Alteration in the gut microbiome influences diabetes development in AID-deficient NOD mice
Brown, Kristina   (Yale University)
Mentor: Wen, Li (Yale University)
Type I diabetes (T1D) is characterized by the autoimmune destruction of insulin-producing beta cells in the pancreas. This destruction is mainly mediated by auto-reactive T lymphocytes, however, B lymphocytes facilitate the autoimmune attack. One enzyme vital to B lymphocyte function is Activation-induced cytidine deaminase (AID). The role of AID is to allow B lymphocytes to produce different classes of immunoglobulins (Igs). Because AID is important in homeostasis of immune system and gut mucosal immunity in particular, there is a reason to believe that AID is involved in T1D development. However, little is known about this relationship. Given that the gut microbiome affects both mucosal immunity and diabetes development, we hypothesize that AID plays a role in modulating T1D pathogenesis. To test the hypothesis, we studied AID deficient (AID-/-) non obese diabetic (NOD) mice. NOD mouse is an excellent animal model of human T1D. To investigate how AID modulates T1D development, we first observed diabetes development in AID-/- NOD mice in different housing conditions: AID-/- NOD mice housed with AID sufficient (AID+/+) NOD mice or AID-/- NOD mice housed with AID-/- NOD mice The former was designated as co-housing (CH) and the latter was designated as non co-housing (NCH) and AID+/+ NOD mice were used as controls. We screened glycosuria weekly and diabetes was confirmed by blood glucose =250mg/dl. Since mice have coprophagia behavior, CH condition can influence the gut microbiome and mucosal Igs of the mice. We collected fecal samples from these mice to investigate the composition of gut microbiota in each group by sequencing bacterial DNA extracted from the stool samples. To evaluate the effect of AID on mucosal anti-microbial peptides, we collected tissue samples from small and large intestine and determined expression levels of five major types of mucosal antimicrobial peptides by quantitative PCR (qPCR). Moreover, we also tested the stimulatory effect of gut bacteria from AID-/- and AID+/+ mice on T lymphocytes. Our results showed that when in CH condition, the AID-/- mice developed significantly accelerated diabetes compared to the CH AID+/+ mice. However, when in NCH condition, AID-/- mice were significantly protected from T1D development. Gut bacterial sequencing results revealed that while Firmicutes was the predominant phlyum in all the groups of mice, AID-/-NCH mice had much reduced phlyum of Bacteroidetes than AID+/+ NCH mice (12.2% vs. 19.0%, p<0.0001). We also found that anti-microbial peptides, Reg3ß and RELMß, from the small intestines were significantly increased in the NCH AID-/-NOD mice as compared to CH AID-/- NOD mice. This suggests that high level of anti-microbial peptides can affect the composition of gut microbiome, which in turn may contribute to diabetes protection seen in NCH AID-/-NOD mice. To this end, we found that gut bacteria from NCH AID-/-NOD mice were significantly less stimulatory to CD8+ T cells compared to the gut bacteria from the mice in other groups. Overall, our results indicate that differences in gut microbial composition and up-regulation of Reg3ß and RELMß in the small intestine of NCH AID-/- mice may explain why these mice are protected from development of T1D while CH AID-/- have accelerated diabetes. Furthermore, the gut microbiome from NCH AID-/- mice was also less stimulatory to CD8+ T cells. In contrast, gut microbiome from CH AID-/- mice showed the strongest stimulatory effect to CD8+ T cells, suggesting that the gut microbiome in CH AID-/- mice may promote an over-responsive immune system that could contribute to accelerated T1D development. In conclusion, our results provide important new knowledge about the role of AID in the pathogenesis of T1D development, mediated by gut microbiome and the homeostasis of mucosal immunity.