Genome scale metabolic modelling of human gut microbes to inform rational community design

Abstract

BackgroundThe human gut microbiome plays a pivotal role in health and disease, influenced significantly by diet, particularly through the intake of digestion-resistant carbohydrates (DRCs). Emerging evidence underscores the potential of DRC supplementation in modulating the gut microbiome towards health-promoting metabolic outputs, notably through the fermentation of DRCs into short-chain fatty acids (SCFAs) including butyrate. However, the effectiveness of such interventions is hindered by the inherent complexity of microbial communities and the variable functional capacity of microbiomes across individuals. Improved understanding of gut ecology is necessary to move past interventions with transient benefits.ResultsThis study leverages genome-scale metabolic models (GEMs) to characterise the metabolic capabilities of 598 stable human gut colonising strains from the AGORA database. We infer the strains’ abilities to utilize dietary carbohydrates of varying complexities and produce metabolites that mediate interactions with other microbes and the host.Our analysis reveals a stratified functional landscape where prominent bacterial families show predispositions towards primary or secondary degrader roles based on their carbohydrate utilisation capabilities. Further, we identify metabolite production profiles that exceed phylogenetic variation in our sample. These results offer a comprehensive functional mapping of carbohydrate metabolism across a wide array of gut microbes, shedding light on the complex trophic networks underpinning the gut ecosystem.Supporting DRC-based interventions with rationally designed microbial communities can better guarantee the delivery of the intended health-promoting metabolic outputs. Applying our functional assessment, and principles of reverse ecology and network analysis, we propose a novel framework for the rational design gut microbial communities, where trophic networks are optimised to produce target metabolites from selected DRCs. Our results further suggest that this framework can predict resilient minimal communities, an important trait in the constantly changing human gut nutritional environment.ConclusionOur work provides novel insights into gut microbial ecology as well as intervention and consortia design. The identified metabolic capabilities of individual strains inform the rational design of a purpose-based microbial community to optimise butyrate production from inulin degradation. The framework we propose herein sets a foundation for future efforts aimed at the rational design of interventions that target the human gut microbiome to improve health.