Role of Endothelium-Derived Hyperpolarizing Factor in ACE Inhibitor-Induced Renal Vasodilation in Vivo

Methods

All experimental procedures in this study complied with the institutional guideline of the Animal Care Committee. Twenty adult male mongrel dogs (10 to 15 kg) were used for the study, with some of the animals undergoing multiple protocols. The dogs were fed a standard diet (Oriental Yeast, Tokyo) and were anesthetized with sodium pentobarbital (30 mg/kg). Animal preparations for study were detailed in our previous publications.^6,14 An electromagnetic flow probe was placed around the renal artery for measurement of renal blood flow (RBF). Data on systemic and renal hemodynamics, including mean arterial pressure (MAP), heart rate (HR), and RBF, were recorded.^16,17

Results

Effect of Cilazaprilat in the Presence of ARB

Cilazaprilat caused a decrease in MAP (Table 1, control) and elicited marked increases in diameters of superficial (afferent, from 16.5±0.4 to 19.3±0.2 μm, P<0.01, n=6; efferent, from 15.1±0.4 to 19.2±0.3 μm, P<0.01, n=6) and juxtamedullary arterioles (afferent, from 20.0±0.4 to 26.0±0.9 μm, P<0.01, n=6; efferent, from 18.4±0.3 to 25.4±0.8 μm, P<0.01, n=6; filled circles; Figure 1) and an elevation in RBF (Table 1).

View this table:

• View inline
• View popup

TABLE 1. Effect of Cilazaprilat on Systemic and Renal Hemodynamics

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Effects of cilazaprilat on renal microvascular diameters in superficial and juxtamedullary nephrons. Cilazaprilat elicited marked dilation of afferent and efferent arterioles (filled circles; control). In the presence of E4177, this vasodilator action was abolished in superficial afferent arterioles and diminished in juxtamedullary arterioles and superficial efferent arterioles (triangles). Finally, the cilazaprilat-induced dilator action observed during the blockade of angiotensin II was completely abrogated by the pretreatment with a bradykinin receptor antagonist, N_α-adamantaneacetyl-d-Arg-[Hyp³, Thi^5,8, D-Phe⁷]bradykinin (open circles). Results are mean±SEM. *P<0.05 vs baseline; **P<0.01 vs baseline. †P<0.05 vs control

The pretreatment with E4177 decreased MAP and increased RBF (Table 1) and renal arteriolar diameters (Table 2). Furthermore, E4177 abolished the cilazaprilat-induced increases in diameters of superficial afferent arterioles (from 18.4±0.4 to 19.1±0.3 μm, n=6; triangles, Figure 1) and the decrease in MAP (Table 1). Similarly, in the presence of E4177, the cilazaprilat-induced vasodilation of superficial efferent arterioles (from 18.2±0.2 to 19.7±0.2 μm, P<0.05, n=8), juxtamedullary afferent arterioles (from 24.0±0.2 to 28.0±0.8 μm, P<0.05, n=8), and juxtamedullary efferent arterioles (from 24.2±0.6 to 27.7±0.7 μm, P<0.05, n=8), as well as the increase in RBF, were diminished when compared with those in the absence of E4177 (8%±3% versus 27±5%, 17%±2% versus 30%±5%, 14%±5% versus 38%±6%, and 7%±2% versus 14%±3%, respectively; P<0.05). Thus, these results precisely coincided with our previous observations.⁶

View this table:

• View inline
• View popup

TABLE 2. Baseline Arteriolar Diameters in the Absence or Presence of Various Inhibitors

The baseline MAP, RBF, and renal arteriolar diameters in the presence of E4177 and NAAB were nearly the same as those in the presence of E4177 alone (P>0.5; Tables 1 and 2⇑). Nevertheless, the combined pretreatment with E4177 plus NAAB completely prevented the cilazaprilat-induced changes in vessel diameter (n=5 for each arteriole; open circles, Figure 1) and RBF (n=7, Table 1).

