Polymorphisms of interleukin-21 and interleukin-21-receptor genes confer risk for autoimmune thyroid diseases

Autoimmune thyroid diseases (AITDs), the thyroid-specific autoimmune disorders, affect about 5% of the population. The etiology of AITDs includes genetic and environmental factors resulting in immune abnormality and the development of AITDs.

In 2000, the interleukin-21(IL-21)-IL21-receptor (IL-21R) system was discovered [1]. IL-21 and IL-21R are located on human chromosome 4q26-27 and 16p11, respectively. The IL-21, preferentially produced by CD4 + T cells, has various functions, such as driving B cells to differentiate into memory cells and ultimately plasma cells, augmenting T cells’ proliferation, and promoting the activity of natural killer (NK) cells [2]. In accordance with these roles of IL-21, IL-21R is mainly expressed in B cells, T cells and NK cells, keratinocytes, and some of myeloid cells. Animal and clinical studies demonstrated the dysregulations of IL-21 and IL-21R in autoimmune diseases. For example, compared with IL-21R-competent BXSB-Yaa mice for multiple parameters of SLE, the IL-21R-deficient Yaa mice showed none of the abnormalities characteristic of systemic lupus erythematosus (SLE) [3]. IL-21R is overexpressed in the inflamed synovial membrane and in peripheral blood or synovial fluid leukocytes of rheumatoid arthritis (RA) patients [4]. And the evident increase of serum IL-21 level in primary Sjogren’s syndrome (pSS) patients has positive correlation with the levels of gamma-globulin and erythrocyte sedimentation rate [5].These suggests that IL-21 and IL-21R may play a critical role in the pathogenesis of autoimmune diseases. Recently, H.Y. Jia [6] also reported that an increase of serum IL-21 level in patients with primary GD.

In recent years, the associations between the IL-21 gene or IL-21R gene polymorphisms and autoimmune diseases have been gradually reported. In a HapMap-CEU population, a large block (480 kb) of linkage disequilibrium encompassing KIAA1109/Tenr/IL2/IL21 showed genetic associations with type 1 diabetes mellitus (T1DM) [7, 8], RA [8], juvenile idiopathic arthritis (JIA) [9], psoriasis and psoriatic arthritis(PA) [10]. Similarly, the polymorphisms of the IL2/IL21 gene region likely to confer to the susceptibility to coeliac disease in Scandinavia families [11] and a US population [12], to ulcerative colitis in a North American and an Italian cohort [13], and to multiple sclerosis in a Spanish population [14]. After the SNPs in the IL-21 gene were analyzed, the significant differences of rs907715 and rs2221903 allele and genotype frequencies were found between SLE and controls [15]. In addition to these reports, an association between microsatellite polymorphisms of the IL-21R and diabetes in Japanese patients, between rs3093301 and rs2285452 in the IL-21R gene region and SLE in the European-derived cohort were documented [16].

Therefore, it is reasonable to speculate that IL-21 and IL-21R are the candidate genes for AITDs. Plagnol [17] reported that the rs2069763 in chromosome 4q27 were associated with the GD susceptibility in a Caucasian cohort. And a recent study that indicated that rs907715 of IL-21 gene was associated with GD in a Chinese population [6].

In this study, additional variants of the IL-21 gene and two SNPs of the IL-21R gene in addition to previously reported variants were investigated in two subtypes of AITDs including GD and HT.

Four SNPs (rs907715, rs4833837, rs2221903 and rs2055979) of the IL-21 gene and two SNPs (rs3093301 and rs2285452) of the IL-21R gene were typed in 405 GD and 228 HT patients and 242 controls. For the tested SNPs, genotyping successful rate was more than 96.9%. All of the 6 genotyped SNPs showed a MAF more than 7% and HWE analysis in the three groups showed P values of more than 0.05 (seen in Table 1). When comparing age and sex distributions between patients and controls, we found the control group matched well with the GD group, but not with HT group (seen in Table 2). The distributions of the SNPs genotypes in HT patients were analyzed using logistic regression analysis adjusted for sex and age when compared with the controls.

