in plane) ⊥ (out of plane) ∥ (in plane) ⊥ (out of plane) thermal conductivity\nof\ncomposite [W/(m·K)] 16.5 ± 2.5 9.8 ± 0.6 8.5 ± 0.5 3.9 ± 0.2 density [g/cm3] 3.24 ± 0.05 3.20 ± 0.05 specific heat capacity [J/(g·K)] 0.87 ± 0.03 0.88 ± 0.02 elastic\nmodulus [MPa] 70 ± 1 30 ± 1 The LCE alumina composite sheet exhibits an average\nvalue of 9.8\nW/(m·K) in the thickness direction. This value is over twice\nas high as that of the acrylic elastomer composite sheet without mesogen.\nUsing Bruggeman’s equation, 2 , 54 − 58 the thermal conductivity of the base resin was estimated to be 0.5\nW/(m·K) from Figure 10 . The apparent thermal conductivity of the LCE in the composite\n(0.50 W/(m·K)) being larger than that of the resin-only sheet\n(0.35 W/(m·K) in Table 1 ) is assumed to be due to the interfacial interaction and\nmolecular orientation of the LCE molecules at the alumina interface\nin the composite. Figure 10 Composite thermal conductivity prediction and validation\nbased\non Bruggeman’s equation."
} | 5,206 |
24429552 | PMC3893645 | pmc | 7,068 | {
"abstract": "Bioelectrochemical systems (BESs) share the principle of the microbially catalyzed anodic substrate oxidation. Creating an electrode interface to promote extracellular electron transfer from microbes to electrode and understanding such mechanisms are crucial for engineering BESs. In this study, significantly promoted electron transfer and a 10-times increase in current generation in a BES were achieved by the utilization of carbon nanotube (CNT) network, compared with carbon paper. The mechanisms for the enhanced current generation with the CNT network were elucidated with both experimental approach and molecular dynamic simulations. The fabricated CNT network was found to be able to substantially enhance the interaction between the c -type cytochromes and solid electron acceptor, indicating that the direct electron transfer from outer-membrane decaheme c -type cytochromes to electrode might occur. The results obtained in this study will benefit for the optimized design of new materials to target the outer membrane proteins for enhanced electron exchanges.",
"discussion": "Discussion In this work, a CNT network-based electrode interface was fabricated to promote microbial extracellular electron transfer from microbes to electrode and the mechanisms for such an enhancement were elucidated by experimental and computational approaches. Nanosized materials have been used to increase current generation in microbial fuel cells 14 20 21 . However in the previous reports, the mechanisms responsible for this improvement have not been investigated and some were attributed to a high biomass colonization on these electrodes 16 . As for S. oneidensis MR-1, the outer membrane c-type cytochrome exhibited a high affinity to nanosized metal oxides 26 . This interaction not only enabled bacteria insert electrons into the metal oxides 25 27 , but also enhanced the bacteria/solid interfacial and interspecies electron transfers 28 29 . However, in a BES how the electron transfer could be modulated rather than only a high biomass colonization by CNTs is unknown yet. Figure 5f and g shows the electron transfer pathway, which is proposed on the basis of our experimental results and previous reports. Cytochromes MtrC and OmcA are located on the extracellular face of the outer membrane, and can be directly involved in electron transfer to an extracellular solid acceptor, such as Fe(III) minerals 4 , or to extracellular electron shuttles 8 . Because of the size effect of CNTs, the direct communication between active site of cytochromes and CNTs is feasible. In addition, the CNT network is rich in redox functional groups, as evidenced by XPS ( Figure 2 ), which are also useful to accelerate electron transfer at the bacteria/electrode interface ( Figure 5f ) 30 . Δ OmcA/MtrC mutant was severely impaired in the sustained electron transfer rate, but it could attach to electrodes in a similar manner 31 . When the Δ OmcA/MtrC mutant was used, the current with a small background level resulted from the interaction between the CNT network and unknown proteins ( Figure 5g ). In addition, molecular dynamic simulations further confirm the experimental results and reveal that the interaction between the CNT network and c -type cytochromes is one of the main mechanisms responsible for the enhanced microbial electron transfer, which is associated with the distorted porphyrin ring, short distance of electron transfer and more negative interaction energy. In summary, we have achieved the enhanced extracellular electron transfer from bacteria to electrode through CNT network, and presented clear evidence for the role of the CNT network to bridge cell cytochromes and electron acceptor for such an enhancement using both experimental and computational approaches. The results obtained in this study will benefit for designing new materials to target the outer membrane protein for enhanced electron exchanges between cells and electrode."
} | 988 |
39858905 | PMC11767486 | pmc | 7,069 | {
"abstract": "Polyhydroxybutyrate (PHB) is a biodegradable natural polymer produced by different prokaryotes as a valuable carbon and energy storage compound. Its biosynthesis pathway requires the sole expression of the phaCAB operon, although auxiliary genes play a role in controlling polymer accumulation, degradation, granule formation and stabilization. Due to its biodegradability, PHB is currently regarded as a promising alternative to synthetic plastics for industrial/biotechnological applications. Azohydromonas lata strain H1 has been reported to accumulate PHB by using simple, inexpensive carbon sources. Here, we present the first de novo genome assembly of the A. lata strain H1. The genome assembly is over 7.7 Mb in size, including a circular megaplasmid of approximately 456 Kbp. In addition to the phaCAB operon, single genes ascribable to PhaC and PhaA functions and auxiliary genes were also detected. A comparative genomic analysis of the available genomes of the genus Azohydromonas revealed the presence of phaCAB and auxiliary genes in all Azohydromonas species investigated, suggesting that the PHB production is a common feature of the genus. Based on sequence identity, we also suggest A. australica as the closest species to which the phaCAB operon of the strain H1, reported in 1998, is similar.",
"conclusion": "5. Conclusions Since the elevated production cost of PHB, compared to petroleum-based plastics, is still a significant barrier to the wide use of this biopolymer in industrial settings, it might be crucial to use an omics approach (genomics, proteomics and transcriptomics) in studies focusing on bacteria PHB accumulation by utilizing organic low-cost wastes as metabolic substrates at different growth conditions [ 10 ]. The results of our study, in addition to the first draft genome sequence of A. lata strain H1, supplies a comprehensive delineation of the genetic repertoire of PHB genes in the genus Azohydromonas and underlines the importance of comparative genomics for informing the design of biotechnological applications based on microbial species. The genome sequencing of PHB producers, such as A. lata , helps to increase knowledge of the genetic network involved in PHB biosynthesis. Our study suggests that the PHB pathway is a common evolutionary feature of the genus Azohydromonas and provides data for further analysis on the ecological and genomic issues of PHB production. This knowledge can be exploited from a biotechnological point of view and for studies (with an omics approach) on PHB production, evolutionary dynamics of PHA operons and auxiliary genes and their possible horizontal gene transfer.",
"introduction": "1. Introduction The exponential increase in fossil-derived plastic waste and the growing demand for plastics [ 1 , 2 ] create the urgency to replace petrochemical-based plastics with biodegradable polymers [ 3 , 4 , 5 ]. Polyhydroxyalkanoates (PHAs) are a group of bio-based polyesters that are biodegradable, resemble synthetic plastics, produced by a range of diverse prokaryotes and accumulated in the cytoplasm as granules reserve of carbon and energy [ 6 , 7 , 8 ]. Based on the carbon atoms content per monomer unit, PHAs can be classified as short-chain-length (scl-PHAs), with 3–5 carbon atoms, and medium-chain-length (mcl-PHAs), containing 6–14 carbon atoms, per unit. Scl-PHAs are rigid and fragile with a high melting temperature and low glass transition temperature; they are the most abundant PHAs among prokaryotes. Mcl-PHAs are elastic with lower melting and glass transition temperatures, as compared to scl-PHAs [ 7 ]. Among the scl-PHAs, the homopolymer polyhydroxybutyrate (PHB) is one of those considered for large-scale production due to its biodegradability and biocompatibility, being proposed in medical and pharmaceutical fields as well [ 9 ]. Thus, PHB is currently regarded as one of the most promising PHAs for biotechnological applications, with increasing studies on PHBs producing bacteria and a growing interest to make PHB more competitive in commercial markets [ 10 , 11 ]. From the view of a circular economy, many efforts target efficient and low-cost PHB production by using, in addition to native producers, bacteria (e.g., Escherichia coli ) engineered with heterologous PHB genes/operons with the objective of producing PHB efficiently from renewable biomass (e.g., whey waste, starch, wastewater) [ 12 ]. Gram-negatives, such as Azohydromonas lata (formerly Alcaligenes latus ) strain H1 (DSM1123; ATCC 29714), Azotobacter spp., Cupriavidus necator (formerly Ralstonia eutropha ) strain H16 (DSM 428; ATCC17699), Pseudomonas spp. and also recombinant Escherichia coli expressing the PHB biosynthetic genes from native producers strains, can accumulate large amounts of PHB and are considered the most promising systems for large-scale PHB production [ 9 , 13 ]. C. necator H16, in particular, represents the model organism for the study of PHB production [ 12 , 14 ]. Three main distinct pathways for PHB synthesis are described [ 7 ]; in C. necator , three genes, usually organized in an operon ( phaCAB ), are deemed sufficient for PHB biosynthesis: PHA synthase ( phaC ), 3-chetothiolase ( phaA ) and acetoacetyl-CoA reductase ( phaB ). PhaCs (the crucial function common to all the known pathways) are generally categorized into four classes (I to IV) based on amino acid sequence, in vivo substrate specificities of the enzymes and the number/composition subunits forming the catalytic complex [ 15 ]. An additional class (class V) was recently proposed by Tan and colleagues, based on the PhaC of Janthinobacterium sp. [ 16 ]. Beyond the three main genes ( phaA , phaB and phaC ), many auxiliary genes have been identified and characterized to encode important functions in controlling polymer accumulation and degradation. These include the regulatory gene phaR , the depolymerase phaZ , the extracellular oligomer hydrolase phaY , and the phasin phaP involved in granule formation/stabilization [ 17 , 18 , 19 ]. Since the availability of the reference C. necator H16 genome [ 20 , 21 ], the phaCAB operon, auxiliary genes associated with PHB production and isologs of PhaA (beta-ketothiolases) and PhaB (reductases) with different substrate specificity, have been characterized and studied in this bacterium [ 21 , 22 ]. However, equivalent considerations do not apply to the A. lata strain H1. Expanding the repertoire of known/characterized PHB genes could provide crucial insights for the optimization of biotechnological applications based on the genetic engineering of the pathway. Here, we present the first de novo genome assembly of A. lata strain H1, whose information might be exploitable for biotechnological applications [ 13 , 14 ]. Although the sequence of the phaCAB operon of A. lata H1 has been previously determined [ 23 ], we reasoned that the assembly of the genome sequence could offer significant advances in the identification of genes associated with PHB production and offer a more complete representation of the organization of the PHB operon, together with the number and configuration of auxiliary genes and potential isologs in A. lata . Moreover, our comparative genomic analyses of phaCAB and related genes within the genus Azohydromonas could advance the general knowledge of the genomic organization and conservation of these genes and inform the design of prospective biotechnological applications.",
"discussion": "4. Discussion The increasing usage of plastic in many anthropic activities has caused a global environmental crisis of plastic pollution, with serious risks for animal and human health. PHB is a bio-based biologically degradable polymer suitable for producing biodegradable plastic, representing an eco-friendly alternative to synthetic plastic [ 46 ]. Being nontoxic, it also has a promising role in designating strategies related to regenerative medicine and tissue engineering [ 9 ]. The improved understanding of both PHB metabolism (e.g., synthesis and degradation) and the genetic repertoire in native and recombinant bacteria producers, might contribute to a widespread industrial application of PHB-based materials to meet the growing global demand of green bioplastics from renewable sources [ 46 ]. Here, we report the first draft genome assembly of the bacterium Azohydromonas lata H1, a PHB producing strain with potential biotechnological applications. The taxonomic assignment of A. lata H1 within the genus Azohydromonas , was confirmed by a sequence similarity matrix based on ANI and dDDH. However, patterns of genome sequence identity, coupled with the observation of substantial differences in the size of the genome, might advocate for a partial revision of the genus Azohydromonas itself. More specifically, A. sediminis (YIM 73032, SYSU G00088) and Calidifontimicrobium sp. (SYSU G02091) presented levels of ANI of around 76% with all the other Azohydromonas species, a value that is considered borderline for the delineation of a bacterial genus; moreover, the size of the genome was significantly reduced in these species (~3.8 Mb compared to ~7.4 Mb of other Azohydromonas species, Table S4 ). Further, since there is a high level of identities between the sequences of A. sediminis (YIM 73032, SYSU G00088) and Calidifontimicrobium sp. (SYSU G02091), they should be reclassified under the same species. An analysis of the genome sequencing also revealed the presence of a 456 Kbp megaplasmid, not reported in the genome assembly of the reference strain A. lata NBRC 102462. The megaplasmid harbors two distinct type II TA systems (here named II-A and II-B). The II-B system is, to the best of our knowledge, the first report of a VapB/RelE/HigA three component system. By focusing on genes implicated in PHB production, in addition to the identification of the phaCAB operon, different auxiliary genes associated with PHB utilization and granule formation were detected. These included phaR , phaP , phaY , as well as extra copies of phaA and phaC and a number of potential phaA and phaB isologs. Based on sequence similarity, four distinct PhaC functions (two PhaC1 and two PhaC2 based on C. necator H16 annotation) were found, of which the two PhaC1 were putatively assigned to class I while the two PhaC2 were assigned to class V. This finding is consistent with the possibility of a broad substrate utilization for PHB production, although experimental investigations are required to validate such a hypothesis. Our comparative genome analyses indicate the presence of PHB and its related genes in all the Azohydromonas genome assemblies considered, suggesting that all the considered species of the genus Azohydromonas are endowed with the molecular machinery for PHB production. According to the observed patterns of PHB gene distribution, A. australica is the species with the largest repertoire of PHB genes within the genus Azohydromonas , and speculations purely based on gene dosage/gene number would suggest high levels of PHB production in this species. Consistent with this hypothesis and in consideration of our sequence similarity analyses, we suggest that the phaCAB described by Choi et al. in 1998 [ 23 ]—which was isolated from a specimen with “high concentration with high productivity” of PHB—is assigned to A. australica and not A. lata H1. By using modern approaches based on genome sequence and identity/similarity metrics, we derive a more precise and unequivocal identification of the species of origin of the phaCAB operon originally described by Choi et al. in 1998 [ 23 ], and we speculate that—at least in A. australica —increased PHB production might depend on PHB gene number/gene dosage, rather than on the optimization of the catalytic activity of a specific enzyme. This observation, coupled with the widespread distribution of PHB-related genes across the genus Azohydromonas , as evidenced by our analyses, prompts for further functional studies for a more accurate characterization of the levels of PHB production, underlying molecular pathways and potential biotechnological applications of Azohydromonas ."
} | 3,052 |
35105487 | null | s2 | 7,071 | {
"abstract": "Engineered microsystems for in vitro studies of cultured cells are evolving from simple 2D platforms to 3D architectures and organoid cultures. Despite advances in reproducing ever more sophisticated biology in these systems, there remain foundational challenges in re-creating key aspects of tissue composition, architecture, and mechanics that are critical to recapitulating in vivo processes. Against the backdrop of current progress in 3D fabrication methods, we evaluate the key requirements for the next generation of cellular platforms. We postulate that these future platforms - apart from building tissue-like structures - will need to have the ability to readily sense and autonomously modulate tissue responses over time, as occurs in natural microenvironments. Such interactive robotic platforms that report and guide cellular events will enable us to probe a previously inaccessible class of questions in cell biology."
} | 232 |
37402639 | PMC10340439 | pmc | 7,072 | {
"abstract": "Abstract Social networks can influence the ecology of gut bacteria, shaping the species composition of the gut microbiome in humans and other animals. Gut commensals evolve and can adapt at a rapid pace when colonizing healthy hosts. Here, we aimed at assessing the impact of host-to-host bacterial transmission on Escherichia coli evolution in the mammalian gut. Using an in vivo experimental evolution approach in mice, we found a transmission rate of 7% (±3% 2× standard error [2SE]) of E. coli cells per day between hosts inhabiting the same household. Consistent with the predictions of a simple population genetics model of mutation–selection–migration, the level of shared events resulting from within host evolution is greatly enhanced in cohoused mice, showing that hosts undergoing the same diet and habit are not only expected to have similar microbiome species compositions but also similar microbiome evolutionary dynamics. Furthermore, we estimated the rate of mutation accumulation of E. coli to be 3.0 × 10 −3 (±0.8 × 10 −3 2SE) mutations/genome/generation, irrespective of the social context of the regime. Our results reveal the impact of bacterial migration across hosts in shaping the adaptive evolution of new strains colonizing gut microbiomes.",
"introduction": "Introduction Individuals that share the same space are known to harbor more similar microbiota compositions than individuals that do not. In humans, substantial strain sharing was found among cohabiting persons, with 12% and 32% strain-sharing rates observed for the gut and oral microbiomes, respectively ( Valles-Colomer et al. 2023 ). This indicates that microbial migration between hosts, which is increased when living in the same household, is an important factor structuring species diversity in the microbiome ( Johnson and Clabots 2006 ; Johnson et al. 2008 ; Siranosian et al. 2021 ). Indeed, it has been shown that cohousing of mice, which are coprophagic, can reduce the diversity of the microbiota species composition among hosts. Cohousing is also a common practice aimed at reducing microbiota variation in many studies of host immune phenotypes ( Ericsson and Franklin 2015 ). Emerging data in both mice and humans show that significant evolutionary change can occur within strains of each microbiota species ( Garud and Pollard 2020 ). In germ-free mice colonized with either a single or multiple strains of commensal bacteria ( Li et al. 2015 ; Barroso-Batista et al. 2020 ; Yilmaz et al. 2021 ) or in mice with a native microbiota ( Barroso-Batista et al. 2014 ; Lescat et al. 2017 ; Frazão et al. 2019 , 2022 ), evolutionary change has been observed within days, weeks, or months. Time series data of human metagenomes also show that several adaptive evolutionary events can occur within months ( Garud et al. 2019 ; Zhao et al. 2019 ). To begin unraveling how bacterial transmission across hosts may affect patterns of molecular evolution of their gut microbes, population genetics theory of metapopulations, where a deme mimics a host, can be useful ( Pannell and Charlesworth 1999 ; Booker et al. 2021 ). Theoretical models of adaptation in the context of metapopulations predict that the rate of adaptive evolution of a clonal population (i.e., of bacteria) resulting from the accumulation of new beneficial mutations should be affected by the amount of migration/transmission between demes (i.e., individual hosts). In particular, for a given level of migration, the rate of adaptation should be increased relative to when no migration occurs ( Gordo and Campos 2006 ; Yeaman and Whitlock 2011 ). The models typically consider a single microbial species and therefore ignore the multispecies complexity characteristic of the gut ecosystem. However, their predictions should be robust for ecosystem level complexities under conditions where the strength of intraspecific competition is much higher than that of interspecies competition. Such conditions are observed when ecological models aiming at explaining the diversity and stability of microbiomes, like the generalized Lotka–Volterra, are fitted to 16s RNA data ( Coyte et al. 2015 ). Here, we study how the evolutionary path of a new lineage colonizing the mouse gut is influenced by its host social environment. We performed both transmission and evolution experiments. The former were designed to estimate Escherichia coli transmission rates among cohoused mice, while in the latter we use in vivo experimental evolution to test the following hypotheses: 1) that cohousing increases the rate of evolution of commensal bacteria and 2) that high migration rates lead to an increase in the number of shared evolutionary events accumulating across hosts. We test this hypothesis for the two key processes of evolution: horizontal gene transfer (HGT) and mutation accumulation. Under HGT, we expect that adaptive events occurring in only one mouse will transmit rapidly to the entire mouse metapopulation. Under mutation accumulation, we expect higher allele sharing in a social regime if adaptive evolution within the host is marked by intense competition between clones carrying different adaptive mutations. In fact, when clonal interference is pervasive within hosts, which live in the same environment (e.g., same diet), migration/transmission may help the bacterial clones which carry the highest combination of beneficial alleles to spread more rapidly across hosts.",
"discussion": "Discussion Bacterial evolution in the mammalian intestine has been studied mainly under asocial conditions, where bacterial transmission among mammalian healthy hosts does not take place ( Barroso-Batista et al. 2014 , 2015 ; Lourenço et al. 2016 ; Lescat et al. 2017 ; Sousa et al. 2017 ; Frazão et al. 2019 , 2022 ; Ghalayini et al. 2019 ; Ramiro et al. 2020 ). To our knowledge, the evolution of E. coli in the presence versus absence of bacterial transmission in mice bearing a microbiota is compared here for the first time. We established a mouse model of bacterial transmission to compare E. coli evolution when colonizing the gut of mice living in a social or asocial regime. We found that in mice inhabiting the same cage 7% (±3% 2SE) of the E. coli colonizing each mouse gut results from daily migration events. It is known that E. coli can adapt to better colonize the mouse intestine in less than a week of colonization, with the mutation and phage-driven HGT events occurring concomitantly ( Frazão et al. 2019 ). Here we observed that E. coli transmission between hosts can still occur even when an adapted E. coli is already colonizing the gut. Interestingly, the first clone to colonize a given mouse was not always the one ending up as dominant after a month of evolution, suggesting that priority effects ( Sprockett et al. 2018 ) should not be dominant in this species. The E. coli population size, a key aspect in determining the level of variability in a population and the effectiveness of selection relative to drift ( Charlesworth 2009 ), was not affected by transmission, with populations isolated from mice living socially or asocially presenting similar loads. The E. coli cells transmitting among microbiome-bearing mice was comparable to the 10% migration inferred in a study using germ-free animals ( Vasquez et al. 2021 ), suggesting that E. coli transmits between hosts at a very similar rate independently of the presence or absence of a microbiome. Interestingly, we show invasion and coexistence of the same strain of E. coli (expressing different fluorescent markers) in mice living in a social regime, suggesting that strain colonization resistance, observed for other important species in the gut such as Bacteroides fragilis ( Lee et al. 2013 ), is absent from the E. coli species. Importantly, whole-genome sequencing of populations revealed that when in a social regime, where host-to-host bacterial transmission is pervasive, the evolution of E. coli is always characterized by an elevated ratio of nonsynonymous to synonymous mutations, indicative of evolution driven by strong selection of adaptive haplotypes, a phenomenon we found to be less common in an asocial regime. Our theoretical model of bacterial transmission predicted that the outcome of bacterial migration between hosts would affect the evolutionary process in terms of the number of shared evolutionary events between hosts. The model simulations predicted that when the number of migrant clones between mice is large enough (≫1), the metapopulation of cohoused mice acts as a single population and genetic changes are shared across hosts. Alternatively, when transmitting a very low number of bacterial migrants, each host behaves as an independent population, with the level of sharing of genetic evolutionary events being extremely low. Indeed, a significantly higher proportion of shared evolutionary events (total and adaptive) was found in E. coli populations from the social condition. These observations relate not only to the mutational process but also to the evolutionary phage-driven events. We were able to study the latter process because our mouse cohort was colonized with a resident E. coli carrying several active prophages that can be transferred to the invader strain, thus conferring an adaptive potential to consume sugars that are common in the gut ( Frazão et al. 2019 ). The adaptive mutational process targeted mostly the same biological processes in both invader E. coli populations, isolated from the social or asocial regime, namely, production of pyrimidine nucleotides from pseudouridine ( psuK/fruA mutation) and fructoselysine consumption ( frlR mutation). Phages are known key players in shaping the composition and diversity of bacterial communities in many environments, namely, in the gut of mice and humans ( Kim and Bae 2018 ; Frazão et al. 2019 ; Sutton and Hill 2019 ). Here, we observed that phage-driven HGT events, namely, the phages Nef and KingRac, were transferred to the invader E. coli genome as observed previously ( Frazão et al. 2019 ). This process was independent of the social or asocial condition, though limited to when cocolonization with both the resident and invader lineages occurred. Nevertheless, under a social regime, bacterial transmission across hosts spread these adaptive phage-driven events across all cohoused animals. We found that under the social regime, total evolutionary events reached significantly higher frequencies in the invader population, with adaptive events following the same tendency. This was indeed predicted in the simulation modeling, where the highest frequency mutations were more common at higher levels of bacterial transmission. This suggests that unlike in asocial contexts, in a social regime, a beneficial mutation reaching high frequency should be more prone to transmit to another host and avoid the drift barrier. Invader populations evolving during transmission did not show any signs of diversifying selection. On the contrary, these populations exhibited genetic homogeneity and an elevated dN/dS ratio indicative of adaptive evolution driven by strong selection. The present study also reveals that the evolution rate of E. coli is unchanged in both social and asocial settings, being on average 3.0 × 10 −3 (±0.8 × 10 −3 2SE) mutations/genome/generation. A possible explanation is that in a social regime, due to E. coli transmission, the number of adaptive mutations in each animal is expected to be higher, given that these can originate not only de novo in each animal but also in other hosts being subsequently transmitted. By sweeping in the population, transmitted adaptive mutations may reduce genetic diversity and the number of nonadaptive mutations in the receiving mouse, thus leading to higher adaptive evolution (high dN/dS ratio) while preserving the same molecular evolution rate as in the asocial context. Interestingly, the evolution rate found is similar to those reported in previous studies in vitro ( Good et al. 2017 ) and in vivo ( Frazão et al. 2022 ), where bacterial host-to-host transmission was not taking place. Thus, as proposed previously ( Frazão et al. 2022 ), E. coli appears to follow a clock-like rate of evolution, which the present data show is independent of the host social context where E. coli evolves. We conclude that selective pressures acting on bacteria depend on the host's social status and that the effect of migration should be taken into account when analyzing the evolutionary history of a bacterium colonizing the mammalian gut. Moreover, in contrast to a social context, social isolation can lead to psychoemotional stress, impair normal development of organs and tissues ( Grigoryan et al. 2022 ), and can alter the gut microbiota ( Donovan et al. 2020 ). These differences related to the host's social status could constitute an additional layer of selection pressure besides bacterial transmission, also contributing differential bacterial evolution in the gut. Our findings demonstrate that the evolution of new bacterial strains invading the gut microbiota can be strongly affected by the host's social environment, which calls for further experimental evolution studies comparing social versus asocial conditions."
} | 3,320 |
22735101 | null | s2 | 7,073 | {
"abstract": "We report here our systematic characterization of a photoinduced electron-transfer (ET) redox cycle in a covalently linked donor-spacer-acceptor flexible system, consisting of N-acetyl-tryptophan methylester as an electron donor and thymine as an electron acceptor in three distinct solvents of water, acetonitrile, and dioxane. With femtosecond resolution, we determined all the ET time scales, forward and backward, by following the complete reaction evolution from reactants to intermediates and finally to products. Surprisingly, we observed two distinct ET dynamics in water, corresponding to a stacked configuration with ultrafast ET in 0.7 ps and back ET in 4.5 ps and a partially folded C-clamp conformation with ET in 322 ps but back ET in 17 ps. In acetonitrile and dioxane, only the C-clamp conformations were observed with ET in 470 and 1068 ps and back ET in 110 and 94 ps, respectively. These relatively slow ET dynamics in hundreds of picoseconds all showed significant conformation heterogeneity and followed a stretched decay behavior. With both forward and back ET rates determined, we derived solvent reorganization energies and coupling constants. Significantly, we found that solvent molecules intercalated in the cleft of the C-clamp structure mediate electron transfer with a tunneling parameter (β) of 1.0-1.4 Å(-1) and the high-frequency vibration modes in the product(s) couple with the back ET process, leading to the ultrafast back ET dynamics in tens of picoseconds. These findings provide mechanistic insights of nonequilibrium ET dynamics modulated by conformation flexibility, mediated by unique solvent configuration, and accelerated by vibrational coupling."
} | 422 |
36976162 | PMC10055941 | pmc | 7,074 | {
"abstract": "The current study emphasizes fungi as an important tool against heavy metals and how isolated fungal species can be used to create a successful strategy for the bioremediation of chromium and arsenic-contaminated sites/soils. Globally, heavy metal pollution is a serious issue. In the current investigation, contaminated sites were chosen, and samples could be taken from various localities of Hisar (29.1492° N, 75.7217° E) and Panipat (29.3909° N, 76.9635° E), India. A total of 19 fungal isolates were obtained from the collected samples through the enrichment culture technique using PDA media supplemented with Cr as chromic chloride hexahydrate (50 mg/L) and As as sodium arsenate (10 mg/L) and the potential of fungal isolates to be used for the removal of heavy metals was examined. The isolates were screened for minimum inhibitory concentrations (MIC) exhibiting tolerance capabilities, and the four best isolates C1, C3, A2, and A6 with the highest MICs (>5000 mg/L), were chosen for further investigations. To use the chosen isolates in the remediation of heavy metals (Cr and As), the culture conditions were optimized. The fungal isolates C1 and C3 estimated the highest removal of 58.60% and 57.00% at 50 mg/L chromium concentration, while the isolates A6 and A2 recorded the highest removal efficiency of 80% and 56% at 10 mg/L arsenic concentration under optimal conditions. Finally, the chosen fungal isolates C1 and A6 were molecularly identified as Aspergillus tamarii and Aspergillus ustus , respectively.",
"conclusion": "4. Conclusions The current study reported the potential of fungal isolates for the removal of heavy metals, i.e., chromium and arsenic. The maximum removal was observed by up to 58.6% and 80% for chromium and arsenic, respectively. Fungal isolates could grow over a wide range of environmental conditions. The effective performance of fungal isolates in the present investigation provided a potent tool for the bioremediation of chromium and arsenic-contaminated sites. Further studies are required to check the metal (Cr & As) removal ability of isolates from actual waste waters to find a more viable alternative. Mycoremediation offers the ability to rehabilitate contaminated ecosystems affordably and successfully. Uncertainty arises from a lack of information regarding how different environmental elements can affect the rate and degree of biodegradation.",
"introduction": "1. Introduction Environment protection has become one of our prime concerns in prevailing conditions. Increased industrialization and urbanization and even specific repair sites have raised the issue of heavy metal pollution in the environment. Heavy metals having a specific density of more than 5 g/cm 3 are considered hazardous pollutants globally [ 1 ]. High toxicity, non-biodegradability, and the subsequent build-up of heavy metals in the environment make the problem more severe. Industrial effluents and municipal wastes are either discharged into water bodies or directly supplied to the fields [ 2 ]. This results in serious health issues in humans due to the accumulation of heavy metals in the human body [ 3 ]. Among various heavy metals, Arsenic (As) and Chromium (Cr) contamination has become a major problem. Chromium holds the first rank among carcinogenic substances [ 4 ]. Based on oxidation, chromium usually occurs in two forms: Cr (III) and Cr (VI). The former is less toxic and mobile than the latter. However, Cr (VI) reacts with other particles that are present in the air and changes to Cr (III), which is more stable than Cr (VI) [ 5 , 6 , 7 ]. There are various sources of chromium emissions in the environment, including anthropogenic sources, which include tanneries, steel industries, and fly ash [ 8 ]. Chromium toxicity leads to various health issues, including fatal chronic diseases [ 9 ]. The presence of arsenic in the groundwater and soil of many developing countries such as India, Thailand, Bangladesh, Nepal, Argentina, and Poland is a major health alarm [ 10 ]. Inorganic forms which are mainly present in the environment are Arsenate (V) and Arsenite (III). Inorganic forms of arsenic are more toxic compared to organic ones and are inter-convertible [ 11 ]. The pentavalent form of arsenic [As (V)] is a structural analog of inorganic phosphate and substitute phosphate in mitochondrial pathways and glycolysis [ 12 ]. The major sources of arsenic contamination are more natural than anthropogenic [ 13 , 14 , 15 , 16 , 17 ]. The tolerance limit of arsenic in drinking water was 10 μg/L as per World Health Organization guidelines, and it was mainly deposited in the nails, hairs, bones, and vital organs, such as the liver and kidneys [ 18 , 19 ]. There are several conventional methods for the elimination of heavy metals, which include chemical precipitation, ion exchange, ultra-filtration, reverse osmosis, electro-winning, carbon adsorption, and solvent extraction. The majority of these are expensive and unfit due to the release of highly hazardous toxic pollutants as by-products. Their high cost and environmental concerns due to the production of toxic by-products make them less effective [ 20 , 21 ]. Many bacteria can bioremediate in wastewaters, especially in the distillery [ 22 ] and textile [ 23 ] effluents. The bacterial cultures remediate these effluents as individuals or as a consortium [ 24 ] for various enzymes produced by microbes which can also be utilized for this purpose [ 25 ]. The plant growth-promoting rhizobacteria (PGPR) can remediate the contaminated sites as they are tolerant to a certain level of heavy metals [ 26 , 27 ]. Plants have the potential, through various molecular and physiological mechanisms, to alleviate abiotic stresses [ 28 ]. Plant–microbial interactions can play a very important role in such remediation processes [ 29 , 30 ]. Thus, the need of the hour is to choose eco-friendly approaches to deal with elevated levels of heavy metals in the environment. Mycoremediation is one of the most promising and eco-friendly approaches to the bioremediation of chromium and arsenic. Fungi have been known for their ability to adapt to harsh environmental conditions, such as pH, nutrient availability, temperature, and high metal concentrations. Fungi secrete various enzymes throughout their life cycle, which also helps in the bioremediation of metals. Fungal cell walls are constituted of various groups such as polysaccharides and proteins with carboxyl, sulfate, amino, hydroxyl, and phosphate groups for the binding of metal ions and which act as the most effective biosorbent for the removal of heavy metals [ 31 , 32 , 33 ]. Until now, several fungal species have been identified as Aspergillus flavus , Aspergillus fumigates, Fusarium proliferatum, Penicillium radicum, Beauvariabassiana, etc., for the bioremediation of these heavy metals [ 34 , 35 , 36 , 37 , 38 , 39 ]. Many fungal strains have been observed for the removal of heavy metals such as Phanerochaete chrysosporium , Aspergillus awamori , Aspergillus flavus , and Trichoderma viride from heavy metal-contaminated wastewater and industrial effluents [ 40 ]. In view of the above problem, the present investigation was taken to investigate the ability of indigenous fungal species to deal with selected heavy metals and to study the efficiency of heavy metal removal from the liquid medium. In short, a heavy metal uptake by fungal isolates was determined.",
"discussion": "3. Results and Discussion 3.1. Sample Collection and Analysis of Physico-Chemical Properties Samples of heavy metal-contaminated soil were collected from different locations in Haryana, and physicochemical analyses were performed. The electrical conductivity of the samples showed a huge variation from 0.51 to 2.50 mS/cm ( Table 1 ). The pH values of the samples did not show much variation. The organic carbon ranged from 0.16% to 0.68%. The organic carbon content was observed at a maximum at site no. 5 due to the high carbon content of sewage waste. It determines the soil’s ability to hold and immobilize heavy metals. The amount of nitrogen, phosphorus, and potassium was highest at site no. 3, 5, 10, and 11, among other sites, due to the high content of organic, inorganic, and nitrogenous compounds in sewage, textile, and tannery wastewater. Similar results were determined by Angin et al. [ 52 ]. High levels of chromium were found at sites no. 10 (1.80 mg/L) and site no. 11 (2.23 mg/L) due to high chromium in the textile and tannery industry effluents. Similar results were reported by Qing et al. [ 53 ] in the steel industrial city (Anshan) of China. Significant levels of arsenic were found at site no. 6 (0.69 mg/L) due to motor vehicle repairs such as bodywork, chemicals in the cleaning and dismantling of vehicles, painting, soldering, hydraulic fluid, engine oil spills, leachates from used oils and greases, and spare parts that are frequently burnt on these sites. Similar results were investigated in a study by [ 54 ]. 3.2. Isolation of Cultures and Their Minimum Inhibitory Concentration (MIC) A total of 19 morphologically distinct fungi were retrieved from different soil samples. It can be assumed that metal-polluted habitats contain a wide range of fungi from all major taxonomic groups [ 55 ]. Fungal colonies were screened for heavy metal resistance in terms of MIC, and the highest tolerance/resistance (>5000 mg/L) was reported for isolates C1 and C3 for chromium and A2 and A6 for arsenic ( Table 2 ). In a similar study which was conducted by Singh et al. [ 30 ], 54 fungal isolates were retrieved from arsenic-contaminated rice fields of middle Indo-Gangetic plains. Out of 54 isolates, 15 were tolerant to 10,000 mg L −1 of arsenates and were capable of bioaccumulation and volatilization. Aspergillus oryzae was the most prominent among them, with a bioaccumulation efficiency of 82%. 3.3. Morphological and Biochemical Characterization The fungal isolates C1, C3, and A2, A6 were selected based on MIC and characterized morphologically and biochemically. The morphological characterization of selected fungal isolates is shown in Table 3 , and microscopic images are shown in Figure 1 . Enzyme bioassays and carbohydrate utilization patterns of selected fungal isolates are shown in Table 4 . All the isolates were found to be positive for the carbohydrate assimilation and pectinase test. The fungal isolates C1 and A2 were found to be positive, and C3 and A6 were negative for the cellulase test. For the amylase test, the fungal isolate A6 was found to be positive, and the rest were negative. In the laccase test, the fungal isolates of C3 and A6 were found to be positive, and C1 and A2 were negative. A similar method of characterization was followed by Mohanty et al. [ 46 ]. The fungal isolates were identified as Aspergillus sp. on the basis of characterization. 3.4. Optimization of Cultural Conditions Fungi require specific growth conditions. The fungal isolates C1, C3, and A2, A6 showed maximum growth when the media was supplemented with 2% ( w / v ) of sucrose. The wet weight of mycelial mass with 2% sucrose was 3.95 g, 3.89 g, 3.27 g, and 4.07 g for the isolates A2, A6, C1, and C3, respectively. In the temperature optimization studies, 30 °C reported the maximum fungal biomass production of 3.72 g, 3.13 g, 3.20 g and 3.73 g for the fungal isolates A2, A6, C1, and C3, respectively. Among the different pH of the medium, maximum growth was found at pH 5 by all the selected fungal isolates, whereas minimum growth was observed at pH 2. At the slightly acidic pH 5, the weight of the fungal pellet was 4.02 g, 3.44 g, 3.84 g, and 3.94 g for the fungal isolates A2, A6, C1, and C3, respectively ( Figure 2 ). 3.5. Heavy Metals Removal Efficacy of Fungal Isolates The fungal isolates C1 and C3 showed the highest removal of 58.60% and 57.00% at 50 mg/L chromium concentration, followed by 53.80% and 50% at 100 mg/L, and the minimum removal of 46.3% and 42.2% at 200 mg/L chromium concentration under the optimized conditions ( Figure 3 ). In a similar study by Kumar and Dwivedi [ 35 ], for the bioremediation of chromium, the fungal isolate Trichoderma lixii CR 700 showed the removal of 99.4% at 50 mg/L of chromium concentration. For arsenic, the fungal isolates A2 and A6 showed the highest removal of 56.00% and 80.00% at ten mg/L arsenic concentration, followed by 53.20% and 67% at 50 mg/L, and the minimum removal of 46.6% and 52% were found at a 200 mg/L chromium concentration. In a similar study by Srivastava et al. [ 56 ], fungal isolates showed arsenic removal from 10.92% to 65.81% at ten mg/L of the arsenic concentration. In another study by Srivastava et al. [ 56 ], ten out of forty-five isolates were highly tolerant to arsenic. The isolates were able to remove 80% of arsenite and 85% of arsenate from the liquid medium. The variability might be due to different resistance tolerance mechanisms such as complexation, extracellular precipitation, crystallization, biosorption, the transformation of metals, efflux, etc., used by different fungal strains against metal [ 55 ]. 3.6. Molecular Characterization of Potent Isolates The phylogenetic trees of the fungal isolates C1 and A6 are shown in Figure 4 and Figure 5 . ITS sequences were submitted in NCBI through GenBank (accession number OQ179906 for C1 and accession number OQ 179908 for A6). The fungal isolate C1 was identified as Aspergillus tamarii, and the isolate A6 was identified as Aspergillus ustus ."
} | 3,370 |
33976129 | PMC8113556 | pmc | 7,075 | {
"abstract": "Realization of a self-assembled, nontoxic and eco-friendly piezoelectric device with high-performance, sensitivity and reliability is highly desirable to complement conventional inorganic and polymer based materials. Hierarchically organized natural materials such as collagen have long been posited to exhibit electromechanical properties that could potentially be amplified via molecular engineering to produce technologically relevant piezoelectricity. Here, by using a simple, minimalistic, building block of collagen, we fabricate a peptide-based piezoelectric generator utilising a radically different helical arrangement of Phe-Phe-derived peptide, Pro-Phe-Phe and Hyp-Phe-Phe, based only on proteinogenic amino acids. The simple addition of a hydroxyl group increases the expected piezoelectric response by an order of magnitude ( d 35 = 27 pm V −1 ). The value is highest predicted to date in short natural peptides. We demonstrate tripeptide-based power generator that produces stable max current >50 nA and potential >1.2 V. Our results provide a promising device demonstration of computationally-guided molecular engineering of piezoelectricity in peptide nanotechnology.",
"introduction": "Introduction Piezoelectric materials generate electrical energy in response to mechanical deformation. Piezoelectric devices are commonly made from a variety of inorganic materials and organic polymers 1 – 5 , which limits their deployment in health monitoring and regenerative medicine due to reliance on toxic starting metals, complicated synthesis procedures, weak oxidation stability and poor sustainability 6 . Green piezoelectric materials that satisfy the requirements of ultra-high mechanosensitivity, flexibility and durability would provide a promising route to biocompatible multifunctional smart energy harvesters. Piezoelectricity has been widely observed in several natural materials including bone, collagen, viruses, cellulose and chitosan 7 – 10 . However, the piezoelectric response exhibited by these biomaterials is typically in the range of 0.1–10 pm V −1 , which is low for many potential applications 11 . Among these various piezoelectric biopolymers, collagen exhibits useful chemical and physical properties of extensibility, high tensile strength and swelling 12 , 13 . Moreover, the piezoelectricity of collagen may play a pivotal role in controlling bone growth 14 , with fibrillar rat tail collagen exhibiting the highest measured shear d 14 value for a biomaterial of 12 pm V −1 (pC N −1 ) 15 . Biomimetics provides potentially disruptive advances in design and fabrication of novel functional materials. Using a minimalistic approach, self-assembled short peptides have emerged as versatile building blocks due to their inherent biocompatibility and highly engineerable properties that provide tailored functionality 16 – 19 . Although nanostructures formed by ultra-short peptide sequences predominantly exhibit β-sheet organization 20 – 22 , the collagen structure is helical 23 . The large number of directionally aligned hydrogen bonds in the helical structure creates a macroscopic dipole that can couple with external electric fields and shear force to produce the piezoelectric response of collagen 24 . Examination of all collagen residues has demonstrated the post-translationally modified amino acid hydroxyproline (Hyp) to show the highest piezoelectric response 25 . Yet, by itself, Hyp does not exhibit the fibrillar assembly and helical structural pattern of collagen, and it has only very recently proved possible to design ultra-short peptides that can mimic the collagen supramolecular architecture 26 . On the other side, earlier findings suggested that the dynamic interactions of aromatic amino acid side chains in peptide-based structures can increase electrical conductivity 27 , 28 . The aromatic Phe-Phe-based β-sheet-rich biomaterials have been investigated and utilized for fabrication of green energy harvester 29 , 30 . However, tailor-made design of ultra-short peptide to achieve high piezoelectric response similar to collagen through mimicking its supramolecular architecture has remained elusive. Molecular modelling, in particular density functional theory (DFT), has emerged as an effective tool for predicting and rationalizing the piezoelectric response of materials, from classic inorganic crystals 31 , 32 to polymers 33 and novel two-dimensional structures 34 , 35 . Our previous work has utilized DFT to accurately predict the elastic and piezoelectric tensors of amino acid 10 , 11 , peptide 25 and biomineral crystals 36 . DFT can be used to complement techniques such as piezoresponse force microscopy (PFM) 37 , 38 or used in standalone predictive modelling studies of nanoscale electromechanical phenomena 39 . Classical molecular dynamics (MD) simulations are also widely used to study the kinetics of piezoelectric systems 40 and are particularly effective at studying systems in a liquid environment 41 and over a range of temperatures 42 . Here, aiming to engineer a minimalistic collagen-mimicking short peptides, we develop radically different organization of Phe-Phe-derived short peptide based only on natural amino acid by including both Hyp and aromatic Phe moieties in the sequence. Both Pro-Phe-Phe and Hyp-Phe-Phe tripeptides assemble into a helical-like sheet that is stabilized by the dry hydrophobic zipper interface of Phe residues. The self-assembled fibrillar biomaterial composed of helical-like molecular arrangement of the tripeptides exhibits excellent mechanical strength and high piezoelectric response (see benchmarks in Supplementary Table 1 ), which is maximized in the strong H-bonding Hyp variant. We designed the material to exhibit markedly higher piezoelectricity, which is achieved by modulating the electromechanical response of the tripeptide via side-chain engineering to optimize supramolecular polarization. When used as the active component in a power generator, the helical tripeptides show much larger short-circuit current and open-circuit voltage output as compared to β-sheet-rich peptide. Our findings demonstrate the rational modulation of peptide self-assembly to create tailored functionality and mark a significant step forward in molecular engineering of peptide piezoelectricity for nanotechnology applications by exemplifying the importance of targeting both primary and secondary structure.",
"discussion": "Results and discussion Characterization of Hyp-Phe-Phe assemblies To engineer collagen-mimicking natural piezoelectric short peptides, we chose the self-assembling tripeptide, Pro-Phe-Phe (Fig. 1a ), an ultra-short helical natural peptide with high mechanical stability similar to that of collagen 26 , 43 . We substituted Pro for Hyp (Fig. 1a ), as it exerts the highest piezoresponse among the collagen component amino acids 25 . Building on the recently solved solution state morphology and secondary structure of Pro-Phe-Phe 26 , we used Fourier transform infrared (FTIR) and circular dichroism (CD) spectroscopy techniques to characterize the solution secondary structure of Hyp-Phe-Phe. The FTIR spectra showed a sharp amide I peak at 1645 cm −1 with a shoulder at 1680 cm −1 , indicating the predominant helical structure (Fig. 1b ) similar to Pro-Phe-Phe. The small peak shift towards lower wavenumber (compared with the standard helical structure, which usually falls near 1650–1655 cm −1 ) agrees with the predicted frequency shift for ultra-short helices 44 . The CD spectrum strongly supported the FTIR data, as it exhibited double negative maxima, characteristic of helical conformations (Fig. 1c ). The maxima at 210 and 230 nm were faintly red shifted compared to canonical double helix, as expected for shorter length peptides 45 . Fig. 1 Self-assembly and structural characterization of Hyp-Phe-Phe. a Chemical structure of Pro-Phe-Phe and Hyp-Phe-Phe. b FTIR analysis of Hyp-Phe-Phe assemblies showing the characteristic peak of helical conformation. c CD spectrum of the tripeptide reveals the presence of helical assemblies in solution. d AFM image of the tripeptide demonstrating fibrillar morphology. e TEM image of the tripeptide fibres. f Single-crystal structure of Hyp-Phe-Phe showing the formation of an elongated structure by stacking of helices through intermolecular hydrogen bonding and aromatic zipper-like packing. CCDC ref. no. 1823367 26 . g , h Mechanical strength of the Hyp-Phe-Phe fibres. g Nanoscale mapping of Young’s modulus ( Z -scale = 140 GPa). h Line section through one fibril as highlighted by the green line in g , showing the periodic variation in stiffness along the fibril. Additional AFM and TEM data are given in Supplementary Figs. 1 – 3 . The supramolecular assembly of Hyp-Phe-Phe was explored using atomic force microscopy (AFM) and transmission electron microscopy (TEM) (Fig. 1d, e and Supplementary Figs. 1 and 2 ). The images revealed that the tripeptide self-assembled into uniform high aspect ratio fibres of 500 nm in diameter that extended for several micrometres (aspect ratio, L/D > 500). Single-crystal X-ray diffraction structures revealed the favourable molecular level interactions that direct the supramolecular organization (Fig. 1f ). The asymmetric unit of the Hyp-Phe-Phe crystal comprised two peptide molecules sharing common structural features 26 . The torsion angles of the Phe 2 moiety were found to be localized within the right-handed helical region of the Ramachandran plot, with φ 2 and ψ 2 values of −71.1°, −70.5° and −43.2°, −41.9°, respectively, comparable to that of the Pro-Phe-Phe crystal. In the crystallographic b -direction, the adjacent molecules were connected through head-to-tail intermolecular H-bonds, generating an extended helical-like molecular organization (Fig. 1f , left and right illustrations). Neighbouring helical-like structural modules connected in a parallel orientation with the interface of the dimer stabilized by the aromatic zipper structure built from π – π interactions between the Phe side chains (Fig. 1f , middle illustration). Such a dry steric zipper interface has long been assumed to provide mechanical rigidity to amyloid fibres 46 and may provide additional stability to the Hyp-Phe-Phe assemblies. To investigate the nanomechanical properties of Hyp-Phe-Phe fibres, we used quantitative nanomechanical mapping AFM (QNM-AFM) (Fig. 1g, h ). The measured Young’s modulus of the fibrils varied in the range of 60–90 GPa, producing the same order of magnitude mechanical stiffness as recorded by AFM nanoindentation in the corresponding single crystal 26 . The preservation of the single-crystal level of mechanical rigidity in the peptide fibre was quite remarkable for a biomaterial and comparable with stiff biological materials such as the bone, collagen and enamel 47 from which we conclude that the presence of the “aromatic zipper” molecular motif led to the fabrication of the stiff macroscopic biomaterial. Definite corrugation was observed perpendicular to the long axis of the fibril (green line in Fig. 1g ), with a peak to peak distance of ~165 nm and a periodicity of 84.9 nm, as determined by direct Fourier transform analysis using the open source software FiberApp (Supplementary Fig. 3 ) 48 . This “mechanical periodicity” cannot be observed in the topography channel but is clear in the Young’s modulus channel. Natural collagen exhibits similar lateral periodicity of 67 nm due to gaps in the hierarchical structure of the triple helix 49 . Piezoelectric response of Pro-Phe-Phe and Hyp-Phe-Phe assemblies Having verified helical conformation, lateral periodicity and high mechanical robustness similar to collagen, we used DFT to predict the elastic, dielectric and piezoelectric constants of the tripeptides (Fig. 2 and Table 1 ). Calculations were carried out on the obtained solid-state XRD crystal structures, which contained no waters of crystallization. Full details of the computational methodology can be found in the ‘Methods’ section. The Pro-Phe-Phe and Hyp-Phe-Phe assemblies have very similar predicted dielectric constants ε r , with only a slight increase in the hydroxylated peptide (3.3 vs. 3.1, see Supplementary Table 2 ). This stems from 25% increase in the ε 3 tensor component for Hyp-Phe-Phe, compared to Pro-Phe-Phe (4.0 vs. 3.2). This axis of highest permittivity also gives the highest predicted piezoelectric charge tensor values. The lower shear elastic stiffness values in Hyp-Phe-Phe (Supplementary Table 3 ) indicate higher shear piezoelectric strain constants and voltage constants for this crystal, as the lower stiffness produces larger ionic displacement under an applied force. Supplementary Tables 4 and 5 show the DFT-computed piezoelectric charge ( e ), strain ( d ) and voltage ( g ) tensors of Pro-Phe-Phe and Hyp-Phe-Phe, respectively. For Pro-Phe-Phe, we observe that low charge tensor components of up to 56 mC/m 2 results in moderate d ij values of up to 3.1 pm V −1 (Fig. 2a ). However, as with most biomaterials 11 , the low dielectric constants produce significant voltage constants of up to 108 mV m/N ( g 22 ). In Hyp-Phe-Phe, hydroxylation lowers the symmetry of the unit cell (monoclinic to triclinic), which increases the number of non-zero piezoelectric constants in each tensor. Hydroxylation also gives a fivefold increase in the highest charge tensor value ( e 33 = 0.1 C m −2 ). Due to the increase in e ij values and decrease in c ij values, we notice a significant increase in the magnitude of the predicted piezoelectric strain constants, with d max = d 35 = −27.3 pm V −1 and d 33 = 4.8 pm V −1 (Fig. 2b ). Looking at the full piezoelectric tensor of the crystals enables more meaningful comparisons to other piezoelectric materials and identifies possible applications for these crystals. The monoclinic Pro-Phe-Phe crystal has eight finite tensor components, whereas the low-symmetry Hyp-Phe-Phe crystal has 17 components. The most commonly exploited piezoelectric response in devices is the longitudinal d 33 value (or crystallographic equivalent d 11 and d 22 ). Both tripeptide crystals have modest longitudinal constants that would see them fit similar applications to aluminium nitride, quartz and zinc oxide (ZnO) as evidenced by the values cited in Supplementary Table 1 . The Pro-Phe-Phe crystal shows a narrow range of tensor values, with coefficients in the range 0.8–2.5 pm V −1 . Hyp-Phe-Phe has a much wider range of values, varying from 0.1 pm V −1 to the maximum 27.3 pm V −1 ( d 16 and d 35 ). The high predicted shear value and doubling of the longitudinal response in the Hyp-Phe-Phe crystal highlights that simple chemical modifications can induce piezoelectric response in biomaterials that exceeds that of many inorganic crystals (Fig. 2e ). The maximum calculated response for Hyp-Phe-Phe is as large as the d 31 value of poled polyvinylidene difluoride, but is predicted to occur without the application of heat or a large external electric field. Given the similar ε r values for Hyp-Phe-Phe and Pro-Phe-Phe, we predict high voltage constants, on the order of 1 V m/N (Hyp-Phe-Phe, g max = g 16 = 1043 mV m/N). For comparison (Fig. 2f ), (K 0:5 Na 0:5 )NbO 3 (KNN)-based ceramics have exhibited voltage constants of g 33 = 40 mV m/N 31 . Other studies have reported values of up to 540 mV m/N ( g 33 ) for single crystals of BiB 3 O 6 32 . Combining our predicted piezoelectric voltage constants with approximate crystal dimensions, we can estimate single-crystal voltage outputs (Fig. 2g ). For example, applying a loading force of 4 μN 50 along the two axis of a Pro-Phe-Phe single crystal will give a predicted voltage output of ~7 mV. For Hyp-Phe-Phe single crystal, the calculated voltage output is 39 mV, fivefold higher under similar conditions. The benefit of predicting the full piezoelectric tensor with DFT and screening for individually high-tensor components is that we can design devices and orientate biomolecular crystals and assemblies in a way that maximizes output. Fig. 2 Piezoelectricity of Pro-Phe-Phe and Hyp-Phe-Phe assemblies. a , b Calculated piezoelectric strain constants for Pro-Phe-Phe ( a ) and Hyp-Phe-Phe ( b ). Selected values are marked above relevant columns. c , d Zoom-in on the electronic structure of the Hyp-Phe-Phe crystal. The π -density of the diphenylalanine motif remains identical across both crystals ( c ) and d shows the hydroxylated pyrrolidine structure of the Hyp residue in the Hyp-Phe-Phe crystal. e Comparison of piezoelectric strain response of different biological and non-biological materials: collagen piezoelectric response at different length scales 10 , 25 , 53 has been emphasized as compared to that of Pro-Phe-Phe and Hyp-Phe-Phe. f The predicted voltage coefficient of Pro-Phe-Phe and Hyp-Phe-Phe in comparison with some currently used inorganic materials. g Calculated output voltage upon applied force of 4 µN. h – j Experimental measurement of coefficients using PFM. h Linear relationship between the vertical piezoresponse of Hyp-Phe-Phe as measured by the photodiode system and applied voltage. Statistical distribution of the vertical d 33 coefficients ( i ) and shear d 34 coefficients ( j ). Additional PFM data are given in Supplementary Figs. 5 – 16 . Table 1 DFT calculated molecular dipoles and crystal dipoles for each tripeptide, and the corresponding experimentally measured piezoelectric response. Peptide Molecule dipole (Debye) Crystal dipole (Debye) Space group no. Longitudinal response permitted? Shear response permitted? Longitudinal response (pm V −1 ) Shear response (pm V −1 ) Pro-Phe-Phe 7.9 2.8 4 Yes Yes 2.2 <0.1 Hyp-Phe-Phe 6.7 1.9 1 Yes Yes 4.0 16 The supramolecular packing modes in the crystal confer the significant d 33 response for Hyp-Phe-Phe, as applying a force along the 3 -axis will align the hydroxylated pyrrolidines and couple to the net dipole in the unit cell. Our DFT and MD molecular models show that by strengthening the H-bonding network, we remodel the aromatic zipper motif and alter the net dipole in the unit cell (Table 1 , Supplementary Table 6 and Supplementary Fig. 4 ) to create more opportunities for charge transfer under stress. We have previously observed that lowering the crystal symmetry lowers the shear elastic constants and allows for more ionic displacement around the unit cell axes, for angles > 90° (in Hyp-Phe-Phe, the angle is 97°) 10 . Finally, the binding energy along the 3 -axis is slightly larger for Pro-Phe-Phe than for Hyp-Phe-Phe in the unstrained crystals (2.3 eV vs. 2.0 eV, Supplementary Table 7 ), which suggests that molecular arrangements that can facilitate an increase in intermolecular interactions under stress may create higher piezoelectric response. To study the hierarchal influence of hydroxylation 51 , we compare the current findings on tripeptide crystals and fibres to previous results obtained on Hyp and proline single crystals 25 . Only very small difference is observed in the predicted stiffness constants between proline and Hyp (24 GPa vs. 28 GPa), with this trend extending to the tripeptide crystals. The lower shear stiffness predicted for Hyp-Phe-Phe results in Young’s modulus of 12 GPa, slightly below the value of 14 GPa for Pro-Phe-Phe. Both tripeptides and amino acids show large increase in charge tensor values after hydroxylation and decrease in stiffness constants, despite decrease in the predicted molecular dipole. The net crystal dipoles for both crystals are along the crystallographic b or piezoelectric 2 axis, with absolute values of 2.8 Debye for Pro-Phe-Phe and 1.9 Debye for Hyp-Phe-Phe. The projected dipole moments along each crystal axis (Supplementary Table 6 ) show that Pro-Phe-Phe has a unidirectional dipole moment, as it only has a 22 longitudinal response. The lowering of symmetry in Hyp-Phe-Phe gives three longitudinal components, which disperses the dipole moment along all three axes (Supplementary Fig. 4 ). The trend for piezoelectric strain tensor values in Pro-Phe-Phe and Hyp-Phe-Phe tracks almost exactly as that for proline and Hyp amino acids with an increase in d max from ~3 to ~30 pm V −1 . To experimentally validate our predictive modelling, we first employed piezoresponse force microscopy (PFM) to characterize the crystals (Fig. 2h–j and Supplementary Figs. 5 – 16 ). Due to the small size of the peptide single crystals, conventional PFM imaging of topography, amplitude, phase, etc., was not possible. The crystals moved during imaging attempts. In line with best practice techniques 34 , 52 , we then carried out PFM point measurements where the probe was brought into contact with the single crystal and held stationary while the applied voltage was varied and the piezoresponse recorded. These were used to generate plots similar to Fig. 2h , which were then used to create statistical distributions of the response. The measurements were carried out with the probe stationary relative to the sample and at a low frequency (21 kHz), which minimized artefacts resulting from topographic crosstalk or resonance enhancement of the signal. Stiff probes with a spring constant of 5–6 N m −1 were used to mitigate electrostatic and flexoelectric contributions 34 . All measurements were carried out at 20 °C and 40% relative humidity (RH) ambient laboratory conditions to ensure uniformity in the measurement conditions. Identical point measurements were carried out on both positive and negative controls, to verify the accuracy of the technique and to rule out any instrumental backgrounds or parasitic effects contributing to the signal, as described in detail in Supporting Information section 3 . The linear relationship between the piezoresponse as measured by the photodiode system and applied voltage, and also the minimal frequency dependence of output piezoresponse are indicative of a genuine piezoelectric property (Fig. 2h and Supplementary Figs. 13 – 16 ). The results revealed the vertical coefficient \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$${d}_{33}^{{\\mathrm{eff}}}$$\\end{document} d 33 eff of Pro-Phe-Phe assemblies to be 2.15 ± 0.86 pm V −1 (Supplementary Fig. 13 ), which rose to 4.03 ± 1.96 pm V −1 (Fig. 2i ) for Hyp-Phe-Phe. Measuring the shear piezoelectricity of Hyp-Phe-Phe yielded an effective shear piezoelectric coefficient \\documentclass[12pt]{minimal}\n\t\t\t\t\\usepackage{amsmath}\n\t\t\t\t\\usepackage{wasysym} \n\t\t\t\t\\usepackage{amsfonts} \n\t\t\t\t\\usepackage{amssymb} \n\t\t\t\t\\usepackage{amsbsy}\n\t\t\t\t\\usepackage{mathrsfs}\n\t\t\t\t\\usepackage{upgreek}\n\t\t\t\t\\setlength{\\oddsidemargin}{-69pt}\n\t\t\t\t\\begin{document}$${d}_{34}^{{\\mathrm{eff}}}$$\\end{document} d 34 eff of 16.12 ± 2.3 pm V −1 (Fig. 2j ), which is higher than the experimentally measured magnitude of LiNbO 3 (13 pm V −1 ), ZnO (12 pm V −1 ), amino acid γ-glycine (10 pm V −1 ) and protein/peptide biomaterials M13 bacteriophage (6–8 pm V −1 ) and collagen film (1 pm V −1 ) 10 , 50 , 53 . To further probe the assemblies, we performed MD simulations (Fig. 3 , Supplementary Figs. 17 – 22 and Supplementary Table 8 ). These simulations are performed on a nanocrystal of each peptide immersed in a large water box to model bulk solvation, as described in the ‘Methods’ section. MD are calculated at constant room temperature and atmospheric pressure at physiological pH of 7.4. The most important finding is that the tightening of the H-bond network in the Hyp variant is associated with increased conformational freedom in the Phe-Phe rings, showing the balance between electrostatic and π – π interactions in the creation of the helical assemblies. Figure 3 shows the computed supramolecular packing in both assemblies with the computed difference map in Fig. 3c comparing the separation vs. contact angle distributions in the Phe-Phe ring contacts for Hyp-Phe-Phe and Pro-Phe-Phe. The angle–distance contour plot makes it possible to identify different conformations between neighbouring phenyl rings, which helps characterize the changes induced by the additional –OH group in the Hyp variant. The difference map is used to highlight these changes. The negative value in the 5–6 Å and small angle region on Fig. 3c show that Pro-Phe-Phe presents larger density of highly ordered π – π contacts. By contrast, the Phe-Phe ring distance is shifted towards larger values in Hyp-Phe-Phe with an overall broadening of the density peaks, illustrating the increased conformational freedom. The evidence for the stability of the zipper for both Pro-Phe-Phe and Hyp-Phe-Phe have been obtained from both MD simulations and crystal structures, which confirm the rigidity and durability of the zipper as a structure-building motif (Supplementary section 4 and Supplementary Fig. 20 – 23 ). Fig. 3 Molecular dynamics (MD) simulations of Pro-Phe-Phe and Hyp-Phe-Phe assemblies. a , b The computed supramolecular packing structure in the assemblies after 0.1 µs of room temperature dynamics of Pro-Phe-Phe ( a ) and Hyp-Phe-Phe ( b ). The balance between increased H-bonding and more flexible π – π contacts is quantified in the Pro-Phe-Phe vs. Hyp-Phe-Phe difference map in c . The distance corresponds to the distance between neighbouring phenyl rings and the angle to the angle between the same two rings, as described in the ‘Methods’ section. The density is the normalized histogram and is given as a percentage. The 2D histogram is built with a 200 × 200 matrix. H-bond distributions are plotted in d . Additional supporting MD data are provided in Supplementary Figs. 14 – 19 and Supplementary Table 8 . Characterisation of peptide-based power generator Finally, to test the potential of the peptide assemblies for energy harvesting and piezoelectric sensing in integrated microdevices, a coin-size power generator was designed and fabricated by tightly sandwiching the tripeptide assembly film between only the two Ag electrodes that were connected to the measuring instrument via copper wires (Fig. 4a and Supplementary Fig. 24 ). The entire device was firmly laminated with kapton tape to protect against mechanical stress, dust and humidity. When the device was compressed and released, the peptide-based nanogenerator converted the mechanical energy into electricity. Mechanical loads were applied to the generators using a dynamic mechanical test system and the resulting electrical output signal was characterized by measuring the short-circuit current and open-circuit voltage. A periodic compressive force was applied on the power generator and the output performance is shown in Fig. 4b, c . The switching-polarity test showed that the current and voltage signals were reversed when the device connection was switched (Fig. 4d, e ) 54 . The switching-polarity test results excluded the errors from the variation of contact resistance or parasitic capacitance and confirmed that the detected electrical signal was truly from the piezoelectric peptide assemblies. Fig. 4 Characterization of the helical peptide-based nanogenerator. a Photograph and schematic configuration of coin-sized piezoelectric energy generation measurement set-up utilized as a direct power source with peptide assemblies as the active component. b , c Open-circuit voltage ( b ) and short-circuit current ( c ) of piezoelectric energy harvester in the forward connection. d , e The generated voltage output ( d ) and current output ( e ) in the reverse connection. f , g Variable external load was applied onto the device. The measured open-circuit voltage ( f ) and short-circuit current output ( g ) of Hyp-Phe-Phe under different applied force. h , i Linear dependence of the voltage ( h ) and current ( i ) output on the applied force. The black and red lines indicate forward and reverse connection, respectively. j – l Comparison of piezoelectric response of the engineered helical tripeptides with β-sheet forming Phe-Phe peptide assemblies. j The self-assembled nanostructures and β-sheet molecular arrangement in the crystal structure of Phe-Phe (CCDC ref. no 16340 83 ). k Short circuit current obtained from the generator using Phe-Phe assemblies as the active component upon applied force = 23 N. l Comparison of output current of Phe-Phe with helical peptides, Pro-Phe-Phe and Hyp-Phe-Phe. Error bars = SD ( n = 3). For Pro-Phe-Phe, under an applied force F = 55 N, the output open-circuit voltage ( V oc ) reached 1.4 V (Supplementary Figs. 25a, c ), which is significantly higher than reported peptide and inorganic alternatives (Supplementary Table 9 ). The corresponding short-circuit current ( I sc ) was 52 nA (Supplementary Figs. 25b, d ), which is significantly higher than the output current obtained from nanogenerators based on M13 bacteriophage virus (6 nA) or fish skin collagen (1.5–20 nA) (Supplementary Table 9 ). Using the Hyp-Phe-Phe assemblies as the active layer (Fig. 4f, g ), similar short-circuit current output (39.3 nA) was achieved when the applied force was only 23 N, half that applied for Pro-Phe-Phe. This validated our theoretical predication of increased piezoelectric response due to simple hydroxylation of the side chain. The corresponding output voltage was 0.45 V. The peak voltage and current increased linearly as a function of the applied force at rates of 15.08 mV N −1 and 1.33 nA N −1 , respectively (Fig. 4h, i ), demonstrating linear piezoelectric response of the peptide assemblies. Furthermore, the high mechanical rigidity suggests that power generation can be sustained under a cyclic force (17 N) (Supplementary Fig. 26 ) and the output voltage showed no degradation over 1000 press/release cycles for more than 60 min, indicating the high durability of the peptide-based devices. Finally, the voltage output characteristic was measured with different external load resistors connected to a nanogenerator, while it was repeatedly deformed, and the result is shown in Supplementary Fig. 27 . The output voltage continually rises with the growth of load resistance, demonstrating the electrical characteristics of the power device and illustrating its potential for practical applications. We expect that the device performance can be further increased in the future by fabricating highly ordered aligned arrays 42 , which will allow to scale up our peptide-based piezoelectric devices to generate a higher energy output. To further understand the relation between atomic level structural organization and macro-scale piezoelectricity, we fabricated a control device using the well-established β-sheet forming dipeptide, Phe-Phe (Fig. 4j and Supplementary Figs. 28 and 29 ). Under the same applied force, F = 23 N, the maximum output open-circuit voltage and short-circuit current was only 0.14 V (Supplementary Fig. 29a, b ) and 3.9 nA (Fig. 4k ), respectively. Although current and voltage output increased linearly with mechanical force (Supplementary Fig. 29c, d ), the value was much lower compared to the helical peptides even at the highest applied force (Fig. 4l ). The significantly decreased performance of the analogous aromatic dipeptide under a similar applied force highlights how differences in the structural organization at the atomic level can bear a dramatic effect on biomaterial piezoelectricity. Structural component, molecular organization and piezoelectric response Currently, the piezoelectricity of biomaterials is not fully understood at the molecular level. The hierarchical structural organization of biomaterials is not included in the classical piezoelectric models, leading to discrepancies between their predictions and experimental results 55 . There is emerging evidence that the piezoelectric response of biomaterials profoundly depends on the atomic level structure and self-assembled hierarchical organization, as well as on the amplitude and sign of the dipoles of the constituent amino acid residues. Both collagen and its tripeptide minimalistic analogues Pro-Phe-Phe and Hyp-Phe-Phe exhibit helical structures. Moreover, the shortest repeating unit of collagen also displays a molecular arrangement and symmetry similar to that of the studied tripeptides 56 . The Phe groups have been assumed to facilitate aromatic molecular orbital overlap in the underlying parent peptide, potentially leading to electron delocalization and generation of a significant band gap 28 . Thus, previous molecular design incorporated aromatic Phe residues in helical peptide sequences in order to engineer electron transport properties in the resulting assembled structures. This is consistent with the lower piezoelectric response of the Hyp-Leu-Phe peptides, which retains the helical organization in the crystal but possesses reduced aromaticity (Supplementary section 6 ). At the same time, among the collagen amino acids, Hyp showed the highest piezoelectric response in a single-crystal form 25 . Our molecular models show that hydroxylation significantly increases the crystal polarizability under stress. Thus, Hyp-Phe-Phe satisfies all the specific requirement of high piezo sensitivity in terms of structure and interactions, thereby exhibiting the highest piezoelectric response among the studied natural short peptides. Exploring the potential of renewable, sustainable and green energy sources to replace fossil fuels is one of the most significant and urgent challenges in energy research. The piezoelectric effect in proteins is an intriguing phenomenon that can potentially allow a better interface between the semiconductor and biological worlds. Molecular engineering of peptide piezoelectricity is critical not only to understand the molecular basis of piezoelectricity in biomaterials, but also to unravel the core recognition motif for modular design of short peptides with a predictable high piezoelectric response. Here we fabricated a simple biopiezoelectric device made from collagen-mimicking ultra-short peptide sequences that could achieve high current and voltage output, similar to that obtained using nanogenerators comprising inorganic materials or organic polymers. We found common helical molecular organization in the packing of collagen and our designed short peptides, which can explain their high piezoelectricity. A significantly enhanced energy output can be obtained by simple replacement of one constituent with another of higher polarizability (in our case, C-H → C-OH on a pyrrolidine ring). This engineered higher response to applied strain emphasizes the importance of incorporating polar building units in piezoelectric biomaterials. Moreover, sequence mutation by replacing Phe for Leu noticeably decreases the piezoelectric response, although having similar molecular organization emphasizing the significant role of aromatic groups. Our demonstration of collagen-level piezoelectricity in rationally designed ultra-short peptides reinforces the value of prediction-led molecular engineering of piezoelectricity to accelerate the deployment of peptides in nanotechnology applications."
} | 8,854 |
38454600 | PMC11052696 | pmc | 7,076 | {
"abstract": "Many bacterial habitats—ranging from gels and tissues in the body to cell-secreted exopolysaccharides in biofilms—are rheologically complex, undergo dynamic external forcing, and have unevenly distributed nutrients. How do these features jointly influence how the resident cells grow and proliferate? Here, we address this question by studying the growth of Escherichia coli dispersed in granular hydrogel matrices with defined and highly tunable structural and rheological properties, under different amounts of external forcing imposed by mechanical shaking, and in both aerobic and anaerobic conditions. Our experiments establish a general principle: that the balance between the yield stress of the environment that the cells inhabit, σ y , and the external stress imposed on the environment, σ , modulates bacterial growth by altering transport of essential nutrients to the cells. In particular, when σ y < σ , the environment is easily fluidized and mixed over large scales, providing nutrients to the cells and sustaining complete cellular growth. By contrast, when σ y > σ , the elasticity of the environment suppresses large-scale fluid mixing, limiting nutrient availability and arresting cellular growth. Our work thus reveals a new mechanism, beyond effects that change cellular behavior via local forcing, by which the rheology of the environment may modulate microbial physiology in diverse natural and industrial settings.",
"introduction": "Introduction Many bacterial environments—e.g., gels and tissues inside hosts, subsurface soils and sediments, exopolysaccharides in biofilms and in the environment, activated sludge in sewage treatment plants, and food products ( 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8 , 9 , 10 , 11 , 12 , 13 )—are neither perfectly elastic solids nor simple viscous fluids. Instead, they are yield stress materials: they behave as viscoelastic solids when exposed to weak mechanical stresses but flow when the imposed stress exceeds a threshold yield stress σ y . Indeed, these habitats are typically not quiescent but are subjected to continuous and dynamic external forcing by moving boundaries, fluid flows, and other mechanical stressors. Thus, depending on the balance between the yield stress σ y and external stresses σ , the environments that bacteria inhabit can vary between being solid-like and liquid-like. A familiar example is mucus, which serves as a habitat for both commensal and pathogenic bacteria in diverse animals. In healthy humans, airway mucus is often a runny solution with a small or negligible yield stress; however, many respiratory disorders are characterized by a more concentrated mucus whose yield stress can be as large as tens of pascals ( 14 , 15 , 16 , 17 ). As a result, the beating cilia that line the airways are less effective at clearing mucus from the lungs—leading, in some cases, to chronic and deadly infections ( 15 , 18 ). Another familiar example is the polymer matrix that encapsulates the cells in a surface-attached biofilm, such as the plaque that we brush off our teeth and the slime that can grow on industrial equipment, medical catheters, and even in our showers. In some cases, this matrix is weak and easily yielded, whereas in others, its yield stress can be as large as thousands of pascals ( 13 , 19 , 20 , 21 , 22 , 23 , 24 )—which is thought to contribute to biofilm virulence and recalcitrance to treatment. Given that the yield stress of the environments that bacteria inhabit can vary so widely ( Table 1 ), with implications for health, environment, and our everyday lives, we ask: How do changes in yield stress influence bacterial behavior? Table 1 Order of magnitude estimates of rheological parameters characterizing bacterial environments in nature and in this study Environment Yield stress σ y (Pa) External stress σ (Pa) Bi ≡ σ y / σ References This study: granular hydrogel matrix ≪ 10 − 2 − 10 2 0 − 1 ≪ 10 − 2 − ≫ 10 2 this study, ( 24 , 25 , 26 , 27 ) Lung mucus: healthy 10 − 1 − 10 10 − 4 − 10 2 10 − 3 − 10 5 ( 14 , 15 , 16 , 17 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 ) Lung mucus: cystic fibrosis 10 − 1 − 10 2 10 − 4 − 10 2 10 − 3 − 10 6 ( 14 , 15 , 16 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 35 , 36 , 37 , 38 ) Biofilm EPS: Pseudomonas aeruginosa 1 − 10 3 10 − 2 − 10 6 10 − 6 − 10 5 ( 13 , 20 , 21 , 23 , 30 , 38 , 39 , 40 , 41 , 42 , 43 ) Biofilm EPS: Staphylococcus epidermidis 1 − 10 10 − 2 − 10 6 10 − 6 − 10 3 ( 22 , 24 , 30 , 38 , 39 , 40 , 41 , 42 , 43 ) Mucus yield stress values are drawn from studies with human and porcine lung and gastric mucus, and biofilm extracellular polymeric substance (EPS) yield stress values are drawn from studies of Pseudomonas aeruginosa and Staphylococcus epidermidis . External stress for mucus is taken to be that arising from coughing, phasic breathing, mechanical ventilation, digestion, and ciliary clearance. External stresses for biofilms include examples of coughing, blood flow, mechanical agitation from tooth brushing, mixed bioreactors, and shear cleaning of biofouling on ship hulls. Prior research has investigated how other aspects of environmental rheology can influence bacterial behavior via local mechanical forcing ( 44 , 45 ). For example, a growing body of work is elucidating the ways in which individual bacterial cells on 2D planar surfaces sense and respond to changes in surface stiffness and topography via local mechanical forces using their flagella, pili, and membrane proteins ( 46 , 47 ). Other work has shown how 3D confinement of dense, multicellular colonies causes cells to rearrange, slow down growth, and even induce biofilm formation due to large cell-cell contact forces ( 48 , 49 , 50 , 51 , 52 , 53 ). In bulk liquids, studies have shown how local stresses generated by fluid flows generated by bacterial swimming alter swimming kinematics ( 54 , 55 , 56 , 57 , 58 , 59 ). Here, we report another, distinct mechanism by which environmental rheology impacts a fundamental aspect of bacterial physiology—cellular growth. By studying Escherichia coli growth inside permeable 3D granular hydrogel matrices, we show that the balance between the yield stress σ y and external stress σ , quantified by the Bingham number Bi ≡ σ y / σ and tuned over a wide range in our experiments, modulates bacterial growth by altering transport of externally supplied essential nutrients to the cells. In particular, when the matrix is fragile enough to be fluidized by shaking (small Bi), mixing transports oxygen from the boundaries of the matrix to the cells; as a result, the bacteria are able to perform aerobic respiration and thereby achieve a high growth yield ( 60 , 61 , 62 , 63 ). In stark contrast, when the matrix is tough enough to withstand shaking (large Bi), its elasticity hinders mixing; consequently, the majority of the bacteria become oxygen depleted, arresting their growth cycle and greatly diminishing the resultant biomass yield ( 60 , 61 , 62 , 63 ). Notably, owing to the unique structure of the hydrogel matrices, this transition between continued and arrested growth is not associated with changes in the local mechanical environment encountered by individual cells. Hence, this mechanism by which environmental rheology modulates bacterial physiology—by altering large-scale transport of nutrients—complements other mechanisms that rely on local mechanical forcing instead. Because many bacterial habitats have complex rheological properties, encounter dynamical external forcing, and have heterogeneous nutrient distributions, we anticipate that our findings are applicable to microbial life in diverse natural and industrial settings.",
"discussion": "Discussion By probing E. coli growth inside permeable 3D granular hydrogel matrices, we have shown that the balance between the yield stress σ y and external stress σ can modulate bacterial growth by altering transport of externally supplied nutrients. This balance can be quantified using Bi ≡ σ y / σ , a dimensionless parameter used in the field of rheology ( 72 ), as summarized in Fig. 4 . In particular, our experiments demonstrate that when the cells inhabit strongly forced (large σ ) and easily fluidized (small σ y ), and thus well-mixed over large length scales, environments, nutrients are more readily available to the cells, and their growth can progress completely ( Bi < 1 shown in the bottom right of Fig. 4 ). By contrast, when the cells inhabit weakly forced (small σ ) and tough (large σ y ) environments, slow diffusion from external boundaries limits the availability of growth-limiting nutrients, and cellular growth is arrested ( Bi > 1 shown in the top left of Fig. 4 ). The transition between these two different growth behaviors across Bi = 1 can be remarkably sharp as a function of small changes in rheology, as shown in Fig. 1 , C and D . This mechanism by which environmental rheology modulates bacterial physiology by altering large-scale nutrient transport does not arise from local mechanical interactions, unlike other mechanisms ( 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 56 , 57 , 58 , 59 ); instead, it likely operates in conjunction with them. Figure 4 State diagram describing how the balance between environmental yield stress σ y and the external stress σ modulates bacterial growth by modulating nutrient transport. When cells inhabit fragile environments that are fluidized by external forcing ( bottom right ), nutrients ( blue ) are well mixed over large length scales, and cellular growth can progress completely. By contrast, when cells inhabit tough environments that are not fluidized ( top left ), nutrients are not well mixed, and their availability is diffusion limited, arresting cellular growth. The boundary between these two growth regimes is described by the diagonal line, Bi ≡ σ y / σ = 1 . As shown in Table 1 , natural bacterial communities inhabit settings that span both regimes of this state diagram. To see this figure in color, go online. The conceptual picture described above, and schematized in Fig. 4 , focuses on nutrient transport over large length scales. However, we note that even when Bi < 1 , nutrient transport can still be limited by diffusion over small length scales. This limitation can be quantified using another dimensionless parameter—the Péclet number. This quantity compares the rates of nutrient transport by advective mixing of the fluidized environment under external forcing, ∼ u / L , and diffusion, ∼ D / L 2 : Pe ≡ u L / D , where u is the characteristic advection speed, L is the characteristic length scale of nutrient transport from its source to the cells, and D is the nutrient diffusivity. That is, nutrient transport is dominated by advective mixing—as assumed above—when the characteristic distance from a nutrient source to the cells L ≥ D / u . By contrast, if Pe < 1 even when Bi < 1 , then we expect that nutrient transport will be diffusion limited as in the Bi > 1 case. For our experiments, Pe ≥ 1 for all length scales ≳ 80 nm, nearly two orders of magnitude smaller than the height of our hydrogel matrices and, indeed, smaller than an individual cell itself—thus, nutrient transport does indeed occur primarily by advection in our experiments when Bi < 1 . However, one could imagine other scenarios involving much slower-diffusing nutrients or weaker external forcing in which, even for Bi < 1 , L < D / u and nutrient transport to the cells is still diffusion limited. Exploring this possibility will be a useful direction for future work. Our experiments used plate reader shaking as a way of testing the influence of external forcing. They also used tunable granular hydrogel matrices as well-defined and well-characterized model materials to test the influence of environmental rheology, enabling us to explore a broad range of Bi in a well-defined and systematic manner. Moreover, as an illustrative example, we used ambient oxygen as a growth-limiting nutrient for cells that grow more efficiently via aerobic respiration. However, natural settings may present additional complexities in, e.g., the nature of external forcing, the microstructure and chemical properties of the bacterial environment, and the biochemical properties of the nutrient. Nevertheless, we expect that our central finding, summarized by Fig. 4 , applies more generally across different forms of forcing, microbial cell types, and nutrient sources in diverse environments where growth-limiting nutrients are not uniformly abundant. In fact, given that bacterial habitats (e.g., mucus in the body, extracellular polymer networks in biofilms) have such widely varying rheological properties and encounter diverse forms of external forcing (e.g., mechanical agitation, imposed fluid flows), Bi spans a broad range of values from much smaller to much larger than one in natural and industrial settings—as summarized in Table 1 . Exploring the generality of our findings through experimental measurements of yield stresses, externally imposed stresses, and growth behavior across diverse bacterial types and environments will therefore be a useful direction for future work. While our study focused on overall bacterial growth, we conjecture that the interplay between environmental yielding and external forcing can modulate other aspects of bacterial physiology as well. For example, we showed that when Bi > 1 , nutrient availability is limited by the competition between diffusion and uptake by cells, with greater limitation for slower diffusion and/or faster uptake, respectively. Under these conditions, nutrients are distributed heterogeneously throughout space, potentially driving collective migration and nonuniform spatial organization of cells in a population ( 73 , 74 ). Such “patchy” nutrient availability can also promote the establishment of phenotypic and genotypic heterogeneity, as well as competition and cooperation via metabolic cross-feeding, in a population ( 75 )—with implications for, e.g., the maintenance of genetic diversity in the population, as well as its resilience to external stressors such as administered antibiotics ( 76 , 77 , 78 , 79 ). Investigating how our findings may translate to other such changes in bacterial physiology will be an interesting avenue for future research."