Role of NO in Cilazaprilat-Induced Vasodilation

When compared with the responses in the presence of E4177 alone, the combined pretreatment with E4177 plus l-NAME elevated MAP and elicited marked reductions in RBF (Table 1) and diameters of all arterioles (Table 2). Furthermore, this pretreatment prevented the cilazaprilat-induced changes in RBF (Table 1) and diameters of superficial efferent arterioles (from 12.2±0.7 to 13.1±0.6 μm, n=5, P>0.5; filled triangles in Figure 2) and juxtamedullary efferent arterioles (from 16.6±0.3 to 16.9±0.3 μm, n=5, P>0.5). Similar to the effect in the presence E4177 alone, cilazaprilat did not alter the superficial afferent arteriolar diameter during the combined pretreatment with E4177 and l-NAME.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Role of nitric oxide and/or EDHF in cilazaprilat-induced vasodilation of superficial and juxtamedullary renal arterioles. The cilazaprilat-induced dilation of superficial and juxtamedullary efferent arterioles observed in the presence of E4177 (open triangles, data duplicated from Figure 1) was completely abolished by the additional pretreatment with nitro-l-arginine methylester (l-NAME) (filled triangles). In juxtamedullary afferent arterioles, the pretreatment with E4177 and l-NAME partially prevented the cilazaprilat-induced vasodilation (filled triangles), and the remaining vasodilator response to cilazaprilat was totally abolished by the further treatment with proadifen (squares). Results are mean±SEM. *P<0.05 vs baseline; †P<0.05 vs E4177.

In the presence of E4177 and l-NAME, juxtamedullary afferent arterioles exhibited a markedly diminished vasodilator response to cilazaprilat (from 18.0±0.4 to 19.6±0.3 μm, n=5, P<0.05; filled triangles, Figure 2) when compared with that in the presence of E4177 alone (open triangles; 8%±3% versus 17±2%, P<0.05). Of note, however, the vasodilator response to cilazaprilat was still retained even under the dual blockade of angiotensin II and NO.

Cilazaprilat-Induced Changes in Renal Bradykinin and NOx Contents

Figure 3 illustrates the effect of cilazaprilat on renal cortical and medullary contents of bradykinin and NOx during the treatment with E4177 or E4177 plus l-NAME. In the absence of E4177 and l-NAME, cilazaprilat elevated renal bradykinin and NOx contents, with greater increments observed in the medulla (n=5) than in the cortex (n=5, filled circles). The pretreatment with E4177 had no effect on the baseline level or the cilazaprilat-induced changes in these substances (n=5, open triangles). Similarly, the combined treatment with E4177 plus l-NAME did not alter the baseline levels or the cilazaprilat-induced changes in renal bradykinin contents (n=5, filled triangles). In contrast, this pretreatment markedly reduced baseline NOx levels and completely prevented the cilazaprilat-induced increase in renal NOx contents.

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Effects of cilazaprilat on renal bradykinin and NOx contents in the cortex and medulla. Cilazaprilat elevated renal contents of bradykinin and NOx (filled circles; control). In the presence of E4177 (open triangles), the cilazaprilat-induced changes in renal bradykinin and NOx were nearly the same as those in the absence of E4177. In the presence of E4177 plus nitro-l-arginine methylester (l-NAME; filled triangles), cilazaprilat increased renal bradykinin, with the increment greater in the medulla than in the cortex. The pretreatment with E4177 plus l-NAME markedly reduced baseline levels of renal NOx contents and prevented the cilazaprilat-induced increase in renal NOx contents. Results are mean±SEM. *P<0.05 vs baseline. †P<0.05 vs control.

In summary, cilazaprilat-induced blockade of angiotensin II activity appears to be a dominant mechanism in superficial afferent arterioles, whereas bradykinin-induced NO dilates superficial efferent arterioles and juxtamedullary afferent and efferent arterioles. Finally, in juxtamedullary afferent arterioles, bradykinin can produce an additional vasodilator factor, most likely an EDHF.

