Background The CCR6-CCL20–mediated migration of Th17 cells to inflamed tissues may represent an important mechanism in the etiology of rheumatoid arthritis (RA). The CCR6 polymorphism rs3093023 is associated with RA disease risk. Variation in the CCR6 locus has been found to affect CCR6 gene expression and influence the IL17 serum concentration in RA patients . An analysis of disease risk stratified by gender revealed opposing effects in Asian subjects: Specifically a CCR6 SNP (rs3093024), which is in almost complete linkage disequilibrium with rs3093023, was associated with an increased risk of RA in women, but appeared to have a protective effect in men .
Objectives We set out to determine whether variation in estrogen levels might influence the expression of CCR6 and other IL-17 pathway related genes. Furthermore, we aimed to clarify if such estrogen dependent regulation might be influenced by the presence of the variant CCR6 genotype that is associated with RA disease risk in Europeans.
Methods We genotyped human lymphoblastoid cells (LCLs) using the Illumina 550K and 510S SNP BeadChip and the Affymetrix SNP array 6.0. We then cultured eight LCLs homozygous for the wild-type (WT) SNP rs3093023 sequence and eight homozygous for the variant (V) allele with increasing concentrations of estradiol (E2). Expression of CCR6, CCL20, IL17A and IL17RA mRNA was measured by qPCR. We then performed siRNA knockdown experiments for CCR6 to explore the downstream effects. To predict putative estrogen receptor (ER) binding sites in the CCR6 gene, we queried the TRANSFEC database. The functional relevance of putative binding sites was confirmed using ChIP assays.
Results The basal expression levels of CCR6, CCL20, IL17A and IL17RA showed no differences when comparing cell lines with CCR6 WT versus V genotypes. Knockdown of CCR6 resulted in upregulation of CCL20 and IL-17A, but downregulation of IL-17RA. Treatment with E2 for 24 hours resulted in a significant increase in CCR6 expression in a dose-dependent manner, but only in cells with the V allele. CCL20, IL-17A and IL17 RA expression was also SNP-dependent: in cells with the variant genotype, E2 treatment resulted in higher IL-17RA expression. Conversely, CCL20 and IL-17A expression decreased with E2 treatment only in cells homozygous for the V allele. The TRANSFEC database predicted the presence of two ER elements in the CCR6 intronic region flanking the rs3093023 SNP. ChIP assays demonstrated increased binding of ERalpha when the V allele was present.
Conclusions Our results indicate that a CCR6 variant, which is associated with RA disease risk, can modulate the CCR6/CCL20/IL-17 axis in an estrogen-dependent manner. Enhanced binding of ERalpha to the variant CCR6 genotype may contribute to the mechanism underlying this observation.
Future studies will be required to show whether the SNP dependent response to estrogen is a phenomenon that also applies to other loci that are associated with RA disease risk. Such a genomic link between variation in sex hormone levels and variation in cytokine/chemokine expression may provide new insights into the gender differences in RA prevalence and prognosis.
Kochi Y, et al. Nat Genet 2010;42: 515-9
Teng E et al.: DNA Cell Biol. 2012 Apr;31:607-10
Disclosure of Interest None declared