Construction of a circRNA-miRNA-mRNA network based on differentially co-expressed circular RNA in gastric cancer tissue and plasma by bioinformatics analysis

Increasing evidence implicates circular RNAs (circRNAs) have been involved in human cancer progression. However, the mechanism remains unclear. In this study, we identified novel circRNAs related to gastric cancer and constructed a circRNA-miRNA-mRNA network.

We propose a new circRNA-miRNA-mRNA network that is associated with the pathogenesis of GC. The network may become a new molecular biomarker and could be used to develop potential therapeutic strategies for gastric cancer.

Microarray datasets GSE83521 and GSE93541 were obtained from the Gene Expression Omnibus (GEO). Then, we used computational biology to identify circRNAs that were differentially expressed in both GC tissue and plasma compared to normal controls; then, we detected the expression of the selected circRNAs in gastric cell lines by quantitative real-time polymerase chain reaction (qRT-PCR). We also identified circRNA-related candidate miRNAs and their target genes with online tools. Combining the predicted miRNAs and target mRNAs, a competing endogenous RNA regulatory network was established. Functional and pathway enrichment analyses were performed, and interactions between proteins were predicted by using String and Cytoscape. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were performed to elucidate the possible functions of these differentially expressed circRNAs. The regulatory network constructed using the microarray datasets (GSE83521 and GSE93541) contained three differentially co-expressed circRNAs (DECs). A circRNA-miRNA-mRNA network was constructed based on 3 circRNAs, 43 miRNAs and 119 mRNAs.

GO and KEGG analysis showed that the regulation of apoptotic signaling pathway and PI3K-Akt signaling pathway were highest degrees of enrichment respectively. We established a protein-protein interaction (PPI) network consisting of 165 nodes and 170 edges and identified hub genes by using MCODE plugin in Cytoscape. Furthermore, a core circRNA-miRNA-mRNA network was constructed based on hub genes. Hsa_circ_0001013 was finally determined to play an important role in the pathogenesis of GC according to the core circRNA-miRNA-mRNA network. Conclusions: We propose a new circRNA-miRNA-mRNA network that is associated with the pathogenesis of GC. The network may become a new molecular biomarker and could be used to develop potential therapeutic strategies for gastric cancer.