Aims/hypothesis. Fulminant Type 1 diabetes is a novel subtype of Type 1 diabetes that involves the abrupt onset of insulin-deficient hyperglycaemia. This subtype appears to be non-autoimmune because of the absence of diabetes-related autoantibodies in the serum, and of insulitis in pancreatic biopsy specimens. The pathogenesis of the disease is still unknown. In this study, we investigated whether T cell autoimmune responses are involved in fulminant Type 1 diabetes. Methods. Cellular immune responses to beta cell autoantigens were studied by enzyme-linked immunospot (ELISPOT) assay in 13 fulminant Type 1 diabetic patients and 49 autoantibody-positive autoimmune Type 1 diabetic patients. Results were compared with those of 18 Type 2 diabetic patients, six secondary diabetic patients (diabetes due to chronic pancreatitis) and 35 healthy controls. Results. Nine of 13 (69.2%) GAD-reactive Th1 cells, and three of 12 (25%) insulin-B9-23-reactive Th1 cells were identified in fulminant Type 1 diabetic patients by ELISPOT, as in autoantibody-positive Type 1 diabetic patients. Four fulminant Type 1 diabetic patients possessed the highly diabetes-resistant allele DR2, three of whom had GAD-reactive Th1 cells in the periphery. Conclusions/interpretation. Peripheral immune reaction was observed in 69.2% of fulminant Type 1 diabetic patients, indicating that autoreactive T cells might contribute, at least in part, to the development of fulminant Type 1 diabetes.

These results suggest a substantial contribution of Fas/ FasL interactions to CD4؉ , but not CD8 ؉ , T-cell-mediated destruction of pancreatic ␤-cells. When anti-FasL antibody was administered to NOD mice between 5 and 15 weeks of age, the onset of diabetes was slightly delayed but the incidence was not decreased. However, administration of anti-FasL antibody at 2-4 weeks of age completely prevented insulitis and diabetes. These results suggest that Fas/FasL interactions contribute to CD4 ؉ T-cell-mediated ␤-cell destruction and play an essential role in the initiation of autoimmune NOD diabetes.

The "gold standard" for evaluation of immunoassays is blinded testing, using coded samples and antigens. Although assays for autoreactive T cells are no exception to this rule, it is nonetheless rarely applied in this context. To facilitate such investigations, we coded a panel of 10 peptides representing T cell epitopes reported to be of relevance to islet autoimmunity. These were deployed in a novel cytokine ELISPOT assay, in which the use of immunoglobulin-free medium reduces background reactivity and thus potentially enhances the specificity and sensitivity of detection of autoreactive T cells. Significant IFN-gamma production against GAD65 (554-575), insulin (B9-23), and IA-2 (709-736) peptides were observed in type 1 diabetic patients, whereas no significant changes from background were detected in healthy controls. These results confirm the utility of the blinded performance of T cell assays as the most robust platform for assessing technologies to T cell autoreactivity.

OBJECTIVE -To investigate the prevalence of ␤-cell autoantigen-reactive peripheral T-cells in type 1 diabetes, we developed an immunoglobulin-free enzyme-linked immunospot (ELIS-POT) assay and assessed its usefulness for diagnosing this disease. RESEARCH DESIGN AND METHODS-Cellular immune responses to ␤-cell autoantigens were studied both by immunoglobulin-free proliferation assays and ELISPOT assays in 33 patients with type 1 diabetes and 15 patients with type 2 diabetes, compared with 23 healthy control subjects. Autoantibodies against GAD65 and IA-2 were measured by radioimmunoassay.RESULTS -Significant proliferative responses to GAD65 were observed in 10 of 31 (32.3%) type 1 diabetic patients (P Ͻ 0.05), whereas GAD65-reactive ␥-interferon (IFN-␥)-secreting cells were detected in 22 of 33 patients (66.7%) by ELISPOT assay (P Ͻ 0.001). Of patients negative for both GAD65 and IA-2, five of six (83.3%) showed IFN-␥ positivity in ELISPOT and two of five (40.0%) showed significant proliferation against GAD65.CONCLUSIONS -Using a newly developed ELISPOT assay, GAD-reactive T-helper 1 cells in PBMC of type 1 diabetic patients could be identified at a higher frequency than by the proliferation assay. Therefore, the immunoglobulin-free ELISPOT assay is an excellent tool for detecting T-cell reactivity to autoantigens with greater specificity and, in combination with ␤-cell autoantibody determination, will improve the diagnosis of type 1 diabetes. Diabetes Care 25:1390 -1397, 2002S everal studies of type 1 diabetes have suggested that the destruction of pancreatic ␤-cells is caused by inflammatory cellular immune responses mediated by T-helper 1 (Th1) cells, which secrete ␥-interferon (IFN-␥) (1-3). In diabetic animal models, a correlation between disease onset and Th1 cytokine production has been observed, and IFN-␥ in particular plays an important role in ␤-cell destruction (4 -6). On the other hand, autoantibodies against GAD65, insulin, IA-2/ICA512, and other islet cell antigens have been used not only to diagnose but also to predict the onset of type 1 diabetes (7-9). However, differences in age at onset of diabetes as well as racial differences have been reported to influence patients' autoantibody status (10,11). In general, circulating levels of autoantibodies are considerably lower in patients with a long disease duration than in patients at the time of onset or in the prediabetic period (12). According to a Japanese multicenter study, positivity for GAD antibody was found in only 35.4% of 921 Japanese type 1 diabetic patients studied, but among those with newly diagnosed diabetes, 50.3% were GAD antibody positive (13). Japanese type 1 diabetic subjects, therefore, seemed to present with lower antibody levels than Caucasians. In addition, it has been suggested that the presence of GAD antibody can identify a subset of latent autoimmune diabetes in adults (LADA) and predict incipient insulin dependency (14,15). Indeed, the patient's autoantibody status is essential for the diagnosis and prediction of type 1 di...

In this study, no association is seen between the Pro12Ala polymorphism in PPAR gamma2 gene and diabetes; a weak association with obesity is seen.