Thermodynamics and H2 Transfer in a Methanogenic, Syntrophic Community

Joshua J. Hamilton; Montserrat Calixto Contreras; Jennifer L. Reed

doi:10.1371/journal.pcbi.1004364

Abstract

Microorganisms in nature do not exist in isolation but rather interact with other species in their environment. Some microbes interact via syntrophic associations, in which the metabolic by-products of one species serve as nutrients for another. These associations sustain a variety of natural communities, including those involved in methanogenesis. In anaerobic syntrophic communities, energy is transferred from one species to another, either through direct contact and exchange of electrons, or through small molecule diffusion. Thermodynamics plays an important role in governing these interactions, as the oxidation reactions carried out by the first community member are only possible because degradation products are consumed by the second community member. This work presents the development and analysis of genome-scale network reconstructions of the bacterium Syntrophobacter fumaroxidans and the methanogenic archaeon Methanospirillum hungatei. The models were used to verify proposed mechanisms of ATP production within each species. We then identified additional constraints and the cellular objective function required to match experimental observations. The thermodynamic S. fumaroxidans model could not explain why S. fumaroxidans does not produce H₂ in monoculture, indicating that current methods might not adequately estimate the thermodynamics, or that other cellular processes (e.g., regulation) play a role. We also developed a thermodynamic coculture model of the association between the organisms. The coculture model correctly predicted the exchange of both H₂ and formate between the two species and suggested conditions under which H₂ and formate produced by S. fumaroxidans would be fully consumed by M. hungatei.

Author Summary

Natural and engineered microbial communities can contain up to hundreds of interacting microbes. These interactions may be positive, negative, or neutral, as well as obligate or facultative. Syntrophy is an obligate, positive interaction, in which one species lives off the metabolic by-products of another. Syntrophic associations play an important role in sustaining a variety of natural communities, including those involved in the breakdown and conversion of short-chain fatty acids (e.g., propionate) to methane. In many syntrophic communities, electrons are transferred from one species to the other through small molecule diffusion. In this work, we expand the study of a two-member syntrophic, methanogenic community through the development and analysis of computational models for both species: the bacterium Syntrophobacter fumaroxidans and the methanogenic archaeon Methanospirillum hungatei. These models were used to analyze energy conservation mechanisms within each species, as well as small molecule exchange between the two organisms in coculture. The coculture model correctly predicted the exchange of both H₂ and formate between the two species and suggested conditions under which these molecules would be fully metabolized within the community.

Citation: Hamilton JJ, Calixto Contreras M, Reed JL (2015) Thermodynamics and H₂ Transfer in a Methanogenic, Syntrophic Community. PLoS Comput Biol 11(7): e1004364. https://doi.org/10.1371/journal.pcbi.1004364

Editor: Costas D. Maranas, The Pennsylvania State University, UNITED STATES

Received: September 11, 2014; Accepted: June 1, 2015; Published: July 6, 2015

Copyright: © 2015 Hamilton et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited

Data Availability: All relevant data are within the paper and its Supporting Information files.

Funding: This work was funded by the National Science Foundation through a CAREER grant to JLR (NSF 1053712) and a Graduate Research Fellowship to JJH (DGE-0718123). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Microorganisms in nature engage in a variety of interactions with other species in their environment. Syntrophy is one such type of inter-species interaction in which one species lives off the metabolic by-products of another [1–3]. Synthetic methanogenic communities [4] are typically tightly constrained by thermodynamics, as the oxidation reactions carried out by the first community member are thermodynamically unfavorable unless the degradation products are maintained at low levels by the second community member [5].

In anaerobic syntrophic communities, electrons are transferred from one partner to the other through direct contact or small molecule diffusion [6]. Traditional biochemistry has elucidated intracellular electron transport mechanisms [3,7–9], but it is difficult to evaluate these pathways in their metabolic and environmental context. Genome-scale metabolic models (GEMs) [10–12] and constraint-based methods are powerful computational tools for understanding individual pathways in a broader metabolic context including both isolated microbial species [13–15] and simple microbial communities [16–26].

One of the earliest microbial community models used flux balance analysis (FBA, [27]) to investigate formate and H₂ exchange between the sulfate-reducing bacterium Desulfovibrio vulgaris and the methanogenic archaeon Methanococcus maripaludis [16]. In this study, each organism was modeled as a compartment within a larger community-scale model. Compartmentalized approaches have been used to study the origins of cooperation and competition [17–19], as well as specific communities [20–24]. These approaches [16–24] have often used a single (joint) objective function to capture community behavior. OptCom [25,26] instead uses a multi-level optimization framework, to capture the trade-offs between individual and community fitness, with separate objective functions for the individual species and the community. In addition, community FBA (cFBA) [28] extends compartmentalized approaches [16–24] to specifically account for individual species’ biomass abundance.

Genome-scale models can also be used to study the relationship between thermodynamics and metabolism, by ensuring that network predictions are consistent with thermodynamic principles [29–34]. In this study, we used thermodynamics-based metabolic flux analysis (TMFA) to develop a thermodynamic, coculture model of the syntrophic association between the anaerobic bacterium Syntrophobacter fumaroxidans and the methanogenic archaeon Methanospirillum hungatei. In association with M. hungatei, S. fumaroxidans converts propionate to acetate, CO₂, and H₂ [35–37]. CO₂ and H₂ can be interconverted to formate [38–40], with H₂ and formate serving as the electron carriers between the two species. H₂ and formate production are only observed during syntrophic growth. Using a thermodynamic, constraint-based model, we set out to test the proposed hypothesis that this behavior is governed by thermodynamics [3,5,6,8,9].

We developed genome-scale metabolic reconstructions of both microorganisms, and verified proposed mechanisms of ATP production within each individual species. Additional constraints and a cellular objective function were identified to predict the proper flux through experimentally characterized carbon and electron transport pathways during monoculture and syntrophic (i.e., coculture) growth. Our analysis revealed that thermodynamic constraints alone are insufficient to explain why S. fumaroxidans does not produce H₂ in monoculture.

We also extended TMFA to model the syntrophic association between the two microorganisms. The association is modeled as a continuous coculture system with constraint-based models for each microbe and a mass balance around the reactor. Similar to cFBA [28], the coculture model accounted for the biomass concentrations of each species. We predicted the behavior of this syntrophic association under a variety of dilution rates, and identified regimes of behavior consistent with experimental observations.

Results

Metabolic reconstructions of S. fumaroxidans (iSfu648) and M. hungatei (iMhu428) were tested and parameterized using experimental data from growth on single substrates in monoculture and coculture. The iSfu648 and iMhu428 models were used to verify proposed mechanisms of energy conservation within each species. A coculture model was then developed to identify conditions where H₂ and formate, produced by S. fumaroxidans, is fully metabolized by M. hungatei.

Testing and Parameterizing the iMhu428 Metabolic Model

The iMhu428 reconstruction of M. hungatei was built from the iMB745 reconstruction of M. acetivorans [41]. A preliminary draft reconstruction was built from iMB745 using the RAVEN Toolbox [42] and the KEGG SSDB [43]; however, M. hungatei orthologs were found for only 428 of the 745 genes in iMB745. To avoid extensive gapfilling, reactions from the iMB745 were copied into the M. hungatei reconstruction, with modifications to reflect key metabolic features of M. hungatei (see S1 Text). As a consequence, the iMhu428 reconstruction is a draft reconstruction requiring further evaluation. A thermodynamic model for iMhu428 was built and TMFA was used to predict ATP generating mechanisms in minimal media monoculture conditions (see S1 Table in S1 Dataset for constraints used).

Experimental evidence suggests that M. hungatei is able to generate 0.5 mole ATP per mole of CO₂ converted to CH₄ [7], via the metabolic route shown in Fig 1. In order for this route to be thermodynamically feasible, the Δ_rG^'0 of one reaction (FMFTSPFT, formylmethanofuran-tetrahydromethanopterin formyltransferase) had to be allowed to vary within a 99% confidence interval of its estimated standard transformed Gibbs free energy of reaction () (rather than the 95% interval used for all other reactions) in order to carry flux in the proper direction. The reactions for carbon source utilization in M. hungatei are well-characterized [7,44–47], but uncertainty remains about the stoichiometry of small ion transport [7]. Na⁺ transport stoichiometries associated with tetrahydromethanopterin S-methyltransferase (MTSPCMMT_CM5HBCMT, E.C. 2.1.1.86) and a Na⁺/H⁺ antiporter (NAT3_1) were selected to give an ATP yield matching the experimental estimates: two Na⁺ ions exported by MTSPCMMT_CM5HBCMT, and a one Na⁺ per H⁺ transported by NAT3_1. However, different stoichiometries for these reactions are also thermodynamically possible (see Discussion).

Download:

Fig 1. Carbon and electron transport pathways in M. hungatei.

The methanogenesis pathway has an overall stoichiometry of CO₂ + 4 H₂ → CH₄ + 2 H₂O. Substrates and products are indicated with purple and blue ovals, respectively, while reactions with and without estimates are indicated with green and orange arrows, respectively. Metabolite and reaction abbreviations are given in S2 Dataset.

https://doi.org/10.1371/journal.pcbi.1004364.g001

Experimental measurements of growth rates, yields, and maintenance costs were also used to identify substrate uptake rates (SUR) for CO₂ and formate, and the growth- (GAM) and non-growth-associated (NGAM) ATP maintenance requirements for M. hungatei, as described in S1 Text. NGAM represents the amount of energy spent to maintain the cell (i.e., maintenance energy), while GAM represents energy spent on growth-related functions (e.g., protein synthesis). For the iMhu428 model, the NGAM was estimated to be 0.6 mmol ATP/gDW/day, GAM was estimated to be 47 mmol ATP/gDW, SUR_CO2 was estimated to be 75.7 mmol/gDW/day, and SUR_formate was estimated to be 955 mmol/gDW/day.

