用系统生物学方法对miRNAs介导的癌细胞死亡网络的研究

Mammalian cells have evolved several elegant mechanisms to die, including apoptosis and autophagy. These two signaling pathways can coordinately modulate the cell death process in cancers. In recent years, substantial research interests have arisen to decipher miRNA function in cell death pathways. Previously we also showed that enforced expression of MIR153 increases apoptosis via targeting B-cell lymphoma 2 (BCL2),myeloid cell leukemia sequence 1 (MCL1) and insulin receptor substrate-2 (IRS2),linking this tumor suppressor to the apoptosis pathway.. Despite enormous success, it is increasingly evident that systems biology approaches are essential to uncover the generic organizing principles of the miRNA mediated cell death network.For the first step toward understanding the network architecture, we have developed miRDeathDB (http://rna-world.org/mirdeathdb/), a knowledge depository that integrates information for experimentally identified miRNAs and their targets in cell death network.In this project, we propose to integrate a variety of omics data centering on cell death related miRNAs. This information includes SNP data,expression profile for both miRNAs and their targets,transcription factor-miRNAs data,protein-protein interaction data, signaling pathways data and Gene Ontology annotations. Then the characteristics of cell death related miRNAs, such as regulating sequences,topological structrues and GO enrichment, were directly mined from above rich resources. Machine learning approaches and information integrations techniques, such as K Nearest Neighbor(KNN) and Support vector machine(SVM),were applied to predict new cell death related miRNAs and their differential function. In a cell based system, the biological significances of the predicted miRNAs were tested via experimental assays, such as tranfection, western blot, immunodection and flow cytometry assays.The overall organization principles are exploited by combining heterogeneous data to construct the global miRNAs mediated cell death network. The outputs from this project will delineate the miRNA mediated crosstalk mechanism between apoptosis and autophagy.In addition, a detailed deciphering of the crucial role of miRNAs in cell death network have profound clinical implications since the evasion of cell death underlies tumorigenesis and represents a major obstacle to successful therapies.Therefore, such efforts are imperative to improve our understanding of miRNAs in tumorigenesis and facilitate the design of next generation theraputics.

癌细胞死亡进程受到包括细胞凋亡和自噬等信号途径的协同调节。采用系统生物学方法分析 miRNAs在癌细胞死亡网络中的分子机制及全局性作用是具有挑战意义的课题。本研究通过有机整合遗传变异、转录调控、表达谱和蛋白互作谱等各种组学数据，分析调控癌细胞死亡的 miRNAs在调控序列、表达谱关系、靶蛋白互作谱以及靶蛋白功能富集类等多方面的特征，利用KNN、SVM等各种机器学习及数据融合方法设计预测算法，从系统的角度识别新的死亡相关 miRNAs及其差异化功能。借助细胞转染、免疫荧光以及流式细胞检测等生物学实验揭示miRNAs在癌细胞死亡进程重要作用。在此基础上，构建包含多层面信息的、miRNAs介导的细胞死亡调控网络,阐明其中调控规律和特征。研究结果将有助于理解凋亡与自噬通路协同调控癌细胞死亡的分子机制，为抗癌治疗提供理论基础。

细胞死亡进程在癌症发生发展中有重要作用。近年来的研究发现miRNAs和编码蛋白基因协同调节细胞死亡进程。本课题首先汇总分析各种生物组学数据，开发构建了miRNAs与细胞死亡数据库。它共包含92个死亡相关miRNAs以及相互作用的106个蛋白；通过整理各种生物数据，开发构建了RNA-protein相互作用数据库。该数据库详细注释了4493对RNA-RNA相互作用和1619对RNA-protein相互作用，总共涉及了2070个各类RNA分子，收录条目6112条。这些数据库资源是开展本课题后续各项工作的重要基础。其次结合食管癌数据，分析了死亡相关miRNAs 的调控序列、功能节点富集类等特征，设计了机器学习及数据融合方法用于识别新的死亡相关miRNAs的新算法并做了初步的实验验证工作，初步阐明了miRNA参与调控细胞死亡网络的框架。此外，采用系统生物学方法分析了非小细胞肺癌数据，构建了miRNA-mRNA互作网络。本课题创建的上述数据库资源可在网络自由索取，已经开始服务于国内外相关领域的其他研究人员。研究成果为研究进一步探索癌症miRNAs与细胞死亡的关系、miRNA-蛋白互作关系奠定了基础。