Association of Interleukin-23 receptor gene polymorphisms with susceptibility to Crohn’s disease: A meta-analysis

Studies investigating the association between Interleukin-23 receptor (IL-23R) gene polymorphisms and Crohn’s disease (CD) report conflicting results. Thus, a meta-analysis was carried out to assess the association between the IL-23R polymorphisms and CD. A systematic literature search was conducted to identify all relevant studies. Pooled odds ratio (ORs) with 95% confidence interval (CIs) was used to estimate the strength of association. Finally, a total of 60 case-control studies in 56 articles, involving 22,820 CD patients and 27,401 healthy controls, were included in the meta-analysis. Overall, a significant association was found between all CD and the rs7517847 polymorphism (OR = 0.699, 95% CI = 0.659 ~ 0.741, P < 0.001). Meta-analysis of the rs11209026, rs1343151, rs10489629 and rs11465804 polymorphisms indicated the same pattern as for rs7517847. Meta-analysis showed an association between the rs10889677A allele and CD (OR = 1.393, 95% CI = 1.328 ~ 1.461, P < 0.001). Similarly, meta-analysis of the rs2201840, rs1004819, rs1495965 and rs11209032 polymorphisms revealed the same pattern as that shown by meta-analysis of rs10889677. Stratification by ethnicity revealed that IL-23R gene polymorphisms were associated with CD in the Caucasian group, but not in Asians. In summary, the meta-analysis suggests a significant association between IL-23R polymorphisms and CD, especially in Caucasians.

In mice models, IL-23/IL-23R, a key component of IL-23/Th17 pathway, is essential for intestinal inflammation of T cell-dependent colitis 18,19 . Furthermore, patients with CD manifested increased IL-23 secretion compared with controls 20 . A genome-wide association study indicated that IL-23R gene is a potential candidate for CD susceptibility 21 . IL-23R gene is mapped to the chromosome 1 (1p31. 3) 13 . Recently, polymorphisms in this locus have been identified and a large number of studies indicated the association between these polymorphisms and CD risk  . However, previous studies in different populations reported conflicting results. The disaccord may be attributed to small sample size, various racial and ethnic backgrounds, uncorrected multiple hypothesis testing and publication bias.
Meta-analysis is a statistical method for augmenting the effective sample size through combining the results of several studies to produce a single estimate of the major effect with enhanced precision. It is considered to be a powerful tool for pooling inconsistent results from different studies 77 . To date, two meta-analysis 52,78 of the association between the IL-23R gene polymorphisms and CD risk have been reported, but these results were still inconclusive. New studies about the role of these SNPs in the IL-23R gene in CD have been published in recent years [53][54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69][70] , and these studies might provide new evidence. Thus, it seems necessary to perform an updated and comprehensive meta-analysis including the latest data to investigate the association of the IL-23R gene polymorphisms and the risk of CD.

Materials and Methods
Publication research. A literature search was conducted for studies that examined association between the IL-23R gene polymorphisms and CD risk in October 13, 2015. We used PubMed database, Elsevier Science Direct, China National Knowledge Infrastructure database (CNKI) and Chinese Biomedical database (CBM) to identify articles with the following terms: "Interleukin-23 receptor, IL-23R, Crohn's disease, Crohn disease, inflammatory bowel disease, CD and IBD". References in the studies were also reviewed to find additional studies. The languages were limited to English and Chinese.
Inclusion and exclusion criteria. A study was included in the analysis if a) it was a case-control or cohort study; b) it included original data (independence among studies); and c) it provided sufficient data to calculate the odds ratio (OR). We excluded the followings: a) studies contained overlapping data; b) studies in which family members had been studied because of the analysis based on linkage considerations. Data extraction. Data were collected by two independent investigators (Qi-Bing Xie and Yi Zhao). The characteristics of the selected articles were in Table 1, including first author, year of publication, study population, ethnicity, numbers of case and control, genotyping method, the concrete polymorphisms investigated in these studies. The study populations comprised Italian, German, Dutch, Hungarian, Korean, Chinese, Malaysian, African American, Australian, Canadian, Algerian and so on. The Asian subgroup included Korean, Chinese, and Malaysian populations. African American population was classified in the African subgroup and others in the Caucasian subgroup. Statistical analysis. Allele frequencies at the IL-23R gene polymorphisms from the respective study were determined by the allele counting method. The contrast of the allelic effect of minor allele versus common allele of the IL-23R gene polymorphisms was tested. The strength of association between the IL-23R gene polymorphisms and CD susceptibility was assessed by odds ratios (OR) and 95% confidence intervals (95% CI).
The heterogeneity of between-studies was evaluated by the Chi-square test based Q-statistic 79 . In addition, the I 2 -statistic also measures the degree of inconsistency in the studies by computing what percentage of the total variation across studies was due to heterogeneity rather than by chance. A high value of I 2 indicated a higher probability of the existence of heterogeneity (I 2 = 0% to 25%, no heterogeneity; I 2 = 25% to 50%, moderate heterogeneity; I 2 = 50% to 75%, large heterogeneity; and I 2 = 75% to 100%, extreme heterogeneity). If a P-value of the heterogeneity Q-statistic was more than 0.10, the pooled OR was calculated by the fixed effects model. Otherwise, a random effects model was adopted. Potential publication bias was analyzed by Egger's linear regression test and the Funnel plot. If P-value was less than 0.05, statistically significant publication bias might exist 80 .
All the statistical analysis of meta-analysis was performed by STATA statistical software (version 11.0 STATA Corp LP, College Station, TX, USA).

