(81d) Improving the Gut Residence Time of Probiotics through Functional Metagenomics

Engineering human-associated microbes holds great promise for treatment of chronic conditions, but the resulting health-promoting strains may not be well-suited for persistence in their target environment. As one solution, human-associated microbial communities may contain genes or pathways which improve the colonization potential of synthetic, therapeutic strains. To test this hypothesis, we focused on the trillions of bacteria which inhabit the human gut. This community performs many functions critical to the maintenance of health, including nutrient extraction, vitamin production, and defense from pathogens. Disruptions to the structure of this bacterial community can have significant negative impacts to the human host, including obesity, malnutrition, and cancer. As one method for correcting these pathologies, probiotics are orally ingested microorganisms which improve health, potentially through provision of additional metabolic functions, modulation of the host immune response, or competition for niche space with pathogens. In particular, E. coli Nissle (EcN) is an approved probiotic with demonstrated efficacy against inflammatory bowel disease. Unfortunately, the residence time of EcN in the gut is quite low, necessitating continual administration for treatment of this chronic condition. In order to discover factors limiting EcN colonization and improve the gut residence time of this important probiotic, we extracted metagenomic DNA from both established and developing gut microbial communities and shotgun-cloned tens of millions of 2-5kb fragments of this DNA into an expression vector. We next selected colonization-enhancing gene fragments via passage of EcN containing this metagenomic library in cell culture, germ-free mice, and mice containing a human-like microbiota consuming one of several human-relevant diets. We observed a reproducible, million-fold reduction in diversity of cloned fragments as our selections proceeded, indicating robust enrichment for factors which improve the persistence of EcN in these environments. Bioinformatic and biochemical analyses of these fragments indicated mechanisms of improved colonization, and specificity of recovered mechanisms to particular dietary contexts was observed. Furthermore, we found that EcN containing these colonization factors exhibited improved persistence over wild-type in a mouse model of Salmonella typhimurium infection, indicating the potential for exclusion of this pathogen via competition for niche space. Taken together, this work has the potential to significantly improve our understanding and use of probiotics while providing a framework for developing robust chassis strains for future synthetic biology efforts in the human gut microbiota.