IL-21 is a member of the common-gamma chain family of cytokines with immunoregulatory activity. IL-21R has been shown to form a heterodimeric receptor complex with the common gamma-chain. IL-21 binding with this receptor leads to the activation of multiple downstream signaling molecules, including JAK1, JAK3, STAT1 and STAT3, therefore affects the innate and adaptive immune responses by inducing the differentiation, proliferation and activity of multiple target cells. The dysregulation of IL-21 and IL-21R plays a role in multiple immune-mediated diseases, including SLE [3], psoriasis [5], RA [4] and other chronic inflammatory diseases [18]. Like other autoimmune diseases, GD and HT are chronic diseases initiated by the loss of immunological tolerance to self-antigens. Previous studies indicate that some immune-related genes may participate in the development of AITDs.

In this study, we found a significant association between GD or HT and rs2221903 located in the IL-21 gene region, and the frequencies of the haplotypes in the IL-21 gene that consists of rs4833837 and rs2221903 in GD and HT patients were significantly different from controls. For rs3093301 and rs2285452 polymorphisms of the IL-21R gene, there were significantly different distributions of these genotypes between HT patients and the controls.

In different from Jia’s study [6], we did not find the rs907715 SNP linkage with GD. In Jia’s report, the AA genotype frequency was lower in GD patients (19.4%) than in controls (30%). The gene frequency in patients of our study was similar to that of Jia’s study, but different in the two control groups. In our study the frequency of AA genotype in the control group was 18.9%, which is similar to the frequency in the HapMap-CHB population. This may be the geographical or selection variation that results in this difference. It should be noted that the lack of association in our study does not completely exclude the possibility of IL-21R as a candidate gene for GD because of the following three reasons: 1) The average MAF of the six SNPs is 0.271, which gives our study genetic power of about 0.8 for GD and 0.7 for HT group with an OR of homozygote 2.0, and of heterozygote 1.5, therefore studies with a larger sample size are necessary to confirm whether patients with these SNPs have more risk for development of GD. 2) As many other immunologic disorders, AITDs is believed to derive from a multiple network of various susceptible loci, which exert synergic or additive effects, but each locus may play a small role [19]. 3) 13 tag-SNPs in the HapMap-CHB population and 17 in the HapMap-CEU population covered a 49.8 kb on 16p11, where IL-21R gene located. And we just explored the association between the two tag-SNPs of them and AITDs in Chinese cohort.

In genetics, a recessive gene is an allele that causes a phenotype (visible or detectable characteristic) that is only seen in a homozygous genotype (an organism that has two copies of the same allele) and never in a heterozygous genotype. While dominance is a relationship between alleles of a gene, in which one allele masks the expression (phenotype) of another allele at the same locus. Our study didn’t find any synergistic effect of the risk genotype. And we were unable to conclude whether these two genes were either recessive or dominant over the other gene.

IL-21 is located on chromosome 4q26-27, which is close to IL-2, a region that has been linked to AITDs susceptibility [17]. Jia’s et al. found rs907715 of IL-21 gene being associated with GD in a Chinese population while Plagnol found only rs2069763 and did not find association of rs2069762 and rs6822844 of IL-2 with GD, maybe there is ethnicity or regional effect contributing to these results differences. Therefore, we can not exclude that the SNPs of the IL-21 and IL-21R genes are only an indicator of a candidate gene contributing to AITDs. Further studies such as screening the rest common variations in IL21R gene, identifying the causal SNP for this association, and exploring gene-environment interaction in GD and HT cohort studies are required to clarify the effect of IL-21 and IL-21R on AITDs susceptibility. Replication of these associations between IL-21 and IL-21R gene and AITDs is also required in larger independent databases of different cohorts.

In conclusion, our study confirmed the synergic effect of the IL-21 SNPs in the development of GD. In addition, to the best of our knowledge, we are the first to report the association of the IL-21 and IL-21R SNPs with an increased risk of HT.