Identification of candidate genes for endometrial cancer in multi-omics: a Mendelian randomization analysis

Endometrial cancer is the most common malignant tumor of the uterus, but the underlying genetic mechanisms of EC remain unclear. To identify candidate genes and investigate genetic mechanisms for endometrial cancer, we utilized the summary-data-based Mendelian randomization (SMR) method to investigate causal associations between genetic variants, gene expression, DNA methylation, and endometrial cancer. Three main analyses were conducted utilizing cis-expression and methylation quantitative trait loci (eQTLs and mQTLs) as instrumental variables to examine causal relationships with endometrial cancer, and assessing the causal relationship between DNA methylation and gene expression. Data sources included genetic association data from O'Mara et al. eQTL data from the GTEx database, and mQTL data from McRae et al. Analysis involved the HEIDI test to distinguish pleiotropy, SMR analysis with multiple testing correction, and colocalization analysis to assess associations driven by linkage disequilibrium. Functional enrichment analysis was performed by the Metascape tool. Our study showed that three genes, SNX11, LINC00243, and EVI2A, were identified as causally related to endometrial cancer. SNX11 exhibited a positive causal relationship, while LINC00243 and EVI2A showed negative ones. Furthermore, 24 CpG sites were identified as causally related to endometrial cancer, with cg14424631 (CYP19A1) being the most significant. The study revealed common genes implicated in endometrial cancer, gene expression, and methylation sites, with LINC00243 playing a key role. Colocalization analysis confirmed significant causal relationships between LINC00243, SNX11, and endometrial cancer. Enrichment analysis uncovered pathways like interferon gamma signaling enriched in both endometrial cancer GWAS and e/mQTL. These findings shed light on the molecular mechanisms underlying endometrial cancer development. The study identified candidate genes and DNA methylation loci causally associated with endometrial cancer, which are expected to serve as potential targets for treatment. <b>Flowcharts of SMR and Mendelian randomization mechanism for eQTL, mQTL, and endometrial cancer GWAS datasets.</b> (A) Mendelian randomization analysis architecture for endometrial cancer. (B) Three parts of SMR analysis using eQTL and mQTL datasets for endometrial cancers were conducted in order to find significant causal correlations in gene expression levels.

子宫内膜癌（Endometrial cancer）是子宫最常见的恶性肿瘤，但其潜在的遗传发病机制仍未阐明。为鉴定子宫内膜癌的候选基因并探究其遗传机制，本研究采用基于汇总数据的孟德尔随机化（summary-data-based Mendelian randomization）方法，探究遗传变异、基因表达、DNA甲基化与子宫内膜癌之间的因果关联。本研究开展三项核心分析：以顺式表达与甲基化数量性状位点（cis-expression and methylation quantitative trait loci，eQTLs和mQTLs）作为工具变量，检验其与子宫内膜癌的因果关系；同时评估DNA甲基化与基因表达间的因果关联。本研究的数据来源包括奥马拉等人（O'Mara et al.）的遗传关联数据、GTEx数据库的表达数量性状位点数据，以及麦克雷等人（McRae et al.）的甲基化数量性状位点数据。分析流程涵盖HEIDI检验以区分多效性、经多重检验校正的SMR分析，以及共定位分析以评估由连锁不平衡驱动的关联。本研究通过Metascape工具完成功能富集分析。本研究鉴定出SNX11、LINC00243与EVI2A三个与子宫内膜癌存在因果关联的基因，其中SNX11呈正向因果关系，而LINC00243和EVI2A呈负向因果关系。此外，本研究共鉴定出24个与子宫内膜癌存在因果关联的CpG位点，其中cg14424631（CYP19A1）的关联最为显著。本研究揭示了参与子宫内膜癌发生、基因表达调控与甲基化位点调控的共同基因，其中LINC00243发挥关键作用。共定位分析证实了LINC00243、SNX11与子宫内膜癌间存在显著因果关联。富集分析显示，γ干扰素信号通路等通路同时富集于子宫内膜癌全基因组关联研究（Genome-Wide Association Study, GWAS）与表达/甲基化数量性状位点数据中。上述研究结果为阐明子宫内膜癌发生的分子机制提供了新的视角。本研究鉴定出与子宫内膜癌存在因果关联的候选基因与DNA甲基化位点，有望成为潜在的治疗靶点。**基于eQTL、mQTL与子宫内膜癌全基因组关联研究（GWAS）数据集的SMR及孟德尔随机化机制流程图**。（A）子宫内膜癌的孟德尔随机化分析架构。（B）本研究利用eQTL与mQTL数据集针对子宫内膜癌开展SMR分析的三个部分，旨在筛选基因表达水平中存在显著因果相关性的关联。