Association between Interleukin-10 Gene Polymorphisms and Behcet's Disease Susceptibility: Evidence from a Meta-Analysis

Epidemiological studies have demonstrated that interleukin-10 (IL-10) polymorphisms may be associated with the development of Behcet's disease (BD). However, the published results were inconsistent. Therefore, this meta-analysis was conducted to derive a more precise relationship between IL-10 polymorphisms and BD susceptibility. Online databases (PubMed, Embase, Science Citation Index (SCI), CNKI, and WanFang) were searched to identify eligible studies. Odds ratio (OR) and a 95% confidence interval (CI) were applied to assess the relationship strength between IL-10 -1082A>G (rs1800896), -819T>C (rs1800871), and -592A>C (rs1800872) polymorphisms and BD susceptibility. Publication bias, sensitivity, and cumulative analyses were conducted to measure the robustness of our findings. Finally, fifteen articles (36 independent case-control studies) involving 5,971 patients and 8,913 controls were included. Overall, significant associations between -819T>C polymorphisms and BD risk were observed in the total population (C vs. T: OR = 0.72, 95%CI = 0.67‐0.77, P < 0.01, I2 = 36.6%; TC vs. TT: OR = 0.73, 95%CI = 0.66‐0.80, P < 0.01, I2 = 23.0%; CC vs. TT: OR = 0.52, 95%CI = 0.39‐0.70, P < 0.01, I2 = 53.7%; TC+CC vs. TT: OR = 0.67, 95%CI = 0.61‐0.71, P < 0.01, I2 = 22.1%; and CC vs. TT+TC: OR = 0.66, 95%CI = 0.53‐0.82, P < 0.01, I2 = 57.8%). Moreover, the IL-10 -592 A>C polymorphism and the ACC haplotype exhibited a significant, protective effect against BD susceptibility. In summary, our meta-analysis suggested that IL-10 gene polymorphisms may play a salient role for BD development.


Introduction
Behcet's disease (BD) is a chronic polygenic autoinflammatory disorder, characterized by multiple affected body parts, with common symptoms including recurrent oral ulcers, ocular inflammation, genital ulcers, and skin lesions [1,2]. In addition, other organs, such as the musculoskeletal, gastrointestinal, central nervous, and pulmonary systems may be affected and present some clinical manifestations [3]. Today, the most authoritative diagnostic criteria were made according to the criteria of the International Study Group of BD: oral ulceration plus any two genital ulcerations, typical defined eye lesions, typical defined skin lesions, or a positive pathergy test [4]. BD is most common in Asian countries, such as China and Japan (East Asian (EA)), as well as in the Middle East (ME) around the regions of the Old Silk Road, such as Turkey and Iran [5]. Epidemiological studies have shown that BD has the highest prevalence in the population of Jordan, a slightly lower prevalence in East Asian countries, and the lowest prevalence in Western countries [6][7][8][9]. Pathologically, the lesions always impair arterioles, venules, and microvessels, resulting in the development of microvasculitis with vascular wall necrosis, dilation, and rupture. To date, the etiology and pathogenesis of BD have not been elucidated; however, environmental risks, immunological aberrations, and multiple genetic factors are considered risk factors that can increase susceptibility to BD [10,11].
Interleukin-10 has been proven to be an important cytokine-suppressing, autoimmune and inflammatory response factor. IL-10 may inhibit the antigen-presenting process by downregulating the expression of HLA molecules on the surface of a cell and suppressing the expression of multiple proinflammatory cytokines, such as TNF-α, IL-1, IL-6, and IL-8, by inhibiting T cells. IL-10 is secreted by T lymphocytes (mainly Th2 subsets), B lymphocytes, NK cells, monocytes, and macrophages and is involved in the pathogenesis of BD [12,13].
A number of studies have indicated that the aberrant expression of IL-10 might lead to an inflammatory state that is more susceptible to BD [14,15]. A recent genome-wide association study (GWAS) demonstrated that a common polymorphic locus of rs1518111 in IL-10 may cause individuals to have a predisposition to BD [16]. The three most common single nucleotide polymorphisms (SNPs) were in the promoter of IL-10 with loci of -1082A>G (rs1800896), -819T>C (rs1800871), and -592A>C (rs1800872), which could influence the gene transcription and expression by changing the promoter binding site, resulting in differing susceptibility of individuals to the disease. To date, a large number of studies have been published on the association between IL-10 polymorphisms and BD susceptibility.
In 2013, Liang et al. conducted the first meta-analysis based on four case-control studies, exploring the potential association between the three IL-10 polymorphisms mentioned and BD risk [17]. In the last six years, new studies have been published on this topic, but the results are still inconsistent and remain inconclusive. Therefore, we conducted an updated meta-analysis to clarify the genetic association between the -1082A>G (rs1800896), -819T>C (rs1800871), and -592A>C (rs1800872) polymorphisms of the IL-10 gene and the risk of BD.

