基于全蛋白组孟德尔随机化方法探究皮肤癌的风险因素

Integration of Large-scale Plasma Proteome with Genomic Information for Identification of Causal Protein in Skin Cancer

目的利用全蛋白组孟德尔随机化(Proteome-wide MR)方法探究与皮肤癌(SC)风险相关的血浆蛋白。方法从8项已发表的血浆蛋白质组全基因组关联研究(GWAS)中获得了13265个蛋白质数量性状位点(pQTL),并在本研究中纳入了4972种独特的血浆蛋白。对包含23694例SC病例和372016例对照的GWAS数据集使用孟德尔随机化(MR)来研究血浆蛋白与SC之间的遗传因果关系。在另外2个GWAS分析中证实这些关联(分别包含25928例病例和466275例对照,以及16531例病例和344663例对照)。利用敏感性和基于汇总数据的MR(SMR)分析以验证因果关系。最后,进行了药物可行性评估,以确定SC的潜在治疗靶点的优先次序。结果共确定了6种与SC风险有因果关系的蛋白质(P<0.05/4972),其中,3种蛋白[二肽酶1(DPEP1)、谷胱甘肽合成酶(GSS)和Copine 1蛋白(CPNE1)]水平升高和3种蛋白[组织蛋白酶S(CTSS)、锚蛋白重复序列-细胞信号抑制因子盒蛋白9(ASB9)和调节蛋白WD重复域蛋白46(WDR46)]水平降低与SC风险增加有关。4种有可用pQTL文件的蛋白质(DPEP1、GSS、CTSS和CPNE1)和3种有表达量性状位点(eQTL)的蛋白质(GSS、CTSS和CPNE1)通过SMR测试。此外,4种蛋白质(DPEP1、GSS、CTSS和CPNE1)有药物可行性。结论确定了与SC风险相关的6种蛋白质生物标志物,为SC的病因和潜在靶点提供了新的见解,可进一步改善基因组和蛋白质组数据的整合,以用于药物开发。

Objective In order to investigate the plasma proteins associated with skin cancer(SC)risk using proteome-wide Mendelian randomization(proteome-wide MR).Methods Totally 13265 protein quantitative trait loci(pQTLs)were derived from eight published genome-wide association studies(GWASs)on plasma proteome and 4972 unique plasma proteins were included in this study.Mendelian randomization(MR)on a GWAS dataset was conducted consisting of 23694 SC cases and 372016 controls to investigate the genetic causal associations between plasma proteins and SC.Two additional GWAS analyses were subsequently used to confirm these associations,one with 25928 cases and 466275 controls,and the other with 16531 cases and 344663 controls.To verify causality,sensitivity and SMR analysis were performed.Finally,a druggability evaluation was conducted to prioritize potential therapeutic targets for SC.Results A total of six proteins were causally associated with SC risk(P<0.05/4972).Elevated levels of three proteins[Dipeptidase 1(DPEP1),Glutathione synthetase(GSS)and Copine 1(CPNE1)]and decreased levels of another three proteins[Cathepsin S(CTSS),Ankyrin repeat and sOCS box protein 9(ASB9)and WD repeat-containing protein 46(WDR46)]were associated with an increased risk of SC.Four proteins(DPEP1,GSS,CTSS and CPNE1)that have available pQTL files and three proteins(GSS,CTSS and CPNE1)with eQTL have passed the SMR test.Furthermore,therapeutics targeting four proteins(DPEP1,GSS,CTSS and CPNE1)are druggable.Conclusion This study has identified six protein biomarkers associated with the risk of SC.The findings provide new insights into the causes of SC and potential targets,which can further improve the integration of genome and proteome data for drug development.