Active suppression of diabetes after oral administration of insulin is determined by antigen dosage.

We have previously demonstrated that feeding six-week-old female mice with 20 units of human insulin every 2 - 3 days for 15 or 30 days induced an active mechanism of suppression through the generation of regulatory T cells that reduced the number of successful diabetic transfers in irradiated NOD recipients. In the present study, we analyzed the effects of antigen dosage and the critical period of cell injection to obtain protection. The effects of the dose of insulin feeding were therefore compared during cotransfer experiments of 5 x 10(6) T cells from diabetic mice and 5 x 10(6) T cells from the spleen of mice receiving 10 units, 20 units, or 40 units of insulin or saline every 2 - 3 days for 15 days. Only T lymphocytes from mice fed with 20 units conferred active cellular protection during adoptive transfer with a significant delay in diabetes onset (p = 0.002). No significant difference was noticed during histological analysis of pancreatic glands, indicating tha insulitis was not prevented. However, mice receiving T lymphocytes from the 20 units of insulin-fed animals had a milder form of inflammation, with a significantly lower percentage of severely infiltrated islets. Injecting regulatory T cells 7 days and 14 days after iv injection of diabetogenic T cells did not modify the incidence curves of diabetes in the recipients, suggesting that cellular interactions and delay in cell trafficking were determinants. These results may have important clinical implications in humans. In conclusion, this study indicates the importance but also the limits of antigen therapy in type I diabetes. Antigen dosage is a critical element for active suppression. Such analysis is important to perform in humans before the initiation of a large-scale prevention trial in prediabetic individuals.