Testing and Parameterizing the iSfu648 Metabolic Model

The iSfu648 reconstruction of S. fumaroxidans was built from the KEGG database using the RAVEN Toolbox [42]. The resulting draft reconstruction was manually refined (see S1 Text), with particular attention paid to ATP production mechanisms. A number of studies have identified gene clusters encoding a variety of hydrogenases, dehydrogenases, and other electron transport enzymes [8,9,48–51], whose expression levels vary across growth conditions [51]. All told, 17 enzymes which catalyze 12 different electron transport reactions have been identified (S3 Table in S1 Dataset). In many cases, the proposed reactions catalyzed by these enzymes differ between studies; a brief description of each reaction and justification for each annotation is given in S1 Text. The draft reconstruction was updated to be consistent with the reported carbon utilization and electron transport reactions, and the resulting stoichiometric model was converted to a thermodynamic model.

ATP Production Mechanisms in S. fumaroxidans

Experimental studies have elucidated five growth modes for S. fumaroxidans: four in monoculture and one in coculture with M. hungatei (S1 Text) [36,48,52]. This work examines the three most commonly studied growth modes (Table 1): monoculture growth on fumarate, monoculture growth on fumarate plus propionate, and coculture growth on propionate.

Download:

Table 1. Experimentally observed and computationally predicted extracellular flux distributions for S. fumaroxidans examined in this study.

https://doi.org/10.1371/journal.pcbi.1004364.t001

A variety of experimental findings were synthesized to develop theoretical flux distributions for these three growth modes (Fig 2 and S1 Text). These experimental findings suggested additional regulatory and flux-coupling constraints for the iSfu648 model, such as coupling between fumarate reductase and the cytosolic hydrogenase due to co-localization in the membrane (see S2 Table in S1 Dataset and S1 Text for the full set of constraints and their justification).

Download:

Fig 2. Carbon and electron transport pathways in S. fumaroxidans under different substrate conditions.

Simulations maximized ATP production with no minimum biomass requirement. The overall stoichiometry of each pathway corresponds to the experimental stoichiometry given in Table 1. (A) Monoculture growth on fumarate alone. (B) Monoculture growth on fumarate and propionate. (C) Syntrophic growth on propionate. Substrates and products are indicated with purple and blue ovals, respectively, while reactions with and without estimates are indicated with green and orange arrows, respectively. Arrow thickness indicates relative flux values. Metabolite and reaction abbreviations are given in S4 Dataset.

https://doi.org/10.1371/journal.pcbi.1004364.g002

During monoculture growth on fumarate alone (Fig 2A), one mole of fumarate gets fully oxidized to CO₂, while six moles of fumarate get reduced to succinate [48,52]: (1)

The oxidation of one fumarate to CO₂ generates one ATP and five reducing equivalents (three NADH and two pairs of reduced ferredoxin) [48,52], while the reduction of additional fumarate to succinate by fumarate reductase (FRD) consumes reducing equivalents (menaquinol) [48]. Electrons are transferred from NADH and reduced ferredoxin to menaquinone through the combined action of the Rnf complex (RNF), the ferredoxin-oxidizing hydrogenase (frH₂ase), the cytosolic hydrogenase (cytH₂ase) and formate hydrogen lyase (FHL). The reduction of fumarate to succinate also generates the proton motive force (PMF) responsible for driving the RNF reaction and producing ATP.

During monoculture growth on fumarate plus propionate (Fig 2B), one mole of propionate gets oxidized to succinate, while one mole of fumarate gets oxidized to acetate and CO₂. Two additional moles of fumarate get reduced to succinate [36,48]: (2)

The oxidation of fumarate to acetate and CO₂ produces one NADH and one pair of reduced ferredoxin, while the reduction of fumarate to succinate by FRD consumes menaquinol. Electrons are transferred from NADH and reduced ferredoxin to menaquinone through the combined action of the confurcating hydrogenase (cH₂ase) and cytH₂ase. Oxidation of propionate to succinate produces one ATP, while FRD generates the PMF necessary for additional ATP production.

During coculture growth on propionate (Fig 2C), propionate gets oxidized to acetate and CO₂ via the methylmalonyl-CoA pathway [48,52]:

(3)

ATP is generated during the oxidation of propionate to succinate, and this ATP establishes the PMF necessary to drive the endergonic oxidation of succinate to fumarate (SDH), producing menaquinone. cytH₂ase then transfers electrons from menaquinol to two protons, generating H₂. The oxidation of fumarate to acetate and CO₂ produces one NADH and one pair of reduced ferredoxin, and cH₂ase couples NADH and ferredoxin re-oxidation with H₂ production. Unlike in the monoculture growth modes, the H₂ is not consumed intracellularly and must diffuse outside the cell. It has been proposed that the net production of H₂ by S. fumaroxidans is only thermodynamically favorable at the low H₂ concentrations maintained by methanogens, thereby explaining why S. fumaroxidans only produces H₂ during coculture growth.

S. fumaroxidans exhibits considerable flexibility in its ATP production mechanisms during coculture growth ([8,9,48,51]), and can produce formate instead of CO₂ (Fig 3) yielding an overall transformation of: (4)

Download:

Fig 3. Alternative carbon and electron transport pathways in S. fumaroxidans under coculture growth conditions.

(A) Formate production via cytFDH. (B) Formate production via cFDH. (A and B) Simulations maximized ATP production with no minimum biomass requirement. The overall stoichiometry of each pathway corresponds to the alternative stoichiometry given in Eq 4. Substrates and products are indicated with purple and blue ovals, respectively, while reactions with and without estimates are indicated with green and orange arrows, respectively. Arrow thickness indicates relative flux values. Metabolite and reaction abbreviations are given in S4 Dataset.

https://doi.org/10.1371/journal.pcbi.1004364.g003

In one mechanism (Fig 3A), activity of the cytosolic formate hydrogenase (cytFDH) substitutes for the activity of cytH₂ase. In a second mechanism (Fig 3B), the confurcating formate dehydrogenase (cFDH) substitutes for cH₂ase. Here, cFDH couples NADH and ferredoxin re-oxidation with the conversion of CO₂ (from propionate oxidation) to formate.

Model Predictions of ATP Production by S. fumaroxidans

Experimental evidence and conceptual models of S. fumaroxidans energy metabolism suggest that the carbon and electron transfer pathways shown in Fig 2 provide the sole source for ATP production in S. fumaroxidans, either by substrate-level phosphorylation or through establishment of a proton gradient used by ATP synthase [5]. To test the computational model’s predictions, TMFA was used to maximize ATP production under each of the three growth modes.

When flux was restricted to a reduced network containing all the reactions shown in Fig 2 (listed in S4 Table in S1 Dataset), the iSfu648 model correctly predicted the flux distributions shown in Fig 2. However, when flux was allowed throughout the entire network, additional flux distributions with higher ATP yields were identified. Additional reaction direction constraints were developed to ensure model-predicted flux distributions matched experimental observations (see S2 Table in S1 Dataset and S1 Text for details). However, the resulting flux distributions are not fully consistent with the hypothesis that S. fumaroxidans has adapted to maximize its energy yield.

Evaluating Thermodynamics of H₂ Production

Experimental studies of S. fumaroxidans have shown that H₂ is not produced during growth in monoculture [36,52,53], and it is widely thought that H₂ production is only thermodynamically favorable at low partial pressures [3,6,8,9]. In particular, methanogens in syntrophic communities enable sustained H₂ production by consuming H₂ and keeping its partial pressure low [3,5,6,8,9]. Indeed, when H₂ production was observed in monoculture, H₂ production ceased at a partial pressure of approximately 10 Pa [53].

However, when maximizing H₂ production under monoculture conditions, simulations reveal that H₂ production remains thermodynamically feasible. For example, during monoculture growth on fumarate, the iSfu648 model predicts that H₂ can be produced via the following mechanism: (5)

In this scenario, H₂ molecules produced by the ferredoxin-oxidizing hydrogenase are exported outside the cell, instead of serving as substrates for the cytosolic hydrogenase. As a result, no PMF is generated by fumarate reductase, and the net ATP yield is zero. Thus, while H₂ production remains thermodynamically possible, H₂ production is only associated with sub-optimal mechanisms of ATP generation. This suggests that thermodynamic considerations alone may not explain the absence of H₂ production during monoculture growth, but that the observed flux distribution may instead be driven by demands for energy generation.

While H₂ production was not predicted for monoculture conditions when ATP production was maximized, H₂ production was initially predicted when growth was instead maximized. To eliminate monoculture H₂ production in the TMFA model, we first sought to constrain ratios of metabolite concentrations with an approach similar to that used to correct TMFA growth predictions [34] (see S1 Text for details). While metabolite ratio constraints could be identified to prevent some H₂ production mechanisms, H₂ production during monoculture growth could not be completely eliminated. If thermodynamics prevents H₂ production in monoculture, then the current thermodynamic model may contain too much uncertainty in its Gibbs free energy estimates (see Discussion). Regulatory effects could also potentially prevent H₂ production in monoculture conditions. To correct the model, all subsequent monoculture simulations were performed by preventing H₂ production.

Parameterization of the iSfu648 Metabolic Model

Model parameters were estimated after reaction direction constraints were added to the iSfu648 model (to be consistent with reported ATP generation and H₂ production mechanisms). Experimental measurements of growth rates, yields, and maintenance costs were used to identify the SURs, GAM, and NGAM parameters for S. fumaroxidans. These parameters were estimated using data from monoculture growth on fumarate alone and coculture growth on propionate alone (see S1 Text). For the iSfu648 model, the following parameters resulted in the best fit of the model to the experimental data: NGAM = 3.36 mmol ATP/gDW/day, GAM = 22.8 mmol ATP/gDW, SUR_propionate = 37.7 mmol/gDW/day, and SUR_fumarate = 27.6 mmol/gDW/day.

Using these parameter values, the in silico growth rates under each growth condition were predicted (Table 1). Not surprisingly, the predicted growth rates for fumarate alone and propionate alone conditions agree with experimental observations (since these were used to estimate the parameter values). However, the model significantly under-predicts the measured growth rate during monoculture growth on fumarate plus propionate (0.55 days^-1 predicted, 0.73 days^-1 observed). This discrepancy could be caused by differences in uptake rates or maintenance costs in the fumarate plus propionate condition compared to the conditions with propionate alone or fumarate alone.