} | 3,586 |
32351469 | PMC7174568 | pmc | 7,078 | {
"abstract": "Rumen fermentation affects ruminants productivity and the environmental impact of ruminant production. The release to the atmosphere of methane produced in the rumen is a loss of energy and a cause of climate change, and the profile of volatile fatty acids produced in the rumen affects the post-absorptive metabolism of the host animal. Rumen fermentation is shaped by intracellular and intercellular flows of metabolic hydrogen centered on the production, interspecies transfer, and incorporation of dihydrogen into competing pathways. Factors that affect the growth of methanogens and the rate of feed fermentation impact dihydrogen concentration in the rumen, which in turn controls the balance between pathways that produce and incorporate metabolic hydrogen, determining methane production and the profile of volatile fatty acids. A basic kinetic model of competition for dihydrogen is presented, and possibilities for intervention to redirect metabolic hydrogen from methanogenesis toward alternative useful electron sinks are discussed. The flows of metabolic hydrogen toward nutritionally beneficial sinks could be enhanced by adding to the rumen fermentation electron acceptors or direct fed microbials. It is proposed to screen hydrogenotrophs for dihydrogen thresholds and affinities, as well as identifying and studying microorganisms that produce and utilize intercellular electron carriers other than dihydrogen. These approaches can allow identifying potential microbial additives to compete with methanogens for metabolic hydrogen. The combination of adequate microbial additives or electron acceptors with inhibitors of methanogenesis can be effective approaches to decrease methane production and simultaneously redirect metabolic hydrogen toward end products of fermentation with a nutritional value for the host animal. The design of strategies to redirect metabolic hydrogen from methane to other sinks should be based on knowledge of the physicochemical control of rumen fermentation pathways. The application of new –omics techniques together with classical biochemistry methods and mechanistic modeling can lead to exciting developments in the understanding and manipulation of the flows of metabolic hydrogen in rumen fermentation.",
"conclusion": "Conclusion and Future Directions Early work in the past century established the foundations to understand fermentation and [H] dynamics in the rumen. Hungate (1967) demonstrated the central role of H 2 in CH 4 production. The principles and importance of interspecies H 2 transfer was illustrated in several ingenious experiments in which H 2 producers were co-cultured with methanogens ( Wolin et al., 1997 ). A model to explain how the diet influences the VFA profile and CH 4 formation through changes in methanogens rate of growth and H 2 concentration ( Janssen, 2010 ) has been an important advancement in this area. Electron confurcation has been incorporated into rumen fermentation models ( Van Lingen et al., 2016 , 2019 ), and recent experimental work with comparative genomics and metatranscriptomics revealed the importance of electron bifurcation and confurcation in H 2 dynamics in the rumen. Shifts in fermentation in defined cultures were studied at the level of gene expression ( Greening et al., 2019 ). In comparison, fewer studies ( Chen and Wolin, 1977 ; Latham and Wolin, 1977 ) have examined the competition for H 2 between methanogens and other hydrogenotrophs, such as succinate and propionate producers, at the basal H 2 concentrations resulting from fermentation-evolving H 2 . A recent experiment studied the competition for [H] between a methanogen and lactate producers Sharpea and Kandleria ( Kumar et al., 2018 ). Pure cultures of rumen hydrogenotrophs, such as those isolated in the Hungate 1000 Project ( Kelly, 2016 ), could be screened for kinetic parameters of H 2 incorporation and H 2 thresholds. This information could be used to predict the outcome of the competition for H 2 between methanogens and other hydrogenotrophs with models similar to the ones generated by Muñoz-Tamayo et al. (2019) and Lynch et al. (2019) for competition between methanogens. The ability of non-methanogenic hydrogenotrophs to compete for H 2 could be evaluated in co-culture with methanogens if they are also fermentative H 2 producers themselves (e.g., R. flavefaciens , S. ruminantium ), or in tri-cultures with methanogens and a H 2 -producing organism if they incorporate H 2 but do not produce it (e.g., F. succinogenes , Succinivibrio dextrinosolvens , Succinimonas amylolytica , reductive acetogens). Understanding the physicochemical control of rumen fermentation can help optimizing the design of strategies to direct [H] from CH 4 to other sinks. In this regard, H 2 concentration is highly influential on the thermodynamics and kinetics of fermentation pathways ( Janssen, 2010 ). Research is needed on dissolved H 2 concentration ( Table 1 ) and H 2 gradients under different conditions, especially when methanogenesis is inhibited. Dissolved H 2 concentration has been generally estimated by measuring the concentration of H 2 gas in the gas phase and assuming equilibrium with dissolved H 2 in the fluid ( Kohn and Boston, 2000 ; Ungerfeld and Kohn, 2006 ; Janssen, 2010 ), but H 2 has been shown to be supersaturated in the rumen ( Wang et al., 2016 ). It would be important to incorporate H 2 supersaturation factors in future models of rumen fermentation, but more results with different diets, time after feeding, and other factors such as methanogenesis inhibition, are needed so that H 2 supersaturation is not modeled as a constant. The application of genomics and transcriptomics has advanced our understanding of the relationships between the abundance and expression of genes encoding for hydrogenases and rumen [H] flows ( Greening et al., 2019 ). The combination of –omics techniques with classical biochemistry and microbiology methods may make possible the isolation and kinetic characterization of H 2 -incorporating hydrogenases. The application of proteomics to understand methanogenesis and flows of [H] through changes in hydrogenases and other enzymes involved in [H] transactions is also of much interest ( Snelling and Wallace, 2017 ). Recently, metabolomics has been applied toward the understanding of differences between dairy cows with high and low feed utilization efficiency associated to high and low CH 4 production ( Shabat et al., 2016 ) and toward understanding the responses to methanogenesis inhibitors ( Martinez-Fernandez et al., 2018 ). Finally, experimental advances must be interpreted in the light of basic physicochemical knowledge of thermodynamics and kinetics to develop mathematical and conceptual mechanistic models ( Janssen, 2010 ; Van Lingen et al., 2016 ) for designing new strategies of manipulation of [H] flows in the rumen and predicting their outcomes.",
"introduction": "Introduction The complex microbial community that inhabits the rumen allows ruminants to digest and transform fibrous carbohydrates unavailable to humans into useful products such as meat, milk, wool and traction. Critical to the symbiosis between the rumen microbiota and the host animal is the anaerobic condition of the rumen, which prevents the complete oxidation of carbohydrates to carbon dioxide (CO 2 ) and water. Instead, carbohydrates are incompletely oxidized to volatile fatty acids (VFA) and gases, with the host animal absorbing and utilizing the VFA as sources and precursors of energy, fat, glucose, and non-essential amino acids ( Armstrong and Blaxter, 1957 ). Rumen fermentation not only provides the ruminant with VFA. Part of the negative Gibbs energy change (Δ G ) associated with fermentation is used by rumen microbes to generate ATP that can be utilized for microbial growth, active transport of substrates, and motility. Microbial growth produces microbial protein, which is the principal ( Wallace et al., 1997 ) and most economical source of amino acids for ruminants. Rumen microorganisms can also synthesize water-soluble vitamins, which thus do not need to be included in most ruminants diets ( Weiss, 2017 ). A product of rumen fermentation is methane (CH 4 ), which is a potent greenhouse gas when released to the atmosphere, and also a loss of energy for ruminants ( Eckard et al., 2010 ; Martin et al., 2010 ). Through the formation of CH 4 and the profile of VFA produced, rumen fermentation has important consequences for animal productivity and the environment. Understanding how rumen fermentation is controlled can help designing strategies to manipulate it in desired directions. Central to rumen metabolism are the dynamics of metabolic hydrogen ([H]) production and utilization. The idea of understanding rumen energy metabolism as [H] flows through different biochemical pathways is not new ( Czerkawski, 1986 ; Hegarty and Gerdes, 1999 ). The objectives of this paper are to review and critically examine [H] flows as the unifying principle to understand rumen fermentation. In particular, this paper will discuss (i) The control of the VFA profile and CH 4 production by dihydrogen (H 2 ), (ii) The principles underlying the competition for H 2 , (iii) The potential inhibitory effects of H 2 and other intercellular electron (e – ) carriers on the rates of fermentation and digestion in the rumen, and (iv) The relationship between the flows of [H] and microbial growth. All of these aspects have implications to animal productivity and the environment mediated by ruminal and post-absorptive metabolism.",
"discussion": "Discussion The Control of the Rumen Fermentation Profile Diets that are rich in fiber produce a profile of VFA high in acetate and low in propionate, with a relatively high production of CH 4 per unit of digested organic matter. On the other hand, concentrates, which are richer in starch, are fermented to more propionate and less CH 4 ( Johnson and Johnson, 1995 ; Janssen, 2010 ; Ungerfeld, 2013 ). The diet effect on the acetate to propionate ratio cannot be explained simply by different chemical composition of cellulose and starch, as they are both hydrolyzed to glucose ( Janssen, 2010 ). Janssen (2010) proposed a mechanism based on methanogens growth rate and the resulting H 2 concentration, to explain how concentrates shift rumen fermentation from acetate to propionate and lower CH 4 production. The replacement of roughages with concentrates induces changes such as increased rumen outflow rates and lower rumen pH. High rumen outflow rates impose methanogens that are not washed out of the rumen faster growth rates. Based on the Monod relationship of microbial growth, H 2 concentration must increase when methanogens grow faster. In turn, greater H 2 concentration would thermodynamically inhibit H 2 production, and by doing so also inhibit acetate production, which is associated with H 2 production. Greater H 2 concentration would conversely favor [H] redirection toward alternative [H] sinks such as propionate. In agreement, the concentration of dissolved H 2 in the rumen has been reported to associate negatively with acetate and positively with propionate molar percentages ( Wang et al., 2017 , 2018 ), although an association with propionate molar percentage was not observed in another study ( Wang et al., 2016 ). Similarly, methanogens are sensitive to the low rumen pH induced by feeding concentrates. A decrease in rumen pH is expected to decrease methanogens maximum growth rates, and, according to the Monod relationship, H 2 concentration would increase if methanogens growth rate is maintained. An increase in H 2 concentration would again thermodynamically shift fermentation from acetate to propionate. A similar explanation was provided for the accumulation of H 2 and shift from acetate to propionate caused by chemical inhibitors of methanogenesis ( Janssen, 2010 ). Several experiments with defined cultures comparing the fermentation profile of pure cultures of H 2 producers with co-cultures of the same organisms growing with methanogens demonstrate the profound influence of H 2 removal by the methanogen on the fermentation profile of the H 2 -producing microorganism ( Wolin et al., 1997 ). These insightful experiments provide a simple proof of concept for the theory proposed by Janssen (2010) : in the absence of methanogens in the mono-culture (i.e., an analogous situation to methanogens completely washed out because a very high rumen outflow rate, or completely inhibited by low pH or an inhibitor of methanogenesis), H 2 accumulates, inhibiting H 2 formation and decreasing acetate production (i.e., a H 2 -releasing pathway). This in turn directs [H] toward reduced intermediates of rumen fermentation such as formate, ethanol, lactate and/or succinate ( Chung, 1976 ; Chen and Wolin, 1977 ; Bauchop and Mountfort, 1981 ; Marvin-Sikkema et al., 1990 ; Pavlostathis et al., 1990 ), and [H] sinks such as propionate ( Chen and Wolin, 1977 ) or butyrate ( Chung, 1976 ). When H 2 producers were co-cultured with methanogens they greatly diminished or stopped producing formate, ethanol, lactate and/or succinate, as well as propionate. The co-cultures accumulated less H 2 , and as CH 4 formation removed H 2 , lower H 2 concentration thermodynamically favored acetate production. Increasing outflow rates in some continuous culture experiments ( Isaacson et al., 1975 ; Stanier and Davies, 1981 ) agree with the predictions of the Janssen (2010) model. Results of the experiment by Wenner et al. (2017) do not fully agree, as, contrary to the model expectations decreasing pH decreased H 2 concentration. Immediately after feed ingestion, the most readily digestible feed components are rapidly digested and fermented, and H 2 concentration rises ( Robinson et al., 1981 ; Janssen, 2010 ; Guyader et al., 2015 ). After feed ingestion, increases in H 2 emissions have been shown to occur earlier and faster than increases in CH 4 ( Rooke et al., 2014 ; Van Lingen et al., 2017 ; Söllinger et al., 2018 ), although this pattern has not occurred in all studies ( Hillman et al., 1985 ). The peak in H 2 emission preceding the evolution of CH 4 production has been modeled by Van Lingen et al. (2019) , and interpreted by Rooke et al. (2014) as the consequence of rapid fermentation and H 2 production exceeding the capacity of methanogens to utilize all the H 2 produced. The lag period between the CH 4 and H 2 peaks observed in some studies suggests that, when fermentation is rapid, methanogens growth and/or the expression of genes encoding for methanogenesis enzymes lags behind rapid H 2 evolution. In that regard, Söllinger et al. (2018) reported that whereas archaeal 16S rRNA genes abundance peaked at 1 h after feeding, methanogenesis mRNA abundance did not peak until 3 h after feeding. The question becomes what impedes or retards methanogens to respond with more rapid gene expression to make use of the elevated H 2 concentrations occurring after a meal. It is possible that temporal increases in outflow rates occurring after feeding episodes ( Van Lingen et al., 2019 ), result in high H 2 concentration by increasing methanogens growth rate, as proposed by Janssen (2010) . However, increasing the concentration of glucose in a chemostat at a constant pH and outflow rate still decreased CH 4 production per mole of glucose fermented ( Isaacson et al., 1975 ), which suggests a limitation of methanogenesis independent of outflow rate or pH to use all of the H 2 made available by the rapid fermentation of glucose at high concentration (although H 2 was not measured in that experiment). In 48 h batch cultures, there were distinct effects of pH and substrate composition (hay or cracked corn) on H 2 , CH 4 and the acetate to propionate ratio ( Russell, 1998 ), which can also be interpreted as an indication of an effect of the rate of fermentation per se independent of outflow rates or pH. It has also been speculated that the evolution of rumen H 2 -incorporating hydrogenases in the rumen environment with low H 2 concentration may have resulted in low K m but also low V max for H 2 ( Ungerfeld, 2015a ). This idea, however, does not agree with the high frequency of genomes of rumen organisms encoding [FeFe]-hydrogenases, and the high abundance of transcripts of various types of [FeFe]-hydrogenases in sheep rumens ( Greening et al., 2019 ), as [FeFe]-hydrogenases have higher V max and K m for H 2 uptake than [NiFe] hydrogenases ( Frey, 2002 ). Effects of Electron Carriers Other Than Dihydrogen on the Rumen Fermentation Profile Eighteen percent of rumen CH 4 was estimated to be produced from formate as [H] donor ( Hungate et al., 1970 ), and formate can be the [H] donor in fumarate reduction to succinate ( Asanuma et al., 1999 ). It is thus possible that, the same as H 2 , formate concentration has an influence on CH 4 production and the VFA profile. Other than an accumulation of formate as a response to methanogenesis inhibitors in some studies ( Ungerfeld et al., 2003 ; Martinez-Fernandez et al., 2016 , 2017 ), the effects of variables such as outflow rates, pH, or rate of fermentation on formate concentration, have not been investigated to the author’s knowledge. Generating information about the relationship between those variables and formate concentration, and how they relate to methanogens growth rate would be important for evaluating the influence of formate on the VFA profile, and integrating formate to a model of [H] flows in rumen fermentation. Lactate is another intercellular e – carrier which, except for lactic acidosis, normally does not accumulate in the rumen and is extensively converted to VFA by various lactate utilizers ( Chen et al., 2019 ). Small amounts of lactate have been reported to accumulate as a consequence of inhibiting methanogenesis in some ( Amgarten et al., 1981 ), but not all ( Božic et al., 2009 ; Martinez-Fernandez et al., 2016 ), studies. In general, lactate accumulation in the rumen is the result of lactate production rate surpassing lactate utilization as a consequence of rapid fermentation. A possible enhancement in the role of lactate as an intermediate of butyrate production in low CH 4 -producing sheep ( Kamke et al., 2016 ) deserves further study (see section “The Competition for Dihydrogen”). Succinate concentration in the rumen is typically low as it is rapidly converted to propionate ( Blackburn and Hungate, 1963 ; Immig, 1996 ). It thus seems that succinate concentration exerts little influence on CH 4 production and the VFA profile, although the finding by Kamke et al. (2016) of greater abundance of genes involved in the conversion of succinate to butyrate in low CH 4 -producing sheep prompts for more investigation. The Competition for Dihydrogen The principle proposed by Janssen (2010) relating rumen H 2 concentration to methanogens growth rates could be in theory extended to other hydrogenotrophs, provided that their pathway of H 2 incorporation is thermodynamically feasible. The rate of H 2 uptake by a methanogen would follow a Michaelis-Menten kinetics-wise function: (1) v m e t = V m a x m e t [ H 2 ] ( K m m e t + [ H 2 ] ) where v met is the rate of H 2 uptake (e.g., mol L –1 min –1 ), V max met is the maximum rate of H 2 uptake with non-limiting H 2 concentration, [H 2 ] is the concentration of H 2 , and K m met is the apparent affinity for H 2 i.e., the concentration of H 2 at which the rate of H 2 uptake is half maximal. If a methanogen was growing in co-culture with another hydrogenotroph whose rate of H 2 uptake was limited by H 2 concentration, and H 2 concentration was above the H 2 thresholds ( Cord-Ruwisch et al., 1988 ) of both organisms, it can be deduced ( Supplementary Appendix S1 ) that the proportion of total H 2 uptake incorporated into methanogenesis would be equal to: (2) v m e t ( v m e t + v a l t ) = V m a x m e t ( K m a l t + [ H 2 ] ) [ V m a x m e t ( K m a l t + [ H 2 ] ) + V m a x a l t ( K m m e t + [ H 2 ] ) ] where v alt and V maxalt are the rate and the maximum rate of H 2 uptake of the alternative hydrogenotroph, respectively, and K m alt is the affinity for H 2 of the alternative hydrogenotroph, with the rest of the variables defined as in Eq. 1. This equation does not take into account possible thermodynamic constraints and differences in the efficiency of microbial growth. Figure 4 shows a simulation of the proportion of H 2 uptake incorporated into methanogenesis as a function of H 2 concentration according to Eq. 2 in co-cultures of mixed methanogens and various hydrogenotrophs that reduce fumarate to succinate. The K m values for methanogens and fumarate reducers used in the simulation were reported by Asanuma et al. (1999) . An equal V max was assumed for methanogenesis and fumarate reduction in this simulation. The range of H 2 concentration in Figure 4 is based on dissolved H 2 concentration measured directly in various in vivo studies ( Table 1 ). It can be seen in Figure 4 that as H 2 concentration increases, the proportion of H 2 taken up by methanogens decrease and approaches 1/2 ( Supplementary Appendix S2 ). It can be shown that if the V max of the alternative hydrogenotroph doubled the V max of the methanogens, the proportion of H 2 taken up by methanogens would trend to 1/3 as H 2 concentration increases ( Supplementary Appendix S3 ). TABLE 1 Dissolved dihydrogen concentration in the rumen. References Treatment or condition Method of measurement H 2 (μM) 1 Hungate (1967) Non-inhibited methanogenesis H 2 extraction procedure 0.19–30.4 Robinson et al. (1981) Non-inhibited methanogenesis H 2 extraction procedure 2–15 Hillman et al. (1985) Non-inhibited methanogenesis Mass spectrometry 0.6–5.8 Smolenski and Robinson (1988) Non-inhibited methanogenesis H 2 sensor 0.36–20.1 Guyader et al. (2015) Non-inhibited methanogenesis H 2 sensor 3.58 Nitrate 45.3 Linseed 4.03 Nitrate + linseed 21 Wang et al. (2016) Oat grass H 2 extraction procedure 6.49 Barley straw 2.34 Wang et al. (2017) Control H 2 extraction procedure 1.02 H 2 released with Mg 1.99 Wang et al. (2018) Control H 2 extraction procedure 2.37 Nitrate 4.79 Ma et al. (2019) Control H 2 extraction procedure 1.76 H 2 released with Mg 2.68 Melgar et al. (2019) Control Gas-stripping 7.3 Methanogenesis inhibited with 3-nitrooxypropanol 43.6 1 Greater ranges in H 2 concentration in some studies in which methanogenesis was not inhibited are due to multiple measurements throughout the day in animals fed once or twice a day, rather than to variation caused by treatments imposed. FIGURE 4 Simulation of the proportion of dihydrogen taken up by methanogens in co-culture with various fumarate reducers as a function of dissolved H 2 concentration. The simulation was conducted based on a kinetic Michaelis-Menten-wise competition for dihydrogen. Apparent K m for H 2 uptake were reported by Asanuma et al. (1999) . An equal V max for dihydrogen uptake is assumed. The range of dissolved H 2 concentration is based on Table 1 . The sky blue area corresponds approximately to baseline dissolved H 2 concentrations (i.e., in between meals). The salmon area corresponds approximately to H 2 concentration peaks occurring closely after feeding. The purple area corresponds approximately to the range of H 2 concentration that could be observed when methanogenesis is inhibited. For n hydrogenotrophs (including m methanogens), the proportion of total H 2 taken up by m methanogens, can be generalized to: ∑ j = 1 j = m v m e t j ∑ i = 1 i = n v i = ∏ i = 1 i = n ( K m i + [ H 2 ] ) ∑ j = 1 j = m v m a x m e t j ( K m m e t j + [ H 2 ] ) ∑ i = 1 i = n v m a x i ( K m i + [ H 2 ] ) (3, Supplementary Appendix S4 ) At the low baseline H 2 concentration prevailing in the rumen ( Hungate, 1967 ), a low K m for H 2 incorporation is key in the competition for H 2 among thermodynamically feasible H 2 -incorporating processes. Methanogens have a lower K m for H 2 than the fumarate reducers depicted in Figure 4 , and consequently they would incorporate most of the H 2 at the low H 2 concentrations occurring between episodes of feed ingestion. Other K m values reported for methanogens are shown in Table 2 , and are similar the K m of methanogens reported by Asanuma et al. (1999) . In agreement with the predictions of Figure 4 , previous co-culture experiments also show that the production of succinate or propionate by Ruminococcus flavefaciens growing on cellulose ( Latham and Wolin, 1977 ) or by Selenomonas ruminantium growing on glucose or lactate ( Chen and Wolin, 1977 ), decreased in the presence of methanogens compared to the pure cultures, although it continued being thermodynamically feasible, as it did not stop. In those co-culture experiments, H 2 concentration (although it was not reported) was kept low by the methanogen, likely situating at the low end of the range of H 2 concentration in Figure 4 . TABLE 2 Apparent K m for dihydrogen of methanogens and fumarate reducers. References Microorganism K m (μM) Hungate et al. (1970) Methanogenesis by a mixed rumen culture 1 Hungate et al. (1970) Methanobrevibacter ruminantium 1 Pavlostathis et al. (1990) Methanobrevibacter smithii 1 Asanuma et al. (1999) Mixed rumen methanogens 1.6 Asanuma et al. (1999) Fibrobacter succinogenes 6.2 Asanuma et al. (1999) Selenomonas ruminantium 7.5 Asanuma et al. (1999) Selenomonas lactylica 4.7 Asanuma et al. (1999) Veillonella parvula 5.8 Asanuma et al. (1999) Wolinella succinogenes 4.0 As H 2 concentration increases, as it occurs after feed ingestion, or when feeding concentrates, or if methanogenesis is inhibited, the K m starts becoming less important to determine the partition of H 2 incorporated into competing pathways, and a greater proportion of H 2 would be incorporated into fumarate reduction to succinate ( Figure 4 ). When H 2 concentration is relatively high, a high V max for H 2 can potentially become very important to determine the flow of H 2 incorporated by a certain microorganism in a particular pathway. If the V max is expressed as the flow of H 2 incorporated per gram of cell DM or cell protein, rather than the flow of H 2 incorporated per volume of culture (or rumen contents), the flow of H 2 into each pathway in the system will also depend on the cell density of each microbial species. The incorporation of [H] into pathways alternative to methanogenesis can be limited by enzyme or substrate kinetics, or thermodynamics ( Ungerfeld, 2015a ). The addition of an e – acceptor that can be metabolized to VFA can help removing substrate kinetics or thermodynamic constraints. In general, adding to rumen fermentation carboxylic acids that are propionate or butyrate precursors as e – acceptors has had small effects on CH 4 production in vitro ( Callaway and Martin, 1996 ; Carro and Ranilla, 2003 ; Ungerfeld et al., 2003 ; Newbold et al., 2005 ; Riede et al., 2013 ) or in vivo ( McGinn et al., 2004 ; Beauchemin and McGinn, 2006 ; Kolver and Aspin, 2006 ; Yang et al., 2012 ), although larger decreases in CH 4 were observed in some experiments ( Li et al., 2009 ; Wood et al., 2009 ). A likely interpretation is that much of the added propionate and butyrate precursor was not metabolized to the expected final product, and thus little [H] was directed away from CH 4 formation ( Carro and Ungerfeld, 2015 ). When methanogenesis was simultaneously inhibited with a chemical compound, increased availability of [H] not incorporated into CH 4 favored the conversion of the added carboxylic acids to the expected end products. In the presence of inhibitors of methanogenesis, the addition of propionate precursors malate ( Mohammed et al., 2004 ) or fumarate ( Tatsuoka et al., 2008 ; Ebrahimi et al., 2011 ) increased propionate production in vitro and decreased H 2 accumulation. In contrast, butyrate precursors did not decrease H 2 accumulation caused by three inhibitors of methanogenesis in batch cultures ( Ungerfeld et al., 2006 ). Martinez-Fernandez et al. (2017) successfully used phlorglucinol as an e – acceptor to decrease the accumulation of H 2 and formate in the rumen of steers whose methanogenesis was inhibited with chloroform. An increase in acetate concentration observed when phlorglucinol was supplemented agrees with previous studies which had shown that phlorglucinol was reduced to acetate by rumen microorganisms using H 2 or formate as e – donors ( Martinez-Fernandez et al., 2017 ). Microbial additives can help removing constraints to the incorporation of [H] into pathways alternative to methanogenesis whose rate is enzyme-limited. Jeyanathan et al. (2014) reviewed the use of direct-fed microbials to manipulate rumen biochemical pathways to decrease CH 4 emissions. They proposed two main avenues to decrease CH 4 formation in the rumen through the use of microbial additives: (i) Microbial additives that incorporate H 2 into pathways alternative to methanogenesis, and (ii) Microbial additives that do not produce H 2 in fermentation. Microbial additives that compete with methanogens for H 2 could be dosed into the rumen ( Jeyanathan et al., 2014 ). It may also be possible to stimulate native rumen non-methanogenic hydrogenotrophs. Some fumarate reducers ( Asanuma et al., 1999 ) and reductive acetogens ( Chaucheyras et al., 1995 ; Joblin, 1999 ) were able to decrease CH 4 production when grown in co-culture with methanogens, but because those experiments were conducted with elevated headspace H 2 , an ability of those organisms to compete for H 2 at low concentration cannot be demonstrated ( Figure 4 ). Methanogens would be expected to prevail over reductive acetogens in a co-culture at low H 2 concentration due to their lower H 2 thresholds ( Cord-Ruwisch et al., 1988 ). Boccazzi and Patterson (2011) isolated rumen reductive acetogens with lower H 2 thresholds than other reductive acetogens previously isolated from the rumen and with similar H 2 thresholds to reductive acetogens from other environments. Yet, they still had higher H 2 thresholds compared to methanogens ( Table 3 ). TABLE 3 Dihydrogen thresholds of methanogens and reductive acetogens from the rumen and other environments. Microorganism Environment H 2 threshold (ppm) References Methanogens Methanospirillum hungatei Sewage sludge 30 Cord-Ruwisch et al. (1988) Methanobrevibacter smithii Primary sewage digester 100 Cord-Ruwisch et al. (1988) Methanobrevibacter arboriphilus Digested sewage sludge 90 Cord-Ruwisch et al. (1988) Methanobacterium formicicum Anaerobic sewage sludge digester 28 Cord-Ruwisch et al. (1988) Methanococcus vannielii Marine mud 75 Cord-Ruwisch et al. (1988) Isolate 10-16B Rumen 126 Le Van et al. (1998) Isolate NI4A Rumen 90 - 92 Boccazzi and Patterson (2011) Reductive acetogens Sporomusa termitida Termite hindgut ∼800 Breznak et al. (1988) Sporomusa termitida Termite hindgut 830 Cord-Ruwisch et al. (1988) Acetobacterium woodii NZ Va 16 Not provided 520 Cord-Ruwisch et al. (1988) Acetobacterium carbinolicum Freshwater mud 950 Cord-Ruwisch et al. (1988) Acetitomaculum ruminis 190A4 Rumen 3830 Le Van et al. (1998) Two reductive acetogenic isolates Rumen ∼750 Joblin (1999) Isolate A2 Rumen 1383–2516 Boccazzi and Patterson (2011) Isolate A4 Rumen 8007 Boccazzi and Patterson (2011) Isolate A9 Rumen 1619–66157 Boccazzi and Patterson (2011) Isolate A10 Rumen 208–1284 Boccazzi and Patterson (2011) Isolate H3HH Rumen 1390 Boccazzi and Patterson (2011) Supplementation of rumen batch cultures with succinate and propionate producers caused mild to moderate decreases in CH 4 production ( Alazzeh et al., 2012 ; Mamuad et al., 2014 ). In another study, supplementing rumen batch cultures with fumarate-reducing enterococci caused large decreases in CH 4 and increases in propionate concentration ( Kim et al., 2016 ). A slight decrease in CH 4 production per kilogram of ingested feed occurred when supplementing Propionibacterium strains to heifers fed a high-forage ( Vyas et al., 2014a ), but not a mixed ( Vyas et al., 2016 ), or a high-concentrate ( Vyas et al., 2014b ), diet. Rumen succinate producers W. succinogenes and Mannheimia succiniciproducens could be interesting candidates to compete with methanogens at low H 2 concentrations, as they possess [NiFe]-hydrogenases for H 2 uptake ( Søndergaard et al., 2016 ). [NiFe]-hydrogenases have K m for H 2 about two orders of magnitude lower than [FeFe] hydrogenases ( Frey, 2002 ). However, the apparent K m of W. succinogenes for H 2 was still higher than that of methanogens ( Asanuma et al., 1999 ; Table 2 ). M. succiniciproducens has been genetically engineered to improve its yield of succinate from glucose ( Lee et al., 2006 ; Choi et al., 2016 ), which could help increasing its V max for H 2 uptake. Nitrate reduction is thermodynamically more favorable than methanogenesis but may result in accumulation of the toxic intermediate nitrite. The addition of nitrite reducers may help avoiding nitrite toxicity while decreasing CH 4 production ( Jeyanathan et al., 2014 ). Nitrate should replace other sources of nitrogen on an isonitrogenous basis to avoid increasing the elimination of nitrogen in urine and the formation of nitrous oxide in the rumen, which is another potent greenhouse gas ( Petersen et al., 2015 ). Sulfate reduction can also thermodynamically outcompete methanogenesis, although it generates the toxic reduced end product hydrogen sulfide ( Jeyanathan et al., 2014 ). Jeyanathan et al. (2014) also proposed that, by avoiding the formation of H 2 , the combined use of added lactate producers and the lactate utilizer Megasphaera elsdenii could channel [H] into propionate production instead of CH 4 . In that regard, lactate producers Sharpea and Kandleria were abundant in the rumens of one of two low CH 4 -producing sheep microbiomes ( Kittelmann et al., 2014 ). Low CH 4 -producing sheep also had higher rumen concentration of lactate, and the lactate dehydrogenases that differed the most between the low- and the high-producing CH 4 sheep associated phylogenetically with S. azabuensis and K. vitulina ( Kamke et al., 2016 ). Several strains of Sharpea and Kandleria that produced predominantly lactate and small amounts of formate, ethanol and acetate, did not change their fermentation products when growing with a methanogen ( Kumar et al., 2018 ). Lactate produced by Sharpea and Kandleria did not accumulate to high concentrations in the rumen because it seemed to be metabolized by Megasphaera spp. mostly to butyrate, and to propionate via the non-randomizing pathway. The conversion of lactate to butyrate would result in less H 2 production compared to acetate production from glucose ( Kamke et al., 2016 ). It seems then that a dual mechanism resulted in lower CH 4 production in the low-CH 4 producing sheep ( Kittelmann et al., 2014 ; Kamke et al., 2016 ): (i) Incorporation of [H] in the reduction of pyruvate to lactate by Sharpea and Kandleria instead of H 2 release, and (ii) Uptake and conversion of lactate to butyrate and propionate by Megasphaera . In this regard, co-culture experiments comparing the kinetics of uptake of lactate and conversion to acetate, propionate and butyrate by Megasphaera and other microorganisms would be of interest. M. elsdenii had a lower affinity for lactate than for glucose ( Russell and Baldwin, 1979 ), but its rate of lactate uptake was not affected by glucose ( Russell and Baldwin, 1978 ). Interestingly, fermentation extracts of the probiotic Aspergillus oryzae stimulated lactate uptake by M. elsdenii and did not affect its fermentation profile ( Waldrip and Martin, 1993 ). This approach toward decreasing CH 4 formation could thus contemplate the addition of a “microbial team” composed by a lactate producer and a lactate utilizer that metabolizes lactate to propionate or butyrate. Another interesting microbial species could be Fibrobacter succinogenes , a fiber degrader which does not produce H 2 and would thus not contribute [H] to methanogenesis and instead incorporate [H] into succinate production ( Morgavi et al., 2010 ). However, the higher K m for H 2 of F. succinogenes compared to methanogens ( Asanuma et al., 1999 ) would imply a lower uptake of H 2 compared to methanogens at low H 2 concentration ( Figure 4 ). Another strategy would be to inhibit methanogenesis with a chemical compound and simultaneously dose hydrogenotrophs that incorporate H 2 into a desirable pathway, or pathways. For example, reductive acetogenesis was enhanced in batch cultures through simultaneous inhibition of methanogenesis and addition of reductive acetogens ( Nollet et al., 1997 ; Le Van et al., 1998 ; Lopez et al., 1999 ). The kinetic parameters for H 2 and the H 2 threshold of the hydrogenotroph of choice would be very important. Compared to the typical rumen fermentation with CH 4 as the main [H] sink ( Figure 5A ), inhibiting methanogenesis results in an increase in the incorporation of [H] into alternative sinks, but also in H 2 as a [H] sink ( Figure 5B ). Adding a hydrogenotroph with a high V max for H 2 can allow a high flow of H 2 incorporation into a desirable fermentation product. However, if the K m of the added hydrogenotroph for H 2 were high, H 2 would still accumulate, and the magnitude of gaseous H 2 losses could be important. A low K m for H 2 would also result in H 2 accumulation if the V max for H 2 was low and the rate of fermentation was high, unless the microorganism was dosed in high numbers. Another possibility would be to combine a hydrogenotroph with high K m and V max for H 2 with another hydrogenotroph with low K m and V max . A theoretically ideal situation in which dissolved H 2 concentration and H 2 emission are at the level of the rumen with functional methanogenesis is depicted in Figure 5C . FIGURE 5 Three hypothetical scenarios of manipulation of metabolic hydrogen ([H]) flows in rumen fermentation: (A) Non-intervened rumen fermentation with functional methanogenesis. Methane (CH 4 ) is the main sink of metabolic hydrogen; (B) Methanogenesis is inhibited with a chemical additive. Part of metabolic hydrogen spared from methane formation is redirected toward alternative sinks that are final fermentation products in the rumen with functional methanogenesis. Redirection of metabolic hydrogen toward alternative sinks is incomplete and the concentration of dissolved dihydrogen increases. The ratio of reduced to oxidized cofactors increases and fermentation, understood as the flow of carbon and the rate of metabolic hydrogen production, is inhibited; (C) A theoretical successful situation in which methanogenesis is inhibited with a chemical additive and an added live hydrogenotrophs redirects a greater proportion of metabolic hydrogen toward alternative sinks potentially beneficial to the host animal. Accumulation of dihydrogen is relieved, cofactors can be re-oxidized as in the rumen with functional methanogenesis, and fermentation is not inhibited. A strategy employing chemical inhibitors of methanogenesis to redirect [H] from CH 4 toward nutritionally useful alternative [H] sinks should evaluate possible direct effects of the chemical inhibitors on non-methanogenic rumen microorganisms, so as to avoid affecting processes such as fiber digestion or propionate production. This aspect cannot be studied in mixed cultures or in vivo , because in these systems, changes in non-methanogenic populations can indirectly result from changes in methanogens and CH 4 production. The potential toxicity of chemical inhibitors of methanogenesis to non-methanogens should instead be studied in pure cultures. Mevastatin and lovastatin inhibited the growth of methanogens but not of major fermentative rumen bacteria, including major fiber degraders and propionate and butyrate producers ( Miller and Wolin, 2001 ). Chloroform at 2 mM inhibited the growth of reductive acetogens and six other rumen bacteria, including fiber degraders and propionate and butyrate producers ( Raju, 2016 ). In contrast, none of the non-methanogenic microorganisms examined were affected by acetylene at 1 mM (aqueous concentration) or 2-bromoethanesulfonate at 10 mM. The effect of n -butylisocyanide and 5,5′-dithio-bis-(2-nitrobenzoic acid) on reductive acetogens was concentration- and species-dependent, whilst 1,10-phenanthroline inhibited reductive acetogens at all of the concentrations studied ( Raju, 2016 ). 3-Nitrooxypropanol at 0.1 mM did not inhibit the growth of 11 functionally diverse rumen bacteria or Escherichia coli , whilst much lower concentrations inhibited rumen and non-rumen methanogens ( Duin et al., 2016 ). From an applied point of view, the addition of a hydrogenotroph to the rumen would ideally target post-feeding peaks of dissolved H 2 with the added microorganism at its exponential phase of growth. This might be difficult if the microbial additive was administered with the feed, as the added microorganism may be at its lag phase of growth at the peak of feed fermentation and H 2 release in the rumen. If adding hydrogenotrophs at their exponential phase of growth in vitro was successful at utilizing H 2 , further developments toward practical application would need to optimize in vivo the timing, means of administration, and doses of added hydrogenotrophs. Recently, Muñoz-Tamayo et al. (2019) conducted a growth and calorimetry experiment with three rumen methanogens. They estimated kinetic, thermodynamic, and growth parameters and predicted that in the long term only one methanogen would survive in tri-cultures. Similar conclusions were reached in another recent theoretical analysis ( Lynch et al., 2019 ). Muñoz-Tamayo et al. (2019) discussed that, as the rumen harbors a diverse community of methanogens, ecological factors such as sensitivity to pH, location in association to fluid or particles, and endosymbiosis with protozoa can contribute to explain the existence of diversity despite of thermodynamic and kinetic advantages of some methanogens over others. Some ecological aspects can cause temporal and spatial variations in H 2 concentration ( Smolenski and Robinson, 1988 ; Janssen, 2010 ), which can affect the partition of H 2 flows among different hydrogenotrophs. In the conceptual model proposed by Leng (2014) , the organization of particle-colonizing microbiota in biofilms results in close proximities between cells releasing and taking up H 2 , resulting in ample variations in H 2 concentration within the rumen. Greening et al. (2019) reported no differences in the expression of the most abundant H 2 -evolving hydrogenases in sheep selected by low and high CH 4 production. In contrast, there were differences between low- and high-CH 4 producing sheep in the expression of H 2 -incorporating hydrogenases. The expression of methanogens hydrogenases and methyl-CoM reductase were lower, and the expression of fumarate reductase and acetyl-CoA synthase (which incorporate H 2 into propionate production and reductive acetogenesis, respectively) were higher, in the low CH 4 -producing sheep. This can be interpreted as those alternative pathways of [H] incorporation decreasing CH 4 formation in low CH 4 -producing sheep by competing with [H] with methanogenesis. Alternatively, it can also be interpreted as those pathways of [H] incorporation alternative to methanogenesis becoming upregulated in low CH 4 -producing animals as a response to less CH 4 production, due perhaps to animal factors such as greater rumen outflow rate or lower rumen pH ( Janssen, 2010 ). Söllinger et al. (2018) reported that, among all bacterial functional genes, the greatest increase in mRNA abundance occurring 1 h after feeding corresponded to the fumarate reductase subunit C transcript, denoting a stimulation of propionate randomizing pathway associated to peaks of H 2 emission after feeding. However, despite of the increase in the abundance of fumarate reductase transcripts, H 2 emissions still increased and propionate concentration did not consistently increase 1 h after feeding. It is possible that the K m of H 2 incorporation into propionate production was relatively high, at least under the conditions of that experiment, which would agree with the higher K m for H 2 of fumarate reducers compared to methanogens reported by Asanuma et al. (1999) in pure cultures. It should be considered that competition for intercellular e – carriers other than H 2 (and lactate) also occurs. For example, the K m for formate was lower for fumarate reducers compared to methanogens ( Asanuma et al., 1999 ). In the rumen, methanol and methylamines resulting from the metabolism of pectin ( Pol and Demeyer, 1988 ) and betaine and choline ( Neill et al., 1978 ; Mitchell et al., 1979 ) can be used by methylotrophic methanogens as substrates for CH 4 production. Importantly, reductive acetogens have also been reported to use methanol and methylamines as [H] donors ( Ragsdale and Pierce, 2008 ; Jeyanathan et al., 2014 ), including the rumen acetogen Eubacterium limosum , a methanol-utilizer ( Genthner et al., 1981 ). It is thought that in the typical rumen fermentation, methanogens drop H 2 pressure below the threshold for reductive acetogenesis ( Ungerfeld and Kohn, 2006 ). However, it is important to examine both in defined and in mixed cultures the competition between methanogens and reductive acetogens for methanol and methylamines. Effects of Dihydrogen Accumulation on the Rates of Fermentation and Digestion The formation of CH 4 in the rumen represents an important loss of energy for the animal. Theoretically, inhibiting rumen methanogenesis could divert [H] toward fermentation products with a nutritional value for the animal, and improve animal productivity ( Czerkawski and Breckenridge, 1975b ; Schulman and Valentino, 1976 ), although this has not been consistently realized ( Ungerfeld, 2018 ). Compared to the rumen with functional methanogenesis ( Figure 5A ), inhibiting rumen methanogenesis results in accumulation of H 2 \n in vitro ( Ungerfeld, 2015b ) and increased H 2 emissions in vivo ( Ungerfeld, 2018 ), increased ratio of NADH to NAD + ( Hino and Russell, 1985 ; Figure 5B ), and decreased reducing potential ( Sauer and Teather, 1987 ). These changes indicate hindering in e – disposal and it is important to understand the consequences that this can have on feed fermentation and digestion. Conceptually, there is little doubt that an imbalance between the rates of reduction and re-oxidation of cofactors can halt fermentation ( Wolin et al., 1997 ), because the turnover rates of cofactors are very high compared to their intracellular concentrations ( de Graef et al., 1999 ). This principle has been experimentally verified as an increase in cellulose degradation when fibrolytic fungi were co-cultured with methanogens or with S. ruminantium as hydrogenotrophs ( Marvin-Sikkema et al., 1990 ). The questions are, at which point increases in the ratios of reduced to oxidized cofactors begin to impair fermentation, and how these ratios are in turn affected by H 2 pressure ( Figure 5B ). Whether accumulated H 2 can be re-channeled into other pathways ( Figure 5C ) has been discussed in the preceding section. High H 2 pressure can thermodynamically inhibit NADH oxidoreductases ( Gottschalk, 1986 ). Van Lingen et al. (2016) modeled the effect of H 2 pressure on the thermodynamic feasibility of NADH oxidation with and without electron confurcation with reduced Fd red 2 – . With an NAD + to NADH ratio of 3, similar to the NAD + to NADH ratio reported by Hino and Russell (1985) for their control treatments, and in the absence of electron confurcation, NADH oxidation was somewhat under thermodynamic control at H 2 partial pressures of between 2 × 10 –4 and 2 × 10 –3 bar, depending on the intracellular pH ( Van Lingen et al., 2016 ). If rumen headspace H 2 was to be at equilibrium with dissolved H 2 , the corresponding range of dissolved H 2 concentrations would be as low as 0.15 to 1.5 μM approximately (calculations not shown), but given the occurrence of H 2 supersaturation ( Wang et al., 2016 ) it would likely be higher. The same calculation conducted with NADH oxidation occurring through electron confurcation would yield a considerable higher range of dissolved H 2 concentration between 6 and 100 μM, again assuming equilibrium between gaseous and aqueous H 2 . Therefore, with electron confurcation, the range of dissolved H 2 concentration at which NADH oxidation becomes thermodynamically controlled coincides or is even higher than previously reported peaks of dissolved H 2 concentration after feed ingestion, or the dissolved H 2 concentration reported by Melgar et al. (2019) for methanogenesis inhibition ( Table 1 ). This agrees with the findings by Greening et al. (2019) regarding the importance of confurcating hydrogenases in H 2 formation in the rumen. The thermodynamic feasibility of NADH oxidation also depends on the intracellular pH ( Van Lingen et al., 2016 ), which in turn depends on the extracellular pH and the bacterial species ( Russell, 1991 ). Inhibiting methanogenesis in vitro results in H 2 accumulation and consistently inhibits hexoses fermentation as estimated through the stoichiometry of VFA production ( Ungerfeld, 2015b ). However, the estimation of fermented hexoses from the stoichiometry of VFA production does not consider carbon in fermented hexoses utilized in microbial biomass accretion. In an in vitro study with several inhibitors of methanogenesis, no consistent effects on true organic matter digestibility were found, with some decreases but also lack of effects with other additives ( Ungerfeld et al., 2019 ). Effects of inhibiting methanogenesis in vivo on digestion and fermentation in the rumen are complex to assess with most animal measurements, as apparent digestibility determinations do not consider microbial biomass and overall tract digestibility could be modified by post-ruminal compensations ( Ungerfeld, 2018 ), and rumen VFA concentrations are affected by, apart from VFA production rates, rates of VFA absorption, passage, incorporation into microbial biomass, and by changes in rumen volume ( Dijkstra et al., 1993 ; Kristensen, 2001 ; Storm et al., 2012 ; Hall et al., 2015 ). It is of course possible that negative effects of inhibiting methanogenesis on fermentation ( Ungerfeld, 2015b ) are not caused by H 2 accumulation per se , and instead some of the chemical inhibitors studied could be toxic to microorganisms other than methanogens. An experimental approach to study the effects of H 2 accumulation on the rate of fermentation without the addition of chemical inhibitors of methanogenesis is the addition of H 2 gas to the headspace of rumen incubations. In general, adding external H 2 to rumen cultures has not consistently resulted in an inhibition of fermentation measured as total VFA concentration or apparent digestibility ( Schulman and Valentino, 1976 ; Patra and Yu, 2013 ; Broudiscou et al., 2014 ; Qiao et al., 2015 ). A factor potentially masking the effects of H 2 gas added to microbial cultures headspace is lack of equilibrium with dissolved H 2 ( Wang et al., 2016 ). In that regard, two in vivo studies in which dissolved H 2 was indirectly delivered through the reaction of elemental magnesium (Mg) with water, reported decreases in total VFA concentration as a result of the augmented dissolved H 2 concentration ( Wang et al., 2017 ; Ma et al., 2019 ), although the limitations of VFA concentration as a metric of rumen fermentation pointed out above are again acknowledged. Ultimately, if means to efficiently redirect [H] to useful sinks ( Figure 5C ) could be designed, the extent to which the accumulation of H 2 can hinder NADH re-oxidation and fermentation would be unimportant from a practical standpoint, as H 2 would not accumulate when inhibiting methanogenesis ( Figure 5C ). A perhaps more realistic scenario intermediate between Figures 5B,C in which dissolved H 2 concentration was only partially relieved, but the rate of digestion and fermentation was not affected, can also be conceived. Pathways of [H] flow alternative to H 2 formation can result in the production of other intercellular e – carriers, such as lactate, ethanol, and formate, or final fermentation products such as propionate, all of which also help NADH oxidation ( Van Lingen et al., 2016 ). Formate, succinate or ethanol have been shown to accumulate along with H 2 when methanogenesis was inhibited in vitro ( Slyter, 1979 ; Asanuma et al., 1998 ; Ungerfeld et al., 2003 ) and in vivo ( Martinez-Fernandez et al., 2016 , 2017 ; Melgar et al., 2019 ), so it is important to understand if the accumulation of those metabolites could potentially inhibit cofactors re-oxidation and fermentation. The effects of lactate, ethanol, formate and propionate on fermentation and digestion are best studied in experiments in which those metabolites are externally added to rumen fermentation as pure compounds. Immig (1996) found that adding formate or succinate to rumen batch cultures did not affect total VFA production. Asanuma et al. (1998) found that added formate was stoichiometrically recovered as CH 4 and total VFA production was unaffected. Infusion of formic acid into the rumen of sheep did not affect overall tract apparent digestibility ( Vercoe and Blaxter, 1965 ). It is possible that as formate has a high rate of diffusion and is rapidly converted to CH 4 ( Leng, 2014 ), its accumulation may not affect cofactors re-oxidation and fermentation rate. Lactate is metabolized to VFA in the rumen ( Chen et al., 2019 ). Even though excess lactate accumulation can inhibit fermentation, this effect would likely be caused by low pH rather than by an impairment of [H] transfer and co-factor re-oxidation. A potential effect of lactate accumulation on fermentation independent of pH would have to be evaluated through, for example, a comparison of the addition of sodium lactate against sodium chloride. The effects of adding ethanol to rumen cultures was dose-depending, and a decrease in cellulose digestibility occurred with the highest dose ( Chalupa et al., 1964 ), whereas no effects on total VFA concentration was found in another in vitro experiment ( Pradhan and Hemken, 1970 ). Emery et al. (1959) found a tendency to decrease OM and N digestibility when feeding ethanol to sheep. However, negative effects of high doses of ethanol on rumen fermentation can be mediated through direct toxicity related to bacterial membranes leakages caused by ethanol ( Ingram, 1990 ), rather than through impairing re-oxidation of cofactors. As an end product of fermentation, propionate is removed by absorption, passage and incorporation into microbial biomass. The removal of propionate would be thought to occur rapidly enough so as to avoid accumulation causing inhibition of fermentation. In agreement, in several experiments intrarruminal infusion of propionate did not affect overall tract digestibility of various feed fractions ( Rook et al., 1963 ; Sheperd and Combs, 1998 ; Noziere et al., 2000 ; Oba and Allen, 2003 ). Effects of Metabolic Hydrogen Flows on Microbial Growth Flows of [H] in the rumen can affect microbial growth through at least three mechanisms: (i) Variation in the generation of ATP; (ii) Provision of precursors for biosynthesis; (iii) Provision of reducing power. The mechanisms through which this might occur will be developed in this section. Hydrolysis of ATP is necessary to drive otherwise thermodynamically unfeasible anabolic processes, such as protein synthesis. The rate of ATP generation in fermentation depends on the rate of fermentation, the fermentation profile, and the ATP generated in each fermentation pathway. Acetate production generates ATP through substrate level phosphorylation, the same as propionate non-randomizing pathway. In propionate randomizing pathway, butyrate production, and methanogenesis, ATP is also generated through ETLP ( Russell and Wallace, 1997 ; Hackmann and Firkins, 2015 ). Production of less reduced fermentation products such as lactate and ethanol generates ATP only in glycolysis and results in less microbial growth per unit of substrate degraded compared to acetate production and methanogenesis ( Wolin et al., 1997 ). One should bear in mind that increasing ATP generation does not necessarily mean maximizing the “efficiency” of fermentation. As the proportion of the Δ G of a pathway coupled to ATP generation increases, the net Δ G approaches zero, and the pathway slows down approaching equilibrium. For example, methanogens possessing cytochromes can generate more ATP per mole of CH 4 produced and have higher growth yields when growing on elevated H 2 concentration compared to methanogens without cytochromes, but on the other hand they have greater H 2 thresholds. Methylotrophic methanogens of the order Methanosarcinales have cytochromes and they have evolved to live in environments with low H 2 concentration by acquiring the capacity of using one carbon compounds as substrates for methanogenesis ( Thauer et al., 2008 ; Vanwonterghem et al., 2016 ; Lynch et al., 2019 ). Organic matter catabolized in the rumen is partitioned into fermentation products i.e., VFA and gases, and microbial biomass. The proportion of carbon in fermented carbohydrates diverted toward microbial cell production increases as the microbial biomass produced per mole of ATP hydrolyzed ( Y ATP ) increases ( Leng and Nolan, 1984 ) and as more moles of ATP are generated per mole of hexoses fermented. Also, each fermentation pathway can contribute different intermediate compounds to microbial anabolism. Microbial biomass is more reduced than substrate fermented, and consequently, alterations in the flows of [H] could affect [H] available for microbial biomass accretion. For example, inhibiting methanogenesis could result in increased [H] disposal into microbial biomass formation ( Czerkawski, 1986 ). Anabolic processes such as amino acids and fatty acids synthesis demand [H] and may be stimulated as a consequence of the inhibition of CH 4 production ( Chalupa, 1977 ; Ungerfeld, 2015b ). Deamination of reduced amino acids was inhibited by reducing power in the form of NADH, and conversely, was stimulated by methylene blue, an oxidizing agent ( Hino and Russell, 1985 ). Later results, however, could not confirm an increase in the incorporation of [H] into microbial amino acids when methanogenesis was inhibited in vitro ( Ungerfeld et al., 2019 )."
} | 14,787 |
30644148 | null | s2 | 7,079 | {
"abstract": "Bacterial type IV pili (T4P) are polymeric protein nanofibers that have diverse biological roles. Their unique physicochemical properties mark them as a candidate biomaterial for various applications, yet difficulties in producing native T4P hinder their utilization. Recent effort to mimic the T4P of the metal-reducing Geobacter sulfurreducens bacterium led to the design of synthetic peptide building blocks, which self-assemble into T4P-like nanofibers. Here, it is reported that the T4P-like peptide nanofibers efficiently bind metal oxide particles and reduce Au ions analogously to their native counterparts, and thus give rise to versatile and multifunctional peptide-metal nanocomposites. Focusing on the interaction with Au ions, a combination of experimental and computational methods provides mechanistic insight into the formation of an exceptionally dense Au nanoparticle (AuNP) decoration of the nanofibers. Characterization of the thus-formed peptide-AuNPs nanocomposite reveals enhanced thermal stability, electrical conductivity from the single-fiber level up, and substrate-selective adhesion. Exploring its potential applications, it is demonstrated that the peptide-AuNPs nanocomposite can act as a reusable catalytic coating or form self-supporting immersible films of desired shapes. The films scaffold the assembly of cardiac cells into synchronized patches, and present static charge detection capabilities at the macroscale. The study presents a novel T4P-inspired biometallic material."
} | 378 |
30644148 | null | s2 | 7,080 | {
"abstract": "Bacterial type IV pili (T4P) are polymeric protein nanofibers that have diverse biological roles. Their unique physicochemical properties mark them as a candidate biomaterial for various applications, yet difficulties in producing native T4P hinder their utilization. Recent effort to mimic the T4P of the metal-reducing Geobacter sulfurreducens bacterium led to the design of synthetic peptide building blocks, which self-assemble into T4P-like nanofibers. Here, it is reported that the T4P-like peptide nanofibers efficiently bind metal oxide particles and reduce Au ions analogously to their native counterparts, and thus give rise to versatile and multifunctional peptide-metal nanocomposites. Focusing on the interaction with Au ions, a combination of experimental and computational methods provides mechanistic insight into the formation of an exceptionally dense Au nanoparticle (AuNP) decoration of the nanofibers. Characterization of the thus-formed peptide-AuNPs nanocomposite reveals enhanced thermal stability, electrical conductivity from the single-fiber level up, and substrate-selective adhesion. Exploring its potential applications, it is demonstrated that the peptide-AuNPs nanocomposite can act as a reusable catalytic coating or form self-supporting immersible films of desired shapes. The films scaffold the assembly of cardiac cells into synchronized patches, and present static charge detection capabilities at the macroscale. The study presents a novel T4P-inspired biometallic material."
} | 378 |
35068628 | null | s2 | 7,081 | {
"abstract": "Grazing is known to affect soil microbial communities, nutrient cycling, and forage quantity and quality over time. However, a paucity of information exists for the immediate changes in the soil physicochemical and microbial environment in response to different grazing strategies. Soil microbes drive nutrient cycling and are involved in plant-soil-microbe relationships, making them potentially vulnerable to plant-driven changes in the soil environment caused by grazing. To test the hypothesis that variable grazing intensities modulate immediate effects on the soil microbial community, we conducted a grazing trial of three management approaches; high-intensity, short-duration grazing (HDG), low-intensity, medium-duration grazing (LDG), and no grazing (NG). Soil and vegetation samples were collected before grazing and 24 hours, 1 week, and 4 weeks after HDG grazing ended. Soil labile carbon (C) and nitrogen (N) pools, vegetation biomass, and soil microbial diversity and functional traits were determined, including extracellular enzymatic assays and high-throughput sequencing of the bacterial 16S rRNA and fungal ITS2 regions. We found that labile soil C and inorganic N increased following the LDG grazing while C-cycling extracellular enzymatic activities increased in response to HDG grazing but both total extracellular enzymatic activity profiles and soil abiotic profiles were mostly affected by temporal fluxes. The soil fungal community composition was strongly affected by the interaction of sampling time and grazing treatment, while the soil bacterial community composition was largely affected by sampling time with a lesser impact from grazing treatment. We identified several key fungal taxa that may influence immediate responses to grazing and modulate plant-soil-microbe interactions. There was strong evidence of temporal influences on soil biogeochemical variables and the soil microbiome, even within our narrow sampling scheme. Our results indicate that the soil ecosystem is dynamic and responsive to different grazing strategies within very short time scales, showing the need for further research to understand plant-soil-microbe interactions and how these feedback mechanisms can inform sustainable land management."
} | 563 |
36842556 | null | s2 | 7,082 | {
"abstract": "Membrane bioreactors (MBRs) suffer from high operational and cleaning costs due to biofouling. The biofouling begins when the adhesins (an anchor-type epitope made up of polar and charged amino acids) on microbial appendages bind to the surface. Two different compounds-dodecyl-β-D-maltoside (DDM) and methyl α-d-mannopyranoside (MeαMan)-were investigated as possible biofilm mitigation tools due to their documented anti-adhesin properties in the biomedical field. DDM prevented up to 56.3, 87.0, and 67.6% of the formation of Pseudomonas putida, Escherichia coli and wastewater culture biofilms, respectively, in microplate experiments. MeαMan increased biofilm in the microplates. In a biofilm reactor setting, DDM was then applied on typical membrane materials, polyvinylidene fluoride, polyamide, polyether-sulfone, and polyacrylonitrile and prevented 79.4, 62.5, 81.3, and 68.2% of the detectable wastewater culture biofilm formation, respectively. The mechanism of anti-adhesion was the binding of the polar head of the DDM to the polar amino acids of the microbial appendages in conjunction with the orientation of the DDM as it binds different membrane materials. If the anti-adhesins are effective at increasing the distance of the bacteria from the membrane materials, they will serve as a new method for delaying biofouling."
} | 334 |
34113832 | PMC8170006 | pmc | 7,084 | {
"abstract": "Summary The use of biomacromolecules is a nascent development in clean alternative energies. In applications of biosensors and biophotovoltaic devices, the bacterial photosynthetic reaction center (RC) is a protein-pigment complex that has been commonly interfaced with electrodes, in large part to take advantage of the long-lived and high efficiency of charge separation. We investigated assemblies of RCs on an electrode that range from monolayer to multilayers by measuring the photocurrent produced when illuminated by an intensity-modulated excitation light source. In addition, atomic force microscopy and modeling of the photocurrent with the Marcus-Hush-Chidsey theory detailed the reorganization energy for the electron transfer process, which also revealed changes in the RC local environment due to the adsorbed conformations. The local environment in which the RCs are embedded significantly influenced photocurrent generation, which has implications for electron transfer of other biomacromolecules deposited on a surface in sensor and photovoltaic applications employing a redox electrolyte.",
"conclusion": "Conclusions The conditions for preparing a gold surface covered with a monolayer of RC covalently tethered via thiol bonds was determined with a modulated illumination intensity method that measured the photocurrent-potential response in addition to ex situ AFM. The photocurrent response under an applied potential for an electrode surface with a partial to full monolayer of adsorbed RCs was fitted to the Marcus-Hush-Chidsey theoretical formalism and a consistent reorganization energy of 0.24 ± 0.03 eV was determined with an equilibrium potential ( E e q p h ) of +257 ± 19 mV/SHE. The photocurrent characteristics were sensitive enough to reveal the initiation of multilayer formation on the gold electrode. Multilayer coated electrode surfaces showed a significant negative shift in the photocurrent E onset and a photocurrent-potential response that was best fit using two components, one with the same reorganization energy as the monolayer samples, and the other component having a close to zero reorganization energy (<0.1 eV). AFM imaging confirmed the presence of isolated adsorbed RCs with 7 nm heights for the monolayer covered surfaces. AFM also showed multilayer formation as aggregated adsorbed species and tall structures. The negative shift in the E onset for multilayers of RCs was hypothesized to be a result of the difference in the local redox conditions present within the adsorbed multilayer. The difference between the local redox potential and the OCP, defined by the redox couple in solution, may create conditions for large DC photocurrents.",
"introduction": "Introduction There has been considerable interest recently in the use of biomacromolecules for cheap, clean, and efficient sources of sustainable energy production ( Ravi and Tan, 2015 ; Jones, 2009 ; Barber, 2008 ). One example is the modification of electrodes with biological photosynthetic protein complexes in biosensors and biophotovoltaic devices for the generation of electrical current. Photosynthetic proteins, such as reaction centers (RCs), are found in plants, algae, and photosynthetic bacteria. In particular, the RC from the bacterium Rhodobacter sphaeroides ( Kontur et al., 2012 ; Mackenzie et al., 2001 ) has been a paradigm for the development of biohybrid solar cells due to its near perfect quantum efficiency in photon-to-electron conversion and the subsequent long-lasting quasi-stable, charge-separated state ( Blankenship, 2018 ; Friebe and Frese, 2017 ; Ravi and Tan, 2015 ). To date, devices incorporating RCs used a variety of methods with the aim of increasing the photocurrent, such as employing plasmonic substrates ( Friebe et al., 2016a ) or embedding RCs within a redox hydrogel matrix to increase surface concentration ( Białek et al., 2020a ). Alignment of protein complexes is possible, such as using electrostatic interactions ( Mirvakili et al., 2014 ), gold-thiol covalent bonds ( Jun et al., 2018 , 2019 ), molecular ( Trammell et al, 2004 , 2006 ; Mahmoudzadeh et al., 2011 ) or protein ( Yaghoubi et al., 2014 ; Friebe et al, 2016b , 2017 ; Lebedev et al., 2006 ) linkages, and Langmuir-Blodgett methods ( Kamran et al., 2015 ) to deposit proteins to produce photocurrents that depended on the orientation of the complexes. However, these approaches resulted in surfaces that were likely composed of multilayers despite preparation strategies to remove non-specifically adsorbed proteins. Often, many use surfaces that are modified to prevent rather than to remove non-specifically adsorbed proteins ( Campuzano et al., 2019 ; Siegers et al., 2004 ). Furthermore, there are challenges in differentiating the small photocurrents from faradaic currents when an overpotential is applied ( Hollander et al., 2011 ). Photocurrents are typically measured at the open circuit potential (OCP), where the net faradaic current is zero for the redox system present in the electrolyte. The small photocurrents are often measured from multilayer assemblies to improve the signal to noise. We have previously established a methodology that is capable of isolating photocurrents from faradaic currents with high sensitivity enabling small photocurrents to be accurately measured even when buried in 10 4 larger faradaic currents at potentials other than the OCP ( Jun et al., 2019 ). This method allows for fundamental studies of electron transfer energetics which require a well-defined, controllable surface arrangement of RCs on the electrode surface; ideally, a single monolayer of RCs adsorbed in a homogeneous manner ( Gerster et al., 2012 ). Assembling a uniform monolayer on an electrode surface has been accomplished for a variety of biomacromolecules, such as DNA and proteins, through self-assembly, gold-thiol covalent bonds, and displacement of non-specifically adsorbed species with a competitive agent. Here, we demonstrate that genetically modifying the RC to introduce two cysteines enables covalent binding to the gold surface ( Figure 1 B) and that adsorbed RCs can be arranged in different surface structure assemblies, such as a monolayer or multilayer configuration. The RC is a membrane-embedded protein-pigment complex that consists of three polypeptide chains that provide a scaffold for ten cofactors. These pigments in the RC are a dimer of bacteriochlorophyll (BChl) molecules called the “special pair” (P), two accessory BChls ( B A , B B ), two bacteriopheophytins (BPhe) ( H A , H B ), two quinones ( Q A , Q B ), a carotenoid, and an iron atom ( Figure 1 A). Light adsorption by P results in the excited state P ∗ , generating an electron that cascades through B A and H A , reaching Q A and forming the P + Q A − charge-separated state. Electrical photocurrent flows if the RCs are oriented appropriately on an electrode for electrons to transfer from Q A − to the electrode ( Jun et al., 2018 ). Figure 1 Schematic of the RC cofactors involved in photocurrent generation and the proposed RC configuration on the gold surface (A) The cofactors in the RC are arranged in two symmetrical branches. The cofactors consist of the special pair bacteriochlorophylls (BChls), P (green); accessory BChls, B A and B B (cyan); bacteriopheophytins (BPhes), H A and H B (yellow); quinones, Q A and Q B (orange); a non-heme iron, Fe (black); and a carotenoid (not shown). The electron flow (black arrows) during charge separation is through the A-branch cofactors (denoted with a subscript A) and terminates at the Q B quinone. The forward electron transfer and relaxation time constants are shown in black and gray respectively ( Jones, 2009 ; Feher, 1971 ). Only the bacteriochlorin macrocycles and quinone headgroups are shown for simplicity. Nitrogen atoms are colored as blue, oxygen as red, magnesium as magenta, and carbon as green, cyan, yellow, or orange. The figure is based on PDB 2J8C . (B) A schematic representation of the cysteine modified RC binding to the gold surface through a gold-thiol covalent bond (yellow circle) and the expected path of the photoexcited electrons through the RC and to the electrode surface. HQH 2 is the sacrificial reactant that reduces P + to P and C 6 OH is 6-mercaptohexanol used to passivate the surface. The structural assemblies of electrode-bound RCs yield characteristic photocurrent-potential (i ph -E) responses that can be modeled by the Marcus-Hush-Chidsey theory, and shift the potentials at which photocurrents are generated. The methodology of increasing photocurrents by increasing protein loading and measurements at the OCP may not take into consideration the underlying mechanistic processes occurring on the surface that influence electron transfer. In contrast, we are able to elucidate the differences in energetics between structural assemblies of RC monolayers or multilayers, and their effects on the relationship between photocurrent generation and the OCP.",
"discussion": "Results and discussion The adsorption of RC proteins onto an electrode with the goal to maximize photocurrent generation has been accomplished by deposition in thick layers ( Plumeré and Nowaczyk, 2016 ; Ravi and Tan, 2015 ; Yehezkeli et al., 2014 ; Kamran et al., 2014 ; Mirvakili et al., 2014 ), or with RCs suspended in a redox hydrogel polymer deposited onto the electrode ( Białek et al., 2020a , 2020b ). Although other approaches to orient RCs have been successful in achieving large photocurrents, the possibility of significant random orientations or multilayers of RCs still remains. Large photocurrents are needed to overcome the background faradaic currents and, consequently, these approaches make accurate energetic measurements challenging which impedes the understanding of photocurrent generation at a molecular level. Creating and measuring the photocurrents from a well-defined monomolecular assembly of RC proteins on an electrode surface are described here with this goal in mind. The RC used in this work was engineered such that all native cysteine residues were replaced with serine or alanine residues ( Dutta et al., 2014a , 2014b ; Jun et al., 2018 ; Mahmoudzadeh et al., 2011 ), and through rational design, two additional cysteine residues were added on the hydrophilic regions of the RC in the vicinity of B A and Q A resulting in the “double mutant” RC (referred to as RC). UV-vis absorbance spectroscopy indicated no changes in the pigment cofactor absorbance bands, inferring that the protein structure was not adversely affected by the mutations (absorbance spectra provided in Supplemental information ). On a low-resolution millisecond timescale, the P + Q A − and P + Q B − charge recombination time constants of the RC were similar to that of previously reported values of an unmodified wild type RC (data in Supplemental information ) ( Dutta et al., 2014b ; Jun et al., 2020 ). Modifying RCs with cysteine thiol moieties enables the use of gold-thiol covalent bonds for binding RCs on a gold electrode surface in a well-defined orientation. Although RCs have been bound to the P-side ( Hollander et al., 2011 ; Kondo et al., 2007 , 2012 ; Yaghoubi et al., 2014 ; Friebe et al., 2016b ; Tan et al., 2012 ) or Q-side ( Jun et al., 2018 , 2019 ; Trammell et al., 2006 ) previously, our approach is different and enables binding of the RC on its side such that Q A is closer to the electrode surface than in other configurations ( Figure 1 B). It is important to note that this approach will not prevent multilayer formation because RCs are large, complex protein entities and the formation of multilayers is most likely due to hydrophobic interactions and entropic energy gains. Therefore, removing multilayers and non-specifically adsorbed RCs requires a competitive adsorption strategy. For example, in thiol-based DNA self-assembled monolayers (SAMs), an approach to minimize non-specific adsorption involved an additional step where the self-assembly of small alkylthiols competitively displaced weakly interacting, non-specifically adsorbed DNA ( Herne and Tarlov, 1997 ). Here, we used a similar strategy to prepare and characterize a monolayer of RCs on a gold electrode. Preparing RC/MCH surfaces The alkylthiol 6-mercaptohexanol (MCH) was used to displace multilayers and non-specifically adsorbed RCs. Additionally, MCH treatment passivated areas of the electrode not covered with RCs from the sacrificial electron donor HQH 2 and limited faradaic current. HQH 2 was used due to the favorable difference between its midpoint potential of ∼250 mV vs. standard hydrogen electrode (SHE) and the P + reduction midpoint potential of ∼500 mV vs. SHE ( Jun et al., 2018 , 2019 ; Lin et al., 1994 ), and has been observed to not affect or interact with the Q B quinone pocket ( Jun et al., 2018 ). Figure 2 shows changes in the faradaic currents from HQH 2 redox and photocurrents (i ph ) after RC deposition and subsequent MCH treatment. Deposition of 5 μM RCs (24 h) on bare gold resulted in HQH 2 redox behavior that was significantly reduced when compared to a clean gold surface (shown in Supplemental information , Figure S1 ) indicating the surface was significantly blocked ( Finklea et al., 1993 ; Diao et al., 1999 ). Subsequent exposure of this same surface to 100 μM MCH (1 h) decreased the faradaic current, as expected, and notably resulted in larger photocurrents. The decrease in the faradaic currents of HQH 2 redox may be due to the displacement of multilayers and non-specifically adsorbed RCs by MCH and thus better passivation of the electrode surface; an MCH SAM significantly blocked faradaic currents as shown in Figure 2 A (green trace). Additionally, removing RC multilayers may allow for a larger number of photons to reach RCs that are close to the electrode surface and more efficient in electron transfer, increasing i ph . Figure 2 Faradaic currents and photocurrents measured on RC-modified gold electrodes (A) Faradaic currents and (B) photocurrents measured from surfaces modified with deposition of RC from a 5 μM RC solution with and without MCH post treatment. An electrode surface covered with an MCH SAM is shown for comparison. Measurements were in PBS with 20 mM HQH 2 , at 2 mV/s and using 810 nm LED illumination (LED characteristics provided in Table 2 ). (—: negative potential scan, - - -: positive potential scan). In contrast, a short deposition time of 5 μM RCs (10 min) followed by a longer incubation in 100 μM MCH (24 h) resulted in decreased photocurrent and a faradaic current that was nearly identical to that of an MCH-only surface. This indicated that an organized RC/MCH monolayer was created, resulting in a well-blocked electrode surface. The decrease in photocurrent relative to the 24 h incubation of RCs suggested that MCH displaced not only the majority of multilayers and non-specifically adsorbed RCs, but possibly also some of those covalently linked to gold, via a thiol exchange mechanism ( Schlenoff et al., 1995 ; Baralia et al., 2005 ). The MCH-only surface generated a small background photocurrent signal due to the leakage of the faradaic current (∼0.01%). The removal of non-specifically adsorbed RCs and multilayers of RCs changed the i ph -E response relative to an organized monolayer of RC/MCH. The i ph -E relationship is addressed in subsequent sections, where we hypothesize that organized monolayers of RC/MCH generate measurable photocurrents at an onset potential (E onset ) more positive than that of RC/MCH multilayers. Crucially, we have established a procedure that consistently generates RC/MCH surfaces that show the same blocking of faradaic currents as an MCH-only surface, which is essential for controlling and replicating the formation of biomolecular surfaces and redox reactions in order to make energetic measurements under well-defined conditions. The faradaic currents shown in Figure 2 A were sensitive to the changes on the surface resulting from differences in RC organization and passivation by MCH. Electrochemical impedance spectroscopy (EIS) was also measured at the OCP (+220 mV/SHE) for these surfaces. EIS provided more evidence of the change in the state of the adsorbed layers. It confirmed the increased blocking of HQH 2 redox corresponded with the increase in MCH coverage. Capacitance of the interface was also estimated though the use of a constant phase element (CPE) circuit component that also reports on the uniformity of the interfacial response to potential perturbation (details in Supplemental information , Table S1 ). The capacitance decreased by almost 50% as the MCH content on the surface increased indicating that adsorbed RCs do not form a low dielectric layer on the electrode surface as compared to MCH, likely due to entrapped electrolyte in the disorganized layer or poor adsorption to the surface. EIS also indicated that the absence of an MCH treatment resulted in a surface that was less ordered and more heterogeneous as compared to the RC-modified surface that was treated with 100 μM MCH (24 h). This behavior was also observed for the considerably shorter 1 hr MCH treatment, which implied that significantly heterogeneous surfaces result if the electrodes are only incubated with RCs and not subsequently immersed in MCH. Additionally, EIS showed that MCH also coated the bare areas of the gold surface in addition to displacing non-specifically adsorbed or multilayers of RCs, as evidenced by the changes in capacitance and the CPE exponent. Unfortunately, EIS was unable to distinguish the characteristics of the surfaces that produced different i ph -E response curves and different E onset in Figure 2 . This is reasonable as the comparison was based on HQH 2 redox, which is much more sensitive to the extent of surface blocking. Therefore, CV and EIS based redox measurements cannot be relied upon to detect different surface structures or explain the differences in the i ph -E response curves because these techniques were neither sensitive nor specific enough. However, identification of the transition from a monolayered to a multilayered surface was revealed by the i ph -E response curves. Distinguishing a monolayer or multilayer of RCs with i ph -E response curves The photocurrents and the resulting i ph -E response curves were more sensitive to the structural differences of a monolayer or multilayer configuration of RCs on the electrode surface when compared to electrochemical characterization. Surfaces were prepared by varying the RC concentration from 0.5 to 10 μM and maintaining the incubation time of 10 min, followed by immersion in a 100 μM MCH solution (24 hr), as previously established, to form a uniform layer of RC/MCH. The photocurrents and i ph -E response curves are shown in Figure 3 . Figure 3 Faradaic currents and photocurrents measured on RC-modified gold electrodes after MCH treatment Faradaic currents (A) and photocurrents (B) measured from gold surfaces prepared by a 10 min deposition into a solution of RC of concentrations from 0.5 μM to 10 μM, followed by post-treatment with immersion in 100 μM MCH for 24 h. An electrode surface covered with an MCH SAM is shown for comparison. Measurements were in PBS with 20 mM HQH 2 , 2 mV/s sweep rate, and 810 nm LED illumination. (—: negative scan, - - -: positive scan) First, Figure 3 confirms that the photocurrents were not contaminated by the faradaic current of HQH 2 redox occurring concurrently because no photocurrent was measured from the MCH-only surface. Furthermore, the faradaic responses of all the RC/MCH surfaces were similar to that of an MCH coated surface, further confirming that the electrochemical response was not effective for analyzing and elucidating the structural configuration of the assembled RC adsorbates. The i ph showed similar E dependence for all the RC concentrations, with a plateau in photocurrent generation between 2.5 and 5 μM RC. A significant increase in photocurrent was observed for the surface prepared with 10 μM RC, which was also accompanied by a change in the shape of the i ph -E response curve. The i ph -E response showed that the onset of photocurrent increase began at a more negative potential when compared to the surfaces prepared with lower concentrations of RCs (+100 mV vs +220 mV/SHE, respectively). The potential at which the photocurrent was above the noise (>0.2 nA/cm 2 ) was designated as the onset potential (E onset ) for photocurrent generation. This negative shift in E onset behavior was similar to the trend previously described with either no MCH treatment or a short time in MCH ( Figure 2 ). Earlier, we hypothesized that the more negative E onset potential appeared to be a characteristic of RC multilayers on the electrode surface relative to a monolayer. Here, the data suggest that the transition from a monolayer to a multilayer occurs between 5 and 10 μM RC solutions. Additionally, at higher RC concentrations, despite incubation with 100 μM MCH (24 h), the negatively shifted E onset value suggests the presence of multilayer features that remain on the surface after treatment. As the other surfaces prepared with 0.5–5 μM RC showed the same E onset value, we hypothesized that they consisted of a monolayer adsorbed configuration rather than a multilayer assembly on the surface. Confirmation of the hypothesis required visualization of the surface using atomic force microscopy (AFM). AFM evaluation of the RC/MCH monolayer surface The electrochemical and photocurrent measurements indirectly showed the differences between monolayer or multilayer structural assemblies of RCs on the electrode surface. A direct measurement using ex situ AFM confirmed the trends in the changes in the RC/MCH surface structures that resulted in the i ph -E response curves. Surfaces were modified with RCs, as previously described, using conditions identified as representative of monolayer formation and of multilayer formation. Previously, the Au(111) facet, of a single crystal gold bead electrode was used for AFM studies as it is atomically flat ( Ogaki and Itaya, 1995 ). In the Supplemental information , Figure S2 shows a 5 × 5 μm AFM image of an MCH-coated Au(111) facet, where features of one or more atomic steps of Au (0.2 nm) are observed. These are substantially smaller than the expected height of adsorbed RCs, making this surface ideal for height determinations. Figure 4 shows AFM images from such an Au(111) facet for three different surface preparations, clearly showing significant differences in the surface structures. Much more flat gold (set as 0 nm), which was most likely MCH covered, was visible in Figure 4 A as compared to the higher coverage surfaces in Figures 4 B and 4C. Essentially no MCH covered gold was observed for the high RC coverage ( Figure 4 C). Figure 4 AFM topography of an RC-modified Au(111) facet after MCH treatment AFM topological images from a Au(111) facet after modification with (A) 2.5 μM RC (10 min) followed by 100 μM MCH (24 h). (B) 5 μM RC (10 min), followed by 100 μM MCH (24 h), (C) 5 μM RC (10 min), followed by 100 μM MCH (1 h). The z scale represents height in nm and is the same for all figures. Topographic images using the full z scale range are provided in Supplemental information . All scale bars are 500 nm. An estimate of the coverage and configuration of adsorbed RCs on the electrode surface was achieved using AFM imaging of the RC monolayer modified Au(111) facet. The RCs can be thought of as an elliptical cylinder that is ∼8 nm in length from the P-side to the Q-side and ∼5 nm in width along the elliptical minor axis; the height is ∼7 nm along the elliptical major axis ( Figure 1 B) ( Trammell et al., 2007 ; Jones, 2009 ) AFM imaging accurately measures the height of well-defined features; however, lateral dimensions are less reliable due to convolution of the tip shape with the entity being imaged. Here the tip radius and the RCs are of similar size, around 10 nm. A 2 × 2 μm topographic AFM image of an Au(111) facet exposed to 2.5 μM RC (1 h), followed by 100 μM MCH (24 h) which is representative of monolayer formation, shown in Figure 4 A. The analysis (outlined in Supplemental information ) of the complete 5 × 5 μm image found >8000 particles (outlined in Figure S3 ) that were easily identified because of the flat gold substrate that surrounded the particles. An example of the individual features observed on this surface is shown in Figure 5 A along with line scans across the features and from the nearby flat Au(111) surface. Complete analysis of the maximum height distribution for all the particles found in the full 5 × 5 μm image is shown in Figure 5 B (details in Supplemental information Figure S4 ). These particles were on average 5–7 nm high, with a broad range of heights up to 20 nm, though only ∼10% were above 10 nm. The observed height corresponded well with the expected RC height of 5–7 nm, if they adsorbed to the surface as designed. The average footprint area of the 5–7 nm high particles was 350 nm 2 , equivalent to a 20 nm diameter. Because tip convolution adds about 10–15 nm to the diameter, the actual diameter of these particles was closer to 6–10 nm. Given the observed heights and diameters, and known RC dimensions, most of the observed particles were single adsorbed RCs arranged in a monolayer. Figure 5 AFM analysis of individually adsorbed RCs on gold Analysis of the AFM images shown in Figures 4 A and 4B. (A) Zoomed-in region from the surface prepared using 2.5 μM RC (10 min) followed by 100 μM MCH (24 h) showing the individual RC particles on a flat MCH covered Au(111) surface with the corresponding height profile taken along the white dashed line. Scale bar is 20 nm. (B) Maximum height distribution analysis for the sample in Figure 4 A. (C) Two zoomed-in regions from the 5 μM RC (10 min) followed by 100 μM MCH (24 h) showing low and medium coverage regions containing chains of RCs. Scale bar is 100 nm. (D) Maximum height distribution from the AFM image in Figure 4 B. Another surface was prepared, which was expected to be representative of a multilayer (5 μM RC (10 min), 100 μM MCH (1 h)), and is shown in Figure 4 C (details in Supplemental information , Figure S8 ). The surface was completely covered with particles and regions of flat gold surface were not visible, making absolute height determinations problematic. This sample was complicated to analyze, as individual isolated RCs were not found in this multilayer configuration. Even so, the number of particles that were taller than their neighbors by 10–20 nm was significantly larger when compared to the monolayer sample (15% vs. 6% respectively). In other experiments, a similar surface was prepared using the same solution of 5 μM RC (10 min), but followed by 100 μM MCH for 24 h, rather than 1 h. The topographic AFM image is shown in Figure 4 B, and the analyses and data are shown in Figures 5 C and 5D (more details in Supplemental information , Figures S6 and S7 ). The average maximum height distribution was ∼5 nm, slightly lower than the surface incubated with 2.5 μM RC, but well within the expected range. A significant fraction (>5%) of these particle heights was greater than 10 nm, suggesting the existence of multilayer formations. Interestingly, the adsorbed species were arranged differently with fewer individual isolated particles and more that appeared to be interconnected. Two different aggregation states were observed and are shown in Figure 5 C. The interconnected nature may be a consequence of RC protein aggregation as a precursor to multilayer formation. Importantly, regions of bare gold were visible again that were previously obscured in the multilayer configuration. From the i ph - E response curves, the surface preparation conditions using 2.5 μM RC (10 min)followed by 100 μM MCH (24 h) and 5 μM RC (10 min) followed by 100 μM MCH (24 h) were hypothesized to generate a monolayer configuration, and a multilayer configuration for 5 μM RC (10 min) followed by 100 μM MCH (1 h). The confirmation with AFM underscores the importance of MCH treatment as necessary to remove multilayer or ill-formed adsorbate configurations. The increase in surface density of adsorbed RCs and the shift from a monolayer to a multilayer configuration corresponded to the changes observed in the photocurrent measurements, namely the shift in the E onset to more negative potentials. This characteristic seems to be a signature of multilayer adsorption and becomes difficult to model due to the numerous factors that are active at the interface and in the adsorbed film (e.g., diffusion of mediator, electron transfer kinetics and thermodynamics, random or unknown orientations of the RC in the multilayer). Therefore, modeling of the energetics of the electron transfer event is focused on the well-defined monolayer RC/MCH surfaces. Modeling i ph -E response curves for monolayer RC/MCH surfaces The photocurrent and AFM analysis of the RC/MCH surfaces showed that under specific deposition conditions, a well-defined monolayer of RCs can be prepared; therefore, an estimate of the electron transfer characteristics is possible. The RC is bound to the electrode surface via two cysteines, and if oriented correctly, Q A is closest to the electrode so that electron transfer can occur. Upon absorption of a photon by P BChls, an excited state is formed, P ∗ , and an electron rapidly cascades through a chain of cofactors to the Q A quinone resulting in the P + Q A − charge-separated state. Concurrently, an HQH 2 reduces P + back to P. The photocurrent measured is the electron transfer from the Q A cofactor to the electrode, driven by the difference in the potential between Q A − and the gold surface (its Fermi level). This process can be modeled by electron transfer theories, most appropriately with the Marcus-Hush-Chidsey model ( Zeng et al., 2014 ). The i ph - E response curves of RC/MCH surfaces were fitted to the Marcus-Hush-Chidsey theory using only photocurrents greater than 0.2 nA/cm 2 (at least 5-fold larger than the smallest reliably-measured photocurrent). These RC modified surfaces were continuously illuminated with a modulation in the intensity range of 10–90% at 13 Hz, and so the photocurrents were representative of processes that were fast enough to respond to the intensity modulation. The resulting fits and parameters, such as the reorganization energy and the equilibrium potential ( E e q p h ), are given in Figure 6 and Table 1 (individual fit results shown in Figure S11 ). The data correlated well with theory (r 2 > 0.995). Figure 6 Marcus-Hush-Chidsey analysis of the photocurrents measured from RC-modified gold electrodes Marcus-Hush-Chidsey fits for the i ph -E curves for data shown in Figure 3 were used to fit for photocurrents >0.2 nA/cm 2 . The experimental data shown are an average of the anodic and cathodic sweeps and shown as symbols (only every 25 th data point) and the line represents the fit. Measurements were in PBS with 20 mM HQH 2 , 2 mV/s sweep rate, and 810 nm LED illumination. Table 1 Parameters from fitting to the Marcus-Hush-Chidsey theory using only photocurrents (>0.2 nA/cm 2 ) [RC] (μM) λ 810 n m (eV) E e q p h (mV/SHE) 0.5 0.271 ± 0.016 +253 ± 8 1.0 0.240 ± 0.001 +241 ± 1 1.0 0.194 ± 0.001 +286 ± 1 2.5 0.248 ± 0.001 +242 ± 1 5.0 0.215 ± 0.001 +262 ± 1 5.0∗ 0.289 ± 0.001 +217 ± 1 10.0 0.301 ± 0.001 +152 ± 1 Uncertainties (1 σ ) are determined using a weighted fitting routine in MATLAB (fitnlm) with weights determined from the moving standard deviation calculated using a 20 mV window. ∗ Large faradaic current (see data in SI: Figure S10 ). For low coverage RC monolayers prepared using 0.5 to 2.5 μM RC, the reorganization energies were on average 0.24 ± 0.03 eV and E e q p h = +257 ± 19 mV/SHE. The reorganization energies are similar to values reported in the literature for surface-bound RCs ( Trammell et al., 2007 ), but smaller than for the Q A − to Q B electron transfer in RCs in solution ( Ptushenko and Krishtalik, 2018 ). Interestingly, one of the surfaces prepared with 5 μM RC showed similar results, but a repeat of this preparation had a larger reorganization energy (0.29 eV) and a more negative E e q p h of +217 mV/SHE (shown in Supplemental information , Figure S10 ). Unlike lower concentrations of RCs, the 5 μM RC preparation was challenging to replicate in a monolayer and as postulated earlier, this may be the concentration at which multilayer formation and protein-protein interactions begin to exert a larger influence on adsorbed surface formation. The surface prepared with 10 μM RC also followed this trend with a reorganization energy of 0.30 eV and E e q p h of +152 mV/SHE. Multilayer characteristics, such as interconnected regions shown in the AFM images, are correlated to fits that yielded higher reorganization energies and more negative E e q p h . The same analysis was performed on the faradaic currents measured in the same region of potential used in the photocurrent analysis. The oxidation of HQH 2 was consistent for all samples with a reorganization energy that ranged from 0.36 to 0.44 eV with an E e q p h of +200 to +230 mV/SHE. All data and fits are shown in Supplemental information ( Table S3 and Figure S9 ). Fitting the photocurrents measured at the same time show distinctly different values of the reorganization energy and E e q p h when compared to the faradaic currents, confirming that the photocurrents measured were not influenced by the faradaic currents. The systematic increase in the fitted reorganization energy and more negative E e q p h with an increase in RC coverage may indicate the presence of another RC configuration. Fitting to only one set of parameters may produce average reorganization energy and E e q p h resulting from two (or more) different populations of RCs on the surface or RCs in different environments. In order to elucidate the change in E onset further, multilayers were evaluated using the same fitting procedure for surfaces modified with 5 μM RC (24 h) without MCH treatment as well as 5 μM RC (24 h) with 100 μM MCH (1 h) treatment. Using a single set of parameters, the fits continued the trend observed for higher concentration RC treated surfaces, namely larger reorganization energies (0.427 eV (no MCH) and 0.247 eV (1 h MCH)) and an E e q p h shift to more negative potentials (−87 mV/SHE (no MCH) and +40 mV/SHE (1 h MCH)). The fit to theory improved with the addition of a second set of parameters (details in Supplemental information , Tables S4 and S5 and Figure S12 ). One component had a reorganization energy of 0.28 ± 0.02 eV which was consistent for both multilayer samples. The E e q p h parameter for both samples ranged from +125 to +75 mV/SHE. This component was present at about 50% of the adsorbate for both surfaces. The second component was the same for both multilayer preparations with a small reorganization energy of 0.06 ± 0.002 eV and E e q p h = +150 ± 7 mV/SHE. The results from the two component fitting were consistent with one part of the surface composed of adsorbed RCs similar to a monolayer configuration and a second component, possibly distorted or denatured RCs, indicating little change in conformation during electron transfer. The same shift toward negative potentials was observed for the E onset for the monolayer and multilayer samples, suggesting that E e q p h is closely related to or can be interpreted as E onset . Although the mechanism behind the shift in E e q p h to negative potentials remains uncertain, we suggest a distortion of proteins that shortens the distance for electron transfer to the electrode, or changes in the local environment within the multilayers, due to restrictions in the diffusion of HQH 2 , HQ, and/or H + . This is supported by evidence of diffusion-limited currents in multilayers observed when measuring the DC photocurrent at the OCP. DC photocurrent measurements at the OCP To facilitate comparison with data in the literature, DC photocurrents were measured using 860 nm illumination with the potential set to the OCP (+200 to +220 mV/SHE) with three surface conditions: 5 μM RC (24 h) without MCH (“multilayer”); the same surface, but then incubated with 100 μM MCH (1 hr) (“monolayer-like”); and 2.5 μM RC (10 min), 100 μM MCH (24 h) (“monolayer”). The DC photocurrent responses are shown in the left column of Figure 7 with the right column comparing the photocurrents measured around the OCP (taken from Figure 2 ) using our modulated method that exclusively measures the photocurrent. Figure 7 DC photocurrents measured from RC-modified gold electrodes DC photocurrents measured for three samples prepared using: (A) 5 μM RC (24 h) with no MCH treatment, (B) same electrode as in (A) but with MCH (1 h) treatment, and (C) 2.5 μM RC (10 min) MCH (24 h) treatment. The photocurrents were measured at OCP (+218, +222, +203 mV/SHE for each respective surface) in PBS with 20 mM HQH 2 . The red background indicates when the interface was illuminated with an 860 nm LED (characteristics in Table 2 ). The photocurrent shown in Figure 2 is replotted on the right panels zoomed into the region around the OCP (arrow) for the corresponding interface on the left panel. A light-dependent response was observed in which the “multilayer” surface sustained a DC photocurrent of 10 nA/cm 2 at OCP (+218 mV/SHE). This was approximately 4-fold larger than the photocurrents of 2.5 nA/cm 2 ( Figure 7 A ((right panel)) measured using the modulation method (i ph ) at +218 mV/SHE (arrow in Figure 7 A(right panel)). The DC photocurrents were stable for many seconds suggesting that the current was not limited by HQH 2 reduction of P + . Turning the light off resulted in a slow decrease in the measured current that lasted tens of seconds and was characterized by a diffusional response (in Supplemental information , Figure S15 ), likely a result of the rate at which HQH 2 or HQ diffused into or out of the layer to reach equilibrium with the bulk solution. The same surface was then treated with MCH, creating a “monolayer-like” adsorbed layer, and a similar photocurrent magnitude was obtained in addition to an improved response time observed by the absence of a diffusional response. This effect was also reflected in an increase in the modulated photocurrent to 4 nA/cm 2 since the faster response would be able to produce greater photocurrents at 13 Hz ( Figure 7 B (right panel)). The displacement of multilayers and non-specifically adsorbed RC by MCH eliminated the diffusion behavior. Diffusion-controlled currents for RC modified electrode surfaces have been reported previously and described in multilayer models ( Takshi et al., 2009 ; Mirvakili et al., 2014 ; Buesen et al., 2019 ). The “monolayer” surface generated a small DC photocurrent transient that rapidly returned to baseline after the light was turned on, which is in agreement with the modulated photocurrent measurement where at OCP, no i ph was observed ( Figure 7 F). Examples of the DC photocurrents measured from other monolayer surfaces are shown in Figure S16 . Comparing photocurrent measurements for monolayer and multilayer adsorbed RCs The DC photocurrent response of the monolayer and multilayer RC surfaces depended on the nature of the adsorbed layer; a similar dependency was observed for E onset where significant i ph was measured. The OCP was essentially independent of the surface configuration and similar for all the samples. The measured OCP was controlled by the HQH 2 redox species in solution (+200 to +220 mV/SHE at pH = 7.2). Classically, the OCP is the potential where the net current is zero, and reflects a balance between the rates of oxidation and reduction, dominated by the largest rates for a mixture of redox species. The use of MCH to remove non-specifically adsorbed RCs does not completely passivate the gold surface, so HQH 2 redox still takes place and controls the OCP. The photocurrent-specific E onset was shifted to more negative potentials for the multilayer compared to the monolayer RC covered surfaces. Therefore, these photocurrent measurements may be sensitive to the local environment experienced by the adsorbed RCs. For example, the difference in the E onset measured from the photocurrents of the monolayer and multilayer surfaces was most likely influenced by the changes in the environment within the multilayer, which sets the local redox potential, independent of the OCP. In other words, a monolayer surface is not constrained by layers of protein that impede the access of the redox species, and therefore, the individually adsorbed RCs experience the same conditions as the MCH covered gold electrode surface; the opposite is the case for multilayer surfaces that may “trap” redox species and change the local redox environment. As the E onset shifted to more negative potentials, the local redox environment within the multilayer RC assembly differed from that in solution, which can be rationalized via the Nernst equation for HQH 2 redox: E = E o − 0.0592 2 l o g [ H Q H 2 ] [ H Q ] [ H + ] 2 . The negative shift in E onset would result from an increase in the [HQH 2 ]/[HQ] ratio and/or an increase in pH within the RC multilayer. Meanwhile, the OCP measured via faradaic redox from the entire electrode would be determined by the redox conditions in the solution. Consequently, in the “multilayer” surface, the DC photocurrent measurement at OCP was performed at an overpotential of +100 mV relative to the local equilibrium potential experienced by the RCs in the adsorbed multilayer. This resulted in large photocurrent generation which also had a diffusional character when illumination was removed. Conversely, the “monolayer” surface had an E onset equal to the OCP revealing that these RCs existed in a local environment in equilibrium with the solution phase. Therefore, creation of P + Q A − via photon absorption did not result in significant photocurrents when measured at the OCP as the measurements were performed at a negligible overpotential ( Figure 8 ). Figure 8 Schematic representation of monolayer and multilayer adsorbed RCs and their corresponding photocurrent-potential relationships Schematic representation of the two adsorbed layers (A) multilayer and (B) monolayer on the gold surface and the HQH 2 access to the electrode surface through the MCH SAM or through the RC multilayer. (C) schematic representation of the OCP and onset potentials to the photocurrents measured. Confirming evidence of the influence of [HQH 2 ] on the photocurrent and E onset was obtained by measurements on a surface that had multilayer characteristics similar to the 5 μM RC sample with a short MCH treatment time ( Figure S13 ). An increase in the [HQH 2 ] from 5 to 30 mM correlated with increased photocurrents, a negative shift in E onset (or E e q p h from fitting), and an increase in reorganization energy that agreed with results previously described for multilayer samples (more details in Supplemental information : Figures S13 and S14 and Table S6 ). As expected, no significant shift in the OCP was observed for the faradaic process taking place through the MCH layer on the electrode surface. Therefore, the negative shift in E onset was correlated to the increase in [HQH 2 ]. Partitioning of HQH 2 into the RC multilayer would increase the local concentration of HQH 2 , consequently shifting the E onset for photocurrents to more negative potentials, and explaining the difference between the E onset values for multilayer and monolayer surfaces. Figure 8 shows a schematic of the multilayer and monolayer samples detailing the local environment and its relationship to the measured photocurrents. The photocurrent measurements are sensitive to the adsorbed states of the RCs and their interfacial environment. This reveals that the DC photocurrent measurements done at the OCP provide a limited evaluation of the surface modifications. Moreover, manipulation of the local redox environment could influence the driving force (overpotential) that controls the magnitude of the generated photocurrents. These factors result in an uncertainty of the measurement of electron transfer energetics that are important for understanding the activity of the adsorbed RC and the electron transfer to an electrode from an adsorbed RC protein. Finally, the influence of the redox conditions in local interfacial regions plays an important role for surfaces modified with molecular adsorbates emphasizing the need for approaches that can distinguish between cases where the conditions of the interface and the bulk solution differ. Conclusions The conditions for preparing a gold surface covered with a monolayer of RC covalently tethered via thiol bonds was determined with a modulated illumination intensity method that measured the photocurrent-potential response in addition to ex situ AFM. The photocurrent response under an applied potential for an electrode surface with a partial to full monolayer of adsorbed RCs was fitted to the Marcus-Hush-Chidsey theoretical formalism and a consistent reorganization energy of 0.24 ± 0.03 eV was determined with an equilibrium potential ( E e q p h ) of +257 ± 19 mV/SHE. The photocurrent characteristics were sensitive enough to reveal the initiation of multilayer formation on the gold electrode. Multilayer coated electrode surfaces showed a significant negative shift in the photocurrent E onset and a photocurrent-potential response that was best fit using two components, one with the same reorganization energy as the monolayer samples, and the other component having a close to zero reorganization energy (<0.1 eV). AFM imaging confirmed the presence of isolated adsorbed RCs with 7 nm heights for the monolayer covered surfaces. AFM also showed multilayer formation as aggregated adsorbed species and tall structures. The negative shift in the E onset for multilayers of RCs was hypothesized to be a result of the difference in the local redox conditions present within the adsorbed multilayer. The difference between the local redox potential and the OCP, defined by the redox couple in solution, may create conditions for large DC photocurrents. Limitations of the study The formation of RC monolayers requires proteins that are not significantly oxidized or aggregated. The use of freshly prepared proteins is ideal, but aggregation occurs over time resulting in some reproducibility challenges in making monolayer RC surfaces. The method for measuring the photocurrents can distinguish between monolayer and multilayer samples. The photocurrent signals are very small, which requires a high performance lock-in-amplifier that is capable of working at frequencies below 20 Hz, and not all amplifiers will be able to measure these signals reliably."
} | 11,780 |
23857726 | null | s2 | 7,086 | {
"abstract": "The term hydrogel describes a type of soft and wet material formed by cross-linked hydrophilic polymers. The distinct feature of hydrogels is their ability to absorb a large amount of water and swell. The properties of a hydrogel are usually determined by the chemical properties of their constituent polymer(s). However, a group of hydrogels, called \"smart hydrogels,\" changes properties in response to environmental changes or external stimuli. Recently, DNA or DNA-inspired responsive hydrogels have attracted considerable attention in construction of smart hydrogels because of the intrinsic advantages of DNA. As a biological polymer, DNA is hydrophilic, biocompatible, and highly programmable by Watson-Crick base pairing. DNA can form a hydrogel by itself under certain conditions, and it can also be incorporated into synthetic polymers to form DNA-hybrid hydrogels. Functional DNAs, such as aptamers and DNAzymes, provide additional molecular recognition capabilities and versatility. In this Review, DNA-based hydrogels are discussed in terms of their stimulus response, as well as their applications."
} | 277 |
34140134 | null | s2 | 7,089 | {
"abstract": "Climate change, with its extreme temperature, weather and precipitation patterns, is a major global concern of dryland farmers, who currently meet the challenges of climate change agronomically and with growth of drought-tolerant crops. Plants themselves compensate for water stress by modifying aerial surfaces to control transpiration and altering root hydraulic conductance to increase water uptake. These responses are complemented by metabolic changes involving phytohormone network-mediated activation of stress response pathways, resulting in decreased photosynthetic activity and the accumulation of metabolites to maintain osmotic and redox homeostasis. Phylogenetically diverse microbial communities sustained by plants contribute to host drought tolerance by modulating phytohormone levels in the rhizosphere and producing water-sequestering biofilms. Drylands of the Inland Pacific Northwest, USA, illustrate the interdependence of dryland crops and their associated microbiota. Indigenous Pseudomonas spp. selected there by long-term wheat monoculture suppress root diseases via the production of antibiotics, with soil moisture a critical determinant of the bacterial distribution, dynamics and activity. Those pseudomonads producing phenazine antibiotics on wheat had more abundant rhizosphere biofilms and provided improved tolerance to drought, suggesting a role of the antibiotic in alleviation of drought stress. The transcriptome and metabolome studies suggest the importance of wheat root exudate-derived osmoprotectants for the adaptation of these pseudomonads to the rhizosphere lifestyle and support the idea that the exchange of metabolites between plant roots and microorganisms profoundly affects and shapes the belowground plant microbiome under water stress."
} | 446 |
34140134 | null | s2 | 7,090 | {
"abstract": "Climate change, with its extreme temperature, weather and precipitation patterns, is a major global concern of dryland farmers, who currently meet the challenges of climate change agronomically and with growth of drought-tolerant crops. Plants themselves compensate for water stress by modifying aerial surfaces to control transpiration and altering root hydraulic conductance to increase water uptake. These responses are complemented by metabolic changes involving phytohormone network-mediated activation of stress response pathways, resulting in decreased photosynthetic activity and the accumulation of metabolites to maintain osmotic and redox homeostasis. Phylogenetically diverse microbial communities sustained by plants contribute to host drought tolerance by modulating phytohormone levels in the rhizosphere and producing water-sequestering biofilms. Drylands of the Inland Pacific Northwest, USA, illustrate the interdependence of dryland crops and their associated microbiota. Indigenous Pseudomonas spp. selected there by long-term wheat monoculture suppress root diseases via the production of antibiotics, with soil moisture a critical determinant of the bacterial distribution, dynamics and activity. Those pseudomonads producing phenazine antibiotics on wheat had more abundant rhizosphere biofilms and provided improved tolerance to drought, suggesting a role of the antibiotic in alleviation of drought stress. The transcriptome and metabolome studies suggest the importance of wheat root exudate-derived osmoprotectants for the adaptation of these pseudomonads to the rhizosphere lifestyle and support the idea that the exchange of metabolites between plant roots and microorganisms profoundly affects and shapes the belowground plant microbiome under water stress."
} | 446 |
33318197 | PMC7776822 | pmc | 7,092 | {
"abstract": "Significance Changes in ocean redox chemistry are frequently observed in Earth’s history and have fundamental implications for the evolution of marine life. These transitions are commonly ascribed to large changes in the supply of iron, sulfur, or organic carbon in the deeper ocean. We propose that small variations in carbon input flux can drive nonreversible redox changes of the ocean interior and other anoxic systems, such as marine sediments. Nonlinear interactions in the iron and sulfur cycles create tipping points where regime shifts can occur between alternative stable states that are either iron dominated or sulfide dominated. The recognition that the biogeochemistry of sediments and oceans embeds intrinsic bistability provides a conceptual framework for understanding past and present anoxic marine systems.",
"conclusion": "Conclusions and Perspectives Our model analysis supports the existence of redox bistability in marine sediments and explains the redox dichotomy observed in modern marine aquatic sediments ( 20 – 22 ). At the same time, it provides a parsimonious explanation both for shifts in the ocean redox state that occurred from the late Archean to the early Phanerozoic ( 9 , 17 ) as well as the short-term redox oscillations observed in the mid-Cretaceous ( 11 ). However, the dependence of ocean redox shifts on export productivity does not obviate the potential impact of changes in sulfur and iron fluxes on the nature of ocean anoxia. Essentially, the interplay between export productivity and the fluxes of sulfur and iron will determine whether a system shifts between ferruginous and euxinic states. If the sulfate flux to the ocean is too low, sulfide production cannot completely remove all ferrous iron from solution, and the ocean interior will not shift to an euxinic state ( 19 ). Conversely, if the sulfate flux is high enough to generate euxinia but productivity is too low, the ocean will also not shift to an euxinic state. Because nutrient availability is strongly dependent on the redox state, the existence of redox bistability has potentially profound implications for the causes and consequences of marine evolutionary innovations and ecological transitions. For instance, phosphate limitation could have been maintained in a ferruginous ocean via scavenging by iron oxides near the ocean surface or precipitation of ferrous phosphate minerals ( 12 , 40 , 41 ), keeping primary productivity low enough to stay below the bistability threshold. In contrast, any transition induced to euxinic conditions could lead to the occurrence of iron or other trace element limitation ( 11 , 13 ), eroding the stability of the euxinic state and driving the system back to ferruginous conditions. This could in large part explain the existence of predominantly ferruginous redox conditions in the ocean interior throughout much of the Precambrian, as implied by the rock record ( 7 , 19 ). The potential for such an interplay and feedback between the ocean redox state, nutrient availability, and marine productivity, and, ultimately, with the evolution of marine life, hints at a hitherto-underappreciated complexity in the dynamics of the Earth system. As such, our work has implications for understanding the causes of past transitions in ocean biogeochemistry and their consequences for the biosphere. The fact that bistability and hysteresis emerge within a simplified ocean model as well as different reactive transport models of sediment biogeochemistry suggests that it is deeply rooted within anoxic systems, and, hence, the phenomenon should also emerge in more complex models (e.g., ref. 11 ). Still, one should be aware of the limitations of the ocean and sediment models employed here. For instance, our sediment models assume that iron oxide minerals only consist of one or two fractions with different reactivities toward sulfide. Accounting for more complex iron mineralogy (e.g., by including green rust, likely an important mineral in Precambrian environments ( 31 )) will likely not prevent the occurrence of bistability but could affect the location or width of the bistability zone. Likewise, our ocean model analysis is based on steady state, so we cannot make any inferences about the frequency of redox shifts or whether these match periodicities observed in Earth’s rock record. Also, the three-box ocean model employed here does not capture any spatial variation in redox conditions, and as a result, we cannot assess whether increasing spatial resolution and heterogeneity of the ocean environment may lead to loss of redox bistability. Therefore, an important next step is to investigate the phenomenon in a dynamical and spatially resolved Earth system model and conduct both transient and steady-state simulations. This will allow for making explicit comparisons with the geological record and its attendant redox proxies.",
"discussion": "Results and Discussion In this study, we systematically examine a series of models that describe the iron and sulfur cycling in different marine environments (both sediments and water column). Drawing upon recent geochemical observations, we first describe that iron and sulfur cycling in modern salt marsh sediments displays a built-in redox dichotomy, where both an iron-rich and sulfide-rich state can occur under identical boundary conditions. Subsequently, we demonstrate that such bistable redox states are not limited to salt marsh sediments but are an inherent feature of models that describe a wider range of natural anoxic sediments. In the third step, we demonstrate that a similar bistability is also present in a model for large-scale ocean redox chemistry. Via these models, we hence progressively link small-scale observations made in salt marsh sediments, through a more generalized marine sediment system, to the ocean interior. In the final step, we uncover the common underlying cause of the bistability, demonstrating that the bistability is uniquely linked to a restricted set of core reactions that feature in all models examined, of which the precipitation of reduced iron monosulfide (FeS) minerals is the key process responsible for bistability. As FeS formation is common to many anoxic systems, our analysis hence demonstrates how redox bistability emerges as an inherent property of the anoxic marine environment, both in modern times and in the geological past. Redox Bistability in Salt Marsh Sediments. We start with an analysis of the sediment geochemistry of salt marsh sediments. Recently, it has been observed that stagnant ponds in salt marshes along the Norfolk coast (East Anglia, UK) segregate into two end members that exhibit highly different sediment geochemistry, with pore waters either rich in dissolved ferrous iron or rich in dissolved sulfide ( 20 – 22 ). In the sulfide-rich pond sediment, >70% of the reactive solid-phase iron is reduced and bound to sulfide, while ferrous iron is undetectable in the pore water and dissolved sulfide accumulates to 10 mM ( Fig. 2 A , C , E , and G ). In contrast, in the iron-rich pond sediment, >60% of the solid-phase iron is present in oxidized form and ferrous iron accumulates up to 0.5 mM, with no accumulation of dissolved sulfide ( Fig. 2 B , D , F , and H ) ( 22 ). Importantly, these two different redox states can be found in neighboring ponds only a few meters apart ( SI Appendix , Materials and Methods , section 1 and Fig. S1 ), and this behavior is seen in different salt marsh systems along the ∼200-km-long coastline of East Anglia ( 20 , 21 ). It has been suggested that local differences in iron supply and organic matter input may drive the redox dichotomy ( 20 – 22 ), although it is unclear how the boundary conditions could vary dramatically and systematically over the small distances (<10 m) between ponds. Here, we propose that intrinsic bistability in the coupled cycles of iron and sulfur gives rise to the observed redox dichotomy in the salt marsh pond sediments. Fig. 2. The redox dichotomy observed in salt marsh sediments. Iron and sulfur speciation data are shown from two nearby ponds in a Norfolk (UK) salt marsh. Panels A – G show data from the sulfidic pond, while B – H are data from the ferruginous pond. ( A and B ) Pore water dissolved iron (Fe pw ). ( C and D ) Solid-phase reactive iron (Fe reac ) = oxidized iron (Fe ox ) + reduced unsulfurized iron (Fe unsulph ) + reduced sulfurized iron (Fe sulph ). ( E and F ) Pore water free sulfide (ΣH 2 S pw = [H 2 S] + [HS − ] + [S 2− ]). ( G and H ) Solid-phase inorganic sulfur (S inorg ) = acid-volatile sulfide (S AVS ) + elemental sulfur (S 0 ) + pyritic sulfur (S PYR ). Data were originally presented in ref. 22 . We use the term “bistability” to describe a dynamical system that can be found in either of two stable equilibrium states for identical boundary conditions ( Fig. 3 A provides the classical conceptual representation of such a bistable system). In the case of the salt marsh pond sediments, these two stable states represent pore waters that are either rich in ferrous iron (ferruginous) or rich in dissolved free sulfide (euxinic). The euxinic state is characterized by high dissolved sulfide concentrations and also a higher degree of sulfurization (most iron minerals are in reduced form and bound to sulfide), whereas the ferruginous state is characterized by high concentrations of ferrous iron in the pore water and a lower degree of sulfurization ( Fig. 2 ). If the system starts in an intermediate unstable state, even a small perturbation will direct it toward one of the two stable states ( Fig. 3 A ). Fig. 3. ( A ) A conceptual illustration of bistability in iron–sulfur geochemistry. For a certain set of boundary conditions (e.g., the flux of organic matter and iron oxides), the dominant redox state of an anoxic system can be either sulfide rich or iron rich. See the article text for details. ( B ) The positive feedback scheme that generates bistability in the geochemistry of anoxic systems. See the article text for details. ( C – E ) The model results for the reactive transport model of salt marsh sediments ( SI Appendix , Materials and Methods , section 2 ). ( C ) The steady-state concentration of reactive iron oxide (C FeOOH ) versus the ratio of organic matter influx (F CH2O ) and iron oxide influx (F FeOOH ). ( D ) The steady-state concentration of dissolved sulfide (C HS− ) versus the ratio of organic matter influx and iron oxide influx. The y axis in C and D are normalized against the maximum concentration in the plot for clarity. Each point represents the results from a single steady-state model simulation. The bistability zone (gray zone) occurs when two divergent steady states are present for identical input parameters. The purple dots indicate the S-rich state; the red dots indicate the Fe-rich state. The arrows indicate the direction of the hysteresis loop. The black box indicates the steady-state model simulation for which the diagenetic profiles are shown in E . ( E ) The diagenetic profiles for the S-rich and Fe-rich stable states. To test whether the experimentally observed dichotomy in pond sediment geochemistry ( Fig. 2 ) is an expression of intrinsic bistability in sedimentary Fe and S cycling, we developed a detailed one-dimensional reactive transport model that describes the salt marsh sediment geochemistry in detail, explicitly accounting for different types of iron oxides and organic matter with different reactivities and including all key reactions of the coupled oxygen, carbon, iron, and sulfur cycles ( 22 – 24 ) ( SI Appendix , Materials and Methods , section 2 ). The model was parameterized using the available data for the East Anglian salt marsh ponds ( 22 ). We dynamically ran the sediment model to steady state for different inputs of reactive carbon to the sediment (F CH2O ), which is the primary driver of geochemical cycling in sediments. In these simulations, we kept the input of reactive iron oxides to the sediment (F FeOOH ) and all other model parameters constant. Note that the F CH2O and F FeOOH fluxes only include those forms of organic carbon and iron oxides that are reactive on early diagenetic timescales (∼100 y), and hence they do not incorporate, for example, highly refractory organic carbon or Fe from more poorly reactive iron oxides. The model output shows bistable behavior and regime shifts ( Fig. 3 C and D ). When running the model for increasing F CH2O values, there is a tipping point past which the system suddenly switches from an Fe-rich state to an S-rich state. When the input of organic carbon is lowered again, the system can be brought back to a ferruginous state, but hysteresis occurs: the forward trajectory differs from the backward trajectory ( Fig. 3 C and D ). In particular, upon decreasing F CH2O /F FeOOH , the organic matter input to the system needs to be reduced considerably more before the system shifts back to the Fe-rich state. The two tipping points delineate the bistability zone (F CH2O /F FeOOH ranging between 17.5 and 21; Fig. 3 C and D ), where two steady-state solutions are present for identical boundary conditions ( Fig. 3 E ). The model-simulated depth profiles associated with these two alternative stable states are comparable to what is experimentally observed in the salt marsh sediments ( Fig. 3 E ). The observed redox bistability in the model can be explained by a positive feedback between known interactions within the sedimentary Fe and S cycles ( Fig. 3 B ). When the F CH2O /F FeOOH ratio is low, the reactive iron oxide input is high enough to suppress sulfide accumulation by limiting sulfate reduction and the reoxidation of sulfide via iron oxides. Under these conditions, the sediment has only a single Fe-rich stable state, characterized by high solid-phase iron oxide (FeOOH) concentrations and ferruginous pore water conditions. If one gradually increases the input of organic matter, the system shows resilience and remains within the Fe-rich state. However, if one increases the input of organic matter sufficiently, the system passes a tipping point beyond which it suddenly shifts from an Fe-rich to a S-rich state, characterized by low FeOOH concentrations and highly sulfidic pore waters ( Fig. 3 C – E ). Near the tipping point, the increased organic matter input increases the quantitative importance of sulfate reduction, which generates free sulfide that removes ferrous iron via the formation of solid-phase FeS. At this point, a positive feedback emerges: as FeS precipitates, less ferrous iron gets reoxidized, and so less FeOOH becomes available for dissimilatory iron reduction. This condition then further stimulates sulfate reduction and sulfide production, which further promotes FeS precipitation ( Fig. 3 B ). The overall result is a positive feedback, which causes the rapid transition from the Fe-rich to the S-rich state beyond the tipping point. Because the estimated F CH2O and F FeOOH values for the Norfolk salt marshes are subject to considerable uncertainty, the empirically estimated ratio of organic matter influx over the iron oxide influx (F CH2O /F FeOOH ) lies within a broad range from 10 to 34 ( SI Appendix , Materials and Methods , section 1 ). Still, the median value falls within or near the model-predicted bistability zone (F CH2O /F FeOOH from 17 to 21), and, therefore, bistability provides a plausible explanation for the observed redox dichotomy in the East Anglian salt marsh systems. For example, consider an initial state in which all ponds start in an Fe-rich condition with a low organic matter input. As the marsh develops and vegetation expands with time, the flux of organic matter to the ponds will increase, and so the F CH2O /F FeOOH ratio may approach the tipping point at the right edge of the bistability zone. In this condition, most ponds remain Fe dominated, but some ponds may receive a small additional input of organic matter for a short while, upon which they pass the tipping point and shift to an S-dominated state. Due to the hysteresis effect, these ponds will not go back to an Fe-dominated state when the temporary additional input of organic matter ceases but will tend to remain in the S-dominated state. As a result, one obtains a marsh where all ponds receive the same input of organic matter but show a redox dichotomy; some ponds will still have the Fe-rich state, while others have irreversibly switched to an S-rich state ( 20 – 22 ). The S-rich ponds can only switch back to an Fe-rich state when the organic matter input to the ponds decreases enough to pass the left tipping point of the bistability zone. The important consequence is that the observed redox dichotomy in the salt marsh sediments may not reflect present differences in boundary conditions but may instead be path dependent. Generalization to Marine Sediments. The phenomenon of Fe/S bistability is not restricted to salt marsh sediments. We demonstrate this in a generalized reactive transport model of a marine sediment ( SI Appendix , Materials and Methods , section 3 ). The model includes a set of transport and reaction processes common to models that have been previously used to describe early diagenesis in a variety of marine sediment environments ( 24 – 28 ). The model is run over a broad range of systematically varied parameter values, and these model simulations provide insight into the prevalence of Fe/S bistability ( SI Appendix , Figs. S2 and S3 ). One crucial constraint is that sediments receive a sufficient input of organic carbon and iron oxides so that they become anoxic, with dissimilatory iron and sulfate reduction becoming competing mineralization pathways. Slope and deep-sea sediments in the modern ocean are generally oxic and dominated by aerobic respiration ( 29 ) and hence are not prone to regime shifts. In contrast, the Fe/S bistability may be an inherent feature of many modern coastal and shelf sediments and generally occurs around a value of F CH2O /F FeOOH ∼15 ( SI Appendix , Fig. S3 ). A compilation of the F CH2O /F FeOOH ratios for several coastal and shelf sediments shows that these sediments indeed fall close to or within the model-predicted bistability ranges ( SI Appendix , Table S10 ). We find that the width of the bistability zone becomes smaller for higher rates of sediment mixing by marine infauna, which suggests that Fe/S bistability could have been more prevalent in shallow seafloor environments without benthic fauna, as was the case before the Phanerozoic. Similarly, the hysteresis decreases for increasing sedimentation rates, as higher sedimentation will increase the burial of FeS minerals, thereby promoting the S-rich state. A wide bistability zone is generally present at sedimentation rates below ∼0.2 cm/yr −1 ( SI Appendix , Fig. S3 ). This transport regime applies to most of the present-day coastal seafloor, apart from regions with high accumulation like the deltaic mobile muds of the Amazon delta ( 30 ). While coastal and shelf sediments appear susceptible to bistability, the existence of hysteresis may explain why the phenomenon has not previously been identified in field studies. Sites residing within the middle of the bistability zone would tend to be “locked” in an Fe-rich or S-rich state, and revisiting the same site multiple times would yield the same observed state. Only when a sediment system is poised at the edge of the bistability zone and external boundary conditions (like organic matter delivery) show suitable spatial or temporal variation will one observe both Fe-rich and S-rich states within a given area. This particular situation is likely not often encountered in the field but could be the case in the East Anglian salt marsh systems, as discussed above. Another potential location is the Arvadi Springs, where S-rich and Fe-rich sediments are found to coexist on spatial scales of <1 m, without apparent differences in organic matter input ( 31 ). The bistability in Fe and S cycling has important ramifications for studies on modern marine sediment geochemistry as well as for the interpretation of the geologic record. Firstly, as a consequence of the existence of hysteresis, the particular geochemical state observed in a sediment may depend on the history of the system. A second ramification is that one-to-one mapping between observed concentrations in the sediment and inferred input fluxes breaks down within the bistability interval, as a system can be either ferruginous or euxinic for the same F CH2O /F FeOOH ratio. Therefore, in ancient deposits, one cannot make strong inferences about the influx of organic matter solely based on sedimentary redox proxies. Sediments with ferruginous and euxinic pore waters may underly water columns with nearly the same productivity. Extension to Ocean Geochemical Dynamics. To explore whether the Fe/S bistability can also be present in natural systems other than sediments, we equipped a three-box ocean model ( 16 , 32 ) with a set of biogeochemical reactions describing the cycling of carbon, iron, and sulfur ( SI Appendix , Materials and Methods , section 4 and Fig. S4 ). The model was parameterized to represent the present-day ocean in terms of large-scale circulation, particle sinking fluxes, and export production. We dynamically ran this ocean model to steady state for different values of ocean export production (J CH2O ; this forms the counterpart of F CH2O in the sediment model) and O 2 partial pressures in the atmosphere, keeping the weathering fluxes of reactive iron oxides and sulfate to the ocean constant. While the spatial scales and transport mechanisms are very different compared to sediments, the ocean model consistently exhibited comparable redox bistability ( SI Appendix , Fig. S5 ). Furthermore, for increasing (and then decreasing) J CH2O values, the model output traces out a similar hysteresis loop ( Fig. 4 A ). Low values of J CH2O lead to ferruginous conditions, which can switch to euxinic conditions when J CH2O is increased past a tipping point. Moreover, we find that there is a coherent scaling in the location of the bistable region as a function of atmospheric p O 2 ( Fig. 4 ). For example, for an atmospheric O 2 concentration of 0.1 present atmospheric level (PAL), the bistability zone emerges between 45 and 47% of present-day oceanic export production, while for 0.01 PAL, the bistability zone arises at lower productivity, between 5 and 6% of present-day oceanic export production ( Fig. 4 A and B ). Fig. 4. Bistability in the ocean interior. ( A ) A sensitivity analysis of a three-box ocean model. The steady-state model results are displayed for an atmospheric O 2 concentration of 0.01 PAL and 0.1 PAL. A bistability zone (gray zone) appears between 0.05 and 0.06 present-day level (PDL) and 0.45 and 0.47 PDL of export production, respectively. The arrows indicate the direction of the hysteresis loop. ( B ) The location of the bistability zone (thin black wedge) for different p O 2 concentrations. The dashed vertical line on the right indicates the p O 2 for which the deep ocean box becomes oxic. The dashed diagonal line shows the minimum ratio of export production (EP) over p O 2 consistent with an anoxic deep ocean as derived from a three-dimensional ocean-based model (cGEnIE) ( 15 ). The gray symbols show the median (circles) and 90% credible intervals (error bars) on export production estimated from an intermediate-complexity ocean biogeochemistry model ( 33 ). The results are shown for both low-O 2 (closed) and high-O 2 (open) scenarios. The black symbols show the median (circles) and 90% credible intervals (error bars) on export production based on the oxygen isotope composition of evaporitic sulfate minerals ( 34 ), assuming 20% of the gross primary production is exported from the photic zone ( 48 ). The results are shown for both low-O 2 (closed) and high-O 2 (open) scenarios. Our ocean box model predicts a much narrower bistability zone (gray zone in Fig. 4 A ; thin black wedge in Fig. 4 B ) than exhibited by a sediment system. In the sediment, the zone of hysteresis becomes smaller when vertical transport (sedimentation) increases ( SI Appendix , Fig. S3 ). In the oceanic water column, the residence time of sinking particles is much shorter (∼10 to 100 d) compared to shelf sediments (>50 y). This strong vertical transport partly explains why, in an oceanic setting, the Fe/S redox bistability falls within a significantly more restricted region of the parameter space. Importantly, the narrow bistability zone implies that oceanic systems are more prone to observable redox shifts than sedimentary systems. Small changes in oceanic export production, on the order of just a few percent, may shift the deep ocean back and forth across the narrow bistability zone, thus inducing repeated switches between ferruginous and euxinic states. The occurrence of oceanic regime shifts crucially depends on the export production at any given partial pressure of atmospheric oxygen ( Fig. 4 B ). Interestingly, recent estimates for productivity and atmospheric oxygen during the Proterozoic ( 33 , 34 ) suggest that the ocean interior could have been within or near the bistable region for much of Proterozoic time ( Fig. 4 B ), which may explain the alternation between ferruginous and euxinic states observed in the rock record. Further model analysis shows that the key parameters determining the location and width of the bistability zone are the strength of ocean circulation (mixing) and the rate at which particles sink in the ocean; the bistability zone becomes larger with increasing strength of ocean circulation and decreasing particle sinking fluxes ( SI Appendix , Fig. S5 ). These parameters are difficult to constrain through time but will have varied throughout Earth’s history as a consequence of plate tectonics and biological evolution (see, e.g., refs. 35 – 37 ). Interestingly, at atmospheric oxygen partial pressures equivalent to 0.1 PAL, oceanic sulfate abundance has only a negligible influence on the location and width of the bistable region. The bistability zone appears at the same level of export production for oceanic sulfate concentrations that range from a few micromolar to millimolar levels ( SI Appendix , Fig. S6 ). This implies that transitions from ferruginous to euxinic conditions need not be linked solely to variations in sulfate runoff from terrestrial weathering. However, at lower atmospheric oxygen levels, the bistability zone disappears for sulfate levels of a few micromolars, which could suggest that the Fe/S bistability was less important in the Archean oceans ( 38 , 39 ). Iron Sulfide Formation Drives Fe/S Bistability. Finally, to illustrate how we can confidently make the stretch from modern observations in salt marsh sediments to potential Precambrian ocean states and transitions, we need to expose the core mechanism that causes redox bistability. To this end, we reduced the sediment and ocean system to its barest biogeochemical essentials and reformulated transport as a zero dimensional (0D) chemostat-type description, while the reaction set was progressively simplified to eventually only retain those reactions that generate the Fe/S bistability ( SI Appendix , Materials and Methods , section 5 ). These key interactions include the following: 1) the oxidation of organic matter by dissimilatory iron and sulfate reduction, 2) the reoxidation of Fe 2+ and H 2 S by oxygen, and 3) the precipitation of FeS. These reactions are common to all ocean and sediment models examined above. After suitable reformulation, we were able to condense the 0D chemostat model into a single ordinary differential equation that describes the dynamic behavior of the reactive iron oxide concentration (C FeOOH ), supplemented with algebraic equations for the concentrations of Fe 2+ and H 2 S. Overall, this highly simplified and abstract model enables classical linear stability analysis ( Fig. 5 ) and generates a very similar dynamical response as the more complex models analyzed above. For low F CH2O inputs, there is only one single ferruginous stable state (high C FeOOH ; low C HS− ), while at higher F CH2O inputs, there is a single euxinic state (low C FeOOH ; high C HS− ). In between, there is a bistability zone that shows hysteresis induced by two saddle-point bifurcations, which is the archetypal fingerprint of a dynamical system exhibiting regime shifts ( Fig. 5 A and B ). Most importantly, this abstract model demonstrates that FeS formation is the key reaction responsible for the positive feedback that generates the bistability ( SI Appendix , Materials and Methods , section 5 and Figs. S7 and S8 ). In a reaction system where FeS precipitation is hypothetically inhibited, there is only a single, Fe-rich stable state ( Fig. 5 C ). However, when FeS formation is kinetically fast, as is the case in natural environments, the reaction system exhibits two alternative stable states ( Fig. 5 D ): the Fe-rich (C FeOOH ∼6, Fig. 5 D ) and S-rich (C FeOOH ∼0, Fig. 5 D ) states described above. Our analysis thus demonstrates that the Fe/S bistability is uniquely associated with the occurrence of FeS formation, and its occurrence is not fundamentally influenced by the inclusion of other Fe and S reactions in the reaction set. These other Fe and S reactions do not determine the presence of the bistability but only modulate the shape of the stability zone. This also explains why the Fe/S bistability is generically present in all ocean and sediment models examined. Fig. 5. A mathematical analysis of bistability based on an abstract 0D model that includes reactions common to all complex ocean and sediment models previously examined ( SI Appendix , Materials and Methods , section 5 ). ( A ) The steady-state FeOOH concentrations (C FeOOH ) versus the ratio of organic matter influx (F CH2O ) and iron (oxyhydr)oxide influx (F FeOOH ). ( B ) The steady-state concentration of dissolved sulfide (C HS− ) versus the ratio of organic matter influx and iron (oxyhydr)oxide influx. ( C ) A phase plot showing the relation between the rate of change of oxidized iron (dC FeOOH /dt) and the concentration of oxidized iron (C FeOOH ) when FeS precipitation is negligible compared to reoxidation of ferrous iron and dissolved sulfide. The system shows a single stable state. ( D ) A similar phase plot when FeS precipitation is much faster than reoxidation of ferrous iron and dissolved sulfide. The system shows two separate stable states and one unstable state. C and D show results for F CH2O /F FeOOH ratio of 4. Concentrations in C and D are in dimensionless form."
