Nitric Oxide Induces Dilation of Rat Aorta via Inhibition of Rho-Kinase Signaling

Methods

Results

Effect of Ca²⁺ Channel Inhibition on PE-Induced Contraction

To determine the contribution of membrane Ca²⁺ channels to the tonic component of agonist-induced contraction, endothelium-denuded aortic rings were first contracted with 10 μmol/L PE or 80 mmol/L KCl to achieve a similar extent of force generation and subsequently treated with a Ca²⁺ channel antagonist (nifedipine or diltiazem). Nifedipine was a significantly less effective inhibitor of force generation in response to PE compared with KCl. At the maximum concentration used, nifedipine relaxed the vessels contracted with PE only 19±6%, whereas the same concentration of nifedipine relaxed the vessels contracted with KCl 80±4%. Diltiazem was also a poor inhibitor of PE-induced contraction, relaxing the vessels contracted with PE only 11±3% at the maximum concentration used (Figure 1, top).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Top, Effect of Ca²⁺ channel antagonism on the force generation response to PE. Endothelium-denuded rat aortic rings were contracted with 10 μmol/L PE or 80 mmol/L KCl to the plateau phase and then treated with the Ca²⁺ channel antagonists nifedipine or diltiazem. Nifedipine was a significantly less effective inhibitor of the force generation response to PE compared with the force generation response to KCl. Diltiazem was also a poor inhibitor of PE-induced contraction. Data are expressed as mean±SEM (n≥4 per group). *P<0.05 compared with the KCl+nifedipine group (ANOVA). Bottom, Effect of endothelium removal on the response to Y-27632. Endothelium-intact and -denuded aortic rings were treated with increasing concentrations of Y-27632 after maximal contraction with PE. Endothelium-intact aortic rings (solid squares), compared with endothelium-denuded aortic rings (open circles), exhibited an increased vasodilator sensitivity to Y-27632. Data are expressed as mean±SEM (n≥4 per group). *P<0.05 compared with the endothelium-denuded group (ANOVA).

Effect of NO Removal on Response to Y-27632

The effect of NO on the response to Y-27632 was examined through endothelium removal or by the use of inhibitors of NO synthase and guanylate cyclase. Treatment with 1 μmol/L Y-27632 resulted in a significant inhibition of force generation to 10 μmol/L PE (compared with the initial control response to 10 μmol/L PE) in endothelium-intact rat aortic rings (Figure 2, top and middle). However, 1 μmol/L Y-27632 was a completely ineffective inhibitor of the contractile response to 10 μmol/L PE in endothelium-denuded vessels (Figure 2, middle). Additionally, in the presence 100 μmol/L L-NNA or (10 μmol/L) ODQ (to inhibit NO synthase or guanylate cyclase, respectively), 1 μmol/L Y-27632 was significantly less effective at inhibiting the contractile response to 10 μmol/L PE, suggesting that endogenous NO/cGMP signaling functions to decrease Rho-kinase activity (Figure 2, middle). In endothelium-denuded rings, 10 μmol/L or 100 μmol/L Y-27632 resulted in a concentration-dependent inhibition of the contractile response to 10 μmol/L PE (Figure 2, bottom).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Top, Representative tracing of the general protocol used. Rat aortic rings were contracted with 10 μmol/L PE and passively rinsed to baseline tension. Rings were then treated with Y-27632 for 20 minutes and contracted again with 10 μmol/L PE. Data are expressed as percentage of inhibition of PE-induced contraction after treatment with Y-27632. Middle, Effect of 1 μmol/L Y-27632 on the force generation response to PE. Y-27632 (1 μmol/L) was a significantly less effective inhibitor of the PE-induced contraction in endothelium-denuded rings (hatched bar) or endothelium-intact rings (solid bars) treated with L-NNA or ODQ compared with untreated endothelium-intact vessels. Data are mean±SEM (n≥4 per group) and are expressed as percentage of inhibition of PE-induced force generation after treatment with Y-27632 (compared with initial contraction to PE). *P<0.05 compared with the Y-27632 alone endothelium-intact group (Student t test with Bonferroni correction). Bottom, Concentration-dependent effect of Y-27632. Treatment with 1, 10, or 100 μmol/L Y-27632 resulted in a concentration-dependent inhibition of the PE-induced contraction in endothelium-denuded rat aorta. Data are mean±SEM (n≥4 per group) and are expressed as percentage of inhibition of PE-induced force generation after treatment with Y-27632 (compared with initial contraction to PE). *P<0.05 compared with the 1 μmol/L Y-27632 group, and #P<0.05 compared with the 10 μmol/L Y-27632 group (ANOVA).

