WS3_1 The Synergistic Effects of HLA-B*52 and Risk Allele at a SNP in IL12B Region on Complication of Aortic Regurgitation in Patients with Takayasu Arteritis

Objectives: Aortic regurgitation (AR) is one of severe complications of Takayasu arteritis (TAK). Several studies have reported that AR is more common in TAK patients with HLA-B*52 than in those without B*52. We previously found a single nucleotide polymorphism (SNP), rs6871626, in IL12B region as a susceptibility gene to TAK and reported that the prevalence of AR was higher in the TAK patients with risk allele at the IL12B SNP than in those without risk allele and that B*52 and IL-12B SNP showed a synergistic effect on susceptibility of TAK itself. We investigated whether the two factors have a synergistic effect in complication of AR in TAK.

Methods: We examined the clinical records of 84 cases with TAK and investigated the association of the genotypes and prevalence of AR.

Results: The prevalence of AR in B*52-positive patients was 55% (24/44), which tended to be higher than 33% (13/40) in B*52-negative patients but the difference was not statistically significant (p=0.0501). The prevalence of AR in patients with risk allele at the SNP was 52% (33/68), significantly higher than 13% (2/16) in patients without risk allele (p=0.00050). The odds ratios of complicating AR were 1.00 in the patients without B*52 nor risk allele, 1.33 with B*52 only, 5.05 with risk allele only and 13.14 with both B*52 and risk allele.

Conclusions: The results suggested that HLA-B*52 and IL12B SNP have a synergistic effect on the complication of AR in TAK patients.

WS3_2 In Anti-GBM Disease, HLA-Peptide Conformation Determines Susceptibility or Protection from Disease

Objectives: Anti-glomerular basement membrane (GBM) disease is a small vessel vasculitis caused by anti-α3(IV)NC1 T and B cell autoreactivity. An immunodominant T cell epitope, α3_135-145, has been defined. HLA-DR15 confers an increased disease risk (OR 8.5) while HLA-DR1 is dominantly protective from this risk. These studies aimed to define the mechanisms by which different HLA-DR alleles lead to susceptibility or protection from disease.

Methods: α3_135-145-specific CD4⁺ T cells were enumerated in anti-GBM patients, HLA-typed healthy humans and HLA-DR transgenic mice using HLA-DR-α3_135-145 tetramers. The crystal structures of HLA-DR15-α3_135-145 and HLA-DR1-α3_135-145 were solved. α3_135-145-specific T cell receptor (TCR) usage was determined by single cell sequencing. T cell phenotype was determined by flow cytometry. T cell responses were determined in vivo, ex vivo and by culturing naive T cells with α3_135-145in vitro. Experimental autoimmune anti-GBM disease was induced by α3_135-145 immunization of HLA-DR transgenic mice (Treg intact or Treg depleted).

Results: α3_135-145-specific CD4⁺ T cells were expanded in anti-GBM patients and infiltrated kidneys of α3_135-145-immunized HLA-DR15 transgenic mice. Mice transgenic for both HLA-DR15 and HLA-DR1 showed similar reactivity to most α3 peptides, but lacked pro-inflammatory responses to α3_135-145. Structurally, HLA-DR15 and HLA-DR1 presented α3_135-145 in different binding registers. Compared to the conformation of α3_135-145 in HLA-DR15, in HLA-DR1 α3_135-145 adopted a more pronounced conformation with a flipped backbone. These different HLA-peptide conformations altered α3_135-145-specific TCR usage and T cell phenotypes. HLA-DR15-α3_135-145-specific T cells exhibited a Foxp3^- conventional T cell phenotype resulting in pro-inflammatory T cell autoreactivity, anti-GBM antibody production and disease. In contrast, HLA-DR1-α3_135-145-specific T cells were predominantly anti-inflammatory Foxp3⁺ Tregs that proliferated and responded to α3_135-145. Furthermore, in HLA transgenic mice expressing both HLA-DR15 and HLA-DR1, autoreactivity to α3_135-145 and disease protection was Treg dependent, but Treg depletion in HLA-DR1 transgenic mice did not result in disease.

Conclusions: HLA-DR15 presents α3_135-145 to induce the development of nephritogenic T cells, while HLA-DR1 protects from disease by presenting α3_135-145 in a conformation that activates dominantly protective antigen-specific Tregs. These results provide a mechanistic basis for the HLA associations in anti-GBM disease and explain the mechanism of HLA-mediated protection in autoimmune disease.

WS3_3 Identification of Functional and Expression Polymorphisms Associated with Risk for Anti-Neutrophil Cytoplasmic Autoantibody-Associated Vasculitis

Objectives: To identify risk alleles relevant to the cause and biology of anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV).

Methods: We conducted a genome-wide association and subsequent replication study including 1986 cases of AAV [granulomatosis with polyangiitis (GPA) or microscopic polyangiitis (MPA)] and 4723 controls. Meta-analysis of these datasets and functional annotation of identified risk loci were performed and candidate disease variants with unknown functional effects investigated for impact on gene expression and/or protein function.

Results: Among the genome-wide significant associations identified, the largest effect on risk came from rs141530235 and rs1042169 SNP variants at the HLA-DPB1 gene locus (odds ratios = 2.99 and 2.82, respectively) which, together with a third (rs386699872) variant, constitute a triallelic risk haplotype encoding a DP beta chain dimorphic polymorphism shown in this study to be associated with reduced HLA-DPB1 gene and HLA-DP protein expression in B cells and monocytes and with an increased frequency of complementary proteinase-3-reactive T cells relative to levels in protective haplotype carriers. Genome-wide significant associations were also identified at the SERPINA1 and PTPN22 loci, the peak associations at each locus coming from a functionally-relevant missense variant (rs28929474 and rs2476601, respectively) and at the PRTN3 gene locus (rs62132293), the highest signal corresponding to an imputed risk variant shown here to correlate with increased PRTN3 expression in neutrophils. While effects of individual loci on risk differed in patients with GPA versus MPA or proteinase 3 versus myeloperoxidase ANCA, the collective population attributable fraction for these variants was substantive (77%). Results of classification and regression tree analysis also revealed a recessive effect of variants at several HLA class II gene loci on risk and the potential for genetic data to identify population subsets at highly elevated risk for developing GPA or MPA.

Conclusion: This study reveals the association of GPA and MPA with biologically-important MHC and non-MHC risk variants that explain much of the genetic etiology of AAV, influence and possibly predict clinical presentation of disease, and engender alterations in immune cell proteins and responses likely key to disease etiopathogenesis.