A large body of data, including our own studies in the Old Order Mennonites (OOM), suggests that living on farms is associated with a lower risk of food allergy, asthma and other atopic diseases, compared to urban living. While increasing evidence indicates that gut dysbiosis precedes the development of atopy, atopic eczema and food allergy/sensitization, the role of infant gut microbiome in protection against atopic diseases in the farm lifestyle communities are lacking. Our preliminary data in the OOM pilot study suggest enrichment of Bifidobacteria, especially Bifidobacterium longum ssp. infantis (i.e., B. infantis) both in the rate of gut colonization as well as in abundance in the OOM compared to urban/suburban Rochester infants. In the United States, the majority of breastfed infants lack colonization with B. infantis, suggesting extinction in modern societies, while in countries such as Bangladesh, colonization with B. infantis is found in >90% of infants in an overwhelming abundance. Access to a North American community, such as the OOM with a lifestyle dating back 100 years and unusually high rates of B. infantis colonization provides an unprecedented opportunity to assess the impact of B. infantis and strain differences for their anti-allergic potential. Bifidobacteria have been shown to predominantly colonize breastfed infants and have major health benefits compared to those lacking Bifidobacteria due to their improved barrier function and ability to metabolize human milk oligosaccharides (HMOs), resulting in lower stool pH and short-chain fatty acid (SCFA) production. Specifically, B. infantis has been shown to be anti-inflammatory in intestinal epithelial cells and to attenuate allergic inflammation in animal models. B. infantis associated with enhanced vaccine responses in infants, and an increase in stool levels of lactate, also seen in our preliminary studies. However, B. infantis strains differ in their metabolic function and strain differences associated with protection against allergic diseases have not been assessed. We hypothesize that despite their genomic (and phylogenetic) similarities, B. infantis strains vary significantly in their metabolic phenotypes with broad implications to microbiome function. This project will utilize stool samples from an active NIH-funded longitudinal infant cohort including OOM farm-life and Rochester urban/suburban infants. Frozen and fresh stool samples will be characterized for B. infantis genomic and functional differences and associated metabolites as well as their anti-inflammatory potential to pick the most promising candidate strains. They will be further down selected in a gnotobiotic animal model for their potential to prevent food allergy. These studies aim to develop strategies for future prevention trials. 1