Virus infection is hypothesized to be an important environmental “trigger” of type 1 diabetes in humans. We used the BBDR rat model to investigate the relationship between viral infection and autoimmune diabetes. BBDR rats are diabetes-free in viral Ab-free housing, but the disease develops in ∼30% of BBDR rats infected with Kilham rat virus (KRV) through a process that does not involve infection of pancreatic β cells. Pretreatment with polyinosinic-polycytidylic (poly(I:C)), a ligand of TLR3, acts synergistically to induce diabetes in 100% of KRV-infected rats. The mechanisms by which KRV induces diabetes and TLR3 ligation facilitates this process are not clear. In this study, we demonstrate that activation of the innate immune system plays a crucial role in diabetes induction. We report that multiple TLR agonists synergize with KRV infection to induce diabetes in BBDR rats, as do heat-killed Escherichia coli or Staphylococcus aureus (natural TLR agonists). KRV infection increases serum IL-12 p40 in a strain-specific manner, and increases IL-12 p40, IFN-γ-inducible protein-10, and IFN-γ mRNA transcript levels, particularly in the pancreatic lymph nodes of BBDR rats. Infection with vaccinia virus or H-1 parvovirus induced less stimulation of the innate immune system and failed to induce diabetes in BBDR rats. Our results suggest that the degree to which the innate immune system is activated by TLRs is important for expression of virus-induced diabetes in genetically susceptible hosts.

Viruses are believed to contribute to the pathogenesis of autoimmune type 1A diabetes in humans. This pathogenic process can be modeled in the BBDR rat, which develops pancreatic insulitis and type 1A-like diabetes after infection with Kilham’s rat virus (RV). The mechanism is unknown, but does not involve infection of the pancreatic islets. We first documented that RV infection of BBDR rats induces diabetes, whereas infection with its close homologue H-1 does not. Both viruses induced similar humoral and cellular immune responses in the host, but only RV also caused a decrease in splenic CD4+CD25+ T cells in both BBDR rats and normal WF rats. Surprisingly, RV infection increased CD4+CD25+ T cells in pancreatic lymph nodes of BBDR but not WF rats. This increase appeared to be due to the accumulation of nonproliferating CD4+CD25+ T cells. The results imply that the reduction in splenic CD4+CD25+ cells observed in RV-infected animals is virus specific, whereas the increase in pancreatic lymph node CD4+CD25+ cells is both virus and rat strain specific. The data suggest that RV but not H-1 infection alters T cell regulation in BBDR rats and permits the expression of autoimmune diabetes. More generally, the results suggest a mechanism that could link an underlying genetic predisposition to environmental perturbation and transform a “regulated predisposition” into autoimmune diabetes, namely, failure to maintain regulatory CD4+CD25+ T cell function.

Viral infections are associated epidemiologically with the expression of type 1 diabetes in humans, but the mechanisms underlying this putative association are unknown. To investigate the role of viruses in diabetes, we used a model of viral induction of autoimmune diabetes in genetically susceptible biobreeding diabetes-resistant (BBDR) rats. BBDR rats do not develop diabetes in viral-Ab-free environments, but ∼25% of animals infected with the parvovirus Kilham rat virus (KRV) develop autoimmune diabetes via a mechanism that does not involve β cell infection. Using this model, we recently documented that TLR agonists synergize with KRV infection and increase disease penetrance. We now report that KRV itself activates innate immunity through TLR ligation. We show that KRV infection strongly stimulates BBDR splenocytes to produce the proinflammatory cytokines IL-6 and IL-12p40 but not TNF-α. KRV infection induces high levels of IL-12p40 by splenic B cells and Flt-3-ligand-induced bone marrow-derived dendritic cells (DCs) but only low levels of IL-12p40 production by thioglycolate-elicited peritoneal macrophages or GM-CSF plus IL-4-induced bone marrow-derived DCs. KRV-induced cytokine production is blocked by pharmacological inhibitors of protein kinase R and NF-κB. Genomic KRV DNA also induces BBDR splenocytes and Flt-3L-induced DCs from wild-type but not TLR9-deficient mice to produce IL-12p40; KRV-induced up-regulation of B lymphocytes can be blocked by TLR9 antagonists including inhibitory CpG and chloroquine. Administration of chloroquine to virus-infected BBDR rats decreases the incidence of diabetes and decreases blood levels of IL-12p40. Our data implicate the TLR9-signaling pathway in KRV-induced innate immune activation and autoimmune diabetes in the BBDR rat.