EFFECT AND MECHANISM OF HYPERHOMOCYSTEINEMIA ON ENDOTHELIAL INSULIN SIGNALING

Abstract

Hyperhomocysteinemia (HHcy) is an independent risk factor for cardiovascular disease (CVD). Both HHcy and insulin resistance (IR) are associated with atherosclerotic CVD. Recent studies have confirmed that insulin is not only a principle regulator of glucose homeostasis but also an important vasoactive hormone involved in the modulation of vascular tone. Epidemiological studies and animal studies have demonstrated the positive correlation of HHcy with IR and diabetes. Nevertheless, the effect and mechanism of HHcy on endothelial insulin signaling and insulin resistance has not been studied. In this study, we investigated the role and mechanism of HHcy on endothelial IR in vivo using transgenic mouse model of HHcy (Tg-hCBS Cbs -/- mice, plasma Hcy levels of 102.6 ± 9.1µmol/L) and in vitro using human aortic endothelial cells (HAEC). Using bioinformatics approach, we found tissue differential expression of Insulin/PI3K pathway genes in human and mouse. Furthermore, we measured tissue Hcy, S-adenosyl methionine (SAM), S-adenosyl homocysteine (SAH) levels in Tg-hCBS Cbs +/+ mice and examined correlation of insulin signaling genes with tissue Hcy, SAH levels and SAM/SAH ratio. We found negative correlation of Insulin/PI3K signaling genes with Hcy and SAH levels and positive correlation of Insulin/PI3K signaling genes with SAM/SAH ratio. These results led us to hypothesize that HHcy might negatively regulate insulin signaling and further contributes to IR. We found that HHcy impaired glucose metabolism (p<0.01 vs controls [CT]) and insulin sensitivity (p<0.05 vs CT) in Tg-hCBS Cbs -/- mice compared to their littermate controls (Tg-hCBS Cbs -/+ or +/+ mice). Furthermore, HHcy impaired insulin-induced vasorelaxation (31% vs CT, p<0.05) and endothelium-dependent relaxation (26% vs CT, p<0.05) in Tg-hCBS Cbs -/- mouse mesenteric arterioles. HHcy did not affect endothelium-independent relaxation and potassium chloride (KCl) & phenylephrine (PE)-induced contraction responses. Moreover, we found that HHcy significantly inhibited insulin-stimulated Akt and eNOS phosphorylation and activation in HAEC, mesenteric arterial tree, and in aorta. Pre-treatment of mesenteric arterioles with Wortmanin (PI3K inhibitor) and L-NAME (Nitric oxide synthase inhibitor) significantly inhibited insulin-induced vasorelaxation in controls (p<0.05 vs vehicle pre-treatment) but not in Tg-hCBS Cbs -/- mice, suggesting that HHcy impairs insulin-induced PI3K/Akt/eNOS signaling pathway. Moreover, we found that HHcy augmented insulin-induced MAPK pathway in HAEC, mesenteric arteries, and in aorta. In addition, pre-treatment of mesenteric arterioles with MEK inhibitor (PD98059) and endothelin-1A receptor blocker (BQ123) significantly improved (p<0.05 vs vehicle pre-treatment) insulin-induced vasorelaxation in Tg-hCBS Cbs -/- mice. Further analysis of upstream insulin signaling genes show that HHcy downregulated insulin receptor substrates (IRS) 1/2 mRNAs and protein expression but did not affect insulin receptor mRNA expression. Moreover, reactive oxygen species (ROS) scavenger restored HHcy induced vascular IR. In summary, our results suggest that HHcy impairs vasodilator actions of insulin by impairing IRS/PI3K/eNOS-dependent signaling pathway and amplifying MAPK-dependent pathway leading to systemic IR and endothelial dysfunction via oxidative stress related mechanism. Our work will greatly improve our understanding by which HHcy contributes to diabetic vascular disease. Our work is supported by grants from the National Institute of Health (NIH).