Probiotic Design through Microbial Community Modelling Using Genome-Scale Metabolic Models

The human gastrointestinal tract harbors a densely populated and sophisticated community of microorganisms. This microbial community outnumbers the human eukaryotic cells by a factor of ten and plays a significant role in maintaining human health by performing different metabolic tasks such as digestion of otherwise indigestible dietary components like complex carbohydrates as well as synthesis of valuable metabolites such as short-chain fatty acids, vitamins and essential amino acids. The gut microbiota seems to be of high relevance for human health, since its dysbiosis is associated with many health problems including cardiovascular disease, obesity and type 2 diabetes. These close relationships between gut microbiota composition and disease, have led to a growing interest in using probiotics to positively modulate the gut microbiota to prevent or treat some diseases.

Mathematical modelling has been proven to be a valuable tool to gain better understanding about the relevant interactions and community behavior. Through mathematical modelling of the gut microbiota, it is possible to evaluate different hypothesis and hereby gain mechanistic insights into how the gut microbiota composition affects host metabolism. GEnome-scale Metabolic (GEM) modelling is particularly well suited for this purpose as it is possible to reconstruct the metabolic networks of gut symbionts based on genomic information and then use constraint-based modelling for simulation of their metabolic functions. GEMs can be used to study microbial communities as well to predict interactions (cooperation and competition) in different media. The objective function of a highly complex community, however, is more complicated as there are several competing factors. We have implemented a novel approach to simulate a microbial community, in which there are two separate organism-level and community-level objective functions. We have used this approach to model the interactions between the human gut representatives and probiotic strains to predict optimal conditions for a stable desired process and towards human health benefits. This optimization algorithm integrates well with diet analysis and can hereby be used to understand the metabolic microbe-microbe, diet-microbe and host-microbe interactions. This method has the potential to design an optimum probiotic community in combination with prebiotics to be used as a therapeutic approach to alleviate digestive maladies such as severe diarrhea, inflammatory bowel disease and pouchitis.