【作者】 韩建文；

【导师】 张学军； 杨森；

【作者基本信息】 安徽医科大学 ， 皮肤病与性病学， 2010， 博士

【摘要】 引言:系统性红斑狼疮(Systemic lupus erythematosus, SLE)是一种经典的自身免疫病,以多种自身抗体产生、补体激活、免疫复合物沉积,进而导致组织和器官的损伤为基本特征。SLE的病因包含遗传和环境两个方面的因素。SLE好发于女性,特别是育龄期妇女,男女的患病率之比约是1:9。SLE的发病率和患病率在世界不同人种、地区间有很大的差异。中国的患病率约为31-70/100000,欧洲人群的患病率约为7-71/100000。不同人种间的SLE临床表现也存在一定差异,在中国人群的SLE患者比欧洲SLE患者更容易发生狼疮肾炎。提示遗传异质性在SLE的临床表现的复杂性方面可能也起着重要的作用。在过去的20年多中,通过遗传连锁分析和关联分析等研究方法确定了许多SLE的易感基因和位点。特别是2008年完成的4项欧洲人群的SLE全基因组关联分析研究(genome-wide association studies, GWAS)确定了20多个SLE的易感基因/位点。流行病学研究以及遗传学研究提示种族差异和遗传异质性的在SLE的易感性、疾病的发生、发展以及引起复杂的临床表现方面中起到重要的作用。但是目前大部分连锁分析、关联分析研究以及所有的GWAS研究都是在欧洲人群中进行的。由于遗传异质性的原因,在中国汉族人群中进行SLE的GWAS研究具有重要的意义。目的:在中国汉族人群,利用全基因组关联分析研究方法在全基因组范围内搜寻与SLE关联的遗传变异,鉴定中国汉族人SLE的易感基因/位点。方法:按照相同的样本收集标准,从全国各地的医院收集SLE患者和对照样本,以地域来源进行匹配,分成3个独立的人群:初筛样本(1099例SLE患者和1254例正常对照,来自中国中部)、验证样本1(1643例SLE患者和5930例正常对照,来自中国中部)和验证样本2(1509例SLE患者和1120例正常对照,来自中国南部)。首先利用Illumina Human 610-Quard BeadChips基因芯片对初筛样本进行全基因组tag SNPs基因分型。经过质控后,进行全基因组关联分析筛选与SLE关联的SNPs,随后在另外两个独立的病例-对照验证人群中进行验证。将既往欧洲人群SLE的GWAS结果与本研究初筛阶段的统计分析结果进行比较。结果:全基因组关联分析结果:在初筛阶段的数据统计分析结果显示3个位点与SLE显著关联,达到全基因组的显著性水平(P<5×10-8),分别是6p21(MHC), 2q32.3(STAT4)和8p23.1(BLK)。在MHC区域,有13个位于HLAⅡ类区域的SNPs的关联性达到全基因组的显著性水平(P<5×10-8)。关联性最为显著的SNP是rs9271100(P=1.42×10-12,OR=1.9),对13个SNPs进行条件回归分析发现它们代表2个关联信号:一个是rs9271100,靠近HLA-DRB1;另外一个是HLA-DQA2附近的rs3997854(P=2.85×10-8,OR= 0.44)。根据初筛结果,选择78个非MHC区域的SNPs(来自67个位点)在另外两个验证人群中进行分析,发现位于16个位点的21个SNPs在两个验证人群中都显示与SLE的关联性(P<0.03),并且在合并三个独立人群的数据分析后显示显著的关联(Pcombined<5×10-8),16个位点分别是:1q25.1, 2p22.3, 2q32.3, 5q33.1, 6q21, 6q23.3, 7p12.2, 7q11.23, 7q32.1, 8p23.1, 10q11.22, 11q23.3, 11q24.3, 12q24.32, 16p11.2和22q11.21(5.17×10?42≤Pcombined≤2.77×10?8)。控制MHC的2个SNPs(rs9271100和rs3997854)后,这21个SNPs的关联性仍然存在。其中1q25.1, 2q32.3, 6q21, 6q23.3, 7q32.1, 8p23.1和22q11.21七个位点与SLE的关联性在既往欧洲人群中的GWAS已经有报道,这些易感位点中包含易感基因分别是TNFSF4, STAT4 , PRDM1-ATG5, TNFAIP3, IRF5, BLK和HIC2-UBE2L3。深入分析新发现的9个关联信号,对每个关联SNP周围区域的基因组重组率以及每个连锁不平衡区域内存在的基因进行分析。在2p22.3(RASGRP3), 7p12.2(IKZF1), 11q24.3(ETS1)和12q24.32(SLC15A4)四个易感位点的LD区域内仅包含一个已知的基因。在易感位点5q33.1的LD区域内,包含2个已知的基因(TNIP1和GPX3),但是从基因功能角度考虑,由于TNIP1在细胞激活、细胞因子信号转导及凋亡过程发挥着重要的作用,更有可能是SLE的易感基因。新发现的5个易感基因在功能学可能也与SLE发病过程中的病理机制相关,比如:SLC15A4可能参与抗原呈递和免疫复合物的处理;TNIP1可能参与Toll样受体功能及1型干扰素的产生;ETS1, RASGRP3和IKZF1可能参与淋巴细胞中免疫信号的转导。在另外4个易感位点(7q11.23, 10q11.22, 11q23.3和16p11.2)的LD区域内包含多个基因,很难确定哪一个是SLE的真正易感基因,还有待于更加深入、全面的精细定位分析来鉴定位于这些区域内的SLE易感基因。值得一提的是,位于16p11.2易感位点距离欧洲人群中发现的SLE易感基因ITGAM-ITGAX非常接近。但是本研究发现的关联信号与欧洲人群中发现的关联信号不在同一个LD区域内。另外,欧洲人群报道的5个SLE关联SNP在中国汉族人群中的等位基因频率都很低(MAF<0.01),这一区域与SLE的易感性关系复杂,有待于进一步多种族、大样本的研究来阐明其与SLE的具体关系。通过分析几项欧洲人群的SLE的GWAS研究报道的48个SNPs(来自22个位点),并对两个人群结果进行比较。可以把所研究的48个SNPs分成4组:第一组是在两个人群都提示有关联性的13个SNPs。其余35个SNPs与SLE的关联性在本项研究中未被发现,根据等位基因频率可以分成3组。第二组的20个SNPs在中国人群中的等位基因频率非常低(MAF<5%),而在欧洲人群中是常见SNPs(MAF>9%),特别是其中有14个SNPs在中国人群中几乎呈单态分布。第三组的4个SNPs在在中国人群中的等位基因频率较低(5%<MAF<10%),而在欧洲人群中是常见SNPs(MAF>10%)。第四组的11个SNPs在两个人群中的等位基因频率大体一致,均为常见SNPs(MAF>10%)。通过查询既往一项GWAS结果数据库,本研究发现的SLE关联SNPs中有23个SNPs(包括13个位于MHC的SNPs)存在于他们的数据库中。只有位于TNIP1基因内的rs10036748在欧洲人群GWAS中提示关联性,它在两个人群中的等位基因频率明显相反(欧洲人群77%,中国汉族人群26%)。结论:本项研究在中国汉族人群进行了大样本量SLE的GWAS研究,在HLA区域的HLA-DRB1和HLA-DQA2附近存在两个独立的SLE关联信号。发现9个新的SLE易感基因/位点,并且验证出既往欧洲人群报道的7个易感基因/位点。另外通过与欧洲人群SLE的GWAS研究结果进行比较,发现影响SLE易感性的遗传变异在不同人种间存在差异,遗传异质性可能在SLE的易感性方面起着重要作用。该项研究将加深我们对SLE遗传基础的认识,为将来SLE机制研究提供启示。更多还原