Results
Study characteristics. The process for selecting the studies was shown in Fig. 1. 424 potentially relevant records were first reviewed, and 56 articles met the inclusion criteria were finally included in themeta-analysis  . Among the 56 articles, 4 were published in Chinese and the others in English. Of the 56 articles, one article included three cohorts 24 , other two 21,68 included two cohorts, thus, each cohort was considered as a single study. Finally, a total of 60 case-control studies in 56 articles involving 22,820 CD patients and 27,401 healthy controls were identified. There were 49 studies on rs11209026, 25 studies on rs7517847, 19 studies on rs2201841 and rs1004819 separately, 18 studies on rs10889677, 15 studies on rs1343151 and rs1495965 separately, 13 studies on rs11209032, 12 studies on rs10489629 and 8 on rs11465804. Forty-nine studies involved Caucasian populations [66][67][68][69] , nine studies involved Asian populations 34,51,61,[70][71][72][73][74][75] , and two study involved African populations 65,76 . The main characteristics of each study included in this meta-analysis were summarized in Table 1.

Evaluation of heterogeneity and publication bias.
Between-study heterogeneity was found in some meta-analysis of the IL-23R gene polymorphisms in CD (Table 2). Therefore, these meta-analyses were performed in a random effects model, and the other meta-analyses were done in a fixed effects model.  The funnel plot asymmetry and Egger's test were performed to assess potential publication bias. If there was asymmetry, the index to determine Egger's test will show P < 0.05. It was shown that there was no significant publication bias between all of the comparisons (all P-value > 0.05).

Discussion
A genome-wide association study by Duerr et al. 21 found strong correlation between IL-23R polymorphisms and inflammatory bowel disease (IBD). Since then, genetic variants of IL-23R have been investigated in numerous autoimmune diseases. It has been shown that the IL-23R gene plays an important role in the pathogenesis of some autoimmune diseases, such as IBD, AS, psoriasis, rheumatoid arthritis, and multiple sclerosis 81 . Oliver et al. 30 and Baptista et al. 36 found rs11209026, rs10889677 and rs1004819 polymorphisms were associated with CD disease susceptibility in Spanish. Similarly, a study on Hungarian with CD revealed that the rs1004819T carrier (TT + TC) had a 53% increasing risk when compared with the CC homozygote 53 . Furthermore, a study in Korean population indicated that the rs1495965G allele frequency was higher in CD cases compared to controls (OR = 1.31, 95% CI = 1.07 ~ 1.60) 51 . However, Okazaki et al. 43 and Chua et al. 61 found no significant differences in allele or genotype frequency of rs11209026 and rs1004819 polymorphisms between CD patients and controls in a Malaysian population. Therefore, to better comprehend the association between polymorphisms in the IL-23R gene and CD susceptibility, a pooled analysis with a larger sample size, subgroup analysis performed and heterogeneity explored is needed. Overall, results of this meta-analysis suggested that IL-23R gene polymorphisms were associated with CD susceptibility (all P value < 0.001). Ethnicity-specific analysis showed that the polymorphisms in the IL-23R gene might confer susceptibility to CD in Caucasians, but not in Asians.
The diverse roles of the same gene polymorphism in ethnicity-specific analysis by ethnicity could be ascribed to the following major aspects. First, Yamazaki et al. 34 showed that IL-23R gene was not a candidate gene to CD in the Japanese population. Furthermore, Lee et al. 82 found that rs1004819, rs10489629, rs1343151, rs1495965, rs11209032 and rs2201841 polymorphisms revealed a different association with ankylosing spondylitis between Caucasians and Asians. Compared to the populations of Caucasian, no significant association was detected for IL-23R polymorphisms in Chinese psoriasis patients 83 . Similarly, IL-23R (rs1004819, rs7517847, rs10489629, rs2201841, rs1343151, rs11209032, and rs1495965) polymorphisms were not associated with systemic lupus erythematosus (SLE) in the Korean population 84 , while rs10889677 and rs7517847 polymorphisms were not associated with Chinese SLE patients 85 . In addition, CD is the most prevalent in North America and Europe, and the least prevalent among African Americans and Asians 86 . These findings suggested that IL-23R gene might have a susceptible nature in the Caucasians, but not in the Asians. Therefore, the inconsistent results of subgroup analysis might be attributed to ethnic differences. Second, for other variants in the IL-23R gene, Zhao et al. 75 found that IL-23R rs11805303 and rs17375018 polymorphisms were not associated with CD disease susceptibility in Chinese. Similarly, a study in another Chinese population showed no significant differences in allele or genotype frequency of the rs11805303 polymorphism between CD patients and controls 71 . However, Chen et al. 72 found that the rs11465788T allele frequency was lower in Chinese CD cases compared to controls (OR = 0.30, 95%CI = 0.15-0.60). Since few studies in Asian population focus on these polymorphisms, these results might not be reliable. Therefore, large sample size GWAS should be considered in future to confirm the role of these variants in the IL-23R gene in Asians. Third, most autoimmune diseases including CD are multifactor diseases, and these diseases were caused by an