Uptake and Secretion at Maximal Growth

When maximizing biomass production on the entire network, the iSfu648 model predicted a wide range of product secretion rates. When the enzyme cost (i.e., total flux) was minimized at the maximum growth (pTMFA [54], see Methods) the model-predicted product yields closely matched reported values for two of the three growth modes (Table 1)—monoculture growth on fumarate alone and coculture growth on propionate alone. These results indicate that the majority of carbon is diverted to fermentation products, consistent with the expectation that high fluxes through the low-energy fermentation pathways are needed to meet cellular energy demands.

However, for monoculture growth on fumarate plus propionate, the model failed to predict that fumarate and propionate should be consumed at the observed ratio of approximately three fumarate per propionate [36]. Instead, the model predicted both substrates would be consumed at their maximum SURs, resulting in a ratio of 0.73 fumarate per propionate. The experimentally observed 3:1 ratio is thought to arise due to coupling within the metabolic network (Fig 2B), since oxidation of one fumarate produces one CO₂ (used to oxidize one propionate) and two pairs of electrons (used to reduce two fumarate). While this coupling arises naturally on the reduced network, the full metabolic network enables alternative coupling mechanisms (not shown) that permit other fumarate to propionate ratios. Since propionate oxidation generates carbon precursors and ATP for biomass, maximizing biomass production results in the model under-predicting the fumarate to propionate uptake ratio. Increasing the ratio of fumarate to propionate uptake (to 3:1) decreases the predicted propionate SUR and growth rate (S1 Fig in S1 Text), implying the experimental ratio is sub-optimal with respect to growth maximization. Instead of constraining SURs (since values were not reported in the literature), we incorporated a fumarate to propionate SUR ratio constraint for this condition.

While the predicted product yields closely matched experimental observations (after imposing the SUR ratio constraint), the pTMFA-predicted intracellular flux distribution during monoculture growth on fumarate and propionate substantially deviated from that shown in Fig 2B. Further constraints on reaction directions were required so that propionate and fumarate were metabolized in the model via the pathways shown in Fig 2B (results not shown). Taken together, the need for constraints on reaction directions and fumarate to propionate uptake ratio suggests that neither maximization of biomass nor minimization of enzyme cost are sufficient to explain the fluxes of S. fumaroxidans growing in this monoculture condition.

Behavior of M. hungatei and S. fumaroxidans in Coculture

During growth in coculture, S. fumaroxidans converts propionate to acetate, H₂, and CO₂ or formate [36,52], while M. hungatei consumes acetate, CO₂, H₂, and formate and produces CH₄ [55]. M. hungatei can also optionally interconvert excess CO₂ and H₂ to formate via a formate dehydrogenase [56].

Cocultures of M. hungatei and S. fumaroxidans have been grown in both batch and continuous (chemostat) systems. During batch growth, H₂ pressure rose during the lag phase and became constant during exponential growth [53]. Continuous cultures also exhibited constant H₂ partial pressure [53]. However, to the best of our knowledge, measurements for the relative ratios of M. hungatei to S. fumaroxidans at constant H₂ pressure (where H₂ consumption and production rates are balanced) have not been reported. Instead, an overall reaction for the coculture of (6) is frequently discussed [5,53], which can occur at a ratio of three M. hungatei to four S. fumaroxidans. Since initial conditions for batch experiments were not reported, a continuous culture model was constructed and first evaluated using a 3:4 relative biomass ratio (M. hungatei: S. fumaroxidans).

The continuous coculture model included constraint-based models for each microbe and mass balances around the reactor (S2 Fig in S1 Text), and accounted for the biomass concentrations of each species. Both species were constrained to grow at the dilution rate, and the model minimized the species-weighted total flux through the two metabolic networks (pTMFA). Propionate was the only substrate in the reactor feed (i.e., it had a net flux into the reactor), ensuring that all carbon and electrons used by M. hungatei were produced by S. fumaroxidans.

Predicted yields around individual species were first evaluated (Fig 4A). At low dilution rates, S. fumaroxidans was predicted to convert propionate to acetate, H₂, and CO₂/formate and M. hungatei was predicted to convert CO₂/formate to CH₄ (Fig 4B and 4C, which show alternate solutions with CO₂ or formate being exchanged). As the reactor dilution rate increased, the predicted species yields of H₂ and CO₂/formate (S. fumaroxidans) and CH₄ (M. hungatei) decreased slightly (Fig 4B and 4C). In addition, species’ uptake rates of propionate and CO₂/formate increased with dilution rate, as biochemical transformation of these substrates provides the energy needed for cellular growth and maintenance.

Download:

Fig 4. Predicted individual and community by-product yields in the coculture system at a ratio of three M. hungatei to four S. fumaroxidans.

(A and D) Diagrams illustrating that yields are calculated around individual species (A) and the entire community (D). The plots show the yields of acetate, H₂, CO₂, and formate for S. fumaroxidans (per propionate) (B and C); CH₄ for M. hungatei (per CO₂) (B and C); and acetate, H₂, CO₂, formate, and CH₄ for the entire reactor (per propionate) (E and F), as a function of the dilution rate of the reactor (X-axis). Plots are shown in which H₂ and CO₂ and exchanged (B and E), and in which formate and H₂ is exchanged (C and F).

https://doi.org/10.1371/journal.pcbi.1004364.g004

Species’ uptake rates, secretion rates, and relative biomass ratio affects overall bioreactor yields, and these bioreactor yields were subsequently investigated (Fig 4D). At a 3:4 relative biomass ratio (M. hungatei: S. fumaroxidans), M. hungatei did not fully utilize all of the H₂ and CO₂/formate produced by S. fumaroxidans, even at high dilution rates (Fig 4E and 4F). The net H₂ production by the community indicates that S. fumaroxidans produces more H₂ than M. hungatei’s needs and suggests the community can maintain higher M. hungatei to S. fumaroxidans ratios or that S. fumaroxidans could support faster growth of M. hungatei. At a dilution rate of 0.05 days^-1, S. fumaroxidans produces H₂ in excess of M. hungatei’s energy needs until the relative M. hungatei to S. fumaroxidans biomass ratio reaches approximately 1.6:1 (Fig 5). These simulations suggest that invariant external H₂ concentration requires high ratios of M. hungatei to S. fumaroxidans (higher than has been proposed in the literature based on overall reaction stoichiometries), M. hungatei growing at faster rates than S. fumaroxidans (e.g., in batch culture), or a combination of the two.

Download:

Fig 5. Predicted community by-product yields by the coculture at a dilution rate of 0.05 days^-1.

The plot shows the yields of acetate, H₂, CO₂, and CH₄ for the entire reactor (per propionate) as a function of the relative biomass ratio of M. hungatei to S. fumaroxidans (X-axis). As in Fig 4, formate could be exchanged in the place of CO₂ and H₂.

https://doi.org/10.1371/journal.pcbi.1004364.g005

Furthermore, studies have shown that in coculture, S. fumaroxidans passes electrons to M. hungatei via formate, as well as H₂ [38,39,56]. Coculture simulations predicted that formate could be exchanged in lieu of CO₂ (Fig 4B and 4C), without affecting the predicted bioreactor yields or species-weighted total flux (pTMFA objectives). When formate is exchanged the formate dehydrogenases of S. fumaroxidans and M. hungatei facilitate the interconversion of formate to CO₂ and H₂.

Finally, steady-state metabolite concentrations in the coculture were predicted using thermodynamic variability analysis [33,34] (S1 Text). Similar to a previous study of E. coli [34], the majority of steady-state metabolite concentrations were not constrained by thermodynamics (i.e., the concentration ranges were the global concentration bounds of 0.01mM and 20 mM). However, hypotheses for some extracellular (Table 2) and intracellular (S5 Table in S1 Dataset) metabolite concentrations could be made. The model predicts that propionate in the media must be greater than 0.36 mM for coculture growth to occur, and that acetate and CO₂ concentrations must be less than 4mM and 0.65 mM, respectively. The model also predicts critical concentrations for H₂ (0.0032 mM) and formate (0.0020 mM, when formate is being exchanged), which must be maintained in order for methanogenesis to occur. All of these predictions were insensitive to both the dilution rate and biomass ratio (M. hungatei: S. fumaroxidans).

Download:

Table 2. Model-predicted steady-state metabolite concentrations for select external metabolites in coculture.

https://doi.org/10.1371/journal.pcbi.1004364.t002

Discussion

The iMhu428 and iSfu648 thermodynamic models were successfully used to verify proposed carbon and ATP production pathways in M. hungatei and S. fumaroxidans; however, some more efficient pathways (based on ATP or biomass yields) were found in some cases. The computational results highlighted topics that necessitate further discussion relating to (1) the stoichiometry of transport reactions, (2) the thermodynamics of H₂ production, and (3) the carbon and electron shuttles between species in coculture.

Ion Transport in the iMhu428 Metabolic Model

The majority of the iMhu428 model content comes from the iMB745 reconstruction of M. acetivorans and still needs to be verified. Changes to model content could affect the conclusions drawn about M. hungatei behavior. Despite containing a complete methanogenesis pathway, the iMhu428 model was unable to identify the H⁺/Na⁺ transport stoichiometry of the energy-converting (Eha- or Ehb-type) hydrogenase (EHA, 1.12.7.2), which is thought to pump H⁺/Na⁺ while reducing ferredoxin [7]. The heterodisulfide reductase (HDR, 1.8.98.1) can also reduce ferredoxin, and the iMhu428 model predicted HDR to be the only ferredoxin-reducing reaction required for methanogenesis. This observation is consistent with the observation that the expression of Eha/Ehb is considerably lower than that of HDR [57].

Additionally, different stoichiometries for other ion transport reactions important to methanogenesis remain thermodynamically possible. For example, the group contribution method predicted that tetrahydromethanopterin S-methyltransferase (MTSPCMMT_CM5HBCMT, E.C. 2.1.1.86) could drive transport of up to 4 Na⁺ ions under standard conditions, instead of the 2 Na⁺ ions used in the iMhu428 reconstruction. Furthermore, some studies suggest the archaeal A₁A₀ ATP synthase is coupled to Na⁺ instead of H⁺ translocation [7,58]. When the iMhu428 model was modified to reflect this coupling, the model predicted the Na⁺/H⁺ antiporter was no longer active, as Na⁺ ions from MTSPCMMT_CM5HBCMT were directly used for ATP synthesis. Thus, while the modeled methanogenesis pathway is consistent with available data, it is not the only possibility.