【Abstract】 Introduction: Systemic lupus erythematosus (SLE) is a prototypic systemic autoimmune disease, characterized by a diverse array of autoantibody production, complement activation, immune complex deposition, and tissue and organ damage and influenced by both genetic and environmental factors. SLE affects predominantly in women (prevalence ratio of women to men is 9 to 1) and particularly during childbearing years. There are marked disparities in SLE incidence and prevalence worldwide, which varies in different ethnic and geographical populations. The prevalence of SLE ranges from 31 to 70 cases per 100,000 persons among Chinese and 7 to 71 in European populations, and lupus nephritis is more prevalent in Chinese than in Europeans. The ethnic and genetic heterogeneity may contribute to the complexity of its clinical manifestation.Over past two decades, numerous studies were performed and identified multiple genetic factors related to SLE by linkage and association study. In particular, recent four genome-wide association studies (GWAS) of SLE in European populations have identified more than twenty robust susceptibility genes and/or loci. Epidemiological and genetic studies the genetic heterogeneity between ethnic populations has been suggested to play an important role in the susceptibility, development and complexity of clinical manifestation of SLE. Up to today, most linkage, association studies and all GWAS for SLE were performed in European population, which called for GWAS in Chinese populations considering the genetic heterogeneity in SLE.Object: To screen SLE associated SNPs in whole genome and indetify susceptibility genes/loci for SLE in Chinese Han population by using GWAS approach. Methods: All samples in this study were recruited from the Chinese Han population through collaboration with multiple hospitals, which were matched by geographic regions. All cases and controls were separated into three independent samples, in initial stage (1,099 SLE cases and 1,254 controls) were recruited from central China, samples in replication studies were recruited from the central (Replication 1: 1,643 cases and 5,930 controls) and southern (Replication 2: 1,509 cases and 1,120 controls) China. Samples in the initial stage were genotyped by Illumina Human 610-Quard BeadChips. After quality controls, genome wide association analysis were performed to screen SLE associated SNPs, which were replicated in other two case-control samples. The results of previous SLE GWAS in European population were compared with our GWAS.Results: The genome wide association analysis revealed association at three loci: 6p21 (MHC), 2q32.3 (STAT4) and 8p23.1 (BLK) with genome-wide significance (P<5×10-8). In the MHC region, 13 SNPs, all located within the HLA class II region, showed genome-wide significant association (P<5×10-8), and the most significant association was identified at rs9271100 (P=1.42×10?12, OR=1.9). Further conditional association analysis of the 13 SNPs confirmed that there were two independent associations within the region at rs9271100 next to HLA-DRB1 and rs3997854 next to HLA-DQA2(P=2.85×10-8, OR= 0.44).We performed a replication study by genotyping 78 non-MHC SNPs (from 67 loci) in two additional cohorts of Chinese Han (Replication 1 and Replication 2). Twenty-one SNPs within 16 loci were validated with independent supporting evidence for association (P<0.03) from the two replication samples and highly significant evidence in the combined sample that surpassed genome-wide significance (Pcombined<5×10-8). The 16 confirmed susceptibility loci were located at 1q25.1, 2p22.3, 2q32.3, 5q33.1, 6q21, 6q23.3, 7p12.2, 7q11.23, 7q32.1, 8p23.1, 10q11.22, 11q23.3, 11q24.3, 12q24.32, 16p11.2 and 22q11.21 (5.17×10-42≤Pcombined≤2.77×10-8). The associations at the 16 loci were independent of the association within the MHC region, because the associations at these loci remained similar after controlling the genetic effect of rs9271100 and rs3997854 within the MHC region. The associations