microRNA-mRNA regulatory network analysis reveals a possible metastatic mechanism of uveal melanoma

Abstract

Abstract Aim: Uveal melanoma (UVM) is the most common primary intraocular malignant tumor in adults and it can develop metastatic melanoma. Therefore, it is crucial to identify biomarkers to provide early diagnosis and therapeutic targets. Methods: The differentially expressed microRNAs (DEmiRNAs) and mRNAs in patients with UVM were identified in The Cancer Genome Atlas (TCGA) database. The target mRNAs regu-lated by DEmiRNAs were obtained from TargetScan and miRDB databases. Gene Ontology (GO) terms and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways were performed using Metascape software. The hub mRNAs used for the construction of protein–protein interaction (PPI) network were identified using the STRING database and CytoHub plug-in. TCGA database and miRNA-targeted mRNAs were used to identify key mRNAs. Hub and key mRNAs were searched PubMed database for verification. Survival analysis was done using Gene Expression Profiling Interactive Analysis (GEPIA). Moreover, the correlations between methylation level and key mRNA expression together with survival rate were analyzed by gene set cancer analysis (GSCA). The miRNA–mRNA network was constructed by integrating mRNAs and miRNAs in-formation. Results: We identified 22 DEmiRNAs and obtained 1436 targeted mRNAs in patients with UVM. Ten hub mRNAs (i.e., HNRNPA1, SRSF1, MATR3, SYNCRIP, TRA2B, TIAL1, FUS, FN1, SFPQ, HNRNPU) were screened and HNRNPA1, SRSF1, TRA2B, TIAL1, FUS, FN1, SFPQ, and HNRNPU were associated with cancer metastasis. KEGG analysis showed FN1 was associated with survival. In addition, CA12, NYNRIN, TDRD10 and WDR72 were associated with survival, while FOXD3, CA12 and SPDEF play pivotal roles in cancer metastasis. The TDRD10, COL11A2 and NYNRIN levels were negatively correlated with methylation, and the methylation level had a significant impact on the prognosis of metastatic UVM. The miRNA–mRNA regulatory network was con-sisted of 10 miRNAs and 14 key mRNAs, and these miRNA targets may have potential links to UVM metastasis. Conclusion: We found that HNRNPA1, SRSF1, TRA2B, TIAL1, FUS, FN1, SFPQ, HNRNPU, FOXD3, CA12 and SPDEF were related to metastatic UVM, and FN1, CA12, NYNRIN, TDRD10 and WDR72 were related to survival in metastatic UVM. These mRNAs may be used as bi-omarkers of metastatic UVM and therapeutic targets.