Materials and Methods
This meta-analysis was conducted according to the guidelines of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [18]. All included data were based on published studies, and no ethical issues were involved.

Eligibility Criteria.
All selected studies met the following criteria: (1) case-control design was employed; (2) some polymorphisms of the IL-10 gene was reported; (3) sufficient data on genotypes of each polymorphism locus were provided to calculate the odds ratios (ORs) and 95% confidence intervals (CIs); (4) the most recent or largest sample sizes were selected when multiple publications were repeatedly reported with the same theme; and (5) the articles were written only in English and Chinese. The studies were excluded if they were (1) letters, reviews, or conference reports; (2) duplicate reports with the same theme and samples; (3) animal models or cell line studies; or (4) studies without sufficient data.

Data Extraction.
Two independent investigators (LXF and HZZ) reviewed the articles and extracted the following information: name of first author in each study, year of publication, study country or region where the study was conducted, ethnicity of research population, design of the control, sample sizes of patients with BD and controls, frequency genotype of distribution data, and genotyping method used. Disagreements were settled through discussion.

Quality Assessment.
Each study included was evaluated via risk assessment of bias by two authors, LXF and HZZ, according to the modified Newcastle-Ottawa Quality Assessment Scale [19]. The score was assigned on the basis of five parameters (representativeness of cases, source of controls, Hardy-Weinberg's equilibrium (HWE) in controls, genotyping examination, and association assessment) and ranked from zero to 11 points. Studies with a score of at least 7 were considered to be of high quality (Table 1).

Statistical
Analysis. The pooled ORs with 95% CIs were used to assess the strength of the associations between the IL-10 gene polymorphisms and BD risk. For example, the genetic models of the -1082A>G (rs1800896) polymorphisms were examined: allele contrast (G vs. A), codominant models (heterozygote model: AG vs. AA; homozygote model: GG vs. AA), dominant model (AG+GG vs. AA), and recessive model (GG vs. AA+AG). Similar genetic models were also examined with the other polymorphic loci (-819T>C (rs1800871) and -592A>C (rs1800872)). Heterogeneity was investigated by Cochran's χ 2 -based Q-statistic among the included studies [20]. A fixed-effect model (Mantel-Haenszel's method) was used to evaluate the results if heterogeneity was not significant (I 2 < 40%) [21]. Otherwise, a random-effects model (DerSimonian and Laird's method) was used [22]. Subgroup analyses were performed according to HWE status, ethnic differences, and control design and genotyping methods. Cumulative analyses were conducted to test the changing trends in the results, and sensitivity analyses were conducted to test the robustness of the original results with the published studies which were added gradually. Publication bias testing was conducted with Egger's bias test and Begg's funnel plots [23,24]. Statistical analysis was performed using STATA version 14.0 (Stata Corporation, College Station, TX, USA), and P values < 0.05 were considered to be significant.

Quantitative Analysis
3.2.1. -1082A>G Polymorphism and BD Risk. Nine casecontrol studies, involving 1,042 patients and 1,389 controls, were identified on the association between the -1082A>G polymorphism and BD risk in this meta-analysis, and these are shown in Table 3. Overall, the synthetic data indicated that there was no significant association between the -1082A>G polymorphism and BD risk in the overall population for all five genetic models (G vs.  Figure 2(a) for the G vs. A model). Moderate heterogeneity was identified in allele contrast, heterozygote model, dominant model, and recessive model. Then, metaregression analyses were conducted and they did not find any factors that might be responsible for the emergence of heterogeneity. In addition, no significant relationship was Association assessment Assess association between genotypes and BD risk with appropriate statistics and adjustment for confounders 2 Assess association between genotypes and BD risk with appropriate statistics and without adjustment for confounders 1 Inappropriate statistics used 0 3 Journal of Immunology Research found in the subgroup analyses based on HWE status, ethnic differences, or control design (Table 3). Cumulative analyses presented a consistent trend of negative results with publication date, except for the second case-control study by Wallace et al. [25] (Figure 2

