Screening and bioinformatics analysis of a ceRNA network based on the circular RNAs, miRNAs, and mRNAs in pan-cancer

The pan-cancer analysis has recently brought us into a novel level of cancer research. Nowadays, the Circular RNAs (circRNAs) is becoming increasingly important in the occurrence and progression of tumors. Nevertheless, the specific expression patterns and functions of circRNAs in the pan-cancer remains unclear. Here we aimed to explore the expression patterns and functions of circRNAs in pan-cancer.

We combined our microarray with seven circRNA arrays from the Gene Expression Omnibus (GEO) database and transcriptome profiles were acquired from The Cancer Genome Atlas (TCGA) database. A circRNA-miRNA-mRNA network was created and analyzed using multiple bioinformatic approaches including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, Search Tool for the Retrieval of Interacting Genes (STRING) database, cytoHubba and MCODE app. Cell function assays including CCK-8 analysis, colony formation, and transwell assay were used to explore pan-circRNAs' functions.

A panel of 6 circRNAs, 11 miRNAs, and 318 mRNAs was found to be differentially expressed (DE) in pan-cancer. A circRNA-miRNA-mRNA network was also constructed. Then, a circRNA-miRNA-hub gene network was created according to 5 pan-circRNAs, 8 pan-miRNAs, and 16 pan-mRNAs. Enrichment analysis pointed out the possible association of DEmRNAs with pan-cancer is transcriptional misregulation in cancer. Overexpression of hsa_circ_0004639 and silence of hsa_circ_0008310 can inhibit the malignant biological properties of cancer cells.

Six pan-circRNAs were discovered and their regulating mechanisms were predicted. Those findings together will give a new insight into pan-cancer research and present potential therapy targeting as well as promising biomarkers.