Changes in autoreactive B cell lifestyle early in development of autoimmunity
Date
2017
Authors
Smith, Mia J., author
Dow, Steven, advisor
Cambier, John, advisor
Avery, Anne, committee member
Webb, Craig, committee member
Journal Title
Journal ISSN
Volume Title
Abstract
Type 1 diabetes (T1D) is an autoimmune disorder characterized by destruction of the pancreatic beta cells, leading to decreased production of insulin and hyperglycemia. Although environmental factors contribute, genetic factors are likely the primary determinants of risk. With recent advances in GWAS studies, hundreds of risk-conferring alleles have been discovered for T1D. For most cases the exact mechanisms by which these genes and their gene products contribute to development of autoimmunity remains to be elucidated. However, given that T1D requires the activation of autoantigen-specific T and B cells that are normally silenced by immune tolerance, it is likely a combination of HLA and non-HLA alleles act in concert to undermine normal tolerance mechanisms, allowing activation of these autoreactive cells. Although T cells are the primary effectors of beta cell destruction in T1D, autoreactive B cells are thought to act primarily as antigen presenting cells. In a healthy individual, autoreactive B cells are normally silenced by one of three mechanisms: receptor editing, clonal deletion, or anergy. In this work I determined B cells bearing antigen receptors with high affinity for insulin are found only in the anergic B cell compartment, termed BND, of healthy individuals. Importantly, these cells leave this compartment in a proportion of first-degree relatives (FDRs), and in all autoantibody positive pre-diabetics and new onset diabetics. We posited people at risk for development of T1D carry autoimmune risk alleles that impair proper silencing of autoreactive B cells by anergy, allowing these cells to become activated and contribute to disease. In order to test this, I analyzed the HLA class II alleles and over 50 high risk non-HLA alleles in BND sufficient and deficient FDRs. I found loss of anergic insulin-binding B cells (IBCs) in FDRs was associated with the high risk T1D HLA alleles and polymorphisms in the high risk non-HLA loci, INS, PTPN2, PTPN22, and IKZF3. The associations of loss of B cell anergy with these particular risk alleles suggest insulin-reactive T cells and changes in negative regulation of B cell signaling contribute to the unstable anergic phenotype observed in autoimmune patients. In our T1D studies, we found loss of anergic IBCs was correlated with loss of the entire anergic B cell population, irrespective of their specificity, suggesting loss of B cell anergy could be a common phenomenon in other autoimmune diseases. In addition, many risk alleles for T1D are shared among other autoimmune diseases, including HLA and PTPN22, suggesting B cell anergy could be compromised in other autoimmune disorders in which similar contributing risk alleles are at play. Hence, I also analyzed the frequency and phenotype of thyroglobulin (Tg) and thyroid peroxidase (TPO) binding B cells, as well as total B cells, in early onset and long standing autoimmune thyroid disease (AITD) patients compared to healthy controls. Similar to studies in T1D, early onset AITD patients had a significant decrease in anergic Tg and TPO-binding B cells that was correlated with a decrease in total anergic B cells. Furthermore, loss of anergic Tg-binding B cells was inversely correlated with Tg autoantibodies and Tg-binding B cells expressed high levels of the activation marker CD86. These findings suggest activation of high affinity thyroid reactive B cells that are normally silenced by anergy, likely leads to production of autoantibodies. In order to further elucidate the possible contribution a breakdown in anergy of autoreactive B cells has in development of autoimmunity, I studied the phenotype and functional status of IBCs in diabetes susceptible (NOD) and diabetes resistant (C57BL/6) mice transgenic for the 125Tg heavy chain. This transgene increases the frequency of peripheral IBCs to a level that is easily detectable (~0.5-2% of total splenic B cells depending on the strain) [33]. In these mice, I found that high affinity IBCs were phenotypically and functionally anergic in C57BL/6 mice, but the equivalent in NOD appeared activated and functionally responsive, accumulated in the pancreas, and expressed insulin peptides in association with MHC II on their cell surface. Accumulation of these B cells in the pancreas correlated with retention and activation of insulin-reactive CD4 T cells. Hence, these mouse studies nicely summarize what I hypothesize occurs in autoimmune humans; namely, anergy is impaired in autoreactive B cells, likely due to genetic risk alleles, which allows them to become activated and provide critical antigen presenting function to cognate antigen-reactive T cells. These studies are significant in that they are the first studies to identify a breach in B cell anergy occurs early in development in multiple autoimmune disorders in humans, which is likely driven by a combination of autoimmune risk alleles that alter thresholds for B cell activation, enabling them to become activated and participate in disease through antigen presentation and autoantibody production. Furthermore, these studies highlight the utility of loss of B cell anergy as a possible biomarker for increased risk for development of autoimmune disorders.