Association of Variants in CCR6 with Susceptibility to Lupus Nephritis in Chinese
Xu-jie Zhou,Rong Mu,Chun Li,Swapan K. Nath,Yue-miao Zhang,Yuan-yuan Qi,Zhan-guo Li,Ming-hui Zhao,Hong Zhang
DOI: https://doi.org/10.1002/art.39268
2015-01-01
Arthritis & Rheumatology
Abstract:Systemic lupus erythematosus (SLE) is a common autoimmune disease with a complex genetic etiology. In the last 2 years, exciting insights supporting the notion that the interleukin17 (IL-17)–producing Th 17 cell subset exerts proinflammatory activity in the pathogenesis of several autoimmune or allergic diseases, including asthma, atopic dermatitis, rheumatoid arthritis (RA), and SLE, have emerged (1,2). High production of IL-17 is observed in SLE patients, and several compounds targeting IL-17 have demonstrated marked clinical efficacy (1,2). Genetic studies have reproducibly demonstrated an association between susceptibility to RA and polymorphisms of the CCR6 gene, a surface marker for Th17 cells (3,4). However, to date no direct link between CCR6 and SLE has been found from genetic studies, although higher percentages of CCR61 T helper cells were observed in patients with SLE, and especially in those with lupus nephritis (LN), compared to controls (5,6). The current study was undertaken to investigate whether there are any associations between CCR6 polymorphisms and susceptibility to LN. The discovery cohort in this study comprised 500 LN patients (mean 6 SD age 32.0 6 11.5 years; 83.6% female) and 900 unrelated healthy individuals (age 31.5 6 8.4 years; 40.1% female). The replication cohort consisted of 626 LN patients (age 32.5 6 12.8 years; 84.0% female) and 1,932 healthy individuals (age 40.9 6 12.7 years; 47.7% female), as well as 1,063 SLE patients without evidence of renal involvement (age 36.6 6 13.4 years; 90.5% female) as an additional control group. All of the SLE patients met the American College of Rheumatology revised criteria for the disease (7). The study was approved by the medical ethics committee of Peking University, and all subjects provided informed consent. Two intronic single-nucleotide polymorphisms (SNPs), rs3093023 and rs3093024, known to be associated with RA with high association signals (3,4), were selected for study, and genotyping was conducted by TaqMan allele discrimination assay (Applied Biosystems) as previously described (8,9). Direct sequencing was performed in 45 samples that were randomly selected on the basis of rs3093024 genotype (15 subjects with the A/A genotype, 15 with the G/A genotype, and 15 with the G/G genotype) to determine whether rs3093024 could tag the recently identified functional CCR6DNP (a triallelic dinucleotide polymorphism of CCR6) (3). The power of the study was calculated using CaTS (http://www.sph.umich.edu/csg/abecasis/CaTS/). As rs3093023 and rs3093024 are in high linkage disequilibrium (LD) (r 5 0.98), indicating that statistical tests performed on each SNP are highly dependent, no multiple correction was applied. Odds ratios (ORs) and 95% confidence intervals (95% CIs) were calculated. Statistical analyses were performed with SPSS16.0 software. Functional annotations of variants were obtained from the HaploReg and regulomeDB databases. A total of 5,021 Chinese subjects were included as described above. The call rates for rs3093024 and rs3093023 were 99.82% and 98.94%, respectively. Both SNPs studied were in Hardy-Weinberg equilibrium in patients and controls (P . 0.05). With the expected frequency of the rs3093024 and rs3093023 minor allele (40%), the study had a power of 96– 99% to detect a 1.20-fold increased risk in the discovery and replication cohorts separately and combined. The frequency of the A allele of rs3093024 was significantly higher in patients with LN as compared to healthy controls both in the discovery cohort and in the replication cohort (P 5 1.32 3 10 and P 5 4.15 3 10, respectively) (Table 1).