Construction and Investigation of an lncRNA-miRNA-mRNA Regulatory Network of Pancreatic Cancer via bioinformatics analysis

Abstract PurposePancreatic cancer causes a malignant digestive tumour with an unfavourable prognosis and insidious onset. The regulation mechanisms of pancreatic cancer were connected to the abnormal accommodate of diverse signaling pathways. Studies had increasingly indicated that long non-coding RNAs (lncRNAs), microRNAs (miRNAs), and messenger RNA (mRNAs) could play crucial roles during tumorigenesis. This study designs to detect functional genes and determine the potential molecular mechanisms of pancreatic cancer through bioinformatics.MethodsWe made a comprehensive comparison on the RNA-sequencing using the Gene Expression Omnibus (GEO) and established a dysregulated lncRNA-miRNA-mRNA network. In addition, several functional analyses in the competing endogenous RNAs (ceRNA) network were conducted by Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG). In addition, survival analysis and clinical feature analysis were made by Gene Expression Profiling Interactive Analysis (GEPIA) and linkedOmics. Proliferation, migration tests were selected to examine the roles of mir-130b in pancreatic cancer. Results23 lncRNAs, 40 miRNAs and 1613 mRNAs were aberrantly expressed in pancreatic cancer. We constructed a dysregulated ceRNA network with 11 lncRNAs, 19 miRNAs, and 66 mRNAs. A group of 121 interconnections was constructed in the ceRNA network. These upregulated DEmRNAs were mainly enriched in the mitotic spindle, were involved the transcription by RNA polymerase II, were related to E-box binding and participated in bacterial invasion of epithelial cells and yersinia infection etc. in contrast, the these upregulated DEmiRNAs were mainly enriched in spanning component of the plasma membrane, were involved in pyrimidine nucleobase catabolic process, were related to pyridoxal phosphate binding and participated in beta-Alanine metabolism and propanoate metabolism etc.. The survival analysis showed that 1 lncRNAs, 4 miRNAs, and 18 mRNAs might be potential factors for the prognostic prediction of pancreatic cancer. The strong positive relationships were found between candidate lncRNAs and candidate mRNAs targeted by miR-130b, CYTOR interacts with FRMD6 and TCF4, while HOXA11-AS interacts with MET regulated by miR-130b. MiR-130b was found that its overexpression could inhibit the proliferation and migration of pancreatic cancer cells, and lower expression of miR-130b was connected with longer survival of pancreatic cancer patients.ConclusionOur research provided further acknowledges the molecular mechanisms underlying pancreatic cancer and laid strong ground for improving pancreatic cancer diagnosis and prognosis by establishing the lncRNA-miRNA-mRNA regulatory network via bioinformatics analysis.