Abstract
Prostate cancer (PCa) is the most common malignant tumor in men, and its incidence increases with age. Serum prostate-specific antigen and tissue biopsy remain the standard for diagnosis of suspected PCa. However, these clinical indicators may lead to aggressive overtreatment in patients who have been treated sufficiently with active surveillance. Circular RNAs (circRNAs) have been recently recognized as a new type of regulatory RNA that is not easily degraded by RNases and other exonucleases because of their covalent closed cyclic structure. Thus, we utilized high-throughput sequencing data and bioinformatics analysis to identify specifically expressed circRNAs in PCa and filtered out five specific circRNAs for further analysis-hsa_circ_0006410, hsa_circ_0003970, hsa_circ_0006754, hsa_circ_0005848, and a novel circRNA, hsa_circ_AKAP7. We constructed a circRNA-miRNA regulatory network and used miRNA and differentially expressed mRNA interactions to predict the function of the selected circRNAs. Furthermore, survival analysis of their cognate genes and PCR verification of these five circRNAs revealed that they are closely related to well-known PCa pathways such as the MAPK signaling pathway, P53 pathway, androgen receptor signaling pathway, cell cycle, hormone-mediated signaling pathway, and cellular lipid metabolic process. By understanding the related metabolism of circRNAs, these circRNAs could act as metabolic biomarkers, and monitoring their levels could help diagnose PCa. Meanwhile, the exact regulatory mechanism for AR-related regulation in PCa is still unclear. The circRNAs we found can provide new solutions for research in this field.