Characterizing the dynamic operations of model human gut communities in gnotobiotic mice and piglets

PROJECT SUMMARY DK70977 retains its focus on gnotobiotic models of the dynamic operations the human gut microbiota, and the development of new experimental designs/tools to analyze how dietary components, specifically different prebiotic polysaccharides added to diets representative of those consumed in the USA, shape microbial community structure/function. It is based on the following advances and leverages a partnership between 2 labs with complementary expertise. (1) Methods for creating clonally arrayed sequenced `personal' culture collections that capture the majority of bacterial taxa present in a human fecal microbiota sample. (2) Finding that culture collections from twins stably discordant for obesity transmit the donor's discordant adiposity phenotypes plus obesity-associated metabolic abnormalities to germfree (GF) mice. Co-housing results in invasion of specific bacterial species from the lean (Ln) community into the guts of cagemates harboring Ob co-twin's culture collection, preventing obesity and associated metabolic abnormalities. Invasion, principally by members of the Bacteroides occurs with a representative USA diet low in saturated fats/high fruits-vegetables diet (LoSF/HiFV) but not with a HiSF/LoFV diet. (3) Methods for performing genome-wide transposon mutagenesis of multiple Bacteroides strains and simultaneously identifying genes/metabolic pathways that confer their fitness in specified diet/community contexts (multi-taxon insertion sequencing, INSeq). (4) Methods for generating GF piglets. We will pursue 2 aims. Aim 1, Perform diet oscillation studies in Ln-colonized mice fed 160 different HiSF/LoFV diets supplemented with 40 different purified plant polysaccharides (PPS) in varying doses, some from the waste-streams of commercial food processing, and identify PPS that produce the greatest increase in abundance of a single targeted Bacteroides (`high selectivity, large effect size lead) or significant increases in the greatest number of targeted Bacteroides (`broad range, significant effect size lead). Followup mechanistic studies using (i) these leads, fed monotonously at different doses, and in diet oscillations of different periodicity, plus (ii) multi-taxon INSeq, microbial RNA-Seq, new methods for multiplex quantitative proteomics, and metabolomics, are intended to provide new definitions/insights about niche, syntrophic partnerships, foodwebs, competitiveness, and their impact on community/host metabolism. Leads will also be tested with the Ob collection supplemented with Bacteroides invaders. Aim 2, Application of these methods to gnotobiotic piglets, colonized with these defined communities, fed HiSF/LoFV ± PPS leads and gavaged with `artificial microscopic nutrient platforms' consisting of fluorescently-labeled paramagnetic beads coated with different PPS, that can be retrieved from gut samples by FACS, should provide an unprecedented level of spatial and temporal detail of how community structure and function varies along the length of the gut, how nutrient sharing relationships are established and operate in different PPS and community contexts and establish the translatability of data obtained from gnotobiotic mice to this second more human-like model.