Abstract
Prostate adenocarcinoma (PCa) is the most common cause of death due to malignancy among men, and bone metastasis is the leading cause of mortality in patients with PCa. Therefore, identifying the causes and molecular mechanism of bone metastasis is important for early detection, diagnosis and personalized therapy. In this study, we systematically analyzed molecular correlates of bone metastasis by bioinformatics analysis. A total of 12 differentially expressed microRNAs (miRNAs) and 102 differentially expressed genes were identified. Five miRNAs had prognostic significance in biochemical recurrence-free survival (miR-636, miR-491-5p, miR-199b-5p, miR-199b-3p, miR-28-3p). The differentially expressed genes were significantly enriched in extracellular matrix, cell-substrate adhesion, collagen and integrin. Seven hub genes (VCAN, COL3A1, COL1A1, APOE, COL1A2, SDC1, THY1) with worse biochemical recurrence-free survival and one hub gene (MMP9) with worse overall survival were detected. miR-636, a novel oncogene, was found to be up-regulated in bone metastatic PCa tissues and also predominately up-regulated in human PCa cell lines. miR-636 promoted cellular invasion and migration, and may promote bone metastasis via targeting MBNL2, TNS1 and STAB1. In conclusion, we have successfully defined molecular signatures of bone metastasis in PCa.