Identification and analysis of circulating long non-coding RNAs with high significance in diabetic cardiomyopathy

Diabetic cardiomyopathy (DCM) lacks diagnostic biomarkers. Circulating long non-coding RNAs (lncRNAs) can serve as valuable diagnostic biomarkers in cardiovascular disease. To seek potential lncRNAs as a diagnostic biomarker for DCM, we investigated the genome-wide expression profiling of circulating lncRNAs and mRNAs in type 2 diabetic db/db mice with and without DCM and performed bioinformatic analyses of the deregulated lncRNA-mRNA co-expression network. Db/db mice had obesity and hyperglycemia with normal cardiac function at 6 weeks of age (diabetes without DCM) but with an impaired cardiac function at 20 weeks of age (DCM) on an isolated Langendorff apparatus. Compared with the age-matched controls, 152 circulating lncRNAs, 127 mRNAs and 3355 lncRNAs, 2580 mRNAs were deregulated in db/db mice without and with DCM, respectively. The lncRNA-mRNA co-expression network analysis showed that five deregulated lncRNAs, XLOC015617, AK035192, Gm10435, TCR-α chain, and MouselincRNA0135, have the maximum connections with differentially expressed mRNAs. Bioinformatic analysis revealed that these five lncRNAs were highly associated with the development and motion of myofilaments, regulation of inflammatory and immune responses, and apoptosis. This finding was validated by the ultrastructural examination of myocardial samples from the db/db mice with DCM using electron microscopy and changes in the expression of myocardial tumor necrosis factor-α and phosphorylated p38 mitogen-activated protein kinase in db/db mice with DCM. These results indicate that XLOC015617, AK035192, Gm10435, TCR-α chain, and MouselincRNA0135 are crucial circulating lncRNAs in the pathogenesis of DCM. These five circulating lncRNAs may have high potential as a diagnostic biomarker for DCM.