-819T>C Polymorphism and BD Risk.
Sixteen casecontrol studies, involving 5,550 cases and 8,469 controls, were identified on the association between the -819T>C polymorphism and BD risk in this meta-analysis, and these are shown in Table 3. Overall, a significant association of pro-  Journal of Immunology Research  (Table 3). In addition, the same protective effects were also observed in both the control design groups and the genotyping groups (Table 3). Cumulative analyses steadily demonstrated a significant association with all studies (Figure 3(b) for the C vs. T model). Sensitivity analysis was conducted and did not present any apparent changes in the five genetic models (Figure 3

-592A>C Polymorphism and BD Risk.
Eleven casecontrol studies, involving 3,052 cases and 3,553 controls, were identified on the association between the -592A>C polymorphism and BD risk in this meta-analysis, and these are shown in Table 3 3.3. IL-10 Haplotype and BD Risk. Among the selected studies, only three studies provided enough data on the association between the IL-10 -1082A>G, -819T>C, and -592A>C haplotypes and BD risk. Three well-known haplotypes of ATA, ACC, and GCC were found and included in this meta-analysis. The haplotype frequency data were examined, and the quantitative synthesis did not demonstrate any significant associations when the ACC and GCC haplotypes were  Table 4).

Discussion
To date, the pathogenesis and etiology of BD have not been elucidated. It is speculated that BD has characteristics of both inflammatory and autoimmune diseases and has a certain genetic susceptibility. Patients with BD have always suffered from serious complications, such as recurrent oral ulcers, genital ulcers, and uveitis. In addition, these microvascular inflammatory injuries are frequently involved in the gastrointestinal tract, central nervous system, large blood vessels, and limb joints in BD patients [12,13].
The IL-10 gene is located on chromosome 1 at 1q31-32, occupies more than 4.7 kb of genomic DNA, and consists of five exons and four introns. These are the three most studied genetic variants: -1082A>G (rs1800896), -819T>C (rs1800871), and -592A>C (rs1800872). These polymorphisms and their haplotypes are associated with abnormal circulating levels of IL-10 and contribute to the pathophysiology of various autoimmune and inflammatory diseases. During the triggering and development of BD, many inflammatory markers have been proposed to be involved in the development of BD. IL-10 is one of the most important immunoregulatory cytokines and is mainly secreted by macrophage cells and other T helper 1 (Th1) and Th2 cells [38]. IL-10 participates in complex inflammatory and immune  Journal of Immunology Research    9 Journal of Immunology Research processes by promoting the widespread suppression of immune effects and the abnormal circulating level of IL-10 which has been associated with an increased risk for the development of BD.
In 2007, Wallace et al. published the first case-control study of the UK population and suggested that subjects with the -1082AA genotype were weakly associated with BD risk in the general population (OR = 1:4, 95%CI = 1:1 -1:9, P = 0:04), and the -819T allele also indicated an increased BD risk (OR = 1:5, 95%CI = 1:1-2:0, P = 0:02). In subsequent years, an increasing number of casecontrol studies have been conducted to explore the association between IL-10 polymorphisms and BD susceptibility; however, the results were inconsistent and confusing. In 2013, Liang et al. published a meta-analysis on IL-10 gene polymorphisms and BD susceptibility and suggested that the IL-10 -1082A>G, -819T>C, and -592A>C polymorphisms were indeed associated with BD susceptibility [17]. However, the results were only based on two, five, and four case-controls on the IL-10 -1082A>G, -819T>C, and -592A>C loci, respectively, and the sample sizes were small as well. In 2015, Jung et al. conducted a meta-analysis on IL-10 polymorphisms and vasculitis susceptibility and reported a significant association between the IL-10 -819T>C and -592A>C polymorphisms and BD susceptibility in the stratified analysis based on only one genetic model (allele contrast). In addition, the two previous meta-analyses did not report any influence from subgroup factors, such as HWE status, ethnic   11 Journal of Immunology Research differences, and the source of control groups. In 2019, two latest meta-analyses were published by Shahriyari et al. and Lee and Song. The two publications had collected all current case-control studies and made an updated conclusion on the association between IL-10 polymorphisms and BD risk [39,40].
