Identification of Potentially Functional Circular RNA/Long Noncoding RNA-MicroRNA-mRNA Regulatory Networks Associated with Vascular Injury in Type 2 Diabetes Mellitus by Integrated Microarray Analysis

Abstract

This research is aimed at figuring out the potential circular RNA (circRNA)/long noncoding RNA- (lncRNA-) microRNA- (miRNA-) mRNA regulatory networks associated with a vascular injury in type 2 diabetes mellitus (T2DM). Differentially expressed genes (DEGs) screened in T2DM-related expression datasets were intersected with genes associated with vascular injury in T2DM to obtain candidate DEGs, followed by the construction of an interaction network of DEGs. The upstream miRNAs of candidate genes were predicted by mirDIP, miRWalk, and DIANA TOOLS databases, and the upstream lncRNAs/circRNAs of miRNAs by DIANA-LncBase/circBank database, followed by the construction of circRNA/lncRNA-miRNA-mRNA regulatory networks. Peripheral blood was attained from T2DM patients with macroangiopathy for clinical validation of expression and correlation of key factors. Differential analysis screened 37 candidate DEGs correlated with vascular injury in T2DM. Besides, MAPK3 was a core gene associated with vascular injury in T2DM. Among the predicted upstream miRNAs of MAPK3, miR-4270, miR-92a-2-5p, miR-423-5p, and miR-613 ranked at the top according to binding scores. The upstream lncRNAs and circRNAs of the 4 miRNAs were further predicted, obtaining 11 candidate lncRNAs and 3 candidate circRNAs. Moreover, KCNQ1OT1, circ_0020316, and MAPK3 were upregulated, but miR-92a-2-5p was downregulated in the peripheral blood of T2DM patients with macroangiopathy. Mechanistically, KCNQ1OT1 and circ_0020316 bound to miR-92a-2-5p that inversely targeted MAPK3. Collectively, KCNQ1OT1/circ_0020316-miR-92a-2-5p-MAPK3 coexpression regulatory networks might promote vascular injury in T2DM.