Discussion

It has been well established that bradykinin and NO contribute substantially to the regulation of the renal microcirculation,^5–8,15 and the renal action of the ACE inhibitor is attributed at least in part to these intrarenal vasoactive substances.^5–8 Kon et al⁷ demonstrated a predominant efferent arteriolar dilation by enalapril in superficial nephrons, and this action was offset by a bradykinin antagonist. Furthermore, a bradykinin antagonist was shown to have only a modest effect on cortical blood flow but caused a marked reduction in papillary blood flow that originates from juxtamedullary nephrons.⁵ Similarly, intrarenal NO serves to counter the angiotensin II-induced afferent arteriolar constriction of superficial nephrons, but this effect is absent in the efferent arteriole.²¹ In addition to the stimulation of NO synthesis, bradykinin is also demonstrated to produce EDHF, which subsequently inhibits voltage-dependent calcium channels and arteriolar constriction.¹² However, no investigations have been conducted examining the zonal (superficial versus juxtamedullary) and segmental (afferent versus efferent) heterogeneity in the action of the ACE inhibitor and its vasodilator mechanisms in the same experimental setting.

The present study demonstrates that multiple mechanisms contribute to the ACE inhibitor-induced vasodilation of the renal microcirculation in an in vivo and in situ setting, with the use of an intravital needle-type CCD camera videomicroscopy technique.^6,14,18 This experimental system is unique in that direct observation of relatively intact renal microvascular effect is readily available by the insertion of a needle-type CCD camera probe without impairment in renal microvascular integrity or disruption of tubuloglomerular feedback mechanism.¹⁸ Using this system, we have observed that the ACE inhibitor not only blocks angiotensin II activity by reducing its synthesis but also enhances the renal bradykinin activity, with the increment greater in the medulla (Figure 3). This action is particularly important in the superficial efferent arterioles, juxtamedullary afferent arterioles and juxtamedullary efferent arterioles, with its contribution greater in juxtamedullary nephrons (Figure 1). In this regard, it was shown that although bradykinin caused a relatively selective dilator action on the efferent arteriole,^7,22 higher concentrations of bradykinin could dilate afferent and efferent arterioles.^6,23 Furthermore, because we have shown that the NO inhibition by l-NAME blunts or abolishes the cilazaprilat-induced renal vasodilation and the increase in renal NOx level during the blockade of angiotensin II activity (Figure 2 and 3⇑), it is most likely that NO conveys a vasodilatory signal for bradykinin and plays an important role in mediating the ACE inhibitor-induced renal vasodilation. Thus, the present observations extend previous findings with regard to vasodilator mechanisms for the ACE inhibitor and further provide detailed information on the intrarenal action of the ACE inhibitor. Because E4177 completely abolished the cilazaprilat-induced changes in MAP, the alterations in afferent arteriolar diameter and RBF are mediated in large part by the direct vascular action of cilazaprilat, per se, but are not related to the autoregulatory (eg, myogenic) response of renal arterioles.

Although the present study clearly demonstrates an important role of NO in mediating the ACE inhibitor-induced vasodilation, there still remains a vasodilator component during the blockade of NO and angiotensin II. Thus, in the presence of E4177 and l-NAME, cilazaprilat increased renal bradykinin contents but did not alter the renal NOx level (Figure 3). In this setting, cilazaprilat failed to dilate superficial arterioles or juxtamedullary efferent arterioles (Figure 2). Nevertheless, this agent caused a significant dilation of juxtamedullary afferent arterioles, although the degree of the cilazaprilat-induced dilation was diminished compared with that in the presence of E4177 alone (8%±3% versus 17±2%). In concert with the fact that the bradykinin blockade completely inhibited the cilazaprilat-induced dilation in the presence of E4177 (Figure 1), it follows that the remaining vasodilator activity is coupled with additional mechanisms associated with bradykinin but is independent of NO.

