Distinct genetic control of autoimmune neuropathy and diabetes in the non-obese diabetic background

Abstract

The non-obese diabetic (NOD) mouse is susceptible to the development of autoimmune diabetes but also multiple other autoimmune diseases. Over twenty susceptibility loci linked to diabetes have been identified in NOD mice and progress has been made in the definition of candidate genes at many of these loci (termed Idd for insulin-dependent diabetes). The susceptibility to multiple autoimmune diseases in the NOD background is a unique opportunity to examine susceptibility genes that confer a general propensity for autoimmunity versus susceptibility genes that control individual autoimmune diseases. We previously showed that NOD mice deficient for the costimulatory molecule B7-2 (NOD-B7-2KO mice) were protected from diabetes but spontaneously developed an autoimmune peripheral neuropathy. Here, we took advantage of multiple NOD mouse strains congenic for Idd loci to test the role of these Idd loci the development of neuropathy and determine if B6 alleles at Idd loci that are protective for diabetes will also be for neuropathy. Thus, we generated NOD-B7-2KO strains congenic at Idd loci and examined the development of neuritis and clinical neuropathy. We found that the NOD-H-2(g7) MHC region is necessary for development of neuropathy in NOD-B7-2KO mice. In contrast, other Idd loci that significantly protect from diabetes did not affect neuropathy when considered individually. However, we found potent genetic interactions of some Idd loci that provided almost complete protection from neuritis and clinical neuropathy. In addition, defective immunoregulation by Tregs could supersede protection by some, but not other, Idd loci in a tissue-specific manner in a model where neuropathy and diabetes occurred concomitantly. Thus, our study helps identify Idd loci that control tissue-specific disease or confer general susceptibility to autoimmunity, and brings insight to the Treg-dependence of autoimmune processes influenced by given Idd region in the NOD background.

Keywords: Autoimmunity; Diabetes; Idd loci; NOD-B7-2KO mice; Peripheral neuropathy; Tregs.

Fig. 1

The NOD H-2^g7 MHC locus is necessary for autoimmune peripheral neuropathy in NOD-B7-2KO mice. NOD-B7-2KO mice (H-2^g7/g7) were crossed to NOD-H-2^h4/h4 or NOD-H-2^b/b congenic mice to generate NOD-B7-2KO-H-2^h4/h4 and NOD-B7-2KO-H-2^b/b mice. A) Congenic NOD-B7-2KO mice that were homozygous or heterozygous for H-2^g7, H-2^h4 or H-2^b were followed for the development of neuropathy. Cumulative incidence of neuropathy is shown for females of indicated genotypes. Incidence curves significantly different from incidence in NOD-B7-2KO mice are indicated (Log-rank (Mantel–Cox) test). B) H&E staining of sciatic nerves isolated from NOD-B7-2KO and NOD-B7-2KO-H2^h4/h4 mice.

Fig. 2

Diabetes-protective alleles at individual Idd loci do not alter the course of autoimmune peripheral neuropathy in NOD-B7-2KO mice. NOD-B7-2KO mice were crossed to NOD mice congenic for individual Idd loci to create congenic NOD-B7-2KO strains that carry introgressed alleles from diabetes-resistant B6, B10 or NOR mice at indicated Idd loci. Introgressed alleles at these loci were individually shown to reduce diabetes incidence by ~40–100% in NOD congenic strains [–12]. Cumulative incidence of neuropathy is shown for NOD-B7-2KO females congenic for indicated Idd loci (triangles) and relevant cohorts of NOD-B7-2KO controls (circles).

Fig. 3

Differential effect of Idd3/5 and Idd3/10/18 combinations on clinical disease and tissue infiltration in NOD-B7-2KO mice. NOD-B7-2KO mice congenic for Idd3 were crossed to NOD-B7-2KO mice congenic for Idd5 or Idd10/18 to generate NOD-B7-2KO mice congenic for both Idd3 and Idd5 (NOD-B7-2KO-Idd3/5) or Idd3 and Idd10/18 (NOD-B7-2KO-Idd3/10/18). A) Congenic females and NOD-B7-2KO control females were followed for neuropathy as in Fig. 1. B) Sciatic nerves were isolated from age-matched mice in indicated strains, stained with H&E and scored for infiltration as indicated in Materials and methods. Statistical comparison of infiltration severity was performed using the Mann–Whitney test (NS: non-significant; *: p = 0.017). C) Spleen and LN cells of NOD-B7-2KO-Idd3/5 congenic mice and NOD-B7-2KOcontrols were stimulated with anti-CD3+/− anti-CD28 mAbs in primary or secondary stimulations in vitro and IFN-γ production was measured by ELISA of culture supernatants.

Fig. 4

Treatment with anti-B7-1 mAbs restores neuropathy but not diabetes in NOD-B7-2KO-Idd3/5 and NOD-B7-2KO-Idd3/10/18 congenic mice. NOD-B7-2KO-Idd3/5 and NOD-B7-2KO-Idd3/10/18 congenic mice as well as NOD-B7-2KO controls were treated with anti-B7-1 mAbs between 2–4 weeks of age. A–B) Cumulative incidence of neuropathy (A) and diabetes (B) in female NOD-B7-2KO-Idd3/5 (n = 5), NOD-B7-2KO-Idd3/10/18 (n = 5), and NOD-B7-2KO (n = 12) mice is shown. C) Cumulative incidence of neuropathy and diabetes in anti-B7-1 mAbs-treated NOD-B7-2KO-Idd3/5 males and females.

Fig. 5

Treatment with anti-B7-1 mAbs does not alter neuropathy and diabetes but exacerbates autoimmune thyroiditis in NOD-B7-2KO-H-2^h4 congenic mice. NOD-B7-2KO-H-2^h4 and NOD-B7-2KO controls were treated with anti-B7-1 mAbs between 2–4 weeks of age. A) Cumulative incidence of neuropathy and diabetes in NOD-B7-2KO-H-2^h4 and NOD-B7-2KO mice. Similar results were observed in males and females and were pooled. B) Sciatic nerves and thyroid glands were isolated from indicated mice, stained with H&E and scored for infiltration as indicated in Materials and methods. Statistical comparison of infiltration severity was performed using the Mann–Whitney test (*: p < 0.05; **: p < 0.005, ****: p < 0.0001). C–D) Representative sections of sciatic nerves (C) and thyroid glands (D) are shown for indicated mice.