} | 7,700 |
37723228 | PMC10507100 | pmc | 7,093 | {
"abstract": "Crop rotation is an important agricultural practice for homeostatic crop cultivation. Here, we applied high-throughput sequencing of ribosomal RNA gene amplicons to investigate soil biota in two fields of central Japan with different histories of maize–cabbage rotation. We identified 3086 eukaryotic and 17,069 prokaryotic sequence variants (SVs) from soil samples from two fields rotating two crops at three different growth stages. The eukaryotic and prokaryotic communities in the four sample groups of two crops and two fields were clearly distinguished using β-diversity analysis. Redundancy analysis showed the relationships of the communities in the fields to pH and nutrient, humus, and/or water content. The complexity of eukaryotic and prokaryotic networks was apparently higher in the cabbage-cultivated soils than those in the maize-cultivated soils. The node SVs (nSVs) of the networks were mainly derived from two eukaryotic phyla: Ascomycota and Cercozoa, and four prokaryotic phyla: Pseudomonadota, Acidobacteriota, Actinomycetota, and Gemmatimonadota. The networks were complexed by cropping from maize to cabbage, suggesting the formation of a flexible network under crop rotation. Ten out of the 16 eukaryotic nSVs were specifically found in the cabbage-cultivated soils were derived from protists, indicating the potential contribution of protists to the formation of complex eukaryotic networks.",
"introduction": "Introduction Soils are complex and invaluable terrestrial media. Soil organisms, such as bacteria, fungi, or protists, play crucial roles in nutrient cycles in terrestrial ecosystems 1 and their communities are influenced by soil environments and plants. In particular, the potential interactions of soil organisms with crops cultivated in agricultural fields can affect the growth and health of crops via microorganism-derived nutrients and plant pathogens 2 , 3 . Therefore, quantitative data of prokaryotic and eukaryotic communities and their taxonomic changes during crop cultivation are useful for monitoring and accessing the biological environment of agricultural soils. High-throughput sequencing of amplicons derived from the 16S ribosomal RNA (rRNA) gene and 18S rRNA gene cluster (i.e., DNA metabarcoding) is a powerful tool for analyzing soil prokaryotic and eukaryotic communities (mainly bacteria and fungi) in agricultural soils because of the crucial roles of these organisms in the pedosphere 4 . Previous DNA metabarcodings clarified the taxonomic compositions of major soil organisms such as bacteria, fungi, and protists in agricultural areas as well as their changes with fertility, tillage, and other types of agricultural practices. Crop rotation has been widely used in crop cultivation to suppress plant diseases and replant failure by changing crops in cultivation cycles to avoid continuous cropping 3 . Several studies using DNA metabarcoding have been reported for soil organisms living in agricultural fields under crop rotation with different crops and cropping sequences, cycles, and periods 5 – 33 and with different agricultural management systems, such as tillage and fertility. However, most of these studies were focused on soil bacteria and fungi and details of whole soil organisms, especially nonfungal eukaryotes, are poorly understood in crop-rotation fields. We assumed that crop rotation could also influence soil nonfungal eukaryotes, and investigated both the eukaryotic and prokaryotic communities in the field soils under crop rotation in this study, to verify the hypothesis. We previously applied 18S rRNA gene-derived amplicon sequencing using Illumina MiSeq to analyze soil nematodes and successfully clarified the nematode communities in sweet potato-cultivated fields 34 . Herein, we applied DNA metabarcoding to investigate both prokaryotes and eukaryotes living in two fields (i.e., field_1 and field_2) cropping maize and cabbage by rotation in central Japan in 2019. The maize–cabbage rotation was performed to prevent clubroot diseases of cabbage 35 in both fields in this year, although each field had a different history of agricultural management in the previous year: field_1 was managed as fallow and was treated with green manure, and field_2 underwent the maize–cabbage rotation in 2018. Thirty-six sample soils from two fields cultivating two crops at three different growth stages were analyzed, and unique sequence variants (SVs) of the 16S and 18S rRNA genes were identified. We then investigated the prokaryotic and eukaryotic taxonomic compositions of the SVs, analyzed the α- and β-diversities of the communities in the samples, and assessed the sample-soil chemical parameter relationships via redundancy analysis (RDA). Finally, we characterized the networks of prokaryotic and eukaryotic SVs in four sample groups (i.e., maize- and cabbage-cropping soils in two fields). Using these analyses, we clarified the biological features of two agricultural fields cropping maize and cabbage by rotation with different management history.",
"discussion": "Discussion Several DNA metabarcoding studies have analyzed soil bacterial and fungal communities in agricultural lands under various types of crop rotation, but understanding remains poor concerning other eukaryotes. In addition, few studies have assessed the soil biota under poaceous crops-brassicaceous crop rotation 7 , 14 , 23 , 24 , 31 . Therefore, we investigated both eukaryotes and prokaryotes and their networks formed in two agricultural fields in central Japan under the 1st and 2nd cycles of maize–cabbage rotation. The major phyla of SVs found in our study were Ascomycota, Basidiomycota, and Cercozoa in eukaryotes and Pseudomonadota, Acidobacteriota, and Actinomycetota in prokaryotes (Fig. 2 ). These fungal and bacterial phyla have been commonly detected in other studies on crop rotations 9 , 12 , 18 , 22 , 24 , 25 , 27 , 32 , 33 , 36 , 37 . We found different bacterial phylum compositions between two field soils, where the relative abundance of the Actinomycetota phylum in field_1 soils was clearly higher than that in field_2 soils (Fig. 2 b). The abundant Actinomycetota-derived nSVs were consistently found in the prokaryotic networks (Fig. 5 m,n). Previous studies showed that the abundance of bacteria in Actinomycetota is increased by treatment with green manure 38 – 40 , and Tao et al. also reported that green manure fertilization altered the topological properties of microbial networks 41 . These studies are consistent with our observations because of the green manure supplied to field_1 in the previous year. Several families were unequally distributed according to field and crop, and their relative abundances were changed by cabbage cultivation (Supplementary Table S1 ). Fungal Mrakiaceae and protist Euglyphida families were more abundant in field_2 vs. field_1 and their relative abundances were differently changed by crop rotation to cabbage. The biased relative abundance of some families was likely accounted for by the different contents of chemical factors in the fields; for example, the Mrakiaceae (SV_2) and Sphingomonadaceae families, which were more abundant in field_2 vs. the other field, were strongly associated with water content in RDA, which was high in field_2 (Fig. 4 a, Supplementary Table S4 ). The prokaryotic families and genera for which the relative abundances were increased by crop rotation in both fields, such as family Vicinamibacteraceae and field_2 families Nitrosomonadaceae and Pyrinomonadaceae and genus RB41 , contained cabbage- and field_2-specific nSVs, respectively. The Chitinophagaceae and Rhodanobacteraceae families and genus Nocardioides , the abundances of which were decreased in both fields, tended to have maize-specific nSVs. These results suggest the involvement of those nSVs in the changes of networks triggered by cabbage cultivation. The maize–cabbage rotation in the fields is used to suppress serious clubroot diseases of cabbage 35 . We identified the SV_97 derived from P. brassicae , which is a pathogenic protist causing clubroot diseases 42 . SV_97 was abundantly detected in the maize-cultivated field_2 soils under the 2nd rotation cycle in 2019, where cabbage was planted in 2018. The relative abundance of SV_97 gradually decreased during maize growth and was significantly reduced in the cabbage-cultivated field_2 soils. These results indicate that P. brassicae that propagated during cabbage cultivation in 2018 was detected in the maize-cultivated soils in 2019, and further expansion of the pathogen was likely suppressed by the practices applied during the maize harvest, i.e., treatments with anti-clubroot disease agent flusulfamide, chemical fertilizer supply, and tillage with maize residues. In addition to effective flusulfamide treatment, precropped maize may contribute to reduce the density of pathogens in the cabbage-cultivated field_2 soils by trapping P. brassicae with roots because maize does not exhibit a clubroot phenotype. Changes in soil microbial diversity by crop rotation (Fig. 3 b) may also influence the suppression of clubroot disease via modulation of the pathogen transcriptomes, as reported by Daval et al. 43 . Regarding the communities in soils under crop rotation, many studies have indicated distinct β-diversities of soil bacteria and/or fungi between mono- and rotation-cropping 5 , 6 , 8 , 10 , 12 , 14 – 18 , 20 – 22 , 24 , 25 , 27 , 33 or among crop rotations with different crops and rotation sequences, places, and/or periods 5 , 10 , 11 , 15 , 18 – 22 , 24 , 25 , 32 , 33 ; however, the details of soil biota that change with plant growth have not been clarified in a crop-rotation cycle. We showed that the taxonomic variations indicated by β-diversities in the four sample groups were clearly distinguished by crops as well as by fields (Fig. 3 ), suggesting the presence of unique eukaryotic and prokaryotic communities in each soil. This result is unsurprising because microbial communities including fungi are known to be influenced by crops and land use history, including crop rotations. This is also the case with eukaryotic communities; for instance, protist communities were changed by fertilization 44 and depth 45 , and unique communities of eukaryotes 46 – 49 , including protists 50 – 53 and nematodes 54 , 55 , were formed in different types of soils, including agricultural soils. Furthermore, the taxonomic variations in β-diversities were almost comparable among the three different growth stages of crops except for prokaryotic variations in field_2 soils cropping maize at the early stage (Fig. 3 ), suggesting that the communities in bulk soils near crops are largely unaffected by plant growth. We also found higher Shannon indexes of eukaryotes for α-diversity in the cabbage-cultivated field soils vs. maize-cultivated soils (Supplementary Fig. S3 a), which may reflect the appearance or propagation of additional species in eukaryotic communities via the cultivation of crops of different families. In particular, the low Shannon index in field_1 soils significantly increased to levels comparable with those in field_2 soils after crop rotation to cabbage (Supplementary Table S2 ). This observation may be accounted for by a study reporting a higher OTU richness of fungi and protists in agricultural fields compared with grasslands and woods 47 : because of fallow in field_1 in the previous year, the Shannon index in the maize-cultivated field_1 soils was low, and then increased with subsequent cabbage cultivation. Conversely, the prokaryotic α-diversities in field_1 soils cropping maize were high and decreased to the levels of those in field_2 soils with subsequent cabbage cultivation (Supplementary Fig. S3 b). This could also be explained by the previous observation reported by Woo and colleagues, who showed that the Shannon index of bacteria in fallow was highest compared with those observed in mono- and rotationally cropped fields with pea and wheat 28 . These data suggest that the distinct communities in the two field soils were formed by different agricultural practices (i.e., fallow with green manure and rotation cropping) before maize cultivation and that a rapid change in eukaryotic and prokaryotic communities in field_1 soils was caused by the first cropping. RDA revealed that the eukaryotes and prokaryotes in field_2 soils were closely related to nutrient ions and phosphorus content, and those in the maize-cultivated field_2 soils to water content (Fig. 4 ). Despite minor differences among investigations, previous studies on crop rotations, including maize or cabbage, showed that soil pH and nutrients (mainly nitrogen) often affect bacterial and/or fungal communities 12 , 17 , 18 , 20 , 23 , 25 , 33 , 36 , 56 . The higher content of these nutrients and of water in field_2 compared with those in field_1 may account for the above relationships (Fig. 4 a). In field_1, the communities in the maize-cultivated soils were related to the humus content, which was likely derived from green manure supplied in the previous year. Regarding the relationship to crops, the communities in the maize-cultivated soils were associated with ammonium and nitrite nitrogen content, and electrical conductivity, and those in the cabbage-cultivated soils to soil pH. Similar RDA profiles of both prokaryotes and eukaryotes were obtained, indicating that the eukaryotic communities are affected by soil chemical factors in a similar manner to prokaryotic communities. RDA identified particular families and major SV groups that were strongly associated with each chemical parameter (Supplementary Table S4 ). For instance, cabbage- and Cercozoa-derived SVs and maize-, Ascomycota- and Nematode-derived SVs were associated with pH and nitrate nitrogen, respectively, indicating that each chemical factor affected each crop and SV group. Moreover, many cabbage-specific nSVs were found in nonplant SVs associated with pH (5 of 11), suggesting that soil pH contributes to the formation of complex eukaryotic networks in cabbage-cultivated soils as described later. Similarly, many eukaryotic and prokaryotic nSVs among the SVs associated with nutrient ions were field_2-specific (5 out of 11 and 11 out of 25 SVs, respectively), and four field_2-specific nSVs out of six prokaryotic SVs were associated with water content. Because of the high content of nutrient ions and water in field_2, these factors may have influenced the nSVs during network formation in field_2. Finally, we investigated the prokaryotic and eukaryotic networks in each sample group. Notably, networks where poorly formed with small numbers of nSVs and links in the maize-cultivated field_1 soils where the first rotation cycle was initiated compared with those in field_2 soils under the second rotation cycle (Fig. 5 , Supplementary Table S6 ), suggesting distinct communities in two field soils under maize cultivation as mentioned above. Under cabbage cultivation, complex networks were formed in field_1 soils as well as in field_2 soils. Furthermore, despite a few core nSVs being shared by the networks in the crop-cultivated or the field soils, most of the eukaryotic and prokaryotic core nSVs were not conserved in the networks throughout the maize–cabbage crop rotation in each field (Supplementary Table S8 ). These observations indicate distinct networks in the maize- and cabbage-cultivated soils in each field and are consistent with the β-diversity results (Fig. 3 ), suggesting that the communities and networks in the field soils were easily changed and affected by cropping. Several studies have shown that rotation cropping produces more complex networks (mainly bacteria and fungi) than monoculture does 12 , 13 , 18 and that the microbial networks are distinct among different crop rotations 19 . Crops and agricultural management practices such as fertilization and green manures also affect microbial communities and their networks 41 , 57 – 62 . Xiong et al. showed a change in microbial networks along with maize developmental stages 63 . Xie et al. reported different microbial networks in each crop under wheat–rice rotation as we observed 64 . These studies suggest that the networks of soil organisms are flexibly reformed by coupling with the changes in soil biota upon crop cultivation and/or soil environmental changes and agree with our observations. Our data suggest the increase in the complexity of networks by cropping with cabbage. Based on previous studies, subsequent cropping with plants in different families (maize in Poaceae and cabbage in Brassicaceae ) and/or tillage with maize residues may contribute to the increased complexity of networks triggered by cabbage cultivation via newly generated plant–soil organism interactions. We identified 13 eukaryotic and 7 prokaryotic common nSVs shared by the networks of 4 sample groups (Supplementary Tables S10 and S11 ), and these Ascomycota- or Cercozoa-derived and Pseudomonadota-derived common nSVs likely have potential roles in forming core networks. We also identified core nSVs for nSVs because of the crucial role of hubs in the network. Few crop- or field-specific eukaryotic core nSVs were present in the nSVs; however, five eukaryotic SVs (SV_2, 4, 28, 43, and 63) were not only common nSVs but also core nSVs in more than two networks (Supplementary Table S10 ). Two abundant SVs (SV_2 and SV_4) were derived from the Basidiomycota Tausonia and Solicoccozyma genera that respectively contains enzyme-producing yeasts 65 and plant-growth promoting microorganisms 66 . SV_28, SV_43, and SV_63 were assigned to the Spongomonas and Heteromita genera in Cercozoa and the Colpoda genus in Ciliophora respectively. Heteromita is an abundant flagellate 67 , Spongomonas and Heteromita are bacterivorous protists 68 , Colpoda is a common free-living terrestrial protist, and Colpoda cucullus has been reported to improve maize growth 69 . These common SVs could act as core hubs of eukaryotic networks in both fields throughout cropping. By contrast, only one of the seven prokaryotic common nSVs (SV_23) was a core nSV in the two networks. Furthermore, although six core nSVs (SV_2, 4, 28, 43, 63, and 84) were conserved in the eukaryotic networks of the maize- or cabbage-cultivated soils, no conserved core nSVs were found in the corresponding prokaryotic networks (Supplementary Table S8 ). This may suggest that networks of soil prokaryotes are more unstable than those of eukaryotes under crop rotation. Protists are an important group of soil eukaryotes 1 and are involved in nutrient cycles in the pedosphere and in plant growth and health in agricultural lands 70 ; however, protists have not been well characterized compared with soil bacteria or fungi. Notably, the numbers of core nSVs, especially protist (phyla Cercozoa and Ciliophora)-derived core nSVs in this study, were markedly increased in the eukaryotic networks under cropping cabbage (Supplementary Table S9 ). Several studies have been conducted on nonfungal eukaryotic networks, including those of protists 52 , 53 , 59 , 71 and nematodes 72 , in agricultural soils, and some of these demonstrated the ecological importance of bacteria-fungi-protist networks in the rhizosphere 52 , 59 , of fungi-protist interactions in paddy soils 71 , and of protist networks in arable soils 53 . Our observations also showed that protists helped increase the complexity of eukaryotic networks in the cabbage-cultivated soils. Protists are drivers for modifying soil microbiomes, and it is therefore important to clarify the taxonomic compositions of whole soil organisms and dynamic changes in their networks for advanced crop cultivation in the future. In conclusion, we have showed that both chemical parameters and crops affect eukaryotic and prokaryotic communities and clarified the structures and changes in eukaryotic and prokaryotic networks under maize–cabbage crop rotation. We have newly demonstrated the involvement of protists in eukaryotic network formation."
} | 5,022 |
37537795 | PMC10832534 | pmc | 7,094 | {
"abstract": "Abstract As one of the main precursors, acetyl‐CoA leads to the predominant production of even‐chain products. From an industrial biotechnology perspective, extending the acyl‐CoA portfolio of a cell factory is vital to producing industrial relevant odd‐chain alcohols, acids, ketones and polyketides. The bioproduction of odd‐chain molecules can be facilitated by incorporating propionyl‐CoA into the metabolic network. The shortest pathway for propionyl‐CoA production, which relies on succinyl‐CoA catabolism encoded by the sleeping beauty mutase operon, was evaluated in Pseudomonas taiwanensis VLB120. A single genomic copy of the sleeping beauty mutase genes scpA , argK and scpB combined with the deletion of the methylcitrate synthase PVLB_08385 was sufficient to observe propionyl‐CoA accumulation in this Pseudomonas . The chassis' capability for odd‐chain product synthesis was assessed by expressing an acyl‐CoA hydrolase, which enabled propionate synthesis. Three fed‐batch strategies during bioreactor fermentations were benchmarked for propionate production, in which a maximal propionate titre of 2.8 g L −1 was achieved. Considering that the fermentations were carried out in mineral salt medium under aerobic conditions and that a single genome copy drove propionyl‐CoA production, this result highlights the potential of Pseudomonas to produce propionyl‐CoA derived, odd‐chain products.",
"conclusion": "CONCLUSION This study presents the engineering of the propionyl‐CoA synthesis capabilities into P. taiwanensis VLB120 and demonstrated the capacity of this strain to produce products derived from propionyl‐CoA from glucose as sole carbon source using propionate as a relevant example. A crucial step to produce this organic acid was the deletion of the methylcitrate synthase, encoded by PVLB_08385, which enabled propionyl‐CoA accumulation in a strain expressing a single genomic integrated copy of the sleeping beauty mutase operon. The capability of this strain to produce propionate when co‐expressing an acyl‐CoA hydrolase was assessed in bioreactor fermentations and showed robust production performance under different fed‐batch schemes. Besides robust behaviour, the titres achieved are promising, considering that production was achieved with a minimal engineering approach, including a single‐copy genome integration of the sleepy beauty mutase operon. Evaluating P. taiwanensis under stress conditions to increase TCA cycle activity and optimising the expression of the sleepy beauty mutase and acyl‐CoA hydrolase should be considered in future work. Nevertheless, this initial work showed the promising and expandable potential of Pseudomonas to produce propionyl‐CoA‐derived products.",
"introduction": "INTRODUCTION Coenzyme A (CoA) activated carboxylic acids play a crucial role in microbial metabolism by acting as a carrier of reactive acyl groups and facilitating enzyme recognition (Brass, 1994 ). The study of CoA‐activated carboxylic acids in bacteria has significantly evolved in microbial biotechnology, shaping a crucial research area with immense importance. Coenzyme A (CoA)‐activated carboxylic acids serve as fundamental building blocks for various metabolic pathways, including the production of biofuels, pharmaceuticals and fine chemicals (Murali et al., 2017 ; Vila‐Santa et al., 2021 ). These advancements have been made possible through the identification and characterisation of key enzymes involved in CoA activation, pathway engineering strategies, and the optimisation of bacterial hosts (Zhu et al., 2022 ). The shortest acyl‐CoA, acetyl‐CoA, is a vital metabolite in the carbon metabolism of all living systems and essential for fuelling the TCA cycle and fatty acid biosynthesis (Krivoruchko et al., 2015 ). All major workhorses in the metabolic engineering field are confined to dependency on acetyl‐CoA, leading to the production of even‐chain products. The production of odd‐chain products requires the incorporation of propionyl‐CoA into the metabolic network. Industrial‐relevant products that rely on propionyl‐CoA are odd‐chain alcohols, acids, ketones and polyketides such as 1‐propanol (Srirangan et al., 2013 ), 1‐pentanol (Tseng & Prather, 2012 ), propionic acid (Akawi et al., 2015 ), valeric acid (Tseng & Prather, 2012 ), 2‐butanone (Srirangan et al., 2016 ), 6‐deoxyerythronolide B (Pfeifer et al., 2001 ) or bio(co)polymers such as poly(3‐hydroxybutyrate‐ co ‐3‐hydroxyvalerate) (Ibrahim et al., 2021 ). Incorporating propionyl‐CoA into a metabolism lacking this acyl‐CoA, like Escherichia coli 's metabolism, can be achieved, for instance, by feeding propionate or through fatty acid activation/degradation by adding odd‐chain fatty acids (Dellomonaco et al., 2010 ). Besides these feeding strategies, three de novo synthesis pathways of propionyl‐CoA have been reported, a catabolic pathway of succinyl‐CoA native to E. coli (Gonzalez‐Garcia et al., 2020 ; Haller et al., 2000 ), a heterologous catabolic pathway of 2‐ketobutyrate (Jun Choi et al., 2012 ), and the 3‐hydroxypropionate cycle from thermoacidophilic crenarchae Metallosphaera sedula and Sulfolobus tokodaii (Yuzawa et al., 2012 ) (Figure 1 ). FIGURE 1 Strategies for propionyl‐CoA production: addition of odd‐chain fatty acids and propionate to the cultivation medium (blue); catabolic pathway of 2‐ketobutyrate (red); catabolic pathway of succinyl‐CoA (yellow) employing the sleeping beauty mutase, and the 3‐hydroxypropionate cycle from Metallosphaera sedula and Sulfolobus tokodaii (green). Asp, aspartate; AcCoA, acetyl‐CoA; AspP, aspartyl‐phosphate; AspAld, aspartate‐semialdehyde; HoSer, homoserine; HoSerP, homoserine‐phosphate; Threo, threonine; 2‐ketobut, 2‐ketobutyrate; PropP, propionyl‐phosphate; Prop, propionate; PropCoA, propionyl‐CoA; SucCoA, succinyl‐CoA; MmCoA, methylmalonyl‐CoA; MalCoA, malonyl‐CoA; MalAld, malonic semialdehyde; 3‐HP, 3‐hydroxypropionate; 3‐HPCoA, 3‐hydroxypropionyl‐CoA; AcrCoA, acryolyl‐CoA; CimA, citramalate synthase; AspC, aspartate aminotransferase; MetL, aspartokinase/homoserine dehydrogenase 2; Asd, aspartate‐semialdehyde dehydrogenase; ThrA, aspartokinase/homoserine dehydrogenase 1; ThrB, homoserine kinase; ThrC, threonine synthase; IlvA, threonine dehydratase; AckA, acetate kinase; AtoA, acetate CoA‐transferase subunit B; AtoD, acetate CoA‐transferase subunit A; FadA, 3‐ketoacyl‐CoA thiolase; FadB, fatty acid oxidation complex subunit alpha; FadE, acyl‐CoA dehydrogenase; FadD, long chain fatty acid‐CoA ligase; ScpA, methylmalonyl‐CoA mutase; ScpB, methylmalonyl‐CoA decarboxylase; Mcr, malonyl‐CoA reductase; Msr, malonic semialdehyde reductase; 3HPCS, 3‐hydroxypropionyl‐CoA synthase; 3HPCD, 3‐hydroxypropionyl‐CoA dehydratase and Acr, acryloyl‐CoA reductase; PVLB_08385, 2‐methylcitrate synthase; Mcit, 2‐methylcitrate. The three de novo synthesis pathways for propionyl‐CoA production diverge from different points in the central carbon metabolism and differ in length. The catabolic succinyl‐CoA pathway catalysed by E. coli 's sleeping beauty mutase ( sbm ) requires only two reactions and is the shortest of the three alternatives. The sbm operon has been successfully used to produce several propionyl‐CoA dependent products by either activation of the operon on the genome (Akawi et al., 2015 ; Srirangan et al., 2014 , 2016 ) or by plasmid‐based overexpression (Gonzalez‐Garcia, McCubbin, Wille, et al., 2017 ; Li et al., 2017 ; Srirangan et al., 2013 ). The sbm operon of E. coli MG1655 comprises four genes encoding a methylmalonyl‐CoA mutase ( scpA ), a membrane‐bound ATP kinase ( argK ), a methylmalonyl‐CoA decarboxylase ( scpB ) and a propionyl‐CoA:succinate‐CoA transferase ( scpC ). The function of the ATP kinase ArgK has not been fully elucidated but has been suggested to interact with ScpA (Haller et al., 2000 ) and was described to contribute to enzyme activity (Srirangan et al., 2013 ). Achievements in producing propionyl‐CoA‐derived products through metabolic engineering approaches have been comprehensively reviewed by (Srirangan et al., 2017 ). \n Pseudomonas strains are generally discussed as promising hosts for industrial biotechnology because they harness distinctive features like broad carbon source utilisation, the ability to proliferate in the presence of organic solvents and non‐fermentative growth (Blombach et al., 2022 ; Köhler et al., 2013 ; Poblete‐Castro et al., 2012 ; Rühl et al., 2009 ; Schwanemann et al., 2020 ). This species has been successfully used to produce acetyl‐CoA‐derived products such as rhamnolipids (Müller et al., 2010 ; Tiso et al., 2017 ), methyl ketones (Nies et al., 2020 ), polyhydroxyalkanoates (Cha et al., 2020 ; Mozejko‐Ciesielska et al., 2019 ; Poblete‐Castro et al., 2014 ) and other natural products like terpenoids (Mi et al., 2014 ), polyketides (Gross, Luniak, et al., 2006 ; Martinez et al., 2004 ) and non‐ribosomal peptides (Loeschcke & Thies, 2015 ). Contrarily, only a few examples of products synthesised from propionyl‐CoA exist for this species and are mostly limited to the degradation of branched‐chain amino acids and L‐methionine (Marshall & Sokatch, 1972 ; Mooney et al., 2002 ). Propionyl‐CoA has not been detected in Pseudomonas , indicating low abundance or missing enzymatic capabilities to synthesise this acyl‐CoA species. (Bannerjee et al., 1970 ; Gross, Ring, et al., 2006 ). The potential of metabolic engineering of P. taiwanensis VLB120 into a platform for acyl‐CoA derived products has been recently proven by increasing the supply of malonyl‐CoA for secondary metabolites like pinosylvin, resveratrol and flaviolin (Schwanemann et al., 2023 ). Here, we argue that Pseudomonas could likewise be turned into a superior host for propionyl‐CoA biosynthesis from succinyl‐CoA by harnessing its capability to supply this precursor at a high rate via its highly active TCA cycle flux reported for diverse strains of this species to reach values up to 4‐times higher than in E. coli (Blank et al., 2008 ; Ebert et al., 2011 ; Long et al., 2017 ; Rühl et al., 2009 ). Accordingly, this work aimed to couple this naturally high TCA cycle flux in Pseudomonas with heterologous expression of the sleeping beauty mutase from E. coli to derive a P. taiwanensis VLB120 (Panke et al., 1998 ) propionyl‐CoA chassis and to showcase its potential for odd‐chain length product synthesis on the example of the important building block propionate. This carboxylic acid has a broad application spectrum, including as an FDA‐approved food preservative or herbicide (Zidwick et al., 2013 ) or additive in lacquer formulations and moulding plastics (Álvarez‐Chávez et al., 2012 ), making it an interesting target for this study.",
"discussion": "RESULTS AND DISCUSSION Extending the acyl‐CoA portfolio of P. taiwanensis \n VLB120 by expressing the sleeping beauty mutase The alternative propionyl‐CoA synthesis pathways shown in Figure 1 were found to provide similar theoretical yields from glucose as carbon source (Table 2 ). We opted for the sleeping beauty mutase pathway due to the shorter length of the pathway when compared to alternative biosynthesis routes and the probability of achieving theoretical maximal yields as shown by an in silico flux balance analysis assuming TCA cycle fluxes previously observed for P. putida K2440 (Ebert et al., 2011 ) (Table 2 ). TABLE 2 Theoretical yields of propionate production from glucose for the three alternative propionyl‐CoA pathways expressed in P. taiwanensis VLB120 (Ebert et al., 2011 ). Pathway Theoretical yield [Mol/Mol] \n a \n \n succinyl‐CoA catabolism 1.59 1.59 1.57 2‐ketobutyrate catabolism 1.59 3‐hydroxypropionate cycle 1.57 \n a \n Simulations are based on a genome‐scale metabolic model of P. taiwanensis VLB120 constrained with physiological data taken from (Ebert et al., 2011 ). Three out of the four sbm operon genes ( scpA , argK , and scpB ) were retrieved from the genome of E. coli MG1655 and assembled into a genomic integration construct, which placed this reduced operon under the control of the salicylate inducible nagR /P \n nagAa \n promoter system and incorporated it into the single, neutral att Tn7 site. The genomic integrated construct of the reduced sbm operon was denominated as Tn7 and is represented in Figure 2 . FIGURE 2 Representation of the reduced sbm operon genomically integrated at the single att Tn7 site (Tn7) and the traditional (pT) and optimised (pO) expression cassettes used to express YciA and AarC in the pTN1 plasmid as described by (Neves et al., 2020 ). The fourth gene, scpC , was omitted in the initial strain to avoid a possible loss of propionyl‐CoA by transfer of the CoA group to succinate. Expressing this integration construct without further modification of P. taiwanensis VLB120 led to a marginal accumulation of propionyl‐CoA (Table 3 ). TABLE 3 Propionyl‐CoA levels in the wild‐type strain P. taiwanensis VLB120 (VLB120), the strain harbouring the single genomic integration cassette at the neutral att Tn7 site expressing scpA , argK and scpB under the control of the inducible nagR /P \n nagAa \n promoter (VLB120 Tn7) and, the single genomic integrated construct in the PVLB_08385 knock out strain (VLB120 Δ Tn7). Strain Propionyl‐CoA (nmol/g CDW ) \n a \n \n VLB120 <0.01 VLB120 Tn7 0.02 ± 0.01 \n b \n \n VLB120 Δ Tn7 3.56 ± 1.03 \n b \n \n \n a \n CDW, cell dry weight. \n b \n Errors indicate the standard deviation of biological duplicates. Propionyl‐CoA is an intermediate of the β‐oxidation of odd‐chain fatty acids and amino acid degradation and is metabolised by the methylcitrate cycle, which it enters by condensation with oxaloacetate to 2‐methylcitrate. Deletion of the gene PVLB_08385, which encodes the corresponding methylcitrate synthase, led to a 158‐fold increase in propionyl‐CoA accumulation, demonstrating the vital role of this knockout for the successful construction of a P. taiwanensis VLB120 propionyl‐CoA chassis strain (Table 3 ). Divergent statements regarding the auto sufficiency of the sleeping beauty mutase operon to convert succinyl‐CoA to propionyl‐CoA have risen in the past. Whereas (Gonzalez‐Garcia, McCubbin, Wille, et al., 2017 ) claimed the need for an epimerase to convert the stereoisomers of methylmalonyl‐CoA, Haller et al. showed the conversion of succinyl‐CoA into propionyl‐CoA in in vitro enzymatic assays with methylmalonyl‐CoA mutase (ScpA) and decarboxylase (ScpB) but lacking the epimerase (Haller et al., 2000 ). Gonzalez‐Garcia further alleged that without an epimerase, the production of propionate, and therefore also propionyl‐CoA, is fuelled by the degradation of amino acid provided by the yeast extract in the medium. In the present study, propionyl‐CoA production was achieved in a mineral salt medium with no addition of yeast extract, giving evidence that ScpA and ScpB are sufficient to enable propionyl‐CoA synthesis in P. taiwanensis VLB120, which to our knowledge, does not possess a methylmalonyl‐CoA epimerase. Propionate production in P. taiwanensis \n VLB120 \n Propionate was previously produced in reasonable amounts by knocking out the ack gene in the natural propionate producer Propionibacterium acidipropionici (Suwannakham et al., 2006 ), activating the sleeping beauty mutase operon in E. coli (Akawi et al., 2015 ) or through L‐threonine degradation in E. coli (Jun Choi et al., 2012 ). Further work with the sleeping beauty mutase for propionate production in E. coli was recently published by Miscevic et al., in which the role of each of the TCA metabolic routes was elucidated, and the deletions of ∆s dhA ∆ iclR were shown to be vital to achieve a 30 g L −1 propionate titre in M9 mineral salt medium supplemented with yeast extract (Miscevic et al., 2020 ). Propionate was also produced in high amounts in P. putida KT2440 by the transformation of exogenous L‐threonine to propionyl‐CoA and further to propionate by action of a thioesterase (Ma et al., 2020 ). However, the sleeping beauty mutase pathway has not been evaluated in this species yet. We chose propionate synthesis as a case study to elucidate the potential of the P. taiwanensis VLB120 propionyl‐CoA chassis because of the industrial relevance of this compound and its short production pathway encompassing only a single recombinant enzymatic propionyl‐CoA hydrolysis step. The biocatalytic performance of two alternative enzymes with different catalytic mechanisms was evaluated. The acyl‐CoA hydrolase YciA from Haemophilus influenzae (Uniprot P44886) was chosen because it does not require a CoA acceptor. The second enzyme, AarC, a CoA transferase from Propionibacterium freudenreichii subsp. shermanii (Uniprot A0A160VNK6) is a monomer (Wang, 2013 ), an advantage from an expression burden point of view, and its overexpression in the native host was found to considerably improve propionate production (Wang et al., 2015 ). The enzyme requires succinate as CoA acceptor, which allows CoA recycling into the TCA cycle, potentially leading to a less detrimental effect on the TCA cycle flux. The selection of yciA over the scpC from the sleeping beauty mutase operon of E. coli was based on superior kinetic enzymatic properties of yciA , namely, lower K \n \n M \n (6 against 7.1 μM of ScpC) and superior k \n \n cat \n (25 against 0.72 s −1 of ScpC) (Haller et al., 2000 ; Zhuang et al., 2008 ). Both enzymes were assembled into the pTN1_ nagR _Tra and Opt plasmids (denominated as pT_YciA/AarC and pO_yciA/AarC, respectively) (Neves et al., 2020 ) to evaluate two inducible expression cassettes with different expression levels and transformed into the propionyl‐CoA chassis strain P. taiwanensis VLB120 Δ Tn7. A schematic of the expression cassettes pT and pO is represented in Figure 2 . The pT plasmid expresses the genes from an operon consisting of the inducible promoter and an RBS. In contrast, the pO plasmid contains additionally a ribozyme, the bicistronic design developed by Mutalik et al. (Mutalik et al., 2013 ) and an RNAse III restriction site which was shown to lead to higher expression levels (Neves et al., 2020 ). Preliminary data showed that propionate production only started when cells reached stationary phase in a medium with limiting nitrogen content and excess glucose (data not shown). Therefore, the four constructed strains were evaluated under two growth conditions. One growth medium contained balanced amounts of the carbon (glucose) and nitrogen (ammonium sulfate) source (C:N of 6:1), whereas the second contained only half of the nitrogen amount (C:N of 12:1) to assess if the excess carbon would be converted into propionate. An initial screen of the four strains under these two conditions showed no relevant difference in propionate production between the two enzymes and that the use of the pO plasmid did not achieve higher productivities of propionate, suggesting that hydrolase activity was not limiting (Figure S2 ). The acyl‐CoA hydrolase YciA from Haemophilus influenzae , expressed in the pT plasmid, was the selected enzyme to be evaluated in the bioreactor setup based on the initial screen and the superior enzymatic properties like K \n \n M \n and k \n \n cat \n . Benchmarking of fed‐batch fermentation strategies for propionate production Bioreactor experiments were conducted with the P. taiwanensis VLB120 Δ Tn7 expressing the pT_yciA plasmid to achieve increased propionate titres by performing the cultivations under controlled pH and prolonged nitrogen‐limited conditions. Three fed‐batch strategies were pursued to evaluate the influence of feeding regimes and feast‐famine switches on the performance of the propionate‐producing strain during bioreactor fermentations. Two of these strategies used a feeding script that added 25 mM glucose when the dissolved oxygen (DO) signal increased above 50%. This procedure was repeated until the additional carbon did not lead to a decrease in the DO signal, indicating that the cells were no longer metabolically active. Two initial glucose concentrations of 50 and 100 mM were evaluated. The evaluation of different initial glucose amounts was motivated by the observation that production only started once the nitrogen source was depleted. The two different initial glucose concentrations allowed to evaluate if the presence of carbon source at the point of nitrogen depletion influenced propionate production since the nitrogen present in the MSM B12 media is balanced for the consumption of 50 mM of glucose. The feeding of the third strategy relied on an online feedback‐loop unit, which monitored the glucose concentration inside the bioreactor through an enzymatic‐amperometric module and continuously fed glucose to maintain a concentration of 25 mM throughout the fermentation. In the DO‐controlled fed‐batch strategy, the cells were submitted to stressful feast‐famine switches, which were avoided in the feedback‐loop controlled fed‐batch fermentation with continuous glucose feed. The P. taiwanensis VLB120 Δ Tn7 pT_yciA exhibited similar growth rates and achieved similar propionate titres in all fed‐batch regimes (Table 4 ). Product yields and productivities achieved with the three fed‐batch fermentation strategies were calculated at the beginning of the stationary phase (period A), at which product formation started, and at a later point (period B) to evaluate the robustness of the strain under the different fermentation conditions. The fermentation time intervals of periods A and B for each fermentation were not selected based on absolute fermentation time but rather to represent the desired fermentation stages and to encompass feeding points and carbon source depletion. In all three fermentations, the productivity declined over the course of the fermentation (Table 4 ). This decrease can be related to a higher metabolic fitness of the cells during the initial stationary phase apparent from faster glucose consumption (Figure 3 ). An opposite trend was observed for the product yields, which improved in all setups at the later fermentation stages. Although in both evaluated periods, the cells are in stationary phase, the increased yield might be explained by the transient phase into the resting cell state at the early time point, where resources were still directed into growth, as apparent from an increase in biomass. TABLE 4 Comparison of propionate yields and productivities of P. taiwanensis VLB120 Δ Tn7 pT_yciA during the three evaluated fed‐batch strategies, DO‐triggered with 50 and 100 mM initial glucose concentration and feedback‐loop controlled fed‐batch (TRACE) at two different fermentation stages. Fermentation setup Parameter Period 50 mM \n a \n \n , \n \n b \n \n 100 mM \n a \n \n , \n \n b \n \n Trace \n c \n \n Max. specific growth rate (h −1 ) – 0.19 ± 0.01 0.20 ± 0.00 0.22 Max. propionate titre (mM) – 36 ± 3 39 ± 6 35 (g L −1 ) 2.6 ± 0.2 2.8 ± 0.4 2.6 Product yield (mg g CDW \n −1 ) A \n d \n \n 83 ± 3 66 ± 19 95 B \n e \n \n 94 ± 11 100 ± 13 120 Productivity (mg L −1 h −1 ) A \n d \n \n 84 ± 3 67 ± 11 82 B \n e \n \n 62 ± 1 55 ± 2 44 \n a \n Initial glucose concentration. \n b \n Data are the mean and standard deviation of biological replicates. \n c \n Data are from a single experiment. \n d \n Time interval A was, on average, between 11.6 ± 0.7 and 24.9 ± 1.8 h. \n e \n Time interval B was, on average, between 37.3 ± 8.1 and 50.5 ± 11.1 h. FIGURE 3 Representative fermentation profiles of the three evaluated fed‐batch modes, DO‐triggered addition of 25 mM of glucose with (A) 50 mM of initial glucose and (B) 100 mM of initial glucose concentration and (C) feedback‐loop controlled fed‐batch maintaining a constant glucose concentration of 25 mM. The time periods highlighted in red represent the initial, (A), and later periods, (B) evaluated for productivities, titres and product yields shown in Table 4 . In both the DO‐controlled fed‐batch starting with 50 mM glucose and the feedback‐loop controlled fermentation, a maximal specific productivity of 11 mg h −1 g −1 was observed. The feedback‐loop controlled fermentation achieved a product yield (mg g CDW \n −1 ) 13% higher than the DO‐controlled 50 mM glucose fed‐batch during period A, showing some benefits of the continuous glucose availability. However, despite the yield differences between the different fermentation strategies, there was no significant difference between the propionate titres. The titres reached with this engineered Pseudomonas strain of maximal 39 ± 6 mM (2.8 ± 0.4 g L −1 ) are far from the highest production levels published using native producers like P. acidipropionici , P. shermanii or P. freudenreichii reported to produce propionate in the range of 191–1259 mM (14–92 g L −1 ) (Gonzalez‐Garcia, McCubbin, Navone, et al., 2017 ). However, in most reported studies, high amounts of yeast extract were added to the production media, contributing to propionate production through amino acid catabolism (Akawi et al., 2015 ; Jun Choi et al., 2012 ; Miscevic et al., 2020 ; Suwannakham et al., 2006 ). For a fair benchmark with other cell factories, comparable conditions, namely, production in mineral salt medium without adding yeast extract and single‐copy expression of the sleeping beauty mutase, need to be employed. Closest to this requirement is the work of (Li et al., 2017 ), who reported the production of 37.2 mM (2.72 g L −1 ) propionate with E. coli expressing the sleeping beauty mutase from the medium‐copy plasmid pET28a (~10–20 copies) under a strong constitutive promoter cultivated in mineral salt medium with glucose as sole carbon source. When a methylmalonyl‐CoA mutase from Methylobacterium extorquens AM1 was co‐expressed from the pACYC184 plasmid (~10 copies) under the control of the same strong promoter, the propionate titre increased to 4.95 g L −1 . Besides the work of Li et al., a comparable approach to propionate production was pursued by (Gonzalez‐Garcia, McCubbin, Wille, et al., 2017 ) by overexpressing on a plasmid base the sbm operon and the methylmalonyl‐CoA epimerase from P. acidipropionici reaching a final titre of 15.5 mM (1.13 g L −1 ) of propionate. Although the titre achieved in the present work stayed below this benchmark, the following considerations underline the potential of Pseudomonas . Propionate production in E. coli was achieved under microaerobic conditions leading to fermentation and significant succinate secretion in the engineered strain void of lactate and formate production pathways. The strain was engineered to run the reverse TCA cycle, and the significant succinate secretion reflects thermodynamic hindrances of its conversion to succinyl‐CoA (Li et al., 2017 ). Pseudomonas fermentations are conducted under aerobic conditions in which the strain runs the oxidative TCA cycle, which might lead to lower yields due to carbon loss in form of CO 2 but is thermodynamically more favourable. Furthermore, the fact that no byproducts are produced might counterbalance the lower yield since, according to Gonzalez‐Garcia et al., a major reason that makes biological propionate production not competitive is the complex downstream process required to remove the undesired fermentation byproducts (Gonzalez‐Garcia, McCubbin, Navone, et al., 2017 ). Besides these metabolic advantages of Pseudomonas , it is important to reckon that minimal genetic engineering was employed in the production strain in this initial work, in contrast to the previous E. coli work, which only led to ca. 10% of the theoretical yield of propionate production in P. taiwanensis VLB120 with glucose as carbon source (Table 2 ). Anticipating that gene expression optimisation has a similar impact on propionate production as in E. coli , there is high potential to significantly improve propionate synthesis in Pseudomonas and propel it into competitive ranges."
} | 6,996 |
37638302 | PMC10450786 | pmc | 7,095 | {
"abstract": "Honey bees use a complex form of spatial referential communication. Their waggle dance communicates to nestmates the direction, distance, and quality of a resource by encoding celestial cues, retinal optic flow, and relative food value into motion and sound within the nest. This protocol was developed to investigate the potential for social learning of this waggle dance. Using this protocol, we showed that correct waggle dancing requires social learning. Bees ( Apis mellifera ) that did not follow any dances before they first danced produced significantly more disordered dances, with larger waggle angle divergence errors, and encoded distance incorrectly. The former deficits improved with experience, but distance encoding was set for life. The first dances of bees that could follow other dancers had none of these impairments. Social learning, therefore, shapes honey bee signaling, as it does communication in human infants, birds, and multiple other vertebrate species. However, much remains to be learned about insects’ social learning, and this protocol will help to address knowledge gaps in the understanding of sophisticated social signal learning, particularly in understanding the molecular bases for such learning. Key features It was unclear if honey bees ( Apis mellifera ) could improve their waggle dance by following experienced dancers before they first waggle dance. Honey bees perform their first waggle dances with more errors if they cannot follow experienced waggle dancers first. Directional and disorder errors improved over time, but distance error was maintained. Bees in experimental colonies continued to communicate longer distances than control bees. Dancing correctly, with less directional error and disorder, requires social learning. Distance encoding in the honey bee dance is largely genetic but may also include a component of cultural transmission."
} | 474 |
35540780 | PMC9079982 | pmc | 7,096 | {
"abstract": "Nitrogen deposition and soil salinization–alkalization have become major environmental problems throughout the world. Leymus chinensis is the dominant, and considered the most valuable, species for grassland restoration in the Northeast of China. However, little information exists concerning the role of arbuscular mycorrhizal fungi (AMF) in the adaptation of seedlings to the interactive effects of nitrogen and salt–alkali stress, especially from the perspective of osmotic adjustment and ion balance. Experiments were conducted in a greenhouse and Leymus chinensis seedlings were cultivated with NaCl/NaHCO 3 under two nitrogen treatments (different concentrations of NH 4 + /NO 3 − ). Root colonization, seedling growth, ion content, and solute accumulation were measured. The results showed that the colonization rate and the dry weights of the seedlings were both decreased with the increasing salt–alkali concentration, and were much lower under alkali stress. Both of the nitrogen treatments decreased the colonization rate and dry weights compared with those of the AM seedlings, especially under the N2 (more NH 4 + –N content) treatment. The Na + content increased but the K + content decreased under salt–alkali stress, and more markedly under alkali stress. AMF colonization decreased the Na + content and increased the K + content to some extent. In addition, the nitrogen treatments had a negative effect on the two ions in the AM seedlings. Under salt stress, the seedlings accumulated abundant Cl − to maintain osmotic and ionic balance, but alkali stress inhibited the absorption of anions and the seedlings accumulated organic acids in order to resist the imbalance of both osmosis and ions, whether under the AM or nitrogen treatments. In addition, proline accumulation is thought to be a typical adaptive feature in both AM and non-AM plants under nitrogen and salt–alkali stress. Our results suggest that the salt–alkali tolerance of Leymus chinensis seedlings is enhanced by association with arbuscular mycorrhizal fungi, and the seedlings can adapt to the nitrogen and salt–alkali conditions by adjusting their osmotic adjustment and ion balance. Excessive nitrogen partly decreased the salt–alkali tolerance of the Leymus chinensis seedlings .",
"conclusion": "Conclusion In brief, the results of the present investigation show that salt stress and alkali stress differ greatly and the physiological adaptive strategy of L. chinensis seedlings to the two stresses is also different. AMF colonization can protect the seedlings against salt–alkali stress and nitrogen deposition by adjusting the osmotic and ion balance (by changing the inorganic cation–anion and organic solute content). In addition, excessive nitrogen exerts negative effects on the mycorrhizal colonization and the salt–alkali tolerance of L. chinensis seedlings. Therefore, AMF inoculation in L. chinensis seedlings could serve as a useful tool for alleviating salinity and alkalinity stress, and could be further applied to a practice such as restoring and reestablishing deteriorated salt–alkali grasslands, as found in northeastern China. In addition, our study also provides an important theoretical basis for the responses of Leymus chinensis –AMF symbionts to nitrogen deposition.",
"introduction": "Introduction Natural processes and human activities, such as fossil fuel combustion and fertilizer use, have strongly enhanced the rate of atmospheric nitrogen deposition and have received much attention. 1 According to the fifth IPCC report, nitrogen input into the ecosystem due to human causes each year has increased tenfold over the past 150 years and it has been predicted to be at least twice the current level by the 2050s. 2 Excess input of nitrogen can have adverse ecological effects, such as changes in soil function and eutrophication. 3,4 Nitrogen is also a limiting resource in the grassland ecosystem of northern China, which exerts a profound effect on plant and soil microorganisms. 5 In addition, not only is the total N content increasing, but the ratio of NH 4 + –N to NO 3 − –N is also changing as a result of N deposition. However, to the best of our knowledge, the physiological effect of NH 4 + /NO 3 − has been always overlooked. Salinity is considered one of the most important environmental factors limiting plant growth and yield throughout the world. 6 The stress caused by soil salinity generally involves osmotic stress and ion-induced injury. 7 These effects can inhibit nutrient absorption and prevent them from being transported within plants. In order to resist environmental stress, most plants have developed a variety of adaptive mechanisms, such as the synthesis of compatible solutes, the accumulation or exclusion of selected ions, and the control of ion uptake by the roots. 8 In addition, soil alkalization is also a marked feature in Northeastern China. The existence of alkali stress has been demonstrated to be more severe than salt stress due to the high pH, which may inhibit ion uptake and disrupt ionic balance. 9–11 Thus, they are actually two totally different types of stress. Yet, in spite of this, people pay little attention to the effect of alkali stress compared with that of salt stress. Arbuscular mycorrhizal fungi (AMF), a kind of ancient soil microorganism, widely occurs in saline soil, and can form mutualistic relationships with over 80% of terrestrial plants. 12 It is known that arbuscular mycorrhiza can enhance plant growth and development, and can also alleviate the adverse effects of salt stress. 13,14 The potential mechanisms for mycorrhizal plants responding to salt stress may include improving the nutrient uptake ability of the plant (especially the uptake of phosphorus), maintaining higher antioxidant enzymatic activities, elevating the K + content, and changing the root functions. 15–17 However, the ecological role of arbuscular mycorrhizal fungi on the osmotic adjustment and ion transport of plants under alkali stress (high pH), especially the interactive effects of alkali stress and nitrogen deposition (different concentrations of NH 4 + /NO 3 − ) conditions, has still rarely been reported. \n Leymus chinensis , also called alkali grass, is a perennial rhizomatous species of the family Poaceae. This grass not only has a high tolerance to salt–alkali soil, but also contains quite a lot of nutrients, such as carbohydrates, minerals, and proteins. 18 Thus, some reports have indicated that Leymus chinensis is considered to be one of the most valuable species for grassland restoration in the Northeast of China. 19,20 Here, we evaluated the contribution of arbuscular mycorrhizal fungi to the growth, ion content, and solute accumulation of Leymus chinensis seedlings under salt–alkali stress and nitrogen deposition. We aimed firstly to clarify the role of AMF during the adaptation of the Leymus chinensis seedlings to salt–alkali stress, and then we further explored the response of the plant–AMF to salt/alkali stress and its interactive effects with nitrogen deposition (different concentrations of NH 4 + /NO 3 − ) from the perspective of osmotic adjustment and ion balance.",
"discussion": "Results and discussion Root colonization and seedling growth The root mycorrhizal colonization was decreased significantly by the nitrogen treatments under salt–alkaline stress ( P < 0.01). Without the nitrogen treatment and stress conditions, the root mycorrhizal colonization can reach a maximum value of 97%. Compared to the N1 treatment, the colonization rate under the N2 treatment was much lower. Moreover, both salt and alkali stresses affected the mycorrhizal colonization, and the inhibition action of alkali stress was much stronger. Under the highest salinity stress concentration (200 mM), the colonization rate reached nearly 90%, but only reached 63% under alkalinity stress at the same stress intensity ( Tables 1 and 2 ). Mycorrhizal colonization rate and seedling dry weight of Leymus chinensis under nitrogen deposition and salinity conditions a Treatment Mycorrhizal colonization rate (%) Dry weight (g per plant) CK − AM 0.00 ± 0.00a 0.34 ± 0.03a CK + AM 97.33 ± 1.33d 0.49 ± 0.03b CK + AM + N1 78.33 ± 1.20c 0.36 ± 0.02a CK + AM + N2 71.67 ± 1.76b 0.33 ± 0.02a S1 − AM 0.00 ± 0.00a 0.27 ± 0.01a S1 + AM 92.00 ± 1.00d 0.37 ± 0.04b S1 + AM + N1 69.33 ± 0.67c 0.31 ± 0.01a S1 + AM + N2 64.33 ± 0.67b 0.28 ± 0.01a S2 − AM 0.00 ± 0.00a 0.24 ± 0.02a S2 + AM 88.67 ± 0.67d 0.31 ± 0.04b S2 + AM + N1 63.67 ± 1.33c 0.29 ± 0.02b S2 + AM + N2 59.33 ± 2.96b 0.28 ± 0.02 ab AM *** *** S *** * AM*S *** N.S. N1 *** ** N1*AM*S *** N.S. N2 *** *** N2*AM*S N.S. N.S. a The different letters indicate significant differences between the treatments (Tukey’s test P < 0.05). *** P < 0.001, ** P < 0.01, * P < 0.05, N.S. = not significant. Mycorrhizal colonization rate and seedling dry weight of Leymus chinensis under nitrogen deposition and alkalinity conditions a Treatment Mycorrhizal colonization rate (%) Dry weight (g per plant) CK − AM 0.00 ± 0.00a 0.34 ± 0.03a CK + AM 97.33 ± 1.33d 0.49 ± 0.03b CK + AM + N1 78.33 ± 1.20c 0.36 ± 0.02a CK + AM + N2 71.67 ± 1.76b 0.33 ± 0.02a A1 − AM 0.00 ± 0.00a 0.25 ± 0.01a A1 + AM 79.33 ± 1.76d 0.25 ± 0.02a A1 + AM + N1 58.67 ± 2.03c 0.26 ± 0.01a A1 + AM + N2 50.67 ± 0.67b 0.27 ± 0.01a A2 − AM 0.00 ± 0.00a 0.20 ± 0.02a A2 + AM 63.33 ± 1.67d 0.22 ± 0.03a A2 + AM + N1 51.33 ± 0.67c 0.20 ± 0.01a A2 + AM + N2 44.00 ± 1.00b 0.20 ± 0.01a AM *** ** A *** *** AM*A *** * N1 * *** N1*AM*A * * N2 *** ** N2*AM*S ** ** a The different letters indicate significant differences between the treatments (Tukey’s test P < 0.05). *** P < 0.001, ** P < 0.01, * P < 0.05, N.S. = not significant. The dry weight of the L. chinensis seedlings was significantly inhibited with the increasing the salt–alkali stress concentration, and alkali stress also further inhibited the dry weight ( P < 0.05, Tables 1 and 2 ). The plants inoculated with the AMF had significantly higher seedling weights under both stress and non-stress conditions. The nitrogen treatments (N1 and N2) decreased the dry weight compared with the inoculated seedlings. The N1 and N2 treatments made no difference to the seedling dry weight under salt stress ( P > 0.05). In addition, when alkali stress reached 100 mM, no significant change was observed among the four treatments ( Table 2 ). It is known that salt stress can affect plants, AMF, and their interactions. 26,27 In the present study, the mycorrhizal colonization decreased with the increasing salt stress intensity, and decreased much more markedly under alkali stress ( Table 2 ), indicating that salt–alkali stress inhibited the AMF growth. Some other reports have also supported this viewpoint. 22 The main reason is that salt stress inhibits spore germination or hyphal growth. 28 However, the much greater inhibition of alkali stress is perhaps due to high pH stress. Under such conditions, the plant root was subjected to much more damage, and the root also released many more chemical components into the soil, and then affected the mycorrhizal colonization. The specific reason requires further study. Our research also showed that nitrogen treatments could decrease the mycorrhizal colonization. One of the main reasons might be that the much higher concentration of nitrogen changed the function of the AMF. In addition, the inhibitory effect of NH 4 + –N was much stronger, and it might be that NO 3 − –N is more easily absorbed by L. chinensis , and the ammonium had a toxic effect to some extent, which also needs further research. Previous studies have demonstrated that plant growth can be inhibited under both salt and alkali stresses because of ion-excess effects. 9,29 However, the plant must spend much more energy to cope with the high pH under alkali stress. Thus, the dry weight of L. chinensis under alkali stress was lower than that under salt stress. Moreover, the inoculated seedlings had higher dry weights under salt stress, showing that AMF could increase the salt tolerance of L. chinensis. Similar results have also been reported by others. 30,31 However, alkali stress did not increase the seedling dry weights, which was also because of the high pH and much more damaging effects, and needs further research. Inorganic ion (cation and anion) content With the increasing intensity of the salt–alkali stress, the Na + content in the shoot increased significantly ( P < 0.05), and this effect was observed more markedly under alkali stress ( Fig. 1A and B ). AMF colonization decreased the Na + content under stress conditions, especially under higher concentration stress. When the salt concentration reached 200 mM, the Na + content of the seedlings inoculated with AMF was decreased by 18.1% and 18.8% compared with that of the non-inoculated seedlings under salt and alkali stresses. Both nitrogen treatments increased the Na + content at each salinity/alkalinity level in the AMF colonization treatments ( P < 0.05), and the N1 and N2 treatments showed no significant differences, except for at 200 mM salt stress. The K + content in the shoot decreased with increasing salt and alkali stresses, and these changes were much greater under alkali stress ( P < 0.05). Except for at 100 mM salt stress, AMF colonization increased the K + content under salt–alkali stress conditions. The nitrogen treatments decreased the K + content under salt–alkali stresses ( P < 0.05), and the N1 and N2 treatments also showed no significant differences ( Fig. 1C and D ). In addition, AMF colonization did not affect the Mg 2+ and Ca 2+ content and, furthermore, the N treatments also had almost no influence on the Mg 2+ and Ca 2+ content ( Fig. 1E–H ). Fig. 1 Na + (A and B), K + (C and D), Ca 2+ (E and F), and Mg 2+ (G and H) content in Leymus chinensis seedlings [non AM ( ), AM ( ), AM + N1 ( ), AM + N2 ( )] under nitrogen deposition, salinity (A, C, E, and G), and alkalinity (B, D, F, and H) conditions. The bars represent mean ± S.E. ( n = 4). The different letters indicate significant differences between the treatments (Tukey’s test P < 0.05). *** P < 0.001, ** P < 0.01, * P < 0.05, NS = not significant. With the increasing intensity of the salt stress, the Cl − content in the shoot increased significantly, and AMF colonization decreased the Cl − content under salt stress ( P < 0.05; Fig. 2A ). However, no significant change was observed in the Cl − content under either alkali stress or AMF colonization with alkali stress ( P > 0.05; Fig. 2B ). In addition, the nitrogen treatments increased the Cl − content in the L. chinensis seedlings. Under the alkali stress treatment, AMF colonization and nitrogen treatments both had no obvious significant effect on the H 2 PO 4 − and SO 4 2− content ( P > 0.05; Fig. 2D and F ). Moreover, the H 2 PO 4 − and SO 4 2− content only increased significantly at 100 mM salt stress with the N2 treatment ( P < 0.05; Fig. 2C and E ). The NO 3 − content in the shoot decreased with increasing salt and alkali stresses, and AMF colonization increased the NO 3 − content, but it only reached a significant level at CK and 200 mM salt stress treatment ( P < 0.05; Fig. 2G ). In addition, the N treatments have almost no influence on the NO 3 − content ( Fig. 2G and H ). Fig. 2 Cl − (A and B), H 2 PO 4 − (C and D), SO 4 2− (E and F), and NO 3 − (G and H) content in Leymus chinensis seedlings [non AM ( ), AM ( ), AM + N1 ( ), AM + N2 ( )] under nitrogen deposition, salinity (A, C, E, and G), and alkalinity (B, D, F, and H) conditions. The bars represent mean ± S.E. ( n = 4). The different letters indicate the significant differences between the treatments (Tukey’s test P < 0.05). *** P < 0.001, ** P < 0.01, * P < 0.05, NS = not significant. Plants growing in salt–alkali soil generally suffer from two mainly distinct stresses: ionic stress and osmotic stress. Na + is one of the most dominating toxic ions in salinized and alkalinized soil. 32 It can disrupt the structure of many macromolecules and their normal physiological metabolism. In addition, ionic imbalance is also a distinguishing feature caused by the influx of superfluous Na + . 33,34 The Na + ions enter the plant cells through a high-affinity K + transporter (HKT) and non-selective cation channels. 35 Thus, for most plants living in salt–alkali environments, Na + often greatly accumulates in the vacuoles and also inhibits K + absorption. 7 In our results, the Na + concentration sharply increased and the K + concentration was simultaneously decreased under both salt and alkali stresses, with both the +AMF and −AMF treatments. However, the change was much greater under alkali stress, indicating that the high pH caused by alkali stress exerts much more damaging effects on the seedlings (higher Na + and lower K + ), which rely on a transmembrane proton gradient. In addition, this result was in contradiction with observations of other halophytes, such as Kochia sieversiana and Chloris virgate , that are also widespread in the Songnen salt–alkali grassland in Northeastern China, 29,36 and showed no competitive inhibition between the two ions in the two species, indicating that different plants may have distinct pathways for the absorption and transfer of K + and Na + , and need further research. Our results also showed that Mg 2+ and Ca 2+ increased under both salt stress and alkali stress, but their amounts were very low and their roles in ionic balance and osmotic adjustment were tiny. It is widely recognized that AM symbiosis is a pivotal component in helping plants cope with adverse environments such as salt–alkali stress. 37 It is evident from the present study that AM seedlings showed lower Na + and higher K + compared with the −AM plants, especially under the highest concentration of stress (200 mM, Fig. 1 ). These results show that the role of the mycorrhizal fungi in alleviating salt stress is mainly due to the inhibition of toxic Na + absorption and transportation. The lower levels of Na + in the AM seedlings may also be explained by the dilution effect because of plant growth enhancement. 27 In addition, potassium is one of the most important essential elements for plant growth as it plays a key role in plant metabolism. The higher K + concentration and its direct effect of establishing a higher K + /Na + in the mycorrhizal plants under salt–alkali stress also relieved the negative impacts by the ionic balance of the cytoplasm or the Na efflux from the plant. 15 Similar results have also been reported for other plants (with or without AM) under salt stress conditions, such as wheat and Trigonella foenum-graecum . 38 Studies carried out by Giri et al. 15 have also indicated that the increased accumulation of K + reduced the translocation of Na + in the shoots of AMF-colonized acacia plants grown in saline soil. Moreover, nitrogen deposition increased the Na + content and also decreased the K + content within the mycorrhizal inoculation treatments of the L. chinensis seedlings, especially under the N2 treatment ( Fig. 1 ). The main reason for this finding is that excessive nitrogen in the soil induces a nutrient imbalance and also changes the distribution and transportation of ions. The ammonium may have a toxic effect to some extent, and the specific mechanism needs further research. The above results also clearly show that nitrogen deposition (different concentrations of NH 4 + /NO 3 − ) decreased the salt–alkali tolerance of the L. chinensis seedlings by reducing the suitability of the colonization. Ionic imbalance within plants has been reported to be mainly caused by the influx of superfluous Na + . 33 Most plants always accumulate the main inorganic anions, such as Cl − , at this point to cope with adverse environments. In our study, Cl − and SO 4 2− were accumulated in the L. chinensis seedlings under salt stress. However, all of the anions showed a decreasing trend under alkali stress ( Fig. 2A ). Our results verified that salt stress and alkali stress had different physiological effects on the plants, and the high pH might inhibit the absorption of inorganic anions such as Cl − , NO 3 − , and SO 4 2− . Similar results were also reported for other plants (both crops and grasses), such as the tomato, 39 seabuckthorn 40 , and Lathyrus quinquenervius . 11 In general, superfluous Cl − can also be toxic to plant growth in saline areas. 41 This problem can also be resolved by using arbuscular mycorrhiza to some extent, which can decrease the Cl − uptake. 42 In our results, AMF colonization decreased the Cl − content under salt stress, which is consistent with the previous report on lettuce and onions, 43 but there was no significant difference between the −AM and mycorrhizal inoculation treatments in L. chinensis seedlings under alkali stress because of the severe influence of the high pH. In addition, the result was in contradiction with observations reported for citrus plants, spring wheat, and winter barley. 44,45 They found that the Cl − content increased due to the mycorrhizal colonization, and the main reason is perhaps the carbon drain imposed by the mycorrhizal hyphae on plants, and then the enhancement of the translocation of Cl − from the saline soil. The differences in the plant species may be a reason for these differences and for their specific causes, both of which should be further investigated. Moreover, similar changes were observed for Cl − and Na + when the L. chinensis seedlings were inoculated with mycorrhizal, further illustrating the negative impact of nitrogen deposition, especially the ammonium nitrogen. Interestingly, L. chinensis seedlings can also maintain constant ionic balance under nitrogen treatment. Organic solute content With the increasing intensity of the salt–alkali stress, the proline content in the shoots of the L. chinensis seedlings increased significantly, and this effect was observed more markedly under salt stress ( P < 0.05, Fig. 3A and B ). AMF colonization decreased the proline content under stress conditions, especially under the highest concentration stress (200 mM), where the proline content decreased by 65.8%. The application of either nitrogen treatment increased the proline content at each salinity/alkalinity level ( Fig. 3A and B ). The soluble sugar content showed a similar tendency to the proline ( Fig. 3C and D ). In addition, AMF colonization only decreased the soluble sugar content to 19.6% at 200 mM alkali stress. The N treatments had no influence on the soluble sugar content of the inoculated seedlings. The soluble sugar content only increased at alkali stress under the N2 treatment. AM colonization did not affect the MDA content under salt and alkali stresses ( Fig. 3E and F ). The N treatments increased the MDA content, much more markedly with the N2 treatment. Fig. 3 Proline (A and B), soluble sugar (C and D), and MDA (E and F) content in Leymus chinensis seedlings [non AM ( ), AM ( ), AM + N1 ( ), AM + N2 ( )] under nitrogen deposition, salinity (A, C, E, and G), and alkalinity (B, D, F, and H) conditions. The different letters indicate significant differences between the treatments (Tukey’s test P < 0.05). *** P < 0.001, ** P < 0.01, * P < 0.05, NS = not significant. Malic acid, citric acid, acetate acid, and oxalate acid were detected in the shoots of the L. chinensis seedlings under salt and alkali stresses ( Fig. 4 ). The citric acid content increased under alkali stress ( Fig. 4B ), and AMF colonization and N treatments did not affect the citric acid content under salt–alkali stress. The malic acid content showed a similar tendency to that of citric acid. AMF colonization decreased the malic acid content by 31.5% under 200 mM alkali stress, but did not affect the content at other stress concentrations ( Fig. 4D and E ). In addition, the N treatments have almost no influence on the malic acid content. AMF colonization also decreased the acetate acid content according to Fig. 4E and F . However, unlike these three organic acid compositions, no change was observed in the oxalate acid content under both the AM or N treatments ( Fig. 4G and H ). Fig. 4 Citric acid (A and B), malic acid (C and D), acetic acid (E and F), and oxalic acid (G and H) content in Leymus chinensis seedlings [non AM ( ), AM ( ), AM + N1 ( ), AM + N2 ( )] under nitrogen deposition, salinity (A, C, E, and G), and alkalinity (B, D, F, and H) conditions. The different letters indicate significant differences between the treatments (Tukey’s test P < 0.05). *** P < 0.001, ** P < 0.01, * P < 0.05, NS = not significant. In salt–alkali environments, plants must face the negative impact of the lower water potential of the soil, which makes it difficult for the plants to acquire adequate water from the surrounding soil. Several organic solutes, such as proline, soluble sugar, and MDA, are accumulated under these conditions. In general, proline accumulation is thought to be an adaptive feature under salt–alkali stress in both AM and non-AM plants. 46 Our results showed that the accumulation of proline in the L. chinensis seedlings increased with increasing salt–alkaline concentration ( Fig. 3A and B ). Moreover, the proline content was detected to decrease in the seedlings with AMF inoculation. This differed to most previous reports, which found that AMF colonization could increase the proline content in plants such as soybean and wheat. 46 However, similar results were found for Vicia faba by Rabie and Almadini. 47 The possible reason is that the proline content can also be used for measuring the degree of plants that were injured, indicating that AMF colonization alleviates salt–alkali stress on the L. chinensis seedlings. The nitrogen deposition results also prove this viewpoint. Organic acids also played a potential role as cell osmolytes in the osmotic adjustment. 48 In our study, we clearly found that salt stress did not affect the amounts of four organic acids (malic acid, citric acid, acetate acid, and oxalate acid), but alkali stress significantly enhanced them, indicating that organic acid accumulation is a specific response to high pH stress. Under alkali stress, superfluous Na + was accumulated, but the Cl − content had no significant change. Thus, we consider that the accumulation of organic acids not only played an important osmotic role but also buffered excess toxic cations and maintained ionic balance. Similar viewpoints have also been reported by others. 9,11 In addition, we also found that AMF colonization did not affect the organic acid content, except for acetate acid, but decreased most of them under alkali stress. Combined with the Na + and Cl − results as described above, we conclude that there exists a correlation between the Na + and organic acid content, and that L. chinensis seedlings can maintain ionic balance under both AM and non-AM conditions. In addition, the organic acid content showed an increasing trend in the AM seedlings that underwent the N treatments to some extent, especially the N2 treatment. The main reason for this finding is that excessive nitrogen in the soil causes a negative effect on the AMF, and the host plant needs to synthesize much more organic acid to resist the stress conditions, which needs further research."
} | 6,874 |
33848489 | null | s2 | 7,099 | {
"abstract": "Arbuscular mycorrhizae (AM) are the most frequent symbioses of land plants. By reisolating a long-lost fungus from nature, a new study cracks the genomics of an enigmatic fungal-cyanobacterial partnership and reestablishes a valuable model for understanding the AM symbiosis."
} | 68 |
21179017 | PMC3018171 | pmc | 7,100 | {
"abstract": "Sensor histidine kinases underlie the regulation of a range of physiological processes in bacterial cells, from chemotaxis to cell division. In the gram-negative bacterium Caulobacter crescentus , the membrane-bound histidine kinase, DivJ, is a polar-localized regulator of cell cycle progression and development. We show that DivJ localizes to the cell pole through a dynamic diffusion and capture mechanism rather than by active localization. Analysis of single C. crescentus cells in microfluidic culture demonstrates that controlled expression of divJ permits facile tuning of both the mean and noise of the cell division period. Simulations of the cell cycle that use a simplified protein interaction network capture previously measured oscillatory protein profiles, and recapitulate the experimental observation that deletion of divJ increases the cell cycle period and noise. We further demonstrate that surface adhesion and swarming motility of C. crescentus in semi-solid media can also be tuned by divJ expression. We propose a model in which pleiotropic control of polar cell development by the DivJ–DivK–PleC signaling pathway underlies divJ -dependent tuning of cell swarming and adhesion behaviors.",
"conclusion": "Concluding remarks In synthetic biology, actuators are engineered genetic circuits used to interface with natural networks and control cellular processes ( Voigt, 2006 ). We have demonstrated a simple multiplex actuator in C. crescentus —a single gene ( divJ ) controlled by an inducible xylose promoter—that tunes cell cycle timing and its variance, swarming motility, and cell surface adhesion. Our measurements of single cells with controlled induction of divJ expression allow a systematic analysis of single-cell dynamics of cell cycle regulation and cell surface adhesion. By integrating a variety of experimental observables with the development of models and simulations, we have developed an improved mechanistic understanding of dynamical cellular processes regulated by the DivJ histidine kinase. Recent advances in controlling the rhythmicity of eukaryotic cell cycle oscillators by forcing ( Battogtokh et al, 2006 ; Charvin et al, 2009 ) demonstrate the feasibility of precise manipulation and synchronization of naturally designed oscillators. Our simplified cell cycle model also suggests the possibility to entrain cell cycle oscillations in a population of prokaryotic cells through periodic genetic perturbations, which has been verified by experiments (Y Lin, Y Li, A Dinner, S Crosson and N Scherer, unpublished). Investigating the connectivity between spatial localization of regulatory components and the function of regulatory networks is clearly a challenging frontier in the study of bacterial cell biology ( McAdams and Shapiro, 2003 ). The microfluidic perturbation and imaging approach presented here allows spatial and temporal resolution of functional connectivities between localized protein components and overall network output (as assessed by monitoring various aspects of cellular growth and physiology at the single-cell level). The combination of acute experimental control and emerging theoretical understanding in this model system offers the possibility of a comprehensive mapping of connectivities in a cellular control network without prior assumptions ( Ross, 2008 ).",
"introduction": "Introduction Simple estimates of diffusion in bacterial cells suggest that cytoplasmic and membrane protein concentrations would become uniform in seconds to minutes ( Mignot and Shaevitz, 2008 ). However, a spatially uniform concentration of protein across a cell precludes a nonequilibrium driving force that can lead to proper differentiation and development in many bacterial species. Indeed, it is known that surface structures such as pili and flagella are asymmetrically distributed on a bacterial cell, and that subcellular structures such as chemoreceptor complexes are selectively localized at the cell pole ( Alley et al, 1992 ; Nelson, 1992 ; Maddock and Shapiro, 1993 ). Recent developments in live cell imaging have shown that cytoskeletal and signaling/regulatory proteins also exhibit complex subcellular localization that varies temporally across the cell cycle ( Gitai et al, 2005 ; Bardy and Maddock, 2007 ; Shapiro et al, 2009 ). Thus, deciphering mechanisms of bacterial cell cycle regulation and development requires in-depth characterization of the organization and activity of proteins in both time and space ( McAdams and Shapiro, 2003 ). In Caulobacter crescentus , a model system for the study of asymmetric cell division and cell cycle regulation, specific regulatory proteins that exhibit temporal polar localization underlie the control of cell cycle progression, cell development, and cell adhesion ( Shapiro et al, 2002 ; Ebersbach and Jacobs-Wagner, 2007 ). C. crescentus begins its life as a non-replicative and motile ‘swarmer’ cell containing a single polar flagellum and polar type-IV pili. The cell occupies this motile developmental phase for a period ranging from as little as 15% ( Keiler and Shapiro, 2003 ; Siegal-Gaskins et al, 2009 ) to more than 50% ( Poindexter and Staley, 1996 ) of its cell cycle depending on the physical and chemical composition of the culture environment. After proceeding through the swarmer phase, C. crescentus differentiates into a replicative and sessile ‘stalked’ cell ( Figure 1A ), which differs both morphologically ( Poindexter, 1964 ) and metabolically ( Felzenberg et al, 1996 ) from its swarmer precursor. This dimorphic developmental cycle is controlled by a suite of signaling/regulatory proteins that exhibit dynamic spatial localization during the cell cycle ( Brown et al, 2009 ). Included among these dynamically localized regulatory proteins is the sensor histidine kinase, DivJ ( Wheeler and Shapiro, 1999 ; Figure 1A ). Co-localized with DivJ in the early stalked phase is the phosphorylated response regulator DivK∼P ( Jacobs et al, 2001 ), and the protease ClpXP ( McGrath et al, 2006 ) ( Figure 1A ), which degrades the master cell cycle regulator, CtrA ( Jenal and Fuchs, 1998 ). CtrA is synthesized and accumulates as the stalked cell develops into a pre-divisional cell. The increased concentration of CtrA drives various processes required for cell replication, development, and division, but also induces the expression of the response regulator DivK. DivK, in turn, controls the stability and activity of CtrA through two essential phosphorelays ( Biondi et al, 2006 ). All of these regulatory proteins are localized to the stalked pole (the former flagellar pole), whereas the nascent pili and flagellum are assembled at the pole opposite the stalk, where the polar development factor PodJ co-localizes ( Viollier et al, 2002 ; Figure 1A ). Thus, constriction of the membrane in the longitudinal middle of the cell body during the late-predivisional stage results in an asymmetric distribution of subcellular structures and proteins between the stalked and swarmer compartments, creating two progeny with distinct morphologies, protein compositions, and developmental programs ( Shapiro et al, 2002 ; Figure 1A ). Recent optical microscopy measurements of single C. crescentus cells have revealed an intriguing role for DivJ in the control of noise in cell division period ( Siegal-Gaskins and Crosson, 2008 ). The variance in interdivision timing of single cells increases abruptly upon disruption of the divJ gene, and is accompanied by a relatively small increase in the mean generation time. Whereas abundant genetic and biochemical data on regulatory/signaling proteins have facilitated modeling the complex transcriptional network underlying the C. crescentus cell cycle ( Li et al, 2008 , 2009 ; Shen et al, 2008 ), the existing cell cycle models do not explain experimental data that demonstrate an increase in cell cycle period and noise in a divJ null strain ( Siegal-Gaskins and Crosson, 2008 ). Moreover, mechanistic descriptions of how DivJ and its signaling partners become localized and how these proteins underlie the control of polar cell development and cell adhesion in C. crescentus remain incomplete. In this study, we study single cells in a microfluidic flow chamber to probe the effects of perturbation of divJ expression on multiple aspects of cell physiology, including the (mean) period of division and noise in the period, swarm behavior in semi-solid medium, and surface adhesion. Moreover, we characterize the mechanism of subcellular localization of DivJ–EGFP, which provides further insight into the regulation of C. crescentus physiology by DivJ. The measured single-cell division periods and the distributions thereof as presented herein reveal a striking tunability in both the mean and variance of the C. crescentus cell cycle, which depends solely on the level of divJ expression. Stochastic simulations with a simplified cell cycle model establish that DivJ-dependent phosphorylation of DivK is critical in maintaining low noise in the C. crescentus cell division period. In addition to its role in regulating the cell cycle period, high divJ expression also affects the frequency of swarmer cell adhesion to a glass surface and the swarm rate of C. crescentus cells in semi-solid growth medium. Control of swarming in semi-solid medium cannot be explained by an altered rate or frequency of swimming motility of individual cells. We propose that divJ -mediated regulation of swarming and adhesion stems from the pleiotropic control of the DivJ–DivK–PleC signaling pathway on multiple aspects of polar cell development and morphology. Finally, we provide direct evidence from experiments and simulations that the DivJ histidine kinase becomes localized to the cell pole through a dynamic diffusion-and-capture mechanism during the C. crescentus cell cycle.",
"discussion": "Discussion Cell cycle model predicts that robust phosphorylation of DivK by DivJ underlies the tight regulation of the cell cycle period The elucidation of the molecular machinery that governs eukaryotic cell growth and division has lead to several successful mathematical models describing the cell cycle(s) in eukaryotic systems, including yeast ( Chen et al, 2004 ) and Xenopus oocytes ( Sha et al, 2003 ). Recent advances in our understanding of bacterial cell regulation ( McAdams and Shapiro, 2003 ; Biondi et al, 2006 ) have prompted cell cycle modeling in prokaryotes, especially in C. crescentus ( Brazhnik and Tyson, 2006 ; Li et al, 2008 , 2009 ; Shen et al, 2008 ). Our single cell division data on wild-type and Δ divJ mutant cells ( Figure 2B ) called for construction of a network for cell cycle regulation that explained the observed differences in the division time statistics between wild-type and Δ divJ . The agreement between the simulated trajectories resulting from in our (simplified) cell cycle model and ensemble experimental data ( Figure 3C ; Supplementary Figure 2 ; Supplementary Table IV ), suggest the sufficiency of this simplified protein regulatory network. Specifically, deterministic (ODE) model simulation yields a longer ST cell cycle time for the Δ divJ mutant than wild type, a finding that is consistent with our experiments. Stochastic simulations with this model offer further information on the noise as reflected in the measured COV of the cell cycle distributions. The finding of a larger simulated COV for CtrA∼P oscillation periods in a Δ divJ mutant than in wild type recapitulates our experimental data and suggests the importance of robust DivJ-mediated phosphorylation of its cognate receiver protein, DivK, in regulating the variance of cell cycle oscillations. In our simulation, increased noise in the period of division in a Δ divJ genetic background arises from increased cell-to-cell variability in the concentration of phosphorylated DivK∼P as driven by an indirect and minor (slow) DivL-dependent phosphorylation pathway ( Reisinger et al, 2007 ) ( Supplementary Figure 4A ). Increased variability in the concentration of DivK∼P at the single-cell level subsequently leads to increased noise in the regulation of CtrA phosphorylation and degradation. Inducing divJ expression from the xylX promoter in a chromosomal Δ divJ null mutant restores the fast pathway for phosphorylation of DivK and thus reduces the noise of cell cycle period ( Supplementary Figure 4B ). It is well established that DivJ-mediated phosphorylation of DivK provides a negative control on the stability of CtrA and thus functions as a negative feedback signal in the C. crescentus cell cycle network. Our experiments and simulations demonstrate that the steady state level of DivK∼P at the single-cell level (as maintained by DivJ) is essential in regulating the timing and coherence of the cell division period in C. crescentus . However, we cannot exclude the existence of a positive feedback loop that also contributes to coherence in cell cycle timing, as has been demonstrated in budding yeast ( Holt et al, 2008 ; Skotheim et al, 2008 ). DivJ signaling and the pleiotropic control of swarming motility and surface adhesion divJ exerts pleiotropic control over multiple aspects of cellular development ( Wheeler and Shapiro, 1999 ; Jacobs et al, 2001 ). As such, the result that high divJ expression decreases swarming in a semi-solid medium (and in the absence of a defect in swimming) may be explained by multiple mechanisms including defects in the assembly of the chemoreceptor machinery or perturbation of the development of surface structures, such as type-IV pili, the stalk, or the adhesive holdfast. Given that decreased swarm rate at high divJ induction is correlated with decreased surface adhesion, it is likely that the adhesion and swarming phenotypes at high levels of divJ expression have a common cause ( Bodenmiller et al, 2004 ; Entcheva-Dimitrov and Spormann, 2004 ; Levi and Jenal, 2006 ). Specifically, the observed swarming and surface adhesion defects caused by overexpression of divJ may be explained by the regulation of PodJ through the DivJ–PleC–DivK signaling pathway. PodJ, a polar development factor, is important for pilus biogenesis, holdfast formation, and chemotaxis ( Wang et al, 1993 ; Viollier et al, 2002 ; Hinz et al, 2003 ; Chen et al, 2006 ). There are two isoforms of PodJ: full-length PodJ (PodJ L ) and cleaved PodJ (PodJ S ). PodJ L peaks at the early predivisional stage and is necessary for pilus biogenesis; its proteolytic product, PodJ S , peaks at late predivisional and early swarmer stage and is required for holdfast formation and chemotaxis ( Viollier et al, 2002 ; Chen et al, 2006 ). Proteolytic conversion from PodJ L into PodJ S requires compartmentalization of DivJ and its partner proteins DivK and PleC upon cytokinesis ( Chen et al, 2006 ). However, the differential compartmentalization of DivJ can be perturbed by overexpression of the divJ gene in both cell compartments, as carried out in our experiments. Therefore, we suggest a model in which increased DivJ concentration in the swarmer compartment due to constitutive overexpression ( Figure 7 ) results in premature localization of DivJ to the flagellar pole and elevated levels of DivK∼P ( Supplementary Figure 8 ). This situation would be predicted to suppress the triggering signal for proteolytic conversion of PodJ L to PodJ S . Phenotypically, decreased PodJ S levels are correlated with deficiencies in holdfast formation and chemotaxis without interfering with swimming motility ( Viollier et al, 2002 ; Hinz et al, 2003 ). A deficiency in holdfast development results in cell-surface-adhesion defects ( Bodenmiller et al, 2004 ; Entcheva-Dimitrov and Spormann, 2004 ; Levi and Jenal, 2006 ), whereas a deficiency in chemotaxis is reflected in reduced swarming motility in semi-solid agar ( Wang et al, 1993 ). Both of these phenotypes are observed in our assays at high divJ expression levels ( Figures 5 and 6 ). Integrated time-lapse in vivo fluorescence measurements and kinetic modeling reveal a diffusion-and-capture pathway for DivJ localization Our time-lapse single-cell fluorescence measurements establish the subcellular distribution of constitutively expressed DivJ–EGFP ( Figure 7 ) and complex dynamics in the appearance of fluorescence ( Figure 7C ). A phenomenological model ( Figure 7D ) is sufficient to explain the time evolution of the single-cell fluorescence time traces. Viewing the membrane-bound molecules as the reservoir and the polar capture matrix as the adsorber ( Supplementary Figure 6A ), the adsorption of the reservoir molecules onto the adsorber is made in analogy to the Langmuir adsorption process of gas molecules onto a solid surface ( Langmuir, 1916 ). The continuous expression of the protein increases the number of molecules in the reservoir which then partition between the reservoir and the adsorber by an adsorption/desorption relationship. The number of molecules adsorbed increases in accordance with the increasing number of reservoir molecules ( Supplementary Figure 6B grey curves) before reaching a steady state. The steady state for the number of reservoir molecule occurs when the rate of reservoir molecule reduction (i.e. by adsorption, cell volume doubling, and degradation) equals its rate of production (i.e. by synthesis and desorption). Similar arguments with the addition of maturation kinetics can account for the plateau behavior of experimental observables (i.e. the bright molecules A * and A * S). The linear dependence of the ratio between the measured steady-state levels of integrated fluorescence intensities for lateral membrane and stalked pole on divJ induction level is also captured by our model ( Supplementary Figure 7 ). This allows us to make a direct comparison between the experiment and simulation (compare Figure 7C versus Figure 7E and Supplementary Figure S5 versus Figure S6). This localization mechanism is consistent with a diffusion-and-capture model. Given its simplicity, diffusion-and-capture is probably a widely used mechanism for protein localization in bacteria ( Thanbichler and Shapiro, 2008 ). This model posits that proteins are randomly distributed and are freely diffusing until they are captured at the site in which they ultimately reside ( Rudner et al, 2002 ; Shapiro et al, 2002 ; Bardy and Maddock, 2007 ). With a diffusion-and-capture pathway, it has been argued that proteins can be adsorbed either dynamically or statically ( Shapiro et al, 2009 ). Our analysis of DivJ–EGFP in single cells supports a dynamic diffuse-and-capture mechanism for DivJ localization. The recent discovery and characterization of the pole-organizing protein, PopZ, at the poles of C. crescentus ( Bowman et al, 2008 ; Ebersbach et al, 2008 ; Shapiro et al, 2009 ) supports the hypothesis that there is a multi-component polar docking station that dynamically sequesters signaling proteins ( Shapiro et al, 2002 ). Our experiments and simulations are supportive evidence for this model of bacterial subcellular organization for the DivJ sensor histidine kinase. Concluding remarks In synthetic biology, actuators are engineered genetic circuits used to interface with natural networks and control cellular processes ( Voigt, 2006 ). We have demonstrated a simple multiplex actuator in C. crescentus —a single gene ( divJ ) controlled by an inducible xylose promoter—that tunes cell cycle timing and its variance, swarming motility, and cell surface adhesion. Our measurements of single cells with controlled induction of divJ expression allow a systematic analysis of single-cell dynamics of cell cycle regulation and cell surface adhesion. By integrating a variety of experimental observables with the development of models and simulations, we have developed an improved mechanistic understanding of dynamical cellular processes regulated by the DivJ histidine kinase. Recent advances in controlling the rhythmicity of eukaryotic cell cycle oscillators by forcing ( Battogtokh et al, 2006 ; Charvin et al, 2009 ) demonstrate the feasibility of precise manipulation and synchronization of naturally designed oscillators. Our simplified cell cycle model also suggests the possibility to entrain cell cycle oscillations in a population of prokaryotic cells through periodic genetic perturbations, which has been verified by experiments (Y Lin, Y Li, A Dinner, S Crosson and N Scherer, unpublished). Investigating the connectivity between spatial localization of regulatory components and the function of regulatory networks is clearly a challenging frontier in the study of bacterial cell biology ( McAdams and Shapiro, 2003 ). The microfluidic perturbation and imaging approach presented here allows spatial and temporal resolution of functional connectivities between localized protein components and overall network output (as assessed by monitoring various aspects of cellular growth and physiology at the single-cell level). The combination of acute experimental control and emerging theoretical understanding in this model system offers the possibility of a comprehensive mapping of connectivities in a cellular control network without prior assumptions ( Ross, 2008 )."
