The complex role of somatic hypermutation in supporting insulin recognition by B lymphocytes in at-risk type 1 diabetes participants

---

Presentation Time: 11:30 AM - 12:45 PM
Poster Board Number: B609
Abstract ID: 4626

Presenting Author:
Lindsay Bass , Graduate student at Vanderbilt Univ.

Abstract:

Insulin autoantibodies predict type 1 diabetes (T1D), yet the molecular basis for B cell receptor (BCR) recognition of insulin in humans has not been well-studied. We used single-cell RNA-seq/BCR-seq/CITE-seq and advanced hybridoma technology to identify insulin-binding B cells from the peripheral blood of n = 18 TrialNet Pathway to Prevention study participants who were positive for ≥ 2 islet autoantibodies and thus at high risk for diabetes. These participants were categorized as Stage 1 or Stage 2 based on normal or impaired glucose tolerance test results at the time of blood draw, respectively. We identified clonally expanded B cells in BCR-seq data based on n ≥ 3 BCRs per clonotype. Clonally expanded B cells expressed an autoreactive-prone CD21^lo CD27⁺ phenotype. Recombinant expression of these clonally expanded BCRs revealed that ~50% recognized insulin. Sequence analysis of insulin-binding BCRs revealed they ranged from 0-11% somatic hypermutation. For one BCR, germline reversion of the single amino acid mutation eliminated insulin recognition, whereas in another, germline reversion of all 22 mutations resulted in only modest diminution of insulin binding. These data highlight clonally expanded, CD21^lo memory B cells as a reservoir for insulin-binding B cells. Further, we show extensive somatic hypermutation of BCRs may not always substantially enhance insulin autoantigen recognition, and rather may simply be a byproduct of a chronic autoimmune response.