NOVEL CONSIDERATIONS ON THE ANTIBODY AUTOANTIGEN SYSTEM IN TYPE-I (INSULIN-DEPENDENT) DIABETES-MELLITUS

Insulin dependent diabetes (IDDM) has an autoimmune pathogenesis. Included is the presence of antibodies to pancreatic islet cells. The first identified were islet cell antibodies (ICA), detected by indirect immunofluorescence, and which react with all cells within islets. Importantly, the autoantibodies are found several years prior to disease and although a pathogenic role for the autoantibodies is unclear, they have become useful markers of prediabetes. A number of studies of twins discordant for IDDM and of first degree relatives of IDDM patients have established that there is an increased risk for disease in individuals who have ICA, especially when ICA levels are high. This high predictive value of ICA decreases in the general population where the incidence of IDDM is lower than in first degree relatives, and both ICA and the disease risk associated with ICA, appear to be influenced by a genetic susceptibility. This has been substantiated in a study of patients with endocrine autoimmunity and ICA (Polyendocrine Study) where the predictive value of very high levels of ICA is less than 50 % in patients without a first degree relative with IDDM. Hence, there remain a substantial number of patients with ICA who do not develop disease. From these patients, it was demonstrated that ICA include at least two distinct specificities, one of which is beta cell specific and is not associated with a high risk for IDDM. These data, and an increasing list of new autoantibodies in IDDM, have increased the search for the relevant autoantigens. That of classic ICA is suggested to be a sialoganglioside, and a number of protein autoantigens have been suggested as other potential autoantigens. Of particular importance has been the identification of glutamic acid decarboxylase (GAD), the biosynthesizing enzyme of the neurotransmitter gamma amino butyric acid, as the autoantigen corresponding to the 64K autoantibodies found in the majority of newly diagnosed IDDM patients. Despite the identification of GAD, specific assays developed so far have not detected anti GAD antibodies at high frequencies in newly diagnosed patients. Like ICA, it appears that there may be distinct specificities of autoantibodies to GAD and associated products. Enzyme digestion of the autoantigen yields 50K, 40K and 37K fragments which appear to be derived from distinct polypeptide chains. Patients may have autoantibodies to either the 50K fragment or the 40/37K fragment or all fragments. Importantly, studies in twins and in the polyendocrine patients indicate that autoantibodies to the 37K fragment are more strongly associated with the progression to disease. In addition to these autoantigens, autoantibodies are detected against the hormone insulin and its precursor proinsulin in up to 40% of newly diagnosed patients. Heat shock protein 65, carboxypeptidase H and glucose transporter have also been reported as autoantigens relevant to IDDM. It is now clear that the autoimmune response associated with IDDM is heterogeneous and against several islet antigens. The search will continue to determine which of these is important in the pathogenesis and which will provide effective markers for prediction which will eventually allow safe intervention therapy for the prevention of IDDM.