} | 5,320 |
35418298 | PMC8764830 | pmc | 7,102 | {
"abstract": "Background Methylacidiphilum sp. IT6 has been validated its C3 substrate assimilation pathway via acetol as a key intermediate using the PmoCAB3, a homolog of the particulate methane monooxygenase (pMMO). From the transcriptomic data, the contribution of PmoD of strain IT6 in acetone oxidation was questioned. Methylomonas sp. DH-1, a type I methanotroph containing pmo operon without the existence of its pmoD , has been deployed as a biocatalyst for the gas-to-liquid bioconversion of methane and propane to methanol and acetone. Thus, Methylomonas sp. DH-1 is a suitable host for investigation. The PmoD-expressed Methylomonas sp. DH-1 can also be deployed for acetol production, a well-known intermediate for various industrial applications. Microbial production of acetol is a sustainable approach attracted attention so far. Results In this study, bioinformatics analyses elucidated that novel protein PmoD is a C-terminal transmembrane–helix membrane with the proposed function as a transport protein. Furthermore, the whole-cell biocatalyst was constructed in Methylomonas sp. DH-1 by co-expression the PmoD of Methylacidiphilum sp. IT6 with the endogenous pMMO to enable acetone oxidation. Under optimal conditions, the maximum accumulation, and specific productivity of acetol were 18.291 mM (1.35 g/L) and 0.317 mmol/g cell/h, respectively. The results showed the first coupling activity of pMMO with a heterologous protein PmoD, validated the involvement of PmoD in acetone oxidation, and demonstrated an unprecedented production of acetol from acetone in type I methanotrophic biocatalyst. From the data achieved in batch cultivation conditions, an assimilation pathway of acetone via acetol as the key intermediate was also proposed. Conclusion Using bioinformatics tools, the protein PmoD has been elucidated as the membrane protein with the proposed function as a transport protein. Furthermore, results from the assays of PmoD-heteroexpressed Methylomonas sp. DH-1 as a whole-cell biocatalyst validated the coupling activity of PmoD with pMMO to convert acetone to acetol, which also unlocks the potential of this recombinant biocatalyst for acetol production. The proposed acetone-assimilated pathway in the recombinant Methylomonas sp. DH-1, once validated, can extend the metabolic flexibility of Methylomonas sp. DH-1. Supplementary Information The online version contains supplementary material available at 10.1186/s13068-022-02105-1.",
"conclusion": "Conclusion In this study, the PmoD from Methylacidiphilum sp. IT6 was elucidated as a membrane protein with the proposed function as a transport protein that captures and delivers acetone to the pMMO for oxidation. The coupling activity of heteroexpressed PmoD with the endogenous pMMO of Methylomonas sp. DH-1 led to a boost in the acetone oxidation to acetol in whole-cell biocatalyst assays with the highest titer of 18.291 mM (1.35 g/L) under optimal conditions. This result illustrates the potential of whole-cell biocatalyst Methylomonas sp. DH-1 in acetol production. Furthermore, the innate capacity of Methylomonas sp. DH-1 wild type to oxidize acetone to acetol as a whole-cell biocatalyst has been demonstrated, which is an unprecedented report in type I methanotrophic bacteria. From the data achieved in batch cultivation conditions and the reference with previous publications, we proposed a novel assimilation pathway of acetone or propane via acetol as the key intermediates. Once validated, this assimilation pathway can extend the metabolic flexibility of Methylomonas sp. DH-1.",
"discussion": "Discussion From the transcriptomic data of the previous study [ 12 ], the novel protein PmoD of Methylacidiphilum sp. IT6 raised the inquiry of its involvement in the acetone oxidation capacity of the PmoCAB3, a pMMO homolog. Thus, in this study, we deployed various bioinformatics tools to achieve some preliminary elucidations of this protein and also heteroexpressed this protein in the methanotrophic host Methylomonas sp. DH-1 to evaluate the coupling activity of this protein PmoD with endogenous pMMO in acetol production. The multiple alignment results revealed a conserved region of this novel protein PmoD with other reference PmoD sequences, composed of two cysteines (Cys31 and Cys63) and one histidine (His42) residue. As described in a previous study, two strictly conserved Cys residues of the surface-exposed region located near the N-terminus of PmoD proteins were suggested to play the main role in the copper-binding of proteobacterial methanotrophs, where each PmoD monomer contributes a Cys ligand to form bridging thiolates at the copper-binding site [ 24 ]. However, at the position of the conserved methionine residue in the PmoD sequences of proteobacterial methanotrophs, it is substituted by alanine residue, which is conserved in the novel protein PmoD of Methylacidiphilum sp. IT6, and other PmoD homologs (PmoD3 and PmoD5) of M. infernorum V4 and M. fumariolicum strain SolV. This replacement has been shown to eliminate the formation of copper-binding sites in proteobacterial methanotrophs [ 24 ], which raises the hypothesis about the function of the novel protein PmoD of strain IT6 and other PmoD proteins of the Methylacidiphilum genus. Furthermore, the phylogenetic tree analysis using MEGA X and structural prediction using InterProScan were to get further insights into the evolutionary and structural characteristics of the novel protein PmoD. The predicted structure of PmoD suggested it as the membrane protein which is similar to the structures of those PmoD sequences in proteobacterial methanotrophs. Still, the distant connection of the PmoD of Methylacidiphilum sp. IT6 with those of proteobacterial methanotrophs demonstrated using the phylogenetic tree analysis, together with the results from the multiple alignment analysis, raised an intriguing question on the functions of this protein, rather than the copper-binding capacity. To date, there have been no reports on the ability of Methylomonas sp. DH-1 to convert acetone to acetol under normal culture conditions. As the whole-cell biocatalyst using propane and 2-propanol as substrates, acetone was the sole product detected [ 17 ]. Thus, by integrating the novel protein PmoD of strain Methylacidiphilum sp. IT6 into the methanotrophic Methylomonas sp. DH-1, we conducted the whole-cell biocatalyst assays to evaluate the coupling activity of PmoD with pMMO of Methylomonas sp. DH-1 to enable the production of acetol from acetone. In this experiment, the reactions of DH-1_WT showed traces of acetol produced from acetone which was unexpected at first. Although Methylomonas sp. DH-1 contains cytochrome P450, an enzyme proved to have the capacity to convert acetone to acetol [ 10 , 11 ], the gene expression level in Methylomonas sp. DH-1 was low with 322 RPKM (reads per kilobase of transcript per million reads mapped) in comparison with 953,554 RPKM of pmoC subunit of pMMO under methane cultured conditions [ 17 ]. Thus, a high concentration of acetone of 10 g/L, in this case, could trigger cytochrome P450 activity in Methylomonas sp. DH-1 to convert acetone to acetol. The mutant DH-1_ΔP450 was deployed in the whole-cell biocatalyst conditions to testify our hypothesis. Under the optimal conditions, although the acetone concentration was reduced, there is no trace of acetol detected in these reactions (data not shown), which reinforces for our conclusion on the role of cytochrome P450 in acetol production. This study is the first to demonstrate the acetone oxidation capacity in type I methanotroph. After the optimization of acetone and cell mass concentrations, the acetol titer achieved in whole-cell biocatalyst is 18.291 mM (1.35 g/L) which is much higher than that in the E. coli expressing MimABCD cluster (1.02 mM under whole-cell conditions) [ 6 ] or even comparable with the result in METabolix Explorer’s patent (1.63 g/L acetol produced from glucose in E. coli ) [ 2 ]. These results demonstrated that the novel protein PmoD was successfully expressed in the Methylomonas sp. DH-1, followed by an enhancement in the oxidation of acetone to acetol. The recent discovery of the acetone oxidation obtained in the copper membrane methane monooxygenase homolog PmoCAB3 in Methylacidiphilum sp. IT6 [ 12 ]. Considering the similar reaction mechanism as PmoCAB3 a homolog of pMMO, the role of novel protein PmoD was evaluated in this study by co-expressing with the pmo operon of Methylomonas sp. DH-1 led to a boost in acetol production from acetone. Together with the bioinformatics analyses, we proposed that PmoD acts as a transport protein to catch and deliver acetone to the active site of the complex to allow the oxidation of acetone. This could explain its similar expression patterns together with that of the proposed acetone monooxygenase in strain Methylacidiphilum sp. IT6. However, further experiments are needed to elucidate the activity of PmoD and its specific mechanism in assisting the oxidation reaction of pMMO complexes. Although we recorded the conversion of acetone to acetol under the whole-cell biocatalyst conditions in DH-1 wild type, the effects of acetone on living cells under batch conditions may be more complicated, and the activity of cytochrome P450 may not be sufficient to detoxify a high acetone concentration. Furthermore, the inhibition of cell growth by acetone could result from various effects on cellular physiology. The adverse effects and the penetration mechanism of acetone into the bacterial cell membranes were elucidated in previous studies [ 28 , 29 ]. The acetone effects on the structure and dynamics of bilayer lipid membranes were intensively studied using molecular dynamics simulations. The study demonstrated a favorable tendency of acetone passively penetrating to the hydrophobic regions of the membrane with the potential energy of − 3.6 kcal/mol. A high concentration of acetone caused adverse effects on the phospholipid membrane organization and fluidity [ 28 ]. Another study showed that the growth of bacterial strains was hindered by the adverse effects of organic solvent, specifically acetone, on the fluidity of cell membranes, thereby affecting the supply of nutrients to the cell [ 29 ]. This led to the hypothesis that the DH-1_IT6 recombinant strain could bypass the inhibitory effects of acetone by converting acetone into other substrates, such as acetol, which is less toxic to bacterial cells. Furthermore, the acetol produced was only detected during the first 3 days after the addition of acetone. Subsequently, although acetone titer continued to decrease to the lowest titer of 1.259 g/L at 168 h, there was no trace of acetol detected in any reactions of both wild-type and recombinant strains. In comparison with the whole-cell biocatalyst experiment, the highest produced acetol titer in batch conditions was nearly 12-fold less (1.443 mM in comparison with 18.291 mM). These different titers should be in consideration of the differences in cell mass concentrations, sampling times, and cultivation conditions. The disappearance of the acetol titer in the following cultivation days could have two possible explanations. The decrease in the acetone titer may have led to a decrease of the acetol titer produced to untraceable levels. However, we observed a decrease in the acetol titer before its disappearance which also indicates the possibility that the acetol converted from acetone to detoxify the inhibitory effects of acetone was further assimilated by Methylomonas sp. DH-1. The hypothesis was reinforced with the data obtained from intracellular pyruvate accumulation. While being cultivated with methane and acetone, the recombinant strains DH-1_IT6 showed a significantly higher amount of pyruvate accumulated inside the cells (4.39 µmol/gDCW in DH-1_IT6 compared to 1.20 µmol/gDCW in DH-1_WT or 1.02 µmol/gDCW in DH-1_ ΔP450), which implies for their ability to assimilate acetol into pyruvate for further metabolism. As mentioned in a previous study, acetol could be converted to methylglyoxal by acetol dehydrogenase, which could be further converted into pyruvate [ 30 ]. The proposed pathway for acetone assimilation in strain IT6 was discussed in a previous study based on the genomic and transcriptomic data [ 12 ]. Their proposed pathway was based on the one described in Methylocella silvestris BL2 [ 11 ], in which glucose–methanol–choline (GMC) oxidoreductase could convert acetol to methylglyoxal, which was further converted to lactate by vicinal oxygen chelate proteins (glyoxalases I and II). Further conversion of lactate to pyruvate, an important intermediate for cellular metabolism, was catalyzed by Fe–S- and FAD-binding motif-containing proteins, which could compose a protein with similar activity as the FAD-dependent lactate dehydrogenase [ 12 , 31 ]. About Methylomonas sp. DH-1 genome, all necessary component proteins of the discussed pathway were able to be identified (Fig. 5 ), especially genes coding for 7 candidate glyoxalases exist in the genome of Methylomonas sp. DH-1 (See the Additional file 1 : Table S3). Although a possible pathway for the assimilation of acetone or even propane exists in Methylomonas sp. DH-1, no reports on the growth of Methylomonas sp. DH-1 on these C3 substrates was reported. This could be due to the toxicity of the intermediates of the pathway, such as acetone and lactate, which could suppress the growth of Methylomonas sp. DH-1 [ 21 ]. With the expression of PmoD and coupling activity of PmoD and pMMO for acetone oxidation, although Methylomonas sp. DH-1 cannot solely assimilate for growth, propane or other C3 intermediates (2-propanol and acetone) could be metabolized together with methane as an auxiliary energy source for Methylomonas sp. DH-1 metabolism via acetol conversion. However, further studies are required to deeply scrutinize this proposed pathway and its potential for future applications. Fig. 5 Proposed acetone assimilation pathway in Methylomonas sp. DH-1. A Organization of related genes in the proposed acetone assimilation pathway on the chromosome of the methanotrophic Methylomonas sp. DH-1. B Particulate methane monooxygenase (pMMO; AYM39_RS18590, AYM39_RS18595, and AYM39_RS18600) co-expressed with novel PmoD (IT6_09410) of Methylacidiphilum sp. IT6 oxidizes acetone to acetol, which could be further oxidized to methylglyoxal by GMC oxidoreductase (AYM39_RS20165). Passive diffusion or active transportation via transporter protein (ABC transporter) may be responsible for the methylglyoxal transportation into the cytoplasm. Seven candidate genes (AYM39_RS20250, AYM39_RS03110, AYM39_RS04650, AYM39_RS09045, AYM39_RS14435, AYM39_RS16860, AYM39_RS21550) annotated as glyoxalase enzymes may convert methylglyoxal into S-lactoylglutathione, which can be further converted into lactate by hydroxyacylglutathione hydrolase (AYM39_RS06120) and then to pyruvate by putative lactate dehydrogenase (AYM39_RS04895, AYM39_RS13600). The generated pyruvate can be either converted to acetyl-CoA to enter the TCA cycle or to phosphoenolpyruvate (PEP) by PEP-synthase (AYM39_RS13170). The related genes are shown using their locus tags obtained from the genome of Methylomonas sp. DH-1 (GenBank accession number NZ_CP014360). The related reactions of the proposed pathway are demonstrated by dashed arrows"
} | 3,870 |
35539103 | PMC9078405 | pmc | 7,103 | {
"abstract": "Recent developments of self-powered devices and systems have attracted much attention. Lead zirconate titanate (PZT) has been regarded as one of the most promising materials for building high-performance nanogenerators. Herein, vertically aligned PZT nanorod arrays were synthesized on a pre-oxidized Ti substrate in the presence of a surfactant by a one-step hydrothermal method. The PZT nanorod arrays consist of an initial layer of a PZT film and well aligned nanorods with (001)-orientated tetragonal single crystalline structures. The PZT nanorods exhibited a high piezoelectric response with a d 33 value of up to 1600 pm V −1 . A piezoelectric energy harvester was fabricated based on the PZT nanorod arrays, which exhibited outstanding energy harvesting performance with an open-circuit output voltage of 3.3 V and 8 V when the devices were pressed by a compressive 10 N force and a finger tapping motion, respectively. Moreover, the average power density generated by those two mechanical stimulations were up to 3.16 and 5.92 μW cm −2 with the external load of 1 MΩ.",
"conclusion": "Conclusions In this work, aligned PZT NRAs with high aspect ratio and uniform distribution were synthesized on pre-oxidized Ti substrates by using a hydrothermal method. The polymer surfactant plays a vital role in the size and morphology of the arrays. The nanorods were confirmed to be the (001) oriented with single-crystal tetragonal perovskite structure. The NRs exhibit high piezoelectric constant along the radial direction, which is ∼1600 pm V −1 . The energy harvesters based on the PZT NRAs could generate impulsive voltage with open-circuit value up to 3.3 V when the device was vertically pressed by a compressive force of 10 N at 10 Hz. The average output power density was up to 3.16 μW cm −2 when the load resistance is 1 M. Those output performance parameters were increased to 8 V and 5.92 μW cm −2 when the devices were knocked by a human finger. The outstanding energy harvesting performance of the PZT NRAs can provide many opportunities for the application of self-powered nano-devices and systems.",
"introduction": "Introduction Rapid development of micro and nano-electronic technology has promoted the research and application of multi-functional personal electronics, wearable devices and smart sensor systems, etc. 1,2 Although the power consumption of such devices and systems has been reduced to a much lower level, the problem of long-term power supply in such miniaturized systems is still limiting their further development and application. 3,4 For instance, the dying battery has become one of the most serious puzzlements of smart phone users. Moreover, the recharging of sensor batteries used in untraversed environments such as underwater or enclosed conditions may also be a great challenge to engineers. Therefore, the research and development of novel types of electrical powering units with miniaturized dimensions, long lifetimes, good long-term stabilities but no recharging problems has become an attractive topic. In recent years, the energy harvesters based on the photovoltaic, thermoelectric, electrochemical and piezoelectric effects have attracted great attentions due to their capability to convert the optical, thermal, chemical and mechanical energies into electrical output. 5–8 Such devices can be utilized for building the self-powered systems, which can independently operate by harvesting the energies in the ambient environment without any external electrical powering systems. Among the various kinds of energies, the mechanical energies such as the air flow, vibration and object movement have wide distributions and low limitation by the environmental conditions, which can be harvested and converted into electricity by using piezoelectric materials. 9–15 Z. L. Wang's group firstly reported the micro-scaled piezoelectric energy harvester based on ZnO nanowire arrays in 2006. 16 After that, several kinds of nanowire-based energy harvesters have been demonstrated, including GaN, BaTiO 3 , (K,Na)NbO 3 (KNN) and Pb(Zr,Ti)O 3 (PZT) nanowires. 12,17–23 Among them, PZT exhibited much better piezoelectric performance than other materials, including high piezoelectric constant ( d 33 ) and electromechanical coupling coefficient. 24–31 For example, Chen et al. has demonstrated a piezoelectric energy harvester based on the electrospun PZT nanofibers with output voltage of 1.6 V. 25 However, the random alignment of the nanofibers limited the output performance of those devices. By integrate the PZT nanofibers into vertically aligned nanofiber arrays, Gu et al. obtained an ultra-high output piezoelectric energy harvester with output voltage of 209 V. 27 Comparing with the polycrystalline nanofibers, the single-crystal nanowires always exhibited much better piezoelectric performance. 32 Lin and co-authors have demonstrated a hydrothermal growth of [110]-oriented vertically aligned PZT nanowire arrays on TiO 2 film. 33 However, there are no further investigation on the piezoelectric property of the nanowire arrays. Moreover, Xu et al. has reported a piezoelectric energy harvester based on the chemical epitaxially grown PZT nanowire array on SrTiO 3 (STO) single-crystal substrates in 2010. 28 However, the output voltage of such devices is lower than 1.0 V, which is much lower than the nanofiber-based devices and shows no obvious superiority than the other kinds of piezoelectric materials such as ZnO and BaTiO 3 . 18,23,34 After that, few reports have been focused on either the synthesis or the piezoelectric performance of the single-crystal PZT nanowire arrays, limiting the application of these excellent piezoelectric materials in the micro-scaled energy harvesting devices. In this work, vertically aligned single-crystal PZT nanorod arrays (NRAs) were synthesized on a pre-oxidized titanium foil by poly(vinyl alcohol) (PVA)-assisted hydrothermal process. The as-synthesized [001]-oriented PZT nanorods exhibited high piezoelectric constant up to 1600 pm V −1 . The as-fabricated vertically aligned nanogenerator (VING) exhibited outstanding energy harvesting performance with open-circuit output voltage ( V OC ) up to 3.3 V under the compressive force of 10 N. The maximal output power density can reach 3.16 μW cm −2 with an external load resistance of 1 MΩ. The outstanding energy harvesting performance of the PZT NRAs provides great potential for the application in building high-performance self-powered systems.",
"discussion": "Results and discussion \n Fig. 1 shows the FE-SEM images of the as-synthesized PZT products with different contents of PVA additives in the hydrothermal precursors. As shown in Fig. 1(a) , no nanorods was formed on the surface of the substrates with the PVA content of 0.10 g. The surface of the product was consisted of condensed film with column-like grains. Moreover, the samples which were not attached on the substrate were consisting of cubic and amorphous particles, as shown in Fig. 1S. † When the PVA contents were lower than 0.10 g, the product was consisted of aligned PZT nanorods standing on the surface of the substrates. With the PVA content decreasing from 0.08 to 0.04 g, the estimated average diameter of the nanorods was ∼200, 400 and 1180 nm, respectively. As reported, the growth of PZT nanorods should be attributed to the adsorption of PVA on the (100) faces of the tetragonal perovskite structures due to the hydrogen bond under the high alkaline hydrothermal condition, which can suppress the growth of the adsorbed surface. 35 As a result, the decrease of PVA content may lower down the limitation of the growth along the radial direction, resulting in the increase of the diameter of the nanorods. Fig. 1 The FE-SEM images of the products with different contents of PVA additives in the hydrothermal precursors. (a) 0.10 g; (b) 0.08 g; (c) 0.06 g; (d) 0.04 g. \n Fig. 2 shows the XRD patterns of the as-synthesized PZT products. All diffraction peaks can be indexed to the tetragonal perovskite phase of PZT and agree with the diffraction data of PbZr 0.52 Ti 0.48 O 3 in JCPDS card no. 33-0784. The sharp peaks indicate that the samples are well crystallized. The diffraction peaks at 27.4° belongs to the (110) face of TiO 2 layer on the substrate according to the JCPDS card no. 21-1276, while peaks at 25.1° and 53° belong to the (100) and (102) faces of the Ti substrate according to the JCPDS card no. 65-9622, respectively. Moreover, the intensity ratio between the (001) and (100) peaks of PZT was increased when the PVA content was decreased from 0.10 to 0.04 g. The results suggested the [001]-oriented growth of the PZT NRAs on the film, which could be further proved by the cross-section SEM image of the NRAs and the TEM characterization results. Fig. 2 The XRD pattern of the as-synthesized products with different PVA contents in the precursors. (a) Substrate; (b) 0.10 g; (c) 0.08 g; (d) 0.06 g; (e) 0.04 g. As shown in Fig. 3 , a layer of PZT film was formed at the initial stage of the hydrothermal process before the growth of the NRAs, which should be attributed to the reaction of the TiO 2 layer on the pre-treated Ti substrates with the hydrothermal precursors. Fig. 4 shows the TEM characterization of an individual PZT nanorod. According to the TEM image and the SAED pattern shows in Fig. 4(a) , the PZT nanorod with ∼200 nm in diameter and 6 μm in length exhibited single-crystalline tetragonal structure. The clear lattice fringes with spacing of 0.40 nm in the HRTEM image shown in Fig. 4(b) , which corresponding to the red area in Fig. 4(a) , confirmed the [001] growth orientation of the PZT nanorod, which agreed with the XRD results. Fig. 3 The cross-section SEM image of the PZT NRAs with 0.1 g PVA. Fig. 4 The TEM results of the PZT nanorod with 0.08 g PVA. (a) TEM image and SAED patterns. (b) HRTEM image. The radial piezoelectric response was measured by using PFM to evaluate the piezoelectric performance of the PZT nanorod. Firstly, the PZT nanorods were dispersed into ethanol and then transferred onto the surface of a piece of Au-coated silicon substrate. After been dried for 3 h at 60 °C, the sample was positioned onto the PFM holder to test the radial piezoelectric response of the nanorods. Fig. 5(a) shows a three-dimensional (3D) topography of an individual PZT nanorod, which is 2–3 μm in length and 200 nm in diameter, respectively. As shown in Fig. 5(b) , the nanorod exhibited obvious symmetrical butterfly-shaped piezoelectric response loop, with maximum deformation of ∼600 pm under bias of 1.5 V. The phase loop along with abrupt changes is shown in Fig. 5(c) , in which the intensive peak at ∼0.3 V was induced by a noise signal during the testing process. The results confirmed the spontaneous polarization behaviour of the PZT nanorod. Fig. 5(d) shows the piezo-response obtained as M = A cos θ , where M , A and θ represented the piezo-response, amplitude and phase angle, respectively. According to these curves, the maximum deformation was ∼600 pm with phase switching of 180°. In addition, the piezoelectric constant could be calculated from the slope of the liner region of the curve. After the calibration by using a standard sample of LiNbO 3 with a nominal d 33 value of 17.3 pm V −1 , the average piezoelectric constant d 33 could be calculated to be ∼1600 pm V −1 . 36,37 The much higher d 33 of PZT nanorods suggested their high superiority in building piezoelectric energy harvesters than the ZnO, BaTiO 3 nanowires. Fig. 5 The piezoresponse of PZT nanorod obtained with 0.08 g PVA. (a) The AFM morphology; (b) the amplitude curve; (c) the phase curve; (d) the piezoresponse curve. \n Fig. 6(a and b) show the schematic diagram and an optical photo image of the energy harvester based on the PZT NRAs. After the hydrothermal growth of the NRAs, a thin layer of PMDS was spin-coated to fill the interspace of the NRAs. Then a layer of Au electrodes of 100 nm in thickness was deposited on the top surface of the coated NRAs. After wire-leading, the whole device was packaged by the PDMS polymer to prevent the piezoelectric materials from physical damage and chemical interference. An outstanding energy harvesting performance can be observed when the device is subject to the axial pressure. According to the piezoelectric theory, the piezoelectric crystal under strain state will generate a piezoelectric field along the polarization direction, which will lead to a transient flow of free electrons in the external circuit to compensate the piezoelectric potential. Thus, an impulsive output voltage signal will be detected. In order to evaluate the device performance, a compressive force with constant amplitude of 10 N and frequency of 10 Hz was applied by using DMA in compressive mode. As shown in the Fig. 6(c) , the PZT NRA-based energy harvester can generate high electrical output with V OC up to 3.3 V. Moreover, much higher open-circuit voltage (up to 8 V) can be generated when the device was knocked by human fingers (as shown in Fig. 2S † ). To further confirm that PZT nanorods is the origin of the energy harvesting behaviour, a similar device without PZT NRAs but only the substrate and packaging layers was also fabricated. It can be seen in Fig. 6(d) that the device without PZT nanorods could generate much lower output voltage with amplitude of 0.14 V, which could be attributed to the release of electrostatic charges due to the capacitance effect of the devices. Therefore, the output voltage generated by the PZT-based devices could mainly be attributed to piezoelectric effect of PZT NRAs. Fig. 6 The (a) schematic diagram, (b) photo image and (c and d) output voltage signal generated by harvesting the mechanical energy when the devices were vertically pressed by periodical tapping. It is well accepted that the property of a power device lies on the capacity for driving the external loads. Fig. 7(a) shows the detected voltage signal applied on the external load resistance, which was varying from 10 kΩ to 1 GΩ. The voltage amplitude increased with the increasing load resistance. However, the signal is unrecognizable when the resistance is lower than or equal to 100 kΩ, which should be attributed to the high internal resistance of the energy harvesters (∼3 MΩ). Fig. 7(b) shows the average voltage and power density of the devices with varying load resistance under the applied force of 10 N. The average power density ( P L ) was calculated according to the method reported by the previous works 25,38 as where T is the time period (0.1 s), R is the load resistance and V O ( t ) is the output voltage, respectively. As shown, the peak of output voltage can reach ∼3 V when the resistance is 1 GΩ. The highest average power density was ∼3.16 μW cm −2 when the resistance is 1 MΩ. Moreover, Fig. 3S † shows the output performance of the devices under finger tapping. The highest V OC was up to ∼8 V, while the average power density was up to ∼5.92 μW cm −2 with the resistance of 1 MΩ. Table 1 lists the reported works about the PZT-based nanogenerators. The device in this work shows higher open-circuit voltage than most of the devices in the previously reported works except the device basing on the nanofiber arrays with integrated electrical output. The outstanding energy harvesting performance with high output voltage and power density indicated the great potential of the PZT NRAs for the application of high-performance piezoelectric energy harvesters and self-powered systems. Fig. 7 (a) The voltage derived under different load resistance. (b) The variation of open-circuit output voltage and average power density with the load resistance. The comparison of output performance of the PZT-related nanogenerators Sample \n V \n OC (V) Maximum power (μW) Maximum power density Mechanical stimulation Reference Single PZT nanofiber 0.65 — Bending by nanomanipulator \n 24 \n PZT nanofibers 1.63 0.03 — Strain of 10% at 250 rad s −1 \n 25 \n PZT nanofiber arrays 209 — — Free-falling object \n 27 \n PZT NRAs 0.7 2.8 mW cm −3 1–50 Hz \n 28 \n PZT ribbons 0.25 0.01 Finger tapping \n 39 \n PZT NRAs 3.3 0.79 3.16 μW cm −2 Compressive force of 10 N at 10 Hz This work 8 1.48 5.92 μW cm −2 Finger tapping"
} | 4,069 |
36739554 | PMC10952891 | pmc | 7,104 | {
"abstract": "Summary \n Most plants form mycorrhizal associations with mutualistic soil fungi. Through these partnerships, resources are exchanged including photosynthetically fixed carbon for fungal‐acquired nutrients. Recently, it was shown that the diversity of associated fungi is greater than previously assumed, extending to Mucoromycotina fungi. These Mucoromycotina ‘fine root endophytes’ (MFRE) are widespread and generally co‐colonise plant roots together with Glomeromycotina ‘coarse’ arbuscular mycorrhizal fungi (AMF). Until now, this co‐occurrence has hindered the determination of the direct function of MFRE symbiosis. To overcome this major barrier, we developed new techniques for fungal isolation and culture and established the first monoxenic in vitro cultures of MFRE colonising a flowering plant, clover. Using radio‐ and stable‐isotope tracers in these in vitro systems, we measured the transfer of 33 P, 15 N and 14 C between MFRE hyphae and the host plant. Our results provide the first unequivocal evidence that MFRE fungi are nutritional mutualists with a flowering plant by showing that clover gained both 15 N and 33 P tracers directly from fungus in exchange for plant‐fixed C in the absence of other micro‐organisms. Our findings and methods pave the way for a new era in mycorrhizal research, firmly establishing MFRE as both mycorrhizal and functionally important in terrestrial ecosystems.",
"introduction": "Introduction Among Earth's most important symbioses are the ancient plant–fungus partnerships known as mycorrhizas, or ‘mycorrhiza‐like’ associations in plants without roots (Read et al ., 2000 ). These mutualisms, underpinned by the bidirectional exchange of plant‐fixed carbon for fungal‐acquired mineral nutrients (Raven & Allen, 2003 ), were instrumental in plant terrestrialisation > 500 million years ago (Morris et al ., 2018 ) by facilitating early, rootless plant access to mineral nutrients held within primeval soils (Pirozynski & Malloch, 1975 ; Field et al ., 2015a ). Thus, fungi played a formative role in the development of Earth's terrestrial ecosystems and climate through their contributions to global carbon and nutrient cycles (Taylor et al ., 2011 ; Mills et al ., 2018 ). Today, these associations are formed between most land plants and a diverse subset of soil fungi (Field & Pressel, 2018 ), including the widespread arbuscular mycorrhizal fungi (AMF) in the subphylum Glomeromycotina (Brundrett & Tedersoo, 2018 ), estimated to occur in > 70% of plants (Smith & Read, 2008 ). Until recently, AMF encompassed the globally distributed ‘fine root endophytes’ (FRE; Glomus tenue ; Thippayrugs et al ., 1999 ), which are known to colonise several vascular plant families (Ali, 1969 ; Abbott, 1982 ; Thippayrugs et al ., 1999 ), but have been largely overlooked due to practical limitations in molecular detection and inability to study them apart from coexisting AMF (Orchard et al ., 2017a ). Through improved molecular detection and identification (Bidartondo et al ., 2011 ), it is now clear that FRE are distinct from Glomeromycotina AMF, belonging instead to the Endogonales in the subphylum Mucoromycotina (Orchard et al ., 2017a ), and recently renamed Planticonsortium tenue (Walker et al ., 2018 ). Thus, FRE previously reported in flowering plants (Ali, 1969 ; Abbott, 1982 ; Thippayrugs et al ., 1999 ) are likely closely related to Endogonales fungal associates previously identified (Bidartondo et al ., 2011 ; Rimington et al ., 2015 ; Hoysted et al ., 2018 , 2019 ) and shown to be mutualistic in terms of carbon‐for‐nutrient exchange in a range of nonflowering plants, albeit using nonsterile, soil‐based experimental systems (Field et al ., 2015b , 2016 , 2019 ; Hoysted et al ., 2019 , 2020 ). Colonisation by FRE is, like AMF, generally characterised by the presence of arbuscules and arbuscule‐like structures (Orchard et al ., 2017b ), while the small diameter of FRE hyphae (< 1.5 μm), with small swellings and ‘fan‐like’ morphologies, is considered a distinctive trait that separates them from AMF (or ‘coarse root endophytes’) which consistently develop wider (> 3 μm in diameter) hyphae and larger vesicles (Orchard et al ., 2017a ). Morphological plasticity has been noted in transmission and scanning electron micrographs of the ultrastructure of Mucoromycotina FRE (MFRE) exclusively associating with liverworts (Field et al ., 2015b , 2016 ) and a vascular plant (Hoysted et al ., 2019 ). Most recently, cryo‐SEM has confirmed uniformly thin hyphae and hyphal ‘ropes’ as potential diagnostic features of MFRE symbioses (Albornoz et al ., 2020 ). Differently from the strictly biotrophic AMF, MFRE are considered facultative saprotrophs as it has been possible to isolate them from host plants – both a nonvascular (Field et al ., 2015b ) and vascular plant (as shown herein) – and to grow them axenically, that is without a host in culture. Latest research indicates that MFRE play a nutritionally complementary role to AMF by facilitating plant nitrogen (N) assimilation alongside AMF‐facilitated plant phosphorus (P) acquisition through co‐colonisation of the same host (Field et al ., 2019 ). Such functional complementarity is further supported by the observation that MFRE transfer significant amounts of 15 N but relatively little 33 P isotope tracers to a host lycophyte in the first experimental demonstration of MFRE nutritional mutualism with a vascular plant (Hoysted et al ., 2019 ). The apparent ability of MFRE, but not AMF, to transfer N also from organic sources to host liverworts in nonsterile soil (Field et al ., 2019 ) together with their presumed facultative saprotrophic nature points to possible functional similarities with ectomycorrhizal fungi, an assumption in line with results from a recent network analysis of symbiotic fungal associations in liverworts (Rimington et al ., 2019 ). However, evidence for the precise role of MFRE, in the absence of other soil micro‐organisms, remains equivocal. To date, detailed research into plant–MFRE associations has been constrained by a lack of in vitro experimental systems that allow indisputable determination of the direct function of the MFRE symbiosis in isolation. Our recent knowledge of MFRE function has been derived largely from experiments using wild soil‐based systems and wild‐collected plants that naturally only associate with MFRE (Field et al ., 2015b , 2016 , 2019 ; Hoysted et al ., 2019 , 2020 ), or from soil culture‐based experimental pots (Orchard et al ., 2017a ; Albornoz et al ., 2020 ). Each of these methods has significantly enhanced our understanding of MFRE form and function, shedding new light on the importance of MFRE associations in nature, and remains useful in the studies of plants associating with mixed microbial communities. However, the development of in vitro experimental systems capable of distinguishing between fungal symbionts in the absence of other soil biota is now critical for the functional significance of MFRE associations to be fully defined (Sinanaj et al ., 2021 ). This is particularly important as evidence is increasingly pointing towards most plants forming simultaneous symbioses with AMF and MFRE (Field et al ., 2015a , b , 2016 ; Hoysted et al ., 2019 , 2020 ) and there being complementarity in function between symbionts (Field et al ., 2019 ; Hoysted et al ., 2019 ). Recently, it was reported that a free‐living Mucoromycotina, Gongronella sp. W5, utilises plant sucrose as a carbon source (Wang et al ., 2021 ); however, data showing mutualistic transfer of carbon‐for‐nutrients between MFRE and a plant host in the absence of other micro‐organisms do not currently exist. The development of an in vitro experimental system is critical to achieve this and further understand function, development and signalling and to identify specific symbiotic structures and interfaces in MFRE, particularly for comparisons with model AM symbioses. Here, we resolve this research challenge by establishing experimentally tractable, monoxenic symbiotic cultures of MFRE and white clover ( Trifolium repens ), a flowering plant genus used in other recent studies of MFRE colonisation (e.g. Orchard et al ., 2017a ; Albornoz et al ., 2020 ), albeit in nonsterile systems. Using radio‐ and stable‐isotope tracers in our in vitro systems, we measured the transfer of 33 P, 15 N and 14 C between MFRE hyphae and the host plant to provide the first unequivocal evidence of mutualistic transfer of MFRE‐assimilated nutrients for plant‐fixed carbon with a flowering plant, in the absence of other microbes.",
"discussion": "Discussion Our results provide the first unequivocal demonstration that symbiosis between a flowering plant, white clover, and a MFRE fungus is mutualistic, with the plant gaining both 15 N and 33 P tracers directly from the fungus while the fungus gains plant‐fixed C, in the absence of other micro‐organisms. While we focussed on a specific association, when considered together with those of previous studies (Field et al ., 2015b , 2016 , 2019 ; Orchard et al ., 2017b ; Hoysted et al ., 2019 , 2020 ; Albornoz et al ., 2020 , 2021 ), our findings indicate that MFRE symbioses are nutritionally mutualistic across diverse land plants. Analysis of the fungus colonising the roots of clover, confirmed molecularly as an Endogonales (Mucoromycotina) isolate from a lycophyte (Hoysted et al ., 2019 , 2020 ), revealed a morphology characterised by fine, irregularly branching hyphae with small intercalary and terminal swellings (Fig. 1g,h ), forming vesicles or spores (Fig. 1g ) as well as hyphal coils and hyphal ropes (Fig. 1h ) alongside arbuscule‐like structures (Fig. 1i ). This morphology matches that described previously in a range of vascular (Orchard et al ., 2017b ; Hoysted et al ., 2019 , 2020 ; Albornoz et al ., 2020 ) and nonvascular plants (Field et al ., 2015b , 2016 , 2019 ) colonised by MFRE. To date, MFRE research has been carried out using unpasteurised soil culture‐based experimental systems (Orchard et al ., 2017a ; Albornoz et al ., 2020 ) or wild‐collected plants (Field et al ., 2015b , 2016 , 2019 ; Hoysted et al ., 2020 ). As such, it is inevitable that these experiments included other soil micro‐organisms alongside MFRE. As there is no information about how rhizosphere bacteria may influence MFRE metabolic characteristics and function, the inclusion of soil micro‐organisms in previous studies made it impossible to determine the direct contribution of MFRE to host plant nutrition. Here, we show for the first time, using a novel in vitro monoxenic system, that MFRE directly assimilate and transfer both 33 P and 15 N to a flowering plant in the absence of other microbes. Our results reveal significant transfer of fungal‐acquired 33 P to clover; however, while we observed a clear trend of greater [ 15 N] in plant shoots when MFRE hyphae are intact than where they are severed, the difference is not statistically significant. A possible explanation is that clover, a legume, is not heavily reliant on fungal symbionts for N assimilation, even in the absence of rhizobia. This could also be due to revolatilisation and recapture of ammonium by the plant, and/or mass flow driven by plant transpiration in these microcosms. Nevertheless, and although the abundance of MFRE in our microcosms was not quantified, our results, when tentatively compared with those of previous studies on AMF (Thirkell et al ., 2020a , b ), indicate that clover (without rhizobia) may assimilate more 15 N tracer via its MFRE symbiont per unit plant biomass than is typically assimilated by plants associated with AMF, albeit in nonsterile systems. This nutritional role, already indicated by studies of MFRE symbioses in nonflowering plants (Field et al ., 2019 ; Hoysted et al ., 2019 ), could help to explain the persistence of MFRE across most modern land plant lineages, facilitating plant access and assimilation of soil N (Howard et al ., 2022 ). The FRE have long been thought to enhance plant P, at least in soils with very low plant‐available P (Crush, 1973 ; Rabatin et al ., 1993 ; Orchard et al ., 2017b ; Albornoz et al ., 2021 ); however, their potential role in plant N uptake has been overlooked. It is therefore important that N transfer and assimilation from MFRE to plant hosts are now investigated in vitro across a range of other flowering plants that do not associate with N‐fixing bacteria. Parallel studies of N and P transfer by AMF are also required before meaningful comparisons between AMF and MFRE function can be made. Because we used an inorganic N source, additional experiments are also needed to assess potential direct organic N utilisation by MFRE (Field et al ., 2019 ). A recent study on the ability of AMF to utilise N from organic sources (Rozmoš et al ., 2022 ) showed, using an in vitro monoxenic experimental system based on Ri T‐DNA transformed chicory roots, that organic nitrogen utilisation by Rhizophagus irregularis was mediated by specific soil bacteria and accelerated by the presence of a protist. These findings, though not based on full plants, may explain the results of previous experiments using organic matter patches labelled with 15 N in soil‐based microcosms, which showed successful transfer of the 15 N by AMF to host plants (Hodge et al ., 2000 , 2001 ; Thirkell et al ., 2016 ). Thus, it is likely that AMF‐associated and free‐living rhizospheric bacteria as well as other soil fungi contained in the nonsterile fungal inoculum used in those experiments may have interacted with AMF (Vivas et al ., 2003 ; Frey‐Klett et al ., 2007 ; Smith & Smith, 2011 ; Jiang et al ., 2021 ), influencing the breakdown, mineralisation and assimilation of 15 N‐labelled organic material. It is now important to determine whether similar processes may also explain organic N utilisation by MFRE (Field et al ., 2019 ) or whether these fungi, by virtue of their putative facultatively saprotrophic nature, can directly access and transfer N from organic sources. Since organic N represents a large proportion of total soil N, direct organic N utilisation by MFRE would have important implications for terrestrial N cycling (Hodge & Storer, 2015 ; Howard et al ., 2022 ). Further research using monoxenic systems is also needed to compare the ‘cost’ in terms of plant‐to‐fungus transfer of C between AMF and MFRE symbioses. Our data show that symbiotic MFRE gain clover‐fixed C (Fig. 4a,b ), and previous experiments using soil culture‐based systems suggested that the ‘cost’ of MFRE‐vascular plant associations is at least on a par with, if not larger than, AMF‐vascular plant associations (Hoysted et al ., 2020 ). However, this has only been compared in one vascular plant species; whether it holds true for in vitro monoxenic systems remains to be tested. Previous research into the function of plant–MFRE symbioses raised fundamental questions about their persistence and ecological relevance in modern terrestrial ecosystems. We can now begin to address such questions with new experimental systems knowing that symbiotic MFRE are nutritionally mutualistic with flowering plants. Field et al . ( 2015b ) demonstrated the ability of MFRE isolates to recolonise host liverworts in vitro ; however, until now, this had not been achieved in vascular plants. Our in vitro system used a fungal isolate that originated from a wild‐collected early‐diverging vascular plant, L. inundata , and was introduced to white clover in vitro . This isolate was molecularly and cytologically confirmed to be colonising the roots of clover used in our experiment. This represents a novel, tractable in vitro experimental system designed to manipulate MFRE isolates and the resynthesis of their mycorrhizas with a flowering plant. It opens a realm of exciting possibilities for further research on MFRE mycorrhizal properties, including cytological, molecular and metabolomic comparisons with AMF where host plants are inoculated singly or co‐colonised with both MFRE and AMF. Furthermore, a fundamental understanding of how MFRE distribution and function are affected by environmental factors such as temperature, water, light and atmospheric CO 2 , in addition to biotic factors such as interactions with other soil microbiota, can now be developed. Successful isolation from wild plants and axenic cultivation of an MFRE isolate offers exciting new opportunities to develop a model system for symbiotic MFRE and for omics in comparison with other fungi."
} | 4,181 |
23258841 | PMC3595025 | pmc | 7,105 | {
"abstract": "An open question regarding the evolution of photosynthesis is how cyanobacteria came to possess the two reaction center (RC) types, Type I reaction center (RCI) and Type II reaction center (RCII). The two main competing theories in the foreground of current thinking on this issue are that either 1) RCI and RCII are related via lineage divergence among anoxygenic photosynthetic bacteria and became merged in cyanobacteria via an event of large-scale lateral gene transfer (also called \"fusion\" theories) or 2) the two RC types are related via gene duplication in an ancestral, anoxygenic but protocyanobacterial phototroph that possessed both RC types before making the transition to using water as an electron donor. To distinguish between these possibilities, we studied the evolution of the core (bacterio)chlorophyll biosynthetic pathway from protoporphyrin IX (Proto IX) up to (bacterio)chlorophyllide a. The results show no dichotomy of chlorophyll biosynthesis genes into RCI- and RCII-specific chlorophyll biosynthetic clades, thereby excluding models of fusion at the origin of cyanobacteria and supporting the selective-loss hypothesis. By considering the cofactor demands of the pathway and the source genes from which several steps in chlorophyll biosynthesis are derived, we infer that the cell that first synthesized chlorophyll was a cobalamin-dependent, heme-synthesizing, diazotrophic anaerobe.",
"conclusion": "Conclusions The phylogeny and evolution of the chlorophyll biosynthesis core pathway were analyzed both at the gene and pathway level. The lack of coevolution of chlorophyll biosynthesis genes with RCIs or RCIIs permits us to exclude the widely discussed possibility that RCI and RCII diverged via lineage splitting and became reunited in cyanobacteria via a large-scale gene transfer (fusion) event. Moreover, it can be concluded that the primordial photosynthetic organism performed nitrogen fixation, synthesized heme, and was cobalamin dependent.",
"introduction": "Introduction The origin of oxygenic photosynthesis introduced a new high-potential electron acceptor into microbial ecosystems ( Holland 2006 ) and enzymatic reaction mechanisms ( Raymond et al. 2002 ), marking the onset of pivotal changes in geochemical cycles and biochemical pathways. Oxygen first began to accumulate in the atmosphere approximately 2.4 billion years ago, and its subsequent accumulation in the oceans was slower, such that full oxic conditions were only reached approximately 635–580 Ma ( Arnold et al. 2004 ; Lyons 2007 ; Canfield et al. 2008 ; Scott et al. 2008 ; Lyons et al. 2009 ; Sahoo et al. 2012 ). This “new chemistry” had far-reaching impact on the evolutionary process. Chlorophyll-based photosynthesis arose among eubacteria, where it is currently found among six phyla. Chlorobia (green sulfur bacteria [GSB]), firmicutes (heliobacteria), and acidobacteria have anoxygenic Type I reaction centers (RCIs), whereas chloroflexi (green nonsulfur bacteria, GNSB of filamentous anoxygenic phototrophs, FAPs) and some proteobacterial organisms (purple nonsulfur bacteria and purple sulfur oxidizing bacteria [ Dahl et al. 2005 ]) perform anoxygenic photosynthesis with Type II reaction centers (RCIIs) ( Xiong and Bauer 2002a ; Bryant et al. 2007 ; Maqueo Chew and Bryant 2007 ; Blankenship 2010 ). Only cyanobacteria and, via endosymbiosis, photosynthetic eukaryotes ( Margulis and Bermudes 1985 ) perform oxygenic photosynthesis, having both the Photosystem I (PSI) and the Photosystem II (PSII) complexes that are homologous to RCI and RCII, respectively ( Michel and Deisenhofer 1988 ; Hauska et al. 2001 ; Neerken and Amesz 2001 ). Although at the sequence level there is almost no detectable similarity between RCI and RCII, their structural and cofactor arrangements are unquestionably homologous and clearly indicate common ancestry ( Schubert et al. 1998 ; Barber et al. 2000 ; Sadekar et al. 2006 ). Chlorophyll (Chl) and bacteriochlorophyll (Bch) are required for oxygenic and anoxygenic photosynthesis, respectively, where they serve two essential functions. As light-harvesting antennae, they act as photon funnels, absorbing light and channeling its energy to reaction centers (RCs), where chlorophylls perform their second function: photochemical charge separation to create strong oxidants and reductants, sending low-potential electrons through the electron transport chain (ETC). In anoxygenic phototrophs, bacteriochlorophylls are the main pigments. In contrast, chloroplasts and cyanobacteria possess only chlorophylls, which can, however, also be residually present in the RC of some anoxygenic photosynthetic bacteria ( Kobayashi et al. 2000 ; Hohmann-Marriott and Blankenship 2011 ; Sarrou et al. 2012 ; Tsukatani et al. 2012 ). With exception of chlorophyll c , chemically, both chlorophylls and bacteriochlorophylls are chlorins; reduced magnesium-containing cyclic tetrapyrroles with an additional fifth isocyclic ring. Their main differences concern the level of unsaturation affecting the system of conjugated double bonds. Chlorophyll has a single bond between carbons C17 and C18 (IUPAC numbering), and bacteriochlorophyll has an additional C7–C8 single bond ( Maqueo Chew and Bryant 2007 ; Niedzwiedzki and Blankenship 2010 ). There are 11 major types of bacteriochlorophyll and chlorophylls ( Hohmann-Marriott and Blankenship 2011 ) differing with respect to the substituents of the tetrapyrrole ring, hence their differing absorption properties, which allow organisms to specialize toward different spectral niches, most notably as a function of depth in the water column ( Glaeser et al. 2002 ; Manske et al. 2005 ; Gomez Maqueo Chew et al. 2007 ; Kiang, Siefert, et al. 2007 ; Kiang et al. 2007 ; Stomp et al. 2007 ; Chen et al. 2010 ). The enzymes involved in chlorophyll metabolism are the only set of photosynthesis-related proteins common to all phototrophs ( Mulkidjanian et al. 2006 ), and thoughts on chlorophyll evolution have a long history. Central to the topic is the Granick hypothesis ( Granick 1965 ), which posits that the evolution of the chlorophyll biosynthetic pathway followed the sequential inventions of new enzymes to generate more stable products. This premise has been widely used to study the evolution of the chlorophyll pathway, in particular as a proxy for the evolution of photosynthesis itself ( Olson and Pierson 1987a , 1987b ; Xiong and Bauer 2002a , 2002b ; Gupta 2012 ). However, other hypotheses are still discussed, and the issue is debated ( Lockhart et al. 1996 ; Blankenship 2001 ). In 2000, Xiong et al. studied phylogenies for 9 of the 17 enzymes then known to be involved in the (bacterio)chlorophyll pathway. They suggested that (bacterio)chlorophylls first arose within purple bacteria (proteobacteria) and that the pathway’s emergence involved the recruitment and duplication of homologous enzymes such as nitrogenase subunits and cobalt chelatase from cobalamin biosynthesis ( Xiong et al. 1998 ; Xiong and Bauer 2002a , 2002b ). More recently, and based on phylogenies and sequence signatures of the (B)ChlNBL complex, which is responsible for catalyzing the last step of the (bacterio)chlorophyll core pathway, an origin of chlorophyll in the Gram-positive heliobacteria lineage has been suggested ( Gupta 2012 ). Here, we address the evolution of the chlorophyll biosynthesis pathway to distinguish between competing theories for the origin of two photosystems in cyanobacteria. The RCs of photosystems I and II are clearly related at the level of three-dimensional structure ( Nitschke and Rutherford 1991 ; Schubert et al. 1998 ; Baymann et al. 2001 ; Allen and Puthiyaveetil 2005 ; Allen et al. 2011 ; Hohmann-Marriott and Blankenship 2011 ), and the issue is how they came to reside within the ancestral cyanobacterial genome. The “fusion” (or “merger”) hypothesis asserts that the two photosystems diverged during the evolution of anoxygenic photosynthetic lineages and became merged in the founding cyanobacterium via lateral gene transfer (LGT) ( Mathis 1990 ; Meyer 1994 ; Xiong and Bauer 2002a ; Hohmann-Marriott and Blankenship 2011 ). In contrast, the “duplication” hypothesis asserts that the photosystems diverged within a protocyanobacterial ancestor and subsequently underwent vertical inheritance and export, via LGT, to diverse anoxygenic photosynthetic lineages ( Olson and Pierson 1987a , 1987b ; Baymann et al. 2001 ; Olson 2001 ; Allen 2005 ; Mulkidjanian et al. 2006 ). Were the merger hypothesis correct, the genes of chlorophyll biosynthesis in anoxygenic photosynthetic lineages should reflect early lineage splittings and hence the same deep divergence as the photosynthetic RCs. In the case that the two photosystems arose via duplication in a protocyanobacterium, the chlorophyll biosynthetic pathway should not reflect ancient lineage splittings, that is, there should be no deep dichotomy into RCI and RCII-specific chlorophyll biosynthetic pathways types. To distinguish between these possibilities, we have studied the evolution of the core (bacterio)chlorophyll biosynthetic pathway from Proto IX up to (bacterio)chlorophyllide a , which is the last chemical intermediate common to all chlorophyll types.",
"discussion": "Discussion Individually, the trees for proteins underlying chlorophyll biosynthesis are complex. Although they do not all tend strongly to reflect a single underlying topology, they do have aspects in common. Their main implication in the context of this article is as follows: taken as a whole, the trees for chlorophyll biosynthesis appear to distinguish between competing hypotheses for the presence of two serially linked photosystems at the origin of water-splitting photosynthesis. This behavior is statistically supported by the AU and SH tests, with rejection of trees where separation of RCI and RCII organisms was imposed. Two Cyanobacterial RCs: Gene Duplication, Not Lineage Merger The most important observation from this study is that there is no coevolutionary pattern linking chlorophyll biosynthesis gene phylogeny with either RCIs or RCIIs; in other words, we observe neither Type I- or Type II-specific chlorophyll biosynthesis genes nor Type I- or Type II-specific chlorophyll biosynthesis gene phylogenies. Because RCs cannot undergo evolution in the absence of chlorophyll, this lack of coevolutionary pattern linking chlorophyll biosynthesis to the divergence of RCI and RCII allows us to exclude the widely discussed possibility that RCI and RCII diverged via lineage splitting and became reunited in cyanobacteria via a large-scale gene transfer event ( Mathis 1990 ; Blankenship 1992 ; Meyer 1994 ; Blankenship and Hartman 1998 ; Xiong et al. 1998 , 2000 ; Blankenship 2001 ; Xiong and Bauer 2002a , 2002b ; Blankenship 2010 ), of the kind that Hohmann-Marriott and Blankenship (2011) call “fusion theories.” Indeed, the consistently close proximity of the GSB (chlorobia, with RCI) and green nonsulfur bacteria (chloroflexi, with RCII) in chlorophyll biosynthesis trees argues strongly against the view that there was a deep evolutionary split in chlorophyll biosynthesis corresponding to a lineage split between RCI and RCII, one that would be expected to have pulled (B)Chl genes in tow. This is all the more true given the tendency for the chlorophyll biosynthesis genes from proteobacteria (with RCII) to branch in close proximity to their cyanobacterial homologs, rather than with homologs from green nonsulfur bacteria. In addition, the chlorophyll biosynthesis genes of C . Chloracidobacterium thermophylum (with RCI) usually branch in close proximity to RCII-containing taxa (proteobacteria or chloroflexi). If we thus exclude ancient lineage splitting and later (re)union in cyanobacteria at the evolutionary origin of two RC types coexisting in the same cell, the simplest competing alternative, and one widely discussed in the literature ( Olson and Pierson 1987a ; Vermaas 1994 ; Olson 2001 ; Allen 2005 ; Mulkidjanian et al. 2006 ), is that gene duplication giving rise to RCI and RCII within the same genome gave rise to the photosystem configuration in oxygenic photosynthesis. This possibility is easily reconciled with chlorophyll biosynthesis phylogenies. An immediately ensuing question is: in which lineage did the gene duplication take place? Occam’s razor clearly favors the premise that the photosystem genes underwent duplication in an ancestral cyanobacterium—a protocyanobacterium—because cyanobacteria are the only group where genes for both RC types have remained present and expressed. It is possible to assume that the duplication took place elsewhere, but there is no obvious alternative location. What Use Might Two Photosystems Be? Olson and Pierson ( 1987a , 1987b ) have suggested an ancient gene duplication event in a cyanobacterium for the origin of the two photosystems. However, that formulation accounted for the distribution of chlorophyll-based photosynthesis and RC types among prokaryotic groups exclusively by vertical inheritance and differential loss. Today, with the number of bacterial lineages having grown, the distribution of photosynthesis and RCs ( fig. 1 ) has become much more sparse than in 1987, such that differential loss alone is unlikely to account in full for these distributions, especially because photosynthesis genes are observed to be mobile in the marine phage metagenome, and because substantial amounts of LGT have indeed been shown for photosynthetic lineages ( Raymond et al. 2002 ; Huang and Gogarten 2007 ; Shi and Falkowski 2008 ). Hence, a mechanistic mixture of some vertical inheritance, some differential loss, and some lateral transfer, relative amounts of which might differ from gene to gene across (B)Chl synthesis, needs to be invoked, because the trees for (B)Chl biosynthesis genes are insufficiently similar for artifacts of phylogenetic reconstruction to account for their differences. However, how much and what kind of LGT might be required to explain the distribution of photosynthesis among prokaryotes? Various possibilities have recently been discussed, for example, by Bryant et al. (2012) and Gupta (2012) . However, that question is not the focus of this article. Rather, the question at our focus is how two photosystems came to reside within a single genome so as to give rise to oxygenic photosynthesis. Gene duplication in a protocyanobacterium is the alternative most compatible with present data. If there were two photosystems, what were they doing? This question has to do with the transition from anoxygenic photosynthesis to water splitting. Blankenship and Hartman (1998) suggested that hydrogen peroxide (H 2 O 2 ) might have been an initial electron donor far more chemically accessible than water, for a linear two-photosystem ETC. Mulkidjanian et al. (2006) suggested that H 2 might have been the initial electron donor for a two-photosystem ETC. Nisbet and coworkers ( Nisbet et al. 1995 ; Nisbet and Sleep 2001 ) suggest that the manganese complex evolved either to handle excess of peroxide or as a toxic weapon against competitors. An alternative suggestion ( Mathis 1990 ; Allen and Puthiyaveetil 2005 ) differs from the foregoing models with respect to the presumed function of the two photosystems. It posits that the two photosystems in the protocyanobacterium operated in a temporally regulated manner ( Allen 2005 ), for example, as an H 2 S-oxidizing and NAD + -reducing RCI when H 2 S was available as it occurs in modern Chlorobium or as a light-driven proton pump (cyclic electron flow thorough RCII as in Rhodobacter ) when H 2 S was not available. Before the water-splitting complex had evolved, what would a bacterium with two different and specialized, nonoxygenic photosystems have done with them? Probably just what modern bacteria do: express them when needed, with the help of a regulatory switch. This model implies relatively strict regulation of the RC genes, because in the event that regulation failed, for example, through mutation of the regulatory protein constituting the redox switch, both RCIs and RCIIs became expressed in the absence of H 2 S, the protocyanobacterium would be exposed to a lethal level of oxidative stress (there is no way to turn off assembled photosystems), unless it could extract electrons from an environmentally available donor. In principle, such a donor could have been H 2 O 2 , or H 2 , or an organic compound such as succinate. However, it also could have been aqueous Mn II/III cations, which have the interesting property of giving off low potential electrons under ultraviolet (UV) radiation (photooxidation) ( Anbar and Holland 1992 ; Hakala et al. 2006 ; Allen and Martin 2007 ; Russell et al. 2008 ). Before the accumulation of atmospheric oxygen (and hence ozone), UV was a larger component of the solar radiation that reaches the Earth’s surface than it is today. Thus, the presence of light for photosynthesis would have also meant the presence of UV radiation for Mn II/III photooxidation. Photooxidation of aqueous, environmental Mn II/III could have thus literally “pushed” electrons into photooxidized chlorophyll of RCII. That would have led to a lethal log-jam of electrons, or over-reduction, in the electron-transport cycle—unless the protocyanobacterium was simultaneously expressing RCI. This complex would redirect the surplus membrane-bound electrons stemming from photosystem II into CO 2 -reduction and thereby create precisely the flow of electrons seen in cyanobacteria today: a linear flow through two photosystems. Obviously, such an ETC would hardly have been perfect from the beginning. And clearly there is a difference between tapping environmental Mn atoms photooxidatively, one at a time, and the establishment of the resident, catalytic Mn 4 Ca center in cyanobacterial RCII water-splitting complex. However, the overall contours of the circuit would have been right, from manganese at RCII and going on through RCI to NADH and ultimately to CO 2 . The key to water splitting would then have entailed the transition from exploiting an environmental supply of soluble manganese, where each electron-donating Mn ion would reach photosystem II by simple diffusion, to holding four manganese atoms (and a calcium of as-yet-unknown function) in place. Dissolved Mn 2+ has long been known to donate electrons to photosystem II, and thus to reconstitute noncyclic electron transport, in isolated chloroplasts biochemically depleted of the capacity to oxidize water to oxygen ( Cheniae 1970 ). Following adsorption of Mn to its donor side, a fine tuning of photosystem II by natural selection to optimize its reduction/oxidation potential could have allowed it to oxidize a now biologically portable manganese reservoir four times in a row. It is notable that the Mn atoms of the water-splitting complex are bound directly by the proteins of the photosystem II RC, without an intervening protein or electron carrier. This suggests that no major evolutionary invention was required for photosystem II to tap environmental Mn as an electron source. In line with that, it has recently been shown ( Allen et al. 2012 ) that an engineered, Mn-binding RCII of R . sphaeroides will produce O 2 from {\\rm O}_{\\rm 2}^- in the presence of Mn in a light-dependent reaction in which photodamage is impeded in comparison with that in a wild-type, Mn-free RC. Allen et al. (2012) interpret this observation as an important clue to the origin of oxygenic photosynthesis. In what sort of setting could suitably high concentrations of Mn II/III have accumulated for such series of events to transpire? Mn is not a good candidate for high enough concentrations to make the model viable in an open ocean setting. However, in a locally circumscribed freshwater setting, sufficiently high Mn II/III concentrations could, in principle, have accumulated. In this context, it is of particular interest that Blank and Sáchez-Baracaldo (2010) recently provided evidence that oxygenic photosynthesis arose in a freshwater environment, based on the basal phylogenetic position of freshwater cyanobacteria and the derived phylogenetic position of marine cyanobacteria. Related Pathways and the Cell That Invented Chlorophyll Chlorophylls coordinate Mg 2+ and belong to the tetrapyrrole family, which includes cobalamin (Co 2+ ), heme (Fe 2+ ), sirohem (Fe 2+ ), heme d1 (Fe 2+ ), and F 430 (Ni 2+ ) ( Heinemann et al. 2008 ; Warren and Smith 2009 ; Zappa et al. 2010 ). All tetrapyrrole biosynthetic pathways are related in the sense that they start from the universal precursor, δ-aminolevulinic acid, and share three enzymatic steps that generate the tetrapyrrole macrocycle uroporphyrinogen III (UroIII) ( fig. 5 ). Proto IX is the common precursor for chlorophyll synthesis and heme synthesis via the classical pathway that occurs in eubacteria and eukaryotes ( Heinemann et al. 2008 ). Precorrin-2 is the common precursor for cobalamin, heme synthesis via the alternative pathway recently discovered in archaebacteria ( Storbeck et al. 2010 ; Bali et al. 2011 ), siroheme (a cofactor in some nitrite and all sulfite reductases ( Tripathy et al. 2010 ), heme d1 (a cofactor only present in the bacterial cd 1 nitrite reductases ( Allen et al. 2005 ), and F 430 pathways. F 430 is a cofactor that is critical to methanogenesis and has only been found in methanogens so far ( Thauer 1998 ). In addition to two routes for heme synthesis, there are also two routes for cobalamin synthesis, one O 2 dependent (often called the late pathway due to the late insertion of Mg 2+ ) and one O 2 independent (early insertion of Mg 2+ ) ( Martens et al. 2002 ; Warren et al. 2002 ). The O 2 -dependent cobalamin pathway involves the Co 2+ chelatase CobNST, the three subunits of which are related to (B)ChlHID. It has been suggested that (B)ChlHID arose from CobNST ( Xiong et al. 2000 ), but the converse might be more likely given the presence of one O 2 -dependent step in the late cobalamin pathway ( Martens et al. 2002 ; McGoldrick et al. 2005 ). In the early (O 2 independent) cobalamin pathway, the CbiK/CbiX L /CbiX s chelatases are related to the SirB and HemH chelatases of the siroheme and classical heme pathways, respectively ( Schubert et al. 1999 ; Brindley et al. 2003 ; Romao et al. 2011 ). The chelatase for the F 430 pathway is unknown, but genes related to (B)ChlHID are present in methanogens and encode candidates for the Ni 2+ -chelatase. In some organisms, the insertion of iron and cobalt into sirohydrochlorin is performed by multifunctional chelatases (CysG and Met8p) ( Spencer et al. 1993 ; Fazzio and Roth 1996 ; Schubert et al. 2002 ) (Class III).\n F ig . 5.— Schematic representation of the bifurcations present in the global tetrapyrrole pathway. The biosynthetic pathway of all tetrapyrroles begins with the condensation of eight molecules of 5-amino-levulinate to uroporphyrinogen-II (UroIII), the first cyclic tetrapyrrole. This compound can be sequentially converted to protoporphyrin IX in three enzymatic steps. This is the last common precursor for (bacterio)chlorophyll synthesis and heme synthesis via the classical pathway (present in most eubacteria and eukaryotes). Alternatively, UroIII can be converted to precorrin-2, the branching point of F430 synthesis, of the O 2 -dependent cobalamin pathway, and of the formation of Sirohydrochlorin. At the level of Sirohydrochlorin, the O 2 -independent cobalamin pathways diverge from the formation of siroheme. Siroheme may be used as a cofactor, further transformed into heme d1 or transformed into heme according to the alternative heme pathway present in archaea and some eubacteria. What did the cell that invented chlorophyll biosynthesis have in terms of tetrapyrrole pathways? It definitely contained cobalamin and might have had a cobalamin biosynthesis pathway (the O 2 -independent type, obviously), because the O 2 -independent route to 3,8-divinyl protochlorophyllide a via (B)ChlE, a cobalamin-dependent reaction ( Fuhrmann et al. 1993 ; Gough et al. 2000 ). Of course, it could also have just been cobalamin dependent; some cyanobacteria synthesize cobalamin, and many, however, acquire it from the environment via high affinity importers ( Tang et al. 2012 ). The cell that invented chlorophyll also had a classical heme pathway, because barring chlorophyll, the three steps from UroIII to Proto IX are specific to classical heme synthesis ( Heinemann et al. 2008 ) and hence preceded the (B)Chl pathway. The first cell with chlorophyll probably also fixed nitrogen, because the subunit of DPOR, (B)ChlNBL, is related to nitrogenase subunits ( Xiong et al. 1998 ). We propose that the cell that invented chlorophyll was a cobalamin-dependent, heme-synthesizing, diazotrophic anaerobe."
} | 6,228 |
39483348 | PMC11527095 | pmc | 7,106 | {
"abstract": "Neural circuits in the brain perform a variety of essential functions, including\ninput classification, pattern completion, and the generation of rhythms and oscillations\nthat support processes such as breathing and locomotion [ 51 ]. There is also substantial evidence that the brain encodes memories and\nprocesses information via sequences of neural activity. In this\ndissertation, we are focused on the general problem of how neural circuits encode rhythmic\nactivity, as in central pattern generators (CPGs), as well as the encoding of sequences.\nTraditionally, rhythmic activity and CPGs have been modeled using coupled oscillators.\nHere we take a different approach, and present models for several different neural\nfunctions using threshold-linear networks. Our approach aims to unify attractor-based\nmodels (e.g., Hopfield networks) which encode static and dynamic patterns as attractors of\nthe network. In the first half of this dissertation, we present several attractor-based\nmodels. These include: a network that can count the number of external inputs it receives;\ntwo models for locomotion, one encoding five different quadruped gaits and another\nencoding the orientation system of a swimming mollusk; and, finally, a model that connects\nthe fixed point sequences with locomotion attractors to obtain a network that steps\nthrough a sequence of dynamic attractors. In the second half of the thesis, we present new\ntheoretical results, some of which have already been published in [ 59 ]. There, we established conditions on network architectures to\nproduce sequential attractors. Here we also include several new theorems relating the\nfixed points of composite networks to those of their component subnetworks, as well as a\nnew architecture for layering networks which produces “fusion” attractors by\nminimizing interference between the attractors of individual layers.",
"introduction": "Chapter 1 | Introduction Attractor neural networks play an important role in computational neuroscience by\nproviding a rich framework for modeling the dynamic behavior of neural systems and giving\ninsights into how the brain might process information and perform computations [ 3 , 43 ]. Originally\ndevised as models of associative memory, these networks were designed to store static\npatterns representing discrete memories as attractors. Their ability to simultaneously\nencode multiple static patterns, represented as fixed points, made them ideal for this\npurpose. Perhaps the most well-known example is that of Hopfield networks, a foundational\nexample of attractor neural networks [ 39 ]. As\nillustrated in Figure 1.1A , in the classical Hopfield\nparadigm, memories are stored in the network as several coexistent stable fixed points, each\none accessible via distinct input pulses (represented as color-coded pulses in the figure).\nThe state space is partitioned into basins of attraction, and each input will start the\ntrajectory into one of these basins. Coexistence of attractors, even if it’s just\nmultiple stable fixed points, requires nonlinear dynamics. While Hopfield networks are well-suited for encoding multiple static patterns, some\npatterns of neural activity are better stored as dynamic attractors. Such\nis the case for the rhythms and oscillations produced by Central Pattern Generator circuits\n(CPGs). CPGs are neural circuits that generate rhythmic patterns that control movements like\nwalking, swimming, chewing, and breathing [ 51 ].\nUnlike static patterns (such as images), these rhythmic processes require attractors whose\nneurons fire sequentially , meaning that neurons take turns to fire.\nFurthermore, a single CPG network should potentially be able to encode multiple such\npatterns. Certainly, animals have several locomotive gaits, all of which activate the same\nlimbs [ 4 , 34 ].\nHow can all these different overlapping dynamic patterns be produced by the same network?\nModeling this using attractors requires multi-stability of dynamic \nattractors, which is known to be a difficult problem [ 24 , 63 ]. Traditional models of locomotion and other CPGs have tackled these challenges by\nusing coupled oscillators [ 12 , 21 , 31 , 41 , 73 , 74 ]. In these models, the parameters are typically adjusted\ndepending on the desired pattern, effectively altering the dynamical system in use. For\nvarious reasons, this approach presents some challenges. For instance, while there is\nevidence of pacemaker neurons, not all neurons are intrinsic oscillators [ 62 ]. Additionally, assuming that synaptic strengths change every\ntime we transition between different locomotive gaits necessitates additional ingredients\nfor the model, such as synaptic plasticity. Despite this, coupled oscillator models have\nremained popular for important reasons. One of the reasons they are so widely used is the\navailability of theoretical results coming from physics and mathematics [ 5 ]. Indeed, many CPG models stemmed from the physics and\nmathematics communities [ 12 , 31 , 73 , 74 ]. The availability of theoretical tools is indeed a very compelling argument for\nusing coupled oscillators to model central pattern generators. Here we aim to take a\ndifferent approach that also leverages recent theoretical advances, but within the world of\nattractor neural networks. The framework of attractor neural networks differs from\ntraditional coupled oscillator models in two key ways. First, neurons do not intrinsically\noscillate; instead, patterns of activity emerge as a result of connectivity. Second, these\npatterns are true attractors of the system, making them more robust and stable to noise and\nother perturbations. Additionally, providing models within this framework would unify our\napproach to CPGs with other classical models, like the aforementioned associative memory\nmodels, ring attractors, etc [ 3 , 43 ]. On the biological side, it has also been suggested that\ncortical circuits involved in associative memory encoding and retrieval have many features\nin common with CPGs [ 79 ]. To accomplish this, we propose the use of Threshold-Linear Networks (TLNs). TLNs\nare recurrent neural networks with simple non oscillating units and piecewise linear\nactivation [ 38 ]. This makes them focus on the role of\nconnectivity in emergent behaviors. TLNs have a rich history as neural models [ 7 , 35 , 46 , 67 , 70 ], and more importantly, they are supported by a wide\narray of theoretical results [ 13 , 14 , 37 , 53 , 77 ], many of them recent\n[ 15 , 16 , 54 , 59 ]. One such key finding is that threshold-linear networks with symmetric connectivity\nmatrices can only have stable fixed point attractors (static patterns) [ 36 , 39 ], which is why here we\nuse non-symmetric TLNs, introduced in Chapter 2 . These\nare known to give rise to a rich variety of non-linear dynamics including multi-stability,\nlimit cycles, chaos and quasi-periodicity. Therefore, we expect it to be possible to\nsimultaneously encode multiple dynamic patterns of activity, whereas in classical Hopfield\nnetworks the stored patterns are all static. Within this unified framework, our aim is to\nprovide attractor models for three broad neural functions, as summarized in Figure 1.1 , and detailed below: First, we propose a simple neural integrator model that is both robust to noise and\ncan count inputs, using fixed point attractors. Neural integration refers to the process in\nwhich information from various sources is combined to create an output. For instance,\ncounting the number of left and right cues to make a decision is quite literally integration\n(summation) in the mathematical sense. Here, we propose to model a discrete counter as a\nsequence of static attractors, as shown in Figure 1.1B .\nWhile this concept is akin to the classical Hopfield model ( Figure 1.1A ), our aim is to internally encode the sequence of\nfixed points, meaning that the input pulses are all identical and contain no information\nabout which fixed point comes next in the sequence (i.e. they are just like pushing a single\ninput button). Second, we aim to devise a (small) network that has attractors corresponding to\n4–5 distinct, but overlapping, quadruped gaits. The goal is for the attractors to\ncoexist in the same network so that they can be accessed by different initial conditions,\nand without changing parameters. This requires the presence of several coexistent\n dynamic attractors, as pictured in Figure\n1.1C , arising as distinct limit cycles in state space. Recall that the simultaneous\nencoding of multiple dynamic attractors (non fixed point attractors) is a network is\nchallenging, especially when the attractors have overlapping units. While classic models\ncircumvent this challenge by adjusting synaptic weights, we aim to obtain a single fixed\nnetwork, which in turn means a simpler model with fewer control parameters. In addition to quadruped gaits, we will also model “Clione’s hunting\nsystem ” [ 58 ], which is a different CPG\nexample, using the same framework. Previous models for it have also used intrinsically\noscillating units and fine-tuned parameters [ 71 ].\nHere we intend to devise a more robust network for this using attractors to avoid the need\nfor finely-tuned parameters. Third, we combine the modeling approaches from panels B and C to devise a network\nthat can step through a set of dynamic attractors sequentially , as in Figure 1.1D . Can different attractors be linked together so\nthat they can be activated in sequence, where the sequence itself is stored within the\nnetwork? This could be useful for modeling sequences of complex movements, such as a\nchoreographed sequence of dance moves, for instance. Sequences of sequential attractors. Note that in Figure 1.1C , we are dealing\nwith dynamic attractors, that are sequential themselves, meaning attractors whose nodes\nactivate in an ordered sequence [ 59 ]. This\ndefinition does not completely exclude attractors in which there is some synchrony in the\nactivations. So for example, we consider both attractors in Figure 1.2 \n sequential attractors , even though the attractor on the right has nodes 2\nand 3 synchronized (we will later formalize the dynamic prescription of the nodes, for now\nnote the sequential of activations of nodes). These are great to model rhythmic activations like those of CPGs. In contrast,\n Figure 1.1D deals with sequences of dynamic\nattractors , which means that different attractors, either static or dynamic,\nare activated in a specific ordered sequence (e.g., attractor A, then attractor B, then\nattractor C). With this distinction clear, our ultimate goal is to achieve an internally\nencoded sequence of sequential attractors. Internally encoded sequences of sequential attractors are good models for\ncomplex sequences of movements, like choreographed dancing. These complex motor behaviors\nhave also been modeled in the past using threshold-linear recurrent networks that choose\nand learn motor motifs, but whose choice mechanism requires plasticity, and where the\nsequence’s order is externally encoded [ 49 ].\nIt is not uncommon to think of broader cognitive sequential processes as a recombination\nof several pre-stored patterns, which offers an efficient alternative to re-encoding\npatterns with each occurrence. Studies in this vein include [ 65 ], where they utilize a combination of discrete metastable\nstates, leveraging winnerless competition of oscillator neurons. Similar mechanisms have\nbeen explored using boolean and spiking networks [ 69 , 72 ]. Desired properties of models. As it turns out, the versatility of threshold linear networks will prove ideal\nfor modeling both sequential attractors and sequences of them. To summarize the discussion above, we want our models to satisfy the following\nproperties: Neurons are not intrinsic\noscillators. Stored static and dynamic patterns should emerge as\n attractors of the network, rather than being fine-tuned\ntrajectories. Static patterns should manifest as fixed points, while dynamic\npatterns should arise from non fixed point attractors, like limit\ncycles. A network’s attractors should be accessible via different\ninitial conditions, easily implemented via input pulses that target subsets of\nneurons. A sequence of attractors should be accessible via a series of\nidentical inputs pulses, with the sequence itself stored within the network\n(possibly in a separate layer, as observed in some biological\nbrains). The models should be mathematically tractable–that is,\nsimple enough to be analyzed mathematically. These properties will distinguish our models from previous coupled oscillator\nmodels and position them within the framework of attractor neural networks. We begin by\nadhering to the last point above, by choosing a framework that fits into the attractor\nneural network paradigm and also provides mathematically tractable models. With\ntractability also come great simplifications. Although TLNs are inspired by networks of\nbiological neurons, real neurons and their interactions are of course far more complex\nthan TLNs paint them to be. TLNs remain useful however because they capture two\nfundamental pieces of biological networks: connectivity and threshold-activation. This is\nwhy here we also focus on another simplification of TLNs, known as Combinatorial\nThreshold-Linear Networks (CTLNs) [ 15 , 53 , 54 ]. CTLNs\nare a special family of TLNs, whose connectivity matrix is defined by a simple directed\ngraph (giving rise to binary connections), as in Figure\n1.2 . Their added simplicity can be used to gain further theoretical results. Summary of models. The table below lists all the models included in the dissertation. Each row\ndefines a single model/network, and each is an example of the attractor behaviors\ndescribed by Figure 1.1 : Model Network function Type of attractor Chapter Model 1a counter sequence of static attractors \n 3 \n Model 1b signed counter sequence of static attractors \n 3 \n Model 1c dynamic attractor chain sequence of identical dynamic attractors \n 3 \n Model 2a quadruped gaits coexisting distinct dynamic attractors \n 4 \n Model 3a molluskan swimming coexisting identical dynamic attractors \n 4 \n Model 2b sequential control of quadruped gaits sequence of distinct dynamic attractors \n 5 \n Model 3b sequential control of molluskan swimming sequence of identical dynamic attractors \n 5 \n In Chapter 3 , we introduce three models\nfor sequences of attractors. Models 1a and 1b are two counter networks that step through\nsequences of fixed points. Models 1a is shown in Figure\n1.3A , where we can see that identical pulses move the network into the next\nstable fixed point, where it stays until it receives another pulse. This network serves as\nrobust discrete neural integrators of inputs, as we show in their chapter by doing a\nthorough robustness analysis. Additionally, we extend our work to dynamic attractors by\npresenting an additional network, Model 1c, capable of encoding a sequence of\n dynamic attractors. These dynamic attractors are all qualitatively\nidentical, and because of this, there are some symmetries in the basins of attractors, and\nthus all attractors easily accessible via distinct input pulses. However, to effectively\nmodel CPGs, we require different types of patterns to coexist\nsimultaneously. How can we achieve this? That is the content of Chapter 4 , where we\nmodel two different CPGs, Model 2a and 3a, which require sequential activation of neurons.\nModel 2a, developed in in Section 4.2 , is pictured\nin Figure 1.3B . It consists of a network encoding\nfive different quadruped gaits as coexistent limit cycles in a 24 unit network. There, we\nsee that all gaits coexist and are accessible via gait-specific pulses. We do a thorough\nanalysis of its dynamics via the set of fixed point supports, and the effect of parameters\nin modulating gait characteristics. Model 3a, in Section\n4.3 , consists of a network encoding swimming orientation of a marine mollusk\n(Clione). Since the attractors are all identical, we manage to prove symmetry of its\nbasins of attraction in Theorem 12 . Finally, in Chapter 5 , we merge the\nconcepts introduced in Chapters 3 and 4 to achieve sequences of sequential\nattractors . From Chapter 3 we get the\ncounter network that will encode the sequence transitions, and from Chapter 4 we get the coexistent sequential attractors. From this\nintegration we obtain Models 2b and 3b: a network for the sequential control of quadruped\nlocomotion and for the sequential control directing swimming movements in Clione. The\nlatter is the one pictured in Figure 1.4 , where we\nobserve the attractors from the CPG network “fuse” with the attractors of\nthe counter network, as both are simultaneously active, and look qualitatively like they\ndid when isolated. Figure 1.4 shows the resulting\nnetwork using Clione’s model, but it can also be done, analogously, with the\nfive-gait network. The fact that we could use the exact same construction with two\ndifferent networks, led us to believe this is an even more general phenomenon, arising\nfrom some structural constraints on these networks, ad they were indeed built with similar\nprinciples. Code to reproduce the plots in Figures 1.3 \nand 1.4 , and also all models listed in the table, can\nbe found at https://github.com/juliana-londono/phd-thesis-basic-plots . Note that in Figure 1.4 , we see a\n“blend” of two different attractors: at the top of they greyscale we see the\ndynamic attractors coming from layer L3, and at the bottom we see fixed points coming from\nlayer L1. This phenomenon was also observed in [ 61 ], where it was called fusion attractors . Fusion attractors\noffer a clean solution for managing sequences of static and dynamic attractors.\nUnderstanding the mechanisms behind this phenomenon motivates us to further explore the\nunderlying structural constraints giving rise to it, from a theoretical standpoint. This\nis why we now transition from models to theory. New network theory. We want to note that all the models we have developed thus far were built within\nthe TLN framework, for which there are plenty of well-established theoretical results.\nThis theoretical foundation made the process of building these models a lot easier.\nHowever, our models have now surpassed the available theory and so now they serve as\nsources of inspiration for the development of new theoretical results. This is why in the\nsecond part of the dissertation, we take a reverse approach: while theory initially guided\nour modeling efforts, now the models are leading the development of new theoretical\nresults. Chapter 6 presents original theoretical\ncontributions, including several results recently published in the paper I co-authored:\n\"Sequential Attractors in Combinatorial Threshold-Linear Networks\" [ 59 ]. This chapter is divided in three parts. First, in Section 6.1 , we establish some necessary technical\nresults, some of which are earlier version of results that we end up generalizing in this\nchapter. Then, in Section 6.2 , we derive new\nstructural theorems for CTLNs supporting sequential attractors. All of the results within\nthis section are my contribution to [ 59 ], which\ncontains several other architectures that support sequential attractors. All theorems I\nproved are in bold. Most of these results relate the fixed point supports of a network to\nthe fixed point supports of component subnetworks, as follows: Theorem 21 for “simply-embedded\npartitions”, constrains the possible fixed point supports of a network to\nunions of fixed points chosen from a particular menu of component\nsubnetwork fixed point supports. This is generalizing results from [ 15 ]. In the same section, Theorem 23 and Corollary 24 give\nconditions on when a node can be removed from a network without changing the set of\nfixed points supports. We include here a new result on removable nodes, that has not\nbeen published: Theorem 25 . Theorem 28 for “simple linear\nchains”, showing that the set of fixed point supports of a simple linear chain\nnetwork is closed under unions of “surviving” component fixed point\nsupports. Theorem 31 for “strongly\nsimply-embedded partitions”, showing that the set of fixed point supports of a\nnetwork can be fully determined from knowledge of the component fixed point supports\ntogether with knowledge of which of those component fixed points\n“survive” in the full network. Finally, in Section 6.3 , we extend some of\nthese theorems to TLNs and provide theoretical explanations for the fusion attractors\nobserved in Chapter 5 , culminating with: Theorem 40 , which is an important\ntechnical result generalizing previous theorems on certain determinant factorizations\nthat control the set of fixed point supports of a network. It relies on a new\ndeterminant factorization lemma, Lemma 38 , which\nI have also proven. Theorem 40 is then used as a\ncrucial ingredient in the proofs of: Theorem 42 ,\ngeneralizing Theorem 21 above. Theorem 44 , explaining how the fixed points of some special\nnetworks are formed from fixed point of smaller component networks. That theorem\ngeneralizes both Theorem 17 (from two components\nto N components) and Theorem 31 \n(from several CTLN components to several TLN components). And finally, we present a\nsimilar result but for “nested” component fixed point supports in Theorem 45 . We also show that the networks in Chapter\n5 satisfy these conditions, thus explaining the fusion attractors observed\nthere. In this dissertation’s final chapter, Chapter 7 , we present partial and further theoretical results derived from\nprojects in sections 4.2 and 6.3 . In Section 7.1 , Lemma 48 , gives conditions under which the same\nattractor can arise from two different networks. This phenomenon is known as degeneracy.\nAll code from this section is available online at https://github.com/juliana-londono/TLN-attractor-interpolation . In Section 7.2 , Lemma\n55 , gives a new way to think about certain Cramer’s determinants, which\nare at the core of the dynamics of TLNs. The rest of this dissertation is organized as follows: Chapter 2 introduces the framework, including firing rate models,\nattractor neural networks, TLNs, and CTLNs. Chapter\n3 provides models for sequences of static and dynamic attractors, both internally\nand externally encoded. In Chapter 4 , we provide two\nCPG models of locomotion, each consisting of several coexistent dynamic attractors, easily\naccessible via initial conditions or inputs. Chapter\n6 explores new architectures and theoretical results, focusing on sequential\ndynamics complex and networks made up of simpler subnetworks. Finally, Chapter 7 discusses further theoretical results derived from the\npresented projects, suggesting avenues for further exploration. Appendix A contains the matrices and parameters\nused to construct all the models, along with some technical calculations of the fixed\npoints of the five-gait quadruped network. That’s all. We hope you enjoy reading\nthis dissertation. We encourage you to keep in mind Figure\n1.1 , which is the road map guiding us through the chapters on models."
} | 5,772 |
33444327 | PMC7808614 | pmc | 7,107 | {
"abstract": "Methanol is often considered as a non-competitive substrate for methanogenic archaea, but an increasing number of sulfate-reducing microorganisms (SRMs) have been reported to be capable of respiring with methanol as an electron donor. A better understanding of the fate of methanol in natural or artificial anaerobic systems thus requires knowledge of the methanol dissimilation by SRMs. In this study, we describe the growth kinetics and sulfur isotope effects of Desulfovibrio carbinolicus , a methanol-oxidizing sulfate-reducing deltaproteobacterium, together with its genome sequence and annotation. D . carbinolicus can grow with a series of alcohols from methanol to butanol. Compared to longer-chain alcohols, however, specific growth and respiration rates decrease by several fold with methanol as an electron donor. Larger sulfur isotope fractionation accompanies slowed growth kinetics, indicating low chemical potential at terminal reductive steps of respiration. In a medium containing both ethanol and methanol, D . carbinolicus does not consume methanol even after the cessation of growth on ethanol. Among the two known methanol dissimilatory systems, the genome of D . carbinolicus contains the genes coding for alcohol dehydrogenase but lacks enzymes analogous to methanol methyltransferase. We analyzed the genomes of 52 additional species of sulfate-reducing bacteria that have been tested for methanol oxidation. There is no apparent relationship between phylogeny and methanol metabolizing capacity, but most gram-negative methanol oxidizers grow poorly, and none carry homologs for methyltransferase (mtaB). Although the amount of available data is limited, it is notable that more than half of the known gram-positive methanol oxidizers have both enzymatic systems, showing enhanced growth relative to the SRMs containing only alcohol dehydrogenase genes. Thus, physiological, genomic, and sulfur isotopic results suggest that D . carbinolicus and close relatives have the ability to metabolize methanol but likely play a limited role in methanol degradation in most natural environments.",
"introduction": "Introduction Sulfate-reducing microorganisms (SRMs) utilize a great variety of organic compounds as an electron donor for energy production, being responsible for most of the terminal carbon mineralization in anoxic environments where sulfate is available [ 1 ]. Methanol is a common C 1 -compound in nature as a product of pectin and lignin degradation and also represents an inexpensive energy source for industrial bioprocesses, which is often considered as a non-competitive substrate for methanogenic archaea in sulfate-rich environments [ 2 ]. However, although less common, several species of SRMs are capable of oxidizing methanol ([ 3 – 9 ] and references therein). Their growth rates are usually slower than those of methanogenic microorganisms at high methanol concentrations, but SRMs have been reported to outcompete methanogens for methanol, where the environmental conditions such as methanol concentration or temperature seem more favorable for SRMs [ 10 – 13 ]. In methanol-fed bioreactors, stimulation of SRMs is particularly problematic because dissimilatory sulfate reduction accumulates toxic and corrosive hydrogen sulfide [ 14 , 15 ]. While a role for SRMs in methanol-containing environments has been shown, knowledge about their underlying physiology remains limited. Chemical investigation of laboratory cultures is valuable for assessing the physiology of a microorganism, but most microorganisms remain uncultured. Application of molecular and stable isotopic techniques offer a view of microbial metabolic activity in situ . An increasing number of genome sequences from SRMs are now available, quite a few of which have been tested for their capacity to degrade methanol. Recently, genomic and proteomic studies using a sulfate-reducing bacterium Desulfofundulus kuznetsovii have revealed two pathways involved in methanol degradation [ 9 , 16 ], providing a solid basis to develop molecular markers for methanol oxidation coupled with sulfate reduction. Stable isotope ratios, albeit less specific, have been also used extensively as recorders of microbial activities. For example, the dominant mode of CO 2 fixation at the given environments can be constrained by measuring the 13 C/ 12 C isotope ratio of organic compounds [ 17 ]. For the anaerobic oxidation of methanol by SRMs, carbon isotope effects have been measured [ 9 ], but no data are currently available for sulfur isotope fractionation. Depletion of heavy sulfur isotopes in sulfide relative to reactant sulfate is a well-established diagnostic for sulfate respiration, where the magnitude of sulfur isotope discrimination primarily reflects the intracellular balance between electron acceptors and donors ([ 18 – 23 ] and references there in). Thus, knowledge of the sulfur isotope fractionation by methanol-degrading SRMs can inform us of the efficiency of respiratory coupling between sulfate reduction and methanol oxidation. Here we presented the full circularized genome sequence of Desulfovibrio carbinolicus , one of the early known sulfate-reducing bacteria capable of oxidizing methanol [ 3 , 24 ] and assessed the metabolic pathways linked to this process. This work was coupled to experiments characterizing growth kinetics and sulfur isotope fractionation, focusing on the influence of alcohol metabolisms by varying the chain length of alcohols or using mixed substrates.",
"discussion": "Discussion Methanol oxidation in D . carbinolicus The nature of electron donors influences the fractionation of sulfur isotopes of the electron acceptors during microbial respiration by altering the route and rate of electron transfer to the terminal reductases [ 20 , 36 – 38 ]. However, an increase in alcohol chain length from ethanol to n-butanol results in negligible changes in the magnitude of sulfur isotope fractionation (< 1‰), suggesting that an identical set of enzymes are likely involved in the oxidation of n-alcohols. Alcohol dehydrogenases isolated from Desulfovibrio species have been shown to catalyze the oxidation of n-alcohols ranging from ethanol to butanol with a lower activity toward butanol [ 39 , 40 ], which might be responsible for a decrease in growth and a slight increase in 34 ε value during the oxidation of butanol ( Fig 1 and Table 1 ). In contrast to the similar results for n-alcohols with 2 to 4 carbon atoms, the specific rates of growth and sulfate reduction with methanol as a sole electron donor is more than an order of magnitude slower than those with other alcohols ( Table 2 ). As described in previous studies [ 20 , 41 , 42 ], such slow metabolism quadruples the magnitude of sulfur isotope fractionation. Theoretically, the overall isotope effect is governed by the reversibility of each enzymatic step in the dissimilatory sulfate reduction pathway: the higher the reversibility is, the larger the sulfur isotope fractionation becomes [ 21 ]. The reversibility of each enzymatic step is dependent on the Gibb’s free energy of the reaction.\n R e v e r s i b i l i t y = b a c k w a r d r e a c t i o n r a t e f o r w a r d r e a c t i o n r a t e = e Δ G R T (4) \nwhere R is the gas constant, T the absolute temperature, and ΔG is the free energy change associated with the reaction. Although the standard reduction potential of methanol from CO 2 (E 0 ’ = -0.37 V) is the same order of that of ethanol from acetate (E 0 ’ = -0.39 V; calculated after [ 43 ]), sulfur isotope fractionations coupled with methanol and ethanol oxidation span a wide range from less than 10‰ to over 40‰ ( Fig 3A ) because ΔG reflects the substrate concentrations and the kinetic properties of the involved enzymes as well as the nature of electron donors [ 21 , 44 , 45 ]. Sulfur isotope fractionation of 36‰ far exceeds that imparted by APS reductase and is close to the sum of the isotope effects in APS and subsequent reduction steps [ 45 , 46 ], implying that the APS reduction should be reversible during the growth on methanol [ 45 ]. A model based on a flux-force relationship ( Eq 4 ) predicts the ΔG of APS reduction in the methanol-grown culture to be about -3 kJ/mole, corresponding to the reversibility of 0.3, while this step is practically irreversible with ethanol as an electron donor ( Fig 3B ). Recent advances in isotope geochemistry have enabled the experimental assessment of reversibility [ 23 , 47 , 48 ], which is beyond the scope of this study. However, future work incorporating oxygen isotope analysis would extend our understanding of reversibility effects during sulfate respiration with methanol. A less negative free energy change suggests that methanol oxidation and subsequent electron transfer processes are sluggish as compared to ethanol. Indeed, it has been shown that alcohol dehydrogenases isolated from D . carbinolicus exhibit marginal catalytic activities for methanol oxidation [ 49 ], and here sulfur isotope data confirm previous in vitro results. 10.1371/journal.pone.0245069.g003 Fig 3 (A) Variations in specific respiration rate and sulfur isotope effect during sulfate reduction coupled to methanol or ethanol oxidation, reported in this study and previous literature [ 18 , 48 , 50 , 51 ]. For the comparison with prior work, the respiration rate normalized by optical density is converted to the approximate cell-specific sulfate reduction rate (csSRR) according to the conversion factor for optical density to total cell volume (A660 of 1.0 as 1.49 μl/ml [ 52 ]) and the average cell volume of 1.68 μm 3 [ 3 ]. (B) Pattern of the sulfur isotope fractionation and the free energy change (ΔG) for APS reduction, predicted based on the model originally proposed by Wing and Halevy [ 21 ] and modified by Sim et al. [ 45 ]. Sulfur isotope fractionation and free energy change are calculated as a function of both csSRR and reduction potential of the electron-donating half reaction. The former varies from 1 fmol/cell/day to 100 fmol/cell/day and the latter from -140 mV to -70 mV. All calculations are made using the constant sulfate and sulfide concentrations of 15 mM and 5 mM, respectively, which approximates when the reaction is half completed. When presented with methanol and ethanol, D . carbinolicus first grows exclusively on ethanol, where growth kinetics and sulfur isotope fractionation are comparable to those with ethanol with a sole electron donor. Microorganisms often exhibit diauxic growth in a batch culture containing a mixture of two substrates, but D . carbinolicus consumes no methanol even after its growth on ethanol ceases. Methanol is thus consumed neither simultaneously nor sequentially, suggesting that a single enzymatic system is unlikely to be responsible for both ethanol and methanol metabolisms. In addition to alcohol dehydrogenase, a few SRMs oxidize methanol to CO 2 \n via a methyl-transfer reaction predominantly used by methanogens and homoacetogens. This methanol methyltransferase is absent from the D . carbinolicus genome, but instead, D . carbinolicus contains multiple genes encoding alcohol dehydrogenases, two of which are homologous to the enzyme involved in methanol oxidation in other methylotrophic bacteria [ 9 , 34 ]. We hypothesize these dehydrogenases are likely responsible for the oxidation of methanol by D . carbinolicus , but the regulatory element that controls the expression of methanol-oxidizing alcohol dehydrogenase currently remains unclear. An unfavorable free energy change with increasing sulfide concentration might hinder the oxidation of methanol coupled with sulfate reduction at the end of growth on ethanol, but sulfide removal by purging with N 2 /CO 2 gas (< 20 μM) failed to resume sulfate respiration in the mixed-substrate culture. Alternatively, even low levels of residual ethanol may suppress the expression of the dehydrogenase responsible for methanol oxidation, which might be related to the energetic state of the cells after fast growth on ethanol switching to a substantially less favorable carbon substrate. Such strong suppression effects by ethanol on methanol metabolism have been described in yeast [ 53 ], although the target enzyme is not alcohol dehydrogenase but alcohol oxidase. Future studies that incorporate transcriptomic and proteomic approaches could help resolve the role of each dehydrogenase in methanol and ethanol oxidation. Methanol is present in nature as a product of pectin and lignin degradation, but also used in industrial wastewater treatment as a cheap carbon and energy source for microbial digestion [ 54 ]. Since methanogenic, homoacetogenic, and sulfate-reducing microorganisms compete for the available methanol under anaerobic conditions, the dominant methylotrophic group may vary across different environments. However, slow methanol metabolism and its strong suppression by alternative electron donors suggest that while having a thermodynamic advantage over methanogens and homoacetogens [ 43 ], D . carbinolicus and presumably its close relatives have a limited role in methanol degradation in nature, where the complex suite of organic substrates is present. In wastewater treatment with methanol as a primary substrate, the addition of a small amount of ethanol might be a promising way to control sulfide production. This is the first report for sulfur isotope fractionation coupled with methanol oxidation, but as fractionation increases with methanol as an electron donor, sulfur isotopes may provide constraints on the role of SRMs in natural or artificial methanol-rich environments. Phylogenetic distribution of methanol metabolism among sulfate-reducing bacteria In methylotrophic SRMs, methanol oxidation is catalyzed by either methanol methyltransferase or alcohol dehydrogenase [ 9 , 49 ]. The genome of the methanol oxidizing D . carbinolicus virtually lacks a methyltransferase coding gene but contains two genes encoding the proteins homologous to the methanol-dissimilating alcohol dehydrogenase reported from methylotrophic microorganisms, including the sulfate-reducing bacterium D . kuznetsovii [ 9 ]. Similar to its closest relative, the D . magneticus genome carries two loci coding for methanol-dissimilating enzyme homologs, and the pairwise comparison reveals that more than 95% of the nucleotides are identical in the corresponding regions of the D . carbinolicus and D . magneticus genomes. Thus, D . magneticus likely has the ability to metabolize methanol, the same as D . carbinolicus , although methanol oxidation by D . magneticus has not been tested experimentally. In addition to D . carbinolicus and D . magneticus , the genomes of an increasing number of SRMs have been sequenced and annotated, over 50 species of which have been tested for their capacity to metabolize methanol in culture. Hence, their phylogenetic analyses based on the 16S rRNA gene and the genes encoding the alcohol dehydrogenase and methyltransferase may provide new insights into the evolutionary and ecological significance of methanol dissimilation by SRMs. Among the 53 species of sulfate-reducing bacteria examined, 12 are able to oxidize methanol. Based on the 16S rRNA phylogeny, methylotrophs occur in multiple clades of sulfate-reducing bacteria with non-methylotrophic sister taxa, although four species each are found in the genera of Desulfovibrio and Desulfosporosinus ( Fig 4A ). The genes homologous to those encoding the methanol-oxidizing alcohol dehydrogenase are identified in 35 genomes of sulfate-reducing bacteria, but only 10 of them have been shown to metabolize methanol. When narrowed down to the species containing the methyl-tranferase gene, 4 out of 6 are identified with the ability to oxidize methanol. Interestingly, two methanol-metabolizing SRMs, Desulfatiglans anilini and Pseudothermotoga lettingae [ 5 , 7 ], have neither of those two, suggesting that there are likely additional, currently unknown pathways for methanol oxidation. 10.1371/journal.pone.0245069.g004 Fig 4 Phylogenetic tree of the selected sulfate-reducing bacteria based on 16S rRNA sequence (A) and of the genes encoding homologous proteins to methanol-oxidizing alcohol dehydrogenase (B) and methanol methyltransferase (C) of D . kuznetsovii [ 9 ]. This analysis involves 52 species of sulfate-reducing bacteria that have their genome sequenced and deposited in NCBI database and have been tested for methanol metabolism [ 3 – 8 , 12 , 49 , 55 – 88 ]. The trees were constructed using MEGA X software with the neighbor-joining method. Numbers before each branch point represents the percentage of bootstrap resampling based on 2,000 trees. Bootstrap values below 50% are not shown. D . carbinolicus examined in this study is highlighted in a blue box. The enzymes homologous to the methanol-oxidizing alcohol dehydrogenase of D . kuznetsovii are distributed in most lineages of SRMs excluding the thermophilic gram-negative bacteria ( Fig 4A ), and their phylogeny is in broad agreement with the 16S rRNA gene tree ( Fig 4B ), indicating vertical transfer of the alcohol dehydrogenase genes. This cytoplasmic enzyme belongs to the PDDH subfamily of the type III NAD-dependent alcohol dehydrogenase and catalyzes the oxidation of methanol to formaldehyde in methylotrophic bacteria; however, the catalytic activity is much higher with multi-carbon alcohols compared to methanol, and it functions more efficiently in a reverse direction, reducing formaldehyde to methanol [ 34 , 49 ]. Given the lack of a simple and consistent relationship between homologous genes and methanol dissimilation ( Fig 4A ), such kinetic properties suggest that instead of oxidizing methanol, the enzyme may have evolved either to metabolize larger alcohols or to detoxify formaldehyde [ 34 ]. Except for two gram-positive bacteria, Desulfosporosinus acididurans and Desulfotomaculum reducens , methanol oxidizing SRMs that are currently known to have alcohol dehydrogenase but no methyltransferase genes fall into the Desulfovibrio genus, and their methylotrophic growth is retarded compared to that with ethanol and other conventional electron donors ( Table 3 ). As seen in D . carbonolicus , Desulfovibrio alcoholivorans and Desulfovibrio salexigens also grow extremely slowly with methanol as an electron donor and require acetate as a carbon source [ 6 , 79 ]. Methanol oxidation by Desulfovibrio fructosivorans does not support any growth even in the presence of acetate [ 4 ]. Thus, the function of these organisms in environmental methanol cycling is likely to be limited. Although the detailed growth kinetics of D . reducens is not currently available, a gram-positive bacterium D . acididurans also showed only weak methylotrophic growth [ 8 ]. 10.1371/journal.pone.0245069.t003 Table 3 Presence of methanol dissimilating enzyme homologs in SRMs capable of methanol oxidation and their growth properties. Methanol-oxidizing SRM PDDH MT Growth with methanol as an electron donor Ref. Gram negative Desulfovibrio alcoholivorans ◯ × + [ 6 ] Desulfovibrio carbinolicus ◯ × + [ 3 ] Desulfovibrio fructosivorans ◯ × - [ 4 ] Desulfovibrio salexigens ◯ × + [ 79 ] Gram positive Desulfotomaculum reducens ◯ × na [ 60 ] Desulfosporosinus acididurans ◯ × + [ 8 ] Desulfosporosinus lacus ◯ ◯ + [ 88 ] Desulfosporosinus meridiei ◯ ◯ ++ [ 66 ] Desulfosporosinus orientis ◯ ◯ ++ [ 57 ] Desulfofundulus kuznetsovii ◯ ◯ ++ [ 68 ] PDDH, 1,3-propanediol dehydrogenase subfamily of the type III alcohol dehydrogenase; MT, methyltransferase; ++, good growth comparable to that with ethanol; +, lesser growth than that with ethanol; -, methanol oxidation without growth; na, no growth kinetic information available. Unlike alcohol dehydrogenases, the occurrence of the methyltransferase gene (mtaB) is restricted to only a few species of SRMs, and their phylogeny is not congruent with 16S rRNA phylogeny. Based on the mtaB gene sequences, the gram-positive D . kuznetsovii is positioned together with the gram-negative Desulfospira joergensenii , while gram-positive Desulfosporosinus species form a separate lineage ( Fig 4C ), suggesting that simple vertical inheritance is unlikely. It has been previously shown that the mtaB genes in sulfate-reducing bacteria reside in two distinct phylogenetic clades: one clade containing sequences from methanogenic archaea, while the other contains sequences from acetogenic bacteria [ 9 ]. A patchy occurrence of the mtaB gene in SRMs and a discordance of this gene tree with the species tree can be the consequence of horizontal gene transfer. Although the number of relevant examples is rather limited, all known SRMs containing the mtaB gene and capable of methanol oxidation belong to gram-positive bacteria and generally utilize methanol better than those only with alcohol dehydrogenase genes ( Table 3 ). For example, the methylotrophic growth of D . kuznetsovii , Desulfosporosinus orientis and Desulfosporosinus meridiei was comparable to the growth with other common substrates [ 57 , 66 , 68 ]. However, because those gram-positive SRMs also carry the homologous genes for methanol-oxidizing alcohol dehydrogenases ( Table 3 ), it remains to be determined whether methyltransferase pathway is advantageous for methanol oxidation as compared to dehydrogenase pathway. Given that the genes encoding the PDDH subfamily of alcohol dehydrogenases in gram-positive and -negative SRMs form two distinct lineages ( Fig 4B ), each might significantly differ in the activity toward methanol. In addition, the rate of methanol consumption by D . kuznetsovii was shown to be equivalent regardless of the pathway of methanol oxidation [ 9 ]. Thus, the contribution of each enzymatic system toward methanol oxidation needs to be further tested, including kinetic studies of the alcohol dehydrogenases from gram-positive and -negative SRMs, but genomic and physiological data available so far suggest that the gram-positive SRMs equipped with both enzymatic pathways should perhaps be considered first in attempts to understand the biogeochemical cycle of methanol in sulfidic environments rather than Desulfovibrio species."