H₂ Production in the iSfu648 Metabolic Model

Analysis of the iSfu648 model revealed that H₂ production is thermodynamically feasible in monoculture, implying there may be other biological reasons why H₂ production is not normally observed under this condition, or that tighter estimates of thermodynamic parameters are needed. The iSfu648 thermodynamic model also has some important limitations. In particular, the iSfu648 model does not contain enough thermodynamic information to predict the directions of important electron transport reactions that involve ferredoxin, including the confurcating hydrogenase and formate dehydrogenase, the ferredoxin-oxidizing hydrogenase and formate dehydrogenase, and the RNF-type oxidoreductase (S3 Table in S1 Dataset). This is because the group contribution method is unable to estimate the standard transformed Gibbs free energy of formation (Δ_fG^'0) of ferredoxin, resulting in no Δ_rG^'0 estimates for these reactions. Fortunately, new quantum chemical approaches for estimating the thermodynamics of metabolism [59] may potentially provide additional Δ_fG^'0 estimates.

This work also raises important questions about the appropriate mathematical basis for representing thermodynamic constraints. Previous studies used the ΔG^'0 of groups directly when modeling thermodynamics [34], and found that introducing uncertainty into a thermodynamic model of E. coli made the model computationally difficult to solve. In this work, using either the ΔG^'0 of molecules or groups to model thermodynamics proved computationally difficult (results not shown). Instead, only using Δ_rG^'0 as the basis for thermodynamic calculations enabled uncertainties in free energy estimates to be handled without any computational difficulties. However, using Δ_rG^'0 as a basis for thermodynamic calculations leads to larger uncertainties in Δ_rG^'0 and greater network flexibility, as it does not account for thermodynamic interconnectivity between reactions with shared metabolites. As a result, the model-predicted feasible Δ_rG^'0 range through a linear combination of reactions considerably exceeds the group-contribution predicted Δ_rG^'0 range of the overall reaction. In the future, thermodynamic interconnectivity should be captured using the ΔG^'0 of molecules or groups, and optimization techniques are needed to improve the runtime performance of the resulting thermodynamic models.

However, the use of Δ_rG^'0 as a basis for thermodynamic calculations is insufficient to explain why the iSfu648 model could still produce H₂ under monoculture growth conditions. As described in S1 Text, Δ_fG^'0 was used as a basis for thermodynamic calculations when attempting to find additional constraints which would prevent H₂ production. The failure to find such constraints indicates either that thermodynamics does not explain the absence of H₂ production in monoculture, or that current thermodynamic models cannot capture this phenomenon. If it is the latter, more accurate group contribution methods with smaller error estimates may eventually be able to explain the role of thermodynamics in syntrophic associations.

Formate and H₂ Transfer in Coculture

The thermodynamic coculture model of the syntrophic association between these species confirmed the role of formate and H₂ in electron transfer in the community, and led us to hypothesize that total H₂ consumption by the community indicates that M. hungatei cells are more abundant and/or faster growing than the S. fumaroxidans cells. The coculture model correctly predicted that both H₂ and formate could shuttle electrons between members of this community. Formate may be preferred over H₂ for electron transfer for thermodynamic reasons, as the Δ_rG^'0 of Eq 4 is more favorable (less positive) than that of Eq 3 in which formate is not exchanged. By exchanging formate in place of CO₂ and H₂, S. fumaxoridans could sustain propionate oxidation at higher extracellular concentrations of formate than of H₂. Formate exchange could also be preferred due to differences in kinetics, diffusion, and/or volatility. These scenarios could stabilize the syntrophic association by enabling faster shuttling of electrons to M. hungatei.

This work highlights some important obstacles to successful modeling of microbial consortia. In order for computational models of microbial communities to make meaningful predictions, individual species models must be integrated into a community model in a biologically relevant manner. Such integration will require an understanding of the objectives and constraints governing the behavior of each community member, and this work demonstrates that identifying the proper constraints and objectives requires extensive experimental characterization of the community. For example, neither maximization of growth rate nor maximization of ATP yield were sufficient for the iSfu648 model to predict the observed behavior of S. fumaroxidans. Additional constraints on reaction directions and flux ratios were required before iSfu648 could be combined with iMhu428 model to simulate the coculture. In addition, the iSfu648 model relied on data from gene expression and ¹³C NMR experiments, suggesting that constraint-based approaches will complement traditional top down (‘omics’) approaches [60] by enabling a mechanistic understanding of microbial interactions [24].

Methods

Reconstruction of the iMhu428 Metabolic Model

The iMhu428 reconstruction of M. hungatei was built from the iMB745 reconstruction of M. acetivorans [41]. A preliminary draft reconstruction was built based on sequence homology (using the RAVEN Toolbox [42]), but the reconstruction contained less than 200 genes (S2 Dataset). Instead of performing extensive gapfilling, reactions from the iMB745 M. acetivorans reconstruction were copied into the M. hungatei reconstruction, with modifications to reflect key metabolic features of M. hungatei (see S1 Text). Results from the RAVEN Toolbox and the KEGG SSDB [43] were used to map genes in M. acetivorans to M. hungatei and identify those reactions which have genomic evidence (S2 Dataset). Finally, blocked reactions lacking genomic evidence were removed from the reconstruction.

The final iMhu428 reconstruction contains 720 reactions, 428 genes (associated with 493 reactions), and 639 metabolites. Of the 428 genes, 351 were added based on sequence homology, and 77 were added manually. The reconstruction is available in S2 Dataset and S3 Dataset in Excel and SBML formats.

Reconstruction of the iSfu648 Metabolic Model

The iSfu648 reconstruction of S. fumaroxidans was built from KEGG [43] (S6 Dataset and S7 Dataset) using the RAVEN Toolbox [42], which uses protein homology to identify the KEGG Orthology (KO) ID for each gene in a genome. The reactions and genes corresponding to that KO ID are then imported into the reconstruction. The resulting draft reconstruction was manually refined as described in S1 Text. The final iSfu648 reconstruction contains 874 reactions, 648 genes (associated with 770 reactions), and 893 metabolites. The reconstruction is available in S4 Dataset and S5 Dataset, in Excel and SBML formats.

Thermodynamics-Based Metabolic Flux Analysis (TMFA)

Flux-balance analysis (FBA) [27] is a constraint-based technique for predicting the state of a metabolic network consistent with physiochemical principles. FBA identifies a flux distribution which maximizes cellular growth (or some other objective function), subject to steady-state mass-balance and enzyme capacity constraints. Thermodynamics-Based Metabolic Flux Analysis (TMFA, [33,34]) extends FBA via the introduction of thermodynamic constraints, which require that the transformed Gibbs free energy of a reaction (Δ_rG^') and its flux (v) have opposite signs. Estimates () and uncertainties () of Δ_rG^'0 for the reactions in the reconstructions were obtained using a group contribution method [61] via the von Bertalanffy 2.0 Toolbox [62]. TMFA was implemented as previously described [34], with additional details given in S1 Text. The mol files for metabolites in iMhu428 and iSfu648 are provided in S1 File and S2 File, respectively.

Parsimonious TMFA (pTMFA)

pFBA [54] is a constraint-based approach which maximizes cellular growth while also minimizing total flux through the network (a proxy for minimizing the total mass of enzymes required to sustain optimal growth through the network). pTMFA uses the same assumptions as pFBA while implementing the thermodynamic constraints of TMFA. pTMFA was implemented as a two-stage optimization process. In the first stage, growth rate is maximized via TMFA. In the second stage, the growth rate is fixed and the total flux through the network is minimized, subject to the same constraints as TMFA. Additional details on implementation can be found in S1 Text.

Coculture Model

For the coculture simulations, a community model of growth in a continuous stirred-tank reactor was developed that accounts for the biomass concentrations of each species. The model minimizes the species-weighted total flux through the metabolic networks subject to TMFA constraints for each species and mass balances around the entire reactor. Details on the specific implementation used in this work can be found in S1 Text. To avoid solving a mixed-integer non-linear program (MINLP), the dilution rate and biomass concentrations for each species were fixed, resulting in a MIP. To explore the community behavior under a variety of operating conditions, the reactor dilution rate was systematically changed, while allowing unlimited propionate uptake by the reactor.

Simulations

The uptake fluxes for carbon and other nutrients used in the simulations are given in S1 Table in S1 Dataset. All simulations were performed using CPLEX 12 (IBM, Armonk, NY) accessed via the General Algebraic Modeling System, Version 23.9.5 (GAMS, GAMS Development Corporation, Washington, DC). Estimates () and uncertainties () of thermodynamic parameters were obtained using version 2.0 of von Bertalanffy and Matlab R2012b (The MathWorks, Inc., Natick, MA).

Supporting Information

S1 Dataset. Supporting tables.

This file contains Supporting Tables S1 to S7 as described in the text.

https://doi.org/10.1371/journal.pcbi.1004364.s001

(XLS)

S2 Dataset. Excel version of the iMhu428 model.

https://doi.org/10.1371/journal.pcbi.1004364.s002

(XLS)

S3 Dataset. SBML version of the iMhu428 model.

https://doi.org/10.1371/journal.pcbi.1004364.s003

(XML)

S4 Dataset. Excel version of the iSfu648 model.

https://doi.org/10.1371/journal.pcbi.1004364.s004

(XLS)

S5 Dataset. SBML version of the iSfu648 model.

https://doi.org/10.1371/journal.pcbi.1004364.s005

(XML)

S6 Dataset. Excel version of the balanced KEGG database.

https://doi.org/10.1371/journal.pcbi.1004364.s006

(XLS)

S7 Dataset. SBML version of the balanced KEGG database.

https://doi.org/10.1371/journal.pcbi.1004364.s007

(XML)

S1 File. Molfile structure files for all metabolites in the iMhu428 GEM.

https://doi.org/10.1371/journal.pcbi.1004364.s008

(ZIP)

S2 File. Molfile structure files for all metabolites in the iSfu648 GEM.

https://doi.org/10.1371/journal.pcbi.1004364.s009

(ZIP)

S1 Text. Supporting text.