In this regard, several in vitro studies suggest the involvement of EDHF in the renal arteriolar action of bradykinin.^24–26 In the present in vivo study, we found that proadifen completely eliminated the remaining vasodilator response of juxtamedullary afferent arterioles observed during the inhibition of angiotensin II and NO. These findings therefore are compatible with previous in vitro observations that a part of the ACE inhibitor-induced vasodilation is associated with EDHF^27,28 and further allow extrapolation to the in vivo actions of this agent. Caveat is in order, because proadifen is reported to inhibit the K channel activity and the cytochrome P450-mediated EDHF production.²⁹ It is of greater importance, however, that the present observation clearly demonstrates in an in vivo and in situ setting that the ACE inhibitor is capable of exerting additional vasodilator actions distinct from NO or the inhibition of angiotensin II. Finally, although the current study does not evaluate the role of prostaglandins, the complete inhibition of cilazaprilat-induced vasodilation by the combined treatment with E4177, l-NAME, and proadifen (Figure 2) suggests that the contribution of prostaglandins to the ACE inhibitor-induced renal arteriolar vasodilation is modest in our experimental setting.

A growing body of evidence has accrued that renal arterioles produce EDHF or EDHF-like substances.^{12,24–26,30–32} In renal arterioles of the rat, acetylcholine-induced afferent arteriolar dilation is attributed in part to EDHF.^12,29,30 Furthermore, bradykinin is demonstrated to elicit afferent arteriolar vasodilation that is mediated in part by EDHF in the isolated perfused rat hydronephrotic kidney²⁶ and in the rat juxtamedullary nephron preparation,³² which therefore is in agreement with our current finding that the vasodilator action of cilazaprilat during the blockade of angiotensin II and NO is restricted to afferent arterioles. Of note, Wang et al³⁰ reported that the efferent arteriole does not exhibit an EDHF-like response to acetylcholine, because the nature of EDHF requires hyperpolarization for its activity on the vascular smooth muscle, which does not appear to affect the efferent arteriolar tone.³³ In contrast, Ren et al²⁵ demonstrated a contribution of cytochrome P450 metabolites as a vasodilator for efferent arterioles. These discrepant observations could be caused by different experimental models used.

In addition to the segmental heterogeneity in the arteriolar response to bradykinin, the present study shows zonal (superficial versus juxtamedullary) differences in the vasodilator mechanism of the ACE inhibitor. Thus, when the effect of angiotensin II is eliminated, the remaining vasodilator action is greater in the juxtamedullary than in the superficial nephrons (Figure 1), and is most likely ascribed to the action of bradykinin and NO. These findings are consistent with the formulation that the medulla contains a greater amount of bradykinin and NO than the cortex in both basal^6,13 and ACE inhibitor-stimulated conditions,^6,34 and are associated with additive natriuresis without changes in GFR that we observed in our previous study.⁶ Furthermore, to the extent that afferent arterioles are reported to dilate in response to bradykinin, and to the extent that a part of this vasodilation is mediated by the EDHF-like activity,^26,32 it is conjectured that the ACE inhibitor-induced elevation in the bradykinin content does reach the level required to elicit the EDHF-mediated vasodilation in the juxtamedullary, but not superficial, cortex in this present study. Of course, this speculation warrants further investigations.

Practical implication of EDHF in normal and diseased kidneys merits comment. ACE inhibitors are characterized by the augmenting action of bradykinin and NO, which contributes to the vasodilator action of the ACE inhibitor in addition to the angiotensin blockade.⁶ In renal injury, however, renal NO production is reported to be suppressed by a variety of mechanisms, including increased levels of free radicals³⁵ and asymmetrical dimethylarginine (an endogenous inhibitor of NO synthase).³⁶ In this setting, several studies demonstrate that the role of EDHF is maintained³⁷ or rather increased.³⁸ It is intriguing, therefore, to surmise that apart from angiotensin II inhibition as a main action, the mechanism of the renal action of ACE inhibitors is shifted from NO to EDHF, and the role of EDHF may be exaggerated in chronic renal disease in which the ACE inhibitor constitutes a primary tool for its treatment.