Bioinformatic and integrated analysis identifies an lncRNA–miRNA–mRNA interaction mechanism in gastric adenocarcinoma

Background lncRNAs–miRNAs–mRNAs networks play an important role in Gastric adenocarcinoma (GA). Identification of these networks provide new insight into the role of these RNAs in gastric cancer. Objectives Biological information databases were screened to characterize and examine the regulatory networks and to further investigate the potential prognostic relationship this regulation has in GA. Methods By mining The Cancer Genome Atlas (TCGA) database, we gathered information on GA-related lncRNAs, miRNAs, and mRNAs. We identified differentially expressed (DE) lncRNAs, miRNAs, and mRNAs using R software. The lncRNA–miRNA–mRNA interaction network was constructed and subsequent survival examination was performed. Representative genes were selected out using The Biological Networks Gene Ontology plug-in tool on Cytoscape. Additional analysis of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) terms were used to screen representative genes for functional enrichment. Reverse transcription quantitative polymerase chain reaction (RT-qPCR) were used to identify the expression of five candidate differential expressed RNAs. Results Information of samples from 375 cases of gastric cancer and 32 healthy cases (normal tissues) were downloaded from the TCGA database. A total of 1632 DE-mRNAs, 1008 DE-lncRNAs and 104 DE-miRNAs were identified and screened. Among them, 65 DE-lncRNAs, 10 DE-miRNAs, and 10 DE-mRNAs form lncRNAs–miRNAs–mRNAs regulatory network. Additionally, 10 lncRNAs and 2 mRNAs were associated with the prognosis of GA. Multivariable COX analysis revealed that AC018781.1 and VCAN-AS1 were independent risk factors for GA. GO functional enrichment analysis found DE-mRNA was significantly enriched TERM (P < 0.05). The KEGG signal regulatory network analysis found 11 significantly enrichment networks, the most prevailing was for the AGE-RAGE signaling pathway associated with Diabetic complications. Results of RT-qPCR was consistent with the in silico results. Conclusions The results of the present study represent a view of GA from a analysis of lncRNA, miRNA and mRNA. The network of lncRNA–miRNA–mRNA interactions revealed here may potentially further experimental studies and may help biomarker development for GA.