Today, we comprehensively summarized the current evidence and conduct this meta-analysis to explore the precise relationships between IL-10 polymorphisms and BD susceptibility with more stratified analysis. Overall, the pooled results demonstrated a significant association between the -819T>C polymorphism and BD susceptibility in all five genetic models, suggesting that the -819T>C polymorphism may decrease the risk of BD from 48% to 28%. Similarly, decreased BD risk was also observed in the subgroup analysis of the ethnic differences, control design, and genotyping method groups. In the -592A>C polymorphism, the pooled analysis of all five genetic models demonstrated that the mutation of -592A>C lowered the risk of developing BD from 48 to 29%. Moreover, similar results were also observed in many genetic models in the subgroup analysis of HWE status, ethnic differences, control design, and genotyping method groups. Regarding the -1082A>G polymorphism, no significant association was observed for BD risk, except for a few scattered cases of decreased BD risk in the G allele and the AG genotype in the EA population.   Previous studies had demonstrated that these three polymorphisms were proposed to affect the transcription efficiency and expression of the IL-10 gene [41]. More than 70% of the individual variability in human IL-10 production has been attributed to genetic variation, which could help to predict patient susceptibility to the disease as well as the severity and progression of the disease [42][43][44]. For the -1082A>G polymorphism, the A allele demonstrated a higher binding affinity to the transcription factor PU.1, resulting in decreased transcriptional activity and lower expression level of IL-10 [45]. In addition, the -819T>C genotype is also a functional polymorphism that could modify NF binding and result in higher levels of IL-10 production with the -819 CC genotype than with the -819 TC and TT    14 Journal of Immunology Research genotypes [46]. Furthermore, the functional analysis indicated that the nucleotide mutation from cytosine (C) to adenine (A) would reduce the inhibitory effect of the transcription factor Sp1/Sp3 complex and increase the expression of IL-10 [47]. Regarding haplotype, no significant difference among the main haplotypes was observed. Due to the small sample sizes and the limited number of studies, this result requires further confirmation before being taken as fact.
To the best of our knowledge, this report describes the most comprehensive meta-analysis on the issue to date. Some advances were demonstrated in this meta-analysis. First, most case-control studies published on the main three polymorphisms were considered, which would increase the statistical power and help to increase understanding of the association between IL-10 gene polymorphisms and BD risk. Second, a stratified analysis based on HWE status, ethnic differences, control design, and genotyping methods was conducted to explore their potential relationships. Third, a more scientific retrieval strategy and statistical methodology were used, including cumulative analyses, sensitivity analyses, and haplotype analyses.
Of course, there were still some inherent limitations that should be addressed in this meta-analysis. First, there were only 7, 12, and 7 case-control studies on IL-10 -1082A>G, -819T>C, and -592A>C loci, respectively, with complete information on the wild homozygotes, heterozygotes, and mutant homozygotes; the others only had the allele data in this meta-analysis. Therefore, the synthetic results and subgroup analysis could not reveal the real association between IL-10 gene polymorphisms and BD susceptibility with limited sample size. Second, most subjects came from the ME, EA, and Europe; therefore, the results of this systematic review and meta-analysis cannot represent all ethnic populations, and the conclusions are only applicable to these ethnicities. Third, other potential factors, such as living habits, mental stress, age, and immunological disarrangement, were not assessed, as sufficient data were not available. Fourth, this meta-analysis is based on published studies; therefore, potential publication bias may be present, even if the tests utilized in this meta-analysis did not find it. Fifth, moderate or high heterogeneity could be observed in these polymorphism sites. To the best of our knowledge, clinical heterogeneity (sometimes called clinical diversity), methodological heterogeneity (sometimes called methodological diversity), and statistical heterogeneity were the three most common sources. In this meta-analysis, the severity of disease, control design, personal habits, and ethnic differences may be the potential factors that contribute to the existing heterogeneity. These above factors would lead to a deviation in the authenticity and reliability of our results. Fortunately, the heterogeneity was partly alleviated in the stratified analysis based on ethnic differences and control design. In terms of the former, the heterogeneity has obviously been relieved in the subgroup analysis of the Middle East populations in -819T>C and -592A>C polymorphism loci. Given the control design, the degree of heterogeneity has decreased in these studies with healthy-based controls in some genetic models of the -1082A>G and -819T>C polymorphisms.

Conclusions
The current evidence indicates that there is a potential association between IL-10 gene polymorphisms and BD susceptibility, especially the -819T>C and -592A>C polymorphisms. In future studies, more case-control studies with larger sample sizes and different ethnicities are needed to verify these conclusions.

Data Availability
The raw data supporting this meta-analysis are from previously reported studies and datasets. All included data are also available from the corresponding author upon request.

Disclosure
The funders had no roles in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.