基于多层次组学数据遗传分析的肺癌转移和复发分子遗传机制的解析

Metastasis and recurrence are the main malignant characters of lung cancer, and are also the leading cause of lung cancer mortality. In the current researches, uncovering the mechanism of metastasis and recurrence of lung cancer mostly focused on single tumor-related gene (or signaling pathway), but it remains relatively rare to systematically reveal the malignant pathological progression of the disease at the omics level. Based on our obtained poly-omic sequencing data generated from large-scale clinical specimens of lung cancer, including exome-seq, methylation-seq, mRNA-seq and miRNA-seq, we will identify the genetic variants affecting clinical phenotypes of metastasis and relapse of lung cancer and the molecular traits, such as methylation level, mRNA and miRNA expression, through QTL mapping; statistically infer the mutual regulatory relationships among the methylated gene, mRNA and miRNA affecting the clinical phenotypes; and finally develop a genetic regulatory network which mechanistically elaborate the metastasis and relapse of lung cancer through integrating the molecular interacting information, such as protein interaction. Additionally，we will also experimentally testify the genetic effect of causal variants and key regulators in the genetic network on controlling the phenotypes of metastasis and recurrence, and predict and experimentally verify novel tumor biomarkers. The project will offer a new research idea and technical pipeline for revealing the underlying mechanism of complex disease by integrating multi-level bio-information, and will also accumulate new knowledge for the metastasis and recurrence of lung cancer.

转移与复发是肺癌的主要恶性特征，也是导致肺癌死亡率一直居高不下的主要原因。当前，对肺癌转移和复发的机制研究大多集中在单个基因或信号通路，而从组学层面系统性揭示该疾病恶性病理进程的研究报道却十分匮乏。本项目基于我们前期积累的成规模临床样本多层次组学测序数据（包括外显子、甲基化、mRNA和miRNA），通过QTL遗传分析，识别控制肿瘤转移和复发临床表型和分子性状（包括甲基化状态、mRNA和miRNA表达水平）的遗传变异；计算推断影响临床表型的分子性状关联基因间的相互调控关系；最终整合数据库中分子间直接互作信息，构建具有机制性解释意义的肺癌转移和复发遗传调控网络。此外，实验验证遗传调控网络中致病遗传变异和关键调控基因对临床表型的生物学效应，并预测和验证新的肺癌分子标记物。本项目的实施将提供一个整合多层次生物信息揭示复杂疾病遗传机制的研究思路和技术路线，也将为肺癌的转移和复发积累新的科学知识。

随着深度测序技术的发展，使得许多生物学现象可以很便捷从组学层面进行解析，肿瘤学的发展也由此获益。然而面对爆炸性增长的组学数据，如何从中高效挖掘有益信息是当前生物学家面临的挑战。本项目利用本实验室收集的大规模肺癌组织多层次组学数据，建立了不同组学数据的分析流程，应用到后续的肺癌和脑肿瘤的组学数据分析中，并由此取得了一系列有意义的研究进展。(1) 通过对肺癌的miRNA组分析并结合分子生物学实验，识别了两个miRNA(miR-205-5p和miR-375)可以用于区分肺腺癌和鳞癌的分子标记物；同时，基于肺癌与正常组织的miRNA比较，并通过血清miRNA的验证，挖掘到3个miRNA（miR-21-5p、miR-451a和miR-486-5p）可以通过血清样本来筛查肺癌病例。（2）通过对肺癌的RNAseq数据分析，鉴别了多个融合基因，并对其中的FGFR3-TACC3融合基因进行了细致的功能分析，发现该融合基因是通过蛋白间相互作用形成液相凝集物，促进了FGFR3的功能，进而促发肿瘤的发生。（3）本项目建立的数据分析方法应用到脑肿瘤的转录组数据分析中，发现YAP的激活可能诱导室管膜瘤的发生。最终通过实验验证，发现病人来源的YAP融合基因具有诱发肿瘤的能力，并进一步揭示了该融合基因是通过液液相分离调控超级增强子的产生，进而驱动多个致癌信号通路的激活。本项目所建立的组学数据分析的方法为后学组学数据分析建立了一个标准的示范；所识别和鉴定的miRNA 分子标记物可以应用于未来的临床实践；所揭示的FGFR3-TACC3融合基因和YAP融合基因诱发肿瘤的分子机制，给抗肿瘤新药的开发提供了新的靶点和设计视角。