Utility of murine models for the study of spontaneous autoimmune type 1 diabetes

The selection process for animals that develop spontaneous autoimmune diabetes has generated models that share many characteristics with the human form of the disease. The pubertal or postpubertal onset in all the four models mimics the clinical onset in humans. Further shared with humans is the paramount role of the MHC Class II alleles, yet the MHC alone is not sufficient, and there are significant contributions of non-MHC genes. While four models do exist, a critical mass of information exists only for two of these: the NOD mouse and the BB rat. These latter two share immunological features with the human disease such as the defective central and peripheral tolerance, as well as impaired dendritic cell maturation and function. To date, the wealth of the research has been performed using the NOD mouse model. Shared features of human and NOD mice are the important roles of environmental and dietary factors. Early gluten exposure has been associated with T1DM onset in human and is essential for T1DM to initiate in the NOD mouse. NOD mice also have immune responses to self-proteins such as GAD65, insulin, IA-2 (ICA512 or protein-tyrosine-phosphatase, receptor type, n), and ICA69 (islet cell autoantigen 1). The NOD mouse has also been utilized to find new autoantigens. In the islets of NOD mice CD8+ T cell respond to peptides generated from the IGRP [glucose-6-phosphatase, catalytic subunit, 2 (G6pc2)] protein. While responses against IGRP have not yet been measured from human T cells, recently, the author's laboratory, in collaboration with Teresa DiLorenzo (Albert Einstein School of Medicine) and David Serreze (The Jackson Laboratory), have used T cells from NOD mice transgenically expressing a human Class I HLA (HLA-A0201) to lyze human islets in a MHC-restricted fashion. This suggests that the T cells are potentially responding to the same peptides. Differences also exist between each of the models and the human disease, but as each of these models represents only one individual with diabetes or a family with a given genetic susceptibility, the similarities and what we can learn from the sum total of all the four models should allow us to better understand the human condition (Table 1). © Blackwell Munksgaard, 2005.