Our previous work demonstrated peptide-pulsed mature myeloid dendritic cells (DC) presenting beta cell antigens induce tolerance. Here we determine whether immature DC (iDC) presenting dominant (insulin beta9-23 chain, proinsulin C19-A3) or ignored (glutamic acid decarboxylase 65(78-97)) antigen determinants promote tolerance. Nonobese diabetic (NOD) mice were given injections of either unpulsed or peptide-pulsed myeloid iDC beginning at 9 weeks of age for 3 consecutive weeks. Diabetes incidence in recipients of unpulsed iDC was comparable to unmanipulated animals ( approximately 80%), whereas GAD65(78-97) pulsed iDC recipients were protected from the disease (P = 0.05). We also analyzed splenic T cell proliferation responses to the panel of studied peptides in diabetic and nondiabetic recipients. When stimulated with insulin or proinsulin peptide, nondiabetic mice receiving the peptide-pulsed iDC had a 21- to 31-fold or 3.9- to 9.0-fold reduction in T cell response, respectively, as compared to the response of diabetic unpulsed recipients. However, only a 2.6- to 3.1-fold reduction in response to beta chain peptide, and a 1.5- to 3.4-fold reduction in proinsulin response were observed in diabetic mice receiving peptide-pulsed iDC. The reduction was not specific to the immunizing peptide, as reduced proliferation was observed to other diabetes-target peptides. We conclude that protective iDC-based therapies require target antigen presentation, and ignored determinants may be preferable perhaps due to an available naïve T cell repertoire. In addition, iDC presenting peptides induce a nonspecific reduction in T cell responses to beta cell antigens, possibly through the induction of regulatory T cells.

Increasing attention is drawn to the contributions of abnormalities in both innate and acquired immune responses to the pathogenesis of autoimmune diseases, such as type 1 diabetes (T1D). Dendritic cells (DC) are critical immune cells linking innate and acquired immune responses and previous studies in NOD mice suggest abnormalities in these cells. To address DC dysregulation we examined kinetic global gene expression in NOD and B6 GM-CSF/IL-4-induced bone marrow-derived DC following lipopolysaccharide (LPS)-stimulation. We identified expression differences in over 300 genes including a cluster of 16 interferon (IFN-alpha/beta) target genes overexpressed in NOD DC. Mechanistically, heightened IFN-alpha/beta responses were not due to increased production of this cytokine, IFN-gamma priming or increased Syk kinase activity. We found, however, heightened responses to IFN-alpha/beta in NOD versus B6 as demonstrated by increased type 1 IFN target gene expression, for example, IRF-7, in NOD DC and macrophages. Analysis of multiple congenic strains demonstrated that the Idd5 susceptibility region largely governed heightened IFN-alpha responses. Of interest, heightened IFN-alpha/beta response in NOD mice was not confined to hematopoietic cells but was also seen in the pancreas and beta cells. Compounding the IFN-alpha response defect, NOD mice harbor significantly more PDC in spleen in comparison to B6 and produce four- to sixfold more IFN-alpha when stimulated with CpG. Finally, treatment of NOD mice with IFN-alpha inducing agents, for example, high-dose poly I:C accelerates diabetes in both female and male mice. The abnormalities in the IFN-alpha/beta axis appear to play a significant role in T1D pathogenesis.