基于microRNA协同调控模块挖掘和验证肾透明细胞癌的联合诊断标记

Early diagnosis of cancer is a difficult problem in the treatment of complex diseases. Its accurate diagnosis must consider regulatory cooperation between genes. MicroRNA provides potential biomarkers for cancer diagnosis. Currently the majority of work in search of microRNA biomarkers is limited to single molecules, ignoring the synergies between them. Therefore the single microRNA molecures are of limited clinical diagnostic value, and the collaborative microRNA-microRNA regulation mechanism remains unclear. This project will identify the synergistic regulatory module of microRNAs on the network level, and accordingly find reliable biomarker combinations for the diagnosis of clear cell renal cell carcinoma (ccRCC). First, we will integrate microRNA-gene interactions from different sources, and optimize multiple regression model to reconstruct a high quality ccRCC-specific microRNA-gene regulatory network. Secondly, we will define and mine the coordinately regulated microRNA modules from the network, with subsequent analysis of the sequence, structure and biological function of the collaborative microRNAs within the module. Based on synergistic microRNA regulatory modules, we will identify a combination of diagnostic markers of ccRCC and evaluate its diagnostic performance. We will finally validate the biomarker combination with downstream biological experiments. This project helps to further understand the synergistic regulatory mechanisms in ccRCC and provides novel means for the identification of reliable diagnostic markers.

癌症的早期诊断是复杂疾病诊疗中的难题,对其准确诊断必须考虑基因间的合作调控。microRNA是肿瘤诊断的潜在生物标记，目前多数研究局限于单分子层面，忽略了它们间的协同作用，临床诊断价值有限，microRNA的协同调控机制仍不明确。本项目将在网络水平上挖掘microRNA协同调控模块，并据此找到可靠的肾透明细胞癌联合诊断标记。内容包括：1）整合多层次的microRNA-gene调控关系，以优化的多元回归模型重构高质量的肾透明细胞癌特异的microRNA-gene调控网络；2）定义并挖掘网络中权重型microRNA功能协同调控模块，分析模块内协同microRNA的序列、结构和生物功能；3) 依据协同调控模块寻找肾透明细胞癌的microRNA联合诊断标记，评价其诊断效能，并进行下游生物学实验验证。本研究有助于深入理解肾透明细胞癌的分子合作调控机制，为发现可靠的诊断标志物提供新的手段。

研究背景：肾透明细胞癌(ccRCC)是肾癌最常见的亚型，目前尚无敏感特异的诊断标志物。癌症生物标志物的发现不应只考虑单独的标记物分子，而应考察其在全局网络之间动态的相互作用。.研究内容：以GSE16441配对miRNA/mRNA芯片表达谱为研究对象，利用LASSO多元线性回归模型构建ccRCC发生过程中miRNA调控网络，并根据靶基因重叠性、靶基因共享功能类和靶基因的调控强度一致性这三个指标挖掘miRNA协同调控模块，并寻找miRNA诊断标记。利用qRT-PCR验证ccRCC细胞株A498中miRNA成熟体表达水平。采用逆转录病毒介导的转染法构建高表达hsa-miR15a-5p的肾癌细胞株，采用流式细胞仪检测细胞凋亡，MTT法检测细胞生长，Transwel法检测癌细胞迁移及侵袭能力。.研究结果：成功构建了ccRCC特异microRNA-mRNA 调控网络，获得6个单分子ccRCC标记物：miR-15a-5p, miR-16-5p miR-25a-3p, miR-34a-5p, miR-224-5p, miR-93-5p，和一对组合标记物：miR-15a-5p/miR-16-5p。miR-15a-5p/miR-16-5p组合的AUC (0.986)高于单一标志物，灵敏度和特异性分别达90%和95%。MetaCore通路富集分析发现TGF, WNT and cytoskeletal remodelling等14条ccRCC相关新通路。qRT-PCR发现所有候选miRNA标记物在癌细胞株中的表达调控方向与芯片一致。体外生物学实验表明，高表达miR-15a-5p可促进肾癌细胞凋亡、抑制癌细胞的生长增殖和侵袭能力。.科学意义：利用LASSO多元回归去除假阳性变量干扰，准确构建了ccRCC特异microRNA-mRNA调控网络，获得可靠的分子标记物，为ccRCC早期诊断提供了理论根据。