This file contains Supporting Methods, Results, and Figs to complement the main text.

https://doi.org/10.1371/journal.pcbi.1004364.s010

(PDF)

Author Contributions

Conceived and designed the experiments: JJH JLR. Performed the experiments: JJH MCC. Analyzed the data: JJH JLR. Wrote the paper: JJH JLR.

References

1. Schink B (2002) Synergistic interactions in the microbial world. Antonie Van Leeuwenhoek 81: 257–261. pmid:12448724
- View Article
- PubMed/NCBI
- Google Scholar
2. McInerney MJ, Struchtemeyer CG, Sieber JR, Mouttaki H, Stams AJM, et al. (2008) Physiology, ecology, phylogeny, and genomics of microorganisms capable of syntrophic metabolism. Ann N Y Acad Sci 1125: 58–72. pmid:18378587
- View Article
- PubMed/NCBI
- Google Scholar
3. Stams AJM, Plugge CM (2009) Electron transfer in syntrophic communities of anaerobic bacteria and archaea. Nat Rev Microbiol 7: 568–577. pmid:19609258
- View Article
- PubMed/NCBI
- Google Scholar
4. McInerney MJ, Sieber JR, Gunsalus RP (2009) Syntrophy in anaerobic global carbon cycles. Curr Opin Biotechnol 20: 623–632. pmid:19897353
- View Article
- PubMed/NCBI
- Google Scholar
5. Schink B (1997) Energetics of syntrophic cooperation in methanogenic degradation. Microbiol Mol Biol Rev 61: 262–280. pmid:9184013
- View Article
- PubMed/NCBI
- Google Scholar
6. Stams AJM, de Bok FAM, Plugge CM, van Eekert MHA, Dolfing J, et al. (2006) Exocellular electron transfer in anaerobic microbial communities. Environ Microbiol 8: 371–382. pmid:16478444
- View Article
- PubMed/NCBI
- Google Scholar
7. Thauer RK, Kaster A-K, Seedorf H, Buckel W, Hedderich R (2008) Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 6: 579–591. pmid:18587410
- View Article
- PubMed/NCBI
- Google Scholar
8. Müller N, Worm P, Schink B, Stams AJM, Plugge CM (2010) Syntrophic butyrate and propionate oxidation processes: from genomes to reaction mechanisms. Environ Microbiol Rep 2: 489–499. pmid:23766220
- View Article
- PubMed/NCBI
- Google Scholar
9. Sieber JR, McInerney MJ, Gunsalus RP (2012) Genomic insights into syntrophy: the paradigm for anaerobic metabolic cooperation. Annu Rev Microbiol 66: 429–452. pmid:22803797
- View Article
- PubMed/NCBI
- Google Scholar
10. Oberhardt MA, Palsson BØ, Papin JA (2009) Applications of genome-scale metabolic reconstructions. Mol Syst Biol 5: 320. pmid:19888215
- View Article
- PubMed/NCBI
- Google Scholar
11. Lewis NE, Nagarajan H, Palsson BØ (2012) Constraining the metabolic genotype-phenotype relationship using a phylogeny of in silico methods. Nat Rev Microbiol 10: 291–305. pmid:22367118
- View Article
- PubMed/NCBI
- Google Scholar
12. Zomorrodi AR, Suthers PF, Ranganathan S, Maranas CD (2012) Mathematical optimization applications in metabolic networks. Metab Eng 14: 672–686. pmid:23026121
- View Article
- PubMed/NCBI
- Google Scholar
13. Mahadevan R, Palsson BØ, Lovley DR (2011) In situ to in silico and back: elucidating the physiology and ecology of Geobacter spp. using genome-scale modelling. Nat Rev Microbiol 9: 39–50. pmid:21132020
- View Article
- PubMed/NCBI
- Google Scholar
14. Osterlund T, Nookaew I, Nielsen J (2012) Fifteen years of large scale metabolic modeling of yeast: Developments and impacts. Biotechnol Adv 30: 979–988. pmid:21846501
- View Article
- PubMed/NCBI
- Google Scholar
15. McCloskey D, Palsson BØ, Feist AM (2013) Basic and applied uses of genome-scale metabolic network reconstructions of Escherichia coli. Mol Syst Biol 9: 661. pmid:23632383
- View Article
- PubMed/NCBI
- Google Scholar
16. Stolyar SM, Van Dien SJ, Hillesland KL, Pinel N, Lie TJ, et al. (2007) Metabolic modeling of a mutualistic microbial community. Mol Syst Biol 3: 92. pmid:17353934
- View Article
- PubMed/NCBI
- Google Scholar
17. Wintermute EH, Silver PA (2010) Emergent cooperation in microbial metabolism. Mol Syst Biol 6: 407. pmid:20823845
- View Article
- PubMed/NCBI
- Google Scholar
18. Freilich S, Zarecki R, Eilam O, Segal ES, Henry CS, et al. (2011) Competitive and cooperative metabolic interactions in bacterial communities. Nat Commun 2: 589. pmid:22158444
- View Article
- PubMed/NCBI
- Google Scholar
19. Klitgord N, Segrè D (2010) Environments that induce synthetic microbial ecosystems. PLoS Comput Biol 6: e1001002. pmid:21124952
- View Article
- PubMed/NCBI
- Google Scholar
20. Bordbar A, Lewis NE, Schellenberger J, Palsson BØ, Jamshidi N (2010) Insight into human alveolar macrophage and M. tuberculosis interactions via metabolic reconstructions. Mol Syst Biol 6: 422. pmid:20959820
- View Article
- PubMed/NCBI
- Google Scholar
21. Lewis NE, Schramm G, Bordbar A, Schellenberger J, Andersen MP, et al. (2010) Large-scale in silico modeling of metabolic interactions between cell types in the human brain. Nat Biotechnol 28: 1279–1285. pmid:21102456
- View Article
- PubMed/NCBI
- Google Scholar
22. Heinken A, Sahoo S, Fleming RMT, Thiele I (2013) Systems-level characterization of a host-microbe metabolic symbiosis in the mammalian gut. Gut Microbes 4: 28–40. pmid:23022739
- View Article
- PubMed/NCBI
- Google Scholar
23. Shoaie S, Karlsson FH, Mardinoglu A, Nookaew I, Bordel S, et al. (2013) Understanding the interactions between bacteria in the human gut through metabolic modeling. Sci Rep 3: 2532. pmid:23982459
- View Article
- PubMed/NCBI
- Google Scholar
24. Nagarajan H, Embree M, Rotaru A-E, Shrestha PM, Feist AM, et al. (2013) Characterization and modelling of interspecies electron transfer mechanisms and microbial community dynamics of a syntrophic association. Nat Commun 4: 2809. pmid:24264237
- View Article
- PubMed/NCBI
- Google Scholar
25. Zomorrodi AR, Maranas CD (2012) OptCom: A Multi-Level Optimization Framework for the Metabolic Modeling and Analysis of Microbial Communities. PLoS Comput Biol 8: e1002363. pmid:22319433
- View Article
- PubMed/NCBI
- Google Scholar
26. Zomorrodi AR, Islam MM, Maranas CD (2014) d-OptCom: Dynamic multi-level and multi-objective metabolic modeling of microbial communities. ACS Synth Biol 3: 247–257. pmid:24742179
- View Article
- PubMed/NCBI
- Google Scholar
27. Orth JD, Thiele I, Palsson BØ (2010) What is flux balance analysis? Nat Biotechnol 28: 245–248. pmid:20212490
- View Article
- PubMed/NCBI
- Google Scholar
28. Khandelwal RA, Olivier BG, Röling WFM, Teusink B, Bruggeman FJ (2013) Community flux balance analysis for microbial consortia at balanced growth. PLoS One 8: e64567. pmid:23741341
- View Article
- PubMed/NCBI
- Google Scholar
29. Beard DA, Liang S, Qian H (2002) Energy balance for analysis of complex metabolic networks. Biophys J 83: 79–86. pmid:12080101
- View Article
- PubMed/NCBI
- Google Scholar
30. Qian H, Beard DA, Liang S (2003) Stoichiometric network theory for nonequilibrium biochemical systems. Eur J Biochem 270: 415–421. pmid:12542691
- View Article
- PubMed/NCBI
- Google Scholar
31. Schellenberger J, Lewis NE, Palsson BØ (2011) Elimination of thermodynamically infeasible loops in steady-state metabolic models. Biophys J 100: 544–553. pmid:21281568
- View Article
- PubMed/NCBI
- Google Scholar
32. Kümmel A, Panke S, Heinemann M, Kummel A (2006) Putative regulatory sites unraveled by network-embedded thermodynamic analysis of metabolome data. Mol Syst Biol 2: 2006.0034.
- View Article
- Google Scholar
33. Henry CS, Broadbelt LJ, Hatzimanikatis V (2007) Thermodynamics-based metabolic flux analysis. Biophys J 92: 1792–1805. pmid:17172310
- View Article
- PubMed/NCBI
- Google Scholar
34. Hamilton JJ, Dwivedi V, Reed JL (2013) Quantitative assessment of thermodynamic constraints on the solution space of genome-scale metabolic models. Biophys J 105: 512–522. pmid:23870272
- View Article
- PubMed/NCBI
- Google Scholar
35. Dolfing J (1987) Microbiological aspects of granular methanogenic sludge. Wageningen Agricultural University.
36. Stams AJM, Van Dijk JB, Dijkema C, Plugge CM (1993) Growth of syntrophic propionate-oxidizing bacteria with fumarate in the absence of methanogenic bacteria. Appl Environ Microbiol 59: 1114–1119. pmid:16348912
- View Article
- PubMed/NCBI
- Google Scholar
37. Harmsen HJM, Van Kuijk BLM, Plugge CM, Akkermans ADL, De Vos WM, et al. (1998) Syntrophobacter fumaroxidans sp. nov., a syntrophic propionate-degrading sulfate-reducing bacterium. Int J Syst Bacteriol 48: 1383–1387. pmid:9828440
- View Article
- PubMed/NCBI
- Google Scholar
38. Dong X, Stams AJM (1995) Evidence for H2 and formate formation during syntrophic butyrate and propionate degradation. Anaerobe 1: 35–39. pmid:16887505
- View Article
- PubMed/NCBI
- Google Scholar
39. Stams AJM, Dong X (1995) Role of formate and hydrogen in the degradation of propionate and butyrate by defined suspended cocultures of acetogenic and methanogenic bacteria. Antonie Van Leeuwenhoek 68: 281–284. pmid:8821782
- View Article
- PubMed/NCBI
- Google Scholar
40. Dong X, Plugge CM, Stams AJM (1994) Anaerobic degradation of propionate by a mesophilic acetogenic bacterium in coculture and triculture with different methanogens. Appl Environ Microbiol 60: 2834–2838. pmid:16349350
- View Article
- PubMed/NCBI
- Google Scholar
41. Benedict MN, Gonnerman MC, Metcalf WW, Price ND (2012) Genome-scale metabolic reconstruction and hypothesis testing in the methanogenic archaeon Methanosarcina acetivorans C2A. J Bacteriol 194: 855–865. pmid:22139506
- View Article
- PubMed/NCBI
- Google Scholar
42. Agren R, Liu L, Shoaie S, Vongsangnak W, Nookaew I, et al. (2013) The RAVEN Toolbox and Its Use for Generating a Genome-scale Metabolic Model for Penicillium chrysogenum. PLoS Comput Biol 9: e1002980. pmid:23555215
- View Article
- PubMed/NCBI
- Google Scholar
43. Kanehisa M, Goto S (2000) KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res 28: 27–30. pmid:10592173
- View Article
- PubMed/NCBI
- Google Scholar
44. Schwörer B, Thauer RK (1991) Activities of formylmethanofuran dehydrogenase, methylenetetrahydromethanopterin dehydrogenase, methylenetetrahydromethanopterin reductase, and heterodisulfide reductase in methanogenic bacteria. Arch Microbiol 155: 459–465.
- View Article
- Google Scholar
45. Thauer RK (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144: 2377–2406. pmid:9782487
- View Article
- PubMed/NCBI
- Google Scholar
46. Ferry JG (1999) Enzymology of one-carbon metabolism in methanogenic pathways. FEMS Microbiol Rev 23: 13–38. pmid:10077852
- View Article
- PubMed/NCBI
- Google Scholar
47. Liu Y, Whitman WB (2008) Metabolic, phylogenetic, and ecological diversity of the methanogenic archaea. Ann N Y Acad Sci 1125: 171–189. pmid:18378594
- View Article
- PubMed/NCBI
- Google Scholar
48. Plugge CM, Henstra AM, Worm P, Swarts DC, Paulitsch-Fuchs AH, et al. (2012) Complete genome sequence of Syntrophobacter fumaroxidans strain (MPOB(T)). Stand Genomic Sci 7: 91–106. pmid:23450070
- View Article
- PubMed/NCBI
- Google Scholar
49. De Bok FAM, Roze EHA, Stams AJM (2002) Hydrogenases and formate dehydrogenases of Syntrophobacter fumaroxidans. Antonie Van Leeuwenhoek 81: 283–291. pmid:12448727
- View Article
- PubMed/NCBI
- Google Scholar
50. De Bok FAM, Hagedoorn P-L, Silva PJ, Hagen WR, Schiltz E, et al. (2003) Two W-containing formate dehydrogenases (CO2-reductases) involved in syntrophic propionate oxidation by Syntrophobacter fumaroxidans. Eur J Biochem 270: 2934–2942.
- View Article
- Google Scholar
51. Worm P, Stams AJM, Cheng X, Plugge CM (2011) Growth- and substrate-dependent transcription of formate dehydrogenase and hydrogenase coding genes in Syntrophobacter fumaroxidans and Methanospirillum hungatei. Microbiology 157: 280–289. pmid:20884694
- View Article
- PubMed/NCBI
- Google Scholar
52. Plugge CM, Dijkema C, Stams AJM (1993) Acetyl-CoA cleavage pathway in a syntrophic propionate oxidizing bacterium growing on fumarate in the absence of methanogens. FEMS Microbiol Lett 110: 71–76.
- View Article
- Google Scholar
53. Scholten JCM, Conrad R (2000) Energetics of syntrophic propionate oxidation in defined batch and chemostat cocultures. Appl Environ Microbiol 66: 2934–2942. pmid:10877789
- View Article
- PubMed/NCBI
- Google Scholar
54. Lewis NE, Hixson KK, Conrad TM, Lerman JA, Charusanti P, et al. (2010) Omic data from evolved E. coli are consistent with computed optimal growth from genome-scale models. Mol Syst Biol 6: 390. pmid:20664636
- View Article
- PubMed/NCBI
- Google Scholar
55. Ekiel I, Smith ICP, Sprott GD (1983) Biosynthetic pathways in Methanospirillum hungatei as determined by 13C nuclear magnetic resonance. J Bacteriol 156: 316–326. pmid:6619097
- View Article
- PubMed/NCBI
- Google Scholar
56. De Bok FAM, Luijten MLGC, Stams AJM (2002) Biochemical Evidence for Formate Transfer in Syntrophic Propionate-Oxidizing Cocultures of Syntrophobacter fumaroxidans and Methanospirillum hungatei. Appl Environ Microbiol 68: 4247–4252. pmid:12200272
- View Article
- PubMed/NCBI
- Google Scholar
57. Tersteegen A, Hedderich R (1999) Methanobacterium thermoautotrophicum encodes two multisubunit membrane-bound [NiFe] hydrogenases. Transcription of the operons and sequence analysis of the deduced proteins. Eur J Biochem 264: 930–943. pmid:10491142
- View Article
- PubMed/NCBI
- Google Scholar
58. Pisa KY, Huber H, Thomm M, Müller V (2007) A sodium ion-dependent A1AO ATP synthase from the hyperthermophilic archaeon Pyrococcus furiosus. FEBS J 274: 3928–3938. pmid:17614964
- View Article
- PubMed/NCBI
- Google Scholar
59. Jinich A, Rappoport D, Dunn I, Sanchez-Lengeling B, Olivares-Amaya R, et al. (2014) Quantum Chemical Approach to Estimating the Thermodynamics of Metabolic Reactions. Sci Rep 4: 7022. pmid:25387603
- View Article
- PubMed/NCBI
- Google Scholar
60. Zarraonaindia I, Smith DP, Gilbert JA (2013) Beyond the genome: community-level analysis of the microbial world. Biol Philos 28: 261–282. pmid:23482824
- View Article
- PubMed/NCBI
- Google Scholar
61. Noor E, Haraldsdóttir HS, Milo R, Fleming RMT (2013) Consistent Estimation of Gibbs Energy Using Component Contributions. PLoS Comput Biol 9: e1003098. pmid:23874165
- View Article
- PubMed/NCBI
- Google Scholar
62. Schellenberger J, Que R, Fleming RMT, Thiele I, Orth JD, et al. (2011) Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox v2.0. Nat Protoc 6: 1290–1307. pmid:21886097
- View Article
- PubMed/NCBI
- Google Scholar

