Candidate genes associated with the effect of rosiglitazone on glycemic control and cardiovascular system in the treatment of type 2 diabetes mellitus.

In the present study, candidate genes affected by rosiglitazone to exert glycemic control in the treatment of type 2 diabetes mellitus (T2DM) and associated with its adverse cardiovascular effects were identified using a bioinformatics analysis. The gene expression profiles of the dataset GSE36875 from the Gene Expression Omnibus database, including heart samples from 5 non-diabetic control mice (NC), 5 untreated diabetic mice (NH) and 5 rosiglitazone-treated diabetic mice (TH), were used to identify differentially expressed genes (DEGs) in the NC vs. NH, NC vs. TH and TH vs. NH groups. Subsequently, Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways enriched by the DEGs were determined. Furthermore, genes associated with the action of rosiglitazone were identified using Short Time-series Expression Miner, which were then subjected to enrichment analysis in gene ontology (GO) terms in the category biological process (BP), and networks of the GO terms, KEGG pathways and genes associated with the action of rosiglitazone were constructed. Finally, biological abnormalities associated with these genes were identified using WebGestalt. A set of 791 DEGs in three groups (NC vs. NH, NC vs. TH and NH vs. TH) were identified. Subsequently, 72 DEGs [e.g., apolipoprotein (Apo)A1, ApoA5, cytochrome P450 (Cyp)2c37, Cyp2J5, Cyp2b9 and Cyp2b10] were identified as genes associated with the action of rosiglitazone. In addition, a network of 13 GO terms in the category BP, 6 KEGG pathways and 41 genes associated with the action of rosiglitazone was constructed, with major terms/pathways including oxidation/reduction, lipid transport, peroxisome proliferator-activated receptor signaling pathway and metabolism of xenobiotics by Cyp. Finally, 15 biological abnormalities (including abnormal triglyceride levels, abnormal cholesterol homeostasis, abnormal lipid homeostasis) associated with these genes were identified. ApoA1, ApoA5, Cyp2c37, Cyp2J5, Cyp2b9 and Cyp2b10 were differently expressed after rosiglitazone treatment, which may be accountable for affecting cardiovascular outcomes and glycemic control in T2DM. The present results may expand the current understanding of the mechanism of action of rosiglitazone to exert glycemic control in T2DM, as well as its effects on the cardiovascular system.