} | 5,577 |
29807839 | null | s2 | 7,108 | {
"abstract": "Evidence is increasing for positive effects of α-diversity on ecosystem functioning. We highlight here the crucial role of β-diversity - a hitherto underexplored facet of biodiversity - for a better process-level understanding of biodiversity change and its consequences for ecosystems. A focus on β-diversity has the potential to improve predictions of natural and anthropogenic influences on diversity and ecosystem functioning. However, linking the causes and consequences of biodiversity change is complex because species assemblages in nature are shaped by many factors simultaneously, including disturbance, environmental heterogeneity, deterministic niche factors, and stochasticity. Because variability and change are ubiquitous in ecosystems, acknowledging these inherent properties of nature is an essential step for further advancing scientific knowledge of biodiversity-ecosystem functioning in theory and practice."
} | 231 |
23977944 | PMC3765991 | pmc | 7,109 | {
"abstract": "Background Fermentative hydrogen production is an attractive means for the sustainable production of this future energy carrier but is hampered by low yields. One possible solution is to create, using metabolic engineering, strains which can bypass the normal metabolic limits to substrate conversion to hydrogen. Escherichia coli can degrade a variety of sugars to hydrogen but can only convert electrons available at the pyruvate node to hydrogen, and is unable to use the electrons available in NADH generated during glycolysis. Results Here, the heterologous expression of the soluble [NiFe] hydrogenase from Ralstonia eutropha H16 (the SH hydrogenase) was used to demonstrate the introduction of a pathway capable of deriving substantial hydrogen from the NADH generated by fermentation. Successful expression was demonstrated by in vitro assay of enzyme activity. Moreover, expression of SH restored anaerobic growth on glucose to adhE strains, normally blocked for growth due to the inability to re-oxidize NADH. Measurement of in vivo hydrogen production showed that several metabolically engineered strains were capable of using the SH hydrogenase to derive 2 mol H 2 per mol of glucose consumed, close to the theoretical maximum. Conclusion Previous introduction of heterologous [NiFe] hydrogenase in E. coli led to NAD(P)H dependent activity, but hydrogen production levels were very low. Here we have shown for the first time substantial in vivo hydrogen production by a heterologously expressed [NiFe] hydrogenase, the soluble NAD-dependent H 2 ase of R. eutropha (SH hydrogenase). This hydrogenase was able to couple metabolically generated NADH to hydrogen production, thus rescuing an alcohol dehydrogenase ( adhE ) mutant. This enlarges the range of metabolism available for hydrogen production, thus potentially opening the door to the creation of greatly improved hydrogen production. Strategies for further increasing yields should revolve around making additional NADH available.",
"conclusion": "Conclusion The work reported here shows convincingly that a pyridine nucleotide dependent [NiFe] hydrogenase can be heterologously expressed in E. coli and produce large amounts of hydrogen from NAD(P)H produced by cellular metabolism. Hydrogen production is at least 50 fold greater than previously reported [ 39 , 41 ] This represents a significant advance in the ability to engineer hydrogen producing pathways in E. coli . Moving forward, a number of improvements could be made. Increasing flux through the system would be required to increase the rates of hydrogen production. In addition, a practical hydrogen production system would require that greater yields be obtained from the substrate, which could be brought about in several different ways. For one thing, more efficient coupling with the native hydrogen producing system, which produces hydrogen indirectly from pyruvate through the pyruvate:formate lyase system, should further increase yields. Another possibility would be to introduce a mechanism whereby additional NADH could be generated through the further metabolism of pyruvate, for example, through the anaerobic functioning of the citric acid cycle.",
"discussion": "Results and discussion Rationale In the absence of exogenous electron acceptors, anaerobically grown E. coli carries out a mixed acid type fermentation. Sugars are degraded to pyruvate by the glycolytic pathway, producing ATP and reducing NAD + to NADH. The amount of NADH that is produced depends upon the redox state of the substrate, and this in turn controls fermentation product distribution. Pyruvate is mainly converted to formate and acetyl-CoA. Under the appropriate conditions, usually acidic pH, formate is broken down via the formate-hydrogen lyase pathway producing CO 2 and H 2 . Thus, E. coli is only capable of the production of a maximum of 2 H 2 per mole of glucose that enters glycolysis (Figure 1 ). The NADH that is generated during anaerobic growth on sugars must be oxidized to NAD + for glycolytic metabolism to continue since NAD + is a necessary cofactor for the oxidation of glyceraldehyde. Although in theory NADH could be oxidized by the reduction of pyruvate to lactate by lactate dehydrogenase, in practice this pathway is only fully expressed under acidic conditions, and does not seem to be sufficient on its own to permit anaerobic growth. Therefore, mutants deleted for alcohol dehydrogenase ( adhE ) cannot grow anaerobically on sugars more reduced than glucuronate [ 20 ]. Thus, adhE mutants might possess excess levels of NADH when incubated anaerobically. Some NADH can be recycled through the oxidation of oxaloacetate to malate, leading ultimately to the formation of succinate, but again, this side pathway is not sufficient in itself to permit anaerobic growth on sugars. Previous attempts to introduce novel hydrogen pathways had only given very low activities. Thus, the goal of the present research was to attempt to introduce a heterologous pathway that would allow the production of substantial amounts of hydrogen by reoxidizing NADH, thus potentially allowing for additional hydrogen production while, depending upon the strain, rescuing growth of some mutant strains incapable of anaerobic growth due to an inability to reoxidize sufficient amounts of NADH (Figure 1 ). Figure 1 Native and engineered metabolic pathways involved in hydrogen production by E. coli. On the right is shown the main multiple pathways of mixed acid fermentation. Key enzymes and enzyme complexes are indicated by either the genetic nomenclature or the commonly used pathway abbreviation: Phosphoenolpyruvate (PEP); Phosphoenolpyruvate carboxylase (PEPC); Fumarate reductase ( frdC ); Lactate dehydrogenase (ldhA); Pyruvate formate lyase (PFL); Formate hydrogen lyase (FHL); Hydrogenase 3 (Hyd 3); formate dehydrogenase-H (FDHH); Uptake hydrogenases; hydrogenase 1 (Hyd 1) and hydrogenase 2 (Hyd 2); fumarase ( fumB ); fumurate reductase ( frdC ). Points where NADH is produced or consumed are noted. On the left is a schematic of the SH hydrogenase, which, if functional, might consume NADH, reducing protons to hydrogen. Initial overexpression of SH hydrogenase In order to examine the potential for engineering E. coli to produce hydrogen from NADH, we chose to express the SH hydrogenase from Ralstonia eutropha H16 (Table 1 ). The SH operon consists of nine genes; hoxFUYHWI and hypA2B2F2 [ 21 ]. HoxHY is the hydrogenase module and HoxFU is a NADH dehydrogenase. hoxW encodes an highly specific endopeptidase required for the C-terminal processing of HoxH during hydrogenase maturation [ 22 ]. Although SH H 2 ase is usually isolated as a heterotetrameric protein (HoxHYFU), HoxI has been shown to provide a NADPH binding domain to a hexameric form of SH that can be isolated under certain conditions [ 23 ]. hypA2B2F2 are duplicate copies of three of the seven R. eutropha hydrogenase maturation genes ( hyp ). Interestingly, they can substitute for hypA1B1F1 in the maturation of both the SH hydrogenase and the MBH (membrane bound) hydrogenase [ 24 ]. A previous attempt to express the SH operon in E. coli from its native promoter (P SH ) was unsuccessful [ 25 ], presumably because there is an absolute requirement for the transcriptional activator HoxA for expression from this promoter [ 21 ]. Therefore, we wished to express the SH operon from a promoter active in E. coli . Plasmid pJWPH5 for expression of the SH operon in E. coli under the control of the inducible trc promoter of the vector pTRC99A was constructed as described in Materials and Methods. Table 1 Strains used Strain Genotype Reference/construction FTD147 Δ hyaB Δ hybC Δ hycE Skibinski et al. 2002 [ 43 ] JW135 Δ hya-Km, Δ hyb-Km Menon et al. 1991 [ 26 ] FTAB1 FTD147/ pJWPH5* This study FTAB4 FTD147, Δ adhE, zch::Tn10 , pJWPH5 P1 DC1048 ( Δ adhE::Tn10 ), FTAB5 FTD147 Δ arcA ::Tn10 , pJWPH5 P1 QC2575 ( Δ arcA::Tn10 ), FTDPH10 FTD147/pJWPH5 adhE, ldhA P1 DC1048 ( Δ adhE::Tn10 ), SE1752 ( Δ ldhA::Tn10 ) DG2 FTD147/pJWPH5 adhE, ldhA, arcA P1 DC1048 ( Δ adhE::Tn10 ), SE1752 ( Δ ldhA::Tn10 ), QC2575 ( Δ arcA::Tn10 ) FTJWDC3 FTD147/pJWPH5 adhE, mdh P1 DC1048 ( Δ adhE::Tn10 ), JW3205-1 ( Δ mdh::Tn5 ) FTGH2 FTD147/pJWPH5 adhE, ldhA, mdh P1 DC1048 ( Δ adhE::Tn10 ), SE1752 ( Δ ldhA::Tn10 ), JW3205-1 ( Δ mdh::Tn5 ) DJ1 JW135/pJWPH5 arcA P1 QC2575 ( Δ arcA::Tn10 ) JWGH1 JW135/pJWPH5 adhE, ldhA, arcA P1 DC1048 ( Δ adhE::Tn10 ), SE1752 ( Δ ldhA::Tn10 ), QC2575 ( Δ arcA::Tn10 ) * Plasmid containing the Ralstonia eutropha SH operon under the control of the trc promoter in pTrc99A. The hydrogen evolution capacity of batch cultures of various strains of E. coli were tested (Table 2 ) under anaerobic conditions with LB medium as previously described [ 17 ]. E. coli possesses four hydrogenases; two of which, Hyd1 ( hya ) and Hyd2 ( hyb ), normally function in hydrogen oxidation, and two others, Hyd3 ( hyc ) and Hyd4 ( hyf ), which function physiologically in proton reduction. Hydrogen was evolved by strain BW535, wild-type for the four hydrogenases [ 26 ]. As expected, hydrogen was also evolved by strain JW135, which, being a Hyd1- Hyd2- (Δ hya -Km Δ hyb -Km) derivative of BW535, only lacks the hydrogenases that function to consume hydrogen. Strain FTD147, which lacks the hydrogen evolving Hyd3 as well as hydrogen consuming Hyd1 and Hyd2 [ 27 ], showed no hydrogen evolution ( E. coli Hyd4 is inactive under these conditions [ 28 ]). The hydrogen evolution of strains containing pJWPH5 that had been altered so that they potentially produced more NADH was also tested. Derivatives of strain FTD147 (H 2 ase-) that lacked alcohol dehydrogenase (Δ adhE ) or the aerobic/anaerobic regulator ArcA (Δ arcA ), were examined for hydrogen evolution in anaerobically incubated LB-glucose (0.4%) medium (+ 0.05 mM IPTG) (Table 2 ). As discussed above, mutants lacking alcohol dehydrogenase should have excess NADH levels when incubated anaerobically and might consequently support hydrogen evolution by SH hydrogenase. However, there was no detectable hydrogen evolution by strain FTAB4 (FTD147/ pJWPH5 Δ adhE ). Another strain, FTAB5, potentially able to produce increased levels of NADH under anaerobic conditions, was also examined. Strain FTAB5, which carries pJWPH5 (SH hydrogenase), is a Δ arcA derivative of strain FTD147. ArcA is a two-component regulator that is responsible for the anaerobic repression of synthesis of enzymes of the TCA cycle. Therefore, a strain mutated in ArcA might be expected to express the TCA cycle under anaerobic conditions, potentially permitting the generation of excess NADH from acetyl-CoA. Indeed, in vitro TCA cycle enzyme activities are greatly increased in ArcA mutants grown under anaerobic conditions, suggesting that additional NADH would become available if it could be more effectively oxidized under these conditions. However, there was no detectable hydrogen evolution by this strain either (Table 2 ). Table 2 Hydrogen evolution by various strains of E. coli incubated under anaerobic conditions E. coli strain Relevant genotype H 2 BW545 Wild type + JW135 Δ hya- Km Δ hyb- Km + FTD147 Δ hyaB Δ hybC Δ hycE - FTAB4 FTD147, Δ adhE, zch:: Tn 10, pJWPH5 - FTAB5 FTD147 Δ arcA :: Tn10 , pJWPH5 - Given these results, we wished to verify that SH hydrogenase protein was present in these cells, even though no activity could be detected. Synthesis of SH hydrogenase proteins was checked by a Western blot (Figure 2 ) of an extract of strain FTD147/pJWPH5 grown under anaerobic conditions (LB + 0.05 mM IPTG). Prominent protein bands at 67 and 55 kDa were observed, corresponding to HoxF and HoxH respectively. Although there was a high level of expressed protein in the supernatant (Figure 2 , lane 2), there appeared to be some inclusion body formation since a significant quantity was also recovered in the pellet obtained after centrifugation of the crude extract, produced by sonication, for 15 min at 10,000 rpm (Figure 2 , lane 3). As a control, a high-speed supernatant of an extract of R. eutropha H16 grown anaerobically in FGN medium mineral salts medium [ 29 ] as previously modified [ 30 ] was included (Figure 2 , lane 1). Thus, sufficient levels of SH hydrogenase appeared to be synthesized under anaerobic conditions in E. coli strain FTD147 strain carrying pJWPH5. Figure 2 Western blot analysis of expression of SH hydrogenase. FTD147/pJWPH5) was cultured overnight at 30°C in LB medium with 0.05 mM IPTG under anaerobic conditions. The culture was harvested by centrifugation, sonicated, and centrifuged (15 min, 10,000 rpm). 25–35 μg pellet (lane 2) and supernatant (lane 3) were electrophoresed on 12% SDS-polyacrylamide gels (Laemmli and Favre (1973)), transferred to PVDF membrane, developed with primary anti-serum to SH hydrogenase, and revealed by chemiluminescence as previously described (Yakunin and Hallenbeck (1998)). A similarly cultured and prepared extract of FTD147 was also analyzed (supernatant, lane 4, pellet, lane 5) and found to be devoid of these protein bands. As a positive control, an aliquot of the supernatant of a 45 min 100,00 g centrifugation of a sonicated extract of R. eutropha H16 grown anaerobically overnight on NB medium at 30°C was also loaded (lane 1). Lack of hydrogen evolution by anaerobically incubated FTD147-derivatives carrying pJWPH5 was therefore not due to lack of SH hydrogenase protein, but might rather be due to inadequate hydrogenase maturation. This was verified by checking the in vitro NAD-linked hydrogenase activity of extracts of anaerobically grown cultures of R. eutropha H16, E. coli FTD147, and E. coli FTAB1 (FTD147/pJWPH5) using a spectrophotometric assay [ 31 ]. As expected, high levels of H 2 -dependent NAD + reduction was observed with the R. eutropha H16 extract, 5.87 ± 0.99 μmol NADH min -1 mg -1 and no reduction was seen with FTD147. However, only a very low level of activity was observed for FTAB1 (FTD147/pJWPH5), 0.39 ± 0.19 μmol NADH min -1 mg -1 . Thus, it indeed appears that maturation of R. eutropha SH hydrogenase to a functional hydrogenase is very inefficient in E. coli grown anaerobically under these conditions. Anaerobic growth of adhE mutants carrying pJWPH5 (SH hydrogenase) on M9 glucose in the presence of nickel and iron Since SH is a [NiFe] hydrogenase, we hypothesized that the low activity observed might be due to the insufficient supply of nickel and iron and therefore we assessed the effect of added nickel and iron on the formation of active SH hydrogenase [ 32 ]. First this was assayed using a growth test run under conditions such that only strains possessing an SH hydrogenase would grow. As described above, strains that are mutated in adhE have been reported to be impaired for growth under anaerobic conditions. We constructed various Δ adhE derivatives of both FTD147 and JW135 (FTGH2, FTJWDC3, JWGH1, DG2, FTDPH10) and verified that they were unable to grow under anaerobic conditions on sorbitol or glucose minimal media (not shown). Reasoning that growth could be restored if a means of reoxidizing NADH were introduced, these mutants were tested for the rescue of anaerobic growth on glucose by the introduction of pJWPH5, plasmid carrying the SH operon under the control of the trc promoter. Indeed, under these growth conditions, M9-glucose + IPTG + Ni + Fe, these derivatives were able to grow (not shown). NAD + reduction in vitro The growth results strongly suggested that SH was expressed and active in anaerobic cultures grown on nickel and iron supplemented M9-glucose. This was verified by assaying the SH hydrogenase activity of these strains. We measured the capacity of extracts to reduce NAD + under a hydrogen atmosphere, i.e. to carry out hydrogen oxidation (Table 3 ). As expected, the positive control, an extract of R. eutropha H16 showed significant NAD + reduction activity whereas the two E. coli strains, FTD147 and JW135, gave insignificant levels of activity. However, extracts of strains carrying pJWPH5 (SH hydrogenase) all showed varying but significant levels of SH hydrogenase activity in vitro. The parental strains, FTD147 and JW135 had specific activities close to 1 μmol NADH /min /mg protein, or 16% of that of an extract of R. eutropha H16. Extracts of strains carrying mutations that could be thought to increase cellular NADH levels, FTGH2, FTJWDC3, JWGH1, DG2, FTDPH10, and DG1, gave even higher specific SH hydrogenase levels, varying from 3.5 ± 0.1 μmol NADH /min /mg protein to 7.1 ± 0.31 μmol NADH /min /mg protein. These results (Table 3 ), obtained with nickel and iron amended M9, demonstrate the importance of medium supplementation with the metals required for cofactor synthesis since, in their absence, very little in vitro activity can be demonstrated. Table 3 In vitro NAD + reduction activity of various strains Strain Relevant genotype μmol NADH min -1 mg -1 R. eutropha H16 Wild type 6.1 ± 0.4 FTD147 Δ hyaB Δ hybC Δ hycE 0 JW135 Δ hya-Km, Δ hyb-Km 0.04 ± 0.009 FTD147 +pJWPH5 Δ hyaB Δ hybC Δ hycE 0.94 ± 0.16 JW135 +pJWPH5 Δ hya-Km, Δ hyb-Km 1.09 ± 0..003 FTGH2 +pJWPH5 FTD147 adhE, ldhA, mdh 3.9 ± 0.06 FTJWDC3 +pJWPH5 FTD147 adhE, mdh 6.4 ± 0.10 JWGH1 +pJWPH5 JW135 adhE, ldhA, arcA 7.1 ± 0.31 DJ1 +pJWPH5 JW135 arcA 4.45 ± 0.003 DG2 +pJWPH5 FTD147 adhE, ldhA, arcA 3.56 ± 0.11 FTDPH10 +pJWPH5 FTD147 adhE, ldhA 3.5 ± 0.1 The in vitro NAD-linked hydrogenase activity of extracts of anaerobically grown cultures of R. eutropha H16, E. coli FTD147, and E. coli FTAB1 were assayed spectrophotometrically (Schneider and Schlegel (1976)). Twenty μg of extract were incubated in a stoppered cuvette containing 1.9 ml of hydrogen-saturated Tris buffer (50 mM, pH 8, 30C°) that had been flushed with hydrogen. The reaction was initiated by the addition of NAD + to 0.8 mM, and the reduction of NAD + followed at 365 nm. The highest activities were observed with JWGH1, a JW135 (H 2 ase 3+) derivative carrying a mutation in the NADH consuming enzyme lactate dehydrogenase and in ArcA in addition to alcohol dehydrogenase, and FTJWDC3, a FTD147 (H 2 ase-) derivative mutated in malate dehydrogenase in addition to alcohol dehydrogenase. Somewhat lower levels of in vitro activity were observed with extracts of DJ1, a JW135 derived strain additionally mutated in arcA , and FTGH2, a FTD147 derivative mutated in both lactate dehydrogenase and malate dehydrogenase. A FTD147 derivative, DG2, carrying the same mutations as the JW135 derived strain JWGH1, gave only about 50% of the in vitro activity of that strain. Finally, FTDPH10, mutated in alcohol dehydrogenase and lactate dehydrogenase, gave only 50% of the highest observed in vitro activity, but even so this was more than three-fold higher than the parental strain, FTD147 carrying pJWPH5. It is clear from these results that, even though transcription is under control of the IPTG inducible trc promoter, higher levels of SH hydrogenase, as measured by in vitro activity, were present in strains in which the ability to reoxidize NADH anaerobically was compromised. The exact mechanism behind this enhancement is unclear, but might be related to general effects on growth. In addition, the results shown in Table 4 suggest that the effect of the introduction of multiple mutations in pathways that oxidize NADH appears to be additive, with abolition of malate dehydrogenase being more effective in a adhE strain than eliminating lactate dehydrogenase activity. At any rate, these results demonstrate the successful heterologous expression in E. coli of R. eutropha SH hydrogenase, multi-subunit [NiFe] hydrogenase capable of interacting with NAD+/NADH. Table 4 In vivo hydrogen yields of strains expressing SH hydrogenase Strain Yield a Background Mutations FTDPH10 FTD147 (Δ hyaB Δ hybC Δ hycE ) /pJWPH5 adhE, ldhA 1.41 ±0.017 DG2 FTD147 (Δ hyaB Δ hybC Δ hycE ) /pJWPH5 adhE, ldhA, arcA 1.46 ±0.015 FTJWDC3 FTD147 (Δ hyaB Δ hybC Δ hycE ) /pJWPH5 adhE, mdh 2.08 ±0.016 FTGH2 FTD147 (Δ hyaB Δ hybC Δ hycE ) /pJWPH5 adhE, ldhA, mdh 1.49 ±0.016 DJ1 JW135/pJWPH5 arcA 1.55 ±0.018 JWGH1 JW135/pJWPH5 adhE, ldhA, arcA 2.11 ±0.014 a mol H 2 /mol glucose consumed. In vivo hydrogen production by E. coli strains expressing SH hydrogenase The results of the in vitro activity assays and the growth studies both provided evidence for the active expression of SH hydrogenase carried by pJWPH5. Therefore it was of interest to determine if these strains could produce hydrogen in vivo from glucose, demonstrating the establishment of a non-native hydrogen producing pathway in E. coli . The different strains were incubated anaerobically in modified M9-glucose. Growth was followed by measuring the OD (600 nm) (Figure 3 A) and the hydrogen produced was assayed using gas chromatography (Figure 3 B). All strains showed significant growth over the experimental period after a variable lag period (Figure 3 A). Growth was highest, and at nearly the same level, in strains FTGH2, FTJWDC3, and JWGH1. FTGH2 and FTJWDC3 both carry adhE and mdh and FTGH2 adhE , mdh and ldhA . Final optical densities were appreciably lower in strains DJ1, DG2, and FTDPH10. Nevertheless, the growth of strains carrying adhE ; FTGH2, FTJWDC3, JWGH1, DG2, and FTDPH10, demonstrates that they were capable of sufficient NADH reoxidation to permit growth. Since growth was only observed in strains carrying pJWPH5, NADH oxidation must have been provided by the action of SH hydrogenase. Figure 3 Growth and in vivo hydrogen production by strains expressing SH hydrogenase. Cultures, pregrown under the same conditions, were incubated at 37°C in anaerobic vials containing modified M9 glucose (+IPTG, Ni and Fe). Samples were taken periodically to measure OD (A) and hydrogen (B) . FTGH2/pJWPH5 (-♦-); FTJWDC3/pJWPH5 (-■-); JWGH1/pJWPH5 (-△-); DJ1/pJWPH5 (-○-); DG2/ pJWPH5 (-□-); FTDPH10/ pJWPH5 (-●-). Hydrogen production by these cultures was also examined (Figure 3 B). All strains tested showed appreciable hydrogen evolution activity with final hydrogen levels of between 2.0 and 3.6 μmol H 2 per vial (2 ml of culture). Strains FTJWDC3 (FTD147/pJWPH5 adhE , mdh ) and JWGH1 (JW135/pJWPH5 adhE, ldhA, arcA ) produced the greatest amount of hydrogen, whereas strains FTGH2 (FTD147/pJWPH5 adhE, ldhA, mdh ) and DG2 (FTD147/pJWPH5 adhE, ldhA, arcA ) produced the least. Interestingly, the two best hydrogen producers were also the strains that were shown to have the highest levels of SH hydrogenase activity in vitro (Table 3 ). Strain DG2, which gave one of the lowest SH hydrogenase activities in vitro, also produced the least amount of hydrogen. Taken together this suggests that hydrogen production levels are controlled by the amount of active SH hydrogenase that is present, but further work would be required to firmly establish this point. In addition, alterations in carbon flux through the different metabolic pathways operating in the various strains may have an influence as well. This could be determined my measurements of all the carbon fluxes involved, but such a study is beyond the scope of the present manuscript. To measure the efficiency of hydrogen production, the amount of glucose consumed at the end point was determined and used to calculate the hydrogen yields, mol H 2 produced / mol glucose consumed, of the different cultures (Table 4 ). All strains showed very good hydrogen yields, varying from 1.41 to 2.1 mol H 2 / mol glucose, with strains FTJWDC3 and JWGH1 being the most efficient. Interestingly, since hydrogen yield and total hydrogen production are not always correlated, these were also the strains that produced the greatest amount of hydrogen. The yields observed here are higher than those normally observed with wild type cultures and are as high, or slightly higher, than the theoretical maximum for E. coli (see Figure 1 and earlier discussion). Of course, without any other metabolic changes, the maximum amount of hydrogen that could be produced by the SH hydrogenase would appear to be 2 H 2 / glucose since two NADH are formed during glycolysis of glucose. However, strains that also contain the native H 2 ase 3 could in theory surpass this by producing in addition hydrogen from formate. Another fashion to exceed the 2 H 2 / glucose limit would be to produce extra NADH by oxidizing some of the pyruvate through the TCA cycle. These yields are also much higher than those obtained in previous studies where heterologous hydrogen producing pathways were introduced into E. coli . In several previous studies, ferredoxin-dependent [FeFe] hydrogenase pathways were introduced along with the enzymes necessary to reduce ferredoxin with either NADH or NADPH. However, yields were disappointingly low; 0.025 [ 33 ], 0.04 [ 34 ], 0.05 [ 35 ] mol H 2 / mol glucose. On the other hand, when a [FeFe] hydrogenase was coupled with metabolism by the expression of a pyruvate:ferredoxin oxidoreductase, yields as high as 1.46 [ 36 ] mol H 2 / mol glucose were obtained. Here we have introduced a [NiFe] hydrogenase dependent pathway and shown that it is capable of higher (44% greater than the highest previously reported) yields than the previously characterized [FeFe] hydrogenase dependent pathways. Others have previously reported the heterologous expression of [NiFe] hydrogenases [ 32 , 37 - 41 ], but only in two reports on the heterologous expression of the cyanobacterial Synechocystis [NiFe] hydrogenase were the in vivo hydrogen yields reported [ 39 , 41 ]. In one report in which a E coli strain devoid of native hydrogenases was used, a total hydrogen production of only 20 μmol H 2 / liter was observed with a yield of only 0.0004 mol H 2 / mol glucose [ 39 ]. The results of the present study represent a nearly 100 and 500 fold improvement respectively. In another study, the cyanobacterial [NiFe] hydrogenase was expressed in an E. coli strain which also possessed a native hydrogenase 3. Thus the two hydrogenase activities are confounded and it is difficult to determine exactly how much was due to the introduced hydrogenase, which might very well have had an indirect effect since its expression increased formate flux through H 2 ase 3 (Hyd3). One estimate, derived from the difference of the native strain and the recombinant strain, is that the heterologous hydrogenase contributed 0.67 mol H 2 / mol of glucose, but again this is an overestimate since the primary effect seemed to be to drive additional hydrogen production from formate by the native H 2 ase 3 [ 41 ]. Here we have unequivocally shown that heterologous expression of the [NiFe] SH hydrogenase can give up to 2 mol H 2 / mol glucose since we used a strain devoid of native hydrogenase activity. It might be thought that this amount of hydrogen from NADH is thermodynamically impossible, but a simple calculation using the Nernst equation shows that at 0.011 atm H 2 (3.6 μmoles H 2 in a head space of 8.3 ml) the equilibrium hydrogen redox potential would be −0.361 mV, within the range of the most recently determined equilibrium redox potential of the NAD/NADH couple, -0.378 mV [ 42 ]. Thus, an unusual NADH/NAD within the cell does not need to be invoked to explain the level of hydrogen production that we obtained in the present study. What the ultimate thermodynamic limits to hydrogen production are remains to be determined. Among other things, the applicability of a NADH/NAD determined for the whole cell to a specific biochemical reaction that may well be compartmentalized, or which may be proceeding under non-equilibrium conditions, remains to be demonstrated."
} | 6,947 |
34123258 | PMC8148073 | pmc | 7,110 | {
"abstract": "Guaiacol is an important feedstock for producing various high-value chemicals. However, the current production route of guaiacol relies heavily on fossil resources. Using lignin as a cheap and renewable feedstock to selectively produce guaiacol has great potential, but it is a challenge because of its heterogeneity and inert reactivity. Herein, we discovered that La(OTf) 3 could catalyze the transformation of lignin with guaiacol as the only liquid product. In the reaction, La(OTf) 3 catalyzed the hydrolysis of lignin ether linkages to form alkyl-syringol and alkyl-guaiacol, which further underwent decarbonization and demethoxylation to produce guaiacol with a yield of up to 25.5 wt%, and the remaining residue was solid. In the scale-up experiment, the isolated yield of guaiacol reached up to 21.2 wt%. To our knowledge, this is the first work to produce pure guaiacol selectively from lignin. The bio-guaiacol may be considered as a platform to promote lignin utilization.",
"conclusion": "Conclusions In summary, guaiacol can be selectively produced from lignin using Lewis acid La(OTf) 3 as a catalyst in methanol/water solvent. The yield of guaiacol reached up to 25.5 wt%, and the remaining residue is solid. As a catalytic system, La(OTf) 3 efficiently catalyzed the hydrolysis of ether linkages in lignin to form alkyl-syringol and alkyl-guaiacol products. Alkyl-guaiacol products further underwent decarbonization to produce guaiacol, while alkyl-syringol was transformed into guaiacol via decarbonization and demethoxylation. In the scale-up experiment, 0.32 g of guaiacol can be obtained from 1.50 g lignin, and the residue is solid. This work opens the way to produce pure guaiacol from lignin selectively and bridges plant-based lignin and fossil-based products, affording a green and sustainable strategy that is of great importance for the valorization of lignin.",
"introduction": "Introduction In the chemical industry, guaiacol is considered as an important raw material for the syntheses of many value-added chemicals, such as spices (vanillin, veratraldehyde and eugenol), 1,2 pesticides, 3 drugs (eugenol, potassium guaiacolsulfonate, guaifenesin, berberine, and isoprenaline), 4–8 a plant growth regulator (2-methoxy-5-nitrophenol sodium salt) and so on. Based on a global market survey, the demand for the above high-value chemicals increases year by year. Typically, annual demand for guaifenesin, vanillin, and eugenol is 37 000, 16 000, and 7300 t per year, respectively, 5,9–12 all of which may be synthesized from guaiacol. Therefore, the demand for guaiacol is large. However, current production of guaiacol is mainly based on the methylation of catechol which is an expensive downstream chemical from fossil resources, using additional methylation reagents, for example, methanol and methyl chloride ( Fig. 1A ). 13 Lignin is an aromatic abundant renewable source that is rich in guaiacyl (G) units, whether in the woody or herbaceous plants. 14–16 Hence, using lignin as a cheap and renewable feedstock to produce guaiacol has considerable potential, and can decrease the dependence on fossil resources for producing this important chemical. 15,17,18 Moreover, guaiacol would be used to produce more chemicals, both in category and quantity, if we could obtain cheap and renewable guaiacol from lignin. Fig. 1 Production route to guaiacol: (A) traditional route, (B) new route in this work. However, lignin has high heterogeneity and low reactivity, because it is predominately bonded by a series of inert C–O and C–C linkages (accounting for approximately 70% and 30%, respectively). 16 Although some technologies, such as hydrogenation, 19–21 oxidation, 22 hydrolysis, 12,23,24 and multiple strategies, 25 have been developed to depolymerize lignin into low-molecular-weight lignin monomers, the primary products from these processes are mixtures of compounds (phenols, arenes, aromatic acids, and so on) due to the complex structure of lignin, 14,26 and guaiacol can be generated in low yield as one of the monomer products in the mixtures. 27–30 The liquid mixtures usually make product separation and purification processes very complicated. It is therefore attractive to selectively transform lignin into a single value-added chemical, such as guaiacol, but this is particularly challenging to achieve efficiently. 31 Herein, we discovered that Lewis acid La(OTf) 3 could efficiently catalyze the selective transformation of lignin into guaiacol ( Fig. 1B ). As a catalytic system, La(OTf) 3 efficiently catalyzed the hydrolysis of ether linkages in lignin to form alkyl-syringol and alkyl-guaiacol products. Alkyl-guaiacol products further underwent decarbonization to produce guaiacol, while alkyl-syringol was transformed into guaiacol via decarbonization and demethoxylation. The unique feature of this methodology is that guaiacol can be directly produced from the transformation of lignin with high yield, and the methylation process is not required in this route.",
"discussion": "Results and discussion In this work, we firstly screened catalytic systems and optimized the reaction conditions for the transformation of organosolv lignin from Eucalyptus . Initially, lignin transformation was studied with various catalysts in methanol ( Table 1 , entries 1–8). It was found that no guaiacol was generated without catalyst ( Table 1 , entry 1). As we know, ether linkages can be efficiently cleaved with Lewis acid catalysts. 12,32–37 However, guaiacol could not be effectively obtained from lignin using weak Lewis acid catalysts, such as La(NO 3 ) 3 , Al(OTf) 3 and LaCl 3 ( Table 1 , entries 2–4). Unlike the above acids, the triflate salts with stronger Lewis acidity could effectively catalyze the production of guaiacol from lignin with various yields ( Table 1 , entries 5–8 and Fig. S1A † ). Methane was the only gaseous product (Fig. S1B † ). We also analysed the reaction mixtures by HPLC, and the spectra clearly indicated that guaiacol was essentially the sole product (Fig. S1C † ), establishing this transformation of lignin, delivering guaiacol as the only liquid product. The results above reveal that catalytic activity of La(OTf) 3 was the highest among these catalysts, and the yield of guaiacol could reach 14.9%. As is well known, a larger water exchange rate constant (WERC) value is beneficial to the fast hydrolysis of the ether linkages. 38 Among the metal cations, the La 3+ ion has one of the largest WERC values. 33,37,39 Furthermore, the higher temperature and stronger acidity cooperatively promoted the decarbonization of the side chain to form guaiacol. 21 Therefore, La(OTf) 3 could effectively hydrolyze the ether linkages and cleave the side chains in lignin. Transformation of lignin under different reaction conditions a \n \n Entry Catalytic system Temperature (°C) Catalyst (mg) Guaiacol yield (wt%) b Catalyst Solvent/water (mL/mL) 1 c — 4.0/0.00 270 20 0.0 2 La(NO 3 ) 3 4.0/0.00 270 20 0.0 3 Al(OTf) 3 4.0/0.00 270 20 0.6 4 LaCl 3 4.0/0.00 270 20 1.2 5 HOTf 4.0/0.00 270 20 6.6 6 Fe(OTf) 3 4.0/0.00 270 20 7.0 7 Yb(OTf) 3 4.0/0.00 270 20 7.8 8 La(OTf) 3 4.0/0.00 270 20 14.9 9 La(OTf) 3 4.0/0.01 270 20 22.5 10 La(OTf) 3 4.0/0.01 250 20 8.6 11 La(OTf) 3 4.0/0.01 255 20 11.4 12 La(OTf) 3 4.0/0.01 260 20 18.9 13 La(OTf) 3 4.0/0.01 265 20 21.7 14 La(OTf) 3 4.0/0.01 275 20 22.5 15 d La(OTf)3 4.0/0.01 270 20 1.2 16 e La(OTf)3 4.0/0.01 270 20 0 17 f La(OTf)3 4.0/0.01 270 20 0 18 g La(OTf) 3 4.0/0.01 270 20 25.5 19 h La(OTf) 3 4.0/0.01 270 20 12.6 a Reaction conditions: 50 mg lignin, 20 mg catalyst, 4 mL methanol, 0.01 mL water if needed, 24 h, 0.1 MPa Ar, and 500 rpm. b Guaiacol yield is calculated based on the weight of lignin. c Without catalyst. d Ethanol/water instead of methanol/water. e Cyclohexane/water instead of methanol/water. f Toluene/water instead of methanol/water. g Substrate is conifer wood enzymatic mild acidolysis lignin from pine (EMAL-p). h Substrate is grass enzymatic mild acidolysis lignin from bamboo (EMAL-b). It is known that hydrolysis of ether linkage is much easier than alcoholysis. 26,31 We found that water could improve the guaiacol yield (Table S1 † ), and a 22.5 wt% yield of guaiacol could be obtained ( Table 1 , entry 9). However, too much water in the catalytic system limited the production of guaiacol (Table S1, † entries 1–5), presumably due to the poor solubility of organosolv lignin in water. 40 According to previous studies, metal triflates exhibit excellent stability at high temperature. 33,37,39 In this strategy, the reaction temperature had a significant effect on the yield of guaiacol. As shown in Table 1 , the yield of guaiacol increased from 8.6 to 22.5 wt% with the temperature increasing from 250 to 270 °C, and then became independent of temperature ( Table 1 , entries 9–14). In addition, the solvents also influence the yield of guaiacol ( Table 1 , entries 9 and 15–17). Guaiacol was hardly obtained in ethanol, cyclohexane, or toluene. This is understandable because methanol is generally an effective hydrogen donor. 41 With methanol steam reforming, methanol could release hydrogen for catalytic transfer hydrogenation of the lignin intermediate products (for details see Table 2 and Fig. S7 † ), whereas ethanol, cyclohexane, and toluene are not good hydrogen donors. Furthermore, lignin has very low solubility in cyclohexane and toluene and is therefore not favorable for the transformation of lignin. 42,43 Therefore, methanol/water was the optimum solvent. In addition, the yield of guaiacol from lignin increased gradually from 3 to 22.5 wt% with the increase of La(OTf) 3 dosage ( Table 1 , entry 9 and Table S1 † entries 6–9). Under the determined optimal conditions, we also investigated the transformation of other lignin samples from conifer wood (EMAL-p) and grass (EMAL-b) lignin using La(OTf) 3 as the catalyst, and the yields of guaiacol were 25.5 and 12.6 wt%, respectively ( Table 1 , entries 18 and 19), further indicating the excellent catalytic performance of La(OTf) 3 . Results showing the effect of reaction temperature on the conversion of the lignin model compounds a \n \n Entry Temperature (°C) Conversion (%) Yield (%) \n 1 \n \n 2 \n \n 3 \n \n 4 \n 1 220 >99 >99 79 — — 2 230 >99 >99 95 4 — 3 240 >99 >99 86 12 — 4 250 >99 >99 71 17 11 5 260 >99 >99 49 18 31 6 270 >99 >99 31 11 49 a Reaction conditions: 50 mg lignin model compound a , 20 mg La(OTf) 3 , 0.1 MPa Ar, 500 rpm. 24 h, and methanol/water (4/0.01, v/v). To gain further information on the transformation of the lignin, lignin structures before and after the reaction were examined using 2D-HSQC NMR. 14,44,45 As shown in Fig. 2 , A α ( δ C / δ H 71.4/4.87, purple), A β ( δ C / δ H for G units: 84.1/4.32, purple; δ C / δ H for S units: 87.1/4.11, purple) and A γ ( δ C / δ H 59.9/3.81, purple) corresponding to the side chain of β- O -4 were almost disappeared. Other units (β–β and β-5) were no longer observed after the catalytic reaction. These results confirmed that the lignin was dissociated in the reactions, and perhaps beyond just the β-ethers. The cross signals correlating with the syringyl (S) ( δ C / δ H 104.0/6.72, red) and oxidized syringyl (S′) ( δ C / δ H 106.3/7.30, red) units were no longer present in the HSQC spectrum after the reaction, and only guaiacyl (G) units ( δ C / δ H 111/6.8, 115/6.7, 119/6.6, green) were observed, suggesting that demethoxylation of syringyl units had occurred to form guaiacyl units. 14 This is consistent with the aforementioned analysis of the liquid products. To further confirm the uniqueness of guaiacol in the liquid product, we used deuterated methanol (methanol- d 4 ) as the solvent in the lignin transformation. The 2D-HSQC NMR spectra of the obtained liquid product revealed that only guaiacol could be detected (Figs. S1D and 1E † ). Fig. 2 The 2D-HSQC NMR spectra of the organosolv lignin from Eucalyptus (top) and reaction residue (bottom) transformation (reaction conditions: Table 1 , entry 9). To further study the catalytic mechanism, the conversion of lignin model compound a was performed, and the results are shown in Table 2 , Fig. 3 and S2. † As shown in Table 2 , La(OTf) 3 could efficiently catalyze the conversion of a into guaiacol. The results illustrated that the β- O -4 linkage was broken to generate 4-ethylguaiacol at 220 °C. Moreover, as the reaction temperature increased, the C–C bond in the ethyl group of 4-ethylguaiacol was gradually cleaved to form 4-methylguaiacol and then further transformed into guaiacol, 21,25,46 which was futher established by the decarbonization of 4-ethylguaiacol ( Fig. 3 ). As shown in Fig. 3A , 4-methylguaiacol was initially the major liquid product due to the cleavage of the C–C bond between C α and C β in the side chain of 4-ethylguaiacol. With an extension of the reaction time, the remaining methyl chain in 4-methylguaiacol was cleaved, and then guaiacol was formed. 21,47 Correspondingly, methane was generated as the main gaseous product ( Fig. 3 ), which confirmed that guaiacol was indeed formed by successive cleavage of the C–C bond in the side chain, as depicted in Fig. 3B . This is different from various reported studies. 12,33,34,36 Although Lewis acids have been used as catalysts 12,36 or one of the active components of the catalytic system 33,34 for transformation of lignin into aromatic chemicals, only a trace of guaiacol was usually generated together with various other compounds, mainly because the reaction was conducted at much lower reaction temperatures in the reported work. In addition, we also used ethanol as a solvent to study the reaction (Fig. S4 † ). The product distribution was different from that in methanol. 4-Vinylguaiacol was the major product in ethanol for the side-chain elimination reaction, which revealed again that methanol was an excellent organic solvent. 40,48,49 Fig. 3 (A) The time-course of the product distributions for the C–C bond cleavage of the ethyl group in 4-ethyl guaiacol with the La(OTf) 3 catalyst. Reaction conditions: 4-ethyl guaiacol (50 mg), La(OTf) 3 (20 mg), methanol 4 mL, 10 μL water, 270 °C, 0.1 MPa Ar, and stirring at 800 rpm. (B) Reaction pathway for C–C bond cleavage of the ethyl group in 4-ethyl guaiacol. Although the amount of S units is higher than that of G units in the organosolv lignin, guaiacol was the only monomeric product observed and the corresponding residual solid contained only G units, as determined by NMR. Therefore, the reactions of S units in lignin model compound b (see ESI † ) and 2, 6-dimethoxy-4-methylphenol were also investigated (Fig. S5 and S6 † ). The yield of alkyl-syringol was almost negligible, whereas guaiacol was generated in high yield. The results of these experiments using model compounds are therefore consistent with the lignin transformation results in revealing that even syringyl units really do produce guaiacol ( via demethoxylation) under these conditions. Under optimal conditions, we also performed a scale-up experiment using a larger amount of organosolv lignin. After the separation and purification, a 21.2 wt% yield of isolated guaiacol (0.32 g) could be obtained from 1.50 g lignin ( Fig. 4 and S7 † ), indicating that pure guaiacol with appreciable yield could be produced from real lignin via this strategy. Fig. 4 Scale-up production of guaiacol from lignin. Reaction conditions: 1.50 g organosolv lignin, 0.6 g La(OTf) 3 , 120 mL methanol, 0.3 mL water, 270 °C, 0.1 MPa Ar, and 24 h."
} | 3,884 |
38226148 | PMC10788777 | pmc | 7,112 | {
"abstract": "Brain-inspired computing systems require a rich variety of neuromorphic devices using multi-functional materials operating at room temperature. Artificial synapses which can be operated using optical and electrical stimuli are in high demand. In this regard, layered materials have attracted a lot of attention due to their tunable energy gap and exotic properties. In the current study, we report the growth of layered MoO 3 using the chemical vapor deposition (CVD) technique. MoO 3 has an energy gap of 3.22 eV and grows with a large aspect ratio, as seen through optical and scanning electron microscopy. We used transmission electron microscopy (TEM) and X-ray photoelectron spectroscopy for complete characterisation. The two-terminal devices using platinum (Pt/MoO 3 /Pt) exhibit superior memory with the high-resistance state (HRS) and low-resistance state (LRS) differing by a large resistance (∼MΩ). The devices also show excellent synaptic characteristics. Both optical and electrical pulses can be utilised to stimulate the synapse. Consistent learning (potentiation) and forgetting (depression) curves are measured. Transition from long term depression to long term potentiation can be achieved using the spike frequency dependent pulsing scheme. We have found that the amplification of postsynaptic current can be tuned using such frequency dependent spikes. This will help us to design neuromorphic devices with the required synaptic amplification.",
"conclusion": "4 Conclusions In summary, we fabricated multilayered, low dimensional MoO 3 based artificial synaptic devices using pre-fabricated Pt electrodes separated by 5 μm. The initial characterisation shows that the CVD-prepared α-MoO 3 are grown in a lamellar structure evidenced by high resolution transmission electron microscopy images. We observe excellent synaptic characteristics, including consistent potentiation and depression curves lasting multiple cycles. The potentiation and depression curves follow a double exponential growth. The excitatory post-synaptic current (EPSC) depends on the spike rate and we find that the device can show transition from short-term memory to long-term memory depending on the spike rate applied. By repeated learning–forgetting–learning operations, the memory retention can be improved. Our studies on lamellar MoO 3 can pave the way to design and fabricate the high performing artificial synaptic devices. By fine tuning the spike frequency of operation, we can achieve 100% enhancement in memory which is essential in designing optimally operating neuromorphic devices in future computing applications.",
"discussion": "1 Results and discussion Lamellar MoO 3 sheets were synthesised using the Chemical Vapor Deposition technique (CVD) where, MoO 3 powder was used as the precursor (see Fig. 1a and b ). 3 g of MoO 3 powder was taken in a quartz crucible and kept inside the hot zone region of the CVD tube under an oxygen atmosphere of 100 sccm flow for 3 h at 750 °C and 2 millibar pressure. We obtained lamellar MoO 3 with sizes ranging from a few millimeters to centimeters in length ( Fig. 1d ) and this was carefully collected from the cold zone of the CVD tube after the chamber was cooled to room temperature. A few milligrams of as-synthesised material was sonicated for a duration of two hours in ethanol. The product obtained was then cleaned and washed with ethanol using centrifugation (1000 rpm/5 min), where the residue was separated from the supernatant and dried. The final product was preserved in a glass vial for further characterisation and device fabrication. Before investigating the electrical properties and checking the synaptic properties, it is necessary to establish the crystallographic phase of the layered MoO 3 prepared. Two of the most commonly known phases of MoO 3 are α-MoO 3 and the metastable β-MoO 3 . 19,46,47 As reported, a spontaneous phase transition between β → α occurs in the temperature range of 375–400 °C. 48 The interesting phase of MoO 3 is the α-phase, so we kept the CVD chamber temperature well above the phase transition temperature so that the product phase is completely the α-phase and ruling out any possibility of mixed phase formation. A typical XRD-pattern obtained for the large multilayered MoO 3 samples is shown in Fig. 1e . The strongest diffraction peaks are found at 2 θ values of 12.3°, 22.9° and 38.5° corresponding to the (020), (040) and (060) crystallographic planes, respectively. As shown in Fig. 1f , the layered MoO 3 grow with very high aspect ratio which is essential for our purpose. The UV-vis spectroscopy done on such layered MoO 3 shows that the energy gap is ∼3.22 eV as shown using the Tauc plot (inset Fig. 1e ). 49 Transmission electron microscopy images shown in Fig. 1g–i reveal that the MoO 3 grow in the form of sheets and atomically resolved images suggest that layers grow along [01] and [10] showing orthorhombic MoO 3 sheets. The Fourier transform of the topography image clearly indicates the bright spots along [01] and [10] clearly indicating the growth of orthorhombic MoO 3 sheets. A detailed calculation of the lattice fringes is shown in ESI Fig. S1. † From the average distance between the spots along the (10) and (01) directions, we find that the real lattice dimensions are 0.276 nm along the (10) direction and 0.274 nm along the (01) direction. A detailed analysis of the TEM images is shown in Fig. S1 in the ESI. † To determine the chemical composition and the chemical states of the MoO 3 surfaces, X-ray photoelectron spectroscopy (XPS) was performed over the lamellar layers and is shown in Fig. 2 . The survey scan shows the characteristics peaks of molybdenum (Mo), oxygen (O), trace amounts of carbon (C) and then the O-KLL Auger peaks. A high resolution scan was conducted in order to determine the valence band edge of the MoO 3 sample (inset Fig. 2a ). The valence band edge is found to be at 2.79 eV ( x -intercept of the red line scan of increasing intensity). The overall intensity at this low energy scan can be fitted with multiple peaks located at 4.68 eV, 6.47 eV and 8.56 eV respectively. The broad intensity observed in the energy range between 1 eV to 6 eV is due to metallic Mo coming from Mo-4d and Mo-5s orbitals,. 50,52 while O-2p orbital contribution is observed between 4 eV to 10 eV indicating there is a strong overlap between the Mo-4d and O-2p orbitals in the MoO 3 as expected in the pure α-MoO 3 . It is to be noted that we do not see the doublet near to the Fermi energy which is associated with the presence of MoO 2 phase in the sample. 50 High resolution narrow scans are done at the binding energy range of the Mo-3d core level and oxygen-1s core levels to identify the chemical states of the elements ( Fig. 2b and c respectively). Well-resolved Mo-3d 5/2 and Mo-3d 3/2 levels due to spin–orbit splitting occur at 232.58 eV and 235.68 eV respectively. These are doublet states of molybdenum in the Mo 6+ state. The energy gap between Mo-3d 5/2 and Mo-3d 3/2 levels is 3.1 eV and the areas under the curve are in the ratio of 3 : 2 indicating clearly that the MoO 3 has molybdenum in the Mo 6+ oxidation state. We do not see the corresponding peak for Mo 5+ occurring at lower binding energies as seen in the literature. 51 A clear peak at 530.48 eV shown in Fig. 2c is due to the O 2− ions present in the MoO 3 layers. The binding energy peak at 530.48 eV of oxygen clearly confirms lattice oxygen rather than oxygen defects. If we consider the small shoulder at 532.68 eV of the oxygen peak, which is coming from the nonlattice oxygen, the concentration of such nonlattice oxygen content is about 9.45% which we consider as very low. This is due to the higher partial pressure of oxygen we have used during the growth of MoO 3 wherein oxygen vacancies are compensated most efficiently. We do not observe any adsorbed water molecules on the MoO 3 layer which would show up as OH–MoO 3 resulting in the reduction of MoO 3 to MoO 2 . 51 The peak at 284.52 eV shown in Fig. 2d corresponding to the C-1s peak is from the standard for the corrections to the data obtained from the XPS (adventitious carbon peaks (AdC)). We can also observe the C–O and C \n \n\n<svg xmlns=\"http://www.w3.org/2000/svg\" version=\"1.0\" width=\"13.200000pt\" height=\"16.000000pt\" viewBox=\"0 0 13.200000 16.000000\" preserveAspectRatio=\"xMidYMid meet\"><metadata>\nCreated by potrace 1.16, written by Peter Selinger 2001-2019\n</metadata><g transform=\"translate(1.000000,15.000000) scale(0.017500,-0.017500)\" fill=\"currentColor\" stroke=\"none\"><path d=\"M0 440 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z M0 280 l0 -40 320 0 320 0 0 40 0 40 -320 0 -320 0 0 -40z\"/></g></svg>\n\n O bonds completely coming from the adventitious carbon present during the investigation. From the XPS spectrum, we can infer that our CVD grown MoO 3 layers are close to MoO 3 stoichiometry with the possibility of having a small oxygen deficiency. 52 Fig. 2 Detailed XPS analysis of layered MoO 3 : (a) survey scan of MoO 3 . The inset shows the detailed scan of the valence band edge. The line fit shows that the valence band edge is 2.79 eV from the Fermi level. Multipeak fitting reveals the presence of peaks at 4.68 eV, 6.47 eV and 8.56 eV which can be assigned to metallic Mo (4d and 5s) orbitals, O-2s orbitals with strong overlapping of Mo-4d and O-2s orbitals. (b) High resolution binding energy scan of Mo-3d orbitals. (c) High resolution binding energy scan of O-1s orbitals. (d) Adventitious carbon peaks (AdC) reveal the energy levels of single bonds and double bonds between carbon and oxygen atoms."
} | 2,413 |
22439012 | PMC3306314 | pmc | 7,113 | {
"abstract": "Elevated ocean temperatures and agrochemical pollution individually threaten inshore coral reefs, but these pressures are likely to occur simultaneously. Experiments were conducted to evaluate the combined effects of elevated temperature and the photosystem II (PSII) inhibiting herbicide diuron on several types of symbiotic algae (diatom, dinoflagellate or rhodophyte) of benthic foraminifera in hospite . Diuron was shown to evoke a direct effect on photosynthetic efficiency (reduced effective PSII quantum yield ΔF / F′ m ), while elevated temperatures (>30°C, only 2°C above current average summer temperatures) were observed to impact photosynthesis more indirectly by causing reductions in maximum PSII quantum yield ( F v / F m ), interpreted as photodamage. Additionally, elevated temperatures were shown to cause bleaching through loss of chlorophyll a in foraminifera hosting either diatoms or dinoflagellates. A significant linear correlation was found between reduced F v / F m and loss of chlorophyll a . In most cases, symbionts within foraminifera proved more sensitive to thermal stress in the presence of diuron (≥1 µg L −1 ). The mixture toxicity model of Independent Action (IA) described the combined effects of temperature and diuron on the photosystem of species hosting diatoms or dinoflagellates convincingly and in agreement with probabilistic statistics, so a response additive joint action can be assumed. We thus demonstrate that improving water quality can improve resilience of symbiotic phototrophs to projected increases in ocean temperatures. As IA described the observed combined effects from elevated temperature and diuron stress it may therefore be employed for prediction of untested mixtures and for assessing the efficacy of management measures.",
"introduction": "Introduction A dramatic decline in coral cover has been recorded in the last three decades [1] , primarily driven by an increasing frequency of climate-related mass mortality events [2] , [3] . Predicted increases in the frequency and duration of high summer temperatures [4] exceeding species' thermal tolerance thresholds present a significant risk to the biodiversity of coral reefs and to the services they provide [5] , [6] . It has been implied that the earliest symptoms of heat damage in reef-building organisms are associated with limitation of photosynthetic electron flow [7] – [11] and carboxylation within the Calvin cycle of symbiotic microalgae [12] , [13] . Excess excitation energy that cannot be utilized in photochemical charge separation subsequently overwhelms photoprotective mechanisms, leading to oxidative stress and photoinhibition [14] . In corals and other symbiotic reef species, this can cause loss of symbiotic algae or reduced pigment concentrations (bleaching). Pollution from terrestrial runoff also negatively affects reef health [15] . In the last century and a half, intensive agriculture and industries along the Queensland coastline have significantly increased the annual input of suspended sediments and nutrients into the Great Barrier Reef (GBR) lagoon [16] . Correspondingly, the use of pesticides in catchments that flow into the GBR has been growing steadily [17] . Recent studies have found contemporary herbicides to be ubiquitous in nearshore areas of the GBR [18] – [20] . Photosystem II (PSII) herbicides are of particular ecological concern with regards to reef systems, as they are relatively mobile and safety margins between chronic environmental concentrations and effect concentrations as determined by laboratory studies are relatively small [21] . These compounds act by inhibiting electron transport through the photosystem in chloroplasts by reversibly binding to a specific electron-acceptor protein (D1-enzyme) in PSII, outcompeting the normal ligand for binding sites [22] . Intracellular microalgae of symbiotic reef species are likewise affected by herbicides and will suffer reduced photosynthetic efficiency, limiting energy flow from symbiont to host [23] . High inhibitor concentrations or sustained blockage of the electron transport chain can cause secondary effects through the reduced availability of ATP and NADPH and the formation of reactive oxygen species, causing chronic photoinhibition [24] . Inhibition of photosynthesis in symbiotic organisms can decrease production and leads to photosystem damage and in turn bleaching, thus disturbing the fragile relationship so important in coral reef ecology. However, not all symbiont bearing organisms display equal vulnerability to environmentally adverse conditions. For example, genetic diversity within dinoflagellates of the genus Symbiodinium that complement the host-symbiont relationship in corals largely influences holobiont resilience to environmental stress such as higher than usual temperatures [25] , [26] or pollution [23] , allowing for ecological adaptation. While the majority of stress-response studies have been performed on corals, other species may also be at risk. Foraminifera are single-celled protists that may form symbiotic relationships with several different microalgal phyla, including diatoms, dinoflagellates, red and green algae [27] . Foraminifera are widespread, sensitive to environmental change [28] , [29] and have been documented to bleach under stressful conditions in the field [30] – [34] . Furthermore, they can be employed as indicator species for water quality assessment [35] , [36] . A wide suit of laboratory studies have demonstrated reduced growth and/or bleaching at elevated temperatures, UV or increased irradiance, elevated nutrient levels or combinations of stressors for several species [37] – [41] . Recently, we demonstrated how several different species of benthic foraminifera, hosting four different microalgal phyla, exhibit widely varying responses to the PSII herbicide diuron [42] , while another recent study has linked changes in foraminiferal community structure to an increase in terrestrial runoff [43] . In the tropics, summer monsoonal rainfall and subsequent river flooding occur when sea surface temperatures (SSTs) approach tolerance threshold levels for many species, thereby simultaneously exposing inshore reefs to combinations of low salinity, high turbidity, nutrients and agrochemical residues during episodes of thermal stress. Despite this, water quality guidelines are based on known thresholds and impacts of single stressors, reflecting the majority of stress-response studies, while environmentally important combinations of stressors are rarely considered. Regulatory agencies have recently recognized the potential for pollution to reduce the resilience of reef systems and have adopted strategies of optimizing water quality in order to protect sensitive species to the effects of global climate change [44] , [45] . However, empirical support for this strategy is limited and heavily relies on results obtained from studies on hard corals [46] – [50] . Increased temperatures may cause conformational changes in the D1-protein and so change a herbicide's binding affinity [51] . Furthermore, as in corals, foraminiferal symbionts reside within their host cells [27] and thus the herbicide will have to cross several membrane layers (of both host and algal origin) to reach its target site on the D1-protein [52] . Temperature affects membrane permeability and internal cellular processes such as protein repair mechanisms or bio-elimination and may therefore enhance or reduce toxicity of pollutants [53] . Since thermal stress and herbicides both target symbiont photochemistry, additive or interactive effects may occur, as has been recently shown in corals [46] . The aims of the present study were to test how the susceptibility (thresholds) of various symbiotic partnerships (benthic foraminifera and their intracellular microalgae) to the adverse effects of elevated SSTs changes in the presence of the PSII herbicide diuron.",
"discussion": "Discussion Synthesis This was the first investigation into combined effects of elevated temperature and low levels of pollution on the photochemistry of different symbiotic partnerships in various species of foraminifera. The negative effects of elevated temperatures on symbiont photochemistry were in most cases more severe in the presence of low concentrations of diuron, as might be expected since both factors are independently capable of impacting PSII. Diuron had a direct effect on photosynthetic efficiency, while elevated temperatures impacted photosynthesis indirectly by causing photodamage. These effective temperatures correspond with not unlikely predictions and may occur by the end of this century. Low concentrations of diuron were found to reduce temperature thresholds for inhibition of photosynthesis and, to a somewhat lesser extent, the onset of photodamage. Additionally, elevated temperatures were shown to cause bleaching through loss of Chl a in both H. depressa and M. vertebralis . A moderately significant correlation was found between reduced F v / F m and loss of Chl a , linking photodamage to bleaching. As the observed combined effects demonstrate a high level of agreement with the predicted combined effects as calculated through the combined effect model of Independent Action (IA), the experimental data are indicative of response additivity for this combination of stressors. Temperature For the species evaluated here, elevated temperatures had an equivalent inhibitory effect on ΔF / F′ m and F v / F m of symbionts in hospite , with the photosystem of diatom-bearing H. depressa and A. quoyi proving most sensitive. In these species, 96 hour exposures to temperatures of 32°C caused a considerable drop in ΔF / F′ m and F v / F m , while 96 hours at 34°C proved lethal. Temperature-induced photodamage in H. depressa and A. quoyi as reported in this study was less extensive than described in a recent heat stress study on related diatom-bearing foraminifera from the Great Barrier Reef [62] . In that study, F v / F m in H. depressa was reduced by 25–45% after 6 days incubations at 32°C (as opposed to 10–20% reduction after 96 hours here), which is a further indication that photodamage caused by heat stress is likely to increase over time as demonstrated here ( Fig. 4C ). Calcarina mayorii , the other species tested hosting diatom symbionts, and M. vertebralis bearing dinoflagellates, were equally affected by temperature and somewhat less sensitive than H. depressa and A. quoyi , with only 96 hours at temperatures over 32°C affecting ΔF / F′ m and F v / F m ( Figs. 1 and 3 ). A recent study observed ∼15% inhibition ΔF / F′ m in M. vertebralis when specimens were incubated at 32°C for 4 days, however higher light intensities were used [39] . Depth distributions of foraminifera vary greatly and are determined by temperature, light attenuation, water movement and substrate [68] , [69] . On the GBR localities where experimental foraminifers were collected, H. depressa and A. quoyi typically exist in more shaded environments at depths between 9–15 m, often hidden deep within coral rubble or macroalgae and are therefore adapted to low irradiance and stable temperatures. Calcarina mayorii and M. vertebralis on the other hand, were collected at less than 7 m depth and are often found in shallower waters (or even on the reef flat) and will therefore be subject to greater temperature fluctuations, wave energy and higher solar irradiance. The dissimilar habitat types and associated adaptive ecology may partly explain the observed differences in sensitivity between species and related symbiont types. In accordance with results obtained in current study, Negri and colleagues [46] recently reported results where 10–15% inhibition of ΔF / F′ m and F v / F m was observed in symbiotic dinoflagellates hosted by the branching coral Acropora millepora after 7 days incubation at 32°C, albeit at much higher levels of irradiance. At higher light intensities, the effects of heat stress may be intensified as absorbed excitation energy that cannot be transferred in photochemical pathways is instead passed on to form additional reactive oxygen [70] . Eventually, photo-oxidative stress renders symbiotic microalgae inefficient, potentially triggering a loss of symbionts (bleaching), possibly by host digestion [31] , [71] or expulsion of the symbionts, a process frequently observed in hard corals [12] , [72] . The fact that chronic photodamage was observed in this study under very low light intensities, further supports the suggestions that foraminifera are vulnerable to overexcitation and that irradiance intensity is an important limiting factor for the distribution and survival of species [42] , [68] , [73] – [75] . Diuron Exposure to 1 and 3 µg L −1 diuron for 96 hours significantly affected ΔF / F′ m but the impact on F v / F m was less pronounced. Both ecotypes of M. vertebralis , hosting dinoflagellate symbionts, were slightly more sensitive to the negative effects of diuron than the species hosting diatoms, of which H. depressa was most vulnerable. Diuron-induced inhibition of ΔF / F′ m after 96 hours as evaluated in this study was more severe than observed in a recent study [42] . However, that study used a lower experimental irradiance intensity (5 versus 10 µmol quanta m −2 s −1 PAR) to examine a more acute effect (over 48 h). Previously we reported 10–25% inhibition ΔF / F′ m in symbiotic diatoms after 48 hours incubation, against 20–35% inhibition observed in this study and this trend was similar for symbiotic dinoflagellates. Results obtained here indicate that maximum inhibition of ΔF / F′ m in response to diuron is reached after 24 hours ( Fig. 4 ), thus implying that different experimental irradiance intensities are primarily responsible for observed differences between the studies. The insensitivity of the red algae in this study can be explained by the fact that red algae can balance the excitation energy distribution between PSI and PSII, restricting herbicide effects and oxidative stress [76] , limiting the usefulness of the saturation pulse method to assess the photosystem of red algae. Combined effects Two stressors are considered biologically independent when the qualitative nature of the mechanism of action of one is not affected by the presence or absence of the other [77] . Moreover, the assessment of potential interaction can be unambiguously made, when taking this explicit non-interaction consideration as a reference model for evaluating observed combination effects. The combined effect model of IA used here revealed very consistent results across species with only minor diversions from the response additive model. The high level of agreement between the observed and predicted (IA) response demonstrates that the underlying simplistic mixture theory for this combination of stressors is valid and provides a useful tool for predictive modeling. ANOVA demonstrated significant individual effects of herbicide exposure and elevated temperature, with the only statistical interaction observed for inhibition ΔF / F′ m in M. vertebralis ( Table 2 ). In that species deviations from the predicted IA model indicated the response to these combined pressures to be sub-additive. Although the results reveal obvious differences in the sensitivity of a variety of symbiotic partnerships to the stressors tested, the experimental data are clearly consistent with IA, thus providing evidence that the risk from this combination of stressors is greater than from individual components. Symbiont type and shell ultrastructure Similar to corals, foraminifera have been reported to host a wide variety of dinoflagellate clades of the genus Symbiodinium \n [78] , [79] and these different clades may confer different stress tolerance characteristics, as has been reported for symbiotic dinoflagellates in corals [25] , [26] . Cantin and colleagues [23] demonstrated that reduced photosynthetic output limits the translocation of carbohydrates from symbiont to host and that this effect was dependent on symbiont type. Reduced energy acquisition could decrease overall fitness and resilience of the host animal to further stressors. The results obtained here suggest that foraminifera hosting diatoms are more vulnerable to temperature stress and species hosting dinoflagellates more vulnerable to the effects of herbicides. Recently, we suggested a species' ultrastructure may influence diuron biokinetics as we observed delayed uptake and effect in porcelaneous (imperforate) species as opposed to hyaline (perforate) species [42] . Despite an equal sensitivity to the effects of elevated temperature, the current study revealed a slightly greater sensitivity to diuron in H. depressa when compared with co-existing A. quoyi (hyaline versus porcelaneous, respectively; both hosting diatoms). Marginopora vertebralis (porcelaneous; dinoflagellates) was more heavily affected by diuron than either hyaline or porcelaneous diatom-bearing species, while at the same time less vulnerable to elevated temperature as assessed by PAM-fluorometry. Following our results, improving water quality (by reducing herbicide levels) will have the greatest effect for the diatom-bearing species H. depressa and A. quoyi , as these species are most likely to suffer the effects from elevated SSTs. The symbiotic dinoflagellates tested here may be less vulnerable to thermal pressure than diatoms, but suffer more stress from herbicides alone. For short-term exposures this may not be a problem, but it remains unclear how foraminifera respond to longer-term exposures of physical or chemical stress. Ecological effects Both ΔF / F′ m and F v / F m have been found to rapidly recover after the responsible stress factor had been removed in various corals [80] , [81] and benthic foraminifera [42] . However, sustained reductions in F v / F m can lead to reduced growth and loss of symbionts or photosynthetic pigmentation, as has been observed in corals [82] and now foraminifera. Corals exposed to diuron for 2–3 months exhibited decreased lipid content, bleaching, curbed reproductive success as well as colony mortality [82] . The correlation between reduced F v / F m and bleaching has often been observed in corals as a consequence of environmental stressors such as high temperatures [83] , high irradiance intensities [70] , reduced salinity [84] , herbicides [85] or combinations thereof [46] . While tissue bleaching in corals is considered a sub-lethal stress response and may be reversed, partial or whole-colony mortality often results [5] . In foraminifera, evidence exists of bleached populations recovering in late summer and fall or over multiple years following mass bleaching events [32] , [86] . On the other hand, environmental stress can induce abnormal reproduction and cellular damage [71] , [87] , affect structural integrity and immune response [28] , [88] , potentially leaving species vulnerable to disease, predation and further stressors. Moreover, previous studies have demonstrated how assemblages can shift from being dominated by large, symbiont-bearing foraminifers to smaller, herbivorous or detrivorous species under the influence of environmental stress [28] , [89] . Implications Whereas thermal stress has been proposed as the main physiological driver behind mass coral bleaching events [2] , [90] , evidence is emerging that water quality may have a strong influence on the sensitivity of reef species to physical stressors as elevated SSTs and ocean acidification [15] , [39] , [46] , [47] , [91] , [92] . In Queensland and other tropical environments, high summer temperatures often coincide with monsoonal rainfall events, responsible for the delivery of the highest annual loads of fresh water, sediments, nutrients and associated pesticides onto nearshore areas of the GBR [19] , [93] . Thus it is likely that inshore primary producers such as corals, seagrasses and foraminifera are simultaneously exposed to chemical and physical stressors. Our results indicate that minimizing pollution can reduce total pressure phototrophic organisms experience under conditions of thermal stress. Water quality guidelines for contaminants as well as laboratory experiments directed at evaluating temperature stress thresholds often do not take into consideration the highly likely scenario that sensitive organisms will be exposed to combined and/or cumulative stressors, potentially underestimating the true extent of environmental pressure. These pressures on reef ecosystems are likely to increase further as a result of expanding coastal development, population growth and climate change. While limiting the effects of climate change is a global challenge, policies minimizing the effects of pollution can contribute towards the survival and sustainable exploitation of our marine resources. Restricting the inflow of suspended sediments, nutrients and chemical contaminants represents a practical local strategy to protect our reef ecosystems in a changing environment [15] , [94] , [95] ."
} | 5,317 |
36199189 | PMC10092097 | pmc | 7,115 | {
"abstract": "Euendolithic, or true‐boring, cyanobacteria actively erode carbonate‐containing substrata in a wide range of environments and pose significant risks to calcareous marine fauna. Their boring activities cause structural damage and increase susceptibility to disease and are projected to only intensify with global climate change. Most research has, however, focused on tropical coral systems, and limited information exists on the global distribution, diversity, and substratum specificity of euendoliths. This metastudy aimed to collate existing 16S rRNA gene surveys along with novel data from the south coast of South Africa to investigate the global distribution and genetic diversity of endoliths to identify a “core endolithic cyanobacterial microbiome” and assess global diversification of euendolithic cyanobacteria. The cyanobacterial families Phormidesmiaceae, Nodosilineaceae, Nostocaceae, and Xenococcaceae were the most prevalent, found in >92% of categories surveyed. All four known euendolith clusters were detected in both intertidal and subtidal habitats, in the North Atlantic, Mediterranean, and South Pacific oceans, across temperate latitudes, and within rock, travertine tiles, coral, shell, and coralline algae substrata. Analysis of the genetic variation within clusters revealed many organisms to be unique to substratum type and location, suggesting high diversity and niche specificity. Euendoliths are known to have important effects on their hosts. This is particularly important when hosts are globally significant ecological engineers or habitat‐forming species. The findings of this study indicate high ubiquity and diversity of euendolithic cyanobacteria, suggesting high adaptability, which may lead to increased community and ecosystem‐level effects with changing climatic conditions favoring the biochemical mechanisms of cyanobacterial bioerosion.",
"conclusion": "CONCLUSIONS This analysis allowed the identification of a core endolith microbiome, with cyanobacteria belonging to known euendolith clusters being ubiquitous; they were detected in both intertidal and subtidal habitats, in all oceans and in all bands of latitude included in the metastudy. In addition, euendoliths were observed in all host substrata with the exception of isopods, though these were represented by a single study. The Leptolyngbya ‐like cluster was detected in all habitats, substrata, and geographic locations, while the Pleurocapsalean cluster and Mastigocoleus testarum‐ like cluster were detected in the majority of these categories. This emphasizes the high diversity and ubiquitous nature of euendolithic cyanobacteria indicating high adaptability of euendoliths and therefore their potential to affect a wide range of globally important substrata and ecosystems. The biochemical changes expected in global aquatic systems with projected climate change favor the mechanisms of bioerosion by euendoliths, which may lead to increased community and ecosystem‐level effects. The few studies that have examined the ecological effects of euendoliths in detail have highlighted both the threats and advantages that they can pose to their hosts (Kaehler and McQuaid 1999 , Zardi et al. 2009 , Ndhlovu et al. 2019 ), while the present study emphasizes their genetic diversity and specificity to substrata and bioregions. This makes an understanding of how host/endolith relations will alter under global climate change particularly challenging.",
"discussion": "DISCUSSION Core cyanobacterial community The first objective of this study was to identify whether there was a core microbiome of lithobiontic cyanobacteria across calcium carbonate structures in different substrata and geographical systems. The Phormidesmiaceae and Nodosilineaceae of the Phormidesmiales, the Xenococcaceae, Nostocaceae, and Chroococcidiopsidaceae of the Nostocales and the Leptolyngbyaceae of the Leptolyngbyales formed a group of cyanobacterial families that were almost ubiquitous over all samples analyzed. In terms of sequence identification using the SILVA database for ASVs falling within the clusters identified by Roush et al. (2020), sequences placed within the euendolith Cluster 1 (the Leptolyngbya‐ like group) comprised the Nodosilineaceae and the Phormidesmiaceae and may also include the well‐documented euendolith Plectonema terebrans (Roush and Garcia‐Pichel 2020 ). Cluster 2, the Pleurocapsalean group, is a diverse group that encompasses known euendoliths including Hyella sp ., Solentia sp., and Hormathonema sp. which fall within the Xenococcaceae. The UBC cluster is a cryptic group whose closest alignment is to the Xenococcaceae (Roush and Garcia‐Pichel 2020 ); however, further investigation is necessary to classify the group. Sequences that were placed within Cluster 3, the Mastigocoleus testarum ‐like group, were contained in the Nostocaceae, the family that includes the known euendoliths Kyrtuthrix dalmatica (Palinska et al. 2017 , Ndhlovu et al. 2019 ) and Scytonema endolithicum (Gektidis et al. 2007 ). It is worth noting that the euendolith clusters from Roush et al. (2020), used to guide this study introduce a bias to the dataset that should be recognized; the clusters were determined from research focused on intertidal rock and travertine tiles in the Caribbean. Distribution of Euendolithic cyanobacteria The second objective was to examine the distribution of the euendoliths. This analysis revealed that all clusters were present in temperate marine systems, on rock, travertine, coral, shell, and coralline algae. Neither latitudinal gradients nor ocean connectivity/proximity was informative in explaining the presence of the euendolith cluster across the biogeographic regions. For example, the UBC cluster was notably absent from several geographical categories, with no obvious pattern to its absence. The Pleurocapsalean cluster and the UBC were both absent from the latitudinal category of 9° N–10° S despite this containing five studies that included samples from coral, coralline algae, and rock, in all of which the UBC cluster was found elsewhere at other latitudes (Fig. 5 ). The UBC cluster was a relatively small group with fewer occurrences and lower diversity than other clusters (Fig. 6 ), which suggests a relatively new group that has not diversified to have the greater temperature and substratum specificity shown by other clusters. The UBC and the Pleurocapsalean cluster were notably absent from equatorial marine samples, possibly as a result of a sampling bias toward more active grazing species in the case of the Pleurocapsalean cluster (Grange et al. 2015 ) and sampling of more mature euendolith communities (Roush and Garcia‐Pichel 2020 ). Nevertheless, their presence in equatorial samples was expected. Some euendolithic species are more easily removed by grazing, such as Hyella sp. (contained within the Pleurocapsalean cluster) and Mastigocoleus testarum (Grange et al. 2015 ) due to their shallow boring or poor attachment. These genera were also absent from echinoderms and polychaete tubes, and the Pleurocapsalean cluster was only absent from chitons. This may be explained by high grazing activity on those substrata by other organisms, or self‐cleaning by the organism, which may remove weakly attached bacteria, allowing deeper borers, or cyanobacteria with greater attachment strength, to persist. The absence of any geographic trends in endolith distribution among oceans or latitudes is presumably a result of the width of the latitudinal brackets used, while oceans include many habitats and bioregions with a wide range of conditions of light intensity and temperature. However, using ocean as a metadata category for the phylogenetic trees of individual clusters allowed the observation of ASVs that were detected more widely than others. For example, the “Ocean” categories demonstrate the larger geographical distribution better than “Latitude” for the branches within the phylogenetic trees. Of particular interest was the genetic diversity analysis of the UBC. Over half of the ASVs were identified in stromatolites, with half found exclusively there. The topography of the tree suggests stromatolites as an evolutionary origin for the cluster, with UBC:A (Fig. 6 ) diverging from stromatolites and being identified exclusively in rock and travertine. The single ASV observed in corals in the North Atlantic is particularly interesting, as it could represent radiation of this recently discovered group into corals as a new substratum, making it of future concern over coral reef degradation should it disperse to other latitudes and geographical areas. Where present, the UBC is an important pioneer cluster in euendolith colonization dynamics (Roush and Garcia‐Pichel 2020 ), so it would be of interest to further investigate the identity, origin, and geographical expansion of the cluster. The cyanobacterial community was most similar among coralline algae, shells, coral, and rock (including travertine), and these substrata collectively contained all the previously identified euendolith clusters. ASVs tended to be quite substratum‐specific, usually being restricted to either rock and/or travertine and one other substratum, with very few being present in both coral and shell. This suggests high substratum (as well as geographic and niche) specificity of euendoliths, contrary to previous suggestions that phototrophic endoliths are generalists (Tribollet 2008 , Marquet et al. 2013 , Ndhlovu et al. 2019 ). Although limited to a single study (Wenzel et al. 2018 ), no euendoliths were detected on isopod carapaces. Samples from isopods contained no photosynthetic cyanobacteria or other euendoliths. This could be a result of frequent molting of the isopods in the included study ( Jaera albifrons ) meaning the protective epicuticle layer over the calcium carbonate layers does not become sufficiently eroded to allow euendolith communities to establish themselves. Coralline algae, many shell‐bearing organisms, and corals are important ecosystem engineers and are important for fostering biodiversity (Buschbaum et al. 2009 , Gómez‐Lemos et al. 2018 ). It is known that microbial dissolution of corals (Tribollet et al. 2009 , Reyes‐Nivia et al. 2013 ) and coralline algae (Reyes‐Nivia et al. 2014 ) increases under conditions of elevated p CO 2 . Consequently, changes in the identity and boring efficiency of the endolithic communities associated with ecological engineers as a result of changes in ocean pH and temperature will have important effects on their fitness (Kaehler and McQuaid 1999 ), including the ability of corals to survive and recover from bleaching events (Pernice et al. 2020 ). This in turn is likely to lead to indirect cascading effects of climate change on the diverse communities that rely on them."
} | 2,710 |
35222314 | PMC8863614 | pmc | 7,116 | {
"abstract": "It is well known that speciation transformations of As(III) vs. As(V) in acid mine drainage (AMD) are mainly driven by microbially mediated redox reactions of Fe and S. However, these processes are rarely investigated. In this study, columns containing mine water were inoculated with two typical acidophilic Fe/S-oxidizing/reducing bacteria [the chemoautotrophic Acidithiobacillus (At.) ferrooxidans and the heterotrophic Acidiphilium (Aph.) acidophilum ], and three typical energy substrates (Fe 2+ , S 0 , and glucose) and two concentrations of As(III) (2.0 and 4.5 mM) were added. The correlation between Fe/S/As speciation transformation and bacterial depth distribution at three different depths, i.e., 15, 55, and 105 cm from the top of the columns, was comparatively investigated. The results show that the cell growth at the top and in the middle of the columns was much more significantly inhibited by the additions of As(III) than at the bottom, where the cell growth was promoted even on days 24–44. At. ferrooxidans dominated over Aph. acidophilum in most samples collected from the three depths, but the elevated proportions of Aph. acidophilum were observed in the top and bottom column samples when 4.5 mM As(III) was added. Fe 2+ bio-oxidation and Fe 3+ reduction coupled to As(III) oxidation occurred for all three column depths. At the column top surfaces, jarosites were formed, and the addition of As(III) could lead to the formation of the amorphous FeAsO 4 ⋅2H 2 O. Furthermore, the higher As(III) concentration could inhibit Fe 2+ bio-oxidation and the formation of FeAsO 4 ⋅2H 2 O and jarosites. S oxidation coupled to Fe 3+ reduction occurred at the bottom of the columns, with the formations of FeAsO 4 ⋅2H 2 O precipitate and S intermediates. The formed FeAsO 4 ⋅2H 2 O and jarosites at the top and bottom of the columns could adsorb to and coprecipitate with As(III) and As(V), resulting in the transfer of As from solution to solid phases, thus further affecting As speciation transformation. The distribution difference of Fe/S energy substrates could apparently affect Fe/S/As speciation transformation and bacterial depth distribution between the top and bottom of the water columns. These findings are valuable for elucidating As fate and toxicity mediated by microbially driven Fe/S redox in AMD environments.",
"conclusion": "Conclusion In the present work, the correlation between Fe/S/As speciation transformation and bacterial depth distribution was studied by using the acidic water columns. Such a correlation, as concluded in Figure 8 , was apparently affected by both the additions of As(III) and the depths of the water columns and was affected differently by both the concentration and species of As. Though chemical reduction of Fe 3+ coupled to As oxidation and Fe 2+ bio-oxidation occurred at all three depths, the Fe/S/As speciation transformation and bacterial depth distribution showed obvious differences between the top and bottom of the water columns. It was apparently dependent on the distribution of energy substrates, Fe 2+ , reducing sulfur compounds and S 0 .",
"introduction": "Introduction Arsenic (As) pollution in acidic environments has been a global concern due to its harmfulness to the ecosystem and human health. The toxicity of As is determined by its speciation and occurrence forms that usually change with the variation of physicochemical properties of the surrounding environments ( Cheng et al., 2009 ; Herath et al., 2016 ). Compared with arsenate [As(V)], arsenite [As(III)] is more difficult to remove from water and is generally considered to be more toxic ( Kochhar et al., 1996 ). Microbially mediated dissolution of As-containing sulfide minerals, such as arsenopyrite (AsFeS), löllingite (FeAs 2 ), and skutterudite [(Co, Ni)As 3– x ] ( Corkhill and Vaughan, 2009 ; Drahota and Filippi, 2009 ; Johnson et al., 2020 ), is the main source for As contamination, in which microbially mediated iron/sulfur (Fe/S) redox reactions drive As speciation transformation to As(III) vs. As(V) as well as As release and thus As toxicity ( Smedley and Kinniburgh, 2002 ; Burton et al., 2013 ; Wang et al., 2020 ). On the one hand, during the oxidative dissolution of As-containing sulfide minerals driven by microbial Fe/S oxidation in acidic environments such as acid mine drainage (AMD) sites, a series of iron/sulfur-containing intermediates and secondary products are formed, e.g., orpiment (As 2 S 3 ), jarosite (M[Fe 3 (OH) 6 (SO 4 ) 2 ], M = K + , Na + , NH 4 + , or H 3 O + ), scorodite (FeAsO 4 ⋅2H 2 O), and schwertmannite [Fe 8 O 8 (OH) 8–2x (SO 4 ) x (where 1 ≤ x ≤ 1.75)], leading to changes in As speciation and occurrence forms due to As adsorption to or coprecipitation with the Fe/S secondary products ( Zhang et al., 2019 ; Tabelin et al., 2020 ; Battistel et al., 2021 ). On the other hand, several acidophilic bacteria and archaea in acidic environments as well as neutrophilic Fe(III)- and sulfate-reducing bacteria (IRB and SRB, e.g., Shewanella and Desulfovibrio , respectively) in pH-neutral and mildly alkaline environments can lead to the reductive dissolution of Fe(III) minerals with the formation of Fe/S secondary products, such as mackinawite (FeS) and siderite (FeCO 3 ), with a release of As species and the reduction of As(V) to As(III) ( Drahota et al., 2013 ; Wang et al., 2016 ; Hedrich and Schippers, 2021 ). It is well known that in acidic AMD environments, the gradients of pH, redox potential (ORP), dissolved oxygen (DO), and ion concentrations can significantly affect the microbial community and Fe/S redox activities ( Chen et al., 2016 ; Han et al., 2018 ; Liu et al., 2019 ). For instance, in mill tailings, which can present a heterogeneous environment with extreme physicochemical properties, pH is primarily responsible for shaping the whole microbial community structure ( Korehi et al., 2014 ; Liu et al., 2014 ). Besides, the secondary products formed due to microbial Fe/S redox cycling in AMD environments vary considerably at different pH values; e.g., schwertmannite is formed at pH 3–4.5, but jarosite is more easily formed at pH < 3, and goethite (α-FeOOH) is generated at pH > 5 ( Acero et al., 2006 ; Liao et al., 2009 ). Many studies have found that Fe-containing secondary minerals are suitable in removing As pollution due to their great adsorption efficiency to As. For example, schwertmannites can remove significant fractions of As(III) (>97%) from contaminated water, and their As(III) sorption behavior was strongly affected by the different synthesis pathways ( Paikaray et al., 2011 ). However, how microbial communities and the Fe/S redox reactions they catalyze affect Fe/S speciation has not yet been fully described. Also, the links between Fe/S speciation transformations and As fate have rarely been studied. The water column system could be an effective way to control water levels ( Tokunaga et al., 2018 ). In the present study, we used acidic water columns to evaluate the links above. To the water columns, we inoculated the mixture of the chemoautotrophic Acidithiobacillus ferrooxidans (At. ferrooxidans) and the heterotrophic Acidiphilium acidophilum (Aph. acidophilum) and added three energy substrates, Fe 2+ , S 0 , and glucose, as well as two concentrations of As(III) (2.0 and 4.5 mM). Through monitoring Fe/S/As speciation transformations and the dynamic change in the microbial community along the depths of the water columns, we comparatively investigated their correlations at three different depths of the water columns. These two strains are widely occurring together in AMD environments ( Hedrich and Schippers, 2021 ) and have been commonly used to simulate the bioleaching of metal sulfide minerals under acidic conditions ( Liu et al., 2011 ; Liu, 2013 ). This work could be valuable in elucidating As transformations mediated by microbially driven Fe/S redox in AMD environments.",
"discussion": "Discussion Bacterial Growth and Depth Distribution of Microbial Communities Responding to As(III) Toxicity It has been widely reported that As(III) could significantly inhibit bacterial growth and the Fe/S oxidation activities ( Hallberg et al., 1996 ; Páez-Espino et al., 2009 ). In the present study, we further found that in the simulated water columns, both the bacterial growth and composition were affected not only by the additions of As(III) but also by the depths of the water columns ( Figure 1 ). For the columns with additions of As(III), the cell growth was significantly inhibited at the top and middle ( Figure 1A ), likely due to the toxicity of dissolved As(III) to bacterial cells. On the other hand, irrespective of the As(III) addition (or a lack of it), the cell growth at the bottom of the water columns quickly entered the exponential phase, which is much different from the top and middle, where the cell growth exhibited typical growth curves with about 40 days of lag phase. The promotion of bacterial growth at the bottom of the columns with the addition of As(III) was probably due to the formation of Fe/As/S-containing precipitates, FeAsO 4 ⋅2H 2 O, and jarosites ( Figures 4–7 ), which can adsorb to or coprecipitate with As species. These processes could significantly mitigate the harmfulness of As(III) and reduce the Fe 3+ concentration and thus mitigate the inhibited feedback of Fe(III) to the ferrous-oxidizing enzyme activity ( Hare et al., 2020 ; Zhang et al., 2020 ). Moreover, the cells detected at the bottom showed somewhat As adsorption according to the results of FT-IR ( Supplementary Figure 8 ), which could also reduce As harmfulness and thus favor cell growth ( Zhang et al., 2021 ). The bacterial compositions of At. ferrooxidans and Aph. acidophilum during cultivation were also significantly affected by the additions of As(III) ( Figures 1D–F ) and the depths of the water columns. When no As(III) was added, At. ferrooxidans was dominant during the most experimental time at the top, middle, and bottom of the water columns. However, for both cases with As(III) added, At. ferrooxidans dominated at the top and middle of the columns, while in the case with a higher concentration of As(III), At. ferrooxidans decreased and became less abundant than Aph. acidophilum at the bottom. This may be due to the differences in the energy substrate depth distribution along with the water columns and the metabolism between the two bacteria. Considering the difference in the solubilities of Fe and S substrates and the phenomenon that S 0 basically deposited at the bottom of the columns during the most experimental time, it was concluded that there was no difference in the glucose distribution but that there is an obvious difference in Fe and S energy substrate distributions along with the depth of the water columns. Therefore, at the top and middle of the columns, there were major Fe 2+ and slight soluble reducing sulfur compounds, S n 2– , as shown by the results of S K-edge XANES and S 2p XPS ( Figures 5 , 6 ). While at the bottom, there were large amounts of S 0 distributed, besides Fe 2+ . Based on the above experimental phenomena and inferences, it was speculated that At. ferrooxidans grew majorly on soluble Fe 2+ at the top and middle, and at the bottom, it grew majorly on S 0 deposited, while Aph. acidophilum grew on glucose at the top and middle and used S 0 and glucose at the bottom of the water columns. Furthermore, both bacteria could use O 2 and Fe 3+ as electron acceptors; however, DO ( Supplementary Figure 6 ) and Fe 2+ ( Figures 2G–I ) concentration measurements indicated that there were aerobic conditions and no lack of Fe 2+ for all three depths of the columns for most of the experimental time; thus, it was very likely that the microbial Fe 2+ oxidation (as opposed to Fe 3+ reduction) played a major role at the top and middle of the columns. Moreover, the higher As(III) concentration may obviously inhibit the growth of At. ferrooxidans . It implies that there were differences in As resistance between the two strains, which may be the reason for the major change in the bacterial distribution for the case with a higher As(III) concentration ( Dopson and Johnson, 2012 ; Mirete et al., 2017 ). Iron and Sulfur Speciation of the Simulated Water Column The variations of physicochemical parameters and the concentrations of dissolved Fe and S at different depths of the water columns were different. The pH values ( Figure 2A ) were gradually decreasing at three depths of all the water columns, which can be mainly caused by sulfur bio-oxidation (Eq. 1) and partly by the formation of jarosites (Eq. 2) ( Xia et al., 2010 ; He et al., 2012 ; Zhang et al., 2021 ). The decrease in pH values slowed down with the additions of As(III), and the slowdown decreased with more additions of As(III), indicating the inhibition of As(III) to sulfur bio-oxidation and/or the formation of jarosites. Moreover, the decrease in the pH values to some extent depended on different depths of the water columns and cultivation time, which is probably related to the evolution of the bacterial composition of At. ferrooxidans and Aph. acidophilum during cultivation. Fe 2+ bio-oxidation was more inhibited with the increase of [As(III)] and solution depths ( Figure 2B ). The [Fe 3+ ] remained very low at the bottom sites for all the water columns, which was likely related to the low Fe 2+ bio-oxidation activity and the Fe 3+ reduction by Aph. acidophilum under microaerobic conditions ( Supplementary Figure 6 ). The changes in the potential of the solution also indicated that the obvious inhibition of Fe 2+ oxidation occurred at the middle and bottom of the water columns. Generally speaking, changes in [Fe 3+ ], [Fe 2+ ], [SO 4 2– ], and potential in solutions at the three water column depths seemed to be very complex ( Figures 2D–I and Supplementary Figures 2 , 3 ), which may be due to the complex interactions (adsorption, complexation, and coprecipitation) between Fe, S, and As species and the formation of jarosites (Eq. 2), scorodite (Eq. 3), and other uncertain complexes. \n (1) \n S + 1.5 O 2 + H 2 O → 4 H + + S O 4 2 - \n \n (2) \n M + + 3 F e 3 + + 2 S O 4 2 - + 6 H 2 O → MFe 3 ( SO 4 ) 2 ( OH ) 6 + 6 H + \n where M is a monovalent cation, such as NH 4 + , K + , Na + , and H 3 O + . \n (3) \n Fe 3 + + AsO 4 3 - → FeAs O 4 \n At the top of the water columns, the Fe 3+ concentrations were relatively high ( Figure 2D ) in the middle and late stages of cultivation due to the apparent Fe 2+ bio-oxidation. The Fe 3+ could react with other monovalent cations and SO 4 2– in the solution to form potassium and ammonium jarosites suspended on the surface of the solution ( Figure 3A ), which was in accordance with the appearance of the yellow surface suspended matter in the acidic environment of AMD ( Supplementary Figure 4 ). The Fe 2p signal ( Figures 3G–I ) further showed that the additions of As(III) could inhibit the Fe 2+ oxidation of the suspended solid matter at the top of the water columns. The higher the As(III) concentration, the more the inhibition to Fe 2+ oxidation. Therefore, the added As(III) could delay the formation of jarosites on the solution surface ( Supplementary Figure 9 ). The formation of amorphous FeAsO 4 ⋅2H 2 O occurred for the case of 2.0 mM As(III) ( Figure 3B ), while its formation was inhibited by the addition of 4.5 mM As(III) ( Figure 3C ). This indicated that the high concentration of As(III) could inhibit As oxidation as well as the formation of FeAsO 4 ⋅2H 2 O. The Fe and S speciation changes at the bottom of the water columns were more complex than those at the top. When no As(III) was added, S oxidation ( Figure 5B ) and Fe reduction ( Figure 4A ) occurred in the early stage of the cultivation, and Fe oxidation was dominant after day 74, while the S speciation on the surface of the residues did not change significantly ( Figure 6A ). However, the addition of 2.0 mM As(III) inhibited Fe reduction ( Figure 4B ) but significantly promoted sulfur oxidation ( Figure 5C ) and produced more diverse sulfide species ( Figure 6B ) in the late experimental stage. When 4.5 mM As(III) was added, the composition change of S speciation was similar to the case with 2.0 mM As(III) added, but the promotion effect of sulfur oxidation was lower, while Fe reduction was promoted at the initial experimental stage ( Figure 4C ) and the proportion of Fe(II) began to decrease after day 100. These results indicated that Fe 3+ reduction mediated by glucose or S oxidation occurred as a more likely process at the bottom of the water columns (Eqs 4, 5), with the S 0 opening the ring and activating in contact with the outer membrane protein thiol group (P-SH, where P presents protein) on the cell membrane to form the polysulfide (Eqs 6, 7) ( Rohwerder and Sand, 2003 ; Xia et al., 2013 ; Liu et al., 2015 ). It could be derived that At. ferrooxidans accounted mainly for S/Fe bio-oxidation at the bottom of the water columns when no As(III) was added, while for the cases with As(III) added, Aph. acidophilum increased obviously, and for the case with 4.5 mM As(III) added, it even became dominant in a certain period of cultivation with S 2 2– , instead of S n 2– , as the main component in the bottom residues. It indicated that the bacterial distribution had an important effect on the Fe/S redox reactions. \n (4) \n C 6 H 12 O 6 + 24 F e 3 + + 6 H 2 O → 6 C O 2 + 24 F e 2 + + 24 H + \n \n (5) \n 12 Fe 3 + + 2 S 0 + 8 H 2 O ⟶ microbe 12 Fe 2 + + 16 H + 2 S O 4 2 - \n \n (6) \n S 8 + P - SH → P - SS 8 H → → P - S m H ( m ≥ 9 ) \n \n (7) \n P - S m H + P - SH → P - S - S m-n - P + S n 2 - + 2 H + ( n ≥ 1 ) \n As Speciation Changes and Their Relationship With Fe/S Redox The complex As speciation transformations occurred in the solution, in the suspended matter at the top, and in the residues at the bottom of the water columns. The [As(III)] in the solution changed in the range of 0.2–0.35 and 0.1–0.15 g/L for the Col_2.0 and Col_4.5, respectively, while [As(V)] changed in the range of 0–0.05 g/L in all water columns ( Figures 2J–L ). It implies that the oxidation of As(III) to As(V) occurred in all depths of the water columns, but the oxidation rate of As(III) was very low, where only < 10% of As(III) was oxidized. Moreover, some of the As species in the solution were transferred to the solid phases by adsorption to and coprecipitation with the Fe(III)-(hydro)oxides and jarosites ( Wang and Zhou, 2012 ; Ahoranta et al., 2016 ). Such processes could be affected by the solution pH and ORP ( Supplementary Figure 7 ), as well as the As forms occurring ( Paikaray et al., 2011 ; Ahoranta et al., 2016 ). Furthermore, As adsorption by bacterial cells ( Supplementary Figure 8 ) could also contribute to the change in the composition of As species in the solution ( Gao et al., 2018 ; Zhang et al., 2021 ). Of note, the As species composition of the suspended matter at the top and the residues at the bottom were quite different for the cases with the additions of 2.0 and 4.5 mM As(III) ( Figures 3 , 7 ). For the suspended matter at the top, the As(III) oxidation and the formation of FeAsO 4 ⋅2H 2 O seemed to be inhibited by the higher concentration of As(III), which was evidenced by the decrease of As(V)-O and FeAsO 4 ⋅2H 2 O contents when 4.5 mM As(III) was added ( Figure 3 ). For the bottom residues of the two cases with As(III) added, the main components of As species changed gradually from As(III) to As(V) species, and the proportion of As(V) was about 74.2–77.5% on day 140 for both cases ( Figure 7 and Supplementary Table 7 ). It implies that As(V) was more easily precipitated onto the solid phases. Moreover, the formation of FeAsO 4 ⋅2H 2 O species at the bottom for the case with 4.5 mM As(III) added was later than that with 2.0 mM As(III) added, and more amounts of FeAsO 4 ⋅2H 2 O species were produced with less addition of As(III). It is noteworthy that there were amounts of As(I)-O species and a little As(0) and As(-I)-S species detected on the solid residues at the bottom, indicating that the reduction of As occurred at the bottom. According to previous studies, the above different As speciation transformations in the water columns could be mostly associated with Fe/S redox reactions ( Karimian et al., 2018 ). In aerobic conditions, As(III) and As(V) were easily released with the bio-oxidation of arsenopyrite by Fe/S-oxidizing bacteria ( Smedley and Kinniburgh, 2002 ; Wang et al., 2020 ). Under anoxic conditions, the bacterial reductions of As(V) and Fe(III) oxides could influence the redox cycling of As, and the adsorbed As(III) on the ferric hydroxide substrate could enhance microbial Fe reduction ( Campbell et al., 2006 ; Smeaton et al., 2012 ). In the present study, the DO changed at different depths during cultivation ( Supplementary Figure 6 ), resulting in different redox environments associated with the depth distribution of At. ferrooxidans and Aph. acidophilum as discussed above. According to the changes in As species with the pH and ORP values ( Supplementary Figure 7 ), As mainly existed as H 3 AsO 3 in the solution at pH 0–4 and ORP 0–360 mV, and it was difficult to oxidize to As(V). This was in accordance with the situations for As(III) additions, the oxidation rate of As(III) was low, and As(III) was dominant in the solutions at different depths ( Figures 2J–L ). Of note, in the present study, it was observed that the higher As(III) concentration could inhibit the Fe 2+ bio-oxidation and the formations of amorphous FeAsO 4 ⋅2H 2 O and jarosites at the top of the water columns ( Figure 3 ). However, the As reduction was enhanced by the higher As(III) concentration at the bottom ( Figure 7 and Supplementary Table 7 ). According to the previous studies, the oxidation of As(III) to As(V) and the formation of FeAsO 4 ⋅2H 2 O could be chemically coupled to the Fe reduction process, and the reduction of As(V) was probably mediated by microbial Fe oxidation ( Campbell et al., 2006 ). It can be derived that the additions of different concentrations of As(III) could affect the Fe/S (bio-)redox behaviors and the Fe/S/As speciation transformation and thus lead to the change in the depth profile of the bacterial community structure. Based on the above discussion, the Fe/S/As speciation transformation and its relationship to the depth distribution of At. ferrooxidans and Aph. acidophilum in the simulated acidic water columns can be outlined ( Figure 8 ). It shows that, on the top of the water columns, As(III) clearly inhibits bacterial growth and Fe 2+ oxidation activity with more time required for the formation of jarosites. The higher the As(III) concentration, the more the inhibition to Fe 2+ oxidation activity. The Fe 3+ reduction coupled to As(III) oxidation occurs with the formation of As(V) and Fe 2+ ; then the As(V) reacts with Fe 3+ to form precipitate FeAsO 4 ⋅2H 2 O, and the Fe 2+ is recycled for further bio-oxidation. These processes can somewhat mitigate the inhibitory effects of As(III). In the middle of the water columns, Fe 2+ bio-oxidation and Fe 3+ reduction coupled to As(III) oxidation also occurred, but the Fe 2+ bio-oxidation activity is more inhibited than that at the top, with no obvious precipitates formed. At the bottom of the water columns, besides the aforementioned similar Fe/As reactions, there also occurs sulfur oxidation coupled to Fe 3+ reduction with the formation of Fe intermediates. These sulfur intermediates can diffuse to the middle and top of the water columns. In addition, As(III) can be reduced to As(I), As(0), and As(-I), which requires further study. These findings are of value for elucidating As fate and toxicity mediated by Fe/S speciation transformation driven by the microbial Fe/S redox in AMD environments. FIGURE 8 The diagram of the correlation between Fe/S/As speciation transformation and bacterial ( At. ferrooxidans and Aph. acidophilum ) depth distribution under different depths of the acidic water column. Dashed arrows show the migration of the dissolved ions, and solid arrows show the Fe/S/As redox reactions."
