Previously, we have reported that activation of Rho-kinase by hypoxia leads to the downregulation of eNOS expression and activity.8
Because activation of eNOS is fundamental to the maintenance of normal vascular function, these data suggest that inhibition of Rho-kinase might exert beneficial effects in diseases in which endothelial function is impaired. A recent study found that long-term treatment with fasudil was effective in preventing the development of atherosclerosis.23
In this study, we found that inhibition of Rho-kinase by HF rapidly increased Akt phosphorylation and activation (EC50
= 7.3 μ
mol/L). At the concentration used, HF is relatively selective for Rho-kinase and had minimal effects on other signaling pathways such as myosin light chain kinase and protein kinase C.24
Furthermore, another chemically distinct Rho-kinase inhibitor, Y-27632, had similar effects on Akt serine-473 phosphorylation as that of HF, suggesting relative selectivity of the observed effects. In contrast to the rapid stimulatory effect of Rho-kinase inhibition on Akt, a recent study could not demonstrate Akt activation using dominant-negative mutants of Rho and Rho-kinase.13
This discrepancy might be explained by the different experimental approaches used between these studies. For example, Akt activation was only examined 18 hours after inhibition of Rho-kinase using adenoviral gene transfer of dominant-negative mutants of Rho and Rho-kinase. At such a late time point, it is possible that Akt is no longer activated. In contrast, we observed activation of Akt within 15 minutes after inhibition of Rho-kinase by HF.
The lipid class I PI3-kinases generate 3′ phosphorylated phosphatidylinositols (eg, phosphatidylinositol 3,4,5-trisphosphate [PIP3
which recruit Akt to the cell membrane, where it is phosphorylated.26
PI3-kinase inhibitors block the activation of Akt by HF, indicating that Rho-kinase tonically inhibits Akt through PI3-kinase. Therefore, our findings support our hypothesis that Rho-kinase negatively regulates Akt via inhibitory effects on PI3-kinase. Interestingly, a potential interaction between Rho-kinase and PI3-kinase/Akt pathway has been reported earlier by Farah et al.27
In this study, using yeast 2-hybrid analyses, a Xenopus homolog of Rho-associated kinase (ROK)-α
was found to bind to the pleckstrein-homology (PH)-phosphotyrosine binding domain of the insulin receptor substrate-1. Micro-injection of mRNA corresponding to a Rho-kinase mutant inhibited insulin-mediated signal transduction. Thus, Rho-kinase may inhibit PI3-kinase through an inactivating serine phosphorylation of insulin receptor substrate-1.27
In addition, the phosphorylation motif of Rho-kinase (RXXS or RXXT) can be found in the catalytic p110 subunit of PI3-kinases as well as in the PH-phosphotyrosine binding domain of the regulatory p85 subunit. However, it is not known whether Rho-kinase can directly phosphorylate PI3-kinase subunits and alter its activity.
A major downstream target of Akt in endothelial cells is eNOS, which is activated by phosphorylation at serine-1179.11,12
We found that HF also stimulated NO production, which was blocked by PI3-kinase and Akt inhibitors. Surprisingly, the NO2
release achieved by HF was comparable to 100 nmol/L of bradykinin. However, because NO2
represents cumulative NO release after 1 hour of incubation, the kinetics of NO release after HF and bradykinin may differ.
NO not only is a potent vasodilator but also inhibits platelet aggregation and leukocyte recruitment to the vascular wall.28
We found that induction of NO by fasudil leads to decreased leukocyte rolling and adhesion after hemorrhage-induced I/R injury in mice. These anti-inflammatory effects of fasudil were completely reversed by the PI3-kinase inhibitors, indicating the involvement of the PI3-kinase/Akt pathway. Similarly, the anti-ischemic effects of HF were mediated by PI3-kinase and NO because the cardioprotective effects of HF were blocked by inhibitors of PI3-kinase and eNOS. Interestingly, we could not observe an effect on myocardial infarct size in animals treated with L-NAME alone, although MAP was significantly increased after L-NAME administration. The lack of effect of L-NAME is in accordance with a recent study from our group, in which L-NAME itself had no effect on infarct size but abolished cardioprotection achieved by statin treatment.19
In addition, these findings agree with those of Ockaili et al,29
who reported no difference in rate pressure product or infarct size in L-NAME–treated versus control rabbits. Although fasudil was given systemically and therefore may inhibit Rho-kinase in the vascular wall as well as inflammatory cells, our findings with L-NAME suggest that the beneficial effects of Rho-kinase inhibition must occur downstream of NO.
In summary, our findings indicate that acute inhibition of Rho-kinase leads to cardiovascular protection mediated by the rapid activation of eNOS. It remains to be determined, however, whether targeting Rho-kinase in cardiovascular disease will yield therapeutic benefits in clinical trials.