【作者】 盛宇俊；

【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， 2013， 博士

【摘要】 研究背景银屑病是一种慢性的多因素的皮肤疾病，该病以界限清楚的红色斑块附着银色的鳞片为特征。目前该病的病因及发病机制尚不清楚，炎症介质的异常产生在银屑病发病中发挥重要作用。多数学者认为银屑病的发生是遗传与环境相互作用的结果，其中遗传因素在银屑病的发病中起到重要作用，并且该病是由多个易感基因共同参与的复杂性疾病。银屑病的患病率在世界各地差异较大，与种族、地理位置、环境等因素有关。总体上，高加索人群的患病率约为2-3％，而日本人的患病率较低且绝大多数银屑病患者为散发。我国银屑病的患病率约为0.123％，总患病人数接近1,000万例，分布趋势表现为男性患病率高于女性，北方高于南方，城市高于农村。国内1043例寻常型银屑病患者及其家族成员的遗传流行病学调查，发现一、二级亲属的发病率明显高于一般人群，一、二级亲属的遗传度分别为67.04%和46.59%，且随亲缘系数的增加呈现降低的趋势。随着遗传学研究技术的发展和研究方法的更新，国内外学者在银屑病遗传学领域研究方面作出了大量的工作，发现了一批有价值的疾病易感基因/区域。“定位克隆”和“定位候选克隆”的方法是研究孟德尔遗传模式疾病遗传学基础、发现疾病致病基因的主要方法。在上个世纪末，科学家们尝试着把遗传连锁分析研究的方法应用到银屑病的易感基因研究中。目前，通过连锁分析方法运用定位克隆和全基因组扫描的方法共计在全基因组范围内发现了10个银屑病易感位点：1q21、1p、3q21、4q、4q31-q34、6p21.33、16q、17q25、18p11.23和19p13。随着分型技术的迅速发展和分型成本的不断降低、覆盖整个人类基因组的高通量商业化SNP检测芯片的出现和不断更新，全基因组关联研究（Genome Wide Association Study，GWAS）得到发展。国内外研究团队通过大规模GWAS发现了超过40个银屑病易感基因或位点。然而，目前发现的风险变异大多数可能仅是标签单核苷酸多态性（SNP），这些易感基因的功能性编码变异，尤其是低频和罕见的变异在GWAS研究中被忽略，因此没有进行系统性研究。目的本研究主要是评估功能性编码变异在整个银屑病遗传度上的贡献，并试图发现银屑病的其他关联信号。方法我们在两个独立中国汉族人群（包括21,309例样本）中对功能性编码变异进行大规模测序分析研究。首先，我们对781例患者和676例对照样本的全基因组外显子测序数据进行全基因非同义单核苷酸变异（SNVs）分析，并在9,946例患者和9,906例对照的1,326个候选基因的靶向测序数据进行验证分析。对于非编码区域SNVs，我们进行深入的数据分析并在大规模测序数据的基础上对22个提示SNVs在另外一个独立中国汉族人群（4,480例患者和6,521例对照）中进行验证实验，以发现银屑病的其他关联信号。结果对于非同义SNVs，我们发现6个常见和低频SNVs在两个独立的测序样本（全基因组外显子测序和靶向测序）中关联性一致，并在联合分析中达到全基因组显著水平，包括IL23R基因上的chr1:67421184（c.G530A）(P_exome-target=1.94×10^-11，OR=0.72)、GJB2基因上的rs72474224（c.G324A）(P_exome-target=7.46×10^-11，OR=1.34)、LCE3D基因上的rs512208（c.C185A）(P_exome-target=2.92×10^-23，OR=1.24)、ERAP1基因上的rs27044（c.C2535G）(P_exome-target=2.16×10^-14，OR=0.86)和rs26653（c.G727C）(P_exome-target=5.27×10^-12，OR=0.87)以及ZNF816A基因上的rs12459008（c.T590A）(P_exome-target=2.25×10^-9，OR=0.88)。在FUT2基因上，我们发现一个常见错义变异rs1047781（c.A418T）(P_exome-target=2.78×10^-6，OR=1.10)和两个罕见错义变异（chr19:53898296/c.C271T和chr19:53898541/c.C516G）(P_exome-target=5.51×10-5，OR=3.61和P_exome-target=2.92×10-3,OR=6.92)具有提示相关性。对于非编码SNVs，我们验证了既往报道的4个银屑病易感基因，包括IL12B(rs2288831，P_exome-target=2.30×10-20，OR=0.83)、IFIH1(rs13431841，P_exome-target=2.96×10^-09，OR=0.83)、ERAP1(rs27043，P_exome-target=6.50×10^-12，OR=0.87)和RNF114(rs4647954，Pcombined=2.41×10^-07，OR=1.09)，并发现了3个新的易感位点达到全基因组显著水平，包括NFKB1(rs1020760，P_exome-target=2.19×10^-08，OR=1.12；rs1609798，P_exome-target=9.87×10^-08，OR=1.12)，12p13.3(rs758739，Pcombined=4.08×10^-08，OR=0.91；rs2243750，Pcombined=4.38×10^-08，OR=0.91)和17q12(rs10852936，Pcombined=1.96×10^-08，OR=1.10；rs12936231，Pcombined=5.02×10^-08，OR=1.10)。此外，我们还发现了两个位点具有提示意义，分别是3p21.31(rs1863837，Pcombined=3.91×10^-07，OR=0.90)和17q25(rs3744017，Pcombined=5.83×10^-07，OR=1.12)。结论本研究结果提示编码区变异，至少是低频或者罕见非同变异对银屑病的整体遗传度贡献非常有限，此外，本研究增加了银屑病确定的风险位点数量。更多还原