Effect of Exogenous NO Addition on Response to Y-27632

The effect of exogenous NO on the response to Rho-kinase inhibition was examined. In the absence of endothelium, Y-27632 (1 μmol/L) was an ineffective inhibitor of PE-induced contraction (Figure 2, middle). However, in endothelium-denuded rings, treatment with the NO donor SNP (30 nmol/L) restored the ability of 1 μmol/L Y-27632 to inhibit significantly the response to PE (10 μmol/L), demonstrating that exogenous NO decreases Rho-kinase signaling (Figure 3, top). Treatment with 30 nmol/L SNP alone resulted in 37±6% inhibition of the contractile response to 10 μmol/L PE in endothelium-denuded rings.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Top, Effect of exogenous NO on the response to Y-27632. Endothelium-denuded vessels were contracted with 10 μmol/L PE and rinsed passively to baseline tension. Rings were then treated with 1 μmol/L Y-27632 in the presence (solid bar) or absence (hatched bar) of 30 nmol/L SNP and contracted again with PE. Treatment with 1 μmol/L Y-27632 resulted in a greater inhibition of the PE-induced contraction in the presence of SNP. Data are mean±SEM (n≥4 per group) and are expressed as percentage of inhibition of PE-induced force generation after treatment with Y-27632 (compared with initial contraction to PE). *P<0.05 (Student t test). Bottom, Effect of Rho-kinase inhibition on relaxation response to SNP. Endothelium-denuded rings were contracted with 10 μmol/L PE in the presence (open circles) or absence (solid squares) of Y-27632 before treatment with increasing concentrations of SNP. An increased vasodilatory sensitivity to SNP was seen in vessels pretreated with the Rho-kinase antagonist. Data are expressed as mean±SEM (n≥4 per group). *P<0.05 compared with control (ANOVA).

Effect of Endothelium Removal on Response to Wortmannin

To examine the specificity of the interaction of NO with the RhoA/Rho-kinase pathway, the effect of endothelium removal on an alternate signaling pathway was evaluated. Endothelium-intact and -denuded rat aortas were maximally contracted with 10 μmol/L PE and relaxed to baseline levels of passive force. Subsequently, the rings were treated with wortmannin (1 μmol/L), an MLC kinase antagonist, for 20 minutes, and the response to 10 μmol/L PE was repeated. The presence or absence of endothelium had no significant effect on the response to MLC kinase inhibition, inasmuch as treatment with 1 μmol/L wortmannin resulted in a similar percent inhibition of the PE-induced contraction (compared with the control response) in endothelium-intact and -denuded rings (Figure 4).

• Download figure
• Open in new tab
• Download powerpoint

Figure 4. Effect of endothelium removal on the response to MLC kinase inhibition. Endothelium-denuded (hatched bar) and endothelium-intact (solid bar) rat aortic rings were maximally contracted with 10 μmol/L PE and rinsed passively to baseline tension. Vessels were then treated with 1 μmol/L wortmannin and contracted again with PE. Wortmannin inhibited the contraction to PE to a similar extent in both endothelium-denuded and endothelium-intact vessels. Data are mean±SEM (n≥4 per group) and are expressed as percentage of inhibition of PE-induced force generation after treatment with wortmannin (compared with the initial contraction to PE). P=NS (Student t test).

Discussion

The findings of the present study confirm the role of Rho-kinase activity in the maintenance of agonist-induced force generation and further support the hypothesis that NO induces the relaxation of vascular smooth muscle through the inhibition of RhoA/Rho-kinase–mediated contraction. Much evidence suggests that the initial phase of agonist-induced force generation is stimulated by increases in intracellular Ca²⁺, whereas the maintained portion of contraction is regulated by a Ca²⁺-sensitization mechanism mediated by the activity of RhoA/Rho-kinase.^7,8,15⇓⇓ In contrast, depolarization-induced contraction is primarily maintained by a sustained increased in intracellular Ca²⁺.¹⁵ In confirmation, we found treatment with the Ca²⁺ channel antagonists diltiazem and nifedipine (a nonselective and L-type Ca²⁺ channel antagonist, respectively) to have little effect on the plateau phase of PE-induced contraction compared KCl-induced contraction (Figure 1, top). These data suggest that Ca²⁺ influx through membrane Ca²⁺ channels does not play a significant role in the sustained contraction to α-adrenergic agonists. However, in agreement with other reports,¹² we found that treatment with the selective inhibitor of Rho-kinase, Y-27632, completely relaxed rat aortic rings contracted with PE, indicative of the role of Rho-kinase activity in agonist-induced force generation (Figure 1, bottom). Interestingly, endothelium-intact rings exhibited an increased vasodilator sensitivity to Y-27632, suggesting an interaction between an endothelium-derived agent and Rho-kinase activity (Figure 1, bottom).