at 1q25.1, 2q32.3, 6q21, 6q23.3, 7q32.1, 8p23.1 and 22q11.21 were reported by previous GWAS in European populations, implicating susceptibility genes TNFSF4, STAT4, PRDM1-ATG5, TNFAIP3, IRF5,BLK and HIC2-UBE2L3 for SLE, respectively.To identify susceptibility gene underlying each of the 9 novel associations, we investigated the patterns of the recombination and LD around the risk-associated SNPs and the gene(s) located within each region harboring the association. Only one gene was implicated by each of the associations at 2p22.3 (RASGRP3), 7p12.2 (IKZF1), 11q24.3 (ETS1), 12q24.32 (SLC15A4), where a single gene was found within the LD block harboring the association. At 5q33.1, two genes, TNIP1 and GPX3, were found within this locus, but TNIP1 is more plausible as susceptibility gene, due to its important role in modulating cell activation, cytokine signaling and apoptosis. These five novel susceptibility genes provide further supporting evidence for their biological function, such as: immune complex processing (SLC15A4); Toll-like receptor function and type I interferon production (TNIP1); and immune signal transduction in lymphocytes (ETS1, RASGRP3 and IKZF1). For the association at 7q11.23, 10q11.22, 11q23.3 and 16p11.2, there were more than one gene or transcript within the each region. Further fine mapping analysis is required to determine the susceptibility genes underlying these novel association loci for SLE. Interestingly, the association at 16p11.2 located close to the susceptibility locus of ITGAM-ITGAX identified in European population. The association identified in our study with a LD pattern that was different from the one where ITGAM and ITGAX located. Furthermore, the MAFs of 5 reported SNPs within ITGAM-ITGAX locus were very low (<0.01) in Chinese Han population. Further study will be needed to confirm the susceptibility gene within this region.The association evidence for 48 SLE-associated SNPs (from 22 loci) identified by the previous GWAS in European populations was also investigated in our GWAS dataset. These SNPs were summarized into four groups. Group 1 included 13 SNPs confirmed by our GWAS. However, our study did not provide evidence for the associations at the remaining 35 SNPs, which were divided into three groups based on their allele frequencies. Group 2 included 20 SNPs are very rare in Chinese Han population (MAF<5%), especially 14 of them are almost monomorphism. However, all these 20 SNPs are common SNPs in Europeans. Group 3 included 4 SNPs with low population frequency in Chinese Han population (5%<MAF<10%), which are common SNPs in Europeans. Group 4 included the remaining 11 SNPs have a similar MAF in both the Chinese Han and European populations, but failed to show any associations in Chinese Han population. None of the top 13 MHC SNPs identified in our GWAS analysis showed significant association in a GWAS of European population. Furthermore, of the 10 novel non-MHC SNPs identified by this study, only rs10036748 within TNIP1 showed significant association (P=10-4~10-3) in European population, but a much higher frequency (77%) in Europeans than in Chinese (26%)Conclusion: This study was the first large scale GWAS of SLE in Asian population (Chinese Han) and identified 9 novel susceptibility genes/loci as well as additional 8 previously reported ones for SLE. By compared with previous GWAS in Europeans, genetic heterogeneity was suggested playing an important role in the disease susceptibility. This study not only advanced our understanding on the genetic basis of SLE susceptibility, but also provided enlightenment for future research of SLE.更多还原