927 MOLECULAR ROLE OF MIR-195/497 CLUSTER IN HUMAN BLADDER CANCER BASED ON MULTIPLE DATASET OF MICRORNA EXPRESSION SIGNATURES BY DEEP SEQUENCING

INTRODUCTION AND OBJECTIVES: Recent our miRNA expression signature of human bladder cancer (BC) by deep sequencing revealed new candidate microRNAs (miRNAs) as tumor suppressors. However, a single miRNA signature from a single study might be unreliable for discriminating its significance. In this study, we adopted in-silico analyses using multiple dataset of miRNAs and mRNA gene expression signatures from the published data and our data to determine tumor suppressive miRNAs and their target genes/molecular pathways in BC. METHODS: We established a miRNA expression signature in 5 BCs and 5 normal bladder epitheliums (NBEs). We also reviewed the miRNAs expression in other deep sequencing studies, which containing a total of 48 BCs and 38 NBEs. To identify the pathways potentially regulated by the miRNAs, we applied TargetScan database and GENECODIS software, which assigned many putative miRNA targets to known pathways in KEGG [http://www.genome.jp/kegg/pathway.html]. Gene expression analyses of all candidate genes were analyzed by using Gene Expression Omnibus (GEO) database. In these microarray expression data, we examined 90 BCs and 6 NBEs. We performed gain-of-function studies (XTT, wound healing, and cell invasion assay) by the mature miRNAs transfection into BC cell lines. RESULTS: Multi-dataset of miRNA deep sequencing analysis revealed that 53 miRNAs were significantly down-regulated in BC. Among the miRNA clusters in the miRNAs list, we focused on miR195/497 cluster because it has not been fully investigated in BC. The expression of miR-195 and miR-497 were significantly reduced in BCs. Significant inhibitions of cell proliferation, migration, and invasion were observed in the miR-195 and miR-497 transfectants (each, P 0.001). TargetScan algorithm exhibited 6730 putative genes targeted by miR195 and miR-497. GENECODIS software assigned these genes into 113 significantly enriched signaling pathways. In the top 21 enriched pathways, 98 genes were actually up-regulated in BC according to the expression data of GEO. In particular, RAF1, MEK1, GRB2 and PIK3R2, which are the key molecule in oncogenic pathways, were highly expressed across more than 12 enriched pathways. CONCLUSIONS: Our results suggested that oncogenic RAF1, MEK1, GRB2, and PIK3R2 might be significant target genes of tumor suppressive miR-195/497 cluster and that activated RAF-MEK-ERK pathway and PI3K pathway might be attribute to the down regulation of miR-195/497 cluster in BC.