In silico analysis of biomarker potentials of miRNA-mediated ceRNAs in prostate cancer

Objective: The objective of this study is to define novel biomarkers for Prostate Cancer (PCa) via in silico analysis that takes PCa-specific miRNAs, finds their combinatorial target genes (potential ceRNAs), selects ones containing Transcribed Ultra Conserved Region (T-UCR) among them and potentiates their relevance with PCa.

Methods: Thirty-four miRNAs of which clinical relevances with PCa were proved experimentally were exported via miRWalk database.Using the ComiR database, 859 genes targeted by these 34 miRNAs simultaneously were identified. Genes with ComiR score above 0.911 were taken into account. Genes containing T-UCR and showing potential ceRNA activity were extracted. Among PCa-associated ceRNAs including T-UCR, we identified genes with significant expression differences between PCa and normal prostate tissue using the GEPIA database. The statistical evaluation of the association of NFAT5 and PTBP2 genes with PCa was performed by Spearman correlation test in GEPIA database.

Results: PCa-associated ceRNAs cross-matching with genes including T-UCR in their exonic regions were NFAT5, CLK3, PTBP2, CPEB4, MIPOL1 and TCF4. We identified genes with significant expression differences between PCa and normal prostate tissues among PCa-associated ceRNAs including T-UCR. According to this analysis, NFAT5 and PTBP2 genes were significantly less expressed in PCa than in normal prostate tissue while the others didn’t show any significant differential expression pattern. NFAT5 and PTBP2 genes were found to be significantly associated with PCa (p=0.000012; R=0.72).

Conclusion: All in all, this is the study associating NFAT5 and PTBP2 genes with PCa and giving them tumor suppressive potential for PCa. Still, larger and more comprehensive studies are needed on this issue.