[ref1] 1. Schink B (2002) Synergistic interactions in the microbial world. Antonie Van Leeuwenhoek 81: 257–261. pmid:12448724
View Article
PubMed/NCBI
Google Scholar

[2] View Article

[3] PubMed/NCBI

[4] Google Scholar

[ref2] 2. McInerney MJ, Struchtemeyer CG, Sieber JR, Mouttaki H, Stams AJM, et al. (2008) Physiology, ecology, phylogeny, and genomics of microorganisms capable of syntrophic metabolism. Ann N Y Acad Sci 1125: 58–72. pmid:18378587
View Article
PubMed/NCBI
Google Scholar

[6] View Article

[7] PubMed/NCBI

[8] Google Scholar

[ref3] 3. Stams AJM, Plugge CM (2009) Electron transfer in syntrophic communities of anaerobic bacteria and archaea. Nat Rev Microbiol 7: 568–577. pmid:19609258
View Article
PubMed/NCBI
Google Scholar

[10] View Article

[11] PubMed/NCBI

[12] Google Scholar

[ref4] 4. McInerney MJ, Sieber JR, Gunsalus RP (2009) Syntrophy in anaerobic global carbon cycles. Curr Opin Biotechnol 20: 623–632. pmid:19897353
View Article
PubMed/NCBI
Google Scholar

[14] View Article

[15] PubMed/NCBI

[16] Google Scholar

[ref5] 5. Schink B (1997) Energetics of syntrophic cooperation in methanogenic degradation. Microbiol Mol Biol Rev 61: 262–280. pmid:9184013
View Article
PubMed/NCBI
Google Scholar

[18] View Article

[19] PubMed/NCBI

[20] Google Scholar

[ref6] 6. Stams AJM, de Bok FAM, Plugge CM, van Eekert MHA, Dolfing J, et al. (2006) Exocellular electron transfer in anaerobic microbial communities. Environ Microbiol 8: 371–382. pmid:16478444
View Article
PubMed/NCBI
Google Scholar

[22] View Article

[23] PubMed/NCBI

[24] Google Scholar

[ref7] 7. Thauer RK, Kaster A-K, Seedorf H, Buckel W, Hedderich R (2008) Methanogenic archaea: ecologically relevant differences in energy conservation. Nat Rev Microbiol 6: 579–591. pmid:18587410
View Article
PubMed/NCBI
Google Scholar

[26] View Article

[27] PubMed/NCBI

[28] Google Scholar

[ref8] 8. Müller N, Worm P, Schink B, Stams AJM, Plugge CM (2010) Syntrophic butyrate and propionate oxidation processes: from genomes to reaction mechanisms. Environ Microbiol Rep 2: 489–499. pmid:23766220
View Article
PubMed/NCBI
Google Scholar

[30] View Article

[31] PubMed/NCBI

[32] Google Scholar

[ref9] 9. Sieber JR, McInerney MJ, Gunsalus RP (2012) Genomic insights into syntrophy: the paradigm for anaerobic metabolic cooperation. Annu Rev Microbiol 66: 429–452. pmid:22803797
View Article
PubMed/NCBI
Google Scholar

