A Computational Modeling Approach Elucidate Metabolic Interactions between the Gut Microbiota Community and Their Impact on Host Metabolism

The human intestine possesses a diverse microbial ecosystem, the gut microbiota. Here, we apply multi-scale constraint-based modeling to predict host-microbe and microbe-microbe interactions in silico. First, a human gut microbe community, consisting of 11 manually curated and validated gut microbe reconstructions spanning three phyla, was joined with the global human reconstruction, Recon2. The effects of the different microbes on host metabolic tasks were systematically explored while simulating four different dietary regimes. Moreover, the 11 microbes were joined pairwise in all possible combinations and pairwise interactions were predicted on six nutrient environments. In a more extensive follow-up work, pairwise interactions in a representative community of 321 gut microbes were investigated. A variety of human body fluid metabolites were influenced by microbial presence, in agreement with the role of the microbiota as an additional organ. Natural interactions including commensalism, mutualism and competition differed for microbe-microbe pairs depending on the nutrient environment. Anoxic conditions promoted mutualistic behavior in certain pairs, which was abolished in the presence of oxygen. Pairs in the representative community of 321 microbes displayed less mutualistic and commensal behavior in the presence of oxygen, confirming our previous finding. We demonstrate that constraint-based modeling successfully predicts natural interspecies interactions and the metabolic effects of a gut microbiota community on its host. We present the most extensive constraint-based model of a mammalian intestinal ecosystem to date. It will prove useful in future efforts to elucidate the metabolism of this complex ecosystem and its effects on human health.