Page 103 - Recognizing axial spondyloarthritis - Janneke de Winter

This study shows a major difference in prevalence of an MHC class I gene in axSpA. In recent years, extensive progress has been made identifying susceptibility alleles in the disease with over 100 additional loci identified (16–21). Using a GWAS approach to identify additional loci in the HLA region remains, however, very challenging because of the huge diversity in this genetic region and the potential linkage disequilibrium with HLA-B*27. In this study we used a ‘reverse’ approach to identify HLA-C*07 as a candidate susceptibility gene: rather than performing large-scale genetic screening and then assessing altered expression and/or function of the gene products of interest, we started by assessing altered expression of a focused set of molecules in a small homogenous sample set in order to select strong signals and avoid false positive data. Subsequently, we analyzed whether altered expression of target molecules was related to genetic variation, allowing us to uncover the association with HLA-C*07.
In sharp contrast to psoriasis, where there is a clear genetic association with HLA-C06 (22,23), indications that HLA-C could play a role in axSpA are sparse. One study in Korean patients showed an increased prevalence of HLA-C15:02 in AS patients, mainly if they were HLA-B*27 negative (24). In PsA, a condition closely related to axSpA, it was reported that patients without HLA-C*0701 had a 42.2% chance of symmetrical sacroiliitis, when compared to 15.8% of PsA patients with an HLA-C*0701 allele (25). This observation is in line with our study, suggesting a link between radiographic damage and HLA-C*07 negativity. In IBD, one Spanish study showed an interaction between a single nucleotide polymorphism (SNP) in endoplasmic reticulum aminopeptidase (ERAP)-1 with the HLA-C*07 antigen (26). Interestingly, ERAP-1 SNPs are also associated with AS (20,21), but the cohorts analyzed in the present study were too small to assess reliably such interaction in axSpA. Nevertheless, these data in PsA and IBD are consistent with our genetic findings in axSpA that HLA-C*07 could play a role in these closely related conditions.
The proposed genetic association between absence of HLA-C*07 and susceptibility to axSpA could potentially be relevant for pathophysiology, for diagnosis, and/or for disease stratification. As to pathophysiology, it is tempting to hypothesize that HLA-C*07 (or its absence) can shape the inflammatory response through its impact on killer cell immunoglobulin-like receptors (KIRs). KIRs regulate natural killer (NK) and T cell function by recognition of HLA class I molecules (27). The regulation of NK and T cell responses appear to depend on KIR genotype, HLA genotype, and the interaction between HLA and KIR products. Depending on the receptor type, KIRs activate or inhibit NK and T cell responses. An imbalance between activating and inhibitory KIRs in the presence of specific HLA genotypes may play a role in susceptibility to or protection against several conditions. Such a mechanism has been shown to play a role in the maintenance of pregnancy (28), in clearance of hepatitis C viral infection (29), as well as in inflammatory diseases such as PsA (30) and psoriasis (22). Interestingly, HLA-B*27 homodimers have been suggested to contribute to axSpA by triggering
HLA-C*07 IN AXIAL SPA
101
SIX