Immune pathogenesis of diabetes in the nonobese diabetic mouse: an overview

Absence of a reductase, NCB5OR, causes insulin-deficient diabetes

NCB5OR is a highly conserved NAD(P)H reductase that contains a cytochrome b5-like domain at the N terminus and a cytochrome b5 reductase-like domain at the C terminus. The enzyme is located in the endoplasmic reticulum (ER) and is widely expressed in organs and tissues. Targeted inactivation of this gene in mice has no impact on embryonic or fetal viability. At 4 weeks of age, Ncb5or-/- mice have normal blood glucose levels but impaired glucose tolerance. Isolated Ncb5or-/- islets have markedly impaired glucose- or arginine-stimulated insulin secretion. By 7 weeks of age, these mice develop severe hyperglycemia with markedly decreased serum insulin levels and nearly normal insulin tolerance. As the animals age, there is a progressive loss of beta cells in pancreatic islets, but there is no loss of alpha, delta, or PP cells. Electron microscopy reveals degranulation of beta cells and hypertrophic and hyperplastic mitochondria, some of which contain electron dense inclusions. Four-week-old Ncb5or-/- mice have enhanced sensitivity to the diabetogenic agent streptozotocin. NCB5OR appears to play a critical role in protecting pancreatic beta cells against oxidant stress.

Radicals and oxidative stress in diabetes

Recent evidence is reviewed indicating increased oxidative damage in Type 1 and Type 2 diabetes mellitus as well as deficits in antioxidant defence enzymes and vitamins. Mechanisms are considered whereby hyperglycaemia can increase oxidative stress, and change the redox potential of glutathione and whereby reactive oxygen species can cause hyperglycaemia. It is argued that oxygen, antioxidant defences, and cellular redox status should now be regarded as central players in diabetes and the metabolic syndrome.

Interferon-gamma is essential for destruction of beta cells and development of insulin-dependent diabetes mellitus

Autoimmune mediated destruction of beta cells of the islets of Langerhans leads to insulin-dependent diabetes mellitus (IDDM). Rat insulin promoter (RIP) lymphocytic choriomeningitis virus (LCMV) transgenic mice that express the nucleoprotein (NP) or glycoprotein (GP) of LCMV under control of the RIP in their beta cells develop IDDM after infection with LCMV and serve as a model for virus-induced IDDM. Recently, Kagi et al. (Kagi, D., B. Odermatt, P. Ohashi, R.M. Zinkernagel, and H. Hengartner, 1996, J. Exp. Med. 183:2143-2149) showed, using RIP LCMV perforin-deficient mice, that IDDM does not occur in the absence of perforin. They concluded that perforin-mediated killing by cytotoxic T lymphocytes (CTLs) is the main factor needed for beta cell injury and destruction. Here we provide evidence that killing of beta cells is more complex and multifactorial. By the use of our RIP LCMV model, we show that in perforin competent but interferon-gamma (IFN-gamma)-deficient mice, beta cell injury is limited and IDDM does not occur. For these studies, double transgenic mice were generated that were genetically deficient in the production of IFN-gamma and express LCMV NP or GP in their beta cells. In such mice, IDDM was aborted despite the generation of LCMV-specific antiself CTLs that displayed normal cytolytic activity in vitro and in vivo and entered the pancreas. However, mononuclear infiltration into the islets did not occur, and upregulation of class I and II molecules usually found in islets of RIP LCMV single transgenic mice after LCMV infection preceding the onset of clinical IDDM was not present in these bigenic mice. Our findings indicate that in addition to perforin, beta cell destruction, development of insulitis, and IDDM also depend on the cytokine INF-gamma, presumably through enhancement of major histocompatibility complex expression and antigen presentation.

Transgenic copper/zinc superoxide dismutase modulates susceptibility to type I diabetes

A growing body of evidence suggests that active oxygen is an important participant in the destruction of the pancreatic beta cell, which, in turn, leads to type I or insulin-dependent diabetes mellitus. Consequently, genetic factors predisposing susceptibility to insulin-dependent diabetes mellitus may include those that determine active oxygen metabolism. A direct test of this hypothesis is provided by a transgenic model for increased activity of Cu/Zn superoxide dismutase (EC 1.15.1.1), a principal radical scavenging enzyme. Here we demonstrate that elevated levels of this enzyme provided by a Cu/Zn superoxide dismutase transgene enhance the tolerance of pancreatic beta cells to oxidative stress-induced diabetogenesis. These results show that this transgenic approach holds promise for revealing the role of reactive oxygen in autoimmune models of diabetogenesis as well as in other models of disease pathology in which active oxygen has been implicated.

Transfer of type 1 (insulin-dependent) diabetes mellitus associated autoimmunity to mice with severe combined immunodeficiency (SCID)

Pancreatic beta-cell destruction and development of Type 1 (insulin-dependent) diabetes mellitus are associated with circulating islet cell antibodies. Mice with severe combined immunodeficiency (SCID mice) were reconstituted with peripheral blood mononuclear cells from Type 1 diabetic patients, one who was antibody positive and one antibody negative, and from healthy individuals. Reconstituted mice were subsequently immunized with rat islets in incomplete Freunds adjuvant or adjuvant alone. Seventeen mice received peripheral blood mononuclear cells obtained at three different time points from the islet cell antibody positive patient. Before immunization with rat islets two mice developed antibodies to glutamic acid decarboxylase, a major target for antibodies in Type 1 diabetes, whereas none were positive for cytoplasmic islet cell antibodies. Following immunization with rat islets, glutamic acid decarboxylase antibodies were detected by immunoprecipitation in three additional mice, two of which also became positive for cytoplasmic islet cell antibodies. Of 22 mice which received peripheral blood mononuclear cells from either the islet cell antibody negative patient (n = 5) or from two healthy individuals (n = 17), none were positive for islet cell autoantibodies before or after immunization. None of the islet cell antibody positive mice became hyperglycaemic, showed impaired glucose tolerance or islet cell damage when studied 40 days after immunization (i.e. 100 days after reconstitution). In conclusion these results show that human B lymphocytes producing diabetes-associated autoantibodies can be transferred to SCID mice and remain antigen sensitive, but also that autoantibodies alone are not sufficient to induce beta-cell destruction.

Genetic analysis of diabetes and insulitis in an interspecific cross of the nonobese diabetic mouse with Mus spretus

The nonobese diabetic (NOD) mouse is a widely used model for genetic studies of insulin-dependent diabetes mellitus due to the similarities between the murine and human diseases. To aid in the localization and identification of diabetes-related susceptibility genes, we have constructed an interspecific backcross between NOD and Mus spretus (SEG/Pas) mice. Although no diabetic animals were observed in the first backcross generation of (SEG/Pas x NOD) x NOD (BC1), the incidence of insulitis (lymphocyte infiltration of the islets of Langerhans) exceeded 20% after injections of cyclophosphamide, a treatment that provokes an acute form of diabetes in NOD mice. Insulitis, a prediabetic condition, is a useful phenotype in studies of diabetes susceptibility. In the second backcross (BC2) generation, 8% of the animals became diabetic and 76% were found to have insulitis. Genetic mapping studies in the BC2 families confirmed the importance of the major histocompatibility complex region on the severity of insulitis and suggested that additional susceptibility loci were linked to markers on mouse chromosomes 3, 6, and 15. Mus spretus crosses have been an important tool in recent advances in murine genetics, and our results extend their usefulness to the study of a multifactorial disease.