Association Of Interleukin-18Polymorphisms With Systemic Lupus Erythematosus

Systemic Lupus Erythematosus(SLE) is a chronic autoimmunity disease resulting tissue damage, and both environment and gene play important roles in the pathogenesis of SLE. Meta-analysis was used to initially determine the association between interleukin-18(IL-18)polymorphisms and the risk of SLE, then we applied case-control study to detect the association of IL-18polymorphisms and the gene-environment interaction with SLE among Chinese Han in South of the Yangtze by the method of restriction fragment length polymorphism (RFLP). Furthermore, we used case-only study to explore the association of IL-18gene polymorphisms with SLE phenotypes. At last, meta-analysis was applied again to detect the association after adding our research.The results are as follows:Part I Meta-analysis of interleukin-18polymorphism and Systemic LupusErythematosus1. There was significant association between IL-18-1297C/T polymorphism and SLE in European origin populations. With A genotype as a reference, people with C allele had a higher risk of SLE (OR=1.388,95%CI=1.242-1.551, P<0.001). Moreover, in European origin populations there was significant difference between SLE patients and healthy controls in the genotype of-1297CC+CT versus TT (OR=1.353,95%CI=1.174-1.561, P<0.001), CC versus CT+TT (OR=2.216,95%CI=1.695-2.898, P<0.001).2. There was significant association between IL-18-607A/C polymorphism and SLE in European origin populations. With C genotype as a reference, people with A allele had a higher risk of SLE (OR=1.158,95%CI=1.014-1.322, P=0.031). However,-607A allele decreased SLE risk in Chinese population (OR=0.606,95%CI=0.396-0.930, P0.001).3. No significant association was found between IL-18-137C/G polymorphism and SLE. There was no significant association between-137C allele and the risk of SLE (OR=1.122,95%CI=0.997-1.264, P=0.057), as well as under the dominant(OR=1.114,95%CI=0.957-1.295, P=0.163) and recessive model (OR=1.521,95%CI=0.856-2.702, P=0.152). 4. No significant association was found between IL-18105A/C polymorphism and SLE. There was no significant association between105A allele and the risk of SLE (OR=1367,95%CI=0.680-2.746, P<0.380).Part II Study on the main effect and interaction effect of Single Nucleotide Polymorphisms(SNPs) and environmental factors for SLEAssociation of interleukin-18Single Nucleotide Polymorphisms (SNPs) with Systemic Lupus Erythematosus1. The univariate logistic regression failed to find any significant association between IL-18-1297C/T or-607A/C polymorphism and SLE (P>0.05).2. The univariate logistic regression showed that there was a significant difference in the allelic distribution of105A/C in IL-18between case and control (P=0.010), the A allele might be a risk factor for SLE, relative to C allele (OR=1.815,95%CI=1.151-2.863). Under the additive and recessive model105A/C polymorphism was associated with SLE (Additive:OR=1.828,95%CI=1.152-2.900; Recessive:OR=1.855,95%CI=1.131-3.044), but under the dominant model the association did not exist. Therefore, with C allele or CC genotype or CC/CT genotype as a reference, people with A allele or AC genotype or AA genotype had a higher risk of SLE.3. The Multiple logistic regression showed that there was no significant association between IL-18-1297C/T or-607A/C polymorphism and SLE after adjusting covariant,like age and gender in all genetic models(P>0.05).4. After adjusting covariant, like age and gender, the Multiple logistic regression showed that the association of IL-18105A/C with SLE still existed (additive model: OR=1.987,95%CI=1.181-3.343; recessive model:OR=1.960,95%CI=1.111-3.456).The association of haplotype with SLE1. Significant linkage disequilibrium was found between-1297C/T and-607A/C,-607A/C and105A/C,-1297C/T and105A/C in IL-18gene (D’=0.821, r~2=0.078, P<0.001; D’=0.608,r~2=0.042, P<0.001; D’=0.654,r~2=0.423, P<0.001).2. HAPSTAT3.1software was applied to analyze the association of haplotype with SLE under all genetic models. The haplotypes T-C-A and T-A-A appeared to be significant’at-risk’haplotypes (OR=2.612, P=0.003; OR=2.575, P=0.004) under additive model. With no T-C-A haplotypes as a reference, people with T-C-A haplotypes had a higher risk of SLE, as well as people with T-A-A haplotypes. However, the association was not existed under the dominant and recessive models.C-A-C haplotypes were not found to be associated with SLE under all genetic models (P>0.05). The additive model was the optimal model according to the principle of AIC.3. HAPSTAT3.1software was applied to discuss