} | 6,081 |
22211106 | PMC3248649 | pmc | 7,118 | {
"abstract": "Mass production of glucosamine (GlcN) using microbial cells is a worthy approach to increase added values and keep safety problems in GlcN production process. Prior to set up a microbial cellular platform, this study was to assess acetate metabolism in Citrobacter sp. BL-4 (BL-4) which has produced a polyglucosamine PGB-2. The LC-MS analysis was conducted after protein separation on the 1D-PAGE to accomplish the purpose of this study. 280 proteins were totally identified and 188 proteins were separated as acetate-related proteins in BL-4. Acetate was converted to acetyl-CoA by acetyl-CoA synthetase up-regulated in the acetate medium. The glyoxylate bypass in the acetate medium was up-regulated with over-expression of isocitrate lyases and 2D-PAGE confirmed this differential expression. Using 1 H-NMR analysis, the product of isocitrate lyases, succinate, increased about 15 times in the acetate medium. During acetate metabolism proteins involved in the lipid metabolism and hexosamine biosynthesis were over-expressed in the acetate medium, while proteins involved in TCA cycle, pentose phosphate cycle and purine metabolism were down-regulated. Taken together, the results from the proteomic analysis can be applied to improve GlcN production and to develop metabolic engineering in BL-4.",
"conclusion": "Conclusions Proteomic analysis of BL-4 using LC-MS/MS after protein separation on 1D-PAGE found a dramatic change in the acetate medium when compared to the glucose medium. According to UNIPROT classification and KEGG regulatory pathways, most up-regulated proteins in relation to acetate feeding were involved in acetate assimilation, the glyoxylate cycle, glycerol metabolism, energy production, and lipid metabolism. On the other hand, BL-4 revealed down-regulation of proteins involved in the TCA cycle, pentose phosphate cycle and purine metabolism. These results were reconfirmed by LC-MS/MS analysis after 2D-PAGE and 1 H-NMR. Higher amount of succinate, glutamate and lactate were found in the acetate medium than in the glucose medium. As BL-4 produced polyglucosamines (PGB-2) in the acetate medium, most proteomic changes were related to the production of PGB-2. Further studies in BL-4 will be necessary to modify metabolic pathways for becoming a cell factory to produce massive PGB-2. Whole proteomic findings are summarized in Fig. 9 .",
"introduction": "Introduction Glucosamine (GlcN) is now widely used to treat osteoarthritis in humans 1 , 2 . However, nearly all GlcNs have been manufactured from shellfish wastes 3 . This chemically treated process on GlcN production may cause shellfish protein contamination and the production amount can be limited by raw materials available. Therefore, newly developed technology based on microbial fermentation for producing GlcN has been introduced 3 - 5 and Citrobacter sp. is one of good candidates for achieving mass production of GlcN 6 . Recently, a new extracellular polyglucosamine biopolymer PGB-2 from Citrobacter sp. BL-4 (BL-4) has been introduced and the structural similarity of PGB-2 with chitosan from crab shells has been determined 6 . For mass production of PGB-2 it is necessary to obtain whole biochemical information in BL-4 when acetate is available as a sole carbon source because the bacteria produced PGB-2 under acetate feeding only. Proteomics technology can easily collect high-throughput information on intercellular changes due to up- and down-regulation of genes by a certain growth environment 7 . As the previous report demonstrated, acetate has been known to be toxic to bacterial cells 8 and inhibit growth because its free acid forms quickly penetrate the cell membrane and acidify the cytoplasm with dissociation, conferring the gradient of protons through the membrane cannot be maintained and the energy generation is decoupled 8 . Interestingly, Citrobacter sp. shows its preference for acetate rather than glucose as a carbon source 6 . Therefore, it is worthwhile to assess how BL-4 survives under acetate feeding and produces PGB-2 using proteomic technology. Herein, we used differential expression proteomics which employed one dimensional polyacrylamide gel electrophoresis (1D-PAGE) coupled with liquid chromatography-quadrupole mass spectrometer (LC-MS/MS) system for producing high-throughput data. Confirmation of up- and down-regulation of proteins was conducted with 2D-PAGE. This methodology has been developed and applied to various biological samples 9 - 11 . Biochemical products mediated by differentially expressed proteins were determined by 1 H-NMR to prove the proteomic results.",
"discussion": "Discussion Acetate metabolism in microorganisms has been well studied 12 - 15 and reviewed 16 - 18 . Proteome differences according to acetate supply in BL-4 were studied in respect of biotechnological aspects for accomplishing industrial purposes as increasing production rate of the target biopolymer, PGB-2. It is the first report for the bacteria to describe proteomic changes due to acetate metabolism. Up-regulation of acetyl-CoA synthetase activity for initiation of acetate activation Acetate assimilation to acetyl-CoA must be occurred when acetate is available as a sole carbon source. This reaction is catalyzed by acetyl-CoA synthetase and its activity is not determined in bacterial cells grown on glucose. In contrast to acetyl-CoA synthetase acetate kinase (AK; ATP acetate phosphotransferase) and phosphotransacetylase (PT; acetyl-CoA: orthophosphate acetyl transferase) are employed for the formation of acetyl-CoA when glucose is used as the carbon source. Levels and expression of these two enzymes vary little with the different carbon sources in E. coli and Salmonella typhimurium 19 , 20 . Therefore, three enzymes are involved in the process of acetyl-CoA production. In our study, acetyl-CoA synthetase was newly expressed in the acetate medium (Table 1 ). According to the previous report the gene FacA is responsible for encoding acetyl-CoA synthetase and the gene has been known to be activated by acetate and deactivated by glucose 21 . Our finding was quite similar to the findings of the previous report 21 . Up-regulation of Glyoxylate pathway Acetate has been known to regulate 37 genes using a newly developed metabolic array in the bacteria 15 . In the differential transcription profile in Corynebacterium glutaricum, two genes aceA (102-fold) and aceB (85-fold) were up-regulated in acetate when compared to glucose 15 . These genes produce isocitrate lyase (ICL) and malate synthase (MS) in the bacteria. ICL and MS proteins are functionally participated in the glyoxylate cycle. The glyoxylate cycle operates to produce one mole of malate from two moles of acetyl-CoA via two unique enzymes, ICL and MS, in bacteria. In this study, ICL dramatically increased in BL-4 in the acetate medium (Table 1 and Fig. 4 ). The enzyme catalyzes reversible cleavage of isocitrate to glyoxylate and succinate as the products. By the analysis of 1 H NMR the concentration of succinic acid increased about 15 times higher in the acetate medium than in the glucose medium (Fig. 8 ). Therefore, the glyoxylate bypass was up-regulated in BL-4 when acetate was used as a sole carbon source. TCA cycle in acetate metabolism in Citrobacter sp. BL-4 In a normal bacterial cell pyruvate is ready to be transferred into citric acid cycle (tricarboxylic acid cycle, TCA cycle) by pyruvate dehydrogenase complex. This enzyme produces carbon dioxide, NADH and acetyl-CoA from pyruvate, CoA and NAD + . As acetyl-CoA synthetase activity was up-regulated in the acetate medium, the role of pyruvate dehydrogenase for producing acetyl-CoA may be minimized in the acetate medium. In the result of the study, pyruvate dehydrogenase was dramatically down-regulated in the acetate medium (Table 1 ). Additionally, excessive acetyl-CoA or over-expression of acetyl-CoA synthetase may be related to the down-regulation of pyruvate formate lyase (PFL; formate acetyltransferase) in the acetate medium (Table 1 and Fig. 5 ). PFL is essential to be activated for forming acetyl-CoA and formate from pyruvate and CoA. For the PFL activation, bacterial cells require PFL activase and radical S -adenosylmethionine (SAM). In this study, radical SAM was down-regulated in the acetate medium, referring the down-regulation of PFL (Table 1 and Fig. 5 ). Using the glyoxylate bypass in the acetate medium, excessive amount of succinate should be recycled in the TCA cycle to produce oxaloacetate to form citrate. Thus, succinate dehydrogenase was up-regulated in the acetate medium (Fig. 6 ). Therefore, up-regulation of glyoxylate bypass may change TCA cycle during acetate metabolism. Up-regulation of lipid metabolism Fatty acid biosynthesis and degradation requires carrier proteins and enzymes involved in the addition and subtraction reactions of acetate unit to a hydrocarbon chain. Acetyl-CoA carboxylases (ACCs) are the key enzymes and mediate a carboxylation reaction to produce malonyl-CoA from acetyl-CoA. Other key enzyme is fatty acid synthase consisting of six enzymatic activities and it is responsible for the reactions of adding acetate unit to a growing fatty acid chain. In this study, ACCs were up-regulated in the acetate medium in contrast to the glucose medium (Table 1 ). This finding is quite similar to the result previous reported as ACC was up-regulated in the acetate medium to generate polyhydroxyalcanoates (PHAs) in Ralstonia eutropha 11 . In R. eutropha , various fatty acid biosynthesizing proteins involved in the PHA production were over-expressed in relation to the use of acetate as a sole carbon source and they were acyl-CoA dehydrogenase, 3-oxoacyl-acyl-carrier protein synthase (FabH), enoyl-[acyl-carrier-protein]reductase (FabI) and fatty acid-CoA ligase 11 . In our study, FabI and β-ketoacyl carrier protein reductase (FabG) were found in both medium and they might be not expressed differently (data not shown). However, FabH was up-regulated in the acetate medium (Table 1 ) and it seemed to be involved in the PHA production in BL-4. Further studies may be needed to find possible PHA production in BL-4 when various organic acids are used as carbon sources. Induction of hexosamine biosynthesis GlcNs are produced by three different enzymes such as L-glutamine: D-fructose-6-phosphate amidotransferase (glucosamine synthase, GlmS), phosphoglucosamine mutase and acetyltransferase. Among them, GlmS is a key enzyme which catalyzes fructose-6-phosphate to glucosamine-6-phosphate in host cells. By the 1D-PAGE and LC-MS/MS analysis, GlmS was up-regulated in acetate-fed BL-4 (Table 1 ). As N -acetylglucosamine has been determined as a unit of peptidoglycan with N -acetylmuramic acid, its primary function is to preserve cell integrity by withstanding the internal osmotic pressure 22 . Therefore, it is likely that BL-4 expresses GlmS involved in the hexosamine biosynthesis to build up peptidoglycan to control osmotic pressure occurred by acetate supply in the growth medium. On the other hands, excessive glucosamine-6-phosphate (GlcN-6-P) produced by metabolic engineered E. coli strongly inhibits GlmS activity in the host cells, whereas GlcN weakly inhibits GlmS activity 4 . GlcN-6-P may be a key compound to control glucosamine production in bacterial cells. Another biosynthetic intermediate, N -acetyl-D-glucosamine-6-phosphate (AcGlcN-6-P), can be catalyzed to GlcN-6-P by AcGlcN-6-P deacetylase. Our study showed that AcGlcN-6-P deacetylase was up-regulated 5 times in the acetate medium than in the glucose medium (Fig. 6 ). From the results, GlcN production is regulated by GlcN-6-P concentration in the bacterial cells. Therefore, GlcN production is definitely improved if those factors are controlled adequately. For instance, an AcGlcN-6-P deacetylase knock-out strain of BL-4 can be developed and this strain may produce massive GlcN under the acetate supply. Presumable biosynthetic mechanisms induced by acetate The expression of glycerol dehydrogenase and glycerol kinase (Table 1 ) was over-expressed in BL-4 in the acetate medium. Glycerol dehydrogenase catalyses the NAD + -dependent oxidation of 2,3-butanediol to acetoin as well as the corresponding reverse reactions. Glycerol kinase is a key enzyme in the regulation of glycerol uptake and metabolism and the enzyme plays an important role in gluconeogenesis to form glucose. Therefore, BL-4 may produce 2,3-butanediol from acetoin when acetate is available in the growth medium. The aldehyde dehydrogenase B (lactaldehyde dehydrogenase) was over-expressed at least five-fold when acetate was available in BL-4 and it catalyzes ( S )-lactaldehyde with NAD + and a molecule of water to form ( S )-lactate and NADH. Lactate concentration in the bacterial cells increased at least 15 times in the acetate medium when compared to the glucose medium (Fig. 8 ). Therefore, acetate plays an important role in the induction of 2,3-butanediol and lactate production. Glutathione S -transferase (GST) is involved in the detoxification reaction when cells are exposed to various toxic compounds. In the acetate medium, BL-4 expressed more GSTs than in the glucose medium (Table 1 ). Glutathione is composed to three amino acids such as cysteine, glutamate and glycine. In this study, cysteine synthase was up-regulated (Table 1 and Fig. 4 ) and the concentration of glutamate in the acetate medium was about 3.1-fold higher than in the glucose medium (Fig. 8 ). Another key enzyme, gamma-glutamyltransferase, was also up-regulated in relation to glutathione biosynthesis (Fig. 6 ). Therefore, GST detoxification system may be needed to suppress oxidative stress in the acetate medium in BL-4. Some of up- and down-regulated proteins analyzed on the 1D-PAGE was reconfirmed using 2D-PAGE were universal stress protein A, Helix-turn-helix, FBKP-type peptidyl-prolyl cis-trans isomerase, and argininosuccinate synthase. Further studies needs to find their role in the acetate medium in BL-4."
} | 3,516 |
20224547 | PMC3144576 | pmc | 7,120 | {
"abstract": "The need for renewable, carbon neutral, and sustainable raw materials for industry and society has become one of the most pressing issues for the 21st century. This has rekindled interest in the use of plant products as industrial raw materials for the production of liquid fuels for transportation 1 and other products such as biocomposite materials 7 . Plant biomass remains one of the greatest untapped reserves on the planet 4 . It is mostly comprised of cell walls that are composed of energy rich polymers including cellulose, various hemicelluloses (matrix polysaccharides, and the polyphenol lignin 6 and thus sometimes termed lignocellulosics. However, plant cell walls have evolved to be recalcitrant to degradation as walls provide tensile strength to cells and the entire plants, ward off pathogens, and allow water to be transported throughout the plant; in the case of trees up to more the 100 m above ground level. Due to the various functions of walls, there is an immense structural diversity within the walls of different plant species and cell types within a single plant 4 . Hence, depending of what crop species, crop variety, or plant tissue is used for a biorefinery, the processing steps for depolymerization by chemical/enzymatic processes and subsequent fermentation of the various sugars to liquid biofuels need to be adjusted and optimized. This fact underpins the need for a thorough characterization of plant biomass feedstocks. Here we describe a comprehensive analytical methodology that enables the determination of the composition of lignocellulosics and is amenable to a medium to high-throughput analysis. In this first part we focus on the analysis of the polyphenol lignin (Figure 1). The method starts of with preparing destarched cell wall material. The resulting lignocellulosics are then split up to determine its lignin content by acetylbromide solubilization 3 , and its lignin composition in terms of its syringyl, guaiacyl- and p-hydroxyphenyl units 5 . The protocol for analyzing the carbohydrates in lignocellulosic biomass including cellulose content and matrix polysaccharide composition is discussed in Part II 2 .",
"discussion": "Discussion The described methods enable a rapid quantitative assessment of the lignin content and composition of lignocellulosic plant biomass. Using the iWall robot approximately 350 samples can be ground and dispensed per day. The throughput of the various analytical methods per person varies. Using the protocols described here, 30 samples can be processed for lignin content, and 15 for lignin composition per day. Due to the quantitative nature of the data optimal feedstock crops, variety or genotypes can be assessed in terms of their suitability for biofuel production."
} | 685 |
33260379 | PMC7760971 | pmc | 7,123 | {
"abstract": "To fabricate an industrial and highly efficient super-hydrophobic brass surface, annealed H59 brass samples have here been textured by using a 1064 nm wavelength nanosecond fiber laser. The effects of different laser parameters (such as laser fluence, scanning speed, and repetition frequency), on the translation to super-hydrophobic surfaces, have been of special interest to study. As a result of these studies, hydrophobic properties, with larger water contact angles (WCA), were observed to appear faster than for samples that had not been heat-treated (after an evolution time of 4 days). This wettability transition, as well as the evolution of surface texture and nanograins, were caused by thermal annealing treatments, in combination with laser texturing. At first, the H59 brass samples were annealed in a Muffle furnace at temperatures of 350 °C, 600 °C, and 800 °C. As a result of these treatments, there were rapid formations of coarse surface morphologies, containing particles of both micro/nano-level dimensions, as well as enlarged distances between the laser-induced grooves. A large number of nanograins were formed on the brass metal surfaces, onto which an increased number of exceedingly small nanoparticles were attached. This combination of fine nanoparticles, with a scattered distribution of nanograins, created a hierarchic Lotus leaf-like morphology containing both micro-and nanostructured material (i.e., micro/nanostructured material). Furthermore, the distances between the nano-clusters and the size of nano-grains were observed, analyzed, and strongly coupled to the wettability transition time. Hence, the formation and evolution of functional groups on the brass surfaces were influenced by the micro/nanostructure formations on the surfaces. As a direct consequence, the surface energies became reduced, which affected the speed of the wettability transition—which became enhanced. The micro/nanostructures on the H59 brass surfaces were analyzed by using Field Emission Scanning Electron Microscopy (FESEM). The chemical compositions of these surfaces were characterized by using an Energy Dispersive Analysis System (EDS). In addition to the wettability, the surface energy was thereby analyzed with respect to the different surface micro/nanostructures as well as to the roughness characteristics. This study has provided a facile method (with an experimental proof thereof) by which it is possible to construct textured H59 brass surfaces with tunable wetting behaviors. It is also expected that these results will effectively extend the industrial applications of brass material.",
"conclusion": "5. Conclusions Different annealing treatments of the brass samples have resulted in different wettability transition times, which in turn were caused by the existence of different amounts and types of nanoparticles, nanograins, and grain boundaries. In addition, the hierarchic micro/nanostructures played an important role during the reversible transition from super-hydrophilicity to super-hydrophobicity. (1) The different distances between the nanograins were observed to affect (a) the reversible wettability time, (b) the nanoparticles, that in turn were related to the progress of the Cu 2 O evolution, and (c) the surface roughness (that would influence the surface energy). After the initial nanosecond laser ablation, the high laser fluence did increase the distances between the nanograins. More CuO materials were formed, and the surfaces received super-hydrophilic properties. The CuO composition infiltrated the grooves and became more easily embedded into the ablated microstructures. After the evolution of Cu 2 O, the surfaces were transferred to super-hydrophobic properties. To reduce the laser repeat repetition, it interacts with the H59 brass surface with a higher power in each pulse, and then, the brass oxidation was extended and the distances between the nanograins were increased. Furthermore, while the laser rate was turned down, the laser was removing an increased volume out of the micro-grooves. All of these methods were found to speed up the evolution of Cu 2 O and WCA. (2) The annealing temperatures were strongly coupled with the evolution of nanoparticles. High temperatures induced much smaller nanoparticles and increased the adhesion of Cu 2 O to these nanoparticles. Moreover, the number of nanoparticles increased rapidly, which was also the situation with the WCAs. (3) The chemical properties of brass played a fundamental role in the evolution of super-hydrophobicity. For instance, the hydrophilic properties of the brass surfaces were related to the number of CuO nanoclusters. The transformation from a hydrophilic to a hydrophobic surface was then related to the existence of Cu 2 O nanoclusters in the micro/nanostructures, and to the surface energies. The more pronounced micro/nanocomposite structures were more easily adsorbed by ions and formed a type of air cushion attached to it. This resulted in liquid–solid bubbles and formed a gas–liquid–solid composite in contact with the surface. With an increase in surface roughness, the solid–liquid contact area did decrease, thereby forming a hydrophobic surface. It has here been assumed that the droplet will be in direct contact with the solid surface, with no air in between. The contact angle increased, and the surface became more hydrophobic. Hence, when the solid surface is made of a hydrophilic material, the increased surface roughness will cause a more hydrophilic surface.",
"introduction": "1. Introduction Today, an anti-corrosion related technology based on brass has increased many people’s interest in mimicking the pattern of the lotus leaf surface. The reason is that a nanosecond fiber laser can be used to create this type of pattern on an H59 brass surface, thereby forming super-hydrophobic surfaces after an incubation period of about 2 weeks. This means that an increase in water contact angle (WCA) of up to 150° can be obtained after about 14 days of evolution [ 1 , 2 , 3 ]. However, the possibility of producing a superhydrophobic H59 brass surface in less than 14 days remains a big challenge. While annealing the brass material, the temperature will continuously increase, and the laser-induced texture will become both broader and wider. This process will simultaneously reduce the surface energy, and speed up the super-hydrophobic properties of the brass surface. As an important engineering metal material, H59 brass has become widely used in various types of applications (in industrial hydraulic systems, for transportation, in ship propellers, etc.). For this brass, a quick transformation from hydrophilicity to hydrophobicity should effectively improve the anti-corrosion properties. Examples of potential applications for this type of modified H59 brass material are self-cleaning surfaces [ 4 , 5 ], oil/water separations [ 6 ], anti-icing/frosting [ 7 , 8 ], drag reduction [ 9 ], and microfluidic devices [ 10 ]. As a result of recent research efforts, appropriate brass surfaces with hierarchic micro/nanostructures (from here on called micro/nanostructures), containing nanosized grains and Cu 2 O nanoclusters, have been manufactured with the purpose of producing hydrophobic surfaces with very short evolution times. These hydrophobic surfaces have been formed by using various types of deposition techniques; the sol–gel method, chemical vapor deposition [ 11 , 12 ], chemical etching [ 13 , 14 ], nanoimprint lithography [ 15 ], and self-assembly [ 16 ]. However, these methods are limited in terms of high cost, complicated processes, and poor mechanical properties of the textured surfaces. Recently, some environmentally friendly chemical solutions (like isopropyl alcohol, sodium bicarbonate solution, etc.) have been used to shorten the wettability transition time (down to about 5 days) [ 17 ]. In recent years, different types of annealing treatments, at lower temperatures, have been used to speed up the super-hydrophobic wettability transition [ 18 ]. The surface energy, roughness, and morphology, as well as the dual-scaled particles in the surface micro/nanostructure, play important roles in this evolution procedure (with an exception for the Cu 2 O nanoclusters) [ 19 , 20 ]. As a matter of fact, the physical profile of the hierarchic micro/nanostructures, in addition to the large oxygen content, will speed up the variation of Cu 2 O material on the surface [ 21 , 22 ]. It is well known that the super-hydrophobic property of a surface is determined by a combination of physical roughness and chemical composition. In fact, super-hydrophobic surfaces do not only depend on dual micro/nanostructures, but also on the existence of low surface energy Hence, key to the evolution of Cu 2 O during the laser processing of the brass surfaces is the reversible wettability induced by different laser parameters [ 13 ]. Different annealing treatments will also induce phase transitions between the alpha and beta phases of brass, with different metallographic sizes. Furthermore, the different WCAs are determined by the different nanograin sizes, which in turn depend on the different hierarchies in the micro/nanostructure (as manufactured by the combination of laser processing and Cu 2 O evolution)."
} | 2,317 |
35844067 | PMC9544590 | pmc | 7,125 | {
"abstract": "Abstract Light availability is the main regulator of primary production, shaping photosynthetic communities and their production of ecologically important biomolecules. In freshwater ecosystems, increasing dissolved organic carbon (DOC) concentrations, commonly known as browning, leads to lower light availability and the proliferation of mixotrophic phytoplankton. Here, a mixotrophic algal species ( Cryptomonas sp.) was grown under five increasing DOC concentrations to uncover the plastic responses behind the success of mixotrophs in browning environments and their effect in the availability of nutritionally important biomolecules. In addition to the browning treatments, phototrophic, heterotrophic and mixotrophic growth conditions were used as controls. Despite reduced light availability, browning did not impair algal growth compared to phototrophic conditions. Comparative transcriptomics showed that genes related to photosynthesis were down‐regulated, whereas phagotrophy gene categories (phagosome, lysosome and endocytosis) were up‐regulated along the browning gradient. Stable isotope analysis of phospholipid fractions validated these results, highlighting that the studied mixotroph increases its reliance on heterotrophic processes with browning. Metabolic pathway reconstruction using transcriptomic data suggests that organic carbon is acquired through phagotrophy and used to provide energy in conjunction with photosynthesis. Although metabolic responses to browning were observed, essential fatty acid content was similar between treatments while sterol content was slightly higher upon browning. Together, our results provide a mechanistic model of how a mixotrophic alga responds to browning and how such responses affect the availability of nutritionally essential biomolecules for higher trophic levels.",
"introduction": "1 INTRODUCTION Primary production is centrally important to ecological processes since it provides energy and essential biomolecules for upper trophic levels. Ultimately, harvesting of light energy depends on light availability and the ability of phototrophic organisms to regulate photosynthesis according to environmental cues (Gerbaud & André, 1980 ). When light becomes limiting, the capacity of a photosynthetic organism to utilize alternative sources of energy can manifest in a competitive advantage over obligate phototrophs. Mixotrophy refers to the capacity of organisms to complement their photosynthetic activity with exogenous organic carbon sources (i.e., uptake of sugars, engulfment of bacteria) to maintain or enhance their fitness. While such metabolic flexibility can be advantageous with transient low light, energetic and nutrient investments in light harvesting machinery are high (Raven, 1997 ). Consequently, mixotrophs present plastic responses to changes in light and organic carbon availability to maximize fitness (González‐Olalla et al., 2021 ). Such changes can be ecologically relevant since they potentially affect the availability of nutritionally important biomolecules for higher trophic levels (Brett et al., 2006 ; Peltomaa et al., 2017 ). In freshwater and costal ecosystems, increases in terrestrial dissolved organic carbon (DOC) loading is a process commonly called browning. The term browning refers to the darkening of water towards a brown colour which leads to reduced penetration of shorter wavelengths of the visible spectrum of solar radiation (Graneli, 2012 ). In lakes, this reduction in light availability has been linked to reduced dissolved oxygen concentrations (Couture et al., 2015 ), and reduced primary production (Thrane et al., 2014 ), as well as higher costs of drinking water purification (Hongve et al., 2004 ). Given that browning is associated with reduced acid deposition, increased precipitation and increased terrestrial primary production in catchment areas (Kritzberg et al., 2020 ), it is expected that climate change will further intensify ongoing brownification processes in surface waters (Björnerås et al., 2017 ; Kritzberg & Ekström, 2012 ; Spilling et al., 2022 ). In particular, the boreal zone could suffer more intense browning given the combination of increased precipitation (12%–20% expected increase; IPCC, 2013 ; Ruosteenoja et al., 2016 ) and the predominance of water bodies with peatland‐dominated catchments. In addition to changes in water colour, mobilization of terrestrial DOC is accompanied by macronutrients such as nitrogen and phosphorus and micronutrients such as iron (Ged & Boyer, 2013 ; Maranger & Pullin, 2003 ; Qualls & Richardson, 2003 ). Together, these modifications in the physical and chemical properties of water are expected to disturb the balance between algal and bacterial production (Creed et al., 2018 ), reduce energy transfer efficiency through food webs (Hessen, 1998 ), and affect the availability of essential fatty acids (FAs) and sterols for consumers (Martin‐Creuzburg et al., 2009 ; Parrish, 2009 ; Peltomaa et al., 2017 ; Sargent et al., 1995 ). Phytoplankton communities observed in highly brown environments (high DOC concentration) are less diverse (Jones, 1992 ) and are usually dominated by mixotrophic flagellated species (Bergström et al., 2003 ; Deininger et al., 2017 ; Wilken et al., 2018 ). Environmental data together with mesocosm studies have shown that mixotrophic cryptophytes, in particular the genus Cryptomonas , thrive in brown freshwater environments (Bergström et al., 2003 ; Deininger et al., 2017 ; Isaksson et al., 1999 ; Wilken et al., 2018 ), making them promising models of plastic and genetic adaptations to browning. Importantly, mesocosm studies suggest that mixotrophs respond to browning by enhancing bacterial predation (Wilken et al., 2018 ). Since bacteria make use of terrestrially derived DOC for growth (Kritzberg et al., 2006 ), and that active bacterial predation in mixotrophic algae has been linked to decreases in oxygen production (González‐Olalla et al., 2021 ; Wilken et al., 2014 ), increasing algal mixotrophy with browning could alter the carbon and oxygen cycling of lakes. In addition, the trophic transition from phototrophic to mixotrophic, as well as variations in the physicochemical properties of water, can affect the FA and sterol content of algae (Boëchat et al., 2007 ; Peltomaa & Taipale, 2020 ; Piepho et al., 2010 ). Nevertheless, other responses to low light (i.e., changes in pigments), together with diel movements across the water column to avoid light limitation (Gervais, 1997 ), could also explain the success of mixotrophic flagellates in brown environments. Therefore, examining plastic responses of mixotrophic algae to browning is necessary to settle the bases for understanding the biogeochemistry, algal physiology and food web ecology of browning lakes. In this study, we used Cryptomonas sp. (isolated from a clear water lake) as a model system for plastic responses to browning environments to ask the following questions: (i) is mixotroph growth negatively impacted by browning; (ii) does browning affect the use of light energy; (iii) does browning affect FA or sterol quality and quantity in mixotrophs; (iv) do mixotrophs change their photosynthetic pigment content with browning; and (v) which genes show differential expression between phototrophic, glucose‐supplemented phototrophic and browning conditions? To answer these questions, we cultivated Cryptomonas sp. under three different control conditions (phototrophic, glucose‐supplemented phototrophic and heterotrophic) in addition to five degrees of browning (DOC concentrations ranging from 1.5 to 90 mg C L −1 ). During exponential growth, we performed a comparative transcriptome analysis followed by differential gene expression analysis between three browning treatments and two of our controls. The balance between phototrophic and heterotrophic activity was evaluated by the incorporation of 13 C‐labelled carbon from NaHCO 3 into membrane lipids. At the end of the experiment, biomass analyses were carried out to observe differences in FAs, sterols and pigments between treatments.",
"discussion": "4 DISCUSSION The impact of browning processes in aquatic ecosystems includes a shift from autotrophic to heterotrophic‐based basal production (Creed et al., 2018 ). This study provides physiological, biochemical and transcriptomic evidence supporting such a prediction, indicating that browning leads to an increased reliance on heterotrophically acquired energy in our model mixotroph ( Cryptomonas sp.). Our growth rate results validated using the selected algae species as a model mixotroph since glucose supplementation in the presence of light enhanced growth, as previously seen with other mixotrophic species (Smith et al., 2015 ; Zhang et al., 2021 ). In addition, fully heterotrophic conditions could not sustain growth, highlighting the need for basal levels of phototrophy for survival. Increases in water colour, and its associated reduction in light availability (Graneli, 2012 ), did not impair growth compared to fully phototrophic conditions. Conversely, two of our browning treatments (DOC 10 and DOC 90) had higher specific growth rates than under purely phototrophic conditions. Nevertheless, since increases in growth rate were modest with browning, we do not believe that such differences alone explain the dominance of mixotrophic flagellates observed in highly brown environments (Bergström et al., 2003 ; Deininger et al., 2017 ). Given that increases in water colour did not negatively impact growth, we investigated if inorganic carbon fixation is altered by browning in the studied mixotrophic algae. Our stable isotope analysis of phospholipid fractions showed that increases in water colour come with a reduction in inorganic carbon utilization for growth. Furthermore, as demonstrated by our transcriptomic results, the down‐regulation of genes related to photosynthesis further supports this claim and indicates a decreased reliance on phototrophy with browning. Notably, even our lowest DOC concentration (DOC 1.5) showed a down‐regulation of photosynthetic genes. These results suggest that when organic carbon is available, mixotrophs may rapidly capitalize on their ability to increase reliance on heterotrophy to sustain growth, possibly to decrease the high energetic costs of maintaining photosynthetic machinery (Raven, 1995 ). Such changes in the balance between phototrophy and heterotrophy with browning could potentially impact oxygen production and carbon cycling in mixotrophic‐dominated environments. Nevertheless, further studies are needed to observe how oxygen production is affected by increases in water colour. Surprisingly, despite a reduction of phototrophy with browning, we did not observe a decrease in the chlorophyll or carotenoid content per cell. Previous studies on plastic responses of chlorophyll levels to mixotrophy suggest that, together with the down‐regulation of photosynthesis, there is a reduction in the production or enhanced degradation of chlorophyll (González‐Olalla et al., 2021 ; Roth et al., 2019 ; Wilken et al., 2014 ). Nevertheless, these studies were carried out using algae species that can grow fully heterotrophically; hence, our model mixotroph (which requires light to survive) might have specific needs to maintain basal photosynthetic function. Enhanced expression of genes associated with phagotrophy (lysosome, endosome and phagosome KO categories) with browning, but not with glucose supplementation, suggests that heterotrophy was supported by predation of bacteria. These results agree with mesocosm experiments by Wilken et al. ( 2018 ), where mixotrophs showed higher bacterial predation along with temperature and water colour increases. Although the phagotrophy‐related genes used in our analysis have been identified in macrophages and could have different functions in algae (Bohdanowicz & Grinstein, 2013 ; Flannagan et al., 2012 ; Settembre et al., 2013 ), other laboratory studies have observed expression of these groups of genes in algae growing under mixotrophic conditions (Li et al., 2021 ; Lie et al., 2017 ; McKie‐Krisberg et al., 2018 ). In addition, the high expressions observed for key phagotrophy enzymes (PEPCK and PEPC) in our browning treatments agree with previous studies on phagotrophic activity of algae (González‐Olalla et al., 2021 ; Liu et al., 2016 ; McKie‐Krisberg et al., 2018 ). PEPCK is believed to be the enzyme by which degraded bacterial proteins are channelled to energy production (González‐Olalla et al., 2021 ; Liu et al., 2016 ). Following this, oxidation of organic substrates has been shown to accelerate in the presence of bacterial prey (González‐Olalla et al., 2021 ), which matches the up‐regulation of the TCA cycle genes observed in our experiment with browning. The availability and quality of FAs and sterols in phytoplankton have been repeatedly demonstrated as necessary for the growth and reproduction of zooplankton (Brett & Müller‐Navarra, 1997 ; Martin‐Creuzburg et al., 2009 ; Peltomaa et al., 2017 ). This has great ecological importance since zooplankton have a key position in aquatic food webs, linking the flow of dietary energy and essential biomolecules from phytoplankton to fish (Taipale et al., 2016 ). The FA concentration remained unaffected in our experiment while the sterol content and composition varied between browning and fully phototrophic conditions. Since the switch from phototrophy to heterotrophy has been previously characterized by plastic changes in both FAs and sterols in algae (Boëchat et al., 2007 ), it is possible that particular FAs and sterols are required for structural adaptations that occur during these transitions. We hypothesize that the higher phagotrophic activity associated with browning could potentially explain the differences in composition and content observed in sterols. We can only speculate that sterol requirements for our glucose‐supplemented treatment were not different from fully phototrophic conditions, explaining why there were no differences between these treatments. Our results suggest that browning in mixotroph‐dominated environments should not reduce the availability of ecologically important biomolecules at a cellular level. Nevertheless, further studies with other mixotrophic species are needed to observe if the regulatory strategies seen in Cryptomonas sp. with browning are common or diversified among mixotrophs. Overall, we report the metabolic plastic changes, and the possible ecological implications, behind the success of a model mixotroph in browning environments. As observed in our study, metabolic plasticity allows Cryptomonas sp. to thrive in low light environments, where obligated phototrophs are limited by photosynthetic output. Furthermore, we showed that an enhanced reliance on bacterial prey accompanies the transition to mixotrophic growth; hence, the consequences of such a shift in aquatic ecosystems could signify higher CO 2 exportations but increased reliance of whole food webs on bacterial production. Further studies could look more directly into the biogeochemical consequences of phototrophy to mixotrophy transitions. In addition, since we studied the plastic adaptations of a mixotroph isolated from a clear‐water lake, investigating the differences between evolved and plastic effects of browning is important in understanding the genetic architecture necessary for algae to thrive in browning environments."
} | 3,886 |
25806595 | null | s2 | 7,126 | {
"abstract": "Methane monooxygenases (MMOs) are enzymes that catalyze the oxidation of methane to methanol in methanotrophic bacteria. As potential targets for new gas-to-liquid methane bioconversion processes, MMOs have attracted intense attention in recent years. There are two distinct types of MMO, a soluble, cytoplasmic MMO (sMMO) and a membrane-bound, particulate MMO (pMMO). Both oxidize methane at metal centers within a complex, multisubunit scaffold, but the structures, active sites, and chemical mechanisms are completely different. This Current Topic review article focuses on the overall architectures, active site structures, substrate reactivities, protein-protein interactions, and chemical mechanisms of both MMOs, with an emphasis on fundamental aspects. In addition, recent advances, including new details of interactions between the sMMO components, characterization of sMMO intermediates, and progress toward understanding the pMMO metal centers are highlighted. The work summarized here provides a guide for those interested in exploiting MMOs for biotechnological applications."
} | 272 |
27742415 | null | s2 | 7,127 | {
"abstract": "For the past 20 years, research on biodiversity and ecosystem functioning (B-EF) has only implicitly considered the underlying role of environmental change. We illustrate that explicitly reintroducing environmental change drivers in B-EF research is needed to predict the functioning of ecosystems facing changes in biodiversity. Next we show how this reintroduction improves experimental control over community composition and structure, which helps to provide mechanistic insight on how multiple aspects of biodiversity relate to function and how biodiversity and function relate in food webs. We also highlight challenges for the proposed reintroduction and suggest analyses and experiments to better understand how random biodiversity changes, as studied by classic approaches in B-EF research, contribute to the shifts in function that follow environmental change."
} | 217 |
33166000 | PMC11469257 | pmc | 7,129 | {
"abstract": "Abstract Future robots and intelligent systems will autonomously navigate in unstructured environments and closely collaborate with humans; integrated with our bodies and minds, they will allow us to surpass our physical limitations. Traditional robots are mostly built from rigid, metallic components and electromagnetic motors, which make them heavy, expensive, unsafe near people, and ill‐suited for unpredictable environments. By contrast, biological organisms make extensive use of soft materials and radically outperform robots in terms of dexterity, agility, and adaptability. Particularly, natural muscle—a masterpiece of evolution—has long inspired researchers to create “artificial muscles” in an attempt to replicate its versatility, seamless integration with sensing, and ability to self‐heal. To date, natural muscle remains unmatched in all‐round performance, but rapid advancements in soft robotics have brought viable alternatives closer than ever. Herein, the recent development of hydraulically amplified self‐healing electrostatic (HASEL) actuators, a new class of high‐performance, self‐sensing artificial muscles that couple electrostatic and hydraulic forces to achieve diverse modes of actuation, is discussed; current designs match or exceed natural muscle in many metrics. Research on materials, designs, fabrication, modeling, and control systems for HASEL actuators is detailed. In each area, research opportunities are identified, which together lays out a roadmap for actuators with drastically improved performance. With their unique versatility and wide potential for further improvement, HASEL actuators are poised to play an important role in a paradigm shift that fundamentally challenges the current limitations of robotic hardware toward future intelligent systems that replicate the vast capabilities of biological organisms.",
"conclusion": "9 Conclusions and Outlook Only two years have passed since the first two publications on HASEL artificial muscles; [ \n \n 14 \n , \n 15 \n \n ] in this short time span, the understanding of HASEL actuators has increased dramatically—from fabrication, materials, design, and modeling to understanding of the fundamentals of electrohydraulic transduction. Electrohydraulic HASEL actuators are extremely versatile as they achieve all three basic modes of actuation (expansion, contraction, and rotation), they feature the ability to self‐sense their deformation state, and they can be made from a variety of material systems in many different form‐factors and sizes. Current designs of HASEL actuators exhibit well‐rounded, muscle‐like performance which already makes them a viable alternative to other types of artificial muscle. Moreover, HASEL artificial muscles are a young field of research and a technology with a clearly evident potential for drastically improved performance. Realizing the full potential of HASEL artificial muscles will require pushing the limits of current knowledge and technology, which will require creative contributions from researchers including materials scientists, physicists, chemists, mechanical and electrical engineers, and roboticists for many years to come. This progress report described some immediate research opportunities that we can foresee, but we expect that with time many other interesting research questions will arise. Considering their wide potential for further improvement, HASEL actuators are a prime candidate to become the first technology where all performance metrics match or exceed the capabilities of mammalian skeletal muscle, thus solving the centuries‐old grand challenge dating back to Robert Hooke, to replicate the remarkable properties of natural muscle. Such artificial muscles will be key building blocks of future robotic systems that mimic the rich multifunctionality of organisms found in nature. These bioinspired robotic systems will increasingly be integrated into our lives, spanning all the way from robots for disaster response to medical and consumer robotics, as well as to wearable robotic devices and lifelike prostheses. Even though there is still some way to go, we are approaching a world in which human and machine are seamlessly integrated to surpass the physical and cognitive limitations of the human body, as seen today only in science‐fiction movies. It is not a question of if this will happen, but rather when.",
"introduction": "1 Introduction Popular culture is rife with examples of robotic assistants that increase our productivity and enhance our quality of life—from aiding humans in dangerous areas such as space exploration, to helping with mundane tasks around the home. This vision of robots that are integral to our daily lives is multifaceted. In part, it is driven by immediate needs, such as addressing the increasing labor shortages in industries such as agriculture, where there are not enough people to harvest crops in time; in other areas, it is driven by the desire to surpass our physical limitations through the integration of humans and machines: wearable robotics or exoskeletons that surpass our innate strength and endurance, or provide assistance to the elderly or those rehabilitating from injury. While robots today are impressive in their capabilities—especially in controlled environments such as an assembly line in a factory—they are often ill‐suited in unstructured environments and can be dangerous for humans in collaborative situations. In order to realize a future in which robots can adapt to changing tasks and environments, we need machines that are extremely versatile in terms of their decision making and that are based on robotic hardware that can dynamically respond in a variety of situations. In the last decades advances in robotics have been driven predominantly by increases in computing power, new sensing concepts and control algorithms, new machine learning approaches, and more generally artificial intelligence. Less focus has been placed on improving the body of robots, so they continue to rely on rigid, mostly metallic components and electromagnetic actuators such as servo and stepper motors. Ample inspiration for the search of new materials, actuation mechanisms, and sensors in robots can be found in nature. Using a variety of soft materials such as muscle and skin, nature has produced organisms that drastically outperform current robots in dexterity, agility, and adaptability. Especially inspirational is the natural muscle, which has evolved into an extremely versatile actuator. Natural muscle enables the rapid wing‐flapping rates of a hummingbird, it is strong enough to move an elephant, and it enables the complex motion of the arm of an octopus, whose versatility is still unmatched by human‐made machines. Additionally, muscle continuously regenerates over the lifetime of an organism, heals after damage, and is seamlessly integrated with nerves for sensing. [ \n \n 1 \n \n ] \n Actuators are a key component of all robotic systems. Not surprisingly, the remarkable capabilities of muscle have inspired scientists and engineers for centuries ( Figure \n 1 \n ). Already in the 17th century, Robert Hooke experimented with gunpowder in the attempt to replicate the performance of natural muscle. [ \n \n 2 \n \n ] Evidently, Robert Hooke's efforts were unsuccessful, and creating actuators that truly match all performance metrics of mammalian skeletal muscle simultaneously—such as actuation strains of 40%, [ \n \n 3 \n \n ] strain rates of 500% s −1 , [ \n \n 3 \n \n ] actuation stresses of 0.35 MPa, [ \n \n 4 \n \n ] energy densities of 40 J kg −1 , [ \n \n 4 \n \n ] and efficiencies of 40% [ \n \n 3 \n \n ] —remains a grand challenge to this day. [ \n \n 5 \n \n ] For example, the July 2019 issue of Science featured three separate research articles dedicated to the development of artificial muscles. [ \n \n 6 \n , \n 7 \n , \n 8 \n \n ] Today, a broad range of artificial muscle actuators exists, including dielectric elastomer actuators (DEAs), [ \n \n 9 \n \n ] fluid‐driven soft actuators, [ \n \n 10 \n \n ] ionic polymer‐metal composites, [ \n \n 11 \n \n ] shape memory alloys, [ \n \n 12 \n \n ] and thermally responsive fiber‐based muscles. [ \n \n 13 \n \n ] Many of these types of actuators excel in specific performance metrics, but exhibit poor performance in others. Figure 1 Selected sources of inspiration for hydraulically amplified self‐healing electrostatic (HASEL) actuators. The timeline shows key sources of inspirations for the invention of the HASEL technology. Photo of the hummingbird and elephant by Günter Oesterling and reproduced with permission. Photo of the octopus by Jeahn Laffitte on Unsplash and reproduced with permission. “Soft, electrostatic actuator studied by Röntgen” image: Reproduced with permission. [ \n \n 137 \n \n ] Copyright 2010, National Academy of Sciences USA. “McKibben actuator” image: Reproduced from ref. [ \n \n 10 \n \n ] . “Pneumatically driven flexible microactuators” image: Reproduced with permission. [ \n \n 59 \n \n ] Copyright 1996, Elsevier Ltd. “Dielectric elastomer actuators with strains > 100%” image: Reproduced with permission. [ \n \n 9 \n \n ] Copyright 2000, AAAS. “Self‐sensing in dielectric elastomer actuators” image: Reproduced with permission. [ \n \n 38 \n \n ] Copyright 2008, Elsevier Ltd. “Hydrostatically coupled dielectric elastomer actuators” image: Reproduced with permission. [ \n \n 39 \n \n ] Copyright 2010, IEEE. “Embedded pneumatic networks (PneuNets) of channels in elastomers” image: Reproduced with permission. [ \n \n 60 \n \n ] Copyright 2011, Wiley‐VCH. “Zipping dielectric elastomer actuators” image: Reproduced with permission. [ \n \n 40 \n \n ] Copyright 2012, SPIE. “PneuNet tentacle” image: Reproduced with permission. [ \n \n 61 \n \n ] Copyright 2013, Wiley‐VCH. “Self‐healing dielectric elastomer actuators” image: Reproduced with permission. [ \n \n 57 \n \n ] Copyright 2014, American Institute of Physics. “Pneumatic pouch motors” image: Reproduced with permission. [ \n \n 63 \n \n ] Copyright 2014, IEEE. “HASEL actuators” top row images; two left images: Reproduced with permission. [ \n \n 14 \n \n ] Copyright 2018, The Authors, published by AAAS; Two right images: Reproduced with permission. [ \n \n 15 \n \n ] Copyright 2018, The Authors, published by AAAS. “HASEL actuators” bottom row, left four images: Reproduced under the terms of the CC‐BY Creative Commons Attribution 4.0 International license ( https://creativecommons.licenses/by/4.0 ). [ \n \n 89 \n \n ] Copyright 2019, The Authors, published by Wiley‐VCH. “HASEL actuators” bottom row, right two images: Reproduced with permission. [ \n \n 97 \n \n ] Copyright 2019, Wiley‐VCH. Herein, we discuss the background, fundamentals, recent progress (including articles appearing online before April 7, 2020), and future research opportunities for hydraulically amplified self‐healing electrostatic (HASEL) artificial muscles. [ \n \n 14 \n , \n 15 \n \n ] HASEL artificial muscles are a versatile new technology that couples electrostatic and hydraulic forces to achieve excellent overall muscle‐like performance, the ability to self‐sense their state of deformation, and the ability to self‐heal from electrical damage. HASEL actuators were first introduced in 2018 by our research group, [ \n \n 14 \n , \n 15 \n \n ] and in this progress report we provide our point of view on the development of HASEL actuators, with the overall intention to provide a comprehensive discussion that will stimulate future research on the HASEL technology but also more generally to help accelerate the development of “robotic materials”—a new class of materials systems that tightly integrate actuation, sensing and even computation to provide physical building blocks for the intelligent systems of the future. While we include background on artificial muscle technologies that inspired HASEL actuators, this is not a comprehensive review of the fields of artificial muscles or soft robotics. Interested readers are referred to other articles, which discuss additional specific aspects or give a more comprehensive overview of these fields of research. [ \n \n 3 \n , \n 4 \n , \n 16 \n , \n 17 \n , \n 18 \n , \n 19 \n , \n 20 \n , \n 21 \n , \n 22 \n , \n 23 \n , \n 24 \n , \n 25 \n , \n 26 \n , \n 27 \n , \n 28 \n , \n 29 \n , \n 30 \n , \n 31 \n , \n 32 \n , \n 33 \n , \n 34 \n , \n 35 \n \n ] \n 1.1 Inspiration for HASEL Artificial Muscles Since Robert Hooke's early attempts, many different types of artificial muscles have been investigated. However, researchers have been unsuccessful in creating an artificial muscle that matches the performance of natural muscle in all metrics. The invention of HASEL actuators began with a desire to create a new type of artificial muscle that combines the benefits of two well‐known technologies—dielectric elastomer actuators (DEAs) and fluid‐driven soft actuators—while also addressing their shortcomings. 1.1.1 Dielectric Elastomer Actuators In the middle of the 19th century James C. Maxwell derived the so‐called Maxwell equations, [ \n \n 36 \n \n ] which laid the foundations for classical electromagnetism. The Maxwell equations showed that electric fields induce mechanical stresses in matter (the Maxwell stress). In 1880, Röntgen sprayed electric charges onto a strip of natural rubber; he observed that the electric field between the charges induced a Maxwell stress in the rubber which caused the rubber to increase in length, thereby discovering the underlying principle of dielectric elastomer actuators (DEAs). [ \n \n 37 \n \n ] However, it was not until 2000 that DEAs gained widespread interest in the research community, when Pelrine et al. [ \n \n 9 \n \n ] discovered that prestretching DEA materials enables voltage‐controlled actuation strains >100%. The seminal work by Pelrine et al. motivated many researchers to understand and improve the performance of DEAs, as large actuation strains combined with the benefits of electrical control make DEAs a compelling technology for soft actuators. DEAs allow for close integration of actuation and sensing—analogous to the proprioceptive capabilities of natural muscle. [ \n \n 1 \n \n ] Since DEAs are deformable capacitors, the state of deformation of a DEA can be determined by measuring its capacitance. Jung et al. [ \n \n 38 \n \n ] showed that both sensing and actuation signals for DEAs can be combined into a single input to enable self‐sensing of actuation. While many configurations of DEAs have been explored, two types in particular have influenced the invention of HASEL actuators: hydrostatically coupled DEAs [ \n \n 39 \n \n ] in which liquids couple the deformation of multiple elastomer membranes and DEA‐based pumps in which electrostatic “zipping” of dielectric elastomer membranes onto rigid substrates is used to pump fluids. [ \n \n 40 \n , \n 41 \n \n ] \n Despite the convenience of electrical control and excellent electromechanical performance, DEAs usually need a prestretch [ \n \n 9 \n , \n 42 \n \n ] and rigid frames to achieve large strains, or stacked configurations to achieve linear, muscle‐like contraction upon activation; [ \n \n 43 \n , \n 44 \n \n ] these constraints limit the design freedom of DEAs. Further, DEAs require stretchable material systems for dielectric layers and electrodes. [ \n \n 45 \n , \n 46 \n , \n 47 \n \n ] Elastomer membranes and stretchable electrodes often suffer from reduction in electrical performance, mechanical fatigue, and degradation under high strains. [ \n \n 45 \n , \n 48 \n , \n 49 \n , \n 50 \n , \n 51 \n \n ] Reliability is a critical challenge to up‐scaling of DEAs for practical applications, such as humanoid robots, as the elastomer layers are prone to catastrophic failure by dielectric breakdown at material defects, and the fabrication of large, thin, defect‐free elastomer membranes is challenging. [ \n \n 48 \n , \n 52 \n , \n 53 \n , \n 54 \n \n ] One approach to address this problem is the use of self‐clearing electrodes, which electrically insulate the damaged zones of the elastomer. [ \n \n 53 \n , \n 55 \n , \n 56 \n \n ] Industrial high voltage transformers use liquid dielectrics that immediately return to an insulating state after dielectric breakdown. Hunt et al. [ \n \n 57 \n \n ] used this principle to build a self‐healing DEA based on a foam swollen with a liquid dielectric. When damaged (due to breakdown or puncturing) the liquid dielectric flowed into the damaged zone and resealed the dielectric layer. The work by Hunt et al. and the self‐healing capabilities of industrial high voltage transformers served as an inspiration for the self‐healing mechanism in HASEL actuators. 1.1.2 Fluid‐Driven Soft Actuators The first type of soft pneumatic actuator was introduced in 1958 by Gaylord. [ \n \n 10 \n \n ] These so‐called McKibben actuators consist of a soft tube wrapped in a fiber mesh and linearly contract when pressurized with a fluid. In the 1990s, Suzumori et al. [ \n \n 58 \n , \n 59 \n \n ] demonstrated more complex modes of actuation with pneumatically‐driven microactuators, in which multiple pneumatic chambers were combined into a single fiber reinforced tube. Wide development of soft pneumatic actuators began in earnest in 2010; with the advent of rapid manufacturing tools such as 3D printing, actuators based on pneumatic networks (PneuNets) in elastomers could be easily fabricated. [ \n \n 60 \n \n ] PneuNets can be designed for a variety of modes of actuation and have led to a diverse range of soft robot configurations and capabilities. [ \n \n 60 \n , \n 61 \n , \n 62 \n \n ] In contrast to the elastomeric materials used for PneuNets, pouch motors or “Peano” actuators [ \n \n 63 \n \n ] consist of pouches made from thin, inextensible plastic films. When inflated with pressurized air, the pouch changes shape, which (depending on orientation and constraints) can be used for contraction, expansion, or bending. Whereas fluid‐driven soft actuators offer a large design freedom and versatile modes of actuation, the need for a source of pressurized fluid (e.g., compressors, pumps, or pressurized reservoirs of fluid) is a distinct disadvantage. Additionally, valves are required to regulate flow into and out of the actuators. Therefore, fluid‐driven soft actuators are often tethered. [ \n \n 64 \n , \n 65 \n , \n 66 \n \n ] Untethered, mobile, fluid‐driven soft robots have been realized, [ \n \n 67 \n \n ] but the capabilities of mobile pumps are limited, and the need for valves leads to bulky and heavy control systems for robots with multiple actuators. Flow control has been successfully integrated directly into the structure soft actuators, [ \n \n 68 \n , \n 69 \n , \n 70 \n , \n 71 \n \n ] but attempts to integrate a pressurized fluid‐reservoir into the soft structure have led to limited performance. [ \n \n 71 \n \n ] Finally, the losses associated with flow of fluids through tubing and valves limit the speed and reduce the efficiency of fluid‐driven soft actuators. 1.2 Fundamental Principles of HASEL Artificial Muscles A basic HASEL actuator consists of a deformable shell (flexible or stretchable) that is covered with a pair of opposing electrodes and filled with a liquid dielectric ( Figure \n 2 a ). [ \n \n 14 \n \n ] When voltage is applied to the electrodes, a Maxwell stress acts on the shell and the liquid dielectric, thereby driving local redistribution of the liquid dielectric, which results in a shape change of the soft hydraulic structure. This working principle enables HASEL actuators to use the principles of hydraulic amplification (Pascal's law) (Figure 2b ). Changing the size of the electrode with respect to the size of the shell modifies stroke and force output (analogous to a hydraulic system of pistons); a large electrode results in larger actuation strain but lower force output whereas an actuator with small electrodes results in a smaller actuation strain but larger force output. The use of liquid dielectrics also enables self‐healing from dielectric breakdown to improve reliability (Figure 2c ); when dielectric breakdown occurs, the liquid redistributes and returns to an insulating state. Additionally, HASEL actuators possess the ability to self‐sense their state of deformation (Figure 2d ). The structure of a HASEL actuator is that of a deformable capacitor. When the actuator deforms due to an external force or applied voltage, the size and spacing of the electrodes changes, resulting in a measurable change of capacitance that is a function of the state of deformation of the actuator. Figure 2 Basic operating principles of HASEL actuators. a) HASEL actuators consist of polymer shells that are coated with opposing electrodes and that are filled with a liquid dielectric. When a voltage is applied to the actuator an electric field arises between the electrodes. The electric field causes a Maxwell stress in the actuator, which leads to redistribution of the liquid dielectric and deformation of the actuator. b) HASEL actuators use the principles of hydraulic amplification to generate forces. c) After a dielectric breakdown event, the liquid dielectric redistributes and returns to its insulating state. This property leads to electrical self‐healing in HASEL actuators. d) The electrodes of HASEL actuators form a capacitor. Shape changes due to applied voltages or external forces can be detected by measuring the capacitance. a–d) Adapted with permission. [ \n \n 14 \n \n ] Copyright 2018, The Authors, published by AAAS. Overall, the fundamental principles of HASEL actuators are inspired by the versatility of fluid‐driven soft actuators as well as the performance and electrical control of DEAs. HASEL actuators use electrostatic forces to locally generate hydraulic pressure in soft structures that are filled with liquid dielectrics. This approach combines the benefits of both DEAs and fluid‐driven soft actuators, while also addressing their challenges. The use of electrostatic forces allows for fast and efficient activation. Because a fluid is used to induce deformation, HASEL actuators can be designed for a variety of actuation modes, whereas the local generation of pressure avoids the need for valves and tubing. The use of liquid dielectrics also improves reliability and allows up‐scaling, as liquid dielectrics can self‐heal from dielectric breakdown. Additionally, many designs of HASEL actuators do not require dielectrics and electrodes that are stretchable, but they can be built from materials that are only flexible instead of stretchable; flexible materials are a much larger class of materials than stretchable materials and include high‐performance electrical insulators (e.g., poly(vinylidene fluoride) terpolymer) [ \n \n 72 \n \n ] and low‐resistance conductors (e.g., thin metal films or conductive silver inks). 1.3 Outline of the Progress Report The Progress Report is organized in topical sections and loosely guided by the order in which the technology developed. Most sections include a summary that describes specific problems and possible avenues for future research. Sections 2 and 3 describe different designs of HASEL actuators whose shells are made with elastomeric or thermoplastic materials, respectively. Section 3 also presents a method for rapid prototyping of thermoplastic actuators. Section 4 discusses the demonstrated performance metrics of HASEL actuators. Section 5 describes the first modeling efforts to understand the electromechanical behavior of HASEL actuators. Section 6 discusses several pathways for substantially improving the performance of HASEL actuators. Section 7 discusses closely related technologies. Section 8 describes high voltage electronics, sensing, and control of HASEL actuators toward implementation in untethered soft robotic systems. Section 9 discusses the significance of HASEL actuators for a new generation of lifelike robots."
} | 5,936 |
38957543 | PMC11215720 | pmc | 7,131 | {
"abstract": "Biomass is an abundantly available, underutilized feedstock\nfor\nthe production of bulk and fine chemicals, polymers, and sustainable\nand biodegradable plastics that are traditionally sourced from petrochemicals.\nAmong potential feedstocks, 2,5-furan dicarboxylic acid (FDCA) stands\nout for its potential to be converted to higher-value polymeric materials\nsuch as polyethylene furandicarboxylate (PEF), a bio-based plastic\nalternative. In this study, the sustainable, electrocatalytic oxidation\nof stable furan molecule 2,5-bis(hydroxymethyl)furan (BHMF) to FDCA\nis investigated using a variety of TEMPO derivative electrocatalysts\nin a mediated electrosynthetic reaction. Three TEMPO\ncatalysts (acetamido-TEMPO, methoxy-TEMPO, and TEMPO) facilitate full\nconversion to FDCA in basic conditions with >90% yield and >100%\nFaradaic efficiency.\nThe remaining three TEMPO catalysts (hydroxy-TEMPO, oxo-TEMPO, and\namino-TEMPO) all perform intermediate oxidation of BHMF in basic conditions\nbut do not facilitate full conversion to FDCA. On the basis of pH\nstudies completed on all TEMPO derivatives to assess their electrochemical\nreversibility and response to substrate, pH and reversibility play\nsignificant roles in the catalytic ability of each catalyst, which\ndirectly influences catalyst turnover and product formation. More\nbroadly, this study also highlights the importance of an effective\nand rapid electroanalytical workflow in mediated electrosynthetic\nreactions, demonstrating how voltammetric catalyst screening can serve\nas a useful tool for predicting the reactivity and efficacy of a catalyst–substrate\nelectrochemical system.",
"conclusion": "Conclusions In summary, we have demonstrated the efficient\nand complete electrochemical\nconversion of BHMF to FDCA via TEMPO mediation. Specifically, we highlight\nthe ability of reversible TEMPO derivatives (acetamido-TEMPO, methoxy-TEMPO,\nand TEMPO) which are highly electrochemically reversible in basic\nconditions to catalyze the complete oxidation of BHMF to FDCA, with\nquasi-reversible or irreversible derivatives (hydroxy-TEMPO, oxo-TEMPO,\nand amino-TEMPO) only accomplishing intermediate oxidation of BHMF.\nIn addition to being an important electrochemical transformation of\nbiomass derived substrates, these results emphasize the importance\nof an effective voltammetry workflow in catalysis studies for predicting\nreactivity and efficacy. The results of this model alcohol oxidation\nstudy confirm hypotheses based on quick and easy CV studies that can\nbe completed in 1 day, supporting the argument that it is essential\nfor an effective electrocatalysis or electrosynthesis study—especially\na mediated study—to include voltammetry data. Compared to traditional\nmethods for organic synthesis, electrosynthesis offers a more streamlined,\neconomic, and environmentally friendly approach. However, using a\nredox mediator in a mediated electrosynthetic reaction can produce\nsignificant separation costs (in energy expenditure, time, and money). 26 For this reason, homogeneous mediated electrosynthetic\nreactions are most suitable for laboratory applications, as significant\nseparation costs are present if a solution-based catalyst is used\nin a commercialized reaction. To address this concern, many commercialized\nelectrosynthesis reactions use modified electrodes where the catalyst\nis immobilized or anchored to the electrode, eliminating the need\nfor catalyst separation from the product. To further adapt this system\nfor commercial scale, highest performing TEMPO catalysts would be\nimmobilized on a carbon electrode, anchored through a linear polyethylenimine\n(L-PEI) polymer backbone in an effort to reduce separation costs and\npromote catalyst recovery and reuse in a scaled-up reaction.",
"introduction": "Introduction Global industrialization has escalated\nthe demand for energy while\nsimultaneously contributing to carbon emissions, global warming, and\nclimate change. To avoid irreparable damage to the atmosphere and\necosystems on earth, there is an increasing urgency for large-scale\ndecarbonization efforts, including: renewable energy generation, electrofuels,\nand fine chemical production derived independently from fossil fuel\nresources (e.g., coal, crude oil, natural gas). Green chemistry is\na prominent effort focused on the sustainable production of chemicals\nto meet environmental goals. Over the last decade, organic electrosynthesis\nhas emerged as a green approach for the synthesis of important molecules\nand chemical feedstocks. Organic electrosynthesis uses electricity\nto synthesize molecules via the oxidizing and/or reducing power from\nan electrode. Using electricity directly to carry out redox transformations\nallows the replacement of toxic oxidizing/reducing chemicals. Not\nonly are electrons a cleaner reagent, but they can be sourced from\nrenewable “green” resources such as wind turbines, solar\npanels, and biomass. In particular, biomass has shown promise as a\nversatile and renewable feedstock and an alternative source for chemicals\nand energy derived from petroleum. 1 Biomass,\nderived from plant matter, is an abundantly available raw material\nfrom natural sources and industrial waste streams, which makes it\nan attractive source for carbon-neutral fuels and precursor chemicals.\nPresently, biomass is underutilized as a feedstock to produce both\nbulk and fine chemicals, polymers, and sustainable and biodegradable\nplastics, 2 all traditionally produced from\na dwindling fossil fuel supply. The United States Department\nof Energy published a list of Top\nValue-Added Chemicals from Biomass to motivate and drive research\nin the field. 3 Two molecules highlighted\nin this pubication for their potential to be converted to higher-value\nchemicals are 5-hydroxymethylfurfural (HMF) and 2,5-furan dicarboxylic\nacid (FDCA). Several important transformations stem from the biomass-derived\nfuran molecule HMF which is derived from C 6 sugars glucose\nand fructose. 1 , 4 A key transformation of HMF is\noxidation into the high value-added product FDCA. FDCA is one of the\n12 building block chemicals highlighted from biomass upgrading because\nit is a precursor for polymeric materials including polyethylene furandicarboxylate\n(PEF), a bio-based alternative to petroleum-derived polyethylene terephthalate\n(PET) which is the most common plastic polymer material used in clothing,\ncontainers, plastic bottles, and manufacturing. 5 , 6 While\nbatch processing can accomplish the initial conversion of biomass-derived\nstarch and cellulose to intermediate sugar platforms such as HMF,\nthe conversion of these building blocks to secondary fine chemicals\nand polymerized materials remains challenging and experimentally complex.\nTraditional methods for the catalytic conversion of furan molecules\nrely on noble metal catalysts, organic solvents, and high temperatures\nand pressures. 7 − 9 Alternatively, electrochemistry offers a streamlined\napproach to the synthesis of furan derivatives by eliminating the\nneed for stoichiometric oxidizing reagents and harsh reaction conditions. 10 , 11 Instead, an applied electric potential serves as thermodynamic driving\nforce for catalysis at lower temperatures. Such an approach\nwas used by Choi and co-workers, where the electrochemical\noxidation of HMF to FDCA was carried out using TEMPO as a homogenous\ncatalyst. 12 Here, authors demonstrated\na highly efficient electrocatalysis in pH 9.2 aqueous media under\nambient conditions, with a ≥ 99% yield of FDCA and a Faradaic\nefficiency of ≥93%. Additionally, a related electrocatalytic\nsystem was designed using an n-type BiVO 4 semiconductor\nas a photoanode where the overpotential required to initiate HMF oxidation\nwas reduced significantly. Here, the photovoltage gained from the\nphotogenerated holes in the BiVO 4 decreased the applied\npotential bias necessary for TEMPO oxidation. This study demonstrated\nthat HMF transformations could be achieved at high yields and efficiencies\nwithout the need for precious-metal catalyst electrodes and stimulated\ninterest in using organic mediators, specifically TEMPO, as homogeneous\ncatalysts in biomass valorization reactions. With this interest in\nTEMPO-mediated oxidation reactions, Choi and co-workers further investigated\nthe electrocatalytic oxidation of HMF to FDCA using a less expensive\nTEMPO derivative, 4-acetamido-TEMPO (ACT), as an alternative catalyst. 13 Here, authors show the advantage of TEMPO- and\nACT-mediated electrooxidation that works efficiently in mildly basic\nconditions (pH 9–10) as compared to heterogeneous catalysts\nthat require the use of more basic media. This is important because\nHMF becomes increasingly more unstable at high pH, polymerizing and\nforming insoluble polymer products. Chadderdon et al. accomplished\nthe high efficiency paired electrochemical conversion of HMF to 2,5-bis(hydroxymethyl)furan\n(BHMF) and FDCA in a divided electrochemical cell. 14 The hydrogenation of HMF to BHMF was catalyzed by carbon-supported\nAg nanoparticles (Ag/C) and the homogeneous oxidation of HMF to FDCA\nwas facilitated by 4-acetamido-TEMPO (ACT) at a carbon felt electrode.\nThe paired cell achieved high yields of BHMF and FDCA (85% and 98%,\nrespectively), demonstrating the utility of electrochemical conversions\nof biomass-derived furan molecules and once again revealing the effectiveness\nof TEMPO derivative catalysts. Recently, Zhu et al. proposed a new\nsubstrate alternative to HMF by studying the electrocatalytic production\nof FDCA from 2,5-bis(hydroxymethyl)furan (BHMF). 15 BHMF is a reduced, more stable furan derivative of HMF\nthat offers increased thermal and chemical stability compared to HMF,\na molecule often subject to degradation at high pH and over time.\nIn this study, CoOOH/Ni electrodes were fabricated and achieved complete\nBHMF conversion with 90.2% FDCA yield and 100% current efficiency\nwhen paired with H 2 evolution reaction at the cathode.\nTraditional thermal catalytic conversion of BHMF to FDCA has also\nbeen explored: Li et al. disclosed oxidation of BHMF to FDCA catalyzed\nby a carbon nanotube-supported Pd catalyst (Pd/CNT). 16 The Pd/CNT catalyst demonstrated a maximum FDCA yield of\n93.0% with a full conversion of BHMF after 60 min at 60 °C. This\nwas compared to the traditional furan oxidation synthesis using HMF\nas a substrate, which only yielded 35.7% FDCA. Liu et al. also recently\ndemonstrated selective production of FDCA from BHMF using porous nitrated\ncarbon-supported bimetallic Au–Pd nanocatalysts that facilitated\nBHMF conversion of 100% and FDCA yield of 95.8% at 100 °C. 17 The success of these thermal approaches for\nBHMF oxidation highlight a promising route to combining the efficiency\nof TEMPO mediated HMF electrochemical oxidation strategies with the\nthermal and pH stability of BHMF demonstrated through heterogeneous\ncatalysis. Here, we use organic electrochemistry as an approach\nthat combines\nthe production demand of renewable, non-toxic, and recyclable polymeric\nplastics with the abundance of biomass feedstocks. Specifically, we\nfocus on an intermediate reaction in this biomass upgrading process:\nthe electrochemical oxidation of a reduced furan derivative, 2,5-bis(hydroxymethyl)furan\n(BHMF) to the polymeric precursor FDCA via mediated electrocatalysis\n( Figure 1 ). BHMF in\nthis study serves as a model substrate for electrochemical oxidation\nof a wide variety of sugar-based substrates. In contrast to HMF, BHMF\nlacks the reactive aldehyde group present on HMF making it a more\nstable intermediate. HMF is notorious for its instability and is prone\nto quickly polymerize and participate in side reactions, such as the\nformation of humins. 18 − 20 Selecting a more stable furan molecule, such as BHMF,\ncould overcome certain industrial bottlenecks relating to storage\nand reaction stability. Utilizing a library of TEMPO redox mediators,\nwe identify the key molecular parameters necessary for catalyzing\nselective alcohol oxidation. We demonstrate the complete electrochemical\noxidation of BHMF to FDCA in near unity yields and high Faradaic efficiencies.\nThese studies pave the way for BHMF as an alternative precursor to\nHMF for the electrosynthesis of FDCA and represent an opportunity\nfor advancement in redox mediated commodity chemical production from\nbiomass, an underutilized strategy in the present literature. Figure 1 Catalytic scheme\nof mediated electrochemical oxidation of BHMF\nto FDCA. Oxidation of TEMPO catalyzes the eight electron oxidation\nwhere catalyst regeneration occurs at the electrode surface and homogeneous\noxidation of substrate occurs in the bulk solution. Listed advantages\nof using BHMF as a substrate over HMF.",
"discussion": "Results and Discussion Direct oxidation of BHMF at\nthe electrode surface is not feasible\ndue to the competing water oxidation reaction. This is demonstrated\nin Figure 2 which shows\nthe cyclic voltammetric response of BHMF in borate buffer (pH 9.2,\nglassy carbon (GC) electrode, dashed green trace) where OER is observed\nat 1.2 V vs SCE. Aside from the onset of OER, no other electrochemical\nfeatures are present in the BHMF voltammogram, indicating that it\nis not capable of being heterogeneously oxidized at the GC electrode\nsurface in aqueous conditions. To enable electrocatalysis, an electrochemical\nmediator was chosen to facilitate homogeneous oxidation of BHMF and\nlower the overpotential required for the reaction. Nitroxyl radical-containing\ncompounds, such as 2,2,6,6-tetramethylpiperidine 1-oxyl (TEMPO) and\nits derivatives are widely utilized for their ability to electrocatalytically\noxidize a wide range of alcohols under mild and environmentally benign\nconditions. The ability of TEMPO to serve as a catalyst for BHMF oxidation\nwas examined using cyclic voltammetry (CV) where TEMPO derivatives\nwere evaluated for catalytic activity in the absence and presence\nof BHMF. One TEMPO derivative, acetamido-TEMPO (ACT) is shown as a\nrepresentation of classic electrocatalytic behavior ( Figure 2 , solid light green trace).\nThe ACT electrochemical current response shows an ideal, reversible\nvoltammogram centered at a redox potential of 0.60 V vs SCE and an\nanodic peak current of 4.47 × 10 –3 mA. Once\nthe substrate (BHMF) is introduced to a solution with catalyst (TEMPO)\npresent, BHMF is oxidized by the electro-generated oxoammonium cation,\n(TEMPO + ). This can be seen in the voltammogram in Figure 2 where catalytic\nbehavior is observed in the voltammogram with BHMF and ACT present\n(solid dark green trace). With BHMF addition, there is an increase\nin the anodic current (peak current 1.34 × 10 –2 mA, three-fold increase over the only TEMPO peak current) with an\nassociated decrease in the cathodic return wave, indicative of homogeneous\ncatalysis which returns the reduced form of TEMPO at the electrode\nsurface. Figure 2 CVs showing electrochemical response of 1 mM BHMF substrate (dashed\nlight green trace), 1 mM ACT (solid light green trace), and electrocatalytic\nresponse of 1 mM BHMF + 1 mM ACT (solid dark green trace). Scans taken\nat 10 mV/s in 0.5 M borate buffer (pH 9.2). Scan Rate Studies To assess the ability of other TEMPO-based\nderivatives to act as competent oxidative mediators, their electrochemical\nreversibility needed to be evaluated at the basic pH necessary for\nFDCA conversion. Cyclic voltammetry (CV) scan rate studies were completed\non six TEMPO derivatives to measure their redox potential and reversibility.\nExamined TEMPO derivatives were acetamido-TEMPO (ACT), methoxy-TEMPO\n(MT), TEMPO (T), hydroxy-TEMPO (HT), oxo-TEMPO (OT), and amino-TEMPO\n(AT). CVs were run in a 0.5 M borate buffer solution at pH 9.2 containing\n1 mM of varying TEMPO catalysts and scan rates ranged from 10 mV/s\nto 1 V/s. Scan rate dependence of an electroactive molecule can give\ninsight into the stability and reversibility of a molecule as well\nas the redox potential at which it is oxidized. Cyclic voltammograms\nfor all six TEMPO catalysts are shown in Figure 3 at varying scan rates to demonstrate their\nelectrochemical behavior. The oxidation potentials for each TEMPO\nderivative are as follows: ACT was 0.62 V, MT was 0.62 V, T was 0.53\nV, HT was 0.62 V, OT was 0.715 V, and AT was 0.53 V. Three out of\nsix catalysts, ACT (peak ratio of 0.90), MT (peak ratio of 0.87),\nand T (peak ratio of 0.97) display electrochemically reversible behavior\nin basic solvent conditions, as shown by equal magnitude anodic and\ncathodic peak currents. In addition, tight peak potential separation\n(Δ E p ) was observed for these catalysts\n(64 mV for ACT, MT, and T) indicating facile heterogeneous electron\ntransfer. Finally, these three catalysts exhibited ideal linear relationship\nof the peak current ( i p ) to square root\nscan rate as predicted by the Randles-Sevcik equation (see Supporting Information , SI , Figures S1–S2 ). This ideal voltammetric\nbehavior indicates that these TEMPO species are stable and freely\ndiffusing in solution. In contrast, the remaining three catalysts\n(HT, OT, AT), do not display electrochemically reversible behavior\nover all scan rates. HT and OT lose reversibility at lower scan rates\ndemonstrated by a departure of cathodic/anodic peak current ratio\nfrom unity, while AT lacks reversibility over all scan rates. OT appears\nquasi-reversible above 250 mV/s scan rate where a reverse peak is\nstill present as indicated by a peak ratio of 0.44 at 1000 mV/s. However,\nas scan rate decreases, the peak current ratio decreases to 0.32 and\n0.24 at 500 and 250 mV/s, respectively. Below a scan rate of 250 mV/s,\nOT completely loses reversibility as indicated by peak ratios that\nrange from 0.20 down to 0.17 at the slowest scan rate, 10 mV/s. Likewise,\nHT shows a similar trend of reversibility where HT is quasi-reversible\nabove 100 mV/s scan rate as indicated by increasing peak ratios of\n0.68, 0.74, 0.80, 0.83, and 0.84 at scan rates 100, 150, 250, 500,\nand 1000 mV/s, respectively. Below 100 mV/s, HT loses reversibility\nand the return peak, demonstrated by a significant decrease in peak\ncurrent ratio to 0.42 at 50 mV/s and 0.24 at 10 mV/s. AT appears irreversible\nat all scan rates, and, as a result, the reverse peak current is difficult\nto reliably detect. The forward anodic peak current is consistent\nwith the Randles-Sevcik linear relationship to square root scan rate.\nA complete list of peak current ratios for all catalysts can be found\nin SI Table S1 . This voltammetric behavior of HT, OT, and AT is consistent with\nan “EC” mechanism where an initial electrochemical step\nis followed by a chemical reaction, indicating the chemical decay\nof the catalytically active oxoammonium intermediate. This chemical\ndecay of the oxidized intermediate prevents the TEMPO species from\nbeing reduced back to its original state which results in the disappearance\nof the reverse peak and gives rise to peak current ratios <1.0. Figure 3 Cyclic\nvoltammograms for a range of TEMPO derivatives with different\nstructures (R groups) illustrating the relationship between the differences\nin R groups, oxidizing activity, and reversibility. (A) acetamido-TEMPO\n(ACT, green traces), (B) methoxy-TEMPO (MT, orange traces), (C) TEMPO\n(T, yellow traces), (D) hydroxy-TEMPO (HT, blue traces), (E) oxo-TEMPO\n(OT, purple traces), and (F) amino-TEMPO (AT, red traces). CVs obtained\nin a 0.5 M borate buffer solution (pH 9.2) containing 1 mM of varying\nTEMPO catalysts at scan rates 10 mV/s–1 V/s. The voltammetric scan rate data in Figure 3 can be used to understand\nthe reversibility\nof each catalyst and, in turn, predict the efficacy of the catalyst\nin an electrosynthetic reaction. CV can be a diagnostic technique\nand predictive model for the outcome of a subsequent bulk electrolysis\nexperiment based on observations in mechanism and reversibility. Electrochemical\nreversibility indicates that a chemical species is capable of “turning\nover” or undergoing many redox cycles with no interfering chemical\nreactions nor other electrochemical events. An effective catalyst\nwill continuously transfer electrons between the electrode and substrate\nwithout degradation or irreversible interference with the reaction\nof interest. For example, reversible catalysts ACT, MT, and T are\npredicted to experience high turnover and readily facilitate the oxidation\nof BHMF to FDCA. Quasi-reversible catalysts HT and OT are predicted\nto facilitate some BHMF oxidation, but due to competing chemical reactions\nin solution, may not perform as well as the reversible TEMPO catalysts.\nAT is predicted to facilitate minimal BHMF oxidation based on the\nirreversible CVs and the chemical reactions occurring in solution\nwith the oxidized oxoammonium intermediate that outcompete the turnover\nof AT back to its reduced state. Substrate Titrations After the initial screening and\nassessment of the electrochemical reversibility of the TEMPO catalysts\nin basic conditions, the effect of adding substrate was examined.\nBriefly, BHMF substrate was added to a 1 mM solution of catalyst in\n1 mM aliquots and the resulting voltammogram recorded. Figure 4 shows voltammograms of all\nsix catalysts comparing the current responses with no substrate present\nto those with increasing substrate concentrations. Over the course\nof several additions of BHMF (1 mM to 10 mM), catalytic behavior was\nobserved for ACT, MT, T, and HT TEMPO derivatives, as indicated by\nthe increased forward current and decreased reverse current. At 10\nmM BHMF and 10 mV/s, the observed peak currents were 5.48 × 10 –2 mA for ACT, 2.81 × 10 –2 mA\nfor MT, 1.76 × 10 –2 mA for T, and 3.07 ×\n10 –2 mA for HT which represent a 12.3, 7.3, 4.1,\nand 5.0-fold increase in oxidative current over each catalyst peak\ncurrent, respectively. In contrast, several TEMPO catalysts show little\nto no catalytic activity with substrate, such as OT and AT. In the\nOT CVs, there is a slight increase in anodic peak current with increasing\nsubstrate contributions as well as in increase in the current of the\ndiffusional tail, but the irreversible voltammogram suggests that\ncatalyst turnover is limited due to the competitive chemical step\noutlined previously. Similarly, the AT CVs show no catalytic behavior\nwith hardly any changes in the voltammogram between substrate additions.\nThe current is low, and the catalyst seems to support little to no\ncatalysis of BHMF. Voltammetric substrate titrations were completed\nat a range of scan rates from 10 mV/s to 1000 mV/s. A selection of\nfaster scan CVs can be found in SI Figure S3 . Specifically, the faster scan rate of 1000 mV/s in Figure S3 shows there is not a significant increase\nin the current with substrate titrations, indicating that at higher\nscan rates, the scanning time scale is faster than the chemical step\nbetween the TEMPO oxoammonium cation and BHMF. Figure 4 Cyclic voltammograms\nof (A) acetamido-TEMPO (ACT, green traces),\n(B) methoxy-TEMPO (MT, orange traces), (C) TEMPO (T, yellow traces),\n(D) hydroxy-TEMPO (HT, blue traces), (E) oxo-TEMPO (OT, purple traces),\nand (F) amino-TEMPO (AT, red traces) in the presence of increasing\nBHMF concentrations. CVs obtained in a 0.5 M borate buffer solution\n(pH 9.2) containing 1 mM of varying TEMPO catalysts at scan rate 10\nmV/s. The results of these experiments correlate with\nthe predictions\nfrom the catalyst scan rate studies where reversible behavior was\npredicted to correspond to high turnover and catalysis ability in\nACT, MT, and T whereas irreversible catalyst behavior was predicted\nto limit substrate catalysis and performance in OT and AT. In the\nsubstrate titrations, HT also showed catalytic behavior which validates\nthe reversible behavior seen from HT at scan rates faster than 100\nmV/s. From the voltammograms in Figure 4 , it is clear that the catalysts that displayed reversible\nelectrochemical behavior and catalytic responses with substrate (ACT,\nMT, and T) will effectively catalyze the oxidation of BHMF to FDCA\nin bulk electrolysis experiments. Catalysts that displayed reversible\nbehavior over most scan rates and displayed catalytic responses with\nsubstrate (HT) are predicted to facilitate moderate oxidation of BHMF,\nperhaps to intermediate oxidized furan species. Conversely, OT and\nAT are hypothesized to not completely catalyze the BHMF oxidation\nreaction, as catalyst CVs demonstrate limited turnover capability\nand voltammetric substrate titrations show minimal to no catalytic\nbehavior when substrate is present. Bulk Electrolysis Voltammetry is a valuable technique\nto determine electrochemical reversibility as well as elucidate mechanism\nand kinetics. CV studies are quick, easy, and can provide valuable\ninformation about catalyst viability within a system before moving\non to exhaustive electrolysis. Most electrosynthesis studies investigate\nthe complete electrochemical conversion of substrate in solution through\nbulk electrolysis (chronopotentiometry and chronoamperometry) to exhaustively\noxidize or reduce the species in solution. In this study, controlled\npotential electrolysis (CPE) experiments were conducted in an H-cell\n(glass frit separator) with a graphite rod working electrode and saturated\ncalomel reference electrode (SCE) in the anode chamber, and a platinum\nmesh counter electrode in the cathode chamber. The mediated oxidation\nof BHMF occurs at the graphite rod anode while the platinum catalyzes\nhydrogen evolution reactions (HER) at the cathode. Experimental setup\npictures are shown in SI Figure S4 . In\ncontrolled potential coulometry, the working electrode is held at\na specific potential, E (120 mV more positive than\nthe oxidation potential of the mediator), and the system is continuously\nstirred so that the solution at the electrode interface is continuously\nrefreshed. As BHMF in solution is oxidized and consumed, the bulk\nconcentration is decreased and thus the oxidation current decreases\nexponentially. Eventually, the electrolysis finishes as the bulk concentration\nof BHMF is diminished and the current reaches background level. The\ntotal number of coulombs consumed in an electrolysis is used to determine\nthe amount of substance electrolyzed. The charge ( Q ) passed during the experiment is given by the area under the i–t curve and can be obtained by integrating the\ncurrent with respect to time ( eq 1 ). 1 The charge can then be converted to\nthe number of moles of species electrolyzed ( N ) using\nthe following eq 2 where n is the number of electrons and F is Faraday’s\nconstant (96 485 C/mol). 2 A representative i–t trace for the electrolysis\nof BHMF catalyzed by ACT is shown in Figure 5 where current is measured over time and\nthe total charge passed during electrolysis is represented by the\nshaded area under the curve. Each bulk electrolysis experiment was\nset up as follows. 10 mM BHMF substrate (5 mL, 5 × 10 –5 mol) was prepared in borate buffer (pH 9.2) and added to the anode\ncell. Plain borate buffer (5 mL) was added to the cathode cell. A\nconstant potential of +120 mV relative to the oxidation potential\nof each TEMPO catalyst was applied to the graphite working electrode\nin solution, and the current was allowed to baseline for ∼15\nmin. Applied potential values for each catalyst are listed in SI Table S2 . An initial spike in the current\nresponse is due to the charging of the electrode which decays to baseline\nquickly. At this point, the applied potential is not enough to directly\noxidize BHMF (without catalyst) so no reaction occurs. 1 mM TEMPO\ncatalyst is then injected into the anode chamber (100 μL from\na 50 mM stock solution) where an immediate current increase is observed\ndue to the increase in concentration of electroactive species in solution.\nAt this point, the TEMPO is able to continuously oxidize BHMF because\nthe electrode is poised at a potential positive enough to re-oxidize\nand “turnover” the TEMPO. An effective catalyst will\ncontinue to oxidize the substrate in solution until the substrate\nis completely consumed which is seen in the electrochemical response\nas a current decay. The electrolysis is complete when the current\nreturns to baseline and no more substrate is present to be oxidized. Figure 5 Bulk electrolysis i – t trace\nfor BHMF oxidation to FDCA, catalyzed by ACT. 10 mM BHMF, 1 mM ACT.\nBorate buffer pH 9.2. Bulk electrolysis experiments were conducted for\neach TEMPO catalyst\nin triplicate to assess how different TEMPO molecules are capable\nof oxidizing BHMF (individual bulk electrolysis traces for each catalyst\navailable in SI Figure S5 ). Representative i–t traces for each catalyst are shown in Figure 6 demonstrating the\nhow the differences in electrochemical reversibility affects electrolysis.\nCatalysts ACT, MT, and T which all displayed electrochemically reversible\nbehavior and catalytic responses with BHMF titrations also show similar i–t traces lasting approximately 17 h (60 000\nsec). The charge passed in these electrolyses fall within the same\nrange of 36.0, 35.6, and 35.7 C of charge consumed using ACT, MT,\nand T, respectively (theoretical amount of charge passed for 5 ×\n10 –5 mol BHMF was 38.6 C, see SI Equation S1 ). The electrolysis catalyzed by HT was also\nrun for ∼17 h but baselined much faster and only consumed 19.7\nC of charge. OT baselined within 3 h and only passed 2.2 C of charge,\nsupporting the hypothesis that this quasi-reversible catalyst that\ndid not promote catalytic activity with substrate present is not capable\nof completely oxidizing BHMF. Generally, shorter i–t trace times indicate that the catalyst does not turnover effectively\nand, as a result, only see partially oxidized products. For example,\nin bulk electrolysis experiments with irreversible catalysts (HT and\nOT), the current drops off rapidly, likely due to the complete use\nof the concentration of catalyst in solution. For these catalysts\nthat have little-to-no turnover, the substrate is only oxidized with\nthe amount of catalyst initially present which causes a quick reaction\nand a rapid decay of current. On the contrary, in experiments with\nreversible catalysts (ACT, MT, and T), the current decays slower as\ncatalyst turnover continuously facilitates the conversion of BHMF\nto FDCA. The bulk electrolysis trace using AT is slightly different\nfrom the other TEMPO catalysts. After the AT is introduced to the\nanode chamber, the current spikes and initially decays similar to\nthe other catalysts. However, the current starts to increase for 15 000\nseconds before it peaks and returns to the expected exponential decay\npattern and baselines after 42 h. This increase in current only occurs\nfor the AT catalyst and is likely due to local pH fluctuations at\nthe electrode surface upon addition with substrate. During this electrolysis,\n27.9 C of charge was passed. Figure 6 Bulk electrolysis traces for the oxidation of\n10 mM BHMF catalyzed\nby 1 mM of various TEMPO catalysts in borate buffer solution (pH 9.2).\n(A) acetamido-TEMPO (ACT, green traces), (B) methoxy-TEMPO (MT, orange\ntraces), (C) TEMPO (T, yellow traces), (D) hydroxy-TEMPO (HT, blue\ntraces), (E) oxo-TEMPO (OT, purple traces), and (F) amino-TEMPO (AT,\nred traces). Product Analysis The coulometry data obtained from\nthe bulk electrolyses was paired with ultra high-performance liquid\nchromatography-mass spectrometry (UHPLC-MS) product analysis data\nto identify the oxidation products and calculate Faradaic efficiency\nof the BHMF oxidation reaction. All products extracted from the reaction\nchamber were separated and detected via LC-MS. Product yields were\nquantified based on a FDCA calibration curve developed via diode array\ndetection (DAD) using commercial standards. Calibration data, linear\nfit equation, and LCMS information for BHMF and FDCA standards are\nfound in SI Figures S6–S12 and SI Tables S3–S5 . On the basis of the MS\nresults, ACT, MT, and T catalyzed the complete oxidation of BHMF to\nFDCA, as evidenced by the presence of a characteristic FDCA mass spectrum\nfrom experimental bulk electrolysis samples. The Faradaic efficiency\nof this oxidation was 102.78 ± 0.09% with ACT, 106.7 ± 0.1%\nwith MT, and 100.6 ± 0.1% with T. A sample calculation for Faradaic\nefficiency, including calculating the amount of moles electrolyzed\nduring electrolysis, can be found in SI Equations S2 and S3 . HT, OT, and AT all facilitated intermediate oxidation\nto a variety of semi-oxidized furan molecules with alcohol, aldehyde,\nand carboxylic acid functional groups ( Figure 7 ), but did not oxidize completely to FDCA.\nThese semi-oxidized furan molecules are HMF (hydroxymethylfurfural),\nDFF (2,5-diformylfuran), HMFCA (5-hydroxymethyl-2-furancarboxylic\nacid), and FFCA (5-formyl-2-furancarboxylic acid). In the eight-electron\noxidation of BHMF to FDCA, only a reversible catalyst is expected\nto facilitate the complete conversion while an irreversible catalyst\nwith low turnover is expected to facilitate the primary oxidations\nof the alcohol groups on BHMF, resulting in mixed, semi-oxidized products.\nA summary of the charge passed and Faradaic efficiency of each reaction\nis found in Table 1 and a complete description of intermediate oxidation products in\nHT, OT, and AT reactions with associated mass spectra can be found\nin SI Figures S13–S25 and Tables S6–S11 . Figure 7 Oxidized products of\nBHMF highlighting the desired product, FDCA. Table 1 Yields for the Electrocatalytic Oxidation\nof BHMF to FDCA Catalyzed by Different TEMPO Catalysts a catalyst charge passed (C) bulk\nelectrolysis yield (%) HPLC yield (%) Faradaic efficiency (%) acetamido-TEMPO 36.0 ± 0.1 93.3 ± 0.4 95 ± 2 102.78 ± 0.09 methoxy-TEMPO 35.6 ± 0.3 92.2 ± 0.9 98 ± 3 106.7 ± 0.1 TEMPO 35.7 ± 0.8 92 ± 2 93.1 ± 0.2 100.6 ± 0.1 hydroxy-TEMPO 19.7 ± 0.1 intermediate oxidation oxo-TEMPO 2.2 ± 0.1 intermediate oxidation amino-TEMPO 27.9 ± 0.2 intermediate oxidation a Included information: coulometry\ndata, bulk electrolysis and HPLC yields, and Faradaic Efficiency (standard\nerror included). The results from the LC-MS product analysis indicate\nthat the FDCA\nproduct is only detected from electrolyses that use ACT, MT, and T\nas catalysts. This outcome lines up precisely with the hypotheses\nmade after the initial scan rate studies where the electrochemical\nreversibility of the catalyst was predicted to correlate with its\neffectiveness in a mediated electrocatalytic reaction. Additionally,\nfurther studies could focus on characterizing the state of the catalysts\nbefore and after bulk electrolysis experiments, leading to a deeper\nunderstanding of catalyst turnover and degradation. These results\nemphasize the importance of voltammetry in electrocatalysis studies,\nespecially those using a mediator as a catalyst. One set of voltammograms\ntaken at a range of scan rates takes less than 30 min and an entire\nlibrary of catalysts can be screened in 1 day. An informed and rational\nselection of a catalyst can save time and improve reaction yields\nwith minimal upfront time commitment compared to running time-consuming\nbulk electrolysis experiments for catalyst selection and optimization. pH Dependent Electrochemical Properties of TEMPO TEMPO\ncatalyzed alcohol oxidation is a strongly pH dependent reaction, where\nsuperior catalytic activity of TEMPO and its derivatives is observed\nin basic conditions. In this reaction, TEMPO reacts with an alcohol\nsubstrate resulting in an overall 2-proton/2-electron oxidation (dehydrogenation)\nprocess where the activation and recycling of the catalyst is accomplished\nby oxidation at the electrode. The catalytic cycle for the oxidation\nof BHMF ( Figure 8 )\nfeatures three electrochemically accessible TEMPO oxidation states:\nnitroxyl radical (T • ), oxoammonium cation (T + ), and hydroxylamine (TH). The catalytically active oxoammonium\nspecies (T + ) is electrochemically generated by the one-electron\noxidation of the inactive T • . The alcohol substrate\n(in this case BHMF) reacts with the catalytically active T + species and forms a TEMPO-alcohol intermediate complex. In basic\nconditions, a homogeneous two electron oxidation of the alcohol generates\nthe corresponding aldehyde species and the reduced hydroxylamine TEMPO\n(TH). The active form of the catalyst (oxoammonium cation, T + ) is then electrochemically regenerated through a one proton coupled-two-electron\noxidation of TH. A full explanation of the catalytic cycle is depicted\nin Figure 8 (adapted\nfrom Hickey et al). 21 In acidic conditions,\nexcess protons promote the disproportionation of the nitroxyl radical\n(T • ) species to produce hydroxylamine (TH) and its\nprotonated hydroxylammonium form (TH 2 + ). 22 These disproportionation reactions are fast\nunder acidic conditions and ultimately lead to the inactivation of\nthe T + species. Thus, the presence of base is necessary\nto facilitate the activation and regeneration of TEMPO catalyst. For\nmore information, interested readers are directed toward studies by\nBailey et al. 23 and Stahl 24 , 25 investigating the mechanism of alcohol oxidation by oxoammonium\ncations. In this work, the redox-mediated oxidation of BHMF to FDCA\nwas conducted in mildly basic (pH 9.2) electrolyte solution, conditions\nthat have been shown to be well-suited for the electrochemical conversion\nof biomass-derived molecules such as HMF that are typically unstable\nin very acidic or basic environments. 14 Figure 8 Catalytic\ncycle of TEMPO catalyzing an alcohol oxidation of BHMF. An investigation into the pH dependent electrochemical\nproperties\nof TEMPO was conducted on all six TEMPO derivatives over a pH range\nof 3–10 to elucidate how changing solution pH affects catalyst\nreversibility and activity toward substrate oxidation. Voltammetry\ndata was collected on each TEMPO species, evaluating pH dependence\non scan rate and substrate additions. In the case of ACT, MT, and\nT, all three catalysts displayed electrochemically reversible voltammograms\nover the entire pH range (see SI Figure S31 ). The redox potentials of these species were also independent of\npH. Though the electrochemical reversibility and oxidation potential\nare independent of pH, the electrocatalytic activity of TEMPO toward\nBHMF alcohol oxidation is still dependent on pH. To demonstrate this,\na substrate addition of 1 mM BHMF was added to each catalyst solution\nimmediately following the scan rate study ( Figure 9 ). Cyclic voltammetry data taken with substrate\nin solution only shows catalytic behavior at pHs ≥ 8 for ACT,\nMT, and T. At solution conditions more acidic than pH 7, no catalytic\nbehavior is observed, despite the TEMPO displaying perfectly reversible\nbehavior (see SI Figures S26–S30 ). Thus, it is important to consider optimal solution conditions\nfor catalyst and substrate. Figure 9 CVs showing the electrochemical response of\n1 mM ACT (light green\ntrace) and comparing the electrocatalytic response of 1 mM BHMF +\n1 mM ACT (dark green trace) at pH 3 and pH 10. Scans taken at 10 mV/s\nin 0.5 M borate buffer. Catalysts HT, OT, and AT have been labeled quasi-\nor irreversible\nthus far in this study and display interesting pH dependence. HT,\nOT, and AT all display electrochemically reversible voltammograms\nbetween pH 3–7 but start to lose reversibility in basic conditions\naround pH 8 and completely lack a reverse peak by pH 10 (see SI Figure S31 ). Upon addition of substrate, no\ncatalytic behavior was observed in neutral or acidic conditions (<\npH 8) and minimal catalytic behavior was revealed in basic conditions\n(pH ≥ 8), as described in previous sections. The redox potential\nof both HT and OT remain independent of pH and do not shift across\nthe pH range. In contrast, the oxidation potential of AT does shift\nwith changes in pH. At pH 3-6 the oxidation potential is 0.71 V vs\nSCE. At pH 7, the potential shifts negative to 0.67 V vs SCE and continues\nto shift negative as pH increases. At pH 8, 9, and 10 the observed\noxidation potentials are 0.56, 0.49, and 0.45 V vs SCE, respectively.\nCyclic voltammograms of AT across the pH range are shown in Figure 10 , highlighting\nthe shifting redox potentials. This pH dependent potential shift is\nconsistent with literature reports that demonstrate the reversible\noxidation potential of amino-TEMPO as strongly pH dependent. 24 This investigation into amino-TEMPO by Stahl\nand co-workers found that the primary amine substituent is what directs\nthe pH sensitivity of amino-TEMPO’s electrochemical behavior.\nSpecifically, the amine protonation introduces an additional pKa at\neach TEMPO oxidation state in the catalytic cycle ( Figure 8 ) which drastically influences\nthe redox properties of the nitroxyl, therefore causing shifts in\nthe redox potential. 24 Stahl and co-workers\nalso observe that the AT oxidation is irreversible at high pH (consistent\nwith data from this study) which they propose is a result of the oxidation\nof the amino group. Figure 10 CVs comparing the electrochemical response of 1 mM AT\nover pH range\n3–10. Scans taken at 10 mV/s in 0.5 M borate buffer. Oxidation\npotentials are reported as a function of pH."
} | 10,239 |
20100071 | null | s2 | 7,132 | {
"abstract": "Hyperbolic discounting of future outcomes is widely observed to underlie choice behavior in animals. Additionally, recent studies (Kobayashi & Schultz, 2008) have reported that hyperbolic discounting is observed even in neural systems underlying choice. However, the most prevalent models of temporal discounting, such as temporal difference learning, assume that future outcomes are discounted exponentially. Exponential discounting has been preferred largely because it can be expressed recursively, whereas hyperbolic discounting has heretofore been thought not to have a recursive definition. In this letter, we define a learning algorithm, hyperbolically discounted temporal difference (HDTD) learning, which constitutes a recursive formulation of the hyperbolic model."
