Autoreactivity to proinsulin (PI) is essential for diabetes in the non-obese diabetic (NOD) mouse and perhaps in humans whereas autoreactivity to other autoantigens eg IGRP is not essential and may represent epitope spreading. However, PI based clinical trials have failed to achieve therapeutic benefit. Timing of antigen-specific therapy may be crucial to achieve clinical efficacy. To define whether a window for inducing lasting tolerance exists, we generated NOD mice with Tetracycline-regulated PI expression (TIP mice) in the antigen presenting cells (APC) allowing induction of tolerance to PI in a temporally controlled manner. Enumeration of antigen specific T cells is a robust method to assess tolerance. Because PI tetramer+ T cells are much more difficult to detect in NOD mice than IGRP tetramer+ T cells, we also made NOD mice with tetracycline-regulated IGRP expression in the APC (TII mice) to track autoreactive T cells.
PI expression in TIP mice was induced from gestation until 5 weeks of age. Significantly, these mice were completely protected from insulitis, diabetes and insulin autoantibody development comparably with mice with PI switched on throughout life.
IGRP expression in TII mice was induced from gestation and this led to deletion of IGRP tetramer+ cells but no impact on insulitis or diabetes was observed as previously described. When IGRP expression was switched off at 5w no recovery of IGRP tetramer+ cells was observed suggesting IGRP-specific T cells emerge from the thymus only in early life.
Our data suggest that anti-islet autoimmunity is determined in early life and that it does not occur de novo in later life. PI-based therapies to prevent diabetes may have lasting benefit when used as a vaccination in early life.