Back

HLA-DQβ57, anti-insulin T cells and insulin mimicry in autoimmune diabetes

Rubio Garcia, A.; Paterou, A.; Powell Doherty, R. D.; Landry, L. G.; Lee, M.; Anderson, A. M.; Slawinski, H.; Ferreira, R. C.; Trzupek, D.; Szypowska, A.; Wicker, L. S.; Teyton, L.; Ternette, N.; Nakayama, M.; Todd, J. A.; Pekalski, M. L.

2022-05-14 allergy and immunology
10.1101/2022.05.11.22274678 medRxiv
Show abstract

Insulin autoimmunity and pancreatic islet {beta}-cell destruction, mediated by T cells, underlie the pathogenesis of type 1 diabetes (T1D). However, the mechanisms by which genetic and environmental risk factors interact to cause T1D remain incompletely understood. Here, we show how variation in the human leukocyte antigen (HLA) class II region, notably DQ{beta}157, the strongest T1D genetic risk factor, induces a positive thymic selection bias that favors amino acid motifs with negatively charged side chains in the peptide-binding CDR3{beta} region of CD4+ T cell receptors (TCRs). This bias was enriched in TCRs from T1D patients and predicted the proportion of anti-insulin infiltrates in pancreatic islets. We link host genetics and immunity to the environment by identifying over 100 gut bacterial proteins with significantly similar sequences to a primary epitope in T1D, insulin B9-25 peptide. We show that CD4+ T cells isolated from the pancreas of a newly diagnosed child with T1D recognised both insulin and a bacterial mimotope peptide, restricted by the most predisposing DQ molecule. Our results point to the evolution of certain microbial antigens towards mimicry in the regulation of immune tolerance to insulin.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.