} | 193 |
19219458 | PMC2757589 | pmc | 7,135 | {
"abstract": "The majority of plants are involved in symbioses with arbuscular mycorrhizal fungi (AMF), and these associations are known to have a strong influence on the performance of both plants and insect herbivores. Little is known about the impact of AMF on complex trophic chains, although such effects are conceivable. In a greenhouse study we examined the effects of two AMF species, Glomus intraradices and G. mosseae on trophic interactions between the grass Phleum pratense , the aphid Rhopalosiphum padi , and the parasitic wasp Aphidius rhopalosiphi . Inoculation with AMF in our study system generally enhanced plant biomass (+5.2%) and decreased aphid population growth (−47%), but there were no fungal species-specific effects. When plants were infested with G. intraradices , the rate of parasitism in aphids increased by 140% relative to the G. mosseae and control treatment. When plants were associated with AMF, the developmental time of the parasitoids decreased by 4.3% and weight at eclosion increased by 23.8%. There were no clear effects of AMF on the concentration of nitrogen and phosphorus in plant foliage. Our study demonstrates that the effects of AMF go beyond a simple amelioration of the plants’ nutritional status and involve rather more complex species-specific cascading effects of AMF in the food chain that have a strong impact not only on the performance of plants but also on higher trophic levels, such as herbivores and parasitoids.",
"conclusion": "Conclusion Our results show that three interacting trophic levels are significantly affected by both the presence and the species identity of AMF. Bottom–up effects of AMF influenced plants, aphids, and their parasitoids differently, with a positive impact on plants and parasitoids and a negative impact on aphids. However, changes in plant nutrient contents C, N, and P) were not driving the observed performance alterations, as these values were equal between mycorrhizal and non-mycorrhizal plants. Therefore, food choice experiments (Prince et al. 2004 ) and stable isotope probing (Langellotto et al. 2006 ) would be useful approaches for monitoring changes in preferences and nutrient fluxes. The observed changes in the trophic interactions due to AMF inoculation emphasize that belowground interactions can have strong implications for aboveground food webs (van der Putten et al. 2001 ). Models of trophic interactions (Hoover and Newman 2004 ; van der Putten et al. 2004 ) should include the impact of a symbiosis as widespread as arbuscular mycorrhiza (Treseder and Cross 2006 ).",
"introduction": "Introduction Multitrophic interactions between above- and belowground organisms are powerful forces shaping the structure and diversity of natural communities (van der Putten et al. 2001 ). For example, belowground herbivores can influence aboveground herbivores via a shared host plant and vice versa (van Dam et al. 2003 ; Wurst and van der Putten 2007 ). One interaction that has been found to affect the performance of both above- and belowground organisms is the symbiosis between plants and arbuscular mycorrhizal fungi (AMF; Bennett et al. 2006 ; Bezemer and van Dam 2005 ; Gehring et al. 2002 ). The infection of plants by AMF affects the interactions of the former with root pathogenic fungi (Newsham et al. 1995 ), Collembola (Gange 2000 ), saprotrophic fungi (Tiunov and Scheu 2005 ), above- and belowground herbivores (Gange 2001 ; Goverde et al. 2000 ), and parasitic plants (Stein et al. 2009 ). Aphids, as one guild of herbivores directly feeding on plant phloem, can be influenced by AMF colonizing the roots of their host plants (e.g., Gange et al. 1999 ; Guerrieri et al. 2004 ; Wurst et al. 2004 ), but the direction of the effects have varied between different experiments. While Gange and West ( 1994 ) and Gange et al. ( 1999 ) found a positive influence of AMF on weight and fecundity of two Myzus species reared on Plantago lanceolata , negative AMF effects were reported with Chaitophorus populicola reared on Populus angustifolia × P. fremontii (Gehring and Whitham 2002 ) and Macrosiphum euphorbiae reared on Lycopersicon esculentum (Guerrieri et al. 2004 ). One possible explanation for this inconsistency in results may be the variability of arbuscular mycorrhizal symbiosis itself, which ranges from mutualism to parasitism depending on various biotic and abiotic factors (Johnson 1993 ; Klironomos 2003 ). In addition, infection by different AMF species can have different effects on several plant traits, such as biomass or nutrient capture (van der Heijden et al. 1998 ). There are also indications that AMF infection of plants can have cascading effects in the food chain up to higher trophic levels (Gange et al. 2003 ). For example, there is evidence that AMF symbioses with plants can affect both the rate of aphid parasitism by parasitoid wasps (Gange et al. 2003 ) and parasitoid preference, where aphid infested non-mycorrhizal plants are as attractive to parasitoid wasps as uninfested mycorrhizal plants (Guerrieri et al. 2004 ). However, both of these studies did not directly assess parasitoid performance, although it is likely that the strong effects of AMF reported on primary producers (plants) and primary consumers (herbivores) cascade upwards in the food chain and thus also affect several traits in predator or parasitoid performance, such as food consumption or reproductive output (Bezemer et al. 2005 ). The objective of this study was to test AMF species effects on the tri-trophic interaction of a typical grassland plant species ( Phleum pratense ), its insect herbivore Rhopalosiphum padi L., and the parasitoid Aphidius rhopalosiphi. In a greenhouse experiment, the grass was inoculated with either one of the two AMF species, Glomus intraradices or G. mosseae , and the results compared to a non-mycorrhizal control. These three treatments were combined with three insect treatments: (1) plants only (no insects), (2) plant + aphid, and (3) plant + aphid + parasitoid. We hypothesized that: the association with AMF improves plant biomass and nutrient capture; there is an increase in food quality which benefits aphid reproduction and supports larger aphid populations on mycorrhizal plants; larger aphid populations allow female parasitoids to choose more suitable aphids for parasitization, which leads to an increase in parasitoid weight and a decrease in parasitoid development time; and that the two AMF species have different effects on the tri-trophic interaction.",
"discussion": "Discussion In contrast to our initial hypotheses, positive effects of AMF inoculation on performance could only be observed at two trophic levels. One is the level of primary producers ( Phleum pratense ), which benefitted from the association in terms of an increase in biomass, and the other level is that of the parasitoids ( Aphidius rhopalosiphi ), which showed increased weight at eclosion and shorter developmental time on mycorrhizal plants (Figs. 1 , 2 ). Population growth rates of aphids ( R. padi ) as primary consumers decreased on plants inoculated with AMF. Although we did not detect significant differences between the two inoculated AMF species concerning aphid population growth rates, such differences were clearly present in terms of plant biomass under aphid attack and in the rates of parasitism. These differences may reflect direct physiological effects of the two AMF species, but they may also result from the observed differences in mean colonization rate between G. intraradices and G. mosseae (42 and 21%, respectively). The different mycorrhization rates, in turn, may be an artefact of our experiment, but we suggest that they instead reflect innate differences between AMF species, as shown by Hart and Reader ( 2002 ), because we used the same amount of inoculum for both AMF species. Nevertheless, the results of our experiment do not allow the indirect effects of different colonization rates to be disentangled from the direct physiological effects of AMF species, and we also have to consider with caution any conclusion regarding species-specific effects on higher trophic levels. It has to be considered that the AMF inoculum (a commercial cultivar) and plant seeds used in this study share no common ecological background and conceivably are not adapted to each other (Fitter et al. 2005 ). Klironomos ( 2003 ) showed that the combination of non-adapted AMF and plants can narrow the range of host plant responses. Yet, our study demonstrates potential effects of different AMF species on plants and higher trophic levels, rather than revealing the actual outcome of these interactions under natural conditions. However, a positive impact of the two AMF isolates on plant biomass was present. Highly adapted AMF can be expected to provide even more benefits to their host plants (Helgason et al. 2007 ); as such, comparable or even stronger effects on higher trophic levels may be expected under more natural conditions. Arbuscular mycorrhizal fungi and aphid effects on plants The positive effect of AMF inoculation on plant biomass was also present at the two interim clippings (data not shown). In the case of G. mosseae , this positive effect was clearly not reflected in foliar N and P contents (Table 2 ), as these values tended to be lower than in the control. Such species-specific effects of AMF on several plant variables are also in accordance with previous studies (see Jansa et al. 2008 ; Maherali and Klironomos 2007 ; van der Heijden et al. 1998 ), which showed that biomass and nutrient capture of a plant community varied independently with the identity of the inoculated AMF species. Although aphid presence had a consistently negative impact on shoot biomass, this reduction was only significant in the control and G. intraradices -inoculated plants, indicating a higher tolerance to aphid feeding in the G. mosseae -inoculated plants (Fig. 1 a). In contrast, an inconsistent pattern was observed in root biomass. Plants inoculated with G. mosseae increased their root biomass under aphid presence, whereas lowered biomass was detected in control and G. intraradices -inoculated plants under aphid herbivory (Fig. 1 b). Such interactive effects of AMF species and aphids have been reported previously (Gange and West 1994 ) and may reflect differences in nutrient allocation within plants under aphid attack (Vestergård et al. 2004 ). AMF effects on aphids The negative effect of AMF inoculation on aphid population growth rates found in our experiment (Fig. 1 c, d) contradicts some results of other studies on AMF-aphid interactions (e.g., Gange et al. 1999 ; Gange and West 1994 ). However, negative interactions have also been reported by other authors (Gehring and Whitham 2002 ; Guerrieri et al. 2004 ; Wurst et al. 2004 ). In most of the respective publications, AMF effects on aphids were measured using the reproductive fitness of individual females (e.g. Gange et al. 1999 ; Wurst et al. 2004 ). In our study, however, we measured aphid response to AMF in terms of population growth rather than in terms of individual fitness. Although individual fitness and population growth rate may be correlated (Ponder et al. 2000 ), we would like to emphasize that we can draw conclusions only for AMF effects on the population growth of aphids. Three potential mechanisms limiting aphid growth in our experiment must be considered: aphid crowding, nutrient limitation, and plant defense compounds. Winged morphs, a good indicator of aphid crowding (Hodgson 2001 ), were rarely detected (in two pots only). Nutrient limitation is also unlikely, as aphid population growth rates (Fig. 1 c) and plant nutrient contents at harvest (Table 2 ) show no correlation. Inoculation with G. intraradices induced the highest decrease in aphid population growth rates, but the respective plants contained as much N as control plants and tended to contain even more P than plants of the two other fungal treatments. This lack of correlation between N contents and aphid performance is in accordance with a field study by Gange and West ( 1994 ), who hypothesized that changes in aphid numbers were more related to a changed leaf morphology (phloem location and size) in mycorrhizal plants than to differences in N content. In contrast, previous studies (Bezemer et al. 2005 ; Ponder et al. 2000 ) reported decreased aphid population sizes in parallel with decreased foliar N concentrations using the same aphid species ( R. padi ) as was used in our study. Another explanation for the high proportion of aphid populations with negative growth rates on mycorrhizal plants might be the presence of defence compounds in the phloem of Phleum pratense , indicating increased plant resistance against aphids induced by AMF (Pozo and Azcón-Aguilar 2007 ). Bezemer et al. ( 2005 ) have recently shown that R. padi might be sensitive to phenolic compounds encountered in the phloem, which may be synthesized at higher rate upon mycorrhizal inoculation (Zhu and Yao 2004 ). However, elicitation of defence compounds by repeated cutting of Phleum pratense is not likely, as plant response mechanisms triggered by wounding and by phloem feeding insects (i.e. aphids) follow different signaling pathways (Pozo and Azcón-Aguilar 2007 ). The missing effect of parasitoid presence on aphid numbers is not surprising as the proportion of parasitized aphids was generally low (<16%) and aphid populations encountered parasitoids only once during a 12-h period. AMF effects on parasitoids First studies concerning the effects of AMF on parasitoid wasps focused on parasitoid preferences. Gange et al. ( 2003 ) provided data on AMF species-dependent variations in the rates of parasitism of the ichneumon Diglyphus isaea parasitizing the leaf mining fly Chromatomyia syngenesiae . Guerrieri et al. ( 2004 ) showed that non-mycorrhizal tomato plants infested with aphids were as attractive to Aphidius ervi as mycorrhizal, non-infested plants. In our study, we revealed that the performance of parasitoid wasps is also influenced by the presence and identity of AMF, as Aphidius rhopalosiphi got heavier and developed faster when their host R. padi was reared on mycorrhizal plants (Fig. 2 ). Changes in weight at eclosion and developmental time are highly correlated to several fitness traits, such as longevity, number of hosts attacked (in case of females), and number of matings achieved (in case of males; Godfray 1994 ). These changes were rather uniform for male and female parasitoids in the case of the G. intraradices treatment, but they varied substantially between sexes in the G. mosseae treatment. While we cannot discern the underlying reasons for the sex-specific difference, one possibility is that females adjusted their behavior when plants were infested by G. mosseae such that fertilized eggs resulting in females were laid in hosts that differed in size from those of unfertilized eggs. Alternatively, the observed sex-specific pattern in the G. mosseae treatment may indicate that larger female parasitoids enjoy a proportionally greater increase in fitness than larger males (Godfray 1994 ). Therefore, female parasitoids may have invested additional resources in an increased weight rather than decreased development time, which is also reflected by the significant influence of the ratio between sexes within pots on parasitoid dry weight (Table 4 ). Despite this sex-specific difference in the G. mosseae treatment, the observed differences in parasitoid dry weight and development time between mycorrhizal and control treatments were not due to changes in sex ratio, as this variable was relatively constant in all AMF treatments (proportion of females 44.8 ± 7.1%). Alternatively, it might be the case that male and female parasitoids can use the resources provided by G. mosseae infection in different ways. This hypothesis needs to be addressed in further studies. Parasitoid developmental time was correlated with aphid density, i.e., parasitoids developed faster when more aphids were available for oviposition. One possible explanation for this relationship is that at higher aphid densities, the parasitoids encounter more aphids of different larval instars and hence are able to carry out more ovipositions in more suitable aphid stages. In the pea aphid ( Acyrthosiphon pisum ), oviposition in intermediate instars reduced the developmental time of Aphidius ervi relative to ovipositions in younger instars or adults (Sequeira and Mackauer 1992 ). While we did not directly study parasitoid oviposition, such selection behavior is conceivable. More generally, there is little information about density dependence in parasitoid host selection behavior and the consequences for offspring fitness. Sequeira and Mackauer ( 1992 ) showed that the values of weight and developmental time covary and are furthermore highly dependent on the age of the parasitized aphids. This association was not present in our study, as the proportion of aphids that died as adults was the same in the two AMF inoculation treatments and the control, indicating that aphids were parasitized at comparable larval stages in all fungal treatments. In accordance to the study by Gange et al. ( 2003 ), we found changes in the rates of parasitism, expressed as a significant higher proportion of parasitized aphids in the G. intraradices treatment (Fig. 3 ). Gange et al. ( 2003 ) partially attributed their observed mycorrhizal effects on the rates of parasitism to a decreased parasitoid searching efficiency due to changes in plant architecture. However, the limited space under the cellophane bags in our experiment surely interfered with this effect. Additional effects, such as the induction of volatiles influencing parasitoid activity, are also likely, as these can be AMF species-specific (Bezemer and van Dam 2005 ). Interactions of belowground organisms with plant roots resulting in contrasting reactions on aboveground aphids and parasitoids were also reported by Bezemer et al. ( 2005 ). They attributed increasing parasitoid performance to a visually observed increase in aphid size, although they did not explicitly quantify this parameter. Another possible explanation for the observed effects on aphid and parasitoid level would be a decrease in the aphid immune answer against parasitoid eggs on mycorrhizal plants, which could have led to an increase in parasitoid performance (W. Völkl, personal communication; see also Godfray 1994 ). However, all of these hypothetical mechanisms do not seem to follow a linear relation, as aphid population growth rates were highest on control plants, intermediate on plants from the G. mosseae treatment, and worst on G. intraradices -inoculated plants. In contrast, parasitoid weight and development time were best on G. mosseae -inoculated plants, worst on control plants, and intermediate with G. intraradices . Conclusion Our results show that three interacting trophic levels are significantly affected by both the presence and the species identity of AMF. Bottom–up effects of AMF influenced plants, aphids, and their parasitoids differently, with a positive impact on plants and parasitoids and a negative impact on aphids. However, changes in plant nutrient contents C, N, and P) were not driving the observed performance alterations, as these values were equal between mycorrhizal and non-mycorrhizal plants. Therefore, food choice experiments (Prince et al. 2004 ) and stable isotope probing (Langellotto et al. 2006 ) would be useful approaches for monitoring changes in preferences and nutrient fluxes. The observed changes in the trophic interactions due to AMF inoculation emphasize that belowground interactions can have strong implications for aboveground food webs (van der Putten et al. 2001 ). Models of trophic interactions (Hoover and Newman 2004 ; van der Putten et al. 2004 ) should include the impact of a symbiosis as widespread as arbuscular mycorrhiza (Treseder and Cross 2006 )."
} | 5,033 |
28401691 | PMC5658586 | pmc | 7,137 | {
"abstract": "Biocatalysis can enable a closed‐loop recycling of post‐consumer PET waste.",
"conclusion": "Concluding remarks Microbial polyester hydrolases have shown their potential in the biocatalytic depolymerization of PET. For an efficient degradation of postconsumer PET plastic waste in an industrial process, the performance of the enzymes still requires substantial improvements. The discovery and engineering of novel polyester hydrolases exhibiting specific catalytic properties towards high crystalline and bottle‐grade PET materials therefore remain key challenges. This can be achieved by applying microbial biotechnology methodologies for identifying novel enzymes from the environment exploiting microbial biodiversity and by generating powerful variants by protein engineering using rational design and enzyme evolution strategies. If successful, these enzymes can make an important contribution towards a future sustainable closed‐loop plastic recycling industry.",
"introduction": "Introduction The global production of fossil‐based plastics has grown more than 20‐fold since 1964 to 322 million tons in 2015, and a slowdown of this rate is not expected (Ellen MacArthur Foundation and World Economic Forum, 2014 ; PlasticsEurope, 2016 ). Many materials derived from synthetic polymers have already replaced their natural counterparts in all areas of human life. The majority of plastics are short‐lived products which are disposed within 1 year after manufacture. However, only 14% of plastic packaging materials used worldwide is currently collected for recycling while another 14% is incinerated for energy recovery (Ellen MacArthur Foundation and World Economic Forum, 2014 ). The remaining 72% of plastic packaging is not recovered with 40% land filled and 32% estimated to completely escape the collection system. This part of plastic waste ends up in diverse natural habitats, especially in oceans where it can cause serious environmental damages (Andrady, 2015 ; Jambeck et al ., 2015 ). Therefore, innovative technologies to improve the recycling of plastics and to reduce the consumption of non‐renewable fossil feed stocks are required. Polyethylene terephthalate (PET) is the most widely used synthetic polyester. It is a thermoplastic of high‐molecular‐weight composed of terephthalic acid (TPA) and ethylene glycol (EG). PET can exist as both an amorphous and a semi‐crystalline polymer (Webb et al ., 2013 ). Owing to its excellent physical and chemical properties, PET finds numerous applications as textile fibres, packaging materials and beverage bottles. PET is generally referred to as ′polyester′ in the textile industry which consumes the majority of the PET produced globally. In 2014, 49.2 million tons of PET fibres was produced worldwide (Fiber Economics Bureau, 2015 ). In 2015, the global production of PET resins was 27.8 million tons which was dominantly used for the manufacture of packaging materials and beverage bottles (Plastic Insight, 2016 ). Almost half of the postconsumer PET bottles worldwide are collected for mechanical recycling to produce polyester fibres (Ellen MacArthur Foundation and World Economic Forum, 2014 ). Polyethylene terephthalate made from renewable biomass (bio‐PET) is becoming of industrial interest lately. EG derived from sugarcane ethanol (Tsiropoulos et al ., 2015 ) and TPA derived from sugar beet paraxylene (Collias et al ., 2014 ; Smith, 2015 ) can be utilized to replace their fossil‐based counterparts to produce bottle‐grade PET. Although fossil feedstocks can be saved by the commercialization of bio‐PET bottles, the challenges for their recycling remain as their recalcitrant properties are the same as those of petroleum‐derived PET bottles (Chen et al ., 2016 )."
} | 930 |
36577119 | null | s2 | 7,138 | {
"abstract": "The DNA-origami technique has enabled the engineering of transmembrane nanopores with programmable size and functionality, showing promise in building biosensors and synthetic cells. However, it remains challenging to build large (>10 nm), functionalizable nanopores that spontaneously perforate lipid membranes. Here, we take advantage of pneumolysin (PLY), a bacterial toxin that potently forms wide ring-like channels on cell membranes, to construct hybrid DNA-protein nanopores. This PLY-DNA-origami complex, in which a DNA-origami ring corrals up to 48 copies of PLY, targets the cholesterol-rich membranes of liposomes and red blood cells, readily forming uniformly sized pores with an average inner diameter of ∼22 nm. Such hybrid nanopores facilitate the exchange of macromolecules between perforated liposomes and their environment, with the exchange rate negatively correlating with the macromolecule size (diameters of gyration: 8-22 nm). Additionally, the DNA ring can be decorated with intrinsically disordered nucleoporins to further restrict the diffusion of traversing molecules, highlighting the programmability of the hybrid nanopores. PLY-DNA pores provide an enabling biophysical tool for studying the cross-membrane translocation of ultralarge molecules and open new opportunities for analytical chemistry, synthetic biology, and nanomedicine."
} | 341 |
24311557 | PMC3892345 | pmc | 7,140 | {
"abstract": "The rhizosphere microbial community in a hydroponics system with multiple parallel mineralization (MPM) can potentially suppress root-borne diseases. This study focused on revealing the biological nature of the suppression against Fusarium wilt disease, which is caused by the fungus Fusarium oxysporum , and describing the factors that may influence the fungal pathogen in the MPM system. We demonstrated that the rhizosphere microbiota that developed in the MPM system could suppress Fusarium wilt disease under in vitro and greenhouse conditions. The microbiological characteristics of the MPM system were able to control the population dynamics of F. oxysporum , but did not eradicate the fungal pathogen. The roles of the microbiological agents underlying the disease suppression and the magnitude of the disease suppression in the MPM system appear to depend on the microbial density. F. oxysporum that survived in the MPM system formed chlamydospores when exposed to the rhizosphere microbiota. These results suggest that the microbiota suppresses proliferation of F. oxysporum by controlling the pathogen's morphogenesis and by developing an ecosystem that permits coexistence with F. oxysporum .",
"introduction": "Introduction A soil-free cultivation system is generally believed to require the elimination of microbial and organic contaminants from the nutrient solution, whether the system uses open or closed circulation (Garland and Mackoqiak 1990 ; Stanghellini and Rasmussen 1994 ; Stanghellini et al. 1996 ; Garland et al. 1997 ; Koohakana et al. 2004 ; Ehret et al. 2005 ; Lee et al. 2006 ). In contrast, multiple parallel mineralization (MPM) is a novel form of hydroponics system in which microbial activity and the presence of organic nutrients in the solution lead to the development of plant–microbe ecosystems in the rhizosphere (Shinohara et al. 2011 ). The rhizosphere microbiota that develops in an MPM system is responsible for two reproducible functions that are necessary for sustainable plant growth: high nutrient production efficiency and the control of root diseases. The former trait results from microbial mineralization of the organic fertilizer used in the hydroponics solution, which promotes the transformation of organic nitrogen into nitrate nitrogen in the hydroponics solution as a result of two sequential microbial processes: ammonification and nitrification. To achieve this process, it is necessary to culture soil microorganisms in the hydroponics solution and develop a microbial community that is capable of mineralizing organic fertilizer into nitrate ions. This approach recreates a microbial environment that promotes coexistence of the plants and microbes in a hydroponics system in a manner similar to that which occurs in soils (Shinohara et al. 2011 ). The second trait means that the MPM system has the potential to control root-borne diseases as a result of the actions of the microbiota community that develops in the rhizosphere. We have previously explored the ability of this system to suppress the bacterial wilt disease caused by Ralstonia solanacearum , a plurivorous phytopathogenic bacterium (Fujiwara et al. 2012 ). Our examination of this suppression demonstrated that the MPM system could suppress this bacterial wilt disease. We have also observed that other root-borne diseases that often damage several plant species in inorganic hydroponics systems, including lettuce, komatsuna, rice, cucumber, and pepper, did not occur in the MPM system, allowing cultivation of the plants without requiring fungicides or other antibiotics before they could be harvested (Shinohara 2006 ; Shinohara et al. 2011 ; M. Shinohara and K. Fujiwara, unpubl. data). These results revealed the importance of the microbiota that colonizes the plant rhizosphere in the suppression of root-borne diseases. Our interest in the suppression of root diseases obtained in an MPM system led us to explore the biological nature of the disease suppression and describe the possible factors influencing root-borne pathogens in this system. Previously, the MPM system showed an ability to suppress bacterial wilt (Fujiwara et al. 2012 ), but the potential for controlling fungal root diseases had not yet been examined. In this study, we used the fungal root pathogen Fusarium oxysporum , which is an important phytopathogenic fungus for many crops (Di Pietro et al. 2003 ). We focused on investigating the microbiological factors that underlie the suppression of F. oxysporum f. sp. lactucae , which causes root rot and wilting of commercial lettuce plants and has become a significant problem in Japan (Fujinaga et al. 2003 , 2005 ). Identifying the factors responsible for the success of this approach will improve our understanding of microbial contributions to the suppression of root-borne diseases and provide insights into previously unknown biological phenomena that occur in the plant rhizosphere.",
"discussion": "Discussion A major objective of this study was to evaluate the suppression of fungal disease by MPM system. Our results demonstrated that the community of rhizosphere microbiota that developed in the MPM system could suppress F. oxysporum in both in vitro studies and in experiments performed under greenhouse conditions. Previous studies provided some evidence that the production of antimicrobial substances, control of fungal morphogenesis, and induction of host systemic resistance play key roles in soils that are capable of suppressing Fusarium wilt (Scher and Baker 1980 ; Alabouvette 1986 ; Weller et al. 2002 ; Haas and Défago 2005 ). In this study, our DGGE profiles revealed that the microbiota, including important bacteria in Bacillus spp., was present in the rhizosphere of the MPM system. Members of genus Bacillus are known to be potential biocontrol agents for F. oxysporum (Weller et al. 2002 ) and produce a wide range of antibiotics from different functional classes (Raaijmakers et al. 2002 ). We further tested the biocontrol abilities against F. oxysporum under in vitro conditions using several treatments and demonstrated that microbial survivability was strongly associated with the biocontrol efficacy, suggesting that antimicrobial substances that are diffusible, degradable, or sensitive to heat and concentration were unlikely to be the primary cause of disease suppression. We are currently investigating possible antimicrobial agents that include volatile organic compounds (Minerdi et al. 2009 ). Control of F. oxysporum morphogenesis by rhizosphere microbes is part of the mechanism responsible for disease suppression by the MPM system. Microbial interactions with F. oxysporum in the microenvironment stimulated chlamydospore formation and inhibited reproduction of F. oxysporum . The chlamydospores of F. oxysporum can form under unfavorable environmental conditions, such as at low temperature, and these resting bodies act as primary inocula in subsequent soil-borne infections, whereas the microconidia and macroconidia formed by F. oxysporum are important in secondary infection (Nelson 1981 ; Couteaudier and Alabouvette 1990 ). In this study, F. oxysporum survived in the MPM system by forming chlamydospores when they were exposed to the rhizosphere microbiota, even though there were few chlamydospores 3 days after inoculation ( Fig. S5 ). The surviving fungal cells did not cause Fusarium wilt disease in the MPM system, but retained their pathogenicity when they were reisolated from the solution. The F. oxysporum cells in the MPM solution had a higher rate of microconidia germination, but a lower hyphal elongation rate than those in the inorganic hydroponics solution. In this species, hyphal elongation is necessary for the production of microconidia and macroconidia (Nelson 1981 ). Thus, the decreased hyphal elongation in the MPM solution may prevent an increase in the numbers of macroconidia and microconidia, resulting in an unchanged F. oxysporum cell density. Ford et al. ( 1970 ) accounted for microbial stimuli of Fusarium solani activity and demonstrated that the inclusion of soil microbes in the solution induced chlamydospore formation. Some bacteria that are known to be biocontrol agents adhere to the hyphae of F. oxysporum (Elvers et al. 2001 ) and alter its hyphal morphology (Bolwerk et al. 2003 ). Minerdi et al. ( 2008 ) reported that microbial symbionts silence the virulence of F. oxysporum and that changes in cell morphogenesis by F. oxysporum underlie this suppression. These findings suggest that the rhizosphere microbial consortia that become established in the MPM system are unsuitable for the growth of F. oxysporum and negatively affect its pathogenic ability as a result of changes in its morphogenesis. Bacterial interactions with plant roots can lead to plant resistance to F. oxysporum . Induced systemic resistance (ISR) against this pathogen can be elicited by some strains in the genera Bacillus , Pseudomonas , Serratia , and Achromobacter (Van Peer et al. 1991 ; Someya et al. 2000 ; Lugtenberg and Kamilova 2009 ). In this study, we demonstrated that one or more potential biocontrol agents in genus Bacillus existed in the MPM system. Although ISR may be one of the causal factors for the suppression of root-borne diseases in the MPM system, there is conflicting evidence as to whether ISR plays a major role in disease suppression. Our results indicated that tomato plants grown in the MPM system were still susceptible to diseases of the aerial parts of the plant, including airborne pathogens such as powdery mildew ( Oidium neolycopersici ), leaf mold ( Cladosporium fulvum ), or gray mold ( Botryotinia fuckeliana ) during cultivation in the greenhouse (M. Shinohara and K. Fujiwara, unpubl. data). Because F. oxysporum exhibited morphological changes that allowed it to survive in the rhizosphere microbiota, but with reduced infectious ability, the disease suppression we observed in the MPM system is most likely to result from microbiological interactions rather than from ISR. Fusarium wilt disease suppression in the MPM system can be ascribed to microbiota that controls the morphological changes in F. oxysporum . However, little is known about the development of disease suppression in the MPM system. In this study, the development of functional abilities of the MPM solution capable of influencing Fusarium wilt disease suppression and altering fungal morphogenesis was not found when the pathogen was inoculated within 4 days after transplanting. We hypothesize that the absence of biocontrol abilities during this early stage results from incomplete biofilm formation in the rhizosphere that prevents the microbiota from conferring their biocontrol ability due to insufficient microbe–microbe or plant–microbe interactions. Our previous study provided evidence that a thick biofilm begins to form about 1 day after transplanting, followed by the development of a thinner biofilm as a result of changes in the rhizosphere structures involving the roots, and that stable biofilm formation occurs around 4 days after transplanting (Fujiwara et al. 2012 ). The present results agree with that previous finding. Because it is well known that biofilm formation does not simply cause the development of disease suppression, we further predict that certain as-yet-unidentified causal factors influence the stability of the rhizosphere microbiota and their biocontrol abilities. Past investigations suggest that microbial communication physically and functionally contributes to the establishment of microbiota that is associated with biocontrol activities (Raaijmakers and Mazzola 2012 ; Chen et al. 2013 ). In future research, we will need to elucidate the causal factors responsible for the development of biocontrol effects that lead to the suppression of root-borne diseases."
} | 2,977 |
37237552 | PMC10215164 | pmc | 7,141 | {
"abstract": "Simple Summary Division of labour is a crucial characteristic of social organisations such as insect colonies and is a key feature in their well-known survival and efficacy. The presence of “laziness”, or inactivity is a widely debated phenomenon that has been observed in some colonies and is puzzling because it goes against the idea that a division of labour would lead to greater efficiency and effectiveness. Inactivity has been previously explained as a by-product of social learning, which is a fundamental type of behavioural adaptation in these colonies. However, this explanation is limited because it is still unclear if social learning governs aspects of colony life. This study explores how inactivity can also emerge similarly from an individual learning paradigm, which is a firmly established paradigm of behaviour learning in insect colonies. Using individual-based simulations backed up by mathematical analysis, the study finds that individual learning can induce the same behavioural patterns as social learning. This is important for understanding the collective behaviour of social insects. The insight that both modes of learning can lead to the same patterns of behaviour opens up new ways of approaching the study of emergent patterns of collective behaviour in a more generalised manner. Abstract Division of labour, or the differentiation of the individuals in a collective across tasks, is a fundamental aspect of social organisations, such as social insect colonies. It allows for efficient resource use and improves the chances of survival for the entire collective. The emergence of large inactive groups of individuals in insect colonies sometimes referred to as laziness , has been a puzzling and hotly debated division-of-labour phenomenon in recent years that is counter to the intuitive notion of effectiveness. It has previously been shown that inactivity can be explained as a by-product of social learning without the need to invoke an adaptive function. While highlighting an interesting and important possibility, this explanation is limited because it is not yet clear whether the relevant aspects of colony life are governed by social learning. In this paper, we explore the two fundamental types of behavioural adaptation that can lead to a division of labour, individual learning and social learning . We find that inactivity can just as well emerge from individual learning alone. We compare the behavioural dynamics in various environmental settings under the social and individual learning assumptions, respectively. We present individual-based simulations backed up by analytic theory, focusing on adaptive dynamics for the social paradigm and cross-learning for the individual paradigm. We find that individual learning can induce the same behavioural patterns previously observed for social learning. This is important for the study of the collective behaviour of social insects because individual learning is a firmly established paradigm of behaviour learning in their colonies. Beyond the study of inactivity, in particular, the insight that both modes of learning can lead to the same patterns of behaviour opens new pathways to approach the study of emergent patterns of collective behaviour from a more generalised perspective.",
"introduction": "1. Introduction Division of labour is fundamental to the functioning of social organisms and has been central to their study for decades [ 1 ]. The separation of tasks among different individuals or groups within a collective allows for the efficient use of resources and increases the chances of survival for the collective as a whole [ 2 , 3 , 4 ]. Studies have shown that division of labour is prevalent in many socially living organisms, such as ants, bees, termites, and even some mammals [ 5 , 6 , 7 , 8 , 9 ]. Social insect colonies are well known for their intricate organisation and their ability to handle a wide range of tasks simultaneously, including foraging, colony defence, nest construction, temperature regulation, and caring for offspring [ 6 , 10 ]. The colony’s ability to effectively allocate its workforce to these different tasks, adapting to changes in both external conditions and internal needs, is often cited as a key to their ecological success [ 11 , 12 , 13 , 14 ]. Understanding the underlying mechanisms of division of labour is fundamental to understanding these social organisations and the emergence of complex social systems in general. It is well established that both developmental and genetic factors significantly influence the division of labour [ 15 , 16 ]. Additionally, studies have shown that faster and self-organised mechanisms for division of labour exist within colonies, enabling them to rapidly and adaptably respond to shifts in task requirements. Environmental factors or internal shifts within the colony may be responsible for these changes [ 14 , 17 ]. The fast changes in labour division arise from a combination of factors including workforce distribution, interaction structures, and environmental influences [ 18 , 19 , 20 , 21 ]. Empirical research has also emphasised the significance of social context and interactions in shaping the task preferences of individuals [ 22 , 23 ]. Individuals generally lack knowledge of the overall state of the colony, thus their behavioural decisions rely on the local information that is readily available to them [ 24 , 25 ]. Interactions among colony members can offer valuable insight into the colony’s condition and act as cues for behaviour, as well as a means for social learning [ 26 , 27 ]. Information acquired through local interactions with other individuals and the environment can often indicate the global state of the colony. While not frequently discussed in connection to social insects, empirical studies for some species have shown that social learning occurs [ 26 , 28 , 29 , 30 , 31 ]. The fundamental idea of social learning is that an individual observes other individuals and changes their behaviour based on the others’ presence or behaviour. This is a very broad notion. Individuals can be directly influenced by the observed behaviour of other individuals (learning), or they can be influenced by environmental social cues, such as pheromones or simply the presence of others [ 32 , 33 , 34 ]. Which behaviours are governed by which type of social influence is generally not well understood. In this study, we are only concerned with the dynamics of behaviour learned through imitation, which is already complex in itself and has not yet been widely investigated with mathematical models for social insects. Combining this with other social cues, for example pheromones, is a matter for future extensions of this framework. Thus, in our context, we apply the specific meaning that an individual copies a behaviour that is observed in others. There is indisputable empirical evidence that this happens [ 26 ]. Direct interaction or observation is necessary for this to occur. Given the possible complexity of social information exchange, we do not make any assumptions about its underlying mechanisms. We simply posit that individuals are more likely to imitate the behaviours of those who are successful. Independently of this, each individual may explore new behavioural variations with some probability. We have previously established that certain empirically observed characteristics of colony behaviour, including task specialisation and the emergence of inactive subgroups can arise as a by-product of social learning mechanisms [ 35 ]. However, despite the well-established existence of social interactions in colonies, it is uncertain whether these behavioural phenomena can be attributed to mechanisms of social learning with certainty. This is because the exact scope and extent of social learning in insect colonies are not yet well understood. In this study, we investigate whether the same inactivity can also emerge if we only assume individual learning mechanisms. We juxtapose pure individual learning and pure social learning , as two distinct methods of information processing by individuals at opposite ends of the spectrum of learning methods. Being based on very different types of information, these learning modes require very different cognitive and sensory capacities. Through a thorough examination of these two extremes, the study aims to gain an understanding of the effect that varying learning assumptions may have on the dynamics of the system. Our study is motivated by the empirical evidence indicating the presence of both social and individual learning mechanisms in social insects. Bumble bees are a common model system that demonstrate both types of learning. An instance of this is the flower selection behavior in bumblebees, which can be a result of both individual learning and behavior copying [ 28 ] and bumblebees exhibit the ability to learn when to use each type of information [ 36 ]. We thus need to understand the differences and similarities of these types of learning mechanisms, including their comparative advantages and disadvantages for colony fitness. To analyse the development of behaviour in a population under the social learning assumption, we employ adaptive dynamics, an analytic approach that originated in Evolutionary Game Theory (EGT) [ 37 , 38 , 39 ]. Agents follow basic rules to adjust their behaviour in response to an environmental signal, typically referred to as payoff [ 40 ]. Adaptive dynamics describes how a group responds to changes by taking into account the actions and interactions of individuals [ 41 ]. Evolutionary game theory was initially developed to model changes across evolutionary timescales, where the payoff represents fitness. However, this conceptual framework is not restricted to this timescale and can also be used to model faster processes that involve changes on colony lifetime scales, where payoffs are interpreted as feedback signals instead of fitness [ 35 , 42 , 43 ]. Our study is explicitly only concerned with these colony lifetime timescales. Interpreted in this way, adaptive dynamics captures how agents modify their task selection by taking into account task performance experience and environmental factors when when working together on multiple tasks. On the opposite end of the spectrum lies individual learning. It is commonly agreed that individual learning plays a crucial role [ 3 , 44 , 45 ] for social insects. It enables individuals to adjust to changing environments and improve their task performance over time. This is vital for the colony’s survival, as it allows individuals to adapt to new challenges and make better decisions about how to allocate resources and solve problems. Individuals can adapt their strategies by utilising previously acquired information in their current context [ 30 , 46 ]. The arguably best-established model of task selection in social insects, the reinforcement response threshold model, is centrally based on this notion [ 47 , 48 , 49 , 50 ]. In this study, we employ a particularly well-studied form of Reinforcement Learning (RL) [ 51 , 52 ] where agents update their action probabilities using the cross rule of RL [ 53 ]. Cross-learning is a relatively simple type of reinforcement learning that is based on individual behaviour and fully aligns with the assumptions of the established adaptive threshold reinforcement model. We compare the behavioural dynamics under these two learning assumptions for different types of environments. Our central aim is to investigate whether specific types of dynamics can be attributed to a specific learning mechanism, i.e., if they only emerge from social learning but not from individual learning or vice versa. We implement both processes in agent-based models to compare the outcomes. We back up the simulation studies with analytic results derived from adaptive dynamics. We are specifically interested in an effect previously referred to as laziness or inactivity in the population. This refers to the fact that in numerous efficient colonies, a significant portion of the workforce is comprised of inactive workers. This is a frequent occurrence in social organisations, including social insects, animals, humans, etc., which has been observed empirically and explained through modelling studies [ 35 , 54 , 55 , 56 , 57 , 58 ]. Our results show that identical behavioural dynamics, including the emergence of inactive workers, are observed independently of the learning mode . We conclude that this inactivity can be a by-product of the collective learning process in a joint environment but is not conditioned by a particular type of learning.",
"discussion": "4. Discussion Division of labour is essential for the survival and ecological success of social organisations. By dividing tasks among individuals, a social organisation can ensure that the most skilled or efficient individuals are performing specific tasks, which can lead to increased productivity and overall success. Dividing labour can also promote specialisation in the population, as individuals are able to focus on specific tasks and develop expertise in those areas. Furthermore, it also allows for flexibility and the ability to respond quickly to changes in the environment and internal requirements. Social insect colonies are examples of the most ecologically successful life forms, and an efficient division of labour is a critical aspect of their success. There has been a significant amount of research on the division of labour in social insects [ 1 ]. However, much of this research has focused on the impact of internal factors such as genetics [ 60 ], morphology [ 61 ], and hormones [ 62 ]. In comparison, there has been relatively less focus on the impact of the environment on task choices at the individual level and the underlying mechanisms of social interactions and their role in regulating the division of labour. We studied two of the most widely used methods for modelling the mechanisms of the division of labour in social organisations: social learning and individual learning. Very few previous studies have focused on comparing the similarities and differences in the outcomes resulting from the different update rules used. A comprehensive comparison of the two frameworks in various environmental settings is crucial in understanding the advantages and limitations of each assumption. It will help in the better understanding of the underlying mechanisms of the division of labour, in general, and more specific phenomena such as the emergence of inactivity as observed in empirical data, in particular. In this study, we have attempted to gain a deeper understanding of the implications of these two different learning paradigms for a specific behavioural phenomenon observed in social colonies: the emergence of inactive subgroups that do not participate in the collective action that sustains the colony. Previous studies have posited that this particular phenomenon, an instance of branching behaviour, is a by-product of the learning mechanism [ 35 ]. However, it was unclear whether these aspects of dynamics were indeed influenced by the presence of social interactions and the learning mechanism itself. By comparing and contrasting the results of both individual and social learning paradigms across different environmental conditions, we found equivalent behavioural outcomes in both cases. Specifically, we demonstrated that an individual, experience-based learning approach can also lead to inactivity in the population. This supports the hypothesis that regardless of the dominant learning mechanism in the colony, this aspect of colony life can arise as an artefact of the collective behaviour modification in a joint environment but is not necessarily restricted to a specific learning mechanism. Using mathematical intuition, this is not entirely surprising. There are deep correspondences between cross-learning and social learning. The seminal contribution by Borgers [ 63 ] first established that cross-learning and replicator dynamics ( Appendix D ), a formal model of learning by imitation, exhibit similar dynamics. However, the important restriction of this result is that it only applies to the learning of a single individual and that it can only be proven in expectation and in the continuous-time limit. The setting of a social insect colony, however, is necessarily population-based learning. The term population-learning is adapted from the reinforcement literature and can refer to any interacting collective, rather than to the specific meaning of population in biology. Börgers and Sarin’s finding was later extended by Lakhar and Seymour to show that population-based cross-learning evolves according to a specific form of the replicator equation under certain conditions, the so-called replicator continuity equation [ 64 ]. This equation is a partial differential equation that describes the changes in the population state over time. This was a very important finding, but it is restricted to replicator dynamics, which can only capture discrete behavioural states. Adaptive dynamics, which we have used here, can capture continuous behaviour parameters (such as an engagement level) and can analytically predict whether a population will split into subgroups. What we have demonstrated in this paper is that adaptive dynamics and population-based cross-learning exhibit qualitatively equivalent dynamics in the context of the study of inactivity. The presence of the studied inactive subgroups is a common occurrence in collective behavior, observed in organisations such as active particles, insects, animals, and humans [ 55 , 65 ]. Social insect colonies, in particular, can have over half of their workers inactive at a given time [ 66 ], which is surprising given the low individual selfishness levels in these colonies [ 13 ]. Despite our hypothesis that inactivity can arise as a by-product of the task allocation process and independent of the learning mode, in certain situations, this inactivity has been proposed to have a functional purpose. The main hypothesis for the functional role of inactive workers is that they serve as a reserve workforce that can be mobilised quickly when there is a sudden loss of workers or unexpectedly high task demands, thereby increasing colony flexibility and resilience [ 67 , 68 ]. Nevertheless, the benefits derived from having a reserve workforce of inactive individuals have never been quantified, either empirically or otherwise, leading many empirical research to still question this hypothesis [ 69 ]. Examples of other explanations include sleep or rest time of individuals [ 70 , 71 ], or delays occurring during task switching and the time the workers require to asses the collected information about task demands without engaging in any work [ 72 ]. However, the variation among individuals in social insect colonies in terms of their amount of inactivity cannot be fully explained by the need for resting periods alone [ 13 ] and although challenging, the purposeful activity of searching for a task, or “patrolling” should be distinguished from aimless and inactive wandering [ 73 ]. This suggests the necessity for additional research on the subject, both through empirical studies and theoretical or modeling work, as proposed in this study. Perhaps the main limitation of our study is that we have solely looked at social learning and individual learning in isolation. Yet, it is likely that, in many circumstances, both may occur simultaneously and even intermingle, possibly for individual task selection or even in a context-dependent manner, as shown in previous research [ 36 ]. While we have focused on analyzing isolated forms of each learning paradigm, investigating such mixed modes is a task for future studies. Our paper’s primary objective was to demonstrate that the overall behavior dynamics in the population are similar under both learning paradigms. Therefore, it is probable that a combination of learning mechanisms would also manifest similar dynamics. It should be relatively straightforward to confirm this with simulations, however a mathematical framework that captures both modes simultaneously is unclear. To conclude, we believe that our approach, beyond the study of this particular phenomenon of collective behaviour (i.e., inactivity), hints at new pathways for studying collective behaviour in animal groups from a generalised perspective without having to assume (or know) a restrictive model of learning. To explore this possibility further, the exact conditions under which these equivalences hold will have to be established formally and we hope to do so in future work."
} | 5,179 |
33973345 | PMC8596668 | pmc | 7,145 | {
"abstract": "Summary Saprotrophic fungi play an important role in ecosystem functioning and plant performance, but their abundance in intensively managed arable soils is low. Saprotrophic fungal biomass in arable soils can be enhanced with amendments of cellulose‐rich materials. Here, we examined if sawdust‐stimulated saprotrophic fungi extend their activity to the rhizosphere of crop seedlings and influence the composition and activity of other rhizosphere and root inhabitants. After growing carrot seedlings in sawdust‐amended arable soil, we determined fungal and bacterial biomass and community structure in roots, rhizosphere and soil. Utilization of root exudates was assessed by stable isotope probing (SIP) following 13 CO 2 ‐pulse‐labelling of seedlings. This was combined with analysis of lipid fatty acids (PLFA/NLFA‐SIP) and nucleic acids (DNA‐SIP). Sawdust‐stimulated Sordariomycetes colonized the seedling's rhizosphere and roots and actively consumed root exudates. This did not reduce the abundance and activity of bacteria, yet higher proportions of α‐Proteobacteria and Bacteroidia were seen. Biomass and activity of mycorrhizal fungi increased with sawdust amendments, whereas exudate consumption and root colonization by functional groups containing plant pathogens did not change. Sawdust amendment of arable soil enhanced abundance and exudate‐consuming activity of saprotrophic fungi in the rhizosphere of crop seedlings and promoted potential beneficial microbial groups in root‐associated microbiomes.",
"conclusion": "Conclusions and perspectives Stimulation of saprotrophic fungi by sawdust in arable soil was not restricted to bulk soil, but extended to the rhizosphere and roots of carrot seedlings. This coincided with increased uptake of root exudates by saprotrophic fungi, which was associated with a higher relative abundance of α‐Proteobacteria, Bacteroidia and a higher abundance of active arbuscular mycorrhizal fungi in rhizosphere and roots (Fig. 6 ). In soil and rhizosphere, sawdust caused a small increase in the estimated abundance of fungal groups containing potential plant pathogens; however, the root exudate consumption and root colonization by these groups did not increase after sawdust amendment. This study shows that an increased biomass and activity of saprotrophic fungi in the rhizosphere and root has a steering effect on other plant‐associated microbes and can potentially promote beneficial microbes. Further research is required for assessing the benefits of such changes on plant growth and health. We highlight the potential of saprotrophic fungi as target group for the design of sustainable agricultural practices, that impact both soil and plant functioning. Fig 6 Effect of sawdust on fungi and bacteria in an arable soil and in the root surroundings of carrot seedlings after 2‐ and 6‐weeks (T1, T2) pre‐sowing incubation times. White boxes report the biomass of bacteria, fungi and AMF in the rhizosphere soil, based on PLFA/NLFA markers. The black arrows represent the incorporation of 13 C in fungal, bacterial and AMF PLFA/NLFA markers. Pie charts report the proportion of bacterial taxa (top) and fungal functional groups (bottom) that responded to sawdust amendment in the soil and rhizosphere soil (external charts) and in in the 13 C‐enriched rhizosphere and root (central charts), based on DNA amplicon sequencing.",
"introduction": "Introduction In intensively managed arable soils, saprotrophic fungal biomass is low as compared with soils of natural ecosystems (Djajakirana et al ., 1996 ; de Vries and Bardgett, 2012 ). This is most likely caused by a lack of decomposable plant residues, due to an efficient removal of crop parts and the predominant use of mineral fertilizers (Clocchiatti et al ., 2020 ). Other contributing factors are the destruction of hyphal networks by tillage (Beare et al ., 1997 ; Frey et al ., 1999 ; Jiang et al ., 2011 ) and inhibition of fungal growth by chemical pesticides and fungicides (Duah‐Yentumi and Johnson, 1986 ; Nettles et al ., 2016 ; Rahman et al ., 2017 ; Shao and Zhang, 2017 ). In natural ecosystems and less intensively managed arable ecosystems, it is known that saprotrophic fungi have major positive contributions to ecosystem functioning, for example, by reducing nitrogen‐losses (de Vries et al ., 2011 ) forming soil aggregates (Beare et al ., 1997 ) and suppressing root‐infecting pathogenic fungi (van der Wal et al ., 2013 ). Therefore, agricultural management practises resulting in stimulation of saprotrophic fungi in arable soils can make an important contribution to improve sustainable crop production (de Vries and Bardgett, 2012 ; Frąc et al ., 2018 ). Fungal biomass stimulation can be achieved by low‐intensity tillage, in the long term (Wang et al ., 2017 ; Chen et al ., 2020 ), whereas the use of cellulose‐rich organic amendments has an immediate effect (Lucas et al ., 2014 ; Clocchiatti et al ., 2020 ). Strategies that act at multiple scales, such as improving soil functioning and engineering the rhizosphere microbiome, are key for maximizing their benefits on crop performance (Chaparro et al ., 2012 ; Bender et al ., 2016 ). In this light, the influence of fungus‐stimulating organic amendments on the performance of crops could become even more pronounced, when the stimulated fungi in bulk soil can extend their growth and activities into the rhizosphere and roots of seedlings. Conventional arable soils are bacterial‐dominated and bacteria are also the major consumers of root exudates in the rhizosphere of young plants (Hünninghaus et al ., 2019 ). Only over time, crop plants recruit and modulate bacterial activities so that they are well suited to benefit the plant (Badri et al ., 2013 ; Chaparro et al ., 2014 ). Similarly, saprotrophic fungal biomass increases in the rhizosphere of crops only at later plant developmental stages (Hannula et al ., 2012 ; Pausch et al ., 2016 ), when they become an important microbial group consuming rhizodeposits (Hannula et al ., 2012 ). This can be ascribed to the presence of larger amounts of cellulose‐rich root debris that become available during maturation of crops (Dennis et al ., 2010 ; Eisenhauer et al ., 2017 ; Pausch and Kuzyakov, 2018 ). Moreover, older roots exude a larger share of complex soluble compounds, such as aromatic acids (Gransee and Wittenmayer, 2000 ; Chaparro et al ., 2013 ; Zhalnina et al ., 2018 ). The low fungal activity in arable soil provides limited support to the initial growth and health of plants, which, conversely, are highly susceptible to the negative effects of soil‐borne pathogens (Lamichhane et al ., 2017 ) and heavily rely on exogenous chemical inputs. In a recent study, we showed that incorporation of deciduous wood sawdust in bare arable soils resulted in a rapid and prolonged increase of saprotrophic ascomycetes in bulk soil (Clocchiatti et al ., 2020 ). The stimulation of ascomycetes was ascribed to the accessibility of cellulose polymers in fragmented wood. Therefore, wood amendment could also represent a strategy for increasing the activity of saprotrophic ascomycetes near roots of young crop plants. Increased utilization of exudates by saprotrophic fungi is expected to have impacts on other members of rhizosphere – and root microbiomes (de Boer et al ., 2008 ; de Menezes et al ., 2017 ). Such interactions could be important to establish competitive suppression of root‐infecting pathogens (Fravel et al ., 2003 ; de Boer et al ., 2015 ; Kepler et al ., 2017 ); however, they could also hamper the colonization of roots by beneficial mycorrhizal fungi (de Jaeger et al ., 2010 ). The aim of the current study was to examine whether fungi stimulated by sawdust in bulk soil can colonize the rhizosphere and root of seedlings and participate in competitive interactions with other microbes for root exudates. In order to test this, we determined abundance and activity of fungi and bacteria in rhizosphere and roots of carrot seedlings that were grown in arable soil amended with beech wood sawdust. For the measurement of fungal and bacterial activity in the rhizosphere, we used whole plant 13 CO 2 pulse‐labelling and determined 13 C accumulation in phospholipid fatty acids (PLFA‐SIP) and microbial DNA markers (DNA‐SIP). We hypothesized that (i) stimulating fungal biomass in the bulk soil causes an increase in fungal biomass in the rhizosphere, which coincides with (ii) increased activity (i.e. uptake of exudates) of saprotrophic fungi in the rhizosphere of crop seedlings and results in (iii) shifts in the composition of rhizosphere – and root microbiomes. We further investigated the persistence of the effects by sowing seedlings 2 (T1) and 6 weeks (T2) after sawdust amendment.",
"discussion": "Discussion Effect of sawdust addition on abundance and composition of fungi in soil, rhizosphere and roots Free‐living, saprotrophic fungi not only occur in soil, they are also common inhabitants of the rhizosphere and are even found inside plant roots (Gkarmiri et al ., 2017 ; Hugoni et al ., 2018 ). However, saprotrophic fungi are negatively affected by intensive agricultural management practises, which tend to favour plant pathogenic fungi (Gao et al ., 2019 ). The addition of organic materials, such as cover crops and straw, can increment the abundance of fungi in soil of arable fields (García‐Orenes et al ., 2013 ; Lucas et al ., 2014 ; Rahman et al ., 2017 ) and fragmented woody material has been shown to be particularly effective (Clocchiatti et al ., 2020 ). The present study not only confirmed the stimulating effect of sawdust on fungi in unplanted arable soil, but also showed that, upon subsequent sowing, fungal stimulation extended to the rhizosphere and even roots of carrot seedlings. Fungal stimulation in rhizosphere and roots was still occurring after 6 weeks of pre‐incubation of sawdust in soil. Saprotrophic Sordariomycetes were the most stimulated fungi in bulk soil, rhizosphere and roots. Stimulation of Sordariomycetes by sawdust is in line with their known ability to efficiently decompose cellulose fractions of organic materials (Koechli et al ., 2019 ). Under conventional agricultural practises, this group tends to associate with older plant roots (Hannula et al ., 2010 ; Han et al ., 2017 ). In order to analyse the impact of sawdust‐amendment on plant‐pathogenic fungi, all functional groups containing 'Pathogen' as the only function or as one of the possible functions were included. This analysis revealed that the relative abundance of the SVs assigned to these functional groups was not increased by sawdust amendment. Yet, the estimation of the absolute abundances of SVs in functional groups containing plant pathogens showed a small increase after sawdust amendment. However, this absolute increase in potential pathogens was limited to the soil and rhizosphere compartments and it was not mirrored by an increase in their uptake of root‐derived 13 C or colonization of root tissues. Conversely, saprotrophic fungal stimulation was consistently high in sawdust‐amended soil, based on both estimated absolute and relative abundance. Such promotion of sawdust‐stimulated saprotrophic fungi extended also to the active root‐associated fungi and to the fungi colonizing the interior of roots. Further quantification of pathogens population abundance, combined with bioassays, is required to test the potential of the present saprotroph‐stimulating approach in counteracting the invasion of roots by disease‐causing fungi. Recently, the importance of the mycobiome in reducing the severity of soil‐borne diseases has been suggested (Busby et al ., 2016 ; Poli et al ., 2016 ; Sarrocco, 2016 ). An increased abundance of AMF in the rhizosphere (NLFA 16:1ω5) and in roots (ITS2 rRNA amplicon sequences) was found following sawdust amendment. In the current study, the soil received N‐fertilization only, thus the addition of sawdust could have raised the competition for available P in the soil due to increased uptake by saprotrophic fungi (Wu et al ., 2007 ; Zhang et al ., 2018 ). Under conditions of low P availability, AMF increase in abundance and support plant nutrient acquisition, in particular when plant litter is added to an arable soil (Xu et al ., 2018 ). In natural ecosystems, AMF are often associated with plant litter (Joner and Jakobsen, 1995 ; Gryndler et al ., 2002 ; Bunn et al ., 2019 ), where they are involved in nutrient mining and have a positive feedback on saprotrophic microbes (Quist et al ., 2016 ). To our knowledge, this is the first observation of simultaneous stimulation of AMF and saprotrophic fungi by an organic amendment in arable soil. We suggest that this finding is worth further exploration, given the well‐known beneficial effects of AMF of plant growth and health (Chandanie et al ., 2009 ; Hu et al ., 2010 ). Effect of sawdust addition on root exudate consumption by fungi Here, we demonstrate that saprotrophic fungi stimulated by sawdust amendment are actively consuming root‐derived C in the rhizosphere of seedlings, which was measured as increased 13 C excess accumulated in the fungal PLFA 18:2ω6,9. Uptake of root‐derived C by saprotrophic fungi was previously observed for mature annual crop plants and perennial crops (Tavi et al ., 2013 ; Pausch et al ., 2016 ), but not for crop seedlings (Hünninghaus et al ., 2019 ). For potato grown in intensively managed soils, it was shown that fungal biomass in the rhizosphere increased in advanced growth stages, probably due to the presence of recalcitrant root deposits (Hannula et al ., 2010 ). Hence, it seems that sawdust amendment creates a situation where the rhizosphere of crops can readily be colonized by root‐exudate consuming saprotrophic fungi at the seedling stage, which normally requires crop maturation. Increased root C uptake after sawdust addition was ascribed mainly to sordariomycetal fungi, as they were prevalent within the 13 C‐enriched fungal community. Due to the short duration of the 13 CO 2 pulse (24 h), followed by immediate harvesting of rhizosphere soil and plants, mainly soluble exudates were probably enriched in 13 C (Kaštovská et al ., 2017 ). Thus, the observed 13 C labelling in the fungal biomarker can be attributed to the uptake of root exudates like sugars and organic acids. The PLFA marker 18:2w6,9 can be found in plants; hence, care was taken when sampling the rhizosphere soil not to include plant cells. Furthermore, by using qPCR on the ITS2 region and ergosterol as additional measurements of fungal biomass, we showed that these reveal a similar pattern (Supporting Information Fig. S1 ) and we concluded that the marker 18:2ω6,9 is a useful indicator of fungal biomass in the rhizosphere (Joergensen and Wichern, 2008 ; Kaiser et al ., 2010 ; Frostegård et al ., 2011 ), especially when combined with 13 C pulse‐labelling. Wood sawdust and exudates greatly differ in biochemical complexity. However, saprotrophic fungi take up soluble sugars released from lignocellulose‐rich organic matter by extra‐cellular enzymes and are, therefore, also able to use labile monomeric exudate compounds (Buée et al ., 2009 ; van der Wal et al ., 2013 ; de Vries and Caruso, 2016 ). Rapid exudate assimilation by ascomycetal fungi was indeed found in earlier studies (Hannula et al ., 2012 ; Marschner et al ., 2012 ; Gkarmiri et al ., 2017 ; Wang et al ., 2019 ). Our study points out that sordariomycetal fungi are well‐equipped for efficient decomposition of both exudates and of cellulose‐rich organic materials in arable soils. The use of fine wood particles could be important to ensure bridging of fungal hyphae between wood particles and plant roots. Earlier studies reported that AMF play a dominant role in rapid acquisition of plant‐C, which is transferred only in a later stage to saprotrophic fungi and bacteria (Drigo et al ., 2010 ; Hünninghaus et al ., 2019 ). In the current study, the contributions of saprotrophic and arbuscular mycorrhizal fungi to plant‐C acquisition were not mutually exclusive. On the contrary, a simultaneous stimulation of active incorporation of plant‐C by AMF ( 13 C excess in the NLFA 16:1ω5) and saprotrophic fungi was observed in the rhizosphere of plants sown 2 weeks after sawdust amendment. In line with what is discussed above, this suggest the possibility that after sawdust addition, in absence of P fertilization, no negative interactions occurred between saprotrophic fungi and AMF in the root surroundings. Effect of sawdust addition on composition and activity of bacteria PLFA‐ and qPCR‐based quantification of bacteria showed an increase in bacterial abundance in sawdust‐amended soil, but this was less pronounced as compared to fungi. Similarly, sawdust caused a smaller shift in bacterial community composition than in that of fungi. Bacterial groups typically associated with cellulose degradation in arable soils, such as Acidobacteria, β‐, γ‐, δ‐Proteobacteria and Actinobacteria (Kramer et al ., 2016 ) did not increase in relative abundance after sawdust addition. This suggests that fungi were the main primary decomposers of sawdust, while increase of bacteria can be ascribed to the consumption of fungal‐derived carbon or to the consumption of breakdown products derived from fungal sawdust decomposition (Schneider et al ., 2010 ; Ballhausen and de Boer, 2016 ; de Menezes et al ., 2017 ). Amplicon sequencing revealed that Bacteroidia and α‐Proteobacteria increased after addition of sawdust in the soil. Although Bacteroidia and α‐Proteobacteria could contribute in part to sawdust decomposition (Schellenberger et al ., 2009 ; Haichar et al ., 2016 ; Kramer et al ., 2016 ), these groups are known to engage in fungal‐bacterial interactions. Bacteroidia harbour an array of chitinolytic enzymes, which confers them the ability to utilize fungal debris or predate on fungi (Wieczorek et al ., 2019 ). The utilization of fungal‐derived C by Bacteroidia is in line with the observation that they were more present in the soil, but not in the 13 C‐enriched bacterial community, meaning that they potentially received a large share of sawdust‐derived C via fungi. α‐Proteobacteria are often associated with decomposing wood, where they are thought to provide an additional source of N to fungal saprotrophs in exchange of wood breakdown products (Hoppe et al ., 2015 ; Johnston et al ., 2016 ). Hence, the ability of α‐Proteobacteria to respond to wood breakdown products may explain their increase, even though their role of N‐suppliers is unlikely to be important in the current experimental setting where mineral N was supplied together with sawdust. Moreover, the presence of fungal hyphae in the bulk and rhizosphere soil could facilitate the movement of α‐Proteobacteria, and Rhizobiales in particular, towards plant roots, as recently demonstrated in a legume crop (Zhang et al ., 2020 ). The 13 C incorporation in bacterial PLFAs was not affected by sawdust amendment. Thus, the increased consumption of root exudates by saprotrophic fungi did not create the hypothesized constraint for the uptake of plant‐derived carbon by bacteria. This suggests that sawdust‐stimulated fungi utilize an additional pool of plant C than what is used by bacteria in both the control and sawdust‐amended soil. Accordingly, it seems that the total uptake of released root C is more efficient in the rhizosphere when active fungi are present. A similar pattern was reported by Morriën et al . ( 2017 ), in a study where an increased activity of fungi led to a better uptake of plant C by the food web of a restored ex‐arable soil. Even though increased fungal activity did not alter the total bacterial activity, the presence of active fungi in the rhizosphere and roots steered the composition of the plant‐associated bacterial community, especially promoting Bacteroidia and α‐Proteobacteria. In particular, sawdust amendment and fungal stimulation increased the abundance of potentially beneficial bacteria within the roots, namely Chitinophagaceae , Sphingobacteriaceae and Rhizobiales. Chitinophagaceae and Sphingobacteriaceae could contribute to resistance against diseases by antagonistic interactions. Chitinophaga spp. were previously found within sugar beet plants challenged by Rhizoctonia solani and were associated with a higher expression of chitinases (Carrión et al ., 2019 ). Spingobacteriaceae were also found in disease suppressive soils and showed antagonistic activity against plant‐pathogenic fungi, when triggered by interaction with other members of the bacterial community (de Boer et al ., 2007 ; Gómez Expósito, 2017 ). Although Rhizobiales are mostly known as N‐fixing mutualists of legumes, they can also establish mutualistic interactions with non‐legume plants and contribute to plant growth stimulation and priming of the plant immune system (Garrido‐Oter et al ., 2018 ). Overall, the use of sawdust and the subsequent fungal stimulation steered the soil and rhizosphere bacterial community and influenced the association of seedlings with potentially beneficial bacterial groups. In this study, changes in size, activity and composition of the fungal and bacterial communities are mainly interpreted as a direct effect of wood particles or sawdust‐stimulated fungi to the resident soil microbiome. In addition to this, it cannot be excluded that wood‐inhabiting microbes were introduced into the soil and contributed to the observed effects (Massart et al ., 2015 ; Mawarda et al ., 2020 ). We argue that their role was minor, as the presence of microbes in commercial sawdust is reduced by drying at high temperatures (Ståhl et al ., 2004 ). Furthermore, we expect that the remaining sawdust‐inhabiting microbes are rapidly outcompeted during wood decomposition in soil (Petrini and Fisher, 1990 ; van der Wal et al ., 2016 )."
} | 5,511 |
29900421 | PMC5995451 | pmc | 7,146 | {
"abstract": "Advances in metabolic engineering and synthetic biology have facilitated the manufacturing of many valuable-added compounds and commodity chemicals using microbial cell factories in the past decade. However, due to complexity of cellular metabolism, the optimization of metabolic pathways for maximal production represents a grand challenge and an unavoidable barrier for metabolic engineering. Recently, cell-free protein synthesis system (CFPS) has been emerging as an enabling alternative to address challenges in biomanufacturing. This review summarizes the recent progresses of CFPS in rapid prototyping of biosynthetic pathways and genetic circuits (biosensors) to speed up design-build-test (DBT) cycles of metabolic engineering and synthetic biology.",
"introduction": "1 Introduction Metabolic engineering and synthetic biology are one of the most promising solutions to address sustainability and global climate change challenges through the development of efficient cell factories for producing fuels, chemicals, and pharmaceutics. Introduction of a biosynthetic pathway containing multiple heterologous genes is generally the first step to enable microbial synthesis. For practical applications, pathway gene expression levels must be carefully fine-tuned and balanced to maximize product titer, rate, and yield (TRY) [ 1 ]. However, iterative design-build-test (DBT) cycles are generally required to optimize metabolic pathways, making the development of efficient cell factories rather time-consuming. Although numerous synthetic biology tools have been developed, such as gene copy number tuning [ 2 , 3 ] and combinatorial transcriptional regulation [ 4 ], the build and test of the combinatorial biosynthetic pathway library remains a grand challenge for metabolic engineering. For example, it is reported that more than 100 person-years are needed to commercialize biobased 1,3-propanediol [ 1 ]. Therefore, the development of novel enabling synthetic biology tools to accelerate DBT cycles will be critical for the construction and optimization of microbial cell factories. Cell-free protein synthesis system (CFPS) has been widely used for recombinant protein expression, particularly toxic proteins and membrane proteins that are difficult to express in vivo [ [5] , [6] , [7] , [8] , [9] ]. Recently, CFPS is further developed as an enabling platform for rapid prototyping of biosynthetic pathways and genetic circuits to address the challenges in metabolic engineering and synthetic biology. Compared with the conventional in vivo systems, CFPS activates biological machineries without the boundary of cell membranes and cell walls, and the open environment allows for direct monitoring and manipulation of transcription, translation and metabolism [ 10 ]. In addition, CFPS uses linear DNAs (i.e. PCR products) as templates for transcription, which bypasses the time-consuming and laborious gene-cloning and microorganism transformation steps and allows rapid prototyping in a high-throughput manner [ 11 ]. Different from the in vivo strategies by manipulating the complex transcription and translation machineries, metabolic pathway gene expression levels can be simply controlled by adjusting linear DNA concentrations supplemented to CFPS with coupled transcription-translation reactions under no resource limitation conditions [ 11 ]. Currently, CFPS has been successfully applied to the construction and optimization of metabolic pathways [ [12] , [13] , [14] , [15] , [16] , [17] ] and genetic circuits [ [18] , [19] , [20] , [21] , [22] ]. It has been reported that the CFPS can reduce the time to build metabolic pathways and genetic circuits from days to hours. Notably, biosensors based on the optimized genetic circuits hold great potentials for rapidly testing the constructed metabolic pathway libraries in both in vivo and in vitro metabolic engineering systems in a high throughout manner [ 23 , 24 ]. The combined rapid prototyping of metabolic pathways (build) and genetic circuits (test) open unique opportunities to accelerate DBT cycles for metabolic engineering and synthetic biology ( Fig. 1 ). Fig. 1 CFPS as an enabling platform to accelerate design-build-test cycles of metabolic engineering and synthetic biology. Fig. 1 As protein synthesis in CFPS has been reviewed in detail [ [25] , [26] , [27] ], this review covers the most recent update on the applications of CFPS in rapid prototyping for metabolic engineering and synthetic biology. More specifically, we focus on the construction and optimization of metabolic pathways and genetic circuits (biosensors) using CFPS. The methods to prepare the CFPS extracts and the corresponding energy regeneration systems will be briefly reviewed as well. Finally, the challenges and future perspectives on CFPS for metabolic engineering and synthetic biology applications will be discussed."
} | 1,222 |
28899034 | PMC5812505 | pmc | 7,147 | {
"abstract": "Abstract Over the last 15 years, several genome-scale metabolic models (GSMMs) were developed for different yeast species, aiding both the elucidation of new biological processes and the shift toward a bio-based economy, through the design of in silico inspired cell factories. Here, an historical perspective of the GSMMs built over time for several yeast species is presented and the main inheritance patterns among the metabolic reconstructions are highlighted. We additionally provide a critical perspective on the overall genome-scale modeling procedure, underlining incomplete model validation and evaluation approaches and the quest for the integration of regulatory and kinetic information into yeast GSMMs. A summary of experimentally validated model-based metabolic engineering applications of yeast species is further emphasized, while the main challenges and future perspectives for the field are finally addressed.",
"introduction": "INTRODUCTION The availability of low-cost whole-genome sequencing techniques led to an explosion of data for several organisms. This, alongside the advent of organism-specific omics data, advanced bioinformatics tools, and an increasing computational performance has paved the way to the reconstruction of metabolic networks at the genome scale. Genome-wide reconstructions of the cell metabolism can be converted into predictive constraint-based models, establishing a complex network of biochemical reactions with information on stoichiometry, compartmentalization, biomass composition, thermodynamics and genes responsible for each reaction (Covert et al. 2001 ; Thiele and Palsson 2010 ). When combined with constraint-based algorithms, genome-scale metabolic models (GSMMs, also known as GEMs) offer an excellent opportunity for studying metabolism and genotype–phenotype relationships (O’Brien, Monk and Palsson 2015 ). Hence, GSMMs have become a key framework in the systems biology field, in particular, for systems metabolic engineering (ME) approaches. After the first GSMM was published nearly 20 years ago (Edwards and Palsson 1999 ), many others have followed. Hitherto, there are GSMMs published and accessible for download in several websites for more than 100 organisms (e.g. www.optflux.org/models and http://systemsbiology.ucsd.edu/InSilicoOrganisms/OtherOrganisms ), and the number keeps rising. The yeast Saccharomyces cerevisiae was the first eukaryotic organism to be fully genome sequenced (Goffeau et al. 1996 ; Cherry et al. 1997 ), and it has been one of the workhorses in cell factory engineering for biotechnological production of several compounds with widespread applications in food (Brochado et al. 2010 ; Li et al. 2013a ), chemical (Hong and Nielsen 2012 ; Nielsen et al. 2013 ) and pharmaceutical industries (Paddon et al. 2013 ). Given the similarity and high number of features conserved with the human functions, it is also a role model for diseases, drug screening and fundamental biology studies (Sturgeon et al. 2006 ; Petranovic and Nielsen 2008 ). So, it is not a surprise that S. cerevisiae has been the first eukaryotic organism to have a GSMM (Förster et al. 2003 ), and is top-ranked if we count the number of available GSMMs per single organism. However, other yeast species constitute important human pathogens or have also proven to be suitable platforms for biotechnological applications and several models have been therefore reconstructed for different yeasts. Here, we review the genome-scale modeling process in yeast, presenting an historical perspective of the GSMMs built along the time for different yeast species beyond the well-characterized S. cerevisiae through the representation of a chronological network containing the inherited features of several yeast models. We then present a critical perspective on the overall genome-scale modeling procedure in yeast, from incomplete model validation and evaluation approaches to the increasing pursuit for the integration of regulatory and kinetic information into metabolic networks. A summary of the main applications of yeasts’ GSMMs in cell factory development is further addressed. Lastly, the future perspectives in the field are discussed."
} | 1,051 |
39805848 | PMC11730660 | pmc | 7,148 | {
"abstract": "Here, we design exotic interfaces within a flexible thermoelectric device, incorporating a polyimide substrate, Ti contact layer, Cu electrode, Ti barrier layer, and thermoelectric thin film. The device features 162 pairs of thin-film legs with high room-temperature performance, using p-Bi 0.5 Sb 1.5 Te 3 and n-Bi 2 Te 2.7 Se 0.3 , with figure-of-merit values of 1.39 and 1.44, respectively. The 10 nm Ti contact layer creates a strong bond between the substrate and the Cu electrode, while the 10 nm Ti barrier layer significantly reduces internal resistance and enhances the tightness between thermoelectric thin films and Cu electrodes. This enables both exceptional flexibility and an impressive power density of 108 μW cm −2 under a temperature difference of just 5 K, with a normalized power density exceeding 4 μW cm −2 K −2 . When attached to a 50 °C irregular heat source, three series-connected devices generate 1.85 V, powering a light-emitting diode without the need for an additional heat sink or booster.",
"introduction": "Introduction Thermoelectric materials and devices offer direct conversion between thermal energy (temperature differences, Δ T ) and electrical energy, holding promise in fields such as waste heat recovery and solid-state cooling, aiming to alleviate pressure on fossil fuel consumption and environmental pollution 1 . Currently, most commercialized thermoelectric devices are based on solid-state configurations, comprising pairs of p-type and n-type inorganic single or polycrystalline bismuth-telluride-based bulks connected electrically in series and thermally in parallel, sandwiched between rigid alumina substrates, with metal electrodes (such as copper) linking them 1 . While these devices demonstrate high thermoelectric conversion efficiency, they struggle to effectively harness waste heat from irregular heat sources like human skin and drainage exhaust pipes 2 . Special device designs (e.g., ring-shaped) can accommodate curved heat sources, but they require specific surface curvature, limiting their adaptability to various and variably curved heat sources 1 . While it is possible to attach a flexible secondary substrate to the hot side of solid-state devices to capture waste heat from irregular heat sources, the secondary substrate introduces a significant loss of Δ T , resulting in lower utilization efficiency and diminished output performance. Thus, developing thermoelectric materials and devices with certain flexibility is crucial and has become a recent research focus 3 . Generally, higher thermoelectric performance of materials leads to greater energy conversion efficiency of the device. The thermoelectric potential of a material can be evaluated by ZT = S 2 σT / κ , where T denotes the absolute temperature, S 2 σ represents the power factor, composed of the Seebeck coefficient ( S ) and electrical conductivity ( σ ), reflecting the overall electrical transport capacity, and κ denotes the thermal conductivity, comprising electronic thermal conductivity ( κ e ) and lattice thermal conductivity ( κ l ) 1 . κ l is usually determined by the micro/nanostructure of the material, and more complex structures with more multi-dimensional lattice defects usually lead to lower κ l , but also lower σ due to carrier scattering at these crystal imperfections 1 . Generally, to enable certain flexibility to thermoelectric devices, organic and organic/inorganic hybrid thermoelectric materials have been explored 4 . While organic materials exhibit ultra-high flexibility and low κ (owing to their macromolecular characteristics) 5 , their low S results in reduced S 2 σ and ZT , limiting device performance 5 . Similarly, carbon-based materials such as single-walled carbon nanotubes (SWCNTs) are flexible but suffer from excessive κ and low S 6 . In contrast, inorganic thermoelectric materials and their devices demonstrate significantly higher thermoelectric performance and stability compared to carbon, organic, and organic/inorganic hybrid materials, leading to the rapid development of flexible devices based on inorganic materials in recent years 7 . To impart flexibility to these devices, besides reducing the size of bulk thermoelectric materials in the device (e.g., miniatures) 8 , approaches include using flexible substrates 7 and special device designs (e.g., hinge-type) 9 , enabling these devices to maintain high output power ( P ) while retaining some flexibility. However, the interface issues between rigid materials and flexible substrates and electrodes continue to limit the flexibility of these devices 7 . Therefore, to further enhance the flexibility of flexible thermoelectric devices based on inorganic materials while maintaining their high thermoelectric performance, flexible inorganic thin-film materials have started to receive significant attention. Currently, thin-film-based inorganic thermoelectric materials include bismuth/antimony tellurides (e.g., Bi 2 Te 3− x Se x and Bi x Sb 2- x Te 3 ) 10 , 11 , silver chalcogenides (e.g., Ag 2 S 12 , Ag 2 Se 13 , and Ag 2 Te 14 ), other chalcogenides (e.g., SnSe 15 , Cu 2 Se 16 , and TiS 2 17 ), and oxide (e.g., ZnO) 18 , among others. So far, thermoelectric materials based on bismuth/antimony tellurides remain one of the best choices for near-room-temperature thermoelectric applications, owing to their narrow bandgap of approximately 0.1 eV, high inherent S , high σ , and low κ 10 , 11 . n-type materials, such as Bi 2 Te 3− x Se x , typically exhibit thermoelectric performance with ZT > 1 in the temperature range of 300–400 K 10 , 11 , while p -type materials, such as Bi x Sb 2- x Te 3 , can achieve ZT values exceeding 1.2 in the same temperature range 10 , 11 . Considering that most flexible thermoelectric devices use flexible substrates such as polyimide (PI) and other organic polymers 7 , which are not tolerant to high temperatures (e.g., over 300 °C which is the glass transition temperature of PI), these thin-film thermoelectric materials can be well-matched with these flexible organic polymer-based substrates in terms of operating temperature. At this point, to maximize the thermoelectric performance and flexibility of flexible devices based on bismuth/antimony telluride thin films, reasonable material design and optimization are required, while the materials themselves need to possess a certain level of flexibility 7 . This has become a focal point of research in recent years 7 . To enhance the flexibility of these materials, aside from nanostructuring their thickness 7 , reinforcement of their orientation structure can also improve flexibility 19 . However, despite the improvement in flexibility of the materials and the devices composed of them, interface issues remain unresolved 7 , resulting in the P of these flexible devices still being unable to match that of commercial solid-state devices. Moreover, whether thin-film materials can maintain the same high thermoelectric performance as bulk materials also poses a challenge to their preparation processes and design strategies 7 . Till now, the structural design of flexible thermoelectric devices based on inorganic flexible films, particularly interface design, remains at a very preliminary stage 7 , 20 . Many reported flexible devices based on thin-film thermoelectric materials simply used Ag paste to connect the thermoelectric legs on the PI substrates 21 – 24 , with little consideration for contact issues between the films, electrodes, and substrates, resulting in impractical device designs and poor sustainability 7 . Some devices utilize inorganic metal materials as functional diffusion barrier layers between thermoelectric materials and electrodes 25 , 26 . However, there are few reports in flexible device design on enhancing the adhesion between electrodes and flexible substrates to improve overall device flexibility and stability 27 – 31 . Additionally, previous flexible devices often applied Δ T s of tens or even hundreds of Kelvins to demonstrate their high power density ( ω ). However, in practical environments such as wearable thermoelectric devices on human skin, Δ T s are typically only a few Kelvin 7 . Moreover, due to the difficulty of mounting a heat sink at the cold side of flexible devices for power generation, the actual Δ T within the flexible devices is often only a few Kelvin 7 . In such small Δ T s, to enable truly practical applications of flexible thermoelectric devices, their ω or normalized power density ( ω T ) must be sufficiently high to power low-grade electronics without the need for a heat sink or amplifier."
} | 2,160 |
36466259 | PMC9712798 | pmc | 7,149 | {
"abstract": "Concern that depletion of fertilizer feedstocks, which are a finite mineral resource, threatens agricultural sustainability has driven the exploration of sustainable methods of soil fertilization. Given that microalgae, which are unicellular photosynthetic organisms, can take up nutrients efficiently from water systems, their application in a biological wastewater purification system followed by the use of their biomass as a fertilizer alternative has attracted attention. Such applications of microalgae would contribute to the accelerated recycling of nutrients from wastewater to farmland. Many previous reports have provided information on the physiological characteristics of microalgae that support their utility. In this review, we focus on recent achievements of studies on microalgal physiology and relevant applications and outline the prospects for the contribution of microalgae to the establishment of sustainable agricultural practices.",
"conclusion": "Conclusions and prospects To achieve rapid growth and efficient nutrient accumulation in water systems, microalgae developed mechanisms such as flexible CCMs and membrane lipid remodeling. Previous research has shed light on the sophisticated molecular interactions underlying the physiological characteristics of microalgae, which support its utility as a wastewater purification system and fertilizer. Applications of microalgae in a wastewater purification system followed by fertilizer use may facilitate the establishment of nutrient recycling. Many studies have shown that application of microalgal biomass can provide nutrients essential for plants and enrich organic carbons in soils. In addition, microalgal biomass contains slowly degradable forms of plant-essential nutrients, reducing the leaching of the nutrients from farmland. Furthermore, microalga-based fertilizers are regarded as suppliers of plant growth regulators. However, challenges remain in the expansion of microalga-based technologies. For example, a life cycle assessment highlighted the detrimental impact of electricity consumption required for microalgal cultivation ( Diniz et al., 2017 ; de Souza et al., 2019 ). In addition, the application of a microalga-based fertilizer can stimulate the emission of greenhouse gases, such as N 2 O and CO 2 , from soils ( Suleiman et al., 2020 ). Thus, further technological advances, as well as a more in-depth understanding of microalgal physiology, are required for wider implementation of microalgal applications for sustainable agriculture.",
"introduction": "Introduction With the increasing threat of mineral resource depletion through human activities, demand for renewable feedstocks is rising dramatically. The utilization of photosynthetic organisms, including land plants and algae, offers one promising solution. For example, lignocellulosic biomass, which is composed predominantly of plant secondary cell walls, represents an abundant and renewable feedstock for materials, chemicals, and fuels ( Ragauskas et al., 2014 ; Umezawa, 2018 ; Miyamoto et al., 2020 ). Promoting the applications of photosynthetic organisms would contribute to the establishment of a sustainable human society. In the context of agricultural sustainability, a renewable alternative to synthetic chemical fertilizers is urgently required. Enhanced utilization of synthetic chemical fertilizers in conjunction with the development of modern crop cultivars, in which the yield is highly responsive to intensive fertilization, has contributed to improved crop productivity worldwide ( Khush, 2001 ). For example, in soils a large portion of phosphorus (P), an essential macronutrient for plants, likely exists as non-available or poorly available forms for crops, which increases the importance of P fertilizer. However, because the raw material of P fertilizers, rock phosphate, is a finite resource distributed unevenly in limited areas of the world, depletion of the reserves is of grave concern ( Desmidt et al., 2015 ). In addition, the manufacture of nitrogen (N) fertilizers requires the burning of fossil fuels to fix atmospheric N 2 and intensive use of N fertilizers enriches reactive N compounds, leading to soil acidification, water eutrophication, and atmospheric pollution ( Hayashi et al., 2021 ). Thus, to establish a sustainable agricultural system worldwide, renewable alternatives to chemical fertilizers and the adoption of eco-friendly soil fertilization practices ( Lin et al., 2019 ), as well as strategies to increase the nutrient use efficiency of crops ( Hu et al., 2015 ; Wu et al., 2020 ; Ochiai et al., 2022 ), should be explored. Microalgae are unicellular photosynthetic organisms commonly found in freshwater and marine ecosystems. They have been used in both experimental and real-world settings to biologically purify wastewater ( Vadiveloo et al., 2021 ). Wastewater purification systems using microalgae represent a promising alternative to conventional wastewater treatment technologies that consume high amounts of energy, discharge sludge, and emit greenhouse gases ( Qiao et al., 2020 ). Microalgae can rapidly grow and proliferate by efficiently acquiring carbon dioxide (CO 2 ) and nutrients, such as P and N, from water systems ( Sukačová et al., 2020 ). Also, the use of microalgal biomass as a biofertilizer as well as a fuel resource can contribute to the enhanced recycling of nutrients ( Das et al., 2019 ; Khan et al., 2019 ; Moges et al., 2020 ). Previous works have revealed many physiological characteristics favorable to the use of microalgae in sustainable agriculture. In addition, empirical evidence on the effectiveness and characteristics of microalga-based fertilizers associated with their physiology has been reported. This review is focused on interactions between basic and applied studies of microalgae, providing insight into a strategy for the establishment of sustainable agriculture."
} | 1,475 |
33286159 | PMC7516857 | pmc | 7,150 | {
"abstract": "How cognitive neural systems process information is largely unknown, in part because of how difficult it is to accurately follow the flow of information from sensors via neurons to actuators. Measuring the flow of information is different from measuring correlations between firing neurons, for which several measures are available, foremost among them the Shannon information, which is an undirected measure. Several information-theoretic notions of “directed information” have been used to successfully detect the flow of information in some systems, in particular in the neuroscience community. However, recent work has shown that directed information measures such as transfer entropy can sometimes inadequately estimate information flow, or even fail to identify manifest directed influences, especially if neurons contribute in a cryptographic manner to influence the effector neuron. Because it is unclear how often such cryptic influences emerge in cognitive systems, the usefulness of transfer entropy measures to reconstruct information flow is unknown. Here, we test how often cryptographic logic emerges in an evolutionary process that generates artificial neural circuits for two fundamental cognitive tasks (motion detection and sound localization). Besides counting the frequency of problematic logic gates, we also test whether transfer entropy applied to an activity time-series recorded from behaving digital brains can infer information flow, compared to a ground-truth model of direct influence constructed from connectivity and circuit logic. Our results suggest that transfer entropy will sometimes fail to infer directed information when it exists, and sometimes suggest a causal connection when there is none. However, the extent of incorrect inference strongly depends on the cognitive task considered. These results emphasize the importance of understanding the fundamental logic processes that contribute to information flow in cognitive processing, and quantifying their relevance in any given nervous system.",
"conclusion": "5. Conclusions Our results imply that using pairwise transfer entropy has its limitations in accurately estimating the information flow, and its accuracy may depend on the type of network or cognitive task it is applied to, as well as the type of data that is used to construct the measure. Higher-order conditional transfer entropies or more sophisticated measures such as partial information decomposition [ 21 ] may be able to alleviate those errors, at the expense of significant computational investments. We also find that simple networks that respond to a low-dimensional set of stimuli (such as the two example tasks investigated here) lead to problems in inferring information flow simply because transfer entropy estimates will be prone to sampling errors. These findings highlight the importance of understanding the frequency and types of fundamental processes and relations in biological nervous systems. For example, one approach would be to examine transfer entropy in known systems, especially in simple biological neural networks in order to shed light on the strengths and deficiencies of current methods. Performing an information flow analysis on brains in vivo will remain a daunting task for the foreseeable future, but advances in the evolution of digital cognitive systems may allow us a glimpse of the circuits in biological brains, and perhaps guide the development of other measures of information flow.",
"introduction": "1. Introduction When searching for common foundations of cortical computation, more and more emphasis is being placed on information-theoretic descriptions of cognitive processing [ 1 , 2 , 3 , 4 , 5 ]. One of the core tasks in the analysis of cognitive processing is to follow the flow of information within the nervous system, by finding cause-effect components. Indeed, understanding causal relationships is considered to be fundamental to all natural sciences [ 6 ]. However, inferring causal relationships and separating them from mere correlations is difficult, and the subject of ongoing research [ 7 , 8 , 9 , 10 , 11 ]. The concept of Granger causality is an established statistical measure that aims to determine directed (causal) functional interactions among components or processes of a system. Schreiber [ 12 ] described Granger causality in terms of information theory by introducing the concept of transfer entropy (TE). The main idea is that if a process X is influencing process Y , then an observer can predict the future state of Y more accurately given the history of both X and Y (written as X t ( k ) and Y t ( ℓ ) , where k and ℓ determine how many states from the past of X and Y are taken into account) compared to only knowing the history of Y . According to Schreiber, the transfer entropy TE X → Y quantifies the flow of information from process X to Y : (1) TE X → Y = I ( Y t + 1 : X t ( k ) | Y t ( ℓ ) ) = H ( Y t + 1 | Y t ( ℓ ) ) − H ( Y t + 1 | Y t ( ℓ ) , X t ( k ) ) = ∑ y t + 1 ∑ x t ( k ) ∑ y t ( ℓ ) p ( y t + 1 , x t ( k ) , y t ( ℓ ) ) log p ( y t + 1 | x t ( k ) , y t ( ℓ ) ) p ( y t + 1 | y t ( ℓ ) ) . Here as before, X t ( k ) and Y t ( ℓ ) refer to the history of the processes X and Y , while Y t + 1 refers to the variable at t + 1 only. Further, p ( y t + 1 , x t ( k ) , y t ( ℓ ) ) is the joint probability of Y t + 1 and the histories X t ( k ) and Y t ( ℓ ) , while p ( y t + 1 | x t ( k ) , y t ( ℓ ) ) and p ( y t + 1 | y t ( ℓ ) ) are conditional probabilities. The transfer entropy ( 1 ) is a conditional mutual entropy, and quantifies what the process Y at time t + 1 knows about the process X up to time t , given the history of Y up to time t (see [ 13 ] for a thorough introduction to the subject). Specifically, TE X → Y measures “how much uncertainty about the future course of Y can be reduced by the past of X , given Y ’s own past.” Transfer entropy reduces to Granger causality for so-called “auto-regressive processes” [ 14 ] (which encompasses most biological dynamics), and has become one of the most widely used directed information measures, especially in neuroscience (see [ 5 , 13 , 15 , 16 ] and references cited therein). While transfer entropy is sometimes used to infer causal influences between susbsystems, it is important to point out that inferring causal relationships is different from inferring information flow [ 17 ]. In complex systems (for example, in computations that a brain performs to choose the correct action given a particular sensory experience) events in the sensory past can causally influence decisions significantly distant in time, and to capture such influences using the transfer entropy concept requires a careful analysis in which not only the history lengths k and ℓ used in Equation ( 1 ) must be optimized, but false influences due to linear mixing of signals (which can mimic causal influences) must also be corrected for [ 13 , 15 ]. In some sense, inferring information flow is a much simpler task than finding all causal influences, as we need only to identify (and quantify) the sources of information transferred to a particular variable. More precisely, for this application the pairwise transfer entropy is used to find candidate sources (in the immediate past) that account for the entropy of a particular neuron. Using transfer entropy to search for and detect directed information was shown to lead to inaccurate assessments in simple case studies [ 18 , 19 ]. For instance, James et al. [ 18 ] presented two examples in which TE underestimates the flow of information from inputs to output in one example, and overestimates it in the other. In the first example, they define a simple system with three binary variables X , Y , and Z where Z t + 1 = X t ⊕ Y t (⊕ is the exclusive OR logic operation) and variables X and Y take states 0 or 1 with equal probabilities, i.e., P ( X = 0 ) = P ( X = 1 ) = P ( Y = 0 ) = P ( Y = 1 ) = 0.5 (this 2-to-1 relation is schematically shown in Figure 1 A). In this network, TE X → Z = TE X → Z = 0 whereas the entropy of the process Z , H ( Z ) = 1 bit , and variables X and Y certainly influence the future state of Z . In this example, the entropy of Z can be reduced by 1 bit but the TE does not attribute this entropy to either variables X or Y and as a consequence the TE underestimates the flow of information from X and Y to Z . In another example, they define a system with two binary variables Y and Z , where Z t + 1 = Y t ⊕ Z t and similar to the previous example, P ( Y = 0 ) = P ( Y = 1 ) = P ( Z = 0 ) = P ( Z = 1 ) = 0.5 (this feedback loop relation is schematically shown in Figure 1 B). In this scenario, TE Y → Z = 1 bit , which implies that the entire 1 bit of entropy in Z is coming from process Y . However, this is not correct since both Y and Z are equally contributing to determine the future state of Z . In this example, TE overestimates the information flow from process Y to Z . It is also noteworthy that in this example the processed information (defined as I ( Z t : Z t + 1 ) ) vanishes, which again does not correctly detect the other source, Z t , from which the information is coming. As acknowledged by the authors in [ 18 ], expecting that the entropy of the output H ( Z t + 1 ) is given simply by the sum of the transfer entropy from each of the inputs independently is a naive interpretation of information flow. Indeed, this is generally not the case, even if the two sources are uncorrelated. Consider for example, the first system described above in which Z t + 1 = f ( X t , Y t ) . Suppose f is a deterministic function of X t and Y t , in which case the conditional entropy H ( Z t + 1 | X t , Y t ) = 0 . Then, the entropy H ( Z t + 1 ) decomposes into the sum of an unconditional and a conditional transfer entropy\n (2) H ( Z t + 1 ) = TE Y → Z + TE X → Z | Y t , \nwhere the conditional transfer entropy is defined as (see [ 13 ], Section 4.2.3)\n (3) TE Y → Z | X t = I ( Y t : Z t + 1 | Z t , X t ) . Using this definition, it is easy to show that\n (4) TE Y → Z = TE Y → Z | X t + I ( X t : Y t : Z t + 1 | Z t ) , \nand Equation ( 2 ) can be rewritten in terms of transfer entropies only, or else conditional transfer entropies only, as\n (5) H ( Z t + 1 ) = TE Y → Z | X t + TE X → Z | Y t + I ( X t : Y t : Z t + 1 | Z t ) = TE Y → Z + TE X → Z − I ( X t : Y t : Z t + 1 | Z t ) . In light of Equation ( 5 ), it then becomes clear that the naive sum of the transfer entropies TE X → Z and TE Y → Z (or naive sum of conditional transfer entropies) must fail to account for the entropy of Z whenever the term I ( X t : Y t : Z t + 1 | Z t ) is non-zero, and therefore will fail to fully and accurately quantify information transferred from sources X and Y . Therefore, the error in information flow estimate when using transfer entropy is simply given by the absolute value of I ( X t : Y t : Z t + 1 | Z t ) (same when using conditional transfer entropies). Now consider the second example system with a feedback loop in which Z t + 1 = f ( Y t , Z t ) , and again suppose f is a deterministic function which implies H ( Z t + 1 | Y t , Z t ) = 0 . In this case, there is a similar information decomposition that now involves a shared entropy I ( Y t : Z t : Z t + 1 ) \n (6) I ( Y t : Z t + 1 ) = TE Y → Z + I ( Y t : Z t : Z t + 1 ) . Here, the entropy H ( Z t + 1 ) can be written in terms of transfer entropy and processed information (recall that H ( Z t + 1 | Z t , Y t ) = 0 )\n (7) H ( Z t + 1 ) = TE Y → Z + I ( Z t : Z t + 1 ) . While Equation ( 7 ) shows that the sum of transfer entropy TE Y → Z and processed information I ( Z t : Z t + 1 ) account for all the entropy Z t + 1 , these two terms do not always individually identify the sources of information flow correctly. For instance, we have seen that in the second example (where Z t + 1 = Y t ⊕ Z t ) the processed information I ( Z t : Z t + 1 ) vanishes even though variable Z t most definitely influences the state of variable Z t + 1 . As discussed earlier, all the information transferred to Z t + 1 in that case is attributed to variable Y t . Note that the processed information can be written as\n (8) I ( Z t : Z t + 1 ) = I ( Z t : Z t + 1 | Y t ) + I ( Z t + 1 : Z t : Y t ) \nwhere I ( Z t : Z t + 1 | Y t ) = 1 and I ( Z t + 1 : Z t : Y t ) = − 1 . Note that for the most general case where function f can be non-deterministic and the network with or without feedback loop, the full entropy decomposition can be written as\n (9) H ( Z t + 1 ) = TE Y → Z | X t + TE X → Z | Y t + I ( X t : Y t : Z t + 1 | Z t ) + I ( Z t + 1 : Z t ) + H ( Z t + 1 | X t , Y t , Z t ) . There is also another key factor in the examples described above that results in misestimating information flow when using transfer entropy. In both examples, the input to output relation is implemented by an XOR function. For instance, in the first example ( Z t + 1 = X t ⊕ Y t ), the transfer entropy TE X → Z considers X in isolation and independent of variable Y . We should make it clear that it is not the formulation of TE that is at the origin of mis-attributing the sources of the transferred information. Rather, by definition Shannon’s mutual information, I ( X : Y ) = H ( X ) + H ( Y ) − H ( X , Y ) is dyadic, and cannot capture polyadic correlations where more than one variable influences another. Consider for example a similar but time-independent process between binary variables X , Y , and Z where Z = X ⊕ Y . As is well-known, the mutual information between X and Z , and also between Y and Z vanishes: I ( X : Z ) = I ( Y : Z ) = 0 (this corresponds to the one-time pad, or Vernam cipher [ 20 ], a common method of encryption that takes advantage of the fact that I ( X : Y : Z ) = − 1 ). Thus, while the TE formulation aims to capture a directed dependency of information, Shannon information measures the undirected (correlational) dependency of two variables only. As a consequence, problems with TE measurements in detecting directed dependencies are unavoidable when using Shannon information, and do not stem from the formulation of transfer entropy [ 12 ] or similar measures such causation entropy [ 10 ] to capture causal relations. Note that methods such as partial information decomposition have been proposed to take into account the synergistic influence of a set of variables on the others [ 21 ]. However, such higher-order calculations are more costly (possibly exponentially so) and require significantly more data in order to perform accurate measurements. Given the observed error in measuring information flow using TE due to logic gates that encrypt, we now set out to examine how well TE measurements capture information flow when the function is implemented with Boolean functions other than XOR. In particular, we examine every first-order Markov process Z t + 1 = f ( X t , Y t ) where function f is implemented by all 16 possible 2-to-1 binary relations ( Figure 1 A) and quantify the error in information transfer estimate for each of them. Similar to previous examples, the state of variable Z is independent of its past, and inputs X and Y take states 0 and 1 with equal probabilities, i.e., P ( X = 0 ) = P ( X = 1 ) = P ( Y = 0 ) = P ( Y = 1 ) = 0.5 . Table 1 shows the results of transfer entropy measurements for all possible 2-to-1 logic gates and the error that would occur if TE measures are used to quantify the information flow from inputs to outputs. This error is the sum of misestimations in information flow quantified by pairwise transfer entropies TE X → Z and TE Y → Z . As we discussed before, for the XOR relation the transfer entropies TE X → Z = TE Y → Z = 0 , and H ( Z t + 1 ) = 1 which means that TE misestimates the information flow from inputs X and Y by 1 bit (the XNOR is exactly the same). We find that in all other polyadic relations where both X and Y influence the future state of Z , TE X → Z and TE Y → Z capture part of the information flow from inputs to outputs, but TE X → Z + TE Y → Z is less than the entropy of the output Z by 0.19 bits ( TE X → Z + TE Y → Z = 0.62 , H ( Z ) = 0.81 ). In the remaining six relations where only one of the inputs or neither of them influences the output, the transfer entropies correctly capture the information flow. The difference between the sum of transfer entropies, TE X → Z + TE Y → Z , and the entropy of the output H ( Z ) in XOR and XNOR relations, stems from the fact that I ( X : Y : Z ) = − 1 , the tell-tale sign of encryption. Furthermore, while other polyadic gates do not implement perfect encryption, they still encrypt partially I ( X : Y : Z ) = − 0.19 , which we call obfuscation . It is this obfuscation that is at the heart of the TE error shown in Table 1 . We repeated similar calculations for the case of a feedback loop network where Z t + 1 = f ( Y t , Z t ) ( Figure 1 B) and function f can be any one of the 16 logic relations shown in Table 1 . These simple calculations show that in 16 relations including XOR and XNOR, the sum of the transfer entropies, TE Y → Z + I ( Z t + 1 : Z t ) (the formulation for transfer entropy of a variable to itself reduces to processed information I ( Z t + 1 : Z t ) ) is equal to the entropy of the output Z t + 1 as was shown in Equation ( 7 ). However, in XOR and XNOR relations transfer entropy incorrectly attributes all the information to one of the input variables and no influence is attributed to the other. Furthermore, in the polyadic relations other than XOR and XNOR, the transfer entropies TE Y → Z and I ( Z t + 1 : Z t ) differ in value while variables X and Y equally influence the state of the output Z , which is why the TE error in these relations is 0.19 bits. Given that pairwise TE measurements (not taking into account higher-order conditional transfer entropies) only fail to correctly identify the sources of information flow in cryptographic gates and demonstrate partial errors in quantifying information flow in polyadic relations, we now set out to determine how often these relations appear in networks that implement basic cognitive tasks, and how much error is introduced when measuring information flow using transfer entropy. If the total error in transfer entropy measurements of information flow in cognitive networks is significant, an analysis of pairwise directed information among neural components (neurons, voxels, cortical columns, etc.) using this concept is bound to be problematical. If, however, these errors are reasonably low within biological control structures because cryptographic logic is rarely used, then treatments using the TE concept can largely be trusted. To answer this question, we use a new tool in computational cognitive neuroscience, namely computational models of cognitive processing that can explain task-performance in terms of plausible dynamic components [ 22 ]. In particular, we use Darwinian evolution to evolve artificial digital brains (also known as Markov Brains or MBs [ 23 ]) that can receive sensory stimuli from the environment, process this information, and take actions in response (In the following we refer to digital brains as “Brains”, while biological brains remain “brains”.). We evolve Markov Brains that perform two different cognitive tasks whose circuitry is thoroughly studied in neuroscience: visual motion detection [ 24 ], as well as sound localization [ 25 , 26 ]. Markov Brains have been shown to be a powerful platform that can unravel the information-theoretic correlates of fitness and network structure in neural networks [ 27 , 28 , 29 , 30 , 31 ]. This computational platform enables us to analyze structure, function, and circuitry of hundreds of evolved digital Brains. As a result, we can obtain statistics on the frequency of different types of relations in evolved circuits (as opposed to studying only a single evolutionary outcome), and further assess how crucial different operators are for each evolved task, by performing knockout experiments in order to measure an operator’s contribution to the task. In particular, we first investigate the composition of different types of logic gates in networks evolved for the two cognitive tasks, and then theoretically estimate how accurate transfer entropy measures could be when applied to quantify the pairwise information flow from one neuron to another in such simple cognitive networks. We then use transfer entropy measures as a statistic to identify information flow between neurons of evolved circuits using the time series of neural recordings obtained from behaving Brains engaged in their task, and evaluate how successful transfer entropy is in detecting this flow. While artificial evolution of control structures (“artificial Brains”) is not a substitute for the analysis of information flow in biological brains, this investigation should provide some insights on how accurate (or inaccurate) transfer entropy measures could be.",
"discussion": "4. Discussion We used an agent-based evolutionary platform to create digital Brains so as to quantitatively evaluate the accuracy of transfer entropy measurements as a proxy for measuring information flow. To this end, we measured the frequency and significance of cryptographic and polyadic 2-to-1 logic gates in evolved digital Brains that perform two fundamental and well-studied cognitive tasks: visual motion detection and sound localization. We evolved 100 populations for each of the cognitive tasks and analyzed the Brain with the highest fitness at the end of each run. Markov Brains evolved a variety of neural architectures that vary in number of neurons and the number of logic gates, as well as the type of logic gates to perform each of the cognitive tasks. In fact, both modeling [ 41 ] and empirical [ 42 ] studies have shown that a wide variety of internal parameters in neural circuits can result in the same functionality [ 43 ]. Thus, it would be informative and perhaps necessary to examine a variety of circuits that perform the same cognitive task [ 34 ]. An analysis of the evolved Brains suggests that selecting for different cognitive tasks leads to significantly different gate type distributions. Using the error estimate for each particular gate due to encryption or polyadicity, we used the gate type distributions for each cognitive task to estimate the total error in information flow stemming from using transfer entropy as a statistic. The transfer entropy misestimate was 1.33 bits (SE = 0.08) per Brain on average for Brains evolved for motion detection, whereas in evolved Brains performing sound localization the misestimate was significantly higher: 2.39 bits (SE = 0.12) per Brain on average. More importantly, the inherent differences between the two tasks result in different levels of accuracy when using transfer entropy measures to identify information flow between neurons. It is important to note that in calculating these misestimates, we only accounted for the misestimates that result from TE measurements in polyadic or cryptographic gates. However, we commonly face several other challenges when applying the transfer entropy concept to components of nervous systems (neurons, voxels, etc.). These challenges range from intrinsic noise in neurons to inaccessibility of recording data for larger populations of neurons which we discuss in more detail later. We also tested how well transfer entropy can identify the existence of information flow between any pair of neurons using the statistics of neural recordings at two subsequent time points only. Because a perfect model for the “ground truth” of information flow is difficult (if not impossible) to establish, we use an approximate ground truth that uses the connectivity of the network, along with information from the (simplified) logic function to provide a comparison. We find that TE captures many of the connections established by the ground truth model, with a true positive rate (hit rate) of 73.1% for motion detection and 78.7% for sound localization (assuming any non-zero value of transfer entropy implies causal relation). The TE measurements miss some relations from the established ground truth while also providing demonstrably false positives, with a false-alarm rate of 7.7% in motion detection and 18.5% for sound localization. For example, some of the information flow estimates in Figure 6 manifestly reverse the actual information flow, suggesting a backwards flow that is causally impossible. Such erroneous backwards influence is possible, for example, when the signal has a periodicity that creates accidental correlations with significant frequency. Besides these false positives, the false negatives (missed inferences) are due to the use of information-hiding (cryptographic or obfuscating) relations, as discussed earlier. It is noteworthy that in the transfer entropy measurements we performed, we benefited from multiple factors that are commonly great challenges in TE analysis of biological neural recordings. First, our TE measurement results were obtained using error-free recordings of noise-free neurons, while biological neurons are intrinsically noisy. We were also able to use the recordings from every neuron in the network, which presumably results in more accurate estimates. In contrast, in biological networks we only have the capacity to record from a finite number of neurons which, in turn, constrains our understanding of how information flows in the network. Furthermore, by focusing only on information flow from one time step to the next we can evade the complex issues posed by estimating causal influence, which requires finding optimal time delays in transfer entropies. For example, while a signal may influence a neuron’s firing three time steps after it was perceived by a sensory neuron, it must be possible to follow this influence step-by-step in a first-order Markov process, as causal signals must be relayed physically (no action-at-a-distance). As a consequence, when using transfer entropy to detect and follow information flow, we can restrict ourselves to history lengths of 1 ( k = l = 1 ), which significantly simplifies the analysis [ 17 ]. Furthermore, complications arising from discretizing continuous signals [ 15 ] do not arise, nor is there a choice in embedding the signal as all our neurons have discrete states. In principle, extending the history lengths (from k = ℓ = 1 to higher) may be used to get rid of false positives in entropy estimates (even for a first-order Markov process), for the simple reason that the higher dimensionality of state space reduces accidental correlations, given a finite sample set. However, such an increase in dimensionality has a drawback: it makes the detection of true positives more difficult (it increases the rate of false negatives) unless the dataset size is also increased. In many dynamical systems such an increase in data size is not an issue, but it may be very difficult (if not impossible) for smaller systems such as the simple cognitive circuits that we evolve. For those, the number of different “sensory experiences” is extremely limited, and increasing the dataset size does not solve the problem because it would simply repeat the same data. In other words, unlike for large probabilistic systems where generating longer time series will almost invariably exhaustively sample the probability space, this is not the case for motion detection and sound localization. For such “small” systems, increasing the history lengths reduces false positives, but increases false negatives at the same time. Finally, in order to precisely calculate transfer entropy from Equation ( 1 ), the summation should be performed over all possible states of variables X t , Y t , Y t + 1 . Using only a subset of those states when calculating the entropy estimate may result in false positives, as well as false negatives. This is another common source of inaccuracy in TE measurements of neural recordings. Here we were able to generate neural recording data for all possible sensory input patterns and included them in our dataset, yet we still observe the described shortcomings in our results. This brings up another important point to notice, namely, even if we introduce every possible sensory pattern to the network, we do not necessarily observe every possible neural firing pattern in the network, and as a result, we do not necessarily sample the entire set of variable states ( Y t + 1 , Y t , X t ) ."
