Epistasis between the MHC and the RCA alpha block in primary Sjögren syndrome

Abstract

Objective: The RCA alpha block (Regulators of Complement Activation, 1q32) contains critical complement regulatory genes such as CR1 and MCP. This study examined RCA alpha block haplotype associations with both disease susceptibility and diversification of the anti-Ro/La autoantibody response in primary Sjögren syndrome (pSS).

Methods: 115 patients with pSS and 98 controls were included in the study. 93 of 109 (85%) of the patients with pSS were seropositive for Ro/La autoantibodies. The Genomic Matching Technique (GMT) was used to define RCA alpha block ancestral haplotypes (AH).

Results: RCA alpha block haplotypes, AH1 and AH3, were both associated with autoantibody-positive pSS (p = 0.0003). Autoantibody associations with both HLA DR3 and DR15 have been previously defined. There was an epistatic interaction (p = 0.023) between RCA alpha AH1 and HLA DR3, and this genotypic combination was present in 48% of autoantibody-positive patients with pSS compared with 8% of controls. This epistasis is most simply attributable to an interaction between C4 and its receptor, CR1, encoded within the RCA alpha block. Both DR3 and a relative C4 deficiency are carried on the major histocompatibility complex 8.1 ancestral haplotype. Only four of 92 (4%) autoantibody-positive patients with pSS did not carry any risk RCA alpha or HLA haplotype, compared with 36 of 96 (38%) controls, and there were differences in haplotype frequencies within autoantibody subsets of pSS.

Conclusions: Normal population variation in the RCA alpha block, in addition to the major histocompatibility complex, contributes genetic susceptibility to systemic autoimmune disease and the autoantibody response. This finding provides evidence for the role of regulation of complement activation in disease pathogenesis.

Figure 1. RCAα ancestral haplotype (AH) 1 and AH3 genotype frequencies in patients with primary Sjögren syndrome who were Ro/La autoantibody positive compared with controls. There was a significant departure from an additive or allele dose model (p = 0.043) by logistic regression analysis, and the data are most consistent with AH1 and AH3 exerting dominant effects of similar size. Under the dominant model, AH1 and AH3 appear to be independently associated with autoantibody-positive primary Sjögren syndrome, ie, with multiplicative risks.

Figure 2. Epistatic interaction between the MHC and RCAα block in Ro/La autoantibody-positive patients with primary Sjögren syndrome (pSS). (A) Odds ratios (y-axis, logarithmic scale) derived by logistic regression for the cross-classification of HLA DR3, DR15 and RCAα AH1, AH3 genotypic combinations (dominant coding) in Ro/La autoantibody-positive patients with pSS (n = 92) relative to controls (n = 98). The vertical bars represent 95% confidence intervals, and the horizontal line represents an odds ratio of 1 (no effect). HLA DR3, DR15 alleles and the RCAα AH3 haplotype were independent risk factors for autoantibody-positive pSS (ie, multiplicative risks), but there was an epistatic interaction between HLA DR3 and RCAα AH1 (interaction term p = 0.021). The genotypic combination of HLA DR3 and RCAα AH1 was the greatest genetic risk factor for autoantibody-positive pSS (OR 15.7, p = 10−8), but in the absence of DR3, there was no effect of RCAα AH1. (B) Pie chart depicting the relative proportions of risk genotypes. The HLA DR3–RCAα AH1 epistatic combination was present in 48% of autoantibody-positive patients with pSS compared with 8% of controls. The majority of other patients with pSS carried any combination of HLA DR3, DR15 and RCAα AH3.

Figure 3. Phenotypic prevalence of HLA DR3 (A), HLA-DR15 (B), RCAα AH1 (C) and RCAα AH3 (D) haplotypes by diversification of the Ro/La autoantibody response within primary Sjögren syndrome (pSS). There were 115 patients with pSS in the study: 19 were seronegative, 19 with anti-Ro only, 22 with anti-Ro+La (ppt−) and 55 with anti-Ro+La (ppt+). The increased prevalence of HLA DR3 in autoantibody-positive pSS is largely B8-DR3 (the prevalence of B8-DR3 in the controls is not known) and there are genetic differences between the pSS autoantibody subgroups.