[34] View Article

[35] PubMed/NCBI

[36] Google Scholar

[ref10] 10. Oberhardt MA, Palsson BØ, Papin JA (2009) Applications of genome-scale metabolic reconstructions. Mol Syst Biol 5: 320. pmid:19888215
View Article
PubMed/NCBI
Google Scholar

[38] View Article

[39] PubMed/NCBI

[40] Google Scholar

[ref11] 11. Lewis NE, Nagarajan H, Palsson BØ (2012) Constraining the metabolic genotype-phenotype relationship using a phylogeny of in silico methods. Nat Rev Microbiol 10: 291–305. pmid:22367118
View Article
PubMed/NCBI
Google Scholar

[42] View Article

[43] PubMed/NCBI

[44] Google Scholar

[ref12] 12. Zomorrodi AR, Suthers PF, Ranganathan S, Maranas CD (2012) Mathematical optimization applications in metabolic networks. Metab Eng 14: 672–686. pmid:23026121
View Article
PubMed/NCBI
Google Scholar

[46] View Article

[47] PubMed/NCBI

[48] Google Scholar

[ref13] 13. Mahadevan R, Palsson BØ, Lovley DR (2011) In situ to in silico and back: elucidating the physiology and ecology of Geobacter spp. using genome-scale modelling. Nat Rev Microbiol 9: 39–50. pmid:21132020
View Article
PubMed/NCBI
Google Scholar

[50] View Article

[51] PubMed/NCBI

[52] Google Scholar

[ref14] 14. Osterlund T, Nookaew I, Nielsen J (2012) Fifteen years of large scale metabolic modeling of yeast: Developments and impacts. Biotechnol Adv 30: 979–988. pmid:21846501
View Article
PubMed/NCBI
Google Scholar

[54] View Article

[55] PubMed/NCBI

[56] Google Scholar

[ref15] 15. McCloskey D, Palsson BØ, Feist AM (2013) Basic and applied uses of genome-scale metabolic network reconstructions of Escherichia coli. Mol Syst Biol 9: 661. pmid:23632383
View Article
PubMed/NCBI
Google Scholar

[58] View Article

[59] PubMed/NCBI

[60] Google Scholar

[ref16] 16. Stolyar SM, Van Dien SJ, Hillesland KL, Pinel N, Lie TJ, et al. (2007) Metabolic modeling of a mutualistic microbial community. Mol Syst Biol 3: 92. pmid:17353934
View Article
PubMed/NCBI
Google Scholar

[62] View Article

[63] PubMed/NCBI

[64] Google Scholar

[ref17] 17. Wintermute EH, Silver PA (2010) Emergent cooperation in microbial metabolism. Mol Syst Biol 6: 407. pmid:20823845
View Article
PubMed/NCBI
Google Scholar

[66] View Article

[67] PubMed/NCBI

[68] Google Scholar

[ref18] 18. Freilich S, Zarecki R, Eilam O, Segal ES, Henry CS, et al. (2011) Competitive and cooperative metabolic interactions in bacterial communities. Nat Commun 2: 589. pmid:22158444
View Article
PubMed/NCBI
Google Scholar

[70] View Article

[71] PubMed/NCBI

[72] Google Scholar

[ref19] 19. Klitgord N, Segrè D (2010) Environments that induce synthetic microbial ecosystems. PLoS Comput Biol 6: e1001002. pmid:21124952
View Article
PubMed/NCBI
Google Scholar

[74] View Article

[75] PubMed/NCBI

[76] Google Scholar

[ref20] 20. Bordbar A, Lewis NE, Schellenberger J, Palsson BØ, Jamshidi N (2010) Insight into human alveolar macrophage and M. tuberculosis interactions via metabolic reconstructions. Mol Syst Biol 6: 422. pmid:20959820
View Article
PubMed/NCBI
Google Scholar

[78] View Article

[79] PubMed/NCBI

[80] Google Scholar

[ref21] 21. Lewis NE, Schramm G, Bordbar A, Schellenberger J, Andersen MP, et al. (2010) Large-scale in silico modeling of metabolic interactions between cell types in the human brain. Nat Biotechnol 28: 1279–1285. pmid:21102456
View Article
PubMed/NCBI
Google Scholar

[82] View Article

[83] PubMed/NCBI

[84] Google Scholar

[ref22] 22. Heinken A, Sahoo S, Fleming RMT, Thiele I (2013) Systems-level characterization of a host-microbe metabolic symbiosis in the mammalian gut. Gut Microbes 4: 28–40. pmid:23022739
View Article
PubMed/NCBI
Google Scholar

[86] View Article

[87] PubMed/NCBI

[88] Google Scholar

[ref23] 23. Shoaie S, Karlsson FH, Mardinoglu A, Nookaew I, Bordel S, et al. (2013) Understanding the interactions between bacteria in the human gut through metabolic modeling. Sci Rep 3: 2532. pmid:23982459
View Article
PubMed/NCBI
Google Scholar

[90] View Article

[91] PubMed/NCBI

[92] Google Scholar

[ref24] 24. Nagarajan H, Embree M, Rotaru A-E, Shrestha PM, Feist AM, et al. (2013) Characterization and modelling of interspecies electron transfer mechanisms and microbial community dynamics of a syntrophic association. Nat Commun 4: 2809. pmid:24264237
View Article
PubMed/NCBI
Google Scholar

[94] View Article

[95] PubMed/NCBI

[96] Google Scholar

[ref25] 25. Zomorrodi AR, Maranas CD (2012) OptCom: A Multi-Level Optimization Framework for the Metabolic Modeling and Analysis of Microbial Communities. PLoS Comput Biol 8: e1002363. pmid:22319433
View Article
PubMed/NCBI
Google Scholar

[98] View Article

[99] PubMed/NCBI

[100] Google Scholar

[ref26] 26. Zomorrodi AR, Islam MM, Maranas CD (2014) d-OptCom: Dynamic multi-level and multi-objective metabolic modeling of microbial communities. ACS Synth Biol 3: 247–257. pmid:24742179
View Article
PubMed/NCBI
Google Scholar

[102] View Article

[103] PubMed/NCBI

[104] Google Scholar

[ref27] 27. Orth JD, Thiele I, Palsson BØ (2010) What is flux balance analysis? Nat Biotechnol 28: 245–248. pmid:20212490
View Article
PubMed/NCBI
Google Scholar

[106] View Article

[107] PubMed/NCBI

[108] Google Scholar

[ref28] 28. Khandelwal RA, Olivier BG, Röling WFM, Teusink B, Bruggeman FJ (2013) Community flux balance analysis for microbial consortia at balanced growth. PLoS One 8: e64567. pmid:23741341
View Article
PubMed/NCBI
Google Scholar

[110] View Article

[111] PubMed/NCBI

[112] Google Scholar

[ref29] 29. Beard DA, Liang S, Qian H (2002) Energy balance for analysis of complex metabolic networks. Biophys J 83: 79–86. pmid:12080101
View Article
PubMed/NCBI
Google Scholar

[114] View Article

[115] PubMed/NCBI

[116] Google Scholar

[ref30] 30. Qian H, Beard DA, Liang S (2003) Stoichiometric network theory for nonequilibrium biochemical systems. Eur J Biochem 270: 415–421. pmid:12542691
View Article
PubMed/NCBI
Google Scholar

[118] View Article

[119] PubMed/NCBI

[120] Google Scholar

[ref31] 31. Schellenberger J, Lewis NE, Palsson BØ (2011) Elimination of thermodynamically infeasible loops in steady-state metabolic models. Biophys J 100: 544–553. pmid:21281568
View Article
PubMed/NCBI
Google Scholar

[122] View Article

[123] PubMed/NCBI

[124] Google Scholar

[ref32] 32. Kümmel A, Panke S, Heinemann M, Kummel A (2006) Putative regulatory sites unraveled by network-embedded thermodynamic analysis of metabolome data. Mol Syst Biol 2: 2006.0034.
View Article
Google Scholar

[126] View Article

[127] Google Scholar

[ref33] 33. Henry CS, Broadbelt LJ, Hatzimanikatis V (2007) Thermodynamics-based metabolic flux analysis. Biophys J 92: 1792–1805. pmid:17172310
View Article
PubMed/NCBI
Google Scholar

[129] View Article

[130] PubMed/NCBI

[131] Google Scholar

[ref34] 34. Hamilton JJ, Dwivedi V, Reed JL (2013) Quantitative assessment of thermodynamic constraints on the solution space of genome-scale metabolic models. Biophys J 105: 512–522. pmid:23870272
View Article
PubMed/NCBI
Google Scholar

[133] View Article

[134] PubMed/NCBI

[135] Google Scholar

[ref35] 35. Dolfing J (1987) Microbiological aspects of granular methanogenic sludge. Wageningen Agricultural University.

