Anti-islet autoantibodies trigger autoimmune diabetes in the presence of an increased frequency of islet-reactive CD4 T cells

OBJECTIVE

To define cellular mechanisms by which B cells promote type 1 diabetes.

RESEARCH DESIGN AND METHODS

The study measured islet-specific CD4 T cell regulation in T-cell receptor transgenic mice with elevated frequencies of CD4 T cells recognizing hen egg lysozyme (HEL) autoantigen expressed in islet β-cells and thymic epithelium under control of the insulin-gene promoter. The effects of a mutation in Roquin that dysregulates T follicular helper (Tfh) cells to promote B-cell activation and anti-islet autoantibodies were studied, as were the effects of HEL antigen–presenting B cells and passively transferred or maternally transmitted anti-islet HEL antibodies.

RESULTS

Mouse anti-islet IgG antibodies—either formed as a consequence of excessive Tfh activity, maternally transmitted, or passively transferred—caused a breakdown of tolerance in islet-reactive CD4⁺ cells and fast progression to diabetes. Progression to diabetes was ameliorated in the absence of B cells or when the B cells could not secrete islet-specific IgG. Anti-islet antibodies increased the survival of proliferating islet-reactive CD4⁺ T cells. FcγR blockade delayed and reduced the incidence of autoimmune diabetes.

CONCLUSIONS

B cells can promote type 1 diabetes by secreting anti-islet autoantibodies that act in an FcγR-mediated manner to enhance the expansion of islet-reactive CD4 T cells and cooperate with inherited defects in thymic and peripheral CD4 T–cell tolerance. Cooperation between inherited variants affecting CD4 T–cell tolerance and anti-islet autoantibodies should be examined in epidemiological studies and in studies examining the efficacy of B-cell depletion.Type 1 diabetes is an autoimmune disease caused by islet-reactive T cells that destroy insulin-producing β-cells. T-cell tolerance mechanisms normally prevent autoimmune diabetes in mice and humans (1,2), and T cell–directed treatment with anti-CD3 monoclonal antibodies slows the loss of insulin production in the 1st year after diabetes diagnosis (3). With the focus on T cells, the role of antibodies and B cells in type 1 diabetes etiopathogenesis is often neglected. A single clinical case of type 1 diabetes in a person with X-linked agammaglobulinemia indicates that type 1 diabetes can develop in the absence of antibodies and B cells (4). However, antibodies to insulin and other islet antigens are predictive of subsequent type 1 diabetes (5), and B cells are required for diabetes development in the nonobese diabetic (NOD) mouse model of type 1 diabetes (6–8). A recent clinical trial of B-cell depletion with anti-CD20 in humans with newly diagnosed type 1 diabetes demonstrated a significant delay in further loss of insulin synthesis and established a role for B cells (9), but how B cells contribute to type 1 diabetes pathology remains unclear. Better understanding of mechanisms by which B cells can promote type 1 diabetes may improve the efficacy of interventions that target B cells.B cells may contribute to type 1 diabetes pathology by capturing and presenting autoantigens to islet-reactive CD4+ T cells (10–12). The effect on CD4 T cells is particularly relevant because B cells present captured antigens to CD4 cells via major histocompatibility complex (MHC) II molecules, and type 1 diabetes is strongly associated with particular HLA-DR haplotypes. However, an alternative possibility is that B cells contribute to type 1 diabetes pathology by secreting autoantibodies. In humans, the high predictive value of anti-islet autoantibodies for progression to type 1 diabetes is well documented (13–17). However, it remains unresolved whether or not autoantibodies are simply a biomarker of breakdown in T-cell tolerance (5). Indirect evidence against a pathogenic role for autoantibodies comes from the decreased incidence of type 1 diabetes in offspring of diabetic mothers compared with diabetic fathers, despite transmission of maternal anti-islet autoantibodies (18–20). Paradoxical protection from diabetes in children with maternally derived islet-autoantibodies has been reported (21), although this protection was not observed in children with the high-risk HLA-DR3/DR4-DQ8 genotype. This observation could indicate that the effects of anti-islet antibodies are influenced by underlying heterogeneity in the efficiency of CD4 T–cell tolerance mechanisms, which are affected by variability in MHC II antigen presentation.Only a few experimental studies have investigated whether or not secreted autoantibodies influence type 1 diabetes progression, and none have examined their influence on cellular mechanisms of tolerance in islet-reactive CD4 T cells. In NOD mice, diabetes was reduced when maternal transmission of antibodies from NOD mothers was prevented (22,23). It is not yet known whether these maternal effects reflect transmission of anti-islet autoantibodies, antibodies against microbial flora, or changes in maternal transmission of microbial commensals themselves. In one study, type 1 diabetes incidence in litters from NOD mothers clustered according to the titer of anti-insulin antibodies in the mother’s serum (24); however, in another study there was no correlation (25). Supporting a pathogenic role for autoantibodies, NOD mice lacking different activating Fcγ receptors were shown to be protected from diabetes (26). Also, passive transfer of rabbit or mouse antibodies against islet-expressed ovalbumin (OVA) has been shown to enhance activation of islet-reactive CD8+ T cells and break tolerance (27). Hence, experimental studies are needed to directly investigate the effects of anti-islet autoantibodies on islet-specific CD4 T cells and how these may interact with inherited defects in islet-specific CD4 T–cell tolerance. Here we find that anti-islet autoantibodies are potent cofactors in type 1 diabetes progression and not simply markers of breakdown in CD4+ T–cell tolerance.