We hypothesized that the endothelium-derived agent is NO and that the principle endogenous mechanism of NO-mediated vasodilation in intact rat aortas is the inhibition of Rho-kinase activity. NO is known to stimulate an increase in cGMP levels, leading to the activation of cGK. Recent studies have reported recombinant cGK to phosphorylate RhoA, destabilizing the membrane binding of RhoA, thus providing a mechanism by which NO may inhibit RhoA/Rho-kinase activity.^5,6⇓ Additionally, cGK and the NO donor SNP have been shown to inhibit the agonist-induced migration of RhoA from the cytosolic to membrane cellular fractions, indicative of the inactivation of RhoA.^5,6⇓ In the present study, we used the Rho-kinase inhibitor Y-27632 to test the hypothesis that NO induces vasorelaxation through the inhibition of Rho-kinase–mediated Ca²⁺ sensitization. Y-27632 is a pyridine derivative that acts as a competitive inhibitor of the ATP-binding site on Rho-kinase and has been shown to be ≈200 to 2000 more specific for Rho-kinase than are MLC kinase, conventional protein kinase C, and cAMP-dependent protein kinase.^12,16⇓

Measuring the isometric force generation of isolated rat aorta, we found 1 μmol/L Y-27632 to inhibit the contraction to PE ≈60% in endothelium-intact vessels (Figure 2). In the absence of NO, however, a condition in which the NO-mediated inhibition of RhoA/Rho-kinase activity is removed, we expected a higher level of agonist-induced Rho-kinase activity. Y-27632 (1 μmol/L) was indeed found to be a significantly less effective inhibitor of the PE-induced force generation in the presence of NO synthase or guanylate cyclase inhibition or in endothelium-denuded rings, supporting the presence of increased Rho-kinase activity in the absence of NO (Figure 2). It is important to note that endothelium removal had no significant effect on the response to the MLC kinase inhibition with wortmannin (Figure 4; the concentration of wortmannin used has been reported to have selectivity for MLC kinase¹⁷; however, wortmannin has also been shown to inhibit the activity of PI3 kinase).

In further support of our hypothesis, we found the exogenous addition of NO with SNP to restore the effect of 1 μmol/L Y-27632 to inhibit PE-induced contraction in endothelium-denuded rings (Figure 3, top). Moreover, an increased vasodilatory sensitivity to SNP was observed in endothelium-denuded aortic rings treated with Y-27632, further demonstrating that the inhibition of Rho-kinase promotes the relaxant effects of NO (Figure 3, bottom).

Altogether, these data support the hypothesis that the principle endogenous mechanism of NO-mediated vasodilation in intact rat aortas is the inhibition of Rho-kinase activity. This novel signaling pathway may add to our understanding of the mechanisms of vasodilation in both physiological and pathological conditions. Numerous studies have demonstrated the attenuation of endothelium-dependent vasodilation in various hypertensive models, which is often associated with a decrease in NO bioavailability.^18,19⇓ On the basis of the findings of the present study and others (supporting the hypothesis that NO inhibits RhoA/Rho-kinase constrictor activity), it is tempting to speculate that decreased NO bioavailability leads to an increase in RhoA/Rho-kinase constrictor activity. Elevated Rho-kinase activity may then mediate the increased vasoconstrictor sensitivity seen in various hypertensive animal models. Indeed, in vivo and in vitro evidence supports an increase in Rho-kinase activity in spontaneously hypertensive, deoxycorticosterone-salt, and renal hypertensive rats.^12,20⇓ Uehata et al¹² found that the in vivo treatment of rats with Y-27632 significantly decreased blood pressure in hypertensive but not normotensive rats. Additionally, arteries from mineralocorticoid and angiotensin II hypertensive rats exhibit increased vasodilator sensitivity to Y-27632.^12,20,21⇓⇓

In summary, the findings of the present study support the hypothesis that endogenous NO-induced relaxation occurs principally through the inhibition of RhoA/Rho-kinase activity in the intact rat aorta, providing novel insight into a prominent vasodilation pathway.