(566b) Metabolic Modeling of Microbial Communities

Despite rapid progress, our ability to probe microbial communities for species-specific information of metabolite pools or metabolic fluxes remains limited. Most of the available information refers to species abundance and overall chemical exchanges with the environment. Computational modeling provides the missing link for explaining the metabolic interactions of microbial partners in communities in the context of conservation laws and optimality principles. While several modeling techniques have been developed for microbial communities, little emphasis has been placed on the need to impose a time-averaged constant growth rate across all members for a community to ensure co-existence and stability. To this end, we developed the SteadyCom optimization framework for predicting metabolic flux distributions consistent with the steady-state requirement. SteadyCom imposes restrictions on the allowable community species membership, composition and phenotypes. In contrast to the direct use of FBA, SteadyCom is able to reveal hidden interactions in a community as demonstrated in a community of four E. coli auxotrophic mutants and predict an abundance profile of a gut microbiota model with a good agreement to experimental data.

In separate efforts, we applied community metabolic modeling to analyze experimental observations in the gut microbiome. A working hypothesis regarding the production of a beneficial estrogen, (S)-equol by gut microbes was assessed in terms of metabolic consequences by simulating the interactions between three representative organisms (Bacteroidetes, Firmicutes and equol producer). The effect of diet and species abundance on the community’s ability to produce (S)-equol was explored and testable predictions were generated. In another case study, a ten-species gut microbiota metabolic model was developed to analyze the effect of bile salt hydrolase (BSH) activity and gut microbial productions under the treatment by an obesity-related drug, glycine-β-muricholic acid. Under the running hypothesis of the studies was that BSH activity expressed in Lactobacillus and Clostridia relieves the intestinal farnesoid X receptor (FXR) antagonism and in turn induces obesity, short-chain fatty acid productions and amino acid consumptions by the gut microbiota correlated to the experimental changes was predicted, supporting the hypothesis. These case studies allude to the promise of genome-scale metabolic modeling approach in microbiome research.