[ref36] 36. Stams AJM, Van Dijk JB, Dijkema C, Plugge CM (1993) Growth of syntrophic propionate-oxidizing bacteria with fumarate in the absence of methanogenic bacteria. Appl Environ Microbiol 59: 1114–1119. pmid:16348912
View Article
PubMed/NCBI
Google Scholar

[138] View Article

[139] PubMed/NCBI

[140] Google Scholar

[ref37] 37. Harmsen HJM, Van Kuijk BLM, Plugge CM, Akkermans ADL, De Vos WM, et al. (1998) Syntrophobacter fumaroxidans sp. nov., a syntrophic propionate-degrading sulfate-reducing bacterium. Int J Syst Bacteriol 48: 1383–1387. pmid:9828440
View Article
PubMed/NCBI
Google Scholar

[142] View Article

[143] PubMed/NCBI

[144] Google Scholar

[ref38] 38. Dong X, Stams AJM (1995) Evidence for H2 and formate formation during syntrophic butyrate and propionate degradation. Anaerobe 1: 35–39. pmid:16887505
View Article
PubMed/NCBI
Google Scholar

[146] View Article

[147] PubMed/NCBI

[148] Google Scholar

[ref39] 39. Stams AJM, Dong X (1995) Role of formate and hydrogen in the degradation of propionate and butyrate by defined suspended cocultures of acetogenic and methanogenic bacteria. Antonie Van Leeuwenhoek 68: 281–284. pmid:8821782
View Article
PubMed/NCBI
Google Scholar

[150] View Article

[151] PubMed/NCBI

[152] Google Scholar

[ref40] 40. Dong X, Plugge CM, Stams AJM (1994) Anaerobic degradation of propionate by a mesophilic acetogenic bacterium in coculture and triculture with different methanogens. Appl Environ Microbiol 60: 2834–2838. pmid:16349350
View Article
PubMed/NCBI
Google Scholar

[154] View Article

[155] PubMed/NCBI

[156] Google Scholar

[ref41] 41. Benedict MN, Gonnerman MC, Metcalf WW, Price ND (2012) Genome-scale metabolic reconstruction and hypothesis testing in the methanogenic archaeon Methanosarcina acetivorans C2A. J Bacteriol 194: 855–865. pmid:22139506
View Article
PubMed/NCBI
Google Scholar

[158] View Article

[159] PubMed/NCBI

[160] Google Scholar

[ref42] 42. Agren R, Liu L, Shoaie S, Vongsangnak W, Nookaew I, et al. (2013) The RAVEN Toolbox and Its Use for Generating a Genome-scale Metabolic Model for Penicillium chrysogenum. PLoS Comput Biol 9: e1002980. pmid:23555215
View Article
PubMed/NCBI
Google Scholar

[162] View Article

[163] PubMed/NCBI

[164] Google Scholar

[ref43] 43. Kanehisa M, Goto S (2000) KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res 28: 27–30. pmid:10592173
View Article
PubMed/NCBI
Google Scholar

[166] View Article

[167] PubMed/NCBI

[168] Google Scholar

[ref44] 44. Schwörer B, Thauer RK (1991) Activities of formylmethanofuran dehydrogenase, methylenetetrahydromethanopterin dehydrogenase, methylenetetrahydromethanopterin reductase, and heterodisulfide reductase in methanogenic bacteria. Arch Microbiol 155: 459–465.
View Article
Google Scholar

[170] View Article

[171] Google Scholar

[ref45] 45. Thauer RK (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144: 2377–2406. pmid:9782487
View Article
PubMed/NCBI
Google Scholar

[173] View Article

[174] PubMed/NCBI

[175] Google Scholar

[ref46] 46. Ferry JG (1999) Enzymology of one-carbon metabolism in methanogenic pathways. FEMS Microbiol Rev 23: 13–38. pmid:10077852
View Article
PubMed/NCBI
Google Scholar

[177] View Article

[178] PubMed/NCBI

[179] Google Scholar

[ref47] 47. Liu Y, Whitman WB (2008) Metabolic, phylogenetic, and ecological diversity of the methanogenic archaea. Ann N Y Acad Sci 1125: 171–189. pmid:18378594
View Article
PubMed/NCBI
Google Scholar

[181] View Article

[182] PubMed/NCBI

[183] Google Scholar

[ref48] 48. Plugge CM, Henstra AM, Worm P, Swarts DC, Paulitsch-Fuchs AH, et al. (2012) Complete genome sequence of Syntrophobacter fumaroxidans strain (MPOB(T)). Stand Genomic Sci 7: 91–106. pmid:23450070
View Article
PubMed/NCBI
Google Scholar

[185] View Article

[186] PubMed/NCBI

[187] Google Scholar

[ref49] 49. De Bok FAM, Roze EHA, Stams AJM (2002) Hydrogenases and formate dehydrogenases of Syntrophobacter fumaroxidans. Antonie Van Leeuwenhoek 81: 283–291. pmid:12448727
View Article
PubMed/NCBI
Google Scholar

[189] View Article

[190] PubMed/NCBI

[191] Google Scholar

[ref50] 50. De Bok FAM, Hagedoorn P-L, Silva PJ, Hagen WR, Schiltz E, et al. (2003) Two W-containing formate dehydrogenases (CO2-reductases) involved in syntrophic propionate oxidation by Syntrophobacter fumaroxidans. Eur J Biochem 270: 2934–2942.
View Article
Google Scholar

[193] View Article

[194] Google Scholar

[ref51] 51. Worm P, Stams AJM, Cheng X, Plugge CM (2011) Growth- and substrate-dependent transcription of formate dehydrogenase and hydrogenase coding genes in Syntrophobacter fumaroxidans and Methanospirillum hungatei. Microbiology 157: 280–289. pmid:20884694
View Article
PubMed/NCBI
Google Scholar

[196] View Article

[197] PubMed/NCBI

[198] Google Scholar

[ref52] 52. Plugge CM, Dijkema C, Stams AJM (1993) Acetyl-CoA cleavage pathway in a syntrophic propionate oxidizing bacterium growing on fumarate in the absence of methanogens. FEMS Microbiol Lett 110: 71–76.
View Article
Google Scholar

[200] View Article

[201] Google Scholar

[ref53] 53. Scholten JCM, Conrad R (2000) Energetics of syntrophic propionate oxidation in defined batch and chemostat cocultures. Appl Environ Microbiol 66: 2934–2942. pmid:10877789
View Article
PubMed/NCBI
Google Scholar

[203] View Article

[204] PubMed/NCBI

[205] Google Scholar

[ref54] 54. Lewis NE, Hixson KK, Conrad TM, Lerman JA, Charusanti P, et al. (2010) Omic data from evolved E. coli are consistent with computed optimal growth from genome-scale models. Mol Syst Biol 6: 390. pmid:20664636
View Article
PubMed/NCBI
Google Scholar

[207] View Article

[208] PubMed/NCBI

[209] Google Scholar

[ref55] 55. Ekiel I, Smith ICP, Sprott GD (1983) Biosynthetic pathways in Methanospirillum hungatei as determined by 13C nuclear magnetic resonance. J Bacteriol 156: 316–326. pmid:6619097
View Article
PubMed/NCBI
Google Scholar

[211] View Article

[212] PubMed/NCBI

[213] Google Scholar

[ref56] 56. De Bok FAM, Luijten MLGC, Stams AJM (2002) Biochemical Evidence for Formate Transfer in Syntrophic Propionate-Oxidizing Cocultures of Syntrophobacter fumaroxidans and Methanospirillum hungatei. Appl Environ Microbiol 68: 4247–4252. pmid:12200272
View Article
PubMed/NCBI
Google Scholar

[215] View Article

[216] PubMed/NCBI

[217] Google Scholar

[ref57] 57. Tersteegen A, Hedderich R (1999) Methanobacterium thermoautotrophicum encodes two multisubunit membrane-bound [NiFe] hydrogenases. Transcription of the operons and sequence analysis of the deduced proteins. Eur J Biochem 264: 930–943. pmid:10491142
View Article
PubMed/NCBI
Google Scholar

[219] View Article

[220] PubMed/NCBI

[221] Google Scholar

[ref58] 58. Pisa KY, Huber H, Thomm M, Müller V (2007) A sodium ion-dependent A1AO ATP synthase from the hyperthermophilic archaeon Pyrococcus furiosus. FEBS J 274: 3928–3938. pmid:17614964
View Article
PubMed/NCBI
Google Scholar

[223] View Article

[224] PubMed/NCBI

[225] Google Scholar

[ref59] 59. Jinich A, Rappoport D, Dunn I, Sanchez-Lengeling B, Olivares-Amaya R, et al. (2014) Quantum Chemical Approach to Estimating the Thermodynamics of Metabolic Reactions. Sci Rep 4: 7022. pmid:25387603
View Article
PubMed/NCBI
Google Scholar

[227] View Article

[228] PubMed/NCBI

[229] Google Scholar

[ref60] 60. Zarraonaindia I, Smith DP, Gilbert JA (2013) Beyond the genome: community-level analysis of the microbial world. Biol Philos 28: 261–282. pmid:23482824
View Article
PubMed/NCBI
Google Scholar

[231] View Article

[232] PubMed/NCBI

[233] Google Scholar

[ref61] 61. Noor E, Haraldsdóttir HS, Milo R, Fleming RMT (2013) Consistent Estimation of Gibbs Energy Using Component Contributions. PLoS Comput Biol 9: e1003098. pmid:23874165
View Article
PubMed/NCBI
Google Scholar

[235] View Article

[236] PubMed/NCBI

[237] Google Scholar

[ref62] 62. Schellenberger J, Que R, Fleming RMT, Thiele I, Orth JD, et al. (2011) Quantitative prediction of cellular metabolism with constraint-based models: the COBRA Toolbox v2.0. Nat Protoc 6: 1290–1307. pmid:21886097
View Article
PubMed/NCBI
Google Scholar

[239] View Article

[240] PubMed/NCBI

[241] Google Scholar

Figures

Abstract

Author Summary

Introduction

Results

Testing and Parameterizing the iMhu428 Metabolic Model

Testing and Parameterizing the iSfu648 Metabolic Model

ATP Production Mechanisms in S. fumaroxidans

Model Predictions of ATP Production by S. fumaroxidans

Evaluating Thermodynamics of H2 Production

Parameterization of the iSfu648 Metabolic Model

Uptake and Secretion at Maximal Growth

Behavior of M. hungatei and S. fumaroxidans in Coculture

Discussion

Ion Transport in the iMhu428 Metabolic Model

H2 Production in the iSfu648 Metabolic Model

Formate and H2 Transfer in Coculture

Methods

Reconstruction of the iMhu428 Metabolic Model

Reconstruction of the iSfu648 Metabolic Model

Thermodynamics-Based Metabolic Flux Analysis (TMFA)

Parsimonious TMFA (pTMFA)

Coculture Model

Simulations

Supporting Information

S1 Dataset. Supporting tables.

S2 Dataset. Excel version of the iMhu428 model.

S3 Dataset. SBML version of the iMhu428 model.

S4 Dataset. Excel version of the iSfu648 model.

S5 Dataset. SBML version of the iSfu648 model.

S6 Dataset. Excel version of the balanced KEGG database.

S7 Dataset. SBML version of the balanced KEGG database.

S1 File. Molfile structure files for all metabolites in the iMhu428 GEM.

S2 File. Molfile structure files for all metabolites in the iSfu648 GEM.

S1 Text. Supporting text.

Author Contributions

References

Evaluating Thermodynamics of H₂ Production

H₂ Production in the iSfu648 Metabolic Model

Formate and H₂ Transfer in Coculture