【Abstract】 Background Psoriasis is a chronic and multifactorial skin diseasecharacterized by sharply demarcated erythematous plaques with adherentsilvery scales. Currently, the etiology and pathogenesis of psoriasis arenot fully understood, abnormal production of inflammatory mediators playan important role in the pathogenesis of this disease. Most scholars believethat psoriasis is the result of genetic and environmental interactions,and genetic factors are essential to psoriasis risk. Furthermore, psoriasisis one of the complex diseases caused by a number of common susceptibilitygene. The prevalence of psoriasis varies between different ethnic andgeographical populations, and ethnicity, geographic location, environmentand other factors are relative to the difference. Overall, the prevalenceof the Caucasian population is about2-3%, while low rate is observed inJapanese with the majority of patients with psoriasis are sporadic. In China,the prevalence of psoriasis is approximately0.123%with the total numberof patients close to10million, the distribution trends showed that theprevalence of male is higher than that of women, the prevalence in northand urban areas is higher than that in south and raral areas. Data for1043patients with psoriasis vulgaris suggests that the prevalence of psoriasisin first-and second-degree relatives of the proband with psoriasis was higher than that in the general population and the heritability of psoriasisin first-and second-degree relatives was67.04and46.59%, respectively.With the development of genetics research technologies and research methods,great progress have been made and a number of valuable diseasesusceptibility genes/regions have been identified."Positional cloning"and "positional candidate cloning" approachs are important methods usedto evaluate the genetic basis of Mendelian diseases and identify the causalgenes. In the end of last century, the genetic linkage analysis methodswere applied to the genetic susceptibility research of psoriasis. So far,10susceptibility loci were identified via the use of positional cloningand genome-wide linkage analysis, such as1q21,1p,3q21,4q,4q31-q34,6p21.33,16q,17q25,18p11.23and19p13. With the rapid development oftechnologies and genotyping costs continue to decrease, high-throughputcommercial SNP detection chip appears and constantly update, which makesgenome-wide association studies （GWAS） to develop. Currently, more than40susceptibility genes/loci have been identified through large-scaleassociation studies, particularly GWAS; however, most of the identifiedrisk variants are expected to be tagging SNPs, and the functional codingvariants of these susceptibility genes, particularly those that are of lowfrequency and rare, are largely refractory to the interrogation by GWASand have therefore not been systematically investigated. Objective This study aimed to investigate the contribution of functionalcoding variants and noncoding variants to the genetic susceptibility ofpsoriasis and identifiy additional association with the disease.Method We carried out a large-scale sequencing analysis of functionalcoding variants in two independent samples comprising21,309individualsof Chinese Han. First, we analyzed whole genome nonsynonymoussingle-nucleotide variants （SNVs） by exome sequencing in781cases and676controls, and following validation on1,326candidate genes by targetedsequencing in9,946cases and9,906controls of Chinese population. Forthe noncoding SNVs, we performed in-depth data analysis and a replicationstudy for22SNVs in an independent cohort of Chinese Han consisting of4,480cases and6,521controls based on the large-scale sequencing data（10,727cases and10,582controls） to find additional association with thedisease.Result For the nonsynonymous SNVs, we discovered6common and low frequencymissense SNVs that showed consistent association in the two independentsamples of exome and targeted sequencing analyses and achieved genome-widesignificance in combined analysis: chr1:67421184（c.G530A） at IL23R(P_exome-target=1.94×10^-11, OR=0.72), rs72474224（c.G324A） at GJB2(P_exome-target=7.46×10^-11, OR=1.34), rs512208（c.C185A） at LCE3D (P_exome-target=2.92×10^-23, OR=1.24), rs27044（c.C2535G） and rs26653（c.G727C） at ERAP1(P_exome-target =2.16×10^-14, OR=0.86and P_exome-target=5.27×10^-12, OR=0.87), and rs12459008（c.T590A） at ZNF816A (P_exome-target=2.25×10^-9, OR=0.88). One common missenseSNV rs1047781（c.A418T） and two rare missense SNVs （chr19:53898296/c.C271Tand chr19:53898541/c.C516G） in FUT2were also identified with suggestiveassociation evidence (P_exome-target=2.78×10^-6, OR=1.10, P_exome-target=5.51×10-5, OR=3.61and P_exome-target=2.92×10-3, OR=6.92, respectively). Forthe noncoding SNVs, we confirmed4previously reported psoriasissusceptibility genes: IL12B (rs2288831, P_exome-target=2.30×10-20, OR=0.83), IFIH1(rs13431841, P_exome-target=2.96×10^-09, OR=0.83), ERAP1(rs27043, P_exome-target=6.50×10^-12, OR=0.87) and RNF114(rs4647954, Pcombined=2.41×10^-07, OR=1.09),and identified3new susceptibility loci exceeded the genome-widesignificance threshold: NFKB1(rs1020760, P08exome-target=2.19×10-, OR=1.12;rs1609798, P_exome-target=9.87×10^-08, OR=1.12),12p13.3(rs758739, Pcombined=4.08×10^-08, OR=0.91; rs2243750, P08combined=4.38×10-, OR=0.91) and17q12(rs10852936, Pcombined=1.96×10^-08, OR=1.10, rs12936231, Pcombined=5.02×10^-08,OR=1.10). We also identified2additional loci exhibiting suggestiveevidence of an association:3p21.31（rs1863837, P7combined=3.91×10-0, OR=0.90） and17q25（rs3744017, P-07combined=5.83×10, OR=1.12）.Conclusion The results of this study indicated that coding variants, atleast nonsynonymous ones with low and rare frequency might have limitedcontribution to the overall genetic risk of psoriasis and increase thenumber of confirmed psoriasis risk loci.更多还原