Gaps in our knowledge: Recently, numerous genetic loci with the potential to initiate or perpetuate auto- immune disease have been detected, many of them found by genome-wide association studies (GWAS). However, GWAS results are sometimes difficult to interpret, and the utility of anonymous SNP association has been questioned. Genes underlying quantitative trait loci (QTL) and GWAS need verification and testing to identify the tissue and the critical functional pathways in which they act. It would be advantageous to have an excellent animal counterpart to test hypotheses about how non-MHC genes control autoimmunity unimpaired by the ethical and practical considerations of human research. Effective rodent models would be those that: (1) display the need for a susceptible allele at a GWAS homologue in the same disease, (2) have variation in expression or coding sequence of the candidate gene, so that resistant and susceptible alleles can be compared, and (3) have measurable gene-by-environment and gene-by-gene interaction that is likely necessary to promote autoimmunity in both the human and rodent populations. Our proposed model fulfils those criteria. Novelty and Risk: Type 1 diabetes (T1D) is a T cell mediated autoimmune disease, and >18 significant non- MHC diabetes-promoting associations have been discovered by GWAS. In our novel rat model of T1D with excellent clinical fidelity and an absolute dependence on environmental perturb ants, we induce T1D in response to exposure to viruses and TLR agonists. We have mapped two highly significant rat QTL to regions containing two orthologues of human GWAS gene candidates: diubiquitin (UBD or FAT10) and UBASH3A. This model meets all three criteria above for potential to study how these genes act in diabetes induction. The risk is that neither species requires a disease-linked allele of UBD or UBASH3A in diabetogenesis, and that both human and rat mapping projects have given two incorrect candidate genes. Potential breakthroughs/relevance to mission: This work will explore the manner in which UBD and/or UBASH3A creates an environment conducive to the onset of diabetes in response to viral infection in susceptible rats, and could provide a faithful model of the way the human orthologues do the same thing. Both these candidates function in the ubiquitin pathway, already implicated in T lymphocyte activation, and the UBD transcripts are also found in the pancreas. One important outcome would be to show that enhanced post-viral ubiquity-mediated degradation of key T cell or beta-cell regulatory pathway members can lead to T1D.
Aim 1. Test the hypothesis that UBD and UBASH3A are T1D candidate genes in recombinant congenic rats.
Aim 2 : Determine if heightened UBD and/or UBASH3A gene expression after virus infection underlie the diabetes phenotype. Determine if human UBD promoters, like the rat UBD promoter, show allelic differences in promoter activity in an in vitro reporter assay.
Aim 3. Determine the tissue in which the two rat diabetes loci act to initiate T1D, with mix-and-match neonatal adoptive transfer of T cells and viral induction of T1D.
Autoimmunity is a serious public health issue that affects nearly 20 million US residents. Type 1 autoimmune diabetes (T1D) is increasing in prevalence and decreasing in time-to-onset;new cases appear at a rate of 1/600 in US youth. Numerous genes predispose to T1D in the context of environmental factors. Our proposal is designed to develop a rat model of virus-induced T1D in which we have mapped two genes required for autoimmune destruction of the pancreatic beta cells. Each has a human diabetes-modifying orthologue. This novel rat model will allow a fresh approach to explore and study these genes and probe potential therapies 9 based on their shared ubiquitin and pathway and T cell localization.