Figure 2. Odds ratios and 95% confidence intervals for individual studies and pooled data for the association between the G versus T allele of the IL-23R rs7517847 polymorphism and Crohn's disease.
interaction of genetic and environmental factors 87 . Therefore, different populations lived in different environments, gene-environment interactions may partly affect the autoimmune diseases susceptibility. It has been widely accepted that genetic and environmental factors play an important role in disease initiation such as SLE, and the progression of the disease. A previous study has confirmed that the IFIH1 rs1990760 polymorphism was associated with SLE in the European population, not in the Asian population 88 , suggesting that interactions between different environments and genes might be different.
Although the IL-23R gene is associated with CD risk, the function of IL-23R gene variants remains unclear. Several possible mechanisms can be suggested by which polymorphisms can regulate the function of the IL-23R gene. First, the standard form of IL-23R is encoded by at least 12 exons. Zhang et al. 89 demonstrated that at least six spliced isoforms of IL-23R (IL-23R1 to 6) can be generated through alternative splicing. Translation prediction revealed that spliced variants led to either premature termination to give rise to a diverse form of receptor ectodomain, or a frameshift to generate various lengths of the IL-23R endodomain 89 . Furthermore, the intronic polymorphisms, such as rs11805303, rs1004819, might exert their influence by regulating differential splicing. Second, the rs11209026 polymorphism, also named as Arg381Gln, is located between the transmembrane domain and the putative JAK2 binding site in the cytoplasmic portion of IL23R protein and highly conserved between species 90 . Replacing conserved Arg381 for Gln381 at this position probably modulates IL-17 and IL-22 expression in response to IL-23 stimulation, which may have a functional influence on IL-23R signaling pathway 91,92 . Third, the rs10889677 polymorphism, located in the 3′ -UTR, might promote overexpression of the receptor through increasing mRNA stability and driving T cells to differentiate towards Th17, and thus, leading to inflammation by increasing release of other cytokines 81 . Forth, the rs10889677 variant enhanced both mRNA and protein expression of IL-23R can lead to a loss of binding capacity to the microRNAs (miRNAs) Let-7e and Let-7f. MiRNA-mediated dysregulation of IL-23R signaling correlated with a single nucleotide polymorphism in the IL-23R gene was strongly associated with CD susceptibility, suggesting that this mutation can lead to altered IL-23R signaling 93 .
Compared with the previous meta-analysis 52,78 , the current study involved a total of 60 case-control studies in 56 articles, which is much larger than the data of the previous meta-analysis. Moreover, we performed subgroup analysis by ethnicity to discuss the ethnic effect on the risk of CD. Thus, our meta-analysis might enhance the statistical power and draw a more reliable conclusion.
Some limitations of the present study should be considered. First, we could not analyze the potential gene-environment interactions and gene susceptibility haplotypes owing to lack of data, such as the data of environmental risk factors and genotypes. Second, our literature search was only dependent on English and Chinese, language bias might be considered. Third, potential publication bias was not found by statistical method, but it might exist because of only published articles included. Fourth, only one published studies in the African origin was included in the meta-analysis, the stratified analysis for Africans might not be reliable. Thus, the results were applicable only to the Asian and Caucasian groups. Finally, different genotyping methods and disease status might affect the data interpretation of the included studies.
In summary, the current study provides a comprehensive examination of the available evidence for the association between polymorphisms in the IL-23R gene and CD. This updated meta-analysis suggests that IL-23R gene polymorphisms are associated with CD susceptibility. However, larger sample size studies taking environmental risk factors into account and including more ethnic groups should be considered in future to confirm the results from our meta-analysis.