interaction of haplotype-environment under all genetic models. The result showed that no significant interaction was found under all genetic models.Study on interaction between gene and environment for SLE.1. Log-linear model was applied to explore the interaction of IL-18gene(-1297T/C、-607A/C or105A/C)and environment (chilblain, humid environment or ultraviolet light) for SLE, no significant interaction was found under all genetic models.2. Logistic regression was applied to explore the interaction of IL-18gene(-1297T/C、-607A/C or105A/C)and environment (chilblain, humid environment or ultraviolet light) for SLE, no significant interaction was found under all genetic models.Part III IL-18polymorphisms and clinical features of SLE1.X~2test was used to analyze the association of IL-18gene-1297T/C polymorphism with clinical features of SLE(organ damage and immune index). The results showed that no significant association was found.2. X~2test was used to analyze the association of IL-18gene-607A/C polymorphism with clinical features of SLE(organ damage and immune index). The results showed that no significant association was found between the clinical features profile of SLE patients with various alleles and genotypes.3. X~2test was used to analyze the association of IL-18gene105A/C polymorphism with clinical features of SLE(organ damage and immune index). The results showed that no significant association was found between105A/C polymorphism and organ damage of SLE, but significant association was detected between105A/C polymorphism and immune index of SLE, SLE patient with higher A allele frequency had a higher risk of lower complement C3and C4(OR=1.737,95%CI=1.001-3.014, P=0.048).4. As for IL-18gene haplotype association study with organ damage, there was no significance in any other haplotypes except haplotypes T-C-A and haplotypes T-A-A. The haplotype frequency of T-C-A (OR=4.681,95%CI=1.076-20.371, P=0.040) and T-A-A (OR=5.295,95%CI=1.215-23.071, P=0.026) were significantly higher in patients with central nervous system disorder under additive model, but we failed to detect the association under dominant and recessive models. According to the principle of AIC, the additive model was the optimal model. Haplotypes T-A-A were observed to have a higher frequency in patients with serosities under recessive model (OR=2.S63,95%CI=1.090-7.522, P=0.033), but under the other two models this association did not exist. According to the principle of AIC, the recessive model was the optimal model.5. As for IL-18gene haplotype association study with immune index, there was no significance in any other haplotypes except haplotypes T-A-A. There was significant association between haplotypes T-A-A and patients with anti-SSA under recessive model (OR=1.920,95%CI=1.020-3.614, P=0.043), but we failed to detect the association under additive and dominant models, According to the principle of AIC, the additive model was the optimal model.Part IV Meta-analysis of interleukin-18polymorphism and Systemic LupusErythematosus(the second time)1. There was significant association between IL-18-1297C/T polymorphism and SLE, and the association was detected in European origin populations, but not in Asians. With T genotype as a reference, people with C allele had a higher risk of SLE(OR=1.261,95%CI=1.031-1.543, P=0.024); With no CC genotype as a reference, people with CC genotype had a higher risk of SLE (OR=2.046,95%CI=1.439-2.911, P<0.001). Moreover, in European origin populations, with T genotype as a reference, people with C allele had a higher risk of SLE(OR=1.388,95%CI=1.242-1.551, P<0.001), and with TT/CT genotype or TT genotype as a reference, people with CC genotype or CC/CT genotype had a higher risk of SLE(CC+CT versus TT: OR=1.353,95%CI=1.174-1.561, P<0.001; CC versus CT+TT:OR=2.216,95%CI=1.695-2.898, P<0.001).2. There was significant association between IL-18-607A/C polymorphism and SLE in European origin populations. With C genotype as a reference, people with A allele had a higher risk of SLE (OR=1.158,95%CI=1.014-1.322, P=0.031). However,-607A allele decreased SLE risk in Chinese population (OR=0.659,95%CI=0.458-0.947, P=0.024), and significant association was also detected under reccesive model (OR=0.792,95%CI=0.639-0.981, P=0.033).3. There was significant association between IL-18105A/C polymorphism and SLE in Asian origin populations. With C genotype as a reference, people with A allele had a higher risk of SLE (OR=1.894,95%CI=1.376-2.606, P<0.001); With no AA genotype as a reference, people with AA genotype had a higher risk of SLE (OR=2.084,95%CI=1.448-2.999, P<0.001).