} | 7,182 |
38257543 | PMC10819133 | pmc | 7,152 | {
"abstract": "The growing demand from the extended reality and wearable electronics market has led to an increased focus on the development of flexible human-machine interfaces (HMI). These interfaces require efficient user input acquisition modules that can realize touch operation, handwriting input, and motion sensing functions. In this paper, we present a systematic review of triboelectric-based contact localization electronics (TCLE) which play a crucial role in enabling the lightweight and long-endurance designs of flexible HMI. We begin by summarizing the mainstream working principles utilized in the design of TCLE, highlighting their respective strengths and weaknesses. Additionally, we discuss the implementation methods of TCLE in realizing advanced functions such as sliding motion detection, handwriting trajectory detection, and artificial intelligence-based user recognition. Furthermore, we review recent works on the applications of TCLE in HMI devices, which provide valuable insights for guiding the design of application scene-specified TCLE devices. Overall, this review aims to contribute to the advancement and understanding of TCLE, facilitating the development of next-generation HMI for various applications.",
"introduction": "1. Introduction The rising consumer demand for enriched and immersive experiences in wearable devices, coupled with a burgeoning need across diverse industries such as gaming, healthcare, and education, has spurred a surge in research devoted to extended reality (XR) technologies [ 1 ] and wearable electronics [ 2 , 3 , 4 ]. The objectives of XR technologies are to deliver multiple sensations, including vision, hearing, touch, temperature, etc., to the user without perception distortion and collect the user’s multi-modal input via the human-machine interface (HMI) [ 5 ]. The seamless interaction and immersive experiences of the XR system heavily rely on the low latency and high accuracy recording of the user’s gesture, body movement, eye movement, voice, and facial expression [ 6 , 7 ]. The current challenges lie in seamlessly integrating with non-flat surfaces, conforming to the contours of the human body, and achieving a compact and space-efficient design. These challenges necessitate the development of flexible HMI [ 8 ]. One of the primary objectives of the intensively studied flexible HMI devices such as touch panels [ 9 ], electronic skin (e-skin) [ 10 , 11 ], and the Internet of Things (IoT) [ 12 , 13 ] is to achieve accurate spatial tracking to implement functions like contact detection [ 14 ], body movement detection [ 15 , 16 ], and handwriting character recognition [ 17 ]. Therefore, contact localization electronics that are crucial in obtaining spatial information have undergone rapid development. Following the pioneering work of Wang et al. in 2012 [ 18 ], who introduced the triboelectric nanogenerator (TENG), the progressive advancements in TENG have demonstrated tremendous potential for application in high-entropy energy collection [ 19 , 20 ]. In addition to energy harvesting, triboelectric sensors also have demonstrated advantages such as low cost, easy fabrication, and a wide range of material options [ 21 , 22 , 23 ]. These advantages benefit the research on triboelectric contact localization electronics (TCLE). Wang et al. pioneered the construction of TCLEs by developing a 6 × 6 triboelectric sensor matrix for pressure mapping [ 24 ] and a 4 × 4 triboelectric sensor matrix for finger touch positioning [ 25 ] in 2013. Building on this foundation, subsequent research has introduced TCLEs utilizing rubber [ 26 ], plastic [ 27 ], and fiber [ 28 ] tribo-surfaces, thereby extending their application to various fields such as touch panels, e-skin, smart fabrics, etc. [ 29 , 30 , 31 , 32 ]. Furthermore, TCLE utilizes contact-induced triboelectric output for contact localization. Therefore, the triboelectric approach has the potential to reduce energy consumption and battery volume in HMI devices, allowing for the development of lightweight and long-endurance designs [ 33 , 34 ]. The key challenge of TCLE lies in improving the resolution while simultaneously addressing the issues of decreased accuracy and an increased number of signal channels. To overcome this challenge, novel working principles such as sensor matrix [ 35 ], parallel line array [ 36 ], and analog localization [ 37 ], and signal processing methods such as deep-learning signal processing [ 38 ] have been proposed. Additionally, as shown in Figure 1 , TCLE has expanded its functionality beyond contact positioning to include sliding detection, input trajectory detection, and handwritten character recognition, stimulating its applications in diverse HMI devices [ 39 , 40 , 41 ]. The current reviews on triboelectric tactile sensors provide a thorough discussion of the theory, structure, and material aspects of the tactile sensor unit [ 42 , 43 , 44 ]. However, these reviews predominantly consider TCLE as an application example of the tactile sensor unit array, lacking in-depth analysis of the contact localization principle and design considerations. Therefore, here, we present a systematic review of TCLE with an arrangement as follows: Section 2 discusses the development path of the working principles of TCLE; Section 3 reviews the design considerations and achievable functions of TCLE; Section 4 presents the various application scenarios that have been developed in HMI devices; and Section 5 provides perspectives on future research of TCLE."
} | 1,381 |
31067695 | PMC6562933 | pmc | 7,153 | {
"abstract": "The halotolerant photoautotrophic marine microalga Dunaliella salina is one of the richest sources of natural carotenoids. Here we investigated the effects of high intensity blue, red and white light from light emitting diodes (LED) on the production of carotenoids by strains of D. salina under nutrient sufficiency and strict temperature control favouring growth. Growth in high intensity red light was associated with carotenoid accumulation and a high rate of oxygen uptake. On transfer to blue light, a massive drop in carotenoid content was recorded along with very high rates of photo-oxidation. In high intensity blue light, growth was maintained at the same rate as in red or white light, but without carotenoid accumulation; transfer to red light stimulated a small increase in carotenoid content. The data support chlorophyll absorption of red light photons to reduce plastoquinone in photosystem II, coupled to phytoene desaturation by plastoquinol:oxygen oxidoreductase, with oxygen as electron acceptor. Partitioning of electrons between photosynthesis and carotenoid biosynthesis would depend on both red photon flux intensity and phytoene synthase upregulation by the red light photoreceptor, phytochrome. Red light control of carotenoid biosynthesis and accumulation reduces the rate of formation of reactive oxygen species (ROS) as well as increases the pool size of anti-oxidant.",
"conclusion": "5. Conclusions This study shows that under conditions of nutrient sufficiency, high intensity red light enhanced the production of carotenoids, mostly β-carotene, by upregulating the entire biosynthetic pathway of carotenoids, and that accumulation of carotenoids was accompanied by the highest rate of O 2 consumption and a low rate of net O 2 evolution. The data support a model of flexible co-operation between photosynthesis and carotenoid production via the plastoquinone pool. Chlorophyll absorption of red light photons and plastoquinone reduction in photosystem II is coupled to oxygen reduction and phytoene desaturation by plastoquinol:oxygen oxidoreductase. Partitioning of electrons between photosynthesis and carotenoid biosynthesis depends on photon flux intensity as well as upregulation of phytoene synthase by the red light photoreceptor phytochrome. Red light control of carotenoid biosynthesis and accumulation reduces the rate of formation of ROS as well as increases the pool size of anti-oxidant. Red light may have industrial value as an energy-efficient light source for carotenoid production by D. salina .",
"introduction": "1. Introduction Carotenoids are orange, yellow or red pigments which are synthesized by all photosynthetic organisms for light-harvesting and for photo-protection, and for stabilising the pigment–protein light-harvesting complexes and photosynthetic reaction centres in the thylakoid membrane. They may also be accumulated by some non-photosynthetic archaea, bacteria, fungi and animals for pigmentation [ 1 , 2 , 3 ]. Carotenoids are also the precursors of a range apocarotenoids of biological and commercial importance, such as the phytohormone abscisic acid, the visual and signalling molecules retinal and retinoic acid, and the aromatic or volatile beta-ionone [ 4 ]. Increasingly sought after as natural colorants, there is accumulating evidence that carotenoids protect humans against ageing and diseases that are caused by harmful free radicals and may also reduce the risks of cataract, macular degeneration, neurodegeneration and some cancers [ 5 , 6 ]. They have also been implicated as the actives for treating diseases associated with retinoids [ 4 ]. In most plants and algae containing chlorophyll a (λ max ~680 nm) and b (λ max ~660 nm), photons with a wavelength of 660–680 nm yield the highest quantum efficiencies. However the solar spectrum at the surface of the Earth is at its maximum intensity in the blue and green regions of the visible spectrum (400–550 nm), which is where carotenoids have strong absorption. In photosynthetic organisms in the light, carotenoids drive photosynthesis by transferring absorbed excitation energy to chlorophylls, which have poor absorption in this range. Carotenoids are also able to protect photosynthetic organisms from the harmful effects of excess exposure to light by permitting triplet–triplet energy transfer from chlorophyll to carotenoid and by quenching reactive oxygen species (ROS) [ 2 ]. Dunaliella salina , a halotolerant chlorophyte, is one of the richest sources of natural carotenoids and, similar to various members of the Chlorophyceae, accumulates a high content (up to 10% of the dry biomass) of carotenoids under conditions that are sub-optimal for growth i.e., high light intensity, sub-optimal temperatures, nutrient limitation and high salt concentrations. In D. salina , the major accumulated carotenoid is β-carotene, which is stored in globules of lipid and proline-rich, carotene globule protein in the inter-thylakoid spaces of the chloroplast (βC-plastoglobuli) [ 7 , 8 , 9 , 10 ]. The pathway for β-carotene synthesis and accumulation in D. salina has been partly mapped out [ 11 , 12 ], but the physiological role and signals triggering its accumulation are not well established. In other members of the Chlorophyceae, such as Haematococcus pluvialis and Chlorella zofingiensis , high levels of oxygen-rich, secondary ketocarotenoids, astaxanthin and canthaxanthin, also accumulate under high light stress or nutrient stress, often in lipid bodies located outside the chloroplast in the cytoplasm. Accumulation of these may also be accompanied by cell encystment. Lemoine and Schoefs [ 13 ] proposed that these carotenoids accumulate as a metabolic means of lowering ROS levels by lowering cellular oxygen concentration, as well as serving as a convenient way to store energy and carbon for further synthesis under less stressful conditions [ 13 , 14 ]. Chemically generated ROS will trigger astaxanthin accumulation [ 15 ] and recently Sharma et al. [ 16 ] showed that a small dose (up to 50 mJ cm 2 ) of short wavelength ultraviolet C (UV-C ) light (100–280 nm) in cultures of either D. salina or H. pluvialis massively increased carotenoids accumulation as well as detached the flagellae to increase cell settling, 24 h after exposure: UV-light exposure is typically accompanied by ROS formation. However in D. salina there may be additional mechanisms leading to carotene accumulation. Jahnke [ 17 ] for example found that whilst supplements to visible radiation of long wavelength ultraviolet A (UV-A) radiation (320–400 nm) specifically increased carotenoid levels and the ratio of carotenoids to chlorophylls in the closely related D. bardawil, neither blue light nor medium wavelength ultraviolet B (UV-B) light (290–320 nm) supplements were similarly effective. In blue light, Loeblich [ 8 ] found that green cells of D. salina with a low carotenoid to chlorophyll ratio had a relatively depressed photosynthetic activity, which was even more exaggerated in red cells with a high carotenoid to chlorophyll ratio. They proposed that blue light, which was absorbed by the accumulated β-carotene, was not available for photosynthetic oxygen evolution. Amotz et al. [ 18 ] on the other hand found a marked photo-inhibition for both red and green cells under high intensity red light, which is absorbed by chlorophylls, but red cells, when transferred to high intensity blue light were seemingly photoprotected. Since the accumulated carotenoids were physically distant from chlorophylls located in thylakoid membranes, Amotz et al. [ 18 ] proposed that in high intensity red light, the carotenoids were unable to provide photoprotection against chlorophyll-generated ROS or quench chlorophyll excited states, supporting the argument that carotene globules may function as a screen against high irradiation in blue light to protect photosynthetic reaction centres in D. salina . Fu et al. [ 19 ] examined the effects of different light intensities of red LED light on carotenoid production in D. salina, and showed that the major carotenoids changed in parallel to the chlorophyll b content and that both carotenoids and chlorophyll b decreased with increasing red light intensity and increased with nitrogen starvation. Light-emitting diodes (LEDs) can be applied to adjust the biochemical composition of the biomass produced by microalgae via single wavelengths at different light intensities [ 20 , 21 , 22 , 23 ] and recently Han et al. [ 20 ] successfully used a low light intensity blue-red LED wavelength-shifting system to increase carotenoid productivity in D. salina . In this paper we explore the effects of red, blue and white LEDs on the growth and content of carotenoids and chlorophyll in four different D. salina strains under nutrient-sufficient conditions using a temperature-controlled photobioreactor (PBR) favouring growth. We show that in this system, cultivation using red LED was particularly effective in supporting a high rate of carotenoid productivity. We suggest that in strains of Dunaliella salina , accumulating carotenoids may be synthesised principally as a mechanism for maintaining cellular homeostasis under conditions which might otherwise lead to over-reduction of electron transport chains, formation ROS and of a hyperoxidant state and ultimately lead to cell death.",
"discussion": "4. Discussion LEDs with different wavelengths have been increasingly used to study the wavelength effects on the growth and productivity of photoautotrophic microalgae, and much effort is being invested to understand the most energy-efficient way to incorporate their use for large-scale algal cultivation [ 20 , 21 , 22 , 23 , 28 , 29 ]. In the present work we explored the effects of using red, white, blue and mixtures of red and blue LEDs at different intensities to evaluate the basis for carotenoid accumulation in strains of Dunaliella salina . The emission spectrum of the red LED used in the present work (625–680 nm) emits photons with the exact range required by molecules of chlorophyll a and chlorophyll b to initiate photosynthesis [ 30 ]. In D. salina, action spectra of O 2 evolution rates show maximum photosynthetic activity within the red absorption bands of the chlorophylls [ 8 ]. Photosystems I and II (PSI, PSII), which both contain chlorophyll a, work together in a series of more than 40 steps that proceed with the efficiency of nearly 100% to transfer electrons from water to nicotinamide adenine dinucleotide phosphate in its oxidised form (NADP + ) [ 2 ]. Consequently the wavelength range of the red LED should be the most efficient emission required for photosynthesis in this alga and deliver the highest specific growth rate. However this also depends on the rate at which the absorbed light energy from any given applied photon flux density is converted to chemical energy: with increasing photon flux density, photosynthesis eventually achieves a light-saturated maximum rate that is limited by the rate of carbon fixation in the Calvin cycle. Spirulina platensis for example exhibited the highest specific growth rate using high intensity red LED [ 28 ]. C. reinhardtii however, showed unstable growth in high intensity orange-red and deep red LED, which ceased completely after a few days and was accompanied by cell agglomeration [ 22 ]: agglomeration is typical of oxidant stress and formation of a hyperoxidant state [ 31 ]. In the present work, we found that in high intensity red light in conditions of nutrient sufficiency, D. salina strains maintained a growth rate at least equal to that in white light or blue LED light, seemingly in contrast with the work of others [ 8 , 18 , 19 ]. However we also found that some but not all strains accumulated carotenoids rapidly, within 48 h of exposure. Carotenoids are known antioxidants that are synthesized by many microalgae as part of the battery of photoprotective mechanisms necessary to prevent photo-inhibition caused by photo-oxidation of photosynthetic reaction centres [ 2 , 32 , 33 ]. Photo-oxidation may occur in photon flux density levels that result in absorption of more light than is required to saturate photosynthesis. At the molecular level, when a photon is absorbed by a chlorophyll molecule, it enters a short-lived singlet excited state ( 1 Chl*): the longer the excitation of 1 Chl* lasts, which increases under saturating light conditions, the greater the chance that the molecule will enter the triplet excited state ( 3 Chl*) via intersystem crossing. 3 Chl* has a longer excitation lifetime and can transfer energy to the ground state of O 2 to form singlet oxygen, 1 O 2 , predominantly at the reaction centre of PSII and, to a lesser extent, in the light-harvesting complexes. Photo-oxidative damage occurs to the photosynthetic apparatus when species such as 1 O 2 react with fatty acids form lipid peroxides, setting up a chain of oxygen activation events that may eventually lead to a hyperoxidant state and cell death. Carotenoids may protect the photosystems by reacting with lipid peroxidation products to terminate these chain reactions; by scavenging 1 O 2 and dissipating the energy as heat; by reacting with 3 Chl* to prevent formation of 1 O 2 or by dissipation of excess excitation energy through the xanthophyll cycle. It is tempting therefore to suppose that the differences observed by different workers simply reflects differences in carotenoids content between different strains, but this does not explain what triggered the differences in carotenoids content. In the D. salina strain CCAP 19/41, accumulation of carotenoids was accompanied by the highest rate of O 2 consumption and a low rate of net O 2 evolution, which might imply 1 O 2 formation and ROS accumulation. In the non-photosynthetic, astaxanthin-accumulating yeast, Phaffia rhodozyma , artificially generated 1 O 2 was proposed to degraded astaxanthin to relieve feedback inhibition of carotenoid biosynthesis and also to induce carotenoid synthesis by gene activation [ 34 ]. However these authors also found that carotenoid biosynthesis was linked to O 2 consumption by a cyanide-insensitive alternative oxidase, serving to consume oxygen without chemiosmotic synthesis of adenosine triphosphate (ATP). In C. reinhardtii, a specific thylakoid-associated, terminal plastoquinol:oxygen oxidoreductase has been identified with homology to the mitochondrial alternative oxidase [ 35 ]. The smaller rate of oxygen uptake compared to mitochondrial respiration suggested a function in directly coupling oxygen uptake and the exergonic reaction of plastoquinol oxidation with plastoquinone reduction by a phytoene/phytoene desaturase couple, to permit endergonic carotene desaturation without ATP involvement [ 35 ]. In D. bardawil , a decrease in oxygen consumption rate coupled to phytoene accumulation caused by norflurazon inhibition of phytoene desaturase also suggests a connection between direct desaturation of phytoene and chloroplastic oxygen dissipation [ 36 ]. In the present work, high intensity red light in conditions of nutrient sufficiency maintained growth at the same rate as in blue or white light, and red light also led to carotenoid accumulation albeit to different extents in different strains. These data support involvement of a plastoquinol:oxygen oxidoreductase as originally proposed for C. reinhardtii [ 35 ], but controlled by red photon flux intensity (see Scheme 1 ). In this scheme, chlorophyll absorption of red light photons is coupled to plastoquinone reduction in photosystem II, and oxygen reduction is coupled to phytoene desaturation by plastoquinol:oxygen oxidoreductase leading to carotenoid accumulation. Partitioning of electrons between photosynthesis and carotenoid biosynthesis would depend on both red photon flux intensity as well as upregulation of phytoene synthase. The observed increase in O 2 consumption coupled to accumulation of carotenoids via the carotenoid biosynthetic pathway would reduce the tendency for 1 O 2 formation under high photon flux and maintain cytosolic redox potential. The coupling of reduction of the plastoquinone pool to carotenoid synthesis driven by chlorophyll absorption of red light may involve the red light photoreceptor phytochrome. Photosynthetic organisms are known to perceive red light signals via phytochrome. The synthesis of phytoene by phytoene synthase is under phytochrome regulation [ 37 , 38 , 39 ] and is upregulated by both red and far-red light [ 37 ]. Red light also lowers the concentration of the transcription factor PIF1, a repressor of carotenoid biosynthesis [ 40 ]. In those strains which do not accumulate carotenoids, alternative mechanisms may serve to consume energy e.g., via NAD(P)H reduction of dihydroxyacetone phosphate to form glycerol [ 41 ]. In support of this model, transfer from high intensity red light to blue with higher energy content caused a massive drop in the accumulated carotenoid content, very high rates of photo-oxidation and low respiratory rates. Carotenoids in both the accumulated pool and in the light harvesting antenna, but not chlorophyll, absorb photons in the range 400–550 nm, exactly overlapping the emission spectrum of the blue LED (440–500 nm). Failure of chlorophyll molecules to use the absorbed energy to reduce the plastoquinone pool would be expected to reduce the rate of electron flux through the plastoquinol:oxygen oxidoreductase as well as uncouple carotenoid synthesis and consequently increase cellular O 2 concentration. This would lead in turn to increased ROS formation by reaction of O 2 with reduced electron transport chains, initiating further oxygen radical chain reactions, and carotenoid oxidation. Furthermore, in continuous high intensity blue LED, growth was maintained without carotenoid accumulation, but transfer to high intensity red LED light stimulated a small increase in carotenoid content, once again putting red light absorption by chlorophylls and transfer of absorbed energy to the plastoquinone pool at the centre of carotenoid biosynthesis. Transfer from blue to red light in the Clark-type electrode would cause absorption of more light than was required to saturate photosynthesis; if upregulation of the carotenoid biosynthetic pathway via phytochrome perception was required before coupling with O 2 uptake via the plastoquinol:oxygen oxidoreductase, this would result in initial increase in the rate of photoinhibition and O 2 uptake in the Clark-type electrode, and consequent loss in chlorophyll content, as was observed. β-carotene accumulation in βC-plastoglobuli has parallels with that for astaxanthin accumulation, serving both as a carbon sink and end-product of an alternative oxygen-consuming biosynthetic pathway that on the one hand, controls over-reduction of photosynthetic (and respiratory) electron transport chains at the same time as removes oxygen from the plastid to limit formation of ROS. It is also able to quench any ROS that form. In blue light it may serve as a screen to absorb excess irradiation [ 7 , 18 ] but clearly offers photoprotection in red light as well. These functions are seen as distinct from its role as an accessory pigment in light-harvesting antennae systems. Recently Davidi et al. [ 10 ] showed that the formation of cytoplasmic triacylglyceride (TAG) under N deprivation preceded that of βC-plastoglobuli, reaching a maximum after 48h of N deprivation and then decreasing. They suggested that βC-plastoglobuli are made in part from hydrolysis of chloroplast membrane lipids and in part by a continual transfer of TAG or of fatty acids derived from cytoplasmic lipid droplets. TAG synthesis represents a pathway for restricting over-reduction of electron transport chains [ 42 ] and its recruitment in formation of βC-plastoglobuli is entirely consistent with steps to dissipate excessive energy absorbed by chlorophyll in high intensity red light. Overall, cultivation with red light may hold potential to enhance carotenoids production in carotenoid-accumulating strains of D. salina . Red light treatment has also been reported as an effective way to accelerate ripening of tomato fruit and increase the content of carotenoids [ 43 ]. Compared to other commonly used approaches to induce carotenogenesis, such as high light stress, high salt stress and addition of hydrogen peroxide or sodium hypochlorite, the use of red light provides a clean, convenient and economic alternative to promote carotenoids production from D. salina in a short time."
} | 5,148 |
39914849 | PMC11843533 | pmc | 7,154 | {
"abstract": "New logic in memory (LiM) architectures, able to unify\nmemory and\nlogic functionalities into a single component, are highly promising\nfor executing self-learning algorithms such as artificial neural networks\n(ANNs), with lower energy requirements. The multigated reconfigurable\nfield effect transistor (RFET) is a novel type of logic device that\ncan be fully reconfigured at run-time, promising to be a very versatile\nplatform for logic applications. If equipped with a memory element,\nthen it would represent the ideal building block for LiM-enabling\nhardware with embedded self-learning capabilities. To reach this goal,\nhere we investigate the integration of a ferroelectric Hf 0.5 Zr 0.5 O 2 (HZO) layer onto dual top gated RFETs.\nWe demonstrate that HZO polarization charges can be successfully employed\nto tune the height of the two Schottky barriers, influencing the injection\nbehavior and allowing the selection of the majority carriers, thus\ndefining the transistor mode and switching it between a fully p-type\ntransport to a prevalently n-type one. Moreover, we show that the\nmodulation strength is strongly dependent on the height of the pulse\nused to polarize the ferroelectric domains, allowing for the selection\nof different current levels. All the different achievable states show\na good retention over time, owing to the stability of the HZO polarization.\nThe limitations of the produced devices are discussed, alongside possible\nmitigation strategies. The presented results demonstrate that ferroelectric\nHZO can modulate the carrier injection across Schottky barriers in\nRFET devices. This approach paves the way for the future realization\nof a fully optimized nonvolatile RFET, an ideal building block for\nnovel LiM hardware, enabling low-power circuits for ANN execution.",
"conclusion": "Conclusions In this work we have presented the fabrication\nand characterization\nof a nonvolatile, SOI-based, reconfigurable field effect transistor\nwhere an HZO ferroelectric segment is precisely placed above the metal–semiconductor\nSchottky barriers. This design choice allows us to investigate the\nimpact of the ferroelectric polarization onto the carrier injection\nacross the metal–semiconductor barriers without altering the\ntransport properties of the semiconducting channel. Importantly, the\nHZO layer is integrated above an ultrathin SiO 2 layer,\nenabling transfer characteristics with very low hysteresis. We have\nshown that, by applying a sufficiently high negative or positive voltage\nto the dedicated program gates through a pulsing sequence, it is possible\nto successfully modulate the carrier injection, changing the observed\nbehavior from unipolar p-type to predominantly n-type and back, without\nrequiring the application of an external voltage. The observed changes\nare proof of a ferroelectric modulation of the Schottky barriers,\nclearly ruling out charge-trapping as the responsible mechanism. We\nhave systematically investigated the device behavior as a function\nof the program pulse height, showing that minimum voltages of −4\nV and +3.5 V are needed in order to fully change the transfer characteristic\nto, respectively, p-type and n-type. The choice of intermediate pulse\nvoltages results in different current levels, i.e., a multitude of\nwell-separated states. Too high of pulse voltages, however, can induce\ncharge trapping, which determines a weakening of the ferroelectric\ninfluence. The examined devices show high stability over time, retaining\nthe stored configuration for at least 5 h. This work paves the way\nfor fully optimized nonvolatile reconfigurable field effect transistors,\na promising building block for scaled, low-power hardware combining\nlogic and memory capabilities.",
"introduction": "Introduction Deep learning algorithms have recently\ndemonstrated extraordinary\nsuccess, rapidly becoming ubiquitous. 1 However,\nenergy requirements and carbon emissions for training and operating\nartificial neural networks (ANNs) are on a steep increase and will\nsoon reach unsustainable levels, given the current hardware. 2 , 3 In modern computing, a great deal of resources are spent moving\ndata from processing units to memory and back, strongly impacting\ndata-intensive applications such as ANNs and ultimately limiting the\ncomputational performances, a problem better known as von Neumann\nbottleneck. 4 Energy efficiency, however,\ncan only be marginally improved through further scaling of the electronic\ncomponents, as this process is becoming ever increasingly complex\nand costly, leading to a clear departure from the exponential trend\nthat was well described by Moore’s law. New logic-in-memory\n(LiM) architectures have been proposed as a\npossible solution to this problem. 5 These\nmodels aim at drastically reducing data transfer by relying on computing\nunits that are also able to locally store information. 6 − 8 Over the recent years, LiM has been experimentally demonstrated\nrecurring to classic charge-based 9 , 10 or resistance-based\nmemories, 6 , 11 , 12 as well as\ninnovative memory concepts based on 2D material heterostructures, 13 ferroelectric field effect transistors, 14 and photonic platforms. 15 In general, a successful LiM-enabling technology must fulfill\nthe\nfundamental requirement of being able to reach a high density of components\nper unit area. This can be achieved by combining scaling with a smart\nchoice of device architecture: while the former option is limited\nby rising costs and complexity, the latter approach has been shown\nto hold great potential for the realization of area-optimized logic\ncircuits. Within this framework, the recent development of a\nreconfigurable\nfield effect transistor (RFET) has introduced a device in which electrical\nbehavior is not fixed at production but can be freely programmed to\nbe either n- or p-type. 16 This flexibility\nis achieved through the addition of two polarity gate metal contacts,\nto tune the Schottky barriers forming between the source (S) and drain\n(D) contacts and the semiconducting channel. This allows one to filter\nthe charge carriers, selecting a hole- or electron-based transport.\nDespite the small penalty introduced by the marginal increase in the\nsingle transistor complexity, RFETs offer a large advantage through\nreconfigurability, allowing for the realization of adaptable circuits\nbased on polymorphic logic gates, 17 with\nan important area reduction compared to the standard CMOS architecture. 18 Thus, it is clear that this innovative\narchitecture represents\na promising building block for the realization of adaptable, highly\ndense, LiM-enabling hardware, if equipped with a nonvolatile memory\nelement. To introduce this functionality, here we investigate the\nintegration of a Hf 0.5 Zr 0.5 O 2 (HZO)\nferroelectric layer, localized precisely below the polarity gates\n(PGs) of a silicon-on-insulator (SOI)-based RFET. Similar approaches,\nalbeit not restricting the ferroelectric influence to the Schottky\nregions only, have been proven viable for the realization of a nonvolatile\nferroelectric memory 19 and artificial synapses. 20 , 21 In this work, we show that by applying programming pulses\nto the\nPGs, it is possible to control the magnitude and direction of the\nferroelectric polarization. In this way, the Schottky barriers that\nregulate carrier injection can be tuned by the remnant polarization\n(P r ) charges without the need for applying\nan external voltage. Thus, when the ferroelectric domains are pinned\nin a specific polarization state, the resulting transport configuration\nis maintained over time. Multiple different states can be achieved\nby tuning the height of the programming pulse, showing a stable separation\nover the chosen time window. In addition to a precise characterization\nof the behavior of the engineered devices, a thorough discussion regarding\ntheir current limitations is also presented, with a focus on some\nviable strategies for their improvement. The data presented\nin this work demonstrate that ferroelectric\nHZO can be successfully employed to modulate carrier injection over\nSchottky barriers. This approach is promising for the future realization\nof a fully optimized nonvolatile reconfigurable field effect transistor.",
"discussion": "Results and Discussion Figure 1 a and b\nshow a SEM image of the realized device, with the highlighted metallic\nterminals (PG, S, D) alongside an inset taken with an optical microscope\nand a schematic representation of the material stack. As described\nin the Experimental Section , the devices are\nbased on a commercial SOI substrate, with a device layer thickness\nof 20 nm on top of a 100 nm thick buried oxide (BOX) and on a Si handling\nlayer, which is used as the back-gate (BG). Once the Si nanosheet\nstructures are defined through laser lithography and reactive ion\netching, the dielectric stack is grown following a two-step process.\nFirst, a 0.9 nm thick SiO 2 layer is grown via chemical\nprocessing, followed by the deposition of an amorphous 8.5 nm thick\nHZO layer via ALD. The importance of the thin SiO 2 layer\nis clarified later in the discussion. Once the dielectric stack is\ncompleted, the metallic contacts are defined using laser lithography.\nThe dielectric layer is removed via Ar + sputtering, followed\nby Al sputtering to form the two S and D contacts. The dual PGs are\ndefined by sputtering TiN, after an additional lithographic step.\nMore details on the fabrication process can be found in the Experimental Section . Once the stack is completed,\na rapid thermal annealing step is performed, with the 2-fold purpose\nof enabling the Al–Si exchange reaction 22 and the crystallization of the HZO layer. 23 This causes Al to diffuse along the Si nanosheet, reaching\nthe area below the two fingers of the PG structure. Thanks to the\ntransparency of the thin TiN layer, the two interfaces formed between\nthe Si segment and the diffused Al are clearly visible below the PG,\nas shown in the inset of Figure 1 a. The precise position and sharpness of both Al–Si\njunctions below the PG are essential to ensure a good electrostatic\ncontrol of the Schottky barriers, fundamental for the functionality\nof the devices. The channel lengths are measured to be 3.9 ±\n0.9 μm. The variability is presumably due to local changes in\nthe temperature and etch rates. It must be noted that classic (i.e.,\nvolatile) RFETs are expected to operate without degradation for channel\nlengths as small as 8 times the screening length, as demonstrated\nvia TCAD simulations. 24 Figure 1 a) Scanning electron\nmicroscope image of one of the realized device.\nLabels identify the source (S), drain (D), and polarity gate (PG)\ncontacts. The inset shows an optical microscope image of the same\ndevice. The Al–Si interfaces can be identified below the transparent\nTiN PG gates. b) Cross-sectional representation of the realized stack.\nThe HZO layer ferroelectric domains are located only under the PG\narea. c) Adopted pulsing and transfer measurement schemes, alongside\nthe associated device configurations. d) Schematic representation\nof the band structure in the Si segment, with emphasis on the Schottky\nbarrier configuration. Depending on the direction of the remnant polarization\nvector (Pr), the barriers are influenced to block the injection of\nelectrons (blue) or holes (red). e) Transfer characteristics obtained\nby sweeping the back gate of the device after a polarization sequence\nhas been applied. Respectively, the blue curve is obtained after\na pulse height of −4.25 V is applied, while the red curve is\nobtained with a pulse height of +4.5 V. The curves are associated\nwith the band structures of the same color depicted in panel d. As mentioned earlier, the annealing process triggers\nthe crystallization\nof the amorphous HZO layer, giving rise to different crystalline phases\ndepending on which material is interfacing the oxide layer. The regions\nof HZO that are in contact with TiN are subjected to an in-plane stress\nthat leads to the formation of a ferroelectric orthorhombic phase,\nin a mix with other non-ferroelectric crystalline structures. Differently,\nthe material not in contact with TiN crystallizes only in the non-ferroelectric\nphases, giving rise to a trivial dielectric layer. 25 As summarized in Figure 1 b, the two ferroelectric segments are therefore located\nbetween the two TiN fingers and the Al–Si interfaces, making\nit possible to control the ferroelectric remnant polarization via\nthe PGs. At the same time, the Schottky barriers at the Al–Si\ninterfaces can be strongly influenced by the HZO remnant polarization,\nas a consequence of their extreme vicinity, determining the overall\nelectrical behavior of the device. To obtain the data presented\nin this work, only two different electrical\noperations are performed, denoted “SET” and “READ”,\nschematically represented in Figure 1 c. As the name suggests, the “SET” operation\nis used to change the polarization state of both ferroelectric segments.\nThis is done by pulsing a voltage of sufficient height ( V PULSE ) to the PG, while all other terminals\nare kept grounded. The “READ” operation, instead, corresponds\nto measuring I DS while\nsweeping the back-gate potential ( V BG ) within values low enough to not trigger a change in the HZO\npolarization state. As clearly indicated in the scheme, during this\noperation, the PG terminals are fixed at 0 V. This is different from\nthe normal operation of a classic, volatile RFET device, where the\npolarity is imposed by applying a fixed voltage at the PG. In the\ncurrent work, however, the net electric dipole of the aligned ferroelectric\ndomains is able to modulate the Schottky barriers at the Al–Si\ninterfaces in a nonvolatile manner, determining the preferential injection\nof holes or electrons without the need of applying, respectively,\na negative or positive external voltage to PG. The energy landscapes\nfor the two described states are shown in Figure 1 d. When a sufficiently negative V PULSE is applied onto the PG, the ferroelectric\ndomains align to form a dipole oriented toward the PG contacts, resulting\nin a net negative charge in the vicinity of the semiconductor interface,\nas shown in the upper portion of the scheme. This determines a local\nrise in energy of the bands, which leads to a modification of the\nSchottky barriers in a way that enables hole tunneling, while impeding\nelectron injection. During the “READ” operation, V BG is swept in a way that enables\nthe flow of charges, as depicted in the band scheme with a solid line,\nor that blocks it, as indicated by the dashed line. It must be noted\nthat the Schottky barriers at the Al–Si interfaces are almost\ntotally influenced by the HZO layer, as these regions are extremely\nsharp and close to the ferroelectric segments. While it is true that V BG affects the whole device,\ninterface regions included, its effect on the HZO segment is negligible,\nthanks to the thickness of the buried insulation layer. This can be\ndeduced from the fact that only minor changes can be observed between\nthe currents measured during the forward and backward sweeps, indicating\nno change in the remnant ferroelectric polarization. The “READ”\noperation results in the measurement of a p-type transfer characteristic,\nas shown in blue in Figure 1 e, as a result of the ferroelectric-mediated band-bending.\nOn the contrary, with the positive and sufficiently high V PULSE , the ferroelectric dipole is oriented\nin the opposite way, resulting in a net positive charge in the vicinity\nof the semiconductor interface, as shown in the bottom part of the\nscheme. This leads to the lowering of the bands in correspondence\nwith the Al–Si interface regions, enabling a preferential injection\nof electrons, resulting in a more n-type-oriented transfer characteristic,\nas shown in red in Figure 1 e. Clearly, the two states produce strongly different outcomes,\nwith defined on and off regions. The measured transfer characteristics\nalways show an asymmetric behavior between the two states with higher I ON on the p-side. In addition\nto this, the threshold voltage of both curves is clearly shifted toward\nhigh positive V BG values.\nThe latter observation is normally explained by the presence of a\ncertain amount of negative fixed charges near the semiconducting channel. 26 At the same time, an abundance of negative charges\ncould determine an upward band-bending at the Al–Si interfaces,\nscreening the effects of the ferroelectric polarization and determining\nthe preferential injection of holes, thus explaining also why the\ndevices are leaning toward a p-type behavior. The presence of fixed\ncharges is a well-known characteristic of the high-k/Si systems, 27 − 30 determined by a wide range of causes. 31 − 34 In the specific case of HfO 2 and ZrO 2 , most trap states are related to the\npresence of defects in the layer, with an important role taken by\noxygen vacancies and oxygen interstitials. 34 − 36 First-principle\ncalculations demonstrate that trap levels associated with oxygen vacancies\nlie close to the Si gap, while those caused by oxygen interstitials\nare located deep below the Si valence band edge. This means that the\nlatter may not act as a carrier trap but could, instead, represent\na source of negative fixed charge. 34 , 36 Generally\nspeaking, the trappy interface between high- k dielectrics\nand Si can seriously influence the transport behavior\nin different ways. 37 − 39 First of all, a high concentration of interface traps\ncan drastically degrade the carrier mobility and increase the hysteresis\nof a device, limiting its performances. Second, due to the brief but\nextreme band bending taking place during a “SET” operation,\neven the deep trap states become accessible to the carriers, getting\ncharged for a long period of time. This is detrimental to the expected\noperation of the device, as the electric field induced by charge trapping\nis always opposed to the ferroelectric-associated one, thus, reducing\nor even nullifying its control over the Schottky barriers. To minimize\nthis problem, a thin SiO 2 layer ( t ∼\n0.9 nm) is formed at the Si surface by chemical treatment, as described\nin the Experimental Section , prior to the\ndeposition of the HZO layer. This helps to obtain a very low hysteresis,\nas seen from the dual-sweep transfer characteristics of Figure 1 e. The thickness of the interface\noxide is kept as low as possible to maximize the influence of the\nferroelectric dipole on the Schottky barriers. At the same time, however,\nthis choice has the disadvantage of keeping the high fixed charges\nclose to the semiconducting channel, producing the effects described\nabove. Nevertheless, due to the extreme thinness of the interface\nlayer, it is reasonable to assume that charges can tunnel through\nit, even at low V PULSE . It must be stressed that this mechanism of charge injection is\nof great importance for stabilization of the ferroelectric effect,\nas it was clearly demonstrated through experiments and simulations.\nIn particular, it was shown that the injection and subsequent trapping\nof charges in the vicinity of the ferroelectric layer is a fundamental\nprerequisite for enabling polarization switching in ferroelectric\ntunnel junctions (FTJs) 40 as well as in\nferroelectric field effect transistors (FeFETs). 41 Moreover, charge trapping at the ferroelectric interface\nimproves the retention of the spontaneous polarization 42 but can also limit the achievable currents. 43 Thus, the density of trap states at the semiconductor–dielectric\nand dielectric–ferroelectric interfaces, together with the\namount of charges injected through the SiO 2 layer, are\nparameters that can greatly influence the device behavior. It is,\nhowever, critically important for the correct behavior of the device\nto control the amount of trapped charges, as their excess can negatively\nimpact the endurance of a device. 44 This\ntype of investigation, however, exceeds the scope of this article\nand will be addressed in the future. Overall, the produced devices\nare all able to achieve clearly separable\nstates, which are, as we will later show, very stable over time, a\nfundamental requirement for the conceived use. It is very important\nto notice that the two states shown in Figure 1 e are characterized by a current ratio larger\nthan 3 orders of magnitude for V BG = 0 V, a very important parameter for exploiting the device\nmemory capability in the most efficient possible way. Before\ndiscussing the stability of the devices, we first focus\non investigating the response to different V PULSE magnitudes. Figure 2 a shows different transfer curves obtained\nafter programming the device using an increasingly higher positive V PULSE . After each measurement,\nthe device is reset by applying a sufficiently high reverse V PULSE of −3 V. Therefore,\nwe always investigate the transition from an almost ambipolar state\nto an n-type one. The device starts to show a behavioral change when V PULSE = +2.5 V is used, with\na strongly increasing I n and a slightly decreasing I p . This denotes that the barriers are already changed by the\nHZO influence, with an increase in probability for electron injection\nand a slightly higher blockage for holes. Thus, a mixture of carriers\ncan be injected into the semiconducting channel, therefore limiting\nthe capability of V BG to efficiently block the transport, explaining why I OFF is still very high. When V PULSE reaches +3.5 V, a more\ndrastic change is observed, with a strong reduction of I p . By further increase of the amplitude\nof V PULSE , a stronger\nreduction of I p takes\nplace. I n , differently,\nslightly increases for V PULSE = +4.5 V, while marginally decreasing when V PULSE reaches +5 V. It is important to\nnotice that the measured curves do not shift horizontally, showing\nthat the polarization charges are accumulating, as expected, only\nat the Al–Si interfaces. The color-map on the right ( Figure 2 b) shows in a more\nimmediate way the change of I DS ( V BG ) as a function\nof V PULSE magnitude.\nAs described earlier, by increasing V PULSE , I p decreases starting from +2.5 V, while I n increases, until a V PULSE of +4.5 V is reached. This observation can be\nexplained by considering different effects: first, it is well-known\nthat the remnant polarization of a ferroelectric layer is limited\nin magnitude and cannot surpass the value achieved when all the domains\nare aligned parallel to each other. Therefore, the influence of the\nferroelectric layer over the Schottky barriers will not increase indefinitely,\nbut will reach a maximum level at a certain V PULSE . In addition to this, it must also be\ntaken into consideration that, when a “SET” operation\nwith a very high V PULSE is performed, the band bending at the interfaces is quite extreme,\npossibly letting some charges tunnel through the SiO 2 interface\nlayer and occupy certain deep traps. This implies that for an increasing V PULSE magnitude, the ferroelectric\neffect will be reduced by the competing trapping field. Moreover,\nextreme V PULSE magnitudes\ncan damage the oxide layer, increasing leakage, and thus hindering\nthe ferroelectric response. Figure 2 c shows the maximum I D currents extracted from the p-side ( V BG = −20 V) and n-side ( V BG = +20 V) of the plots shown\nin Figure 2 a. As described\npreviously, the p-side current is suppressed with increasing V PULSE , while the n-side follows\nthe opposite behavior. Clearly, the increase and suppression of the\ncurrents follow an exponential trend as a function of applied V PULSE . Nevertheless, the maximum\nchange observed on the n-side is smaller, with an increase of 2 orders\nof magnitude, compared to the 4 orders of magnitude observable on\nthe p-side. Figure 2 a) Transfer characteristics obtained after applying positive pulses\nof varying magnitudes to the polarity gate. From blue to red, the\nmagnitude of the applied pulses increases from +1.5 V to +5 V. b)\nThe color map shows the measured drain current as a function of the\nback gate voltage and of the chosen pulse height, clearly indicating\na transition from p- to n-type behavior. c) The maximum drain currents\nat the n-side ( V BG =\n+20 V) and at the p-side ( V BG = −20 V) are plotted with respect to the pulse height,\nshowing an exponential modulation. d–f) Same type of plots\nas in a,b,c but for negative polarization pulses. Figure 2 d shows\nthe transfer characteristics obtained after programming the device\nusing an increasingly large negative V PULSE , from −1.5 V to −5 V. In a similar\nfashion, the device shows drastic changes when the pulsed voltage\nis larger than −2.5 V, with both a strong decrease of I n and an increase of I p . Once again, if V PULSE is too large, carriers can tunnel\nthrough the SiO 2 layer and get trapped in deep levels at\nthe high- k interface. In this case, as also noticeable\nfrom the color-map on the right side, the reduction of I p is very evident for V PULSE < −4.5 V. This stronger\ntrapping effect for holes could be determined by the fact that, as\nmentioned before, the HZO defects are acting like donors, thus repelling\nelectrons but favoring holes. Overall, the analysis presented in Figure 2 clearly shows that\ndifferent intermediate states can be obtained by carefully choosing V PULSE , resulting in defined\nand well-separated values of I DS from which to choose. Once again, a color-map is shown in Figure 2 e, to clearly represent\nthe change of I DS ( V BG ) as a function of V PULSE magnitude. The maximum I D currents from the p-side\nand n-side are reported in Figure 2 f for increasingly negative V PULSE levels. From this figure, an exponential\nmodulation of the current is observable. Clearly, the peak reached\non the p-side is slightly reduced for the largest values of V PULSE due to the trapping mechanism\nthat was addressed just above. A very important characteristic\nof the developed devices is their\nobserved stability over time. Figure 3 shows the data acquired over a period of ∼6\nh. It must be noted that the data shown in Figure 3 are obtained using a different device compared\nwith the previous plots. This was necessary in order to observe the\nbehavior of a pristine device, as the previously used one was affected\nby a certain amount of charge trapping due to the application of a\nhigh V PULSE , as shown\nin Figure 2 . In order\nto perform the measurements shown in Figure 3 , the device is set into one of the states\ndescribed earlier, by performing a “SET” operation with\neither +5 V or −3 V V PULSE . After the device is set into a specific state, its transfer\ncharacteristic is measured every 10 min. In between measurements,\nthe device is kept connected to the probe station with grounded voltages.\nThe measured data are shown in Figure 3 a and b. Clearly, the device shows remarkable stability\nwithin the probed time window, with only minor changes in the I ON and I OFF or the I n / I p ratio.\nThis shows that almost no depolarization happens during this time,\nkeeping the influence of the ferroelectric layer on the barriers unchanged.\nAn interesting parameter to track over time is represented by the I ON for different V PULSE levels. This is reported in Figure 3 c and d for I ON at V BG = −20 V ( I p ) and at V BG = 20 V ( I n ),\nrespectively. Similarly to what was observed in Figure 2 , different levels of I n and I p can be observed as a result of certain choices for V PULSE . Importantly, all of\nthe observed states are similarly stable over time, keeping a clear\nseparation through the whole measurement window. However, as time\npasses by, I p shows a\nslightly upward trend, especially marked for negative values of V PULSE . This behavior is in\ncontrast with the expected decrease of the ferroelectric effect over\ntime due to the onset of depolarization. Despite not having grasped\nyet a full explanation for the observed behavior, we speculate that\nthis is the result of a competition between the electrostatic effect\nof trapped charges and ferroelectric polarization. As observed in Figure 2 , charge trapping\ncan occur during the “SET” operation due to the extreme\nband bending, allowing charges to tunnel through the SiO 2 interface and to occupy empty levels. Thus, at the beginning of\na retention test, the Schottky barriers feel the combined influence\nof these two effects, which are opposing each other, as explained\nearlier. As time passes, the trapped charges are partially released\nat a rate faster than that of the depolarization of the ferroelectric\nlayer, thus resulting in a perceived increase of the ferroelectric\neffect. This, however, is only visible for I p , implying that the trapping of holes is\nmore favorable, in accordance with the earlier argumentation. Figure 3 a) The device\nis polarized using a pulse height of +5 V. The panel\nreports several transfer curves measured every 15 min, up to 6 h,\nshowing little change. b) Same as a, but after polarizing the device\nwith a pulse height of −3.5 V. c) Time-dependency of the drain\ncurrents measured at the p-side ( V BG = −20 V) for different polarization pulse heights.\nd) Time-dependency of the drain currents measured at the n-side ( V BG = +20 V) for different polarization\npulse heights."
} | 7,319 |
36626561 | PMC9934207 | pmc | 7,159 | {
"abstract": "Significance Understanding the mechanisms structuring soil bacterial diversity is central to predicting how organisms and communities respond to biotic/abiotic disturbances. Metabolic theory has provided a framework to explain patterns of physiology and diversity in ecological communities. We established a quantitative model to incorporate pH into metabolic theory to capture some of the unexplained variation in bacterial diversity across scales. We combined laboratory experiments at the level of a single species with meta-analysis at the level of community at continental and global scales to build predictive models of species and community diversity. The conceptual framework firstly incorporated pH into metabolic theory to advance accuracy in model predictions of bacterial diversity. Our study allows for further incorporation of multiple factors into MTE-based models.",
"discussion": "Discussion Our study provides the advances in quantitative models aiming at exploring the influence of temperature and pH—and their interactive effects—structuring patterns of bacterial diversity in soils across distinct systems and scales ( Fig. 1 ). The explicit consideration of temperature and pH ranges into the models was based on their known effects on bacterial cell metabolism ( 5 , 6 ). These models were empirically tested across distinct scales of biological organization, ranging from single species at the level of individual bacterium strain to soil bacterial communities within an ecosystem gradient and across global ecosystems. Advancing research on the development of quantitative models able to explain bacterial diversity across spatial scales is still a challenge. For example, Okie et al. (2015) proposed a model primarily based on environmental filtering associated with metabolic theory ( 11 ). In line with this study, our models provide a new synthesis that further integrates diversification rate and pH across scales with principles of the MTE. Despite the fundamental importance of temperature and pH in structuring bacterial diversity in soils has long been recognized ( 1 , 2 ), relatively less attention has been given to integrating these variables into predictive models of biodiversity. A mechanistic understanding of how these factors jointly structure patterns of soil diversity can enhance our ability to prospectively predict dynamic changes in bacterial communities and the impacts on their functioning across distinct spatial scales and systems. The outcome results of the models confirmed our hypothesis that the growth rate and diversity of P. fluorescens were highly correlated and displayed similarly a high dependence on temperature and pH ( Fig. 3 ). Interestingly, corroborating our single species level experiment, these relationships were also ubiquitous across communities both at the continental and global scales. Together, our models provide a new framework that successfully and more accurately integrates the theoretical predictions of cell metabolism as the foundation of all life processes with patterns of biodiversity. As predicted by Model I, we observed the logarithmic of the diversity index at the species level to have a strong Boltzmann exponential relationship with the reciprocal of temperature (1/ KT ) ranging from 5 °C to 28 °C ( P < 0.01), with a slope value (− E a ) of −0.62 ( Fig. 2 A ). Higher temperatures within this range tend to accelerate metabolic rates and biochemical processes and thereby should promote bacterial diversity. Particularly for microbes, this would increase ‘effective’ evolutionary time, given their general shorter generation times, faster mutation rates, and faster selection ( 12 , 30 ). Considering higher temperature would impose a physiological constraint on organisms ( 12 , 31 ), our model assumed that both growth and mutation rates increased exponentially with temperature over a biologically realistic range of temperature. However, the Boltzmann relationship between temperature and species life processes involved in growth rates and diversity disappeared once the temperature exceeded 28 °C ( Fig. 2 A ). Furthermore, the enhanced spontaneous mutation rate of P. fluorescens SBW25 was observed at high temperature ( Fig. 3 E ). Although the MTE holds that spontaneous mutation rate is independent of temperature, numerous studies have shown that high temperature can frequently facilitate spontaneous mutation rates of various bacterial species ( 22 , 32 , 33 ). This occurs possibly due to the trade-off between genetic conservatism and variability of bacteria once exposed to stringent habitat selection ( 34 ). Consequently, speciation rate would be expected to augment the temperature dependence of diversity (i.e., activation energy) compared to the temperature dependence of metabolic rate (and generation time) alone. As such, the speciation rate strengthens the relationship between the logarithm of species diversity and the inverse of absolute temperature, as indicated by the steeper slope. However, our results showed that the slope values (e.g., the inverse number of activation energy, − E a ) of negative diversity-temperature relationships in natural ecosystems are lower or even opposite when compared to the model’s expectation at the community level at both continental and global scales, which does not follow the Model I predictions. This result can be explained by distinct non-mutually exclusive points. That is, the variable diversity is not proportional to the speciation rate, and the effective activation energy for metabolic rate (or speciation rate) (here parametrized as 0.65 eV) may likely differ across distinct microbial taxa. This also may be due to other abiotic factors—in particular pH and salinity—known to explain the variation in microbial community in soils ( 35 , 36 ), potentially weakening the temperature-diversity relationship at the community level ( SI Appendix , Fig. S2 ). For instance, temperature and pH exhibited counteractive effects in the paddy soil samples ( R 2 = 0.63, P < 0.001, SI Appendix , Fig. S14 ), with higher bacterial diversity detected in colder samples being found at closer-to-optimal pH values, which led to a positive diversity (ln H )-temperature (1/ KT ) relationship. By using Model II, we found the diversity of P. fluorescens SBW25 to follow a hump-shaped function over a wide range of pH values ( Fig. 2 B ). This is consistent with previous research ( 37 , 38 ), and our results also suggested that the membrane protein activity, the bacterial growth rate, and the bacterial metabolic rate are all expected to display a hump-shaped function relationship with pH ( Fig. 3 G and H and SI Appendix , Figs. S4 and S5 ). Thus, the consistent relationship across ontogeny to ecological processes likely implies a certain unified mechanism of metabolic processes. The metabolism of individual living cells is largely dependent on extracellular pH, and this occurs because distinct ranges of pH will impact the functional performance of a variety of proteins in the cell membrane ( 25 ). It is worth mentioning that the pH-metabolic model is supported by the significant correlation between metabolic rates and the number of cell membrane proteins. However, the observed correlation could possibly reflect the increased energy cell expended at non-neutral pH to maintain homeostasis. As such, caution is warranted in interpreting this model, as it is primarily based on correlation and does not aim at providing an integrative mechanistic metabolic understanding. Importantly, pH also affects the PMF, which is involved in the maintenance of cellular homeostasis and provides energy requirements for essential bacterial metabolism processes ( 39 ). The PMF and cell membrane carrier proteins synergistically affect the bacterial metabolic processes when pH changes. However, the diverse and non-uniform PMF responses to pH variation hinder the incorporation of this variable into our metabolic model. As for the community-level comparison, although the R 2 of Model II (both at the continental and global scales) was much lower than that at the single species level under controlled conditions, the model still strongly predicted the relationship between pH and soil bacterial diversity. There were no differences between Model I and Model II for all diversity indexes in terms of R 2 ( Fig. 5 A and B ), and our results do not support an absolute dominance of temperature ( 15 ) or pH 6 in structuring bacterial diversity. In fact, our results suggest that their relative importance may vary across ecosystem types or at the detailed level of distinct bacterial phyla. The observed optimal pH range for ecosystem types and bacterial phyla ranged widely from 6.10 to 9.35 ( Fig. 4 B ). In addition, soil pH also dynamically affects other soil parameters, e.g., nutrient cycling dynamics, organic carbon transformation, soil moisture regimes, and salinity ( 1 , 3 ). And, worth mentioning, at a lower phylogenetic resolution, different microbial taxa are also expected to have distinct pH optimum values. The combined influence of temperature and pH on the patterns of soil bacterial diversity was more accurately predicted by Model III, with R 2 ranging from 7 to 66%, for the patterns of community diversity at the continental and global scales. The two predictor variables improved the accuracy of the model prediction of bacterial diversity patterns ( Fig. 5 ). High temperature (in this case, within the operational temperature range) and optimum pH were found to directly and indirectly result in high bacterial diversity in soils. First, as predicted by our modeling approach, higher temperatures and optimal pH promote higher metabolic rates, growth rates, and shorter population growth cycles ( 12 ). These biological rates set the pace of population dynamics and underlie nearly all biochemical activities at multiple levels of organismal organization ( 12 , 40 , 41 ). Second, it is important to acknowledge that other factors might also affect these observed biodiversity patterns. For example, the growth rate can—to some extent—enhance colonization by favoring dispersal, which would augment richness at optimal pH and warmer temperatures. Besides, optimal pH ranges can be associated with higher speciation rates due to there being a wide variety of environments or a greater number of niches or stable enzymes at this pH. Last, both pH and temperature modulate multiple parameters in soils (e.g., resource availability, organic carbon types, soil moisture regimes, and salinity), all of which exert an effect on bacterial diversity. It is worth discussing that the R 2 values of all three models at the community level were lower than that obtained at the population level under laboratory conditions. The low R 2 value is likely a reflection of the complex interplay of ecological processes and mechanisms operating in structuring bacterial communities from populations to communities and from local to global scales. The different abiotic variables have distinct levels of stringency in imposing species selection. In line with that, pH has been suggested to be a more stringent filter in soil bacterial taxa ( 6 , 17 , 18 ), whereas selection imposed by variation in temperature tends to be weaker. This corroborates the fact that microbial thermal sensitivity is lower than that of macro-organisms and the fact that microbes can persist under harsh—albeit non-stringent—environmental conditions under low metabolic states, dormant, via spore formation ( 15 , 35 ). It can also be reasoned non-deterministic factors—in this case, random dispersal—can partially counterbalance local selection and be a more important factor in structuring communities at broader scales. This nicely aligns with a recent consensus in ecology stating that most of the factors explaining variation in community dissimilarities tend to be scale-dependent ( 42 ). For example, while pH was a strong predictor of community assembly processes at a local scale in salt marsh soil bacterial communities, variation in the content and concentration of organic carbon better explained the assembly processes when the model was applied at a regional scale ( 43 ). Collectively, it is thus expected a steady decrease in the model fit as communities become progressively more diverse and models attempt to cover broader scales of biological organization ( 15 , 16 , 35 ). In conclusion, the novel quantitative models devised here enable reliable quantitative prediction of bacterial diversity in soil, by adding another parameter (i.e., pH) to the current MTE, which solely focuses on the influences of temperature. While it is generally accepted that temperature and pH are two of the most important environmental factors that determine bacterial diversity in soil, their relative importance remains controversial and their potential interactions have rarely been addressed ( 15 – 18 ). We show that the new models fit well the experimental evolution data with a single bacterial and can also explain the microbial community data from previous field studies at regional and global scales. However, given the diversity of soil environments, other environmental factors such as soil moisture, salinity, and resource availability may play dominant roles under certain conditions, it is thus important to further develop the models, allowing the integration of multiple environmental factors. To the best of our knowledge, this study is the first to explicitly incorporate pH into existing metabolism theory, highlighting the capacity to improve our mechanistic understanding of microbial biodiversity biogeography. Improving this mechanistic understanding of the variation in microbial communities across broad environmental gradients will be essential in our efforts to model and forecast the variation in microbial biodiversity under current and future climate scenarios."
} | 3,492 |
20591477 | null | s2 | 7,162 | {
"abstract": "A peptide-based hydrogel has been designed that directs the formation of hydroxyapatite. MDG1, a twenty-seven residue peptide, undergoes triggered folding to form an unsymmetrical beta-hairpin that self-assembles in response to an increase in solution ionic strength to yield a mechanically rigid, self supporting hydrogel. The C-terminal portion of MDG1 contains a heptapeptide (MLPHHGA) capable of directing the mineralization process. Circular dichroism spectroscopy indicates that the peptide folds and assembles to form a hydrogel network rich in beta-sheet secondary structure. Oscillatory rheology indicates that the hydrogel is mechanically rigid (G' 2500Pa) before mineralization. In separate experiments, mineralization was induced both biochemically and with cementoblast cells. Mineralization-domain had little effect on the mechanical rigidity of the gel. SEM and EDXS show that MDG1 gels are capable of directing the formation of hydroxapatite. Control hydrogels, prepared by peptides either lacking the mineral-directing portion or reversing its sequence, indicated that the heptapeptide is necessary and its actions are sequence specific."
} | 288 |
32614188 | null | s2 | 7,163 | {
"abstract": "Red-light bacteriophytochromes regulate many physiological functions through photoisomerization of a linear tetrapyrrole chromophore. In this work, we mapped out femtosecond-resolved fluorescence spectra of the excited Pr state and observed unique active-site relaxations on the picosecond time scale with unusual spectral tuning of rises on the blue side and decays on the red side of the emission. We also observed initial wavepacket dynamics in femtoseconds with two low-frequency modes of 38 and 181 cm"
} | 126 |
38030907 | PMC10686828 | pmc | 7,165 | {
"abstract": "Dimethylsulfoxonium propionate (DMSOP) is a recently identified and abundant marine organosulfur compound with roles in oxidative stress protection, global carbon and sulfur cycling and, as shown here, potentially in osmotolerance. Microbial DMSOP cleavage yields dimethyl sulfoxide, a ubiquitous marine metabolite, and acrylate, but the enzymes responsible, and their environmental importance, were unknown. Here we report DMSOP cleavage mechanisms in diverse heterotrophic bacteria, fungi and phototrophic algae not previously known to have this activity, and highlight the unappreciated importance of this process in marine sediment environments. These diverse organisms, including Roseobacter , SAR11 bacteria and Emiliania huxleyi , utilized their dimethylsulfoniopropionate lyase ‘Ddd’ or ‘Alma’ enzymes to cleave DMSOP via similar catalytic mechanisms to those for dimethylsulfoniopropionate. Given the annual teragram predictions for DMSOP production and its prevalence in marine sediments, our results highlight that DMSOP cleavage is likely a globally significant process influencing carbon and sulfur fluxes and ecological interactions.",
"discussion": "Discussion Before this study, the scale, mechanism(s) and importance of DMSOP cycling in organisms and marine environments were unknown. We found DMSOP at millimolar levels in saltmarsh sediments, which were uniquely higher than DMSP and far more abundant than the 0.14 nM average reported seawater values 14 . These data highlight surface marine sediments, which contain far higher cell densities than seawater 20 , as potential niches for DMSOP production. Thus, the predicted teragram budget for DMSOP 14 was probably vastly underestimated. Above this, a potential role for DMSOP in osmoregulation was elucidated. The role of DMSOP in organisms that accumulate it will probably depend on its concentration, cellular location and catabolism in the host, like DMSP 2 . To be a major osmolyte, DMSOP would have to accumulate to high intracellular concentrations, which is rare in known DMSOP producers 14 , and the ~500-fold higher DMSP seawater concentration over DMSOP would largely favour the former as being imported for osmoregulation. Indeed, DMSOP probably has an antioxidant role in the pelagic DMSOP-producing bacterium P. bermudensis 15 . DMSOP may be more commonly used for osmoregulation in marine sediments where DMSOP was more abundant than DMSP. This study vastly extended the magnitude and biodiversity of DMSOP cleavage, from previously being confined to some marine bacteria, to being present in the most abundant marine bacterial groups and other domains of life, namely bloom-forming algae and pathogenic fungi. We elucidated exactly how these organisms cleave DMSOP, which is via their DMSP lyase enzymes that had varied catalytic efficiencies but similar mechanisms for DMSP and DMSOP cleavage. Moreover, clarification was provided on the potential importance of DMSOP, with DMSP/DMSOP lyase genes being very abundant (in 10–13% of marine prokaryotes) and transcribed in Earth’s marine waters and sediments, particularly from Roseobacter and SAR11, which together can account for ~45% of marine bacteria 35 – 37 and who could use DMSOP as a carbon and sulfur source. Ultimately, this work highlights DMSOP cleavage as a potentially important cog in marine and global sulfur and nutrient cycling, and as a major source of DMSO. It also challenges future studies to gain vital knowledge on the range of DMSOP-producing organisms, their DMSOP synthesis mechanism(s) and the environmental levels of DMSOP, unknown factors at large that are required to fully comprehend the global significance of this recently discovered organosulfur compound."
} | 928 |
28225763 | PMC5337119 | pmc | 7,167 | {
"abstract": "Summary Methane biogenesis in methanogens is mediated by methyl-coenzyme M reductase, an enzyme that is also responsible for the utilisation of methane through anaerobic methane oxidation. The enzyme employs an ancillary factor called coenzyme F 430 , a nickel-containing modified tetrapyrrole that promotes catalysis through a novel methyl radical/Ni(II)-thiolate intermediate. However, the biosynthesis of coenzyme F 430 from the common primogenitor uroporphyrinoge III, incorporating 11 steric centres into the macrocycle, has remained poorly understood although the pathway must involve chelation, amidation, macrocyclic ring reduction, lactamisation and carbocyclic ring formation. We have now identified the proteins that catalyse coenzyme F 430 biosynthesis from sirohydrochlorin, termed CfbA-E, and shown their activity. The research completes our understanding of how nature is able to construct its repertoire of tetrapyrrole-based life pigments, permitting the development of recombinant systems to utilise these metalloprosthetic groups more widely.",
"conclusion": "Conclusion The elucidation of the pathway for coenzyme F 430 biosynthesis ( Figure 5 ) completes our understanding of how the major members of the modified tetrapyrrole family are constructed. By using a rich tapestry of enzymes Nature has shown how it is possible to construct a broad range of complex small molecules, such as heme, chlorophyll, vitamin B 12 and coenzyme F 430 , that are all derived from a common tetrapyrrole template and which are all involved in fundamental cellular processes, ranging from photosynthesis through to respiration. Although the biosynthesis of molecules such as heme and chlorophyll have been understood for some time 12 recent research has led to the determination of the aerobic 25 , 33 and anaerobic 34 pathways for vitamin B 12 biosynthesis and the unexpected discovery of alternative routes for heme synthesis 35 , 36 . By identifying the enzymes responsible for the transformation of sirohydrochlorin into coenzyme F 430 we have been able to show how the assembly of the molecular framework that is used to house nickel is orchestrated and optimised for its role in methanogenesis. Three of these biosynthetic steps require MgATP reflecting the high energetic cost in making this specialised metallo-prosthetic group. Our understanding of F 430 synthesis will not only allow the opportunity to explore the development of recombinant MCR systems, a key component of which requires the synthesis of the essential F 430 coenzyme, but also lead to mechanistic studies of some very interesting enzymes.",
"introduction": "Introduction Coenzyme F 430 is a modified tetrapyrrole that is required by methyl-coenzyme M reductase (MCR), the terminal enzyme in the process of methanogenesis ( Figure 1 ) 1 , 2 . This cofactor is responsible for the generation of about a billion tons of methane gas per annum, roughly one third of which escapes into the atmosphere where it is photochemically converted into CO 2 2 , thus contributing to the greenhouse effect and global warming. More recently, MCR has also been implicated in the process of reverse methanogenesis (anaerobic methane oxidation) 3 – 6 , which is mediated by bacterial/archaeal mats on the ocean floor. MCR is an enzyme ensemble consisting of a dimer of heterotrimers (α 2 β 2 γ 2 ), catalyzing the reversible reduction of methyl-coenzyme M (CH 3 -S-CoM) and coenzyme B (HS-CoB) into the heterodisulfide CoM-S-S-CoB and methane 7 . Central to the mechanism of this powerful redox catalyst 8 , 9 is the nickel porphinoid, coenzyme F 430 (-650 mV Ni + / 2+ redox couple). Despite the indispensable role played by coenzyme F 430 in the process of methanogenesis and carbon cycling, the assembly of this unique cofactor had not been determined 10 . As a modified tetrapyrrole the synthesis of coenzyme F 430 is based upon the macrocyclic template of uroporphyrinogen III 11 , 12 , from which all hemes, chlorophylls, sirohemes, corrins, bilins and heme d 1 are derived. However, coenzyme F 430 differs from these other modified tetrapyrroles in the nature of the centrally chelated metal ion and in the oxidation state of the macrocycle, a tetrahydroporphyrinogen, the most reduced member of the family 13 . As well as the four pyrrole-derived rings found in all modified tetrapyrroles (labelled A-D; Figure 1 ), coenzyme F 430 also contains two extra rings (E and F; Figure 1 ). Ring E is a lactam derived from the amidated acetic acid side chain attached to ring B, whilst the keto-containing ring F originates from the propionic acid side chain on ring D. Radiolabelling experiments indicated that the biosynthesis of coenzyme F 430 proceeds via sirohydrochlorin, the metal-free precursor of siroheme 14 . Moreover, under depleted nickel growth conditions, Methanothermobacter marburgensis was found to accumulate a 15,17 3 - seco intermediate ( seco -F 430 ) missing ring F 15 . This intermediate could be converted into coenzyme F 430 by cell-free extracts in the presence of ATP 15 indicating that this seco -F 430 may represent the penultimate intermediate in the biosynthetic pathway."
} | 1,288 |
28421166 | PMC5376567 | pmc | 7,168 | {
"abstract": "The significance of polymicrobial infections is increasingly being recognized especially in a biofilm context wherein multiple bacterial species—including both potential pathogens and members of the commensal flora—communicate, cooperate, and compete with each other. Two important bacterial pathogens that have developed a complex network of evasion, counter-inhibition, and subjugation in their battle for space and nutrients are Pseudomonas aeruginosa and Staphylococcus aureus . Their strain- and environment-specific interactions, for instance in the cystic fibrosis lung or in wound infections, show severe competition that is generally linked to worse patient outcomes. For instance, the extracellular factors secreted by P. aeruginosa have been shown to subjugate S. aureus to persist as small colony variants (SCVs). On the other hand, data also exist where S. aureus inhibits biofilm formation by P. aeruginosa but also protects the pathogen by inhibiting its phagocytosis. Interestingly, such interspecies interactions differ between the planktonic and biofilm phenotype, with the extracellular matrix components of the latter likely being a key, and largely underexplored, influence. This review attempts to understand the complex relationship between P. aeruginosa and Staphylococcus spp., focusing on S. aureus , that not only is interesting from the bacterial evolution point of view, but also has important consequences for our understanding of the disease pathogenesis for better patient management.",
"conclusion": "Concluding remarks Both P. aeruginosa and staphylococci are highly versatile organisms, which readily adapt to a wide variety of environments and stress factors. In the first glance, these bacteria seem to have an antagonistic relationship as P. aeruginosa produces a wide variety of molecules inhibiting staphylococci and frequently outcompetes S. aureus and S. epidermidis during co-culture. This antagonistic behavior is mainly shown during planktonic growth and under traditional culture conditions, where no host factors or antibiotics are present. However, under some in vitro and in vivo circumstances, both bacteria are able to co-exist and form dual species biofilms. These circumstances are dependent on a combination of strain-dependent properties of both species and the presence or absence of certain environmental factors like antibiotics or host factors, the sum of which might tip the balance toward either killing or co-existence. The presence of some sort of selection pressure or presence of a preformed matrix seems to favor dual species biofilm formation whereas planktonic co-culture without selection pressure leads to domination of P. aeruginosa . Interspecies competition often leads to an increased production of virulence factors in both P. aeruginosa and S. aureus , which are also harmful to the human host. In addition, escaping the antistaphylococcal compounds results in a more stress-resistant phenotype of S. aureus , which is more difficult to be cleared by the immune system, to be eradicated by antibiotics and to be detected in diagnostic cultures. Furthermore, the presence of an extracellular matrix was shown to be beneficial for all biofilm inhabitants, providing protection against classical antibiotics and the host immune system, although the exact composition might be variable depending on the species present and the environment. The role of these matrix components (exopolysaccharides, eDNA, matrix proteins, host-derived factors etc.) in interspecies interactions and their role in disease pathogenesis provides an exciting opportunity for future research toward better patient care. When strains are co-existing for a longer time, they might evolve to a phenotype that is better adapted to the presence of the other. For example, non-pathogenic staphylococci that are frequently encountering P. aeruginosa have developed strategies to continue growing in the presence of P. aeruginosa antistaphylcoccal compounds, indicating parallel evolution. Moreover, S. aureus and P. aeruginosa strains isolated from the same chronic CF lung infection are less sensitive to, and produce less, HQNO, respectively. This strain adaptation and the underestimation of the co-existence of P. aeruginosa and S. aureus might still have a large impact on the clinical outcome of a patient and therefore should be a subject of continuing investigation.",
"introduction": "Introduction Over the past decade there is a growing appreciation that the biofilm mode of growth is the most common lifestyle adopted by bacteria (Hall-Stoodley et al., 2004 ; Burmolle et al., 2014 ). Biofilms can be defined as surface-associated, structured bacterial communities embedded in an extracellular matrix (Hall-Stoodley et al., 2004 ). Living in a biofilm provides protection in a stressful environment where mechanical stress, desiccation, and biocides are common threats (Donlan and Costerton, 2002 ; Flemming and Wingender, 2010 ). Multiple species frequently exist together in a single biofilm, where they either improve the fitness of one another or compete for space and nutrients (Jefferson, 2004 ; Billings et al., 2013 ; Burmolle et al., 2014 ; DeLeon et al., 2014 ). Most bacteria have developed interaction strategies to communicate within and between species in a cell density-dependent manner, for example, by using small diffusible molecules in a process called quorum-sensing (Federle and Bassler, 2003 ; Li and Tian, 2012 ). Furthermore, many bacteria excrete antimicrobial components, also often regulated by quorum-sensing, to eliminate competitors (Federle and Bassler, 2003 ; Li and Tian, 2012 ). Indeed, these multispecies interactions within the biofilm are important for the inhabiting bacteria and, given the increasing evidence of the link between biofilm-associated pathogens and disease, also from a clinical point of view (Donlan and Costerton, 2002 ; Li and Tian, 2012 ). S. aureus and P. aeruginosa are important pathogens causing a wide variety of infections, including pneumonia in cystic fibrosis (CF) patients, healthcare associated pneumonia and chronic wounds (Harrison, 2007 ; Fazli et al., 2009 ; Cystic Fibrosis Foundation Patient Registry, 2012 ). Initially, only an antagonistic relationship between both organisms was described as the presence of one is associated with the absence of the other in CF and both are rarely found in close association in chronic wounds. S. aureus mostly resides on the wound surface whereas P. aeruginosa is found in the deep layers (Kirketerp-Moller et al., 2008 ; Fazli et al., 2009 ). We also recently showed a negative correlation between presence of P. aeruginosa and the total species diversity in in vivo endotracheal tube biofilms and a low co-occurrence of P. aeruginosa with Staphylococcus epidermidis (Hotterbeekx et al., 2016 ). Nonetheless, recent studies have also co-isolated P. aeruginosa and Gram-positive bacteria, including S. aureus , from the same infection site where increased virulence and/or antibiotic resistance is described (Duan et al., 2003 ; Kirketerp-Moller et al., 2008 ; Fazli et al., 2009 ; Dalton et al., 2011 ; Korgaonkar et al., 2013 ). After describing first in vivo observations occurring in human diseases, we will discuss and summarize in vitro data from the current literature on potential mechanisms of interactions between P. aeruginosa and Staphylococcus spp., primarily S. aureus ."
} | 1,876 |
33920295 | PMC8069537 | pmc | 7,169 | {
"abstract": "This paper focuses on the energy generating capacity of polyvinylidene difluoride (PVDF) piezoelectric material through a number of prototype sensors with different geometric and loading characteristics. The effect of sensor configuration, surface area, dielectric thickness, aspect ratio, loading frequency and strain on electrical power output was investigated systematically. Results showed that parallel bimorph sensor was found to be the best energy harvester, with measured capacitance being reasonably acceptable. Power output increased with the increase of sensor’s surface area, loading frequency, and mechanical strain, but decreased with the increase of the sensor thickness. For all scenarios, sensors under flicking loading exhibited higher power output than that under bending. A widely used energy harvesting circuit had been utilized successfully to convert the AC signal to DC, but at the sacrifice of some losses in power output. This study provided a useful insight and experimental validation into the optimization process for an energy harvester based on human movement for future development.",
"conclusion": "5. Conclusions and Outlook In this paper, the effects of PVDF sensor type, geometry, loading frequency and strain on the electrical power generation capability were studied. Two types of movements were emulated, namely ‘bending’ and ‘flicking’. The major findings from the study are summarized as follows. Among the three sensor configurations, parallel bimorph sensor was shown to be the best energy harvester in that it generated the highest amount of electrical power. Practically measured capacitance values for PVDF sensors with large thicknesses were found to be close to the theoretical values [ 23 ]. Given the range of PVDF thickness studied, power output seems to decrease with increasing PVDF thickness. Such trend could be because the stiffness of the PVDF dielectric might have dominated power generation, and the sensor thickness passed the optimum value after which the power output decreased. Power output increased consistently with the increase of sensor surface area. Flicking movement showed greater power output than bending for all surface areas as well as far greater increase in power output with surface area. As for the effect of geometry, power output increased initially with increasing aspect ratio, reaching the maximum value at an aspect ratio of around 2:1, and then decreasing for higher aspect ratios. Mean current and power output showed a significant increase with the increase in loading frequency for flicking. However, for bending, modest increases were noticed initially (3 Hz and 4 Hz) with the output dropping for higher frequency (5 Hz). Power output also followed an increasing trend with the increase of mechanical strain up to a certain value of strain. In addition, an increase in loading frequency resulted in an increase in power generation, however, it did not increase linearly as “wave-cutting” decreased the potential for charge generation at higher frequencies. A simple energy harvesting circuit was found to successfully convert the AC signal generated by the sensor into DC signal, but at the cost of a significant loss in power output. The above findings for PVDF sensors were largely in line with the trends observed using sensors made from other piezoelectric materials, for example, PZT. Although the PVDF sensors studied were not able to generate significant power output to be readily useful in practical applications, some credible relationships and insights on the performance of PVDF sensors were obtained. To realise the full potential of PVDF sensors, future work will focus on more precise preparation of the PVDF sensors with poling, increasing the efficiency of the energy harvesting circuit, optimizing the geometry of the parallel bimorph configuration, and deploying efficient noise reduction techniques. In this study, the resistance load was kept constant for all parametric analysis of the sensors. However, the change in resistance load in the energy harvesting circuit has a significant influence on the current and power output. Using a coupled piezoelectric circuit-finite element method (CPC-FEM) on vibration-based PZT energy harvesting devices, Zhu et al. [ 24 ] reported that when the resistive load within the circuit increases, power output increases up to a peak (maximum) point, and after which, decreases. On the other hand, vibrational amplitude decreases and then picks ups. They found that maximum power occurs at a resistive load of 488 kΩ and vibrational amplitude of 100 µm. The findings imply that the electrical power generation is dependent on the resistive load. Frequency and phase response of the output current and power with respect to the changing resistive load should be analysed to better understand, explain and optimize the power generation capacity of the PVDF sensor [ 25 ]. In addition to noise reduction via grounding through the body, the variation of resistive load must be taken into account to elude the efficiency of the energy harvesting PZT sensors. Furthermore, our future work will aim to address theoretical background and simulation to reveal the efficacy of the sensors. Though, in this study, all experiments were conducted in ambient temperature, the efficiency of power generation can be impacted by the change in temperature. For instance, a recent study by Bernard et al. [ 26 ] reported the increase of dielectric constant of the PVDF with the increase of temperature from 30 °C to 60 °C and strain up to 2.5%, which caused a decrease in efficiency down to 60–75%, as opposed to ideal conversion efficiency of 90%. They suggested that the circuit output capacitance must be decreased in proportion to be adapted to operation conditions in order to restore higher power generation capacity.",
"introduction": "1. Introduction The use of portable electronics has been growing rapidly. As the emerging portable electronic devices are embedding more functionalities, they are becoming complex in nature, requiring higher amount of electrical energy for their operation. With higher levels of energy consumption, the batteries used to supply power to the portable devices deplete their energy quite quickly. Hence, there is increasing demand for alternative energy, and this has facilitated growth in the area of renewable energy sources, which can be harnessed to power portable devices ‘on the go’. Current postulated application areas for portable energy harvesting include wireless sensor networks, biomedical (e.g., pacemakers) and military use [ 1 , 2 ]. In particular, energy harvesting via human body movements (e.g., running, jogging) is seen as a promising means to power portable devices carried by people. In this regard, different types of piezoelectric materials such as piezoelectric (PZT), polyvinylidene difluoride (PVDF), have been studied to understand and evaluate the efficacy of electrical power generation from the mechanical energy harvested from human movements. For example, piezoelectric materials tapped on shoe sole and wearable fabrics have been found to generate power [ 3 ]. As the amount of power generated through such movements is relatively small, the selection of appropriate piezoelectric material along with design of the energy harvesting circuits is crucial to generate usable power. In the past, with the aid of experimental and simulation tools, tremendous efforts have been made to study the design, implementation and evaluation of piezo sensors with the aim of increasing the amount of power generation [ 4 ]. Furthermore, PVDF is flexible, resistant to mechanical shock and highly compatible with the environment, which make it suitable for full PZT beam without needing any substrate layer. As such, researchers focused on gross energy scavenging capacity of PVDF based sensors in different applications, e.g., human walking, elbow bending. However, little research is reported on comprehensive parametric study of PVDF sensors in a holistic way. For example, studies focusing on the effects of different sensor parameters including geometry, mechanical loading, frequency and straining on electrical power generation capability of PVDF sensors have not been reported. This paper presents the effect of the type and geometry of PVDF (polyvinyl difluoride) piezoelectric sensors on the amount of power generated, along with power conversion circuitry to be embedded in wearable-fabrics to harvest energy via human body movements. PVDF was chosen due to its flexibility, low cost, and good piezo-properties, such as high electric charge accumulation under small mechanical strain. Three piezoelectric sensor configurations, namely-parallel bimorph, series bimorph and unimorph sensors were investigated in this study. The effect of dielectric thickness, surface area and aspect ratio, loading type and frequency on the electrical power output of each sensor configuration were studied, with the aim of optimizing the senor parameters. A simple power conversion circuit without a voltage regulator was designed and implemented, and its efficiency in terms of power output and potential issues was discussed in this paper. The results are analyzed with reference to available literature on energy harvesting.",
"discussion": "4. Results and Discussion 4.1. Capacitance As a measure of charge storing capability, the capacitances of the sensors with different geometric configurations were measured using a digital multi-meter. For the purpose of comparison, capacitance values were also estimated using the capacitive circuit model of the sensors [ 20 ]. Table 2 summarizes the measured and calculated capacitances of the sensors. Clearly, for small PVDF thickness (t = 0.0254 mm), the calculated capacitance values were considerably larger than the measured capacitances. This is expected because the theoretical calculation does not take into account the impedance of the epoxy joiner. Sensors with a larger surface area showed a higher capacitance, which was to be expected, and sensors with the same surface area but different aspect ratios had very similar capacitances. For much larger PVDF thickness (t = 0.254 and 0.508 mm), the calculated capacitance value is very close to the measured capacitance. This is due to the fact that in the practical sensor the impedance of the epoxy joiner has lesser effect on the overall capacitance. The capacitance can give us an indication of charge decay time of the actuator and hence its energy harvesting capability. For instance, the largest measured capacitance in Table 2 was 0.34 nF for the PVDF sensor with A = 25 mm × 25 mm and t = 0.0254 mm. This capacitance was approximately 2% of the capacitance of the PVDF sample with A = 240 mm × 240 mm and t = 0.027 mm studied in [ 15 ]. However, it is noted that the surface area of our sensor is only 1.1% of that of [ 15 ], which resulted in much lower capacitance. Therefore, it can be argued that the capacitance generated by the proposed sensor is consistent and reasonable. Results on measured capacitance per unit area as shown in Table 2 follow the similar trend and are consistent to that of [ 15 ] as well. 4.2. Effect of Sensor Type To measure the effect of sensor type on the electrical power generation capability, three different sensor types with the same dielectric thickness and surface area were tested under ‘Flicking’ and ‘Bending’ loading at a frequency of 3 Hz. Figure 5 shows the mean current (Irms) and power outputs (Pout) for PB, SB, and U sensors. For both loading types, clearly the electrical output for parallel bimorph sensor was the highest among all three sensor types. The unimorph sensor exhibited the second highest electrical output. The results are consistent the work of [ 12 ]. Since the parallel bimorph sensor showed the highest current and power output, we used this sensor type for the remainder of the experiments, i.e., for parametric studies, which are described in the following sections. 4.3. Effect of PVDF Thickness To measure the effect of PVDF thickness, parallel bimorph sensors with three different thicknesses (0.0254 mm, 0.254 mm and 0.508 mm) were tested for the same surface area under ‘bending’ and ‘flicking’ loading at a frequency of 3 Hz. Figure 6 shows the mean current and power output with respect to the thickness. It is seen that the current ( Figure 6 a) and power output ( Figure 6 b) decreases with the increase in dielectric thickness under both types of loading. The results are comparable with the work of [ 14 ] who reported that at an optimal dielectric thickness of the sensor, after which, the stored electrical energy decreases with the increase of dielectric thickness. They found that the optimal piezoelectric thickness to substrate thickness ratio was 1.05 for a fixed (200 µm thick) steel substrate, which was independent of geometric dimensions of the sensor. It was postulated that the maximum power output was achieved before the rigidity of the piezoelectric material had a significant effect on the overall rigidity of the sensor. As can be seen from Figure 6 , the power output decreases with the increase of dielectric thickness and the maximum output occurs at substrate thickness of 0.025 mm (first data point in Figure 6 ). PVDF thicknesses we studied were beyond the optimal ratio, which caused a decreasing trend of power output due to the potential dominant stiffness effect of the dielectric material on the overall stiffness of the sensor. For flexible substrate similar to copper shims used in our study, a sharp decreasing trend of power output with the increase of stiffness of polyvinylidene fluoride-trifluoroethylene (PVDF-TrFE) films was reported in [ 21 ]. Therefore, the overall trend of power generation with respect to the dielectric thickness seems reasonably accurate. Note that only three PVDF thickness values were studied in this paper and thus the results obtained are not exhaustive, and a detailed study encapsulating wider range of thickness including thinner PVDF films (<0.025 mm) would be of greater interest to conclusively determine an optimum characteristic of the sensors for future work. 4.4. Effect of Aspect Ratio To measure the effect of aspect ratio (L/W) on electrical power generation capability, three sensors with different ratios (4:3, 25:12, and 3:1) at constant thickness, surface area and configuration were tested at a frequency of 3 Hz for both bending and flicking loads. Mean current and power output with respect to aspect ratio are shown in Figure 7 . As can be seen from the figure, power output increases up to a maximum value and then decreases as the aspect ratio increases. There are two factors to consider when analyzing the power output characteristics: one is the moment, which is the product of the force vector multiplied by the beam length, and the other is the width of the sensor. A longer beam length results in a greater moment, which generates larger electrical energy. On the other hand, with a smaller width, the applied load is distributed across a smaller area at the edge of the bend line, which results in a smaller power output. This would explain the parabolic output characteristics of the sensors ( Figure 7 ). The power increased and reached a peak point, followed by a decreasing trend, as L/W ratio increases. The ideal L/W ratio was found to be about 2:1. The findings are slightly in contrast with that of [ 16 ], where they reported a noticeable increasing trend of power output only when the L/W ratio was greater than 3. 4.5. Effect of Surface Area To measure the effect of surface area on electrical power generation capability, sensors with four different surface areas of 100, 225, 400, and 625 mm 2 with the same thickness and configuration were tested at a frequency of 3 Hz. As can be seen in Figure 8 , mean current and power increase with the increase of surface area. Interestingly, the relationship between power output and surface area is slightly different to that found from the simulation in [ 16 ]. The authors reported that the power output was greatest for an area of 400 mm 2 and then decreased with increasing surface area. They further suggested that this was due to the increased flexural stiffness for larger areas, which would result in less strain for the same applied load. It must be noted that they applied a constant magnitude of loading, whereas our method of loading was aimed at maintaining constant strain. This fundamental difference in loading method could be responsible for the discrepancy in the trends of the results. 4.6. Effect of Loading Frequency To measure the effect of loading frequency on electrical power generation capability, a sensor was tested at frequencies ranging from 1 to 5 Hz, which encompasses equivalent loading frequency on wearable fabrics due to human body movements. All other geometric parameters of the sensor were kept constant. Figure 9 shows the test results for different loading frequencies. The effect of loading frequency had a similar effect to that of surface area for “bending”, the output current increased modestly with frequency up to a point (4 Hz) and then dropped slightly at higher frequencies. The power output increased slightly with frequency up to point (4 Hz) before dropping marginally at higher frequencies. With “flicking”, the current and power output increased almost linearly with frequency all the way up to 5 Hz. The increases were much more pronounced than seen in the case of “bending”. The trends can be explained by carefully observing the instantaneous current waveforms for 3, 4, and 5 Hz, as can be seen in Figure 10 . When the sensor was stressed, the current spiked initially and then discharged. This is consistent with the observations of [ 13 ], although the force application is slightly different. The discharging of a sensor can be characterised by the time constant τ, which indicates how quickly the sensor is able to discharge. The time constants for both ‘flicking’ and ‘bending’ loading at 3–5 Hz were calculated by analyzing the data associated with the instantaneous waveforms shown in Figure 10 . There was a distinct time gap between the positive and negative currents for the “up” and “down” motions with ‘flicking’, whereas the two regions were much closer together for ‘bending’. In the case of “bending”, when the loading was done at 3 Hz, the induced current had enough time between each ‘down’ and ‘up’ motion to discharge. Once the loading frequency reached 5 Hz, there was not enough time for the current to discharge naturally. Therefore, potential energy from that motion was not completely transferred into electrical energy. This led to the logarithmic increase in power output with the increase of frequency. Indeed, the power even dropped at 5 Hz due to this effect. As Figure 10 demonstrates, with the “Flicking” motion, the loading was much more sudden and the induced current had ample time to return to 0 before the next load is applied, so the ‘pulse-cutting’ was not seen. As summarized in Table 3 , the time constants for 3 Hz and 4 Hz of ‘bending’ were larger than that for 5 Hz. This reflects the fact that the 3 Hz and 4 Hz pulses are longer than the 5 Hz pulse, and are able to impart a greater amount of energy per pulse. For all frequencies, the time constants for ‘bending’ were consistently greater than that for ‘flicking’. The results are in line with the findings of [ 13 ] which reported that an increase in static pressure or stress in PZT sensor results in a decrease in decay time constant, and thus suggest that ‘flicking’ imparts greater pressure than ‘bending’. Moreover, the mechanical stability of the flexible electrodes can be compromised under both loading types when subjected to large number of cycles of operation. Eventually, this can cause permanent deformation of the sensor device, which can affect the overall power generation efficiency of the sensor [ 22 ]. This warrants further investigation and remains within the scope of our future work. 4.7. Effect of Mechanical Strain To measure the effect of strain on electrical power generation capability, two sensors with the same PVDF thickness (of 0.0254 mm) and two different surface areas (of 400 and 625 mm 2 ) were tested with different loading in terms of maximum displacements of PVDF sensor end at 1.5, 3.0, 5, and 6.5 mm, which corresponds to strains of 0.00602, 0.0122, 0.0209 and 0.0283 for the 400 mm 2 sample, and 0.00603, 0.0123, 0.0216, and 0.0303 for the 625 mm 2 sample. The values of strains are chosen based on human body movements at key flex points such as the inside of the elbow and behind the knee. The strain was calculated using a simple formula to find the angle of separation between the bent sensor and the bottom step when fully loaded, which is as follows.\n (1) ∆ l = 2 π t s θ 360 \nwhere θ is the angle of separation and, ∆ l is the change in length, and t s is the total sensor thickness. Figure 11 shows the mean current and power output with respect to mechanical strain. It can be seen from the figure that for the sensor with surface area of 400 mm 2 , clearly, mean voltage and current increased when the strain increases from 0.00602 to 0.0122 (first two data points on the graphs). They, however, did not increase as rapidly beyond past the strain of 0.0122. For the sensor with surface area of 625 mm 2 , the voltage and current curves started to flatten after the strain of 0.0216. The limiting strain seems to be higher for sensor with larger surface area. Our results suggest that an upper limit exists on the amount of deformation a material may experience before reaching its optimum charge generating capability. Defects in the sensor construction may also limit the maximum strain for power generation. With a large deformation, the dielectric material may become detached from the conducting substrate, hence increasing the impedance of the gap between the two materials and decreasing the energy generating potential. Therefore, a more thorough analysis would be required to investigate the full effect of this. 4.8. Analysis of Energy Harvesting Circuit Efficiency To test the efficiency of the energy harvesting circuit, the electrical output parameters were measured for different system configurations with a 625 mm 2 sensor with 3 Hz ‘Bending’ loading at strains of 0.0123 and 0.0216. Table 4 summarizes the test results on the performance of the energy harvesting circuit that was presented in Figure 4 b. As the capacitors take a certain amount of time to charge as determined by the RC time constant, τ R C = R l C l , where R l = 10 M Ω and C l = 470 nF and 1 μ F , the measurements could not be taken from the start of the loading, but rather taken from the time at which the capacitor became fully charged. The charging time was determined by collecting data from a “cold start” and determining the point at which the charge stopped increasing in post-analysis as shown in Figure 12 (for ε = 0.0216). As less strain generates less current, the charging time for the capacitors was higher for the lower strain measurement (ε = 0.0123). After the capacitor is fully charged (the time of which is dependent on the energy of the signal being fed into the capacitor) we would expect the I rms output to hold steady under continuous loading of the same frequency and magnitude. This concept was reiterated in [ 12 ] for energy. However, what actually occurs is that after the current stops increasing due to the capacitor’s energy storage capacity, it starts to decrease, as shown in Figure 12 . It may be due to some parasitic element of the capacitor itself, such as leakage resistance or current leakage through the rest of the system. It is clear that once the capacitor reaches its maximum energy storage capability, the current is being drained faster than it is being produced under constant loading. As shown in Table 4 , at ε = 0.0123, the power generated at the output of the rectifier (refer to Figure 4 b) was slightly higher than the power generated with no rectifier. As the values were very close, we can put this down to small changes in the applied force. For ε = 0.0216, the power generated at the output of the rectifier is less than the power generated with no rectifier, which is what we would expect due to the voltage drop and associated power dissipation through the diodes. As Table 4 shows, the greatest constant DC current generated in this study was 34.7 nA corresponding to a generated power of 12 nW when the load capacitor was 470 nF. If such sensor device was used to charge a typical 3.7 V, 1000 mAh phone battery from flat with a 100% efficient voltage step up circuit, it would take approximately 307 million hours to charge fully. Charging a 2.8 V, 1000 mAh pacemaker battery would take 233 million hours. The largest AC current generated was 49.4 nA with a corresponding power output of 24.5 nW. However, the power output reduces to almost half for the same strain (0.0216) when a full-wave rectifier is used with a 470 nF capacitor. This power output capability is far lower than that predicted (1800 µW) in [ 16 ], and the theoretical recoverable power of 0.33 W from upper arm motion calculated by [ 6 ]. It is however to be noted that the simulated PVDF patch in [ 16 ] had a much larger surface area (1600 mm 2 as opposed to 625 mm 2 in our study), which should explain the reduced power generated in our case. The main reason behind our generator’s poor power generation capability is likely to be the absence of poling in the test procedure. Noise in the system was a big issue during testing. Table 5 displays the different earthing configurations with the associated noise levels. Figure 13 a shows direct grounding of the sensor, while Figure 13 b shows grounding through the body. The sensor used was a 20 × 20 mm parallel bimorph with 0.0254 mm thick PVDF. Clearly, the best configuration to minimize system noise is with the sensor connected to ground through the body. The most noise occurs when the sensor is connected to the body with no grounding at all (shown as “Direct” in Table 5 ). As the sensor is unlikely to remain completely out of contact from the human body, the best solution is to ground the body. For future applications, this may require a ground point integrated into the clothing or grounding directly from the electronic device, which is being charged."
} | 6,621 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.