Characterize the gut microbiota during the first 3 years of life and elucidate a mechanism by which it can shape the immune system and impact the incidence of autoimmune diseases such as type 1 diabetes (T1D).

A subcohort from the DIABIMMUNE (http://www.diabimmune.org) study of 1000 infants from Finland, Estonia, and Russia matched by sex, age, and human leukocyte antigen–associated risk factors. In Finland and Estonia, autoimmunity is more common than in Russia.

Stool and serum samples were collected periodically from birth up to 3 years of age. Stool samples were sequenced and their functional potential was compared by HUMAn2 software (http://huttenhower.sph.harvard.edu/humann2). NF-κB–dependent cytokines were determined in peripheral blood mononuclear cells after Escherichia coli or Bacteroides dorei lipopolysaccharide (LPS) stimulation. A nonobese-diabetic mouse model was used to evaluate effects of these 2 types of LPSs in the development of autoimmunity.

Sequencing of stool samples revealed a substantially different composition between Russian and Finnish or Estonian microbiota. Finns and Estonians had a greater abundance of B dorei, and the amount of B dorei correlated directly with insulin autoantibody levels. In terms of functional potential, glycolytic functions associated to milk oligosaccharide metabolism were increased in Russian samples, whereas LPS and lipid A biosynthetic processes were increased in the Finnish infants. Lipid A was mainly derived from E coli, but in Finland and Estonia it was also derived from B dorei. The authors show that E coli and B dorei LPSs are structurally and functionally different and that B dorei LPS inhibits NF-κB–dependent cytokine production induced by E coli LPS in a dose-dependent manner. Thus, in peripheral blood mononuclear cells, B dorei dampens the immunomodulatory effect of E coli. This was translated into an in vivo model because mice injected with E coli LPS had a lower incidence and later onset of T1D than mice injected with B dorei LPS. In these latter mice, their splenocytes were hyporesponsive 24 hours after E coli LPS injection.

Despite their geographical proximity, there is a significant difference in the composition and function of gut microbiota in infants from 3 different countries with distinct prevalence of autoimmune and allergic diseases. The presence of B dorei during early infancy could inhibit endotoxin tolerance induced by E coli LPS and predispose for autoimmunity.

The incidence of T1D is increasing worldwide, with an exponential increase in prevalence over the last 40 years. Despite extensive research, the specific causes of this rise are unknown, and primary prevention strategies are lacking. Recent evidence highlights the relevance of gut microbes and its impact on immunity and immunopathology. However, the definition of a protective microbiota is poorly defined. The authors of this study identify E coli as a protective element in microbiota and B dorei as a deleterious microbe and potential driver of autoimmune disease. Moreover, it provides direct evidence of a mechanistic link between these 2 components of the microbiota and the selective disruption of a specific immunomodulatory mechanism by which the risk of developing an autoimmune disease, specifically T1D, could increase. Future research should be focused on establishing specific causes of detrimental changes in the composition of gut microbiota, understanding its full range of implications, and developing strategies to favor the presence of protective microbes in an attempt to modify the increasing incidence of autoimmune and allergic diseases.