(497b) Prediction of Therapeutic Microrna Based On Human Metabolic Network

Wu, M., Michigan State University
Chan, C., Michigan State University

A global decrease in microRNA (miRNA) levels and alterations of miRNA regulation has been observed in human cancers, which suggests miRNAs could have an intrinsic function in tumor suppression. Thus, there is interest in the cancer research community to use miRNA as alternative therapeutic targets. Novel approaches have been developed to target and deliver miRNAs to mammalian system that can successfully inhibit tumor cell proliferation and induced cancer-specific apoptosis. Although miRNAs could be good alternative targets for cancer treatment, it has been difficult to identify which miRNA to target for a particular type of cancer due to our limited knowledge of the regulatory roles of miRNAs in cancer. In this study we propose a novel approach to tackle this challenge, by integrating miRNA target prediction, metabolic modeling and context-specific gene expression data to predict therapeutic miRNAs that can reduce the growth of cancer. Our analysis focuses on the human metabolic system, since abnormal metabolic functions are known to be involved in supporting tumor growth and proliferation. We hypothesized that miRNAs could be implicated in the metabolic regulation of cancer, and developed a novel approach to simulate context dependent metabolic states upon perturbation of gene expression (e.g. induced by miRNA). We incorporate tissue/cancer-specific gene expression information with the generic human metabolic network to reconstruct a context-specific cancer metabolic network, in which miRNAs are perturbed in silico to predict their function on metabolic processes and cancer cell growth. Compared with experimental evidences collected from the literature, our approach achieves 89 percent accuracy in predicting miRNAs that can suppress metastasis and progression of human hepatocellular carcinoma (HCC). Our methodology predicts how perturbations in miRNAs impact the cancer metabolic states, thereby suggesting novel mechanisms by which miRNAs regulate cancer cell growth through modulating metabolic functions. We also applied our system to human breast cancer to identify therapeutic miRNA targets for breast cancer.To the best of our knowledge, this is the first computational approach implemented to predict therapeutic miRNAs for human cancer based upon their functional role in cancer.